From 8de35efb6afa75f25415d54e21221eea759db261 Mon Sep 17 00:00:00 2001 From: Russell King Date: Mon, 20 Nov 2006 15:59:10 +0000 Subject: [ARM] ebsa110: fix warnings generated by asm/arch/io.h Remove two warnings: drivers/serial/8250_early.c:136: warning: unused variable 'mapsize' include/linux/io.h:47: warning: passing argument 1 of '__readb' discards qualifiers from pointer target type Signed-off-by: Russell King --- include/asm-arm/arch-ebsa110/io.h | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/asm-arm/arch-ebsa110/io.h b/include/asm-arm/arch-ebsa110/io.h index ae048441c9e..722c5e08628 100644 --- a/include/asm-arm/arch-ebsa110/io.h +++ b/include/asm-arm/arch-ebsa110/io.h @@ -27,9 +27,9 @@ void __outw(u16 val, unsigned int port); u32 __inl(unsigned int port); void __outl(u32 val, unsigned int port); -u8 __readb(void __iomem *addr); -u16 __readw(void __iomem *addr); -u32 __readl(void __iomem *addr); +u8 __readb(const volatile void __iomem *addr); +u16 __readw(const volatile void __iomem *addr); +u32 __readl(const volatile void __iomem *addr); void __writeb(u8 val, void __iomem *addr); void __writew(u16 val, void __iomem *addr); @@ -64,8 +64,14 @@ void __writel(u32 val, void __iomem *addr); #define writew(v,b) __writew(v,b) #define writel(v,b) __writel(v,b) -#define __arch_ioremap(cookie,sz,c) ((void __iomem *)(cookie)) -#define __arch_iounmap(cookie) do { } while (0) +static inline void __iomem *__arch_ioremap(unsigned long cookie, size_t size, + unsigned int flags) +{ + return (void __iomem *)cookie; +} + +#define __arch_ioremap __arch_ioremap +#define __arch_iounmap(cookie) do { } while (0) extern void insb(unsigned int port, void *buf, int sz); extern void insw(unsigned int port, void *buf, int sz); -- cgit v1.2.3-70-g09d2 From 105ef9a0af1469a9dd906839dd4628aa9c013f58 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Tue, 21 Nov 2006 22:57:23 +0100 Subject: [ARM] 3942/1: ARM: comment: consistent_sync should not be called directly /* * Note: Drivers should NOT use this function directly, as it will break * platforms with CONFIG_DMABOUNCE. * Use the driver DMA support - see dma-mapping.h (dma_sync_*) */ Signed-off-by: Dan Williams Signed-off-by: Russell King --- arch/arm/mm/consistent.c | 3 +++ include/asm-arm/dma-mapping.h | 4 ++++ 2 files changed, 7 insertions(+) (limited to 'include') diff --git a/arch/arm/mm/consistent.c b/arch/arm/mm/consistent.c index 50e6b6bfb2e..b797217e82b 100644 --- a/arch/arm/mm/consistent.c +++ b/arch/arm/mm/consistent.c @@ -476,6 +476,9 @@ core_initcall(consistent_init); /* * Make an area consistent for devices. + * Note: Drivers should NOT use this function directly, as it will break + * platforms with CONFIG_DMABOUNCE. + * Use the driver DMA support - see dma-mapping.h (dma_sync_*) */ void consistent_sync(void *vaddr, size_t size, int direction) { diff --git a/include/asm-arm/dma-mapping.h b/include/asm-arm/dma-mapping.h index 55eb4dc3253..666617711c8 100644 --- a/include/asm-arm/dma-mapping.h +++ b/include/asm-arm/dma-mapping.h @@ -12,6 +12,10 @@ * uncached, unwrite-buffered mapped memory space for use with DMA * devices. This is the "generic" version. The PCI specific version * is in pci.h + * + * Note: Drivers should NOT use this function directly, as it will break + * platforms with CONFIG_DMABOUNCE. + * Use the driver DMA support - see dma-mapping.h (dma_sync_*) */ extern void consistent_sync(void *kaddr, size_t size, int rw); -- cgit v1.2.3-70-g09d2 From ec7080d185a9b79581bf1dbe300e877719c0b1a9 Mon Sep 17 00:00:00 2001 From: Manuel Lauss Date: Wed, 22 Nov 2006 14:51:32 +0100 Subject: [PATCH] make au1xxx-ide compile again The Au1xx IDE controller driver doesn't compile: CC drivers/ide/mips/au1xxx-ide.o /linux-2.6.19-rc6-work/drivers/ide/mips/au1xxx-ide.c:480: error: conflicting types for 'auide_ddma_tx_callback' include2/asm/mach-au1x00/au1xxx_ide.h:174: error: previous declaration of 'auide_ddma_tx_callback' was here /linux-2.6.19-rc6-work/drivers/ide/mips/au1xxx-ide.c:486: error: conflicting types for 'auide_ddma_rx_callback' include2/asm/mach-au1x00/au1xxx_ide.h:176: error: previous declaration of 'auide_ddma_rx_callback' was here Signed-off-by: Manuel Lauss Signed-off-by: Ralf Baechle --- include/asm-mips/mach-au1x00/au1xxx_ide.h | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/asm-mips/mach-au1x00/au1xxx_ide.h b/include/asm-mips/mach-au1x00/au1xxx_ide.h index 301e7130077..e9fa252f8a3 100644 --- a/include/asm-mips/mach-au1x00/au1xxx_ide.h +++ b/include/asm-mips/mach-au1x00/au1xxx_ide.h @@ -170,10 +170,8 @@ int __init auide_probe(void); static int auide_dma_host_on(ide_drive_t *drive); static int auide_dma_lostirq(ide_drive_t *drive); static int auide_dma_on(ide_drive_t *drive); - static void auide_ddma_tx_callback(int irq, void *param, - struct pt_regs *regs); - static void auide_ddma_rx_callback(int irq, void *param, - struct pt_regs *regs); + static void auide_ddma_tx_callback(int irq, void *param); + static void auide_ddma_rx_callback(int irq, void *param); static int auide_dma_off_quietly(ide_drive_t *drive); #endif /* end CONFIG_BLK_DEV_IDE_AU1XXX_MDMA2_DBDMA */ -- cgit v1.2.3-70-g09d2 From 701e054e0c2db82359f0454c7ed4fd24346d52eb Mon Sep 17 00:00:00 2001 From: Vasily Tarasov Date: Sat, 25 Nov 2006 11:09:22 -0800 Subject: [PATCH] mounstats NULL pointer dereference OpenVZ developers team has encountered the following problem in 2.6.19-rc6 kernel. After some seconds of running script while [[ 1 ]] do find /proc -name mountstats | xargs cat done this Oops appears: BUG: unable to handle kernel NULL pointer dereference at virtual address 00000010 printing eip: c01a6b70 *pde = 00000000 Oops: 0000 [#1] SMP Modules linked in: xt_length ipt_ttl xt_tcpmss ipt_TCPMSS iptable_mangle iptable_filter xt_multiport xt_limit ipt_tos ipt_REJECT ip_tables x_tables parport_pc lp parport sunrpc af_packet thermal processor fan button battery asus_acpi ac ohci_hcd ehci_hcd usbcore i2c_nforce2 i2c_core tg3 floppy pata_amd ide_cd cdrom sata_nv libata CPU: 1 EIP: 0060:[] Not tainted VLI EFLAGS: 00010246 (2.6.19-rc6 #2) EIP is at mountstats_open+0x70/0xf0 eax: 00000000 ebx: e6247030 ecx: e62470f8 edx: 00000000 esi: 00000000 edi: c01a6b00 ebp: c33b83c0 esp: f4105eb4 ds: 007b es: 007b ss: 0068 Process cat (pid: 6044, ti=f4105000 task=f4104a70 task.ti=f4105000) Stack: c33b83c0 c04ee940 f46a4a80 c33b83c0 e4df31b4 c01a6b00 f4105000 c0169231 e4df31b4 c33b83c0 c33b83c0 f4105f20 00000003 f4105000 c0169445 f2503cf0 f7f8c4c0 00008000 c33b83c0 00000000 00008000 c0169350 f4105f20 00008000 Call Trace: [] mountstats_open+0x0/0xf0 [] __dentry_open+0x181/0x250 [] nameidata_to_filp+0x35/0x50 [] do_filp_open+0x50/0x60 [] seq_read+0xc6/0x300 [] get_unused_fd+0x31/0xc0 [] do_sys_open+0x63/0x110 [] sys_open+0x27/0x30 [] sysenter_past_esp+0x56/0x79 ======================= Code: 45 74 8b 54 24 20 89 44 24 08 8b 42 f0 31 d2 e8 47 cb f8 ff 85 c0 89 c3 74 51 8d 80 a0 04 00 00 e8 46 06 2c 00 8b 83 48 04 00 00 <8b> 78 10 85 ff 74 03 f0 ff 07 b0 01 86 83 a0 04 00 00 f0 ff 4b EIP: [] mountstats_open+0x70/0xf0 SS:ESP 0068:f4105eb4 The problem is that task->nsproxy can be equal NULL for some time during task exit. This patch fixes the BUG. Signed-off-by: Vasily Tarasov Cc: Herbert Poetzl Cc: "Serge E. Hallyn" Cc: "Eric W. Biederman" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/base.c | 3 ++- include/linux/nsproxy.h | 4 +++- 2 files changed, 5 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/fs/proc/base.c b/fs/proc/base.c index 8df27401d29..795319c54f7 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -442,7 +442,8 @@ static int mountstats_open(struct inode *inode, struct file *file) if (task) { task_lock(task); - namespace = task->nsproxy->namespace; + if (task->nsproxy) + namespace = task->nsproxy->namespace; if (namespace) get_namespace(namespace); task_unlock(task); diff --git a/include/linux/nsproxy.h b/include/linux/nsproxy.h index f6baecdeecd..971d1c6dfc4 100644 --- a/include/linux/nsproxy.h +++ b/include/linux/nsproxy.h @@ -45,8 +45,10 @@ static inline void exit_task_namespaces(struct task_struct *p) { struct nsproxy *ns = p->nsproxy; if (ns) { - put_nsproxy(ns); + task_lock(p); p->nsproxy = NULL; + task_unlock(p); + put_nsproxy(ns); } } #endif -- cgit v1.2.3-70-g09d2 From ee3ce191e8eaa4cc15c51a28b34143b36404c4f5 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Sat, 25 Nov 2006 11:09:36 -0800 Subject: [PATCH] Enforce "unsigned long flags;" when spinlocking Make it break or warn if you pass to spin_lock_irqsave() and friends something different from "unsigned long flags;". Suprisingly large amount of these was caught by recent commit c53421b18f205c5f97c604ae55c6a921f034b0f6 and others. Idea is largely from FRV typechecking. Suggestions from Andrew Morton. All stupid typos in first version fixed. Passes allmodconfig on i386, x86_64, alpha, arm as well as my usual config. Note #1: checking with sparse is still needed, because a driver can save and pass around flags or something. So far patch is very intrusive. Note #2: techically, we should break only if sizeof(flags) < sizeof(unsigned long), however, the more pain for getting suspicious code into kernel, the better. Signed-off-by: Alexey Dobriyan Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/irqflags.h | 37 ++++++++++++++++++++++++++++----- include/linux/spinlock.h | 53 ++++++++++++++++++++++++++++++++++++++++-------- 2 files changed, 76 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/include/linux/irqflags.h b/include/linux/irqflags.h index 412e025bc5c..4fe740bf4ea 100644 --- a/include/linux/irqflags.h +++ b/include/linux/irqflags.h @@ -11,6 +11,12 @@ #ifndef _LINUX_TRACE_IRQFLAGS_H #define _LINUX_TRACE_IRQFLAGS_H +#define BUILD_CHECK_IRQ_FLAGS(flags) \ + do { \ + BUILD_BUG_ON(sizeof(flags) != sizeof(unsigned long)); \ + typecheck(unsigned long, flags); \ + } while (0) + #ifdef CONFIG_TRACE_IRQFLAGS extern void trace_hardirqs_on(void); extern void trace_hardirqs_off(void); @@ -50,10 +56,15 @@ #define local_irq_disable() \ do { raw_local_irq_disable(); trace_hardirqs_off(); } while (0) #define local_irq_save(flags) \ - do { raw_local_irq_save(flags); trace_hardirqs_off(); } while (0) + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + raw_local_irq_save(flags); \ + trace_hardirqs_off(); \ + } while (0) #define local_irq_restore(flags) \ do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ if (raw_irqs_disabled_flags(flags)) { \ raw_local_irq_restore(flags); \ trace_hardirqs_off(); \ @@ -69,8 +80,16 @@ */ # define raw_local_irq_disable() local_irq_disable() # define raw_local_irq_enable() local_irq_enable() -# define raw_local_irq_save(flags) local_irq_save(flags) -# define raw_local_irq_restore(flags) local_irq_restore(flags) +# define raw_local_irq_save(flags) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + local_irq_save(flags); \ + } while (0) +# define raw_local_irq_restore(flags) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + local_irq_restore(flags); \ + } while (0) #endif /* CONFIG_TRACE_IRQFLAGS_SUPPORT */ #ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT @@ -80,7 +99,11 @@ raw_safe_halt(); \ } while (0) -#define local_save_flags(flags) raw_local_save_flags(flags) +#define local_save_flags(flags) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + raw_local_save_flags(flags); \ + } while (0) #define irqs_disabled() \ ({ \ @@ -90,7 +113,11 @@ raw_irqs_disabled_flags(flags); \ }) -#define irqs_disabled_flags(flags) raw_irqs_disabled_flags(flags) +#define irqs_disabled_flags(flags) \ +({ \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + raw_irqs_disabled_flags(flags); \ +}) #endif /* CONFIG_X86 */ #endif diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index b800d2d68b3..54ad37089c4 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -52,6 +52,7 @@ #include #include #include +#include #include @@ -183,13 +184,37 @@ do { \ #define read_lock(lock) _read_lock(lock) #if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) -#define spin_lock_irqsave(lock, flags) flags = _spin_lock_irqsave(lock) -#define read_lock_irqsave(lock, flags) flags = _read_lock_irqsave(lock) -#define write_lock_irqsave(lock, flags) flags = _write_lock_irqsave(lock) +#define spin_lock_irqsave(lock, flags) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + flags = _spin_lock_irqsave(lock); \ + } while (0) +#define read_lock_irqsave(lock, flags) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + flags = _read_lock_irqsave(lock); \ + } while (0) +#define write_lock_irqsave(lock, flags) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + flags = _write_lock_irqsave(lock); \ + } while (0) #else -#define spin_lock_irqsave(lock, flags) _spin_lock_irqsave(lock, flags) -#define read_lock_irqsave(lock, flags) _read_lock_irqsave(lock, flags) -#define write_lock_irqsave(lock, flags) _write_lock_irqsave(lock, flags) +#define spin_lock_irqsave(lock, flags) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + _spin_lock_irqsave(lock, flags); \ + } while (0) +#define read_lock_irqsave(lock, flags) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + _read_lock_irqsave(lock, flags); \ + } while (0) +#define write_lock_irqsave(lock, flags) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + _write_lock_irqsave(lock, flags); \ + } while (0) #endif #define spin_lock_irq(lock) _spin_lock_irq(lock) @@ -225,15 +250,24 @@ do { \ #endif #define spin_unlock_irqrestore(lock, flags) \ - _spin_unlock_irqrestore(lock, flags) + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + _spin_unlock_irqrestore(lock, flags); \ + } while (0) #define spin_unlock_bh(lock) _spin_unlock_bh(lock) #define read_unlock_irqrestore(lock, flags) \ - _read_unlock_irqrestore(lock, flags) + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + _read_unlock_irqrestore(lock, flags); \ + } while (0) #define read_unlock_bh(lock) _read_unlock_bh(lock) #define write_unlock_irqrestore(lock, flags) \ - _write_unlock_irqrestore(lock, flags) + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + _write_unlock_irqrestore(lock, flags); \ + } while (0) #define write_unlock_bh(lock) _write_unlock_bh(lock) #define spin_trylock_bh(lock) __cond_lock(lock, _spin_trylock_bh(lock)) @@ -247,6 +281,7 @@ do { \ #define spin_trylock_irqsave(lock, flags) \ ({ \ + BUILD_CHECK_IRQ_FLAGS(flags); \ local_irq_save(flags); \ spin_trylock(lock) ? \ 1 : ({ local_irq_restore(flags); 0; }); \ -- cgit v1.2.3-70-g09d2 From cfd3ef2346f924d6c0e82236c20fdb3a8840136a Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Sat, 25 Nov 2006 11:09:37 -0800 Subject: [PATCH] lockdep: spin_lock_irqsave_nested() Introduce spin_lock_irqsave_nested(); implementation from: http://lkml.org/lkml/2006/6/1/122 Patch from: http://lkml.org/lkml/2006/9/13/258 [akpm@osdl.org: two compile fixes] Signed-off-by: Arjan van de Ven Signed-off-by: Jiri Kosina Signed-off-by: Peter Zijlstra Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/spinlock.h | 18 ++++++++++++++++++ include/linux/spinlock_api_smp.h | 2 ++ kernel/spinlock.c | 21 +++++++++++++++++++++ 3 files changed, 41 insertions(+) (limited to 'include') diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index 54ad37089c4..57f670d78f7 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -199,6 +199,21 @@ do { \ BUILD_CHECK_IRQ_FLAGS(flags); \ flags = _write_lock_irqsave(lock); \ } while (0) + +#ifdef CONFIG_DEBUG_LOCK_ALLOC +#define spin_lock_irqsave_nested(lock, flags, subclass) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + flags = _spin_lock_irqsave_nested(lock, subclass); \ + } while (0) +#else +#define spin_lock_irqsave_nested(lock, flags, subclass) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + flags = _spin_lock_irqsave(lock); \ + } while (0) +#endif + #else #define spin_lock_irqsave(lock, flags) \ do { \ @@ -215,6 +230,9 @@ do { \ BUILD_CHECK_IRQ_FLAGS(flags); \ _write_lock_irqsave(lock, flags); \ } while (0) +#define spin_lock_irqsave_nested(lock, flags, subclass) \ + spin_lock_irqsave(lock, flags) + #endif #define spin_lock_irq(lock) _spin_lock_irq(lock) diff --git a/include/linux/spinlock_api_smp.h b/include/linux/spinlock_api_smp.h index 8828b8155e9..8a2307ce729 100644 --- a/include/linux/spinlock_api_smp.h +++ b/include/linux/spinlock_api_smp.h @@ -32,6 +32,8 @@ void __lockfunc _read_lock_irq(rwlock_t *lock) __acquires(lock); void __lockfunc _write_lock_irq(rwlock_t *lock) __acquires(lock); unsigned long __lockfunc _spin_lock_irqsave(spinlock_t *lock) __acquires(lock); +unsigned long __lockfunc _spin_lock_irqsave_nested(spinlock_t *lock, int subclass) + __acquires(lock); unsigned long __lockfunc _read_lock_irqsave(rwlock_t *lock) __acquires(lock); unsigned long __lockfunc _write_lock_irqsave(rwlock_t *lock) diff --git a/kernel/spinlock.c b/kernel/spinlock.c index 476c3741511..2c6c2bf8551 100644 --- a/kernel/spinlock.c +++ b/kernel/spinlock.c @@ -293,6 +293,27 @@ void __lockfunc _spin_lock_nested(spinlock_t *lock, int subclass) } EXPORT_SYMBOL(_spin_lock_nested); +unsigned long __lockfunc _spin_lock_irqsave_nested(spinlock_t *lock, int subclass) +{ + unsigned long flags; + + local_irq_save(flags); + preempt_disable(); + spin_acquire(&lock->dep_map, subclass, 0, _RET_IP_); + /* + * On lockdep we dont want the hand-coded irq-enable of + * _raw_spin_lock_flags() code, because lockdep assumes + * that interrupts are not re-enabled during lock-acquire: + */ +#ifdef CONFIG_PROVE_SPIN_LOCKING + _raw_spin_lock(lock); +#else + _raw_spin_lock_flags(lock, &flags); +#endif + return flags; +} + +EXPORT_SYMBOL(_spin_lock_irqsave_nested); #endif -- cgit v1.2.3-70-g09d2 From dc9b334622bff6d22456917a034c2e2d194b9328 Mon Sep 17 00:00:00 2001 From: Paul Bonser Date: Thu, 23 Nov 2006 17:56:13 -0800 Subject: [NET]: Re-fix of doc-comment in sock.h Restoring old, correct comment for sk_filter_release, moving it to where it should actually be, and changing new comment into proper comment for sk_filter_rcu_free, where it actually makes sense. The original fix submitted for this on Oct 23 mistakenly documented the wrong function. Signed-off-by: Paul Bonser Signed-off-by: David S. Miller --- include/net/sock.h | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/net/sock.h b/include/net/sock.h index ac286a35303..9cdbae2a53a 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -883,18 +883,23 @@ static inline int sk_filter(struct sock *sk, struct sk_buff *skb) } /** - * sk_filter_release: Release a socket filter - * @rcu: rcu_head that contains the sk_filter info to remove - * - * Remove a filter from a socket and release its resources. + * sk_filter_rcu_free: Free a socket filter + * @rcu: rcu_head that contains the sk_filter to free */ - static inline void sk_filter_rcu_free(struct rcu_head *rcu) { struct sk_filter *fp = container_of(rcu, struct sk_filter, rcu); kfree(fp); } +/** + * sk_filter_release: Release a socket filter + * @sk: socket + * @fp: filter to remove + * + * Remove a filter from a socket and release its resources. + */ + static inline void sk_filter_release(struct sock *sk, struct sk_filter *fp) { unsigned int size = sk_filter_len(fp); -- cgit v1.2.3-70-g09d2 From b8e6ec865fd1d8838b6ce9516977b65e9f08f876 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 26 Nov 2006 16:27:17 -0800 Subject: Revert "[PATCH] Enforce "unsigned long flags;" when spinlocking" This reverts commit ee3ce191e8eaa4cc15c51a28b34143b36404c4f5, since it broke on at least ARM, MIPS and PA-RISC due to complicated header file dependencies. Conflicts in include/linux/spinlock.h (due to the "nested" variety fixes) fixed up by hand. Cc: Alexey Dobriyan Cc: Ralf Baechle Cc: Kyle McMartin Cc: Russell King Signed-off-by: Linus Torvalds --- include/linux/irqflags.h | 37 ++++---------------------- include/linux/spinlock.h | 69 +++++++++++------------------------------------- 2 files changed, 20 insertions(+), 86 deletions(-) (limited to 'include') diff --git a/include/linux/irqflags.h b/include/linux/irqflags.h index 4fe740bf4ea..412e025bc5c 100644 --- a/include/linux/irqflags.h +++ b/include/linux/irqflags.h @@ -11,12 +11,6 @@ #ifndef _LINUX_TRACE_IRQFLAGS_H #define _LINUX_TRACE_IRQFLAGS_H -#define BUILD_CHECK_IRQ_FLAGS(flags) \ - do { \ - BUILD_BUG_ON(sizeof(flags) != sizeof(unsigned long)); \ - typecheck(unsigned long, flags); \ - } while (0) - #ifdef CONFIG_TRACE_IRQFLAGS extern void trace_hardirqs_on(void); extern void trace_hardirqs_off(void); @@ -56,15 +50,10 @@ #define local_irq_disable() \ do { raw_local_irq_disable(); trace_hardirqs_off(); } while (0) #define local_irq_save(flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - raw_local_irq_save(flags); \ - trace_hardirqs_off(); \ - } while (0) + do { raw_local_irq_save(flags); trace_hardirqs_off(); } while (0) #define local_irq_restore(flags) \ do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ if (raw_irqs_disabled_flags(flags)) { \ raw_local_irq_restore(flags); \ trace_hardirqs_off(); \ @@ -80,16 +69,8 @@ */ # define raw_local_irq_disable() local_irq_disable() # define raw_local_irq_enable() local_irq_enable() -# define raw_local_irq_save(flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - local_irq_save(flags); \ - } while (0) -# define raw_local_irq_restore(flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - local_irq_restore(flags); \ - } while (0) +# define raw_local_irq_save(flags) local_irq_save(flags) +# define raw_local_irq_restore(flags) local_irq_restore(flags) #endif /* CONFIG_TRACE_IRQFLAGS_SUPPORT */ #ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT @@ -99,11 +80,7 @@ raw_safe_halt(); \ } while (0) -#define local_save_flags(flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - raw_local_save_flags(flags); \ - } while (0) +#define local_save_flags(flags) raw_local_save_flags(flags) #define irqs_disabled() \ ({ \ @@ -113,11 +90,7 @@ raw_irqs_disabled_flags(flags); \ }) -#define irqs_disabled_flags(flags) \ -({ \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - raw_irqs_disabled_flags(flags); \ -}) +#define irqs_disabled_flags(flags) raw_irqs_disabled_flags(flags) #endif /* CONFIG_X86 */ #endif diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index 57f670d78f7..8451052ca66 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -52,7 +52,6 @@ #include #include #include -#include #include @@ -184,52 +183,24 @@ do { \ #define read_lock(lock) _read_lock(lock) #if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) -#define spin_lock_irqsave(lock, flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - flags = _spin_lock_irqsave(lock); \ - } while (0) -#define read_lock_irqsave(lock, flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - flags = _read_lock_irqsave(lock); \ - } while (0) -#define write_lock_irqsave(lock, flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - flags = _write_lock_irqsave(lock); \ - } while (0) + +#define spin_lock_irqsave(lock, flags) flags = _spin_lock_irqsave(lock) +#define read_lock_irqsave(lock, flags) flags = _read_lock_irqsave(lock) +#define write_lock_irqsave(lock, flags) flags = _write_lock_irqsave(lock) #ifdef CONFIG_DEBUG_LOCK_ALLOC -#define spin_lock_irqsave_nested(lock, flags, subclass) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - flags = _spin_lock_irqsave_nested(lock, subclass); \ - } while (0) +#define spin_lock_irqsave_nested(lock, flags, subclass) \ + flags = _spin_lock_irqsave_nested(lock, subclass) #else -#define spin_lock_irqsave_nested(lock, flags, subclass) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - flags = _spin_lock_irqsave(lock); \ - } while (0) +#define spin_lock_irqsave_nested(lock, flags, subclass) \ + flags = _spin_lock_irqsave(lock) #endif #else -#define spin_lock_irqsave(lock, flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - _spin_lock_irqsave(lock, flags); \ - } while (0) -#define read_lock_irqsave(lock, flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - _read_lock_irqsave(lock, flags); \ - } while (0) -#define write_lock_irqsave(lock, flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - _write_lock_irqsave(lock, flags); \ - } while (0) + +#define spin_lock_irqsave(lock, flags) _spin_lock_irqsave(lock, flags) +#define read_lock_irqsave(lock, flags) _read_lock_irqsave(lock, flags) +#define write_lock_irqsave(lock, flags) _write_lock_irqsave(lock, flags) #define spin_lock_irqsave_nested(lock, flags, subclass) \ spin_lock_irqsave(lock, flags) @@ -268,24 +239,15 @@ do { \ #endif #define spin_unlock_irqrestore(lock, flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - _spin_unlock_irqrestore(lock, flags); \ - } while (0) + _spin_unlock_irqrestore(lock, flags) #define spin_unlock_bh(lock) _spin_unlock_bh(lock) #define read_unlock_irqrestore(lock, flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - _read_unlock_irqrestore(lock, flags); \ - } while (0) + _read_unlock_irqrestore(lock, flags) #define read_unlock_bh(lock) _read_unlock_bh(lock) #define write_unlock_irqrestore(lock, flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - _write_unlock_irqrestore(lock, flags); \ - } while (0) + _write_unlock_irqrestore(lock, flags) #define write_unlock_bh(lock) _write_unlock_bh(lock) #define spin_trylock_bh(lock) __cond_lock(lock, _spin_trylock_bh(lock)) @@ -299,7 +261,6 @@ do { \ #define spin_trylock_irqsave(lock, flags) \ ({ \ - BUILD_CHECK_IRQ_FLAGS(flags); \ local_irq_save(flags); \ spin_trylock(lock) ? \ 1 : ({ local_irq_restore(flags); 0; }); \ -- cgit v1.2.3-70-g09d2 From c9c3b86f2ab79f7f6e87eb735f9cc4508b73fc48 Mon Sep 17 00:00:00 2001 From: Kyle McMartin Date: Sun, 26 Nov 2006 18:56:56 -0500 Subject: [PATCH] Fix incorrent type of flags in I still think using BUILD_BUG_ON() is unacceptable, especially given how vague the error message was. Signed-off-by: Kyle McMartin [ And I already removed gthe BUILD_BUG_ON() in the previous commit ] Signed-off-by: Linus Torvalds --- include/asm-parisc/semaphore.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/asm-parisc/semaphore.h b/include/asm-parisc/semaphore.h index c9ee41cd070..d45827a21f9 100644 --- a/include/asm-parisc/semaphore.h +++ b/include/asm-parisc/semaphore.h @@ -115,7 +115,8 @@ extern __inline__ int down_interruptible(struct semaphore * sem) */ extern __inline__ int down_trylock(struct semaphore * sem) { - int flags, count; + unsigned long flags; + int count; spin_lock_irqsave(&sem->sentry, flags); count = sem->count - 1; @@ -131,7 +132,8 @@ extern __inline__ int down_trylock(struct semaphore * sem) */ extern __inline__ void up(struct semaphore * sem) { - int flags; + unsigned long flags; + spin_lock_irqsave(&sem->sentry, flags); if (sem->count < 0) { __up(sem); -- cgit v1.2.3-70-g09d2 From 2ea5814472c3c910aed5c5b60f1f3b1000e353f1 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 26 Nov 2006 19:05:22 -0800 Subject: Fix 'ALIGN()' macro, take 2 You wouldn't think that doing an ALIGN() macro that aligns something up to a power-of-two boundary would be likely to have bugs, would you? But hey, in the wonderful world of mixing integer types, you have to be careful. This just makes sure that the alignment is interpreted in the same type as the thing to be aligned. Thanks to Roland Dreier, who noticed that the amso1100 driver got broken by the previous fix (that just extended the mask to "unsigned long", but was still broken in "unsigned long long" - it just happened to be the same on 64-bit architectures). See commit 4c8bd7eeee4c8f157fb61fb64b57500990b42e0e for the history of bugs here... Acked-by: Roland Dreier Cc: Andrew Morton Cc: David Miller Cc: Al Viro Signed-off-by: Linus Torvalds --- include/linux/kernel.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 24b611147ad..b9b5e4ba166 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -30,8 +30,10 @@ extern const char linux_banner[]; #define STACK_MAGIC 0xdeadbeef +#define ALIGN(x,a) __ALIGN_MASK(x,(typeof(x))(a)-1) +#define __ALIGN_MASK(x,mask) (((x)+(mask))&~(mask)) + #define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0])) -#define ALIGN(x,a) (((x)+(a)-1UL)&~((a)-1UL)) #define FIELD_SIZEOF(t, f) (sizeof(((t*)0)->f)) #define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d)) #define roundup(x, y) ((((x) + ((y) - 1)) / (y)) * (y)) -- cgit v1.2.3-70-g09d2 From c4423cccc0846a812013f39c8f8cae1d2d9dff9d Mon Sep 17 00:00:00 2001 From: Jaroslav Kysela Date: Tue, 28 Nov 2006 15:07:33 +0100 Subject: [ALSA] version 1.0.13 Signed-off-by: Jaroslav Kysela --- include/sound/version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/sound/version.h b/include/sound/version.h index 52fd6879b86..17137f3a3b6 100644 --- a/include/sound/version.h +++ b/include/sound/version.h @@ -1,3 +1,3 @@ /* include/version.h. Generated by alsa/ksync script. */ #define CONFIG_SND_VERSION "1.0.13" -#define CONFIG_SND_DATE " (Sun Oct 22 08:56:16 2006 UTC)" +#define CONFIG_SND_DATE " (Tue Nov 28 14:07:24 2006 UTC)" -- cgit v1.2.3-70-g09d2 From 38b5b036b91248be8033d42dd0778b1c75c5af58 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 28 Nov 2006 20:12:59 +0100 Subject: [PATCH] i386: Fix compilation with UP genericarch Fix arch/i386/mach-generic/built-in.o: In function `apicid_to_node': summit.c:(.text+0x2f): undefined reference to `apicid_2_node' with CONFIG_GENERICH_ARCH and !CONFIG_SMP Signed-off-by: Andi Kleen --- include/asm-i386/mach-summit/mach_apic.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include') diff --git a/include/asm-i386/mach-summit/mach_apic.h b/include/asm-i386/mach-summit/mach_apic.h index ef0671e5d5c..43e5bd8f4a1 100644 --- a/include/asm-i386/mach-summit/mach_apic.h +++ b/include/asm-i386/mach-summit/mach_apic.h @@ -88,7 +88,11 @@ static inline void clustered_apic_check(void) static inline int apicid_to_node(int logical_apicid) { +#ifdef CONFIG_SMP return apicid_2_node[hard_smp_processor_id()]; +#else + return 0; +#endif } /* Mapping from cpu number to logical apicid */ -- cgit v1.2.3-70-g09d2 From 4195bdbea42151c04485b0d131eed6dd97309cee Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Mon, 27 Nov 2006 23:58:11 -0500 Subject: [PATCH] add missing libsas include to fix s390 compilation. include/scsi/libsas.h:479: error: field 'smp_req' has incomplete type include/scsi/libsas.h:480: error: field 'smp_resp' has incomplete type Signed-off-by: Dave Jones Signed-off-by: Linus Torvalds --- include/scsi/libsas.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h index 9582e840166..1d77b63c5ea 100644 --- a/include/scsi/libsas.h +++ b/include/scsi/libsas.h @@ -35,6 +35,7 @@ #include #include #include +#include struct block_device; -- cgit v1.2.3-70-g09d2 From e81c73596704793e73e6dbb478f41686f15a4b34 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 28 Nov 2006 20:53:39 -0800 Subject: [NET]: Fix MAX_HEADER setting. MAX_HEADER is either set to LL_MAX_HEADER or LL_MAX_HEADER + 48, and this is controlled by a set of CONFIG_* ifdef tests. It is trying to use LL_MAX_HEADER + 48 when any of the tunnels are enabled which set hard_header_len like this: dev->hard_header_len = LL_MAX_HEADER + sizeof(struct xxx); The correct set of tunnel drivers which do this are: ipip ip_gre ip6_tunnel sit so make the ifdef test match. Noticed by Patrick McHardy and with help from Herbert Xu. Signed-off-by: David S. Miller --- include/linux/netdevice.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 9264139bd8d..83b8c4f1d69 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -93,8 +93,10 @@ struct netpoll_info; #endif #endif -#if !defined(CONFIG_NET_IPIP) && \ - !defined(CONFIG_IPV6) && !defined(CONFIG_IPV6_MODULE) +#if !defined(CONFIG_NET_IPIP) && !defined(CONFIG_NET_IPIP_MODULE) && \ + !defined(CONFIG_NET_IPGRE) && !defined(CONFIG_NET_IPGRE_MODULE) && \ + !defined(CONFIG_IPV6_SIT) && !defined(CONFIG_IPV6_SIT_MODULE) && \ + !defined(CONFIG_IPV6_TUNNEL) && !defined(CONFIG_IPV6_TUNNEL_MODULE) #define MAX_HEADER LL_MAX_HEADER #else #define MAX_HEADER (LL_MAX_HEADER + 48) -- cgit v1.2.3-70-g09d2 From aed6fad8beca30a71a6950b6f650822254efa8c6 Mon Sep 17 00:00:00 2001 From: Milan Svoboda Date: Wed, 29 Nov 2006 12:09:52 +0100 Subject: [ARM] 3943/1: share declaration of struct pxa2xx_udc_mach_info between multiple platforms Move declaration of struct pxa2xx_udc_mach_info from include/asm-arm/arch-pxa/udc.h to new file include/asm-arm/mach/udc_pxa2xx.h. This allow us to use this structure with multiple platforms - pxa and ixp4xx. USB device controller used in pxa25x is the same as controller used in ixp4xx. Signed-off-by: Milan Svoboda Signed-off-by: Russell King --- include/asm-arm/arch-pxa/udc.h | 17 +---------------- include/asm-arm/mach/udc_pxa2xx.h | 26 ++++++++++++++++++++++++++ 2 files changed, 27 insertions(+), 16 deletions(-) create mode 100644 include/asm-arm/mach/udc_pxa2xx.h (limited to 'include') diff --git a/include/asm-arm/arch-pxa/udc.h b/include/asm-arm/arch-pxa/udc.h index 121cd241115..646480d3725 100644 --- a/include/asm-arm/arch-pxa/udc.h +++ b/include/asm-arm/arch-pxa/udc.h @@ -4,23 +4,8 @@ * This supports machine-specific differences in how the PXA2xx * USB Device Controller (UDC) is wired. * - * It is set in linux/arch/arm/mach-pxa/.c and used in - * the probe routine of linux/drivers/usb/gadget/pxa2xx_udc.c */ -struct pxa2xx_udc_mach_info { - int (*udc_is_connected)(void); /* do we see host? */ - void (*udc_command)(int cmd); -#define PXA2XX_UDC_CMD_CONNECT 0 /* let host see us */ -#define PXA2XX_UDC_CMD_DISCONNECT 1 /* so host won't see us */ - - /* Boards following the design guidelines in the developer's manual, - * with on-chip GPIOs not Lubbock's wierd hardware, can have a sane - * VBUS IRQ and omit the methods above. Store the GPIO number - * here; for GPIO 0, also mask in one of the pxa_gpio_mode() bits. - */ - u16 gpio_vbus; /* high == vbus present */ - u16 gpio_pullup; /* high == pullup activated */ -}; +#include extern void pxa_set_udc_info(struct pxa2xx_udc_mach_info *info); diff --git a/include/asm-arm/mach/udc_pxa2xx.h b/include/asm-arm/mach/udc_pxa2xx.h new file mode 100644 index 00000000000..ff0a95715a0 --- /dev/null +++ b/include/asm-arm/mach/udc_pxa2xx.h @@ -0,0 +1,26 @@ +/* + * linux/include/asm-arm/mach/udc_pxa2xx.h + * + * This supports machine-specific differences in how the PXA2xx + * USB Device Controller (UDC) is wired. + * + * It is set in linux/arch/arm/mach-pxa/.c or in + * linux/arch/mach-ixp4xx/.c and used in + * the probe routine of linux/drivers/usb/gadget/pxa2xx_udc.c + */ + +struct pxa2xx_udc_mach_info { + int (*udc_is_connected)(void); /* do we see host? */ + void (*udc_command)(int cmd); +#define PXA2XX_UDC_CMD_CONNECT 0 /* let host see us */ +#define PXA2XX_UDC_CMD_DISCONNECT 1 /* so host won't see us */ + + /* Boards following the design guidelines in the developer's manual, + * with on-chip GPIOs not Lubbock's wierd hardware, can have a sane + * VBUS IRQ and omit the methods above. Store the GPIO number + * here; for GPIO 0, also mask in one of the pxa_gpio_mode() bits. + */ + u16 gpio_vbus; /* high == vbus present */ + u16 gpio_pullup; /* high == pullup activated */ +}; + -- cgit v1.2.3-70-g09d2 From e1444b5a163e81138754cab27c4fa1637b5a2239 Mon Sep 17 00:00:00 2001 From: Sean Hefty Date: Tue, 28 Nov 2006 14:57:13 -0800 Subject: IB/cm: Fix automatic path migration support The ib_cm_establish() function is replaced with a more generic ib_cm_notify(). This routine is used to notify the CM that failover has occurred, so that future CM messages (LAP, DREQ) reach the remote CM. (Currently, we continue to use the original path) This bumps the userspace CM ABI. New alternate path information is captured when a LAP message is sent or received. This allows QP attributes to be initialized for the user when a new path is loaded after failover occurs. Signed-off-by: Sean Hefty Signed-off-by: Roland Dreier --- drivers/infiniband/core/cm.c | 115 ++++++++++++++++++++++++++++++++---------- drivers/infiniband/core/ucm.c | 12 ++--- include/rdma/ib_cm.h | 16 ++++-- include/rdma/ib_user_cm.h | 7 +-- 4 files changed, 110 insertions(+), 40 deletions(-) (limited to 'include') diff --git a/drivers/infiniband/core/cm.c b/drivers/infiniband/core/cm.c index 78d9c0c3314..e5dc4530808 100644 --- a/drivers/infiniband/core/cm.c +++ b/drivers/infiniband/core/cm.c @@ -147,12 +147,12 @@ struct cm_id_private { __be32 rq_psn; int timeout_ms; enum ib_mtu path_mtu; + __be16 pkey; u8 private_data_len; u8 max_cm_retries; u8 peer_to_peer; u8 responder_resources; u8 initiator_depth; - u8 local_ack_timeout; u8 retry_count; u8 rnr_retry_count; u8 service_timeout; @@ -690,7 +690,7 @@ static void cm_enter_timewait(struct cm_id_private *cm_id_priv) * timewait before notifying the user that we've exited timewait. */ cm_id_priv->id.state = IB_CM_TIMEWAIT; - wait_time = cm_convert_to_ms(cm_id_priv->local_ack_timeout); + wait_time = cm_convert_to_ms(cm_id_priv->av.packet_life_time + 1); queue_delayed_work(cm.wq, &cm_id_priv->timewait_info->work.work, msecs_to_jiffies(wait_time)); cm_id_priv->timewait_info = NULL; @@ -1009,6 +1009,7 @@ int ib_send_cm_req(struct ib_cm_id *cm_id, cm_id_priv->responder_resources = param->responder_resources; cm_id_priv->retry_count = param->retry_count; cm_id_priv->path_mtu = param->primary_path->mtu; + cm_id_priv->pkey = param->primary_path->pkey; cm_id_priv->qp_type = param->qp_type; ret = cm_alloc_msg(cm_id_priv, &cm_id_priv->msg); @@ -1023,8 +1024,6 @@ int ib_send_cm_req(struct ib_cm_id *cm_id, cm_id_priv->local_qpn = cm_req_get_local_qpn(req_msg); cm_id_priv->rq_psn = cm_req_get_starting_psn(req_msg); - cm_id_priv->local_ack_timeout = - cm_req_get_primary_local_ack_timeout(req_msg); spin_lock_irqsave(&cm_id_priv->lock, flags); ret = ib_post_send_mad(cm_id_priv->msg, NULL); @@ -1409,9 +1408,8 @@ static int cm_req_handler(struct cm_work *work) cm_id_priv->initiator_depth = cm_req_get_resp_res(req_msg); cm_id_priv->responder_resources = cm_req_get_init_depth(req_msg); cm_id_priv->path_mtu = cm_req_get_path_mtu(req_msg); + cm_id_priv->pkey = req_msg->pkey; cm_id_priv->sq_psn = cm_req_get_starting_psn(req_msg); - cm_id_priv->local_ack_timeout = - cm_req_get_primary_local_ack_timeout(req_msg); cm_id_priv->retry_count = cm_req_get_retry_count(req_msg); cm_id_priv->rnr_retry_count = cm_req_get_rnr_retry_count(req_msg); cm_id_priv->qp_type = cm_req_get_qp_type(req_msg); @@ -1715,7 +1713,7 @@ static int cm_establish_handler(struct cm_work *work) unsigned long flags; int ret; - /* See comment in ib_cm_establish about lookup. */ + /* See comment in cm_establish about lookup. */ cm_id_priv = cm_acquire_id(work->local_id, work->remote_id); if (!cm_id_priv) return -EINVAL; @@ -2401,11 +2399,16 @@ int ib_send_cm_lap(struct ib_cm_id *cm_id, cm_id_priv = container_of(cm_id, struct cm_id_private, id); spin_lock_irqsave(&cm_id_priv->lock, flags); if (cm_id->state != IB_CM_ESTABLISHED || - cm_id->lap_state != IB_CM_LAP_IDLE) { + (cm_id->lap_state != IB_CM_LAP_UNINIT && + cm_id->lap_state != IB_CM_LAP_IDLE)) { ret = -EINVAL; goto out; } + ret = cm_init_av_by_path(alternate_path, &cm_id_priv->alt_av); + if (ret) + goto out; + ret = cm_alloc_msg(cm_id_priv, &msg); if (ret) goto out; @@ -2430,7 +2433,8 @@ out: spin_unlock_irqrestore(&cm_id_priv->lock, flags); } EXPORT_SYMBOL(ib_send_cm_lap); -static void cm_format_path_from_lap(struct ib_sa_path_rec *path, +static void cm_format_path_from_lap(struct cm_id_private *cm_id_priv, + struct ib_sa_path_rec *path, struct cm_lap_msg *lap_msg) { memset(path, 0, sizeof *path); @@ -2442,10 +2446,10 @@ static void cm_format_path_from_lap(struct ib_sa_path_rec *path, path->hop_limit = lap_msg->alt_hop_limit; path->traffic_class = cm_lap_get_traffic_class(lap_msg); path->reversible = 1; - /* pkey is same as in REQ */ + path->pkey = cm_id_priv->pkey; path->sl = cm_lap_get_sl(lap_msg); path->mtu_selector = IB_SA_EQ; - /* mtu is same as in REQ */ + path->mtu = cm_id_priv->path_mtu; path->rate_selector = IB_SA_EQ; path->rate = cm_lap_get_packet_rate(lap_msg); path->packet_life_time_selector = IB_SA_EQ; @@ -2471,7 +2475,7 @@ static int cm_lap_handler(struct cm_work *work) param = &work->cm_event.param.lap_rcvd; param->alternate_path = &work->path[0]; - cm_format_path_from_lap(param->alternate_path, lap_msg); + cm_format_path_from_lap(cm_id_priv, param->alternate_path, lap_msg); work->cm_event.private_data = &lap_msg->private_data; spin_lock_irqsave(&cm_id_priv->lock, flags); @@ -2479,6 +2483,7 @@ static int cm_lap_handler(struct cm_work *work) goto unlock; switch (cm_id_priv->id.lap_state) { + case IB_CM_LAP_UNINIT: case IB_CM_LAP_IDLE: break; case IB_CM_MRA_LAP_SENT: @@ -2501,6 +2506,10 @@ static int cm_lap_handler(struct cm_work *work) cm_id_priv->id.lap_state = IB_CM_LAP_RCVD; cm_id_priv->tid = lap_msg->hdr.tid; + cm_init_av_for_response(work->port, work->mad_recv_wc->wc, + work->mad_recv_wc->recv_buf.grh, + &cm_id_priv->av); + cm_init_av_by_path(param->alternate_path, &cm_id_priv->alt_av); ret = atomic_inc_and_test(&cm_id_priv->work_count); if (!ret) list_add_tail(&work->list, &cm_id_priv->work_list); @@ -3039,7 +3048,7 @@ static void cm_work_handler(void *data) cm_free_work(work); } -int ib_cm_establish(struct ib_cm_id *cm_id) +static int cm_establish(struct ib_cm_id *cm_id) { struct cm_id_private *cm_id_priv; struct cm_work *work; @@ -3087,7 +3096,44 @@ int ib_cm_establish(struct ib_cm_id *cm_id) out: return ret; } -EXPORT_SYMBOL(ib_cm_establish); + +static int cm_migrate(struct ib_cm_id *cm_id) +{ + struct cm_id_private *cm_id_priv; + unsigned long flags; + int ret = 0; + + cm_id_priv = container_of(cm_id, struct cm_id_private, id); + spin_lock_irqsave(&cm_id_priv->lock, flags); + if (cm_id->state == IB_CM_ESTABLISHED && + (cm_id->lap_state == IB_CM_LAP_UNINIT || + cm_id->lap_state == IB_CM_LAP_IDLE)) { + cm_id->lap_state = IB_CM_LAP_IDLE; + cm_id_priv->av = cm_id_priv->alt_av; + } else + ret = -EINVAL; + spin_unlock_irqrestore(&cm_id_priv->lock, flags); + + return ret; +} + +int ib_cm_notify(struct ib_cm_id *cm_id, enum ib_event_type event) +{ + int ret; + + switch (event) { + case IB_EVENT_COMM_EST: + ret = cm_establish(cm_id); + break; + case IB_EVENT_PATH_MIG: + ret = cm_migrate(cm_id); + break; + default: + ret = -EINVAL; + } + return ret; +} +EXPORT_SYMBOL(ib_cm_notify); static void cm_recv_handler(struct ib_mad_agent *mad_agent, struct ib_mad_recv_wc *mad_recv_wc) @@ -3220,6 +3266,9 @@ static int cm_init_qp_rtr_attr(struct cm_id_private *cm_id_priv, if (cm_id_priv->alt_av.ah_attr.dlid) { *qp_attr_mask |= IB_QP_ALT_PATH; qp_attr->alt_port_num = cm_id_priv->alt_av.port->port_num; + qp_attr->alt_pkey_index = cm_id_priv->alt_av.pkey_index; + qp_attr->alt_timeout = + cm_id_priv->alt_av.packet_life_time + 1; qp_attr->alt_ah_attr = cm_id_priv->alt_av.ah_attr; } ret = 0; @@ -3246,19 +3295,31 @@ static int cm_init_qp_rts_attr(struct cm_id_private *cm_id_priv, case IB_CM_REP_SENT: case IB_CM_MRA_REP_RCVD: case IB_CM_ESTABLISHED: - *qp_attr_mask = IB_QP_STATE | IB_QP_SQ_PSN; - qp_attr->sq_psn = be32_to_cpu(cm_id_priv->sq_psn); - if (cm_id_priv->qp_type == IB_QPT_RC) { - *qp_attr_mask |= IB_QP_TIMEOUT | IB_QP_RETRY_CNT | - IB_QP_RNR_RETRY | - IB_QP_MAX_QP_RD_ATOMIC; - qp_attr->timeout = cm_id_priv->local_ack_timeout; - qp_attr->retry_cnt = cm_id_priv->retry_count; - qp_attr->rnr_retry = cm_id_priv->rnr_retry_count; - qp_attr->max_rd_atomic = cm_id_priv->initiator_depth; - } - if (cm_id_priv->alt_av.ah_attr.dlid) { - *qp_attr_mask |= IB_QP_PATH_MIG_STATE; + if (cm_id_priv->id.lap_state == IB_CM_LAP_UNINIT) { + *qp_attr_mask = IB_QP_STATE | IB_QP_SQ_PSN; + qp_attr->sq_psn = be32_to_cpu(cm_id_priv->sq_psn); + if (cm_id_priv->qp_type == IB_QPT_RC) { + *qp_attr_mask |= IB_QP_TIMEOUT | IB_QP_RETRY_CNT | + IB_QP_RNR_RETRY | + IB_QP_MAX_QP_RD_ATOMIC; + qp_attr->timeout = + cm_id_priv->av.packet_life_time + 1; + qp_attr->retry_cnt = cm_id_priv->retry_count; + qp_attr->rnr_retry = cm_id_priv->rnr_retry_count; + qp_attr->max_rd_atomic = + cm_id_priv->initiator_depth; + } + if (cm_id_priv->alt_av.ah_attr.dlid) { + *qp_attr_mask |= IB_QP_PATH_MIG_STATE; + qp_attr->path_mig_state = IB_MIG_REARM; + } + } else { + *qp_attr_mask = IB_QP_ALT_PATH | IB_QP_PATH_MIG_STATE; + qp_attr->alt_port_num = cm_id_priv->alt_av.port->port_num; + qp_attr->alt_pkey_index = cm_id_priv->alt_av.pkey_index; + qp_attr->alt_timeout = + cm_id_priv->alt_av.packet_life_time + 1; + qp_attr->alt_ah_attr = cm_id_priv->alt_av.ah_attr; qp_attr->path_mig_state = IB_MIG_REARM; } ret = 0; diff --git a/drivers/infiniband/core/ucm.c b/drivers/infiniband/core/ucm.c index b4894ba223b..1f4f2d2cfa2 100644 --- a/drivers/infiniband/core/ucm.c +++ b/drivers/infiniband/core/ucm.c @@ -683,11 +683,11 @@ out: return result; } -static ssize_t ib_ucm_establish(struct ib_ucm_file *file, - const char __user *inbuf, - int in_len, int out_len) +static ssize_t ib_ucm_notify(struct ib_ucm_file *file, + const char __user *inbuf, + int in_len, int out_len) { - struct ib_ucm_establish cmd; + struct ib_ucm_notify cmd; struct ib_ucm_context *ctx; int result; @@ -698,7 +698,7 @@ static ssize_t ib_ucm_establish(struct ib_ucm_file *file, if (IS_ERR(ctx)) return PTR_ERR(ctx); - result = ib_cm_establish(ctx->cm_id); + result = ib_cm_notify(ctx->cm_id, (enum ib_event_type) cmd.event); ib_ucm_ctx_put(ctx); return result; } @@ -1105,7 +1105,7 @@ static ssize_t (*ucm_cmd_table[])(struct ib_ucm_file *file, [IB_USER_CM_CMD_DESTROY_ID] = ib_ucm_destroy_id, [IB_USER_CM_CMD_ATTR_ID] = ib_ucm_attr_id, [IB_USER_CM_CMD_LISTEN] = ib_ucm_listen, - [IB_USER_CM_CMD_ESTABLISH] = ib_ucm_establish, + [IB_USER_CM_CMD_NOTIFY] = ib_ucm_notify, [IB_USER_CM_CMD_SEND_REQ] = ib_ucm_send_req, [IB_USER_CM_CMD_SEND_REP] = ib_ucm_send_rep, [IB_USER_CM_CMD_SEND_RTU] = ib_ucm_send_rtu, diff --git a/include/rdma/ib_cm.h b/include/rdma/ib_cm.h index c9b4738be9d..5c070176d9a 100644 --- a/include/rdma/ib_cm.h +++ b/include/rdma/ib_cm.h @@ -60,6 +60,7 @@ enum ib_cm_state { }; enum ib_cm_lap_state { + IB_CM_LAP_UNINIT, IB_CM_LAP_IDLE, IB_CM_LAP_SENT, IB_CM_LAP_RCVD, @@ -443,13 +444,20 @@ int ib_send_cm_drep(struct ib_cm_id *cm_id, u8 private_data_len); /** - * ib_cm_establish - Forces a connection state to established. + * ib_cm_notify - Notifies the CM of an event reported to the consumer. * @cm_id: Connection identifier to transition to established. + * @event: Type of event. * - * This routine should be invoked by users who receive messages on a - * connected QP before an RTU has been received. + * This routine should be invoked by users to notify the CM of relevant + * communication events. Events that should be reported to the CM and + * when to report them are: + * + * IB_EVENT_COMM_EST - Used when a message is received on a connected + * QP before an RTU has been received. + * IB_EVENT_PATH_MIG - Notifies the CM that the connection has failed over + * to the alternate path. */ -int ib_cm_establish(struct ib_cm_id *cm_id); +int ib_cm_notify(struct ib_cm_id *cm_id, enum ib_event_type event); /** * ib_send_cm_rej - Sends a connection rejection message to the diff --git a/include/rdma/ib_user_cm.h b/include/rdma/ib_user_cm.h index 066c20b7cdf..37650afb982 100644 --- a/include/rdma/ib_user_cm.h +++ b/include/rdma/ib_user_cm.h @@ -38,7 +38,7 @@ #include -#define IB_USER_CM_ABI_VERSION 4 +#define IB_USER_CM_ABI_VERSION 5 enum { IB_USER_CM_CMD_CREATE_ID, @@ -46,7 +46,7 @@ enum { IB_USER_CM_CMD_ATTR_ID, IB_USER_CM_CMD_LISTEN, - IB_USER_CM_CMD_ESTABLISH, + IB_USER_CM_CMD_NOTIFY, IB_USER_CM_CMD_SEND_REQ, IB_USER_CM_CMD_SEND_REP, @@ -117,8 +117,9 @@ struct ib_ucm_listen { __u32 reserved; }; -struct ib_ucm_establish { +struct ib_ucm_notify { __u32 id; + __u32 event; }; struct ib_ucm_private_data { -- cgit v1.2.3-70-g09d2 From 99e3b942c62f42c8d5added63305e12372b06daf Mon Sep 17 00:00:00 2001 From: Franck Bui-Huu Date: Thu, 19 Oct 2006 13:19:59 +0200 Subject: [MIPS] page.h: remove __pa() usages. __pa() was used by virt_to_page() and virt_addr_valid(). These latter are used when kernel is initialised so __pa() is not appropriate, we use virt_to_phys() instead. Futhermore __pa() is going to take care of CKSEG0/XKPHYS address mix for 64 bit kernels. This makes __pa() more complex than virt_to_phys() and this extra work is not needed by virt_to_page() and virt_addr_valid(). Eventually it consolidates virt_to_phys() prototype by making its argument 'const'. this avoids some warnings that was due to some virt_to_page() usages which pass const pointer. Signed-off-by: Franck Bui-Huu Signed-off-by: Ralf Baechle --- arch/mips/mm/init.c | 8 ++++---- include/asm-mips/io.h | 2 +- include/asm-mips/page.h | 6 ++++-- include/asm-mips/pgtable.h | 2 +- 4 files changed, 10 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/arch/mips/mm/init.c b/arch/mips/mm/init.c index 2de4d3c367a..304991263f6 100644 --- a/arch/mips/mm/init.c +++ b/arch/mips/mm/init.c @@ -90,9 +90,9 @@ unsigned long setup_zero_pages(void) if (!empty_zero_page) panic("Oh boy, that early out of memory?"); - page = virt_to_page(empty_zero_page); + page = virt_to_page((void *)empty_zero_page); split_page(page, order); - while (page < virt_to_page(empty_zero_page + (PAGE_SIZE << order))) { + while (page < virt_to_page((void *)(empty_zero_page + (PAGE_SIZE << order)))) { SetPageReserved(page); page++; } @@ -448,8 +448,8 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end) unsigned long addr; for (addr = begin; addr < end; addr += PAGE_SIZE) { - ClearPageReserved(virt_to_page(addr)); - init_page_count(virt_to_page(addr)); + ClearPageReserved(virt_to_page((void *)addr)); + init_page_count(virt_to_page((void *)addr)); memset((void *)addr, 0xcc, PAGE_SIZE); free_page(addr); totalram_pages++; diff --git a/include/asm-mips/io.h b/include/asm-mips/io.h index bc5f3c53155..d77b657c09c 100644 --- a/include/asm-mips/io.h +++ b/include/asm-mips/io.h @@ -113,7 +113,7 @@ static inline void set_io_port_base(unsigned long base) * almost all conceivable cases a device driver should not be using * this function */ -static inline unsigned long virt_to_phys(volatile void * address) +static inline unsigned long virt_to_phys(volatile const void *address) { return (unsigned long)address - PAGE_OFFSET; } diff --git a/include/asm-mips/page.h b/include/asm-mips/page.h index 85b258ee709..edb33a14ae9 100644 --- a/include/asm-mips/page.h +++ b/include/asm-mips/page.h @@ -34,7 +34,9 @@ #ifndef __ASSEMBLY__ +#include #include +#include extern void clear_page(void * page); extern void copy_page(void * to, void * from); @@ -160,8 +162,8 @@ typedef struct { unsigned long pgprot; } pgprot_t; #endif -#define virt_to_page(kaddr) pfn_to_page(__pa(kaddr) >> PAGE_SHIFT) -#define virt_addr_valid(kaddr) pfn_valid(__pa(kaddr) >> PAGE_SHIFT) +#define virt_to_page(kaddr) pfn_to_page(PFN_DOWN(virt_to_phys(kaddr))) +#define virt_addr_valid(kaddr) pfn_valid(PFN_DOWN(virt_to_phys(kaddr))) #define VM_DATA_DEFAULT_FLAGS (VM_READ | VM_WRITE | VM_EXEC | \ VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) diff --git a/include/asm-mips/pgtable.h b/include/asm-mips/pgtable.h index 1ca4d1e185c..f2e1325fec6 100644 --- a/include/asm-mips/pgtable.h +++ b/include/asm-mips/pgtable.h @@ -67,7 +67,7 @@ extern unsigned long empty_zero_page; extern unsigned long zero_page_mask; #define ZERO_PAGE(vaddr) \ - (virt_to_page(empty_zero_page + (((unsigned long)(vaddr)) & zero_page_mask))) + (virt_to_page((void *)(empty_zero_page + (((unsigned long)(vaddr)) & zero_page_mask)))) #define __HAVE_ARCH_MOVE_PTE #define move_pte(pte, prot, old_addr, new_addr) \ -- cgit v1.2.3-70-g09d2 From 620a4802be8ee7989b1b6684b7198ebae02af854 Mon Sep 17 00:00:00 2001 From: Franck Bui-Huu Date: Thu, 19 Oct 2006 13:20:00 +0200 Subject: [MIPS] Make __pa() aware of XKPHYS/CKSEG0 address mix for 64 bit kernels During early boot mem init, some configs couldn't use __pa() to convert virtual into physical addresses. Specially for 64 bit kernel cases when CONFIG_BUILD_ELF64=n. This patch make __pa() work for _all_ configs and thus make CPHYSADDR() useless. Signed-off-by: Franck Bui-Huu Signed-off-by: Ralf Baechle --- include/asm-mips/page.h | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/asm-mips/page.h b/include/asm-mips/page.h index edb33a14ae9..c3b13dedf21 100644 --- a/include/asm-mips/page.h +++ b/include/asm-mips/page.h @@ -136,8 +136,13 @@ typedef struct { unsigned long pgprot; } pgprot_t; /* to align the pointer to the (next) page boundary */ #define PAGE_ALIGN(addr) (((addr) + PAGE_SIZE - 1) & PAGE_MASK) -#define __pa(x) ((unsigned long) (x) - PAGE_OFFSET) -#define __va(x) ((void *)((unsigned long) (x) + PAGE_OFFSET)) +#if defined(CONFIG_64BIT) && !defined(CONFIG_BUILD_ELF64) +#define __pa_page_offset(x) ((unsigned long)(x) < CKSEG0 ? PAGE_OFFSET : CKSEG0) +#else +#define __pa_page_offset(x) PAGE_OFFSET +#endif +#define __pa(x) ((unsigned long)(x) - __pa_page_offset(x)) +#define __va(x) ((void *)((unsigned long)(x) + PAGE_OFFSET)) #define pfn_to_kaddr(pfn) __va((pfn) << PAGE_SHIFT) -- cgit v1.2.3-70-g09d2 From 8431fd094d625b94d364fe393076ccef88e6ce18 Mon Sep 17 00:00:00 2001 From: Franck Bui-Huu Date: Thu, 19 Oct 2006 13:20:02 +0200 Subject: [MIPS] Introduce __pa_symbol() This patch introduces __pa_symbol() macro which should be used to calculate the physical address of kernel symbols. It also relies on RELOC_HIDE() to avoid any compiler's oddities when doing arithmetics on symbols. Signed-off-by: Franck Bui-Huu Signed-off-by: Ralf Baechle --- include/asm-mips/page.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/asm-mips/page.h b/include/asm-mips/page.h index c3b13dedf21..0dc1a45c27e 100644 --- a/include/asm-mips/page.h +++ b/include/asm-mips/page.h @@ -142,6 +142,7 @@ typedef struct { unsigned long pgprot; } pgprot_t; #define __pa_page_offset(x) PAGE_OFFSET #endif #define __pa(x) ((unsigned long)(x) - __pa_page_offset(x)) +#define __pa_symbol(x) __pa(RELOC_HIDE((unsigned long)(x),0)) #define __va(x) ((void *)((unsigned long)(x) + PAGE_OFFSET)) #define pfn_to_kaddr(pfn) __va((pfn) << PAGE_SHIFT) -- cgit v1.2.3-70-g09d2 From 4e3884fc83f40b5daabceeee3a428a8ebebbbe4a Mon Sep 17 00:00:00 2001 From: Atsushi Nemoto Date: Sat, 21 Oct 2006 01:28:26 +0900 Subject: [MIPS] Use "long" for 64-bit values on 64-bit kernel. This would get rid of some warnings about "long" vs. "long long". Signed-off-by: Atsushi Nemoto Signed-off-by: Ralf Baechle --- include/asm-mips/addrspace.h | 40 ++++++++++++++++++++++------------------ 1 file changed, 22 insertions(+), 18 deletions(-) (limited to 'include') diff --git a/include/asm-mips/addrspace.h b/include/asm-mips/addrspace.h index 45c706e34df..c6275088cf6 100644 --- a/include/asm-mips/addrspace.h +++ b/include/asm-mips/addrspace.h @@ -19,12 +19,16 @@ #define _ATYPE_ #define _ATYPE32_ #define _ATYPE64_ -#define _LLCONST_(x) x +#define _CONST64_(x) x #else #define _ATYPE_ __PTRDIFF_TYPE__ #define _ATYPE32_ int -#define _ATYPE64_ long long -#define _LLCONST_(x) x ## LL +#define _ATYPE64_ __s64 +#ifdef CONFIG_64BIT +#define _CONST64_(x) x ## L +#else +#define _CONST64_(x) x ## LL +#endif #endif /* @@ -48,7 +52,7 @@ */ #define CPHYSADDR(a) ((_ACAST32_(a)) & 0x1fffffff) #define XPHYSADDR(a) ((_ACAST64_(a)) & \ - _LLCONST_(0x000000ffffffffff)) + _CONST64_(0x000000ffffffffff)) #ifdef CONFIG_64BIT @@ -57,14 +61,14 @@ * The compatibility segments use the full 64-bit sign extended value. Note * the R8000 doesn't have them so don't reference these in generic MIPS code. */ -#define XKUSEG _LLCONST_(0x0000000000000000) -#define XKSSEG _LLCONST_(0x4000000000000000) -#define XKPHYS _LLCONST_(0x8000000000000000) -#define XKSEG _LLCONST_(0xc000000000000000) -#define CKSEG0 _LLCONST_(0xffffffff80000000) -#define CKSEG1 _LLCONST_(0xffffffffa0000000) -#define CKSSEG _LLCONST_(0xffffffffc0000000) -#define CKSEG3 _LLCONST_(0xffffffffe0000000) +#define XKUSEG _CONST64_(0x0000000000000000) +#define XKSSEG _CONST64_(0x4000000000000000) +#define XKPHYS _CONST64_(0x8000000000000000) +#define XKSEG _CONST64_(0xc000000000000000) +#define CKSEG0 _CONST64_(0xffffffff80000000) +#define CKSEG1 _CONST64_(0xffffffffa0000000) +#define CKSSEG _CONST64_(0xffffffffc0000000) +#define CKSEG3 _CONST64_(0xffffffffe0000000) #define CKSEG0ADDR(a) (CPHYSADDR(a) | CKSEG0) #define CKSEG1ADDR(a) (CPHYSADDR(a) | CKSEG1) @@ -122,7 +126,7 @@ #define PHYS_TO_XKSEG_UNCACHED(p) PHYS_TO_XKPHYS(K_CALG_UNCACHED,(p)) #define PHYS_TO_XKSEG_CACHED(p) PHYS_TO_XKPHYS(K_CALG_COH_SHAREABLE,(p)) #define XKPHYS_TO_PHYS(p) ((p) & TO_PHYS_MASK) -#define PHYS_TO_XKPHYS(cm,a) (_LLCONST_(0x8000000000000000) | \ +#define PHYS_TO_XKPHYS(cm,a) (_CONST64_(0x8000000000000000) | \ ((cm)<<59) | (a)) #if defined (CONFIG_CPU_R4300) \ @@ -132,20 +136,20 @@ || defined (CONFIG_CPU_NEVADA) \ || defined (CONFIG_CPU_TX49XX) \ || defined (CONFIG_CPU_MIPS64) -#define TO_PHYS_MASK _LLCONST_(0x0000000fffffffff) /* 2^^36 - 1 */ +#define TO_PHYS_MASK _CONST64_(0x0000000fffffffff) /* 2^^36 - 1 */ #endif #if defined (CONFIG_CPU_R8000) /* We keep KUSIZE consistent with R4000 for now (2^^40) instead of (2^^48) */ -#define TO_PHYS_MASK _LLCONST_(0x000000ffffffffff) /* 2^^40 - 1 */ +#define TO_PHYS_MASK _CONST64_(0x000000ffffffffff) /* 2^^40 - 1 */ #endif #if defined (CONFIG_CPU_R10000) -#define TO_PHYS_MASK _LLCONST_(0x000000ffffffffff) /* 2^^40 - 1 */ +#define TO_PHYS_MASK _CONST64_(0x000000ffffffffff) /* 2^^40 - 1 */ #endif #if defined(CONFIG_CPU_SB1) || defined(CONFIG_CPU_SB1A) -#define TO_PHYS_MASK _LLCONST_(0x00000fffffffffff) /* 2^^44 - 1 */ +#define TO_PHYS_MASK _CONST64_(0x00000fffffffffff) /* 2^^44 - 1 */ #endif #ifndef CONFIG_CPU_R8000 @@ -155,7 +159,7 @@ * in order to catch bugs in the source code. */ -#define COMPAT_K1BASE32 _LLCONST_(0xffffffffa0000000) +#define COMPAT_K1BASE32 _CONST64_(0xffffffffa0000000) #define PHYS_TO_COMPATK1(x) ((x) | COMPAT_K1BASE32) /* 32-bit compat k1 */ #endif -- cgit v1.2.3-70-g09d2 From 56ae58333031bb0564c141f955d1e42276cade55 Mon Sep 17 00:00:00 2001 From: Yoichi Yuasa Date: Sat, 14 Oct 2006 00:25:04 +0900 Subject: [MIPS] Rewrite GALILEO_INL/GALILEO_OUTL to GT_READ/GT_WRITE This patch has rewritten GALILEO_INL/GALILEO_OUTL using GT_READ/GT_WRITE. This patch tested on Cobalt Qube2. Signed-off-by: Yoichi Yuasa Signed-off-by: Ralf Baechle --- arch/mips/cobalt/irq.c | 31 +++++++++++++---------------- arch/mips/cobalt/setup.c | 16 +++++++-------- arch/mips/pci/fixup-cobalt.c | 11 +++++----- arch/mips/pci/ops-gt64111.c | 16 +++++++-------- include/asm-mips/gt64120.h | 14 +++++++++++++ include/asm-mips/mach-cobalt/cobalt.h | 29 ++------------------------- include/asm-mips/mach-cobalt/mach-gt64120.h | 28 +++++++++++++++++++++++++- 7 files changed, 78 insertions(+), 67 deletions(-) (limited to 'include') diff --git a/arch/mips/cobalt/irq.c b/arch/mips/cobalt/irq.c index 82e569d5b02..4c46f0e7378 100644 --- a/arch/mips/cobalt/irq.c +++ b/arch/mips/cobalt/irq.c @@ -45,25 +45,22 @@ static inline void galileo_irq(void) { unsigned int mask, pending, devfn; - mask = GALILEO_INL(GT_INTRMASK_OFS); - pending = GALILEO_INL(GT_INTRCAUSE_OFS) & mask; + mask = GT_READ(GT_INTRMASK_OFS); + pending = GT_READ(GT_INTRCAUSE_OFS) & mask; - if (pending & GALILEO_INTR_T0EXP) { - - GALILEO_OUTL(~GALILEO_INTR_T0EXP, GT_INTRCAUSE_OFS); + if (pending & GT_INTR_T0EXP_MSK) { + GT_WRITE(GT_INTRCAUSE_OFS, ~GT_INTR_T0EXP_MSK); do_IRQ(COBALT_GALILEO_IRQ); - - } else if (pending & GALILEO_INTR_RETRY_CTR) { - - devfn = GALILEO_INL(GT_PCI0_CFGADDR_OFS) >> 8; - GALILEO_OUTL(~GALILEO_INTR_RETRY_CTR, GT_INTRCAUSE_OFS); - printk(KERN_WARNING "Galileo: PCI retry count exceeded (%02x.%u)\n", - PCI_SLOT(devfn), PCI_FUNC(devfn)); - + } else if (pending & GT_INTR_RETRYCTR0_MSK) { + devfn = GT_READ(GT_PCI0_CFGADDR_OFS) >> 8; + GT_WRITE(GT_INTRCAUSE_OFS, ~GT_INTR_RETRYCTR0_MSK); + printk(KERN_WARNING + "Galileo: PCI retry count exceeded (%02x.%u)\n", + PCI_SLOT(devfn), PCI_FUNC(devfn)); } else { - - GALILEO_OUTL(mask & ~pending, GT_INTRMASK_OFS); - printk(KERN_WARNING "Galileo: masking unexpected interrupt %08x\n", pending); + GT_WRITE(GT_INTRMASK_OFS, mask & ~pending); + printk(KERN_WARNING + "Galileo: masking unexpected interrupt %08x\n", pending); } } @@ -104,7 +101,7 @@ void __init arch_init_irq(void) * Mask all Galileo interrupts. The Galileo * handler is set in cobalt_timer_setup() */ - GALILEO_OUTL(0, GT_INTRMASK_OFS); + GT_WRITE(GT_INTRMASK_OFS, 0); init_i8259_irqs(); /* 0 ... 15 */ mips_cpu_irq_init(COBALT_CPU_IRQ); /* 16 ... 23 */ diff --git a/arch/mips/cobalt/setup.c b/arch/mips/cobalt/setup.c index bf9dc72b972..e8f0f20b852 100644 --- a/arch/mips/cobalt/setup.c +++ b/arch/mips/cobalt/setup.c @@ -51,23 +51,23 @@ const char *get_system_type(void) void __init plat_timer_setup(struct irqaction *irq) { /* Load timer value for HZ (TCLK is 50MHz) */ - GALILEO_OUTL(50*1000*1000 / HZ, GT_TC0_OFS); + GT_WRITE(GT_TC0_OFS, 50*1000*1000 / HZ); /* Enable timer */ - GALILEO_OUTL(GALILEO_ENTC0 | GALILEO_SELTC0, GT_TC_CONTROL_OFS); + GT_WRITE(GT_TC_CONTROL_OFS, GT_TC_CONTROL_ENTC0_MSK | GT_TC_CONTROL_SELTC0_MSK); /* Register interrupt */ setup_irq(COBALT_GALILEO_IRQ, irq); /* Enable interrupt */ - GALILEO_OUTL(GALILEO_INTR_T0EXP | GALILEO_INL(GT_INTRMASK_OFS), GT_INTRMASK_OFS); + GT_WRITE(GT_INTRMASK_OFS, GT_INTR_T0EXP_MSK | GT_READ(GT_INTRMASK_OFS)); } extern struct pci_ops gt64111_pci_ops; static struct resource cobalt_mem_resource = { - .start = GT64111_MEM_BASE, - .end = GT64111_MEM_END, + .start = GT_DEF_PCI0_MEM0_BASE, + .end = GT_DEF_PCI0_MEM0_BASE + GT_DEF_PCI0_MEM0_SIZE - 1, .name = "PCI memory", .flags = IORESOURCE_MEM }; @@ -115,7 +115,7 @@ static struct pci_controller cobalt_pci_controller = { .mem_resource = &cobalt_mem_resource, .mem_offset = 0, .io_resource = &cobalt_io_resource, - .io_offset = 0 - GT64111_IO_BASE + .io_offset = 0 - GT_DEF_PCI0_IO_BASE, }; void __init plat_mem_setup(void) @@ -128,7 +128,7 @@ void __init plat_mem_setup(void) _machine_halt = cobalt_machine_halt; pm_power_off = cobalt_machine_power_off; - set_io_port_base(CKSEG1ADDR(GT64111_IO_BASE)); + set_io_port_base(CKSEG1ADDR(GT_DEF_PCI0_IO_BASE)); /* I/O port resource must include UART and LCD/buttons */ ioport_resource.end = 0x0fffffff; @@ -139,7 +139,7 @@ void __init plat_mem_setup(void) /* Read the cobalt id register out of the PCI config space */ PCI_CFG_SET(devfn, (VIA_COBALT_BRD_ID_REG & ~0x3)); - cobalt_board_id = GALILEO_INL(GT_PCI0_CFGDATA_OFS); + cobalt_board_id = GT_READ(GT_PCI0_CFGDATA_OFS); cobalt_board_id >>= ((VIA_COBALT_BRD_ID_REG & 3) * 8); cobalt_board_id = VIA_COBALT_BRD_REG_to_ID(cobalt_board_id); diff --git a/arch/mips/pci/fixup-cobalt.c b/arch/mips/pci/fixup-cobalt.c index 75a01e76489..7d5f6bbf7a9 100644 --- a/arch/mips/pci/fixup-cobalt.c +++ b/arch/mips/pci/fixup-cobalt.c @@ -94,22 +94,21 @@ static void qube_raq_galileo_fixup(struct pci_dev *dev) #if 0 if (galileo_id >= 0x10) { /* New Galileo, assumes PCI stop line to VIA is connected. */ - GALILEO_OUTL(0x4020, GT_PCI0_TOR_OFS); + GT_WRITE(GT_PCI0_TOR_OFS, 0x4020); } else if (galileo_id == 0x1 || galileo_id == 0x2) #endif { signed int timeo; /* XXX WE MUST DO THIS ELSE GALILEO LOCKS UP! -DaveM */ - timeo = GALILEO_INL(GT_PCI0_TOR_OFS); + timeo = GT_READ(GT_PCI0_TOR_OFS); /* Old Galileo, assumes PCI STOP line to VIA is disconnected. */ - GALILEO_OUTL( + GT_WRITE(GT_PCI0_TOR_OFS, (0xff << 16) | /* retry count */ (0xff << 8) | /* timeout 1 */ - 0xff, /* timeout 0 */ - GT_PCI0_TOR_OFS); + 0xff); /* timeout 0 */ /* enable PCI retry exceeded interrupt */ - GALILEO_OUTL(GALILEO_INTR_RETRY_CTR | GALILEO_INL(GT_INTRMASK_OFS), GT_INTRMASK_OFS); + GT_WRITE(GT_INTRMASK_OFS, GT_INTR_RETRYCTR0_MSK | GT_READ(GT_INTRMASK_OFS)); } } diff --git a/arch/mips/pci/ops-gt64111.c b/arch/mips/pci/ops-gt64111.c index 13de45940b1..ecd3991bd0e 100644 --- a/arch/mips/pci/ops-gt64111.c +++ b/arch/mips/pci/ops-gt64111.c @@ -38,18 +38,18 @@ static int gt64111_pci_read_config(struct pci_bus *bus, unsigned int devfn, switch (size) { case 4: PCI_CFG_SET(devfn, where); - *val = GALILEO_INL(GT_PCI0_CFGDATA_OFS); + *val = GT_READ(GT_PCI0_CFGDATA_OFS); return PCIBIOS_SUCCESSFUL; case 2: PCI_CFG_SET(devfn, (where & ~0x3)); - *val = GALILEO_INL(GT_PCI0_CFGDATA_OFS) + *val = GT_READ(GT_PCI0_CFGDATA_OFS) >> ((where & 3) * 8); return PCIBIOS_SUCCESSFUL; case 1: PCI_CFG_SET(devfn, (where & ~0x3)); - *val = GALILEO_INL(GT_PCI0_CFGDATA_OFS) + *val = GT_READ(GT_PCI0_CFGDATA_OFS) >> ((where & 3) * 8); return PCIBIOS_SUCCESSFUL; } @@ -68,25 +68,25 @@ static int gt64111_pci_write_config(struct pci_bus *bus, unsigned int devfn, switch (size) { case 4: PCI_CFG_SET(devfn, where); - GALILEO_OUTL(val, GT_PCI0_CFGDATA_OFS); + GT_WRITE(GT_PCI0_CFGDATA_OFS, val); return PCIBIOS_SUCCESSFUL; case 2: PCI_CFG_SET(devfn, (where & ~0x3)); - tmp = GALILEO_INL(GT_PCI0_CFGDATA_OFS); + tmp = GT_READ(GT_PCI0_CFGDATA_OFS); tmp &= ~(0xffff << ((where & 0x3) * 8)); tmp |= (val << ((where & 0x3) * 8)); - GALILEO_OUTL(tmp, GT_PCI0_CFGDATA_OFS); + GT_WRITE(GT_PCI0_CFGDATA_OFS, tmp); return PCIBIOS_SUCCESSFUL; case 1: PCI_CFG_SET(devfn, (where & ~0x3)); - tmp = GALILEO_INL(GT_PCI0_CFGDATA_OFS); + tmp = GT_READ(GT_PCI0_CFGDATA_OFS); tmp &= ~(0xff << ((where & 0x3) * 8)); tmp |= (val << ((where & 0x3) * 8)); - GALILEO_OUTL(tmp, GT_PCI0_CFGDATA_OFS); + GT_WRITE(GT_PCI0_CFGDATA_OFS, tmp); return PCIBIOS_SUCCESSFUL; } diff --git a/include/asm-mips/gt64120.h b/include/asm-mips/gt64120.h index 2edd171bb6c..4bf8e28f885 100644 --- a/include/asm-mips/gt64120.h +++ b/include/asm-mips/gt64120.h @@ -451,6 +451,13 @@ #define GT_SDRAM_OPMODE_OP_MODE 3 #define GT_SDRAM_OPMODE_OP_CBR 4 +#define GT_TC_CONTROL_ENTC0_SHF 0 +#define GT_TC_CONTROL_ENTC0_MSK (MSK(1) << GT_TC_CONTROL_ENTC0_SHF) +#define GT_TC_CONTROL_ENTC0_BIT GT_TC_CONTROL_ENTC0_MSK +#define GT_TC_CONTROL_SELTC0_SHF 1 +#define GT_TC_CONTROL_SELTC0_MSK (MSK(1) << GT_TC_CONTROL_SELTC0_SHF) +#define GT_TC_CONTROL_SELTC0_BIT GT_TC_CONTROL_SELTC0_MSK + #define GT_PCI0_BARE_SWSCS3BOOTDIS_SHF 0 #define GT_PCI0_BARE_SWSCS3BOOTDIS_MSK (MSK(1) << GT_PCI0_BARE_SWSCS3BOOTDIS_SHF) @@ -523,6 +530,13 @@ #define GT_PCI0_CMD_SWORDSWAP_MSK (MSK(1) << GT_PCI0_CMD_SWORDSWAP_SHF) #define GT_PCI0_CMD_SWORDSWAP_BIT GT_PCI0_CMD_SWORDSWAP_MSK +#define GT_INTR_T0EXP_SHF 8 +#define GT_INTR_T0EXP_MSK (MSK(1) << GT_INTR_T0EXP_SHF) +#define GT_INTR_T0EXP_BIT GT_INTR_T0EXP_MSK +#define GT_INTR_RETRYCTR0_SHF 20 +#define GT_INTR_RETRYCTR0_MSK (MSK(1) << GT_INTR_RETRYCTR0_SHF) +#define GT_INTR_RETRYCTR0_BIT GT_INTR_RETRYCTR0_MSK + /* * Misc */ diff --git a/include/asm-mips/mach-cobalt/cobalt.h b/include/asm-mips/mach-cobalt/cobalt.h index b3c5ecbec03..00b0fc68d5c 100644 --- a/include/asm-mips/mach-cobalt/cobalt.h +++ b/include/asm-mips/mach-cobalt/cobalt.h @@ -67,34 +67,9 @@ #define COBALT_BRD_ID_QUBE2 0x5 #define COBALT_BRD_ID_RAQ2 0x6 -/* - * Galileo chipset access macros for the Cobalt. The base address for - * the GT64111 chip is 0x14000000 - * - * Most of this really should go into a separate GT64111 header file. - */ -#define GT64111_IO_BASE 0x10000000UL -#define GT64111_IO_END 0x11ffffffUL -#define GT64111_MEM_BASE 0x12000000UL -#define GT64111_MEM_END 0x13ffffffUL -#define GT64111_BASE 0x14000000UL -#define GALILEO_REG(ofs) CKSEG1ADDR(GT64111_BASE + (unsigned long)(ofs)) - -#define GALILEO_INL(port) (*(volatile unsigned int *) GALILEO_REG(port)) -#define GALILEO_OUTL(val, port) \ -do { \ - *(volatile unsigned int *) GALILEO_REG(port) = (val); \ -} while (0) - -#define GALILEO_INTR_T0EXP (1 << 8) -#define GALILEO_INTR_RETRY_CTR (1 << 20) - -#define GALILEO_ENTC0 0x01 -#define GALILEO_SELTC0 0x02 - #define PCI_CFG_SET(devfn,where) \ - GALILEO_OUTL((0x80000000 | (PCI_SLOT (devfn) << 11) | \ - (PCI_FUNC (devfn) << 8) | (where)), GT_PCI0_CFGADDR_OFS) + GT_WRITE(GT_PCI0_CFGADDR_OFS, (0x80000000 | (PCI_SLOT (devfn) << 11) | \ + (PCI_FUNC (devfn) << 8) | (where))) #define COBALT_LED_PORT (*(volatile unsigned char *) CKSEG1ADDR(0x1c000000)) # define COBALT_LED_BAR_LEFT (1 << 0) /* Qube */ diff --git a/include/asm-mips/mach-cobalt/mach-gt64120.h b/include/asm-mips/mach-cobalt/mach-gt64120.h index 587fc4378f4..ae9c5523c7e 100644 --- a/include/asm-mips/mach-cobalt/mach-gt64120.h +++ b/include/asm-mips/mach-cobalt/mach-gt64120.h @@ -1 +1,27 @@ -/* there's something here ... in the dark */ +/* + * Copyright (C) 2006 Yoichi Yuasa + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef _COBALT_MACH_GT64120_H +#define _COBALT_MACH_GT64120_H + +/* + * Cobalt uses GT64111. GT64111 is almost the same as GT64120. + */ + +#define GT64120_BASE CKSEG1ADDR(GT_DEF_BASE) + +#endif /* _COBALT_MACH_GT64120_H */ -- cgit v1.2.3-70-g09d2 From 656be92f9ae194ed62bc81310a4589a7cd765f13 Mon Sep 17 00:00:00 2001 From: Atsushi Nemoto Date: Thu, 26 Oct 2006 00:08:31 +0900 Subject: [MIPS] Load modules to CKSEG0 if CONFIG_BUILD_ELF64=n This is a patch to load 64-bit modules to CKSEG0 so that can be compiled with -msym32 option. This makes each module ~10% smaller. * introduce MODULE_START and MODULE_END * custom module_alloc() * PGD for modules * change XTLB refill handler synthesizer * enable -msym32 for modules again (revert ca78b1a5c6a6e70e052d3ea253828e49b5d07c8a) New XTLB refill handler looks like this: 80000080 dmfc0 k0,C0_BADVADDR 80000084 bltz k0,800000e4 # goto l_module_alloc 80000088 lui k1,0x8046 # %high(pgd_current) 8000008c ld k1,24600(k1) # %low(pgd_current) 80000090 dsrl k0,k0,0x1b # l_vmalloc_done: 80000094 andi k0,k0,0x1ff8 80000098 daddu k1,k1,k0 8000009c dmfc0 k0,C0_BADVADDR 800000a0 ld k1,0(k1) 800000a4 dsrl k0,k0,0x12 800000a8 andi k0,k0,0xff8 800000ac daddu k1,k1,k0 800000b0 dmfc0 k0,C0_XCONTEXT 800000b4 ld k1,0(k1) 800000b8 andi k0,k0,0xff0 800000bc daddu k1,k1,k0 800000c0 ld k0,0(k1) 800000c4 ld k1,8(k1) 800000c8 dsrl k0,k0,0x6 800000cc mtc0 k0,C0_ENTRYLO0 800000d0 dsrl k1,k1,0x6 800000d4 mtc0 k1,C0_ENTRYL01 800000d8 nop 800000dc tlbwr 800000e0 eret 800000e4 dsll k1,k0,0x2 # l_module_alloc: 800000e8 bgez k1,80000008 # goto l_vmalloc 800000ec lui k1,0xc000 800000f0 dsubu k0,k0,k1 800000f4 lui k1,0x8046 # %high(module_pg_dir) 800000f8 beq zero,zero,80000000 800000fc nop 80000000 beq zero,zero,80000090 # goto l_vmalloc_done 80000004 daddiu k1,k1,0x4000 80000008 dsll32 k1,k1,0x0 # l_vmalloc: 8000000c dsubu k0,k0,k1 80000010 beq zero,zero,80000090 # goto l_vmalloc_done 80000014 lui k1,0x8046 # %high(swapper_pg_dir) Signed-off-by: Atsushi Nemoto Signed-off-by: Ralf Baechle --- arch/mips/Makefile | 4 +--- arch/mips/kernel/head.S | 3 +++ arch/mips/kernel/module.c | 15 ++++++++++++ arch/mips/mm/fault.c | 4 ++++ arch/mips/mm/pgtable-64.c | 3 +++ arch/mips/mm/tlbex.c | 55 +++++++++++++++++++++++++++++++++++++++++++ include/asm-mips/pgtable-64.h | 13 ++++++++++ 7 files changed, 94 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/arch/mips/Makefile b/arch/mips/Makefile index d580d46f967..641aa30b363 100644 --- a/arch/mips/Makefile +++ b/arch/mips/Makefile @@ -63,9 +63,7 @@ cflags-y += -mabi=64 ifdef CONFIG_BUILD_ELF64 cflags-y += $(call cc-option,-mno-explicit-relocs) else -# -msym32 can not be used for modules since they are loaded into XKSEG -CFLAGS_MODULE += $(call cc-option,-mno-explicit-relocs) -CFLAGS_KERNEL += $(call cc-option,-msym32) +cflags-y += $(call cc-option,-msym32) endif endif diff --git a/arch/mips/kernel/head.S b/arch/mips/kernel/head.S index ddc1b71c937..a2e095adaa3 100644 --- a/arch/mips/kernel/head.S +++ b/arch/mips/kernel/head.S @@ -250,6 +250,9 @@ NESTED(smp_bootstrap, 16, sp) */ page swapper_pg_dir, _PGD_ORDER #ifdef CONFIG_64BIT +#if defined(CONFIG_MODULES) && !defined(CONFIG_BUILD_ELF64) + page module_pg_dir, _PGD_ORDER +#endif page invalid_pmd_table, _PMD_ORDER #endif page invalid_pte_table, _PTE_ORDER diff --git a/arch/mips/kernel/module.c b/arch/mips/kernel/module.c index d7bf0215bc1..cb0801437b6 100644 --- a/arch/mips/kernel/module.c +++ b/arch/mips/kernel/module.c @@ -29,6 +29,7 @@ #include #include #include +#include /* MODULE_START */ struct mips_hi16 { struct mips_hi16 *next; @@ -43,9 +44,23 @@ static DEFINE_SPINLOCK(dbe_lock); void *module_alloc(unsigned long size) { +#ifdef MODULE_START + struct vm_struct *area; + + size = PAGE_ALIGN(size); + if (!size) + return NULL; + + area = __get_vm_area(size, VM_ALLOC, MODULE_START, MODULE_END); + if (!area) + return NULL; + + return __vmalloc_area(area, GFP_KERNEL, PAGE_KERNEL); +#else if (size == 0) return NULL; return vmalloc(size); +#endif } /* Free memory returned from module_alloc */ diff --git a/arch/mips/mm/fault.c b/arch/mips/mm/fault.c index 8423d859077..6f90e7ef66a 100644 --- a/arch/mips/mm/fault.c +++ b/arch/mips/mm/fault.c @@ -60,6 +60,10 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long write, */ if (unlikely(address >= VMALLOC_START && address <= VMALLOC_END)) goto vmalloc_fault; +#ifdef MODULE_START + if (unlikely(address >= MODULE_START && address < MODULE_END)) + goto vmalloc_fault; +#endif /* * If we're in an interrupt or have no user diff --git a/arch/mips/mm/pgtable-64.c b/arch/mips/mm/pgtable-64.c index 8d600d307d5..c46eb651bf0 100644 --- a/arch/mips/mm/pgtable-64.c +++ b/arch/mips/mm/pgtable-64.c @@ -58,6 +58,9 @@ void __init pagetable_init(void) /* Initialize the entire pgd. */ pgd_init((unsigned long)swapper_pg_dir); +#ifdef MODULE_START + pgd_init((unsigned long)module_pg_dir); +#endif pmd_init((unsigned long)invalid_pmd_table, (unsigned long)invalid_pte_table); pgd_base = swapper_pg_dir; diff --git a/arch/mips/mm/tlbex.c b/arch/mips/mm/tlbex.c index fec318a1c8c..492c518e7ba 100644 --- a/arch/mips/mm/tlbex.c +++ b/arch/mips/mm/tlbex.c @@ -423,6 +423,9 @@ enum label_id { label_invalid, label_second_part, label_leave, +#ifdef MODULE_START + label_module_alloc, +#endif label_vmalloc, label_vmalloc_done, label_tlbw_hazard, @@ -455,6 +458,9 @@ static __init void build_label(struct label **lab, u32 *addr, L_LA(_second_part) L_LA(_leave) +#ifdef MODULE_START +L_LA(_module_alloc) +#endif L_LA(_vmalloc) L_LA(_vmalloc_done) L_LA(_tlbw_hazard) @@ -686,6 +692,13 @@ static void __init il_bgezl(u32 **p, struct reloc **r, unsigned int reg, i_bgezl(p, reg, 0); } +static void __init __attribute__((unused)) +il_bgez(u32 **p, struct reloc **r, unsigned int reg, enum label_id l) +{ + r_mips_pc16(r, *p, l); + i_bgez(p, reg, 0); +} + /* The only general purpose registers allowed in TLB handlers. */ #define K0 26 #define K1 27 @@ -970,7 +983,11 @@ build_get_pmde64(u32 **p, struct label **l, struct reloc **r, * The vmalloc handling is not in the hotpath. */ i_dmfc0(p, tmp, C0_BADVADDR); +#ifdef MODULE_START + il_bltz(p, r, tmp, label_module_alloc); +#else il_bltz(p, r, tmp, label_vmalloc); +#endif /* No i_nop needed here, since the next insn doesn't touch TMP. */ #ifdef CONFIG_SMP @@ -1023,8 +1040,46 @@ build_get_pgd_vmalloc64(u32 **p, struct label **l, struct reloc **r, { long swpd = (long)swapper_pg_dir; +#ifdef MODULE_START + long modd = (long)module_pg_dir; + + l_module_alloc(l, *p); + /* + * Assumption: + * VMALLOC_START >= 0xc000000000000000UL + * MODULE_START >= 0xe000000000000000UL + */ + i_SLL(p, ptr, bvaddr, 2); + il_bgez(p, r, ptr, label_vmalloc); + + if (in_compat_space_p(MODULE_START) && !rel_lo(MODULE_START)) { + i_lui(p, ptr, rel_hi(MODULE_START)); /* delay slot */ + } else { + /* unlikely configuration */ + i_nop(p); /* delay slot */ + i_LA(p, ptr, MODULE_START); + } + i_dsubu(p, bvaddr, bvaddr, ptr); + + if (in_compat_space_p(modd) && !rel_lo(modd)) { + il_b(p, r, label_vmalloc_done); + i_lui(p, ptr, rel_hi(modd)); + } else { + i_LA_mostly(p, ptr, modd); + il_b(p, r, label_vmalloc_done); + i_daddiu(p, ptr, ptr, rel_lo(modd)); + } + + l_vmalloc(l, *p); + if (in_compat_space_p(MODULE_START) && !rel_lo(MODULE_START) && + MODULE_START << 32 == VMALLOC_START) + i_dsll32(p, ptr, ptr, 0); /* typical case */ + else + i_LA(p, ptr, VMALLOC_START); +#else l_vmalloc(l, *p); i_LA(p, ptr, VMALLOC_START); +#endif i_dsubu(p, bvaddr, bvaddr, ptr); if (in_compat_space_p(swpd) && !rel_lo(swpd)) { diff --git a/include/asm-mips/pgtable-64.h b/include/asm-mips/pgtable-64.h index 7e7320300aa..b9b1e86493e 100644 --- a/include/asm-mips/pgtable-64.h +++ b/include/asm-mips/pgtable-64.h @@ -14,6 +14,7 @@ #include #include #include +#include #include @@ -103,6 +104,13 @@ #define VMALLOC_START MAP_BASE #define VMALLOC_END \ (VMALLOC_START + PTRS_PER_PGD * PTRS_PER_PMD * PTRS_PER_PTE * PAGE_SIZE) +#if defined(CONFIG_MODULES) && !defined(CONFIG_BUILD_ELF64) && \ + VMALLOC_START != CKSSEG +/* Load modules into 32bit-compatible segment. */ +#define MODULE_START CKSSEG +#define MODULE_END (FIXADDR_START-2*PAGE_SIZE) +extern pgd_t module_pg_dir[PTRS_PER_PGD]; +#endif #define pte_ERROR(e) \ printk("%s:%d: bad pte %016lx.\n", __FILE__, __LINE__, pte_val(e)) @@ -174,7 +182,12 @@ static inline void pud_clear(pud_t *pudp) #define __pmd_offset(address) pmd_index(address) /* to find an entry in a kernel page-table-directory */ +#ifdef MODULE_START +#define pgd_offset_k(address) \ + ((address) >= MODULE_START ? module_pg_dir : pgd_offset(&init_mm, 0UL)) +#else #define pgd_offset_k(address) pgd_offset(&init_mm, 0UL) +#endif #define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1)) #define pmd_index(address) (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1)) -- cgit v1.2.3-70-g09d2 From 583bb86fbb9e85287f020fe4eb5352a0ec3c66a3 Mon Sep 17 00:00:00 2001 From: Nicolas Schichan Date: Wed, 18 Oct 2006 15:14:55 +0200 Subject: [MIPS] Add support for kexec A tiny userland application loading the kernel and invoking kexec_load for mips is available here: http://chac.le-poulpe.net/~nico/kexec/kexec-2006-10-18.tar.gz Signed-off-by: Ralf Baechle --- arch/mips/Kconfig | 17 ++++++++ arch/mips/kernel/Makefile | 2 + arch/mips/kernel/machine_kexec.c | 85 ++++++++++++++++++++++++++++++++++++++ arch/mips/kernel/relocate_kernel.S | 80 +++++++++++++++++++++++++++++++++++ arch/mips/kernel/scall32-o32.S | 2 +- arch/mips/kernel/scall64-64.S | 2 +- arch/mips/kernel/scall64-n32.S | 2 +- arch/mips/kernel/scall64-o32.S | 2 +- include/asm-mips/kexec.h | 32 ++++++++++++++ include/linux/kexec.h | 2 + 10 files changed, 222 insertions(+), 4 deletions(-) create mode 100644 arch/mips/kernel/machine_kexec.c create mode 100644 arch/mips/kernel/relocate_kernel.S create mode 100644 include/asm-mips/kexec.h (limited to 'include') diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 1443024b1c7..c0da0ffe8d5 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -766,6 +766,23 @@ config TOSHIBA_RBTX4938 endchoice +config KEXEC + bool "Kexec system call (EXPERIMENTAL)" + depends on EXPERIMENTAL + help + kexec is a system call that implements the ability to shutdown your + current kernel, and to start another kernel. It is like a reboot + but it is indepedent of the system firmware. And like a reboot + you can start any kernel with it, not just Linux. + + The name comes from the similiarity to the exec system call. + + It is an ongoing process to be certain the hardware in a machine + is properly shutdown, so do not be surprised if this code does not + initially work for you. It may help to enable device hotplugging + support. As of this writing the exact hardware interface is + strongly in flux, so no good recommendation can be made. + source "arch/mips/ddb5xxx/Kconfig" source "arch/mips/gt64120/ev64120/Kconfig" source "arch/mips/jazz/Kconfig" diff --git a/arch/mips/kernel/Makefile b/arch/mips/kernel/Makefile index 6bfbbed0897..f35b739d0a1 100644 --- a/arch/mips/kernel/Makefile +++ b/arch/mips/kernel/Makefile @@ -67,6 +67,8 @@ obj-$(CONFIG_64BIT) += cpu-bugs64.o obj-$(CONFIG_I8253) += i8253.o +obj-$(CONFIG_KEXEC) += machine_kexec.o relocate_kernel.o + CFLAGS_cpu-bugs64.o = $(shell if $(CC) $(CFLAGS) -Wa,-mdaddi -c -o /dev/null -xc /dev/null >/dev/null 2>&1; then echo "-DHAVE_AS_SET_DADDI"; fi) EXTRA_AFLAGS := $(CFLAGS) diff --git a/arch/mips/kernel/machine_kexec.c b/arch/mips/kernel/machine_kexec.c new file mode 100644 index 00000000000..e0ad754c7ed --- /dev/null +++ b/arch/mips/kernel/machine_kexec.c @@ -0,0 +1,85 @@ +/* + * machine_kexec.c for kexec + * Created by on Thu Oct 12 15:15:06 2006 + * + * This source code is licensed under the GNU General Public License, + * Version 2. See the file COPYING for more details. + */ + +#include +#include +#include + +#include +#include + +const extern unsigned char relocate_new_kernel[]; +const extern unsigned int relocate_new_kernel_size; + +extern unsigned long kexec_start_address; +extern unsigned long kexec_indirection_page; + +int +machine_kexec_prepare(struct kimage *kimage) +{ + return 0; +} + +void +machine_kexec_cleanup(struct kimage *kimage) +{ +} + +void +machine_shutdown(void) +{ +} + +void +machine_crash_shutdown(struct pt_regs *regs) +{ +} + +void +machine_kexec(struct kimage *image) +{ + unsigned long reboot_code_buffer; + unsigned long entry; + unsigned long *ptr; + + reboot_code_buffer = + (unsigned long)page_address(image->control_code_page); + + kexec_start_address = image->start; + kexec_indirection_page = phys_to_virt(image->head & PAGE_MASK); + + memcpy((void*)reboot_code_buffer, relocate_new_kernel, + relocate_new_kernel_size); + + /* + * The generic kexec code builds a page list with physical + * addresses. they are directly accessible through KSEG0 (or + * CKSEG0 or XPHYS if on 64bit system), hence the + * pys_to_virt() call. + */ + for (ptr = &image->head; (entry = *ptr) && !(entry &IND_DONE); + ptr = (entry & IND_INDIRECTION) ? + phys_to_virt(entry & PAGE_MASK) : ptr + 1) { + if (*ptr & IND_SOURCE || *ptr & IND_INDIRECTION || + *ptr & IND_DESTINATION) + *ptr = phys_to_virt(*ptr); + } + + /* + * we do not want to be bothered. + */ + local_irq_disable(); + + flush_icache_range(reboot_code_buffer, + reboot_code_buffer + KEXEC_CONTROL_CODE_SIZE); + + printk("Will call new kernel at %08x\n", image->start); + printk("Bye ...\n"); + flush_cache_all(); + ((void (*)(void))reboot_code_buffer)(); +} diff --git a/arch/mips/kernel/relocate_kernel.S b/arch/mips/kernel/relocate_kernel.S new file mode 100644 index 00000000000..a3f0d00c133 --- /dev/null +++ b/arch/mips/kernel/relocate_kernel.S @@ -0,0 +1,80 @@ +/* + * relocate_kernel.S for kexec + * Created by on Thu Oct 12 17:49:57 2006 + * + * This source code is licensed under the GNU General Public License, + * Version 2. See the file COPYING for more details. + */ + +#include +#include +#include +#include +#include +#include +#include + + .globl relocate_new_kernel +relocate_new_kernel: + + PTR_L s0, kexec_indirection_page + PTR_L s1, kexec_start_address + +process_entry: + PTR_L s2, (s0) + PTR_ADD s0, s0, SZREG + + /* destination page */ + and s3, s2, 0x1 + beq s3, zero, 1f + and s4, s2, ~0x1 /* store destination addr in s4 */ + move a0, s4 + b process_entry + +1: + /* indirection page, update s0 */ + and s3, s2, 0x2 + beq s3, zero, 1f + and s0, s2, ~0x2 + b process_entry + +1: + /* done page */ + and s3, s2, 0x4 + beq s3, zero, 1f + b done +1: + /* source page */ + and s3, s2, 0x8 + beq s3, zero, process_entry + and s2, s2, ~0x8 + li s6, (1 << PAGE_SHIFT) / SZREG + +copy_word: + /* copy page word by word */ + REG_L s5, (s2) + REG_S s5, (s4) + INT_ADD s4, s4, SZREG + INT_ADD s2, s2, SZREG + INT_SUB s6, s6, 1 + beq s6, zero, process_entry + b copy_word + b process_entry + +done: + /* jump to kexec_start_address */ + j s1 + + .globl kexec_start_address +kexec_start_address: + .long 0x0 + + .globl kexec_indirection_page +kexec_indirection_page: + .long 0x0 + +relocate_new_kernel_end: + + .globl relocate_new_kernel_size +relocate_new_kernel_size: + .long relocate_new_kernel_end - relocate_new_kernel diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S index a95f37de080..7c0b3936ba4 100644 --- a/arch/mips/kernel/scall32-o32.S +++ b/arch/mips/kernel/scall32-o32.S @@ -653,7 +653,7 @@ einval: li v0, -EINVAL sys sys_move_pages 6 sys sys_set_robust_list 2 sys sys_get_robust_list 3 /* 4310 */ - sys sys_ni_syscall 0 + sys sys_kexec_load 4 sys sys_getcpu 3 sys sys_epoll_pwait 6 .endm diff --git a/arch/mips/kernel/scall64-64.S b/arch/mips/kernel/scall64-64.S index 8fb0f60f657..e569b846e9a 100644 --- a/arch/mips/kernel/scall64-64.S +++ b/arch/mips/kernel/scall64-64.S @@ -468,6 +468,6 @@ sys_call_table: PTR sys_move_pages PTR sys_set_robust_list PTR sys_get_robust_list - PTR sys_ni_syscall /* 5270 */ + PTR sys_kexec_load /* 5270 */ PTR sys_getcpu PTR sys_epoll_pwait diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S index 0da5ca2040f..5b18f265d75 100644 --- a/arch/mips/kernel/scall64-n32.S +++ b/arch/mips/kernel/scall64-n32.S @@ -394,6 +394,6 @@ EXPORT(sysn32_call_table) PTR sys_move_pages PTR compat_sys_set_robust_list PTR compat_sys_get_robust_list - PTR sys_ni_syscall + PTR compat_sys_kexec_load PTR sys_getcpu PTR sys_epoll_pwait diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S index b9d00cae8b5..e91379c1be1 100644 --- a/arch/mips/kernel/scall64-o32.S +++ b/arch/mips/kernel/scall64-o32.S @@ -516,7 +516,7 @@ sys_call_table: PTR compat_sys_move_pages PTR compat_sys_set_robust_list PTR compat_sys_get_robust_list /* 4310 */ - PTR sys_ni_syscall + PTR compat_sys_kexec_load PTR sys_getcpu PTR sys_epoll_pwait .size sys_call_table,.-sys_call_table diff --git a/include/asm-mips/kexec.h b/include/asm-mips/kexec.h new file mode 100644 index 00000000000..b25267ebcb0 --- /dev/null +++ b/include/asm-mips/kexec.h @@ -0,0 +1,32 @@ +/* + * kexec.h for kexec + * Created by on Thu Oct 12 14:59:34 2006 + * + * This source code is licensed under the GNU General Public License, + * Version 2. See the file COPYING for more details. + */ + +#ifndef _MIPS_KEXEC +# define _MIPS_KEXEC + +/* Maximum physical address we can use pages from */ +#define KEXEC_SOURCE_MEMORY_LIMIT (0x20000000) +/* Maximum address we can reach in physical address mode */ +#define KEXEC_DESTINATION_MEMORY_LIMIT (0x20000000) + /* Maximum address we can use for the control code buffer */ +#define KEXEC_CONTROL_MEMORY_LIMIT (0x20000000) + +#define KEXEC_CONTROL_CODE_SIZE 4096 + +/* The native architecture */ +#define KEXEC_ARCH KEXEC_ARCH_MIPS + +#define MAX_NOTE_BYTES 1024 + +static inline void crash_setup_regs(struct pt_regs *newregs, + struct pt_regs *oldregs) +{ + /* Dummy implementation for now */ +} + +#endif /* !_MIPS_KEXEC */ diff --git a/include/linux/kexec.h b/include/linux/kexec.h index 6427949ddf9..a4ede62b339 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -122,6 +122,8 @@ extern struct kimage *kexec_crash_image; #define KEXEC_ARCH_IA_64 (50 << 16) #define KEXEC_ARCH_S390 (22 << 16) #define KEXEC_ARCH_SH (42 << 16) +#define KEXEC_ARCH_MIPS_LE (10 << 16) +#define KEXEC_ARCH_MIPS ( 8 << 16) #define KEXEC_FLAGS (KEXEC_ON_CRASH) /* List of defined/legal kexec flags */ -- cgit v1.2.3-70-g09d2 From 2472d0b519c9634d89420064315c0926149947aa Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 30 Nov 2006 01:14:45 +0000 Subject: [MIPS] Remove unused R10000 performance counter definitions. Signed-off-by: Ralf Baechle --- include/asm-mips/mipsregs.h | 56 --------------------------------------------- 1 file changed, 56 deletions(-) (limited to 'include') diff --git a/include/asm-mips/mipsregs.h b/include/asm-mips/mipsregs.h index 1f318d70799..9985cb7c16e 100644 --- a/include/asm-mips/mipsregs.h +++ b/include/asm-mips/mipsregs.h @@ -545,62 +545,6 @@ #define MIPS_FPIR_L (_ULCAST_(1) << 21) #define MIPS_FPIR_F64 (_ULCAST_(1) << 22) -/* - * R10000 performance counter definitions. - * - * FIXME: The R10000 performance counter opens a nice way to implement CPU - * time accounting with a precission of one cycle. I don't have - * R10000 silicon but just a manual, so ... - */ - -/* - * Events counted by counter #0 - */ -#define CE0_CYCLES 0 -#define CE0_INSN_ISSUED 1 -#define CE0_LPSC_ISSUED 2 -#define CE0_S_ISSUED 3 -#define CE0_SC_ISSUED 4 -#define CE0_SC_FAILED 5 -#define CE0_BRANCH_DECODED 6 -#define CE0_QW_WB_SECONDARY 7 -#define CE0_CORRECTED_ECC_ERRORS 8 -#define CE0_ICACHE_MISSES 9 -#define CE0_SCACHE_I_MISSES 10 -#define CE0_SCACHE_I_WAY_MISSPREDICTED 11 -#define CE0_EXT_INTERVENTIONS_REQ 12 -#define CE0_EXT_INVALIDATE_REQ 13 -#define CE0_VIRTUAL_COHERENCY_COND 14 -#define CE0_INSN_GRADUATED 15 - -/* - * Events counted by counter #1 - */ -#define CE1_CYCLES 0 -#define CE1_INSN_GRADUATED 1 -#define CE1_LPSC_GRADUATED 2 -#define CE1_S_GRADUATED 3 -#define CE1_SC_GRADUATED 4 -#define CE1_FP_INSN_GRADUATED 5 -#define CE1_QW_WB_PRIMARY 6 -#define CE1_TLB_REFILL 7 -#define CE1_BRANCH_MISSPREDICTED 8 -#define CE1_DCACHE_MISS 9 -#define CE1_SCACHE_D_MISSES 10 -#define CE1_SCACHE_D_WAY_MISSPREDICTED 11 -#define CE1_EXT_INTERVENTION_HITS 12 -#define CE1_EXT_INVALIDATE_REQ 13 -#define CE1_SP_HINT_TO_CEXCL_SC_BLOCKS 14 -#define CE1_SP_HINT_TO_SHARED_SC_BLOCKS 15 - -/* - * These flags define in which privilege mode the counters count events - */ -#define CEB_USER 8 /* Count events in user mode, EXL = ERL = 0 */ -#define CEB_SUPERVISOR 4 /* Count events in supvervisor mode EXL = ERL = 0 */ -#define CEB_KERNEL 2 /* Count events in kernel mode EXL = ERL = 0 */ -#define CEB_EXL 1 /* Count events with EXL = 1, ERL = 0 */ - #ifndef __ASSEMBLY__ /* -- cgit v1.2.3-70-g09d2 From 6f2c3fa022312d5381f44359984395761e375f1b Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 30 Nov 2006 01:14:45 +0000 Subject: [MIPS] Fix alignment hole in struct cache_desc; shrink struct. Signed-off-by: Ralf Baechle --- include/asm-mips/cpu-info.h | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/asm-mips/cpu-info.h b/include/asm-mips/cpu-info.h index a2f0c8ea916..610d0cdeaa9 100644 --- a/include/asm-mips/cpu-info.h +++ b/include/asm-mips/cpu-info.h @@ -22,12 +22,12 @@ * Descriptor for a cache */ struct cache_desc { - unsigned short linesz; /* Size of line in bytes */ - unsigned short ways; /* Number of ways */ - unsigned short sets; /* Number of lines per set */ unsigned int waysize; /* Bytes per way */ - unsigned int waybit; /* Bits to select in a cache set */ - unsigned int flags; /* Flags describing cache properties */ + unsigned short sets; /* Number of lines per set */ + unsigned char ways; /* Number of ways */ + unsigned char linesz; /* Size of line in bytes */ + unsigned char waybit; /* Bits to select in a cache set */ + unsigned char flags; /* Flags describing cache properties */ }; /* -- cgit v1.2.3-70-g09d2 From c87b6ebaea034c0e0ce86127870cf1511a307b64 Mon Sep 17 00:00:00 2001 From: Atsushi Nemoto Date: Sat, 28 Oct 2006 01:14:37 +0900 Subject: [MIPS] mips hpt cleanup: get rid of mips_hpt_init Currently nobody outside time.c require mips_hpt_init(). Remove it and call c0_hpt_timer_init() directly if R4k counter was used for timer interrupt. Signed-off-by: Atsushi Nemoto Signed-off-by: Ralf Baechle --- arch/mips/kernel/time.c | 22 ++++++++-------------- arch/mips/pmc-sierra/yosemite/smp.c | 2 -- include/asm-mips/time.h | 1 - 3 files changed, 8 insertions(+), 17 deletions(-) (limited to 'include') diff --git a/arch/mips/kernel/time.c b/arch/mips/kernel/time.c index e535f86efa2..111d1baa3b2 100644 --- a/arch/mips/kernel/time.c +++ b/arch/mips/kernel/time.c @@ -88,12 +88,6 @@ static unsigned int null_hpt_read(void) return 0; } -static void __init null_hpt_init(void) -{ - /* nothing */ -} - - /* * Timer ack for an R4k-compatible timer of a known frequency. */ @@ -133,7 +127,6 @@ static void __init c0_hpt_timer_init(void) int (*mips_timer_state)(void); void (*mips_timer_ack)(void); unsigned int (*mips_hpt_read)(void); -void (*mips_hpt_init)(void) __initdata = null_hpt_init; unsigned int mips_hpt_mask = 0xffffffff; /* last time when xtime and rtc are sync'ed up */ @@ -383,16 +376,20 @@ void __init time_init(void) if (!mips_timer_state) { /* No external timer interrupt -- use R4k. */ - mips_hpt_init = c0_hpt_timer_init; mips_timer_ack = c0_timer_ack; + /* Calculate cache parameters. */ + cycles_per_jiffy = + (mips_hpt_frequency + HZ / 2) / HZ; + /* + * This sets up the high precision + * timer for the first interrupt. + */ + c0_hpt_timer_init(); } } if (!mips_hpt_frequency) mips_hpt_frequency = calibrate_hpt(); - /* Calculate cache parameters. */ - cycles_per_jiffy = (mips_hpt_frequency + HZ / 2) / HZ; - /* Report the high precision timer rate for a reference. */ printk("Using %u.%03u MHz high precision timer.\n", ((mips_hpt_frequency + 500) / 1000) / 1000, @@ -403,9 +400,6 @@ void __init time_init(void) /* No timer interrupt ack (e.g. i8254). */ mips_timer_ack = null_timer_ack; - /* This sets up the high precision timer for the first interrupt. */ - mips_hpt_init(); - /* * Call board specific timer interrupt setup. * diff --git a/arch/mips/pmc-sierra/yosemite/smp.c b/arch/mips/pmc-sierra/yosemite/smp.c index 3cc0436db6c..305491e74db 100644 --- a/arch/mips/pmc-sierra/yosemite/smp.c +++ b/arch/mips/pmc-sierra/yosemite/smp.c @@ -99,8 +99,6 @@ void prom_cpus_done(void) */ void prom_init_secondary(void) { - mips_hpt_init(); - set_c0_status(ST0_CO | ST0_IE | ST0_IM); } diff --git a/include/asm-mips/time.h b/include/asm-mips/time.h index 625acd337bc..b58665efb1d 100644 --- a/include/asm-mips/time.h +++ b/include/asm-mips/time.h @@ -48,7 +48,6 @@ extern void (*mips_timer_ack)(void); * If mips_hpt_read is NULL, an R4k-compatible timer setup is attempted. */ extern unsigned int (*mips_hpt_read)(void); -extern void (*mips_hpt_init)(void); extern unsigned int mips_hpt_mask; /* -- cgit v1.2.3-70-g09d2 From 1603b5aca4f15b34848fb5594d0c7b6333b99144 Mon Sep 17 00:00:00 2001 From: Atsushi Nemoto Date: Thu, 2 Nov 2006 02:08:36 +0900 Subject: [MIPS] IRQ cleanups This is a big irq cleanup patch. * Use set_irq_chip() to register irq_chip. * Initialize .mask, .unmask, .mask_ack field. Functions for these method are already exist in most case. * Do not initialize .startup, .shutdown, .enable, .disable fields if default routines provided by irq_chip_set_defaults() were suitable. * Remove redundant irq_desc initializations. * Remove unnecessary local_irq_save/local_irq_restore, spin_lock. With this cleanup, it would be easy to switch to slightly lightwait irq flow handlers (handle_level_irq(), etc.) instead of __do_IRQ(). Though whole this patch is quite large, changes in each irq_chip are not quite simple. Please review and test on your platform. Thanks. Signed-off-by: Atsushi Nemoto Signed-off-by: Ralf Baechle --- arch/mips/au1000/common/irq.c | 63 +++---- arch/mips/ddb5xxx/ddb5477/irq_5477.c | 23 +-- arch/mips/dec/ecc-berr.c | 6 - arch/mips/dec/ioasic-irq.c | 72 ++------ arch/mips/dec/kn02-irq.c | 53 +----- arch/mips/emma2rh/common/irq_emma2rh.c | 34 +--- arch/mips/emma2rh/markeins/irq_markeins.c | 55 ++----- arch/mips/gt64120/ev64120/irq.c | 40 +---- arch/mips/jazz/irq.c | 27 +-- arch/mips/jmr3927/rbhma3100/irq.c | 32 +--- arch/mips/kernel/i8259.c | 21 +-- arch/mips/kernel/irq-msc01.c | 45 +---- arch/mips/kernel/irq-mv6434x.c | 53 +----- arch/mips/kernel/irq-rm7000.c | 53 +----- arch/mips/kernel/irq-rm9000.c | 45 ++--- arch/mips/kernel/irq.c | 13 -- arch/mips/kernel/irq_cpu.c | 77 ++------- arch/mips/lasat/interrupt.c | 36 +--- arch/mips/mips-boards/atlas/atlas_int.c | 28 +--- arch/mips/momentum/ocelot_c/cpci-irq.c | 53 +----- arch/mips/momentum/ocelot_c/uart-irq.c | 56 +------ arch/mips/philips/pnx8550/common/int.c | 66 ++------ arch/mips/sgi-ip22/ip22-eisa.c | 33 +--- arch/mips/sgi-ip22/ip22-int.c | 109 ++---------- arch/mips/sgi-ip27/ip27-irq.c | 17 +- arch/mips/sgi-ip27/ip27-timer.c | 27 +-- arch/mips/sgi-ip32/ip32-irq.c | 133 +++------------ arch/mips/sibyte/bcm1480/irq.c | 30 +--- arch/mips/sibyte/sb1250/irq.c | 30 +--- arch/mips/sni/irq.c | 40 +---- arch/mips/tx4927/common/tx4927_irq.c | 162 ++---------------- .../tx4927/toshiba_rbtx4927/toshiba_rbtx4927_irq.c | 183 ++------------------- arch/mips/tx4938/common/irq.c | 113 ++----------- arch/mips/tx4938/toshiba_rbtx4938/irq.c | 53 +----- arch/mips/vr41xx/common/icu.c | 46 ++---- arch/mips/vr41xx/nec-cmbvr4133/irq.c | 20 +-- include/asm-mips/dec/kn02.h | 2 - 37 files changed, 292 insertions(+), 1657 deletions(-) (limited to 'include') diff --git a/arch/mips/au1000/common/irq.c b/arch/mips/au1000/common/irq.c index 2abe132bb07..9cf7b671583 100644 --- a/arch/mips/au1000/common/irq.c +++ b/arch/mips/au1000/common/irq.c @@ -70,7 +70,6 @@ extern irq_cpustat_t irq_stat [NR_CPUS]; extern void mips_timer_interrupt(void); static void setup_local_irq(unsigned int irq, int type, int int_req); -static unsigned int startup_irq(unsigned int irq); static void end_irq(unsigned int irq_nr); static inline void mask_and_ack_level_irq(unsigned int irq_nr); static inline void mask_and_ack_rise_edge_irq(unsigned int irq_nr); @@ -84,20 +83,6 @@ void (*board_init_irq)(void); static DEFINE_SPINLOCK(irq_lock); -static unsigned int startup_irq(unsigned int irq_nr) -{ - local_enable_irq(irq_nr); - return 0; -} - - -static void shutdown_irq(unsigned int irq_nr) -{ - local_disable_irq(irq_nr); - return; -} - - inline void local_enable_irq(unsigned int irq_nr) { if (irq_nr > AU1000_LAST_INTC0_INT) { @@ -249,41 +234,37 @@ void restore_local_and_enable(int controller, unsigned long mask) static struct irq_chip rise_edge_irq_type = { .typename = "Au1000 Rise Edge", - .startup = startup_irq, - .shutdown = shutdown_irq, - .enable = local_enable_irq, - .disable = local_disable_irq, .ack = mask_and_ack_rise_edge_irq, + .mask = local_disable_irq, + .mask_ack = mask_and_ack_rise_edge_irq, + .unmask = local_enable_irq, .end = end_irq, }; static struct irq_chip fall_edge_irq_type = { .typename = "Au1000 Fall Edge", - .startup = startup_irq, - .shutdown = shutdown_irq, - .enable = local_enable_irq, - .disable = local_disable_irq, .ack = mask_and_ack_fall_edge_irq, + .mask = local_disable_irq, + .mask_ack = mask_and_ack_fall_edge_irq, + .unmask = local_enable_irq, .end = end_irq, }; static struct irq_chip either_edge_irq_type = { .typename = "Au1000 Rise or Fall Edge", - .startup = startup_irq, - .shutdown = shutdown_irq, - .enable = local_enable_irq, - .disable = local_disable_irq, .ack = mask_and_ack_either_edge_irq, + .mask = local_disable_irq, + .mask_ack = mask_and_ack_either_edge_irq, + .unmask = local_enable_irq, .end = end_irq, }; static struct irq_chip level_irq_type = { .typename = "Au1000 Level", - .startup = startup_irq, - .shutdown = shutdown_irq, - .enable = local_enable_irq, - .disable = local_disable_irq, .ack = mask_and_ack_level_irq, + .mask = local_disable_irq, + .mask_ack = mask_and_ack_level_irq, + .unmask = local_enable_irq, .end = end_irq, }; @@ -328,31 +309,31 @@ static void setup_local_irq(unsigned int irq_nr, int type, int int_req) au_writel(1<<(irq_nr-32), IC1_CFG2CLR); au_writel(1<<(irq_nr-32), IC1_CFG1CLR); au_writel(1<<(irq_nr-32), IC1_CFG0SET); - irq_desc[irq_nr].chip = &rise_edge_irq_type; + set_irq_chip(irq_nr, &rise_edge_irq_type); break; case INTC_INT_FALL_EDGE: /* 0:1:0 */ au_writel(1<<(irq_nr-32), IC1_CFG2CLR); au_writel(1<<(irq_nr-32), IC1_CFG1SET); au_writel(1<<(irq_nr-32), IC1_CFG0CLR); - irq_desc[irq_nr].chip = &fall_edge_irq_type; + set_irq_chip(irq_nr, &fall_edge_irq_type); break; case INTC_INT_RISE_AND_FALL_EDGE: /* 0:1:1 */ au_writel(1<<(irq_nr-32), IC1_CFG2CLR); au_writel(1<<(irq_nr-32), IC1_CFG1SET); au_writel(1<<(irq_nr-32), IC1_CFG0SET); - irq_desc[irq_nr].chip = &either_edge_irq_type; + set_irq_chip(irq_nr, &either_edge_irq_type); break; case INTC_INT_HIGH_LEVEL: /* 1:0:1 */ au_writel(1<<(irq_nr-32), IC1_CFG2SET); au_writel(1<<(irq_nr-32), IC1_CFG1CLR); au_writel(1<<(irq_nr-32), IC1_CFG0SET); - irq_desc[irq_nr].chip = &level_irq_type; + set_irq_chip(irq_nr, &level_irq_type); break; case INTC_INT_LOW_LEVEL: /* 1:1:0 */ au_writel(1<<(irq_nr-32), IC1_CFG2SET); au_writel(1<<(irq_nr-32), IC1_CFG1SET); au_writel(1<<(irq_nr-32), IC1_CFG0CLR); - irq_desc[irq_nr].chip = &level_irq_type; + set_irq_chip(irq_nr, &level_irq_type); break; case INTC_INT_DISABLED: /* 0:0:0 */ au_writel(1<<(irq_nr-32), IC1_CFG0CLR); @@ -380,31 +361,31 @@ static void setup_local_irq(unsigned int irq_nr, int type, int int_req) au_writel(1< #include #include -#include #include #include @@ -231,13 +230,10 @@ irqreturn_t dec_ecc_be_interrupt(int irq, void *dev_id) static inline void dec_kn02_be_init(void) { volatile u32 *csr = (void *)CKSEG1ADDR(KN02_SLOT_BASE + KN02_CSR); - unsigned long flags; kn0x_erraddr = (void *)CKSEG1ADDR(KN02_SLOT_BASE + KN02_ERRADDR); kn0x_chksyn = (void *)CKSEG1ADDR(KN02_SLOT_BASE + KN02_CHKSYN); - spin_lock_irqsave(&kn02_lock, flags); - /* Preset write-only bits of the Control Register cache. */ cached_kn02_csr = *csr | KN02_CSR_LEDS; @@ -247,8 +243,6 @@ static inline void dec_kn02_be_init(void) cached_kn02_csr |= KN02_CSR_CORRECT; *csr = cached_kn02_csr; iob(); - - spin_unlock_irqrestore(&kn02_lock, flags); } static inline void dec_kn03_be_init(void) diff --git a/arch/mips/dec/ioasic-irq.c b/arch/mips/dec/ioasic-irq.c index 41cd2a96148..d0af08bdbb4 100644 --- a/arch/mips/dec/ioasic-irq.c +++ b/arch/mips/dec/ioasic-irq.c @@ -13,7 +13,6 @@ #include #include -#include #include #include @@ -21,8 +20,6 @@ #include -static DEFINE_SPINLOCK(ioasic_lock); - static int ioasic_irq_base; @@ -52,65 +49,31 @@ static inline void clear_ioasic_irq(unsigned int irq) ioasic_write(IO_REG_SIR, sir); } -static inline void enable_ioasic_irq(unsigned int irq) -{ - unsigned long flags; - - spin_lock_irqsave(&ioasic_lock, flags); - unmask_ioasic_irq(irq); - spin_unlock_irqrestore(&ioasic_lock, flags); -} - -static inline void disable_ioasic_irq(unsigned int irq) -{ - unsigned long flags; - - spin_lock_irqsave(&ioasic_lock, flags); - mask_ioasic_irq(irq); - spin_unlock_irqrestore(&ioasic_lock, flags); -} - - -static inline unsigned int startup_ioasic_irq(unsigned int irq) -{ - enable_ioasic_irq(irq); - return 0; -} - -#define shutdown_ioasic_irq disable_ioasic_irq - static inline void ack_ioasic_irq(unsigned int irq) { - spin_lock(&ioasic_lock); mask_ioasic_irq(irq); - spin_unlock(&ioasic_lock); fast_iob(); } static inline void end_ioasic_irq(unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS))) - enable_ioasic_irq(irq); + unmask_ioasic_irq(irq); } static struct irq_chip ioasic_irq_type = { .typename = "IO-ASIC", - .startup = startup_ioasic_irq, - .shutdown = shutdown_ioasic_irq, - .enable = enable_ioasic_irq, - .disable = disable_ioasic_irq, .ack = ack_ioasic_irq, + .mask = mask_ioasic_irq, + .mask_ack = ack_ioasic_irq, + .unmask = unmask_ioasic_irq, .end = end_ioasic_irq, }; -#define startup_ioasic_dma_irq startup_ioasic_irq - -#define shutdown_ioasic_dma_irq shutdown_ioasic_irq - -#define enable_ioasic_dma_irq enable_ioasic_irq +#define unmask_ioasic_dma_irq unmask_ioasic_irq -#define disable_ioasic_dma_irq disable_ioasic_irq +#define mask_ioasic_dma_irq mask_ioasic_irq #define ack_ioasic_dma_irq ack_ioasic_irq @@ -123,11 +86,10 @@ static inline void end_ioasic_dma_irq(unsigned int irq) static struct irq_chip ioasic_dma_irq_type = { .typename = "IO-ASIC-DMA", - .startup = startup_ioasic_dma_irq, - .shutdown = shutdown_ioasic_dma_irq, - .enable = enable_ioasic_dma_irq, - .disable = disable_ioasic_dma_irq, .ack = ack_ioasic_dma_irq, + .mask = mask_ioasic_dma_irq, + .mask_ack = ack_ioasic_dma_irq, + .unmask = unmask_ioasic_dma_irq, .end = end_ioasic_dma_irq, }; @@ -140,18 +102,10 @@ void __init init_ioasic_irqs(int base) ioasic_write(IO_REG_SIMR, 0); fast_iob(); - for (i = base; i < base + IO_INR_DMA; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 1; - irq_desc[i].chip = &ioasic_irq_type; - } - for (; i < base + IO_IRQ_LINES; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 1; - irq_desc[i].chip = &ioasic_dma_irq_type; - } + for (i = base; i < base + IO_INR_DMA; i++) + set_irq_chip(i, &ioasic_irq_type); + for (; i < base + IO_IRQ_LINES; i++) + set_irq_chip(i, &ioasic_dma_irq_type); ioasic_irq_base = base; } diff --git a/arch/mips/dec/kn02-irq.c b/arch/mips/dec/kn02-irq.c index 04a367a60a5..c761d97787e 100644 --- a/arch/mips/dec/kn02-irq.c +++ b/arch/mips/dec/kn02-irq.c @@ -14,7 +14,6 @@ #include #include -#include #include #include @@ -29,7 +28,6 @@ * There is no default value -- it has to be initialized. */ u32 cached_kn02_csr; -DEFINE_SPINLOCK(kn02_lock); static int kn02_irq_base; @@ -53,54 +51,24 @@ static inline void mask_kn02_irq(unsigned int irq) *csr = cached_kn02_csr; } -static inline void enable_kn02_irq(unsigned int irq) -{ - unsigned long flags; - - spin_lock_irqsave(&kn02_lock, flags); - unmask_kn02_irq(irq); - spin_unlock_irqrestore(&kn02_lock, flags); -} - -static inline void disable_kn02_irq(unsigned int irq) -{ - unsigned long flags; - - spin_lock_irqsave(&kn02_lock, flags); - mask_kn02_irq(irq); - spin_unlock_irqrestore(&kn02_lock, flags); -} - - -static unsigned int startup_kn02_irq(unsigned int irq) -{ - enable_kn02_irq(irq); - return 0; -} - -#define shutdown_kn02_irq disable_kn02_irq - static void ack_kn02_irq(unsigned int irq) { - spin_lock(&kn02_lock); mask_kn02_irq(irq); - spin_unlock(&kn02_lock); iob(); } static void end_kn02_irq(unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS))) - enable_kn02_irq(irq); + unmask_kn02_irq(irq); } static struct irq_chip kn02_irq_type = { .typename = "KN02-CSR", - .startup = startup_kn02_irq, - .shutdown = shutdown_kn02_irq, - .enable = enable_kn02_irq, - .disable = disable_kn02_irq, .ack = ack_kn02_irq, + .mask = mask_kn02_irq, + .mask_ack = ack_kn02_irq, + .unmask = unmask_kn02_irq, .end = end_kn02_irq, }; @@ -109,22 +77,15 @@ void __init init_kn02_irqs(int base) { volatile u32 *csr = (volatile u32 *)CKSEG1ADDR(KN02_SLOT_BASE + KN02_CSR); - unsigned long flags; int i; /* Mask interrupts. */ - spin_lock_irqsave(&kn02_lock, flags); cached_kn02_csr &= ~KN02_CSR_IOINTEN; *csr = cached_kn02_csr; iob(); - spin_unlock_irqrestore(&kn02_lock, flags); - - for (i = base; i < base + KN02_IRQ_LINES; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 1; - irq_desc[i].chip = &kn02_irq_type; - } + + for (i = base; i < base + KN02_IRQ_LINES; i++) + set_irq_chip(i, &kn02_irq_type); kn02_irq_base = base; } diff --git a/arch/mips/emma2rh/common/irq_emma2rh.c b/arch/mips/emma2rh/common/irq_emma2rh.c index 197ed4c2ba0..bf1b83ba925 100644 --- a/arch/mips/emma2rh/common/irq_emma2rh.c +++ b/arch/mips/emma2rh/common/irq_emma2rh.c @@ -56,22 +56,6 @@ static void emma2rh_irq_disable(unsigned int irq) ll_emma2rh_irq_disable(irq - emma2rh_irq_base); } -static unsigned int emma2rh_irq_startup(unsigned int irq) -{ - emma2rh_irq_enable(irq); - return 0; -} - -#define emma2rh_irq_shutdown emma2rh_irq_disable - -static void emma2rh_irq_ack(unsigned int irq) -{ - /* disable interrupt - some handler will re-enable the irq - * and if the interrupt is leveled, we will have infinite loop - */ - ll_emma2rh_irq_disable(irq - emma2rh_irq_base); -} - static void emma2rh_irq_end(unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS))) @@ -80,25 +64,19 @@ static void emma2rh_irq_end(unsigned int irq) struct irq_chip emma2rh_irq_controller = { .typename = "emma2rh_irq", - .startup = emma2rh_irq_startup, - .shutdown = emma2rh_irq_shutdown, - .enable = emma2rh_irq_enable, - .disable = emma2rh_irq_disable, - .ack = emma2rh_irq_ack, + .ack = emma2rh_irq_disable, + .mask = emma2rh_irq_disable, + .mask_ack = emma2rh_irq_disable, + .unmask = emma2rh_irq_enable, .end = emma2rh_irq_end, - .set_affinity = NULL /* no affinity stuff for UP */ }; void emma2rh_irq_init(u32 irq_base) { u32 i; - for (i = irq_base; i < irq_base + NUM_EMMA2RH_IRQ; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = NULL; - irq_desc[i].depth = 1; - irq_desc[i].chip = &emma2rh_irq_controller; - } + for (i = irq_base; i < irq_base + NUM_EMMA2RH_IRQ; i++) + set_irq_chip(i, &emma2rh_irq_controller); emma2rh_irq_base = irq_base; } diff --git a/arch/mips/emma2rh/markeins/irq_markeins.c b/arch/mips/emma2rh/markeins/irq_markeins.c index 0b36eb001e6..8e5f08a4245 100644 --- a/arch/mips/emma2rh/markeins/irq_markeins.c +++ b/arch/mips/emma2rh/markeins/irq_markeins.c @@ -48,19 +48,6 @@ static void emma2rh_sw_irq_disable(unsigned int irq) ll_emma2rh_sw_irq_disable(irq - emma2rh_sw_irq_base); } -static unsigned int emma2rh_sw_irq_startup(unsigned int irq) -{ - emma2rh_sw_irq_enable(irq); - return 0; -} - -#define emma2rh_sw_irq_shutdown emma2rh_sw_irq_disable - -static void emma2rh_sw_irq_ack(unsigned int irq) -{ - ll_emma2rh_sw_irq_disable(irq - emma2rh_sw_irq_base); -} - static void emma2rh_sw_irq_end(unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS))) @@ -69,25 +56,19 @@ static void emma2rh_sw_irq_end(unsigned int irq) struct irq_chip emma2rh_sw_irq_controller = { .typename = "emma2rh_sw_irq", - .startup = emma2rh_sw_irq_startup, - .shutdown = emma2rh_sw_irq_shutdown, - .enable = emma2rh_sw_irq_enable, - .disable = emma2rh_sw_irq_disable, - .ack = emma2rh_sw_irq_ack, + .ack = emma2rh_sw_irq_disable, + .mask = emma2rh_sw_irq_disable, + .mask_ack = emma2rh_sw_irq_disable, + .unmask = emma2rh_sw_irq_enable, .end = emma2rh_sw_irq_end, - .set_affinity = NULL, }; void emma2rh_sw_irq_init(u32 irq_base) { u32 i; - for (i = irq_base; i < irq_base + NUM_EMMA2RH_IRQ_SW; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = NULL; - irq_desc[i].depth = 2; - irq_desc[i].chip = &emma2rh_sw_irq_controller; - } + for (i = irq_base; i < irq_base + NUM_EMMA2RH_IRQ_SW; i++) + set_irq_chip(i, &emma2rh_sw_irq_controller); emma2rh_sw_irq_base = irq_base; } @@ -126,14 +107,6 @@ static void emma2rh_gpio_irq_disable(unsigned int irq) ll_emma2rh_gpio_irq_disable(irq - emma2rh_gpio_irq_base); } -static unsigned int emma2rh_gpio_irq_startup(unsigned int irq) -{ - emma2rh_gpio_irq_enable(irq); - return 0; -} - -#define emma2rh_gpio_irq_shutdown emma2rh_gpio_irq_disable - static void emma2rh_gpio_irq_ack(unsigned int irq) { irq -= emma2rh_gpio_irq_base; @@ -149,25 +122,19 @@ static void emma2rh_gpio_irq_end(unsigned int irq) struct irq_chip emma2rh_gpio_irq_controller = { .typename = "emma2rh_gpio_irq", - .startup = emma2rh_gpio_irq_startup, - .shutdown = emma2rh_gpio_irq_shutdown, - .enable = emma2rh_gpio_irq_enable, - .disable = emma2rh_gpio_irq_disable, .ack = emma2rh_gpio_irq_ack, + .mask = emma2rh_gpio_irq_disable, + .mask_ack = emma2rh_gpio_irq_ack, + .unmask = emma2rh_gpio_irq_enable, .end = emma2rh_gpio_irq_end, - .set_affinity = NULL, }; void emma2rh_gpio_irq_init(u32 irq_base) { u32 i; - for (i = irq_base; i < irq_base + NUM_EMMA2RH_IRQ_GPIO; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = NULL; - irq_desc[i].depth = 2; - irq_desc[i].chip = &emma2rh_gpio_irq_controller; - } + for (i = irq_base; i < irq_base + NUM_EMMA2RH_IRQ_GPIO; i++) + set_irq_chip(i, &emma2rh_gpio_irq_controller); emma2rh_gpio_irq_base = irq_base; } diff --git a/arch/mips/gt64120/ev64120/irq.c b/arch/mips/gt64120/ev64120/irq.c index ed4d82b9a24..b3e5796c81d 100644 --- a/arch/mips/gt64120/ev64120/irq.c +++ b/arch/mips/gt64120/ev64120/irq.c @@ -66,38 +66,21 @@ asmlinkage void plat_irq_dispatch(void) static void disable_ev64120_irq(unsigned int irq_nr) { - unsigned long flags; - - local_irq_save(flags); if (irq_nr >= 8) { // All PCI interrupts are on line 5 or 2 clear_c0_status(9 << 10); } else { clear_c0_status(1 << (irq_nr + 8)); } - local_irq_restore(flags); } static void enable_ev64120_irq(unsigned int irq_nr) { - unsigned long flags; - - local_irq_save(flags); if (irq_nr >= 8) // All PCI interrupts are on line 5 or 2 set_c0_status(9 << 10); else set_c0_status(1 << (irq_nr + 8)); - local_irq_restore(flags); -} - -static unsigned int startup_ev64120_irq(unsigned int irq) -{ - enable_ev64120_irq(irq); - return 0; /* Never anything pending */ } -#define shutdown_ev64120_irq disable_ev64120_irq -#define mask_and_ack_ev64120_irq disable_ev64120_irq - static void end_ev64120_irq(unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS))) @@ -106,13 +89,11 @@ static void end_ev64120_irq(unsigned int irq) static struct irq_chip ev64120_irq_type = { .typename = "EV64120", - .startup = startup_ev64120_irq, - .shutdown = shutdown_ev64120_irq, - .enable = enable_ev64120_irq, - .disable = disable_ev64120_irq, - .ack = mask_and_ack_ev64120_irq, + .ack = disable_ev64120_irq, + .mask = disable_ev64120_irq, + .mask_ack = disable_ev64120_irq, + .unmask = enable_ev64120_irq, .end = end_ev64120_irq, - .set_affinity = NULL }; void gt64120_irq_setup(void) @@ -122,8 +103,6 @@ void gt64120_irq_setup(void) */ clear_c0_status(ST0_IM); - local_irq_disable(); - /* * Enable timer. Other interrupts will be enabled as they are * registered. @@ -133,16 +112,5 @@ void gt64120_irq_setup(void) void __init arch_init_irq(void) { - int i; - - /* Let's initialize our IRQ descriptors */ - for (i = 0; i < NR_IRQS; i++) { - irq_desc[i].status = 0; - irq_desc[i].chip = &no_irq_chip; - irq_desc[i].action = NULL; - irq_desc[i].depth = 0; - spin_lock_init(&irq_desc[i].lock); - } - gt64120_irq_setup(); } diff --git a/arch/mips/jazz/irq.c b/arch/mips/jazz/irq.c index d5bd6b3a093..4bbb6cb08d6 100644 --- a/arch/mips/jazz/irq.c +++ b/arch/mips/jazz/irq.c @@ -28,14 +28,6 @@ static void enable_r4030_irq(unsigned int irq) spin_unlock_irqrestore(&r4030_lock, flags); } -static unsigned int startup_r4030_irq(unsigned int irq) -{ - enable_r4030_irq(irq); - return 0; /* never anything pending */ -} - -#define shutdown_r4030_irq disable_r4030_irq - void disable_r4030_irq(unsigned int irq) { unsigned int mask = ~(1 << (irq - JAZZ_PARALLEL_IRQ)); @@ -47,8 +39,6 @@ void disable_r4030_irq(unsigned int irq) spin_unlock_irqrestore(&r4030_lock, flags); } -#define mask_and_ack_r4030_irq disable_r4030_irq - static void end_r4030_irq(unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS))) @@ -57,11 +47,10 @@ static void end_r4030_irq(unsigned int irq) static struct irq_chip r4030_irq_type = { .typename = "R4030", - .startup = startup_r4030_irq, - .shutdown = shutdown_r4030_irq, - .enable = enable_r4030_irq, - .disable = disable_r4030_irq, - .ack = mask_and_ack_r4030_irq, + .ack = disable_r4030_irq, + .mask = disable_r4030_irq, + .mask_ack = disable_r4030_irq, + .unmask = enable_r4030_irq, .end = end_r4030_irq, }; @@ -69,12 +58,8 @@ void __init init_r4030_ints(void) { int i; - for (i = JAZZ_PARALLEL_IRQ; i <= JAZZ_TIMER_IRQ; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 1; - irq_desc[i].chip = &r4030_irq_type; - } + for (i = JAZZ_PARALLEL_IRQ; i <= JAZZ_TIMER_IRQ; i++) + set_irq_chip(i, &r4030_irq_type); r4030_write_reg16(JAZZ_IO_IRQ_ENABLE, 0); r4030_read_reg16(JAZZ_IO_IRQ_SOURCE); /* clear pending IRQs */ diff --git a/arch/mips/jmr3927/rbhma3100/irq.c b/arch/mips/jmr3927/rbhma3100/irq.c index de4a238c28b..3da49c5aaf4 100644 --- a/arch/mips/jmr3927/rbhma3100/irq.c +++ b/arch/mips/jmr3927/rbhma3100/irq.c @@ -90,17 +90,6 @@ static unsigned char irc_level[TX3927_NUM_IR] = { static void jmr3927_irq_disable(unsigned int irq_nr); static void jmr3927_irq_enable(unsigned int irq_nr); -static DEFINE_SPINLOCK(jmr3927_irq_lock); - -static unsigned int jmr3927_irq_startup(unsigned int irq) -{ - jmr3927_irq_enable(irq); - - return 0; -} - -#define jmr3927_irq_shutdown jmr3927_irq_disable - static void jmr3927_irq_ack(unsigned int irq) { if (irq == JMR3927_IRQ_IRC_TMR0) @@ -118,9 +107,7 @@ static void jmr3927_irq_end(unsigned int irq) static void jmr3927_irq_disable(unsigned int irq_nr) { struct tb_irq_space* sp; - unsigned long flags; - spin_lock_irqsave(&jmr3927_irq_lock, flags); for (sp = tb_irq_spaces; sp; sp = sp->next) { if (sp->start_irqno <= irq_nr && irq_nr < sp->start_irqno + sp->nr_irqs) { @@ -130,15 +117,12 @@ static void jmr3927_irq_disable(unsigned int irq_nr) break; } } - spin_unlock_irqrestore(&jmr3927_irq_lock, flags); } static void jmr3927_irq_enable(unsigned int irq_nr) { struct tb_irq_space* sp; - unsigned long flags; - spin_lock_irqsave(&jmr3927_irq_lock, flags); for (sp = tb_irq_spaces; sp; sp = sp->next) { if (sp->start_irqno <= irq_nr && irq_nr < sp->start_irqno + sp->nr_irqs) { @@ -148,7 +132,6 @@ static void jmr3927_irq_enable(unsigned int irq_nr) break; } } - spin_unlock_irqrestore(&jmr3927_irq_lock, flags); } /* @@ -457,11 +440,10 @@ void __init arch_init_irq(void) static struct irq_chip jmr3927_irq_controller = { .typename = "jmr3927_irq", - .startup = jmr3927_irq_startup, - .shutdown = jmr3927_irq_shutdown, - .enable = jmr3927_irq_enable, - .disable = jmr3927_irq_disable, .ack = jmr3927_irq_ack, + .mask = jmr3927_irq_disable, + .mask_ack = jmr3927_irq_ack, + .unmask = jmr3927_irq_enable, .end = jmr3927_irq_end, }; @@ -469,12 +451,8 @@ void jmr3927_irq_init(u32 irq_base) { u32 i; - for (i= irq_base; i< irq_base + JMR3927_NR_IRQ_IRC + JMR3927_NR_IRQ_IOC; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = NULL; - irq_desc[i].depth = 1; - irq_desc[i].chip = &jmr3927_irq_controller; - } + for (i= irq_base; i< irq_base + JMR3927_NR_IRQ_IRC + JMR3927_NR_IRQ_IOC; i++) + set_irq_chip(i, &jmr3927_irq_controller); jmr3927_irq_base = irq_base; } diff --git a/arch/mips/kernel/i8259.c b/arch/mips/kernel/i8259.c index 48e3418c217..2526c0ca4d8 100644 --- a/arch/mips/kernel/i8259.c +++ b/arch/mips/kernel/i8259.c @@ -40,21 +40,10 @@ static void end_8259A_irq (unsigned int irq) enable_8259A_irq(irq); } -#define shutdown_8259A_irq disable_8259A_irq - void mask_and_ack_8259A(unsigned int); -static unsigned int startup_8259A_irq(unsigned int irq) -{ - enable_8259A_irq(irq); - - return 0; /* never anything pending */ -} - static struct irq_chip i8259A_irq_type = { .typename = "XT-PIC", - .startup = startup_8259A_irq, - .shutdown = shutdown_8259A_irq, .enable = enable_8259A_irq, .disable = disable_8259A_irq, .ack = mask_and_ack_8259A, @@ -120,7 +109,7 @@ int i8259A_irq_pending(unsigned int irq) void make_8259A_irq(unsigned int irq) { disable_irq_nosync(irq); - irq_desc[irq].chip = &i8259A_irq_type; + set_irq_chip(irq, &i8259A_irq_type); enable_irq(irq); } @@ -323,12 +312,8 @@ void __init init_i8259_irqs (void) init_8259A(0); - for (i = 0; i < 16; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = NULL; - irq_desc[i].depth = 1; - irq_desc[i].chip = &i8259A_irq_type; - } + for (i = 0; i < 16; i++) + set_irq_chip(i, &i8259A_irq_type); setup_irq(2, &irq2); } diff --git a/arch/mips/kernel/irq-msc01.c b/arch/mips/kernel/irq-msc01.c index 650a80ca374..e1880b27381 100644 --- a/arch/mips/kernel/irq-msc01.c +++ b/arch/mips/kernel/irq-msc01.c @@ -44,31 +44,6 @@ static inline void unmask_msc_irq(unsigned int irq) MSCIC_WRITE(MSC01_IC_ENAH, 1<<(irq - irq_base - 32)); } -/* - * Enables the IRQ on SOC-it - */ -static void enable_msc_irq(unsigned int irq) -{ - unmask_msc_irq(irq); -} - -/* - * Initialize the IRQ on SOC-it - */ -static unsigned int startup_msc_irq(unsigned int irq) -{ - unmask_msc_irq(irq); - return 0; -} - -/* - * Disables the IRQ on SOC-it - */ -static void disable_msc_irq(unsigned int irq) -{ - mask_msc_irq(irq); -} - /* * Masks and ACKs an IRQ */ @@ -136,25 +111,21 @@ msc_bind_eic_interrupt (unsigned int irq, unsigned int set) (irq<im_type) { case MSC01_IRQ_EDGE: - irq_desc[base+n].chip = &msc_edgeirq_type; + set_irq_chip(base+n, &msc_edgeirq_type); if (cpu_has_veic) MSCIC_WRITE(MSC01_IC_SUP+n*8, MSC01_IC_SUP_EDGE_BIT); else MSCIC_WRITE(MSC01_IC_SUP+n*8, MSC01_IC_SUP_EDGE_BIT | imp->im_lvl); break; case MSC01_IRQ_LEVEL: - irq_desc[base+n].chip = &msc_levelirq_type; + set_irq_chip(base+n, &msc_levelirq_type); if (cpu_has_veic) MSCIC_WRITE(MSC01_IC_SUP+n*8, 0); else diff --git a/arch/mips/kernel/irq-mv6434x.c b/arch/mips/kernel/irq-mv6434x.c index 37d106202b8..5012b9df1b5 100644 --- a/arch/mips/kernel/irq-mv6434x.c +++ b/arch/mips/kernel/irq-mv6434x.c @@ -66,39 +66,6 @@ static inline void unmask_mv64340_irq(unsigned int irq) } } -/* - * Enables the IRQ on Marvell Chip - */ -static void enable_mv64340_irq(unsigned int irq) -{ - unmask_mv64340_irq(irq); -} - -/* - * Initialize the IRQ on Marvell Chip - */ -static unsigned int startup_mv64340_irq(unsigned int irq) -{ - unmask_mv64340_irq(irq); - return 0; -} - -/* - * Disables the IRQ on Marvell Chip - */ -static void disable_mv64340_irq(unsigned int irq) -{ - mask_mv64340_irq(irq); -} - -/* - * Masks and ACKs an IRQ - */ -static void mask_and_ack_mv64340_irq(unsigned int irq) -{ - mask_mv64340_irq(irq); -} - /* * End IRQ processing */ @@ -133,15 +100,12 @@ void ll_mv64340_irq(void) do_IRQ(ls1bit32(irq_src_high) + irq_base + 32); } -#define shutdown_mv64340_irq disable_mv64340_irq - struct irq_chip mv64340_irq_type = { .typename = "MV-64340", - .startup = startup_mv64340_irq, - .shutdown = shutdown_mv64340_irq, - .enable = enable_mv64340_irq, - .disable = disable_mv64340_irq, - .ack = mask_and_ack_mv64340_irq, + .ack = mask_mv64340_irq, + .mask = mask_mv64340_irq, + .mask_ack = mask_mv64340_irq, + .unmask = unmask_mv64340_irq, .end = end_mv64340_irq, }; @@ -149,13 +113,8 @@ void __init mv64340_irq_init(unsigned int base) { int i; - /* Reset irq handlers pointers to NULL */ - for (i = base; i < base + 64; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 2; - irq_desc[i].chip = &mv64340_irq_type; - } + for (i = base; i < base + 64; i++) + set_irq_chip(i, &mv64340_irq_type); irq_base = base; } diff --git a/arch/mips/kernel/irq-rm7000.c b/arch/mips/kernel/irq-rm7000.c index 6b54c7109e2..6a297e3b889 100644 --- a/arch/mips/kernel/irq-rm7000.c +++ b/arch/mips/kernel/irq-rm7000.c @@ -29,42 +29,6 @@ static inline void mask_rm7k_irq(unsigned int irq) clear_c0_intcontrol(0x100 << (irq - irq_base)); } -static inline void rm7k_cpu_irq_enable(unsigned int irq) -{ - unsigned long flags; - - local_irq_save(flags); - unmask_rm7k_irq(irq); - local_irq_restore(flags); -} - -static void rm7k_cpu_irq_disable(unsigned int irq) -{ - unsigned long flags; - - local_irq_save(flags); - mask_rm7k_irq(irq); - local_irq_restore(flags); -} - -static unsigned int rm7k_cpu_irq_startup(unsigned int irq) -{ - rm7k_cpu_irq_enable(irq); - - return 0; -} - -#define rm7k_cpu_irq_shutdown rm7k_cpu_irq_disable - -/* - * While we ack the interrupt interrupts are disabled and thus we don't need - * to deal with concurrency issues. Same for rm7k_cpu_irq_end. - */ -static void rm7k_cpu_irq_ack(unsigned int irq) -{ - mask_rm7k_irq(irq); -} - static void rm7k_cpu_irq_end(unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS))) @@ -73,11 +37,10 @@ static void rm7k_cpu_irq_end(unsigned int irq) static struct irq_chip rm7k_irq_controller = { .typename = "RM7000", - .startup = rm7k_cpu_irq_startup, - .shutdown = rm7k_cpu_irq_shutdown, - .enable = rm7k_cpu_irq_enable, - .disable = rm7k_cpu_irq_disable, - .ack = rm7k_cpu_irq_ack, + .ack = mask_rm7k_irq, + .mask = mask_rm7k_irq, + .mask_ack = mask_rm7k_irq, + .unmask = unmask_rm7k_irq, .end = rm7k_cpu_irq_end, }; @@ -87,12 +50,8 @@ void __init rm7k_cpu_irq_init(int base) clear_c0_intcontrol(0x00000f00); /* Mask all */ - for (i = base; i < base + 4; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = NULL; - irq_desc[i].depth = 1; - irq_desc[i].chip = &rm7k_irq_controller; - } + for (i = base; i < base + 4; i++) + set_irq_chip(i, &rm7k_irq_controller); irq_base = base; } diff --git a/arch/mips/kernel/irq-rm9000.c b/arch/mips/kernel/irq-rm9000.c index 62f011ba97a..977538445cf 100644 --- a/arch/mips/kernel/irq-rm9000.c +++ b/arch/mips/kernel/irq-rm9000.c @@ -48,15 +48,6 @@ static void rm9k_cpu_irq_disable(unsigned int irq) local_irq_restore(flags); } -static unsigned int rm9k_cpu_irq_startup(unsigned int irq) -{ - rm9k_cpu_irq_enable(irq); - - return 0; -} - -#define rm9k_cpu_irq_shutdown rm9k_cpu_irq_disable - /* * Performance counter interrupts are global on all processors. */ @@ -89,16 +80,6 @@ static void rm9k_perfcounter_irq_shutdown(unsigned int irq) on_each_cpu(local_rm9k_perfcounter_irq_shutdown, (void *) irq, 0, 1); } - -/* - * While we ack the interrupt interrupts are disabled and thus we don't need - * to deal with concurrency issues. Same for rm9k_cpu_irq_end. - */ -static void rm9k_cpu_irq_ack(unsigned int irq) -{ - mask_rm9k_irq(irq); -} - static void rm9k_cpu_irq_end(unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS))) @@ -107,11 +88,10 @@ static void rm9k_cpu_irq_end(unsigned int irq) static struct irq_chip rm9k_irq_controller = { .typename = "RM9000", - .startup = rm9k_cpu_irq_startup, - .shutdown = rm9k_cpu_irq_shutdown, - .enable = rm9k_cpu_irq_enable, - .disable = rm9k_cpu_irq_disable, - .ack = rm9k_cpu_irq_ack, + .ack = mask_rm9k_irq, + .mask = mask_rm9k_irq, + .mask_ack = mask_rm9k_irq, + .unmask = unmask_rm9k_irq, .end = rm9k_cpu_irq_end, }; @@ -119,9 +99,10 @@ static struct irq_chip rm9k_perfcounter_irq = { .typename = "RM9000", .startup = rm9k_perfcounter_irq_startup, .shutdown = rm9k_perfcounter_irq_shutdown, - .enable = rm9k_cpu_irq_enable, - .disable = rm9k_cpu_irq_disable, - .ack = rm9k_cpu_irq_ack, + .ack = mask_rm9k_irq, + .mask = mask_rm9k_irq, + .mask_ack = mask_rm9k_irq, + .unmask = unmask_rm9k_irq, .end = rm9k_cpu_irq_end, }; @@ -135,15 +116,11 @@ void __init rm9k_cpu_irq_init(int base) clear_c0_intcontrol(0x0000f000); /* Mask all */ - for (i = base; i < base + 4; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = NULL; - irq_desc[i].depth = 1; - irq_desc[i].chip = &rm9k_irq_controller; - } + for (i = base; i < base + 4; i++) + set_irq_chip(i, &rm9k_irq_controller); rm9000_perfcount_irq = base + 1; - irq_desc[rm9000_perfcount_irq].chip = &rm9k_perfcounter_irq; + set_irq_chip(rm9000_perfcount_irq, &rm9k_perfcounter_irq); irq_base = base; } diff --git a/arch/mips/kernel/irq.c b/arch/mips/kernel/irq.c index 9b0e49d63d7..e997c9454f4 100644 --- a/arch/mips/kernel/irq.c +++ b/arch/mips/kernel/irq.c @@ -172,19 +172,6 @@ __setup("nokgdb", nokgdb); void __init init_IRQ(void) { - int i; - - for (i = 0; i < NR_IRQS; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = NULL; - irq_desc[i].depth = 1; - irq_desc[i].chip = &no_irq_chip; - spin_lock_init(&irq_desc[i].lock); -#ifdef CONFIG_MIPS_MT_SMTC - irq_hwmask[i] = 0; -#endif /* CONFIG_MIPS_MT_SMTC */ - } - arch_init_irq(); #ifdef CONFIG_KGDB diff --git a/arch/mips/kernel/irq_cpu.c b/arch/mips/kernel/irq_cpu.c index 9bb21c7f214..3b7cfa407e8 100644 --- a/arch/mips/kernel/irq_cpu.c +++ b/arch/mips/kernel/irq_cpu.c @@ -50,44 +50,6 @@ static inline void mask_mips_irq(unsigned int irq) irq_disable_hazard(); } -static inline void mips_cpu_irq_enable(unsigned int irq) -{ - unsigned long flags; - - local_irq_save(flags); - unmask_mips_irq(irq); - back_to_back_c0_hazard(); - local_irq_restore(flags); -} - -static void mips_cpu_irq_disable(unsigned int irq) -{ - unsigned long flags; - - local_irq_save(flags); - mask_mips_irq(irq); - back_to_back_c0_hazard(); - local_irq_restore(flags); -} - -static unsigned int mips_cpu_irq_startup(unsigned int irq) -{ - mips_cpu_irq_enable(irq); - - return 0; -} - -#define mips_cpu_irq_shutdown mips_cpu_irq_disable - -/* - * While we ack the interrupt interrupts are disabled and thus we don't need - * to deal with concurrency issues. Same for mips_cpu_irq_end. - */ -static void mips_cpu_irq_ack(unsigned int irq) -{ - mask_mips_irq(irq); -} - static void mips_cpu_irq_end(unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS))) @@ -96,11 +58,10 @@ static void mips_cpu_irq_end(unsigned int irq) static struct irq_chip mips_cpu_irq_controller = { .typename = "MIPS", - .startup = mips_cpu_irq_startup, - .shutdown = mips_cpu_irq_shutdown, - .enable = mips_cpu_irq_enable, - .disable = mips_cpu_irq_disable, - .ack = mips_cpu_irq_ack, + .ack = mask_mips_irq, + .mask = mask_mips_irq, + .mask_ack = mask_mips_irq, + .unmask = unmask_mips_irq, .end = mips_cpu_irq_end, }; @@ -110,8 +71,6 @@ static struct irq_chip mips_cpu_irq_controller = { #define unmask_mips_mt_irq unmask_mips_irq #define mask_mips_mt_irq mask_mips_irq -#define mips_mt_cpu_irq_enable mips_cpu_irq_enable -#define mips_mt_cpu_irq_disable mips_cpu_irq_disable static unsigned int mips_mt_cpu_irq_startup(unsigned int irq) { @@ -119,13 +78,11 @@ static unsigned int mips_mt_cpu_irq_startup(unsigned int irq) clear_c0_cause(0x100 << (irq - mips_cpu_irq_base)); evpe(vpflags); - mips_mt_cpu_irq_enable(irq); + unmask_mips_mt_irq(irq); return 0; } -#define mips_mt_cpu_irq_shutdown mips_mt_cpu_irq_disable - /* * While we ack the interrupt interrupts are disabled and thus we don't need * to deal with concurrency issues. Same for mips_cpu_irq_end. @@ -143,10 +100,10 @@ static void mips_mt_cpu_irq_ack(unsigned int irq) static struct irq_chip mips_mt_cpu_irq_controller = { .typename = "MIPS", .startup = mips_mt_cpu_irq_startup, - .shutdown = mips_mt_cpu_irq_shutdown, - .enable = mips_mt_cpu_irq_enable, - .disable = mips_mt_cpu_irq_disable, .ack = mips_mt_cpu_irq_ack, + .mask = mask_mips_mt_irq, + .mask_ack = mips_mt_cpu_irq_ack, + .unmask = unmask_mips_mt_irq, .end = mips_mt_cpu_irq_end, }; @@ -163,19 +120,11 @@ void __init mips_cpu_irq_init(int irq_base) * leave them uninitialized for other processors. */ if (cpu_has_mipsmt) - for (i = irq_base; i < irq_base + 2; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = NULL; - irq_desc[i].depth = 1; - irq_desc[i].chip = &mips_mt_cpu_irq_controller; - } - - for (i = irq_base + 2; i < irq_base + 8; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = NULL; - irq_desc[i].depth = 1; - irq_desc[i].chip = &mips_cpu_irq_controller; - } + for (i = irq_base; i < irq_base + 2; i++) + set_irq_chip(i, &mips_mt_cpu_irq_controller); + + for (i = irq_base + 2; i < irq_base + 8; i++) + set_irq_chip(i, &mips_cpu_irq_controller); mips_cpu_irq_base = irq_base; } diff --git a/arch/mips/lasat/interrupt.c b/arch/mips/lasat/interrupt.c index a144a002dcc..cac82afe5eb 100644 --- a/arch/mips/lasat/interrupt.c +++ b/arch/mips/lasat/interrupt.c @@ -36,33 +36,14 @@ static volatile int lasat_int_mask_shift; void disable_lasat_irq(unsigned int irq_nr) { - unsigned long flags; - - local_irq_save(flags); *lasat_int_mask &= ~(1 << irq_nr) << lasat_int_mask_shift; - local_irq_restore(flags); } void enable_lasat_irq(unsigned int irq_nr) { - unsigned long flags; - - local_irq_save(flags); *lasat_int_mask |= (1 << irq_nr) << lasat_int_mask_shift; - local_irq_restore(flags); } -static unsigned int startup_lasat_irq(unsigned int irq) -{ - enable_lasat_irq(irq); - - return 0; /* never anything pending */ -} - -#define shutdown_lasat_irq disable_lasat_irq - -#define mask_and_ack_lasat_irq disable_lasat_irq - static void end_lasat_irq(unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS))) @@ -71,11 +52,10 @@ static void end_lasat_irq(unsigned int irq) static struct irq_chip lasat_irq_type = { .typename = "Lasat", - .startup = startup_lasat_irq, - .shutdown = shutdown_lasat_irq, - .enable = enable_lasat_irq, - .disable = disable_lasat_irq, - .ack = mask_and_ack_lasat_irq, + .ack = disable_lasat_irq, + .mask = disable_lasat_irq, + .mask_ack = disable_lasat_irq, + .unmask = enable_lasat_irq, .end = end_lasat_irq, }; @@ -152,10 +132,6 @@ void __init arch_init_irq(void) panic("arch_init_irq: mips_machtype incorrect"); } - for (i = 0; i <= LASATINT_END; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 1; - irq_desc[i].chip = &lasat_irq_type; - } + for (i = 0; i <= LASATINT_END; i++) + set_irq_chip(i, &lasat_irq_type); } diff --git a/arch/mips/mips-boards/atlas/atlas_int.c b/arch/mips/mips-boards/atlas/atlas_int.c index be624b8c3b0..7c710040d3f 100644 --- a/arch/mips/mips-boards/atlas/atlas_int.c +++ b/arch/mips/mips-boards/atlas/atlas_int.c @@ -62,16 +62,6 @@ void enable_atlas_irq(unsigned int irq_nr) iob(); } -static unsigned int startup_atlas_irq(unsigned int irq) -{ - enable_atlas_irq(irq); - return 0; /* never anything pending */ -} - -#define shutdown_atlas_irq disable_atlas_irq - -#define mask_and_ack_atlas_irq disable_atlas_irq - static void end_atlas_irq(unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS))) @@ -80,11 +70,10 @@ static void end_atlas_irq(unsigned int irq) static struct irq_chip atlas_irq_type = { .typename = "Atlas", - .startup = startup_atlas_irq, - .shutdown = shutdown_atlas_irq, - .enable = enable_atlas_irq, - .disable = disable_atlas_irq, - .ack = mask_and_ack_atlas_irq, + .ack = disable_atlas_irq, + .mask = disable_atlas_irq, + .mask_ack = disable_atlas_irq, + .unmask = enable_atlas_irq, .end = end_atlas_irq, }; @@ -217,13 +206,8 @@ static inline void init_atlas_irqs (int base) */ atlas_hw0_icregs->intrsten = 0xffffffff; - for (i = ATLAS_INT_BASE; i <= ATLAS_INT_END; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 1; - irq_desc[i].chip = &atlas_irq_type; - spin_lock_init(&irq_desc[i].lock); - } + for (i = ATLAS_INT_BASE; i <= ATLAS_INT_END; i++) + set_irq_chip(i, &atlas_irq_type); } static struct irqaction atlasirq = { diff --git a/arch/mips/momentum/ocelot_c/cpci-irq.c b/arch/mips/momentum/ocelot_c/cpci-irq.c index 47e3fa32b07..7723f099894 100644 --- a/arch/mips/momentum/ocelot_c/cpci-irq.c +++ b/arch/mips/momentum/ocelot_c/cpci-irq.c @@ -65,39 +65,6 @@ static inline void unmask_cpci_irq(unsigned int irq) value = OCELOT_FPGA_READ(INTMASK); } -/* - * Enables the IRQ in the FPGA - */ -static void enable_cpci_irq(unsigned int irq) -{ - unmask_cpci_irq(irq); -} - -/* - * Initialize the IRQ in the FPGA - */ -static unsigned int startup_cpci_irq(unsigned int irq) -{ - unmask_cpci_irq(irq); - return 0; -} - -/* - * Disables the IRQ in the FPGA - */ -static void disable_cpci_irq(unsigned int irq) -{ - mask_cpci_irq(irq); -} - -/* - * Masks and ACKs an IRQ - */ -static void mask_and_ack_cpci_irq(unsigned int irq) -{ - mask_cpci_irq(irq); -} - /* * End IRQ processing */ @@ -125,15 +92,12 @@ void ll_cpci_irq(void) do_IRQ(ls1bit8(irq_src) + CPCI_IRQ_BASE); } -#define shutdown_cpci_irq disable_cpci_irq - struct irq_chip cpci_irq_type = { .typename = "CPCI/FPGA", - .startup = startup_cpci_irq, - .shutdown = shutdown_cpci_irq, - .enable = enable_cpci_irq, - .disable = disable_cpci_irq, - .ack = mask_and_ack_cpci_irq, + .ack = mask_cpci_irq, + .mask = mask_cpci_irq, + .mask_ack = mask_cpci_irq, + .unmask = unmask_cpci_irq, .end = end_cpci_irq, }; @@ -141,11 +105,6 @@ void cpci_irq_init(void) { int i; - /* Reset irq handlers pointers to NULL */ - for (i = CPCI_IRQ_BASE; i < (CPCI_IRQ_BASE + 8); i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 2; - irq_desc[i].chip = &cpci_irq_type; - } + for (i = CPCI_IRQ_BASE; i < (CPCI_IRQ_BASE + 8); i++) + set_irq_chip(i, &cpci_irq_type); } diff --git a/arch/mips/momentum/ocelot_c/uart-irq.c b/arch/mips/momentum/ocelot_c/uart-irq.c index 510257dc205..72faf81b36c 100644 --- a/arch/mips/momentum/ocelot_c/uart-irq.c +++ b/arch/mips/momentum/ocelot_c/uart-irq.c @@ -59,39 +59,6 @@ static inline void unmask_uart_irq(unsigned int irq) value = OCELOT_FPGA_READ(UART_INTMASK); } -/* - * Enables the IRQ in the FPGA - */ -static void enable_uart_irq(unsigned int irq) -{ - unmask_uart_irq(irq); -} - -/* - * Initialize the IRQ in the FPGA - */ -static unsigned int startup_uart_irq(unsigned int irq) -{ - unmask_uart_irq(irq); - return 0; -} - -/* - * Disables the IRQ in the FPGA - */ -static void disable_uart_irq(unsigned int irq) -{ - mask_uart_irq(irq); -} - -/* - * Masks and ACKs an IRQ - */ -static void mask_and_ack_uart_irq(unsigned int irq) -{ - mask_uart_irq(irq); -} - /* * End IRQ processing */ @@ -118,28 +85,17 @@ void ll_uart_irq(void) do_IRQ(ls1bit8(irq_src) + 74); } -#define shutdown_uart_irq disable_uart_irq - struct irq_chip uart_irq_type = { .typename = "UART/FPGA", - .startup = startup_uart_irq, - .shutdown = shutdown_uart_irq, - .enable = enable_uart_irq, - .disable = disable_uart_irq, - .ack = mask_and_ack_uart_irq, + .ack = mask_uart_irq, + .mask = mask_uart_irq, + .mask_ack = mask_uart_irq, + .unmask = unmask_uart_irq, .end = end_uart_irq, }; void uart_irq_init(void) { - /* Reset irq handlers pointers to NULL */ - irq_desc[80].status = IRQ_DISABLED; - irq_desc[80].action = 0; - irq_desc[80].depth = 2; - irq_desc[80].chip = &uart_irq_type; - - irq_desc[81].status = IRQ_DISABLED; - irq_desc[81].action = 0; - irq_desc[81].depth = 2; - irq_desc[81].chip = &uart_irq_type; + set_irq_chip(80, &uart_irq_type); + set_irq_chip(81, &uart_irq_type); } diff --git a/arch/mips/philips/pnx8550/common/int.c b/arch/mips/philips/pnx8550/common/int.c index 710611615ca..e4bf494dd43 100644 --- a/arch/mips/philips/pnx8550/common/int.c +++ b/arch/mips/philips/pnx8550/common/int.c @@ -38,8 +38,6 @@ #include #include -static DEFINE_SPINLOCK(irq_lock); - /* default prio for interrupts */ /* first one is a no-no so therefore always prio 0 (disabled) */ static char gic_prio[PNX8550_INT_GIC_TOTINT] = { @@ -149,38 +147,6 @@ static inline void unmask_irq(unsigned int irq_nr) } } -#define pnx8550_disable pnx8550_ack -static void pnx8550_ack(unsigned int irq) -{ - unsigned long flags; - - spin_lock_irqsave(&irq_lock, flags); - mask_irq(irq); - spin_unlock_irqrestore(&irq_lock, flags); -} - -#define pnx8550_enable pnx8550_unmask -static void pnx8550_unmask(unsigned int irq) -{ - unsigned long flags; - - spin_lock_irqsave(&irq_lock, flags); - unmask_irq(irq); - spin_unlock_irqrestore(&irq_lock, flags); -} - -static unsigned int startup_irq(unsigned int irq_nr) -{ - pnx8550_unmask(irq_nr); - return 0; -} - -static void shutdown_irq(unsigned int irq_nr) -{ - pnx8550_ack(irq_nr); - return; -} - int pnx8550_set_gic_priority(int irq, int priority) { int gic_irq = irq-PNX8550_INT_GIC_MIN; @@ -192,26 +158,19 @@ int pnx8550_set_gic_priority(int irq, int priority) return prev_priority; } -static inline void mask_and_ack_level_irq(unsigned int irq) -{ - pnx8550_disable(irq); - return; -} - static void end_irq(unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS))) { - pnx8550_enable(irq); + unmask_irq(irq); } } static struct irq_chip level_irq_type = { .typename = "PNX Level IRQ", - .startup = startup_irq, - .shutdown = shutdown_irq, - .enable = pnx8550_enable, - .disable = pnx8550_disable, - .ack = mask_and_ack_level_irq, + .ack = mask_irq, + .mask = mask_irq, + .mask_ack = mask_irq, + .unmask = unmask_irq, .end = end_irq, }; @@ -233,8 +192,8 @@ void __init arch_init_irq(void) int configPR; for (i = 0; i < PNX8550_INT_CP0_TOTINT; i++) { - irq_desc[i].chip = &level_irq_type; - pnx8550_ack(i); /* mask the irq just in case */ + set_irq_chip(i, &level_irq_type); + mask_irq(i); /* mask the irq just in case */ } /* init of GIC/IPC interrupts */ @@ -270,7 +229,7 @@ void __init arch_init_irq(void) /* mask/priority is still 0 so we will not get any * interrupts until it is unmasked */ - irq_desc[i].chip = &level_irq_type; + set_irq_chip(i, &level_irq_type); } /* Priority level 0 */ @@ -279,20 +238,19 @@ void __init arch_init_irq(void) /* Set int vector table address */ PNX8550_GIC_VECTOR_0 = PNX8550_GIC_VECTOR_1 = 0; - irq_desc[MIPS_CPU_GIC_IRQ].chip = &level_irq_type; + set_irq_chip(MIPS_CPU_GIC_IRQ, &level_irq_type); setup_irq(MIPS_CPU_GIC_IRQ, &gic_action); /* init of Timer interrupts */ - for (i = PNX8550_INT_TIMER_MIN; i <= PNX8550_INT_TIMER_MAX; i++) { - irq_desc[i].chip = &level_irq_type; - } + for (i = PNX8550_INT_TIMER_MIN; i <= PNX8550_INT_TIMER_MAX; i++) + set_irq_chip(i, &level_irq_type); /* Stop Timer 1-3 */ configPR = read_c0_config7(); configPR |= 0x00000038; write_c0_config7(configPR); - irq_desc[MIPS_CPU_TIMER_IRQ].chip = &level_irq_type; + set_irq_chip(MIPS_CPU_TIMER_IRQ, &level_irq_type); setup_irq(MIPS_CPU_TIMER_IRQ, &timer_action); } diff --git a/arch/mips/sgi-ip22/ip22-eisa.c b/arch/mips/sgi-ip22/ip22-eisa.c index 0d18ed47c47..a1a9af6da7b 100644 --- a/arch/mips/sgi-ip22/ip22-eisa.c +++ b/arch/mips/sgi-ip22/ip22-eisa.c @@ -95,16 +95,11 @@ static irqreturn_t ip22_eisa_intr(int irq, void *dev_id) static void enable_eisa1_irq(unsigned int irq) { - unsigned long flags; u8 mask; - local_irq_save(flags); - mask = inb(EISA_INT1_MASK); mask &= ~((u8) (1 << irq)); outb(mask, EISA_INT1_MASK); - - local_irq_restore(flags); } static unsigned int startup_eisa1_irq(unsigned int irq) @@ -130,8 +125,6 @@ static void disable_eisa1_irq(unsigned int irq) outb(mask, EISA_INT1_MASK); } -#define shutdown_eisa1_irq disable_eisa1_irq - static void mask_and_ack_eisa1_irq(unsigned int irq) { disable_eisa1_irq(irq); @@ -148,25 +141,20 @@ static void end_eisa1_irq(unsigned int irq) static struct irq_chip ip22_eisa1_irq_type = { .typename = "IP22 EISA", .startup = startup_eisa1_irq, - .shutdown = shutdown_eisa1_irq, - .enable = enable_eisa1_irq, - .disable = disable_eisa1_irq, .ack = mask_and_ack_eisa1_irq, + .mask = disable_eisa1_irq, + .mask_ack = mask_and_ack_eisa1_irq, + .unmask = enable_eisa1_irq, .end = end_eisa1_irq, }; static void enable_eisa2_irq(unsigned int irq) { - unsigned long flags; u8 mask; - local_irq_save(flags); - mask = inb(EISA_INT2_MASK); mask &= ~((u8) (1 << (irq - 8))); outb(mask, EISA_INT2_MASK); - - local_irq_restore(flags); } static unsigned int startup_eisa2_irq(unsigned int irq) @@ -192,8 +180,6 @@ static void disable_eisa2_irq(unsigned int irq) outb(mask, EISA_INT2_MASK); } -#define shutdown_eisa2_irq disable_eisa2_irq - static void mask_and_ack_eisa2_irq(unsigned int irq) { disable_eisa2_irq(irq); @@ -210,10 +196,10 @@ static void end_eisa2_irq(unsigned int irq) static struct irq_chip ip22_eisa2_irq_type = { .typename = "IP22 EISA", .startup = startup_eisa2_irq, - .shutdown = shutdown_eisa2_irq, - .enable = enable_eisa2_irq, - .disable = disable_eisa2_irq, .ack = mask_and_ack_eisa2_irq, + .mask = disable_eisa2_irq, + .mask_ack = mask_and_ack_eisa2_irq, + .unmask = enable_eisa2_irq, .end = end_eisa2_irq, }; @@ -275,13 +261,10 @@ int __init ip22_eisa_init(void) outb(0, EISA_DMA2_WRITE_SINGLE); for (i = SGINT_EISA; i < (SGINT_EISA + EISA_MAX_IRQ); i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 1; if (i < (SGINT_EISA + 8)) - irq_desc[i].chip = &ip22_eisa1_irq_type; + set_irq_chip(i, &ip22_eisa1_irq_type); else - irq_desc[i].chip = &ip22_eisa2_irq_type; + set_irq_chip(i, &ip22_eisa2_irq_type); } /* Cannot use request_irq because of kmalloc not being ready at such diff --git a/arch/mips/sgi-ip22/ip22-int.c b/arch/mips/sgi-ip22/ip22-int.c index af518898eaa..8e2074b4ce4 100644 --- a/arch/mips/sgi-ip22/ip22-int.c +++ b/arch/mips/sgi-ip22/ip22-int.c @@ -40,34 +40,17 @@ extern int ip22_eisa_init(void); static void enable_local0_irq(unsigned int irq) { - unsigned long flags; - - local_irq_save(flags); /* don't allow mappable interrupt to be enabled from setup_irq, * we have our own way to do so */ if (irq != SGI_MAP_0_IRQ) sgint->imask0 |= (1 << (irq - SGINT_LOCAL0)); - local_irq_restore(flags); -} - -static unsigned int startup_local0_irq(unsigned int irq) -{ - enable_local0_irq(irq); - return 0; /* Never anything pending */ } static void disable_local0_irq(unsigned int irq) { - unsigned long flags; - - local_irq_save(flags); sgint->imask0 &= ~(1 << (irq - SGINT_LOCAL0)); - local_irq_restore(flags); } -#define shutdown_local0_irq disable_local0_irq -#define mask_and_ack_local0_irq disable_local0_irq - static void end_local0_irq (unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS))) @@ -76,44 +59,26 @@ static void end_local0_irq (unsigned int irq) static struct irq_chip ip22_local0_irq_type = { .typename = "IP22 local 0", - .startup = startup_local0_irq, - .shutdown = shutdown_local0_irq, - .enable = enable_local0_irq, - .disable = disable_local0_irq, - .ack = mask_and_ack_local0_irq, + .ack = disable_local0_irq, + .mask = disable_local0_irq, + .mask_ack = disable_local0_irq, + .unmask = enable_local0_irq, .end = end_local0_irq, }; static void enable_local1_irq(unsigned int irq) { - unsigned long flags; - - local_irq_save(flags); /* don't allow mappable interrupt to be enabled from setup_irq, * we have our own way to do so */ if (irq != SGI_MAP_1_IRQ) sgint->imask1 |= (1 << (irq - SGINT_LOCAL1)); - local_irq_restore(flags); -} - -static unsigned int startup_local1_irq(unsigned int irq) -{ - enable_local1_irq(irq); - return 0; /* Never anything pending */ } void disable_local1_irq(unsigned int irq) { - unsigned long flags; - - local_irq_save(flags); sgint->imask1 &= ~(1 << (irq - SGINT_LOCAL1)); - local_irq_restore(flags); } -#define shutdown_local1_irq disable_local1_irq -#define mask_and_ack_local1_irq disable_local1_irq - static void end_local1_irq (unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS))) @@ -122,44 +87,26 @@ static void end_local1_irq (unsigned int irq) static struct irq_chip ip22_local1_irq_type = { .typename = "IP22 local 1", - .startup = startup_local1_irq, - .shutdown = shutdown_local1_irq, - .enable = enable_local1_irq, - .disable = disable_local1_irq, - .ack = mask_and_ack_local1_irq, + .ack = disable_local1_irq, + .mask = disable_local1_irq, + .mask_ack = disable_local1_irq, + .unmask = enable_local1_irq, .end = end_local1_irq, }; static void enable_local2_irq(unsigned int irq) { - unsigned long flags; - - local_irq_save(flags); sgint->imask0 |= (1 << (SGI_MAP_0_IRQ - SGINT_LOCAL0)); sgint->cmeimask0 |= (1 << (irq - SGINT_LOCAL2)); - local_irq_restore(flags); -} - -static unsigned int startup_local2_irq(unsigned int irq) -{ - enable_local2_irq(irq); - return 0; /* Never anything pending */ } void disable_local2_irq(unsigned int irq) { - unsigned long flags; - - local_irq_save(flags); sgint->cmeimask0 &= ~(1 << (irq - SGINT_LOCAL2)); if (!sgint->cmeimask0) sgint->imask0 &= ~(1 << (SGI_MAP_0_IRQ - SGINT_LOCAL0)); - local_irq_restore(flags); } -#define shutdown_local2_irq disable_local2_irq -#define mask_and_ack_local2_irq disable_local2_irq - static void end_local2_irq (unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS))) @@ -168,44 +115,26 @@ static void end_local2_irq (unsigned int irq) static struct irq_chip ip22_local2_irq_type = { .typename = "IP22 local 2", - .startup = startup_local2_irq, - .shutdown = shutdown_local2_irq, - .enable = enable_local2_irq, - .disable = disable_local2_irq, - .ack = mask_and_ack_local2_irq, + .ack = disable_local2_irq, + .mask = disable_local2_irq, + .mask_ack = disable_local2_irq, + .unmask = enable_local2_irq, .end = end_local2_irq, }; static void enable_local3_irq(unsigned int irq) { - unsigned long flags; - - local_irq_save(flags); sgint->imask1 |= (1 << (SGI_MAP_1_IRQ - SGINT_LOCAL1)); sgint->cmeimask1 |= (1 << (irq - SGINT_LOCAL3)); - local_irq_restore(flags); -} - -static unsigned int startup_local3_irq(unsigned int irq) -{ - enable_local3_irq(irq); - return 0; /* Never anything pending */ } void disable_local3_irq(unsigned int irq) { - unsigned long flags; - - local_irq_save(flags); sgint->cmeimask1 &= ~(1 << (irq - SGINT_LOCAL3)); if (!sgint->cmeimask1) sgint->imask1 &= ~(1 << (SGI_MAP_1_IRQ - SGINT_LOCAL1)); - local_irq_restore(flags); } -#define shutdown_local3_irq disable_local3_irq -#define mask_and_ack_local3_irq disable_local3_irq - static void end_local3_irq (unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS))) @@ -214,11 +143,10 @@ static void end_local3_irq (unsigned int irq) static struct irq_chip ip22_local3_irq_type = { .typename = "IP22 local 3", - .startup = startup_local3_irq, - .shutdown = shutdown_local3_irq, - .enable = enable_local3_irq, - .disable = disable_local3_irq, - .ack = mask_and_ack_local3_irq, + .ack = disable_local3_irq, + .mask = disable_local3_irq, + .mask_ack = disable_local3_irq, + .unmask = enable_local3_irq, .end = end_local3_irq, }; @@ -430,10 +358,7 @@ void __init arch_init_irq(void) else handler = &ip22_local3_irq_type; - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 1; - irq_desc[i].chip = handler; + set_irq_chip(i, handler); } /* vector handler. this register the IRQ as non-sharable */ diff --git a/arch/mips/sgi-ip27/ip27-irq.c b/arch/mips/sgi-ip27/ip27-irq.c index 270ecd3e6b4..824320281a3 100644 --- a/arch/mips/sgi-ip27/ip27-irq.c +++ b/arch/mips/sgi-ip27/ip27-irq.c @@ -332,11 +332,6 @@ static inline void disable_bridge_irq(unsigned int irq) intr_disconnect_level(cpu, swlevel); } -static void mask_and_ack_bridge_irq(unsigned int irq) -{ - disable_bridge_irq(irq); -} - static void end_bridge_irq(unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS)) && @@ -348,18 +343,16 @@ static struct irq_chip bridge_irq_type = { .typename = "bridge", .startup = startup_bridge_irq, .shutdown = shutdown_bridge_irq, - .enable = enable_bridge_irq, - .disable = disable_bridge_irq, - .ack = mask_and_ack_bridge_irq, + .ack = disable_bridge_irq, + .mask = disable_bridge_irq, + .mask_ack = disable_bridge_irq, + .unmask = enable_bridge_irq, .end = end_bridge_irq, }; void __devinit register_bridge_irq(unsigned int irq) { - irq_desc[irq].status = IRQ_DISABLED; - irq_desc[irq].action = 0; - irq_desc[irq].depth = 1; - irq_desc[irq].chip = &bridge_irq_type; + set_irq_chip(irq, &bridge_irq_type); } int __devinit request_bridge_irq(struct bridge_controller *bc) diff --git a/arch/mips/sgi-ip27/ip27-timer.c b/arch/mips/sgi-ip27/ip27-timer.c index 5e82a268e3c..86ba7fc10c3 100644 --- a/arch/mips/sgi-ip27/ip27-timer.c +++ b/arch/mips/sgi-ip27/ip27-timer.c @@ -172,15 +172,6 @@ static __init unsigned long get_m48t35_time(void) return mktime(year, month, date, hour, min, sec); } -static unsigned int startup_rt_irq(unsigned int irq) -{ - return 0; -} - -static void shutdown_rt_irq(unsigned int irq) -{ -} - static void enable_rt_irq(unsigned int irq) { } @@ -189,21 +180,16 @@ static void disable_rt_irq(unsigned int irq) { } -static void mask_and_ack_rt(unsigned int irq) -{ -} - static void end_rt_irq(unsigned int irq) { } static struct irq_chip rt_irq_type = { .typename = "SN HUB RT timer", - .startup = startup_rt_irq, - .shutdown = shutdown_rt_irq, - .enable = enable_rt_irq, - .disable = disable_rt_irq, - .ack = mask_and_ack_rt, + .ack = disable_rt_irq, + .mask = disable_rt_irq, + .mask_ack = disable_rt_irq, + .unmask = enable_rt_irq, .end = end_rt_irq, }; @@ -221,10 +207,7 @@ void __init plat_timer_setup(struct irqaction *irq) if (irqno < 0) panic("Can't allocate interrupt number for timer interrupt"); - irq_desc[irqno].status = IRQ_DISABLED; - irq_desc[irqno].action = NULL; - irq_desc[irqno].depth = 1; - irq_desc[irqno].chip = &rt_irq_type; + set_irq_chip(irqno, &rt_irq_type); /* over-write the handler, we use our own way */ irq->handler = no_action; diff --git a/arch/mips/sgi-ip32/ip32-irq.c b/arch/mips/sgi-ip32/ip32-irq.c index c9acadd0846..ae063864c02 100644 --- a/arch/mips/sgi-ip32/ip32-irq.c +++ b/arch/mips/sgi-ip32/ip32-irq.c @@ -113,12 +113,6 @@ static void inline flush_mace_bus(void) * is quite different anyway. */ -/* - * IRQ spinlock - Ralf says not to disable CPU interrupts, - * and I think he knows better. - */ -static DEFINE_SPINLOCK(ip32_irq_lock); - /* Some initial interrupts to set up */ extern irqreturn_t crime_memerr_intr(int irq, void *dev_id); extern irqreturn_t crime_cpuerr_intr(int irq, void *dev_id); @@ -138,12 +132,6 @@ static void enable_cpu_irq(unsigned int irq) set_c0_status(STATUSF_IP7); } -static unsigned int startup_cpu_irq(unsigned int irq) -{ - enable_cpu_irq(irq); - return 0; -} - static void disable_cpu_irq(unsigned int irq) { clear_c0_status(STATUSF_IP7); @@ -155,16 +143,12 @@ static void end_cpu_irq(unsigned int irq) enable_cpu_irq (irq); } -#define shutdown_cpu_irq disable_cpu_irq -#define mask_and_ack_cpu_irq disable_cpu_irq - static struct irq_chip ip32_cpu_interrupt = { .typename = "IP32 CPU", - .startup = startup_cpu_irq, - .shutdown = shutdown_cpu_irq, - .enable = enable_cpu_irq, - .disable = disable_cpu_irq, - .ack = mask_and_ack_cpu_irq, + .ack = disable_cpu_irq, + .mask = disable_cpu_irq, + .mask_ack = disable_cpu_irq, + .unmask = enable_cpu_irq, .end = end_cpu_irq, }; @@ -177,45 +161,27 @@ static uint64_t crime_mask; static void enable_crime_irq(unsigned int irq) { - unsigned long flags; - - spin_lock_irqsave(&ip32_irq_lock, flags); crime_mask |= 1 << (irq - 1); crime->imask = crime_mask; - spin_unlock_irqrestore(&ip32_irq_lock, flags); -} - -static unsigned int startup_crime_irq(unsigned int irq) -{ - enable_crime_irq(irq); - return 0; /* This is probably not right; we could have pending irqs */ } static void disable_crime_irq(unsigned int irq) { - unsigned long flags; - - spin_lock_irqsave(&ip32_irq_lock, flags); crime_mask &= ~(1 << (irq - 1)); crime->imask = crime_mask; flush_crime_bus(); - spin_unlock_irqrestore(&ip32_irq_lock, flags); } static void mask_and_ack_crime_irq(unsigned int irq) { - unsigned long flags; - /* Edge triggered interrupts must be cleared. */ if ((irq >= CRIME_GBE0_IRQ && irq <= CRIME_GBE3_IRQ) || (irq >= CRIME_RE_EMPTY_E_IRQ && irq <= CRIME_RE_IDLE_E_IRQ) || (irq >= CRIME_SOFT0_IRQ && irq <= CRIME_SOFT2_IRQ)) { uint64_t crime_int; - spin_lock_irqsave(&ip32_irq_lock, flags); crime_int = crime->hard_int; crime_int &= ~(1 << (irq - 1)); crime->hard_int = crime_int; - spin_unlock_irqrestore(&ip32_irq_lock, flags); } disable_crime_irq(irq); } @@ -226,15 +192,12 @@ static void end_crime_irq(unsigned int irq) enable_crime_irq(irq); } -#define shutdown_crime_irq disable_crime_irq - static struct irq_chip ip32_crime_interrupt = { .typename = "IP32 CRIME", - .startup = startup_crime_irq, - .shutdown = shutdown_crime_irq, - .enable = enable_crime_irq, - .disable = disable_crime_irq, .ack = mask_and_ack_crime_irq, + .mask = disable_crime_irq, + .mask_ack = mask_and_ack_crime_irq, + .unmask = enable_crime_irq, .end = end_crime_irq, }; @@ -248,34 +211,20 @@ static unsigned long macepci_mask; static void enable_macepci_irq(unsigned int irq) { - unsigned long flags; - - spin_lock_irqsave(&ip32_irq_lock, flags); macepci_mask |= MACEPCI_CONTROL_INT(irq - 9); mace->pci.control = macepci_mask; crime_mask |= 1 << (irq - 1); crime->imask = crime_mask; - spin_unlock_irqrestore(&ip32_irq_lock, flags); -} - -static unsigned int startup_macepci_irq(unsigned int irq) -{ - enable_macepci_irq (irq); - return 0; } static void disable_macepci_irq(unsigned int irq) { - unsigned long flags; - - spin_lock_irqsave(&ip32_irq_lock, flags); crime_mask &= ~(1 << (irq - 1)); crime->imask = crime_mask; flush_crime_bus(); macepci_mask &= ~MACEPCI_CONTROL_INT(irq - 9); mace->pci.control = macepci_mask; flush_mace_bus(); - spin_unlock_irqrestore(&ip32_irq_lock, flags); } static void end_macepci_irq(unsigned int irq) @@ -284,16 +233,12 @@ static void end_macepci_irq(unsigned int irq) enable_macepci_irq(irq); } -#define shutdown_macepci_irq disable_macepci_irq -#define mask_and_ack_macepci_irq disable_macepci_irq - static struct irq_chip ip32_macepci_interrupt = { .typename = "IP32 MACE PCI", - .startup = startup_macepci_irq, - .shutdown = shutdown_macepci_irq, - .enable = enable_macepci_irq, - .disable = disable_macepci_irq, - .ack = mask_and_ack_macepci_irq, + .ack = disable_macepci_irq, + .mask = disable_macepci_irq, + .mask_ack = disable_macepci_irq, + .unmask = enable_macepci_irq, .end = end_macepci_irq, }; @@ -339,7 +284,6 @@ static unsigned long maceisa_mask; static void enable_maceisa_irq (unsigned int irq) { unsigned int crime_int = 0; - unsigned long flags; DBG ("maceisa enable: %u\n", irq); @@ -355,26 +299,16 @@ static void enable_maceisa_irq (unsigned int irq) break; } DBG ("crime_int %08x enabled\n", crime_int); - spin_lock_irqsave(&ip32_irq_lock, flags); crime_mask |= crime_int; crime->imask = crime_mask; maceisa_mask |= 1 << (irq - 33); mace->perif.ctrl.imask = maceisa_mask; - spin_unlock_irqrestore(&ip32_irq_lock, flags); -} - -static unsigned int startup_maceisa_irq(unsigned int irq) -{ - enable_maceisa_irq(irq); - return 0; } static void disable_maceisa_irq(unsigned int irq) { unsigned int crime_int = 0; - unsigned long flags; - spin_lock_irqsave(&ip32_irq_lock, flags); maceisa_mask &= ~(1 << (irq - 33)); if(!(maceisa_mask & MACEISA_AUDIO_INT)) crime_int |= MACE_AUDIO_INT; @@ -387,23 +321,20 @@ static void disable_maceisa_irq(unsigned int irq) flush_crime_bus(); mace->perif.ctrl.imask = maceisa_mask; flush_mace_bus(); - spin_unlock_irqrestore(&ip32_irq_lock, flags); } static void mask_and_ack_maceisa_irq(unsigned int irq) { - unsigned long mace_int, flags; + unsigned long mace_int; switch (irq) { case MACEISA_PARALLEL_IRQ: case MACEISA_SERIAL1_TDMAPR_IRQ: case MACEISA_SERIAL2_TDMAPR_IRQ: /* edge triggered */ - spin_lock_irqsave(&ip32_irq_lock, flags); mace_int = mace->perif.ctrl.istat; mace_int &= ~(1 << (irq - 33)); mace->perif.ctrl.istat = mace_int; - spin_unlock_irqrestore(&ip32_irq_lock, flags); break; } disable_maceisa_irq(irq); @@ -415,15 +346,12 @@ static void end_maceisa_irq(unsigned irq) enable_maceisa_irq(irq); } -#define shutdown_maceisa_irq disable_maceisa_irq - static struct irq_chip ip32_maceisa_interrupt = { .typename = "IP32 MACE ISA", - .startup = startup_maceisa_irq, - .shutdown = shutdown_maceisa_irq, - .enable = enable_maceisa_irq, - .disable = disable_maceisa_irq, .ack = mask_and_ack_maceisa_irq, + .mask = disable_maceisa_irq, + .mask_ack = mask_and_ack_maceisa_irq, + .unmask = enable_maceisa_irq, .end = end_maceisa_irq, }; @@ -433,29 +361,15 @@ static struct irq_chip ip32_maceisa_interrupt = { static void enable_mace_irq(unsigned int irq) { - unsigned long flags; - - spin_lock_irqsave(&ip32_irq_lock, flags); crime_mask |= 1 << (irq - 1); crime->imask = crime_mask; - spin_unlock_irqrestore(&ip32_irq_lock, flags); -} - -static unsigned int startup_mace_irq(unsigned int irq) -{ - enable_mace_irq(irq); - return 0; } static void disable_mace_irq(unsigned int irq) { - unsigned long flags; - - spin_lock_irqsave(&ip32_irq_lock, flags); crime_mask &= ~(1 << (irq - 1)); crime->imask = crime_mask; flush_crime_bus(); - spin_unlock_irqrestore(&ip32_irq_lock, flags); } static void end_mace_irq(unsigned int irq) @@ -464,16 +378,12 @@ static void end_mace_irq(unsigned int irq) enable_mace_irq(irq); } -#define shutdown_mace_irq disable_mace_irq -#define mask_and_ack_mace_irq disable_mace_irq - static struct irq_chip ip32_mace_interrupt = { .typename = "IP32 MACE", - .startup = startup_mace_irq, - .shutdown = shutdown_mace_irq, - .enable = enable_mace_irq, - .disable = disable_mace_irq, - .ack = mask_and_ack_mace_irq, + .ack = disable_mace_irq, + .mask = disable_mace_irq, + .mask_ack = disable_mace_irq, + .unmask = enable_mace_irq, .end = end_mace_irq, }; @@ -586,10 +496,7 @@ void __init arch_init_irq(void) else controller = &ip32_maceisa_interrupt; - irq_desc[irq].status = IRQ_DISABLED; - irq_desc[irq].action = 0; - irq_desc[irq].depth = 0; - irq_desc[irq].chip = controller; + set_irq_chip(irq, controller); } setup_irq(CRIME_MEMERR_IRQ, &memerr_irq); setup_irq(CRIME_CPUERR_IRQ, &cpuerr_irq); diff --git a/arch/mips/sibyte/bcm1480/irq.c b/arch/mips/sibyte/bcm1480/irq.c index 8b1f4148492..2e8f6b2e242 100644 --- a/arch/mips/sibyte/bcm1480/irq.c +++ b/arch/mips/sibyte/bcm1480/irq.c @@ -45,11 +45,9 @@ */ -#define shutdown_bcm1480_irq disable_bcm1480_irq static void end_bcm1480_irq(unsigned int irq); static void enable_bcm1480_irq(unsigned int irq); static void disable_bcm1480_irq(unsigned int irq); -static unsigned int startup_bcm1480_irq(unsigned int irq); static void ack_bcm1480_irq(unsigned int irq); #ifdef CONFIG_SMP static void bcm1480_set_affinity(unsigned int irq, cpumask_t mask); @@ -85,11 +83,10 @@ extern char sb1250_duart_present[]; static struct irq_chip bcm1480_irq_type = { .typename = "BCM1480-IMR", - .startup = startup_bcm1480_irq, - .shutdown = shutdown_bcm1480_irq, - .enable = enable_bcm1480_irq, - .disable = disable_bcm1480_irq, .ack = ack_bcm1480_irq, + .mask = disable_bcm1480_irq, + .mask_ack = ack_bcm1480_irq, + .unmask = enable_bcm1480_irq, .end = end_bcm1480_irq, #ifdef CONFIG_SMP .set_affinity = bcm1480_set_affinity @@ -188,14 +185,6 @@ static void bcm1480_set_affinity(unsigned int irq, cpumask_t mask) /*****************************************************************************/ -static unsigned int startup_bcm1480_irq(unsigned int irq) -{ - bcm1480_unmask_irq(bcm1480_irq_owner[irq], irq); - - return 0; /* never anything pending */ -} - - static void disable_bcm1480_irq(unsigned int irq) { bcm1480_mask_irq(bcm1480_irq_owner[irq], irq); @@ -270,16 +259,9 @@ void __init init_bcm1480_irqs(void) { int i; - for (i = 0; i < NR_IRQS; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 1; - if (i < BCM1480_NR_IRQS) { - irq_desc[i].chip = &bcm1480_irq_type; - bcm1480_irq_owner[i] = 0; - } else { - irq_desc[i].chip = &no_irq_chip; - } + for (i = 0; i < BCM1480_NR_IRQS; i++) { + set_irq_chip(i, &bcm1480_irq_type); + bcm1480_irq_owner[i] = 0; } } diff --git a/arch/mips/sibyte/sb1250/irq.c b/arch/mips/sibyte/sb1250/irq.c index d5d26770daf..82ce7533053 100644 --- a/arch/mips/sibyte/sb1250/irq.c +++ b/arch/mips/sibyte/sb1250/irq.c @@ -44,11 +44,9 @@ */ -#define shutdown_sb1250_irq disable_sb1250_irq static void end_sb1250_irq(unsigned int irq); static void enable_sb1250_irq(unsigned int irq); static void disable_sb1250_irq(unsigned int irq); -static unsigned int startup_sb1250_irq(unsigned int irq); static void ack_sb1250_irq(unsigned int irq); #ifdef CONFIG_SMP static void sb1250_set_affinity(unsigned int irq, cpumask_t mask); @@ -70,11 +68,10 @@ extern char sb1250_duart_present[]; static struct irq_chip sb1250_irq_type = { .typename = "SB1250-IMR", - .startup = startup_sb1250_irq, - .shutdown = shutdown_sb1250_irq, - .enable = enable_sb1250_irq, - .disable = disable_sb1250_irq, .ack = ack_sb1250_irq, + .mask = disable_sb1250_irq, + .mask_ack = ack_sb1250_irq, + .unmask = enable_sb1250_irq, .end = end_sb1250_irq, #ifdef CONFIG_SMP .set_affinity = sb1250_set_affinity @@ -163,14 +160,6 @@ static void sb1250_set_affinity(unsigned int irq, cpumask_t mask) /*****************************************************************************/ -static unsigned int startup_sb1250_irq(unsigned int irq) -{ - sb1250_unmask_irq(sb1250_irq_owner[irq], irq); - - return 0; /* never anything pending */ -} - - static void disable_sb1250_irq(unsigned int irq) { sb1250_mask_irq(sb1250_irq_owner[irq], irq); @@ -239,16 +228,9 @@ void __init init_sb1250_irqs(void) { int i; - for (i = 0; i < NR_IRQS; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 1; - if (i < SB1250_NR_IRQS) { - irq_desc[i].chip = &sb1250_irq_type; - sb1250_irq_owner[i] = 0; - } else { - irq_desc[i].chip = &no_irq_chip; - } + for (i = 0; i < SB1250_NR_IRQS; i++) { + set_irq_chip(i, &sb1250_irq_type); + sb1250_irq_owner[i] = 0; } } diff --git a/arch/mips/sni/irq.c b/arch/mips/sni/irq.c index 48fb74a7aae..8511bcc6d99 100644 --- a/arch/mips/sni/irq.c +++ b/arch/mips/sni/irq.c @@ -11,44 +11,25 @@ #include #include #include -#include #include #include #include -DEFINE_SPINLOCK(pciasic_lock); - static void enable_pciasic_irq(unsigned int irq) { unsigned int mask = 1 << (irq - PCIMT_IRQ_INT2); - unsigned long flags; - spin_lock_irqsave(&pciasic_lock, flags); *(volatile u8 *) PCIMT_IRQSEL |= mask; - spin_unlock_irqrestore(&pciasic_lock, flags); -} - -static unsigned int startup_pciasic_irq(unsigned int irq) -{ - enable_pciasic_irq(irq); - return 0; /* never anything pending */ } -#define shutdown_pciasic_irq disable_pciasic_irq - void disable_pciasic_irq(unsigned int irq) { unsigned int mask = ~(1 << (irq - PCIMT_IRQ_INT2)); - unsigned long flags; - spin_lock_irqsave(&pciasic_lock, flags); *(volatile u8 *) PCIMT_IRQSEL &= mask; - spin_unlock_irqrestore(&pciasic_lock, flags); } -#define mask_and_ack_pciasic_irq disable_pciasic_irq - static void end_pciasic_irq(unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS))) @@ -57,11 +38,10 @@ static void end_pciasic_irq(unsigned int irq) static struct irq_chip pciasic_irq_type = { .typename = "ASIC-PCI", - .startup = startup_pciasic_irq, - .shutdown = shutdown_pciasic_irq, - .enable = enable_pciasic_irq, - .disable = disable_pciasic_irq, - .ack = mask_and_ack_pciasic_irq, + .ack = disable_pciasic_irq, + .mask = disable_pciasic_irq, + .mask_ack = disable_pciasic_irq, + .unmask = enable_pciasic_irq, .end = end_pciasic_irq, }; @@ -178,12 +158,8 @@ asmlinkage void plat_irq_dispatch(void) void __init init_pciasic(void) { - unsigned long flags; - - spin_lock_irqsave(&pciasic_lock, flags); * (volatile u8 *) PCIMT_IRQSEL = IT_EISA | IT_INTA | IT_INTB | IT_INTC | IT_INTD; - spin_unlock_irqrestore(&pciasic_lock, flags); } /* @@ -199,12 +175,8 @@ void __init arch_init_irq(void) init_pciasic(); /* Actually we've got more interrupts to handle ... */ - for (i = PCIMT_IRQ_INT2; i <= PCIMT_IRQ_ETHERNET; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 1; - irq_desc[i].chip = &pciasic_irq_type; - } + for (i = PCIMT_IRQ_INT2; i <= PCIMT_IRQ_ETHERNET; i++) + set_irq_chip(i, &pciasic_irq_type); change_c0_status(ST0_IM, IE_IRQ1|IE_IRQ2|IE_IRQ3|IE_IRQ4); } diff --git a/arch/mips/tx4927/common/tx4927_irq.c b/arch/mips/tx4927/common/tx4927_irq.c index 8266a88a3f8..2c57ced5c68 100644 --- a/arch/mips/tx4927/common/tx4927_irq.c +++ b/arch/mips/tx4927/common/tx4927_irq.c @@ -64,19 +64,13 @@ #define TX4927_IRQ_NEST4 ( 1 << 9 ) #define TX4927_IRQ_CP0_INIT ( 1 << 10 ) -#define TX4927_IRQ_CP0_STARTUP ( 1 << 11 ) -#define TX4927_IRQ_CP0_SHUTDOWN ( 1 << 12 ) #define TX4927_IRQ_CP0_ENABLE ( 1 << 13 ) #define TX4927_IRQ_CP0_DISABLE ( 1 << 14 ) -#define TX4927_IRQ_CP0_MASK ( 1 << 15 ) #define TX4927_IRQ_CP0_ENDIRQ ( 1 << 16 ) #define TX4927_IRQ_PIC_INIT ( 1 << 20 ) -#define TX4927_IRQ_PIC_STARTUP ( 1 << 21 ) -#define TX4927_IRQ_PIC_SHUTDOWN ( 1 << 22 ) #define TX4927_IRQ_PIC_ENABLE ( 1 << 23 ) #define TX4927_IRQ_PIC_DISABLE ( 1 << 24 ) -#define TX4927_IRQ_PIC_MASK ( 1 << 25 ) #define TX4927_IRQ_PIC_ENDIRQ ( 1 << 26 ) #define TX4927_IRQ_ALL 0xffffffff @@ -87,18 +81,12 @@ static const u32 tx4927_irq_debug_flag = (TX4927_IRQ_NONE | TX4927_IRQ_INFO | TX4927_IRQ_WARN | TX4927_IRQ_EROR // | TX4927_IRQ_CP0_INIT -// | TX4927_IRQ_CP0_STARTUP -// | TX4927_IRQ_CP0_SHUTDOWN // | TX4927_IRQ_CP0_ENABLE // | TX4927_IRQ_CP0_DISABLE -// | TX4927_IRQ_CP0_MASK // | TX4927_IRQ_CP0_ENDIRQ // | TX4927_IRQ_PIC_INIT -// | TX4927_IRQ_PIC_STARTUP -// | TX4927_IRQ_PIC_SHUTDOWN // | TX4927_IRQ_PIC_ENABLE // | TX4927_IRQ_PIC_DISABLE -// | TX4927_IRQ_PIC_MASK // | TX4927_IRQ_PIC_ENDIRQ // | TX4927_IRQ_INIT // | TX4927_IRQ_NEST1 @@ -124,49 +112,36 @@ static const u32 tx4927_irq_debug_flag = (TX4927_IRQ_NONE * Forwad definitions for all pic's */ -static unsigned int tx4927_irq_cp0_startup(unsigned int irq); -static void tx4927_irq_cp0_shutdown(unsigned int irq); static void tx4927_irq_cp0_enable(unsigned int irq); static void tx4927_irq_cp0_disable(unsigned int irq); -static void tx4927_irq_cp0_mask_and_ack(unsigned int irq); static void tx4927_irq_cp0_end(unsigned int irq); -static unsigned int tx4927_irq_pic_startup(unsigned int irq); -static void tx4927_irq_pic_shutdown(unsigned int irq); static void tx4927_irq_pic_enable(unsigned int irq); static void tx4927_irq_pic_disable(unsigned int irq); -static void tx4927_irq_pic_mask_and_ack(unsigned int irq); static void tx4927_irq_pic_end(unsigned int irq); /* * Kernel structs for all pic's */ -static DEFINE_SPINLOCK(tx4927_cp0_lock); -static DEFINE_SPINLOCK(tx4927_pic_lock); - #define TX4927_CP0_NAME "TX4927-CP0" static struct irq_chip tx4927_irq_cp0_type = { .typename = TX4927_CP0_NAME, - .startup = tx4927_irq_cp0_startup, - .shutdown = tx4927_irq_cp0_shutdown, - .enable = tx4927_irq_cp0_enable, - .disable = tx4927_irq_cp0_disable, - .ack = tx4927_irq_cp0_mask_and_ack, + .ack = tx4927_irq_cp0_disable, + .mask = tx4927_irq_cp0_disable, + .mask_ack = tx4927_irq_cp0_disable, + .unmask = tx4927_irq_cp0_enable, .end = tx4927_irq_cp0_end, - .set_affinity = NULL }; #define TX4927_PIC_NAME "TX4927-PIC" static struct irq_chip tx4927_irq_pic_type = { .typename = TX4927_PIC_NAME, - .startup = tx4927_irq_pic_startup, - .shutdown = tx4927_irq_pic_shutdown, - .enable = tx4927_irq_pic_enable, - .disable = tx4927_irq_pic_disable, - .ack = tx4927_irq_pic_mask_and_ack, + .ack = tx4927_irq_pic_disable, + .mask = tx4927_irq_pic_disable, + .mask_ack = tx4927_irq_pic_disable, + .unmask = tx4927_irq_pic_enable, .end = tx4927_irq_pic_end, - .set_affinity = NULL }; #define TX4927_PIC_ACTION(s) { no_action, 0, CPU_MASK_NONE, s, NULL, NULL } @@ -211,8 +186,6 @@ tx4927_irq_cp0_modify(unsigned cp0_reg, unsigned clr_bits, unsigned set_bits) break; } } - - return; } static void __init tx4927_irq_cp0_init(void) @@ -222,71 +195,22 @@ static void __init tx4927_irq_cp0_init(void) TX4927_IRQ_DPRINTK(TX4927_IRQ_CP0_INIT, "beg=%d end=%d\n", TX4927_IRQ_CP0_BEG, TX4927_IRQ_CP0_END); - for (i = TX4927_IRQ_CP0_BEG; i <= TX4927_IRQ_CP0_END; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 1; - irq_desc[i].chip = &tx4927_irq_cp0_type; - } - - return; -} - -static unsigned int tx4927_irq_cp0_startup(unsigned int irq) -{ - TX4927_IRQ_DPRINTK(TX4927_IRQ_CP0_STARTUP, "irq=%d \n", irq); - - tx4927_irq_cp0_enable(irq); - - return (0); -} - -static void tx4927_irq_cp0_shutdown(unsigned int irq) -{ - TX4927_IRQ_DPRINTK(TX4927_IRQ_CP0_SHUTDOWN, "irq=%d \n", irq); - - tx4927_irq_cp0_disable(irq); - - return; + for (i = TX4927_IRQ_CP0_BEG; i <= TX4927_IRQ_CP0_END; i++) + set_irq_chip(i, &tx4927_irq_cp0_type); } static void tx4927_irq_cp0_enable(unsigned int irq) { - unsigned long flags; - TX4927_IRQ_DPRINTK(TX4927_IRQ_CP0_ENABLE, "irq=%d \n", irq); - spin_lock_irqsave(&tx4927_cp0_lock, flags); - tx4927_irq_cp0_modify(CCP0_STATUS, 0, tx4927_irq_cp0_mask(irq)); - - spin_unlock_irqrestore(&tx4927_cp0_lock, flags); - - return; } static void tx4927_irq_cp0_disable(unsigned int irq) { - unsigned long flags; - TX4927_IRQ_DPRINTK(TX4927_IRQ_CP0_DISABLE, "irq=%d \n", irq); - spin_lock_irqsave(&tx4927_cp0_lock, flags); - tx4927_irq_cp0_modify(CCP0_STATUS, tx4927_irq_cp0_mask(irq), 0); - - spin_unlock_irqrestore(&tx4927_cp0_lock, flags); - - return; -} - -static void tx4927_irq_cp0_mask_and_ack(unsigned int irq) -{ - TX4927_IRQ_DPRINTK(TX4927_IRQ_CP0_MASK, "irq=%d \n", irq); - - tx4927_irq_cp0_disable(irq); - - return; } static void tx4927_irq_cp0_end(unsigned int irq) @@ -296,8 +220,6 @@ static void tx4927_irq_cp0_end(unsigned int irq) if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS))) { tx4927_irq_cp0_enable(irq); } - - return; } /* @@ -418,94 +340,38 @@ static void tx4927_irq_pic_modify(unsigned pic_reg, unsigned clr_bits, val &= (~clr_bits); val |= (set_bits); TX4927_WR(pic_reg, val); - - return; } static void __init tx4927_irq_pic_init(void) { - unsigned long flags; int i; TX4927_IRQ_DPRINTK(TX4927_IRQ_PIC_INIT, "beg=%d end=%d\n", TX4927_IRQ_PIC_BEG, TX4927_IRQ_PIC_END); - for (i = TX4927_IRQ_PIC_BEG; i <= TX4927_IRQ_PIC_END; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 2; - irq_desc[i].chip = &tx4927_irq_pic_type; - } + for (i = TX4927_IRQ_PIC_BEG; i <= TX4927_IRQ_PIC_END; i++) + set_irq_chip(i, &tx4927_irq_pic_type); setup_irq(TX4927_IRQ_NEST_PIC_ON_CP0, &tx4927_irq_pic_action); - spin_lock_irqsave(&tx4927_pic_lock, flags); - TX4927_WR(0xff1ff640, 0x6); /* irq level mask -- only accept hightest */ TX4927_WR(0xff1ff600, TX4927_RD(0xff1ff600) | 0x1); /* irq enable */ - - spin_unlock_irqrestore(&tx4927_pic_lock, flags); - - return; -} - -static unsigned int tx4927_irq_pic_startup(unsigned int irq) -{ - TX4927_IRQ_DPRINTK(TX4927_IRQ_PIC_STARTUP, "irq=%d\n", irq); - - tx4927_irq_pic_enable(irq); - - return (0); -} - -static void tx4927_irq_pic_shutdown(unsigned int irq) -{ - TX4927_IRQ_DPRINTK(TX4927_IRQ_PIC_SHUTDOWN, "irq=%d\n", irq); - - tx4927_irq_pic_disable(irq); - - return; } static void tx4927_irq_pic_enable(unsigned int irq) { - unsigned long flags; - TX4927_IRQ_DPRINTK(TX4927_IRQ_PIC_ENABLE, "irq=%d\n", irq); - spin_lock_irqsave(&tx4927_pic_lock, flags); - tx4927_irq_pic_modify(tx4927_irq_pic_addr(irq), 0, tx4927_irq_pic_mask(irq)); - - spin_unlock_irqrestore(&tx4927_pic_lock, flags); - - return; } static void tx4927_irq_pic_disable(unsigned int irq) { - unsigned long flags; - TX4927_IRQ_DPRINTK(TX4927_IRQ_PIC_DISABLE, "irq=%d\n", irq); - spin_lock_irqsave(&tx4927_pic_lock, flags); - tx4927_irq_pic_modify(tx4927_irq_pic_addr(irq), tx4927_irq_pic_mask(irq), 0); - - spin_unlock_irqrestore(&tx4927_pic_lock, flags); - - return; -} - -static void tx4927_irq_pic_mask_and_ack(unsigned int irq) -{ - TX4927_IRQ_DPRINTK(TX4927_IRQ_PIC_MASK, "irq=%d\n", irq); - - tx4927_irq_pic_disable(irq); - - return; } static void tx4927_irq_pic_end(unsigned int irq) @@ -515,8 +381,6 @@ static void tx4927_irq_pic_end(unsigned int irq) if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS))) { tx4927_irq_pic_enable(irq); } - - return; } /* @@ -533,8 +397,6 @@ void __init tx4927_irq_init(void) tx4927_irq_pic_init(); TX4927_IRQ_DPRINTK(TX4927_IRQ_INIT, "+\n"); - - return; } static int tx4927_irq_nested(void) diff --git a/arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_irq.c b/arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_irq.c index 0c3c3f66823..1fdace89ae6 100644 --- a/arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_irq.c +++ b/arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_irq.c @@ -151,16 +151,11 @@ JP7 is not bus master -- do NOT use -- only 4 pci bus master's allowed -- SouthB #define TOSHIBA_RBTX4927_IRQ_EROR ( 1 << 2 ) #define TOSHIBA_RBTX4927_IRQ_IOC_INIT ( 1 << 10 ) -#define TOSHIBA_RBTX4927_IRQ_IOC_STARTUP ( 1 << 11 ) -#define TOSHIBA_RBTX4927_IRQ_IOC_SHUTDOWN ( 1 << 12 ) #define TOSHIBA_RBTX4927_IRQ_IOC_ENABLE ( 1 << 13 ) #define TOSHIBA_RBTX4927_IRQ_IOC_DISABLE ( 1 << 14 ) -#define TOSHIBA_RBTX4927_IRQ_IOC_MASK ( 1 << 15 ) #define TOSHIBA_RBTX4927_IRQ_IOC_ENDIRQ ( 1 << 16 ) #define TOSHIBA_RBTX4927_IRQ_ISA_INIT ( 1 << 20 ) -#define TOSHIBA_RBTX4927_IRQ_ISA_STARTUP ( 1 << 21 ) -#define TOSHIBA_RBTX4927_IRQ_ISA_SHUTDOWN ( 1 << 22 ) #define TOSHIBA_RBTX4927_IRQ_ISA_ENABLE ( 1 << 23 ) #define TOSHIBA_RBTX4927_IRQ_ISA_DISABLE ( 1 << 24 ) #define TOSHIBA_RBTX4927_IRQ_ISA_MASK ( 1 << 25 ) @@ -175,15 +170,10 @@ static const u32 toshiba_rbtx4927_irq_debug_flag = (TOSHIBA_RBTX4927_IRQ_NONE | TOSHIBA_RBTX4927_IRQ_INFO | TOSHIBA_RBTX4927_IRQ_WARN | TOSHIBA_RBTX4927_IRQ_EROR // | TOSHIBA_RBTX4927_IRQ_IOC_INIT -// | TOSHIBA_RBTX4927_IRQ_IOC_STARTUP -// | TOSHIBA_RBTX4927_IRQ_IOC_SHUTDOWN // | TOSHIBA_RBTX4927_IRQ_IOC_ENABLE // | TOSHIBA_RBTX4927_IRQ_IOC_DISABLE -// | TOSHIBA_RBTX4927_IRQ_IOC_MASK // | TOSHIBA_RBTX4927_IRQ_IOC_ENDIRQ // | TOSHIBA_RBTX4927_IRQ_ISA_INIT -// | TOSHIBA_RBTX4927_IRQ_ISA_STARTUP -// | TOSHIBA_RBTX4927_IRQ_ISA_SHUTDOWN // | TOSHIBA_RBTX4927_IRQ_ISA_ENABLE // | TOSHIBA_RBTX4927_IRQ_ISA_DISABLE // | TOSHIBA_RBTX4927_IRQ_ISA_MASK @@ -231,35 +221,25 @@ extern void disable_8259A_irq(unsigned int irq); extern void mask_and_ack_8259A(unsigned int irq); #endif -static unsigned int toshiba_rbtx4927_irq_ioc_startup(unsigned int irq); -static void toshiba_rbtx4927_irq_ioc_shutdown(unsigned int irq); static void toshiba_rbtx4927_irq_ioc_enable(unsigned int irq); static void toshiba_rbtx4927_irq_ioc_disable(unsigned int irq); -static void toshiba_rbtx4927_irq_ioc_mask_and_ack(unsigned int irq); static void toshiba_rbtx4927_irq_ioc_end(unsigned int irq); #ifdef CONFIG_TOSHIBA_FPCIB0 -static unsigned int toshiba_rbtx4927_irq_isa_startup(unsigned int irq); -static void toshiba_rbtx4927_irq_isa_shutdown(unsigned int irq); static void toshiba_rbtx4927_irq_isa_enable(unsigned int irq); static void toshiba_rbtx4927_irq_isa_disable(unsigned int irq); static void toshiba_rbtx4927_irq_isa_mask_and_ack(unsigned int irq); static void toshiba_rbtx4927_irq_isa_end(unsigned int irq); #endif -static DEFINE_SPINLOCK(toshiba_rbtx4927_ioc_lock); - - #define TOSHIBA_RBTX4927_IOC_NAME "RBTX4927-IOC" static struct irq_chip toshiba_rbtx4927_irq_ioc_type = { .typename = TOSHIBA_RBTX4927_IOC_NAME, - .startup = toshiba_rbtx4927_irq_ioc_startup, - .shutdown = toshiba_rbtx4927_irq_ioc_shutdown, - .enable = toshiba_rbtx4927_irq_ioc_enable, - .disable = toshiba_rbtx4927_irq_ioc_disable, - .ack = toshiba_rbtx4927_irq_ioc_mask_and_ack, + .ack = toshiba_rbtx4927_irq_ioc_disable, + .mask = toshiba_rbtx4927_irq_ioc_disable, + .mask_ack = toshiba_rbtx4927_irq_ioc_disable, + .unmask = toshiba_rbtx4927_irq_ioc_enable, .end = toshiba_rbtx4927_irq_ioc_end, - .set_affinity = NULL }; #define TOSHIBA_RBTX4927_IOC_INTR_ENAB 0xbc002000 #define TOSHIBA_RBTX4927_IOC_INTR_STAT 0xbc002006 @@ -269,13 +249,11 @@ static struct irq_chip toshiba_rbtx4927_irq_ioc_type = { #define TOSHIBA_RBTX4927_ISA_NAME "RBTX4927-ISA" static struct irq_chip toshiba_rbtx4927_irq_isa_type = { .typename = TOSHIBA_RBTX4927_ISA_NAME, - .startup = toshiba_rbtx4927_irq_isa_startup, - .shutdown = toshiba_rbtx4927_irq_isa_shutdown, - .enable = toshiba_rbtx4927_irq_isa_enable, - .disable = toshiba_rbtx4927_irq_isa_disable, .ack = toshiba_rbtx4927_irq_isa_mask_and_ack, + .mask = toshiba_rbtx4927_irq_isa_disable, + .mask_ack = toshiba_rbtx4927_irq_isa_mask_and_ack, + .unmask = toshiba_rbtx4927_irq_isa_enable, .end = toshiba_rbtx4927_irq_isa_end, - .set_affinity = NULL }; #endif @@ -363,58 +341,15 @@ static void __init toshiba_rbtx4927_irq_ioc_init(void) TOSHIBA_RBTX4927_IRQ_IOC_END); for (i = TOSHIBA_RBTX4927_IRQ_IOC_BEG; - i <= TOSHIBA_RBTX4927_IRQ_IOC_END; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 3; - irq_desc[i].chip = &toshiba_rbtx4927_irq_ioc_type; - } + i <= TOSHIBA_RBTX4927_IRQ_IOC_END; i++) + set_irq_chip(i, &toshiba_rbtx4927_irq_ioc_type); setup_irq(TOSHIBA_RBTX4927_IRQ_NEST_IOC_ON_PIC, &toshiba_rbtx4927_irq_ioc_action); - - return; } -static unsigned int toshiba_rbtx4927_irq_ioc_startup(unsigned int irq) -{ - TOSHIBA_RBTX4927_IRQ_DPRINTK(TOSHIBA_RBTX4927_IRQ_IOC_STARTUP, - "irq=%d\n", irq); - - if (irq < TOSHIBA_RBTX4927_IRQ_IOC_BEG - || irq > TOSHIBA_RBTX4927_IRQ_IOC_END) { - TOSHIBA_RBTX4927_IRQ_DPRINTK(TOSHIBA_RBTX4927_IRQ_EROR, - "bad irq=%d\n", irq); - panic("\n"); - } - - toshiba_rbtx4927_irq_ioc_enable(irq); - - return (0); -} - - -static void toshiba_rbtx4927_irq_ioc_shutdown(unsigned int irq) -{ - TOSHIBA_RBTX4927_IRQ_DPRINTK(TOSHIBA_RBTX4927_IRQ_IOC_SHUTDOWN, - "irq=%d\n", irq); - - if (irq < TOSHIBA_RBTX4927_IRQ_IOC_BEG - || irq > TOSHIBA_RBTX4927_IRQ_IOC_END) { - TOSHIBA_RBTX4927_IRQ_DPRINTK(TOSHIBA_RBTX4927_IRQ_EROR, - "bad irq=%d\n", irq); - panic("\n"); - } - - toshiba_rbtx4927_irq_ioc_disable(irq); - - return; -} - - static void toshiba_rbtx4927_irq_ioc_enable(unsigned int irq) { - unsigned long flags; volatile unsigned char v; TOSHIBA_RBTX4927_IRQ_DPRINTK(TOSHIBA_RBTX4927_IRQ_IOC_ENABLE, @@ -427,21 +362,14 @@ static void toshiba_rbtx4927_irq_ioc_enable(unsigned int irq) panic("\n"); } - spin_lock_irqsave(&toshiba_rbtx4927_ioc_lock, flags); - v = TX4927_RD08(TOSHIBA_RBTX4927_IOC_INTR_ENAB); v |= (1 << (irq - TOSHIBA_RBTX4927_IRQ_IOC_BEG)); TOSHIBA_RBTX4927_WR08(TOSHIBA_RBTX4927_IOC_INTR_ENAB, v); - - spin_unlock_irqrestore(&toshiba_rbtx4927_ioc_lock, flags); - - return; } static void toshiba_rbtx4927_irq_ioc_disable(unsigned int irq) { - unsigned long flags; volatile unsigned char v; TOSHIBA_RBTX4927_IRQ_DPRINTK(TOSHIBA_RBTX4927_IRQ_IOC_DISABLE, @@ -454,36 +382,11 @@ static void toshiba_rbtx4927_irq_ioc_disable(unsigned int irq) panic("\n"); } - spin_lock_irqsave(&toshiba_rbtx4927_ioc_lock, flags); - v = TX4927_RD08(TOSHIBA_RBTX4927_IOC_INTR_ENAB); v &= ~(1 << (irq - TOSHIBA_RBTX4927_IRQ_IOC_BEG)); TOSHIBA_RBTX4927_WR08(TOSHIBA_RBTX4927_IOC_INTR_ENAB, v); - - spin_unlock_irqrestore(&toshiba_rbtx4927_ioc_lock, flags); - - return; } - -static void toshiba_rbtx4927_irq_ioc_mask_and_ack(unsigned int irq) -{ - TOSHIBA_RBTX4927_IRQ_DPRINTK(TOSHIBA_RBTX4927_IRQ_IOC_MASK, - "irq=%d\n", irq); - - if (irq < TOSHIBA_RBTX4927_IRQ_IOC_BEG - || irq > TOSHIBA_RBTX4927_IRQ_IOC_END) { - TOSHIBA_RBTX4927_IRQ_DPRINTK(TOSHIBA_RBTX4927_IRQ_EROR, - "bad irq=%d\n", irq); - panic("\n"); - } - - toshiba_rbtx4927_irq_ioc_disable(irq); - - return; -} - - static void toshiba_rbtx4927_irq_ioc_end(unsigned int irq) { TOSHIBA_RBTX4927_IRQ_DPRINTK(TOSHIBA_RBTX4927_IRQ_IOC_ENDIRQ, @@ -499,8 +402,6 @@ static void toshiba_rbtx4927_irq_ioc_end(unsigned int irq) if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS))) { toshiba_rbtx4927_irq_ioc_enable(irq); } - - return; } @@ -520,13 +421,8 @@ static void __init toshiba_rbtx4927_irq_isa_init(void) TOSHIBA_RBTX4927_IRQ_ISA_END); for (i = TOSHIBA_RBTX4927_IRQ_ISA_BEG; - i <= TOSHIBA_RBTX4927_IRQ_ISA_END; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = - ((i < TOSHIBA_RBTX4927_IRQ_ISA_MID) ? (4) : (5)); - irq_desc[i].chip = &toshiba_rbtx4927_irq_isa_type; - } + i <= TOSHIBA_RBTX4927_IRQ_ISA_END; i++) + set_irq_chip(i, &toshiba_rbtx4927_irq_isa_type); setup_irq(TOSHIBA_RBTX4927_IRQ_NEST_ISA_ON_IOC, &toshiba_rbtx4927_irq_isa_master); @@ -536,48 +432,6 @@ static void __init toshiba_rbtx4927_irq_isa_init(void) /* make sure we are looking at IRR (not ISR) */ outb(0x0A, 0x20); outb(0x0A, 0xA0); - - return; -} -#endif - - -#ifdef CONFIG_TOSHIBA_FPCIB0 -static unsigned int toshiba_rbtx4927_irq_isa_startup(unsigned int irq) -{ - TOSHIBA_RBTX4927_IRQ_DPRINTK(TOSHIBA_RBTX4927_IRQ_ISA_STARTUP, - "irq=%d\n", irq); - - if (irq < TOSHIBA_RBTX4927_IRQ_ISA_BEG - || irq > TOSHIBA_RBTX4927_IRQ_ISA_END) { - TOSHIBA_RBTX4927_IRQ_DPRINTK(TOSHIBA_RBTX4927_IRQ_EROR, - "bad irq=%d\n", irq); - panic("\n"); - } - - toshiba_rbtx4927_irq_isa_enable(irq); - - return (0); -} -#endif - - -#ifdef CONFIG_TOSHIBA_FPCIB0 -static void toshiba_rbtx4927_irq_isa_shutdown(unsigned int irq) -{ - TOSHIBA_RBTX4927_IRQ_DPRINTK(TOSHIBA_RBTX4927_IRQ_ISA_SHUTDOWN, - "irq=%d\n", irq); - - if (irq < TOSHIBA_RBTX4927_IRQ_ISA_BEG - || irq > TOSHIBA_RBTX4927_IRQ_ISA_END) { - TOSHIBA_RBTX4927_IRQ_DPRINTK(TOSHIBA_RBTX4927_IRQ_EROR, - "bad irq=%d\n", irq); - panic("\n"); - } - - toshiba_rbtx4927_irq_isa_disable(irq); - - return; } #endif @@ -596,8 +450,6 @@ static void toshiba_rbtx4927_irq_isa_enable(unsigned int irq) } enable_8259A_irq(irq); - - return; } #endif @@ -616,8 +468,6 @@ static void toshiba_rbtx4927_irq_isa_disable(unsigned int irq) } disable_8259A_irq(irq); - - return; } #endif @@ -636,8 +486,6 @@ static void toshiba_rbtx4927_irq_isa_mask_and_ack(unsigned int irq) } mask_and_ack_8259A(irq); - - return; } #endif @@ -658,8 +506,6 @@ static void toshiba_rbtx4927_irq_isa_end(unsigned int irq) if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS))) { toshiba_rbtx4927_irq_isa_enable(irq); } - - return; } #endif @@ -668,8 +514,6 @@ void __init arch_init_irq(void) { extern void tx4927_irq_init(void); - local_irq_disable(); - tx4927_irq_init(); toshiba_rbtx4927_irq_ioc_init(); #ifdef CONFIG_TOSHIBA_FPCIB0 @@ -681,8 +525,6 @@ void __init arch_init_irq(void) #endif wbflush(); - - return; } void toshiba_rbtx4927_irq_dump(char *key) @@ -715,7 +557,6 @@ void toshiba_rbtx4927_irq_dump(char *key) } } #endif - return; } void toshiba_rbtx4927_irq_dump_pics(char *s) @@ -780,6 +621,4 @@ void toshiba_rbtx4927_irq_dump_pics(char *s) level5_s); TOSHIBA_RBTX4927_IRQ_DPRINTK(TOSHIBA_RBTX4927_IRQ_INFO, "[%s]\n", s); - - return; } diff --git a/arch/mips/tx4938/common/irq.c b/arch/mips/tx4938/common/irq.c index 77fe2454f5b..19c9ee9e3d0 100644 --- a/arch/mips/tx4938/common/irq.c +++ b/arch/mips/tx4938/common/irq.c @@ -37,48 +37,36 @@ /* Forwad definitions for all pic's */ /**********************************************************************************/ -static unsigned int tx4938_irq_cp0_startup(unsigned int irq); -static void tx4938_irq_cp0_shutdown(unsigned int irq); static void tx4938_irq_cp0_enable(unsigned int irq); static void tx4938_irq_cp0_disable(unsigned int irq); -static void tx4938_irq_cp0_mask_and_ack(unsigned int irq); static void tx4938_irq_cp0_end(unsigned int irq); -static unsigned int tx4938_irq_pic_startup(unsigned int irq); -static void tx4938_irq_pic_shutdown(unsigned int irq); static void tx4938_irq_pic_enable(unsigned int irq); static void tx4938_irq_pic_disable(unsigned int irq); -static void tx4938_irq_pic_mask_and_ack(unsigned int irq); static void tx4938_irq_pic_end(unsigned int irq); /**********************************************************************************/ /* Kernel structs for all pic's */ /**********************************************************************************/ -DEFINE_SPINLOCK(tx4938_cp0_lock); -DEFINE_SPINLOCK(tx4938_pic_lock); #define TX4938_CP0_NAME "TX4938-CP0" static struct irq_chip tx4938_irq_cp0_type = { .typename = TX4938_CP0_NAME, - .startup = tx4938_irq_cp0_startup, - .shutdown = tx4938_irq_cp0_shutdown, - .enable = tx4938_irq_cp0_enable, - .disable = tx4938_irq_cp0_disable, - .ack = tx4938_irq_cp0_mask_and_ack, + .ack = tx4938_irq_cp0_disable, + .mask = tx4938_irq_cp0_disable, + .mask_ack = tx4938_irq_cp0_disable, + .unmask = tx4938_irq_cp0_enable, .end = tx4938_irq_cp0_end, - .set_affinity = NULL }; #define TX4938_PIC_NAME "TX4938-PIC" static struct irq_chip tx4938_irq_pic_type = { .typename = TX4938_PIC_NAME, - .startup = tx4938_irq_pic_startup, - .shutdown = tx4938_irq_pic_shutdown, - .enable = tx4938_irq_pic_enable, - .disable = tx4938_irq_pic_disable, - .ack = tx4938_irq_pic_mask_and_ack, + .ack = tx4938_irq_pic_disable, + .mask = tx4938_irq_pic_disable, + .mask_ack = tx4938_irq_pic_disable, + .unmask = tx4938_irq_pic_enable, .end = tx4938_irq_pic_end, - .set_affinity = NULL }; static struct irqaction tx4938_irq_pic_action = { @@ -99,56 +87,20 @@ tx4938_irq_cp0_init(void) { int i; - for (i = TX4938_IRQ_CP0_BEG; i <= TX4938_IRQ_CP0_END; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 1; - irq_desc[i].chip = &tx4938_irq_cp0_type; - } -} - -static unsigned int -tx4938_irq_cp0_startup(unsigned int irq) -{ - tx4938_irq_cp0_enable(irq); - - return 0; -} - -static void -tx4938_irq_cp0_shutdown(unsigned int irq) -{ - tx4938_irq_cp0_disable(irq); + for (i = TX4938_IRQ_CP0_BEG; i <= TX4938_IRQ_CP0_END; i++) + set_irq_chip(i, &tx4938_irq_cp0_type); } static void tx4938_irq_cp0_enable(unsigned int irq) { - unsigned long flags; - - spin_lock_irqsave(&tx4938_cp0_lock, flags); - set_c0_status(tx4938_irq_cp0_mask(irq)); - - spin_unlock_irqrestore(&tx4938_cp0_lock, flags); } static void tx4938_irq_cp0_disable(unsigned int irq) { - unsigned long flags; - - spin_lock_irqsave(&tx4938_cp0_lock, flags); - clear_c0_status(tx4938_irq_cp0_mask(irq)); - - spin_unlock_irqrestore(&tx4938_cp0_lock, flags); -} - -static void -tx4938_irq_cp0_mask_and_ack(unsigned int irq) -{ - tx4938_irq_cp0_disable(irq); } static void @@ -290,70 +242,29 @@ tx4938_irq_pic_modify(unsigned pic_reg, unsigned clr_bits, unsigned set_bits) static void __init tx4938_irq_pic_init(void) { - unsigned long flags; int i; - for (i = TX4938_IRQ_PIC_BEG; i <= TX4938_IRQ_PIC_END; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 2; - irq_desc[i].chip = &tx4938_irq_pic_type; - } + for (i = TX4938_IRQ_PIC_BEG; i <= TX4938_IRQ_PIC_END; i++) + set_irq_chip(i, &tx4938_irq_pic_type); setup_irq(TX4938_IRQ_NEST_PIC_ON_CP0, &tx4938_irq_pic_action); - spin_lock_irqsave(&tx4938_pic_lock, flags); - TX4938_WR(0xff1ff640, 0x6); /* irq level mask -- only accept hightest */ TX4938_WR(0xff1ff600, TX4938_RD(0xff1ff600) | 0x1); /* irq enable */ - - spin_unlock_irqrestore(&tx4938_pic_lock, flags); -} - -static unsigned int -tx4938_irq_pic_startup(unsigned int irq) -{ - tx4938_irq_pic_enable(irq); - - return 0; -} - -static void -tx4938_irq_pic_shutdown(unsigned int irq) -{ - tx4938_irq_pic_disable(irq); } static void tx4938_irq_pic_enable(unsigned int irq) { - unsigned long flags; - - spin_lock_irqsave(&tx4938_pic_lock, flags); - tx4938_irq_pic_modify(tx4938_irq_pic_addr(irq), 0, tx4938_irq_pic_mask(irq)); - - spin_unlock_irqrestore(&tx4938_pic_lock, flags); } static void tx4938_irq_pic_disable(unsigned int irq) { - unsigned long flags; - - spin_lock_irqsave(&tx4938_pic_lock, flags); - tx4938_irq_pic_modify(tx4938_irq_pic_addr(irq), tx4938_irq_pic_mask(irq), 0); - - spin_unlock_irqrestore(&tx4938_pic_lock, flags); -} - -static void -tx4938_irq_pic_mask_and_ack(unsigned int irq) -{ - tx4938_irq_pic_disable(irq); } static void diff --git a/arch/mips/tx4938/toshiba_rbtx4938/irq.c b/arch/mips/tx4938/toshiba_rbtx4938/irq.c index 102e473c10a..2735ffe9ec2 100644 --- a/arch/mips/tx4938/toshiba_rbtx4938/irq.c +++ b/arch/mips/tx4938/toshiba_rbtx4938/irq.c @@ -87,25 +87,18 @@ IRQ Device #include #include -static unsigned int toshiba_rbtx4938_irq_ioc_startup(unsigned int irq); -static void toshiba_rbtx4938_irq_ioc_shutdown(unsigned int irq); static void toshiba_rbtx4938_irq_ioc_enable(unsigned int irq); static void toshiba_rbtx4938_irq_ioc_disable(unsigned int irq); -static void toshiba_rbtx4938_irq_ioc_mask_and_ack(unsigned int irq); static void toshiba_rbtx4938_irq_ioc_end(unsigned int irq); -DEFINE_SPINLOCK(toshiba_rbtx4938_ioc_lock); - #define TOSHIBA_RBTX4938_IOC_NAME "RBTX4938-IOC" static struct irq_chip toshiba_rbtx4938_irq_ioc_type = { .typename = TOSHIBA_RBTX4938_IOC_NAME, - .startup = toshiba_rbtx4938_irq_ioc_startup, - .shutdown = toshiba_rbtx4938_irq_ioc_shutdown, - .enable = toshiba_rbtx4938_irq_ioc_enable, - .disable = toshiba_rbtx4938_irq_ioc_disable, - .ack = toshiba_rbtx4938_irq_ioc_mask_and_ack, + .ack = toshiba_rbtx4938_irq_ioc_disable, + .mask = toshiba_rbtx4938_irq_ioc_disable, + .mask_ack = toshiba_rbtx4938_irq_ioc_disable, + .unmask = toshiba_rbtx4938_irq_ioc_enable, .end = toshiba_rbtx4938_irq_ioc_end, - .set_affinity = NULL }; #define TOSHIBA_RBTX4938_IOC_INTR_ENAB 0xb7f02000 @@ -142,69 +135,35 @@ toshiba_rbtx4938_irq_ioc_init(void) int i; for (i = TOSHIBA_RBTX4938_IRQ_IOC_BEG; - i <= TOSHIBA_RBTX4938_IRQ_IOC_END; i++) { - irq_desc[i].status = IRQ_DISABLED; - irq_desc[i].action = 0; - irq_desc[i].depth = 3; - irq_desc[i].chip = &toshiba_rbtx4938_irq_ioc_type; - } + i <= TOSHIBA_RBTX4938_IRQ_IOC_END; i++) + set_irq_chip(i, &toshiba_rbtx4938_irq_ioc_type); setup_irq(RBTX4938_IRQ_IOCINT, &toshiba_rbtx4938_irq_ioc_action); } -static unsigned int -toshiba_rbtx4938_irq_ioc_startup(unsigned int irq) -{ - toshiba_rbtx4938_irq_ioc_enable(irq); - - return 0; -} - -static void -toshiba_rbtx4938_irq_ioc_shutdown(unsigned int irq) -{ - toshiba_rbtx4938_irq_ioc_disable(irq); -} - static void toshiba_rbtx4938_irq_ioc_enable(unsigned int irq) { - unsigned long flags; volatile unsigned char v; - spin_lock_irqsave(&toshiba_rbtx4938_ioc_lock, flags); - v = TX4938_RD08(TOSHIBA_RBTX4938_IOC_INTR_ENAB); v |= (1 << (irq - TOSHIBA_RBTX4938_IRQ_IOC_BEG)); TX4938_WR08(TOSHIBA_RBTX4938_IOC_INTR_ENAB, v); mmiowb(); TX4938_RD08(TOSHIBA_RBTX4938_IOC_INTR_ENAB); - - spin_unlock_irqrestore(&toshiba_rbtx4938_ioc_lock, flags); } static void toshiba_rbtx4938_irq_ioc_disable(unsigned int irq) { - unsigned long flags; volatile unsigned char v; - spin_lock_irqsave(&toshiba_rbtx4938_ioc_lock, flags); - v = TX4938_RD08(TOSHIBA_RBTX4938_IOC_INTR_ENAB); v &= ~(1 << (irq - TOSHIBA_RBTX4938_IRQ_IOC_BEG)); TX4938_WR08(TOSHIBA_RBTX4938_IOC_INTR_ENAB, v); mmiowb(); TX4938_RD08(TOSHIBA_RBTX4938_IOC_INTR_ENAB); - - spin_unlock_irqrestore(&toshiba_rbtx4938_ioc_lock, flags); -} - -static void -toshiba_rbtx4938_irq_ioc_mask_and_ack(unsigned int irq) -{ - toshiba_rbtx4938_irq_ioc_disable(irq); } static void diff --git a/arch/mips/vr41xx/common/icu.c b/arch/mips/vr41xx/common/icu.c index c215c0d39fa..33d70a6547a 100644 --- a/arch/mips/vr41xx/common/icu.c +++ b/arch/mips/vr41xx/common/icu.c @@ -417,14 +417,7 @@ void vr41xx_disable_bcuint(void) EXPORT_SYMBOL(vr41xx_disable_bcuint); -static unsigned int startup_sysint1_irq(unsigned int irq) -{ - icu1_set(MSYSINT1REG, 1 << SYSINT1_IRQ_TO_PIN(irq)); - - return 0; /* never anything pending */ -} - -static void shutdown_sysint1_irq(unsigned int irq) +static void disable_sysint1_irq(unsigned int irq) { icu1_clear(MSYSINT1REG, 1 << SYSINT1_IRQ_TO_PIN(irq)); } @@ -434,9 +427,6 @@ static void enable_sysint1_irq(unsigned int irq) icu1_set(MSYSINT1REG, 1 << SYSINT1_IRQ_TO_PIN(irq)); } -#define disable_sysint1_irq shutdown_sysint1_irq -#define ack_sysint1_irq shutdown_sysint1_irq - static void end_sysint1_irq(unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS))) @@ -445,22 +435,14 @@ static void end_sysint1_irq(unsigned int irq) static struct irq_chip sysint1_irq_type = { .typename = "SYSINT1", - .startup = startup_sysint1_irq, - .shutdown = shutdown_sysint1_irq, - .enable = enable_sysint1_irq, - .disable = disable_sysint1_irq, - .ack = ack_sysint1_irq, + .ack = disable_sysint1_irq, + .mask = disable_sysint1_irq, + .mask_ack = disable_sysint1_irq, + .unmask = enable_sysint1_irq, .end = end_sysint1_irq, }; -static unsigned int startup_sysint2_irq(unsigned int irq) -{ - icu2_set(MSYSINT2REG, 1 << SYSINT2_IRQ_TO_PIN(irq)); - - return 0; /* never anything pending */ -} - -static void shutdown_sysint2_irq(unsigned int irq) +static void disable_sysint2_irq(unsigned int irq) { icu2_clear(MSYSINT2REG, 1 << SYSINT2_IRQ_TO_PIN(irq)); } @@ -470,9 +452,6 @@ static void enable_sysint2_irq(unsigned int irq) icu2_set(MSYSINT2REG, 1 << SYSINT2_IRQ_TO_PIN(irq)); } -#define disable_sysint2_irq shutdown_sysint2_irq -#define ack_sysint2_irq shutdown_sysint2_irq - static void end_sysint2_irq(unsigned int irq) { if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS))) @@ -481,11 +460,10 @@ static void end_sysint2_irq(unsigned int irq) static struct irq_chip sysint2_irq_type = { .typename = "SYSINT2", - .startup = startup_sysint2_irq, - .shutdown = shutdown_sysint2_irq, - .enable = enable_sysint2_irq, - .disable = disable_sysint2_irq, - .ack = ack_sysint2_irq, + .ack = disable_sysint2_irq, + .mask = disable_sysint2_irq, + .mask_ack = disable_sysint2_irq, + .unmask = enable_sysint2_irq, .end = end_sysint2_irq, }; @@ -723,10 +701,10 @@ static int __init vr41xx_icu_init(void) icu2_write(MGIUINTHREG, 0xffff); for (i = SYSINT1_IRQ_BASE; i <= SYSINT1_IRQ_LAST; i++) - irq_desc[i].chip = &sysint1_irq_type; + set_irq_chip(i, &sysint1_irq_type); for (i = SYSINT2_IRQ_BASE; i <= SYSINT2_IRQ_LAST; i++) - irq_desc[i].chip = &sysint2_irq_type; + set_irq_chip(i, &sysint2_irq_type); cascade_irq(INT0_IRQ, icu_get_irq); cascade_irq(INT1_IRQ, icu_get_irq); diff --git a/arch/mips/vr41xx/nec-cmbvr4133/irq.c b/arch/mips/vr41xx/nec-cmbvr4133/irq.c index 2483487344c..a039bb7251f 100644 --- a/arch/mips/vr41xx/nec-cmbvr4133/irq.c +++ b/arch/mips/vr41xx/nec-cmbvr4133/irq.c @@ -30,17 +30,6 @@ extern void init_8259A(int hoge); extern int vr4133_rockhopper; -static unsigned int startup_i8259_irq(unsigned int irq) -{ - enable_8259A_irq(irq - I8259_IRQ_BASE); - return 0; -} - -static void shutdown_i8259_irq(unsigned int irq) -{ - disable_8259A_irq(irq - I8259_IRQ_BASE); -} - static void enable_i8259_irq(unsigned int irq) { enable_8259A_irq(irq - I8259_IRQ_BASE); @@ -64,11 +53,10 @@ static void end_i8259_irq(unsigned int irq) static struct irq_chip i8259_irq_type = { .typename = "XT-PIC", - .startup = startup_i8259_irq, - .shutdown = shutdown_i8259_irq, - .enable = enable_i8259_irq, - .disable = disable_i8259_irq, .ack = ack_i8259_irq, + .mask = disable_i8259_irq, + .mask_ack = ack_i8259_irq, + .unmask = enable_i8259_irq, .end = end_i8259_irq, }; @@ -104,7 +92,7 @@ void __init rockhopper_init_irq(void) } for (i = I8259_IRQ_BASE; i <= I8259_IRQ_LAST; i++) - irq_desc[i].chip = &i8259_irq_type; + set_irq_chip(i, &i8259_irq_type); setup_irq(I8259_SLAVE_IRQ, &i8259_slave_cascade); diff --git a/include/asm-mips/dec/kn02.h b/include/asm-mips/dec/kn02.h index 8319ad77b25..93430b5f472 100644 --- a/include/asm-mips/dec/kn02.h +++ b/include/asm-mips/dec/kn02.h @@ -82,11 +82,9 @@ #ifndef __ASSEMBLY__ -#include #include extern u32 cached_kn02_csr; -extern spinlock_t kn02_lock; extern void init_kn02_irqs(int base); #endif -- cgit v1.2.3-70-g09d2 From 1417836e81c0ab8f5a0bfeafa90d3eaa41b2a067 Mon Sep 17 00:00:00 2001 From: Atsushi Nemoto Date: Tue, 14 Nov 2006 01:13:18 +0900 Subject: [MIPS] use generic_handle_irq, handle_level_irq, handle_percpu_irq Further incorporation of generic irq framework. Replacing __do_IRQ() by proper flow handler would make the irq handling path a bit simpler and faster. * use generic_handle_irq() instead of __do_IRQ(). * use handle_level_irq for obvious level-type irq chips. * use handle_percpu_irq for irqs marked as IRQ_PER_CPU. * setup .eoi routine for irq chips possibly used with handle_percpu_irq. Signed-off-by: Atsushi Nemoto Signed-off-by: Ralf Baechle --- arch/mips/dec/ioasic-irq.c | 6 ++++-- arch/mips/dec/kn02-irq.c | 2 +- arch/mips/emma2rh/common/irq_emma2rh.c | 3 ++- arch/mips/emma2rh/markeins/irq_markeins.c | 3 ++- arch/mips/jazz/irq.c | 2 +- arch/mips/kernel/irq-msc01.c | 2 ++ arch/mips/kernel/irq-mv6434x.c | 3 ++- arch/mips/kernel/irq-rm7000.c | 3 ++- arch/mips/kernel/irq-rm9000.c | 6 ++++-- arch/mips/kernel/irq_cpu.c | 5 ++++- arch/mips/kernel/smp-mt.c | 2 ++ arch/mips/kernel/smtc.c | 1 + arch/mips/lasat/interrupt.c | 2 +- arch/mips/mips-boards/atlas/atlas_int.c | 3 ++- arch/mips/mips-boards/generic/time.c | 1 + arch/mips/mips-boards/sim/sim_time.c | 3 ++- arch/mips/momentum/ocelot_c/cpci-irq.c | 2 +- arch/mips/momentum/ocelot_c/uart-irq.c | 4 ++-- arch/mips/philips/pnx8550/common/int.c | 12 +++++++----- arch/mips/sgi-ip22/ip22-int.c | 2 +- arch/mips/sgi-ip27/ip27-irq.c | 2 +- arch/mips/sgi-ip27/ip27-timer.c | 3 ++- arch/mips/tx4927/common/tx4927_irq.c | 6 ++++-- arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_irq.c | 3 ++- arch/mips/tx4938/common/irq.c | 6 ++++-- arch/mips/tx4938/toshiba_rbtx4938/irq.c | 3 ++- arch/mips/vr41xx/common/icu.c | 6 ++++-- include/asm-mips/irq.h | 2 +- 28 files changed, 64 insertions(+), 34 deletions(-) (limited to 'include') diff --git a/arch/mips/dec/ioasic-irq.c b/arch/mips/dec/ioasic-irq.c index d0af08bdbb4..269b22b3431 100644 --- a/arch/mips/dec/ioasic-irq.c +++ b/arch/mips/dec/ioasic-irq.c @@ -103,9 +103,11 @@ void __init init_ioasic_irqs(int base) fast_iob(); for (i = base; i < base + IO_INR_DMA; i++) - set_irq_chip(i, &ioasic_irq_type); + set_irq_chip_and_handler(i, &ioasic_irq_type, + handle_level_irq); for (; i < base + IO_IRQ_LINES; i++) - set_irq_chip(i, &ioasic_dma_irq_type); + set_irq_chip_and_handler(i, &ioasic_dma_irq_type, + handle_level_irq); ioasic_irq_base = base; } diff --git a/arch/mips/dec/kn02-irq.c b/arch/mips/dec/kn02-irq.c index c761d97787e..5a9be4c9358 100644 --- a/arch/mips/dec/kn02-irq.c +++ b/arch/mips/dec/kn02-irq.c @@ -85,7 +85,7 @@ void __init init_kn02_irqs(int base) iob(); for (i = base; i < base + KN02_IRQ_LINES; i++) - set_irq_chip(i, &kn02_irq_type); + set_irq_chip_and_handler(i, &kn02_irq_type, handle_level_irq); kn02_irq_base = base; } diff --git a/arch/mips/emma2rh/common/irq_emma2rh.c b/arch/mips/emma2rh/common/irq_emma2rh.c index bf1b83ba925..59b98299c89 100644 --- a/arch/mips/emma2rh/common/irq_emma2rh.c +++ b/arch/mips/emma2rh/common/irq_emma2rh.c @@ -76,7 +76,8 @@ void emma2rh_irq_init(u32 irq_base) u32 i; for (i = irq_base; i < irq_base + NUM_EMMA2RH_IRQ; i++) - set_irq_chip(i, &emma2rh_irq_controller); + set_irq_chip_and_handler(i, &emma2rh_irq_controller, + handle_level_irq); emma2rh_irq_base = irq_base; } diff --git a/arch/mips/emma2rh/markeins/irq_markeins.c b/arch/mips/emma2rh/markeins/irq_markeins.c index 8e5f08a4245..3ac4e405ecd 100644 --- a/arch/mips/emma2rh/markeins/irq_markeins.c +++ b/arch/mips/emma2rh/markeins/irq_markeins.c @@ -68,7 +68,8 @@ void emma2rh_sw_irq_init(u32 irq_base) u32 i; for (i = irq_base; i < irq_base + NUM_EMMA2RH_IRQ_SW; i++) - set_irq_chip(i, &emma2rh_sw_irq_controller); + set_irq_chip_and_handler(i, &emma2rh_sw_irq_controller, + handle_level_irq); emma2rh_sw_irq_base = irq_base; } diff --git a/arch/mips/jazz/irq.c b/arch/mips/jazz/irq.c index 4bbb6cb08d6..5c4f50cdf15 100644 --- a/arch/mips/jazz/irq.c +++ b/arch/mips/jazz/irq.c @@ -59,7 +59,7 @@ void __init init_r4030_ints(void) int i; for (i = JAZZ_PARALLEL_IRQ; i <= JAZZ_TIMER_IRQ; i++) - set_irq_chip(i, &r4030_irq_type); + set_irq_chip_and_handler(i, &r4030_irq_type, handle_level_irq); r4030_write_reg16(JAZZ_IO_IRQ_ENABLE, 0); r4030_read_reg16(JAZZ_IO_IRQ_SOURCE); /* clear pending IRQs */ diff --git a/arch/mips/kernel/irq-msc01.c b/arch/mips/kernel/irq-msc01.c index e1880b27381..bcaad669608 100644 --- a/arch/mips/kernel/irq-msc01.c +++ b/arch/mips/kernel/irq-msc01.c @@ -117,6 +117,7 @@ struct irq_chip msc_levelirq_type = { .mask = mask_msc_irq, .mask_ack = level_mask_and_ack_msc_irq, .unmask = unmask_msc_irq, + .eoi = unmask_msc_irq, .end = end_msc_irq, }; @@ -126,6 +127,7 @@ struct irq_chip msc_edgeirq_type = { .mask = mask_msc_irq, .mask_ack = edge_mask_and_ack_msc_irq, .unmask = unmask_msc_irq, + .eoi = unmask_msc_irq, .end = end_msc_irq, }; diff --git a/arch/mips/kernel/irq-mv6434x.c b/arch/mips/kernel/irq-mv6434x.c index 5012b9df1b5..6cfb31cafde 100644 --- a/arch/mips/kernel/irq-mv6434x.c +++ b/arch/mips/kernel/irq-mv6434x.c @@ -114,7 +114,8 @@ void __init mv64340_irq_init(unsigned int base) int i; for (i = base; i < base + 64; i++) - set_irq_chip(i, &mv64340_irq_type); + set_irq_chip_and_handler(i, &mv64340_irq_type, + handle_level_irq); irq_base = base; } diff --git a/arch/mips/kernel/irq-rm7000.c b/arch/mips/kernel/irq-rm7000.c index 6a297e3b889..ddcc2a5f8a0 100644 --- a/arch/mips/kernel/irq-rm7000.c +++ b/arch/mips/kernel/irq-rm7000.c @@ -51,7 +51,8 @@ void __init rm7k_cpu_irq_init(int base) clear_c0_intcontrol(0x00000f00); /* Mask all */ for (i = base; i < base + 4; i++) - set_irq_chip(i, &rm7k_irq_controller); + set_irq_chip_and_handler(i, &rm7k_irq_controller, + handle_level_irq); irq_base = base; } diff --git a/arch/mips/kernel/irq-rm9000.c b/arch/mips/kernel/irq-rm9000.c index 977538445cf..ba6440c88ab 100644 --- a/arch/mips/kernel/irq-rm9000.c +++ b/arch/mips/kernel/irq-rm9000.c @@ -117,10 +117,12 @@ void __init rm9k_cpu_irq_init(int base) clear_c0_intcontrol(0x0000f000); /* Mask all */ for (i = base; i < base + 4; i++) - set_irq_chip(i, &rm9k_irq_controller); + set_irq_chip_and_handler(i, &rm9k_irq_controller, + handle_level_irq); rm9000_perfcount_irq = base + 1; - set_irq_chip(rm9000_perfcount_irq, &rm9k_perfcounter_irq); + set_irq_chip_and_handler(rm9000_perfcount_irq, &rm9k_perfcounter_irq, + handle_level_irq); irq_base = base; } diff --git a/arch/mips/kernel/irq_cpu.c b/arch/mips/kernel/irq_cpu.c index 3b7cfa407e8..be5ac23d381 100644 --- a/arch/mips/kernel/irq_cpu.c +++ b/arch/mips/kernel/irq_cpu.c @@ -62,6 +62,7 @@ static struct irq_chip mips_cpu_irq_controller = { .mask = mask_mips_irq, .mask_ack = mask_mips_irq, .unmask = unmask_mips_irq, + .eoi = unmask_mips_irq, .end = mips_cpu_irq_end, }; @@ -104,6 +105,7 @@ static struct irq_chip mips_mt_cpu_irq_controller = { .mask = mask_mips_mt_irq, .mask_ack = mips_mt_cpu_irq_ack, .unmask = unmask_mips_mt_irq, + .eoi = unmask_mips_mt_irq, .end = mips_mt_cpu_irq_end, }; @@ -124,7 +126,8 @@ void __init mips_cpu_irq_init(int irq_base) set_irq_chip(i, &mips_mt_cpu_irq_controller); for (i = irq_base + 2; i < irq_base + 8; i++) - set_irq_chip(i, &mips_cpu_irq_controller); + set_irq_chip_and_handler(i, &mips_cpu_irq_controller, + handle_level_irq); mips_cpu_irq_base = irq_base; } diff --git a/arch/mips/kernel/smp-mt.c b/arch/mips/kernel/smp-mt.c index 2ac19a6cbf6..1ee689c0e0c 100644 --- a/arch/mips/kernel/smp-mt.c +++ b/arch/mips/kernel/smp-mt.c @@ -278,7 +278,9 @@ void __init plat_prepare_cpus(unsigned int max_cpus) /* need to mark IPI's as IRQ_PER_CPU */ irq_desc[cpu_ipi_resched_irq].status |= IRQ_PER_CPU; + set_irq_handler(cpu_ipi_resched_irq, handle_percpu_irq); irq_desc[cpu_ipi_call_irq].status |= IRQ_PER_CPU; + set_irq_handler(cpu_ipi_call_irq, handle_percpu_irq); } /* diff --git a/arch/mips/kernel/smtc.c b/arch/mips/kernel/smtc.c index 3b78caf112f..802febed7df 100644 --- a/arch/mips/kernel/smtc.c +++ b/arch/mips/kernel/smtc.c @@ -1009,6 +1009,7 @@ void setup_cross_vpe_interrupts(void) setup_irq_smtc(cpu_ipi_irq, &irq_ipi, (0x100 << MIPS_CPU_IPI_IRQ)); irq_desc[cpu_ipi_irq].status |= IRQ_PER_CPU; + set_irq_handler(cpu_ipi_irq, handle_percpu_irq); } /* diff --git a/arch/mips/lasat/interrupt.c b/arch/mips/lasat/interrupt.c index cac82afe5eb..4a84a7beac5 100644 --- a/arch/mips/lasat/interrupt.c +++ b/arch/mips/lasat/interrupt.c @@ -133,5 +133,5 @@ void __init arch_init_irq(void) } for (i = 0; i <= LASATINT_END; i++) - set_irq_chip(i, &lasat_irq_type); + set_irq_chip_and_handler(i, &lasat_irq_type, handle_level_irq); } diff --git a/arch/mips/mips-boards/atlas/atlas_int.c b/arch/mips/mips-boards/atlas/atlas_int.c index 7c710040d3f..43dba6ce660 100644 --- a/arch/mips/mips-boards/atlas/atlas_int.c +++ b/arch/mips/mips-boards/atlas/atlas_int.c @@ -74,6 +74,7 @@ static struct irq_chip atlas_irq_type = { .mask = disable_atlas_irq, .mask_ack = disable_atlas_irq, .unmask = enable_atlas_irq, + .eoi = enable_atlas_irq, .end = end_atlas_irq, }; @@ -207,7 +208,7 @@ static inline void init_atlas_irqs (int base) atlas_hw0_icregs->intrsten = 0xffffffff; for (i = ATLAS_INT_BASE; i <= ATLAS_INT_END; i++) - set_irq_chip(i, &atlas_irq_type); + set_irq_chip_and_handler(i, &atlas_irq_type, handle_level_irq); } static struct irqaction atlasirq = { diff --git a/arch/mips/mips-boards/generic/time.c b/arch/mips/mips-boards/generic/time.c index d817c60c5ca..e4604c73f02 100644 --- a/arch/mips/mips-boards/generic/time.c +++ b/arch/mips/mips-boards/generic/time.c @@ -288,6 +288,7 @@ void __init plat_timer_setup(struct irqaction *irq) The effect is that the int remains disabled on the second cpu. Mark the interrupt with IRQ_PER_CPU to avoid any confusion */ irq_desc[mips_cpu_timer_irq].status |= IRQ_PER_CPU; + set_irq_handler(mips_cpu_timer_irq, handle_percpu_irq); #endif /* to generate the first timer interrupt */ diff --git a/arch/mips/mips-boards/sim/sim_time.c b/arch/mips/mips-boards/sim/sim_time.c index 24a4ed00cc0..f2d998d2c16 100644 --- a/arch/mips/mips-boards/sim/sim_time.c +++ b/arch/mips/mips-boards/sim/sim_time.c @@ -203,7 +203,8 @@ void __init plat_timer_setup(struct irqaction *irq) on seperate cpu's the first one tries to handle the second interrupt. The effect is that the int remains disabled on the second cpu. Mark the interrupt with IRQ_PER_CPU to avoid any confusion */ - irq_desc[mips_cpu_timer_irq].status |= IRQ_PER_CPU; + irq_desc[mips_cpu_timer_irq].flags |= IRQ_PER_CPU; + set_irq_handler(mips_cpu_timer_irq, handle_percpu_irq); #endif /* to generate the first timer interrupt */ diff --git a/arch/mips/momentum/ocelot_c/cpci-irq.c b/arch/mips/momentum/ocelot_c/cpci-irq.c index 7723f099894..e5a4a0a8a7f 100644 --- a/arch/mips/momentum/ocelot_c/cpci-irq.c +++ b/arch/mips/momentum/ocelot_c/cpci-irq.c @@ -106,5 +106,5 @@ void cpci_irq_init(void) int i; for (i = CPCI_IRQ_BASE; i < (CPCI_IRQ_BASE + 8); i++) - set_irq_chip(i, &cpci_irq_type); + set_irq_chip_and_handler(i, &cpci_irq_type, handle_level_irq); } diff --git a/arch/mips/momentum/ocelot_c/uart-irq.c b/arch/mips/momentum/ocelot_c/uart-irq.c index 72faf81b36c..0029f0008de 100644 --- a/arch/mips/momentum/ocelot_c/uart-irq.c +++ b/arch/mips/momentum/ocelot_c/uart-irq.c @@ -96,6 +96,6 @@ struct irq_chip uart_irq_type = { void uart_irq_init(void) { - set_irq_chip(80, &uart_irq_type); - set_irq_chip(81, &uart_irq_type); + set_irq_chip_and_handler(80, &uart_irq_type, handle_level_irq); + set_irq_chip_and_handler(81, &uart_irq_type, handle_level_irq); } diff --git a/arch/mips/philips/pnx8550/common/int.c b/arch/mips/philips/pnx8550/common/int.c index e4bf494dd43..0dc23930edb 100644 --- a/arch/mips/philips/pnx8550/common/int.c +++ b/arch/mips/philips/pnx8550/common/int.c @@ -192,7 +192,7 @@ void __init arch_init_irq(void) int configPR; for (i = 0; i < PNX8550_INT_CP0_TOTINT; i++) { - set_irq_chip(i, &level_irq_type); + set_irq_chip_and_handler(i, &level_irq_type, handle_level_irq); mask_irq(i); /* mask the irq just in case */ } @@ -229,7 +229,7 @@ void __init arch_init_irq(void) /* mask/priority is still 0 so we will not get any * interrupts until it is unmasked */ - set_irq_chip(i, &level_irq_type); + set_irq_chip_and_handler(i, &level_irq_type, handle_level_irq); } /* Priority level 0 */ @@ -238,19 +238,21 @@ void __init arch_init_irq(void) /* Set int vector table address */ PNX8550_GIC_VECTOR_0 = PNX8550_GIC_VECTOR_1 = 0; - set_irq_chip(MIPS_CPU_GIC_IRQ, &level_irq_type); + set_irq_chip_and_handler(MIPS_CPU_GIC_IRQ, &level_irq_type, + handle_level_irq); setup_irq(MIPS_CPU_GIC_IRQ, &gic_action); /* init of Timer interrupts */ for (i = PNX8550_INT_TIMER_MIN; i <= PNX8550_INT_TIMER_MAX; i++) - set_irq_chip(i, &level_irq_type); + set_irq_chip_and_handler(i, &level_irq_type, handle_level_irq); /* Stop Timer 1-3 */ configPR = read_c0_config7(); configPR |= 0x00000038; write_c0_config7(configPR); - set_irq_chip(MIPS_CPU_TIMER_IRQ, &level_irq_type); + set_irq_chip_and_handler(MIPS_CPU_TIMER_IRQ, &level_irq_type, + handle_level_irq); setup_irq(MIPS_CPU_TIMER_IRQ, &timer_action); } diff --git a/arch/mips/sgi-ip22/ip22-int.c b/arch/mips/sgi-ip22/ip22-int.c index 8e2074b4ce4..c7b13805315 100644 --- a/arch/mips/sgi-ip22/ip22-int.c +++ b/arch/mips/sgi-ip22/ip22-int.c @@ -358,7 +358,7 @@ void __init arch_init_irq(void) else handler = &ip22_local3_irq_type; - set_irq_chip(i, handler); + set_irq_chip_and_handler(i, handler, handle_level_irq); } /* vector handler. this register the IRQ as non-sharable */ diff --git a/arch/mips/sgi-ip27/ip27-irq.c b/arch/mips/sgi-ip27/ip27-irq.c index 824320281a3..5f8835b4e84 100644 --- a/arch/mips/sgi-ip27/ip27-irq.c +++ b/arch/mips/sgi-ip27/ip27-irq.c @@ -352,7 +352,7 @@ static struct irq_chip bridge_irq_type = { void __devinit register_bridge_irq(unsigned int irq) { - set_irq_chip(irq, &bridge_irq_type); + set_irq_chip_and_handler(irq, &bridge_irq_type, handle_level_irq); } int __devinit request_bridge_irq(struct bridge_controller *bc) diff --git a/arch/mips/sgi-ip27/ip27-timer.c b/arch/mips/sgi-ip27/ip27-timer.c index 86ba7fc10c3..e5441c3a0b0 100644 --- a/arch/mips/sgi-ip27/ip27-timer.c +++ b/arch/mips/sgi-ip27/ip27-timer.c @@ -190,6 +190,7 @@ static struct irq_chip rt_irq_type = { .mask = disable_rt_irq, .mask_ack = disable_rt_irq, .unmask = enable_rt_irq, + .eoi = enable_rt_irq, .end = end_rt_irq, }; @@ -207,7 +208,7 @@ void __init plat_timer_setup(struct irqaction *irq) if (irqno < 0) panic("Can't allocate interrupt number for timer interrupt"); - set_irq_chip(irqno, &rt_irq_type); + set_irq_chip_and_handler(irqno, &rt_irq_type, handle_percpu_irq); /* over-write the handler, we use our own way */ irq->handler = no_action; diff --git a/arch/mips/tx4927/common/tx4927_irq.c b/arch/mips/tx4927/common/tx4927_irq.c index 2c57ced5c68..21873de49aa 100644 --- a/arch/mips/tx4927/common/tx4927_irq.c +++ b/arch/mips/tx4927/common/tx4927_irq.c @@ -196,7 +196,8 @@ static void __init tx4927_irq_cp0_init(void) TX4927_IRQ_CP0_BEG, TX4927_IRQ_CP0_END); for (i = TX4927_IRQ_CP0_BEG; i <= TX4927_IRQ_CP0_END; i++) - set_irq_chip(i, &tx4927_irq_cp0_type); + set_irq_chip_and_handler(i, &tx4927_irq_cp0_type, + handle_level_irq); } static void tx4927_irq_cp0_enable(unsigned int irq) @@ -350,7 +351,8 @@ static void __init tx4927_irq_pic_init(void) TX4927_IRQ_PIC_BEG, TX4927_IRQ_PIC_END); for (i = TX4927_IRQ_PIC_BEG; i <= TX4927_IRQ_PIC_END; i++) - set_irq_chip(i, &tx4927_irq_pic_type); + set_irq_chip_and_handler(i, &tx4927_irq_pic_type, + handle_level_irq); setup_irq(TX4927_IRQ_NEST_PIC_ON_CP0, &tx4927_irq_pic_action); diff --git a/arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_irq.c b/arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_irq.c index 1fdace89ae6..34cdb2a240e 100644 --- a/arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_irq.c +++ b/arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_irq.c @@ -342,7 +342,8 @@ static void __init toshiba_rbtx4927_irq_ioc_init(void) for (i = TOSHIBA_RBTX4927_IRQ_IOC_BEG; i <= TOSHIBA_RBTX4927_IRQ_IOC_END; i++) - set_irq_chip(i, &toshiba_rbtx4927_irq_ioc_type); + set_irq_chip_and_handler(i, &toshiba_rbtx4927_irq_ioc_type, + handle_level_irq); setup_irq(TOSHIBA_RBTX4927_IRQ_NEST_IOC_ON_PIC, &toshiba_rbtx4927_irq_ioc_action); diff --git a/arch/mips/tx4938/common/irq.c b/arch/mips/tx4938/common/irq.c index 19c9ee9e3d0..42e127683ae 100644 --- a/arch/mips/tx4938/common/irq.c +++ b/arch/mips/tx4938/common/irq.c @@ -88,7 +88,8 @@ tx4938_irq_cp0_init(void) int i; for (i = TX4938_IRQ_CP0_BEG; i <= TX4938_IRQ_CP0_END; i++) - set_irq_chip(i, &tx4938_irq_cp0_type); + set_irq_chip_and_handler(i, &tx4938_irq_cp0_type, + handle_level_irq); } static void @@ -245,7 +246,8 @@ tx4938_irq_pic_init(void) int i; for (i = TX4938_IRQ_PIC_BEG; i <= TX4938_IRQ_PIC_END; i++) - set_irq_chip(i, &tx4938_irq_pic_type); + set_irq_chip_and_handler(i, &tx4938_irq_pic_type, + handle_level_irq); setup_irq(TX4938_IRQ_NEST_PIC_ON_CP0, &tx4938_irq_pic_action); diff --git a/arch/mips/tx4938/toshiba_rbtx4938/irq.c b/arch/mips/tx4938/toshiba_rbtx4938/irq.c index 2735ffe9ec2..8c87a35f306 100644 --- a/arch/mips/tx4938/toshiba_rbtx4938/irq.c +++ b/arch/mips/tx4938/toshiba_rbtx4938/irq.c @@ -136,7 +136,8 @@ toshiba_rbtx4938_irq_ioc_init(void) for (i = TOSHIBA_RBTX4938_IRQ_IOC_BEG; i <= TOSHIBA_RBTX4938_IRQ_IOC_END; i++) - set_irq_chip(i, &toshiba_rbtx4938_irq_ioc_type); + set_irq_chip_and_handler(i, &toshiba_rbtx4938_irq_ioc_type, + handle_level_irq); setup_irq(RBTX4938_IRQ_IOCINT, &toshiba_rbtx4938_irq_ioc_action); diff --git a/arch/mips/vr41xx/common/icu.c b/arch/mips/vr41xx/common/icu.c index 33d70a6547a..54b92a74c7a 100644 --- a/arch/mips/vr41xx/common/icu.c +++ b/arch/mips/vr41xx/common/icu.c @@ -701,10 +701,12 @@ static int __init vr41xx_icu_init(void) icu2_write(MGIUINTHREG, 0xffff); for (i = SYSINT1_IRQ_BASE; i <= SYSINT1_IRQ_LAST; i++) - set_irq_chip(i, &sysint1_irq_type); + set_irq_chip_and_handler(i, &sysint1_irq_type, + handle_level_irq); for (i = SYSINT2_IRQ_BASE; i <= SYSINT2_IRQ_LAST; i++) - set_irq_chip(i, &sysint2_irq_type); + set_irq_chip_and_handler(i, &sysint2_irq_type, + handle_level_irq); cascade_irq(INT0_IRQ, icu_get_irq); cascade_irq(INT1_IRQ, icu_get_irq); diff --git a/include/asm-mips/irq.h b/include/asm-mips/irq.h index 35a05ca5560..aed37077022 100644 --- a/include/asm-mips/irq.h +++ b/include/asm-mips/irq.h @@ -57,7 +57,7 @@ do { \ do { \ irq_enter(); \ __DO_IRQ_SMTC_HOOK(); \ - __do_IRQ((irq)); \ + generic_handle_irq(irq); \ irq_exit(); \ } while (0) -- cgit v1.2.3-70-g09d2 From 187933f23679c413706030aefad9e85e79164c44 Mon Sep 17 00:00:00 2001 From: Atsushi Nemoto Date: Wed, 25 Oct 2006 23:57:04 +0900 Subject: [MIPS] do_IRQ cleanup Now we have both function and macro version of do_IRQ() and the former is used only by DEC and non-preemptive kernel. This patch makes everyone use the macro version and removes the function version. Signed-off-by: Atsushi Nemoto Signed-off-by: Ralf Baechle --- arch/mips/dec/int-handler.S | 2 +- arch/mips/dec/setup.c | 6 ++++++ arch/mips/kernel/irq.c | 19 ------------------- include/asm-mips/irq.h | 6 ------ 4 files changed, 7 insertions(+), 26 deletions(-) (limited to 'include') diff --git a/arch/mips/dec/int-handler.S b/arch/mips/dec/int-handler.S index 31dd47d1002..b251ef864c3 100644 --- a/arch/mips/dec/int-handler.S +++ b/arch/mips/dec/int-handler.S @@ -267,7 +267,7 @@ handle_it: LONG_L s0, TI_REGS($28) LONG_S sp, TI_REGS($28) PTR_LA ra, ret_from_irq - j do_IRQ + j dec_irq_dispatch nop #ifdef CONFIG_32BIT diff --git a/arch/mips/dec/setup.c b/arch/mips/dec/setup.c index 6b7481e97be..d34032ac492 100644 --- a/arch/mips/dec/setup.c +++ b/arch/mips/dec/setup.c @@ -761,3 +761,9 @@ void __init arch_init_irq(void) if (dec_interrupt[DEC_IRQ_HALT] >= 0) setup_irq(dec_interrupt[DEC_IRQ_HALT], &haltirq); } + +asmlinkage unsigned int dec_irq_dispatch(unsigned int irq) +{ + do_IRQ(irq); + return 0; +} diff --git a/arch/mips/kernel/irq.c b/arch/mips/kernel/irq.c index e997c9454f4..b339798b317 100644 --- a/arch/mips/kernel/irq.c +++ b/arch/mips/kernel/irq.c @@ -88,25 +88,6 @@ atomic_t irq_err_count; unsigned long irq_hwmask[NR_IRQS]; #endif /* CONFIG_MIPS_MT_SMTC */ -#undef do_IRQ - -/* - * do_IRQ handles all normal device IRQ's (the special - * SMP cross-CPU interrupts have their own specific - * handlers). - */ -asmlinkage unsigned int do_IRQ(unsigned int irq) -{ - irq_enter(); - - __DO_IRQ_SMTC_HOOK(); - __do_IRQ(irq); - - irq_exit(); - - return 1; -} - /* * Generic, controller-independent functions: */ diff --git a/include/asm-mips/irq.h b/include/asm-mips/irq.h index aed37077022..67657089efa 100644 --- a/include/asm-mips/irq.h +++ b/include/asm-mips/irq.h @@ -24,8 +24,6 @@ static inline int irq_canonicalize(int irq) #define irq_canonicalize(irq) (irq) /* Sane hardware, sane code ... */ #endif -extern asmlinkage unsigned int do_IRQ(unsigned int irq); - #ifdef CONFIG_MIPS_MT_SMTC /* * Clear interrupt mask handling "backstop" if irq_hwmask @@ -43,8 +41,6 @@ do { \ #define __DO_IRQ_SMTC_HOOK() do { } while (0) #endif -#ifdef CONFIG_PREEMPT - /* * do_IRQ handles all normal device IRQ's (the special * SMP cross-CPU interrupts have their own specific @@ -61,8 +57,6 @@ do { \ irq_exit(); \ } while (0) -#endif - extern void arch_init_irq(void); extern void spurious_interrupt(void); -- cgit v1.2.3-70-g09d2 From 005985609ff72df3257fde6b29aa9d71342c2a6b Mon Sep 17 00:00:00 2001 From: Atsushi Nemoto Date: Sun, 12 Nov 2006 00:10:28 +0900 Subject: [MIPS] mips HPT cleanup: make clocksource_mips public Make clocksource_mips public and get rid of mips_hpt_read, mips_hpt_mask. Signed-off-by: Atsushi Nemoto Signed-off-by: Ralf Baechle --- arch/mips/dec/time.c | 4 ++-- arch/mips/jmr3927/rbhma3100/setup.c | 4 ++-- arch/mips/kernel/time.c | 42 ++++++++++++++----------------------- arch/mips/sgi-ip27/ip27-timer.c | 4 ++-- arch/mips/sibyte/bcm1480/time.c | 4 ++-- arch/mips/sibyte/sb1250/time.c | 8 +++---- include/asm-mips/time.h | 8 +++---- 7 files changed, 32 insertions(+), 42 deletions(-) (limited to 'include') diff --git a/arch/mips/dec/time.c b/arch/mips/dec/time.c index 69e424e9ab6..8b7e0c17ac3 100644 --- a/arch/mips/dec/time.c +++ b/arch/mips/dec/time.c @@ -151,7 +151,7 @@ static void dec_timer_ack(void) CMOS_READ(RTC_REG_C); /* Ack the RTC interrupt. */ } -static unsigned int dec_ioasic_hpt_read(void) +static cycle_t dec_ioasic_hpt_read(void) { /* * The free-running counter is 32-bit which is good for about @@ -171,7 +171,7 @@ void __init dec_time_init(void) if (!cpu_has_counter && IOASIC) /* For pre-R4k systems we use the I/O ASIC's counter. */ - mips_hpt_read = dec_ioasic_hpt_read; + clocksource_mips.read = dec_ioasic_hpt_read; /* Set up the rate of periodic DS1287 interrupts. */ CMOS_WRITE(RTC_REF_CLCK_32KHZ | (16 - __ffs(HZ)), RTC_REG_A); diff --git a/arch/mips/jmr3927/rbhma3100/setup.c b/arch/mips/jmr3927/rbhma3100/setup.c index 16e5dfe7aa8..138f25efe38 100644 --- a/arch/mips/jmr3927/rbhma3100/setup.c +++ b/arch/mips/jmr3927/rbhma3100/setup.c @@ -170,7 +170,7 @@ static void jmr3927_machine_power_off(void) while (1); } -static unsigned int jmr3927_hpt_read(void) +static cycle_t jmr3927_hpt_read(void) { /* We assume this function is called xtime_lock held. */ return jiffies * (JMR3927_TIMER_CLK / HZ) + jmr3927_tmrptr->trr; @@ -182,7 +182,7 @@ extern void rtc_ds1742_init(unsigned long base); #endif static void __init jmr3927_time_init(void) { - mips_hpt_read = jmr3927_hpt_read; + clocksource_mips.read = jmr3927_hpt_read; mips_hpt_frequency = JMR3927_TIMER_CLK; #ifdef USE_RTC_DS1742 if (jmr3927_have_nvram()) { diff --git a/arch/mips/kernel/time.c b/arch/mips/kernel/time.c index 111d1baa3b2..11aab6d6bfe 100644 --- a/arch/mips/kernel/time.c +++ b/arch/mips/kernel/time.c @@ -11,7 +11,6 @@ * Free Software Foundation; either version 2 of the License, or (at your * option) any later version. */ -#include #include #include #include @@ -83,7 +82,7 @@ static void null_timer_ack(void) { /* nothing */ } /* * Null high precision timer functions for systems lacking one. */ -static unsigned int null_hpt_read(void) +static cycle_t null_hpt_read(void) { return 0; } @@ -112,7 +111,7 @@ static void c0_timer_ack(void) /* * High precision timer functions for a R4k-compatible timer. */ -static unsigned int c0_hpt_read(void) +static cycle_t c0_hpt_read(void) { return read_c0_count(); } @@ -126,8 +125,6 @@ static void __init c0_hpt_timer_init(void) int (*mips_timer_state)(void); void (*mips_timer_ack)(void); -unsigned int (*mips_hpt_read)(void); -unsigned int mips_hpt_mask = 0xffffffff; /* last time when xtime and rtc are sync'ed up */ static long last_rtc_update; @@ -269,8 +266,7 @@ static struct irqaction timer_irqaction = { static unsigned int __init calibrate_hpt(void) { - u64 frequency; - u32 hpt_start, hpt_end, hpt_count, hz; + cycle_t frequency, hpt_start, hpt_end, hpt_count, hz; const int loops = HZ / 10; int log_2_loops = 0; @@ -296,28 +292,23 @@ static unsigned int __init calibrate_hpt(void) * during the calculated number of periods between timer * interrupts. */ - hpt_start = mips_hpt_read(); + hpt_start = clocksource_mips.read(); do { while (mips_timer_state()); while (!mips_timer_state()); } while (--i); - hpt_end = mips_hpt_read(); + hpt_end = clocksource_mips.read(); - hpt_count = (hpt_end - hpt_start) & mips_hpt_mask; + hpt_count = (hpt_end - hpt_start) & clocksource_mips.mask; hz = HZ; - frequency = (u64)hpt_count * (u64)hz; + frequency = hpt_count * hz; return frequency >> log_2_loops; } -static cycle_t read_mips_hpt(void) -{ - return (cycle_t)mips_hpt_read(); -} - -static struct clocksource clocksource_mips = { +struct clocksource clocksource_mips = { .name = "MIPS", - .read = read_mips_hpt, + .mask = 0xffffffff, .is_continuous = 1, }; @@ -326,7 +317,7 @@ static void __init init_mips_clocksource(void) u64 temp; u32 shift; - if (!mips_hpt_frequency || mips_hpt_read == null_hpt_read) + if (!mips_hpt_frequency || clocksource_mips.read == null_hpt_read) return; /* Calclate a somewhat reasonable rating value */ @@ -340,7 +331,6 @@ static void __init init_mips_clocksource(void) } clocksource_mips.shift = shift; clocksource_mips.mult = (u32)temp; - clocksource_mips.mask = mips_hpt_mask; clocksource_register(&clocksource_mips); } @@ -360,19 +350,19 @@ void __init time_init(void) -xtime.tv_sec, -xtime.tv_nsec); /* Choose appropriate high precision timer routines. */ - if (!cpu_has_counter && !mips_hpt_read) + if (!cpu_has_counter && !clocksource_mips.read) /* No high precision timer -- sorry. */ - mips_hpt_read = null_hpt_read; + clocksource_mips.read = null_hpt_read; else if (!mips_hpt_frequency && !mips_timer_state) { /* A high precision timer of unknown frequency. */ - if (!mips_hpt_read) + if (!clocksource_mips.read) /* No external high precision timer -- use R4k. */ - mips_hpt_read = c0_hpt_read; + clocksource_mips.read = c0_hpt_read; } else { /* We know counter frequency. Or we can get it. */ - if (!mips_hpt_read) { + if (!clocksource_mips.read) { /* No external high precision timer -- use R4k. */ - mips_hpt_read = c0_hpt_read; + clocksource_mips.read = c0_hpt_read; if (!mips_timer_state) { /* No external timer interrupt -- use R4k. */ diff --git a/arch/mips/sgi-ip27/ip27-timer.c b/arch/mips/sgi-ip27/ip27-timer.c index e5441c3a0b0..7d361726bbf 100644 --- a/arch/mips/sgi-ip27/ip27-timer.c +++ b/arch/mips/sgi-ip27/ip27-timer.c @@ -223,14 +223,14 @@ void __init plat_timer_setup(struct irqaction *irq) setup_irq(irqno, &rt_irqaction); } -static unsigned int ip27_hpt_read(void) +static cycle_t ip27_hpt_read(void) { return REMOTE_HUB_L(cputonasid(0), PI_RT_COUNT); } void __init ip27_time_init(void) { - mips_hpt_read = ip27_hpt_read; + clocksource_mips.read = ip27_hpt_read; mips_hpt_frequency = CYCLES_PER_SEC; xtime.tv_sec = get_m48t35_time(); xtime.tv_nsec = 0; diff --git a/arch/mips/sibyte/bcm1480/time.c b/arch/mips/sibyte/bcm1480/time.c index 03d5822d3a7..6f3f71bf424 100644 --- a/arch/mips/sibyte/bcm1480/time.c +++ b/arch/mips/sibyte/bcm1480/time.c @@ -117,7 +117,7 @@ void bcm1480_timer_interrupt(void) } } -static unsigned int bcm1480_hpt_read(void) +static cycle_t bcm1480_hpt_read(void) { /* We assume this function is called xtime_lock held. */ unsigned long count = @@ -127,6 +127,6 @@ static unsigned int bcm1480_hpt_read(void) void __init bcm1480_hpt_setup(void) { - mips_hpt_read = bcm1480_hpt_read; + clocksource_mips.read = bcm1480_hpt_read; mips_hpt_frequency = BCM1480_HPT_VALUE; } diff --git a/arch/mips/sibyte/sb1250/time.c b/arch/mips/sibyte/sb1250/time.c index bcb74f2c194..2efffe15ff2 100644 --- a/arch/mips/sibyte/sb1250/time.c +++ b/arch/mips/sibyte/sb1250/time.c @@ -51,7 +51,7 @@ extern int sb1250_steal_irq(int irq); -static unsigned int sb1250_hpt_read(void); +static cycle_t sb1250_hpt_read(void); void __init sb1250_hpt_setup(void) { @@ -66,8 +66,8 @@ void __init sb1250_hpt_setup(void) IOADDR(A_SCD_TIMER_REGISTER(SB1250_HPT_NUM, R_SCD_TIMER_CFG))); mips_hpt_frequency = V_SCD_TIMER_FREQ; - mips_hpt_read = sb1250_hpt_read; - mips_hpt_mask = M_SCD_TIMER_INIT; + clocksource_mips.read = sb1250_hpt_read; + clocksource_mips.mask = M_SCD_TIMER_INIT; } } @@ -143,7 +143,7 @@ void sb1250_timer_interrupt(void) * The HPT is free running from SB1250_HPT_VALUE down to 0 then starts over * again. */ -static unsigned int sb1250_hpt_read(void) +static cycle_t sb1250_hpt_read(void) { unsigned int count; diff --git a/include/asm-mips/time.h b/include/asm-mips/time.h index b58665efb1d..a632cef830a 100644 --- a/include/asm-mips/time.h +++ b/include/asm-mips/time.h @@ -21,6 +21,7 @@ #include #include #include +#include extern spinlock_t rtc_lock; @@ -44,11 +45,10 @@ extern int (*mips_timer_state)(void); extern void (*mips_timer_ack)(void); /* - * High precision timer functions. - * If mips_hpt_read is NULL, an R4k-compatible timer setup is attempted. + * High precision timer clocksource. + * If .read is NULL, an R4k-compatible timer setup is attempted. */ -extern unsigned int (*mips_hpt_read)(void); -extern unsigned int mips_hpt_mask; +extern struct clocksource clocksource_mips; /* * to_tm() converts system time back to (year, mon, day, hour, min, sec). -- cgit v1.2.3-70-g09d2 From 63dc68a8cf60cb110b147dab1704d990808b39e2 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 16 Oct 2006 01:38:50 +0100 Subject: [MIPS] Use conditional traps for BUG_ON on MIPS II and better. This shaves of around 4kB and a few cycles for the average kernel that has CONFIG_BUG enabled. Signed-off-by: Ralf Baechle --- arch/mips/kernel/traps.c | 14 ++++++++++---- include/asm-mips/bug.h | 12 ++++++++++++ include/asm-mips/ptrace.h | 2 -- 3 files changed, 22 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c index 0ae19e1fa86..2a932cada24 100644 --- a/arch/mips/kernel/traps.c +++ b/arch/mips/kernel/traps.c @@ -669,8 +669,6 @@ asmlinkage void do_bp(struct pt_regs *regs) unsigned int opcode, bcode; siginfo_t info; - die_if_kernel("Break instruction in kernel code", regs); - if (get_user(opcode, (unsigned int __user *) exception_epc(regs))) goto out_sigsegv; @@ -693,6 +691,7 @@ asmlinkage void do_bp(struct pt_regs *regs) switch (bcode) { case BRK_OVERFLOW << 10: case BRK_DIVZERO << 10: + die_if_kernel("Break instruction in kernel code", regs); if (bcode == (BRK_DIVZERO << 10)) info.si_code = FPE_INTDIV; else @@ -702,7 +701,11 @@ asmlinkage void do_bp(struct pt_regs *regs) info.si_addr = (void __user *) regs->cp0_epc; force_sig_info(SIGFPE, &info, current); break; + case BRK_BUG: + die("Kernel bug detected", regs); + break; default: + die_if_kernel("Break instruction in kernel code", regs); force_sig(SIGTRAP, current); } @@ -715,8 +718,6 @@ asmlinkage void do_tr(struct pt_regs *regs) unsigned int opcode, tcode = 0; siginfo_t info; - die_if_kernel("Trap instruction in kernel code", regs); - if (get_user(opcode, (unsigned int __user *) exception_epc(regs))) goto out_sigsegv; @@ -733,6 +734,7 @@ asmlinkage void do_tr(struct pt_regs *regs) switch (tcode) { case BRK_OVERFLOW: case BRK_DIVZERO: + die_if_kernel("Trap instruction in kernel code", regs); if (tcode == BRK_DIVZERO) info.si_code = FPE_INTDIV; else @@ -742,7 +744,11 @@ asmlinkage void do_tr(struct pt_regs *regs) info.si_addr = (void __user *) regs->cp0_epc; force_sig_info(SIGFPE, &info, current); break; + case BRK_BUG: + die("Kernel bug detected", regs); + break; default: + die_if_kernel("Trap instruction in kernel code", regs); force_sig(SIGTRAP, current); } diff --git a/include/asm-mips/bug.h b/include/asm-mips/bug.h index 7b4739dc8f3..4d560a53394 100644 --- a/include/asm-mips/bug.h +++ b/include/asm-mips/bug.h @@ -1,6 +1,7 @@ #ifndef __ASM_BUG_H #define __ASM_BUG_H +#include #ifdef CONFIG_BUG @@ -13,6 +14,17 @@ do { \ #define HAVE_ARCH_BUG +#if (_MIPS_ISA > _MIPS_ISA_MIPS1) + +#define BUG_ON(condition) \ +do { \ + __asm__ __volatile__("tne $0, %0" : : "r" (condition)); \ +} while (0) + +#define HAVE_ARCH_BUG_ON + +#endif /* _MIPS_ISA > _MIPS_ISA_MIPS1 */ + #endif #include diff --git a/include/asm-mips/ptrace.h b/include/asm-mips/ptrace.h index 5f3a9075cd2..30bf555faea 100644 --- a/include/asm-mips/ptrace.h +++ b/include/asm-mips/ptrace.h @@ -80,8 +80,6 @@ struct pt_regs { #define instruction_pointer(regs) ((regs)->cp0_epc) #define profile_pc(regs) instruction_pointer(regs) -extern void show_regs(struct pt_regs *); - extern asmlinkage void do_syscall_trace(struct pt_regs *regs, int entryexit); #endif -- cgit v1.2.3-70-g09d2 From e303e088f25dc7d8bafc0d1942314214a3a57b44 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 30 Nov 2006 01:14:50 +0000 Subject: [MIPS] Remove old junk left from old atomic_lock. Signed-off-by: Ralf Baechle --- include/asm-mips/atomic.h | 10 +--------- 1 file changed, 1 insertion(+), 9 deletions(-) (limited to 'include') diff --git a/include/asm-mips/atomic.h b/include/asm-mips/atomic.h index e64abc0d822..7978d8e1164 100644 --- a/include/asm-mips/atomic.h +++ b/include/asm-mips/atomic.h @@ -9,16 +9,8 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (C) 1996, 97, 99, 2000, 03, 04 by Ralf Baechle + * Copyright (C) 1996, 97, 99, 2000, 03, 04, 06 by Ralf Baechle */ - -/* - * As workaround for the ATOMIC_DEC_AND_LOCK / atomic_dec_and_lock mess in - * we have to include outside the - * main big wrapper ... - */ -#include - #ifndef _ASM_ATOMIC_H #define _ASM_ATOMIC_H -- cgit v1.2.3-70-g09d2 From 4ffd8b3838f22c34b21a25b7612795ca45d14db6 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 30 Nov 2006 01:14:50 +0000 Subject: [MIPS] Remove userspace proofing from . Signed-off-by: Ralf Baechle --- include/asm-mips/bitops.h | 63 ++++++++++++++++------------------------------- 1 file changed, 21 insertions(+), 42 deletions(-) (limited to 'include') diff --git a/include/asm-mips/bitops.h b/include/asm-mips/bitops.h index 1bb89c5a10e..8cd61c19705 100644 --- a/include/asm-mips/bitops.h +++ b/include/asm-mips/bitops.h @@ -10,10 +10,13 @@ #define _ASM_BITOPS_H #include +#include #include #include #include /* sigh ... */ #include +#include +#include #if (_MIPS_SZLONG == 32) #define SZLONG_LOG 5 @@ -29,32 +32,12 @@ #define cpu_to_lelongp(x) cpu_to_le64p((__u64 *) (x)) #endif -#ifdef __KERNEL__ - -#include -#include -#include - /* * clear_bit() doesn't provide any barrier for the compiler. */ #define smp_mb__before_clear_bit() smp_mb() #define smp_mb__after_clear_bit() smp_mb() -/* - * Only disable interrupt for kernel mode stuff to keep usermode stuff - * that dares to use kernel include files alive. - */ - -#define __bi_flags unsigned long flags -#define __bi_local_irq_save(x) local_irq_save(x) -#define __bi_local_irq_restore(x) local_irq_restore(x) -#else -#define __bi_flags -#define __bi_local_irq_save(x) -#define __bi_local_irq_restore(x) -#endif /* __KERNEL__ */ - /* * set_bit - Atomically set a bit in memory * @nr: the bit to set @@ -93,13 +76,13 @@ static inline void set_bit(unsigned long nr, volatile unsigned long *addr) } else { volatile unsigned long *a = addr; unsigned long mask; - __bi_flags; + unsigned long flags; a += nr >> SZLONG_LOG; mask = 1UL << (nr & SZLONG_MASK); - __bi_local_irq_save(flags); + local_irq_save(flags); *a |= mask; - __bi_local_irq_restore(flags); + local_irq_restore(flags); } } @@ -141,13 +124,13 @@ static inline void clear_bit(unsigned long nr, volatile unsigned long *addr) } else { volatile unsigned long *a = addr; unsigned long mask; - __bi_flags; + unsigned long flags; a += nr >> SZLONG_LOG; mask = 1UL << (nr & SZLONG_MASK); - __bi_local_irq_save(flags); + local_irq_save(flags); *a &= ~mask; - __bi_local_irq_restore(flags); + local_irq_restore(flags); } } @@ -191,13 +174,13 @@ static inline void change_bit(unsigned long nr, volatile unsigned long *addr) } else { volatile unsigned long *a = addr; unsigned long mask; - __bi_flags; + unsigned long flags; a += nr >> SZLONG_LOG; mask = 1UL << (nr & SZLONG_MASK); - __bi_local_irq_save(flags); + local_irq_save(flags); *a ^= mask; - __bi_local_irq_restore(flags); + local_irq_restore(flags); } } @@ -258,14 +241,14 @@ static inline int test_and_set_bit(unsigned long nr, volatile unsigned long *a = addr; unsigned long mask; int retval; - __bi_flags; + unsigned long flags; a += nr >> SZLONG_LOG; mask = 1UL << (nr & SZLONG_MASK); - __bi_local_irq_save(flags); + local_irq_save(flags); retval = (mask & *a) != 0; *a |= mask; - __bi_local_irq_restore(flags); + local_irq_restore(flags); return retval; } @@ -330,14 +313,14 @@ static inline int test_and_clear_bit(unsigned long nr, volatile unsigned long *a = addr; unsigned long mask; int retval; - __bi_flags; + unsigned long flags; a += nr >> SZLONG_LOG; mask = 1UL << (nr & SZLONG_MASK); - __bi_local_irq_save(flags); + local_irq_save(flags); retval = (mask & *a) != 0; *a &= ~mask; - __bi_local_irq_restore(flags); + local_irq_restore(flags); return retval; } @@ -399,23 +382,19 @@ static inline int test_and_change_bit(unsigned long nr, } else { volatile unsigned long *a = addr; unsigned long mask, retval; - __bi_flags; + unsigned long flags; a += nr >> SZLONG_LOG; mask = 1UL << (nr & SZLONG_MASK); - __bi_local_irq_save(flags); + local_irq_save(flags); retval = (mask & *a) != 0; *a ^= mask; - __bi_local_irq_restore(flags); + local_irq_restore(flags); return retval; } } -#undef __bi_flags -#undef __bi_local_irq_save -#undef __bi_local_irq_restore - #include /* -- cgit v1.2.3-70-g09d2 From 0b7883f49810ec91755caa222b3b28f047b8c93b Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 30 Nov 2006 01:14:51 +0000 Subject: [MIPS] Remove unused definition of cpu_to_lelongp() Signed-off-by: Ralf Baechle --- include/asm-mips/bitops.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include') diff --git a/include/asm-mips/bitops.h b/include/asm-mips/bitops.h index 8cd61c19705..b9007411b60 100644 --- a/include/asm-mips/bitops.h +++ b/include/asm-mips/bitops.h @@ -23,13 +23,11 @@ #define SZLONG_MASK 31UL #define __LL "ll " #define __SC "sc " -#define cpu_to_lelongp(x) cpu_to_le32p((__u32 *) (x)) #elif (_MIPS_SZLONG == 64) #define SZLONG_LOG 6 #define SZLONG_MASK 63UL #define __LL "lld " #define __SC "scd " -#define cpu_to_lelongp(x) cpu_to_le64p((__u64 *) (x)) #endif /* -- cgit v1.2.3-70-g09d2 From aa414dff4f7bef29457592414551becdca72dd6b Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 30 Nov 2006 01:14:51 +0000 Subject: [MIPS] Remove duplicate ISA DMA code for 0 DMA channel case. Signed-off-by: Ralf Baechle --- arch/mips/Kconfig | 11 ++++++----- arch/mips/kernel/Makefile | 1 - arch/mips/kernel/dma-no-isa.c | 28 ---------------------------- include/asm-mips/dma.h | 2 ++ 4 files changed, 8 insertions(+), 34 deletions(-) delete mode 100644 arch/mips/kernel/dma-no-isa.c (limited to 'include') diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index c0da0ffe8d5..96143670759 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -266,8 +266,8 @@ config MIPS_MALTA select BOOT_ELF32 select HAVE_STD_PC_SERIAL_PORT select DMA_NONCOHERENT - select IRQ_CPU select GENERIC_ISA_DMA + select IRQ_CPU select HW_HAS_PCI select I8259 select MIPS_BOARDS_GEN @@ -534,7 +534,7 @@ config SGI_IP22 select HW_HAS_EISA select IP22_CPU_SCACHE select IRQ_CPU - select NO_ISA if ISA + select GENERIC_ISA_DMA_SUPPORT_BROKEN select SWAP_IO_SPACE select SYS_HAS_CPU_R4X00 select SYS_HAS_CPU_R5000 @@ -881,6 +881,9 @@ config MIPS_NILE4 config MIPS_DISABLE_OBSOLETE_IDE bool +config GENERIC_ISA_DMA_SUPPORT_BROKEN + bool + # # Endianess selection. Suffiently obscure so many users don't know what to # answer,so we try hard to limit the available choices. Also the use of a @@ -1852,13 +1855,11 @@ source "drivers/pci/Kconfig" config ISA bool -config NO_ISA - bool - config EISA bool "EISA support" depends on HW_HAS_EISA select ISA + select GENERIC_ISA_DMA ---help--- The Extended Industry Standard Architecture (EISA) bus was developed as an open alternative to the IBM MicroChannel bus. diff --git a/arch/mips/kernel/Makefile b/arch/mips/kernel/Makefile index f35b739d0a1..bbbb8d7cb89 100644 --- a/arch/mips/kernel/Makefile +++ b/arch/mips/kernel/Makefile @@ -45,7 +45,6 @@ obj-$(CONFIG_MIPS_APSP_KSPD) += kspd.o obj-$(CONFIG_MIPS_VPE_LOADER) += vpe.o obj-$(CONFIG_MIPS_VPE_APSP_API) += rtlx.o -obj-$(CONFIG_NO_ISA) += dma-no-isa.o obj-$(CONFIG_I8259) += i8259.o obj-$(CONFIG_IRQ_CPU) += irq_cpu.o obj-$(CONFIG_IRQ_CPU_RM7K) += irq-rm7000.o diff --git a/arch/mips/kernel/dma-no-isa.c b/arch/mips/kernel/dma-no-isa.c deleted file mode 100644 index 6df8b07741e..00000000000 --- a/arch/mips/kernel/dma-no-isa.c +++ /dev/null @@ -1,28 +0,0 @@ -/* - * This file is subject to the terms and conditions of the GNU General Public - * License. See the file "COPYING" in the main directory of this archive - * for more details. - * - * Copyright (C) 2004 by Ralf Baechle - * - * Dummy ISA DMA functions for systems that don't have ISA but share drivers - * with ISA such as legacy free PCI. - */ -#include -#include -#include - -DEFINE_SPINLOCK(dma_spin_lock); - -int request_dma(unsigned int dmanr, const char * device_id) -{ - return -EINVAL; -} - -void free_dma(unsigned int dmanr) -{ -} - -EXPORT_SYMBOL(dma_spin_lock); -EXPORT_SYMBOL(request_dma); -EXPORT_SYMBOL(free_dma); diff --git a/include/asm-mips/dma.h b/include/asm-mips/dma.h index e85849ac165..23f789c8084 100644 --- a/include/asm-mips/dma.h +++ b/include/asm-mips/dma.h @@ -74,7 +74,9 @@ * */ +#ifndef GENERIC_ISA_DMA_SUPPORT_BROKEN #define MAX_DMA_CHANNELS 8 +#endif /* * The maximum address in KSEG0 that we can perform a DMA transfer to on this -- cgit v1.2.3-70-g09d2 From 139a7bdc2b9391a4d0362190d9e5625dcf580105 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 30 Nov 2006 04:40:22 +0100 Subject: mqueue.h: don't include linux/types.h This #include is not required. Signed-off-by: Alexey Dobriyan Signed-off-by: Adrian Bunk --- include/linux/mqueue.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include') diff --git a/include/linux/mqueue.h b/include/linux/mqueue.h index 8db9d75541a..8b5a79615fb 100644 --- a/include/linux/mqueue.h +++ b/include/linux/mqueue.h @@ -18,8 +18,6 @@ #ifndef _LINUX_MQUEUE_H #define _LINUX_MQUEUE_H -#include - #define MQ_PRIO_MAX 32768 /* per-uid limit of kernel memory used by mqueue, in bytes */ #define MQ_BYTES_MAX 819200 -- cgit v1.2.3-70-g09d2 From e20ec9911bfef897459b9f8aeaf6eadb0920299a Mon Sep 17 00:00:00 2001 From: Jim Cromie Date: Thu, 30 Nov 2006 04:46:13 +0100 Subject: fix spelling error in include/linux/kernel.h Signed-off-by: Adrian Bunk --- include/linux/kernel.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index b9b5e4ba166..6738283ac38 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -65,7 +65,7 @@ struct user; * context (spinlock, irq-handler, ...). * * This is a useful debugging help to be able to catch problems early and not - * be biten later when the calling function happens to sleep when it is not + * be bitten later when the calling function happens to sleep when it is not * supposed to. */ #ifdef CONFIG_PREEMPT_VOLUNTARY -- cgit v1.2.3-70-g09d2 From 98c4f0c336afe4318c12397bc74910d86ee036a2 Mon Sep 17 00:00:00 2001 From: Chase Venters Date: Thu, 30 Nov 2006 04:53:49 +0100 Subject: Fix jiffies.h comment jiffies.h includes a comment informing that jiffies_64 must be read with the assistance of the xtime_lock seqlock. The comment text, however, calls jiffies_64 "not volatile", which should probably read "not atomic". Signed-off-by: Chase Venters Signed-off-by: Adrian Bunk --- include/linux/jiffies.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/jiffies.h b/include/linux/jiffies.h index c8d5f207c3d..0ec6e28bccd 100644 --- a/include/linux/jiffies.h +++ b/include/linux/jiffies.h @@ -74,7 +74,7 @@ #define __jiffy_data __attribute__((section(".data"))) /* - * The 64-bit value is not volatile - you MUST NOT read it + * The 64-bit value is not atomic - you MUST NOT read it * without sampling the sequence number in xtime_lock. * get_jiffies_64() will do this for you as appropriate. */ -- cgit v1.2.3-70-g09d2 From 0779bf2d2ecc4d9b1e9437ae659f50e6776a7666 Mon Sep 17 00:00:00 2001 From: Matt LaPlante Date: Thu, 30 Nov 2006 05:24:39 +0100 Subject: Fix misc .c/.h comment typos Fix various .c/.h typos in comments (no code changes). Signed-off-by: Matt LaPlante Signed-off-by: Adrian Bunk --- arch/cris/arch-v10/drivers/eeprom.c | 6 +++--- arch/cris/arch-v10/drivers/i2c.c | 2 +- arch/cris/arch-v10/kernel/kgdb.c | 2 +- arch/ia64/hp/common/sba_iommu.c | 8 ++++---- arch/sh64/lib/dbg.c | 2 +- drivers/atm/iphase.c | 2 +- drivers/char/rio/riocmd.c | 2 +- drivers/char/rio/rioinit.c | 2 +- drivers/char/rio/rioparam.c | 6 +++--- drivers/ide/ide-floppy.c | 2 +- drivers/isdn/hardware/eicon/os_4bri.c | 2 +- drivers/isdn/hisax/hfc4s8s_l1.h | 2 +- drivers/media/dvb/ttpci/budget-patch.c | 8 ++++---- drivers/net/e100.c | 2 +- drivers/net/e1000/e1000_hw.c | 2 +- drivers/net/sk98lin/h/skdrv2nd.h | 2 +- drivers/net/sk98lin/skdim.c | 4 ++-- drivers/net/wireless/ipw2200.c | 4 ++-- drivers/parisc/ccio-dma.c | 2 +- drivers/parisc/iosapic.c | 6 +++--- drivers/pci/hotplug/ibmphp_hpc.c | 2 +- drivers/s390/net/claw.h | 2 +- drivers/scsi/aic94xx/aic94xx_reg_def.h | 2 +- drivers/scsi/aic94xx/aic94xx_sds.c | 4 ++-- drivers/scsi/ncr53c8xx.c | 14 +++++++------- drivers/scsi/ncr53c8xx.h | 6 +++--- drivers/usb/host/u132-hcd.c | 6 +++--- drivers/usb/misc/usb_u132.h | 2 +- drivers/usb/serial/digi_acceleport.c | 2 +- fs/reiserfs/journal.c | 6 +++--- include/asm-m68knommu/mcfmbus.h | 2 +- include/asm-parisc/dma.h | 6 +++--- include/asm-parisc/pci.h | 2 +- include/asm-parisc/ropes.h | 2 +- include/linux/ixjuser.h | 2 +- include/linux/reiserfs_fs_sb.h | 2 +- net/wanrouter/af_wanpipe.c | 4 ++-- net/wanrouter/wanmain.c | 2 +- sound/oss/cs46xx.c | 6 +++--- 39 files changed, 71 insertions(+), 71 deletions(-) (limited to 'include') diff --git a/arch/cris/arch-v10/drivers/eeprom.c b/arch/cris/arch-v10/drivers/eeprom.c index 6e1f191a71e..284ebfda03f 100644 --- a/arch/cris/arch-v10/drivers/eeprom.c +++ b/arch/cris/arch-v10/drivers/eeprom.c @@ -1,7 +1,7 @@ /*!***************************************************************************** *! -*! Implements an interface for i2c compatible eeproms to run under linux. -*! Supports 2k, 8k(?) and 16k. Uses adaptive timing adjustents by +*! Implements an interface for i2c compatible eeproms to run under Linux. +*! Supports 2k, 8k(?) and 16k. Uses adaptive timing adjustments by *! Johan.Adolfsson@axis.com *! *! Probing results: @@ -51,7 +51,7 @@ *! Revision 1.8 2001/06/15 13:24:29 jonashg *! * Added verification of pointers from userspace in read and write. *! * Made busy counter volatile. -*! * Added define for inital write delay. +*! * Added define for initial write delay. *! * Removed warnings by using loff_t instead of unsigned long. *! *! Revision 1.7 2001/06/14 15:26:54 jonashg diff --git a/arch/cris/arch-v10/drivers/i2c.c b/arch/cris/arch-v10/drivers/i2c.c index 6114596c3b3..092c724a645 100644 --- a/arch/cris/arch-v10/drivers/i2c.c +++ b/arch/cris/arch-v10/drivers/i2c.c @@ -47,7 +47,7 @@ *! Update Port B register and shadow even when running with hardware support *! to avoid glitches when reading bits *! Never set direction to out in i2c_inbyte -*! Removed incorrect clock togling at end of i2c_inbyte +*! Removed incorrect clock toggling at end of i2c_inbyte *! *! Revision 1.8 2002/08/13 06:31:53 starvik *! Made SDA and SCL line configurable diff --git a/arch/cris/arch-v10/kernel/kgdb.c b/arch/cris/arch-v10/kernel/kgdb.c index 34528da9881..07628a13c6c 100644 --- a/arch/cris/arch-v10/kernel/kgdb.c +++ b/arch/cris/arch-v10/kernel/kgdb.c @@ -33,7 +33,7 @@ *! *! Revision 1.2 2002/11/19 14:35:24 starvik *! Changes from linux 2.4 -*! Changed struct initializer syntax to the currently prefered notation +*! Changed struct initializer syntax to the currently preferred notation *! *! Revision 1.1 2001/12/17 13:59:27 bjornw *! Initial revision diff --git a/arch/ia64/hp/common/sba_iommu.c b/arch/ia64/hp/common/sba_iommu.c index db8e1fcfa04..14691cda05c 100644 --- a/arch/ia64/hp/common/sba_iommu.c +++ b/arch/ia64/hp/common/sba_iommu.c @@ -75,7 +75,7 @@ ** If a device prefetches beyond the end of a valid pdir entry, it will cause ** a hard failure, ie. MCA. Version 3.0 and later of the zx1 LBA should ** disconnect on 4k boundaries and prevent such issues. If the device is -** particularly agressive, this option will keep the entire pdir valid such +** particularly aggressive, this option will keep the entire pdir valid such ** that prefetching will hit a valid address. This could severely impact ** error containment, and is therefore off by default. The page that is ** used for spill-over is poisoned, so that should help debugging somewhat. @@ -258,10 +258,10 @@ static u64 prefetch_spill_page; /* ** DMA_CHUNK_SIZE is used by the SCSI mid-layer to break up -** (or rather not merge) DMA's into managable chunks. +** (or rather not merge) DMAs into manageable chunks. ** On parisc, this is more of the software/tuning constraint -** rather than the HW. I/O MMU allocation alogorithms can be -** faster with smaller size is (to some degree). +** rather than the HW. I/O MMU allocation algorithms can be +** faster with smaller sizes (to some degree). */ #define DMA_CHUNK_SIZE (BITS_PER_LONG*iovp_size) diff --git a/arch/sh64/lib/dbg.c b/arch/sh64/lib/dbg.c index 1326f45f31e..4310fc87444 100644 --- a/arch/sh64/lib/dbg.c +++ b/arch/sh64/lib/dbg.c @@ -383,7 +383,7 @@ void show_excp_regs(char *from, int trapnr, int signr, struct pt_regs *regs) /* ======================================================================= */ /* -** Depending on scan the MMU, Data or Instrction side +** Depending on scan the MMU, Data or Instruction side ** looking for a valid mapping matching Eaddr & asid. ** Return -1 if not found or the TLB id entry otherwise. ** Note: it works only for 4k pages! diff --git a/drivers/atm/iphase.c b/drivers/atm/iphase.c index 9ed1c60048f..bb7ef570514 100644 --- a/drivers/atm/iphase.c +++ b/drivers/atm/iphase.c @@ -305,7 +305,7 @@ static void clear_lockup (struct atm_vcc *vcc, IADEV *dev) { ** | R | NZ | 5-bit exponent | 9-bit mantissa | ** +----+----+------------------+-------------------------------+ ** -** R = reserverd (written as 0) +** R = reserved (written as 0) ** NZ = 0 if 0 cells/sec; 1 otherwise ** ** if NZ = 1, rate = 1.mmmmmmmmm x 2^(eeeee) cells/sec diff --git a/drivers/char/rio/riocmd.c b/drivers/char/rio/riocmd.c index 4df6ab2206a..167ebc84e8d 100644 --- a/drivers/char/rio/riocmd.c +++ b/drivers/char/rio/riocmd.c @@ -922,7 +922,7 @@ int RIOUnUse(unsigned long iPortP, struct CmdBlk *CmdBlkP) ** ** Packet is an actual packet structure to be filled in with the packet ** information associated with the command. You need to fill in everything, -** as the command processore doesn't process the command packet in any way. +** as the command processor doesn't process the command packet in any way. ** ** The PreFuncP is called before the packet is enqueued on the host rup. ** PreFuncP is called as (*PreFuncP)(PreArg, CmdBlkP);. PreFuncP must diff --git a/drivers/char/rio/rioinit.c b/drivers/char/rio/rioinit.c index 99f3df02b61..0794844369d 100644 --- a/drivers/char/rio/rioinit.c +++ b/drivers/char/rio/rioinit.c @@ -222,7 +222,7 @@ int RIOBoardTest(unsigned long paddr, void __iomem *caddr, unsigned char type, i ** which value will be written into memory. ** Call with op set to zero means that the RAM will not be read and checked ** before it is written. -** Call with op not zero, and the RAM will be read and compated with val[op-1] +** Call with op not zero and the RAM will be read and compared with val[op-1] ** to check that the data from the previous phase was retained. */ diff --git a/drivers/char/rio/rioparam.c b/drivers/char/rio/rioparam.c index 1066d976070..bb498d24adc 100644 --- a/drivers/char/rio/rioparam.c +++ b/drivers/char/rio/rioparam.c @@ -87,8 +87,8 @@ static char *_rioparam_c_sccs_ = "@(#)rioparam.c 1.3"; ** command bit set onto the port. The command bit is in the len field, ** and gets ORed in with the actual byte count. ** -** When you send a packet with the command bit set, then the first -** data byte ( data[0] ) is interpretted as the command to execute. +** When you send a packet with the command bit set the first +** data byte (data[0]) is interpreted as the command to execute. ** It also governs what data structure overlay should accompany the packet. ** Commands are defined in cirrus/cirrus.h ** @@ -103,7 +103,7 @@ static char *_rioparam_c_sccs_ = "@(#)rioparam.c 1.3"; ** ** Most commands do not use the remaining bytes in the data array. The ** exceptions are OPEN MOPEN and CONFIG. (NB. As with the SI CONFIG and -** OPEN are currently analagous). With these three commands the following +** OPEN are currently analogous). With these three commands the following ** 11 data bytes are all used to pass config information such as baud rate etc. ** The fields are also defined in cirrus.h. Some contain straightforward ** information such as the transmit XON character. Two contain the transmit and diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c index 8ccee9c769f..e3a267622bb 100644 --- a/drivers/ide/ide-floppy.c +++ b/drivers/ide/ide-floppy.c @@ -1635,7 +1635,7 @@ static int idefloppy_begin_format(ide_drive_t *drive, int __user *arg) /* ** Get ATAPI_FORMAT_UNIT progress indication. ** -** Userland gives a pointer to an int. The int is set to a progresss +** Userland gives a pointer to an int. The int is set to a progress ** indicator 0-65536, with 65536=100%. ** ** If the drive does not support format progress indication, we just check diff --git a/drivers/isdn/hardware/eicon/os_4bri.c b/drivers/isdn/hardware/eicon/os_4bri.c index 11e6f937c1e..7b4ec3f60db 100644 --- a/drivers/isdn/hardware/eicon/os_4bri.c +++ b/drivers/isdn/hardware/eicon/os_4bri.c @@ -464,7 +464,7 @@ int diva_4bri_init_card(diva_os_xdi_adapter_t * a) /* ** Cleanup function will be called for master adapter only -** this is garanteed by design: cleanup callback is set +** this is guaranteed by design: cleanup callback is set ** by master adapter only */ static int diva_4bri_cleanup_adapter(diva_os_xdi_adapter_t * a) diff --git a/drivers/isdn/hisax/hfc4s8s_l1.h b/drivers/isdn/hisax/hfc4s8s_l1.h index e8f9c077fa8..9d5d2a56b4e 100644 --- a/drivers/isdn/hisax/hfc4s8s_l1.h +++ b/drivers/isdn/hisax/hfc4s8s_l1.h @@ -16,7 +16,7 @@ /* * include Genero generated HFC-4S/8S header file hfc48scu.h -* for comlete register description. This will define _HFC48SCU_H_ +* for complete register description. This will define _HFC48SCU_H_ * to prevent redefinitions */ diff --git a/drivers/media/dvb/ttpci/budget-patch.c b/drivers/media/dvb/ttpci/budget-patch.c index fc1267b8c89..9a155396d6a 100644 --- a/drivers/media/dvb/ttpci/budget-patch.c +++ b/drivers/media/dvb/ttpci/budget-patch.c @@ -500,14 +500,14 @@ static int budget_patch_attach (struct saa7146_dev* dev, struct saa7146_pci_exte /* New design (By Emard) ** this rps1 code will copy internal HS event to GPIO3 pin. -** GPIO3 is in budget-patch hardware connectd to port B VSYNC +** GPIO3 is in budget-patch hardware connected to port B VSYNC ** HS is an internal event of 7146, accessible with RPS ** and temporarily raised high every n lines ** (n in defined in the RPS_THRESH1 counter threshold) ** I think HS is raised high on the beginning of the n-th line ** and remains high until this n-th line that triggered -** it is completely received. When the receiption of n-th line +** it is completely received. When the reception of n-th line ** ends, HS is lowered. ** To transmit data over DMA, 7146 needs changing state at @@ -541,7 +541,7 @@ static int budget_patch_attach (struct saa7146_dev* dev, struct saa7146_pci_exte ** hardware debug note: a working budget card (including budget patch) ** with vpeirq() interrupt setup in mode "0x90" (every 64K) will ** generate 3 interrupts per 25-Hz DMA frame of 2*188*512 bytes -** and that means 3*25=75 Hz of interrupt freqency, as seen by +** and that means 3*25=75 Hz of interrupt frequency, as seen by ** watch cat /proc/interrupts ** ** If this frequency is 3x lower (and data received in the DMA @@ -550,7 +550,7 @@ static int budget_patch_attach (struct saa7146_dev* dev, struct saa7146_pci_exte ** this means VSYNC line is not connected in the hardware. ** (check soldering pcb and pins) ** The same behaviour of missing VSYNC can be duplicated on budget -** cards, by seting DD1_INIT trigger mode 7 in 3rd nibble. +** cards, by setting DD1_INIT trigger mode 7 in 3rd nibble. */ // Setup RPS1 "program" (p35) diff --git a/drivers/net/e100.c b/drivers/net/e100.c index 19ab3441269..3a8df479cbd 100644 --- a/drivers/net/e100.c +++ b/drivers/net/e100.c @@ -1215,7 +1215,7 @@ static void e100_setup_ucode(struct nic *nic, struct cb *cb, struct sk_buff *skb * the literal in the instruction before the code is loaded, the * driver can change the algorithm. * -* INTDELAY - This loads the dead-man timer with its inital value. +* INTDELAY - This loads the dead-man timer with its initial value. * When this timer expires the interrupt is asserted, and the * timer is reset each time a new packet is received. (see * BUNDLEMAX below to set the limit on number of chained packets) diff --git a/drivers/net/e1000/e1000_hw.c b/drivers/net/e1000/e1000_hw.c index 65077f39da6..796c4f7d426 100644 --- a/drivers/net/e1000/e1000_hw.c +++ b/drivers/net/e1000/e1000_hw.c @@ -3868,7 +3868,7 @@ e1000_phy_hw_reset(struct e1000_hw *hw) * * hw - Struct containing variables accessed by shared code * -* Sets bit 15 of the MII Control regiser +* Sets bit 15 of the MII Control register ******************************************************************************/ int32_t e1000_phy_reset(struct e1000_hw *hw) diff --git a/drivers/net/sk98lin/h/skdrv2nd.h b/drivers/net/sk98lin/h/skdrv2nd.h index 778d9e618eb..3fa67171e83 100644 --- a/drivers/net/sk98lin/h/skdrv2nd.h +++ b/drivers/net/sk98lin/h/skdrv2nd.h @@ -160,7 +160,7 @@ struct s_IOCTL { /* ** Interim definition of SK_DRV_TIMER placed in this file until -** common modules have boon finallized +** common modules have been finalized */ #define SK_DRV_TIMER 11 #define SK_DRV_MODERATION_TIMER 1 diff --git a/drivers/net/sk98lin/skdim.c b/drivers/net/sk98lin/skdim.c index 07c1b4c8699..37ce03fb8de 100644 --- a/drivers/net/sk98lin/skdim.c +++ b/drivers/net/sk98lin/skdim.c @@ -252,7 +252,7 @@ SkDimEnableModerationIfNeeded(SK_AC *pAC) { /******************************************************************************* ** Function : SkDimDisplayModerationSettings -** Description : Displays the current settings regaring interrupt moderation +** Description : Displays the current settings regarding interrupt moderation ** Programmer : Ralph Roesler ** Last Modified: 22-mar-03 ** Returns : void (!) @@ -510,7 +510,7 @@ EnableIntMod(SK_AC *pAC) { /******************************************************************************* ** Function : DisableIntMod() -** Description : Disbles the interrupt moderation independent of what inter- +** Description : Disables the interrupt moderation independent of what inter- ** rupts are running or not ** Programmer : Ralph Roesler ** Last Modified: 23-mar-03 diff --git a/drivers/net/wireless/ipw2200.c b/drivers/net/wireless/ipw2200.c index 1f742814a01..72120d5c2f7 100644 --- a/drivers/net/wireless/ipw2200.c +++ b/drivers/net/wireless/ipw2200.c @@ -6920,8 +6920,8 @@ static int ipw_qos_association(struct ipw_priv *priv, } /* -* handling the beaconing responces. if we get different QoS setting -* of the network from the the associated setting adjust the QoS +* handling the beaconing responses. if we get different QoS setting +* off the network from the associated setting, adjust the QoS * setting */ static int ipw_qos_association_resp(struct ipw_priv *priv, diff --git a/drivers/parisc/ccio-dma.c b/drivers/parisc/ccio-dma.c index 68cb3a08005..fe3f5f5365c 100644 --- a/drivers/parisc/ccio-dma.c +++ b/drivers/parisc/ccio-dma.c @@ -486,7 +486,7 @@ typedef unsigned long space_t; ** This bit tells U2 to do R/M/W for partial cachelines. "Streaming" ** data can avoid this if the mapping covers full cache lines. ** o STOP_MOST is needed for atomicity across cachelines. -** Apperently only "some EISA devices" need this. +** Apparently only "some EISA devices" need this. ** Using CONFIG_ISA is hack. Only the IOA with EISA under it needs ** to use this hint iff the EISA devices needs this feature. ** According to the U2 ERS, STOP_MOST enabled pages hurt performance. diff --git a/drivers/parisc/iosapic.c b/drivers/parisc/iosapic.c index c2949b4367e..12bab64a62a 100644 --- a/drivers/parisc/iosapic.c +++ b/drivers/parisc/iosapic.c @@ -50,12 +50,12 @@ ** ** PA Firmware ** ----------- -** PA-RISC platforms have two fundementally different types of firmware. +** PA-RISC platforms have two fundamentally different types of firmware. ** For PCI devices, "Legacy" PDC initializes the "INTERRUPT_LINE" register ** and BARs similar to a traditional PC BIOS. ** The newer "PAT" firmware supports PDC calls which return tables. -** PAT firmware only initializes PCI Console and Boot interface. -** With these tables, the OS can progam all other PCI devices. +** PAT firmware only initializes the PCI Console and Boot interface. +** With these tables, the OS can program all other PCI devices. ** ** One such PAT PDC call returns the "Interrupt Routing Table" (IRT). ** The IRT maps each PCI slot's INTA-D "output" line to an I/O SAPIC diff --git a/drivers/pci/hotplug/ibmphp_hpc.c b/drivers/pci/hotplug/ibmphp_hpc.c index c3ac98a0a6a..f55ac3885cb 100644 --- a/drivers/pci/hotplug/ibmphp_hpc.c +++ b/drivers/pci/hotplug/ibmphp_hpc.c @@ -531,7 +531,7 @@ static u8 hpc_readcmdtoindex (u8 cmd, u8 index) * * Action: issue a READ command to HPC * -* Input: pslot - can not be NULL for READ_ALLSTAT +* Input: pslot - cannot be NULL for READ_ALLSTAT * pstatus - can be NULL for READ_ALLSTAT * * Return 0 or error codes diff --git a/drivers/s390/net/claw.h b/drivers/s390/net/claw.h index 969be465309..1ee9a6f0654 100644 --- a/drivers/s390/net/claw.h +++ b/drivers/s390/net/claw.h @@ -29,7 +29,7 @@ #define CLAW_COMPLETE 0xff /* flag to indicate i/o completed */ /*-----------------------------------------------------* -* CLAW control comand code * +* CLAW control command code * *------------------------------------------------------*/ #define SYSTEM_VALIDATE_REQUEST 0x01 /* System Validate request */ diff --git a/drivers/scsi/aic94xx/aic94xx_reg_def.h b/drivers/scsi/aic94xx/aic94xx_reg_def.h index b79f45f3ad4..a11f4e6d8bd 100644 --- a/drivers/scsi/aic94xx/aic94xx_reg_def.h +++ b/drivers/scsi/aic94xx/aic94xx_reg_def.h @@ -2000,7 +2000,7 @@ * The host accesses this scratch in a different manner from the * central sequencer. The sequencer has to use CSEQ registers CSCRPAGE * and CMnSCRPAGE to access the scratch memory. A flat mapping of the - * scratch memory is avaliable for software convenience and to prevent + * scratch memory is available for software convenience and to prevent * corruption while the sequencer is running. This memory is mapped * onto addresses 800h - BFFh, total of 400h bytes. * diff --git a/drivers/scsi/aic94xx/aic94xx_sds.c b/drivers/scsi/aic94xx/aic94xx_sds.c index de7c04d4254..e5a0ec37e95 100644 --- a/drivers/scsi/aic94xx/aic94xx_sds.c +++ b/drivers/scsi/aic94xx/aic94xx_sds.c @@ -64,7 +64,7 @@ struct asd_ocm_dir { #define OCM_INIT_DIR_ENTRIES 5 /*************************************************************************** -* OCM dircetory default +* OCM directory default ***************************************************************************/ static struct asd_ocm_dir OCMDirInit = { @@ -73,7 +73,7 @@ static struct asd_ocm_dir OCMDirInit = }; /*************************************************************************** -* OCM dircetory Entries default +* OCM directory Entries default ***************************************************************************/ static struct asd_ocm_dir_ent OCMDirEntriesInit[OCM_INIT_DIR_ENTRIES] = { diff --git a/drivers/scsi/ncr53c8xx.c b/drivers/scsi/ncr53c8xx.c index 6cc2bc2f62b..adb8eb4f5fd 100644 --- a/drivers/scsi/ncr53c8xx.c +++ b/drivers/scsi/ncr53c8xx.c @@ -185,7 +185,7 @@ static inline struct list_head *ncr_list_pop(struct list_head *head) ** power of 2 cache line size. ** Enhanced in linux-2.3.44 to provide a memory pool ** per pcidev to support dynamic dma mapping. (I would -** have preferred a real bus astraction, btw). +** have preferred a real bus abstraction, btw). ** **========================================================== */ @@ -1438,7 +1438,7 @@ struct head { ** The first four bytes (scr_st[4]) are used inside the script by ** "COPY" commands. ** Because source and destination must have the same alignment -** in a DWORD, the fields HAVE to be at the choosen offsets. +** in a DWORD, the fields HAVE to be at the chosen offsets. ** xerr_st 0 (0x34) scratcha ** sync_st 1 (0x05) sxfer ** wide_st 3 (0x03) scntl3 @@ -1498,7 +1498,7 @@ struct head { ** the DSA (data structure address) register points ** to this substructure of the ccb. ** This substructure contains the header with -** the script-processor-changable data and +** the script-processor-changeable data and ** data blocks for the indirect move commands. ** **---------------------------------------------------------- @@ -5107,7 +5107,7 @@ void ncr_complete (struct ncb *np, struct ccb *cp) /* ** This CCB has been skipped by the NCR. -** Queue it in the correponding unit queue. +** Queue it in the corresponding unit queue. */ static void ncr_ccb_skipped(struct ncb *np, struct ccb *cp) { @@ -5896,8 +5896,8 @@ static void ncr_log_hard_error(struct ncb *np, u16 sist, u_char dstat) ** ** In normal cases, interrupt conditions occur one at a ** time. The ncr is able to stack in some extra registers -** other interrupts that will occurs after the first one. -** But severall interrupts may occur at the same time. +** other interrupts that will occur after the first one. +** But, several interrupts may occur at the same time. ** ** We probably should only try to deal with the normal ** case, but it seems that multiple interrupts occur in @@ -6796,7 +6796,7 @@ void ncr_int_sir (struct ncb *np) ** The host status field is set to HS_NEGOTIATE to mark this ** situation. ** -** If the target doesn't answer this message immidiately +** If the target doesn't answer this message immediately ** (as required by the standard), the SIR_NEGO_FAIL interrupt ** will be raised eventually. ** The handler removes the HS_NEGOTIATE status, and sets the diff --git a/drivers/scsi/ncr53c8xx.h b/drivers/scsi/ncr53c8xx.h index cb8b7701431..b39357d9af8 100644 --- a/drivers/scsi/ncr53c8xx.h +++ b/drivers/scsi/ncr53c8xx.h @@ -218,7 +218,7 @@ ** Same as option 1, but also deal with ** misconfigured interrupts. ** -** - Edge triggerred instead of level sensitive. +** - Edge triggered instead of level sensitive. ** - No interrupt line connected. ** - IRQ number misconfigured. ** @@ -549,7 +549,7 @@ struct ncr_driver_setup { /* ** Initial setup. -** Can be overriden at startup by a command line. +** Can be overridden at startup by a command line. */ #define SCSI_NCR_DRIVER_SETUP \ { \ @@ -1093,7 +1093,7 @@ struct scr_tblsel { **----------------------------------------------------------- ** On 810A, 860, 825A, 875, 895 and 896 chips the content ** of SFBR register can be used as data (SCR_SFBR_DATA). -** The 896 has additionnal IO registers starting at +** The 896 has additional IO registers starting at ** offset 0x80. Bit 7 of register offset is stored in ** bit 7 of the SCRIPTS instruction first DWORD. **----------------------------------------------------------- diff --git a/drivers/usb/host/u132-hcd.c b/drivers/usb/host/u132-hcd.c index 32c635ecbf3..a00d1595656 100644 --- a/drivers/usb/host/u132-hcd.c +++ b/drivers/usb/host/u132-hcd.c @@ -211,7 +211,7 @@ int usb_ftdi_elan_read_pcimem(struct platform_device *pdev, u8 addressofs, int usb_ftdi_elan_write_pcimem(struct platform_device *pdev, u8 addressofs, u8 width, u32 data); /* -* these can not be inlines because we need the structure offset!! +* these cannot be inlines because we need the structure offset!! * Does anyone have a better way????? */ #define u132_read_pcimem(u132, member, data) \ @@ -3045,7 +3045,7 @@ static struct hc_driver u132_hc_driver = { * This function may be called by the USB core whilst the "usb_all_devices_rwsem" * is held for writing, thus this module must not call usb_remove_hcd() * synchronously - but instead should immediately stop activity to the -* device and ansynchronously call usb_remove_hcd() +* device and asynchronously call usb_remove_hcd() */ static int __devexit u132_remove(struct platform_device *pdev) { @@ -3241,7 +3241,7 @@ static int u132_resume(struct platform_device *pdev) #define u132_resume NULL #endif /* -* this driver is loaded explicitely by ftdi_u132 +* this driver is loaded explicitly by ftdi_u132 * * the platform_driver struct is static because it is per type of module */ diff --git a/drivers/usb/misc/usb_u132.h b/drivers/usb/misc/usb_u132.h index 551ba8906d6..5b5a3e621da 100644 --- a/drivers/usb/misc/usb_u132.h +++ b/drivers/usb/misc/usb_u132.h @@ -52,7 +52,7 @@ * the kernel to load the "u132-hcd" module. * * The "ftdi-u132" module provides the interface to the inserted -* PC card and the "u132-hcd" module uses the API to send and recieve +* PC card and the "u132-hcd" module uses the API to send and receive * data. The API features call-backs, so that part of the "u132-hcd" * module code will run in the context of one of the kernel threads * of the "ftdi-u132" module. diff --git a/drivers/usb/serial/digi_acceleport.c b/drivers/usb/serial/digi_acceleport.c index bdb58100fc1..5e3ac281a2f 100644 --- a/drivers/usb/serial/digi_acceleport.c +++ b/drivers/usb/serial/digi_acceleport.c @@ -157,7 +157,7 @@ * to TASK_RUNNING will be lost and write_chan's subsequent call to * schedule() will never return (unless it catches a signal). * This race condition occurs because write_bulk_callback() (and thus -* the wakeup) are called asynchonously from an interrupt, rather than +* the wakeup) are called asynchronously from an interrupt, rather than * from the scheduler. We can avoid the race by calling the wakeup * from the scheduler queue and that's our fix: Now, at the end of * write_bulk_callback() we queue up a wakeup call on the scheduler diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 85ce2326830..ac93174c963 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -1464,7 +1464,7 @@ static int flush_journal_list(struct super_block *s, } /* if someone has this block in a newer transaction, just make - ** sure they are commited, and don't try writing it to disk + ** sure they are committed, and don't try writing it to disk */ if (pjl) { if (atomic_read(&pjl->j_commit_left)) @@ -3384,7 +3384,7 @@ static int remove_from_transaction(struct super_block *p_s_sb, /* ** for any cnode in a journal list, it can only be dirtied of all the -** transactions that include it are commited to disk. +** transactions that include it are committed to disk. ** this checks through each transaction, and returns 1 if you are allowed to dirty, ** and 0 if you aren't ** @@ -3426,7 +3426,7 @@ static int can_dirty(struct reiserfs_journal_cnode *cn) } /* syncs the commit blocks, but does not force the real buffers to disk -** will wait until the current transaction is done/commited before returning +** will wait until the current transaction is done/committed before returning */ int journal_end_sync(struct reiserfs_transaction_handle *th, struct super_block *p_s_sb, unsigned long nblocks) diff --git a/include/asm-m68knommu/mcfmbus.h b/include/asm-m68knommu/mcfmbus.h index 13df9d41bd1..319899c47a2 100644 --- a/include/asm-m68knommu/mcfmbus.h +++ b/include/asm-m68knommu/mcfmbus.h @@ -37,7 +37,7 @@ #define MCFMBUS_MFDR_MBC(a) ((a)&0x3F) /*M-Bus Clock*/ /* -* Define bit flags in Controll Register +* Define bit flags in Control Register */ #define MCFMBUS_MBCR_MEN (0x80) /* M-Bus Enable */ diff --git a/include/asm-parisc/dma.h b/include/asm-parisc/dma.h index da2cf373e31..31ad0f05af3 100644 --- a/include/asm-parisc/dma.h +++ b/include/asm-parisc/dma.h @@ -17,10 +17,10 @@ /* ** DMA_CHUNK_SIZE is used by the SCSI mid-layer to break up -** (or rather not merge) DMA's into managable chunks. +** (or rather not merge) DMAs into manageable chunks. ** On parisc, this is more of the software/tuning constraint -** rather than the HW. I/O MMU allocation alogorithms can be -** faster with smaller size is (to some degree). +** rather than the HW. I/O MMU allocation algorithms can be +** faster with smaller sizes (to some degree). */ #define DMA_CHUNK_SIZE (BITS_PER_LONG*PAGE_SIZE) diff --git a/include/asm-parisc/pci.h b/include/asm-parisc/pci.h index 7b8ad118d2f..7b3be9ac0dd 100644 --- a/include/asm-parisc/pci.h +++ b/include/asm-parisc/pci.h @@ -149,7 +149,7 @@ extern int parisc_bus_is_phys; /* in arch/parisc/kernel/setup.c */ /* ** Most PCI devices (eg Tulip, NCR720) also export the same registers ** to both MMIO and I/O port space. Due to poor performance of I/O Port -** access under HP PCI bus adapters, strongly reccomend use of MMIO +** access under HP PCI bus adapters, strongly recommend the use of MMIO ** address space. ** ** While I'm at it more PA programming notes: diff --git a/include/asm-parisc/ropes.h b/include/asm-parisc/ropes.h index 5542dd00472..007a880615e 100644 --- a/include/asm-parisc/ropes.h +++ b/include/asm-parisc/ropes.h @@ -14,7 +14,7 @@ #endif /* -** The number of pdir entries to "free" before issueing +** The number of pdir entries to "free" before issuing ** a read to PCOM register to flush out PCOM writes. ** Interacts with allocation granularity (ie 4 or 8 entries ** allocated and free'd/purged at a time might make this diff --git a/include/linux/ixjuser.h b/include/linux/ixjuser.h index fd1756d3a47..88b45895746 100644 --- a/include/linux/ixjuser.h +++ b/include/linux/ixjuser.h @@ -315,7 +315,7 @@ typedef struct { * structures. If the freq0 variable is non-zero, the tone table contents * for the tone_index are updated to the frequencies and gains defined. It * should be noted that DTMF tones cannot be reassigned, so if DTMF tone -* table indexs are used in a cadence the frequency and gain variables will +* table indexes are used in a cadence the frequency and gain variables will * be ignored. * * If the array elements contain frequency parameters the driver will diff --git a/include/linux/reiserfs_fs_sb.h b/include/linux/reiserfs_fs_sb.h index 73e0becec08..62a7169aed8 100644 --- a/include/linux/reiserfs_fs_sb.h +++ b/include/linux/reiserfs_fs_sb.h @@ -429,7 +429,7 @@ enum reiserfs_mount_options { /* -o hash={tea, rupasov, r5, detect} is meant for properly mounting ** reiserfs disks from 3.5.19 or earlier. 99% of the time, this option ** is not required. If the normal autodection code can't determine which -** hash to use (because both hases had the same value for a file) +** hash to use (because both hashes had the same value for a file) ** use this option to force a specific hash. It won't allow you to override ** the existing hash on the FS, so if you have a tea hash disk, and mount ** with -o hash=rupasov, the mount will fail. diff --git a/net/wanrouter/af_wanpipe.c b/net/wanrouter/af_wanpipe.c index 6f39faa1583..c2059733e15 100644 --- a/net/wanrouter/af_wanpipe.c +++ b/net/wanrouter/af_wanpipe.c @@ -13,7 +13,7 @@ * Due Credit: * Wanpipe socket layer is based on Packet and * the X25 socket layers. The above sockets were -* used for the specific use of Sangoma Technoloiges +* used for the specific use of Sangoma Technologies * API programs. * Packet socket Authors: Ross Biro, Fred N. van Kempen and * Alan Cox. @@ -23,7 +23,7 @@ * Apr 25, 2000 Nenad Corbic o Added the ability to send zero length packets. * Mar 13, 2000 Nenad Corbic o Added a tx buffer check via ioctl call. * Mar 06, 2000 Nenad Corbic o Fixed the corrupt sock lcn problem. -* Server and client applicaton can run +* Server and client application can run * simultaneously without conflicts. * Feb 29, 2000 Nenad Corbic o Added support for PVC protocols, such as * CHDLC, Frame Relay and HDLC API. diff --git a/net/wanrouter/wanmain.c b/net/wanrouter/wanmain.c index 9479659277a..316211d9f17 100644 --- a/net/wanrouter/wanmain.c +++ b/net/wanrouter/wanmain.c @@ -3,7 +3,7 @@ * * This module is completely hardware-independent and provides * the following common services for the WAN Link Drivers: -* o WAN device managenment (registering, unregistering) +* o WAN device management (registering, unregistering) * o Network interface management * o Physical connection management (dial-up, incoming calls) * o Logical connection management (switched virtual circuits) diff --git a/sound/oss/cs46xx.c b/sound/oss/cs46xx.c index 6e3c41f530e..b1c5d8286e4 100644 --- a/sound/oss/cs46xx.c +++ b/sound/oss/cs46xx.c @@ -779,7 +779,7 @@ static unsigned int cs_set_adc_rate(struct cs_state *state, unsigned int rate) rate = 48000 / 9; /* - * We can not capture at at rate greater than the Input Rate (48000). + * We cannot capture at at rate greater than the Input Rate (48000). * Return an error if an attempt is made to stray outside that limit. */ if (rate > 48000) @@ -4754,8 +4754,8 @@ static int cs_hardware_init(struct cs_card *card) mdelay(5 * cs_laptop_wait); /* Shouldnt be needed ?? */ /* -* If we are resuming under 2.2.x then we can not schedule a timeout. -* so, just spin the CPU. +* If we are resuming under 2.2.x then we cannot schedule a timeout, +* so just spin the CPU. */ if (card->pm.flags & CS46XX_PM_IDLE) { /* -- cgit v1.2.3-70-g09d2 From 03a67a46af8647b2c7825107045ecae641e103d3 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Thu, 30 Nov 2006 05:32:19 +0100 Subject: Fix typos in doc and comments Changes persistant -> persistent. www.dictionary.com does not know persistant (with an A), but should it be one of those things you can spell in more than one correct way, let me know. Signed-off-by: Jan Engelhardt Signed-off-by: Adrian Bunk --- Documentation/Changes | 2 +- Documentation/power/states.txt | 2 +- arch/um/drivers/chan_user.c | 2 +- drivers/message/fusion/mptbase.c | 2 +- drivers/mtd/maps/cfi_flagadm.c | 2 +- drivers/pci/Kconfig | 2 +- fs/Kconfig | 2 +- fs/jfs/jfs_filsys.h | 2 +- include/linux/textsearch.h | 4 ++-- lib/textsearch.c | 2 +- 10 files changed, 11 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/Documentation/Changes b/Documentation/Changes index abee7f58c1e..73a8617f186 100644 --- a/Documentation/Changes +++ b/Documentation/Changes @@ -201,7 +201,7 @@ udev ---- udev is a userspace application for populating /dev dynamically with only entries for devices actually present. udev replaces the basic -functionality of devfs, while allowing persistant device naming for +functionality of devfs, while allowing persistent device naming for devices. FUSE diff --git a/Documentation/power/states.txt b/Documentation/power/states.txt index 3e5e5d3ff41..0931a330d36 100644 --- a/Documentation/power/states.txt +++ b/Documentation/power/states.txt @@ -62,7 +62,7 @@ setup via another operating system for it to use. Despite the inconvenience, this method requires minimal work by the kernel, since the firmware will also handle restoring memory contents on resume. -If the kernel is responsible for persistantly saving state, a mechanism +If the kernel is responsible for persistently saving state, a mechanism called 'swsusp' (Swap Suspend) is used to write memory contents to free swap space. swsusp has some restrictive requirements, but should work in most cases. Some, albeit outdated, documentation can be found diff --git a/arch/um/drivers/chan_user.c b/arch/um/drivers/chan_user.c index 2f880cb167a..0cad3546cb8 100644 --- a/arch/um/drivers/chan_user.c +++ b/arch/um/drivers/chan_user.c @@ -120,7 +120,7 @@ static int winch_thread(void *arg) /* These are synchronization calls between various UML threads on the * host - since they are not different kernel threads, we cannot use * kernel semaphores. We don't use SysV semaphores because they are - * persistant. */ + * persistent. */ count = os_read_file(pipe_fd, &c, sizeof(c)); if(count != sizeof(c)) printk("winch_thread : failed to read synchronization byte, " diff --git a/drivers/message/fusion/mptbase.c b/drivers/message/fusion/mptbase.c index e5c72719deb..051b7c5b8f0 100644 --- a/drivers/message/fusion/mptbase.c +++ b/drivers/message/fusion/mptbase.c @@ -6185,7 +6185,7 @@ mpt_spi_log_info(MPT_ADAPTER *ioc, u32 log_info) "Abort", /* 12h */ "IO Not Yet Executed", /* 13h */ "IO Executed", /* 14h */ - "Persistant Reservation Out Not Affiliation Owner", /* 15h */ + "Persistent Reservation Out Not Affiliation Owner", /* 15h */ "Open Transmit DMA Abort", /* 16h */ "IO Device Missing Delay Retry", /* 17h */ NULL, /* 18h */ diff --git a/drivers/mtd/maps/cfi_flagadm.c b/drivers/mtd/maps/cfi_flagadm.c index 92b5d883d7b..65e5ee55201 100644 --- a/drivers/mtd/maps/cfi_flagadm.c +++ b/drivers/mtd/maps/cfi_flagadm.c @@ -80,7 +80,7 @@ struct mtd_partition flagadm_parts[] = { .size = FLASH_PARTITION2_SIZE }, { - .name = "Persistant storage", + .name = "Persistent storage", .offset = FLASH_PARTITION3_ADDR, .size = FLASH_PARTITION3_SIZE } diff --git a/drivers/pci/Kconfig b/drivers/pci/Kconfig index d0ba112355c..3cfb0a3575e 100644 --- a/drivers/pci/Kconfig +++ b/drivers/pci/Kconfig @@ -34,7 +34,7 @@ config PCI_MULTITHREAD_PROBE It is very unwise to use this option if you are not using a boot process that can handle devices being created in any - order. A program that can create persistant block and network + order. A program that can create persistent block and network device names (like udev) is a good idea if you wish to use this option. diff --git a/fs/Kconfig b/fs/Kconfig index 8bec76bbc0c..b3b5aa0edff 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -972,7 +972,7 @@ config SYSFS Some system agents rely on the information in sysfs to operate. /sbin/hotplug uses device and object attributes in sysfs to assist in - delegating policy decisions, like persistantly naming devices. + delegating policy decisions, like persistently naming devices. sysfs is currently used by the block subsystem to mount the root partition. If sysfs is disabled you must specify the boot device on diff --git a/fs/jfs/jfs_filsys.h b/fs/jfs/jfs_filsys.h index 9901928668c..eb550b339bb 100644 --- a/fs/jfs/jfs_filsys.h +++ b/fs/jfs/jfs_filsys.h @@ -81,7 +81,7 @@ #define JFS_SWAP_BYTES 0x00100000 /* running on big endian computer */ /* Directory index */ -#define JFS_DIR_INDEX 0x00200000 /* Persistant index for */ +#define JFS_DIR_INDEX 0x00200000 /* Persistent index for */ /* directory entries */ diff --git a/include/linux/textsearch.h b/include/linux/textsearch.h index 7dac8f04d28..004808a6df1 100644 --- a/include/linux/textsearch.h +++ b/include/linux/textsearch.h @@ -20,7 +20,7 @@ struct ts_config; /** * struct ts_state - search state * @offset: offset for next match - * @cb: control buffer, for persistant variables of get_next_block() + * @cb: control buffer, for persistent variables of get_next_block() */ struct ts_state { @@ -71,7 +71,7 @@ struct ts_config * Called repeatedly until 0 is returned. Must assign the * head of the next block of data to &*dst and return the length * of the block or 0 if at the end. consumed == 0 indicates - * a new search. May store/read persistant values in state->cb. + * a new search. May store/read persistent values in state->cb. */ unsigned int (*get_next_block)(unsigned int consumed, const u8 **dst, diff --git a/lib/textsearch.c b/lib/textsearch.c index 2cb4a437942..98bcadc0118 100644 --- a/lib/textsearch.c +++ b/lib/textsearch.c @@ -40,7 +40,7 @@ * configuration according to the specified parameters. * (3) User starts the search(es) by calling _find() or _next() to * fetch subsequent occurrences. A state variable is provided - * to the algorihtm to store persistant variables. + * to the algorihtm to store persistent variables. * (4) Core eventually resets the search offset and forwards the find() * request to the algorithm. * (5) Algorithm calls get_next_block() provided by the user continously -- cgit v1.2.3-70-g09d2 From be1c63411addba3ad750eb4fdfc50b97bc82825e Mon Sep 17 00:00:00 2001 From: Olaf Kirch Date: Fri, 1 Dec 2006 10:39:12 +0100 Subject: [PATCH] blktrace: add timestamp message This adds a new timestamp message to blktrace, giving the timeofday when we starting tracing. This helps user space correlate block trace events with eg an application strace. This requires a (compatible) update to blkparse. The changed blkparse is still able to process traces generated by older kernels, and older versions of blkparse should silently ignore the new records (because they have a pid of 0). Signed-off-by: Olaf Kirch Signed-off-by: Jens Axboe --- block/blktrace.c | 57 ++++++++++++++++++++++++++++++++++---------- include/linux/blktrace_api.h | 12 ++++++++++ 2 files changed, 57 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/block/blktrace.c b/block/blktrace.c index 135593c8e45..562ca7cbf85 100644 --- a/block/blktrace.c +++ b/block/blktrace.c @@ -22,30 +22,61 @@ #include #include #include +#include #include static DEFINE_PER_CPU(unsigned long long, blk_trace_cpu_offset) = { 0, }; static unsigned int blktrace_seq __read_mostly = 1; +/* + * Send out a notify message. + */ +static inline unsigned int trace_note(struct blk_trace *bt, + pid_t pid, int action, + const void *data, size_t len) +{ + struct blk_io_trace *t; + int cpu = smp_processor_id(); + + t = relay_reserve(bt->rchan, sizeof(*t) + len); + if (t == NULL) + return 0; + + t->magic = BLK_IO_TRACE_MAGIC | BLK_IO_TRACE_VERSION; + t->time = sched_clock() - per_cpu(blk_trace_cpu_offset, cpu); + t->device = bt->dev; + t->action = action; + t->pid = pid; + t->cpu = cpu; + t->pdu_len = len; + memcpy((void *) t + sizeof(*t), data, len); + return blktrace_seq; +} + /* * Send out a notify for this process, if we haven't done so since a trace * started */ static void trace_note_tsk(struct blk_trace *bt, struct task_struct *tsk) { - struct blk_io_trace *t; + tsk->btrace_seq = trace_note(bt, tsk->pid, + BLK_TN_PROCESS, + tsk->comm, sizeof(tsk->comm)); +} - t = relay_reserve(bt->rchan, sizeof(*t) + sizeof(tsk->comm)); - if (t) { - t->magic = BLK_IO_TRACE_MAGIC | BLK_IO_TRACE_VERSION; - t->device = bt->dev; - t->action = BLK_TC_ACT(BLK_TC_NOTIFY); - t->pid = tsk->pid; - t->cpu = smp_processor_id(); - t->pdu_len = sizeof(tsk->comm); - memcpy((void *) t + sizeof(*t), tsk->comm, t->pdu_len); - tsk->btrace_seq = blktrace_seq; - } +static void trace_note_time(struct blk_trace *bt) +{ + struct timespec now; + unsigned long flags; + u32 words[2]; + + getnstimeofday(&now); + words[0] = now.tv_sec; + words[1] = now.tv_nsec; + + local_irq_save(flags); + trace_note(bt, 0, BLK_TN_TIMESTAMP, words, sizeof(words)); + local_irq_restore(flags); } static int act_log_check(struct blk_trace *bt, u32 what, sector_t sector, @@ -394,6 +425,8 @@ static int blk_trace_startstop(request_queue_t *q, int start) blktrace_seq++; smp_mb(); bt->trace_state = Blktrace_running; + + trace_note_time(bt); ret = 0; } } else { diff --git a/include/linux/blktrace_api.h b/include/linux/blktrace_api.h index b99a714fcac..3680ff9a30e 100644 --- a/include/linux/blktrace_api.h +++ b/include/linux/blktrace_api.h @@ -49,6 +49,15 @@ enum blktrace_act { __BLK_TA_REMAP, /* bio was remapped */ }; +/* + * Notify events. + */ +enum blktrace_notify { + __BLK_TN_PROCESS = 0, /* establish pid/name mapping */ + __BLK_TN_TIMESTAMP, /* include system clock */ +}; + + /* * Trace actions in full. Additionally, read or write is masked */ @@ -68,6 +77,9 @@ enum blktrace_act { #define BLK_TA_BOUNCE (__BLK_TA_BOUNCE) #define BLK_TA_REMAP (__BLK_TA_REMAP | BLK_TC_ACT(BLK_TC_QUEUE)) +#define BLK_TN_PROCESS (__BLK_TN_PROCESS | BLK_TC_ACT(BLK_TC_NOTIFY)) +#define BLK_TN_TIMESTAMP (__BLK_TN_TIMESTAMP | BLK_TC_ACT(BLK_TC_NOTIFY)) + #define BLK_IO_TRACE_MAGIC 0x65617400 #define BLK_IO_TRACE_VERSION 0x07 -- cgit v1.2.3-70-g09d2 From 0e75f9063f5c55fb0b0b546a7c356f8ec186825e Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Fri, 1 Dec 2006 10:40:55 +0100 Subject: [PATCH] block: support larger block pc requests This patch modifies blk_rq_map/unmap_user() and the cdrom and scsi_ioctl.c users so that it supports requests larger than bio by chaining them together. Signed-off-by: Mike Christie Signed-off-by: Jens Axboe --- block/ll_rw_blk.c | 166 +++++++++++++++++++++++++++++++++++++------------ block/scsi_ioctl.c | 53 ++++++++-------- drivers/cdrom/cdrom.c | 6 +- fs/bio.c | 18 +----- include/linux/blkdev.h | 7 ++- 5 files changed, 160 insertions(+), 90 deletions(-) (limited to 'include') diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c index 9eaee664053..0f82e12f7b6 100644 --- a/block/ll_rw_blk.c +++ b/block/ll_rw_blk.c @@ -2322,6 +2322,84 @@ void blk_insert_request(request_queue_t *q, struct request *rq, EXPORT_SYMBOL(blk_insert_request); +static int __blk_rq_unmap_user(struct bio *bio) +{ + int ret = 0; + + if (bio) { + if (bio_flagged(bio, BIO_USER_MAPPED)) + bio_unmap_user(bio); + else + ret = bio_uncopy_user(bio); + } + + return ret; +} + +static int __blk_rq_map_user(request_queue_t *q, struct request *rq, + void __user *ubuf, unsigned int len) +{ + unsigned long uaddr; + struct bio *bio, *orig_bio; + int reading, ret; + + reading = rq_data_dir(rq) == READ; + + /* + * if alignment requirement is satisfied, map in user pages for + * direct dma. else, set up kernel bounce buffers + */ + uaddr = (unsigned long) ubuf; + if (!(uaddr & queue_dma_alignment(q)) && !(len & queue_dma_alignment(q))) + bio = bio_map_user(q, NULL, uaddr, len, reading); + else + bio = bio_copy_user(q, uaddr, len, reading); + + if (IS_ERR(bio)) { + return PTR_ERR(bio); + } + + orig_bio = bio; + blk_queue_bounce(q, &bio); + /* + * We link the bounce buffer in and could have to traverse it + * later so we have to get a ref to prevent it from being freed + */ + bio_get(bio); + + /* + * for most (all? don't know of any) queues we could + * skip grabbing the queue lock here. only drivers with + * funky private ->back_merge_fn() function could be + * problematic. + */ + spin_lock_irq(q->queue_lock); + if (!rq->bio) + blk_rq_bio_prep(q, rq, bio); + else if (!q->back_merge_fn(q, rq, bio)) { + ret = -EINVAL; + spin_unlock_irq(q->queue_lock); + goto unmap_bio; + } else { + rq->biotail->bi_next = bio; + rq->biotail = bio; + + rq->nr_sectors += bio_sectors(bio); + rq->hard_nr_sectors = rq->nr_sectors; + rq->data_len += bio->bi_size; + } + spin_unlock_irq(q->queue_lock); + + return bio->bi_size; + +unmap_bio: + /* if it was boucned we must call the end io function */ + bio_endio(bio, bio->bi_size, 0); + __blk_rq_unmap_user(orig_bio); + bio_put(bio); + return ret; +} + /** * blk_rq_map_user - map user data to a request, for REQ_BLOCK_PC usage * @q: request queue where request should be inserted @@ -2343,42 +2421,44 @@ EXPORT_SYMBOL(blk_insert_request); * unmapping. */ int blk_rq_map_user(request_queue_t *q, struct request *rq, void __user *ubuf, - unsigned int len) + unsigned long len) { - unsigned long uaddr; - struct bio *bio; - int reading; + unsigned long bytes_read = 0; + int ret; if (len > (q->max_hw_sectors << 9)) return -EINVAL; if (!len || !ubuf) return -EINVAL; - reading = rq_data_dir(rq) == READ; + while (bytes_read != len) { + unsigned long map_len, end, start; - /* - * if alignment requirement is satisfied, map in user pages for - * direct dma. else, set up kernel bounce buffers - */ - uaddr = (unsigned long) ubuf; - if (!(uaddr & queue_dma_alignment(q)) && !(len & queue_dma_alignment(q))) - bio = bio_map_user(q, NULL, uaddr, len, reading); - else - bio = bio_copy_user(q, uaddr, len, reading); + map_len = min_t(unsigned long, len - bytes_read, BIO_MAX_SIZE); + end = ((unsigned long)ubuf + map_len + PAGE_SIZE - 1) + >> PAGE_SHIFT; + start = (unsigned long)ubuf >> PAGE_SHIFT; - if (!IS_ERR(bio)) { - rq->bio = rq->biotail = bio; - blk_rq_bio_prep(q, rq, bio); + /* + * A bad offset could cause us to require BIO_MAX_PAGES + 1 + * pages. If this happens we just lower the requested + * mapping len by a page so that we can fit + */ + if (end - start > BIO_MAX_PAGES) + map_len -= PAGE_SIZE; - rq->buffer = rq->data = NULL; - rq->data_len = len; - return 0; + ret = __blk_rq_map_user(q, rq, ubuf, map_len); + if (ret < 0) + goto unmap_rq; + bytes_read += ret; + ubuf += ret; } - /* - * bio is the err-ptr - */ - return PTR_ERR(bio); + rq->buffer = rq->data = NULL; + return 0; +unmap_rq: + blk_rq_unmap_user(rq); + return ret; } EXPORT_SYMBOL(blk_rq_map_user); @@ -2404,7 +2484,7 @@ EXPORT_SYMBOL(blk_rq_map_user); * unmapping. */ int blk_rq_map_user_iov(request_queue_t *q, struct request *rq, - struct sg_iovec *iov, int iov_count) + struct sg_iovec *iov, int iov_count, unsigned int len) { struct bio *bio; @@ -2418,10 +2498,15 @@ int blk_rq_map_user_iov(request_queue_t *q, struct request *rq, if (IS_ERR(bio)) return PTR_ERR(bio); - rq->bio = rq->biotail = bio; + if (bio->bi_size != len) { + bio_endio(bio, bio->bi_size, 0); + bio_unmap_user(bio); + return -EINVAL; + } + + bio_get(bio); blk_rq_bio_prep(q, rq, bio); rq->buffer = rq->data = NULL; - rq->data_len = bio->bi_size; return 0; } @@ -2429,23 +2514,26 @@ EXPORT_SYMBOL(blk_rq_map_user_iov); /** * blk_rq_unmap_user - unmap a request with user data - * @bio: bio to be unmapped - * @ulen: length of user buffer + * @rq: rq to be unmapped * * Description: - * Unmap a bio previously mapped by blk_rq_map_user(). + * Unmap a rq previously mapped by blk_rq_map_user(). + * rq->bio must be set to the original head of the request. */ -int blk_rq_unmap_user(struct bio *bio, unsigned int ulen) +int blk_rq_unmap_user(struct request *rq) { - int ret = 0; + struct bio *bio, *mapped_bio; - if (bio) { - if (bio_flagged(bio, BIO_USER_MAPPED)) - bio_unmap_user(bio); + while ((bio = rq->bio)) { + if (bio_flagged(bio, BIO_BOUNCED)) + mapped_bio = bio->bi_private; else - ret = bio_uncopy_user(bio); - } + mapped_bio = bio; + __blk_rq_unmap_user(mapped_bio); + rq->bio = bio->bi_next; + bio_put(bio); + } return 0; } @@ -2476,11 +2564,8 @@ int blk_rq_map_kern(request_queue_t *q, struct request *rq, void *kbuf, if (rq_data_dir(rq) == WRITE) bio->bi_rw |= (1 << BIO_RW); - rq->bio = rq->biotail = bio; blk_rq_bio_prep(q, rq, bio); - rq->buffer = rq->data = NULL; - rq->data_len = len; return 0; } @@ -3495,6 +3580,7 @@ void blk_rq_bio_prep(request_queue_t *q, struct request *rq, struct bio *bio) rq->hard_cur_sectors = rq->current_nr_sectors; rq->hard_nr_sectors = rq->nr_sectors = bio_sectors(bio); rq->buffer = bio_data(bio); + rq->data_len = bio->bi_size; rq->bio = rq->biotail = bio; } diff --git a/block/scsi_ioctl.c b/block/scsi_ioctl.c index e55a7562143..5493c2fbbab 100644 --- a/block/scsi_ioctl.c +++ b/block/scsi_ioctl.c @@ -226,7 +226,6 @@ static int sg_io(struct file *file, request_queue_t *q, unsigned long start_time; int writing = 0, ret = 0; struct request *rq; - struct bio *bio; char sense[SCSI_SENSE_BUFFERSIZE]; unsigned char cmd[BLK_MAX_CDB]; @@ -258,30 +257,6 @@ static int sg_io(struct file *file, request_queue_t *q, if (!rq) return -ENOMEM; - if (hdr->iovec_count) { - const int size = sizeof(struct sg_iovec) * hdr->iovec_count; - struct sg_iovec *iov; - - iov = kmalloc(size, GFP_KERNEL); - if (!iov) { - ret = -ENOMEM; - goto out; - } - - if (copy_from_user(iov, hdr->dxferp, size)) { - kfree(iov); - ret = -EFAULT; - goto out; - } - - ret = blk_rq_map_user_iov(q, rq, iov, hdr->iovec_count); - kfree(iov); - } else if (hdr->dxfer_len) - ret = blk_rq_map_user(q, rq, hdr->dxferp, hdr->dxfer_len); - - if (ret) - goto out; - /* * fill in request structure */ @@ -294,7 +269,6 @@ static int sg_io(struct file *file, request_queue_t *q, rq->sense_len = 0; rq->cmd_type = REQ_TYPE_BLOCK_PC; - bio = rq->bio; /* * bounce this after holding a reference to the original bio, it's @@ -309,6 +283,31 @@ static int sg_io(struct file *file, request_queue_t *q, if (!rq->timeout) rq->timeout = BLK_DEFAULT_TIMEOUT; + if (hdr->iovec_count) { + const int size = sizeof(struct sg_iovec) * hdr->iovec_count; + struct sg_iovec *iov; + + iov = kmalloc(size, GFP_KERNEL); + if (!iov) { + ret = -ENOMEM; + goto out; + } + + if (copy_from_user(iov, hdr->dxferp, size)) { + kfree(iov); + ret = -EFAULT; + goto out; + } + + ret = blk_rq_map_user_iov(q, rq, iov, hdr->iovec_count, + hdr->dxfer_len); + kfree(iov); + } else if (hdr->dxfer_len) + ret = blk_rq_map_user(q, rq, hdr->dxferp, hdr->dxfer_len); + + if (ret) + goto out; + rq->retries = 0; start_time = jiffies; @@ -339,7 +338,7 @@ static int sg_io(struct file *file, request_queue_t *q, hdr->sb_len_wr = len; } - if (blk_rq_unmap_user(bio, hdr->dxfer_len)) + if (blk_rq_unmap_user(rq)) ret = -EFAULT; /* may not have succeeded, but output values written to control diff --git a/drivers/cdrom/cdrom.c b/drivers/cdrom/cdrom.c index 7ea0f48f8fa..2df5cf4ec74 100644 --- a/drivers/cdrom/cdrom.c +++ b/drivers/cdrom/cdrom.c @@ -2133,16 +2133,14 @@ static int cdrom_read_cdda_bpc(struct cdrom_device_info *cdi, __u8 __user *ubuf, rq->timeout = 60 * HZ; bio = rq->bio; - if (rq->bio) - blk_queue_bounce(q, &rq->bio); - if (blk_execute_rq(q, cdi->disk, rq, 0)) { struct request_sense *s = rq->sense; ret = -EIO; cdi->last_sense = s->sense_key; } - if (blk_rq_unmap_user(bio, len)) + rq->bio = bio; + if (blk_rq_unmap_user(rq)) ret = -EFAULT; if (ret) diff --git a/fs/bio.c b/fs/bio.c index d91cfbf7ebc..aa4d09bd4e7 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -560,10 +560,8 @@ struct bio *bio_copy_user(request_queue_t *q, unsigned long uaddr, break; } - if (bio_add_pc_page(q, bio, page, bytes, 0) < bytes) { - ret = -EINVAL; + if (bio_add_pc_page(q, bio, page, bytes, 0) < bytes) break; - } len -= bytes; } @@ -750,7 +748,6 @@ struct bio *bio_map_user_iov(request_queue_t *q, struct block_device *bdev, int write_to_vm) { struct bio *bio; - int len = 0, i; bio = __bio_map_user_iov(q, bdev, iov, iov_count, write_to_vm); @@ -765,18 +762,7 @@ struct bio *bio_map_user_iov(request_queue_t *q, struct block_device *bdev, */ bio_get(bio); - for (i = 0; i < iov_count; i++) - len += iov[i].iov_len; - - if (bio->bi_size == len) - return bio; - - /* - * don't support partial mappings - */ - bio_endio(bio, bio->bi_size, 0); - bio_unmap_user(bio); - return ERR_PTR(-EINVAL); + return bio; } static void __bio_unmap_user(struct bio *bio) diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 7bfcde2d557..e1c7286165f 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -678,10 +678,11 @@ extern void __blk_stop_queue(request_queue_t *q); extern void blk_run_queue(request_queue_t *); extern void blk_start_queueing(request_queue_t *); extern void blk_queue_activity_fn(request_queue_t *, activity_fn *, void *); -extern int blk_rq_map_user(request_queue_t *, struct request *, void __user *, unsigned int); -extern int blk_rq_unmap_user(struct bio *, unsigned int); +extern int blk_rq_map_user(request_queue_t *, struct request *, void __user *, unsigned long); +extern int blk_rq_unmap_user(struct request *); extern int blk_rq_map_kern(request_queue_t *, struct request *, void *, unsigned int, gfp_t); -extern int blk_rq_map_user_iov(request_queue_t *, struct request *, struct sg_iovec *, int); +extern int blk_rq_map_user_iov(request_queue_t *, struct request *, + struct sg_iovec *, int, unsigned int); extern int blk_execute_rq(request_queue_t *, struct gendisk *, struct request *, int); extern void blk_execute_rq_nowait(request_queue_t *, struct gendisk *, -- cgit v1.2.3-70-g09d2 From bb37b94c68e7b37eecea8576039ae9396ca07839 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Fri, 1 Dec 2006 10:42:33 +0100 Subject: [BLOCK] Cleanup unused variable passing - ->init_queue() does not need the elevator passed in - ->put_request() is a hot path and need not have the queue passed in - cfq_update_io_seektime() does not need cfqd passed in Signed-off-by: Jens Axboe --- block/as-iosched.c | 2 +- block/cfq-iosched.c | 9 ++++----- block/deadline-iosched.c | 2 +- block/elevator.c | 4 ++-- block/noop-iosched.c | 2 +- include/linux/elevator.h | 4 ++-- 6 files changed, 11 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/block/as-iosched.c b/block/as-iosched.c index 50b95e4c142..00242111a45 100644 --- a/block/as-iosched.c +++ b/block/as-iosched.c @@ -1317,7 +1317,7 @@ static void as_exit_queue(elevator_t *e) /* * initialize elevator private data (as_data). */ -static void *as_init_queue(request_queue_t *q, elevator_t *e) +static void *as_init_queue(request_queue_t *q) { struct as_data *ad; diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 1d9c3c70a9a..e9019ed39b7 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -1464,8 +1464,7 @@ cfq_update_io_thinktime(struct cfq_data *cfqd, struct cfq_io_context *cic) } static void -cfq_update_io_seektime(struct cfq_data *cfqd, struct cfq_io_context *cic, - struct request *rq) +cfq_update_io_seektime(struct cfq_io_context *cic, struct request *rq) { sector_t sdist; u64 total; @@ -1617,7 +1616,7 @@ cfq_rq_enqueued(struct cfq_data *cfqd, struct cfq_queue *cfqq, } cfq_update_io_thinktime(cfqd, cic); - cfq_update_io_seektime(cfqd, cic, rq); + cfq_update_io_seektime(cic, rq); cfq_update_idle_window(cfqd, cfqq, cic); cic->last_queue = jiffies; @@ -1770,7 +1769,7 @@ static int cfq_may_queue(request_queue_t *q, int rw) /* * queue lock held here */ -static void cfq_put_request(request_queue_t *q, struct request *rq) +static void cfq_put_request(struct request *rq) { struct cfq_queue *cfqq = RQ_CFQQ(rq); @@ -1951,7 +1950,7 @@ static void cfq_exit_queue(elevator_t *e) kfree(cfqd); } -static void *cfq_init_queue(request_queue_t *q, elevator_t *e) +static void *cfq_init_queue(request_queue_t *q) { struct cfq_data *cfqd; int i; diff --git a/block/deadline-iosched.c b/block/deadline-iosched.c index b7c5b34cb7b..6d673e938d3 100644 --- a/block/deadline-iosched.c +++ b/block/deadline-iosched.c @@ -356,7 +356,7 @@ static void deadline_exit_queue(elevator_t *e) /* * initialize elevator private data (deadline_data). */ -static void *deadline_init_queue(request_queue_t *q, elevator_t *e) +static void *deadline_init_queue(request_queue_t *q) { struct deadline_data *dd; diff --git a/block/elevator.c b/block/elevator.c index 8ccd163254b..c0063f345c5 100644 --- a/block/elevator.c +++ b/block/elevator.c @@ -129,7 +129,7 @@ static struct elevator_type *elevator_get(const char *name) static void *elevator_init_queue(request_queue_t *q, struct elevator_queue *eq) { - return eq->ops->elevator_init_fn(q, eq); + return eq->ops->elevator_init_fn(q); } static void elevator_attach(request_queue_t *q, struct elevator_queue *eq, @@ -810,7 +810,7 @@ void elv_put_request(request_queue_t *q, struct request *rq) elevator_t *e = q->elevator; if (e->ops->elevator_put_req_fn) - e->ops->elevator_put_req_fn(q, rq); + e->ops->elevator_put_req_fn(rq); } int elv_may_queue(request_queue_t *q, int rw) diff --git a/block/noop-iosched.c b/block/noop-iosched.c index 79af4317942..1c3de2b9a6b 100644 --- a/block/noop-iosched.c +++ b/block/noop-iosched.c @@ -65,7 +65,7 @@ noop_latter_request(request_queue_t *q, struct request *rq) return list_entry(rq->queuelist.next, struct request, queuelist); } -static void *noop_init_queue(request_queue_t *q, elevator_t *e) +static void *noop_init_queue(request_queue_t *q) { struct noop_data *nd; diff --git a/include/linux/elevator.h b/include/linux/elevator.h index 2fa9f114422..a24931d2440 100644 --- a/include/linux/elevator.h +++ b/include/linux/elevator.h @@ -21,11 +21,11 @@ typedef void (elevator_completed_req_fn) (request_queue_t *, struct request *); typedef int (elevator_may_queue_fn) (request_queue_t *, int); typedef int (elevator_set_req_fn) (request_queue_t *, struct request *, gfp_t); -typedef void (elevator_put_req_fn) (request_queue_t *, struct request *); +typedef void (elevator_put_req_fn) (struct request *); typedef void (elevator_activate_req_fn) (request_queue_t *, struct request *); typedef void (elevator_deactivate_req_fn) (request_queue_t *, struct request *); -typedef void *(elevator_init_fn) (request_queue_t *, elevator_t *); +typedef void *(elevator_init_fn) (request_queue_t *); typedef void (elevator_exit_fn) (elevator_t *); struct elevator_ops -- cgit v1.2.3-70-g09d2 From bce40a36de574376f41f1ff3c4d212a7da2a3c90 Mon Sep 17 00:00:00 2001 From: Philip Langdale Date: Sat, 21 Oct 2006 12:35:02 +0200 Subject: [PATCH] mmc: Add support for mmc v4 high speed mode This adds support for the high-speed modes defined by mmc v4 (assuming the host controller is up to it). On a TI sdhci controller, it improves read speed from 1.3MBps to 2.3MBps. The TI controller can only go up to 24MHz, but everything helps. Another person has taken this basic patch and used it on a Nokia 770 to get a bigger boost because that controller can run at 48MHZ. Signed-off-by: Philip Langdale Signed-off-by: Pierre Ossman --- drivers/mmc/mmc.c | 121 ++++++++++++++++++++++++++++++++++++++++++- include/linux/mmc/card.h | 8 +++ include/linux/mmc/protocol.h | 47 +++++++++++++++-- 3 files changed, 171 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/drivers/mmc/mmc.c b/drivers/mmc/mmc.c index 766bc54406e..2d5b93000de 100644 --- a/drivers/mmc/mmc.c +++ b/drivers/mmc/mmc.c @@ -4,6 +4,7 @@ * Copyright (C) 2003-2004 Russell King, All Rights Reserved. * SD support Copyright (C) 2004 Ian Molton, All Rights Reserved. * SD support Copyright (C) 2005 Pierre Ossman, All Rights Reserved. + * MMCv4 support Copyright (C) 2006 Philip Langdale, All Rights Reserved. * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -953,6 +954,114 @@ static void mmc_read_csds(struct mmc_host *host) } } +static void mmc_process_ext_csds(struct mmc_host *host) +{ + int err; + struct mmc_card *card; + + struct mmc_request mrq; + struct mmc_command cmd; + struct mmc_data data; + + struct scatterlist sg; + + /* + * As the ext_csd is so large and mostly unused, we don't store the + * raw block in mmc_card. + */ + u8 *ext_csd; + ext_csd = kmalloc(512, GFP_KERNEL); + if (!ext_csd) { + printk("%s: could not allocate a buffer to receive the ext_csd." + "mmc v4 cards will be treated as v3.\n", + mmc_hostname(host)); + return; + } + + list_for_each_entry(card, &host->cards, node) { + if (card->state & (MMC_STATE_DEAD|MMC_STATE_PRESENT)) + continue; + if (mmc_card_sd(card)) + continue; + if (card->csd.mmca_vsn < CSD_SPEC_VER_4) + continue; + + err = mmc_select_card(host, card); + if (err != MMC_ERR_NONE) { + mmc_card_set_dead(card); + continue; + } + + memset(&cmd, 0, sizeof(struct mmc_command)); + + cmd.opcode = MMC_SEND_EXT_CSD; + cmd.arg = 0; + cmd.flags = MMC_RSP_R1 | MMC_CMD_ADTC; + + memset(&data, 0, sizeof(struct mmc_data)); + + mmc_set_data_timeout(&data, card, 0); + + data.blksz = 512; + data.blocks = 1; + data.flags = MMC_DATA_READ; + data.sg = &sg; + data.sg_len = 1; + + memset(&mrq, 0, sizeof(struct mmc_request)); + + mrq.cmd = &cmd; + mrq.data = &data; + + sg_init_one(&sg, ext_csd, 512); + + mmc_wait_for_req(host, &mrq); + + if (cmd.error != MMC_ERR_NONE || data.error != MMC_ERR_NONE) { + mmc_card_set_dead(card); + continue; + } + + switch (ext_csd[EXT_CSD_CARD_TYPE]) { + case EXT_CSD_CARD_TYPE_52 | EXT_CSD_CARD_TYPE_26: + card->ext_csd.hs_max_dtr = 52000000; + break; + case EXT_CSD_CARD_TYPE_26: + card->ext_csd.hs_max_dtr = 26000000; + break; + default: + /* MMC v4 spec says this cannot happen */ + printk("%s: card is mmc v4 but doesn't support " + "any high-speed modes.\n", + mmc_hostname(card->host)); + mmc_card_set_bad(card); + continue; + } + + /* Activate highspeed support. */ + cmd.opcode = MMC_SWITCH; + cmd.arg = (MMC_SWITCH_MODE_WRITE_BYTE << 24) | + (EXT_CSD_HS_TIMING << 16) | + (1 << 8) | + EXT_CSD_CMD_SET_NORMAL; + cmd.flags = MMC_RSP_R1B | MMC_CMD_AC; + + err = mmc_wait_for_cmd(host, &cmd, CMD_RETRIES); + if (err != MMC_ERR_NONE) { + printk("%s: failed to switch card to mmc v4 " + "high-speed mode.\n", + mmc_hostname(card->host)); + continue; + } + + mmc_card_set_highspeed(card); + } + + kfree(ext_csd); + + mmc_deselect_cards(host); +} + static void mmc_read_scrs(struct mmc_host *host) { int err; @@ -1031,8 +1140,14 @@ static unsigned int mmc_calculate_clock(struct mmc_host *host) unsigned int max_dtr = host->f_max; list_for_each_entry(card, &host->cards, node) - if (!mmc_card_dead(card) && max_dtr > card->csd.max_dtr) - max_dtr = card->csd.max_dtr; + if (!mmc_card_dead(card)) { + if (mmc_card_highspeed(card)) { + if (max_dtr > card->ext_csd.hs_max_dtr) + max_dtr = card->ext_csd.hs_max_dtr; + } else if (max_dtr > card->csd.max_dtr) { + max_dtr = card->csd.max_dtr; + } + } pr_debug("%s: selected %d.%03dMHz transfer rate\n", mmc_hostname(host), @@ -1152,6 +1267,8 @@ static void mmc_setup(struct mmc_host *host) if (host->mode == MMC_MODE_SD) mmc_read_scrs(host); + else + mmc_process_ext_csds(host); } diff --git a/include/linux/mmc/card.h b/include/linux/mmc/card.h index 991a37382a2..ce25256f80d 100644 --- a/include/linux/mmc/card.h +++ b/include/linux/mmc/card.h @@ -39,6 +39,10 @@ struct mmc_csd { write_misalign:1; }; +struct mmc_ext_csd { + unsigned int hs_max_dtr; +}; + struct sd_scr { unsigned char sda_vsn; unsigned char bus_widths; @@ -62,11 +66,13 @@ struct mmc_card { #define MMC_STATE_BAD (1<<2) /* unrecognised device */ #define MMC_STATE_SDCARD (1<<3) /* is an SD card */ #define MMC_STATE_READONLY (1<<4) /* card is read-only */ +#define MMC_STATE_HIGHSPEED (1<<5) /* card is in mmc4 highspeed mode */ u32 raw_cid[4]; /* raw card CID */ u32 raw_csd[4]; /* raw card CSD */ u32 raw_scr[2]; /* raw card SCR */ struct mmc_cid cid; /* card identification */ struct mmc_csd csd; /* card specific */ + struct mmc_ext_csd ext_csd; /* mmc v4 extended card specific */ struct sd_scr scr; /* extra SD information */ }; @@ -75,12 +81,14 @@ struct mmc_card { #define mmc_card_bad(c) ((c)->state & MMC_STATE_BAD) #define mmc_card_sd(c) ((c)->state & MMC_STATE_SDCARD) #define mmc_card_readonly(c) ((c)->state & MMC_STATE_READONLY) +#define mmc_card_highspeed(c) ((c)->state & MMC_STATE_HIGHSPEED) #define mmc_card_set_present(c) ((c)->state |= MMC_STATE_PRESENT) #define mmc_card_set_dead(c) ((c)->state |= MMC_STATE_DEAD) #define mmc_card_set_bad(c) ((c)->state |= MMC_STATE_BAD) #define mmc_card_set_sd(c) ((c)->state |= MMC_STATE_SDCARD) #define mmc_card_set_readonly(c) ((c)->state |= MMC_STATE_READONLY) +#define mmc_card_set_highspeed(c) ((c)->state |= MMC_STATE_HIGHSPEED) #define mmc_card_name(c) ((c)->cid.prod_name) #define mmc_card_id(c) ((c)->dev.bus_id) diff --git a/include/linux/mmc/protocol.h b/include/linux/mmc/protocol.h index 08dec8d9e70..311b6547f56 100644 --- a/include/linux/mmc/protocol.h +++ b/include/linux/mmc/protocol.h @@ -25,14 +25,16 @@ #ifndef MMC_MMC_PROTOCOL_H #define MMC_MMC_PROTOCOL_H -/* Standard MMC commands (3.1) type argument response */ +/* Standard MMC commands (4.1) type argument response */ /* class 1 */ #define MMC_GO_IDLE_STATE 0 /* bc */ #define MMC_SEND_OP_COND 1 /* bcr [31:0] OCR R3 */ #define MMC_ALL_SEND_CID 2 /* bcr R2 */ #define MMC_SET_RELATIVE_ADDR 3 /* ac [31:16] RCA R1 */ #define MMC_SET_DSR 4 /* bc [31:16] RCA */ +#define MMC_SWITCH 6 /* ac [31:0] See below R1b */ #define MMC_SELECT_CARD 7 /* ac [31:16] RCA R1 */ +#define MMC_SEND_EXT_CSD 8 /* adtc R1 */ #define MMC_SEND_CSD 9 /* ac [31:16] RCA R2 */ #define MMC_SEND_CID 10 /* ac [31:16] RCA R2 */ #define MMC_READ_DAT_UNTIL_STOP 11 /* adtc [31:0] dadr R1 */ @@ -87,6 +89,17 @@ #define SD_APP_OP_COND 41 /* bcr [31:0] OCR R3 */ #define SD_APP_SEND_SCR 51 /* adtc R1 */ +/* + * MMC_SWITCH argument format: + * + * [31:26] Always 0 + * [25:24] Access Mode + * [23:16] Location of target Byte in EXT_CSD + * [15:08] Value Byte + * [07:03] Always 0 + * [02:00] Command Set + */ + /* MMC status in R1 Type @@ -230,13 +243,41 @@ struct _mmc_csd { #define CSD_STRUCT_VER_1_0 0 /* Valid for system specification 1.0 - 1.2 */ #define CSD_STRUCT_VER_1_1 1 /* Valid for system specification 1.4 - 2.2 */ -#define CSD_STRUCT_VER_1_2 2 /* Valid for system specification 3.1 */ +#define CSD_STRUCT_VER_1_2 2 /* Valid for system specification 3.1 - 3.2 - 3.31 - 4.0 - 4.1 */ +#define CSD_STRUCT_EXT_CSD 3 /* Version is coded in CSD_STRUCTURE in EXT_CSD */ #define CSD_SPEC_VER_0 0 /* Implements system specification 1.0 - 1.2 */ #define CSD_SPEC_VER_1 1 /* Implements system specification 1.4 */ #define CSD_SPEC_VER_2 2 /* Implements system specification 2.0 - 2.2 */ -#define CSD_SPEC_VER_3 3 /* Implements system specification 3.1 */ +#define CSD_SPEC_VER_3 3 /* Implements system specification 3.1 - 3.2 - 3.31 */ +#define CSD_SPEC_VER_4 4 /* Implements system specification 4.0 - 4.1 */ + +/* + * EXT_CSD fields + */ + +#define EXT_CSD_HS_TIMING 185 /* R/W */ +#define EXT_CSD_CARD_TYPE 196 /* RO */ + +/* + * EXT_CSD field definitions + */ + +#define EXT_CSD_CMD_SET_NORMAL (1<<0) +#define EXT_CSD_CMD_SET_SECURE (1<<1) +#define EXT_CSD_CMD_SET_CPSECURE (1<<2) + +#define EXT_CSD_CARD_TYPE_26 (1<<0) /* Card can run at 26MHz */ +#define EXT_CSD_CARD_TYPE_52 (1<<1) /* Card can run at 52MHz */ + +/* + * MMC_SWITCH access modes + */ +#define MMC_SWITCH_MODE_CMD_SET 0x00 /* Change the command set */ +#define MMC_SWITCH_MODE_SET_BITS 0x01 /* Set bits which are 1 in value */ +#define MMC_SWITCH_MODE_CLEAR_BITS 0x02 /* Clear bits which are 1 in value */ +#define MMC_SWITCH_MODE_WRITE_BYTE 0x03 /* Set target to value */ /* * SD bus widths -- cgit v1.2.3-70-g09d2 From e45a1bd20fa5b920901879e85cdf5eda21f78d7c Mon Sep 17 00:00:00 2001 From: Philip Langdale Date: Sun, 29 Oct 2006 10:14:19 +0100 Subject: mmc: Add support for mmc v4 wide-bus modes This change adds support for the mmc4 4-bit wide-bus mode. The mmc4 spec defines 8-bit and 4-bit transfer modes. As we do not support any 8-bit hardware, this patch only adds support for the 4-bit mode, but it can easily be built upon when the time comes. The 4-bit mode is electrically compatible with SD's 4-bit mode but the procedure for turning it on is different. This patch implements only the essential parts of the procedure as defined by the spec. Two additional steps are recommended but not compulsory. I am documenting them here so that there's a record. 1) A bus-test mechanism is implemented using dedicated mmc commands which allow for testing the functionality of the data bus at the electrical level. This is pretty paranoid and they way the commands work is not compatible with the mmc subsystem (they don't set valid CRC values). 2) MMC v4 cards can indicate they would like to draw more than the default amount of current in wide-bus modes. We currently will never switch the card into a higher draw mode. Supposedly, allowing the card to draw more current will let it perform better, but the specs seem to indicate that the card will function correctly without the mode change. Empirical testing supports this interpretation. Signed-off-by: Philip Langdale Signed-off-by: Pierre Ossman --- drivers/mmc/mmc.c | 51 ++++++++++++++++++++++++++++++++------------ include/linux/mmc/protocol.h | 5 +++++ 2 files changed, 42 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/drivers/mmc/mmc.c b/drivers/mmc/mmc.c index 2d5b93000de..1593a6a632c 100644 --- a/drivers/mmc/mmc.c +++ b/drivers/mmc/mmc.c @@ -397,23 +397,23 @@ static int mmc_select_card(struct mmc_host *host, struct mmc_card *card) return err; /* - * Default bus width is 1 bit. - */ - host->ios.bus_width = MMC_BUS_WIDTH_1; - - /* - * We can only change the bus width of the selected - * card so therefore we have to put the handling + * We can only change the bus width of SD cards when + * they are selected so we have to put the handling * here. + * + * The card is in 1 bit mode by default so + * we only need to change if it supports the + * wider version. */ - if (host->caps & MMC_CAP_4_BIT_DATA) { + if (mmc_card_sd(card) && + (card->scr.bus_widths & SD_SCR_BUS_WIDTH_4)) { + /* - * The card is in 1 bit mode by default so - * we only need to change if it supports the - * wider version. - */ - if (mmc_card_sd(card) && - (card->scr.bus_widths & SD_SCR_BUS_WIDTH_4)) { + * Default bus width is 1 bit. + */ + host->ios.bus_width = MMC_BUS_WIDTH_1; + + if (host->caps & MMC_CAP_4_BIT_DATA) { struct mmc_command cmd; cmd.opcode = SD_APP_SET_BUS_WIDTH; cmd.arg = SD_BUS_WIDTH_4; @@ -1055,6 +1055,29 @@ static void mmc_process_ext_csds(struct mmc_host *host) } mmc_card_set_highspeed(card); + + /* Check for host support for wide-bus modes. */ + if (!(host->caps & MMC_CAP_4_BIT_DATA)) { + continue; + } + + /* Activate 4-bit support. */ + cmd.opcode = MMC_SWITCH; + cmd.arg = (MMC_SWITCH_MODE_WRITE_BYTE << 24) | + (EXT_CSD_BUS_WIDTH << 16) | + (EXT_CSD_BUS_WIDTH_4 << 8) | + EXT_CSD_CMD_SET_NORMAL; + cmd.flags = MMC_RSP_R1B | MMC_CMD_AC; + + err = mmc_wait_for_cmd(host, &cmd, CMD_RETRIES); + if (err != MMC_ERR_NONE) { + printk("%s: failed to switch card to " + "mmc v4 4-bit bus mode.\n", + mmc_hostname(card->host)); + continue; + } + + host->ios.bus_width = MMC_BUS_WIDTH_4; } kfree(ext_csd); diff --git a/include/linux/mmc/protocol.h b/include/linux/mmc/protocol.h index 311b6547f56..45c51fd8578 100644 --- a/include/linux/mmc/protocol.h +++ b/include/linux/mmc/protocol.h @@ -256,6 +256,7 @@ struct _mmc_csd { * EXT_CSD fields */ +#define EXT_CSD_BUS_WIDTH 183 /* R/W */ #define EXT_CSD_HS_TIMING 185 /* R/W */ #define EXT_CSD_CARD_TYPE 196 /* RO */ @@ -270,6 +271,10 @@ struct _mmc_csd { #define EXT_CSD_CARD_TYPE_26 (1<<0) /* Card can run at 26MHz */ #define EXT_CSD_CARD_TYPE_52 (1<<1) /* Card can run at 52MHz */ +#define EXT_CSD_BUS_WIDTH_1 0 /* Card is in 1 bit mode */ +#define EXT_CSD_BUS_WIDTH_4 1 /* Card is in 4 bit mode */ +#define EXT_CSD_BUS_WIDTH_8 2 /* Card is in 8 bit mode */ + /* * MMC_SWITCH access modes */ -- cgit v1.2.3-70-g09d2 From 7ccd266e676a3f0c6f8f897f58b684cac3dd1650 Mon Sep 17 00:00:00 2001 From: Pierre Ossman Date: Wed, 8 Nov 2006 23:03:10 +0100 Subject: mmc: Support for high speed SD cards Modern SD cards support a clock speed of 50 MHz. Make sure we test for this capability and do the song and dance required to activate it. Activating high speed support actually modifies the TRAN_SPEED field of the CSD. But as the spec says that the cards must report 50 MHz, we might as well skip re-reading the CSD. Signed-off-by: Pierre Ossman --- drivers/mmc/mmc.c | 122 +++++++++++++++++++++++++++++++++++++++++-- include/linux/mmc/card.h | 7 ++- include/linux/mmc/protocol.h | 22 ++++++++ 3 files changed, 146 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/drivers/mmc/mmc.c b/drivers/mmc/mmc.c index 82b7643c165..9d190022a49 100644 --- a/drivers/mmc/mmc.c +++ b/drivers/mmc/mmc.c @@ -1157,6 +1157,116 @@ static void mmc_read_scrs(struct mmc_host *host) mmc_deselect_cards(host); } +static void mmc_read_switch_caps(struct mmc_host *host) +{ + int err; + struct mmc_card *card; + struct mmc_request mrq; + struct mmc_command cmd; + struct mmc_data data; + unsigned char *status; + struct scatterlist sg; + + status = kmalloc(64, GFP_KERNEL); + if (!status) { + printk(KERN_WARNING "%s: Unable to allocate buffer for " + "reading switch capabilities.\n", + mmc_hostname(host)); + return; + } + + list_for_each_entry(card, &host->cards, node) { + if (card->state & (MMC_STATE_DEAD|MMC_STATE_PRESENT)) + continue; + if (!mmc_card_sd(card)) + continue; + if (card->scr.sda_vsn < SCR_SPEC_VER_1) + continue; + + err = mmc_select_card(host, card); + if (err != MMC_ERR_NONE) { + mmc_card_set_dead(card); + continue; + } + + memset(&cmd, 0, sizeof(struct mmc_command)); + + cmd.opcode = SD_SWITCH; + cmd.arg = 0x00FFFFF1; + cmd.flags = MMC_RSP_R1 | MMC_CMD_ADTC; + + memset(&data, 0, sizeof(struct mmc_data)); + + mmc_set_data_timeout(&data, card, 0); + + data.blksz = 64; + data.blocks = 1; + data.flags = MMC_DATA_READ; + data.sg = &sg; + data.sg_len = 1; + + memset(&mrq, 0, sizeof(struct mmc_request)); + + mrq.cmd = &cmd; + mrq.data = &data; + + sg_init_one(&sg, status, 64); + + mmc_wait_for_req(host, &mrq); + + if (cmd.error != MMC_ERR_NONE || data.error != MMC_ERR_NONE) { + mmc_card_set_dead(card); + continue; + } + + if (status[13] & 0x02) + card->sw_caps.hs_max_dtr = 50000000; + + memset(&cmd, 0, sizeof(struct mmc_command)); + + cmd.opcode = SD_SWITCH; + cmd.arg = 0x80FFFFF1; + cmd.flags = MMC_RSP_R1 | MMC_CMD_ADTC; + + memset(&data, 0, sizeof(struct mmc_data)); + + mmc_set_data_timeout(&data, card, 0); + + data.blksz = 64; + data.blocks = 1; + data.flags = MMC_DATA_READ; + data.sg = &sg; + data.sg_len = 1; + + memset(&mrq, 0, sizeof(struct mmc_request)); + + mrq.cmd = &cmd; + mrq.data = &data; + + sg_init_one(&sg, status, 64); + + mmc_wait_for_req(host, &mrq); + + if (cmd.error != MMC_ERR_NONE || data.error != MMC_ERR_NONE) { + mmc_card_set_dead(card); + continue; + } + + if ((status[16] & 0xF) != 1) { + printk(KERN_WARNING "%s: Problem switching card " + "into high-speed mode!\n", + mmc_hostname(host)); + continue; + } + + mmc_card_set_highspeed(card); + } + + kfree(status); + + mmc_deselect_cards(host); +} + static unsigned int mmc_calculate_clock(struct mmc_host *host) { struct mmc_card *card; @@ -1164,9 +1274,12 @@ static unsigned int mmc_calculate_clock(struct mmc_host *host) list_for_each_entry(card, &host->cards, node) if (!mmc_card_dead(card)) { - if (mmc_card_highspeed(card)) { + if (mmc_card_highspeed(card) && mmc_card_sd(card)) { + if (max_dtr > card->sw_caps.hs_max_dtr) + max_dtr = card->sw_caps.hs_max_dtr; + } else if (mmc_card_highspeed(card) && !mmc_card_sd(card)) { if (max_dtr > card->ext_csd.hs_max_dtr) - max_dtr = card->ext_csd.hs_max_dtr; + max_dtr = card->ext_csd.hs_max_dtr; } else if (max_dtr > card->csd.max_dtr) { max_dtr = card->csd.max_dtr; } @@ -1288,9 +1401,10 @@ static void mmc_setup(struct mmc_host *host) mmc_read_csds(host); - if (host->mode == MMC_MODE_SD) + if (host->mode == MMC_MODE_SD) { mmc_read_scrs(host); - else + mmc_read_switch_caps(host); + } else mmc_process_ext_csds(host); } diff --git a/include/linux/mmc/card.h b/include/linux/mmc/card.h index ce25256f80d..d0e6a549761 100644 --- a/include/linux/mmc/card.h +++ b/include/linux/mmc/card.h @@ -50,6 +50,10 @@ struct sd_scr { #define SD_SCR_BUS_WIDTH_4 (1<<2) }; +struct sd_switch_caps { + unsigned int hs_max_dtr; +}; + struct mmc_host; /* @@ -66,7 +70,7 @@ struct mmc_card { #define MMC_STATE_BAD (1<<2) /* unrecognised device */ #define MMC_STATE_SDCARD (1<<3) /* is an SD card */ #define MMC_STATE_READONLY (1<<4) /* card is read-only */ -#define MMC_STATE_HIGHSPEED (1<<5) /* card is in mmc4 highspeed mode */ +#define MMC_STATE_HIGHSPEED (1<<5) /* card is in high speed mode */ u32 raw_cid[4]; /* raw card CID */ u32 raw_csd[4]; /* raw card CSD */ u32 raw_scr[2]; /* raw card SCR */ @@ -74,6 +78,7 @@ struct mmc_card { struct mmc_csd csd; /* card specific */ struct mmc_ext_csd ext_csd; /* mmc v4 extended card specific */ struct sd_scr scr; /* extra SD information */ + struct sd_switch_caps sw_caps; /* switch (CMD6) caps */ }; #define mmc_card_present(c) ((c)->state & MMC_STATE_PRESENT) diff --git a/include/linux/mmc/protocol.h b/include/linux/mmc/protocol.h index 45c51fd8578..2dce60c43f4 100644 --- a/include/linux/mmc/protocol.h +++ b/include/linux/mmc/protocol.h @@ -82,6 +82,7 @@ /* class 8 */ /* This is basically the same command as for MMC with some quirks. */ #define SD_SEND_RELATIVE_ADDR 3 /* bcr R6 */ +#define SD_SWITCH 6 /* adtc [31:0] See below R1 */ /* Application commands */ #define SD_APP_SET_BUS_WIDTH 6 /* ac [1:0] bus width R1 */ @@ -100,6 +101,19 @@ * [02:00] Command Set */ +/* + * SD_SWITCH argument format: + * + * [31] Check (0) or switch (1) + * [30:24] Reserved (0) + * [23:20] Function group 6 + * [19:16] Function group 5 + * [15:12] Function group 4 + * [11:8] Function group 3 + * [7:4] Function group 2 + * [3:0] Function group 1 + */ + /* MMC status in R1 Type @@ -284,6 +298,14 @@ struct _mmc_csd { #define MMC_SWITCH_MODE_CLEAR_BITS 0x02 /* Clear bits which are 1 in value */ #define MMC_SWITCH_MODE_WRITE_BYTE 0x03 /* Set target to value */ +/* + * SCR field definitions + */ + +#define SCR_SPEC_VER_0 0 /* Implements system specification 1.0 - 1.01 */ +#define SCR_SPEC_VER_1 1 /* Implements system specification 1.10 */ +#define SCR_SPEC_VER_2 2 /* Implements system specification 2.00 */ + /* * SD bus widths */ -- cgit v1.2.3-70-g09d2 From 437052516779fea608261a50682b124315f48f01 Mon Sep 17 00:00:00 2001 From: "inaky@linux.intel.com" Date: Wed, 11 Oct 2006 20:05:58 -0700 Subject: usb/hub: allow hubs up to 31 children Current Wireless USB host hardware (Intel i1480 for example) allows up to 22 devices to connect, thus bringing up the max number of children in the WUSB Host Controller to 22 'fake' ports. Upcoming hardware might raise that limit. Makes almost no difference to go to 31, as the bit arrays are byte-aligned (plus an extra bit in general), so 22 bits fit in 4 bytes as 31 do. As well, the only other array that depends on USB_MAXCHILDREN is 'struct usb_hub->indicator'. By declaring it 'u8' instead of 'enum hub_led_mode', we reduce the size of each entry from 4 bytes (in i386) to 1, which will add as we when are doubling USB_MAXCHILDREN (with 16 the size of that array is 64 bytes, with 31 would be 128; by using u8 that goes down to 31 bytes). Signed-off-by: Inaky Perez-Gonzalez Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/hub.h | 2 +- include/linux/usb.h | 7 ++++++- 2 files changed, 7 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/usb/core/hub.h b/drivers/usb/core/hub.h index 0f8e82a4d48..1b05994de4d 100644 --- a/drivers/usb/core/hub.h +++ b/drivers/usb/core/hub.h @@ -229,7 +229,7 @@ struct usb_hub { unsigned resume_root_hub:1; unsigned has_indicators:1; - enum hub_led_mode indicator[USB_MAXCHILDREN]; + u8 indicator[USB_MAXCHILDREN]; struct work_struct leds; }; diff --git a/include/linux/usb.h b/include/linux/usb.h index 5482bfb3303..e5cb1690975 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -313,8 +313,13 @@ struct usb_bus { /* This is arbitrary. * From USB 2.0 spec Table 11-13, offset 7, a hub can * have up to 255 ports. The most yet reported is 10. + * + * Current Wireless USB host hardware (Intel i1480 for example) allows + * up to 22 devices to connect. Upcoming hardware might raise that + * limit. Because the arrays need to add a bit for hub status data, we + * do 31, so plus one evens out to four bytes. */ -#define USB_MAXCHILDREN (16) +#define USB_MAXCHILDREN (31) struct usb_tt; -- cgit v1.2.3-70-g09d2 From 0c1ac4f25f894f9df0ffe9b912c165fb6a185a3c Mon Sep 17 00:00:00 2001 From: "Luiz Fernando N. Capitulino" Date: Mon, 30 Oct 2006 14:53:03 -0300 Subject: USB: makes usb_endpoint_* functions inline. We have no benefits of having the usb_endpoint_* functions as functions, but making them inline saves text and data segment sizes: text data bss dec hex filename 14893634 3108770 1108840 19111244 1239d4c vmlinux.func 14893185 3108566 1108840 19110591 1239abf vmlinux.inline This is the result of a 2.6.19-rc3 kernel compiled with GCC 4.1.1 without CONFIG_MODULES, CONFIG_CC_OPTIMIZE_FOR_SIZE, CONFIG_REGPARM options set. USB support is fully enabled (while most of the other drivers are not), and that kernel has most of the USB code ported to use the endpoint functions. That happens because a call to those functions are expensive (in terms of bytes), while the function's size is smaller or have the same 'size' of the call. Signed-off-by: Luiz Fernando N. Capitulino Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/usb.c | 144 ------------------------------------------------- include/linux/usb.h | 142 ++++++++++++++++++++++++++++++++++++++++++++---- 2 files changed, 131 insertions(+), 155 deletions(-) (limited to 'include') diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c index 467cb02832f..a83c2d5065c 100644 --- a/drivers/usb/core/usb.c +++ b/drivers/usb/core/usb.c @@ -537,138 +537,6 @@ int usb_get_current_frame_number(struct usb_device *dev) return usb_hcd_get_frame_number (dev); } -/** - * usb_endpoint_dir_in - check if the endpoint has IN direction - * @epd: endpoint to be checked - * - * Returns true if the endpoint is of type IN, otherwise it returns false. - */ -int usb_endpoint_dir_in(const struct usb_endpoint_descriptor *epd) -{ - return ((epd->bEndpointAddress & USB_ENDPOINT_DIR_MASK) == USB_DIR_IN); -} - -/** - * usb_endpoint_dir_out - check if the endpoint has OUT direction - * @epd: endpoint to be checked - * - * Returns true if the endpoint is of type OUT, otherwise it returns false. - */ -int usb_endpoint_dir_out(const struct usb_endpoint_descriptor *epd) -{ - return ((epd->bEndpointAddress & USB_ENDPOINT_DIR_MASK) == USB_DIR_OUT); -} - -/** - * usb_endpoint_xfer_bulk - check if the endpoint has bulk transfer type - * @epd: endpoint to be checked - * - * Returns true if the endpoint is of type bulk, otherwise it returns false. - */ -int usb_endpoint_xfer_bulk(const struct usb_endpoint_descriptor *epd) -{ - return ((epd->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) == - USB_ENDPOINT_XFER_BULK); -} - -/** - * usb_endpoint_xfer_int - check if the endpoint has interrupt transfer type - * @epd: endpoint to be checked - * - * Returns true if the endpoint is of type interrupt, otherwise it returns - * false. - */ -int usb_endpoint_xfer_int(const struct usb_endpoint_descriptor *epd) -{ - return ((epd->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) == - USB_ENDPOINT_XFER_INT); -} - -/** - * usb_endpoint_xfer_isoc - check if the endpoint has isochronous transfer type - * @epd: endpoint to be checked - * - * Returns true if the endpoint is of type isochronous, otherwise it returns - * false. - */ -int usb_endpoint_xfer_isoc(const struct usb_endpoint_descriptor *epd) -{ - return ((epd->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) == - USB_ENDPOINT_XFER_ISOC); -} - -/** - * usb_endpoint_is_bulk_in - check if the endpoint is bulk IN - * @epd: endpoint to be checked - * - * Returns true if the endpoint has bulk transfer type and IN direction, - * otherwise it returns false. - */ -int usb_endpoint_is_bulk_in(const struct usb_endpoint_descriptor *epd) -{ - return (usb_endpoint_xfer_bulk(epd) && usb_endpoint_dir_in(epd)); -} - -/** - * usb_endpoint_is_bulk_out - check if the endpoint is bulk OUT - * @epd: endpoint to be checked - * - * Returns true if the endpoint has bulk transfer type and OUT direction, - * otherwise it returns false. - */ -int usb_endpoint_is_bulk_out(const struct usb_endpoint_descriptor *epd) -{ - return (usb_endpoint_xfer_bulk(epd) && usb_endpoint_dir_out(epd)); -} - -/** - * usb_endpoint_is_int_in - check if the endpoint is interrupt IN - * @epd: endpoint to be checked - * - * Returns true if the endpoint has interrupt transfer type and IN direction, - * otherwise it returns false. - */ -int usb_endpoint_is_int_in(const struct usb_endpoint_descriptor *epd) -{ - return (usb_endpoint_xfer_int(epd) && usb_endpoint_dir_in(epd)); -} - -/** - * usb_endpoint_is_int_out - check if the endpoint is interrupt OUT - * @epd: endpoint to be checked - * - * Returns true if the endpoint has interrupt transfer type and OUT direction, - * otherwise it returns false. - */ -int usb_endpoint_is_int_out(const struct usb_endpoint_descriptor *epd) -{ - return (usb_endpoint_xfer_int(epd) && usb_endpoint_dir_out(epd)); -} - -/** - * usb_endpoint_is_isoc_in - check if the endpoint is isochronous IN - * @epd: endpoint to be checked - * - * Returns true if the endpoint has isochronous transfer type and IN direction, - * otherwise it returns false. - */ -int usb_endpoint_is_isoc_in(const struct usb_endpoint_descriptor *epd) -{ - return (usb_endpoint_xfer_isoc(epd) && usb_endpoint_dir_in(epd)); -} - -/** - * usb_endpoint_is_isoc_out - check if the endpoint is isochronous OUT - * @epd: endpoint to be checked - * - * Returns true if the endpoint has isochronous transfer type and OUT direction, - * otherwise it returns false. - */ -int usb_endpoint_is_isoc_out(const struct usb_endpoint_descriptor *epd) -{ - return (usb_endpoint_xfer_isoc(epd) && usb_endpoint_dir_out(epd)); -} - /*-------------------------------------------------------------------*/ /* * __usb_get_extra_descriptor() finds a descriptor of specific type in the @@ -1102,18 +970,6 @@ EXPORT_SYMBOL(__usb_get_extra_descriptor); EXPORT_SYMBOL(usb_find_device); EXPORT_SYMBOL(usb_get_current_frame_number); -EXPORT_SYMBOL_GPL(usb_endpoint_dir_in); -EXPORT_SYMBOL_GPL(usb_endpoint_dir_out); -EXPORT_SYMBOL_GPL(usb_endpoint_xfer_bulk); -EXPORT_SYMBOL_GPL(usb_endpoint_xfer_int); -EXPORT_SYMBOL_GPL(usb_endpoint_xfer_isoc); -EXPORT_SYMBOL_GPL(usb_endpoint_is_bulk_in); -EXPORT_SYMBOL_GPL(usb_endpoint_is_bulk_out); -EXPORT_SYMBOL_GPL(usb_endpoint_is_int_in); -EXPORT_SYMBOL_GPL(usb_endpoint_is_int_out); -EXPORT_SYMBOL_GPL(usb_endpoint_is_isoc_in); -EXPORT_SYMBOL_GPL(usb_endpoint_is_isoc_out); - EXPORT_SYMBOL (usb_buffer_alloc); EXPORT_SYMBOL (usb_buffer_free); diff --git a/include/linux/usb.h b/include/linux/usb.h index e5cb1690975..e732e024a14 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -495,17 +495,137 @@ static inline int usb_make_path (struct usb_device *dev, char *buf, /*-------------------------------------------------------------------------*/ -extern int usb_endpoint_dir_in(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_dir_out(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_xfer_bulk(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_xfer_int(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_xfer_isoc(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_is_bulk_in(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_is_bulk_out(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_is_int_in(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_is_int_out(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_is_isoc_in(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_is_isoc_out(const struct usb_endpoint_descriptor *epd); +/** + * usb_endpoint_dir_in - check if the endpoint has IN direction + * @epd: endpoint to be checked + * + * Returns true if the endpoint is of type IN, otherwise it returns false. + */ +static inline int usb_endpoint_dir_in(const struct usb_endpoint_descriptor *epd) +{ + return ((epd->bEndpointAddress & USB_ENDPOINT_DIR_MASK) == USB_DIR_IN); +} + +/** + * usb_endpoint_dir_out - check if the endpoint has OUT direction + * @epd: endpoint to be checked + * + * Returns true if the endpoint is of type OUT, otherwise it returns false. + */ +static inline int usb_endpoint_dir_out(const struct usb_endpoint_descriptor *epd) +{ + return ((epd->bEndpointAddress & USB_ENDPOINT_DIR_MASK) == USB_DIR_OUT); +} + +/** + * usb_endpoint_xfer_bulk - check if the endpoint has bulk transfer type + * @epd: endpoint to be checked + * + * Returns true if the endpoint is of type bulk, otherwise it returns false. + */ +static inline int usb_endpoint_xfer_bulk(const struct usb_endpoint_descriptor *epd) +{ + return ((epd->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) == + USB_ENDPOINT_XFER_BULK); +} + +/** + * usb_endpoint_xfer_int - check if the endpoint has interrupt transfer type + * @epd: endpoint to be checked + * + * Returns true if the endpoint is of type interrupt, otherwise it returns + * false. + */ +static inline int usb_endpoint_xfer_int(const struct usb_endpoint_descriptor *epd) +{ + return ((epd->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) == + USB_ENDPOINT_XFER_INT); +} + +/** + * usb_endpoint_xfer_isoc - check if the endpoint has isochronous transfer type + * @epd: endpoint to be checked + * + * Returns true if the endpoint is of type isochronous, otherwise it returns + * false. + */ +static inline int usb_endpoint_xfer_isoc(const struct usb_endpoint_descriptor *epd) +{ + return ((epd->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) == + USB_ENDPOINT_XFER_ISOC); +} + +/** + * usb_endpoint_is_bulk_in - check if the endpoint is bulk IN + * @epd: endpoint to be checked + * + * Returns true if the endpoint has bulk transfer type and IN direction, + * otherwise it returns false. + */ +static inline int usb_endpoint_is_bulk_in(const struct usb_endpoint_descriptor *epd) +{ + return (usb_endpoint_xfer_bulk(epd) && usb_endpoint_dir_in(epd)); +} + +/** + * usb_endpoint_is_bulk_out - check if the endpoint is bulk OUT + * @epd: endpoint to be checked + * + * Returns true if the endpoint has bulk transfer type and OUT direction, + * otherwise it returns false. + */ +static inline int usb_endpoint_is_bulk_out(const struct usb_endpoint_descriptor *epd) +{ + return (usb_endpoint_xfer_bulk(epd) && usb_endpoint_dir_out(epd)); +} + +/** + * usb_endpoint_is_int_in - check if the endpoint is interrupt IN + * @epd: endpoint to be checked + * + * Returns true if the endpoint has interrupt transfer type and IN direction, + * otherwise it returns false. + */ +static inline int usb_endpoint_is_int_in(const struct usb_endpoint_descriptor *epd) +{ + return (usb_endpoint_xfer_int(epd) && usb_endpoint_dir_in(epd)); +} + +/** + * usb_endpoint_is_int_out - check if the endpoint is interrupt OUT + * @epd: endpoint to be checked + * + * Returns true if the endpoint has interrupt transfer type and OUT direction, + * otherwise it returns false. + */ +static inline int usb_endpoint_is_int_out(const struct usb_endpoint_descriptor *epd) +{ + return (usb_endpoint_xfer_int(epd) && usb_endpoint_dir_out(epd)); +} + +/** + * usb_endpoint_is_isoc_in - check if the endpoint is isochronous IN + * @epd: endpoint to be checked + * + * Returns true if the endpoint has isochronous transfer type and IN direction, + * otherwise it returns false. + */ +static inline int usb_endpoint_is_isoc_in(const struct usb_endpoint_descriptor *epd) +{ + return (usb_endpoint_xfer_isoc(epd) && usb_endpoint_dir_in(epd)); +} + +/** + * usb_endpoint_is_isoc_out - check if the endpoint is isochronous OUT + * @epd: endpoint to be checked + * + * Returns true if the endpoint has isochronous transfer type and OUT direction, + * otherwise it returns false. + */ +static inline int usb_endpoint_is_isoc_out(const struct usb_endpoint_descriptor *epd) +{ + return (usb_endpoint_xfer_isoc(epd) && usb_endpoint_dir_out(epd)); +} /*-------------------------------------------------------------------------*/ -- cgit v1.2.3-70-g09d2 From 692a186c9d5f12d43cef28d40c25247dc4f302f0 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Mon, 30 Oct 2006 17:07:51 -0500 Subject: USB: expand autosuspend/autoresume API This patch (as814) adds usb_autopm_set_interface() to the autosuspend API. It also provides convenient wrapper routines, usb_autopm_enable() and usb_autopm_disable(), for drivers that want to specify directly whether autosuspend should be allowed. Signed-off-by: Alan Stern Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/driver.c | 22 ++++++++++++++++++++++ include/linux/usb.h | 29 ++++++++++++++++++++++++++--- 2 files changed, 48 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c index ca0e40ed2b7..204495fa6b3 100644 --- a/drivers/usb/core/driver.c +++ b/drivers/usb/core/driver.c @@ -1322,6 +1322,28 @@ int usb_autopm_get_interface(struct usb_interface *intf) } EXPORT_SYMBOL_GPL(usb_autopm_get_interface); +/** + * usb_autopm_set_interface - set a USB interface's autosuspend state + * @intf: the usb_interface whose state should be set + * + * This routine sets the autosuspend state of @intf's device according + * to @intf's usage counter, which the caller must have set previously. + * If the counter is <= 0, the device is autosuspended (if it isn't + * already suspended and if nothing else prevents the autosuspend). If + * the counter is > 0, the device is autoresumed (if it isn't already + * awake). + */ +int usb_autopm_set_interface(struct usb_interface *intf) +{ + int status; + + status = usb_autopm_do_interface(intf, 0); + // dev_dbg(&intf->dev, "%s: status %d cnt %d\n", + // __FUNCTION__, status, intf->pm_usage_cnt); + return status; +} +EXPORT_SYMBOL_GPL(usb_autopm_set_interface); + #endif /* CONFIG_USB_SUSPEND */ static int usb_suspend(struct device *dev, pm_message_t message) diff --git a/include/linux/usb.h b/include/linux/usb.h index e732e024a14..864c6c21c21 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -415,14 +415,37 @@ extern struct usb_device *usb_find_device(u16 vendor_id, u16 product_id); /* USB autosuspend and autoresume */ #ifdef CONFIG_USB_SUSPEND +extern int usb_autopm_set_interface(struct usb_interface *intf); extern int usb_autopm_get_interface(struct usb_interface *intf); extern void usb_autopm_put_interface(struct usb_interface *intf); +static inline void usb_autopm_enable(struct usb_interface *intf) +{ + intf->pm_usage_cnt = 0; + usb_autopm_set_interface(intf); +} + +static inline void usb_autopm_disable(struct usb_interface *intf) +{ + intf->pm_usage_cnt = 1; + usb_autopm_set_interface(intf); +} + #else -#define usb_autopm_get_interface(intf) 0 -#define usb_autopm_put_interface(intf) do {} while (0) -#endif +static inline int usb_autopm_set_interface(struct usb_interface *intf) +{ return 0; } + +static inline int usb_autopm_get_interface(struct usb_interface *intf) +{ return 0; } + +static inline void usb_autopm_put_interface(struct usb_interface *intf) +{ } +static inline void usb_autopm_enable(struct usb_interface *intf) +{ } +static inline void usb_autopm_disable(struct usb_interface *intf) +{ } +#endif /*-------------------------------------------------------------------------*/ -- cgit v1.2.3-70-g09d2 From ce3615879ae85373c03744b45b7c2d7ae5e29b2a Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Mon, 20 Nov 2006 11:12:22 -0500 Subject: USB: struct usb_device: change flag to bitflag This patch (as816) changes an existing flag in the usb_device structure to a bitflag, preparing the way for more bitflags to come in the future. Signed-off-by: Alan Stern Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/message.c | 2 +- include/linux/usb.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c index 7729c074488..5684d872292 100644 --- a/drivers/usb/core/message.c +++ b/drivers/usb/core/message.c @@ -764,7 +764,7 @@ int usb_string(struct usb_device *dev, int index, char *buf, size_t size) err = -EINVAL; goto errout; } else { - dev->have_langid = -1; + dev->have_langid = 1; dev->string_langid = tbuf[2] | (tbuf[3]<< 8); /* always use the first langid listed */ dev_dbg (&dev->dev, "default language 0x%04x\n", diff --git a/include/linux/usb.h b/include/linux/usb.h index 864c6c21c21..5634a2d91ec 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -362,7 +362,7 @@ struct usb_device { u8 portnum; /* Parent port number (origin 1) */ u8 level; /* Number of USB hub ancestors */ - int have_langid; /* whether string_langid is valid */ + unsigned have_langid:1; /* whether string_langid is valid */ int string_langid; /* language ID for strings */ /* static strings from the device */ -- cgit v1.2.3-70-g09d2 From ee49fb5dc89d34f1794ac9362fa97c1a640f7ddd Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Wed, 22 Nov 2006 16:55:54 -0500 Subject: USB: keep count of unsuspended children This patch (as818b) simplifies autosuspend processing by keeping track of the number of unsuspended children of each USB hub. This will permit us to avoid a good deal of unnecessary work all the time; we will no longer have to create a bunch of workqueue entries to carry out autosuspend requests, only to have them fail because one of the hub's children isn't suspended. The basic idea is simple. There already is a usage counter in the usb_device structure for preventing autosuspends. The patch just increments that counter for every unsuspended child. There's only one tricky part: When a device disconnects we need to remember whether it was suspended at the time (leave the counter alone) or not (decrement the counter). Signed-off-by: Alan Stern Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/driver.c | 34 +++++++++++++++++++++++----------- drivers/usb/core/hub.c | 14 ++++++++++++++ include/linux/usb.h | 1 + 3 files changed, 38 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c index 40c1bf09b2b..0fa15bd62c4 100644 --- a/drivers/usb/core/driver.c +++ b/drivers/usb/core/driver.c @@ -1048,7 +1048,7 @@ int usb_suspend_both(struct usb_device *udev, pm_message_t msg) /* If the suspend succeeded, propagate it up the tree */ } else if (parent) - usb_autosuspend_device(parent, 0); + usb_autosuspend_device(parent, 1); // dev_dbg(&udev->dev, "%s: status %d\n", __FUNCTION__, status); return status; @@ -1096,9 +1096,25 @@ int usb_resume_both(struct usb_device *udev) /* Propagate the resume up the tree, if necessary */ if (udev->state == USB_STATE_SUSPENDED) { if (parent) { - usb_pm_lock(parent); - parent->auto_pm = 1; - status = usb_resume_both(parent); + status = usb_autoresume_device(parent, 1); + if (status == 0) { + status = usb_resume_device(udev); + if (status) { + usb_autosuspend_device(parent, 1); + + /* It's possible usb_resume_device() + * failed after the port was + * unsuspended, causing udev to be + * logically disconnected. We don't + * want usb_disconnect() to autosuspend + * the parent again, so tell it that + * udev disconnected while still + * suspended. */ + if (udev->state == + USB_STATE_NOTATTACHED) + udev->discon_suspended = 1; + } + } } else { /* We can't progagate beyond the USB subsystem, @@ -1107,11 +1123,9 @@ int usb_resume_both(struct usb_device *udev) if (udev->dev.parent->power.power_state.event != PM_EVENT_ON) status = -EHOSTUNREACH; - } - if (status == 0) - status = usb_resume_device(udev); - if (parent) - usb_pm_unlock(parent); + else + status = usb_resume_device(udev); + } } else { /* Needed only for setting udev->dev.power.power_state.event @@ -1119,8 +1133,6 @@ int usb_resume_both(struct usb_device *udev) status = usb_resume_device(udev); } - /* Now the parent won't suspend until we are finished */ - if (status == 0 && udev->actconfig) { for (i = 0; i < udev->actconfig->desc.bNumInterfaces; i++) { intf = udev->actconfig->interface[i]; diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index 46df5e60764..e46d38b1824 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -1039,6 +1039,8 @@ static void recursively_mark_NOTATTACHED(struct usb_device *udev) if (udev->children[i]) recursively_mark_NOTATTACHED(udev->children[i]); } + if (udev->state == USB_STATE_SUSPENDED) + udev->discon_suspended = 1; udev->state = USB_STATE_NOTATTACHED; } @@ -1228,6 +1230,14 @@ void usb_disconnect(struct usb_device **pdev) *pdev = NULL; spin_unlock_irq(&device_state_lock); + /* Decrement the parent's count of unsuspended children */ + if (udev->parent) { + usb_pm_lock(udev); + if (!udev->discon_suspended) + usb_autosuspend_device(udev->parent, 1); + usb_pm_unlock(udev); + } + put_device(&udev->dev); } @@ -1356,6 +1366,10 @@ static int __usb_new_device(void *void_data) goto fail; } + /* Increment the parent's count of unsuspended children */ + if (udev->parent) + usb_autoresume_device(udev->parent, 1); + exit: module_put(THIS_MODULE); return err; diff --git a/include/linux/usb.h b/include/linux/usb.h index 5634a2d91ec..0cd73edeef1 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -362,6 +362,7 @@ struct usb_device { u8 portnum; /* Parent port number (origin 1) */ u8 level; /* Number of USB hub ancestors */ + unsigned discon_suspended:1; /* Disconnected while suspended */ unsigned have_langid:1; /* whether string_langid is valid */ int string_langid; /* language ID for strings */ -- cgit v1.2.3-70-g09d2 From e65e5fb5ceb02aaea7b65bf8b3b0d0c9057718b6 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Tue, 7 Nov 2006 18:21:21 +1100 Subject: PCI: Make some MSI-X #defines generic Move some MSI-X #defines into pci_regs.h so they can be used outside of drivers/pci. Signed-off-by: Michael Ellerman Signed-off-by: Greg Kroah-Hartman --- drivers/pci/msi.h | 8 -------- include/linux/pci_regs.h | 6 ++++++ 2 files changed, 6 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/drivers/pci/msi.h b/drivers/pci/msi.h index f0cca1772f9..3898f523714 100644 --- a/drivers/pci/msi.h +++ b/drivers/pci/msi.h @@ -6,14 +6,6 @@ #ifndef MSI_H #define MSI_H -/* - * MSI-X Address Register - */ -#define PCI_MSIX_FLAGS_QSIZE 0x7FF -#define PCI_MSIX_FLAGS_ENABLE (1 << 15) -#define PCI_MSIX_FLAGS_BIRMASK (7 << 0) -#define PCI_MSIX_FLAGS_BITMASK (1 << 0) - #define PCI_MSIX_ENTRY_SIZE 16 #define PCI_MSIX_ENTRY_LOWER_ADDR_OFFSET 0 #define PCI_MSIX_ENTRY_UPPER_ADDR_OFFSET 4 diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h index c321316f1bc..064b1dc71c2 100644 --- a/include/linux/pci_regs.h +++ b/include/linux/pci_regs.h @@ -292,6 +292,12 @@ #define PCI_MSI_DATA_64 12 /* 16 bits of data for 64-bit devices */ #define PCI_MSI_MASK_BIT 16 /* Mask bits register */ +/* MSI-X registers (these are at offset PCI_MSI_FLAGS) */ +#define PCI_MSIX_FLAGS_QSIZE 0x7FF +#define PCI_MSIX_FLAGS_ENABLE (1 << 15) +#define PCI_MSIX_FLAGS_BIRMASK (7 << 0) +#define PCI_MSIX_FLAGS_BITMASK (1 << 0) + /* CompactPCI Hotswap Register */ #define PCI_CHSWP_CSR 2 /* Control and Status Register */ -- cgit v1.2.3-70-g09d2 From 3efe2d84c8d909567c7976a7106114127b8c3470 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Tue, 10 Oct 2006 08:01:19 -0600 Subject: PCI: Use pci_generic_prep_mwi on ia64 The pci_generic_prep_mwi() code does everything that pcibios_prep_mwi() does on ia64. All we need to do is be sure that pci_cache_line_size is set appropriately, and we can delete pcibios_prep_mwi(). Using SMP_CACHE_BYTES as the default was wrong on uniprocessor machines as it is only 8 bytes. The default in the generic code of L1_CACHE_BYTES is at least as good. Signed-off-by: Matthew Wilcox Acked-by: Jeff Garzik Signed-off-by: Greg Kroah-Hartman --- arch/ia64/pci/pci.c | 71 ++++++++++++++------------------------------------ include/asm-ia64/pci.h | 21 +++++++-------- 2 files changed, 30 insertions(+), 62 deletions(-) (limited to 'include') diff --git a/arch/ia64/pci/pci.c b/arch/ia64/pci/pci.c index b30be7c48ba..4f774725346 100644 --- a/arch/ia64/pci/pci.c +++ b/arch/ia64/pci/pci.c @@ -738,75 +738,44 @@ int ia64_pci_legacy_write(struct pci_bus *bus, u16 port, u32 val, u8 size) return ret; } +/* It's defined in drivers/pci/pci.c */ +extern u8 pci_cache_line_size; + /** - * pci_cacheline_size - determine cacheline size for PCI devices - * @dev: void + * set_pci_cacheline_size - determine cacheline size for PCI devices * * We want to use the line-size of the outer-most cache. We assume * that this line-size is the same for all CPUs. * * Code mostly taken from arch/ia64/kernel/palinfo.c:cache_info(). - * - * RETURNS: An appropriate -ERRNO error value on eror, or zero for success. */ -static unsigned long -pci_cacheline_size (void) +static void __init set_pci_cacheline_size(void) { u64 levels, unique_caches; s64 status; pal_cache_config_info_t cci; - static u8 cacheline_size; - - if (cacheline_size) - return cacheline_size; status = ia64_pal_cache_summary(&levels, &unique_caches); if (status != 0) { - printk(KERN_ERR "%s: ia64_pal_cache_summary() failed (status=%ld)\n", - __FUNCTION__, status); - return SMP_CACHE_BYTES; + printk(KERN_ERR "%s: ia64_pal_cache_summary() failed " + "(status=%ld)\n", __FUNCTION__, status); + return; } - status = ia64_pal_cache_config_info(levels - 1, /* cache_type (data_or_unified)= */ 2, - &cci); + status = ia64_pal_cache_config_info(levels - 1, + /* cache_type (data_or_unified)= */ 2, &cci); if (status != 0) { - printk(KERN_ERR "%s: ia64_pal_cache_config_info() failed (status=%ld)\n", - __FUNCTION__, status); - return SMP_CACHE_BYTES; + printk(KERN_ERR "%s: ia64_pal_cache_config_info() failed " + "(status=%ld)\n", __FUNCTION__, status); + return; } - cacheline_size = 1 << cci.pcci_line_size; - return cacheline_size; + pci_cache_line_size = (1 << cci.pcci_line_size) / 4; } -/** - * pcibios_prep_mwi - helper function for drivers/pci/pci.c:pci_set_mwi() - * @dev: the PCI device for which MWI is enabled - * - * For ia64, we can get the cacheline sizes from PAL. - * - * RETURNS: An appropriate -ERRNO error value on eror, or zero for success. - */ -int -pcibios_prep_mwi (struct pci_dev *dev) -{ - unsigned long desired_linesize, current_linesize; - int rc = 0; - u8 pci_linesize; - - desired_linesize = pci_cacheline_size(); - - pci_read_config_byte(dev, PCI_CACHE_LINE_SIZE, &pci_linesize); - current_linesize = 4 * pci_linesize; - if (desired_linesize != current_linesize) { - printk(KERN_WARNING "PCI: slot %s has incorrect PCI cache line size of %lu bytes,", - pci_name(dev), current_linesize); - if (current_linesize > desired_linesize) { - printk(" expected %lu bytes instead\n", desired_linesize); - rc = -EINVAL; - } else { - printk(" correcting to %lu\n", desired_linesize); - pci_write_config_byte(dev, PCI_CACHE_LINE_SIZE, desired_linesize / 4); - } - } - return rc; +static int __init pcibios_init(void) +{ + set_pci_cacheline_size(); + return 0; } + +subsys_initcall(pcibios_init); diff --git a/include/asm-ia64/pci.h b/include/asm-ia64/pci.h index ef616fd4cb1..825eb7d882e 100644 --- a/include/asm-ia64/pci.h +++ b/include/asm-ia64/pci.h @@ -26,16 +26,18 @@ void pcibios_config_init(void); struct pci_dev; /* - * PCI_DMA_BUS_IS_PHYS should be set to 1 if there is _necessarily_ a direct correspondence - * between device bus addresses and CPU physical addresses. Platforms with a hardware I/O - * MMU _must_ turn this off to suppress the bounce buffer handling code in the block and - * network device layers. Platforms with separate bus address spaces _must_ turn this off - * and provide a device DMA mapping implementation that takes care of the necessary + * PCI_DMA_BUS_IS_PHYS should be set to 1 if there is _necessarily_ a direct + * correspondence between device bus addresses and CPU physical addresses. + * Platforms with a hardware I/O MMU _must_ turn this off to suppress the + * bounce buffer handling code in the block and network device layers. + * Platforms with separate bus address spaces _must_ turn this off and provide + * a device DMA mapping implementation that takes care of the necessary * address translation. * - * For now, the ia64 platforms which may have separate/multiple bus address spaces all - * have I/O MMUs which support the merging of physically discontiguous buffers, so we can - * use that as the sole factor to determine the setting of PCI_DMA_BUS_IS_PHYS. + * For now, the ia64 platforms which may have separate/multiple bus address + * spaces all have I/O MMUs which support the merging of physically + * discontiguous buffers, so we can use that as the sole factor to determine + * the setting of PCI_DMA_BUS_IS_PHYS. */ extern unsigned long ia64_max_iommu_merge_mask; #define PCI_DMA_BUS_IS_PHYS (ia64_max_iommu_merge_mask == ~0UL) @@ -52,9 +54,6 @@ pcibios_penalize_isa_irq (int irq, int active) /* We don't do dynamic PCI IRQ allocation */ } -#define HAVE_ARCH_PCI_MWI 1 -extern int pcibios_prep_mwi (struct pci_dev *); - #include /* pci_unmap_{single,page} is not a nop, thus... */ -- cgit v1.2.3-70-g09d2 From ebf5a24829def5d066922ceebde61dd57fdc6b1e Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Tue, 10 Oct 2006 08:01:20 -0600 Subject: PCI: Use pci_generic_prep_mwi on sparc64 The setting of the CACHE_LINE_SIZE register in sparc64's pci initialisation code isn't quite adequate as the device may have incompatible requirements. The generic code tests for this, so switch sparc64 over to using it. Since sparc64 has different L1 cache line size and PCI cache line size, it would need to override the generic code like i386 and ia64 do. We know what the cache line size is at compile time though, so introduce a new optional constant PCI_CACHE_LINE_BYTES. Signed-off-by: Matthew Wilcox Signed-off-by: David Miller Acked-by: Jeff Garzik Signed-off-by: Greg Kroah-Hartman --- arch/sparc64/kernel/pci.c | 9 --------- drivers/pci/pci.c | 8 +++++++- include/asm-sparc64/pci.h | 6 ++---- 3 files changed, 9 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/arch/sparc64/kernel/pci.c b/arch/sparc64/kernel/pci.c index e02f01b644a..dfc41cd4bb5 100644 --- a/arch/sparc64/kernel/pci.c +++ b/arch/sparc64/kernel/pci.c @@ -646,13 +646,4 @@ int pci_domain_nr(struct pci_bus *pbus) } EXPORT_SYMBOL(pci_domain_nr); -int pcibios_prep_mwi(struct pci_dev *dev) -{ - /* We set correct PCI_CACHE_LINE_SIZE register values for every - * device probed on this platform. So there is nothing to check - * and this always succeeds. - */ - return 0; -} - #endif /* !(CONFIG_PCI) */ diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 0eaf381ae93..bc88c30a418 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -876,8 +876,14 @@ pci_set_master(struct pci_dev *dev) } #ifndef HAVE_ARCH_PCI_MWI + +#ifndef PCI_CACHE_LINE_BYTES +#define PCI_CACHE_LINE_BYTES L1_CACHE_BYTES +#endif + /* This can be overridden by arch code. */ -u8 pci_cache_line_size = L1_CACHE_BYTES >> 2; +/* Don't forget this is measured in 32-bit words, not bytes */ +u8 pci_cache_line_size = PCI_CACHE_LINE_BYTES / 4; /** * pci_generic_prep_mwi - helper function for pci_set_mwi diff --git a/include/asm-sparc64/pci.h b/include/asm-sparc64/pci.h index e1ea67bc32f..ca6560288ae 100644 --- a/include/asm-sparc64/pci.h +++ b/include/asm-sparc64/pci.h @@ -18,6 +18,8 @@ #define PCI_IRQ_NONE 0xffffffff +#define PCI_CACHE_LINE_BYTES 64 + static inline void pcibios_set_master(struct pci_dev *dev) { /* No special bus mastering setup handling */ @@ -291,10 +293,6 @@ extern int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma, enum pci_mmap_state mmap_state, int write_combine); -/* Platform specific MWI support. */ -#define HAVE_ARCH_PCI_MWI -extern int pcibios_prep_mwi(struct pci_dev *dev); - extern void pcibios_resource_to_bus(struct pci_dev *dev, struct pci_bus_region *region, struct resource *res); -- cgit v1.2.3-70-g09d2 From edb2d97eb57b7a21c9256260562de6a65dda86cc Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Tue, 10 Oct 2006 08:01:21 -0600 Subject: PCI: Replace HAVE_ARCH_PCI_MWI with PCI_DISABLE_MWI pSeries is the only architecture left using HAVE_ARCH_PCI_MWI and it's really inappropriate for its needs. It really wants to disable MWI altogether. So here are a pair of stub implementations for pci_set_mwi and pci_clear_mwi. Also rename pci_generic_prep_mwi to pci_set_cacheline_size since that better reflects what it does. Signed-off-by: Matthew Wilcox Cc: Paul Mackerras Acked-by: Jeff Garzik Signed-off-by: Greg Kroah-Hartman --- drivers/pci/pci.c | 31 ++++++++++++++++++------------- include/asm-powerpc/pci.h | 20 +++++++------------- 2 files changed, 25 insertions(+), 26 deletions(-) (limited to 'include') diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index bc88c30a418..427991741cf 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -875,7 +875,17 @@ pci_set_master(struct pci_dev *dev) pcibios_set_master(dev); } -#ifndef HAVE_ARCH_PCI_MWI +#ifdef PCI_DISABLE_MWI +int pci_set_mwi(struct pci_dev *dev) +{ + return 0; +} + +void pci_clear_mwi(struct pci_dev *dev) +{ +} + +#else #ifndef PCI_CACHE_LINE_BYTES #define PCI_CACHE_LINE_BYTES L1_CACHE_BYTES @@ -886,17 +896,17 @@ pci_set_master(struct pci_dev *dev) u8 pci_cache_line_size = PCI_CACHE_LINE_BYTES / 4; /** - * pci_generic_prep_mwi - helper function for pci_set_mwi - * @dev: the PCI device for which MWI is enabled + * pci_set_cacheline_size - ensure the CACHE_LINE_SIZE register is programmed + * @dev: the PCI device for which MWI is to be enabled * - * Helper function for generic implementation of pcibios_prep_mwi - * function. Originally copied from drivers/net/acenic.c. + * Helper function for pci_set_mwi. + * Originally copied from drivers/net/acenic.c. * Copyright 1998-2001 by Jes Sorensen, . * * RETURNS: An appropriate -ERRNO error value on error, or zero for success. */ static int -pci_generic_prep_mwi(struct pci_dev *dev) +pci_set_cacheline_size(struct pci_dev *dev) { u8 cacheline_size; @@ -922,7 +932,6 @@ pci_generic_prep_mwi(struct pci_dev *dev) return -EINVAL; } -#endif /* !HAVE_ARCH_PCI_MWI */ /** * pci_set_mwi - enables memory-write-invalidate PCI transaction @@ -940,12 +949,7 @@ pci_set_mwi(struct pci_dev *dev) int rc; u16 cmd; -#ifdef HAVE_ARCH_PCI_MWI - rc = pcibios_prep_mwi(dev); -#else - rc = pci_generic_prep_mwi(dev); -#endif - + rc = pci_set_cacheline_size(dev); if (rc) return rc; @@ -976,6 +980,7 @@ pci_clear_mwi(struct pci_dev *dev) pci_write_config_word(dev, PCI_COMMAND, cmd); } } +#endif /* ! PCI_DISABLE_MWI */ /** * pci_intx - enables/disables PCI INTx for device dev diff --git a/include/asm-powerpc/pci.h b/include/asm-powerpc/pci.h index 46afd29b904..721c97f09b2 100644 --- a/include/asm-powerpc/pci.h +++ b/include/asm-powerpc/pci.h @@ -62,19 +62,13 @@ static inline int pci_get_legacy_ide_irq(struct pci_dev *dev, int channel) } #ifdef CONFIG_PPC64 -#define HAVE_ARCH_PCI_MWI 1 -static inline int pcibios_prep_mwi(struct pci_dev *dev) -{ - /* - * We would like to avoid touching the cacheline size or MWI bit - * but we cant do that with the current pcibios_prep_mwi - * interface. pSeries firmware sets the cacheline size (which is not - * the cpu cacheline size in all cases) and hardware treats MWI - * the same as memory write. So we dont touch the cacheline size - * here and allow the generic code to set the MWI bit. - */ - return 0; -} + +/* + * We want to avoid touching the cacheline size or MWI bit. + * pSeries firmware sets the cacheline size (which is not the cpu cacheline + * size in all cases) and hardware treats MWI the same as memory write. + */ +#define PCI_DISABLE_MWI extern struct dma_mapping_ops pci_dma_ops; -- cgit v1.2.3-70-g09d2 From 8ea6091f500162e97687d7acf925f84202066b8d Mon Sep 17 00:00:00 2001 From: John Keller Date: Wed, 4 Oct 2006 16:49:25 -0500 Subject: Altix: Add initial ACPI IO support First phase in introducing ACPI support to SN. In this phase, when running with an ACPI capable PROM, the DSDT will define the root busses and all SN nodes (SGIHUB, SGITIO). An ACPI bus driver will be registered for the node devices, with the acpi_pci_root_driver being used for the root busses. An ACPI vendor descriptor is now used to pass platform specific information for both nodes and busses, eliminating the need for the current SAL calls. Also, with ACPI support, SN fixup code is no longer needed to initiate the PCI bus scans, as the acpi_pci_root_driver does that. However, to maintain backward compatibility with non-ACPI capable PROMs, none of the current 'fixup' code can been deleted, though much restructuring has been done. For example, the bulk of the code in io_common.c is relocated code that is now common regardless of what PROM is running, while io_acpi_init.c and io_init.c contain routines specific to an ACPI or non ACPI capable PROM respectively. A new pci bus fixup platform vector has been created to provide a hook for invoking platform specific bus fixup from pcibios_fixup_bus(). The size of io_space[] has been increased to support systems with large IO configurations. Signed-off-by: John Keller Signed-off-by: Greg Kroah-Hartman --- arch/ia64/pci/pci.c | 4 +- arch/ia64/sn/kernel/Makefile | 5 +- arch/ia64/sn/kernel/io_acpi_init.c | 198 ++++++++++ arch/ia64/sn/kernel/io_common.c | 612 +++++++++++++++++++++++++++++++ arch/ia64/sn/kernel/io_init.c | 630 +++++--------------------------- arch/ia64/sn/kernel/iomv.c | 11 +- arch/ia64/sn/kernel/setup.c | 18 + arch/ia64/sn/kernel/tiocx.c | 2 +- arch/ia64/sn/pci/pcibr/pcibr_provider.c | 17 +- arch/ia64/sn/pci/tioce_provider.c | 18 - include/asm-ia64/io.h | 2 +- include/asm-ia64/machvec.h | 12 + include/asm-ia64/machvec_sn2.h | 2 + include/asm-ia64/sn/acpi.h | 16 + include/asm-ia64/sn/pcidev.h | 22 +- include/asm-ia64/sn/sn_feature_sets.h | 6 + include/asm-ia64/sn/sn_sal.h | 1 + 17 files changed, 992 insertions(+), 584 deletions(-) create mode 100644 arch/ia64/sn/kernel/io_acpi_init.c create mode 100644 arch/ia64/sn/kernel/io_common.c create mode 100644 include/asm-ia64/sn/acpi.h (limited to 'include') diff --git a/arch/ia64/pci/pci.c b/arch/ia64/pci/pci.c index 4f774725346..f4edfbf2713 100644 --- a/arch/ia64/pci/pci.c +++ b/arch/ia64/pci/pci.c @@ -469,10 +469,11 @@ pcibios_fixup_resources(struct pci_dev *dev, int start, int limit) } } -static void __devinit pcibios_fixup_device_resources(struct pci_dev *dev) +void __devinit pcibios_fixup_device_resources(struct pci_dev *dev) { pcibios_fixup_resources(dev, 0, PCI_BRIDGE_RESOURCES); } +EXPORT_SYMBOL_GPL(pcibios_fixup_device_resources); static void __devinit pcibios_fixup_bridge_resources(struct pci_dev *dev) { @@ -493,6 +494,7 @@ pcibios_fixup_bus (struct pci_bus *b) } list_for_each_entry(dev, &b->devices, bus_list) pcibios_fixup_device_resources(dev); + platform_pci_fixup_bus(b); return; } diff --git a/arch/ia64/sn/kernel/Makefile b/arch/ia64/sn/kernel/Makefile index 2d78f34dd76..0a59371d347 100644 --- a/arch/ia64/sn/kernel/Makefile +++ b/arch/ia64/sn/kernel/Makefile @@ -4,13 +4,14 @@ # License. See the file "COPYING" in the main directory of this archive # for more details. # -# Copyright (C) 1999,2001-2005 Silicon Graphics, Inc. All Rights Reserved. +# Copyright (C) 1999,2001-2006 Silicon Graphics, Inc. All Rights Reserved. # CPPFLAGS += -I$(srctree)/arch/ia64/sn/include obj-y += setup.o bte.o bte_error.o irq.o mca.o idle.o \ - huberror.o io_init.o iomv.o klconflib.o pio_phys.o \ + huberror.o io_acpi_init.o io_common.o \ + io_init.o iomv.o klconflib.o pio_phys.o \ sn2/ obj-$(CONFIG_IA64_GENERIC) += machvec.o obj-$(CONFIG_SGI_TIOCX) += tiocx.o diff --git a/arch/ia64/sn/kernel/io_acpi_init.c b/arch/ia64/sn/kernel/io_acpi_init.c new file mode 100644 index 00000000000..a9dc36901b1 --- /dev/null +++ b/arch/ia64/sn/kernel/io_acpi_init.c @@ -0,0 +1,198 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (C) 2006 Silicon Graphics, Inc. All rights reserved. + */ + +#include +#include +#include +#include +#include +#include "xtalk/hubdev.h" +#include + + +/* + * The code in this file will only be executed when running with + * a PROM that has ACPI IO support. (i.e., SN_ACPI_BASE_SUPPORT() == 1) + */ + + +/* + * This value must match the UUID the PROM uses + * (io/acpi/defblk.c) when building a vendor descriptor. + */ +struct acpi_vendor_uuid sn_uuid = { + .subtype = 0, + .data = { 0x2c, 0xc6, 0xa6, 0xfe, 0x9c, 0x44, 0xda, 0x11, + 0xa2, 0x7c, 0x08, 0x00, 0x69, 0x13, 0xea, 0x51 }, +}; + +/* + * Perform the early IO init in PROM. + */ +static s64 +sal_ioif_init(u64 *result) +{ + struct ia64_sal_retval isrv = {0,0,0,0}; + + SAL_CALL_NOLOCK(isrv, + SN_SAL_IOIF_INIT, 0, 0, 0, 0, 0, 0, 0); + *result = isrv.v0; + return isrv.status; +} + +/* + * sn_hubdev_add - The 'add' function of the acpi_sn_hubdev_driver. + * Called for every "SGIHUB" or "SGITIO" device defined + * in the ACPI namespace. + */ +static int __init +sn_hubdev_add(struct acpi_device *device) +{ + struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL }; + u64 addr; + struct hubdev_info *hubdev; + struct hubdev_info *hubdev_ptr; + int i; + u64 nasid; + struct acpi_resource *resource; + int ret = 0; + acpi_status status; + struct acpi_resource_vendor_typed *vendor; + extern void sn_common_hubdev_init(struct hubdev_info *); + + status = acpi_get_vendor_resource(device->handle, METHOD_NAME__CRS, + &sn_uuid, &buffer); + if (ACPI_FAILURE(status)) { + printk(KERN_ERR + "sn_hubdev_add: acpi_get_vendor_resource() failed: %d\n", + status); + return 1; + } + + resource = buffer.pointer; + vendor = &resource->data.vendor_typed; + if ((vendor->byte_length - sizeof(struct acpi_vendor_uuid)) != + sizeof(struct hubdev_info *)) { + printk(KERN_ERR + "sn_hubdev_add: Invalid vendor data length: %d\n", + vendor->byte_length); + ret = 1; + goto exit; + } + + memcpy(&addr, vendor->byte_data, sizeof(struct hubdev_info *)); + hubdev_ptr = __va((struct hubdev_info *) addr); + + nasid = hubdev_ptr->hdi_nasid; + i = nasid_to_cnodeid(nasid); + hubdev = (struct hubdev_info *)(NODEPDA(i)->pdinfo); + *hubdev = *hubdev_ptr; + sn_common_hubdev_init(hubdev); + +exit: + kfree(buffer.pointer); + return ret; +} + +/* + * sn_get_bussoft_ptr() - The pcibus_bussoft pointer is found in + * the ACPI Vendor resource for this bus. + */ +static struct pcibus_bussoft * +sn_get_bussoft_ptr(struct pci_bus *bus) +{ + u64 addr; + struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL }; + acpi_handle handle; + struct pcibus_bussoft *prom_bussoft_ptr; + struct acpi_resource *resource; + acpi_status status; + struct acpi_resource_vendor_typed *vendor; + + + handle = PCI_CONTROLLER(bus)->acpi_handle; + status = acpi_get_vendor_resource(handle, METHOD_NAME__CRS, + &sn_uuid, &buffer); + if (ACPI_FAILURE(status)) { + printk(KERN_ERR "get_acpi_pcibus_ptr: " + "get_acpi_bussoft_info() failed: %d\n", + status); + return NULL; + } + resource = buffer.pointer; + vendor = &resource->data.vendor_typed; + + if ((vendor->byte_length - sizeof(struct acpi_vendor_uuid)) != + sizeof(struct pcibus_bussoft *)) { + printk(KERN_ERR + "get_acpi_bussoft_ptr: Invalid vendor data " + "length %d\n", vendor->byte_length); + kfree(buffer.pointer); + return NULL; + } + memcpy(&addr, vendor->byte_data, sizeof(struct pcibus_bussoft *)); + prom_bussoft_ptr = __va((struct pcibus_bussoft *) addr); + kfree(buffer.pointer); + + return prom_bussoft_ptr; +} + +/* + * sn_acpi_bus_fixup + */ +void +sn_acpi_bus_fixup(struct pci_bus *bus) +{ + struct pci_dev *pci_dev = NULL; + struct pcibus_bussoft *prom_bussoft_ptr; + extern void sn_common_bus_fixup(struct pci_bus *, + struct pcibus_bussoft *); + + if (!bus->parent) { /* If root bus */ + prom_bussoft_ptr = sn_get_bussoft_ptr(bus); + if (prom_bussoft_ptr == NULL) { + printk(KERN_ERR + "sn_pci_fixup_bus: 0x%04x:0x%02x Unable to " + "obtain prom_bussoft_ptr\n", + pci_domain_nr(bus), bus->number); + return; + } + sn_common_bus_fixup(bus, prom_bussoft_ptr); + } + list_for_each_entry(pci_dev, &bus->devices, bus_list) { + sn_pci_fixup_slot(pci_dev); + } +} + +static struct acpi_driver acpi_sn_hubdev_driver = { + .name = "SGI HUBDEV Driver", + .ids = "SGIHUB,SGITIO", + .ops = { + .add = sn_hubdev_add, + }, +}; + + +/* + * sn_io_acpi_init - PROM has ACPI support for IO, defining at a minimum the + * nodes and root buses in the DSDT. As a result, bus scanning + * will be initiated by the Linux ACPI code. + */ + +void __init +sn_io_acpi_init(void) +{ + u64 result; + s64 status; + + acpi_bus_register_driver(&acpi_sn_hubdev_driver); + status = sal_ioif_init(&result); + if (status || result) + panic("sal_ioif_init failed: [%lx] %s\n", + status, ia64_sal_strerror(status)); +} diff --git a/arch/ia64/sn/kernel/io_common.c b/arch/ia64/sn/kernel/io_common.c new file mode 100644 index 00000000000..12531de6754 --- /dev/null +++ b/arch/ia64/sn/kernel/io_common.c @@ -0,0 +1,612 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (C) 2006 Silicon Graphics, Inc. All rights reserved. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "xtalk/hubdev.h" +#include "xtalk/xwidgetdev.h" +#include +#include +#include + +extern void sn_init_cpei_timer(void); +extern void register_sn_procfs(void); +extern void sn_acpi_bus_fixup(struct pci_bus *); +extern void sn_bus_fixup(struct pci_bus *); +extern void sn_acpi_slot_fixup(struct pci_dev *, struct pcidev_info *); +extern void sn_more_slot_fixup(struct pci_dev *, struct pcidev_info *); +extern void sn_legacy_pci_window_fixup(struct pci_controller *, u64, u64); +extern void sn_io_acpi_init(void); +extern void sn_io_init(void); + + +static struct list_head sn_sysdata_list; + +/* sysdata list struct */ +struct sysdata_el { + struct list_head entry; + void *sysdata; +}; + +int sn_ioif_inited; /* SN I/O infrastructure initialized? */ + +struct sn_pcibus_provider *sn_pci_provider[PCIIO_ASIC_MAX_TYPES]; /* indexed by asic type */ + +/* + * Hooks and struct for unsupported pci providers + */ + +static dma_addr_t +sn_default_pci_map(struct pci_dev *pdev, unsigned long paddr, size_t size, int type) +{ + return 0; +} + +static void +sn_default_pci_unmap(struct pci_dev *pdev, dma_addr_t addr, int direction) +{ + return; +} + +static void * +sn_default_pci_bus_fixup(struct pcibus_bussoft *soft, struct pci_controller *controller) +{ + return NULL; +} + +static struct sn_pcibus_provider sn_pci_default_provider = { + .dma_map = sn_default_pci_map, + .dma_map_consistent = sn_default_pci_map, + .dma_unmap = sn_default_pci_unmap, + .bus_fixup = sn_default_pci_bus_fixup, +}; + +/* + * Retrieve the DMA Flush List given nasid, widget, and device. + * This list is needed to implement the WAR - Flush DMA data on PIO Reads. + */ +static inline u64 +sal_get_device_dmaflush_list(u64 nasid, u64 widget_num, u64 device_num, + u64 address) +{ + struct ia64_sal_retval ret_stuff; + ret_stuff.status = 0; + ret_stuff.v0 = 0; + + SAL_CALL_NOLOCK(ret_stuff, + (u64) SN_SAL_IOIF_GET_DEVICE_DMAFLUSH_LIST, + (u64) nasid, (u64) widget_num, + (u64) device_num, (u64) address, 0, 0, 0); + return ret_stuff.status; +} + +/* + * Retrieve the pci device information given the bus and device|function number. + */ +static inline u64 +sal_get_pcidev_info(u64 segment, u64 bus_number, u64 devfn, u64 pci_dev, + u64 sn_irq_info) +{ + struct ia64_sal_retval ret_stuff; + ret_stuff.status = 0; + ret_stuff.v0 = 0; + + SAL_CALL_NOLOCK(ret_stuff, + (u64) SN_SAL_IOIF_GET_PCIDEV_INFO, + (u64) segment, (u64) bus_number, (u64) devfn, + (u64) pci_dev, + sn_irq_info, 0, 0); + return ret_stuff.v0; +} + +/* + * sn_pcidev_info_get() - Retrieve the pcidev_info struct for the specified + * device. + */ +inline struct pcidev_info * +sn_pcidev_info_get(struct pci_dev *dev) +{ + struct pcidev_info *pcidev; + + list_for_each_entry(pcidev, + &(SN_PLATFORM_DATA(dev)->pcidev_info), pdi_list) { + if (pcidev->pdi_linux_pcidev == dev) + return pcidev; + } + return NULL; +} + +/* Older PROM flush WAR + * + * 01/16/06 -- This war will be in place until a new official PROM is released. + * Additionally note that the struct sn_flush_device_war also has to be + * removed from arch/ia64/sn/include/xtalk/hubdev.h + */ +static u8 war_implemented = 0; + +static s64 sn_device_fixup_war(u64 nasid, u64 widget, int device, + struct sn_flush_device_common *common) +{ + struct sn_flush_device_war *war_list; + struct sn_flush_device_war *dev_entry; + struct ia64_sal_retval isrv = {0,0,0,0}; + + if (!war_implemented) { + printk(KERN_WARNING "PROM version < 4.50 -- implementing old " + "PROM flush WAR\n"); + war_implemented = 1; + } + + war_list = kzalloc(DEV_PER_WIDGET * sizeof(*war_list), GFP_KERNEL); + if (!war_list) + BUG(); + + SAL_CALL_NOLOCK(isrv, SN_SAL_IOIF_GET_WIDGET_DMAFLUSH_LIST, + nasid, widget, __pa(war_list), 0, 0, 0 ,0); + if (isrv.status) + panic("sn_device_fixup_war failed: %s\n", + ia64_sal_strerror(isrv.status)); + + dev_entry = war_list + device; + memcpy(common,dev_entry, sizeof(*common)); + kfree(war_list); + + return isrv.status; +} + +/* + * sn_common_hubdev_init() - This routine is called to initialize the HUB data + * structure for each node in the system. + */ +void __init +sn_common_hubdev_init(struct hubdev_info *hubdev) +{ + + struct sn_flush_device_kernel *sn_flush_device_kernel; + struct sn_flush_device_kernel *dev_entry; + s64 status; + int widget, device, size; + + /* Attach the error interrupt handlers */ + if (hubdev->hdi_nasid & 1) /* If TIO */ + ice_error_init(hubdev); + else + hub_error_init(hubdev); + + for (widget = 0; widget <= HUB_WIDGET_ID_MAX; widget++) + hubdev->hdi_xwidget_info[widget].xwi_hubinfo = hubdev; + + if (!hubdev->hdi_flush_nasid_list.widget_p) + return; + + size = (HUB_WIDGET_ID_MAX + 1) * + sizeof(struct sn_flush_device_kernel *); + hubdev->hdi_flush_nasid_list.widget_p = + kzalloc(size, GFP_KERNEL); + if (!hubdev->hdi_flush_nasid_list.widget_p) + BUG(); + + for (widget = 0; widget <= HUB_WIDGET_ID_MAX; widget++) { + size = DEV_PER_WIDGET * + sizeof(struct sn_flush_device_kernel); + sn_flush_device_kernel = kzalloc(size, GFP_KERNEL); + if (!sn_flush_device_kernel) + BUG(); + + dev_entry = sn_flush_device_kernel; + for (device = 0; device < DEV_PER_WIDGET; + device++, dev_entry++) { + size = sizeof(struct sn_flush_device_common); + dev_entry->common = kzalloc(size, GFP_KERNEL); + if (!dev_entry->common) + BUG(); + if (sn_prom_feature_available(PRF_DEVICE_FLUSH_LIST)) + status = sal_get_device_dmaflush_list( + hubdev->hdi_nasid, widget, device, + (u64)(dev_entry->common)); + else + status = sn_device_fixup_war(hubdev->hdi_nasid, + widget, device, + dev_entry->common); + if (status != SALRET_OK) + panic("SAL call failed: %s\n", + ia64_sal_strerror(status)); + + spin_lock_init(&dev_entry->sfdl_flush_lock); + } + + if (sn_flush_device_kernel) + hubdev->hdi_flush_nasid_list.widget_p[widget] = + sn_flush_device_kernel; + } +} + +void sn_pci_unfixup_slot(struct pci_dev *dev) +{ + struct pci_dev *host_pci_dev = SN_PCIDEV_INFO(dev)->host_pci_dev; + + sn_irq_unfixup(dev); + pci_dev_put(host_pci_dev); + pci_dev_put(dev); +} + +/* + * sn_pci_fixup_slot() - This routine sets up a slot's resources consistent + * with the Linux PCI abstraction layer. Resources + * acquired from our PCI provider include PIO maps + * to BAR space and interrupt objects. + */ +void sn_pci_fixup_slot(struct pci_dev *dev) +{ + int segment = pci_domain_nr(dev->bus); + int status = 0; + struct pcibus_bussoft *bs; + struct pci_bus *host_pci_bus; + struct pci_dev *host_pci_dev; + struct pcidev_info *pcidev_info; + struct sn_irq_info *sn_irq_info; + unsigned int bus_no, devfn; + + pci_dev_get(dev); /* for the sysdata pointer */ + pcidev_info = kzalloc(sizeof(struct pcidev_info), GFP_KERNEL); + if (!pcidev_info) + BUG(); /* Cannot afford to run out of memory */ + + sn_irq_info = kzalloc(sizeof(struct sn_irq_info), GFP_KERNEL); + if (!sn_irq_info) + BUG(); /* Cannot afford to run out of memory */ + + /* Call to retrieve pci device information needed by kernel. */ + status = sal_get_pcidev_info((u64) segment, (u64) dev->bus->number, + dev->devfn, + (u64) __pa(pcidev_info), + (u64) __pa(sn_irq_info)); + if (status) + BUG(); /* Cannot get platform pci device information */ + + /* Add pcidev_info to list in pci_controller.platform_data */ + list_add_tail(&pcidev_info->pdi_list, + &(SN_PLATFORM_DATA(dev->bus)->pcidev_info)); + + if (!SN_ACPI_BASE_SUPPORT()) + sn_more_slot_fixup(dev, pcidev_info); + + /* + * Using the PROMs values for the PCI host bus, get the Linux + * PCI host_pci_dev struct and set up host bus linkages + */ + + bus_no = (pcidev_info->pdi_slot_host_handle >> 32) & 0xff; + devfn = pcidev_info->pdi_slot_host_handle & 0xffffffff; + host_pci_bus = pci_find_bus(segment, bus_no); + host_pci_dev = pci_get_slot(host_pci_bus, devfn); + + pcidev_info->host_pci_dev = host_pci_dev; + pcidev_info->pdi_linux_pcidev = dev; + pcidev_info->pdi_host_pcidev_info = SN_PCIDEV_INFO(host_pci_dev); + bs = SN_PCIBUS_BUSSOFT(dev->bus); + pcidev_info->pdi_pcibus_info = bs; + + if (bs && bs->bs_asic_type < PCIIO_ASIC_MAX_TYPES) { + SN_PCIDEV_BUSPROVIDER(dev) = sn_pci_provider[bs->bs_asic_type]; + } else { + SN_PCIDEV_BUSPROVIDER(dev) = &sn_pci_default_provider; + } + + /* Only set up IRQ stuff if this device has a host bus context */ + if (bs && sn_irq_info->irq_irq) { + pcidev_info->pdi_sn_irq_info = sn_irq_info; + dev->irq = pcidev_info->pdi_sn_irq_info->irq_irq; + sn_irq_fixup(dev, sn_irq_info); + } else { + pcidev_info->pdi_sn_irq_info = NULL; + kfree(sn_irq_info); + } +} + +/* + * sn_common_bus_fixup - Perform platform specific bus fixup. + * Execute the ASIC specific fixup routine + * for this bus. + */ +void +sn_common_bus_fixup(struct pci_bus *bus, + struct pcibus_bussoft *prom_bussoft_ptr) +{ + int cnode; + struct pci_controller *controller; + struct hubdev_info *hubdev_info; + int nasid; + void *provider_soft; + struct sn_pcibus_provider *provider; + struct sn_platform_data *sn_platform_data; + + controller = PCI_CONTROLLER(bus); + /* + * Per-provider fixup. Copies the bus soft structure from prom + * to local area and links SN_PCIBUS_BUSSOFT(). + */ + + if (prom_bussoft_ptr->bs_asic_type >= PCIIO_ASIC_MAX_TYPES) { + printk(KERN_WARNING "sn_common_bus_fixup: Unsupported asic type, %d", + prom_bussoft_ptr->bs_asic_type); + return; + } + + if (prom_bussoft_ptr->bs_asic_type == PCIIO_ASIC_TYPE_PPB) + return; /* no further fixup necessary */ + + provider = sn_pci_provider[prom_bussoft_ptr->bs_asic_type]; + if (provider == NULL) + panic("sn_common_bus_fixup: No provider registered for this asic type, %d", + prom_bussoft_ptr->bs_asic_type); + + if (provider->bus_fixup) + provider_soft = (*provider->bus_fixup) (prom_bussoft_ptr, + controller); + else + provider_soft = NULL; + + /* + * Generic bus fixup goes here. Don't reference prom_bussoft_ptr + * after this point. + */ + controller->platform_data = kzalloc(sizeof(struct sn_platform_data), + GFP_KERNEL); + if (controller->platform_data == NULL) + BUG(); + sn_platform_data = + (struct sn_platform_data *) controller->platform_data; + sn_platform_data->provider_soft = provider_soft; + INIT_LIST_HEAD(&((struct sn_platform_data *) + controller->platform_data)->pcidev_info); + nasid = NASID_GET(SN_PCIBUS_BUSSOFT(bus)->bs_base); + cnode = nasid_to_cnodeid(nasid); + hubdev_info = (struct hubdev_info *)(NODEPDA(cnode)->pdinfo); + SN_PCIBUS_BUSSOFT(bus)->bs_xwidget_info = + &(hubdev_info->hdi_xwidget_info[SN_PCIBUS_BUSSOFT(bus)->bs_xid]); + + /* + * If the node information we obtained during the fixup phase is + * invalid then set controller->node to -1 (undetermined) + */ + if (controller->node >= num_online_nodes()) { + struct pcibus_bussoft *b = SN_PCIBUS_BUSSOFT(bus); + + printk(KERN_WARNING "Device ASIC=%u XID=%u PBUSNUM=%u" + "L_IO=%lx L_MEM=%lx BASE=%lx\n", + b->bs_asic_type, b->bs_xid, b->bs_persist_busnum, + b->bs_legacy_io, b->bs_legacy_mem, b->bs_base); + printk(KERN_WARNING "on node %d but only %d nodes online." + "Association set to undetermined.\n", + controller->node, num_online_nodes()); + controller->node = -1; + } +} + +void sn_bus_store_sysdata(struct pci_dev *dev) +{ + struct sysdata_el *element; + + element = kzalloc(sizeof(struct sysdata_el), GFP_KERNEL); + if (!element) { + dev_dbg(dev, "%s: out of memory!\n", __FUNCTION__); + return; + } + element->sysdata = SN_PCIDEV_INFO(dev); + list_add(&element->entry, &sn_sysdata_list); +} + +void sn_bus_free_sysdata(void) +{ + struct sysdata_el *element; + struct list_head *list, *safe; + + list_for_each_safe(list, safe, &sn_sysdata_list) { + element = list_entry(list, struct sysdata_el, entry); + list_del(&element->entry); + list_del(&(((struct pcidev_info *) + (element->sysdata))->pdi_list)); + kfree(element->sysdata); + kfree(element); + } + return; +} + +/* + * hubdev_init_node() - Creates the HUB data structure and link them to it's + * own NODE specific data area. + */ +void hubdev_init_node(nodepda_t * npda, cnodeid_t node) +{ + struct hubdev_info *hubdev_info; + int size; + pg_data_t *pg; + + size = sizeof(struct hubdev_info); + + if (node >= num_online_nodes()) /* Headless/memless IO nodes */ + pg = NODE_DATA(0); + else + pg = NODE_DATA(node); + + hubdev_info = (struct hubdev_info *)alloc_bootmem_node(pg, size); + + npda->pdinfo = (void *)hubdev_info; +} + +geoid_t +cnodeid_get_geoid(cnodeid_t cnode) +{ + struct hubdev_info *hubdev; + + hubdev = (struct hubdev_info *)(NODEPDA(cnode)->pdinfo); + return hubdev->hdi_geoid; +} + +void sn_generate_path(struct pci_bus *pci_bus, char *address) +{ + nasid_t nasid; + cnodeid_t cnode; + geoid_t geoid; + moduleid_t moduleid; + u16 bricktype; + + nasid = NASID_GET(SN_PCIBUS_BUSSOFT(pci_bus)->bs_base); + cnode = nasid_to_cnodeid(nasid); + geoid = cnodeid_get_geoid(cnode); + moduleid = geo_module(geoid); + + sprintf(address, "module_%c%c%c%c%.2d", + '0'+RACK_GET_CLASS(MODULE_GET_RACK(moduleid)), + '0'+RACK_GET_GROUP(MODULE_GET_RACK(moduleid)), + '0'+RACK_GET_NUM(MODULE_GET_RACK(moduleid)), + MODULE_GET_BTCHAR(moduleid), MODULE_GET_BPOS(moduleid)); + + /* Tollhouse requires slot id to be displayed */ + bricktype = MODULE_GET_BTYPE(moduleid); + if ((bricktype == L1_BRICKTYPE_191010) || + (bricktype == L1_BRICKTYPE_1932)) + sprintf(address, "%s^%d", address, geo_slot(geoid)); +} + +/* + * sn_pci_fixup_bus() - Perform SN specific setup of software structs + * (pcibus_bussoft, pcidev_info) and hardware + * registers, for the specified bus and devices under it. + */ +void __devinit +sn_pci_fixup_bus(struct pci_bus *bus) +{ + + if (SN_ACPI_BASE_SUPPORT()) + sn_acpi_bus_fixup(bus); + else + sn_bus_fixup(bus); +} + +/* + * sn_io_early_init - Perform early IO (and some non-IO) initialization. + * In particular, setup the sn_pci_provider[] array. + * This needs to be done prior to any bus scanning + * (acpi_scan_init()) in the ACPI case, as the SN + * bus fixup code will reference the array. + */ +static int __init +sn_io_early_init(void) +{ + int i; + + if (!ia64_platform_is("sn2") || IS_RUNNING_ON_FAKE_PROM()) + return 0; + + /* + * prime sn_pci_provider[]. Individial provider init routines will + * override their respective default entries. + */ + + for (i = 0; i < PCIIO_ASIC_MAX_TYPES; i++) + sn_pci_provider[i] = &sn_pci_default_provider; + + pcibr_init_provider(); + tioca_init_provider(); + tioce_init_provider(); + + /* + * This is needed to avoid bounce limit checks in the blk layer + */ + ia64_max_iommu_merge_mask = ~PAGE_MASK; + + sn_irq_lh_init(); + INIT_LIST_HEAD(&sn_sysdata_list); + sn_init_cpei_timer(); + +#ifdef CONFIG_PROC_FS + register_sn_procfs(); +#endif + + printk(KERN_INFO "ACPI DSDT OEM Rev 0x%x\n", + acpi_gbl_DSDT->oem_revision); + if (SN_ACPI_BASE_SUPPORT()) + sn_io_acpi_init(); + else + sn_io_init(); + return 0; +} + +arch_initcall(sn_io_early_init); + +/* + * sn_io_late_init() - Perform any final platform specific IO initialization. + */ + +int __init +sn_io_late_init(void) +{ + struct pci_bus *bus; + struct pcibus_bussoft *bussoft; + cnodeid_t cnode; + nasid_t nasid; + cnodeid_t near_cnode; + + if (!ia64_platform_is("sn2") || IS_RUNNING_ON_FAKE_PROM()) + return 0; + + /* + * Setup closest node in pci_controller->node for + * PIC, TIOCP, TIOCE (TIOCA does it during bus fixup using + * info from the PROM). + */ + bus = NULL; + while ((bus = pci_find_next_bus(bus)) != NULL) { + bussoft = SN_PCIBUS_BUSSOFT(bus); + nasid = NASID_GET(bussoft->bs_base); + cnode = nasid_to_cnodeid(nasid); + if ((bussoft->bs_asic_type == PCIIO_ASIC_TYPE_TIOCP) || + (bussoft->bs_asic_type == PCIIO_ASIC_TYPE_TIOCE)) { + /* TIO PCI Bridge: find nearest node with CPUs */ + int e = sn_hwperf_get_nearest_node(cnode, NULL, + &near_cnode); + if (e < 0) { + near_cnode = (cnodeid_t)-1; /* use any node */ + printk(KERN_WARNING "pcibr_bus_fixup: failed " + "to find near node with CPUs to TIO " + "node %d, err=%d\n", cnode, e); + } + PCI_CONTROLLER(bus)->node = near_cnode; + } else if (bussoft->bs_asic_type == PCIIO_ASIC_TYPE_PIC) { + PCI_CONTROLLER(bus)->node = cnode; + } + } + + sn_ioif_inited = 1; /* SN I/O infrastructure now initialized */ + + return 0; +} + +fs_initcall(sn_io_late_init); + +EXPORT_SYMBOL(sn_pci_fixup_slot); +EXPORT_SYMBOL(sn_pci_unfixup_slot); +EXPORT_SYMBOL(sn_bus_store_sysdata); +EXPORT_SYMBOL(sn_bus_free_sysdata); +EXPORT_SYMBOL(sn_generate_path); + diff --git a/arch/ia64/sn/kernel/io_init.c b/arch/ia64/sn/kernel/io_init.c index dc09a6a28a3..990224a4412 100644 --- a/arch/ia64/sn/kernel/io_init.c +++ b/arch/ia64/sn/kernel/io_init.c @@ -3,103 +3,28 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (C) 1992 - 1997, 2000-2005 Silicon Graphics, Inc. All rights reserved. + * Copyright (C) 1992 - 1997, 2000-2006 Silicon Graphics, Inc. All rights reserved. */ -#include -#include #include #include -#include -#include #include -#include #include -#include +#include #include #include -#include #include -#include -#include #include "xtalk/hubdev.h" -#include "xtalk/xwidgetdev.h" - - -extern void sn_init_cpei_timer(void); -extern void register_sn_procfs(void); - -static struct list_head sn_sysdata_list; - -/* sysdata list struct */ -struct sysdata_el { - struct list_head entry; - void *sysdata; -}; - -struct slab_info { - struct hubdev_info hubdev; -}; - -struct brick { - moduleid_t id; /* Module ID of this module */ - struct slab_info slab_info[MAX_SLABS + 1]; -}; - -int sn_ioif_inited; /* SN I/O infrastructure initialized? */ - -struct sn_pcibus_provider *sn_pci_provider[PCIIO_ASIC_MAX_TYPES]; /* indexed by asic type */ - -static int max_segment_number; /* Default highest segment number */ -static int max_pcibus_number = 255; /* Default highest pci bus number */ /* - * Hooks and struct for unsupported pci providers + * The code in this file will only be executed when running with + * a PROM that does _not_ have base ACPI IO support. + * (i.e., SN_ACPI_BASE_SUPPORT() == 0) */ -static dma_addr_t -sn_default_pci_map(struct pci_dev *pdev, unsigned long paddr, size_t size, int type) -{ - return 0; -} - -static void -sn_default_pci_unmap(struct pci_dev *pdev, dma_addr_t addr, int direction) -{ - return; -} - -static void * -sn_default_pci_bus_fixup(struct pcibus_bussoft *soft, struct pci_controller *controller) -{ - return NULL; -} - -static struct sn_pcibus_provider sn_pci_default_provider = { - .dma_map = sn_default_pci_map, - .dma_map_consistent = sn_default_pci_map, - .dma_unmap = sn_default_pci_unmap, - .bus_fixup = sn_default_pci_bus_fixup, -}; - -/* - * Retrieve the DMA Flush List given nasid, widget, and device. - * This list is needed to implement the WAR - Flush DMA data on PIO Reads. - */ -static inline u64 -sal_get_device_dmaflush_list(u64 nasid, u64 widget_num, u64 device_num, - u64 address) -{ - struct ia64_sal_retval ret_stuff; - ret_stuff.status = 0; - ret_stuff.v0 = 0; +static int max_segment_number; /* Default highest segment number */ +static int max_pcibus_number = 255; /* Default highest pci bus number */ - SAL_CALL_NOLOCK(ret_stuff, - (u64) SN_SAL_IOIF_GET_DEVICE_DMAFLUSH_LIST, - (u64) nasid, (u64) widget_num, - (u64) device_num, (u64) address, 0, 0, 0); - return ret_stuff.status; -} /* * Retrieve the hub device info structure for the given nasid. @@ -131,93 +56,20 @@ static inline u64 sal_get_pcibus_info(u64 segment, u64 busnum, u64 address) return ret_stuff.v0; } -/* - * Retrieve the pci device information given the bus and device|function number. - */ -static inline u64 -sal_get_pcidev_info(u64 segment, u64 bus_number, u64 devfn, u64 pci_dev, - u64 sn_irq_info) -{ - struct ia64_sal_retval ret_stuff; - ret_stuff.status = 0; - ret_stuff.v0 = 0; - - SAL_CALL_NOLOCK(ret_stuff, - (u64) SN_SAL_IOIF_GET_PCIDEV_INFO, - (u64) segment, (u64) bus_number, (u64) devfn, - (u64) pci_dev, - sn_irq_info, 0, 0); - return ret_stuff.v0; -} - -/* - * sn_pcidev_info_get() - Retrieve the pcidev_info struct for the specified - * device. - */ -inline struct pcidev_info * -sn_pcidev_info_get(struct pci_dev *dev) -{ - struct pcidev_info *pcidev; - - list_for_each_entry(pcidev, - &(SN_PCI_CONTROLLER(dev)->pcidev_info), pdi_list) { - if (pcidev->pdi_linux_pcidev == dev) { - return pcidev; - } - } - return NULL; -} - -/* Older PROM flush WAR - * - * 01/16/06 -- This war will be in place until a new official PROM is released. - * Additionally note that the struct sn_flush_device_war also has to be - * removed from arch/ia64/sn/include/xtalk/hubdev.h - */ -static u8 war_implemented = 0; - -static s64 sn_device_fixup_war(u64 nasid, u64 widget, int device, - struct sn_flush_device_common *common) -{ - struct sn_flush_device_war *war_list; - struct sn_flush_device_war *dev_entry; - struct ia64_sal_retval isrv = {0,0,0,0}; - - if (!war_implemented) { - printk(KERN_WARNING "PROM version < 4.50 -- implementing old " - "PROM flush WAR\n"); - war_implemented = 1; - } - - war_list = kzalloc(DEV_PER_WIDGET * sizeof(*war_list), GFP_KERNEL); - if (!war_list) - BUG(); - - SAL_CALL_NOLOCK(isrv, SN_SAL_IOIF_GET_WIDGET_DMAFLUSH_LIST, - nasid, widget, __pa(war_list), 0, 0, 0 ,0); - if (isrv.status) - panic("sn_device_fixup_war failed: %s\n", - ia64_sal_strerror(isrv.status)); - - dev_entry = war_list + device; - memcpy(common,dev_entry, sizeof(*common)); - kfree(war_list); - - return isrv.status; -} /* - * sn_fixup_ionodes() - This routine initializes the HUB data strcuture for - * each node in the system. + * sn_fixup_ionodes() - This routine initializes the HUB data structure for + * each node in the system. This function is only + * executed when running with a non-ACPI capable PROM. */ static void __init sn_fixup_ionodes(void) { - struct sn_flush_device_kernel *sn_flush_device_kernel; - struct sn_flush_device_kernel *dev_entry; + struct hubdev_info *hubdev; u64 status; u64 nasid; - int i, widget, device, size; + int i; + extern void sn_common_hubdev_init(struct hubdev_info *); /* * Get SGI Specific HUB chipset information. @@ -240,70 +92,47 @@ static void __init sn_fixup_ionodes(void) max_segment_number = hubdev->max_segment_number; max_pcibus_number = hubdev->max_pcibus_number; } + sn_common_hubdev_init(hubdev); + } +} - /* Attach the error interrupt handlers */ - if (nasid & 1) - ice_error_init(hubdev); - else - hub_error_init(hubdev); - - for (widget = 0; widget <= HUB_WIDGET_ID_MAX; widget++) - hubdev->hdi_xwidget_info[widget].xwi_hubinfo = hubdev; - - if (!hubdev->hdi_flush_nasid_list.widget_p) - continue; - - size = (HUB_WIDGET_ID_MAX + 1) * - sizeof(struct sn_flush_device_kernel *); - hubdev->hdi_flush_nasid_list.widget_p = - kzalloc(size, GFP_KERNEL); - if (!hubdev->hdi_flush_nasid_list.widget_p) +/* + * sn_pci_legacy_window_fixup - Create PCI controller windows for + * legacy IO and MEM space. This needs to + * be done here, as the PROM does not have + * ACPI support defining the root buses + * and their resources (_CRS), + */ +static void +sn_legacy_pci_window_fixup(struct pci_controller *controller, + u64 legacy_io, u64 legacy_mem) +{ + controller->window = kcalloc(2, sizeof(struct pci_window), + GFP_KERNEL); + if (controller->window == NULL) BUG(); - - for (widget = 0; widget <= HUB_WIDGET_ID_MAX; widget++) { - size = DEV_PER_WIDGET * - sizeof(struct sn_flush_device_kernel); - sn_flush_device_kernel = kzalloc(size, GFP_KERNEL); - if (!sn_flush_device_kernel) - BUG(); - - dev_entry = sn_flush_device_kernel; - for (device = 0; device < DEV_PER_WIDGET; - device++,dev_entry++) { - size = sizeof(struct sn_flush_device_common); - dev_entry->common = kzalloc(size, GFP_KERNEL); - if (!dev_entry->common) - BUG(); - - if (sn_prom_feature_available( - PRF_DEVICE_FLUSH_LIST)) - status = sal_get_device_dmaflush_list( - nasid, widget, device, - (u64)(dev_entry->common)); - else - status = sn_device_fixup_war(nasid, - widget, device, - dev_entry->common); - if (status != SALRET_OK) - panic("SAL call failed: %s\n", - ia64_sal_strerror(status)); - - spin_lock_init(&dev_entry->sfdl_flush_lock); - } - - if (sn_flush_device_kernel) - hubdev->hdi_flush_nasid_list.widget_p[widget] = - sn_flush_device_kernel; - } - } + controller->window[0].offset = legacy_io; + controller->window[0].resource.name = "legacy_io"; + controller->window[0].resource.flags = IORESOURCE_IO; + controller->window[0].resource.start = legacy_io; + controller->window[0].resource.end = + controller->window[0].resource.start + 0xffff; + controller->window[0].resource.parent = &ioport_resource; + controller->window[1].offset = legacy_mem; + controller->window[1].resource.name = "legacy_mem"; + controller->window[1].resource.flags = IORESOURCE_MEM; + controller->window[1].resource.start = legacy_mem; + controller->window[1].resource.end = + controller->window[1].resource.start + (1024 * 1024) - 1; + controller->window[1].resource.parent = &iomem_resource; + controller->windows = 2; } /* * sn_pci_window_fixup() - Create a pci_window for each device resource. - * Until ACPI support is added, we need this code - * to setup pci_windows for use by - * pcibios_bus_to_resource(), - * pcibios_resource_to_bus(), etc. + * It will setup pci_windows for use by + * pcibios_bus_to_resource(), pcibios_resource_to_bus(), + * etc. */ static void sn_pci_window_fixup(struct pci_dev *dev, unsigned int count, @@ -342,60 +171,22 @@ sn_pci_window_fixup(struct pci_dev *dev, unsigned int count, controller->window = new_window; } -void sn_pci_unfixup_slot(struct pci_dev *dev) -{ - struct pci_dev *host_pci_dev = SN_PCIDEV_INFO(dev)->host_pci_dev; - - sn_irq_unfixup(dev); - pci_dev_put(host_pci_dev); - pci_dev_put(dev); -} - /* - * sn_pci_fixup_slot() - This routine sets up a slot's resources - * consistent with the Linux PCI abstraction layer. Resources acquired - * from our PCI provider include PIO maps to BAR space and interrupt - * objects. + * sn_more_slot_fixup() - We are not running with an ACPI capable PROM, + * and need to convert the pci_dev->resource + * 'start' and 'end' addresses to mapped addresses, + * and setup the pci_controller->window array entries. */ -void sn_pci_fixup_slot(struct pci_dev *dev) +void +sn_more_slot_fixup(struct pci_dev *dev, struct pcidev_info *pcidev_info) { unsigned int count = 0; int idx; - int segment = pci_domain_nr(dev->bus); - int status = 0; - struct pcibus_bussoft *bs; - struct pci_bus *host_pci_bus; - struct pci_dev *host_pci_dev; - struct pcidev_info *pcidev_info; s64 pci_addrs[PCI_ROM_RESOURCE + 1]; - struct sn_irq_info *sn_irq_info; - unsigned long size; - unsigned int bus_no, devfn; - - pci_dev_get(dev); /* for the sysdata pointer */ - pcidev_info = kzalloc(sizeof(struct pcidev_info), GFP_KERNEL); - if (!pcidev_info) - BUG(); /* Cannot afford to run out of memory */ - - sn_irq_info = kzalloc(sizeof(struct sn_irq_info), GFP_KERNEL); - if (!sn_irq_info) - BUG(); /* Cannot afford to run out of memory */ - - /* Call to retrieve pci device information needed by kernel. */ - status = sal_get_pcidev_info((u64) segment, (u64) dev->bus->number, - dev->devfn, - (u64) __pa(pcidev_info), - (u64) __pa(sn_irq_info)); - if (status) - BUG(); /* Cannot get platform pci device information */ - - /* Add pcidev_info to list in sn_pci_controller struct */ - list_add_tail(&pcidev_info->pdi_list, - &(SN_PCI_CONTROLLER(dev->bus)->pcidev_info)); + unsigned long addr, end, size, start; /* Copy over PIO Mapped Addresses */ for (idx = 0; idx <= PCI_ROM_RESOURCE; idx++) { - unsigned long start, end, addr; if (!pcidev_info->pdi_pio_mapped_addr[idx]) { pci_addrs[idx] = -1; @@ -425,54 +216,19 @@ void sn_pci_fixup_slot(struct pci_dev *dev) */ if (count > 0) sn_pci_window_fixup(dev, count, pci_addrs); - - /* - * Using the PROMs values for the PCI host bus, get the Linux - * PCI host_pci_dev struct and set up host bus linkages - */ - - bus_no = (pcidev_info->pdi_slot_host_handle >> 32) & 0xff; - devfn = pcidev_info->pdi_slot_host_handle & 0xffffffff; - host_pci_bus = pci_find_bus(segment, bus_no); - host_pci_dev = pci_get_slot(host_pci_bus, devfn); - - pcidev_info->host_pci_dev = host_pci_dev; - pcidev_info->pdi_linux_pcidev = dev; - pcidev_info->pdi_host_pcidev_info = SN_PCIDEV_INFO(host_pci_dev); - bs = SN_PCIBUS_BUSSOFT(dev->bus); - pcidev_info->pdi_pcibus_info = bs; - - if (bs && bs->bs_asic_type < PCIIO_ASIC_MAX_TYPES) { - SN_PCIDEV_BUSPROVIDER(dev) = sn_pci_provider[bs->bs_asic_type]; - } else { - SN_PCIDEV_BUSPROVIDER(dev) = &sn_pci_default_provider; - } - - /* Only set up IRQ stuff if this device has a host bus context */ - if (bs && sn_irq_info->irq_irq) { - pcidev_info->pdi_sn_irq_info = sn_irq_info; - dev->irq = pcidev_info->pdi_sn_irq_info->irq_irq; - sn_irq_fixup(dev, sn_irq_info); - } else { - pcidev_info->pdi_sn_irq_info = NULL; - kfree(sn_irq_info); - } } /* * sn_pci_controller_fixup() - This routine sets up a bus's resources - * consistent with the Linux PCI abstraction layer. + * consistent with the Linux PCI abstraction layer. */ -void sn_pci_controller_fixup(int segment, int busnum, struct pci_bus *bus) +static void +sn_pci_controller_fixup(int segment, int busnum, struct pci_bus *bus) { - int status; - int nasid, cnode; + s64 status = 0; struct pci_controller *controller; - struct sn_pci_controller *sn_controller; struct pcibus_bussoft *prom_bussoft_ptr; - struct hubdev_info *hubdev_info; - void *provider_soft; - struct sn_pcibus_provider *provider; + status = sal_get_pcibus_info((u64) segment, (u64) busnum, (u64) ia64_tpa(&prom_bussoft_ptr)); @@ -480,261 +236,77 @@ void sn_pci_controller_fixup(int segment, int busnum, struct pci_bus *bus) return; /*bus # does not exist */ prom_bussoft_ptr = __va(prom_bussoft_ptr); - /* Allocate a sn_pci_controller, which has a pci_controller struct - * as the first member. - */ - sn_controller = kzalloc(sizeof(struct sn_pci_controller), GFP_KERNEL); - if (!sn_controller) + controller = kzalloc(sizeof(*controller), GFP_KERNEL); + if (!controller) BUG(); - INIT_LIST_HEAD(&sn_controller->pcidev_info); - controller = &sn_controller->pci_controller; controller->segment = segment; - if (bus == NULL) { - bus = pci_scan_bus(busnum, &pci_root_ops, controller); - if (bus == NULL) - goto error_return; /* error, or bus already scanned */ - bus->sysdata = NULL; - } - - if (bus->sysdata) - goto error_return; /* sysdata already alloc'd */ - /* - * Per-provider fixup. Copies the contents from prom to local - * area and links SN_PCIBUS_BUSSOFT(). + * Temporarily save the prom_bussoft_ptr for use by sn_bus_fixup(). + * (platform_data will be overwritten later in sn_common_bus_fixup()) */ + controller->platform_data = prom_bussoft_ptr; - if (prom_bussoft_ptr->bs_asic_type >= PCIIO_ASIC_MAX_TYPES) - goto error_return; /* unsupported asic type */ - - if (prom_bussoft_ptr->bs_asic_type == PCIIO_ASIC_TYPE_PPB) - goto error_return; /* no further fixup necessary */ - - provider = sn_pci_provider[prom_bussoft_ptr->bs_asic_type]; - if (provider == NULL) - goto error_return; /* no provider registerd for this asic */ + bus = pci_scan_bus(busnum, &pci_root_ops, controller); + if (bus == NULL) + goto error_return; /* error, or bus already scanned */ bus->sysdata = controller; - if (provider->bus_fixup) - provider_soft = (*provider->bus_fixup) (prom_bussoft_ptr, controller); - else - provider_soft = NULL; - - if (provider_soft == NULL) { - /* fixup failed or not applicable */ - bus->sysdata = NULL; - goto error_return; - } - - /* - * Setup pci_windows for legacy IO and MEM space. - * (Temporary until ACPI support is in place.) - */ - controller->window = kcalloc(2, sizeof(struct pci_window), GFP_KERNEL); - if (controller->window == NULL) - BUG(); - controller->window[0].offset = prom_bussoft_ptr->bs_legacy_io; - controller->window[0].resource.name = "legacy_io"; - controller->window[0].resource.flags = IORESOURCE_IO; - controller->window[0].resource.start = prom_bussoft_ptr->bs_legacy_io; - controller->window[0].resource.end = - controller->window[0].resource.start + 0xffff; - controller->window[0].resource.parent = &ioport_resource; - controller->window[1].offset = prom_bussoft_ptr->bs_legacy_mem; - controller->window[1].resource.name = "legacy_mem"; - controller->window[1].resource.flags = IORESOURCE_MEM; - controller->window[1].resource.start = prom_bussoft_ptr->bs_legacy_mem; - controller->window[1].resource.end = - controller->window[1].resource.start + (1024 * 1024) - 1; - controller->window[1].resource.parent = &iomem_resource; - controller->windows = 2; - - /* - * Generic bus fixup goes here. Don't reference prom_bussoft_ptr - * after this point. - */ - - PCI_CONTROLLER(bus)->platform_data = provider_soft; - nasid = NASID_GET(SN_PCIBUS_BUSSOFT(bus)->bs_base); - cnode = nasid_to_cnodeid(nasid); - hubdev_info = (struct hubdev_info *)(NODEPDA(cnode)->pdinfo); - SN_PCIBUS_BUSSOFT(bus)->bs_xwidget_info = - &(hubdev_info->hdi_xwidget_info[SN_PCIBUS_BUSSOFT(bus)->bs_xid]); - /* - * If the node information we obtained during the fixup phase is invalid - * then set controller->node to -1 (undetermined) - */ - if (controller->node >= num_online_nodes()) { - struct pcibus_bussoft *b = SN_PCIBUS_BUSSOFT(bus); - - printk(KERN_WARNING "Device ASIC=%u XID=%u PBUSNUM=%u" - "L_IO=%lx L_MEM=%lx BASE=%lx\n", - b->bs_asic_type, b->bs_xid, b->bs_persist_busnum, - b->bs_legacy_io, b->bs_legacy_mem, b->bs_base); - printk(KERN_WARNING "on node %d but only %d nodes online." - "Association set to undetermined.\n", - controller->node, num_online_nodes()); - controller->node = -1; - } return; error_return: - kfree(sn_controller); + kfree(controller); return; } -void sn_bus_store_sysdata(struct pci_dev *dev) +/* + * sn_bus_fixup + */ +void +sn_bus_fixup(struct pci_bus *bus) { - struct sysdata_el *element; - - element = kzalloc(sizeof(struct sysdata_el), GFP_KERNEL); - if (!element) { - dev_dbg(dev, "%s: out of memory!\n", __FUNCTION__); - return; - } - element->sysdata = SN_PCIDEV_INFO(dev); - list_add(&element->entry, &sn_sysdata_list); -} + struct pci_dev *pci_dev = NULL; + struct pcibus_bussoft *prom_bussoft_ptr; + extern void sn_common_bus_fixup(struct pci_bus *, + struct pcibus_bussoft *); + + + if (!bus->parent) { /* If root bus */ + prom_bussoft_ptr = PCI_CONTROLLER(bus)->platform_data; + if (prom_bussoft_ptr == NULL) { + printk(KERN_ERR + "sn_bus_fixup: 0x%04x:0x%02x Unable to " + "obtain prom_bussoft_ptr\n", + pci_domain_nr(bus), bus->number); + return; + } + sn_common_bus_fixup(bus, prom_bussoft_ptr); + sn_legacy_pci_window_fixup(PCI_CONTROLLER(bus), + prom_bussoft_ptr->bs_legacy_io, + prom_bussoft_ptr->bs_legacy_mem); + } + list_for_each_entry(pci_dev, &bus->devices, bus_list) { + sn_pci_fixup_slot(pci_dev); + } -void sn_bus_free_sysdata(void) -{ - struct sysdata_el *element; - struct list_head *list, *safe; - - list_for_each_safe(list, safe, &sn_sysdata_list) { - element = list_entry(list, struct sysdata_el, entry); - list_del(&element->entry); - list_del(&(((struct pcidev_info *) - (element->sysdata))->pdi_list)); - kfree(element->sysdata); - kfree(element); - } - return; } /* - * Ugly hack to get PCI setup until we have a proper ACPI namespace. + * sn_io_init - PROM does not have ACPI support to define nodes or root buses, + * so we need to do things the hard way, including initiating the + * bus scanning ourselves. */ -#define PCI_BUSES_TO_SCAN 256 - -static int __init sn_pci_init(void) +void __init sn_io_init(void) { int i, j; - struct pci_dev *pci_dev = NULL; - - if (!ia64_platform_is("sn2") || IS_RUNNING_ON_FAKE_PROM()) - return 0; - - /* - * prime sn_pci_provider[]. Individial provider init routines will - * override their respective default entries. - */ - - for (i = 0; i < PCIIO_ASIC_MAX_TYPES; i++) - sn_pci_provider[i] = &sn_pci_default_provider; - pcibr_init_provider(); - tioca_init_provider(); - tioce_init_provider(); - - /* - * This is needed to avoid bounce limit checks in the blk layer - */ - ia64_max_iommu_merge_mask = ~PAGE_MASK; sn_fixup_ionodes(); - sn_irq_lh_init(); - INIT_LIST_HEAD(&sn_sysdata_list); - sn_init_cpei_timer(); - -#ifdef CONFIG_PROC_FS - register_sn_procfs(); -#endif /* busses are not known yet ... */ for (i = 0; i <= max_segment_number; i++) for (j = 0; j <= max_pcibus_number; j++) sn_pci_controller_fixup(i, j, NULL); - - /* - * Generic Linux PCI Layer has created the pci_bus and pci_dev - * structures - time for us to add our SN PLatform specific - * information. - */ - - while ((pci_dev = - pci_get_device(PCI_ANY_ID, PCI_ANY_ID, pci_dev)) != NULL) - sn_pci_fixup_slot(pci_dev); - - sn_ioif_inited = 1; /* sn I/O infrastructure now initialized */ - - return 0; -} - -/* - * hubdev_init_node() - Creates the HUB data structure and link them to it's - * own NODE specific data area. - */ -void hubdev_init_node(nodepda_t * npda, cnodeid_t node) -{ - struct hubdev_info *hubdev_info; - int size; - pg_data_t *pg; - - size = sizeof(struct hubdev_info); - - if (node >= num_online_nodes()) /* Headless/memless IO nodes */ - pg = NODE_DATA(0); - else - pg = NODE_DATA(node); - - hubdev_info = (struct hubdev_info *)alloc_bootmem_node(pg, size); - - npda->pdinfo = (void *)hubdev_info; } - -geoid_t -cnodeid_get_geoid(cnodeid_t cnode) -{ - struct hubdev_info *hubdev; - - hubdev = (struct hubdev_info *)(NODEPDA(cnode)->pdinfo); - return hubdev->hdi_geoid; -} - -void sn_generate_path(struct pci_bus *pci_bus, char *address) -{ - nasid_t nasid; - cnodeid_t cnode; - geoid_t geoid; - moduleid_t moduleid; - u16 bricktype; - - nasid = NASID_GET(SN_PCIBUS_BUSSOFT(pci_bus)->bs_base); - cnode = nasid_to_cnodeid(nasid); - geoid = cnodeid_get_geoid(cnode); - moduleid = geo_module(geoid); - - sprintf(address, "module_%c%c%c%c%.2d", - '0'+RACK_GET_CLASS(MODULE_GET_RACK(moduleid)), - '0'+RACK_GET_GROUP(MODULE_GET_RACK(moduleid)), - '0'+RACK_GET_NUM(MODULE_GET_RACK(moduleid)), - MODULE_GET_BTCHAR(moduleid), MODULE_GET_BPOS(moduleid)); - - /* Tollhouse requires slot id to be displayed */ - bricktype = MODULE_GET_BTYPE(moduleid); - if ((bricktype == L1_BRICKTYPE_191010) || - (bricktype == L1_BRICKTYPE_1932)) - sprintf(address, "%s^%d", address, geo_slot(geoid)); -} - -subsys_initcall(sn_pci_init); -EXPORT_SYMBOL(sn_pci_fixup_slot); -EXPORT_SYMBOL(sn_pci_unfixup_slot); -EXPORT_SYMBOL(sn_pci_controller_fixup); -EXPORT_SYMBOL(sn_bus_store_sysdata); -EXPORT_SYMBOL(sn_bus_free_sysdata); -EXPORT_SYMBOL(sn_generate_path); diff --git a/arch/ia64/sn/kernel/iomv.c b/arch/ia64/sn/kernel/iomv.c index 7ce3cdad627..4aa4f301d56 100644 --- a/arch/ia64/sn/kernel/iomv.c +++ b/arch/ia64/sn/kernel/iomv.c @@ -3,10 +3,11 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (C) 2000-2003 Silicon Graphics, Inc. All rights reserved. + * Copyright (C) 2000-2003, 2006 Silicon Graphics, Inc. All rights reserved. */ #include +#include #include #include #include @@ -15,6 +16,7 @@ #include #include #include +#include #define IS_LEGACY_VGA_IOPORT(p) \ (((p) >= 0x3b0 && (p) <= 0x3bb) || ((p) >= 0x3c0 && (p) <= 0x3df)) @@ -31,11 +33,14 @@ void *sn_io_addr(unsigned long port) { if (!IS_RUNNING_ON_SIMULATOR()) { if (IS_LEGACY_VGA_IOPORT(port)) - port += vga_console_iobase; + return (__ia64_mk_io_addr(port)); /* On sn2, legacy I/O ports don't point at anything */ if (port < (64 * 1024)) return NULL; - return ((void *)(port | __IA64_UNCACHED_OFFSET)); + if (SN_ACPI_BASE_SUPPORT()) + return (__ia64_mk_io_addr(port)); + else + return ((void *)(port | __IA64_UNCACHED_OFFSET)); } else { /* but the simulator uses them... */ unsigned long addr; diff --git a/arch/ia64/sn/kernel/setup.c b/arch/ia64/sn/kernel/setup.c index 7a2d824c5ce..1d009f93244 100644 --- a/arch/ia64/sn/kernel/setup.c +++ b/arch/ia64/sn/kernel/setup.c @@ -388,6 +388,14 @@ void __init sn_setup(char **cmdline_p) ia64_sn_plat_set_error_handling_features(); // obsolete ia64_sn_set_os_feature(OSF_MCA_SLV_TO_OS_INIT_SLV); ia64_sn_set_os_feature(OSF_FEAT_LOG_SBES); + /* + * Note: The calls to notify the PROM of ACPI and PCI Segment + * support must be done prior to acpi_load_tables(), as + * an ACPI capable PROM will rebuild the DSDT as result + * of the call. + */ + ia64_sn_set_os_feature(OSF_PCISEGMENT_ENABLE); + ia64_sn_set_os_feature(OSF_ACPI_ENABLE); #if defined(CONFIG_VT) && defined(CONFIG_VGA_CONSOLE) @@ -413,6 +421,16 @@ void __init sn_setup(char **cmdline_p) if (! vga_console_membase) sn_scan_pcdp(); + /* + * Setup legacy IO space. + * vga_console_iobase maps to PCI IO Space address 0 on the + * bus containing the VGA console. + */ + if (vga_console_iobase) { + io_space[0].mmio_base = vga_console_iobase; + io_space[0].sparse = 0; + } + if (vga_console_membase) { /* usable vga ... make tty0 the preferred default console */ if (!strstr(*cmdline_p, "console=")) diff --git a/arch/ia64/sn/kernel/tiocx.c b/arch/ia64/sn/kernel/tiocx.c index feaf1a6e810..493380b2c05 100644 --- a/arch/ia64/sn/kernel/tiocx.c +++ b/arch/ia64/sn/kernel/tiocx.c @@ -552,7 +552,7 @@ static void __exit tiocx_exit(void) bus_unregister(&tiocx_bus_type); } -subsys_initcall(tiocx_init); +fs_initcall(tiocx_init); module_exit(tiocx_exit); /************************************************************************ diff --git a/arch/ia64/sn/pci/pcibr/pcibr_provider.c b/arch/ia64/sn/pci/pcibr/pcibr_provider.c index 27dd7df0f44..6846dc9b432 100644 --- a/arch/ia64/sn/pci/pcibr/pcibr_provider.c +++ b/arch/ia64/sn/pci/pcibr/pcibr_provider.c @@ -3,7 +3,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (C) 2001-2004 Silicon Graphics, Inc. All rights reserved. + * Copyright (C) 2001-2004, 2006 Silicon Graphics, Inc. All rights reserved. */ #include @@ -109,7 +109,6 @@ void * pcibr_bus_fixup(struct pcibus_bussoft *prom_bussoft, struct pci_controller *controller) { int nasid, cnode, j; - cnodeid_t near_cnode; struct hubdev_info *hubdev_info; struct pcibus_info *soft; struct sn_flush_device_kernel *sn_flush_device_kernel; @@ -186,20 +185,6 @@ pcibr_bus_fixup(struct pcibus_bussoft *prom_bussoft, struct pci_controller *cont return NULL; } - if (prom_bussoft->bs_asic_type == PCIIO_ASIC_TYPE_TIOCP) { - /* TIO PCI Bridge: find nearest node with CPUs */ - int e = sn_hwperf_get_nearest_node(cnode, NULL, &near_cnode); - - if (e < 0) { - near_cnode = (cnodeid_t)-1; /* use any node */ - printk(KERN_WARNING "pcibr_bus_fixup: failed to find " - "near node with CPUs to TIO node %d, err=%d\n", - cnode, e); - } - controller->node = near_cnode; - } - else - controller->node = cnode; return soft; } diff --git a/arch/ia64/sn/pci/tioce_provider.c b/arch/ia64/sn/pci/tioce_provider.c index 46e16dcf597..35f854fb612 100644 --- a/arch/ia64/sn/pci/tioce_provider.c +++ b/arch/ia64/sn/pci/tioce_provider.c @@ -15,7 +15,6 @@ #include #include #include -#include /* * 1/26/2006 @@ -990,8 +989,6 @@ tioce_target_interrupt(struct sn_irq_info *sn_irq_info) static void * tioce_bus_fixup(struct pcibus_bussoft *prom_bussoft, struct pci_controller *controller) { - int my_nasid; - cnodeid_t my_cnode, mem_cnode; struct tioce_common *tioce_common; struct tioce_kernel *tioce_kern; struct tioce __iomem *tioce_mmr; @@ -1035,21 +1032,6 @@ tioce_bus_fixup(struct pcibus_bussoft *prom_bussoft, struct pci_controller *cont tioce_common->ce_pcibus.bs_persist_segment, tioce_common->ce_pcibus.bs_persist_busnum); - /* - * identify closest nasid for memory allocations - */ - - my_nasid = NASID_GET(tioce_common->ce_pcibus.bs_base); - my_cnode = nasid_to_cnodeid(my_nasid); - - if (sn_hwperf_get_nearest_node(my_cnode, &mem_cnode, NULL) < 0) { - printk(KERN_WARNING "tioce_bus_fixup: failed to find " - "closest node with MEM to TIO node %d\n", my_cnode); - mem_cnode = (cnodeid_t)-1; /* use any node */ - } - - controller->node = mem_cnode; - return tioce_common; } diff --git a/include/asm-ia64/io.h b/include/asm-ia64/io.h index 855c30af72a..6311e168cd3 100644 --- a/include/asm-ia64/io.h +++ b/include/asm-ia64/io.h @@ -32,7 +32,7 @@ */ #define IO_SPACE_LIMIT 0xffffffffffffffffUL -#define MAX_IO_SPACES_BITS 4 +#define MAX_IO_SPACES_BITS 8 #define MAX_IO_SPACES (1UL << MAX_IO_SPACES_BITS) #define IO_SPACE_BITS 24 #define IO_SPACE_SIZE (1UL << IO_SPACE_BITS) diff --git a/include/asm-ia64/machvec.h b/include/asm-ia64/machvec.h index 7ffbddf5306..8f784f8e45b 100644 --- a/include/asm-ia64/machvec.h +++ b/include/asm-ia64/machvec.h @@ -36,6 +36,7 @@ typedef int ia64_mv_pci_legacy_read_t (struct pci_bus *, u16 port, u32 *val, typedef int ia64_mv_pci_legacy_write_t (struct pci_bus *, u16 port, u32 val, u8 size); typedef void ia64_mv_migrate_t(struct task_struct * task); +typedef void ia64_mv_pci_fixup_bus_t (struct pci_bus *); /* DMA-mapping interface: */ typedef void ia64_mv_dma_init (void); @@ -95,6 +96,11 @@ machvec_noop_task (struct task_struct *task) { } +static inline void +machvec_noop_bus (struct pci_bus *bus) +{ +} + extern void machvec_setup (char **); extern void machvec_timer_interrupt (int, void *); extern void machvec_dma_sync_single (struct device *, dma_addr_t, size_t, int); @@ -159,6 +165,7 @@ extern void machvec_tlb_migrate_finish (struct mm_struct *); # define platform_migrate ia64_mv.migrate # define platform_setup_msi_irq ia64_mv.setup_msi_irq # define platform_teardown_msi_irq ia64_mv.teardown_msi_irq +# define platform_pci_fixup_bus ia64_mv.pci_fixup_bus # endif /* __attribute__((__aligned__(16))) is required to make size of the @@ -210,6 +217,7 @@ struct ia64_machine_vector { ia64_mv_migrate_t *migrate; ia64_mv_setup_msi_irq_t *setup_msi_irq; ia64_mv_teardown_msi_irq_t *teardown_msi_irq; + ia64_mv_pci_fixup_bus_t *pci_fixup_bus; } __attribute__((__aligned__(16))); /* align attrib? see above comment */ #define MACHVEC_INIT(name) \ @@ -257,6 +265,7 @@ struct ia64_machine_vector { platform_migrate, \ platform_setup_msi_irq, \ platform_teardown_msi_irq, \ + platform_pci_fixup_bus, \ } extern struct ia64_machine_vector ia64_mv; @@ -416,5 +425,8 @@ extern int ia64_pci_legacy_write(struct pci_bus *bus, u16 port, u32 val, u8 size #ifndef platform_teardown_msi_irq # define platform_teardown_msi_irq ((ia64_mv_teardown_msi_irq_t*)NULL) #endif +#ifndef platform_pci_fixup_bus +# define platform_pci_fixup_bus machvec_noop_bus +#endif #endif /* _ASM_IA64_MACHVEC_H */ diff --git a/include/asm-ia64/machvec_sn2.h b/include/asm-ia64/machvec_sn2.h index c54b165b1c1..83325f6db03 100644 --- a/include/asm-ia64/machvec_sn2.h +++ b/include/asm-ia64/machvec_sn2.h @@ -69,6 +69,7 @@ extern ia64_mv_dma_supported sn_dma_supported; extern ia64_mv_migrate_t sn_migrate; extern ia64_mv_setup_msi_irq_t sn_setup_msi_irq; extern ia64_mv_teardown_msi_irq_t sn_teardown_msi_irq; +extern ia64_mv_pci_fixup_bus_t sn_pci_fixup_bus; /* @@ -127,6 +128,7 @@ extern ia64_mv_teardown_msi_irq_t sn_teardown_msi_irq; #define platform_setup_msi_irq ((ia64_mv_setup_msi_irq_t*)NULL) #define platform_teardown_msi_irq ((ia64_mv_teardown_msi_irq_t*)NULL) #endif +#define platform_pci_fixup_bus sn_pci_fixup_bus #include diff --git a/include/asm-ia64/sn/acpi.h b/include/asm-ia64/sn/acpi.h new file mode 100644 index 00000000000..2850a7ef5e7 --- /dev/null +++ b/include/asm-ia64/sn/acpi.h @@ -0,0 +1,16 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (C) 2006 Silicon Graphics, Inc. All rights reserved. + */ + +#ifndef _ASM_IA64_SN_ACPI_H +#define _ASM_IA64_SN_ACPI_H + +#include "acpi/acglobal.h" + +#define SN_ACPI_BASE_SUPPORT() (acpi_gbl_DSDT->oem_revision >= 0x20101) + +#endif /* _ASM_IA64_SN_ACPI_H */ diff --git a/include/asm-ia64/sn/pcidev.h b/include/asm-ia64/sn/pcidev.h index eac3561574b..9fe89a93d88 100644 --- a/include/asm-ia64/sn/pcidev.h +++ b/include/asm-ia64/sn/pcidev.h @@ -3,7 +3,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (C) 1992 - 1997, 2000-2005 Silicon Graphics, Inc. All rights reserved. + * Copyright (C) 1992 - 1997, 2000-2006 Silicon Graphics, Inc. All rights reserved. */ #ifndef _ASM_IA64_SN_PCI_PCIDEV_H #define _ASM_IA64_SN_PCI_PCIDEV_H @@ -12,31 +12,29 @@ /* * In ia64, pci_dev->sysdata must be a *pci_controller. To provide access to - * the pcidev_info structs for all devices under a controller, we extend the - * definition of pci_controller, via sn_pci_controller, to include a list - * of pcidev_info. + * the pcidev_info structs for all devices under a controller, we keep a + * list of pcidev_info under pci_controller->platform_data. */ -struct sn_pci_controller { - struct pci_controller pci_controller; +struct sn_platform_data { + void *provider_soft; struct list_head pcidev_info; }; -#define SN_PCI_CONTROLLER(dev) ((struct sn_pci_controller *) dev->sysdata) +#define SN_PLATFORM_DATA(busdev) \ + ((struct sn_platform_data *)(PCI_CONTROLLER(busdev)->platform_data)) #define SN_PCIDEV_INFO(dev) sn_pcidev_info_get(dev) -#define SN_PCIBUS_BUSSOFT_INFO(pci_bus) \ - (struct pcibus_info *)((struct pcibus_bussoft *)(PCI_CONTROLLER((pci_bus))->platform_data)) /* * Given a pci_bus, return the sn pcibus_bussoft struct. Note that * this only works for root busses, not for busses represented by PPB's. */ #define SN_PCIBUS_BUSSOFT(pci_bus) \ - ((struct pcibus_bussoft *)(PCI_CONTROLLER((pci_bus))->platform_data)) + ((struct pcibus_bussoft *)(SN_PLATFORM_DATA(pci_bus)->provider_soft)) #define SN_PCIBUS_BUSSOFT_INFO(pci_bus) \ - (struct pcibus_info *)((struct pcibus_bussoft *)(PCI_CONTROLLER((pci_bus))->platform_data)) + ((struct pcibus_info *)(SN_PLATFORM_DATA(pci_bus)->provider_soft)) /* * Given a struct pci_dev, return the sn pcibus_bussoft struct. Note * that this is not equivalent to SN_PCIBUS_BUSSOFT(pci_dev->bus) due @@ -72,8 +70,6 @@ extern void sn_irq_fixup(struct pci_dev *pci_dev, struct sn_irq_info *sn_irq_info); extern void sn_irq_unfixup(struct pci_dev *pci_dev); extern struct pcidev_info * sn_pcidev_info_get(struct pci_dev *); -extern void sn_pci_controller_fixup(int segment, int busnum, - struct pci_bus *bus); extern void sn_bus_store_sysdata(struct pci_dev *dev); extern void sn_bus_free_sysdata(void); extern void sn_generate_path(struct pci_bus *pci_bus, char *address); diff --git a/include/asm-ia64/sn/sn_feature_sets.h b/include/asm-ia64/sn/sn_feature_sets.h index 30dcfa442e5..bfdc36273ed 100644 --- a/include/asm-ia64/sn/sn_feature_sets.h +++ b/include/asm-ia64/sn/sn_feature_sets.h @@ -44,8 +44,14 @@ extern int sn_prom_feature_available(int id); * Once enabled, a feature cannot be disabled. * * By default, features are disabled unless explicitly enabled. + * + * These defines must be kept in sync with the corresponding + * PROM definitions in feature_sets.h. */ #define OSF_MCA_SLV_TO_OS_INIT_SLV 0 #define OSF_FEAT_LOG_SBES 1 +#define OSF_ACPI_ENABLE 2 +#define OSF_PCISEGMENT_ENABLE 3 + #endif /* _ASM_IA64_SN_FEATURE_SETS_H */ diff --git a/include/asm-ia64/sn/sn_sal.h b/include/asm-ia64/sn/sn_sal.h index ba826b3f75b..be5d83ad7cb 100644 --- a/include/asm-ia64/sn/sn_sal.h +++ b/include/asm-ia64/sn/sn_sal.h @@ -77,6 +77,7 @@ #define SN_SAL_IOIF_GET_WIDGET_DMAFLUSH_LIST 0x02000058 // deprecated #define SN_SAL_IOIF_GET_DEVICE_DMAFLUSH_LIST 0x0200005a +#define SN_SAL_IOIF_INIT 0x0200005f #define SN_SAL_HUB_ERROR_INTERRUPT 0x02000060 #define SN_SAL_BTE_RECOVER 0x02000061 #define SN_SAL_RESERVED_DO_NOT_USE 0x02000062 -- cgit v1.2.3-70-g09d2 From a2302c68d923537436b1114aa207787c1a31bd50 Mon Sep 17 00:00:00 2001 From: John Keller Date: Wed, 4 Oct 2006 16:49:52 -0500 Subject: Altix: Initial ACPI support - ROM shadowing. Support a shadowed ROM when running with an ACPI capable PROM. Define a new dev.resource flag IORESOURCE_ROM_BIOS_COPY to describe the case of a BIOS shadowed ROM, which can then be used to avoid pci_map_rom() making an unneeded call to pci_enable_rom(). Signed-off-by: John Keller Signed-off-by: Greg Kroah-Hartman --- arch/ia64/sn/kernel/io_acpi_init.c | 33 +++++++++++++++++++++++++++++++++ arch/ia64/sn/kernel/io_common.c | 5 +++-- arch/ia64/sn/kernel/io_init.c | 3 +++ drivers/pci/rom.c | 9 ++++++--- include/linux/ioport.h | 1 + 5 files changed, 46 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/arch/ia64/sn/kernel/io_acpi_init.c b/arch/ia64/sn/kernel/io_acpi_init.c index a9dc36901b1..99d7f278612 100644 --- a/arch/ia64/sn/kernel/io_acpi_init.c +++ b/arch/ia64/sn/kernel/io_acpi_init.c @@ -169,6 +169,39 @@ sn_acpi_bus_fixup(struct pci_bus *bus) } } +/* + * sn_acpi_slot_fixup - Perform any SN specific slot fixup. + * At present there does not appear to be + * any generic way to handle a ROM image + * that has been shadowed by the PROM, so + * we pass a pointer to it within the + * pcidev_info structure. + */ + +void +sn_acpi_slot_fixup(struct pci_dev *dev, struct pcidev_info *pcidev_info) +{ + void __iomem *addr; + size_t size; + + if (pcidev_info->pdi_pio_mapped_addr[PCI_ROM_RESOURCE]) { + /* + * A valid ROM image exists and has been shadowed by the + * PROM. Setup the pci_dev ROM resource to point to + * the shadowed copy. + */ + size = dev->resource[PCI_ROM_RESOURCE].end - + dev->resource[PCI_ROM_RESOURCE].start; + addr = + ioremap(pcidev_info->pdi_pio_mapped_addr[PCI_ROM_RESOURCE], + size); + dev->resource[PCI_ROM_RESOURCE].start = (unsigned long) addr; + dev->resource[PCI_ROM_RESOURCE].end = + (unsigned long) addr + size; + dev->resource[PCI_ROM_RESOURCE].flags |= IORESOURCE_ROM_BIOS_COPY; + } +} + static struct acpi_driver acpi_sn_hubdev_driver = { .name = "SGI HUBDEV Driver", .ids = "SGIHUB,SGITIO", diff --git a/arch/ia64/sn/kernel/io_common.c b/arch/ia64/sn/kernel/io_common.c index 12531de6754..d4dd8f4b6b8 100644 --- a/arch/ia64/sn/kernel/io_common.c +++ b/arch/ia64/sn/kernel/io_common.c @@ -286,9 +286,10 @@ void sn_pci_fixup_slot(struct pci_dev *dev) list_add_tail(&pcidev_info->pdi_list, &(SN_PLATFORM_DATA(dev->bus)->pcidev_info)); - if (!SN_ACPI_BASE_SUPPORT()) + if (SN_ACPI_BASE_SUPPORT()) + sn_acpi_slot_fixup(dev, pcidev_info); + else sn_more_slot_fixup(dev, pcidev_info); - /* * Using the PROMs values for the PCI host bus, get the Linux * PCI host_pci_dev struct and set up host bus linkages diff --git a/arch/ia64/sn/kernel/io_init.c b/arch/ia64/sn/kernel/io_init.c index 990224a4412..9ad843e0383 100644 --- a/arch/ia64/sn/kernel/io_init.c +++ b/arch/ia64/sn/kernel/io_init.c @@ -210,6 +210,9 @@ sn_more_slot_fixup(struct pci_dev *dev, struct pcidev_info *pcidev_info) dev->resource[idx].parent = &ioport_resource; else dev->resource[idx].parent = &iomem_resource; + /* If ROM, mark as shadowed in PROM */ + if (idx == PCI_ROM_RESOURCE) + dev->resource[idx].flags |= IORESOURCE_ROM_BIOS_COPY; } /* Create a pci_window in the pci_controller struct for * each device resource. diff --git a/drivers/pci/rom.c b/drivers/pci/rom.c index e1dcefc69bb..d087e081771 100644 --- a/drivers/pci/rom.c +++ b/drivers/pci/rom.c @@ -81,7 +81,8 @@ void __iomem *pci_map_rom(struct pci_dev *pdev, size_t *size) start = (loff_t)0xC0000; *size = 0x20000; /* cover C000:0 through E000:0 */ } else { - if (res->flags & IORESOURCE_ROM_COPY) { + if (res->flags & + (IORESOURCE_ROM_COPY | IORESOURCE_ROM_BIOS_COPY)) { *size = pci_resource_len(pdev, PCI_ROM_RESOURCE); return (void __iomem *)(unsigned long) pci_resource_start(pdev, PCI_ROM_RESOURCE); @@ -165,7 +166,8 @@ void __iomem *pci_map_rom_copy(struct pci_dev *pdev, size_t *size) if (!rom) return NULL; - if (res->flags & (IORESOURCE_ROM_COPY | IORESOURCE_ROM_SHADOW)) + if (res->flags & (IORESOURCE_ROM_COPY | IORESOURCE_ROM_SHADOW | + IORESOURCE_ROM_BIOS_COPY)) return rom; res->start = (unsigned long)kmalloc(*size, GFP_KERNEL); @@ -191,7 +193,7 @@ void pci_unmap_rom(struct pci_dev *pdev, void __iomem *rom) { struct resource *res = &pdev->resource[PCI_ROM_RESOURCE]; - if (res->flags & IORESOURCE_ROM_COPY) + if (res->flags & (IORESOURCE_ROM_COPY | IORESOURCE_ROM_BIOS_COPY)) return; iounmap(rom); @@ -215,6 +217,7 @@ void pci_remove_rom(struct pci_dev *pdev) sysfs_remove_bin_file(&pdev->dev.kobj, pdev->rom_attr); if (!(res->flags & (IORESOURCE_ROM_ENABLE | IORESOURCE_ROM_SHADOW | + IORESOURCE_ROM_BIOS_COPY | IORESOURCE_ROM_COPY))) pci_disable_rom(pdev); } diff --git a/include/linux/ioport.h b/include/linux/ioport.h index d42c8339907..cf8696d4a13 100644 --- a/include/linux/ioport.h +++ b/include/linux/ioport.h @@ -89,6 +89,7 @@ struct resource_list { #define IORESOURCE_ROM_ENABLE (1<<0) /* ROM is enabled, same as PCI_ROM_ADDRESS_ENABLE */ #define IORESOURCE_ROM_SHADOW (1<<1) /* ROM is copy at C000:0 */ #define IORESOURCE_ROM_COPY (1<<2) /* ROM is alloc'd copy, resource field overlaid */ +#define IORESOURCE_ROM_BIOS_COPY (1<<3) /* ROM is BIOS copy, resource field overlaid */ /* PC/ISA/whatever - the normal PC address spaces: IO and memory */ extern struct resource ioport_resource; -- cgit v1.2.3-70-g09d2 From bae94d02371c402408a4edfb95e71e88dbd3e973 Mon Sep 17 00:00:00 2001 From: Inaky Perez-Gonzalez Date: Wed, 22 Nov 2006 12:40:31 -0800 Subject: PCI: switch pci_{enable,disable}_device() to be nestable Changes the pci_{enable,disable}_device() functions to work in a nested basis, so that eg, three calls to enable_device() require three calls to disable_device(). The reason for this is to simplify PCI drivers for multi-interface/capability devices. These are devices that cram more than one interface in a single function. A relevant example of that is the Wireless [USB] Host Controller Interface (similar to EHCI) [see http://www.intel.com/technology/comms/wusb/whci.htm]. In these kind of devices, multiple interfaces are accessed through a single bar and IRQ line. For that, the drivers map only the smallest area of the bar to access their register banks and use shared IRQ handlers. However, because the order at which those drivers load cannot be known ahead of time, the sequence in which the calls to pci_enable_device() and pci_disable_device() cannot be predicted. Thus: 1. driverA starts pci_enable_device() 2. driverB starts pci_enable_device() 3. driverA shutdown pci_disable_device() 4. driverB shutdown pci_disable_device() between steps 3 and 4, driver B would loose access to it's device, even if it didn't intend to. By using this modification, the device won't be disabled until all the callers to enable() have called disable(). This is implemented by replacing 'struct pci_dev->is_enabled' from a bitfield to an atomic use count. Each caller to enable increments it, each caller to disable decrements it. When the count increments from 0 to 1, __pci_enable_device() is called to actually enable the device. When it drops to zero, pci_disable_device() actually does the disabling. We keep the backend __pci_enable_device() for pci_default_resume() to use and also change the sysfs method implementation, so that userspace enabling/disabling the device doesn't disable it one time too much. Signed-off-by: Inaky Perez-Gonzalez Signed-off-by: Greg Kroah-Hartman --- drivers/pci/pci-driver.c | 4 ++-- drivers/pci/pci-sysfs.c | 33 +++++++++++++++++++++------------ drivers/pci/pci.c | 40 +++++++++++++++++++++++++++++++++------- drivers/pci/pci.h | 1 + include/linux/pci.h | 3 ++- 5 files changed, 59 insertions(+), 22 deletions(-) (limited to 'include') diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index 84ec9c8f670..e5ae3a0c13b 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -329,8 +329,8 @@ static int pci_default_resume(struct pci_dev *pci_dev) /* restore the PCI config space */ pci_restore_state(pci_dev); /* if the device was enabled before suspend, reenable */ - if (pci_dev->is_enabled) - retval = pci_enable_device(pci_dev); + if (atomic_read(&pci_dev->enable_cnt)) + retval = __pci_enable_device(pci_dev); /* if the device was busmaster before the suspend, make it busmaster again */ if (pci_dev->is_busmaster) pci_set_master(pci_dev); diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index f952bfea48a..7a94076752d 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -42,7 +42,6 @@ pci_config_attr(subsystem_vendor, "0x%04x\n"); pci_config_attr(subsystem_device, "0x%04x\n"); pci_config_attr(class, "0x%06x\n"); pci_config_attr(irq, "%u\n"); -pci_config_attr(is_enabled, "%u\n"); static ssize_t broken_parity_status_show(struct device *dev, struct device_attribute *attr, @@ -112,26 +111,36 @@ static ssize_t modalias_show(struct device *dev, struct device_attribute *attr, (u8)(pci_dev->class >> 16), (u8)(pci_dev->class >> 8), (u8)(pci_dev->class)); } -static ssize_t -is_enabled_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) + +static ssize_t is_enabled_store(struct device *dev, + struct device_attribute *attr, const char *buf, + size_t count) { + ssize_t result = -EINVAL; struct pci_dev *pdev = to_pci_dev(dev); - int retval = 0; /* this can crash the machine when done on the "wrong" device */ if (!capable(CAP_SYS_ADMIN)) return count; - if (*buf == '0') - pci_disable_device(pdev); + if (*buf == '0') { + if (atomic_read(&pdev->enable_cnt) != 0) + pci_disable_device(pdev); + else + result = -EIO; + } else if (*buf == '1') + result = pci_enable_device(pdev); + + return result < 0 ? result : count; +} - if (*buf == '1') - retval = pci_enable_device(pdev); +static ssize_t is_enabled_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct pci_dev *pdev; - if (retval) - return retval; - return count; + pdev = to_pci_dev (dev); + return sprintf (buf, "%u\n", atomic_read(&pdev->enable_cnt)); } static ssize_t diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 427991741cf..5a14b73cf3a 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -612,29 +612,50 @@ pci_enable_device_bars(struct pci_dev *dev, int bars) } /** - * pci_enable_device - Initialize device before it's used by a driver. + * __pci_enable_device - Initialize device before it's used by a driver. * @dev: PCI device to be initialized * * Initialize device before it's used by a driver. Ask low-level code * to enable I/O and memory. Wake up the device if it was suspended. * Beware, this function can fail. + * + * Note this function is a backend and is not supposed to be called by + * normal code, use pci_enable_device() instead. */ int -pci_enable_device(struct pci_dev *dev) +__pci_enable_device(struct pci_dev *dev) { int err; - if (dev->is_enabled) - return 0; - err = pci_enable_device_bars(dev, (1 << PCI_NUM_RESOURCES) - 1); if (err) return err; pci_fixup_device(pci_fixup_enable, dev); - dev->is_enabled = 1; return 0; } +/** + * pci_enable_device - Initialize device before it's used by a driver. + * @dev: PCI device to be initialized + * + * Initialize device before it's used by a driver. Ask low-level code + * to enable I/O and memory. Wake up the device if it was suspended. + * Beware, this function can fail. + * + * Note we don't actually enable the device many times if we call + * this function repeatedly (we just increment the count). + */ +int pci_enable_device(struct pci_dev *dev) +{ + int result; + if (atomic_add_return(1, &dev->enable_cnt) > 1) + return 0; /* already enabled */ + result = __pci_enable_device(dev); + if (result < 0) + atomic_dec(&dev->enable_cnt); + return result; +} + /** * pcibios_disable_device - disable arch specific PCI resources for device dev * @dev: the PCI device to disable @@ -651,12 +672,18 @@ void __attribute__ ((weak)) pcibios_disable_device (struct pci_dev *dev) {} * * Signal to the system that the PCI device is not in use by the system * anymore. This only involves disabling PCI bus-mastering, if active. + * + * Note we don't actually disable the device until all callers of + * pci_device_enable() have called pci_device_disable(). */ void pci_disable_device(struct pci_dev *dev) { u16 pci_command; + if (atomic_sub_return(1, &dev->enable_cnt) != 0) + return; + if (dev->msi_enabled) disable_msi_mode(dev, pci_find_capability(dev, PCI_CAP_ID_MSI), PCI_CAP_ID_MSI); @@ -672,7 +699,6 @@ pci_disable_device(struct pci_dev *dev) dev->is_busmaster = 0; pcibios_disable_device(dev); - dev->is_enabled = 0; } /** diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 6bf327db5c5..398852f526a 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -1,5 +1,6 @@ /* Functions internal to the PCI core code */ +extern int __must_check __pci_enable_device(struct pci_dev *); extern int pci_uevent(struct device *dev, char **envp, int num_envp, char *buffer, int buffer_size); extern int pci_create_sysfs_dev_files(struct pci_dev *pdev); diff --git a/include/linux/pci.h b/include/linux/pci.h index 09be0f81b27..01c707261f9 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -51,6 +51,7 @@ #include #include #include +#include #include /* File state for mmap()s on /proc/bus/pci/X/Y */ @@ -159,7 +160,6 @@ struct pci_dev { unsigned int transparent:1; /* Transparent PCI bridge */ unsigned int multifunction:1;/* Part of multi-function device */ /* keep track of device state */ - unsigned int is_enabled:1; /* pci_enable_device has been called */ unsigned int is_busmaster:1; /* device is busmaster */ unsigned int no_msi:1; /* device may not use msi */ unsigned int no_d1d2:1; /* only allow d0 or d3 */ @@ -167,6 +167,7 @@ struct pci_dev { unsigned int broken_parity_status:1; /* Device generates false positive parity */ unsigned int msi_enabled:1; unsigned int msix_enabled:1; + atomic_t enable_cnt; /* pci_enable_device has been called */ u32 saved_config_space[16]; /* config space saved at suspend time */ struct hlist_head saved_cap_space; -- cgit v1.2.3-70-g09d2 From 3b59d52d8c7925e7a9a396f2e31a66eb060c6c37 Mon Sep 17 00:00:00 2001 From: Jason Gaston Date: Wed, 22 Nov 2006 15:15:08 -0800 Subject: PCI: irq: irq and pci_ids patch for Intel ICH9 This updated patch adds the Intel ICH9 LPC and SMBus Controller DID's. Signed-off-by: Jason Gaston Signed-off-by: Greg Kroah-Hartman --- arch/i386/pci/irq.c | 6 ++++++ include/linux/pci_ids.h | 7 +++++++ 2 files changed, 13 insertions(+) (limited to 'include') diff --git a/arch/i386/pci/irq.c b/arch/i386/pci/irq.c index 69163998ade..e65551cd821 100644 --- a/arch/i386/pci/irq.c +++ b/arch/i386/pci/irq.c @@ -543,6 +543,12 @@ static __init int intel_router_probe(struct irq_router *r, struct pci_dev *route case PCI_DEVICE_ID_INTEL_ICH8_2: case PCI_DEVICE_ID_INTEL_ICH8_3: case PCI_DEVICE_ID_INTEL_ICH8_4: + case PCI_DEVICE_ID_INTEL_ICH9_0: + case PCI_DEVICE_ID_INTEL_ICH9_1: + case PCI_DEVICE_ID_INTEL_ICH9_2: + case PCI_DEVICE_ID_INTEL_ICH9_3: + case PCI_DEVICE_ID_INTEL_ICH9_4: + case PCI_DEVICE_ID_INTEL_ICH9_5: r->name = "PIIX/ICH"; r->get = pirq_piix_get; r->set = pirq_piix_set; diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index fa4e1d79978..e060a763794 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2211,6 +2211,13 @@ #define PCI_DEVICE_ID_INTEL_ICH8_4 0x2815 #define PCI_DEVICE_ID_INTEL_ICH8_5 0x283e #define PCI_DEVICE_ID_INTEL_ICH8_6 0x2850 +#define PCI_DEVICE_ID_INTEL_ICH9_0 0x2910 +#define PCI_DEVICE_ID_INTEL_ICH9_1 0x2911 +#define PCI_DEVICE_ID_INTEL_ICH9_2 0x2912 +#define PCI_DEVICE_ID_INTEL_ICH9_3 0x2913 +#define PCI_DEVICE_ID_INTEL_ICH9_4 0x2914 +#define PCI_DEVICE_ID_INTEL_ICH9_5 0x2915 +#define PCI_DEVICE_ID_INTEL_ICH9_6 0x2930 #define PCI_DEVICE_ID_INTEL_82855PM_HB 0x3340 #define PCI_DEVICE_ID_INTEL_82830_HB 0x3575 #define PCI_DEVICE_ID_INTEL_82830_CGC 0x3577 -- cgit v1.2.3-70-g09d2 From 116af378201ef793424cd10508ccf18b06d8a021 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 25 Oct 2006 13:44:59 +1000 Subject: Driver core: add notification of bus events I finally did as you suggested and added the notifier to the struct bus_type itself. There are still problems to be expected is something attaches to a bus type where the code can hook in different struct device sub-classes (which is imho a big bogosity but I won't even try to argue that case now) but it will solve nicely a number of issues I've had so far. That also means that clients interested in registering for such notifications have to do it before devices are added and after bus types are registered. Fortunately, most bus types that matter for the various usage scenarios I have in mind are registerd at postcore_initcall time, which means I have a really nice spot at arch_initcall time to add my notifiers. There are 4 notifications provided. Device being added (before hooked to the bus) and removed (failure of previous case or after being unhooked from the bus), along with driver being bound to a device and about to be unbound. The usage I have for these are: - The 2 first ones are used to maintain a struct device_ext that is hooked to struct device.firmware_data. This structure contains for now a pointer to the Open Firmware node related to the device (if any), the NUMA node ID (for quick access to it) and the DMA operations pointers & iommu table instance for DMA to/from this device. For bus types I own (like IBM VIO or EBUS), I just maintain that structure directly from the bus code when creating the devices. But for bus types managed by generic code like PCI or platform (actually, of_platform which is a variation of platform linked to Open Firmware device-tree), I need this notifier. - The other two ones have a completely different usage scenario. I have cases where multiple devices and their drivers depend on each other. For example, the IBM EMAC network driver needs to attach to a MAL DMA engine which is a separate device, and a PHY interface which is also a separate device. They are all of_platform_device's (well, about to be with my upcoming patches) but there is no say in what precise order the core will "probe" them and instanciate the various modules. The solution I found for that is to have the drivers for emac to use multithread_probe, and wait for a driver to be bound to the target MAL and PHY control devices (the device-tree contains reference to the MAL and PHY interface nodes, which I can then match to of_platform_devices). Right now, I've been polling, but with that notifier, I can more cleanly wait (with a timeout of course). Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Greg Kroah-Hartman --- drivers/base/bus.c | 14 ++++++++++++++ drivers/base/core.c | 12 ++++++++++++ drivers/base/dd.c | 10 ++++++++++ include/linux/device.h | 25 +++++++++++++++++++++++++ 4 files changed, 61 insertions(+) (limited to 'include') diff --git a/drivers/base/bus.c b/drivers/base/bus.c index 7d8a7ce73fb..ed3e8a2be64 100644 --- a/drivers/base/bus.c +++ b/drivers/base/bus.c @@ -724,6 +724,8 @@ int bus_register(struct bus_type * bus) { int retval; + BLOCKING_INIT_NOTIFIER_HEAD(&bus->bus_notifier); + retval = kobject_set_name(&bus->subsys.kset.kobj, "%s", bus->name); if (retval) goto out; @@ -782,6 +784,18 @@ void bus_unregister(struct bus_type * bus) subsystem_unregister(&bus->subsys); } +int bus_register_notifier(struct bus_type *bus, struct notifier_block *nb) +{ + return blocking_notifier_chain_register(&bus->bus_notifier, nb); +} +EXPORT_SYMBOL_GPL(bus_register_notifier); + +int bus_unregister_notifier(struct bus_type *bus, struct notifier_block *nb) +{ + return blocking_notifier_chain_unregister(&bus->bus_notifier, nb); +} +EXPORT_SYMBOL_GPL(bus_unregister_notifier); + int __init buses_init(void) { return subsystem_register(&bus_subsys); diff --git a/drivers/base/core.c b/drivers/base/core.c index 002fde46d38..d4f35d8902a 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -17,6 +17,7 @@ #include #include #include +#include #include @@ -428,6 +429,11 @@ int device_add(struct device *dev) if (platform_notify) platform_notify(dev); + /* notify clients of device entry (new way) */ + if (dev->bus) + blocking_notifier_call_chain(&dev->bus->bus_notifier, + BUS_NOTIFY_ADD_DEVICE, dev); + dev->uevent_attr.attr.name = "uevent"; dev->uevent_attr.attr.mode = S_IWUSR; if (dev->driver) @@ -504,6 +510,9 @@ int device_add(struct device *dev) BusError: device_pm_remove(dev); PMError: + if (dev->bus) + blocking_notifier_call_chain(&dev->bus->bus_notifier, + BUS_NOTIFY_DEL_DEVICE, dev); device_remove_groups(dev); GroupError: device_remove_attrs(dev); @@ -622,6 +631,9 @@ void device_del(struct device * dev) */ if (platform_notify_remove) platform_notify_remove(dev); + if (dev->bus) + blocking_notifier_call_chain(&dev->bus->bus_notifier, + BUS_NOTIFY_DEL_DEVICE, dev); bus_remove_device(dev); device_pm_remove(dev); kobject_uevent(&dev->kobj, KOBJ_REMOVE); diff --git a/drivers/base/dd.c b/drivers/base/dd.c index c5d6bb4290a..9c88b1e34bc 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -52,6 +52,11 @@ int device_bind_driver(struct device *dev) pr_debug("bound device '%s' to driver '%s'\n", dev->bus_id, dev->driver->name); + + if (dev->bus) + blocking_notifier_call_chain(&dev->bus->bus_notifier, + BUS_NOTIFY_BOUND_DRIVER, dev); + klist_add_tail(&dev->knode_driver, &dev->driver->klist_devices); ret = sysfs_create_link(&dev->driver->kobj, &dev->kobj, kobject_name(&dev->kobj)); @@ -288,6 +293,11 @@ static void __device_release_driver(struct device * dev) sysfs_remove_link(&dev->kobj, "driver"); klist_remove(&dev->knode_driver); + if (dev->bus) + blocking_notifier_call_chain(&dev->bus->bus_notifier, + BUS_NOTIFY_UNBIND_DRIVER, + dev); + if (dev->bus && dev->bus->remove) dev->bus->remove(dev); else if (drv->remove) diff --git a/include/linux/device.h b/include/linux/device.h index 9d4f6a96393..b00e0271139 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -42,6 +42,8 @@ struct bus_type { struct klist klist_devices; struct klist klist_drivers; + struct blocking_notifier_head bus_notifier; + struct bus_attribute * bus_attrs; struct device_attribute * dev_attrs; struct driver_attribute * drv_attrs; @@ -75,6 +77,29 @@ int __must_check bus_for_each_drv(struct bus_type *bus, struct device_driver *start, void *data, int (*fn)(struct device_driver *, void *)); +/* + * Bus notifiers: Get notified of addition/removal of devices + * and binding/unbinding of drivers to devices. + * In the long run, it should be a replacement for the platform + * notify hooks. + */ +struct notifier_block; + +extern int bus_register_notifier(struct bus_type *bus, + struct notifier_block *nb); +extern int bus_unregister_notifier(struct bus_type *bus, + struct notifier_block *nb); + +/* All 4 notifers below get called with the target struct device * + * as an argument. Note that those functions are likely to be called + * with the device semaphore held in the core, so be careful. + */ +#define BUS_NOTIFY_ADD_DEVICE 0x00000001 /* device added */ +#define BUS_NOTIFY_DEL_DEVICE 0x00000002 /* device removed */ +#define BUS_NOTIFY_BOUND_DRIVER 0x00000003 /* driver bound to device */ +#define BUS_NOTIFY_UNBIND_DRIVER 0x00000004 /* driver about to be + unbound */ + /* driverfs interface for exporting bus attributes */ struct bus_attribute { -- cgit v1.2.3-70-g09d2 From f0ee61a6cecd100301a60d99feb187776533b2a2 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 23 Oct 2006 10:40:54 -0700 Subject: Driver Core: Move virtual_device_parent() to core.c It doesn't need to be global or in device.h Cc: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- drivers/base/class.c | 17 ----------------- drivers/base/core.c | 17 +++++++++++++++++ include/linux/device.h | 2 -- 3 files changed, 17 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/drivers/base/class.c b/drivers/base/class.c index 0ff267a248d..2e705f6abb4 100644 --- a/drivers/base/class.c +++ b/drivers/base/class.c @@ -893,23 +893,6 @@ void class_interface_unregister(struct class_interface *class_intf) class_put(parent); } -int virtual_device_parent(struct device *dev) -{ - if (!dev->class) - return -ENODEV; - - if (!dev->class->virtual_dir) { - static struct kobject *virtual_dir = NULL; - - if (!virtual_dir) - virtual_dir = kobject_add_dir(&devices_subsys.kset.kobj, "virtual"); - dev->class->virtual_dir = kobject_add_dir(virtual_dir, dev->class->name); - } - - dev->kobj.parent = dev->class->virtual_dir; - return 0; -} - int __init classes_init(void) { int retval; diff --git a/drivers/base/core.c b/drivers/base/core.c index d4f35d8902a..dbcd40b987d 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -384,6 +384,23 @@ void device_initialize(struct device *dev) device_init_wakeup(dev, 0); } +static int virtual_device_parent(struct device *dev) +{ + if (!dev->class) + return -ENODEV; + + if (!dev->class->virtual_dir) { + static struct kobject *virtual_dir = NULL; + + if (!virtual_dir) + virtual_dir = kobject_add_dir(&devices_subsys.kset.kobj, "virtual"); + dev->class->virtual_dir = kobject_add_dir(virtual_dir, dev->class->name); + } + + dev->kobj.parent = dev->class->virtual_dir; + return 0; +} + /** * device_add - add device to device hierarchy. * @dev: device. diff --git a/include/linux/device.h b/include/linux/device.h index b00e0271139..00b29e0c5ce 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -440,8 +440,6 @@ extern struct device *device_create(struct class *cls, struct device *parent, __attribute__((format(printf,4,5))); extern void device_destroy(struct class *cls, dev_t devt); -extern int virtual_device_parent(struct device *dev); - /* * Platform "fixup" functions - allow the platform to have their say * about devices and actions that the general device layer doesn't -- cgit v1.2.3-70-g09d2 From 94fbcded4ea0dc14cbfb222a5c68372f150d1476 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 27 Jul 2006 16:16:04 -0700 Subject: Driver core: change misc class_devices to be real devices This also ment that some of the misc drivers had to also be fixed up as they were assuming the device was a class_device. Signed-off-by: Greg Kroah-Hartman --- drivers/char/hw_random/core.c | 38 +++++++++++++++++++------------------- drivers/char/misc.c | 13 ++++--------- drivers/char/tpm/tpm.c | 2 +- drivers/input/serio/serio_raw.c | 2 +- include/linux/miscdevice.h | 5 ++--- 5 files changed, 27 insertions(+), 33 deletions(-) (limited to 'include') diff --git a/drivers/char/hw_random/core.c b/drivers/char/hw_random/core.c index 154a81d328c..ebace201bec 100644 --- a/drivers/char/hw_random/core.c +++ b/drivers/char/hw_random/core.c @@ -162,7 +162,8 @@ static struct miscdevice rng_miscdev = { }; -static ssize_t hwrng_attr_current_store(struct class_device *class, +static ssize_t hwrng_attr_current_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t len) { int err; @@ -192,7 +193,8 @@ static ssize_t hwrng_attr_current_store(struct class_device *class, return err ? : len; } -static ssize_t hwrng_attr_current_show(struct class_device *class, +static ssize_t hwrng_attr_current_show(struct device *dev, + struct device_attribute *attr, char *buf) { int err; @@ -210,7 +212,8 @@ static ssize_t hwrng_attr_current_show(struct class_device *class, return ret; } -static ssize_t hwrng_attr_available_show(struct class_device *class, +static ssize_t hwrng_attr_available_show(struct device *dev, + struct device_attribute *attr, char *buf) { int err; @@ -234,20 +237,18 @@ static ssize_t hwrng_attr_available_show(struct class_device *class, return ret; } -static CLASS_DEVICE_ATTR(rng_current, S_IRUGO | S_IWUSR, - hwrng_attr_current_show, - hwrng_attr_current_store); -static CLASS_DEVICE_ATTR(rng_available, S_IRUGO, - hwrng_attr_available_show, - NULL); +static DEVICE_ATTR(rng_current, S_IRUGO | S_IWUSR, + hwrng_attr_current_show, + hwrng_attr_current_store); +static DEVICE_ATTR(rng_available, S_IRUGO, + hwrng_attr_available_show, + NULL); static void unregister_miscdev(void) { - class_device_remove_file(rng_miscdev.class, - &class_device_attr_rng_available); - class_device_remove_file(rng_miscdev.class, - &class_device_attr_rng_current); + device_remove_file(rng_miscdev.this_device, &dev_attr_rng_available); + device_remove_file(rng_miscdev.this_device, &dev_attr_rng_current); misc_deregister(&rng_miscdev); } @@ -258,20 +259,19 @@ static int register_miscdev(void) err = misc_register(&rng_miscdev); if (err) goto out; - err = class_device_create_file(rng_miscdev.class, - &class_device_attr_rng_current); + err = device_create_file(rng_miscdev.this_device, + &dev_attr_rng_current); if (err) goto err_misc_dereg; - err = class_device_create_file(rng_miscdev.class, - &class_device_attr_rng_available); + err = device_create_file(rng_miscdev.this_device, + &dev_attr_rng_available); if (err) goto err_remove_current; out: return err; err_remove_current: - class_device_remove_file(rng_miscdev.class, - &class_device_attr_rng_current); + device_remove_file(rng_miscdev.this_device, &dev_attr_rng_current); err_misc_dereg: misc_deregister(&rng_miscdev); goto out; diff --git a/drivers/char/misc.c b/drivers/char/misc.c index 62ebe09656e..7a484fc7cb9 100644 --- a/drivers/char/misc.c +++ b/drivers/char/misc.c @@ -169,11 +169,6 @@ fail: return err; } -/* - * TODO for 2.7: - * - add a struct kref to struct miscdevice and make all usages of - * them dynamic. - */ static struct class *misc_class; static const struct file_operations misc_fops = { @@ -228,10 +223,10 @@ int misc_register(struct miscdevice * misc) misc_minors[misc->minor >> 3] |= 1 << (misc->minor & 7); dev = MKDEV(MISC_MAJOR, misc->minor); - misc->class = class_device_create(misc_class, NULL, dev, misc->dev, + misc->this_device = device_create(misc_class, misc->parent, dev, "%s", misc->name); - if (IS_ERR(misc->class)) { - err = PTR_ERR(misc->class); + if (IS_ERR(misc->this_device)) { + err = PTR_ERR(misc->this_device); goto out; } @@ -264,7 +259,7 @@ int misc_deregister(struct miscdevice * misc) down(&misc_sem); list_del(&misc->list); - class_device_destroy(misc_class, MKDEV(MISC_MAJOR, misc->minor)); + device_destroy(misc_class, MKDEV(MISC_MAJOR, misc->minor)); if (i < DYNAMIC_MINORS && i>0) { misc_minors[i>>3] &= ~(1 << (misc->minor & 7)); } diff --git a/drivers/char/tpm/tpm.c b/drivers/char/tpm/tpm.c index 6ad2d3bb945..6e1329d404d 100644 --- a/drivers/char/tpm/tpm.c +++ b/drivers/char/tpm/tpm.c @@ -1130,7 +1130,7 @@ struct tpm_chip *tpm_register_hardware(struct device *dev, const struct tpm_vend scnprintf(devname, DEVNAME_SIZE, "%s%d", "tpm", chip->dev_num); chip->vendor.miscdev.name = devname; - chip->vendor.miscdev.dev = dev; + chip->vendor.miscdev.parent = dev; chip->dev = get_device(dev); if (misc_register(&chip->vendor.miscdev)) { diff --git a/drivers/input/serio/serio_raw.c b/drivers/input/serio/serio_raw.c index ba2a2035d64..7c8d0399ae8 100644 --- a/drivers/input/serio/serio_raw.c +++ b/drivers/input/serio/serio_raw.c @@ -297,7 +297,7 @@ static int serio_raw_connect(struct serio *serio, struct serio_driver *drv) serio_raw->dev.minor = PSMOUSE_MINOR; serio_raw->dev.name = serio_raw->name; - serio_raw->dev.dev = &serio->dev; + serio_raw->dev.parent = &serio->dev; serio_raw->dev.fops = &serio_raw_fops; err = misc_register(&serio_raw->dev); diff --git a/include/linux/miscdevice.h b/include/linux/miscdevice.h index b03cfb91e22..326da7d500c 100644 --- a/include/linux/miscdevice.h +++ b/include/linux/miscdevice.h @@ -31,15 +31,14 @@ #define HPET_MINOR 228 struct device; -struct class_device; struct miscdevice { int minor; const char *name; const struct file_operations *fops; struct list_head list; - struct device *dev; - struct class_device *class; + struct device *parent; + struct device *this_device; }; extern int misc_register(struct miscdevice * misc); -- cgit v1.2.3-70-g09d2 From 01107d343076c34b9e1ce5d073292bd7f3097fda Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 7 Aug 2006 22:19:37 -0700 Subject: Driver core: convert tty core to use struct device Converts from using struct "class_device" to "struct device" making everything show up properly in /sys/devices/ with symlinks from the /sys/class directory. Also fixes up the isdn drivers that were putting something in the class device's directory. Signed-off-by: Greg Kroah-Hartman --- drivers/char/tty_io.c | 19 ++++++++++--------- drivers/isdn/gigaset/common.c | 2 +- drivers/isdn/gigaset/gigaset.h | 2 +- drivers/isdn/gigaset/interface.c | 10 +++++----- drivers/isdn/gigaset/proc.c | 19 ++++++++++--------- include/linux/tty.h | 5 ++--- 6 files changed, 29 insertions(+), 28 deletions(-) (limited to 'include') diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index e90ea39c7c4..50dc49205a2 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -3612,7 +3612,8 @@ static struct class *tty_class; * This field is optional, if there is no known struct device * for this tty device it can be set to NULL safely. * - * Returns a pointer to the class device (or ERR_PTR(-EFOO) on error). + * Returns a pointer to the struct device for this tty device + * (or ERR_PTR(-EFOO) on error). * * This call is required to be made to register an individual tty device * if the tty driver's flags have the TTY_DRIVER_DYNAMIC_DEV bit set. If @@ -3622,8 +3623,8 @@ static struct class *tty_class; * Locking: ?? */ -struct class_device *tty_register_device(struct tty_driver *driver, - unsigned index, struct device *device) +struct device *tty_register_device(struct tty_driver *driver, unsigned index, + struct device *device) { char name[64]; dev_t dev = MKDEV(driver->major, driver->minor_start) + index; @@ -3639,7 +3640,7 @@ struct class_device *tty_register_device(struct tty_driver *driver, else tty_line_name(driver, index, name); - return class_device_create(tty_class, NULL, dev, device, "%s", name); + return device_create(tty_class, device, dev, name); } /** @@ -3655,7 +3656,7 @@ struct class_device *tty_register_device(struct tty_driver *driver, void tty_unregister_device(struct tty_driver *driver, unsigned index) { - class_device_destroy(tty_class, MKDEV(driver->major, driver->minor_start) + index); + device_destroy(tty_class, MKDEV(driver->major, driver->minor_start) + index); } EXPORT_SYMBOL(tty_register_device); @@ -3895,20 +3896,20 @@ static int __init tty_init(void) if (cdev_add(&tty_cdev, MKDEV(TTYAUX_MAJOR, 0), 1) || register_chrdev_region(MKDEV(TTYAUX_MAJOR, 0), 1, "/dev/tty") < 0) panic("Couldn't register /dev/tty driver\n"); - class_device_create(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 0), NULL, "tty"); + device_create(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 0), "tty"); cdev_init(&console_cdev, &console_fops); if (cdev_add(&console_cdev, MKDEV(TTYAUX_MAJOR, 1), 1) || register_chrdev_region(MKDEV(TTYAUX_MAJOR, 1), 1, "/dev/console") < 0) panic("Couldn't register /dev/console driver\n"); - class_device_create(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 1), NULL, "console"); + device_create(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 1), "console"); #ifdef CONFIG_UNIX98_PTYS cdev_init(&ptmx_cdev, &ptmx_fops); if (cdev_add(&ptmx_cdev, MKDEV(TTYAUX_MAJOR, 2), 1) || register_chrdev_region(MKDEV(TTYAUX_MAJOR, 2), 1, "/dev/ptmx") < 0) panic("Couldn't register /dev/ptmx driver\n"); - class_device_create(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 2), NULL, "ptmx"); + device_create(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 2), "ptmx"); #endif #ifdef CONFIG_VT @@ -3916,7 +3917,7 @@ static int __init tty_init(void) if (cdev_add(&vc0_cdev, MKDEV(TTY_MAJOR, 0), 1) || register_chrdev_region(MKDEV(TTY_MAJOR, 0), 1, "/dev/vc/0") < 0) panic("Couldn't register /dev/tty0 driver\n"); - class_device_create(tty_class, NULL, MKDEV(TTY_MAJOR, 0), NULL, "tty0"); + device_create(tty_class, NULL, MKDEV(TTY_MAJOR, 0), "tty0"); vty_init(); #endif diff --git a/drivers/isdn/gigaset/common.c b/drivers/isdn/gigaset/common.c index 5800beeebb8..defd5743dba 100644 --- a/drivers/isdn/gigaset/common.c +++ b/drivers/isdn/gigaset/common.c @@ -702,7 +702,7 @@ struct cardstate *gigaset_initcs(struct gigaset_driver *drv, int channels, cs->open_count = 0; cs->dev = NULL; cs->tty = NULL; - cs->class = NULL; + cs->tty_dev = NULL; cs->cidmode = cidmode != 0; //if(onechannel) { //FIXME diff --git a/drivers/isdn/gigaset/gigaset.h b/drivers/isdn/gigaset/gigaset.h index 884bd72c1bf..06298cc52bf 100644 --- a/drivers/isdn/gigaset/gigaset.h +++ b/drivers/isdn/gigaset/gigaset.h @@ -444,7 +444,7 @@ struct cardstate { struct gigaset_driver *driver; unsigned minor_index; struct device *dev; - struct class_device *class; + struct device *tty_dev; const struct gigaset_ops *ops; diff --git a/drivers/isdn/gigaset/interface.c b/drivers/isdn/gigaset/interface.c index 596f3aebe2f..7edea015867 100644 --- a/drivers/isdn/gigaset/interface.c +++ b/drivers/isdn/gigaset/interface.c @@ -625,13 +625,13 @@ void gigaset_if_init(struct cardstate *cs) return; tasklet_init(&cs->if_wake_tasklet, &if_wake, (unsigned long) cs); - cs->class = tty_register_device(drv->tty, cs->minor_index, NULL); + cs->tty_dev = tty_register_device(drv->tty, cs->minor_index, NULL); - if (!IS_ERR(cs->class)) - class_set_devdata(cs->class, cs); + if (!IS_ERR(cs->tty_dev)) + dev_set_drvdata(cs->tty_dev, cs); else { warn("could not register device to the tty subsystem"); - cs->class = NULL; + cs->tty_dev = NULL; } } @@ -645,7 +645,7 @@ void gigaset_if_free(struct cardstate *cs) tasklet_disable(&cs->if_wake_tasklet); tasklet_kill(&cs->if_wake_tasklet); - cs->class = NULL; + cs->tty_dev = NULL; tty_unregister_device(drv->tty, cs->minor_index); } diff --git a/drivers/isdn/gigaset/proc.c b/drivers/isdn/gigaset/proc.c index 9ad840e95db..e767afa55ab 100644 --- a/drivers/isdn/gigaset/proc.c +++ b/drivers/isdn/gigaset/proc.c @@ -16,11 +16,12 @@ #include "gigaset.h" #include -static ssize_t show_cidmode(struct class_device *class, char *buf) +static ssize_t show_cidmode(struct device *dev, + struct device_attribute *attr, char *buf) { int ret; unsigned long flags; - struct cardstate *cs = class_get_devdata(class); + struct cardstate *cs = dev_get_drvdata(dev); spin_lock_irqsave(&cs->lock, flags); ret = sprintf(buf, "%u\n", cs->cidmode); @@ -29,10 +30,10 @@ static ssize_t show_cidmode(struct class_device *class, char *buf) return ret; } -static ssize_t set_cidmode(struct class_device *class, +static ssize_t set_cidmode(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct cardstate *cs = class_get_devdata(class); + struct cardstate *cs = dev_get_drvdata(dev); long int value; char *end; @@ -64,25 +65,25 @@ static ssize_t set_cidmode(struct class_device *class, return count; } -static CLASS_DEVICE_ATTR(cidmode, S_IRUGO|S_IWUSR, show_cidmode, set_cidmode); +static DEVICE_ATTR(cidmode, S_IRUGO|S_IWUSR, show_cidmode, set_cidmode); /* free sysfs for device */ void gigaset_free_dev_sysfs(struct cardstate *cs) { - if (!cs->class) + if (!cs->tty_dev) return; gig_dbg(DEBUG_INIT, "removing sysfs entries"); - class_device_remove_file(cs->class, &class_device_attr_cidmode); + device_remove_file(cs->tty_dev, &dev_attr_cidmode); } /* initialize sysfs for device */ void gigaset_init_dev_sysfs(struct cardstate *cs) { - if (!cs->class) + if (!cs->tty_dev) return; gig_dbg(DEBUG_INIT, "setting up sysfs"); - if (class_device_create_file(cs->class, &class_device_attr_cidmode)) + if (device_create_file(cs->tty_dev, &dev_attr_cidmode)) dev_err(cs->dev, "could not create sysfs attribute\n"); } diff --git a/include/linux/tty.h b/include/linux/tty.h index 44091c0db0b..65321f911c1 100644 --- a/include/linux/tty.h +++ b/include/linux/tty.h @@ -276,9 +276,8 @@ extern int tty_register_ldisc(int disc, struct tty_ldisc *new_ldisc); extern int tty_unregister_ldisc(int disc); extern int tty_register_driver(struct tty_driver *driver); extern int tty_unregister_driver(struct tty_driver *driver); -extern struct class_device *tty_register_device(struct tty_driver *driver, - unsigned index, - struct device *dev); +extern struct device *tty_register_device(struct tty_driver *driver, + unsigned index, struct device *dev); extern void tty_unregister_device(struct tty_driver *driver, unsigned index); extern int tty_read_raw_data(struct tty_struct *tty, unsigned char *bufp, int buflen); -- cgit v1.2.3-70-g09d2 From fcaf71fd51f9cfc504455d3e19ec242e4b2073ed Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Tue, 12 Sep 2006 17:00:10 +0200 Subject: Driver core: convert mmc code to use struct device Converts from using struct "class_device" to "struct device" making everything show up properly in /sys/devices/ with symlinks from the /sys/class directory. Signed-off-by: Greg Kroah-Hartman --- drivers/mmc/mmc_queue.c | 4 ++-- drivers/mmc/mmc_sysfs.c | 20 ++++++++++---------- drivers/mmc/wbsd.c | 6 +++--- include/linux/mmc/host.h | 8 ++++---- 4 files changed, 19 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/drivers/mmc/mmc_queue.c b/drivers/mmc/mmc_queue.c index 4ccdd82b680..61a1de85cb2 100644 --- a/drivers/mmc/mmc_queue.c +++ b/drivers/mmc/mmc_queue.c @@ -130,8 +130,8 @@ int mmc_init_queue(struct mmc_queue *mq, struct mmc_card *card, spinlock_t *lock u64 limit = BLK_BOUNCE_HIGH; int ret; - if (host->dev->dma_mask && *host->dev->dma_mask) - limit = *host->dev->dma_mask; + if (mmc_dev(host)->dma_mask && *mmc_dev(host)->dma_mask) + limit = *mmc_dev(host)->dma_mask; mq->card = card; mq->queue = blk_init_queue(mmc_request, lock); diff --git a/drivers/mmc/mmc_sysfs.c b/drivers/mmc/mmc_sysfs.c index 10cc9734eaa..ac532963604 100644 --- a/drivers/mmc/mmc_sysfs.c +++ b/drivers/mmc/mmc_sysfs.c @@ -199,7 +199,7 @@ void mmc_init_card(struct mmc_card *card, struct mmc_host *host) memset(card, 0, sizeof(struct mmc_card)); card->host = host; device_initialize(&card->dev); - card->dev.parent = card->host->dev; + card->dev.parent = mmc_dev(host); card->dev.bus = &mmc_bus_type; card->dev.release = mmc_release_card; } @@ -242,7 +242,7 @@ void mmc_remove_card(struct mmc_card *card) } -static void mmc_host_classdev_release(struct class_device *dev) +static void mmc_host_classdev_release(struct device *dev) { struct mmc_host *host = cls_dev_to_mmc_host(dev); kfree(host); @@ -250,7 +250,7 @@ static void mmc_host_classdev_release(struct class_device *dev) static struct class mmc_host_class = { .name = "mmc_host", - .release = mmc_host_classdev_release, + .dev_release = mmc_host_classdev_release, }; static DEFINE_IDR(mmc_host_idr); @@ -267,10 +267,10 @@ struct mmc_host *mmc_alloc_host_sysfs(int extra, struct device *dev) if (host) { memset(host, 0, sizeof(struct mmc_host) + extra); - host->dev = dev; - host->class_dev.dev = host->dev; + host->parent = dev; + host->class_dev.parent = dev; host->class_dev.class = &mmc_host_class; - class_device_initialize(&host->class_dev); + device_initialize(&host->class_dev); } return host; @@ -292,10 +292,10 @@ int mmc_add_host_sysfs(struct mmc_host *host) if (err) return err; - snprintf(host->class_dev.class_id, BUS_ID_SIZE, + snprintf(host->class_dev.bus_id, BUS_ID_SIZE, "mmc%d", host->index); - return class_device_add(&host->class_dev); + return device_add(&host->class_dev); } /* @@ -303,7 +303,7 @@ int mmc_add_host_sysfs(struct mmc_host *host) */ void mmc_remove_host_sysfs(struct mmc_host *host) { - class_device_del(&host->class_dev); + device_del(&host->class_dev); spin_lock(&mmc_host_lock); idr_remove(&mmc_host_idr, host->index); @@ -315,7 +315,7 @@ void mmc_remove_host_sysfs(struct mmc_host *host) */ void mmc_free_host_sysfs(struct mmc_host *host) { - class_device_put(&host->class_dev); + put_device(&host->class_dev); } static struct workqueue_struct *workqueue; diff --git a/drivers/mmc/wbsd.c b/drivers/mmc/wbsd.c index ced309b37a8..682e62b0b09 100644 --- a/drivers/mmc/wbsd.c +++ b/drivers/mmc/wbsd.c @@ -1488,7 +1488,7 @@ static void __devinit wbsd_request_dma(struct wbsd_host *host, int dma) /* * Translate the address to a physical address. */ - host->dma_addr = dma_map_single(host->mmc->dev, host->dma_buffer, + host->dma_addr = dma_map_single(mmc_dev(host->mmc), host->dma_buffer, WBSD_DMA_SIZE, DMA_BIDIRECTIONAL); /* @@ -1512,7 +1512,7 @@ kfree: */ BUG_ON(1); - dma_unmap_single(host->mmc->dev, host->dma_addr, + dma_unmap_single(mmc_dev(host->mmc), host->dma_addr, WBSD_DMA_SIZE, DMA_BIDIRECTIONAL); host->dma_addr = (dma_addr_t)NULL; @@ -1530,7 +1530,7 @@ err: static void __devexit wbsd_release_dma(struct wbsd_host *host) { if (host->dma_addr) { - dma_unmap_single(host->mmc->dev, host->dma_addr, + dma_unmap_single(mmc_dev(host->mmc), host->dma_addr, WBSD_DMA_SIZE, DMA_BIDIRECTIONAL); } kfree(host->dma_buffer); diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h index 587264a58d5..528e7d3fecb 100644 --- a/include/linux/mmc/host.h +++ b/include/linux/mmc/host.h @@ -74,8 +74,8 @@ struct mmc_card; struct device; struct mmc_host { - struct device *dev; - struct class_device class_dev; + struct device *parent; + struct device class_dev; int index; const struct mmc_host_ops *ops; unsigned int f_min; @@ -125,8 +125,8 @@ static inline void *mmc_priv(struct mmc_host *host) return (void *)host->private; } -#define mmc_dev(x) ((x)->dev) -#define mmc_hostname(x) ((x)->class_dev.class_id) +#define mmc_dev(x) ((x)->parent) +#define mmc_hostname(x) ((x)->class_dev.bus_id) extern int mmc_suspend_host(struct mmc_host *, pm_message_t); extern int mmc_resume_host(struct mmc_host *); -- cgit v1.2.3-70-g09d2 From 78cde0887930f5d11a56fc51b013f2672fba0e6f Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 14 Sep 2006 07:30:59 -0700 Subject: Driver core: convert fb code to use struct device Converts from using struct "class_device" to "struct device" making everything show up properly in /sys/devices/ with symlinks from the /sys/class directory. Signed-off-by: Greg Kroah-Hartman --- drivers/video/fbmem.c | 16 ++--- drivers/video/fbsysfs.c | 163 +++++++++++++++++++++++++++--------------------- include/linux/fb.h | 8 +-- 3 files changed, 103 insertions(+), 84 deletions(-) (limited to 'include') diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c index 93ffcdd95f5..e973a87fbb0 100644 --- a/drivers/video/fbmem.c +++ b/drivers/video/fbmem.c @@ -1296,14 +1296,14 @@ register_framebuffer(struct fb_info *fb_info) break; fb_info->node = i; - fb_info->class_device = class_device_create(fb_class, NULL, MKDEV(FB_MAJOR, i), - fb_info->device, "fb%d", i); - if (IS_ERR(fb_info->class_device)) { + fb_info->dev = device_create(fb_class, fb_info->device, + MKDEV(FB_MAJOR, i), "fb%d", i); + if (IS_ERR(fb_info->dev)) { /* Not fatal */ - printk(KERN_WARNING "Unable to create class_device for framebuffer %d; errno = %ld\n", i, PTR_ERR(fb_info->class_device)); - fb_info->class_device = NULL; + printk(KERN_WARNING "Unable to create device for framebuffer %d; errno = %ld\n", i, PTR_ERR(fb_info->dev)); + fb_info->dev = NULL; } else - fb_init_class_device(fb_info); + fb_init_device(fb_info); if (fb_info->pixmap.addr == NULL) { fb_info->pixmap.addr = kmalloc(FBPIXMAPSIZE, GFP_KERNEL); @@ -1356,8 +1356,8 @@ unregister_framebuffer(struct fb_info *fb_info) fb_destroy_modelist(&fb_info->modelist); registered_fb[i]=NULL; num_registered_fb--; - fb_cleanup_class_device(fb_info); - class_device_destroy(fb_class, MKDEV(FB_MAJOR, i)); + fb_cleanup_device(fb_info); + device_destroy(fb_class, MKDEV(FB_MAJOR, i)); event.info = fb_info; fb_notifier_call_chain(FB_EVENT_FB_UNREGISTERED, &event); return 0; diff --git a/drivers/video/fbsysfs.c b/drivers/video/fbsysfs.c index d3a50417ed9..323bdf6fc7d 100644 --- a/drivers/video/fbsysfs.c +++ b/drivers/video/fbsysfs.c @@ -73,7 +73,7 @@ EXPORT_SYMBOL(framebuffer_alloc); * * @info: frame buffer info structure * - * Drop the reference count of the class_device embedded in the + * Drop the reference count of the device embedded in the * framebuffer info structure. * */ @@ -120,10 +120,10 @@ static int mode_string(char *buf, unsigned int offset, m, mode->xres, mode->yres, v, mode->refresh); } -static ssize_t store_mode(struct class_device *class_device, const char * buf, - size_t count) +static ssize_t store_mode(struct device *device, struct device_attribute *attr, + const char *buf, size_t count) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); char mstr[100]; struct fb_var_screeninfo var; struct fb_modelist *modelist; @@ -151,9 +151,10 @@ static ssize_t store_mode(struct class_device *class_device, const char * buf, return -EINVAL; } -static ssize_t show_mode(struct class_device *class_device, char *buf) +static ssize_t show_mode(struct device *device, struct device_attribute *attr, + char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); if (!fb_info->mode) return 0; @@ -161,10 +162,11 @@ static ssize_t show_mode(struct class_device *class_device, char *buf) return mode_string(buf, 0, fb_info->mode); } -static ssize_t store_modes(struct class_device *class_device, const char * buf, - size_t count) +static ssize_t store_modes(struct device *device, + struct device_attribute *attr, + const char *buf, size_t count) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); LIST_HEAD(old_list); int i = count / sizeof(struct fb_videomode); @@ -186,9 +188,10 @@ static ssize_t store_modes(struct class_device *class_device, const char * buf, return 0; } -static ssize_t show_modes(struct class_device *class_device, char *buf) +static ssize_t show_modes(struct device *device, struct device_attribute *attr, + char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); unsigned int i; struct list_head *pos; struct fb_modelist *modelist; @@ -203,10 +206,10 @@ static ssize_t show_modes(struct class_device *class_device, char *buf) return i; } -static ssize_t store_bpp(struct class_device *class_device, const char * buf, - size_t count) +static ssize_t store_bpp(struct device *device, struct device_attribute *attr, + const char *buf, size_t count) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); struct fb_var_screeninfo var; char ** last = NULL; int err; @@ -218,16 +221,18 @@ static ssize_t store_bpp(struct class_device *class_device, const char * buf, return count; } -static ssize_t show_bpp(struct class_device *class_device, char *buf) +static ssize_t show_bpp(struct device *device, struct device_attribute *attr, + char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); return snprintf(buf, PAGE_SIZE, "%d\n", fb_info->var.bits_per_pixel); } -static ssize_t store_rotate(struct class_device *class_device, const char *buf, - size_t count) +static ssize_t store_rotate(struct device *device, + struct device_attribute *attr, + const char *buf, size_t count) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); struct fb_var_screeninfo var; char **last = NULL; int err; @@ -242,17 +247,19 @@ static ssize_t store_rotate(struct class_device *class_device, const char *buf, } -static ssize_t show_rotate(struct class_device *class_device, char *buf) +static ssize_t show_rotate(struct device *device, + struct device_attribute *attr, char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); return snprintf(buf, PAGE_SIZE, "%d\n", fb_info->var.rotate); } -static ssize_t store_virtual(struct class_device *class_device, - const char * buf, size_t count) +static ssize_t store_virtual(struct device *device, + struct device_attribute *attr, + const char *buf, size_t count) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); struct fb_var_screeninfo var; char *last = NULL; int err; @@ -269,23 +276,26 @@ static ssize_t store_virtual(struct class_device *class_device, return count; } -static ssize_t show_virtual(struct class_device *class_device, char *buf) +static ssize_t show_virtual(struct device *device, + struct device_attribute *attr, char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); return snprintf(buf, PAGE_SIZE, "%d,%d\n", fb_info->var.xres_virtual, fb_info->var.yres_virtual); } -static ssize_t show_stride(struct class_device *class_device, char *buf) +static ssize_t show_stride(struct device *device, + struct device_attribute *attr, char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); return snprintf(buf, PAGE_SIZE, "%d\n", fb_info->fix.line_length); } -static ssize_t store_blank(struct class_device *class_device, const char * buf, - size_t count) +static ssize_t store_blank(struct device *device, + struct device_attribute *attr, + const char *buf, size_t count) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); char *last = NULL; int err; @@ -299,42 +309,48 @@ static ssize_t store_blank(struct class_device *class_device, const char * buf, return count; } -static ssize_t show_blank(struct class_device *class_device, char *buf) +static ssize_t show_blank(struct device *device, + struct device_attribute *attr, char *buf) { -// struct fb_info *fb_info = class_get_devdata(class_device); +// struct fb_info *fb_info = dev_get_drvdata(device); return 0; } -static ssize_t store_console(struct class_device *class_device, - const char * buf, size_t count) +static ssize_t store_console(struct device *device, + struct device_attribute *attr, + const char *buf, size_t count) { -// struct fb_info *fb_info = class_get_devdata(class_device); +// struct fb_info *fb_info = dev_get_drvdata(device); return 0; } -static ssize_t show_console(struct class_device *class_device, char *buf) +static ssize_t show_console(struct device *device, + struct device_attribute *attr, char *buf) { -// struct fb_info *fb_info = class_get_devdata(class_device); +// struct fb_info *fb_info = dev_get_drvdata(device); return 0; } -static ssize_t store_cursor(struct class_device *class_device, - const char * buf, size_t count) +static ssize_t store_cursor(struct device *device, + struct device_attribute *attr, + const char *buf, size_t count) { -// struct fb_info *fb_info = class_get_devdata(class_device); +// struct fb_info *fb_info = dev_get_drvdata(device); return 0; } -static ssize_t show_cursor(struct class_device *class_device, char *buf) +static ssize_t show_cursor(struct device *device, + struct device_attribute *attr, char *buf) { -// struct fb_info *fb_info = class_get_devdata(class_device); +// struct fb_info *fb_info = dev_get_drvdata(device); return 0; } -static ssize_t store_pan(struct class_device *class_device, const char * buf, - size_t count) +static ssize_t store_pan(struct device *device, + struct device_attribute *attr, + const char *buf, size_t count) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); struct fb_var_screeninfo var; char *last = NULL; int err; @@ -355,24 +371,27 @@ static ssize_t store_pan(struct class_device *class_device, const char * buf, return count; } -static ssize_t show_pan(struct class_device *class_device, char *buf) +static ssize_t show_pan(struct device *device, + struct device_attribute *attr, char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); return snprintf(buf, PAGE_SIZE, "%d,%d\n", fb_info->var.xoffset, fb_info->var.xoffset); } -static ssize_t show_name(struct class_device *class_device, char *buf) +static ssize_t show_name(struct device *device, + struct device_attribute *attr, char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); return snprintf(buf, PAGE_SIZE, "%s\n", fb_info->fix.id); } -static ssize_t store_fbstate(struct class_device *class_device, - const char *buf, size_t count) +static ssize_t store_fbstate(struct device *device, + struct device_attribute *attr, + const char *buf, size_t count) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); u32 state; char *last = NULL; @@ -385,17 +404,19 @@ static ssize_t store_fbstate(struct class_device *class_device, return count; } -static ssize_t show_fbstate(struct class_device *class_device, char *buf) +static ssize_t show_fbstate(struct device *device, + struct device_attribute *attr, char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); return snprintf(buf, PAGE_SIZE, "%d\n", fb_info->state); } #ifdef CONFIG_FB_BACKLIGHT -static ssize_t store_bl_curve(struct class_device *class_device, - const char *buf, size_t count) +static ssize_t store_bl_curve(struct device *device, + struct device_attribute *attr, + const char *buf, size_t count) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); u8 tmp_curve[FB_BACKLIGHT_LEVELS]; unsigned int i; @@ -432,9 +453,10 @@ static ssize_t store_bl_curve(struct class_device *class_device, return count; } -static ssize_t show_bl_curve(struct class_device *class_device, char *buf) +static ssize_t show_bl_curve(struct device *device, + struct device_attribute *attr, char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); ssize_t len = 0; unsigned int i; @@ -465,7 +487,7 @@ static ssize_t show_bl_curve(struct class_device *class_device, char *buf) /* When cmap is added back in it should be a binary attribute * not a text one. Consideration should also be given to converting * fbdev to use configfs instead of sysfs */ -static struct class_device_attribute class_device_attrs[] = { +static struct device_attribute device_attrs[] = { __ATTR(bits_per_pixel, S_IRUGO|S_IWUSR, show_bpp, store_bpp), __ATTR(blank, S_IRUGO|S_IWUSR, show_blank, store_blank), __ATTR(console, S_IRUGO|S_IWUSR, show_console, store_console), @@ -483,17 +505,16 @@ static struct class_device_attribute class_device_attrs[] = { #endif }; -int fb_init_class_device(struct fb_info *fb_info) +int fb_init_device(struct fb_info *fb_info) { int i, error = 0; - class_set_devdata(fb_info->class_device, fb_info); + dev_set_drvdata(fb_info->dev, fb_info); fb_info->class_flag |= FB_SYSFS_FLAG_ATTR; - for (i = 0; i < ARRAY_SIZE(class_device_attrs); i++) { - error = class_device_create_file(fb_info->class_device, - &class_device_attrs[i]); + for (i = 0; i < ARRAY_SIZE(device_attrs); i++) { + error = device_create_file(fb_info->dev, &device_attrs[i]); if (error) break; @@ -501,22 +522,20 @@ int fb_init_class_device(struct fb_info *fb_info) if (error) { while (--i >= 0) - class_device_remove_file(fb_info->class_device, - &class_device_attrs[i]); + device_remove_file(fb_info->dev, &device_attrs[i]); fb_info->class_flag &= ~FB_SYSFS_FLAG_ATTR; } return 0; } -void fb_cleanup_class_device(struct fb_info *fb_info) +void fb_cleanup_device(struct fb_info *fb_info) { unsigned int i; if (fb_info->class_flag & FB_SYSFS_FLAG_ATTR) { - for (i = 0; i < ARRAY_SIZE(class_device_attrs); i++) - class_device_remove_file(fb_info->class_device, - &class_device_attrs[i]); + for (i = 0; i < ARRAY_SIZE(device_attrs); i++) + device_remove_file(fb_info->dev, &device_attrs[i]); fb_info->class_flag &= ~FB_SYSFS_FLAG_ATTR; } diff --git a/include/linux/fb.h b/include/linux/fb.h index 3e69241e6a8..fa23e0671bb 100644 --- a/include/linux/fb.h +++ b/include/linux/fb.h @@ -774,8 +774,8 @@ struct fb_info { #endif struct fb_ops *fbops; - struct device *device; - struct class_device *class_device; /* sysfs per device attrs */ + struct device *device; /* This is the parent */ + struct device *dev; /* This is this fb device */ int class_flag; /* private sysfs flags */ #ifdef CONFIG_FB_TILEBLITTING struct fb_tile_ops *tileops; /* Tile Blitting */ @@ -910,8 +910,8 @@ static inline void __fb_pad_aligned_buffer(u8 *dst, u32 d_pitch, /* drivers/video/fbsysfs.c */ extern struct fb_info *framebuffer_alloc(size_t size, struct device *dev); extern void framebuffer_release(struct fb_info *info); -extern int fb_init_class_device(struct fb_info *fb_info); -extern void fb_cleanup_class_device(struct fb_info *head); +extern int fb_init_device(struct fb_info *fb_info); +extern void fb_cleanup_device(struct fb_info *head); extern void fb_bl_default_curve(struct fb_info *fb_info, u8 off, u8 min, u8 max); /* drivers/video/fbmon.c */ -- cgit v1.2.3-70-g09d2 From d80f19fab89cba8a6d16193154c8ff3edab00942 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 7 Aug 2006 22:19:37 -0700 Subject: Driver core: convert sound core to use struct device Converts from using struct "class_device" to "struct device" making everything show up properly in /sys/devices/ with symlinks from the /sys/class directory. It also makes the struct sound_card to show up as a "real" device where all the different sound class devices are placed as childs and different card attribute files can hang off of. /sys/class/sound is still a flat directory, but the symlink targets of all devices belonging to the same card, point the the /sys/devices tree below the new card device object. Thanks to Kay for the updates to this patch. Signed-off-by: Kay Sievers Acked-by: Jaroslav Kysela Signed-off-by: Greg Kroah-Hartman --- include/sound/core.h | 8 +++++--- sound/core/init.c | 8 ++++++++ sound/core/pcm.c | 7 ++++--- sound/core/sound.c | 22 +++++++++------------- sound/oss/soundcard.c | 16 ++++++++-------- sound/sound_core.c | 6 +++--- 6 files changed, 37 insertions(+), 30 deletions(-) (limited to 'include') diff --git a/include/sound/core.h b/include/sound/core.h index fa1ca0127ba..a994bea09cd 100644 --- a/include/sound/core.h +++ b/include/sound/core.h @@ -132,6 +132,7 @@ struct snd_card { int shutdown; /* this card is going down */ int free_on_last_close; /* free in context of file_release */ wait_queue_head_t shutdown_sleep; + struct device *parent; struct device *dev; #ifdef CONFIG_PM @@ -187,13 +188,14 @@ struct snd_minor { int device; /* device number */ const struct file_operations *f_ops; /* file operations */ void *private_data; /* private data for f_ops->open */ - struct class_device *class_dev; /* class device for sysfs */ + struct device *dev; /* device for sysfs */ }; /* sound.c */ extern int snd_major; extern int snd_ecards_limit; +extern struct class *sound_class; void snd_request_card(int card); @@ -203,7 +205,7 @@ int snd_register_device(int type, struct snd_card *card, int dev, int snd_unregister_device(int type, struct snd_card *card, int dev); void *snd_lookup_minor_data(unsigned int minor, int type); int snd_add_device_sysfs_file(int type, struct snd_card *card, int dev, - const struct class_device_attribute *attr); + struct device_attribute *attr); #ifdef CONFIG_SND_OSSEMUL int snd_register_oss_device(int type, struct snd_card *card, int dev, @@ -255,7 +257,7 @@ int snd_card_file_add(struct snd_card *card, struct file *file); int snd_card_file_remove(struct snd_card *card, struct file *file); #ifndef snd_card_set_dev -#define snd_card_set_dev(card,devptr) ((card)->dev = (devptr)) +#define snd_card_set_dev(card,devptr) ((card)->parent = (devptr)) #endif /* device.c */ diff --git a/sound/core/init.c b/sound/core/init.c index 3058d626a90..6152a7554df 100644 --- a/sound/core/init.c +++ b/sound/core/init.c @@ -361,6 +361,8 @@ static int snd_card_do_free(struct snd_card *card) snd_printk(KERN_WARNING "unable to free card info\n"); /* Not fatal error */ } + if (card->dev) + device_unregister(card->dev); kfree(card); return 0; } @@ -495,6 +497,12 @@ int snd_card_register(struct snd_card *card) int err; snd_assert(card != NULL, return -EINVAL); + if (!card->dev) { + card->dev = device_create(sound_class, card->parent, 0, + "card%i", card->number); + if (IS_ERR(card->dev)) + card->dev = NULL; + } if ((err = snd_device_register_all(card)) < 0) return err; mutex_lock(&snd_card_mutex); diff --git a/sound/core/pcm.c b/sound/core/pcm.c index fbbbcd20c4c..5ac6e19ccb4 100644 --- a/sound/core/pcm.c +++ b/sound/core/pcm.c @@ -910,7 +910,8 @@ void snd_pcm_detach_substream(struct snd_pcm_substream *substream) substream->pstr->substream_opened--; } -static ssize_t show_pcm_class(struct class_device *class_device, char *buf) +static ssize_t show_pcm_class(struct device *dev, + struct device_attribute *attr, char *buf) { struct snd_pcm *pcm; const char *str; @@ -921,7 +922,7 @@ static ssize_t show_pcm_class(struct class_device *class_device, char *buf) [SNDRV_PCM_CLASS_DIGITIZER] = "digitizer", }; - if (! (pcm = class_get_devdata(class_device)) || + if (! (pcm = dev_get_drvdata(dev)) || pcm->dev_class > SNDRV_PCM_CLASS_LAST) str = "none"; else @@ -929,7 +930,7 @@ static ssize_t show_pcm_class(struct class_device *class_device, char *buf) return snprintf(buf, PAGE_SIZE, "%s\n", str); } -static struct class_device_attribute pcm_attrs = +static struct device_attribute pcm_attrs = __ATTR(pcm_class, S_IRUGO, show_pcm_class, NULL); static int snd_pcm_dev_register(struct snd_device *device) diff --git a/sound/core/sound.c b/sound/core/sound.c index efa476c5210..282742022de 100644 --- a/sound/core/sound.c +++ b/sound/core/sound.c @@ -61,9 +61,6 @@ EXPORT_SYMBOL(snd_ecards_limit); static struct snd_minor *snd_minors[SNDRV_OS_MINORS]; static DEFINE_MUTEX(sound_mutex); -extern struct class *sound_class; - - #ifdef CONFIG_KMOD /** @@ -268,11 +265,10 @@ int snd_register_device(int type, struct snd_card *card, int dev, snd_minors[minor] = preg; if (card) device = card->dev; - preg->class_dev = class_device_create(sound_class, NULL, - MKDEV(major, minor), - device, "%s", name); - if (preg->class_dev) - class_set_devdata(preg->class_dev, private_data); + preg->dev = device_create(sound_class, device, MKDEV(major, minor), + "%s", name); + if (preg->dev) + dev_set_drvdata(preg->dev, private_data); mutex_unlock(&sound_mutex); return 0; @@ -320,7 +316,7 @@ int snd_unregister_device(int type, struct snd_card *card, int dev) return -EINVAL; } - class_device_destroy(sound_class, MKDEV(major, minor)); + device_destroy(sound_class, MKDEV(major, minor)); kfree(snd_minors[minor]); snd_minors[minor] = NULL; @@ -331,15 +327,15 @@ int snd_unregister_device(int type, struct snd_card *card, int dev) EXPORT_SYMBOL(snd_unregister_device); int snd_add_device_sysfs_file(int type, struct snd_card *card, int dev, - const struct class_device_attribute *attr) + struct device_attribute *attr) { int minor, ret = -EINVAL; - struct class_device *cdev; + struct device *d; mutex_lock(&sound_mutex); minor = find_snd_minor(type, card, dev); - if (minor >= 0 && (cdev = snd_minors[minor]->class_dev) != NULL) - ret = class_device_create_file(cdev, attr); + if (minor >= 0 && (d = snd_minors[minor]->dev) != NULL) + ret = device_create_file(d, attr); mutex_unlock(&sound_mutex); return ret; diff --git a/sound/oss/soundcard.c b/sound/oss/soundcard.c index 2344d09c711..75c5e745705 100644 --- a/sound/oss/soundcard.c +++ b/sound/oss/soundcard.c @@ -557,17 +557,17 @@ static int __init oss_init(void) sound_dmap_flag = (dmabuf > 0 ? 1 : 0); for (i = 0; i < sizeof (dev_list) / sizeof *dev_list; i++) { - class_device_create(sound_class, NULL, - MKDEV(SOUND_MAJOR, dev_list[i].minor), - NULL, "%s", dev_list[i].name); + device_create(sound_class, NULL, + MKDEV(SOUND_MAJOR, dev_list[i].minor), + "%s", dev_list[i].name); if (!dev_list[i].num) continue; for (j = 1; j < *dev_list[i].num; j++) - class_device_create(sound_class, NULL, - MKDEV(SOUND_MAJOR, dev_list[i].minor + (j*0x10)), - NULL, "%s%d", dev_list[i].name, j); + device_create(sound_class, NULL, + MKDEV(SOUND_MAJOR, dev_list[i].minor + (j*0x10)), + "%s%d", dev_list[i].name, j); } if (sound_nblocks >= 1024) @@ -581,11 +581,11 @@ static void __exit oss_cleanup(void) int i, j; for (i = 0; i < sizeof (dev_list) / sizeof *dev_list; i++) { - class_device_destroy(sound_class, MKDEV(SOUND_MAJOR, dev_list[i].minor)); + device_destroy(sound_class, MKDEV(SOUND_MAJOR, dev_list[i].minor)); if (!dev_list[i].num) continue; for (j = 1; j < *dev_list[i].num; j++) - class_device_destroy(sound_class, MKDEV(SOUND_MAJOR, dev_list[i].minor + (j*0x10))); + device_destroy(sound_class, MKDEV(SOUND_MAJOR, dev_list[i].minor + (j*0x10))); } unregister_sound_special(1); diff --git a/sound/sound_core.c b/sound/sound_core.c index 5322c50c961..8f1ced4ab34 100644 --- a/sound/sound_core.c +++ b/sound/sound_core.c @@ -170,8 +170,8 @@ static int sound_insert_unit(struct sound_unit **list, const struct file_operati else sprintf(s->name, "sound/%s%d", name, r / SOUND_STEP); - class_device_create(sound_class, NULL, MKDEV(SOUND_MAJOR, s->unit_minor), - dev, s->name+6); + device_create(sound_class, dev, MKDEV(SOUND_MAJOR, s->unit_minor), + s->name+6); return r; fail: @@ -193,7 +193,7 @@ static void sound_remove_unit(struct sound_unit **list, int unit) p = __sound_remove_unit(list, unit); spin_unlock(&sound_loader_lock); if (p) { - class_device_destroy(sound_class, MKDEV(SOUND_MAJOR, p->unit_minor)); + device_destroy(sound_class, MKDEV(SOUND_MAJOR, p->unit_minor)); kfree(p); } } -- cgit v1.2.3-70-g09d2 From c6dbaef22a2f78700e242915a13218dd780c89ff Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Sat, 11 Nov 2006 17:18:39 +1100 Subject: Driver core: add dev_archdata to struct device Add arch specific dev_archdata to struct device Adds an arch specific struct dev_arch to struct device. This enables architecture to add specific fields to every device in the system, like DMA operation pointers, NUMA node ID, firmware specific data, etc... Signed-off-by: Benjamin Herrenschmidt Acked-by: Andi Kleen Acked-By: David Howells Signed-off-by: Greg Kroah-Hartman --- include/asm-alpha/device.h | 7 +++++++ include/asm-arm/device.h | 7 +++++++ include/asm-arm26/device.h | 7 +++++++ include/asm-avr32/device.h | 7 +++++++ include/asm-cris/device.h | 7 +++++++ include/asm-frv/device.h | 7 +++++++ include/asm-generic/device.h | 12 ++++++++++++ include/asm-h8300/device.h | 7 +++++++ include/asm-i386/device.h | 7 +++++++ include/asm-ia64/device.h | 7 +++++++ include/asm-m32r/device.h | 7 +++++++ include/asm-m68k/device.h | 7 +++++++ include/asm-m68knommu/device.h | 7 +++++++ include/asm-mips/device.h | 7 +++++++ include/asm-parisc/device.h | 7 +++++++ include/asm-powerpc/device.h | 7 +++++++ include/asm-ppc/device.h | 7 +++++++ include/asm-s390/device.h | 7 +++++++ include/asm-sh/device.h | 7 +++++++ include/asm-sh64/device.h | 7 +++++++ include/asm-sparc/device.h | 7 +++++++ include/asm-sparc64/device.h | 7 +++++++ include/asm-um/device.h | 7 +++++++ include/asm-v850/device.h | 7 +++++++ include/asm-x86_64/device.h | 7 +++++++ include/asm-xtensa/device.h | 7 +++++++ include/linux/device.h | 3 +++ 27 files changed, 190 insertions(+) create mode 100644 include/asm-alpha/device.h create mode 100644 include/asm-arm/device.h create mode 100644 include/asm-arm26/device.h create mode 100644 include/asm-avr32/device.h create mode 100644 include/asm-cris/device.h create mode 100644 include/asm-frv/device.h create mode 100644 include/asm-generic/device.h create mode 100644 include/asm-h8300/device.h create mode 100644 include/asm-i386/device.h create mode 100644 include/asm-ia64/device.h create mode 100644 include/asm-m32r/device.h create mode 100644 include/asm-m68k/device.h create mode 100644 include/asm-m68knommu/device.h create mode 100644 include/asm-mips/device.h create mode 100644 include/asm-parisc/device.h create mode 100644 include/asm-powerpc/device.h create mode 100644 include/asm-ppc/device.h create mode 100644 include/asm-s390/device.h create mode 100644 include/asm-sh/device.h create mode 100644 include/asm-sh64/device.h create mode 100644 include/asm-sparc/device.h create mode 100644 include/asm-sparc64/device.h create mode 100644 include/asm-um/device.h create mode 100644 include/asm-v850/device.h create mode 100644 include/asm-x86_64/device.h create mode 100644 include/asm-xtensa/device.h (limited to 'include') diff --git a/include/asm-alpha/device.h b/include/asm-alpha/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-alpha/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-arm/device.h b/include/asm-arm/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-arm/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-arm26/device.h b/include/asm-arm26/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-arm26/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-avr32/device.h b/include/asm-avr32/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-avr32/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-cris/device.h b/include/asm-cris/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-cris/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-frv/device.h b/include/asm-frv/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-frv/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-generic/device.h b/include/asm-generic/device.h new file mode 100644 index 00000000000..c17c9600f22 --- /dev/null +++ b/include/asm-generic/device.h @@ -0,0 +1,12 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#ifndef _ASM_GENERIC_DEVICE_H +#define _ASM_GENERIC_DEVICE_H + +struct dev_archdata { +}; + +#endif /* _ASM_GENERIC_DEVICE_H */ diff --git a/include/asm-h8300/device.h b/include/asm-h8300/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-h8300/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-i386/device.h b/include/asm-i386/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-i386/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-ia64/device.h b/include/asm-ia64/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-ia64/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-m32r/device.h b/include/asm-m32r/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-m32r/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-m68k/device.h b/include/asm-m68k/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-m68k/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-m68knommu/device.h b/include/asm-m68knommu/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-m68knommu/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-mips/device.h b/include/asm-mips/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-mips/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-parisc/device.h b/include/asm-parisc/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-parisc/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-powerpc/device.h b/include/asm-powerpc/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-powerpc/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-ppc/device.h b/include/asm-ppc/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-ppc/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-s390/device.h b/include/asm-s390/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-s390/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-sh/device.h b/include/asm-sh/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-sh/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-sh64/device.h b/include/asm-sh64/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-sh64/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-sparc/device.h b/include/asm-sparc/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-sparc/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-sparc64/device.h b/include/asm-sparc64/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-sparc64/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-um/device.h b/include/asm-um/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-um/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-v850/device.h b/include/asm-v850/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-v850/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-x86_64/device.h b/include/asm-x86_64/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-x86_64/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-xtensa/device.h b/include/asm-xtensa/device.h new file mode 100644 index 00000000000..d8f9872b0e2 --- /dev/null +++ b/include/asm-xtensa/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/linux/device.h b/include/linux/device.h index 00b29e0c5ce..5b54d756cd5 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -21,6 +21,7 @@ #include #include #include +#include #define DEVICE_NAME_SIZE 50 #define DEVICE_NAME_HALF __stringify(20) /* Less than half to accommodate slop */ @@ -383,6 +384,8 @@ struct device { struct dma_coherent_mem *dma_mem; /* internal for coherent mem override */ + /* arch specific additions */ + struct dev_archdata archdata; /* class_device migration path */ struct list_head node; -- cgit v1.2.3-70-g09d2 From 465ae641e4a3e5028aa9c85d3843259aa28a22ce Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Sat, 11 Nov 2006 17:18:42 +1100 Subject: ACPI: Change ACPI to use dev_archdata instead of firmware_data Change ACPI to use dev_archdata instead of firmware_data This patch changes ACPI to use the new dev_archdata on i386, x86_64 and ia64 (is there any other arch using ACPI ?) to store it's acpi_handle. It also removes the firmware_data field from struct device as this was the only user. Only build-tested on x86 Signed-off-by: Benjamin Herrenschmidt Cc: Len Brown Signed-off-by: Greg Kroah-Hartman --- drivers/acpi/glue.c | 20 +++++++++++--------- include/acpi/acpi_bus.h | 2 +- include/asm-i386/device.h | 10 +++++++++- include/asm-ia64/device.h | 10 +++++++++- include/asm-x86_64/device.h | 10 +++++++++- include/linux/device.h | 2 -- 6 files changed, 39 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c index 10f160dc75b..a2f46d587d5 100644 --- a/drivers/acpi/glue.c +++ b/drivers/acpi/glue.c @@ -267,9 +267,9 @@ static int acpi_bind_one(struct device *dev, acpi_handle handle) { acpi_status status; - if (dev->firmware_data) { + if (dev->archdata.acpi_handle) { printk(KERN_WARNING PREFIX - "Drivers changed 'firmware_data' for %s\n", dev->bus_id); + "Drivers changed 'acpi_handle' for %s\n", dev->bus_id); return -EINVAL; } get_device(dev); @@ -278,25 +278,26 @@ static int acpi_bind_one(struct device *dev, acpi_handle handle) put_device(dev); return -EINVAL; } - dev->firmware_data = handle; + dev->archdata.acpi_handle = handle; return 0; } static int acpi_unbind_one(struct device *dev) { - if (!dev->firmware_data) + if (!dev->archdata.acpi_handle) return 0; - if (dev == acpi_get_physical_device(dev->firmware_data)) { + if (dev == acpi_get_physical_device(dev->archdata.acpi_handle)) { /* acpi_get_physical_device increase refcnt by one */ put_device(dev); - acpi_detach_data(dev->firmware_data, acpi_glue_data_handler); - dev->firmware_data = NULL; + acpi_detach_data(dev->archdata.acpi_handle, + acpi_glue_data_handler); + dev->archdata.acpi_handle = NULL; /* acpi_bind_one increase refcnt by one */ put_device(dev); } else { printk(KERN_ERR PREFIX - "Oops, 'firmware_data' corrupt for %s\n", dev->bus_id); + "Oops, 'acpi_handle' corrupt for %s\n", dev->bus_id); } return 0; } @@ -328,7 +329,8 @@ static int acpi_platform_notify(struct device *dev) if (!ret) { struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL }; - acpi_get_name(dev->firmware_data, ACPI_FULL_PATHNAME, &buffer); + acpi_get_name(dev->archdata.acpi_handle, + ACPI_FULL_PATHNAME, &buffer); DBG("Device %s -> %s\n", dev->bus_id, (char *)buffer.pointer); kfree(buffer.pointer); } else diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h index f338e40bd54..fdd10953b2b 100644 --- a/include/acpi/acpi_bus.h +++ b/include/acpi/acpi_bus.h @@ -357,7 +357,7 @@ struct device *acpi_get_physical_device(acpi_handle); /* helper */ acpi_handle acpi_get_child(acpi_handle, acpi_integer); acpi_handle acpi_get_pci_rootbridge_handle(unsigned int, unsigned int); -#define DEVICE_ACPI_HANDLE(dev) ((acpi_handle)((dev)->firmware_data)) +#define DEVICE_ACPI_HANDLE(dev) ((acpi_handle)((dev)->archdata.acpi_handle)) #endif /* CONFIG_ACPI */ diff --git a/include/asm-i386/device.h b/include/asm-i386/device.h index d8f9872b0e2..849604c70e6 100644 --- a/include/asm-i386/device.h +++ b/include/asm-i386/device.h @@ -3,5 +3,13 @@ * * This file is released under the GPLv2 */ -#include +#ifndef _ASM_I386_DEVICE_H +#define _ASM_I386_DEVICE_H +struct dev_archdata { +#ifdef CONFIG_ACPI + void *acpi_handle; +#endif +}; + +#endif /* _ASM_I386_DEVICE_H */ diff --git a/include/asm-ia64/device.h b/include/asm-ia64/device.h index d8f9872b0e2..3db6daf7f25 100644 --- a/include/asm-ia64/device.h +++ b/include/asm-ia64/device.h @@ -3,5 +3,13 @@ * * This file is released under the GPLv2 */ -#include +#ifndef _ASM_IA64_DEVICE_H +#define _ASM_IA64_DEVICE_H +struct dev_archdata { +#ifdef CONFIG_ACPI + void *acpi_handle; +#endif +}; + +#endif /* _ASM_IA64_DEVICE_H */ diff --git a/include/asm-x86_64/device.h b/include/asm-x86_64/device.h index d8f9872b0e2..3afa03f33a3 100644 --- a/include/asm-x86_64/device.h +++ b/include/asm-x86_64/device.h @@ -3,5 +3,13 @@ * * This file is released under the GPLv2 */ -#include +#ifndef _ASM_X86_64_DEVICE_H +#define _ASM_X86_64_DEVICE_H +struct dev_archdata { +#ifdef CONFIG_ACPI + void *acpi_handle; +#endif +}; + +#endif /* _ASM_X86_64_DEVICE_H */ diff --git a/include/linux/device.h b/include/linux/device.h index 5b54d756cd5..2d9dc358c02 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -369,8 +369,6 @@ struct device { void *driver_data; /* data private to the driver */ void *platform_data; /* Platform specific data, device core doesn't touch it */ - void *firmware_data; /* Firmware specific data (e.g. ACPI, - BIOS data),reserved for device core*/ struct dev_pm_info power; u64 *dma_mask; /* dma mask (if dma'able device) */ -- cgit v1.2.3-70-g09d2 From 5ab699810d46011ad2195c5916f3cbc684bfe3ee Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Thu, 16 Nov 2006 15:42:07 +0100 Subject: driver core: Introduce device_find_child(). Introduce device_find_child() to match device_for_each_child(). Signed-off-by: Cornelia Huck Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 33 +++++++++++++++++++++++++++++++++ include/linux/device.h | 2 ++ 2 files changed, 35 insertions(+) (limited to 'include') diff --git a/drivers/base/core.c b/drivers/base/core.c index 5d11bbdfbd2..a29e6854546 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -750,12 +750,45 @@ int device_for_each_child(struct device * parent, void * data, return error; } +/** + * device_find_child - device iterator for locating a particular device. + * @parent: parent struct device + * @data: Data to pass to match function + * @match: Callback function to check device + * + * This is similar to the device_for_each_child() function above, but it + * returns a reference to a device that is 'found' for later use, as + * determined by the @match callback. + * + * The callback should return 0 if the device doesn't match and non-zero + * if it does. If the callback returns non-zero and a reference to the + * current device can be obtained, this function will return to the caller + * and not iterate over any more devices. + */ +struct device * device_find_child(struct device *parent, void *data, + int (*match)(struct device *, void *)) +{ + struct klist_iter i; + struct device *child; + + if (!parent) + return NULL; + + klist_iter_init(&parent->klist_children, &i); + while ((child = next_device(&i))) + if (match(child, data) && get_device(child)) + break; + klist_iter_exit(&i); + return child; +} + int __init devices_init(void) { return subsystem_register(&devices_subsys); } EXPORT_SYMBOL_GPL(device_for_each_child); +EXPORT_SYMBOL_GPL(device_find_child); EXPORT_SYMBOL_GPL(device_initialize); EXPORT_SYMBOL_GPL(device_add); diff --git a/include/linux/device.h b/include/linux/device.h index 2d9dc358c02..0a0370c7418 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -421,6 +421,8 @@ extern int __must_check device_add(struct device * dev); extern void device_del(struct device * dev); extern int device_for_each_child(struct device *, void *, int (*fn)(struct device *, void *)); +extern struct device *device_find_child(struct device *, void *data, + int (*match)(struct device *, void *)); extern int device_rename(struct device *dev, char *new_name); /* -- cgit v1.2.3-70-g09d2 From 8a82472f86bf693b8e91ed56c9ca4f62fbbdcfa3 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Mon, 20 Nov 2006 17:07:51 +0100 Subject: driver core: Introduce device_move(): move a device to a new parent. Provide a function device_move() to move a device to a new parent device. Add auxilliary functions kobject_move() and sysfs_move_dir(). kobject_move() generates a new uevent of type KOBJ_MOVE, containing the previous path (DEVPATH_OLD) in addition to the usual values. For this, a new interface kobject_uevent_env() is created that allows to add further environmental data to the uevent at the kobject layer. Signed-off-by: Cornelia Huck Acked-by: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 92 +++++++++++++++++++++++++++++++++++++++++++++++++ fs/sysfs/dir.c | 45 ++++++++++++++++++++++++ include/linux/device.h | 1 + include/linux/kobject.h | 8 +++++ include/linux/sysfs.h | 8 +++++ lib/kobject.c | 50 +++++++++++++++++++++++++++ lib/kobject_uevent.c | 28 ++++++++++++--- 7 files changed, 228 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/drivers/base/core.c b/drivers/base/core.c index 75b45a10935..e4eaf46c4d9 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -955,3 +955,95 @@ int device_rename(struct device *dev, char *new_name) return error; } + + +static int device_move_class_links(struct device *dev, + struct device *old_parent, + struct device *new_parent) +{ +#ifdef CONFIG_SYSFS_DEPRECATED + int error; + char *class_name; + + class_name = make_class_name(dev->class->name, &dev->kobj); + if (!class_name) { + error = PTR_ERR(class_name); + class_name = NULL; + goto out; + } + if (old_parent) { + sysfs_remove_link(&dev->kobj, "device"); + sysfs_remove_link(&old_parent->kobj, class_name); + } + error = sysfs_create_link(&dev->kobj, &new_parent->kobj, "device"); + if (error) + goto out; + error = sysfs_create_link(&new_parent->kobj, &dev->kobj, class_name); + if (error) + sysfs_remove_link(&dev->kobj, "device"); +out: + kfree(class_name); + return error; +#else + return 0; +#endif +} + +/** + * device_move - moves a device to a new parent + * @dev: the pointer to the struct device to be moved + * @new_parent: the new parent of the device + */ +int device_move(struct device *dev, struct device *new_parent) +{ + int error; + struct device *old_parent; + + dev = get_device(dev); + if (!dev) + return -EINVAL; + + if (!device_is_registered(dev)) { + error = -EINVAL; + goto out; + } + new_parent = get_device(new_parent); + if (!new_parent) { + error = -EINVAL; + goto out; + } + pr_debug("DEVICE: moving '%s' to '%s'\n", dev->bus_id, + new_parent->bus_id); + error = kobject_move(&dev->kobj, &new_parent->kobj); + if (error) { + put_device(new_parent); + goto out; + } + old_parent = dev->parent; + dev->parent = new_parent; + if (old_parent) + klist_del(&dev->knode_parent); + klist_add_tail(&dev->knode_parent, &new_parent->klist_children); + if (!dev->class) + goto out_put; + error = device_move_class_links(dev, old_parent, new_parent); + if (error) { + /* We ignore errors on cleanup since we're hosed anyway... */ + device_move_class_links(dev, new_parent, old_parent); + if (!kobject_move(&dev->kobj, &old_parent->kobj)) { + klist_del(&dev->knode_parent); + if (old_parent) + klist_add_tail(&dev->knode_parent, + &old_parent->klist_children); + } + put_device(new_parent); + goto out; + } +out_put: + put_device(old_parent); +out: + put_device(dev); + return error; +} + +EXPORT_SYMBOL_GPL(device_move); diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c index 3aa3434621c..a5782e8c7f0 100644 --- a/fs/sysfs/dir.c +++ b/fs/sysfs/dir.c @@ -372,6 +372,51 @@ int sysfs_rename_dir(struct kobject * kobj, const char *new_name) return error; } +int sysfs_move_dir(struct kobject *kobj, struct kobject *new_parent) +{ + struct dentry *old_parent_dentry, *new_parent_dentry, *new_dentry; + struct sysfs_dirent *new_parent_sd, *sd; + int error; + + if (!new_parent) + return -EINVAL; + + old_parent_dentry = kobj->parent ? + kobj->parent->dentry : sysfs_mount->mnt_sb->s_root; + new_parent_dentry = new_parent->dentry; + +again: + mutex_lock(&old_parent_dentry->d_inode->i_mutex); + if (!mutex_trylock(&new_parent_dentry->d_inode->i_mutex)) { + mutex_unlock(&old_parent_dentry->d_inode->i_mutex); + goto again; + } + + new_parent_sd = new_parent_dentry->d_fsdata; + sd = kobj->dentry->d_fsdata; + + new_dentry = lookup_one_len(kobj->name, new_parent_dentry, + strlen(kobj->name)); + if (IS_ERR(new_dentry)) { + error = PTR_ERR(new_dentry); + goto out; + } else + error = 0; + d_add(new_dentry, NULL); + d_move(kobj->dentry, new_dentry); + dput(new_dentry); + + /* Remove from old parent's list and insert into new parent's list. */ + list_del_init(&sd->s_sibling); + list_add(&sd->s_sibling, &new_parent_sd->s_children); + +out: + mutex_unlock(&new_parent_dentry->d_inode->i_mutex); + mutex_unlock(&old_parent_dentry->d_inode->i_mutex); + + return error; +} + static int sysfs_dir_open(struct inode *inode, struct file *file) { struct dentry * dentry = file->f_dentry; diff --git a/include/linux/device.h b/include/linux/device.h index 0a0370c7418..583a341e016 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -424,6 +424,7 @@ extern int device_for_each_child(struct device *, void *, extern struct device *device_find_child(struct device *, void *data, int (*match)(struct device *, void *)); extern int device_rename(struct device *dev, char *new_name); +extern int device_move(struct device *dev, struct device *new_parent); /* * Manual binding of a device to driver. See drivers/base/bus.c diff --git a/include/linux/kobject.h b/include/linux/kobject.h index bcd9cd173c2..d1c8d28fa92 100644 --- a/include/linux/kobject.h +++ b/include/linux/kobject.h @@ -47,6 +47,7 @@ enum kobject_action { KOBJ_UMOUNT = (__force kobject_action_t) 0x05, /* umount event for block devices (broken) */ KOBJ_OFFLINE = (__force kobject_action_t) 0x06, /* device offline */ KOBJ_ONLINE = (__force kobject_action_t) 0x07, /* device online */ + KOBJ_MOVE = (__force kobject_action_t) 0x08, /* device move */ }; struct kobject { @@ -76,6 +77,7 @@ extern int __must_check kobject_add(struct kobject *); extern void kobject_del(struct kobject *); extern int __must_check kobject_rename(struct kobject *, const char *new_name); +extern int __must_check kobject_move(struct kobject *, struct kobject *); extern int __must_check kobject_register(struct kobject *); extern void kobject_unregister(struct kobject *); @@ -264,6 +266,8 @@ extern int __must_check subsys_create_file(struct subsystem * , #if defined(CONFIG_HOTPLUG) void kobject_uevent(struct kobject *kobj, enum kobject_action action); +void kobject_uevent_env(struct kobject *kobj, enum kobject_action action, + char *envp[]); int add_uevent_var(char **envp, int num_envp, int *cur_index, char *buffer, int buffer_size, int *cur_len, @@ -271,6 +275,10 @@ int add_uevent_var(char **envp, int num_envp, int *cur_index, __attribute__((format (printf, 7, 8))); #else static inline void kobject_uevent(struct kobject *kobj, enum kobject_action action) { } +static inline void kobject_uevent_env(struct kobject *kobj, + enum kobject_action action, + char *envp[]) +{ } static inline int add_uevent_var(char **envp, int num_envp, int *cur_index, char *buffer, int buffer_size, int *cur_len, diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h index 6d5c43d31de..2129d1b6c87 100644 --- a/include/linux/sysfs.h +++ b/include/linux/sysfs.h @@ -96,6 +96,9 @@ sysfs_remove_dir(struct kobject *); extern int __must_check sysfs_rename_dir(struct kobject *, const char *new_name); +extern int __must_check +sysfs_move_dir(struct kobject *, struct kobject *); + extern int __must_check sysfs_create_file(struct kobject *, const struct attribute *); @@ -142,6 +145,11 @@ static inline int sysfs_rename_dir(struct kobject * k, const char *new_name) return 0; } +static inline int sysfs_move_dir(struct kobject * k, struct kobject * new_parent) +{ + return 0; +} + static inline int sysfs_create_file(struct kobject * k, const struct attribute * a) { return 0; diff --git a/lib/kobject.c b/lib/kobject.c index 7dd5c0e9d99..744a4b102c7 100644 --- a/lib/kobject.c +++ b/lib/kobject.c @@ -310,6 +310,56 @@ int kobject_rename(struct kobject * kobj, const char *new_name) return error; } +/** + * kobject_move - move object to another parent + * @kobj: object in question. + * @new_parent: object's new parent + */ + +int kobject_move(struct kobject *kobj, struct kobject *new_parent) +{ + int error; + struct kobject *old_parent; + const char *devpath = NULL; + char *devpath_string = NULL; + char *envp[2]; + + kobj = kobject_get(kobj); + if (!kobj) + return -EINVAL; + new_parent = kobject_get(new_parent); + if (!new_parent) { + error = -EINVAL; + goto out; + } + /* old object path */ + devpath = kobject_get_path(kobj, GFP_KERNEL); + if (!devpath) { + error = -ENOMEM; + goto out; + } + devpath_string = kmalloc(strlen(devpath) + 15, GFP_KERNEL); + if (!devpath_string) { + error = -ENOMEM; + goto out; + } + sprintf(devpath_string, "DEVPATH_OLD=%s", devpath); + envp[0] = devpath_string; + envp[1] = NULL; + error = sysfs_move_dir(kobj, new_parent); + if (error) + goto out; + old_parent = kobj->parent; + kobj->parent = new_parent; + kobject_put(old_parent); + kobject_uevent_env(kobj, KOBJ_MOVE, envp); +out: + kobject_put(kobj); + kfree(devpath_string); + kfree(devpath); + return error; +} + /** * kobject_del - unlink kobject from hierarchy. * @kobj: object. diff --git a/lib/kobject_uevent.c b/lib/kobject_uevent.c index 7f20e7b857c..a1922765ff3 100644 --- a/lib/kobject_uevent.c +++ b/lib/kobject_uevent.c @@ -50,18 +50,22 @@ static char *action_to_string(enum kobject_action action) return "offline"; case KOBJ_ONLINE: return "online"; + case KOBJ_MOVE: + return "move"; default: return NULL; } } /** - * kobject_uevent - notify userspace by ending an uevent + * kobject_uevent_env - send an uevent with environmental data * - * @action: action that is happening (usually KOBJ_ADD and KOBJ_REMOVE) + * @action: action that is happening (usually KOBJ_MOVE) * @kobj: struct kobject that the action is happening to + * @envp_ext: pointer to environmental data */ -void kobject_uevent(struct kobject *kobj, enum kobject_action action) +void kobject_uevent_env(struct kobject *kobj, enum kobject_action action, + char *envp_ext[]) { char **envp; char *buffer; @@ -76,6 +80,7 @@ void kobject_uevent(struct kobject *kobj, enum kobject_action action) char *seq_buff; int i = 0; int retval; + int j; pr_debug("%s\n", __FUNCTION__); @@ -134,7 +139,8 @@ void kobject_uevent(struct kobject *kobj, enum kobject_action action) scratch += sprintf (scratch, "DEVPATH=%s", devpath) + 1; envp [i++] = scratch; scratch += sprintf(scratch, "SUBSYSTEM=%s", subsystem) + 1; - + for (j = 0; envp_ext && envp_ext[j]; j++) + envp[i++] = envp_ext[j]; /* just reserve the space, overwrite it after kset call has returned */ envp[i++] = seq_buff = scratch; scratch += strlen("SEQNUM=18446744073709551616") + 1; @@ -200,6 +206,20 @@ exit: kfree(envp); return; } + +EXPORT_SYMBOL_GPL(kobject_uevent_env); + +/** + * kobject_uevent - notify userspace by ending an uevent + * + * @action: action that is happening (usually KOBJ_ADD and KOBJ_REMOVE) + * @kobj: struct kobject that the action is happening to + */ +void kobject_uevent(struct kobject *kobj, enum kobject_action action) +{ + kobject_uevent_env(kobj, action, NULL); +} + EXPORT_SYMBOL_GPL(kobject_uevent); /** -- cgit v1.2.3-70-g09d2 From c67334fbdfbba533af767610cf3fde8a49710e62 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Thu, 16 Nov 2006 23:28:47 -0800 Subject: Driver core: platform_driver_probe(), can save codespace This defines a new platform_driver_probe() method allowing the driver's probe() method, and its support code+data, to safely live in __init sections for typical system configurations. Many system-on-chip processors could benefit from this API, to the tune of recovering hundreds to thousands of bytes per driver. That's memory which is currently wasted holding code which can never be called after system startup, yet can not be removed. It can't be removed because of the linkage requirement that pointers to init section code (like, ideally, probe support) must not live in other sections (like driver method tables) after those pointers would be invalid. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman --- drivers/base/platform.c | 48 +++++++++++++++++++++++++++++++++++++++++ include/linux/platform_device.h | 6 ++++++ 2 files changed, 54 insertions(+) (limited to 'include') diff --git a/drivers/base/platform.c b/drivers/base/platform.c index 940ce41f188..d1df4a08792 100644 --- a/drivers/base/platform.c +++ b/drivers/base/platform.c @@ -388,6 +388,11 @@ static int platform_drv_probe(struct device *_dev) return drv->probe(dev); } +static int platform_drv_probe_fail(struct device *_dev) +{ + return -ENXIO; +} + static int platform_drv_remove(struct device *_dev) { struct platform_driver *drv = to_platform_driver(_dev->driver); @@ -451,6 +456,49 @@ void platform_driver_unregister(struct platform_driver *drv) } EXPORT_SYMBOL_GPL(platform_driver_unregister); +/** + * platform_driver_probe - register driver for non-hotpluggable device + * @drv: platform driver structure + * @probe: the driver probe routine, probably from an __init section + * + * Use this instead of platform_driver_register() when you know the device + * is not hotpluggable and has already been registered, and you want to + * remove its run-once probe() infrastructure from memory after the driver + * has bound to the device. + * + * One typical use for this would be with drivers for controllers integrated + * into system-on-chip processors, where the controller devices have been + * configured as part of board setup. + * + * Returns zero if the driver registered and bound to a device, else returns + * a negative error code and with the driver not registered. + */ +int platform_driver_probe(struct platform_driver *drv, + int (*probe)(struct platform_device *)) +{ + int retval, code; + + /* temporary section violation during probe() */ + drv->probe = probe; + retval = code = platform_driver_register(drv); + + /* Fixup that section violation, being paranoid about code scanning + * the list of drivers in order to probe new devices. Check to see + * if the probe was successful, and make sure any forced probes of + * new devices fail. + */ + spin_lock(&platform_bus_type.klist_drivers.k_lock); + drv->probe = NULL; + if (code == 0 && list_empty(&drv->driver.klist_devices.k_list)) + retval = -ENODEV; + drv->driver.probe = platform_drv_probe_fail; + spin_unlock(&platform_bus_type.klist_drivers.k_lock); + + if (code != retval) + platform_driver_unregister(drv); + return retval; +} +EXPORT_SYMBOL_GPL(platform_driver_probe); /* modalias support enables more hands-off userspace setup: * (a) environment variable lets new-style hotplug events work once system is diff --git a/include/linux/platform_device.h b/include/linux/platform_device.h index 29cd6dee13d..20f47b81d3f 100644 --- a/include/linux/platform_device.h +++ b/include/linux/platform_device.h @@ -58,6 +58,12 @@ struct platform_driver { extern int platform_driver_register(struct platform_driver *); extern void platform_driver_unregister(struct platform_driver *); +/* non-hotpluggable platform devices may use this so that probe() and + * its support may live in __init sections, conserving runtime memory. + */ +extern int platform_driver_probe(struct platform_driver *driver, + int (*probe)(struct platform_device *)); + #define platform_get_drvdata(_dev) dev_get_drvdata(&(_dev)->dev) #define platform_set_drvdata(_dev,data) dev_set_drvdata(&(_dev)->dev, (data)) -- cgit v1.2.3-70-g09d2 From e17e0f51aeea4e59c7e450a1c0f26605b91c1260 Mon Sep 17 00:00:00 2001 From: Kay Sievers Date: Fri, 24 Nov 2006 12:15:25 +0100 Subject: Driver core: show drivers in /sys/module/ Show the drivers, which belong to the module: $ ls -l /sys/module/usbcore/drivers/ hub -> ../../../bus/usb/drivers/hub usb -> ../../../bus/usb/drivers/usb usbfs -> ../../../bus/usb/drivers/usbfs Signed-off-by: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- include/linux/module.h | 1 + kernel/module.c | 31 +++++++++++++++++++++++++------ 2 files changed, 26 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/module.h b/include/linux/module.h index d1d00ce8f4e..9258ffd8a7f 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -264,6 +264,7 @@ struct module struct module_attribute *modinfo_attrs; const char *version; const char *srcversion; + struct kobject *drivers_dir; /* Exported symbols */ const struct kernel_symbol *syms; diff --git a/kernel/module.c b/kernel/module.c index f0166563c60..45e01cb6010 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -1086,22 +1086,35 @@ static int mod_sysfs_setup(struct module *mod, goto out; kobj_set_kset_s(&mod->mkobj, module_subsys); mod->mkobj.mod = mod; - err = kobject_register(&mod->mkobj.kobj); + + /* delay uevent until full sysfs population */ + kobject_init(&mod->mkobj.kobj); + err = kobject_add(&mod->mkobj.kobj); if (err) goto out; + mod->drivers_dir = kobject_add_dir(&mod->mkobj.kobj, "drivers"); + if (!mod->drivers_dir) + goto out_unreg; + err = module_param_sysfs_setup(mod, kparam, num_params); if (err) - goto out_unreg; + goto out_unreg_drivers; err = module_add_modinfo_attrs(mod); if (err) - goto out_unreg; + goto out_unreg_param; + kobject_uevent(&mod->mkobj.kobj, KOBJ_ADD); return 0; +out_unreg_drivers: + kobject_unregister(mod->drivers_dir); +out_unreg_param: + module_param_sysfs_remove(mod); out_unreg: - kobject_unregister(&mod->mkobj.kobj); + kobject_del(&mod->mkobj.kobj); + kobject_put(&mod->mkobj.kobj); out: return err; } @@ -1110,6 +1123,7 @@ static void mod_kobject_remove(struct module *mod) { module_remove_modinfo_attrs(mod); module_param_sysfs_remove(mod); + kobject_unregister(mod->drivers_dir); kobject_unregister(&mod->mkobj.kobj); } @@ -2275,11 +2289,14 @@ void print_modules(void) void module_add_driver(struct module *mod, struct device_driver *drv) { + int no_warn; + if (!mod || !drv) return; - /* Don't check return code; this call is idempotent */ - sysfs_create_link(&drv->kobj, &mod->mkobj.kobj, "module"); + /* Don't check return codes; these calls are idempotent */ + no_warn = sysfs_create_link(&drv->kobj, &mod->mkobj.kobj, "module"); + no_warn = sysfs_create_link(mod->drivers_dir, &drv->kobj, drv->name); } EXPORT_SYMBOL(module_add_driver); @@ -2288,6 +2305,8 @@ void module_remove_driver(struct device_driver *drv) if (!drv) return; sysfs_remove_link(&drv->kobj, "module"); + if (drv->owner && drv->owner->drivers_dir) + sysfs_remove_link(drv->owner->drivers_dir, drv->name); } EXPORT_SYMBOL(module_remove_driver); -- cgit v1.2.3-70-g09d2 From 13df29f69749a61b5209d52b71fcbf7300e5d6fb Mon Sep 17 00:00:00 2001 From: "Maciej W. Rozycki" Date: Tue, 3 Oct 2006 16:18:28 +0100 Subject: [PATCH] 2.6.18: sb1250-mac: Missing inclusions from The uses some types and macros defined in , , and , but fails to include these headers. Signed-off-by: Maciej W. Rozycki patch-mips-2.6.18-20060920-include-phy-16 Signed-off-by: Jeff Garzik --- include/linux/phy.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include') diff --git a/include/linux/phy.h b/include/linux/phy.h index 9447a57ee8a..ce8bc80b3c8 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -20,6 +20,10 @@ #include #include +#include +#include +#include +#include #define PHY_BASIC_FEATURES (SUPPORTED_10baseT_Half | \ SUPPORTED_10baseT_Full | \ -- cgit v1.2.3-70-g09d2 From c9308b06c049a107edfbd4e5271771564eb6024d Mon Sep 17 00:00:00 2001 From: Daniel Drake Date: Wed, 27 Sep 2006 03:50:31 +0100 Subject: [PATCH] ieee80211: Move IV/ICV stripping into ieee80211_rx This patch adds a host_strip_iv_icv flag to ieee80211 which indicates that ieee80211_rx should strip the IV/ICV/other security features from the payload. This saves on some memmove() calls in the driver and seems like something that belongs in the stack as it can be used by bcm43xx, ipw2200, and zd1211rw I will submit the ipw2200 patch separately as it needs testing. This patch also adds some sensible variable reuse (idx vs keyidx) in ieee80211_rx Signed-off-by: Daniel Drake Acked-by: Johannes Berg Signed-off-by: John W. Linville --- drivers/net/wireless/bcm43xx/bcm43xx_wx.c | 1 + drivers/net/wireless/bcm43xx/bcm43xx_xmit.c | 19 ---------- include/net/ieee80211.h | 4 +++ net/ieee80211/ieee80211_rx.c | 56 +++++++++++++++++++++++++---- 4 files changed, 55 insertions(+), 25 deletions(-) (limited to 'include') diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_wx.c b/drivers/net/wireless/bcm43xx/bcm43xx_wx.c index 12043f8be1b..a659442b9c1 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx_wx.c +++ b/drivers/net/wireless/bcm43xx/bcm43xx_wx.c @@ -690,6 +690,7 @@ static int bcm43xx_wx_set_swencryption(struct net_device *net_dev, bcm->ieee->host_encrypt = !!on; bcm->ieee->host_decrypt = !!on; bcm->ieee->host_build_iv = !on; + bcm->ieee->host_strip_iv_icv = !on; spin_unlock_irqrestore(&bcm->irq_lock, flags); mutex_unlock(&bcm->mutex); diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_xmit.c b/drivers/net/wireless/bcm43xx/bcm43xx_xmit.c index 0159e4e9320..a957bc86138 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx_xmit.c +++ b/drivers/net/wireless/bcm43xx/bcm43xx_xmit.c @@ -543,25 +543,6 @@ int bcm43xx_rx(struct bcm43xx_private *bcm, break; } - frame_ctl = le16_to_cpu(wlhdr->frame_ctl); - if ((frame_ctl & IEEE80211_FCTL_PROTECTED) && !bcm->ieee->host_decrypt) { - frame_ctl &= ~IEEE80211_FCTL_PROTECTED; - wlhdr->frame_ctl = cpu_to_le16(frame_ctl); - /* trim IV and ICV */ - /* FIXME: this must be done only for WEP encrypted packets */ - if (skb->len < 32) { - dprintkl(KERN_ERR PFX "RX packet dropped (PROTECTED flag " - "set and length < 32)\n"); - return -EINVAL; - } else { - memmove(skb->data + 4, skb->data, 24); - skb_pull(skb, 4); - skb_trim(skb, skb->len - 4); - stats.len -= 8; - } - wlhdr = (struct ieee80211_hdr_4addr *)(skb->data); - } - switch (WLAN_FC_GET_TYPE(frame_ctl)) { case IEEE80211_FTYPE_MGMT: ieee80211_rx_mgt(bcm->ieee, wlhdr, &stats); diff --git a/include/net/ieee80211.h b/include/net/ieee80211.h index b174ebb277a..cb255432e4e 100644 --- a/include/net/ieee80211.h +++ b/include/net/ieee80211.h @@ -1037,6 +1037,10 @@ struct ieee80211_device { /* host performs multicast decryption */ int host_mc_decrypt; + /* host should strip IV and ICV from protected frames */ + /* meaningful only when hardware decryption is being used */ + int host_strip_iv_icv; + int host_open_frag; int host_build_iv; int ieee802_1x; /* is IEEE 802.1X used */ diff --git a/net/ieee80211/ieee80211_rx.c b/net/ieee80211/ieee80211_rx.c index 2759312a420..d9265195656 100644 --- a/net/ieee80211/ieee80211_rx.c +++ b/net/ieee80211/ieee80211_rx.c @@ -415,17 +415,16 @@ int ieee80211_rx(struct ieee80211_device *ieee, struct sk_buff *skb, ieee->host_mc_decrypt : ieee->host_decrypt; if (can_be_decrypted) { - int idx = 0; if (skb->len >= hdrlen + 3) { /* Top two-bits of byte 3 are the key index */ - idx = skb->data[hdrlen + 3] >> 6; + keyidx = skb->data[hdrlen + 3] >> 6; } - /* ieee->crypt[] is WEP_KEY (4) in length. Given that idx - * is only allowed 2-bits of storage, no value of idx can - * be provided via above code that would result in idx + /* ieee->crypt[] is WEP_KEY (4) in length. Given that keyidx + * is only allowed 2-bits of storage, no value of keyidx can + * be provided via above code that would result in keyidx * being out of range */ - crypt = ieee->crypt[idx]; + crypt = ieee->crypt[keyidx]; #ifdef NOT_YET sta = NULL; @@ -655,6 +654,51 @@ int ieee80211_rx(struct ieee80211_device *ieee, struct sk_buff *skb, goto rx_dropped; } + /* If the frame was decrypted in hardware, we may need to strip off + * any security data (IV, ICV, etc) that was left behind */ + if (!can_be_decrypted && (fc & IEEE80211_FCTL_PROTECTED) && + ieee->host_strip_iv_icv) { + int trimlen = 0; + + /* Top two-bits of byte 3 are the key index */ + if (skb->len >= hdrlen + 3) + keyidx = skb->data[hdrlen + 3] >> 6; + + /* To strip off any security data which appears before the + * payload, we simply increase hdrlen (as the header gets + * chopped off immediately below). For the security data which + * appears after the payload, we use skb_trim. */ + + switch (ieee->sec.encode_alg[keyidx]) { + case SEC_ALG_WEP: + /* 4 byte IV */ + hdrlen += 4; + /* 4 byte ICV */ + trimlen = 4; + break; + case SEC_ALG_TKIP: + /* 4 byte IV, 4 byte ExtIV */ + hdrlen += 8; + /* 8 byte MIC, 4 byte ICV */ + trimlen = 12; + break; + case SEC_ALG_CCMP: + /* 8 byte CCMP header */ + hdrlen += 8; + /* 8 byte MIC */ + trimlen = 8; + break; + } + + if (skb->len < trimlen) + goto rx_dropped; + + __skb_trim(skb, skb->len - trimlen); + + if (skb->len < hdrlen) + goto rx_dropped; + } + /* skb: hdr + (possible reassembled) full plaintext payload */ payload = skb->data + hdrlen; -- cgit v1.2.3-70-g09d2 From 837925df022a667c302b24aad9d6a58f94efd959 Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Tue, 3 Oct 2006 18:49:32 -0500 Subject: [PATCH] ieee80211: Drop and count duplicate data frames to remove 'replay detected' log messages In the SoftMAC version of the IEEE 802.11 stack, not all duplicate messages are detected. For the most part, there is no difficulty; however for TKIP and CCMP encryption, the duplicates result in a "replay detected" log message where the received and previous values of the TSC are identical. This change adds a new variable to the ieee80211_device structure that holds the 'seq_ctl' value for the previous frame. When a new frame repeats the value, the frame is dropped and the appropriate counter is incremented. Signed-off-by: Larry Finger Signed-off-by: John W. Linville --- include/net/ieee80211.h | 2 ++ net/ieee80211/ieee80211_rx.c | 5 +++++ 2 files changed, 7 insertions(+) (limited to 'include') diff --git a/include/net/ieee80211.h b/include/net/ieee80211.h index cb255432e4e..e6af381e206 100644 --- a/include/net/ieee80211.h +++ b/include/net/ieee80211.h @@ -1080,6 +1080,8 @@ struct ieee80211_device { int perfect_rssi; int worst_rssi; + u16 prev_seq_ctl; /* used to drop duplicate frames */ + /* Callback functions */ void (*set_security) (struct net_device * dev, struct ieee80211_security * sec); diff --git a/net/ieee80211/ieee80211_rx.c b/net/ieee80211/ieee80211_rx.c index d9265195656..ce28d571afa 100644 --- a/net/ieee80211/ieee80211_rx.c +++ b/net/ieee80211/ieee80211_rx.c @@ -478,6 +478,11 @@ int ieee80211_rx(struct ieee80211_device *ieee, struct sk_buff *skb, goto rx_exit; } #endif + /* drop duplicate 802.11 retransmissions (IEEE 802.11 Chap. 9.29) */ + if (sc == ieee->prev_seq_ctl) + goto rx_dropped; + else + ieee->prev_seq_ctl = sc; /* Data frame - extract src/dst addresses */ if (skb->len < IEEE80211_3ADDR_LEN) -- cgit v1.2.3-70-g09d2 From fe75f7471ba5604fe65435f717e3612a482c28cb Mon Sep 17 00:00:00 2001 From: Christian Lamparter Date: Mon, 2 Oct 2006 19:55:22 +0200 Subject: [PATCH] wext: extend MLME support This patch adds two new defines for the SIOCSIWMLME to cover all kinds MLMEs (well, except REASSOC) through a ioctl. Signed-off-by: Christian Lamparter Signed-off-by: John W. Linville --- include/linux/wireless.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/linux/wireless.h b/include/linux/wireless.h index a50a0130fd9..7c269f4992e 100644 --- a/include/linux/wireless.h +++ b/include/linux/wireless.h @@ -546,6 +546,8 @@ /* MLME requests (SIOCSIWMLME / struct iw_mlme) */ #define IW_MLME_DEAUTH 0 #define IW_MLME_DISASSOC 1 +#define IW_MLME_AUTH 2 +#define IW_MLME_ASSOC 3 /* SIOCSIWAUTH/SIOCGIWAUTH struct iw_param flags */ #define IW_AUTH_INDEX 0x0FFF -- cgit v1.2.3-70-g09d2 From 52d78d63310d9818458fd9800d24a4d5425aeac7 Mon Sep 17 00:00:00 2001 From: Ayaz Abdulla Date: Mon, 6 Nov 2006 00:43:39 -0800 Subject: [PATCH] forcedeth: add new NVIDIA pci ids Add pci device ids for the NVIDIA MCP67 chip. Signed-off-by: Ayaz Abdulla Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- include/linux/pci_ids.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index e060a763794..fd5033b8a92 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1213,6 +1213,10 @@ #define PCI_DEVICE_ID_NVIDIA_NVENET_21 0x0451 #define PCI_DEVICE_ID_NVIDIA_NVENET_22 0x0452 #define PCI_DEVICE_ID_NVIDIA_NVENET_23 0x0453 +#define PCI_DEVICE_ID_NVIDIA_NVENET_24 0x054C +#define PCI_DEVICE_ID_NVIDIA_NVENET_25 0x054D +#define PCI_DEVICE_ID_NVIDIA_NVENET_26 0x054E +#define PCI_DEVICE_ID_NVIDIA_NVENET_27 0x054F #define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP67_IDE 0x0560 #define PCI_VENDOR_ID_IMS 0x10e0 -- cgit v1.2.3-70-g09d2 From f789dfdc44d5bbc04fb7f06e1e4eb682169acaaf Mon Sep 17 00:00:00 2001 From: Mariusz Kozlowski Date: Thu, 30 Nov 2006 04:27:00 -0700 Subject: [PATCH] mv643xx_eth: fix unbalanced parentheses in macros Signed-off-by: Mariusz Kozlowski Signed-off-by: Dale Farnsworth Signed-off-by: Jeff Garzik --- include/linux/mv643xx.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/mv643xx.h b/include/linux/mv643xx.h index edfa012fad3..aff25c000ab 100644 --- a/include/linux/mv643xx.h +++ b/include/linux/mv643xx.h @@ -724,7 +724,7 @@ #define MV643XX_ETH_RX_FIFO_URGENT_THRESHOLD_REG(port) (0x2470 + (port<<10)) #define MV643XX_ETH_TX_FIFO_URGENT_THRESHOLD_REG(port) (0x2474 + (port<<10)) #define MV643XX_ETH_RX_MINIMAL_FRAME_SIZE_REG(port) (0x247c + (port<<10)) -#define MV643XX_ETH_RX_DISCARDED_FRAMES_COUNTER(port) (0x2484 + (port<<10) +#define MV643XX_ETH_RX_DISCARDED_FRAMES_COUNTER(port) (0x2484 + (port<<10)) #define MV643XX_ETH_PORT_DEBUG_0_REG(port) (0x248c + (port<<10)) #define MV643XX_ETH_PORT_DEBUG_1_REG(port) (0x2490 + (port<<10)) #define MV643XX_ETH_PORT_INTERNAL_ADDR_ERROR_REG(port) (0x2494 + (port<<10)) @@ -1135,7 +1135,7 @@ struct mv64xxx_i2c_pdata { #define MV643XX_ETH_DEFAULT_RX_UDP_QUEUE_1 (1<<19) #define MV643XX_ETH_DEFAULT_RX_UDP_QUEUE_2 (1<<20) #define MV643XX_ETH_DEFAULT_RX_UDP_QUEUE_3 ((1<<20) | (1<<19)) -#define MV643XX_ETH_DEFAULT_RX_UDP_QUEUE_4 ((1<<21) +#define MV643XX_ETH_DEFAULT_RX_UDP_QUEUE_4 (1<<21) #define MV643XX_ETH_DEFAULT_RX_UDP_QUEUE_5 ((1<<21) | (1<<19)) #define MV643XX_ETH_DEFAULT_RX_UDP_QUEUE_6 ((1<<21) | (1<<20)) #define MV643XX_ETH_DEFAULT_RX_UDP_QUEUE_7 ((1<<21) | (1<<20) | (1<<19)) -- cgit v1.2.3-70-g09d2 From e8a2b6a4207332a2d59628a12cece9e8c1d769e4 Mon Sep 17 00:00:00 2001 From: Andy Fleming Date: Fri, 1 Dec 2006 12:01:06 -0600 Subject: [PATCH] PHY: Add support for configuring the PHY connection interface Most PHYs connect to an ethernet controller over a GMII or MII interface. However, a growing number are connected over different interfaces, such as RGMII or SGMII. The ethernet driver will tell the PHY what type of connection it is by setting it manually, or passing it in through phy_connect (or phy_attach). Changes include: * Updates to documentation * Updates to PHY Lib consumers * Changes to PHY Lib to add interface support * Some minor changes to whitespace in phy.h * gianfar driver now detects interface and passes appropriate value to PHY Lib Signed-off-by: Andrew Fleming Signed-off-by: Jeff Garzik --- Documentation/networking/phy.txt | 13 +++++++++---- drivers/net/au1000_eth.c | 3 ++- drivers/net/fs_enet/fs_enet-main.c | 3 ++- drivers/net/gianfar.c | 39 ++++++++++++++++++++++++++++++++++++-- drivers/net/gianfar.h | 3 +++ drivers/net/phy/phy_device.c | 29 +++++++++++++++++++--------- include/linux/phy.h | 26 +++++++++++++++++++------ 7 files changed, 93 insertions(+), 23 deletions(-) (limited to 'include') diff --git a/Documentation/networking/phy.txt b/Documentation/networking/phy.txt index 29ccae40903..0bc95eab151 100644 --- a/Documentation/networking/phy.txt +++ b/Documentation/networking/phy.txt @@ -1,7 +1,7 @@ ------- PHY Abstraction Layer -(Updated 2005-07-21) +(Updated 2006-11-30) Purpose @@ -97,11 +97,12 @@ Letting the PHY Abstraction Layer do Everything Next, you need to know the device name of the PHY connected to this device. The name will look something like, "phy0:0", where the first number is the - bus id, and the second is the PHY's address on that bus. + bus id, and the second is the PHY's address on that bus. Typically, + the bus is responsible for making its ID unique. Now, to connect, just call this function: - phydev = phy_connect(dev, phy_name, &adjust_link, flags); + phydev = phy_connect(dev, phy_name, &adjust_link, flags, interface); phydev is a pointer to the phy_device structure which represents the PHY. If phy_connect is successful, it will return the pointer. dev, here, is the @@ -115,6 +116,10 @@ Letting the PHY Abstraction Layer do Everything This is useful if the system has put hardware restrictions on the PHY/controller, of which the PHY needs to be aware. + interface is a u32 which specifies the connection type used + between the controller and the PHY. Examples are GMII, MII, + RGMII, and SGMII. For a full list, see include/linux/phy.h + Now just make sure that phydev->supported and phydev->advertising have any values pruned from them which don't make sense for your controller (a 10/100 controller may be connected to a gigabit capable PHY, so you would need to @@ -191,7 +196,7 @@ Doing it all yourself start, or disables then frees them for stop. struct phy_device * phy_attach(struct net_device *dev, const char *phy_id, - u32 flags); + u32 flags, phy_interface_t interface); Attaches a network device to a particular PHY, binding the PHY to a generic driver if none was found during bus initialization. Passes in diff --git a/drivers/net/au1000_eth.c b/drivers/net/au1000_eth.c index 7db3c8af089..f0b6879a1c7 100644 --- a/drivers/net/au1000_eth.c +++ b/drivers/net/au1000_eth.c @@ -360,7 +360,8 @@ static int mii_probe (struct net_device *dev) BUG_ON(!phydev); BUG_ON(phydev->attached_dev); - phydev = phy_connect(dev, phydev->dev.bus_id, &au1000_adjust_link, 0); + phydev = phy_connect(dev, phydev->dev.bus_id, &au1000_adjust_link, 0, + PHY_INTERFACE_MODE_MII); if (IS_ERR(phydev)) { printk(KERN_ERR "%s: Could not attach to PHY\n", dev->name); diff --git a/drivers/net/fs_enet/fs_enet-main.c b/drivers/net/fs_enet/fs_enet-main.c index cb3958704a8..889d3a13e95 100644 --- a/drivers/net/fs_enet/fs_enet-main.c +++ b/drivers/net/fs_enet/fs_enet-main.c @@ -779,7 +779,8 @@ static int fs_init_phy(struct net_device *dev) fep->oldspeed = 0; fep->oldduplex = -1; if(fep->fpi->bus_id) - phydev = phy_connect(dev, fep->fpi->bus_id, &fs_adjust_link, 0); + phydev = phy_connect(dev, fep->fpi->bus_id, &fs_adjust_link, 0, + PHY_INTERFACE_MODE_MII); else { printk("No phy bus ID specified in BSP code\n"); return -EINVAL; diff --git a/drivers/net/gianfar.c b/drivers/net/gianfar.c index 6bf18c82083..baa35144134 100644 --- a/drivers/net/gianfar.c +++ b/drivers/net/gianfar.c @@ -9,7 +9,7 @@ * Author: Andy Fleming * Maintainer: Kumar Gala * - * Copyright (c) 2002-2004 Freescale Semiconductor, Inc. + * Copyright (c) 2002-2006 Freescale Semiconductor, Inc. * * This program is free software; you can redistribute it and/or modify it * under the terms of the GNU General Public License as published by the @@ -398,6 +398,38 @@ static int gfar_remove(struct platform_device *pdev) } +/* Reads the controller's registers to determine what interface + * connects it to the PHY. + */ +static phy_interface_t gfar_get_interface(struct net_device *dev) +{ + struct gfar_private *priv = netdev_priv(dev); + u32 ecntrl = gfar_read(&priv->regs->ecntrl); + + if (ecntrl & ECNTRL_SGMII_MODE) + return PHY_INTERFACE_MODE_SGMII; + + if (ecntrl & ECNTRL_TBI_MODE) { + if (ecntrl & ECNTRL_REDUCED_MODE) + return PHY_INTERFACE_MODE_RTBI; + else + return PHY_INTERFACE_MODE_TBI; + } + + if (ecntrl & ECNTRL_REDUCED_MODE) { + if (ecntrl & ECNTRL_REDUCED_MII_MODE) + return PHY_INTERFACE_MODE_RMII; + else + return PHY_INTERFACE_MODE_RGMII; + } + + if (priv->einfo->device_flags & FSL_GIANFAR_DEV_HAS_GIGABIT) + return PHY_INTERFACE_MODE_GMII; + + return PHY_INTERFACE_MODE_MII; +} + + /* Initializes driver's PHY state, and attaches to the PHY. * Returns 0 on success. */ @@ -409,6 +441,7 @@ static int init_phy(struct net_device *dev) SUPPORTED_1000baseT_Full : 0; struct phy_device *phydev; char phy_id[BUS_ID_SIZE]; + phy_interface_t interface; priv->oldlink = 0; priv->oldspeed = 0; @@ -416,7 +449,9 @@ static int init_phy(struct net_device *dev) snprintf(phy_id, BUS_ID_SIZE, PHY_ID_FMT, priv->einfo->bus_id, priv->einfo->phy_id); - phydev = phy_connect(dev, phy_id, &adjust_link, 0); + interface = gfar_get_interface(dev); + + phydev = phy_connect(dev, phy_id, &adjust_link, 0, interface); if (IS_ERR(phydev)) { printk(KERN_ERR "%s: Could not attach to PHY\n", dev->name); diff --git a/drivers/net/gianfar.h b/drivers/net/gianfar.h index 9e81a50cf2b..39e9e321fcb 100644 --- a/drivers/net/gianfar.h +++ b/drivers/net/gianfar.h @@ -160,7 +160,10 @@ extern const char gfar_driver_version[]; #define ECNTRL_INIT_SETTINGS 0x00001000 #define ECNTRL_TBI_MODE 0x00000020 +#define ECNTRL_REDUCED_MODE 0x00000010 #define ECNTRL_R100 0x00000008 +#define ECNTRL_REDUCED_MII_MODE 0x00000004 +#define ECNTRL_SGMII_MODE 0x00000002 #define MRBLR_INIT_SETTINGS DEFAULT_RX_BUFFER_SIZE diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 2a08b2b62c4..b01fc70a57d 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -59,6 +59,7 @@ struct phy_device* phy_device_create(struct mii_bus *bus, int addr, int phy_id) dev->duplex = -1; dev->pause = dev->asym_pause = 0; dev->link = 1; + dev->interface = PHY_INTERFACE_MODE_GMII; dev->autoneg = AUTONEG_ENABLE; @@ -137,11 +138,12 @@ void phy_prepare_link(struct phy_device *phydev, * the desired functionality. */ struct phy_device * phy_connect(struct net_device *dev, const char *phy_id, - void (*handler)(struct net_device *), u32 flags) + void (*handler)(struct net_device *), u32 flags, + u32 interface) { struct phy_device *phydev; - phydev = phy_attach(dev, phy_id, flags); + phydev = phy_attach(dev, phy_id, flags, interface); if (IS_ERR(phydev)) return phydev; @@ -186,7 +188,7 @@ static int phy_compare_id(struct device *dev, void *data) } struct phy_device *phy_attach(struct net_device *dev, - const char *phy_id, u32 flags) + const char *phy_id, u32 flags, u32 interface) { struct bus_type *bus = &mdio_bus_type; struct phy_device *phydev; @@ -231,6 +233,20 @@ struct phy_device *phy_attach(struct net_device *dev, phydev->dev_flags = flags; + phydev->interface = interface; + + /* Do initial configuration here, now that + * we have certain key parameters + * (dev_flags and interface) */ + if (phydev->drv->config_init) { + int err; + + err = phydev->drv->config_init(phydev); + + if (err < 0) + return ERR_PTR(err); + } + return phydev; } EXPORT_SYMBOL(phy_attach); @@ -612,13 +628,8 @@ static int phy_probe(struct device *dev) spin_unlock(&phydev->lock); - if (err < 0) - return err; - - if (phydev->drv->config_init) - err = phydev->drv->config_init(phydev); - return err; + } static int phy_remove(struct device *dev) diff --git a/include/linux/phy.h b/include/linux/phy.h index ce8bc80b3c8..edd4c88ca7d 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -47,15 +47,26 @@ #define PHY_HAS_INTERRUPT 0x00000001 #define PHY_HAS_MAGICANEG 0x00000002 +/* Interface Mode definitions */ +typedef enum { + PHY_INTERFACE_MODE_MII, + PHY_INTERFACE_MODE_GMII, + PHY_INTERFACE_MODE_SGMII, + PHY_INTERFACE_MODE_TBI, + PHY_INTERFACE_MODE_RMII, + PHY_INTERFACE_MODE_RGMII, + PHY_INTERFACE_MODE_RTBI +} phy_interface_t; + #define MII_BUS_MAX 4 -#define PHY_INIT_TIMEOUT 100000 +#define PHY_INIT_TIMEOUT 100000 #define PHY_STATE_TIME 1 #define PHY_FORCE_TIMEOUT 10 #define PHY_AN_TIMEOUT 10 -#define PHY_MAX_ADDR 32 +#define PHY_MAX_ADDR 32 /* Used when trying to connect to a specific phy (mii bus id:phy device id) */ #define PHY_ID_FMT "%x:%02x" @@ -87,8 +98,8 @@ struct mii_bus { int *irq; }; -#define PHY_INTERRUPT_DISABLED 0x0 -#define PHY_INTERRUPT_ENABLED 0x80000000 +#define PHY_INTERRUPT_DISABLED 0x0 +#define PHY_INTERRUPT_ENABLED 0x80000000 /* PHY state machine states: * @@ -230,6 +241,8 @@ struct phy_device { u32 dev_flags; + phy_interface_t interface; + /* Bus address of the PHY (0-32) */ int addr; @@ -345,9 +358,10 @@ struct phy_device* get_phy_device(struct mii_bus *bus, int addr); int phy_clear_interrupt(struct phy_device *phydev); int phy_config_interrupt(struct phy_device *phydev, u32 interrupts); struct phy_device * phy_attach(struct net_device *dev, - const char *phy_id, u32 flags); + const char *phy_id, u32 flags, phy_interface_t interface); struct phy_device * phy_connect(struct net_device *dev, const char *phy_id, - void (*handler)(struct net_device *), u32 flags); + void (*handler)(struct net_device *), u32 flags, + phy_interface_t interface); void phy_disconnect(struct phy_device *phydev); void phy_detach(struct phy_device *phydev); void phy_start(struct phy_device *phydev); -- cgit v1.2.3-70-g09d2 From 3e6c8cd5669c1202fe806ce3e13d701f20a71c7e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:19:09 -0800 Subject: [TIPC]: endianness annotations Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/tipc_config.h | 32 ++++++++++++++++---------------- include/net/tipc/tipc_bearer.h | 2 +- include/net/tipc/tipc_msg.h | 2 +- net/tipc/config.c | 32 +++++++++++--------------------- net/tipc/dbg.c | 3 +-- net/tipc/name_distr.c | 10 +++++----- net/tipc/node.c | 9 +++------ 7 files changed, 38 insertions(+), 52 deletions(-) (limited to 'include') diff --git a/include/linux/tipc_config.h b/include/linux/tipc_config.h index 33a653913d9..b0c916d1f37 100644 --- a/include/linux/tipc_config.h +++ b/include/linux/tipc_config.h @@ -194,34 +194,34 @@ struct tipc_node_info { - __u32 addr; /* network address of node */ - __u32 up; /* 0=down, 1= up */ + __be32 addr; /* network address of node */ + __be32 up; /* 0=down, 1= up */ }; struct tipc_link_info { - __u32 dest; /* network address of peer node */ - __u32 up; /* 0=down, 1=up */ + __be32 dest; /* network address of peer node */ + __be32 up; /* 0=down, 1=up */ char str[TIPC_MAX_LINK_NAME]; /* link name */ }; struct tipc_bearer_config { - __u32 priority; /* Range [1,31]. Override per link */ - __u32 detect_scope; + __be32 priority; /* Range [1,31]. Override per link */ + __be32 detect_scope; char name[TIPC_MAX_BEARER_NAME]; }; struct tipc_link_config { - __u32 value; + __be32 value; char name[TIPC_MAX_LINK_NAME]; }; #define TIPC_NTQ_ALLTYPES 0x80000000 struct tipc_name_table_query { - __u32 depth; /* 1:type, 2:+name info, 3:+port info, 4+:+debug info */ - __u32 type; /* {t,l,u} info ignored if high bit of "depth" is set */ - __u32 lowbound; /* (i.e. displays all entries of name table) */ - __u32 upbound; + __be32 depth; /* 1:type, 2:+name info, 3:+port info, 4+:+debug info */ + __be32 type; /* {t,l,u} info ignored if high bit of "depth" is set */ + __be32 lowbound; /* (i.e. displays all entries of name table) */ + __be32 upbound; }; /* @@ -262,8 +262,8 @@ struct tipc_route_info { */ struct tlv_desc { - __u16 tlv_len; /* TLV length (descriptor + value) */ - __u16 tlv_type; /* TLV identifier */ + __be16 tlv_len; /* TLV length (descriptor + value) */ + __be16 tlv_type; /* TLV identifier */ }; #define TLV_ALIGNTO 4 @@ -377,9 +377,9 @@ struct tipc_genlmsghdr { struct tipc_cfg_msg_hdr { - __u32 tcm_len; /* Message length (including header) */ - __u16 tcm_type; /* Command type */ - __u16 tcm_flags; /* Additional flags */ + __be32 tcm_len; /* Message length (including header) */ + __be16 tcm_type; /* Command type */ + __be16 tcm_flags; /* Additional flags */ char tcm_reserved[8]; /* Unused */ }; diff --git a/include/net/tipc/tipc_bearer.h b/include/net/tipc/tipc_bearer.h index e07136d74c2..2151a80cdf3 100644 --- a/include/net/tipc/tipc_bearer.h +++ b/include/net/tipc/tipc_bearer.h @@ -58,7 +58,7 @@ */ struct tipc_media_addr { - __u32 type; /* bearer type (network byte order) */ + __be32 type; /* bearer type (network byte order) */ union { __u8 eth_addr[6]; /* 48 bit Ethernet addr (byte array) */ #if 0 diff --git a/include/net/tipc/tipc_msg.h b/include/net/tipc/tipc_msg.h index 4d096eebc93..fb42eb7a86a 100644 --- a/include/net/tipc/tipc_msg.h +++ b/include/net/tipc/tipc_msg.h @@ -40,7 +40,7 @@ #ifdef __KERNEL__ struct tipc_msg { - u32 hdr[15]; + __be32 hdr[15]; }; diff --git a/net/tipc/config.c b/net/tipc/config.c index ed1351ed05e..458a2c46cef 100644 --- a/net/tipc/config.c +++ b/net/tipc/config.c @@ -107,7 +107,7 @@ int tipc_cfg_append_tlv(struct sk_buff *buf, int tlv_type, struct sk_buff *tipc_cfg_reply_unsigned_type(u16 tlv_type, u32 value) { struct sk_buff *buf; - u32 value_net; + __be32 value_net; buf = tipc_cfg_reply_alloc(TLV_SPACE(sizeof(value))); if (buf) { @@ -284,8 +284,7 @@ static struct sk_buff *cfg_set_own_addr(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NET_ADDR)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - addr = *(u32 *)TLV_DATA(req_tlv_area); - addr = ntohl(addr); + addr = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (addr == tipc_own_addr) return tipc_cfg_reply_none(); if (!tipc_addr_node_valid(addr)) @@ -319,8 +318,7 @@ static struct sk_buff *cfg_set_remote_mng(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); tipc_remote_management = (value != 0); return tipc_cfg_reply_none(); } @@ -332,8 +330,7 @@ static struct sk_buff *cfg_set_max_publications(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value != delimit(value, 1, 65535)) return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE " (max publications must be 1-65535)"); @@ -348,8 +345,7 @@ static struct sk_buff *cfg_set_max_subscriptions(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value != delimit(value, 1, 65535)) return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE " (max subscriptions must be 1-65535"); @@ -363,8 +359,7 @@ static struct sk_buff *cfg_set_max_ports(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value == tipc_max_ports) return tipc_cfg_reply_none(); if (value != delimit(value, 127, 65535)) @@ -383,8 +378,7 @@ static struct sk_buff *cfg_set_max_zones(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value == tipc_max_zones) return tipc_cfg_reply_none(); if (value != delimit(value, 1, 255)) @@ -403,8 +397,7 @@ static struct sk_buff *cfg_set_max_clusters(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value != delimit(value, 1, 1)) return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE " (max clusters fixed at 1)"); @@ -417,8 +410,7 @@ static struct sk_buff *cfg_set_max_nodes(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value == tipc_max_nodes) return tipc_cfg_reply_none(); if (value != delimit(value, 8, 2047)) @@ -437,8 +429,7 @@ static struct sk_buff *cfg_set_max_slaves(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value != 0) return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED " (max secondary nodes fixed at 0)"); @@ -451,8 +442,7 @@ static struct sk_buff *cfg_set_netid(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value == tipc_net_id) return tipc_cfg_reply_none(); if (value != delimit(value, 1, 9999)) diff --git a/net/tipc/dbg.c b/net/tipc/dbg.c index d8af4c28695..627f99b7afd 100644 --- a/net/tipc/dbg.c +++ b/net/tipc/dbg.c @@ -393,8 +393,7 @@ struct sk_buff *tipc_log_resize(const void *req_tlv_area, int req_tlv_space) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value != delimit(value, 0, 32768)) return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE " (log size must be 0-32768)"); diff --git a/net/tipc/name_distr.c b/net/tipc/name_distr.c index 03bd659c43c..7bf87cb26ef 100644 --- a/net/tipc/name_distr.c +++ b/net/tipc/name_distr.c @@ -66,11 +66,11 @@ */ struct distr_item { - u32 type; - u32 lower; - u32 upper; - u32 ref; - u32 key; + __be32 type; + __be32 lower; + __be32 upper; + __be32 ref; + __be32 key; }; /** diff --git a/net/tipc/node.c b/net/tipc/node.c index 886bda5e88d..106cd0dfac7 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -597,8 +597,7 @@ struct sk_buff *tipc_node_get_nodes(const void *req_tlv_area, int req_tlv_space) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NET_ADDR)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - domain = *(u32 *)TLV_DATA(req_tlv_area); - domain = ntohl(domain); + domain = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (!tipc_addr_domain_valid(domain)) return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE " (network address)"); @@ -642,8 +641,7 @@ struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NET_ADDR)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - domain = *(u32 *)TLV_DATA(req_tlv_area); - domain = ntohl(domain); + domain = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (!tipc_addr_domain_valid(domain)) return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE " (network address)"); @@ -664,8 +662,7 @@ struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space) /* Add TLV for broadcast link */ - link_info.dest = tipc_own_addr & 0xfffff00; - link_info.dest = htonl(link_info.dest); + link_info.dest = htonl(tipc_own_addr & 0xfffff00); link_info.up = htonl(1); sprintf(link_info.str, tipc_bclink_name); tipc_cfg_append_tlv(buf, TIPC_TLV_LINK_INFO, &link_info, sizeof(link_info)); -- cgit v1.2.3-70-g09d2 From 5a874db4d9bfd8a4c6324d844a4d1c7cfa5cf2c4 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:19:38 -0800 Subject: [NET]: ipconfig and nfsroot annotations Signed-off-by: Al Viro Signed-off-by: David S. Miller --- fs/nfs/nfsroot.c | 13 +++--- include/linux/nfs_fs.h | 2 +- include/net/ipconfig.h | 8 ++-- net/ipv4/ipconfig.c | 105 +++++++++++++++++++++++++------------------------ 4 files changed, 64 insertions(+), 64 deletions(-) (limited to 'include') diff --git a/fs/nfs/nfsroot.c b/fs/nfs/nfsroot.c index 8dfefe41a8d..75f819dc025 100644 --- a/fs/nfs/nfsroot.c +++ b/fs/nfs/nfsroot.c @@ -98,7 +98,7 @@ static char nfs_root_name[256] __initdata = ""; /* Address of NFS server */ -static __u32 servaddr __initdata = 0; +static __be32 servaddr __initdata = 0; /* Name of directory to mount */ static char nfs_path[NFS_MAXPATHLEN] __initdata = { 0, }; @@ -327,7 +327,7 @@ static int __init root_nfs_name(char *name) */ static int __init root_nfs_addr(void) { - if ((servaddr = root_server_addr) == INADDR_NONE) { + if ((servaddr = root_server_addr) == htonl(INADDR_NONE)) { printk(KERN_ERR "Root-NFS: No NFS server available, giving up.\n"); return -1; } @@ -411,7 +411,7 @@ __setup("nfsroot=", nfs_root_setup); * Construct sockaddr_in from address and port number. */ static inline void -set_sockaddr(struct sockaddr_in *sin, __u32 addr, __u16 port) +set_sockaddr(struct sockaddr_in *sin, __be32 addr, __be16 port) { sin->sin_family = AF_INET; sin->sin_addr.s_addr = addr; @@ -468,14 +468,13 @@ static int __init root_nfs_ports(void) dprintk("Root-NFS: Portmapper on server returned %d " "as nfsd port\n", port); } - nfs_port = htons(nfs_port); if ((port = root_nfs_getport(NFS_MNT_PROGRAM, mountd_ver, proto)) < 0) { printk(KERN_ERR "Root-NFS: Unable to get mountd port " "number from server, using default\n"); port = mountd_port; } - mount_port = htons(port); + mount_port = port; dprintk("Root-NFS: mountd port is %d\n", port); return 0; @@ -496,7 +495,7 @@ static int __init root_nfs_get_handle(void) int version = (nfs_data.flags & NFS_MOUNT_VER3) ? NFS_MNT3_VERSION : NFS_MNT_VERSION; - set_sockaddr(&sin, servaddr, mount_port); + set_sockaddr(&sin, servaddr, htons(mount_port)); status = nfsroot_mount(&sin, nfs_path, &fh, version, protocol); if (status < 0) printk(KERN_ERR "Root-NFS: Server returned error %d " @@ -519,6 +518,6 @@ void * __init nfs_root_data(void) || root_nfs_ports() < 0 || root_nfs_get_handle() < 0) return NULL; - set_sockaddr((struct sockaddr_in *) &nfs_data.addr, servaddr, nfs_port); + set_sockaddr((struct sockaddr_in *) &nfs_data.addr, servaddr, htons(nfs_port)); return (void*)&nfs_data; } diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 45228c1a119..625ffea9856 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -318,7 +318,7 @@ extern void put_nfs_open_context(struct nfs_open_context *ctx); extern struct nfs_open_context *nfs_find_open_context(struct inode *inode, struct rpc_cred *cred, int mode); /* linux/net/ipv4/ipconfig.c: trims ip addr off front of name, too. */ -extern u32 root_nfs_parse_addr(char *name); /*__init*/ +extern __be32 root_nfs_parse_addr(char *name); /*__init*/ static inline void nfs_fattr_init(struct nfs_fattr *fattr) { diff --git a/include/net/ipconfig.h b/include/net/ipconfig.h index 2a1fe996fbc..3924d7d2cb1 100644 --- a/include/net/ipconfig.h +++ b/include/net/ipconfig.h @@ -11,12 +11,12 @@ extern int ic_proto_enabled; /* Protocols enabled (see IC_xxx) */ extern int ic_set_manually; /* IPconfig parameters set manually */ -extern u32 ic_myaddr; /* My IP address */ -extern u32 ic_gateway; /* Gateway IP address */ +extern __be32 ic_myaddr; /* My IP address */ +extern __be32 ic_gateway; /* Gateway IP address */ -extern u32 ic_servaddr; /* Boot server IP address */ +extern __be32 ic_servaddr; /* Boot server IP address */ -extern u32 root_server_addr; /* Address of NFS server */ +extern __be32 root_server_addr; /* Address of NFS server */ extern u8 root_server_path[]; /* Path to mount as root */ diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c index 955a07abb91..afa60b9a003 100644 --- a/net/ipv4/ipconfig.c +++ b/net/ipv4/ipconfig.c @@ -101,6 +101,7 @@ #define CONF_NAMESERVERS_MAX 3 /* Maximum number of nameservers - '3' from resolv.h */ +#define NONE __constant_htonl(INADDR_NONE) /* * Public IP configuration @@ -129,19 +130,19 @@ int ic_proto_enabled __initdata = 0 static int ic_host_name_set __initdata = 0; /* Host name set by us? */ -u32 ic_myaddr = INADDR_NONE; /* My IP address */ -static u32 ic_netmask = INADDR_NONE; /* Netmask for local subnet */ -u32 ic_gateway = INADDR_NONE; /* Gateway IP address */ +__be32 ic_myaddr = NONE; /* My IP address */ +static __be32 ic_netmask = NONE; /* Netmask for local subnet */ +__be32 ic_gateway = NONE; /* Gateway IP address */ -u32 ic_servaddr = INADDR_NONE; /* Boot server IP address */ +__be32 ic_servaddr = NONE; /* Boot server IP address */ -u32 root_server_addr = INADDR_NONE; /* Address of NFS server */ +__be32 root_server_addr = NONE; /* Address of NFS server */ u8 root_server_path[256] = { 0, }; /* Path to mount as root */ /* Persistent data: */ static int ic_proto_used; /* Protocol used, if any */ -static u32 ic_nameservers[CONF_NAMESERVERS_MAX]; /* DNS Server IP addresses */ +static __be32 ic_nameservers[CONF_NAMESERVERS_MAX]; /* DNS Server IP addresses */ static u8 ic_domain[64]; /* DNS (not NIS) domain name */ /* @@ -172,7 +173,7 @@ struct ic_device { struct net_device *dev; unsigned short flags; short able; - u32 xid; + __be32 xid; }; static struct ic_device *ic_first_dev __initdata = NULL;/* List of open device */ @@ -223,7 +224,7 @@ static int __init ic_open_devs(void) d->flags = oflags; d->able = able; if (able & IC_BOOTP) - get_random_bytes(&d->xid, sizeof(u32)); + get_random_bytes(&d->xid, sizeof(__be32)); else d->xid = 0; ic_proto_have_if |= able; @@ -269,7 +270,7 @@ static void __init ic_close_devs(void) */ static inline void -set_sockaddr(struct sockaddr_in *sin, u32 addr, u16 port) +set_sockaddr(struct sockaddr_in *sin, __be32 addr, __be16 port) { sin->sin_family = AF_INET; sin->sin_addr.s_addr = addr; @@ -332,7 +333,7 @@ static int __init ic_setup_routes(void) { /* No need to setup device routes, only the default route... */ - if (ic_gateway != INADDR_NONE) { + if (ic_gateway != NONE) { struct rtentry rm; int err; @@ -368,10 +369,10 @@ static int __init ic_defaults(void) if (!ic_host_name_set) sprintf(init_utsname()->nodename, "%u.%u.%u.%u", NIPQUAD(ic_myaddr)); - if (root_server_addr == INADDR_NONE) + if (root_server_addr == NONE) root_server_addr = ic_servaddr; - if (ic_netmask == INADDR_NONE) { + if (ic_netmask == NONE) { if (IN_CLASSA(ntohl(ic_myaddr))) ic_netmask = htonl(IN_CLASSA_NET); else if (IN_CLASSB(ntohl(ic_myaddr))) @@ -420,7 +421,7 @@ ic_rarp_recv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt { struct arphdr *rarp; unsigned char *rarp_ptr; - u32 sip, tip; + __be32 sip, tip; unsigned char *sha, *tha; /* s for "source", t for "target" */ struct ic_device *d; @@ -485,12 +486,12 @@ ic_rarp_recv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt goto drop_unlock; /* Discard packets which are not from specified server. */ - if (ic_servaddr != INADDR_NONE && ic_servaddr != sip) + if (ic_servaddr != NONE && ic_servaddr != sip) goto drop_unlock; /* We have a winner! */ ic_dev = dev; - if (ic_myaddr == INADDR_NONE) + if (ic_myaddr == NONE) ic_myaddr = tip; ic_servaddr = sip; ic_got_reply = IC_RARP; @@ -530,13 +531,13 @@ struct bootp_pkt { /* BOOTP packet format */ u8 htype; /* HW address type */ u8 hlen; /* HW address length */ u8 hops; /* Used only by gateways */ - u32 xid; /* Transaction ID */ - u16 secs; /* Seconds since we started */ - u16 flags; /* Just what it says */ - u32 client_ip; /* Client's IP address if known */ - u32 your_ip; /* Assigned IP address */ - u32 server_ip; /* (Next, e.g. NFS) Server's IP address */ - u32 relay_ip; /* IP address of BOOTP relay */ + __be32 xid; /* Transaction ID */ + __be16 secs; /* Seconds since we started */ + __be16 flags; /* Just what it says */ + __be32 client_ip; /* Client's IP address if known */ + __be32 your_ip; /* Assigned IP address */ + __be32 server_ip; /* (Next, e.g. NFS) Server's IP address */ + __be32 relay_ip; /* IP address of BOOTP relay */ u8 hw_addr[16]; /* Client's HW address */ u8 serv_name[64]; /* Server host name */ u8 boot_file[128]; /* Name of boot file */ @@ -576,7 +577,7 @@ static const u8 ic_bootp_cookie[4] = { 99, 130, 83, 99 }; static void __init ic_dhcp_init_options(u8 *options) { - u8 mt = ((ic_servaddr == INADDR_NONE) + u8 mt = ((ic_servaddr == NONE) ? DHCPDISCOVER : DHCPREQUEST); u8 *e = options; @@ -666,7 +667,7 @@ static inline void ic_bootp_init(void) int i; for (i = 0; i < CONF_NAMESERVERS_MAX; i++) - ic_nameservers[i] = INADDR_NONE; + ic_nameservers[i] = NONE; dev_add_pack(&bootp_packet_type); } @@ -708,7 +709,7 @@ static void __init ic_bootp_send_if(struct ic_device *d, unsigned long jiffies_d h->frag_off = htons(IP_DF); h->ttl = 64; h->protocol = IPPROTO_UDP; - h->daddr = INADDR_BROADCAST; + h->daddr = htonl(INADDR_BROADCAST); h->check = ip_fast_csum((unsigned char *) h, h->ihl); /* Construct UDP header */ @@ -730,8 +731,8 @@ static void __init ic_bootp_send_if(struct ic_device *d, unsigned long jiffies_d b->htype = dev->type; /* can cause undefined behavior */ } b->hlen = dev->addr_len; - b->your_ip = INADDR_NONE; - b->server_ip = INADDR_NONE; + b->your_ip = NONE; + b->server_ip = NONE; memcpy(b->hw_addr, dev->dev_addr, dev->addr_len); b->secs = htons(jiffies_diff / HZ); b->xid = d->xid; @@ -788,11 +789,11 @@ static void __init ic_do_bootp_ext(u8 *ext) switch (*ext++) { case 1: /* Subnet mask */ - if (ic_netmask == INADDR_NONE) + if (ic_netmask == NONE) memcpy(&ic_netmask, ext+1, 4); break; case 3: /* Default gateway */ - if (ic_gateway == INADDR_NONE) + if (ic_gateway == NONE) memcpy(&ic_gateway, ext+1, 4); break; case 6: /* DNS server */ @@ -800,7 +801,7 @@ static void __init ic_do_bootp_ext(u8 *ext) if (servers > CONF_NAMESERVERS_MAX) servers = CONF_NAMESERVERS_MAX; for (i = 0; i < servers; i++) { - if (ic_nameservers[i] == INADDR_NONE) + if (ic_nameservers[i] == NONE) memcpy(&ic_nameservers[i], ext+1+4*i, 4); } break; @@ -917,7 +918,7 @@ static int __init ic_bootp_recv(struct sk_buff *skb, struct net_device *dev, str #ifdef IPCONFIG_DHCP if (ic_proto_enabled & IC_USE_DHCP) { - u32 server_id = INADDR_NONE; + __be32 server_id = NONE; int mt = 0; ext = &b->exten[4]; @@ -949,7 +950,7 @@ static int __init ic_bootp_recv(struct sk_buff *skb, struct net_device *dev, str /* While in the process of accepting one offer, * ignore all others. */ - if (ic_myaddr != INADDR_NONE) + if (ic_myaddr != NONE) goto drop_unlock; /* Let's accept that offer. */ @@ -965,7 +966,7 @@ static int __init ic_bootp_recv(struct sk_buff *skb, struct net_device *dev, str * precedence over the bootp header one if * they are different. */ - if ((server_id != INADDR_NONE) && + if ((server_id != NONE) && (b->server_ip != server_id)) b->server_ip = ic_servaddr; break; @@ -979,8 +980,8 @@ static int __init ic_bootp_recv(struct sk_buff *skb, struct net_device *dev, str default: /* Urque. Forget it*/ - ic_myaddr = INADDR_NONE; - ic_servaddr = INADDR_NONE; + ic_myaddr = NONE; + ic_servaddr = NONE; goto drop_unlock; }; @@ -1004,9 +1005,9 @@ static int __init ic_bootp_recv(struct sk_buff *skb, struct net_device *dev, str ic_dev = dev; ic_myaddr = b->your_ip; ic_servaddr = b->server_ip; - if (ic_gateway == INADDR_NONE && b->relay_ip) + if (ic_gateway == NONE && b->relay_ip) ic_gateway = b->relay_ip; - if (ic_nameservers[0] == INADDR_NONE) + if (ic_nameservers[0] == NONE) ic_nameservers[0] = ic_servaddr; ic_got_reply = IC_BOOTP; @@ -1150,7 +1151,7 @@ static int __init ic_dynamic(void) #endif if (!ic_got_reply) { - ic_myaddr = INADDR_NONE; + ic_myaddr = NONE; return -1; } @@ -1182,12 +1183,12 @@ static int pnp_seq_show(struct seq_file *seq, void *v) seq_printf(seq, "domain %s\n", ic_domain); for (i = 0; i < CONF_NAMESERVERS_MAX; i++) { - if (ic_nameservers[i] != INADDR_NONE) + if (ic_nameservers[i] != NONE) seq_printf(seq, "nameserver %u.%u.%u.%u\n", NIPQUAD(ic_nameservers[i])); } - if (ic_servaddr != INADDR_NONE) + if (ic_servaddr != NONE) seq_printf(seq, "bootserver %u.%u.%u.%u\n", NIPQUAD(ic_servaddr)); @@ -1213,9 +1214,9 @@ static struct file_operations pnp_seq_fops = { * need to have root_server_addr set _before_ IPConfig gets called as it * can override it. */ -u32 __init root_nfs_parse_addr(char *name) +__be32 __init root_nfs_parse_addr(char *name) { - u32 addr; + __be32 addr; int octets = 0; char *cp, *cq; @@ -1237,7 +1238,7 @@ u32 __init root_nfs_parse_addr(char *name) addr = in_aton(name); memmove(name, cp, strlen(cp) + 1); } else - addr = INADDR_NONE; + addr = NONE; return addr; } @@ -1248,7 +1249,7 @@ u32 __init root_nfs_parse_addr(char *name) static int __init ip_auto_config(void) { - u32 addr; + __be32 addr; #ifdef CONFIG_PROC_FS proc_net_fops_create("pnp", S_IRUGO, &pnp_seq_fops); @@ -1277,11 +1278,11 @@ static int __init ip_auto_config(void) * interfaces and no default was set), use BOOTP or RARP to get the * missing values. */ - if (ic_myaddr == INADDR_NONE || + if (ic_myaddr == NONE || #ifdef CONFIG_ROOT_NFS (MAJOR(ROOT_DEV) == UNNAMED_MAJOR - && root_server_addr == INADDR_NONE - && ic_servaddr == INADDR_NONE) || + && root_server_addr == NONE + && ic_servaddr == NONE) || #endif ic_first_dev->next) { #ifdef IPCONFIG_DYNAMIC @@ -1334,7 +1335,7 @@ static int __init ip_auto_config(void) } addr = root_nfs_parse_addr(root_server_path); - if (root_server_addr == INADDR_NONE) + if (root_server_addr == NONE) root_server_addr = addr; /* @@ -1461,19 +1462,19 @@ static int __init ip_auto_config_setup(char *addrs) switch (num) { case 0: if ((ic_myaddr = in_aton(ip)) == INADDR_ANY) - ic_myaddr = INADDR_NONE; + ic_myaddr = NONE; break; case 1: if ((ic_servaddr = in_aton(ip)) == INADDR_ANY) - ic_servaddr = INADDR_NONE; + ic_servaddr = NONE; break; case 2: if ((ic_gateway = in_aton(ip)) == INADDR_ANY) - ic_gateway = INADDR_NONE; + ic_gateway = NONE; break; case 3: if ((ic_netmask = in_aton(ip)) == INADDR_ANY) - ic_netmask = INADDR_NONE; + ic_netmask = NONE; break; case 4: if ((dp = strchr(ip, '.'))) { -- cgit v1.2.3-70-g09d2 From d2ecd9ccd0694278a8c1fa179d1d058ee66e2297 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:20:00 -0800 Subject: [IPV6]: annotate inet6_hashtables Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/inet6_hashtables.h | 16 ++++++++-------- net/ipv6/inet6_hashtables.c | 6 +++--- 2 files changed, 11 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/net/inet6_hashtables.h b/include/net/inet6_hashtables.h index bc6a71dce98..c28e424f53d 100644 --- a/include/net/inet6_hashtables.h +++ b/include/net/inet6_hashtables.h @@ -26,11 +26,11 @@ struct inet_hashinfo; /* I have no idea if this is a good hash for v6 or not. -DaveM */ static inline unsigned int inet6_ehashfn(const struct in6_addr *laddr, const u16 lport, - const struct in6_addr *faddr, const u16 fport) + const struct in6_addr *faddr, const __be16 fport) { - unsigned int hashent = (lport ^ fport); + unsigned int hashent = (lport ^ (__force u16)fport); - hashent ^= (laddr->s6_addr32[3] ^ faddr->s6_addr32[3]); + hashent ^= (__force u32)(laddr->s6_addr32[3] ^ faddr->s6_addr32[3]); hashent ^= hashent >> 16; hashent ^= hashent >> 8; return hashent; @@ -43,7 +43,7 @@ static inline int inet6_sk_ehashfn(const struct sock *sk) const struct in6_addr *laddr = &np->rcv_saddr; const struct in6_addr *faddr = &np->daddr; const __u16 lport = inet->num; - const __u16 fport = inet->dport; + const __be16 fport = inet->dport; return inet6_ehashfn(laddr, lport, faddr, fport); } @@ -57,7 +57,7 @@ extern void __inet6_hash(struct inet_hashinfo *hashinfo, struct sock *sk); */ extern struct sock *__inet6_lookup_established(struct inet_hashinfo *hashinfo, const struct in6_addr *saddr, - const u16 sport, + const __be16 sport, const struct in6_addr *daddr, const u16 hnum, const int dif); @@ -69,7 +69,7 @@ extern struct sock *inet6_lookup_listener(struct inet_hashinfo *hashinfo, static inline struct sock *__inet6_lookup(struct inet_hashinfo *hashinfo, const struct in6_addr *saddr, - const u16 sport, + const __be16 sport, const struct in6_addr *daddr, const u16 hnum, const int dif) @@ -83,8 +83,8 @@ static inline struct sock *__inet6_lookup(struct inet_hashinfo *hashinfo, } extern struct sock *inet6_lookup(struct inet_hashinfo *hashinfo, - const struct in6_addr *saddr, const u16 sport, - const struct in6_addr *daddr, const u16 dport, + const struct in6_addr *saddr, const __be16 sport, + const struct in6_addr *daddr, const __be16 dport, const int dif); #endif /* defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE) */ #endif /* _INET6_HASHTABLES_H */ diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c index 8accd1fbeed..b7e5bae0e34 100644 --- a/net/ipv6/inet6_hashtables.c +++ b/net/ipv6/inet6_hashtables.c @@ -57,7 +57,7 @@ EXPORT_SYMBOL(__inet6_hash); */ struct sock *__inet6_lookup_established(struct inet_hashinfo *hashinfo, const struct in6_addr *saddr, - const u16 sport, + const __be16 sport, const struct in6_addr *daddr, const u16 hnum, const int dif) @@ -146,8 +146,8 @@ struct sock *inet6_lookup_listener(struct inet_hashinfo *hashinfo, EXPORT_SYMBOL_GPL(inet6_lookup_listener); struct sock *inet6_lookup(struct inet_hashinfo *hashinfo, - const struct in6_addr *saddr, const u16 sport, - const struct in6_addr *daddr, const u16 dport, + const struct in6_addr *saddr, const __be16 sport, + const struct in6_addr *daddr, const __be16 dport, const int dif) { struct sock *sk; -- cgit v1.2.3-70-g09d2 From 8c689a6eae2d83970e4f34753d513e96fb97a025 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:20:21 -0800 Subject: [XFRM]: misc annotations Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/xfrm.h | 6 +++--- net/ipv4/xfrm4_policy.c | 6 +++--- net/ipv6/xfrm6_policy.c | 2 +- net/ipv6/xfrm6_tunnel.c | 6 +++--- 4 files changed, 10 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 737fdb2ee8a..a63e277bacf 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -951,9 +951,9 @@ extern int xfrm6_input_addr(struct sk_buff *skb, xfrm_address_t *daddr, xfrm_address_t *saddr, u8 proto); extern int xfrm6_tunnel_register(struct xfrm6_tunnel *handler); extern int xfrm6_tunnel_deregister(struct xfrm6_tunnel *handler); -extern u32 xfrm6_tunnel_alloc_spi(xfrm_address_t *saddr); +extern __be32 xfrm6_tunnel_alloc_spi(xfrm_address_t *saddr); extern void xfrm6_tunnel_free_spi(xfrm_address_t *saddr); -extern u32 xfrm6_tunnel_spi_lookup(xfrm_address_t *saddr); +extern __be32 xfrm6_tunnel_spi_lookup(xfrm_address_t *saddr); extern int xfrm6_output(struct sk_buff *skb); extern int xfrm6_find_1stfragopt(struct xfrm_state *x, struct sk_buff *skb, u8 **prevhdr); @@ -1033,7 +1033,7 @@ static inline int xfrm_addr_cmp(xfrm_address_t *a, xfrm_address_t *b, switch (family) { default: case AF_INET: - return a->a4 - b->a4; + return (__force __u32)a->a4 - (__force __u32)b->a4; case AF_INET6: return ipv6_addr_cmp((struct in6_addr *)a, (struct in6_addr *)b); diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index 1bed0cdf53e..da766234607 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -72,8 +72,8 @@ __xfrm4_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int struct dst_entry *dst, *dst_prev; struct rtable *rt0 = (struct rtable*)(*dst_p); struct rtable *rt = rt0; - u32 remote = fl->fl4_dst; - u32 local = fl->fl4_src; + __be32 remote = fl->fl4_dst; + __be32 local = fl->fl4_src; struct flowi fl_tunnel = { .nl_u = { .ip4_u = { @@ -203,7 +203,7 @@ _decode_session4(struct sk_buff *skb, struct flowi *fl) case IPPROTO_SCTP: case IPPROTO_DCCP: if (pskb_may_pull(skb, xprth + 4 - skb->data)) { - u16 *ports = (u16 *)xprth; + __be16 *ports = (__be16 *)xprth; fl->fl_ip_sport = ports[0]; fl->fl_ip_dport = ports[1]; diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index d400f8fae12..2fba1f0739a 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -278,7 +278,7 @@ _decode_session6(struct sk_buff *skb, struct flowi *fl) case IPPROTO_SCTP: case IPPROTO_DCCP: if (pskb_may_pull(skb, skb->nh.raw + offset + 4 - skb->data)) { - u16 *ports = (u16 *)exthdr; + __be16 *ports = (__be16 *)exthdr; fl->fl_ip_sport = ports[0]; fl->fl_ip_dport = ports[1]; diff --git a/net/ipv6/xfrm6_tunnel.c b/net/ipv6/xfrm6_tunnel.c index 7931e4f898d..f26d81b6342 100644 --- a/net/ipv6/xfrm6_tunnel.c +++ b/net/ipv6/xfrm6_tunnel.c @@ -62,7 +62,7 @@ static unsigned inline xfrm6_tunnel_spi_hash_byaddr(xfrm_address_t *addr) { unsigned h; - h = addr->a6[0] ^ addr->a6[1] ^ addr->a6[2] ^ addr->a6[3]; + h = (__force u32)(addr->a6[0] ^ addr->a6[1] ^ addr->a6[2] ^ addr->a6[3]); h ^= h >> 16; h ^= h >> 8; h &= XFRM6_TUNNEL_SPI_BYADDR_HSIZE - 1; @@ -126,7 +126,7 @@ static struct xfrm6_tunnel_spi *__xfrm6_tunnel_spi_lookup(xfrm_address_t *saddr) return NULL; } -u32 xfrm6_tunnel_spi_lookup(xfrm_address_t *saddr) +__be32 xfrm6_tunnel_spi_lookup(xfrm_address_t *saddr) { struct xfrm6_tunnel_spi *x6spi; u32 spi; @@ -196,7 +196,7 @@ out: return spi; } -u32 xfrm6_tunnel_alloc_spi(xfrm_address_t *saddr) +__be32 xfrm6_tunnel_alloc_spi(xfrm_address_t *saddr) { struct xfrm6_tunnel_spi *x6spi; u32 spi; -- cgit v1.2.3-70-g09d2 From 04ce69093f91547d3a7c4fc815d2868195591340 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:21:01 -0800 Subject: [IPV6]: 'info' argument of ipv6 ->err_handler() is net-endian Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/protocol.h | 2 +- include/net/rawv6.h | 2 +- include/net/xfrm.h | 3 +-- net/ipv6/ah6.c | 2 +- net/ipv6/esp6.c | 2 +- net/ipv6/icmp.c | 2 +- net/ipv6/ip6_tunnel.c | 2 +- net/ipv6/ipcomp6.c | 2 +- net/ipv6/raw.c | 2 +- net/ipv6/tcp_ipv6.c | 2 +- net/ipv6/tunnel6.c | 2 +- net/ipv6/udp.c | 2 +- net/ipv6/xfrm6_tunnel.c | 2 +- net/sctp/ipv6.c | 2 +- 14 files changed, 14 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/include/net/protocol.h b/include/net/protocol.h index c643bce64e5..28c4cac3f8a 100644 --- a/include/net/protocol.h +++ b/include/net/protocol.h @@ -50,7 +50,7 @@ struct inet6_protocol void (*err_handler)(struct sk_buff *skb, struct inet6_skb_parm *opt, int type, int code, int offset, - __u32 info); + __be32 info); int (*gso_send_check)(struct sk_buff *skb); struct sk_buff *(*gso_segment)(struct sk_buff *skb, diff --git a/include/net/rawv6.h b/include/net/rawv6.h index 14476a71725..af8960878ef 100644 --- a/include/net/rawv6.h +++ b/include/net/rawv6.h @@ -21,7 +21,7 @@ extern void rawv6_err(struct sock *sk, struct sk_buff *skb, struct inet6_skb_parm *opt, int type, int code, - int offset, u32 info); + int offset, __be32 info); #endif diff --git a/include/net/xfrm.h b/include/net/xfrm.h index a63e277bacf..9d30c08b9ef 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -887,8 +887,7 @@ struct xfrm_tunnel { struct xfrm6_tunnel { int (*handler)(struct sk_buff *skb); int (*err_handler)(struct sk_buff *skb, struct inet6_skb_parm *opt, - int type, int code, int offset, __u32 info); - + int type, int code, int offset, __be32 info); struct xfrm6_tunnel *next; int priority; }; diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c index b0d83e8e425..2d9e495cf69 100644 --- a/net/ipv6/ah6.c +++ b/net/ipv6/ah6.c @@ -397,7 +397,7 @@ out: } static void ah6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, - int type, int code, int offset, __u32 info) + int type, int code, int offset, __be32 info) { struct ipv6hdr *iph = (struct ipv6hdr*)skb->data; struct ip_auth_hdr *ah = (struct ip_auth_hdr*)(skb->data+offset); diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index e78680a9985..25dcf69cd80 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -256,7 +256,7 @@ static u32 esp6_get_max_size(struct xfrm_state *x, int mtu) } static void esp6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, - int type, int code, int offset, __u32 info) + int type, int code, int offset, __be32 info) { struct ipv6hdr *iph = (struct ipv6hdr*)skb->data; struct ipv6_esp_hdr *esph = (struct ipv6_esp_hdr*)(skb->data+offset); diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 4ec876066b3..81bd45b26c9 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -555,7 +555,7 @@ out: icmpv6_xmit_unlock(); } -static void icmpv6_notify(struct sk_buff *skb, int type, int code, u32 info) +static void icmpv6_notify(struct sk_buff *skb, int type, int code, __be32 info) { struct in6_addr *saddr, *daddr; struct inet6_protocol *ipprot; diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index b9f40290d12..6a3026dd865 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -391,7 +391,7 @@ parse_tlv_tnl_enc_lim(struct sk_buff *skb, __u8 * raw) static int ip6ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, - int type, int code, int offset, __u32 info) + int type, int code, int offset, __be32 info) { struct ipv6hdr *ipv6h = (struct ipv6hdr *) skb->data; struct ip6_tnl *t; diff --git a/net/ipv6/ipcomp6.c b/net/ipv6/ipcomp6.c index 71f59f18ede..511730b67e9 100644 --- a/net/ipv6/ipcomp6.c +++ b/net/ipv6/ipcomp6.c @@ -176,7 +176,7 @@ out_ok: } static void ipcomp6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, - int type, int code, int offset, __u32 info) + int type, int code, int offset, __be32 info) { __be32 spi; struct ipv6hdr *iph = (struct ipv6hdr*)skb->data; diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index d6dedc4aec7..6bc66552442 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -290,7 +290,7 @@ out: void rawv6_err(struct sock *sk, struct sk_buff *skb, struct inet6_skb_parm *opt, - int type, int code, int offset, u32 info) + int type, int code, int offset, __be32 info) { struct inet_sock *inet = inet_sk(sk); struct ipv6_pinfo *np = inet6_sk(sk); diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 4c2a7c0cafe..eb6d145ecfd 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -310,7 +310,7 @@ failure: } static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, - int type, int code, int offset, __u32 info) + int type, int code, int offset, __be32 info) { struct ipv6hdr *hdr = (struct ipv6hdr*)skb->data; const struct tcphdr *th = (struct tcphdr *)(skb->data+offset); diff --git a/net/ipv6/tunnel6.c b/net/ipv6/tunnel6.c index 0ef9a35798d..918d07dd121 100644 --- a/net/ipv6/tunnel6.c +++ b/net/ipv6/tunnel6.c @@ -104,7 +104,7 @@ drop: } static void tunnel6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, - int type, int code, int offset, __u32 info) + int type, int code, int offset, __be32 info) { struct xfrm6_tunnel *handler; diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index c83f23e51c4..886300d13a5 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -238,7 +238,7 @@ csum_copy_err: } static void udpv6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, - int type, int code, int offset, __u32 info) + int type, int code, int offset, __be32 info) { struct ipv6_pinfo *np; struct ipv6hdr *hdr = (struct ipv6hdr*)skb->data; diff --git a/net/ipv6/xfrm6_tunnel.c b/net/ipv6/xfrm6_tunnel.c index f26d81b6342..01a5c52a2be 100644 --- a/net/ipv6/xfrm6_tunnel.c +++ b/net/ipv6/xfrm6_tunnel.c @@ -265,7 +265,7 @@ static int xfrm6_tunnel_rcv(struct sk_buff *skb) } static int xfrm6_tunnel_err(struct sk_buff *skb, struct inet6_skb_parm *opt, - int type, int code, int offset, __u32 info) + int type, int code, int offset, __be32 info) { /* xfrm6_tunnel native err handling */ switch (type) { diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 78071c6e6cf..3c316c688ad 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -84,7 +84,7 @@ static struct notifier_block sctp_inet6addr_notifier = { /* ICMP error handler. */ SCTP_STATIC void sctp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, - int type, int code, int offset, __u32 info) + int type, int code, int offset, __be32 info) { struct inet6_dev *idev; struct ipv6hdr *iph = (struct ipv6hdr *)skb->data; -- cgit v1.2.3-70-g09d2 From a27ee7a4dd30feda1954950b5840455a51ae1507 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:21:21 -0800 Subject: [IPV6]: annotate icmpv6 headers Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/icmpv6.h | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h index c771a7db987..dc79396aac2 100644 --- a/include/linux/icmpv6.h +++ b/include/linux/icmpv6.h @@ -7,17 +7,17 @@ struct icmp6hdr { __u8 icmp6_type; __u8 icmp6_code; - __u16 icmp6_cksum; + __be16 icmp6_cksum; union { - __u32 un_data32[1]; - __u16 un_data16[2]; + __be32 un_data32[1]; + __be16 un_data16[2]; __u8 un_data8[4]; struct icmpv6_echo { - __u16 identifier; - __u16 sequence; + __be16 identifier; + __be16 sequence; } u_echo; struct icmpv6_nd_advt { @@ -53,7 +53,7 @@ struct icmp6hdr { #else #error "Please fix " #endif - __u16 rt_lifetime; + __be16 rt_lifetime; } u_nd_ra; } icmp6_dataun; -- cgit v1.2.3-70-g09d2 From 44473a6b279794f058955aec5c36975e74deb541 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:21:46 -0800 Subject: [IPV6]: annotate struct frag_hdr Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/ipv6.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 8223c4410b4..4953fac1d07 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -95,10 +95,10 @@ */ struct frag_hdr { - unsigned char nexthdr; - unsigned char reserved; - unsigned short frag_off; - __u32 identification; + __u8 nexthdr; + __u8 reserved; + __be16 frag_off; + __be32 identification; }; #define IP6_MF 0x0001 -- cgit v1.2.3-70-g09d2 From d5a0a1e3109339090769e40fdaa62482fcf2a717 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:23:14 -0800 Subject: [IPV4]: encapsulation annotations Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/if_tunnel.h | 8 ++++---- include/linux/xfrm.h | 4 ++-- net/ipv4/ah4.c | 4 ++-- net/ipv4/esp4.c | 4 ++-- net/ipv4/ip_gre.c | 52 +++++++++++++++++++++++------------------------ net/ipv4/ipip.c | 16 +++++++-------- 6 files changed, 44 insertions(+), 44 deletions(-) (limited to 'include') diff --git a/include/linux/if_tunnel.h b/include/linux/if_tunnel.h index bef9f8fd93b..8de079ba110 100644 --- a/include/linux/if_tunnel.h +++ b/include/linux/if_tunnel.h @@ -19,10 +19,10 @@ struct ip_tunnel_parm { char name[IFNAMSIZ]; int link; - __u16 i_flags; - __u16 o_flags; - __u32 i_key; - __u32 o_key; + __be16 i_flags; + __be16 o_flags; + __be32 i_key; + __be32 o_key; struct iphdr iph; }; diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h index 8ae7f744917..7907c42bd4e 100644 --- a/include/linux/xfrm.h +++ b/include/linux/xfrm.h @@ -211,8 +211,8 @@ struct xfrm_user_tmpl { struct xfrm_encap_tmpl { __u16 encap_type; - __u16 encap_sport; - __u16 encap_dport; + __be16 encap_sport; + __be16 encap_dport; xfrm_address_t encap_oa; }; diff --git a/net/ipv4/ah4.c b/net/ipv4/ah4.c index 99542977e47..67a5509e26f 100644 --- a/net/ipv4/ah4.c +++ b/net/ipv4/ah4.c @@ -14,7 +14,7 @@ * into IP header for icv calculation. Options are already checked * for validity, so paranoia is not required. */ -static int ip_clear_mutable_options(struct iphdr *iph, u32 *daddr) +static int ip_clear_mutable_options(struct iphdr *iph, __be32 *daddr) { unsigned char * optptr = (unsigned char*)(iph+1); int l = iph->ihl*4 - sizeof(struct iphdr); @@ -162,7 +162,7 @@ static int ah_input(struct xfrm_state *x, struct sk_buff *skb) iph->frag_off = 0; iph->check = 0; if (ihl > sizeof(*iph)) { - u32 dummy; + __be32 dummy; if (ip_clear_mutable_options(iph, &dummy)) goto out; } diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index b5c205b5766..f2c6776ea0e 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -67,7 +67,7 @@ static int esp_output(struct xfrm_state *x, struct sk_buff *skb) if (x->encap) { struct xfrm_encap_tmpl *encap = x->encap; struct udphdr *uh; - u32 *udpdata32; + __be32 *udpdata32; uh = (struct udphdr *)esph; uh->source = encap->encap_sport; @@ -81,7 +81,7 @@ static int esp_output(struct xfrm_state *x, struct sk_buff *skb) esph = (struct ip_esp_hdr *)(uh + 1); break; case UDP_ENCAP_ESPINUDP_NON_IKE: - udpdata32 = (u32 *)(uh + 1); + udpdata32 = (__be32 *)(uh + 1); udpdata32[0] = udpdata32[1] = 0; esph = (struct ip_esp_hdr *)(udpdata32 + 2); break; diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index d5b5dec075b..25221146deb 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -144,7 +144,7 @@ static struct net_device *ipgre_fb_tunnel_dev; */ #define HASH_SIZE 16 -#define HASH(addr) ((addr^(addr>>4))&0xF) +#define HASH(addr) (((__force u32)addr^((__force u32)addr>>4))&0xF) static struct ip_tunnel *tunnels[4][HASH_SIZE]; @@ -157,7 +157,7 @@ static DEFINE_RWLOCK(ipgre_lock); /* Given src, dst and key, find appropriate for input tunnel. */ -static struct ip_tunnel * ipgre_tunnel_lookup(u32 remote, u32 local, u32 key) +static struct ip_tunnel * ipgre_tunnel_lookup(__be32 remote, __be32 local, __be32 key) { unsigned h0 = HASH(remote); unsigned h1 = HASH(key); @@ -194,9 +194,9 @@ static struct ip_tunnel * ipgre_tunnel_lookup(u32 remote, u32 local, u32 key) static struct ip_tunnel **ipgre_bucket(struct ip_tunnel *t) { - u32 remote = t->parms.iph.daddr; - u32 local = t->parms.iph.saddr; - u32 key = t->parms.i_key; + __be32 remote = t->parms.iph.daddr; + __be32 local = t->parms.iph.saddr; + __be32 key = t->parms.i_key; unsigned h = HASH(key); int prio = 0; @@ -236,9 +236,9 @@ static void ipgre_tunnel_unlink(struct ip_tunnel *t) static struct ip_tunnel * ipgre_tunnel_locate(struct ip_tunnel_parm *parms, int create) { - u32 remote = parms->iph.daddr; - u32 local = parms->iph.saddr; - u32 key = parms->i_key; + __be32 remote = parms->iph.daddr; + __be32 local = parms->iph.saddr; + __be32 key = parms->i_key; struct ip_tunnel *t, **tp, *nt; struct net_device *dev; unsigned h = HASH(key); @@ -319,12 +319,12 @@ static void ipgre_err(struct sk_buff *skb, u32 info) */ struct iphdr *iph = (struct iphdr*)skb->data; - u16 *p = (u16*)(skb->data+(iph->ihl<<2)); + __be16 *p = (__be16*)(skb->data+(iph->ihl<<2)); int grehlen = (iph->ihl<<2) + 4; int type = skb->h.icmph->type; int code = skb->h.icmph->code; struct ip_tunnel *t; - u16 flags; + __be16 flags; flags = p[0]; if (flags&(GRE_CSUM|GRE_KEY|GRE_SEQ|GRE_ROUTING|GRE_VERSION)) { @@ -370,7 +370,7 @@ static void ipgre_err(struct sk_buff *skb, u32 info) } read_lock(&ipgre_lock); - t = ipgre_tunnel_lookup(iph->daddr, iph->saddr, (flags&GRE_KEY) ? *(((u32*)p) + (grehlen>>2) - 1) : 0); + t = ipgre_tunnel_lookup(iph->daddr, iph->saddr, (flags&GRE_KEY) ? *(((__be32*)p) + (grehlen>>2) - 1) : 0); if (t == NULL || t->parms.iph.daddr == 0 || MULTICAST(t->parms.iph.daddr)) goto out; @@ -388,14 +388,14 @@ out: #else struct iphdr *iph = (struct iphdr*)dp; struct iphdr *eiph; - u16 *p = (u16*)(dp+(iph->ihl<<2)); + __be16 *p = (__be16*)(dp+(iph->ihl<<2)); int type = skb->h.icmph->type; int code = skb->h.icmph->code; int rel_type = 0; int rel_code = 0; __be32 rel_info = 0; __u32 n = 0; - u16 flags; + __be16 flags; int grehlen = (iph->ihl<<2) + 4; struct sk_buff *skb2; struct flowi fl; @@ -556,9 +556,9 @@ static int ipgre_rcv(struct sk_buff *skb) { struct iphdr *iph; u8 *h; - u16 flags; + __be16 flags; u16 csum = 0; - u32 key = 0; + __be32 key = 0; u32 seqno = 0; struct ip_tunnel *tunnel; int offset = 4; @@ -568,7 +568,7 @@ static int ipgre_rcv(struct sk_buff *skb) iph = skb->nh.iph; h = skb->data; - flags = *(u16*)h; + flags = *(__be16*)h; if (flags&(GRE_CSUM|GRE_KEY|GRE_ROUTING|GRE_SEQ|GRE_VERSION)) { /* - Version must be 0. @@ -592,11 +592,11 @@ static int ipgre_rcv(struct sk_buff *skb) offset += 4; } if (flags&GRE_KEY) { - key = *(u32*)(h + offset); + key = *(__be32*)(h + offset); offset += 4; } if (flags&GRE_SEQ) { - seqno = ntohl(*(u32*)(h + offset)); + seqno = ntohl(*(__be32*)(h + offset)); offset += 4; } } @@ -605,7 +605,7 @@ static int ipgre_rcv(struct sk_buff *skb) if ((tunnel = ipgre_tunnel_lookup(iph->saddr, iph->daddr, key)) != NULL) { secpath_reset(skb); - skb->protocol = *(u16*)(h + 2); + skb->protocol = *(__be16*)(h + 2); /* WCCP version 1 and 2 protocol decoding. * - Change protocol to IP * - When dealing with WCCPv2, Skip extra 4 bytes in GRE header @@ -673,13 +673,13 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) struct iphdr *old_iph = skb->nh.iph; struct iphdr *tiph; u8 tos; - u16 df; + __be16 df; struct rtable *rt; /* Route to the other host */ struct net_device *tdev; /* Device to other host */ struct iphdr *iph; /* Our new IP header */ int max_headroom; /* The extra header space needed */ int gre_hlen; - u32 dst; + __be32 dst; int mtu; if (tunnel->recursion++) { @@ -860,11 +860,11 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) iph->ttl = dst_metric(&rt->u.dst, RTAX_HOPLIMIT); } - ((u16*)(iph+1))[0] = tunnel->parms.o_flags; - ((u16*)(iph+1))[1] = skb->protocol; + ((__be16*)(iph+1))[0] = tunnel->parms.o_flags; + ((__be16*)(iph+1))[1] = skb->protocol; if (tunnel->parms.o_flags&(GRE_KEY|GRE_CSUM|GRE_SEQ)) { - u32 *ptr = (u32*)(((u8*)iph) + tunnel->hlen - 4); + __be32 *ptr = (__be32*)(((u8*)iph) + tunnel->hlen - 4); if (tunnel->parms.o_flags&GRE_SEQ) { ++tunnel->o_seqno; @@ -877,7 +877,7 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) } if (tunnel->parms.o_flags&GRE_CSUM) { *ptr = 0; - *(__u16*)ptr = ip_compute_csum((void*)(iph+1), skb->len - sizeof(struct iphdr)); + *(__be16*)ptr = ip_compute_csum((void*)(iph+1), skb->len - sizeof(struct iphdr)); } } @@ -1068,7 +1068,7 @@ static int ipgre_header(struct sk_buff *skb, struct net_device *dev, unsigned sh { struct ip_tunnel *t = netdev_priv(dev); struct iphdr *iph = (struct iphdr *)skb_push(skb, t->hlen); - u16 *p = (u16*)(iph+1); + __be16 *p = (__be16*)(iph+1); memcpy(iph, &t->parms.iph, sizeof(struct iphdr)); p[0] = t->parms.o_flags; diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index 0c455652922..9d719d664e5 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -118,7 +118,7 @@ #include #define HASH_SIZE 16 -#define HASH(addr) ((addr^(addr>>4))&0xF) +#define HASH(addr) (((__force u32)addr^((__force u32)addr>>4))&0xF) static int ipip_fb_tunnel_init(struct net_device *dev); static int ipip_tunnel_init(struct net_device *dev); @@ -134,7 +134,7 @@ static struct ip_tunnel **tunnels[4] = { tunnels_wc, tunnels_l, tunnels_r, tunne static DEFINE_RWLOCK(ipip_lock); -static struct ip_tunnel * ipip_tunnel_lookup(u32 remote, u32 local) +static struct ip_tunnel * ipip_tunnel_lookup(__be32 remote, __be32 local) { unsigned h0 = HASH(remote); unsigned h1 = HASH(local); @@ -160,8 +160,8 @@ static struct ip_tunnel * ipip_tunnel_lookup(u32 remote, u32 local) static struct ip_tunnel **ipip_bucket(struct ip_tunnel *t) { - u32 remote = t->parms.iph.daddr; - u32 local = t->parms.iph.saddr; + __be32 remote = t->parms.iph.daddr; + __be32 local = t->parms.iph.saddr; unsigned h = 0; int prio = 0; @@ -203,8 +203,8 @@ static void ipip_tunnel_link(struct ip_tunnel *t) static struct ip_tunnel * ipip_tunnel_locate(struct ip_tunnel_parm *parms, int create) { - u32 remote = parms->iph.daddr; - u32 local = parms->iph.saddr; + __be32 remote = parms->iph.daddr; + __be32 local = parms->iph.saddr; struct ip_tunnel *t, **tp, *nt; struct net_device *dev; unsigned h = 0; @@ -519,13 +519,13 @@ static int ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) struct net_device_stats *stats = &tunnel->stat; struct iphdr *tiph = &tunnel->parms.iph; u8 tos = tunnel->parms.iph.tos; - u16 df = tiph->frag_off; + __be16 df = tiph->frag_off; struct rtable *rt; /* Route to the other host */ struct net_device *tdev; /* Device to other host */ struct iphdr *old_iph = skb->nh.iph; struct iphdr *iph; /* Our new IP header */ int max_headroom; /* The extra header space needed */ - u32 dst = tiph->daddr; + __be32 dst = tiph->daddr; int mtu; if (tunnel->recursion++) { -- cgit v1.2.3-70-g09d2 From d29ef86b0a497dd2c4f9601644a15392f7e21f73 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:23:42 -0800 Subject: [AF_KEY]: annotate Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/pfkeyv2.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/pfkeyv2.h b/include/linux/pfkeyv2.h index d5dd471da22..0f0b880c428 100644 --- a/include/linux/pfkeyv2.h +++ b/include/linux/pfkeyv2.h @@ -32,7 +32,7 @@ struct sadb_ext { struct sadb_sa { uint16_t sadb_sa_len; uint16_t sadb_sa_exttype; - uint32_t sadb_sa_spi; + __be32 sadb_sa_spi; uint8_t sadb_sa_replay; uint8_t sadb_sa_state; uint8_t sadb_sa_auth; @@ -211,7 +211,7 @@ struct sadb_x_nat_t_type { struct sadb_x_nat_t_port { uint16_t sadb_x_nat_t_port_len; uint16_t sadb_x_nat_t_port_exttype; - uint16_t sadb_x_nat_t_port_port; + __be16 sadb_x_nat_t_port_port; uint16_t sadb_x_nat_t_port_reserved; } __attribute__((packed)); /* sizeof(struct sadb_x_nat_t_port) == 8 */ -- cgit v1.2.3-70-g09d2 From 5d36b1803d875cf101fdb972ff9c56663e508e39 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:24:06 -0800 Subject: [XFRM]: annotate ->new_mapping() Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/xfrm.h | 4 ++-- net/key/af_key.c | 2 +- net/xfrm/xfrm_state.c | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 9d30c08b9ef..dbcd2922ff3 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -379,7 +379,7 @@ struct xfrm_mgr int (*notify)(struct xfrm_state *x, struct km_event *c); int (*acquire)(struct xfrm_state *x, struct xfrm_tmpl *, struct xfrm_policy *xp, int dir); struct xfrm_policy *(*compile_policy)(struct sock *sk, int opt, u8 *data, int len, int *dir); - int (*new_mapping)(struct xfrm_state *x, xfrm_address_t *ipaddr, u16 sport); + int (*new_mapping)(struct xfrm_state *x, xfrm_address_t *ipaddr, __be16 sport); int (*notify_policy)(struct xfrm_policy *x, int dir, struct km_event *c); int (*report)(u8 proto, struct xfrm_selector *sel, xfrm_address_t *addr); }; @@ -999,7 +999,7 @@ extern int xfrm_bundle_ok(struct xfrm_policy *pol, struct xfrm_dst *xdst, extern void xfrm_init_pmtu(struct dst_entry *dst); extern wait_queue_head_t km_waitq; -extern int km_new_mapping(struct xfrm_state *x, xfrm_address_t *ipaddr, u16 sport); +extern int km_new_mapping(struct xfrm_state *x, xfrm_address_t *ipaddr, __be16 sport); extern void km_policy_expired(struct xfrm_policy *pol, int dir, int hard, u32 pid); extern int km_report(u8 proto, struct xfrm_selector *sel, xfrm_address_t *addr); diff --git a/net/key/af_key.c b/net/key/af_key.c index 20ff7cca1d0..4e183099948 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -2938,7 +2938,7 @@ out: return NULL; } -static int pfkey_send_new_mapping(struct xfrm_state *x, xfrm_address_t *ipaddr, u16 sport) +static int pfkey_send_new_mapping(struct xfrm_state *x, xfrm_address_t *ipaddr, __be16 sport) { struct sk_buff *skb; struct sadb_msg *hdr; diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 899de9ed22a..864962bbda9 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -1304,7 +1304,7 @@ int km_query(struct xfrm_state *x, struct xfrm_tmpl *t, struct xfrm_policy *pol) } EXPORT_SYMBOL(km_query); -int km_new_mapping(struct xfrm_state *x, xfrm_address_t *ipaddr, u16 sport) +int km_new_mapping(struct xfrm_state *x, xfrm_address_t *ipaddr, __be16 sport) { int err = -EINVAL; struct xfrm_mgr *km; -- cgit v1.2.3-70-g09d2 From 8a9ae2110baf39b2dc53c2891e0705895a8829a9 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:24:26 -0800 Subject: [NET]: annotate dsfield.h Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/dsfield.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/net/dsfield.h b/include/net/dsfield.h index a79c9e075f7..dae8453385b 100644 --- a/include/net/dsfield.h +++ b/include/net/dsfield.h @@ -20,7 +20,7 @@ static inline __u8 ipv4_get_dsfield(struct iphdr *iph) static inline __u8 ipv6_get_dsfield(struct ipv6hdr *ipv6h) { - return ntohs(*(__u16 *) ipv6h) >> 4; + return ntohs(*(__be16 *) ipv6h) >> 4; } @@ -45,9 +45,9 @@ static inline void ipv6_change_dsfield(struct ipv6hdr *ipv6h,__u8 mask, { __u16 tmp; - tmp = ntohs(*(__u16 *) ipv6h); + tmp = ntohs(*(__be16 *) ipv6h); tmp = (tmp & ((mask << 4) | 0xf00f)) | (value << 4); - *(__u16 *) ipv6h = htons(tmp); + *(__be16 *) ipv6h = htons(tmp); } -- cgit v1.2.3-70-g09d2 From 92d9ece7af9c84bfbd1ff640926fac5b573a09f7 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:24:47 -0800 Subject: [INET]: annotate inet_ecn.h Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/inet_ecn.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/inet_ecn.h b/include/net/inet_ecn.h index 7849844a491..cadc0eab54f 100644 --- a/include/net/inet_ecn.h +++ b/include/net/inet_ecn.h @@ -95,13 +95,13 @@ static inline int IP6_ECN_set_ce(struct ipv6hdr *iph) { if (INET_ECN_is_not_ect(ipv6_get_dsfield(iph))) return 0; - *(u32*)iph |= htonl(INET_ECN_CE << 20); + *(__be32*)iph |= htonl(INET_ECN_CE << 20); return 1; } static inline void IP6_ECN_clear(struct ipv6hdr *iph) { - *(u32*)iph &= ~htonl(INET_ECN_MASK << 20); + *(__be32*)iph &= ~htonl(INET_ECN_MASK << 20); } static inline void ipv6_copy_dscp(struct ipv6hdr *outer, struct ipv6hdr *inner) -- cgit v1.2.3-70-g09d2 From 90bcaf7b4a33bb9b100cc06869f0c033a870d4a0 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:25:17 -0800 Subject: [IPV6]: flowlabels are net-endian Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/in6.h | 4 ++-- include/linux/ip6_tunnel.h | 2 +- include/linux/ipv6.h | 2 +- include/net/flow.h | 2 +- include/net/ipv6.h | 4 ++-- net/ipv6/af_inet6.c | 2 +- net/ipv6/datagram.c | 10 +++++----- net/ipv6/ip6_flowlabel.c | 8 ++++---- net/ipv6/ip6_output.c | 4 ++-- net/ipv6/ip6_tunnel.c | 6 +++--- net/ipv6/route.c | 2 +- 11 files changed, 23 insertions(+), 23 deletions(-) (limited to 'include') diff --git a/include/linux/in6.h b/include/linux/in6.h index f28621f638e..4e8350ae886 100644 --- a/include/linux/in6.h +++ b/include/linux/in6.h @@ -54,7 +54,7 @@ extern const struct in6_addr in6addr_loopback; struct sockaddr_in6 { unsigned short int sin6_family; /* AF_INET6 */ __be16 sin6_port; /* Transport layer port # */ - __u32 sin6_flowinfo; /* IPv6 flow information */ + __be32 sin6_flowinfo; /* IPv6 flow information */ struct in6_addr sin6_addr; /* IPv6 address */ __u32 sin6_scope_id; /* scope id (new in RFC2553) */ }; @@ -72,7 +72,7 @@ struct ipv6_mreq { struct in6_flowlabel_req { struct in6_addr flr_dst; - __u32 flr_label; + __be32 flr_label; __u8 flr_action; __u8 flr_share; __u16 flr_flags; diff --git a/include/linux/ip6_tunnel.h b/include/linux/ip6_tunnel.h index 5c23aeb104c..af3f4a70f3d 100644 --- a/include/linux/ip6_tunnel.h +++ b/include/linux/ip6_tunnel.h @@ -25,7 +25,7 @@ struct ip6_tnl_parm { __u8 proto; /* tunnel protocol */ __u8 encap_limit; /* encapsulation limit for tunnel */ __u8 hop_limit; /* hop limit for tunnel */ - __u32 flowinfo; /* traffic class and flowlabel for tunnel */ + __be32 flowinfo; /* traffic class and flowlabel for tunnel */ __u32 flags; /* tunnel flags */ struct in6_addr laddr; /* local tunnel end-point address */ struct in6_addr raddr; /* remote tunnel end-point address */ diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index 4f435c59de0..f8241130f5e 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -274,7 +274,7 @@ struct ipv6_pinfo { struct in6_addr *saddr_cache; #endif - __u32 flow_label; + __be32 flow_label; __u32 frag_size; __s16 hop_limit; __s16 mcast_hops; diff --git a/include/net/flow.h b/include/net/flow.h index 5cda27cd9de..270d4c1761a 100644 --- a/include/net/flow.h +++ b/include/net/flow.h @@ -27,7 +27,7 @@ struct flowi { struct in6_addr daddr; struct in6_addr saddr; __u32 fwmark; - __u32 flowlabel; + __be32 flowlabel; } ip6_u; struct { diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 4953fac1d07..0b8c9b990ac 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -191,7 +191,7 @@ struct ipv6_txoptions struct ip6_flowlabel { struct ip6_flowlabel *next; - u32 label; + __be32 label; struct in6_addr dst; struct ipv6_txoptions *opt; atomic_t users; @@ -211,7 +211,7 @@ struct ipv6_fl_socklist struct ip6_flowlabel *fl; }; -extern struct ip6_flowlabel *fl6_sock_lookup(struct sock *sk, u32 label); +extern struct ip6_flowlabel *fl6_sock_lookup(struct sock *sk, __be32 label); extern struct ipv6_txoptions *fl6_merge_options(struct ipv6_txoptions * opt_space, struct ip6_flowlabel * fl, struct ipv6_txoptions * fopt); diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 858cae29581..92bfccf62cb 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -678,7 +678,7 @@ int ipv6_opt_accepted(struct sock *sk, struct sk_buff *skb) if (np->rxopt.all) { if ((opt->hop && (np->rxopt.bits.hopopts || np->rxopt.bits.ohopopts)) || - ((IPV6_FLOWINFO_MASK & *(u32*)skb->nh.raw) && + ((IPV6_FLOWINFO_MASK & *(__be32*)skb->nh.raw) && np->rxopt.bits.rxflow) || (opt->srcrt && (np->rxopt.bits.srcrt || np->rxopt.bits.osrcrt)) || diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c index 7206747022f..a67434af691 100644 --- a/net/ipv6/datagram.c +++ b/net/ipv6/datagram.c @@ -318,7 +318,7 @@ int ipv6_recv_error(struct sock *sk, struct msghdr *msg, int len) ipv6_addr_copy(&sin->sin6_addr, (struct in6_addr *)(skb->nh.raw + serr->addr_offset)); if (np->sndflow) - sin->sin6_flowinfo = *(u32*)(skb->nh.raw + serr->addr_offset - 24) & IPV6_FLOWINFO_MASK; + sin->sin6_flowinfo = *(__be32*)(skb->nh.raw + serr->addr_offset - 24) & IPV6_FLOWINFO_MASK; if (ipv6_addr_type(&sin->sin6_addr) & IPV6_ADDR_LINKLOCAL) sin->sin6_scope_id = IP6CB(skb)->iif; } else { @@ -401,8 +401,8 @@ int datagram_recv_ctl(struct sock *sk, struct msghdr *msg, struct sk_buff *skb) put_cmsg(msg, SOL_IPV6, IPV6_TCLASS, sizeof(tclass), &tclass); } - if (np->rxopt.bits.rxflow && (*(u32*)skb->nh.raw & IPV6_FLOWINFO_MASK)) { - u32 flowinfo = *(u32*)skb->nh.raw & IPV6_FLOWINFO_MASK; + if (np->rxopt.bits.rxflow && (*(__be32*)skb->nh.raw & IPV6_FLOWINFO_MASK)) { + __be32 flowinfo = *(__be32*)skb->nh.raw & IPV6_FLOWINFO_MASK; put_cmsg(msg, SOL_IPV6, IPV6_FLOWINFO, sizeof(flowinfo), &flowinfo); } @@ -560,12 +560,12 @@ int datagram_send_ctl(struct msghdr *msg, struct flowi *fl, } if (fl->fl6_flowlabel&IPV6_FLOWINFO_MASK) { - if ((fl->fl6_flowlabel^*(u32 *)CMSG_DATA(cmsg))&~IPV6_FLOWINFO_MASK) { + if ((fl->fl6_flowlabel^*(__be32 *)CMSG_DATA(cmsg))&~IPV6_FLOWINFO_MASK) { err = -EINVAL; goto exit_f; } } - fl->fl6_flowlabel = IPV6_FLOWINFO_MASK & *(u32 *)CMSG_DATA(cmsg); + fl->fl6_flowlabel = IPV6_FLOWINFO_MASK & *(__be32 *)CMSG_DATA(cmsg); break; case IPV6_2292HOPOPTS: diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c index 6d4533b58dc..624fae251f4 100644 --- a/net/ipv6/ip6_flowlabel.c +++ b/net/ipv6/ip6_flowlabel.c @@ -61,7 +61,7 @@ static DEFINE_RWLOCK(ip6_fl_lock); static DEFINE_RWLOCK(ip6_sk_fl_lock); -static __inline__ struct ip6_flowlabel * __fl_lookup(u32 label) +static __inline__ struct ip6_flowlabel * __fl_lookup(__be32 label) { struct ip6_flowlabel *fl; @@ -72,7 +72,7 @@ static __inline__ struct ip6_flowlabel * __fl_lookup(u32 label) return NULL; } -static struct ip6_flowlabel * fl_lookup(u32 label) +static struct ip6_flowlabel * fl_lookup(__be32 label) { struct ip6_flowlabel *fl; @@ -153,7 +153,7 @@ static void ip6_fl_gc(unsigned long dummy) write_unlock(&ip6_fl_lock); } -static int fl_intern(struct ip6_flowlabel *fl, __u32 label) +static int fl_intern(struct ip6_flowlabel *fl, __be32 label) { fl->label = label & IPV6_FLOWLABEL_MASK; @@ -182,7 +182,7 @@ static int fl_intern(struct ip6_flowlabel *fl, __u32 label) /* Socket flowlabel lists */ -struct ip6_flowlabel * fl6_sock_lookup(struct sock *sk, u32 label) +struct ip6_flowlabel * fl6_sock_lookup(struct sock *sk, __be32 label) { struct ipv6_fl_socklist *sfl; struct ipv6_pinfo *np = inet6_sk(sk); diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 66716911962..9ff1be1bc2e 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -217,7 +217,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl, if (tclass < 0) tclass = 0; - *(u32 *)hdr = htonl(0x60000000 | (tclass << 20)) | fl->fl6_flowlabel; + *(__be32 *)hdr = htonl(0x60000000 | (tclass << 20)) | fl->fl6_flowlabel; hdr->payload_len = htons(seg_len); hdr->nexthdr = proto; @@ -1311,7 +1311,7 @@ int ip6_push_pending_frames(struct sock *sk) skb->nh.ipv6h = hdr = (struct ipv6hdr*) skb_push(skb, sizeof(struct ipv6hdr)); - *(u32*)hdr = fl->fl6_flowlabel | + *(__be32*)hdr = fl->fl6_flowlabel | htonl(0x60000000 | ((int)np->cork.tclass << 20)); if (skb->len <= sizeof(struct ipv6hdr) + IPV6_MAXPLEN) diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 6a3026dd865..4919f9294e2 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -665,9 +665,9 @@ ip6ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) dsfield = ipv6_get_dsfield(ipv6h); if ((t->parms.flags & IP6_TNL_F_USE_ORIG_TCLASS)) - fl.fl6_flowlabel |= (*(__u32 *) ipv6h & IPV6_TCLASS_MASK); + fl.fl6_flowlabel |= (*(__be32 *) ipv6h & IPV6_TCLASS_MASK); if ((t->parms.flags & IP6_TNL_F_USE_ORIG_FLOWLABEL)) - fl.fl6_flowlabel |= (*(__u32 *) ipv6h & IPV6_FLOWLABEL_MASK); + fl.fl6_flowlabel |= (*(__be32 *) ipv6h & IPV6_FLOWLABEL_MASK); if (encap_limit >= 0 && (opt = create_tel(encap_limit)) == NULL) goto tx_err; @@ -735,7 +735,7 @@ ip6ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) skb->nh.raw = skb_push(skb, sizeof(struct ipv6hdr)); ipv6h = skb->nh.ipv6h; - *(u32*)ipv6h = fl.fl6_flowlabel | htonl(0x60000000); + *(__be32*)ipv6h = fl.fl6_flowlabel | htonl(0x60000000); dsfield = INET_ECN_encapsulate(0, dsfield); ipv6_change_dsfield(ipv6h, ~INET_ECN_MASK, dsfield); ipv6h->payload_len = htons(skb->len - sizeof(struct ipv6hdr)); diff --git a/net/ipv6/route.c b/net/ipv6/route.c index b39ae99122d..e9c1fc5f21b 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -714,7 +714,7 @@ void ip6_route_input(struct sk_buff *skb) #ifdef CONFIG_IPV6_ROUTE_FWMARK .fwmark = skb->nfmark, #endif - .flowlabel = (* (u32 *) iph)&IPV6_FLOWINFO_MASK, + .flowlabel = (* (__be32 *) iph)&IPV6_FLOWINFO_MASK, }, }, .proto = iph->nexthdr, -- cgit v1.2.3-70-g09d2 From fede70b9862635ab1bed84ab3d765e9069616b02 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:25:41 -0800 Subject: [IPV6]: annotate inet6_csk_search_req() Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/inet6_connection_sock.h | 2 +- net/ipv6/inet6_connection_sock.c | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/net/inet6_connection_sock.h b/include/net/inet6_connection_sock.h index b33b438bffc..f13ddc2543b 100644 --- a/include/net/inet6_connection_sock.h +++ b/include/net/inet6_connection_sock.h @@ -27,7 +27,7 @@ extern int inet6_csk_bind_conflict(const struct sock *sk, extern struct request_sock *inet6_csk_search_req(const struct sock *sk, struct request_sock ***prevp, - const __u16 rport, + const __be16 rport, const struct in6_addr *raddr, const struct in6_addr *laddr, const int iif); diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c index 827f41d1478..e48fb4c53c4 100644 --- a/net/ipv6/inet6_connection_sock.c +++ b/net/ipv6/inet6_connection_sock.c @@ -52,7 +52,7 @@ EXPORT_SYMBOL_GPL(inet6_csk_bind_conflict); /* * request_sock (formerly open request) hash tables. */ -static u32 inet6_synq_hash(const struct in6_addr *raddr, const u16 rport, +static u32 inet6_synq_hash(const struct in6_addr *raddr, const __be16 rport, const u32 rnd, const u16 synq_hsize) { u32 a = raddr->s6_addr32[0]; @@ -65,7 +65,7 @@ static u32 inet6_synq_hash(const struct in6_addr *raddr, const u16 rport, __jhash_mix(a, b, c); a += raddr->s6_addr32[3]; - b += (u32)rport; + b += (__force u32)rport; __jhash_mix(a, b, c); return c & (synq_hsize - 1); @@ -73,7 +73,7 @@ static u32 inet6_synq_hash(const struct in6_addr *raddr, const u16 rport, struct request_sock *inet6_csk_search_req(const struct sock *sk, struct request_sock ***prevp, - const __u16 rport, + const __be16 rport, const struct in6_addr *raddr, const struct in6_addr *laddr, const int iif) -- cgit v1.2.3-70-g09d2 From 3fbd418acc264e7069483d04165633ed3bfad80c Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:26:05 -0800 Subject: [LLC]: anotations Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/llc_pdu.h | 10 +++++----- net/llc/af_llc.c | 2 +- net/llc/llc_input.c | 4 ++-- 3 files changed, 8 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/net/llc_pdu.h b/include/net/llc_pdu.h index 8f6306581fa..aa33a477c3f 100644 --- a/include/net/llc_pdu.h +++ b/include/net/llc_pdu.h @@ -252,9 +252,9 @@ static inline void llc_pdu_header_init(struct sk_buff *skb, u8 type, */ static inline void llc_pdu_decode_sa(struct sk_buff *skb, u8 *sa) { - if (skb->protocol == ntohs(ETH_P_802_2)) + if (skb->protocol == htons(ETH_P_802_2)) memcpy(sa, eth_hdr(skb)->h_source, ETH_ALEN); - else if (skb->protocol == ntohs(ETH_P_TR_802_2)) { + else if (skb->protocol == htons(ETH_P_TR_802_2)) { memcpy(sa, tr_hdr(skb)->saddr, ETH_ALEN); *sa &= 0x7F; } @@ -269,9 +269,9 @@ static inline void llc_pdu_decode_sa(struct sk_buff *skb, u8 *sa) */ static inline void llc_pdu_decode_da(struct sk_buff *skb, u8 *da) { - if (skb->protocol == ntohs(ETH_P_802_2)) + if (skb->protocol == htons(ETH_P_802_2)) memcpy(da, eth_hdr(skb)->h_dest, ETH_ALEN); - else if (skb->protocol == ntohs(ETH_P_TR_802_2)) + else if (skb->protocol == htons(ETH_P_TR_802_2)) memcpy(da, tr_hdr(skb)->daddr, ETH_ALEN); } @@ -345,7 +345,7 @@ static inline void llc_pdu_init_as_test_rsp(struct sk_buff *skb, pdu->ctrl_1 = LLC_PDU_TYPE_U; pdu->ctrl_1 |= LLC_1_PDU_CMD_TEST; pdu->ctrl_1 |= LLC_U_PF_BIT_MASK; - if (ev_skb->protocol == ntohs(ETH_P_802_2)) { + if (ev_skb->protocol == htons(ETH_P_802_2)) { struct llc_pdu_un *ev_pdu = llc_pdu_un_hdr(ev_skb); int dsize; diff --git a/net/llc/af_llc.c b/net/llc/af_llc.c index 2652ead96c6..190bb3e0518 100644 --- a/net/llc/af_llc.c +++ b/net/llc/af_llc.c @@ -64,7 +64,7 @@ static inline u16 llc_ui_next_link_no(int sap) * * Given an ARP header type return the corresponding ethernet protocol. */ -static inline u16 llc_proto_type(u16 arphrd) +static inline __be16 llc_proto_type(u16 arphrd) { return arphrd == ARPHRD_IEEE802_TR ? htons(ETH_P_TR_802_2) : htons(ETH_P_802_2); diff --git a/net/llc/llc_input.c b/net/llc/llc_input.c index 94d2368ade9..db82aff6e40 100644 --- a/net/llc/llc_input.c +++ b/net/llc/llc_input.c @@ -115,8 +115,8 @@ static inline int llc_fixup_skb(struct sk_buff *skb) skb->h.raw += llc_len; skb_pull(skb, llc_len); if (skb->protocol == htons(ETH_P_802_2)) { - u16 pdulen = eth_hdr(skb)->h_proto, - data_size = ntohs(pdulen) - llc_len; + __be16 pdulen = eth_hdr(skb)->h_proto; + u16 data_size = ntohs(pdulen) - llc_len; if (unlikely(pskb_trim_rcsum(skb, data_size))) return 0; -- cgit v1.2.3-70-g09d2 From 0e11c91e1e912bc4db5b71607d149e7e9a77e756 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:26:29 -0800 Subject: [AF_PACKET]: annotate Weirdness: the third argument of socket() is net-endian here. Oh, well - it's documented in packet(7). Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/if_packet.h | 6 ++++-- net/packet/af_packet.c | 18 ++++++++++-------- 2 files changed, 14 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/linux/if_packet.h b/include/linux/if_packet.h index b92558549d2..99393ef3af3 100644 --- a/include/linux/if_packet.h +++ b/include/linux/if_packet.h @@ -1,17 +1,19 @@ #ifndef __LINUX_IF_PACKET_H #define __LINUX_IF_PACKET_H +#include + struct sockaddr_pkt { unsigned short spkt_family; unsigned char spkt_device[14]; - unsigned short spkt_protocol; + __be16 spkt_protocol; }; struct sockaddr_ll { unsigned short sll_family; - unsigned short sll_protocol; + __be16 sll_protocol; int sll_ifindex; unsigned short sll_hatype; unsigned char sll_pkttype; diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index f4ccb90e673..271d2eed069 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -201,7 +201,7 @@ struct packet_sock { spinlock_t bind_lock; char running; /* prot_hook is attached*/ int ifindex; /* bound device */ - unsigned short num; + __be16 num; #ifdef CONFIG_PACKET_MULTICAST struct packet_mclist *mclist; #endif @@ -331,7 +331,7 @@ static int packet_sendmsg_spkt(struct kiocb *iocb, struct socket *sock, struct sockaddr_pkt *saddr=(struct sockaddr_pkt *)msg->msg_name; struct sk_buff *skb; struct net_device *dev; - unsigned short proto=0; + __be16 proto=0; int err; /* @@ -704,7 +704,7 @@ static int packet_sendmsg(struct kiocb *iocb, struct socket *sock, struct sockaddr_ll *saddr=(struct sockaddr_ll *)msg->msg_name; struct sk_buff *skb; struct net_device *dev; - unsigned short proto; + __be16 proto; unsigned char *addr; int ifindex, err, reserve = 0; @@ -858,7 +858,7 @@ static int packet_release(struct socket *sock) * Attach a packet hook. */ -static int packet_do_bind(struct sock *sk, struct net_device *dev, int protocol) +static int packet_do_bind(struct sock *sk, struct net_device *dev, __be16 protocol) { struct packet_sock *po = pkt_sk(sk); /* @@ -983,6 +983,7 @@ static int packet_create(struct socket *sock, int protocol) { struct sock *sk; struct packet_sock *po; + __be16 proto = (__force __be16)protocol; /* weird, but documented */ int err; if (!capable(CAP_NET_RAW)) @@ -1010,7 +1011,7 @@ static int packet_create(struct socket *sock, int protocol) po = pkt_sk(sk); sk->sk_family = PF_PACKET; - po->num = protocol; + po->num = proto; sk->sk_destruct = packet_sock_destruct; atomic_inc(&packet_socks_nr); @@ -1027,8 +1028,8 @@ static int packet_create(struct socket *sock, int protocol) #endif po->prot_hook.af_packet_priv = sk; - if (protocol) { - po->prot_hook.type = protocol; + if (proto) { + po->prot_hook.type = proto; dev_add_pack(&po->prot_hook); sock_hold(sk); po->running = 1; @@ -1624,7 +1625,8 @@ static int packet_set_ring(struct sock *sk, struct tpacket_req *req, int closing { char **pg_vec = NULL; struct packet_sock *po = pkt_sk(sk); - int was_running, num, order = 0; + int was_running, order = 0; + __be16 num; int err = 0; if (req->tp_block_nr) { -- cgit v1.2.3-70-g09d2 From 98a4a86128d7179b22365e16bf880e849e20bc7d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:26:51 -0800 Subject: [NETFILTER]: trivial annotations Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/netfilter/nfnetlink.h | 2 +- include/linux/netfilter/nfnetlink_log.h | 10 +++++----- include/linux/netfilter/nfnetlink_queue.h | 18 +++++++++--------- include/linux/netfilter/xt_conntrack.h | 4 ++-- include/linux/netfilter/xt_policy.h | 2 +- include/linux/types.h | 2 ++ net/netfilter/nfnetlink_log.c | 22 ++++++++++------------ net/netfilter/nfnetlink_queue.c | 9 ++++----- net/netfilter/xt_multiport.c | 4 ++-- 9 files changed, 36 insertions(+), 37 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nfnetlink.h b/include/linux/netfilter/nfnetlink.h index 6d8e3e5a80e..1e9c821f152 100644 --- a/include/linux/netfilter/nfnetlink.h +++ b/include/linux/netfilter/nfnetlink.h @@ -78,7 +78,7 @@ struct nfattr struct nfgenmsg { u_int8_t nfgen_family; /* AF_xxx */ u_int8_t version; /* nfnetlink version */ - u_int16_t res_id; /* resource id */ + __be16 res_id; /* resource id */ }; #define NFNETLINK_V0 0 diff --git a/include/linux/netfilter/nfnetlink_log.h b/include/linux/netfilter/nfnetlink_log.h index 87b92f8b988..55a2a2b814e 100644 --- a/include/linux/netfilter/nfnetlink_log.h +++ b/include/linux/netfilter/nfnetlink_log.h @@ -16,20 +16,20 @@ enum nfulnl_msg_types { }; struct nfulnl_msg_packet_hdr { - u_int16_t hw_protocol; /* hw protocol (network order) */ + __be16 hw_protocol; /* hw protocol (network order) */ u_int8_t hook; /* netfilter hook */ u_int8_t _pad; }; struct nfulnl_msg_packet_hw { - u_int16_t hw_addrlen; + __be16 hw_addrlen; u_int16_t _pad; u_int8_t hw_addr[8]; }; struct nfulnl_msg_packet_timestamp { - aligned_u64 sec; - aligned_u64 usec; + aligned_be64 sec; + aligned_be64 usec; }; #define NFULNL_PREFIXLEN 30 /* just like old log target */ @@ -67,7 +67,7 @@ struct nfulnl_msg_config_cmd { } __attribute__ ((packed)); struct nfulnl_msg_config_mode { - u_int32_t copy_range; + __be32 copy_range; u_int8_t copy_mode; u_int8_t _pad; } __attribute__ ((packed)); diff --git a/include/linux/netfilter/nfnetlink_queue.h b/include/linux/netfilter/nfnetlink_queue.h index 36af0360b56..4beea3d6fda 100644 --- a/include/linux/netfilter/nfnetlink_queue.h +++ b/include/linux/netfilter/nfnetlink_queue.h @@ -13,20 +13,20 @@ enum nfqnl_msg_types { }; struct nfqnl_msg_packet_hdr { - u_int32_t packet_id; /* unique ID of packet in queue */ - u_int16_t hw_protocol; /* hw protocol (network order) */ + __be32 packet_id; /* unique ID of packet in queue */ + __be16 hw_protocol; /* hw protocol (network order) */ u_int8_t hook; /* netfilter hook */ } __attribute__ ((packed)); struct nfqnl_msg_packet_hw { - u_int16_t hw_addrlen; + __be16 hw_addrlen; u_int16_t _pad; u_int8_t hw_addr[8]; }; struct nfqnl_msg_packet_timestamp { - aligned_u64 sec; - aligned_u64 usec; + aligned_be64 sec; + aligned_be64 usec; }; enum nfqnl_attr_type { @@ -47,8 +47,8 @@ enum nfqnl_attr_type { #define NFQA_MAX (__NFQA_MAX - 1) struct nfqnl_msg_verdict_hdr { - u_int32_t verdict; - u_int32_t id; + __be32 verdict; + __be32 id; }; @@ -63,7 +63,7 @@ enum nfqnl_msg_config_cmds { struct nfqnl_msg_config_cmd { u_int8_t command; /* nfqnl_msg_config_cmds */ u_int8_t _pad; - u_int16_t pf; /* AF_xxx for PF_[UN]BIND */ + __be16 pf; /* AF_xxx for PF_[UN]BIND */ }; enum nfqnl_config_mode { @@ -73,7 +73,7 @@ enum nfqnl_config_mode { }; struct nfqnl_msg_config_params { - u_int32_t copy_range; + __be32 copy_range; u_int8_t copy_mode; /* enum nfqnl_config_mode */ } __attribute__ ((packed)); diff --git a/include/linux/netfilter/xt_conntrack.h b/include/linux/netfilter/xt_conntrack.h index 4c2d9945ca5..70b6f718cf4 100644 --- a/include/linux/netfilter/xt_conntrack.h +++ b/include/linux/netfilter/xt_conntrack.h @@ -29,14 +29,14 @@ struct ip_conntrack_old_tuple { struct { - __u32 ip; + __be32 ip; union { __u16 all; } u; } src; struct { - __u32 ip; + __be32 ip; union { __u16 all; } u; diff --git a/include/linux/netfilter/xt_policy.h b/include/linux/netfilter/xt_policy.h index a8132ec076f..45654d359a6 100644 --- a/include/linux/netfilter/xt_policy.h +++ b/include/linux/netfilter/xt_policy.h @@ -39,7 +39,7 @@ struct xt_policy_elem union xt_policy_addr smask; union xt_policy_addr daddr; union xt_policy_addr dmask; - u_int32_t spi; + __be32 spi; u_int32_t reqid; u_int8_t proto; u_int8_t mode; diff --git a/include/linux/types.h b/include/linux/types.h index 750f085fa56..9f11fdd2bd7 100644 --- a/include/linux/types.h +++ b/include/linux/types.h @@ -128,6 +128,8 @@ typedef __s64 int64_t; /* this is a special 64bit data type that is 8-byte aligned */ #define aligned_u64 unsigned long long __attribute__((aligned(8))) +#define aligned_be64 __be64 __attribute__((aligned(8))) +#define aligned_le64 __le64 __attribute__((aligned(8))) /** * The type used for indexing onto a disc or disc partition. diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index 1e5207b80fe..856ed0d1997 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -414,7 +414,7 @@ __build_packet_message(struct nfulnl_instance *inst, struct nfulnl_msg_packet_hdr pmsg; struct nlmsghdr *nlh; struct nfgenmsg *nfmsg; - u_int32_t tmp_uint; + __be32 tmp_uint; UDEBUG("entered\n"); @@ -508,11 +508,9 @@ __build_packet_message(struct nfulnl_instance *inst, if (indev && skb->dev && skb->dev->hard_header_parse) { struct nfulnl_msg_packet_hw phw; - - phw.hw_addrlen = - skb->dev->hard_header_parse((struct sk_buff *)skb, + int len = skb->dev->hard_header_parse((struct sk_buff *)skb, phw.hw_addr); - phw.hw_addrlen = htons(phw.hw_addrlen); + phw.hw_addrlen = htons(len); NFA_PUT(inst->skb, NFULA_HWADDR, sizeof(phw), &phw); } @@ -529,7 +527,7 @@ __build_packet_message(struct nfulnl_instance *inst, if (skb->sk) { read_lock_bh(&skb->sk->sk_callback_lock); if (skb->sk->sk_socket && skb->sk->sk_socket->file) { - u_int32_t uid = htonl(skb->sk->sk_socket->file->f_uid); + __be32 uid = htonl(skb->sk->sk_socket->file->f_uid); /* need to unlock here since NFA_PUT may goto */ read_unlock_bh(&skb->sk->sk_callback_lock); NFA_PUT(inst->skb, NFULA_UID, sizeof(uid), &uid); @@ -882,15 +880,15 @@ nfulnl_recv_config(struct sock *ctnl, struct sk_buff *skb, } if (nfula[NFULA_CFG_TIMEOUT-1]) { - u_int32_t timeout = - *(u_int32_t *)NFA_DATA(nfula[NFULA_CFG_TIMEOUT-1]); + __be32 timeout = + *(__be32 *)NFA_DATA(nfula[NFULA_CFG_TIMEOUT-1]); nfulnl_set_timeout(inst, ntohl(timeout)); } if (nfula[NFULA_CFG_NLBUFSIZ-1]) { - u_int32_t nlbufsiz = - *(u_int32_t *)NFA_DATA(nfula[NFULA_CFG_NLBUFSIZ-1]); + __be32 nlbufsiz = + *(__be32 *)NFA_DATA(nfula[NFULA_CFG_NLBUFSIZ-1]); nfulnl_set_nlbufsiz(inst, ntohl(nlbufsiz)); } @@ -903,8 +901,8 @@ nfulnl_recv_config(struct sock *ctnl, struct sk_buff *skb, } if (nfula[NFULA_CFG_FLAGS-1]) { - u_int16_t flags = - *(u_int16_t *)NFA_DATA(nfula[NFULA_CFG_FLAGS-1]); + __be16 flags = + *(__be16 *)NFA_DATA(nfula[NFULA_CFG_FLAGS-1]); nfulnl_set_flags(inst, ntohs(flags)); } diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index e815a9aa6e9..4ab7b1416bb 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -349,7 +349,7 @@ nfqnl_build_packet_message(struct nfqnl_instance *queue, struct sk_buff *entskb = entry->skb; struct net_device *indev; struct net_device *outdev; - unsigned int tmp_uint; + __be32 tmp_uint; QDEBUG("entered\n"); @@ -489,10 +489,9 @@ nfqnl_build_packet_message(struct nfqnl_instance *queue, && entskb->dev->hard_header_parse) { struct nfqnl_msg_packet_hw phw; - phw.hw_addrlen = - entskb->dev->hard_header_parse(entskb, + int len = entskb->dev->hard_header_parse(entskb, phw.hw_addr); - phw.hw_addrlen = htons(phw.hw_addrlen); + phw.hw_addrlen = htons(len); NFA_PUT(skb, NFQA_HWADDR, sizeof(phw), &phw); } @@ -835,7 +834,7 @@ nfqnl_recv_verdict(struct sock *ctnl, struct sk_buff *skb, } if (nfqa[NFQA_MARK-1]) - entry->skb->nfmark = ntohl(*(u_int32_t *) + entry->skb->nfmark = ntohl(*(__be32 *) NFA_DATA(nfqa[NFQA_MARK-1])); issue_verdict(entry, verdict); diff --git a/net/netfilter/xt_multiport.c b/net/netfilter/xt_multiport.c index d3aefd38093..b4293058c6f 100644 --- a/net/netfilter/xt_multiport.c +++ b/net/netfilter/xt_multiport.c @@ -104,7 +104,7 @@ match(const struct sk_buff *skb, unsigned int protoff, int *hotdrop) { - u16 _ports[2], *pptr; + __be16 _ports[2], *pptr; const struct xt_multiport *multiinfo = matchinfo; if (offset) @@ -135,7 +135,7 @@ match_v1(const struct sk_buff *skb, unsigned int protoff, int *hotdrop) { - u16 _ports[2], *pptr; + __be16 _ports[2], *pptr; const struct xt_multiport_v1 *multiinfo = matchinfo; if (offset) -- cgit v1.2.3-70-g09d2 From ae08e1f092210619fe49551aa3ed0dc0003d5880 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:27:11 -0800 Subject: [IPV6]: ip6_output annotations Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/skbuff.h | 2 +- net/ipv6/ip6_output.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 85577a4ffa6..7fc9a3aaa1c 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -139,7 +139,7 @@ struct skb_shared_info { /* Warning: this field is not always filled in (UFO)! */ unsigned short gso_segs; unsigned short gso_type; - unsigned int ip6_frag_id; + __be32 ip6_frag_id; struct sk_buff *frag_list; skb_frag_t frags[MAX_SKB_FRAGS]; }; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 9ff1be1bc2e..93330685adf 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -267,7 +267,7 @@ int ip6_nd_hdr(struct sock *sk, struct sk_buff *skb, struct net_device *dev, hdr = (struct ipv6hdr *) skb_put(skb, sizeof(struct ipv6hdr)); skb->nh.ipv6h = hdr; - *(u32*)hdr = htonl(0x60000000); + *(__be32*)hdr = htonl(0x60000000); hdr->payload_len = htons(len); hdr->nexthdr = proto; @@ -571,7 +571,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) struct ipv6hdr *tmp_hdr; struct frag_hdr *fh; unsigned int mtu, hlen, left, len; - u32 frag_id = 0; + __be32 frag_id = 0; int ptr, offset = 0, err=0; u8 *prevhdr, nexthdr = 0; -- cgit v1.2.3-70-g09d2 From 6ba9c755e5f044c6a37296481bf6a861e4193a37 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:28:19 -0800 Subject: [BLUETOOTH]: rfcomm endianness annotations Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/bluetooth/rfcomm.h | 4 ++-- net/bluetooth/rfcomm/core.c | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/rfcomm.h b/include/net/bluetooth/rfcomm.h index 89d743cfdfd..3c563f02907 100644 --- a/include/net/bluetooth/rfcomm.h +++ b/include/net/bluetooth/rfcomm.h @@ -124,7 +124,7 @@ struct rfcomm_pn { u8 flow_ctrl; u8 priority; u8 ack_timer; - u16 mtu; + __le16 mtu; u8 max_retrans; u8 credits; } __attribute__ ((packed)); @@ -136,7 +136,7 @@ struct rfcomm_rpn { u8 flow_ctrl; u8 xon_char; u8 xoff_char; - u16 param_mask; + __le16 param_mask; } __attribute__ ((packed)); struct rfcomm_rls { diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c index ddc4e9d5963..f3e5b7e2e04 100644 --- a/net/bluetooth/rfcomm/core.c +++ b/net/bluetooth/rfcomm/core.c @@ -1018,7 +1018,7 @@ static void rfcomm_make_uih(struct sk_buff *skb, u8 addr) if (len > 127) { hdr = (void *) skb_push(skb, 4); - put_unaligned(htobs(__len16(len)), (u16 *) &hdr->len); + put_unaligned(htobs(__len16(len)), (__le16 *) &hdr->len); } else { hdr = (void *) skb_push(skb, 3); hdr->len = __len8(len); -- cgit v1.2.3-70-g09d2 From c1a856c9640c9ff3d70bbd8214b6a0974609eef8 Mon Sep 17 00:00:00 2001 From: Venkat Yekkirala Date: Wed, 8 Nov 2006 17:03:44 -0600 Subject: SELinux: Various xfrm labeling fixes Since the upstreaming of the mlsxfrm modification a few months back, testing has resulted in the identification of the following issues/bugs that are resolved in this patch set. 1. Fix the security context used in the IKE negotiation to be the context of the socket as opposed to the context of the SPD rule. 2. Fix SO_PEERSEC for tcp sockets to return the security context of the peer as opposed to the source. 3. Fix the selection of an SA for an outgoing packet to be at the same context as the originating socket/flow. The following would be the result of applying this patchset: - SO_PEERSEC will now correctly return the peer's context. - IKE deamons will receive the context of the source socket/flow as opposed to the SPD rule's context so that the negotiated SA will be at the same context as the source socket/flow. - The SELinux policy will require one or more of the following for a socket to be able to communicate with/without SAs: 1. To enable a socket to communicate without using labeled-IPSec SAs: allow socket_t unlabeled_t:association { sendto recvfrom } 2. To enable a socket to communicate with labeled-IPSec SAs: allow socket_t self:association { sendto }; allow socket_t peer_sa_t:association { recvfrom }; This Patch: Pass correct security context to IKE for use in negotiation Fix the security context passed to IKE for use in negotiation to be the context of the socket as opposed to the context of the SPD rule so that the SA carries the label of the originating socket/flow. Signed-off-by: Venkat Yekkirala Signed-off-by: James Morris --- include/linux/security.h | 21 ++++++++++----------- security/dummy.c | 4 ++-- security/selinux/include/xfrm.h | 4 ++-- security/selinux/xfrm.c | 35 +++++++++-------------------------- 4 files changed, 23 insertions(+), 41 deletions(-) (limited to 'include') diff --git a/include/linux/security.h b/include/linux/security.h index b200b9856f3..a509329a669 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -836,10 +836,8 @@ struct request_sock; * used by the XFRM system. * @sec_ctx contains the security context information being provided by * the user-level policy update program (e.g., setkey). - * @sk refers to the sock from which to derive the security context. * Allocate a security structure to the xp->security field; the security - * field is initialized to NULL when the xfrm_policy is allocated. Only - * one of sec_ctx or sock can be specified. + * field is initialized to NULL when the xfrm_policy is allocated. * Return 0 if operation was successful (memory to allocate, legal context) * @xfrm_policy_clone_security: * @old contains an existing xfrm_policy in the SPD. @@ -858,9 +856,6 @@ struct request_sock; * Database by the XFRM system. * @sec_ctx contains the security context information being provided by * the user-level SA generation program (e.g., setkey or racoon). - * @polsec contains the security context information associated with a xfrm - * policy rule from which to take the base context. polsec must be NULL - * when sec_ctx is specified. * @secid contains the secid from which to take the mls portion of the context. * Allocate a security structure to the x->security field; the security * field is initialized to NULL when the xfrm_state is allocated. Set the @@ -1378,12 +1373,12 @@ struct security_operations { #ifdef CONFIG_SECURITY_NETWORK_XFRM int (*xfrm_policy_alloc_security) (struct xfrm_policy *xp, - struct xfrm_user_sec_ctx *sec_ctx, struct sock *sk); + struct xfrm_user_sec_ctx *sec_ctx); int (*xfrm_policy_clone_security) (struct xfrm_policy *old, struct xfrm_policy *new); void (*xfrm_policy_free_security) (struct xfrm_policy *xp); int (*xfrm_policy_delete_security) (struct xfrm_policy *xp); int (*xfrm_state_alloc_security) (struct xfrm_state *x, - struct xfrm_user_sec_ctx *sec_ctx, struct xfrm_sec_ctx *polsec, + struct xfrm_user_sec_ctx *sec_ctx, u32 secid); void (*xfrm_state_free_security) (struct xfrm_state *x); int (*xfrm_state_delete_security) (struct xfrm_state *x); @@ -3120,7 +3115,7 @@ static inline void security_inet_csk_clone(struct sock *newsk, #ifdef CONFIG_SECURITY_NETWORK_XFRM static inline int security_xfrm_policy_alloc(struct xfrm_policy *xp, struct xfrm_user_sec_ctx *sec_ctx) { - return security_ops->xfrm_policy_alloc_security(xp, sec_ctx, NULL); + return security_ops->xfrm_policy_alloc_security(xp, sec_ctx); } static inline int security_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new) @@ -3141,7 +3136,7 @@ static inline int security_xfrm_policy_delete(struct xfrm_policy *xp) static inline int security_xfrm_state_alloc(struct xfrm_state *x, struct xfrm_user_sec_ctx *sec_ctx) { - return security_ops->xfrm_state_alloc_security(x, sec_ctx, NULL, 0); + return security_ops->xfrm_state_alloc_security(x, sec_ctx, 0); } static inline int security_xfrm_state_alloc_acquire(struct xfrm_state *x, @@ -3149,7 +3144,11 @@ static inline int security_xfrm_state_alloc_acquire(struct xfrm_state *x, { if (!polsec) return 0; - return security_ops->xfrm_state_alloc_security(x, NULL, polsec, secid); + /* + * We want the context to be taken from secid which is usually + * from the sock. + */ + return security_ops->xfrm_state_alloc_security(x, NULL, secid); } static inline int security_xfrm_state_delete(struct xfrm_state *x) diff --git a/security/dummy.c b/security/dummy.c index 43874c1e6e2..838d8442cf3 100644 --- a/security/dummy.c +++ b/security/dummy.c @@ -836,7 +836,7 @@ static inline void dummy_req_classify_flow(const struct request_sock *req, #ifdef CONFIG_SECURITY_NETWORK_XFRM static int dummy_xfrm_policy_alloc_security(struct xfrm_policy *xp, - struct xfrm_user_sec_ctx *sec_ctx, struct sock *sk) + struct xfrm_user_sec_ctx *sec_ctx) { return 0; } @@ -856,7 +856,7 @@ static int dummy_xfrm_policy_delete_security(struct xfrm_policy *xp) } static int dummy_xfrm_state_alloc_security(struct xfrm_state *x, - struct xfrm_user_sec_ctx *sec_ctx, struct xfrm_sec_ctx *pol, u32 secid) + struct xfrm_user_sec_ctx *sec_ctx, u32 secid) { return 0; } diff --git a/security/selinux/include/xfrm.h b/security/selinux/include/xfrm.h index 526b28019ac..8e329ddb5e3 100644 --- a/security/selinux/include/xfrm.h +++ b/security/selinux/include/xfrm.h @@ -8,12 +8,12 @@ #define _SELINUX_XFRM_H_ int selinux_xfrm_policy_alloc(struct xfrm_policy *xp, - struct xfrm_user_sec_ctx *sec_ctx, struct sock *sk); + struct xfrm_user_sec_ctx *sec_ctx); int selinux_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new); void selinux_xfrm_policy_free(struct xfrm_policy *xp); int selinux_xfrm_policy_delete(struct xfrm_policy *xp); int selinux_xfrm_state_alloc(struct xfrm_state *x, - struct xfrm_user_sec_ctx *sec_ctx, struct xfrm_sec_ctx *pol, u32 secid); + struct xfrm_user_sec_ctx *sec_ctx, u32 secid); void selinux_xfrm_state_free(struct xfrm_state *x); int selinux_xfrm_state_delete(struct xfrm_state *x); int selinux_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir); diff --git a/security/selinux/xfrm.c b/security/selinux/xfrm.c index 675b995a67c..4d5a043cdfa 100644 --- a/security/selinux/xfrm.c +++ b/security/selinux/xfrm.c @@ -226,16 +226,15 @@ int selinux_xfrm_decode_session(struct sk_buff *skb, u32 *sid, int ckall) * CTX does not have a meaningful value on input */ static int selinux_xfrm_sec_ctx_alloc(struct xfrm_sec_ctx **ctxp, - struct xfrm_user_sec_ctx *uctx, struct xfrm_sec_ctx *pol, u32 sid) + struct xfrm_user_sec_ctx *uctx, u32 sid) { int rc = 0; struct task_security_struct *tsec = current->security; struct xfrm_sec_ctx *ctx = NULL; char *ctx_str = NULL; u32 str_len; - u32 ctx_sid; - BUG_ON(uctx && pol); + BUG_ON(uctx && sid); if (!uctx) goto not_from_user; @@ -279,15 +278,7 @@ static int selinux_xfrm_sec_ctx_alloc(struct xfrm_sec_ctx **ctxp, return rc; not_from_user: - if (pol) { - rc = security_sid_mls_copy(pol->ctx_sid, sid, &ctx_sid); - if (rc) - goto out; - } - else - ctx_sid = sid; - - rc = security_sid_to_context(ctx_sid, &ctx_str, &str_len); + rc = security_sid_to_context(sid, &ctx_str, &str_len); if (rc) goto out; @@ -302,7 +293,7 @@ not_from_user: ctx->ctx_doi = XFRM_SC_DOI_LSM; ctx->ctx_alg = XFRM_SC_ALG_SELINUX; - ctx->ctx_sid = ctx_sid; + ctx->ctx_sid = sid; ctx->ctx_len = str_len; memcpy(ctx->ctx_str, ctx_str, @@ -323,22 +314,14 @@ out2: * xfrm_policy. */ int selinux_xfrm_policy_alloc(struct xfrm_policy *xp, - struct xfrm_user_sec_ctx *uctx, struct sock *sk) + struct xfrm_user_sec_ctx *uctx) { int err; - u32 sid; BUG_ON(!xp); - BUG_ON(uctx && sk); - - if (sk) { - struct sk_security_struct *ssec = sk->sk_security; - sid = ssec->sid; - } - else - sid = SECSID_NULL; + BUG_ON(!uctx); - err = selinux_xfrm_sec_ctx_alloc(&xp->security, uctx, NULL, sid); + err = selinux_xfrm_sec_ctx_alloc(&xp->security, uctx, 0); return err; } @@ -399,13 +382,13 @@ int selinux_xfrm_policy_delete(struct xfrm_policy *xp) * xfrm_state. */ int selinux_xfrm_state_alloc(struct xfrm_state *x, struct xfrm_user_sec_ctx *uctx, - struct xfrm_sec_ctx *pol, u32 secid) + u32 secid) { int err; BUG_ON(!x); - err = selinux_xfrm_sec_ctx_alloc(&x->security, uctx, pol, secid); + err = selinux_xfrm_sec_ctx_alloc(&x->security, uctx, secid); return err; } -- cgit v1.2.3-70-g09d2 From 6b877699c6f1efede4545bcecc367786a472eedb Mon Sep 17 00:00:00 2001 From: Venkat Yekkirala Date: Wed, 8 Nov 2006 17:04:09 -0600 Subject: SELinux: Return correct context for SO_PEERSEC Fix SO_PEERSEC for tcp sockets to return the security context of the peer (as represented by the SA from the peer) as opposed to the SA used by the local/source socket. Signed-off-by: Venkat Yekkirala Signed-off-by: James Morris --- include/linux/security.h | 16 +++++++++++++++- include/net/request_sock.h | 1 + net/ipv4/tcp_input.c | 2 ++ security/dummy.c | 6 ++++++ security/selinux/hooks.c | 21 +++++++++++++++++---- security/selinux/include/xfrm.h | 12 ++++++------ security/selinux/xfrm.c | 40 +++------------------------------------- 7 files changed, 50 insertions(+), 48 deletions(-) (limited to 'include') diff --git a/include/linux/security.h b/include/linux/security.h index a509329a669..84cebcdb3f8 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -826,6 +826,8 @@ struct request_sock; * Sets the openreq's sid to socket's sid with MLS portion taken from peer sid. * @inet_csk_clone: * Sets the new child socket's sid to the openreq sid. + * @inet_conn_established: + * Sets the connection's peersid to the secmark on skb. * @req_classify_flow: * Sets the flow's sid to the openreq sid. * @@ -1368,6 +1370,7 @@ struct security_operations { int (*inet_conn_request)(struct sock *sk, struct sk_buff *skb, struct request_sock *req); void (*inet_csk_clone)(struct sock *newsk, const struct request_sock *req); + void (*inet_conn_established)(struct sock *sk, struct sk_buff *skb); void (*req_classify_flow)(const struct request_sock *req, struct flowi *fl); #endif /* CONFIG_SECURITY_NETWORK */ @@ -2961,9 +2964,15 @@ static inline void security_inet_csk_clone(struct sock *newsk, { security_ops->inet_csk_clone(newsk, req); } + +static inline void security_inet_conn_established(struct sock *sk, + struct sk_buff *skb) +{ + security_ops->inet_conn_established(sk, skb); +} #else /* CONFIG_SECURITY_NETWORK */ static inline int security_unix_stream_connect(struct socket * sock, - struct socket * other, + struct socket * other, struct sock * newsk) { return 0; @@ -3110,6 +3119,11 @@ static inline void security_inet_csk_clone(struct sock *newsk, const struct request_sock *req) { } + +static inline void security_inet_conn_established(struct sock *sk, + struct sk_buff *skb) +{ +} #endif /* CONFIG_SECURITY_NETWORK */ #ifdef CONFIG_SECURITY_NETWORK_XFRM diff --git a/include/net/request_sock.h b/include/net/request_sock.h index 8e165ca16bd..f743a941a4f 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -54,6 +54,7 @@ struct request_sock { struct request_sock_ops *rsk_ops; struct sock *sk; u32 secid; + u32 peer_secid; }; static inline struct request_sock *reqsk_alloc(struct request_sock_ops *ops) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index cf06accbe68..4a8c96cdec7 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -4230,6 +4230,8 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, mb(); tcp_set_state(sk, TCP_ESTABLISHED); + security_inet_conn_established(sk, skb); + /* Make sure socket is routed, for correct metrics. */ icsk->icsk_af_ops->rebuild_header(sk); diff --git a/security/dummy.c b/security/dummy.c index 838d8442cf3..0148d1518dd 100644 --- a/security/dummy.c +++ b/security/dummy.c @@ -828,6 +828,11 @@ static inline void dummy_inet_csk_clone(struct sock *newsk, { } +static inline void dummy_inet_conn_established(struct sock *sk, + struct sk_buff *skb) +{ +} + static inline void dummy_req_classify_flow(const struct request_sock *req, struct flowi *fl) { @@ -1108,6 +1113,7 @@ void security_fixup_ops (struct security_operations *ops) set_to_dummy_if_null(ops, sock_graft); set_to_dummy_if_null(ops, inet_conn_request); set_to_dummy_if_null(ops, inet_csk_clone); + set_to_dummy_if_null(ops, inet_conn_established); set_to_dummy_if_null(ops, req_classify_flow); #endif /* CONFIG_SECURITY_NETWORK */ #ifdef CONFIG_SECURITY_NETWORK_XFRM diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 28ee187ed22..5bbd599a447 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -3535,8 +3535,10 @@ static int selinux_socket_getpeersec_stream(struct socket *sock, char __user *op } else if (isec->sclass == SECCLASS_TCP_SOCKET) { peer_sid = selinux_netlbl_socket_getpeersec_stream(sock); - if (peer_sid == SECSID_NULL) - peer_sid = selinux_socket_getpeer_stream(sock->sk); + if (peer_sid == SECSID_NULL) { + ssec = sock->sk->sk_security; + peer_sid = ssec->peer_sid; + } if (peer_sid == SECSID_NULL) { err = -ENOPROTOOPT; goto out; @@ -3647,11 +3649,11 @@ static int selinux_inet_conn_request(struct sock *sk, struct sk_buff *skb, return 0; } - err = selinux_xfrm_decode_session(skb, &peersid, 0); - BUG_ON(err); + selinux_skb_xfrm_sid(skb, &peersid); if (peersid == SECSID_NULL) { req->secid = sksec->sid; + req->peer_secid = 0; return 0; } @@ -3660,6 +3662,7 @@ static int selinux_inet_conn_request(struct sock *sk, struct sk_buff *skb, return err; req->secid = newsid; + req->peer_secid = peersid; return 0; } @@ -3669,6 +3672,7 @@ static void selinux_inet_csk_clone(struct sock *newsk, struct sk_security_struct *newsksec = newsk->sk_security; newsksec->sid = req->secid; + newsksec->peer_sid = req->peer_secid; /* NOTE: Ideally, we should also get the isec->sid for the new socket in sync, but we don't have the isec available yet. So we will wait until sock_graft to do it, by which @@ -3677,6 +3681,14 @@ static void selinux_inet_csk_clone(struct sock *newsk, selinux_netlbl_sk_security_init(newsksec, req->rsk_ops->family); } +static void selinux_inet_conn_established(struct sock *sk, + struct sk_buff *skb) +{ + struct sk_security_struct *sksec = sk->sk_security; + + selinux_skb_xfrm_sid(skb, &sksec->peer_sid); +} + static void selinux_req_classify_flow(const struct request_sock *req, struct flowi *fl) { @@ -4739,6 +4751,7 @@ static struct security_operations selinux_ops = { .sock_graft = selinux_sock_graft, .inet_conn_request = selinux_inet_conn_request, .inet_csk_clone = selinux_inet_csk_clone, + .inet_conn_established = selinux_inet_conn_established, .req_classify_flow = selinux_req_classify_flow, #ifdef CONFIG_SECURITY_NETWORK_XFRM diff --git a/security/selinux/include/xfrm.h b/security/selinux/include/xfrm.h index 8e329ddb5e3..27502365d70 100644 --- a/security/selinux/include/xfrm.h +++ b/security/selinux/include/xfrm.h @@ -39,7 +39,6 @@ int selinux_xfrm_sock_rcv_skb(u32 sid, struct sk_buff *skb, struct avc_audit_data *ad); int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb, struct avc_audit_data *ad); -u32 selinux_socket_getpeer_stream(struct sock *sk); u32 selinux_socket_getpeer_dgram(struct sk_buff *skb); int selinux_xfrm_decode_session(struct sk_buff *skb, u32 *sid, int ckall); #else @@ -55,11 +54,6 @@ static inline int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb, return 0; } -static inline int selinux_socket_getpeer_stream(struct sock *sk) -{ - return SECSID_NULL; -} - static inline int selinux_socket_getpeer_dgram(struct sk_buff *skb) { return SECSID_NULL; @@ -71,4 +65,10 @@ static inline int selinux_xfrm_decode_session(struct sk_buff *skb, u32 *sid, int } #endif +static inline void selinux_skb_xfrm_sid(struct sk_buff *skb, u32 *sid) +{ + int err = selinux_xfrm_decode_session(skb, sid, 0); + BUG_ON(err); +} + #endif /* _SELINUX_XFRM_H_ */ diff --git a/security/selinux/xfrm.c b/security/selinux/xfrm.c index 4d5a043cdfa..8fef74271f2 100644 --- a/security/selinux/xfrm.c +++ b/security/selinux/xfrm.c @@ -184,7 +184,8 @@ int selinux_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm, } /* - * LSM hook implementation that determines the sid for the session. + * LSM hook implementation that checks and/or returns the xfrm sid for the + * incoming packet. */ int selinux_xfrm_decode_session(struct sk_buff *skb, u32 *sid, int ckall) @@ -402,44 +403,9 @@ void selinux_xfrm_state_free(struct xfrm_state *x) kfree(ctx); } -/* - * SELinux internal function to retrieve the context of a connected - * (sk->sk_state == TCP_ESTABLISHED) TCP socket based on its security - * association used to connect to the remote socket. - * - * Retrieve via getsockopt SO_PEERSEC. - */ -u32 selinux_socket_getpeer_stream(struct sock *sk) -{ - struct dst_entry *dst, *dst_test; - u32 peer_sid = SECSID_NULL; - - if (sk->sk_state != TCP_ESTABLISHED) - goto out; - - dst = sk_dst_get(sk); - if (!dst) - goto out; - - for (dst_test = dst; dst_test != 0; - dst_test = dst_test->child) { - struct xfrm_state *x = dst_test->xfrm; - - if (x && selinux_authorizable_xfrm(x)) { - struct xfrm_sec_ctx *ctx = x->security; - peer_sid = ctx->ctx_sid; - break; - } - } - dst_release(dst); - -out: - return peer_sid; -} - /* * SELinux internal function to retrieve the context of a UDP packet - * based on its security association used to connect to the remote socket. + * based on its security association. * * Retrieve via setsockopt IP_PASSSEC and recvmsg with control message * type SCM_SECURITY. -- cgit v1.2.3-70-g09d2 From 67f83cbf081a70426ff667e8d14f94e13ed3bdca Mon Sep 17 00:00:00 2001 From: Venkat Yekkirala Date: Wed, 8 Nov 2006 17:04:26 -0600 Subject: SELinux: Fix SA selection semantics Fix the selection of an SA for an outgoing packet to be at the same context as the originating socket/flow. This eliminates the SELinux policy's ability to use/sendto SAs with contexts other than the socket's. With this patch applied, the SELinux policy will require one or more of the following for a socket to be able to communicate with/without SAs: 1. To enable a socket to communicate without using labeled-IPSec SAs: allow socket_t unlabeled_t:association { sendto recvfrom } 2. To enable a socket to communicate with labeled-IPSec SAs: allow socket_t self:association { sendto }; allow socket_t peer_sa_t:association { recvfrom }; Signed-off-by: Venkat Yekkirala Signed-off-by: James Morris --- include/linux/security.h | 19 -------- net/xfrm/xfrm_policy.c | 3 +- security/dummy.c | 7 --- security/selinux/hooks.c | 26 +++++++---- security/selinux/include/xfrm.h | 7 +-- security/selinux/xfrm.c | 101 +++++++++++++++++++--------------------- 6 files changed, 70 insertions(+), 93 deletions(-) (limited to 'include') diff --git a/include/linux/security.h b/include/linux/security.h index 84cebcdb3f8..83cdefae993 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -886,11 +886,6 @@ struct request_sock; * @xp contains the policy to check for a match. * @fl contains the flow to check for a match. * Return 1 if there is a match. - * @xfrm_flow_state_match: - * @fl contains the flow key to match. - * @xfrm points to the xfrm_state to match. - * @xp points to the xfrm_policy to match. - * Return 1 if there is a match. * @xfrm_decode_session: * @skb points to skb to decode. * @secid points to the flow key secid to set. @@ -1388,8 +1383,6 @@ struct security_operations { int (*xfrm_policy_lookup)(struct xfrm_policy *xp, u32 fl_secid, u8 dir); int (*xfrm_state_pol_flow_match)(struct xfrm_state *x, struct xfrm_policy *xp, struct flowi *fl); - int (*xfrm_flow_state_match)(struct flowi *fl, struct xfrm_state *xfrm, - struct xfrm_policy *xp); int (*xfrm_decode_session)(struct sk_buff *skb, u32 *secid, int ckall); #endif /* CONFIG_SECURITY_NETWORK_XFRM */ @@ -3186,12 +3179,6 @@ static inline int security_xfrm_state_pol_flow_match(struct xfrm_state *x, return security_ops->xfrm_state_pol_flow_match(x, xp, fl); } -static inline int security_xfrm_flow_state_match(struct flowi *fl, - struct xfrm_state *xfrm, struct xfrm_policy *xp) -{ - return security_ops->xfrm_flow_state_match(fl, xfrm, xp); -} - static inline int security_xfrm_decode_session(struct sk_buff *skb, u32 *secid) { return security_ops->xfrm_decode_session(skb, secid, 1); @@ -3255,12 +3242,6 @@ static inline int security_xfrm_state_pol_flow_match(struct xfrm_state *x, return 1; } -static inline int security_xfrm_flow_state_match(struct flowi *fl, - struct xfrm_state *xfrm, struct xfrm_policy *xp) -{ - return 1; -} - static inline int security_xfrm_decode_session(struct sk_buff *skb, u32 *secid) { return 0; diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 7736b23c3f0..b88b038530c 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -1894,7 +1894,8 @@ int xfrm_bundle_ok(struct xfrm_policy *pol, struct xfrm_dst *first, if (fl && !xfrm_selector_match(&dst->xfrm->sel, fl, family)) return 0; - if (fl && !security_xfrm_flow_state_match(fl, dst->xfrm, pol)) + if (fl && pol && + !security_xfrm_state_pol_flow_match(dst->xfrm, pol, fl)) return 0; if (dst->xfrm->km.state != XFRM_STATE_VALID) return 0; diff --git a/security/dummy.c b/security/dummy.c index 0148d1518dd..558795b237d 100644 --- a/security/dummy.c +++ b/security/dummy.c @@ -886,12 +886,6 @@ static int dummy_xfrm_state_pol_flow_match(struct xfrm_state *x, return 1; } -static int dummy_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm, - struct xfrm_policy *xp) -{ - return 1; -} - static int dummy_xfrm_decode_session(struct sk_buff *skb, u32 *fl, int ckall) { return 0; @@ -1126,7 +1120,6 @@ void security_fixup_ops (struct security_operations *ops) set_to_dummy_if_null(ops, xfrm_state_delete_security); set_to_dummy_if_null(ops, xfrm_policy_lookup); set_to_dummy_if_null(ops, xfrm_state_pol_flow_match); - set_to_dummy_if_null(ops, xfrm_flow_state_match); set_to_dummy_if_null(ops, xfrm_decode_session); #endif /* CONFIG_SECURITY_NETWORK_XFRM */ #ifdef CONFIG_KEYS diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 5bbd599a447..956137baf3e 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -2889,7 +2889,8 @@ static void selinux_task_to_inode(struct task_struct *p, } /* Returns error only if unable to parse addresses */ -static int selinux_parse_skb_ipv4(struct sk_buff *skb, struct avc_audit_data *ad) +static int selinux_parse_skb_ipv4(struct sk_buff *skb, + struct avc_audit_data *ad, u8 *proto) { int offset, ihlen, ret = -EINVAL; struct iphdr _iph, *ih; @@ -2907,6 +2908,9 @@ static int selinux_parse_skb_ipv4(struct sk_buff *skb, struct avc_audit_data *ad ad->u.net.v4info.daddr = ih->daddr; ret = 0; + if (proto) + *proto = ih->protocol; + switch (ih->protocol) { case IPPROTO_TCP: { struct tcphdr _tcph, *th; @@ -2950,7 +2954,8 @@ out: #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) /* Returns error only if unable to parse addresses */ -static int selinux_parse_skb_ipv6(struct sk_buff *skb, struct avc_audit_data *ad) +static int selinux_parse_skb_ipv6(struct sk_buff *skb, + struct avc_audit_data *ad, u8 *proto) { u8 nexthdr; int ret = -EINVAL, offset; @@ -2971,6 +2976,9 @@ static int selinux_parse_skb_ipv6(struct sk_buff *skb, struct avc_audit_data *ad if (offset < 0) goto out; + if (proto) + *proto = nexthdr; + switch (nexthdr) { case IPPROTO_TCP: { struct tcphdr _tcph, *th; @@ -3007,13 +3015,13 @@ out: #endif /* IPV6 */ static int selinux_parse_skb(struct sk_buff *skb, struct avc_audit_data *ad, - char **addrp, int *len, int src) + char **addrp, int *len, int src, u8 *proto) { int ret = 0; switch (ad->u.net.family) { case PF_INET: - ret = selinux_parse_skb_ipv4(skb, ad); + ret = selinux_parse_skb_ipv4(skb, ad, proto); if (ret || !addrp) break; *len = 4; @@ -3023,7 +3031,7 @@ static int selinux_parse_skb(struct sk_buff *skb, struct avc_audit_data *ad, #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) case PF_INET6: - ret = selinux_parse_skb_ipv6(skb, ad); + ret = selinux_parse_skb_ipv6(skb, ad, proto); if (ret || !addrp) break; *len = 16; @@ -3494,7 +3502,7 @@ static int selinux_socket_sock_rcv_skb(struct sock *sk, struct sk_buff *skb) ad.u.net.netif = skb->dev ? skb->dev->name : "[unknown]"; ad.u.net.family = family; - err = selinux_parse_skb(skb, &ad, &addrp, &len, 1); + err = selinux_parse_skb(skb, &ad, &addrp, &len, 1, NULL); if (err) goto out; @@ -3820,6 +3828,7 @@ static unsigned int selinux_ip_postroute_last(unsigned int hooknum, struct avc_audit_data ad; struct net_device *dev = (struct net_device *)out; struct sk_security_struct *sksec; + u8 proto; sk = skb->sk; if (!sk) @@ -3831,7 +3840,7 @@ static unsigned int selinux_ip_postroute_last(unsigned int hooknum, ad.u.net.netif = dev->name; ad.u.net.family = family; - err = selinux_parse_skb(skb, &ad, &addrp, &len, 0); + err = selinux_parse_skb(skb, &ad, &addrp, &len, 0, &proto); if (err) goto out; @@ -3845,7 +3854,7 @@ static unsigned int selinux_ip_postroute_last(unsigned int hooknum, if (err) goto out; - err = selinux_xfrm_postroute_last(sksec->sid, skb, &ad); + err = selinux_xfrm_postroute_last(sksec->sid, skb, &ad, proto); out: return err ? NF_DROP : NF_ACCEPT; } @@ -4764,7 +4773,6 @@ static struct security_operations selinux_ops = { .xfrm_state_delete_security = selinux_xfrm_state_delete, .xfrm_policy_lookup = selinux_xfrm_policy_lookup, .xfrm_state_pol_flow_match = selinux_xfrm_state_pol_flow_match, - .xfrm_flow_state_match = selinux_xfrm_flow_state_match, .xfrm_decode_session = selinux_xfrm_decode_session, #endif diff --git a/security/selinux/include/xfrm.h b/security/selinux/include/xfrm.h index 27502365d70..ebd7246a4be 100644 --- a/security/selinux/include/xfrm.h +++ b/security/selinux/include/xfrm.h @@ -19,9 +19,6 @@ int selinux_xfrm_state_delete(struct xfrm_state *x); int selinux_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir); int selinux_xfrm_state_pol_flow_match(struct xfrm_state *x, struct xfrm_policy *xp, struct flowi *fl); -int selinux_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm, - struct xfrm_policy *xp); - /* * Extract the security blob from the sock (it's actually on the socket) @@ -38,7 +35,7 @@ static inline struct inode_security_struct *get_sock_isec(struct sock *sk) int selinux_xfrm_sock_rcv_skb(u32 sid, struct sk_buff *skb, struct avc_audit_data *ad); int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb, - struct avc_audit_data *ad); + struct avc_audit_data *ad, u8 proto); u32 selinux_socket_getpeer_dgram(struct sk_buff *skb); int selinux_xfrm_decode_session(struct sk_buff *skb, u32 *sid, int ckall); #else @@ -49,7 +46,7 @@ static inline int selinux_xfrm_sock_rcv_skb(u32 isec_sid, struct sk_buff *skb, } static inline int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb, - struct avc_audit_data *ad) + struct avc_audit_data *ad, u8 proto) { return 0; } diff --git a/security/selinux/xfrm.c b/security/selinux/xfrm.c index 8fef74271f2..9b777140068 100644 --- a/security/selinux/xfrm.c +++ b/security/selinux/xfrm.c @@ -115,71 +115,40 @@ int selinux_xfrm_state_pol_flow_match(struct xfrm_state *x, struct xfrm_policy * struct flowi *fl) { u32 state_sid; - u32 pol_sid; - int err; + int rc; - if (xp->security) { - if (!x->security) - /* unlabeled SA and labeled policy can't match */ - return 0; - else - state_sid = x->security->ctx_sid; - pol_sid = xp->security->ctx_sid; - } else + if (!xp->security) if (x->security) /* unlabeled policy and labeled SA can't match */ return 0; else /* unlabeled policy and unlabeled SA match all flows */ return 1; - - err = avc_has_perm(state_sid, pol_sid, SECCLASS_ASSOCIATION, - ASSOCIATION__POLMATCH, - NULL); - - if (err) - return 0; - - err = avc_has_perm(fl->secid, state_sid, SECCLASS_ASSOCIATION, - ASSOCIATION__SENDTO, - NULL)? 0:1; - - return err; -} - -/* - * LSM hook implementation that authorizes that a particular outgoing flow - * can use a given security association. - */ - -int selinux_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm, - struct xfrm_policy *xp) -{ - int rc = 0; - u32 sel_sid = SECINITSID_UNLABELED; - struct xfrm_sec_ctx *ctx; - - if (!xp->security) - if (!xfrm->security) - return 1; - else - return 0; else - if (!xfrm->security) + if (!x->security) + /* unlabeled SA and labeled policy can't match */ return 0; + else + if (!selinux_authorizable_xfrm(x)) + /* Not a SELinux-labeled SA */ + return 0; - /* Context sid is either set to label or ANY_ASSOC */ - if ((ctx = xfrm->security)) { - if (!selinux_authorizable_ctx(ctx)) - return 0; + state_sid = x->security->ctx_sid; - sel_sid = ctx->ctx_sid; - } + if (fl->secid != state_sid) + return 0; - rc = avc_has_perm(fl->secid, sel_sid, SECCLASS_ASSOCIATION, + rc = avc_has_perm(fl->secid, state_sid, SECCLASS_ASSOCIATION, ASSOCIATION__SENDTO, NULL)? 0:1; + /* + * We don't need a separate SA Vs. policy polmatch check + * since the SA is now of the same label as the flow and + * a flow Vs. policy polmatch check had already happened + * in selinux_xfrm_policy_lookup() above. + */ + return rc; } @@ -481,6 +450,13 @@ int selinux_xfrm_sock_rcv_skb(u32 isec_sid, struct sk_buff *skb, } } + /* + * This check even when there's no association involved is + * intended, according to Trent Jaeger, to make sure a + * process can't engage in non-ipsec communication unless + * explicitly allowed by policy. + */ + rc = avc_has_perm(isec_sid, sel_sid, SECCLASS_ASSOCIATION, ASSOCIATION__RECVFROM, ad); @@ -492,10 +468,10 @@ int selinux_xfrm_sock_rcv_skb(u32 isec_sid, struct sk_buff *skb, * If we have no security association, then we need to determine * whether the socket is allowed to send to an unlabelled destination. * If we do have a authorizable security association, then it has already been - * checked in xfrm_policy_lookup hook. + * checked in the selinux_xfrm_state_pol_flow_match hook above. */ int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb, - struct avc_audit_data *ad) + struct avc_audit_data *ad, u8 proto) { struct dst_entry *dst; int rc = 0; @@ -514,6 +490,27 @@ int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb, } } + switch (proto) { + case IPPROTO_AH: + case IPPROTO_ESP: + case IPPROTO_COMP: + /* + * We should have already seen this packet once before + * it underwent xfrm(s). No need to subject it to the + * unlabeled check. + */ + goto out; + default: + break; + } + + /* + * This check even when there's no association involved is + * intended, according to Trent Jaeger, to make sure a + * process can't engage in non-ipsec communication unless + * explicitly allowed by policy. + */ + rc = avc_has_perm(isec_sid, SECINITSID_UNLABELED, SECCLASS_ASSOCIATION, ASSOCIATION__SENDTO, ad); out: -- cgit v1.2.3-70-g09d2 From fcc70d5fdc9b0bd3e99c9dacb8198224af2b4b42 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 8 Nov 2006 22:44:35 -0800 Subject: [BLUETOOTH] lockdep: annotate sk_lock nesting in AF_BLUETOOTH ============================================= [ INFO: possible recursive locking detected ] 2.6.18-1.2726.fc6 #1 --- include/net/sock.h | 8 +++++++- net/bluetooth/l2cap.c | 4 ++-- net/core/sock.c | 6 +++--- 3 files changed, 12 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/net/sock.h b/include/net/sock.h index 9cdbae2a53a..35ffbdd35d3 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -745,7 +745,13 @@ static inline int sk_stream_wmem_schedule(struct sock *sk, int size) */ #define sock_owned_by_user(sk) ((sk)->sk_lock.owner) -extern void FASTCALL(lock_sock(struct sock *sk)); +extern void FASTCALL(lock_sock_nested(struct sock *sk, int subclass)); + +static inline void lock_sock(struct sock *sk) +{ + lock_sock_nested(sk, 0); +} + extern void FASTCALL(release_sock(struct sock *sk)); /* BH context may only use the following locking interface. */ diff --git a/net/bluetooth/l2cap.c b/net/bluetooth/l2cap.c index bbf78e6a7bc..29a8fa4d372 100644 --- a/net/bluetooth/l2cap.c +++ b/net/bluetooth/l2cap.c @@ -770,7 +770,7 @@ static int l2cap_sock_accept(struct socket *sock, struct socket *newsock, int fl long timeo; int err = 0; - lock_sock(sk); + lock_sock_nested(sk, SINGLE_DEPTH_NESTING); if (sk->sk_state != BT_LISTEN) { err = -EBADFD; @@ -792,7 +792,7 @@ static int l2cap_sock_accept(struct socket *sock, struct socket *newsock, int fl release_sock(sk); timeo = schedule_timeout(timeo); - lock_sock(sk); + lock_sock_nested(sk, SINGLE_DEPTH_NESTING); if (sk->sk_state != BT_LISTEN) { err = -EBADFD; diff --git a/net/core/sock.c b/net/core/sock.c index ee6cd2541d3..32ff1c551d6 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1527,7 +1527,7 @@ void sock_init_data(struct socket *sock, struct sock *sk) atomic_set(&sk->sk_refcnt, 1); } -void fastcall lock_sock(struct sock *sk) +void fastcall lock_sock_nested(struct sock *sk, int subclass) { might_sleep(); spin_lock_bh(&sk->sk_lock.slock); @@ -1538,11 +1538,11 @@ void fastcall lock_sock(struct sock *sk) /* * The sk_lock has mutex_lock() semantics here: */ - mutex_acquire(&sk->sk_lock.dep_map, 0, 0, _RET_IP_); + mutex_acquire(&sk->sk_lock.dep_map, subclass, 0, _RET_IP_); local_bh_enable(); } -EXPORT_SYMBOL(lock_sock); +EXPORT_SYMBOL(lock_sock_nested); void fastcall release_sock(struct sock *sk) { -- cgit v1.2.3-70-g09d2 From 776810217ac558351cdcca01c4c6a9474e4a68c2 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 8 Nov 2006 22:46:26 -0800 Subject: [XFRM]: uninline xfrm_selector_match() Six callsites, huge. Signed-off-by: Andrew Morton Signed-off-by: David S. Miller --- include/net/xfrm.h | 36 ++---------------------------------- net/xfrm/xfrm_policy.c | 34 ++++++++++++++++++++++++++++++++++ 2 files changed, 36 insertions(+), 34 deletions(-) (limited to 'include') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index dbcd2922ff3..81c91e8a328 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -506,40 +506,8 @@ __be16 xfrm_flowi_dport(struct flowi *fl) return port; } -static inline int -__xfrm4_selector_match(struct xfrm_selector *sel, struct flowi *fl) -{ - return addr_match(&fl->fl4_dst, &sel->daddr, sel->prefixlen_d) && - addr_match(&fl->fl4_src, &sel->saddr, sel->prefixlen_s) && - !((xfrm_flowi_dport(fl) ^ sel->dport) & sel->dport_mask) && - !((xfrm_flowi_sport(fl) ^ sel->sport) & sel->sport_mask) && - (fl->proto == sel->proto || !sel->proto) && - (fl->oif == sel->ifindex || !sel->ifindex); -} - -static inline int -__xfrm6_selector_match(struct xfrm_selector *sel, struct flowi *fl) -{ - return addr_match(&fl->fl6_dst, &sel->daddr, sel->prefixlen_d) && - addr_match(&fl->fl6_src, &sel->saddr, sel->prefixlen_s) && - !((xfrm_flowi_dport(fl) ^ sel->dport) & sel->dport_mask) && - !((xfrm_flowi_sport(fl) ^ sel->sport) & sel->sport_mask) && - (fl->proto == sel->proto || !sel->proto) && - (fl->oif == sel->ifindex || !sel->ifindex); -} - -static inline int -xfrm_selector_match(struct xfrm_selector *sel, struct flowi *fl, - unsigned short family) -{ - switch (family) { - case AF_INET: - return __xfrm4_selector_match(sel, fl); - case AF_INET6: - return __xfrm6_selector_match(sel, fl); - } - return 0; -} +extern int xfrm_selector_match(struct xfrm_selector *sel, struct flowi *fl, + unsigned short family); #ifdef CONFIG_SECURITY_NETWORK_XFRM /* If neither has a context --> match diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index b88b038530c..e5ea1347a4f 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -50,6 +50,40 @@ static void xfrm_policy_put_afinfo(struct xfrm_policy_afinfo *afinfo); static struct xfrm_policy_afinfo *xfrm_policy_lock_afinfo(unsigned int family); static void xfrm_policy_unlock_afinfo(struct xfrm_policy_afinfo *afinfo); +static inline int +__xfrm4_selector_match(struct xfrm_selector *sel, struct flowi *fl) +{ + return addr_match(&fl->fl4_dst, &sel->daddr, sel->prefixlen_d) && + addr_match(&fl->fl4_src, &sel->saddr, sel->prefixlen_s) && + !((xfrm_flowi_dport(fl) ^ sel->dport) & sel->dport_mask) && + !((xfrm_flowi_sport(fl) ^ sel->sport) & sel->sport_mask) && + (fl->proto == sel->proto || !sel->proto) && + (fl->oif == sel->ifindex || !sel->ifindex); +} + +static inline int +__xfrm6_selector_match(struct xfrm_selector *sel, struct flowi *fl) +{ + return addr_match(&fl->fl6_dst, &sel->daddr, sel->prefixlen_d) && + addr_match(&fl->fl6_src, &sel->saddr, sel->prefixlen_s) && + !((xfrm_flowi_dport(fl) ^ sel->dport) & sel->dport_mask) && + !((xfrm_flowi_sport(fl) ^ sel->sport) & sel->sport_mask) && + (fl->proto == sel->proto || !sel->proto) && + (fl->oif == sel->ifindex || !sel->ifindex); +} + +int xfrm_selector_match(struct xfrm_selector *sel, struct flowi *fl, + unsigned short family) +{ + switch (family) { + case AF_INET: + return __xfrm4_selector_match(sel, fl); + case AF_INET6: + return __xfrm6_selector_match(sel, fl); + } + return 0; +} + int xfrm_register_type(struct xfrm_type *type, unsigned short family) { struct xfrm_policy_afinfo *afinfo = xfrm_policy_lock_afinfo(family); -- cgit v1.2.3-70-g09d2 From 82e91ffef60e6eba9848fe149ce1eecd2b5aef12 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 9 Nov 2006 15:19:14 -0800 Subject: [NET]: Turn nfmark into generic mark nfmark is being used in various subsystems and has become the defacto mark field for all kinds of packets. Therefore it makes sense to rename it to `mark' and remove the dependency on CONFIG_NETFILTER. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/skbuff.h | 4 ++-- net/bridge/netfilter/ebt_mark.c | 8 ++++---- net/bridge/netfilter/ebt_mark_m.c | 4 ++-- net/bridge/netfilter/ebt_ulog.c | 2 +- net/core/skbuff.c | 4 ++-- net/decnet/dn_route.c | 4 ++-- net/ipv4/ip_output.c | 2 +- net/ipv4/ipvs/ip_vs_proto_tcp.c | 2 +- net/ipv4/ipvs/ip_vs_proto_udp.c | 2 +- net/ipv4/netfilter.c | 2 +- net/ipv4/netfilter/ip_queue.c | 2 +- net/ipv4/netfilter/ipt_REJECT.c | 2 +- net/ipv4/netfilter/ipt_ULOG.c | 2 +- net/ipv4/netfilter/iptable_mangle.c | 6 +++--- net/ipv4/route.c | 10 +++++----- net/ipv6/ip6_output.c | 2 +- net/ipv6/netfilter/ip6_queue.c | 2 +- net/ipv6/netfilter/ip6table_mangle.c | 9 ++++----- net/ipv6/route.c | 2 +- net/netfilter/nfnetlink_log.c | 4 ++-- net/netfilter/nfnetlink_queue.c | 8 ++++---- net/netfilter/xt_CONNMARK.c | 10 +++++----- net/netfilter/xt_MARK.c | 12 ++++++------ net/netfilter/xt_mark.c | 2 +- net/sched/Kconfig | 2 +- net/sched/cls_fw.c | 6 +----- net/sched/cls_u32.c | 2 +- net/sched/em_meta.c | 10 +++------- 28 files changed, 59 insertions(+), 68 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 7fc9a3aaa1c..e3ae544b395 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -216,7 +216,7 @@ enum { * @tail: Tail pointer * @end: End pointer * @destructor: Destruct function - * @nfmark: Can be used for communication between hooks + * @mark: Generic packet mark * @nfct: Associated connection, if any * @ipvs_property: skbuff is owned by ipvs * @nfctinfo: Relationship of this skb to the connection @@ -295,7 +295,6 @@ struct sk_buff { #ifdef CONFIG_BRIDGE_NETFILTER struct nf_bridge_info *nf_bridge; #endif - __u32 nfmark; #endif /* CONFIG_NETFILTER */ #ifdef CONFIG_NET_SCHED __u16 tc_index; /* traffic control index */ @@ -310,6 +309,7 @@ struct sk_buff { __u32 secmark; #endif + __u32 mark; /* These elements must be at the end, see alloc_skb() for details. */ unsigned int truesize; diff --git a/net/bridge/netfilter/ebt_mark.c b/net/bridge/netfilter/ebt_mark.c index b54306a934e..2458638561c 100644 --- a/net/bridge/netfilter/ebt_mark.c +++ b/net/bridge/netfilter/ebt_mark.c @@ -25,13 +25,13 @@ static int ebt_target_mark(struct sk_buff **pskb, unsigned int hooknr, int action = info->target & -16; if (action == MARK_SET_VALUE) - (*pskb)->nfmark = info->mark; + (*pskb)->mark = info->mark; else if (action == MARK_OR_VALUE) - (*pskb)->nfmark |= info->mark; + (*pskb)->mark |= info->mark; else if (action == MARK_AND_VALUE) - (*pskb)->nfmark &= info->mark; + (*pskb)->mark &= info->mark; else - (*pskb)->nfmark ^= info->mark; + (*pskb)->mark ^= info->mark; return info->target | -16; } diff --git a/net/bridge/netfilter/ebt_mark_m.c b/net/bridge/netfilter/ebt_mark_m.c index a6413e4b498..025869ee0b6 100644 --- a/net/bridge/netfilter/ebt_mark_m.c +++ b/net/bridge/netfilter/ebt_mark_m.c @@ -19,8 +19,8 @@ static int ebt_filter_mark(const struct sk_buff *skb, struct ebt_mark_m_info *info = (struct ebt_mark_m_info *) data; if (info->bitmask & EBT_MARK_OR) - return !(!!(skb->nfmark & info->mask) ^ info->invert); - return !(((skb->nfmark & info->mask) == info->mark) ^ info->invert); + return !(!!(skb->mark & info->mask) ^ info->invert); + return !(((skb->mark & info->mask) == info->mark) ^ info->invert); } static int ebt_mark_check(const char *tablename, unsigned int hookmask, diff --git a/net/bridge/netfilter/ebt_ulog.c b/net/bridge/netfilter/ebt_ulog.c index 9f950db3b76..c1af68b5a29 100644 --- a/net/bridge/netfilter/ebt_ulog.c +++ b/net/bridge/netfilter/ebt_ulog.c @@ -168,7 +168,7 @@ static void ebt_ulog_packet(unsigned int hooknr, const struct sk_buff *skb, if (ub->qlen == 1) skb_set_timestamp(ub->skb, &pm->stamp); pm->data_len = copy_len; - pm->mark = skb->nfmark; + pm->mark = skb->mark; pm->hook = hooknr; if (uloginfo->prefix != NULL) strcpy(pm->prefix, uloginfo->prefix); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index b8b10635804..b3dea1ef953 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -473,8 +473,8 @@ struct sk_buff *skb_clone(struct sk_buff *skb, gfp_t gfp_mask) #endif C(protocol); n->destructor = NULL; + C(mark); #ifdef CONFIG_NETFILTER - C(nfmark); C(nfct); nf_conntrack_get(skb->nfct); C(nfctinfo); @@ -534,8 +534,8 @@ static void copy_skb_header(struct sk_buff *new, const struct sk_buff *old) new->pkt_type = old->pkt_type; new->tstamp = old->tstamp; new->destructor = NULL; + new->mark = old->mark; #ifdef CONFIG_NETFILTER - new->nfmark = old->nfmark; new->nfct = old->nfct; nf_conntrack_get(old->nfct); new->nfctinfo = old->nfctinfo; diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index 23489f7232d..3482839af28 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -1236,7 +1236,7 @@ static int dn_route_input_slow(struct sk_buff *skb) .saddr = cb->src, .scope = RT_SCOPE_UNIVERSE, #ifdef CONFIG_DECNET_ROUTE_FWMARK - .fwmark = skb->nfmark + .fwmark = skb->mark #endif } }, .iif = skb->dev->ifindex }; @@ -1458,7 +1458,7 @@ int dn_route_input(struct sk_buff *skb) (rt->fl.fld_dst == cb->dst) && (rt->fl.oif == 0) && #ifdef CONFIG_DECNET_ROUTE_FWMARK - (rt->fl.fld_fwmark == skb->nfmark) && + (rt->fl.fld_fwmark == skb->mark) && #endif (rt->fl.iif == cb->iif)) { rt->u.dst.lastuse = jiffies; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index fc195a44fc2..23633bf042b 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -386,6 +386,7 @@ static void ip_copy_metadata(struct sk_buff *to, struct sk_buff *from) dst_release(to->dst); to->dst = dst_clone(from->dst); to->dev = from->dev; + to->mark = from->mark; /* Copy the flags to each fragment. */ IPCB(to)->flags = IPCB(from)->flags; @@ -394,7 +395,6 @@ static void ip_copy_metadata(struct sk_buff *to, struct sk_buff *from) to->tc_index = from->tc_index; #endif #ifdef CONFIG_NETFILTER - to->nfmark = from->nfmark; /* Connection association is same as pre-frag packet */ nf_conntrack_put(to->nfct); to->nfct = from->nfct; diff --git a/net/ipv4/ipvs/ip_vs_proto_tcp.c b/net/ipv4/ipvs/ip_vs_proto_tcp.c index 6ff05c3a32e..7de385267b3 100644 --- a/net/ipv4/ipvs/ip_vs_proto_tcp.c +++ b/net/ipv4/ipvs/ip_vs_proto_tcp.c @@ -84,7 +84,7 @@ tcp_conn_schedule(struct sk_buff *skb, } if (th->syn && - (svc = ip_vs_service_get(skb->nfmark, skb->nh.iph->protocol, + (svc = ip_vs_service_get(skb->mark, skb->nh.iph->protocol, skb->nh.iph->daddr, th->dest))) { if (ip_vs_todrop()) { /* diff --git a/net/ipv4/ipvs/ip_vs_proto_udp.c b/net/ipv4/ipvs/ip_vs_proto_udp.c index 691c8b637b2..452cb9c384b 100644 --- a/net/ipv4/ipvs/ip_vs_proto_udp.c +++ b/net/ipv4/ipvs/ip_vs_proto_udp.c @@ -89,7 +89,7 @@ udp_conn_schedule(struct sk_buff *skb, struct ip_vs_protocol *pp, return 0; } - if ((svc = ip_vs_service_get(skb->nfmark, skb->nh.iph->protocol, + if ((svc = ip_vs_service_get(skb->mark, skb->nh.iph->protocol, skb->nh.iph->daddr, uh->dest))) { if (ip_vs_todrop()) { /* diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c index e2005c6810a..bfc8d753a23 100644 --- a/net/ipv4/netfilter.c +++ b/net/ipv4/netfilter.c @@ -28,7 +28,7 @@ int ip_route_me_harder(struct sk_buff **pskb, unsigned addr_type) fl.nl_u.ip4_u.tos = RT_TOS(iph->tos); fl.oif = (*pskb)->sk ? (*pskb)->sk->sk_bound_dev_if : 0; #ifdef CONFIG_IP_ROUTE_FWMARK - fl.nl_u.ip4_u.fwmark = (*pskb)->nfmark; + fl.nl_u.ip4_u.fwmark = (*pskb)->mark; #endif if (ip_route_output_key(&rt, &fl) != 0) return -1; diff --git a/net/ipv4/netfilter/ip_queue.c b/net/ipv4/netfilter/ip_queue.c index 97556cc2e4e..cd520df4dcf 100644 --- a/net/ipv4/netfilter/ip_queue.c +++ b/net/ipv4/netfilter/ip_queue.c @@ -243,7 +243,7 @@ ipq_build_packet_message(struct ipq_queue_entry *entry, int *errp) pmsg->data_len = data_len; pmsg->timestamp_sec = entry->skb->tstamp.off_sec; pmsg->timestamp_usec = entry->skb->tstamp.off_usec; - pmsg->mark = entry->skb->nfmark; + pmsg->mark = entry->skb->mark; pmsg->hook = entry->info->hook; pmsg->hw_protocol = entry->skb->protocol; diff --git a/net/ipv4/netfilter/ipt_REJECT.c b/net/ipv4/netfilter/ipt_REJECT.c index 264763adc39..f0319e5ee43 100644 --- a/net/ipv4/netfilter/ipt_REJECT.c +++ b/net/ipv4/netfilter/ipt_REJECT.c @@ -76,7 +76,7 @@ static void send_reset(struct sk_buff *oldskb, int hook) /* This packet will not be the same as the other: clear nf fields */ nf_reset(nskb); - nskb->nfmark = 0; + nskb->mark = 0; skb_init_secmark(nskb); tcph = (struct tcphdr *)((u_int32_t*)nskb->nh.iph + nskb->nh.iph->ihl); diff --git a/net/ipv4/netfilter/ipt_ULOG.c b/net/ipv4/netfilter/ipt_ULOG.c index 2b104ea54f4..dbd34783a64 100644 --- a/net/ipv4/netfilter/ipt_ULOG.c +++ b/net/ipv4/netfilter/ipt_ULOG.c @@ -239,7 +239,7 @@ static void ipt_ulog_packet(unsigned int hooknum, pm->data_len = copy_len; pm->timestamp_sec = skb->tstamp.off_sec; pm->timestamp_usec = skb->tstamp.off_usec; - pm->mark = skb->nfmark; + pm->mark = skb->mark; pm->hook = hooknum; if (prefix != NULL) strncpy(pm->prefix, prefix, sizeof(pm->prefix)); diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c index b91f3582359..62d4ccc259c 100644 --- a/net/ipv4/netfilter/iptable_mangle.c +++ b/net/ipv4/netfilter/iptable_mangle.c @@ -132,7 +132,7 @@ ipt_local_hook(unsigned int hook, unsigned int ret; u_int8_t tos; __be32 saddr, daddr; - unsigned long nfmark; + u_int32_t mark; /* root is playing with raw sockets. */ if ((*pskb)->len < sizeof(struct iphdr) @@ -143,7 +143,7 @@ ipt_local_hook(unsigned int hook, } /* Save things which could affect route */ - nfmark = (*pskb)->nfmark; + mark = (*pskb)->mark; saddr = (*pskb)->nh.iph->saddr; daddr = (*pskb)->nh.iph->daddr; tos = (*pskb)->nh.iph->tos; @@ -154,7 +154,7 @@ ipt_local_hook(unsigned int hook, && ((*pskb)->nh.iph->saddr != saddr || (*pskb)->nh.iph->daddr != daddr #ifdef CONFIG_IP_ROUTE_FWMARK - || (*pskb)->nfmark != nfmark + || (*pskb)->mark != mark #endif || (*pskb)->nh.iph->tos != tos)) if (ip_route_me_harder(pskb, RTN_UNSPEC)) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 925ee4dfc32..4de3e38fa1a 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1644,7 +1644,7 @@ static int ip_route_input_mc(struct sk_buff *skb, __be32 daddr, __be32 saddr, rth->rt_dst = daddr; rth->fl.fl4_tos = tos; #ifdef CONFIG_IP_ROUTE_FWMARK - rth->fl.fl4_fwmark= skb->nfmark; + rth->fl.fl4_fwmark= skb->mark; #endif rth->fl.fl4_src = saddr; rth->rt_src = saddr; @@ -1790,7 +1790,7 @@ static inline int __mkroute_input(struct sk_buff *skb, rth->rt_dst = daddr; rth->fl.fl4_tos = tos; #ifdef CONFIG_IP_ROUTE_FWMARK - rth->fl.fl4_fwmark= skb->nfmark; + rth->fl.fl4_fwmark= skb->mark; #endif rth->fl.fl4_src = saddr; rth->rt_src = saddr; @@ -1921,7 +1921,7 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, .tos = tos, .scope = RT_SCOPE_UNIVERSE, #ifdef CONFIG_IP_ROUTE_FWMARK - .fwmark = skb->nfmark + .fwmark = skb->mark #endif } }, .iif = dev->ifindex }; @@ -2035,7 +2035,7 @@ local_input: rth->rt_dst = daddr; rth->fl.fl4_tos = tos; #ifdef CONFIG_IP_ROUTE_FWMARK - rth->fl.fl4_fwmark= skb->nfmark; + rth->fl.fl4_fwmark= skb->mark; #endif rth->fl.fl4_src = saddr; rth->rt_src = saddr; @@ -2114,7 +2114,7 @@ int ip_route_input(struct sk_buff *skb, __be32 daddr, __be32 saddr, rth->fl.iif == iif && rth->fl.oif == 0 && #ifdef CONFIG_IP_ROUTE_FWMARK - rth->fl.fl4_fwmark == skb->nfmark && + rth->fl.fl4_fwmark == skb->mark && #endif rth->fl.fl4_tos == tos) { rth->u.dst.lastuse = jiffies; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 93330685adf..1bde3aca346 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -499,12 +499,12 @@ static void ip6_copy_metadata(struct sk_buff *to, struct sk_buff *from) dst_release(to->dst); to->dst = dst_clone(from->dst); to->dev = from->dev; + to->mark = from->mark; #ifdef CONFIG_NET_SCHED to->tc_index = from->tc_index; #endif #ifdef CONFIG_NETFILTER - to->nfmark = from->nfmark; /* Connection association is same as pre-frag packet */ nf_conntrack_put(to->nfct); to->nfct = from->nfct; diff --git a/net/ipv6/netfilter/ip6_queue.c b/net/ipv6/netfilter/ip6_queue.c index 9fec832ee08..21908c9a10d 100644 --- a/net/ipv6/netfilter/ip6_queue.c +++ b/net/ipv6/netfilter/ip6_queue.c @@ -241,7 +241,7 @@ ipq_build_packet_message(struct ipq_queue_entry *entry, int *errp) pmsg->data_len = data_len; pmsg->timestamp_sec = entry->skb->tstamp.off_sec; pmsg->timestamp_usec = entry->skb->tstamp.off_usec; - pmsg->mark = entry->skb->nfmark; + pmsg->mark = entry->skb->mark; pmsg->hook = entry->info->hook; pmsg->hw_protocol = entry->skb->protocol; diff --git a/net/ipv6/netfilter/ip6table_mangle.c b/net/ipv6/netfilter/ip6table_mangle.c index 386ea260e76..6250e86a6dd 100644 --- a/net/ipv6/netfilter/ip6table_mangle.c +++ b/net/ipv6/netfilter/ip6table_mangle.c @@ -149,11 +149,10 @@ ip6t_local_hook(unsigned int hook, int (*okfn)(struct sk_buff *)) { - unsigned long nfmark; unsigned int ret; struct in6_addr saddr, daddr; u_int8_t hop_limit; - u_int32_t flowlabel; + u_int32_t flowlabel, mark; #if 0 /* root is playing with raw sockets. */ @@ -165,10 +164,10 @@ ip6t_local_hook(unsigned int hook, } #endif - /* save source/dest address, nfmark, hoplimit, flowlabel, priority, */ + /* save source/dest address, mark, hoplimit, flowlabel, priority, */ memcpy(&saddr, &(*pskb)->nh.ipv6h->saddr, sizeof(saddr)); memcpy(&daddr, &(*pskb)->nh.ipv6h->daddr, sizeof(daddr)); - nfmark = (*pskb)->nfmark; + mark = (*pskb)->mark; hop_limit = (*pskb)->nh.ipv6h->hop_limit; /* flowlabel and prio (includes version, which shouldn't change either */ @@ -179,7 +178,7 @@ ip6t_local_hook(unsigned int hook, if (ret != NF_DROP && ret != NF_STOLEN && (memcmp(&(*pskb)->nh.ipv6h->saddr, &saddr, sizeof(saddr)) || memcmp(&(*pskb)->nh.ipv6h->daddr, &daddr, sizeof(daddr)) - || (*pskb)->nfmark != nfmark + || (*pskb)->mark != mark || (*pskb)->nh.ipv6h->hop_limit != hop_limit)) return ip6_route_me_harder(*pskb) == 0 ? ret : NF_DROP; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index e9c1fc5f21b..aaabb1fad1c 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -712,7 +712,7 @@ void ip6_route_input(struct sk_buff *skb) .daddr = iph->daddr, .saddr = iph->saddr, #ifdef CONFIG_IPV6_ROUTE_FWMARK - .fwmark = skb->nfmark, + .fwmark = skb->mark, #endif .flowlabel = (* (__be32 *) iph)&IPV6_FLOWINFO_MASK, }, diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index 856ed0d1997..bd3ffa6f1a6 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -501,8 +501,8 @@ __build_packet_message(struct nfulnl_instance *inst, #endif } - if (skb->nfmark) { - tmp_uint = htonl(skb->nfmark); + if (skb->mark) { + tmp_uint = htonl(skb->mark); NFA_PUT(inst->skb, NFULA_MARK, sizeof(tmp_uint), &tmp_uint); } diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index 4ab7b1416bb..82e4454659b 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -480,8 +480,8 @@ nfqnl_build_packet_message(struct nfqnl_instance *queue, #endif } - if (entskb->nfmark) { - tmp_uint = htonl(entskb->nfmark); + if (entskb->mark) { + tmp_uint = htonl(entskb->mark); NFA_PUT(skb, NFQA_MARK, sizeof(u_int32_t), &tmp_uint); } @@ -834,8 +834,8 @@ nfqnl_recv_verdict(struct sock *ctnl, struct sk_buff *skb, } if (nfqa[NFQA_MARK-1]) - entry->skb->nfmark = ntohl(*(__be32 *) - NFA_DATA(nfqa[NFQA_MARK-1])); + entry->skb->mark = ntohl(*(__be32 *) + NFA_DATA(nfqa[NFQA_MARK-1])); issue_verdict(entry, verdict); instance_put(queue); diff --git a/net/netfilter/xt_CONNMARK.c b/net/netfilter/xt_CONNMARK.c index c01524f817f..67ed5315299 100644 --- a/net/netfilter/xt_CONNMARK.c +++ b/net/netfilter/xt_CONNMARK.c @@ -42,7 +42,7 @@ target(struct sk_buff **pskb, { const struct xt_connmark_target_info *markinfo = targinfo; u_int32_t diff; - u_int32_t nfmark; + u_int32_t mark; u_int32_t newmark; u_int32_t ctinfo; u_int32_t *ctmark = nf_ct_get_mark(*pskb, &ctinfo); @@ -62,7 +62,7 @@ target(struct sk_buff **pskb, break; case XT_CONNMARK_SAVE: newmark = (*ctmark & ~markinfo->mask) | - ((*pskb)->nfmark & markinfo->mask); + ((*pskb)->mark & markinfo->mask); if (*ctmark != newmark) { *ctmark = newmark; #if defined(CONFIG_IP_NF_CONNTRACK) || defined(CONFIG_IP_NF_CONNTRACK_MODULE) @@ -73,10 +73,10 @@ target(struct sk_buff **pskb, } break; case XT_CONNMARK_RESTORE: - nfmark = (*pskb)->nfmark; - diff = (*ctmark ^ nfmark) & markinfo->mask; + mark = (*pskb)->mark; + diff = (*ctmark ^ mark) & markinfo->mask; if (diff != 0) - (*pskb)->nfmark = nfmark ^ diff; + (*pskb)->mark = mark ^ diff; break; } } diff --git a/net/netfilter/xt_MARK.c b/net/netfilter/xt_MARK.c index c6e860a7114..0b48547e8d6 100644 --- a/net/netfilter/xt_MARK.c +++ b/net/netfilter/xt_MARK.c @@ -31,8 +31,8 @@ target_v0(struct sk_buff **pskb, { const struct xt_mark_target_info *markinfo = targinfo; - if((*pskb)->nfmark != markinfo->mark) - (*pskb)->nfmark = markinfo->mark; + if((*pskb)->mark != markinfo->mark) + (*pskb)->mark = markinfo->mark; return XT_CONTINUE; } @@ -54,16 +54,16 @@ target_v1(struct sk_buff **pskb, break; case XT_MARK_AND: - mark = (*pskb)->nfmark & markinfo->mark; + mark = (*pskb)->mark & markinfo->mark; break; case XT_MARK_OR: - mark = (*pskb)->nfmark | markinfo->mark; + mark = (*pskb)->mark | markinfo->mark; break; } - if((*pskb)->nfmark != mark) - (*pskb)->nfmark = mark; + if((*pskb)->mark != mark) + (*pskb)->mark = mark; return XT_CONTINUE; } diff --git a/net/netfilter/xt_mark.c b/net/netfilter/xt_mark.c index 934dddfbcd2..dfa1ee6914c 100644 --- a/net/netfilter/xt_mark.c +++ b/net/netfilter/xt_mark.c @@ -31,7 +31,7 @@ match(const struct sk_buff *skb, { const struct xt_mark_info *info = matchinfo; - return ((skb->nfmark & info->mask) == info->mark) ^ info->invert; + return ((skb->mark & info->mask) == info->mark) ^ info->invert; } static int diff --git a/net/sched/Kconfig b/net/sched/Kconfig index 8298ea9ffe1..b2437092978 100644 --- a/net/sched/Kconfig +++ b/net/sched/Kconfig @@ -320,7 +320,7 @@ config CLS_U32_PERF config CLS_U32_MARK bool "Netfilter marks support" - depends on NET_CLS_U32 && NETFILTER + depends on NET_CLS_U32 ---help--- Say Y here to be able to use netfilter marks as u32 key. diff --git a/net/sched/cls_fw.c b/net/sched/cls_fw.c index e54acc6bccc..f59a2c4aa03 100644 --- a/net/sched/cls_fw.c +++ b/net/sched/cls_fw.c @@ -101,11 +101,7 @@ static int fw_classify(struct sk_buff *skb, struct tcf_proto *tp, struct fw_head *head = (struct fw_head*)tp->root; struct fw_filter *f; int r; -#ifdef CONFIG_NETFILTER - u32 id = skb->nfmark & head->mask; -#else - u32 id = 0; -#endif + u32 id = skb->mark & head->mask; if (head != NULL) { for (f=head->ht[fw_hash(id)]; f; f=f->next) { diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c index 0a6cfa0005b..8b519480199 100644 --- a/net/sched/cls_u32.c +++ b/net/sched/cls_u32.c @@ -143,7 +143,7 @@ next_knode: #endif #ifdef CONFIG_CLS_U32_MARK - if ((skb->nfmark & n->mark.mask) != n->mark.val) { + if ((skb->mark & n->mark.mask) != n->mark.val) { n = n->next; goto next_knode; } else { diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c index 61e3b740ab1..d3ff3503326 100644 --- a/net/sched/em_meta.c +++ b/net/sched/em_meta.c @@ -208,13 +208,9 @@ META_COLLECTOR(int_maclen) * Netfilter **************************************************************************/ -META_COLLECTOR(int_nfmark) +META_COLLECTOR(int_mark) { -#ifdef CONFIG_NETFILTER - dst->value = skb->nfmark; -#else - dst->value = 0; -#endif + dst->value = skb->mark; } /************************************************************************** @@ -490,7 +486,7 @@ static struct meta_ops __meta_ops[TCF_META_TYPE_MAX+1][TCF_META_ID_MAX+1] = { [META_ID(PKTLEN)] = META_FUNC(int_pktlen), [META_ID(DATALEN)] = META_FUNC(int_datalen), [META_ID(MACLEN)] = META_FUNC(int_maclen), - [META_ID(NFMARK)] = META_FUNC(int_nfmark), + [META_ID(NFMARK)] = META_FUNC(int_mark), [META_ID(TCINDEX)] = META_FUNC(int_tcindex), [META_ID(RTCLASSID)] = META_FUNC(int_rtclassid), [META_ID(RTIIF)] = META_FUNC(int_rtiif), -- cgit v1.2.3-70-g09d2 From 47dcf0cb1005e86d0eea780f2984b2e7490f63cd Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 9 Nov 2006 15:20:38 -0800 Subject: [NET]: Rethink mark field in struct flowi Now that all protocols have been made aware of the mark field it can be moved out of the union thus simplyfing its usage. The config options in the IPv4/IPv6/DECnet subsystems to enable respectively disable mark based routing only obfuscate the code with ifdefs, the cost for the additional comparison in the flow key is insignificant, and most distributions have all these options enabled by default anyway. Therefore it makes sense to remove the config options and enable mark based routing by default. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/net/flow.h | 7 +------ include/net/ip_mp_alg.h | 4 +--- net/decnet/Kconfig | 8 -------- net/decnet/dn_route.c | 28 ++++++++-------------------- net/decnet/dn_rules.c | 12 +----------- net/ipv4/Kconfig | 7 ------- net/ipv4/fib_frontend.c | 4 ++-- net/ipv4/fib_rules.c | 12 +----------- net/ipv4/netfilter.c | 4 +--- net/ipv4/netfilter/iptable_mangle.c | 2 -- net/ipv4/route.c | 36 +++++++++--------------------------- net/ipv6/Kconfig | 7 ------- net/ipv6/fib6_rules.c | 12 +----------- net/ipv6/route.c | 4 +--- 14 files changed, 26 insertions(+), 121 deletions(-) (limited to 'include') diff --git a/include/net/flow.h b/include/net/flow.h index 270d4c1761a..ce4b10d8b41 100644 --- a/include/net/flow.h +++ b/include/net/flow.h @@ -13,12 +13,12 @@ struct flowi { int oif; int iif; + __u32 mark; union { struct { __be32 daddr; __be32 saddr; - __u32 fwmark; __u8 tos; __u8 scope; } ip4_u; @@ -26,28 +26,23 @@ struct flowi { struct { struct in6_addr daddr; struct in6_addr saddr; - __u32 fwmark; __be32 flowlabel; } ip6_u; struct { __le16 daddr; __le16 saddr; - __u32 fwmark; __u8 scope; } dn_u; } nl_u; #define fld_dst nl_u.dn_u.daddr #define fld_src nl_u.dn_u.saddr -#define fld_fwmark nl_u.dn_u.fwmark #define fld_scope nl_u.dn_u.scope #define fl6_dst nl_u.ip6_u.daddr #define fl6_src nl_u.ip6_u.saddr -#define fl6_fwmark nl_u.ip6_u.fwmark #define fl6_flowlabel nl_u.ip6_u.flowlabel #define fl4_dst nl_u.ip4_u.daddr #define fl4_src nl_u.ip4_u.saddr -#define fl4_fwmark nl_u.ip4_u.fwmark #define fl4_tos nl_u.ip4_u.tos #define fl4_scope nl_u.ip4_u.scope diff --git a/include/net/ip_mp_alg.h b/include/net/ip_mp_alg.h index beffdd66ad7..25b56571e54 100644 --- a/include/net/ip_mp_alg.h +++ b/include/net/ip_mp_alg.h @@ -88,9 +88,7 @@ static inline int multipath_comparekeys(const struct flowi *flp1, return flp1->fl4_dst == flp2->fl4_dst && flp1->fl4_src == flp2->fl4_src && flp1->oif == flp2->oif && -#ifdef CONFIG_IP_ROUTE_FWMARK - flp1->fl4_fwmark == flp2->fl4_fwmark && -#endif + flp1->mark == flp2->mark && !((flp1->fl4_tos ^ flp2->fl4_tos) & (IPTOS_RT_MASK | RTO_ONLINK)); } diff --git a/net/decnet/Kconfig b/net/decnet/Kconfig index 36e72cb145b..7914fd619c5 100644 --- a/net/decnet/Kconfig +++ b/net/decnet/Kconfig @@ -41,11 +41,3 @@ config DECNET_ROUTER See for more information. -config DECNET_ROUTE_FWMARK - bool "DECnet: use FWMARK value as routing key (EXPERIMENTAL)" - depends on DECNET_ROUTER && NETFILTER - help - If you say Y here, you will be able to specify different routes for - packets with different FWMARK ("firewalling mark") values - (see ipchains(8), "-m" argument). - diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index 3482839af28..f759d6f422e 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -269,9 +269,7 @@ static inline int compare_keys(struct flowi *fl1, struct flowi *fl2) { return ((fl1->nl_u.dn_u.daddr ^ fl2->nl_u.dn_u.daddr) | (fl1->nl_u.dn_u.saddr ^ fl2->nl_u.dn_u.saddr) | -#ifdef CONFIG_DECNET_ROUTE_FWMARK - (fl1->nl_u.dn_u.fwmark ^ fl2->nl_u.dn_u.fwmark) | -#endif + (fl1->mark ^ fl2->mark) | (fl1->nl_u.dn_u.scope ^ fl2->nl_u.dn_u.scope) | (fl1->oif ^ fl2->oif) | (fl1->iif ^ fl2->iif)) == 0; @@ -882,10 +880,8 @@ static int dn_route_output_slow(struct dst_entry **pprt, const struct flowi *old { .daddr = oldflp->fld_dst, .saddr = oldflp->fld_src, .scope = RT_SCOPE_UNIVERSE, -#ifdef CONFIG_DECNET_ROUTE_FWMARK - .fwmark = oldflp->fld_fwmark -#endif } }, + .mark = oldflp->mark, .iif = loopback_dev.ifindex, .oif = oldflp->oif }; struct dn_route *rt = NULL; @@ -903,7 +899,7 @@ static int dn_route_output_slow(struct dst_entry **pprt, const struct flowi *old "dn_route_output_slow: dst=%04x src=%04x mark=%d" " iif=%d oif=%d\n", dn_ntohs(oldflp->fld_dst), dn_ntohs(oldflp->fld_src), - oldflp->fld_fwmark, loopback_dev.ifindex, oldflp->oif); + oldflp->mark, loopback_dev.ifindex, oldflp->oif); /* If we have an output interface, verify its a DECnet device */ if (oldflp->oif) { @@ -1108,9 +1104,7 @@ make_route: rt->fl.fld_dst = oldflp->fld_dst; rt->fl.oif = oldflp->oif; rt->fl.iif = 0; -#ifdef CONFIG_DECNET_ROUTE_FWMARK - rt->fl.fld_fwmark = oldflp->fld_fwmark; -#endif + rt->fl.mark = oldflp->mark; rt->rt_saddr = fl.fld_src; rt->rt_daddr = fl.fld_dst; @@ -1178,9 +1172,7 @@ static int __dn_route_output_key(struct dst_entry **pprt, const struct flowi *fl rt = rcu_dereference(rt->u.rt_next)) { if ((flp->fld_dst == rt->fl.fld_dst) && (flp->fld_src == rt->fl.fld_src) && -#ifdef CONFIG_DECNET_ROUTE_FWMARK - (flp->fld_fwmark == rt->fl.fld_fwmark) && -#endif + (flp->mark == rt->fl.mark) && (rt->fl.iif == 0) && (rt->fl.oif == flp->oif)) { rt->u.dst.lastuse = jiffies; @@ -1235,10 +1227,8 @@ static int dn_route_input_slow(struct sk_buff *skb) { .daddr = cb->dst, .saddr = cb->src, .scope = RT_SCOPE_UNIVERSE, -#ifdef CONFIG_DECNET_ROUTE_FWMARK - .fwmark = skb->mark -#endif } }, + .mark = skb->mark, .iif = skb->dev->ifindex }; struct dn_fib_res res = { .fi = NULL, .type = RTN_UNREACHABLE }; int err = -EINVAL; @@ -1385,7 +1375,7 @@ make_route: rt->fl.fld_dst = cb->dst; rt->fl.oif = 0; rt->fl.iif = in_dev->ifindex; - rt->fl.fld_fwmark = fl.fld_fwmark; + rt->fl.mark = fl.mark; rt->u.dst.flags = DST_HOST; rt->u.dst.neighbour = neigh; @@ -1457,9 +1447,7 @@ int dn_route_input(struct sk_buff *skb) if ((rt->fl.fld_src == cb->src) && (rt->fl.fld_dst == cb->dst) && (rt->fl.oif == 0) && -#ifdef CONFIG_DECNET_ROUTE_FWMARK - (rt->fl.fld_fwmark == skb->mark) && -#endif + (rt->fl.mark == skb->mark) && (rt->fl.iif == cb->iif)) { rt->u.dst.lastuse = jiffies; dst_hold(&rt->u.dst); diff --git a/net/decnet/dn_rules.c b/net/decnet/dn_rules.c index 590e0a72495..a09c09f5572 100644 --- a/net/decnet/dn_rules.c +++ b/net/decnet/dn_rules.c @@ -45,10 +45,8 @@ struct dn_fib_rule __le16 dstmask; __le16 srcmap; u8 flags; -#ifdef CONFIG_DECNET_ROUTE_FWMARK u32 fwmark; u32 fwmask; -#endif }; static struct dn_fib_rule default_rule = { @@ -131,10 +129,8 @@ static int dn_fib_rule_match(struct fib_rule *rule, struct flowi *fl, int flags) ((daddr ^ r->dst) & r->dstmask)) return 0; -#ifdef CONFIG_DECNET_ROUTE_FWMARK - if ((r->fwmark ^ fl->fld_fwmark) & r->fwmask) + if ((r->fwmark ^ fl->mark) & r->fwmask) return 0; -#endif return 1; } @@ -169,7 +165,6 @@ static int dn_fib_rule_configure(struct fib_rule *rule, struct sk_buff *skb, if (tb[FRA_DST]) r->dst = nla_get_u16(tb[FRA_DST]); -#ifdef CONFIG_DECNET_ROUTE_FWMARK if (tb[FRA_FWMARK]) { r->fwmark = nla_get_u32(tb[FRA_FWMARK]); if (r->fwmark) @@ -181,7 +176,6 @@ static int dn_fib_rule_configure(struct fib_rule *rule, struct sk_buff *skb, if (tb[FRA_FWMASK]) r->fwmask = nla_get_u32(tb[FRA_FWMASK]); -#endif r->src_len = frh->src_len; r->srcmask = dnet_make_mask(r->src_len); @@ -203,13 +197,11 @@ static int dn_fib_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh, if (frh->dst_len && (r->dst_len != frh->dst_len)) return 0; -#ifdef CONFIG_DECNET_ROUTE_FWMARK if (tb[FRA_FWMARK] && (r->fwmark != nla_get_u32(tb[FRA_FWMARK]))) return 0; if (tb[FRA_FWMASK] && (r->fwmask != nla_get_u32(tb[FRA_FWMASK]))) return 0; -#endif if (tb[FRA_SRC] && (r->src != nla_get_u16(tb[FRA_SRC]))) return 0; @@ -248,12 +240,10 @@ static int dn_fib_rule_fill(struct fib_rule *rule, struct sk_buff *skb, frh->src_len = r->src_len; frh->tos = 0; -#ifdef CONFIG_DECNET_ROUTE_FWMARK if (r->fwmark) NLA_PUT_U32(skb, FRA_FWMARK, r->fwmark); if (r->fwmask || r->fwmark) NLA_PUT_U32(skb, FRA_FWMASK, r->fwmask); -#endif if (r->dst_len) NLA_PUT_U16(skb, FRA_DST, r->dst); if (r->src_len) diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig index 5572071af73..bc298bcc344 100644 --- a/net/ipv4/Kconfig +++ b/net/ipv4/Kconfig @@ -104,13 +104,6 @@ config IP_MULTIPLE_TABLES If unsure, say N. -config IP_ROUTE_FWMARK - bool "IP: use netfilter MARK value as routing key" - depends on IP_MULTIPLE_TABLES && NETFILTER - help - If you say Y here, you will be able to specify different routes for - packets with different mark values (see iptables(8), MARK target). - config IP_ROUTE_MULTIPATH bool "IP: equal cost multipath" depends on IP_ADVANCED_ROUTER diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index af0190d8b6c..ee8daaebcfa 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -768,8 +768,8 @@ static void nl_fib_lookup(struct fib_result_nl *frn, struct fib_table *tb ) { struct fib_result res; - struct flowi fl = { .nl_u = { .ip4_u = { .daddr = frn->fl_addr, - .fwmark = frn->fl_fwmark, + struct flowi fl = { .mark = frn->fl_fwmark, + .nl_u = { .ip4_u = { .daddr = frn->fl_addr, .tos = frn->fl_tos, .scope = frn->fl_scope } } }; if (tb) { diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c index 0852b9cd065..de8d5dd7099 100644 --- a/net/ipv4/fib_rules.c +++ b/net/ipv4/fib_rules.c @@ -44,10 +44,8 @@ struct fib4_rule __be32 srcmask; __be32 dst; __be32 dstmask; -#ifdef CONFIG_IP_ROUTE_FWMARK u32 fwmark; u32 fwmask; -#endif #ifdef CONFIG_NET_CLS_ROUTE u32 tclassid; #endif @@ -160,10 +158,8 @@ static int fib4_rule_match(struct fib_rule *rule, struct flowi *fl, int flags) if (r->tos && (r->tos != fl->fl4_tos)) return 0; -#ifdef CONFIG_IP_ROUTE_FWMARK - if ((r->fwmark ^ fl->fl4_fwmark) & r->fwmask) + if ((r->fwmark ^ fl->mark) & r->fwmask) return 0; -#endif return 1; } @@ -220,7 +216,6 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb, if (tb[FRA_DST]) rule4->dst = nla_get_be32(tb[FRA_DST]); -#ifdef CONFIG_IP_ROUTE_FWMARK if (tb[FRA_FWMARK]) { rule4->fwmark = nla_get_u32(tb[FRA_FWMARK]); if (rule4->fwmark) @@ -232,7 +227,6 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb, if (tb[FRA_FWMASK]) rule4->fwmask = nla_get_u32(tb[FRA_FWMASK]); -#endif #ifdef CONFIG_NET_CLS_ROUTE if (tb[FRA_FLOW]) @@ -264,13 +258,11 @@ static int fib4_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh, if (frh->tos && (rule4->tos != frh->tos)) return 0; -#ifdef CONFIG_IP_ROUTE_FWMARK if (tb[FRA_FWMARK] && (rule4->fwmark != nla_get_u32(tb[FRA_FWMARK]))) return 0; if (tb[FRA_FWMASK] && (rule4->fwmask != nla_get_u32(tb[FRA_FWMASK]))) return 0; -#endif #ifdef CONFIG_NET_CLS_ROUTE if (tb[FRA_FLOW] && (rule4->tclassid != nla_get_u32(tb[FRA_FLOW]))) @@ -296,13 +288,11 @@ static int fib4_rule_fill(struct fib_rule *rule, struct sk_buff *skb, frh->src_len = rule4->src_len; frh->tos = rule4->tos; -#ifdef CONFIG_IP_ROUTE_FWMARK if (rule4->fwmark) NLA_PUT_U32(skb, FRA_FWMARK, rule4->fwmark); if (rule4->fwmask || rule4->fwmark) NLA_PUT_U32(skb, FRA_FWMASK, rule4->fwmask); -#endif if (rule4->dst_len) NLA_PUT_BE32(skb, FRA_DST, rule4->dst); diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c index bfc8d753a23..e49441ac357 100644 --- a/net/ipv4/netfilter.c +++ b/net/ipv4/netfilter.c @@ -27,9 +27,7 @@ int ip_route_me_harder(struct sk_buff **pskb, unsigned addr_type) fl.nl_u.ip4_u.saddr = iph->saddr; fl.nl_u.ip4_u.tos = RT_TOS(iph->tos); fl.oif = (*pskb)->sk ? (*pskb)->sk->sk_bound_dev_if : 0; -#ifdef CONFIG_IP_ROUTE_FWMARK - fl.nl_u.ip4_u.fwmark = (*pskb)->mark; -#endif + fl.mark = (*pskb)->mark; if (ip_route_output_key(&rt, &fl) != 0) return -1; diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c index 62d4ccc259c..af293988944 100644 --- a/net/ipv4/netfilter/iptable_mangle.c +++ b/net/ipv4/netfilter/iptable_mangle.c @@ -153,9 +153,7 @@ ipt_local_hook(unsigned int hook, if (ret != NF_DROP && ret != NF_STOLEN && ret != NF_QUEUE && ((*pskb)->nh.iph->saddr != saddr || (*pskb)->nh.iph->daddr != daddr -#ifdef CONFIG_IP_ROUTE_FWMARK || (*pskb)->mark != mark -#endif || (*pskb)->nh.iph->tos != tos)) if (ip_route_me_harder(pskb, RTN_UNSPEC)) ret = NF_DROP; diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 4de3e38fa1a..d7152b2b2c6 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -568,9 +568,7 @@ static inline int compare_keys(struct flowi *fl1, struct flowi *fl2) { return ((fl1->nl_u.ip4_u.daddr ^ fl2->nl_u.ip4_u.daddr) | (fl1->nl_u.ip4_u.saddr ^ fl2->nl_u.ip4_u.saddr) | -#ifdef CONFIG_IP_ROUTE_FWMARK - (fl1->nl_u.ip4_u.fwmark ^ fl2->nl_u.ip4_u.fwmark) | -#endif + (fl1->mark ^ fl2->mark) | (*(u16 *)&fl1->nl_u.ip4_u.tos ^ *(u16 *)&fl2->nl_u.ip4_u.tos) | (fl1->oif ^ fl2->oif) | @@ -1643,9 +1641,7 @@ static int ip_route_input_mc(struct sk_buff *skb, __be32 daddr, __be32 saddr, rth->fl.fl4_dst = daddr; rth->rt_dst = daddr; rth->fl.fl4_tos = tos; -#ifdef CONFIG_IP_ROUTE_FWMARK - rth->fl.fl4_fwmark= skb->mark; -#endif + rth->fl.mark = skb->mark; rth->fl.fl4_src = saddr; rth->rt_src = saddr; #ifdef CONFIG_NET_CLS_ROUTE @@ -1789,9 +1785,7 @@ static inline int __mkroute_input(struct sk_buff *skb, rth->fl.fl4_dst = daddr; rth->rt_dst = daddr; rth->fl.fl4_tos = tos; -#ifdef CONFIG_IP_ROUTE_FWMARK - rth->fl.fl4_fwmark= skb->mark; -#endif + rth->fl.mark = skb->mark; rth->fl.fl4_src = saddr; rth->rt_src = saddr; rth->rt_gateway = daddr; @@ -1920,10 +1914,8 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, .saddr = saddr, .tos = tos, .scope = RT_SCOPE_UNIVERSE, -#ifdef CONFIG_IP_ROUTE_FWMARK - .fwmark = skb->mark -#endif } }, + .mark = skb->mark, .iif = dev->ifindex }; unsigned flags = 0; u32 itag = 0; @@ -2034,9 +2026,7 @@ local_input: rth->fl.fl4_dst = daddr; rth->rt_dst = daddr; rth->fl.fl4_tos = tos; -#ifdef CONFIG_IP_ROUTE_FWMARK - rth->fl.fl4_fwmark= skb->mark; -#endif + rth->fl.mark = skb->mark; rth->fl.fl4_src = saddr; rth->rt_src = saddr; #ifdef CONFIG_NET_CLS_ROUTE @@ -2113,9 +2103,7 @@ int ip_route_input(struct sk_buff *skb, __be32 daddr, __be32 saddr, rth->fl.fl4_src == saddr && rth->fl.iif == iif && rth->fl.oif == 0 && -#ifdef CONFIG_IP_ROUTE_FWMARK - rth->fl.fl4_fwmark == skb->mark && -#endif + rth->fl.mark == skb->mark && rth->fl.fl4_tos == tos) { rth->u.dst.lastuse = jiffies; dst_hold(&rth->u.dst); @@ -2239,9 +2227,7 @@ static inline int __mkroute_output(struct rtable **result, rth->fl.fl4_tos = tos; rth->fl.fl4_src = oldflp->fl4_src; rth->fl.oif = oldflp->oif; -#ifdef CONFIG_IP_ROUTE_FWMARK - rth->fl.fl4_fwmark= oldflp->fl4_fwmark; -#endif + rth->fl.mark = oldflp->mark; rth->rt_dst = fl->fl4_dst; rth->rt_src = fl->fl4_src; rth->rt_iif = oldflp->oif ? : dev_out->ifindex; @@ -2385,10 +2371,8 @@ static int ip_route_output_slow(struct rtable **rp, const struct flowi *oldflp) .scope = ((tos & RTO_ONLINK) ? RT_SCOPE_LINK : RT_SCOPE_UNIVERSE), -#ifdef CONFIG_IP_ROUTE_FWMARK - .fwmark = oldflp->fl4_fwmark -#endif } }, + .mark = oldflp->mark, .iif = loopback_dev.ifindex, .oif = oldflp->oif }; struct fib_result res; @@ -2583,9 +2567,7 @@ int __ip_route_output_key(struct rtable **rp, const struct flowi *flp) rth->fl.fl4_src == flp->fl4_src && rth->fl.iif == 0 && rth->fl.oif == flp->oif && -#ifdef CONFIG_IP_ROUTE_FWMARK - rth->fl.fl4_fwmark == flp->fl4_fwmark && -#endif + rth->fl.mark == flp->mark && !((rth->fl.fl4_tos ^ flp->fl4_tos) & (IPTOS_RT_MASK | RTO_ONLINK))) { diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig index 6e48f52e197..deb4101a2a8 100644 --- a/net/ipv6/Kconfig +++ b/net/ipv6/Kconfig @@ -196,10 +196,3 @@ config IPV6_SUBTREES If unsure, say N. -config IPV6_ROUTE_FWMARK - bool "IPv6: use netfilter MARK value as routing key" - depends on IPV6_MULTIPLE_TABLES && NETFILTER - ---help--- - If you say Y here, you will be able to specify different routes for - packets with different mark values (see iptables(8), MARK target). - diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index 1896ecb5289..89bea64eee1 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -25,10 +25,8 @@ struct fib6_rule struct fib_rule common; struct rt6key src; struct rt6key dst; -#ifdef CONFIG_IPV6_ROUTE_FWMARK u32 fwmark; u32 fwmask; -#endif u8 tclass; }; @@ -130,10 +128,8 @@ static int fib6_rule_match(struct fib_rule *rule, struct flowi *fl, int flags) if (r->tclass && r->tclass != ((ntohl(fl->fl6_flowlabel) >> 20) & 0xff)) return 0; -#ifdef CONFIG_IPV6_ROUTE_FWMARK - if ((r->fwmark ^ fl->fl6_fwmark) & r->fwmask) + if ((r->fwmark ^ fl->mark) & r->fwmask) return 0; -#endif return 1; } @@ -177,7 +173,6 @@ static int fib6_rule_configure(struct fib_rule *rule, struct sk_buff *skb, nla_memcpy(&rule6->dst.addr, tb[FRA_DST], sizeof(struct in6_addr)); -#ifdef CONFIG_IPV6_ROUTE_FWMARK if (tb[FRA_FWMARK]) { rule6->fwmark = nla_get_u32(tb[FRA_FWMARK]); if (rule6->fwmark) { @@ -192,7 +187,6 @@ static int fib6_rule_configure(struct fib_rule *rule, struct sk_buff *skb, if (tb[FRA_FWMASK]) rule6->fwmask = nla_get_u32(tb[FRA_FWMASK]); -#endif rule6->src.plen = frh->src_len; rule6->dst.plen = frh->dst_len; @@ -225,13 +219,11 @@ static int fib6_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh, nla_memcmp(tb[FRA_DST], &rule6->dst.addr, sizeof(struct in6_addr))) return 0; -#ifdef CONFIG_IPV6_ROUTE_FWMARK if (tb[FRA_FWMARK] && (rule6->fwmark != nla_get_u32(tb[FRA_FWMARK]))) return 0; if (tb[FRA_FWMASK] && (rule6->fwmask != nla_get_u32(tb[FRA_FWMASK]))) return 0; -#endif return 1; } @@ -254,13 +246,11 @@ static int fib6_rule_fill(struct fib_rule *rule, struct sk_buff *skb, NLA_PUT(skb, FRA_SRC, sizeof(struct in6_addr), &rule6->src.addr); -#ifdef CONFIG_IPV6_ROUTE_FWMARK if (rule6->fwmark) NLA_PUT_U32(skb, FRA_FWMARK, rule6->fwmark); if (rule6->fwmask || rule6->fwmark) NLA_PUT_U32(skb, FRA_FWMASK, rule6->fwmask); -#endif return 0; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index aaabb1fad1c..e3eab152963 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -711,12 +711,10 @@ void ip6_route_input(struct sk_buff *skb) .ip6_u = { .daddr = iph->daddr, .saddr = iph->saddr, -#ifdef CONFIG_IPV6_ROUTE_FWMARK - .fwmark = skb->mark, -#endif .flowlabel = (* (__be32 *) iph)&IPV6_FLOWINFO_MASK, }, }, + .mark = skb->mark, .proto = iph->nexthdr, }; -- cgit v1.2.3-70-g09d2 From 5f300893fdd3b6e30a226c9a848eaa39b99a6431 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 9 Nov 2006 15:21:41 -0800 Subject: [IPV4] nl_fib_lookup: Rename fl_fwmark to fl_mark For the sake of consistency. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/net/ip_fib.h | 2 +- net/ipv4/fib_frontend.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h index 949b932d2f0..36c635ca1aa 100644 --- a/include/net/ip_fib.h +++ b/include/net/ip_fib.h @@ -115,7 +115,7 @@ struct fib_result { struct fib_result_nl { __be32 fl_addr; /* To be looked up*/ - u32 fl_fwmark; + u32 fl_mark; unsigned char fl_tos; unsigned char fl_scope; unsigned char tb_id_in; diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index ee8daaebcfa..14025345cc5 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -768,7 +768,7 @@ static void nl_fib_lookup(struct fib_result_nl *frn, struct fib_table *tb ) { struct fib_result res; - struct flowi fl = { .mark = frn->fl_fwmark, + struct flowi fl = { .mark = frn->fl_mark, .nl_u = { .ip4_u = { .daddr = frn->fl_addr, .tos = frn->fl_tos, .scope = frn->fl_scope } } }; -- cgit v1.2.3-70-g09d2 From b8964ed9fa727109c9084abc807652ebfb681c18 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 9 Nov 2006 15:22:18 -0800 Subject: [NET] rules: Protocol independant mark selector Move mark selector currently implemented per protocol into the protocol independant part. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/fib_rules.h | 2 +- include/net/fib_rules.h | 2 ++ net/core/fib_rules.c | 29 +++++++++++++++++++++++++++++ net/decnet/dn_rules.c | 27 --------------------------- net/ipv4/fib_rules.c | 29 ----------------------------- net/ipv6/fib6_rules.c | 32 -------------------------------- 6 files changed, 32 insertions(+), 89 deletions(-) (limited to 'include') diff --git a/include/linux/fib_rules.h b/include/linux/fib_rules.h index 4418c8d9d47..adcdfbdd14d 100644 --- a/include/linux/fib_rules.h +++ b/include/linux/fib_rules.h @@ -34,7 +34,7 @@ enum FRA_UNUSED3, FRA_UNUSED4, FRA_UNUSED5, - FRA_FWMARK, /* netfilter mark */ + FRA_FWMARK, /* mark */ FRA_FLOW, /* flow/class id */ FRA_UNUSED6, FRA_UNUSED7, diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h index 8e2f473d3e8..68542b565cf 100644 --- a/include/net/fib_rules.h +++ b/include/net/fib_rules.h @@ -13,6 +13,8 @@ struct fib_rule atomic_t refcnt; int ifindex; char ifname[IFNAMSIZ]; + u32 mark; + u32 mark_mask; u32 pref; u32 flags; u32 table; diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index 6b0e63cacd9..da91bf2e615 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -119,6 +119,9 @@ int fib_rules_lookup(struct fib_rules_ops *ops, struct flowi *fl, if (rule->ifindex && (rule->ifindex != fl->iif)) continue; + if ((rule->mark ^ fl->mark) & rule->mark_mask) + continue; + if (!ops->match(rule, fl, flags)) continue; @@ -179,6 +182,18 @@ int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) rule->ifindex = dev->ifindex; } + if (tb[FRA_FWMARK]) { + rule->mark = nla_get_u32(tb[FRA_FWMARK]); + if (rule->mark) + /* compatibility: if the mark value is non-zero all bits + * are compared unless a mask is explicitly specified. + */ + rule->mark_mask = 0xFFFFFFFF; + } + + if (tb[FRA_FWMASK]) + rule->mark_mask = nla_get_u32(tb[FRA_FWMASK]); + rule->action = frh->action; rule->flags = frh->flags; rule->table = frh_get_table(frh, tb); @@ -250,6 +265,14 @@ int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) nla_strcmp(tb[FRA_IFNAME], rule->ifname)) continue; + if (tb[FRA_FWMARK] && + (rule->mark != nla_get_u32(tb[FRA_FWMARK]))) + continue; + + if (tb[FRA_FWMASK] && + (rule->mark_mask != nla_get_u32(tb[FRA_FWMASK]))) + continue; + if (!ops->compare(rule, frh, tb)) continue; @@ -298,6 +321,12 @@ static int fib_nl_fill_rule(struct sk_buff *skb, struct fib_rule *rule, if (rule->pref) NLA_PUT_U32(skb, FRA_PRIORITY, rule->pref); + if (rule->mark) + NLA_PUT_U32(skb, FRA_FWMARK, rule->mark); + + if (rule->mark_mask || rule->mark) + NLA_PUT_U32(skb, FRA_FWMASK, rule->mark_mask); + if (ops->fill(rule, skb, nlh, frh) < 0) goto nla_put_failure; diff --git a/net/decnet/dn_rules.c b/net/decnet/dn_rules.c index a09c09f5572..1f5d23c9668 100644 --- a/net/decnet/dn_rules.c +++ b/net/decnet/dn_rules.c @@ -45,8 +45,6 @@ struct dn_fib_rule __le16 dstmask; __le16 srcmap; u8 flags; - u32 fwmark; - u32 fwmask; }; static struct dn_fib_rule default_rule = { @@ -129,9 +127,6 @@ static int dn_fib_rule_match(struct fib_rule *rule, struct flowi *fl, int flags) ((daddr ^ r->dst) & r->dstmask)) return 0; - if ((r->fwmark ^ fl->mark) & r->fwmask) - return 0; - return 1; } @@ -165,18 +160,6 @@ static int dn_fib_rule_configure(struct fib_rule *rule, struct sk_buff *skb, if (tb[FRA_DST]) r->dst = nla_get_u16(tb[FRA_DST]); - if (tb[FRA_FWMARK]) { - r->fwmark = nla_get_u32(tb[FRA_FWMARK]); - if (r->fwmark) - /* compatibility: if the mark value is non-zero all bits - * are compared unless a mask is explicitly specified. - */ - r->fwmask = 0xFFFFFFFF; - } - - if (tb[FRA_FWMASK]) - r->fwmask = nla_get_u32(tb[FRA_FWMASK]); - r->src_len = frh->src_len; r->srcmask = dnet_make_mask(r->src_len); r->dst_len = frh->dst_len; @@ -197,12 +180,6 @@ static int dn_fib_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh, if (frh->dst_len && (r->dst_len != frh->dst_len)) return 0; - if (tb[FRA_FWMARK] && (r->fwmark != nla_get_u32(tb[FRA_FWMARK]))) - return 0; - - if (tb[FRA_FWMASK] && (r->fwmask != nla_get_u32(tb[FRA_FWMASK]))) - return 0; - if (tb[FRA_SRC] && (r->src != nla_get_u16(tb[FRA_SRC]))) return 0; @@ -240,10 +217,6 @@ static int dn_fib_rule_fill(struct fib_rule *rule, struct sk_buff *skb, frh->src_len = r->src_len; frh->tos = 0; - if (r->fwmark) - NLA_PUT_U32(skb, FRA_FWMARK, r->fwmark); - if (r->fwmask || r->fwmark) - NLA_PUT_U32(skb, FRA_FWMASK, r->fwmask); if (r->dst_len) NLA_PUT_U16(skb, FRA_DST, r->dst); if (r->src_len) diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c index de8d5dd7099..d2a190a35d6 100644 --- a/net/ipv4/fib_rules.c +++ b/net/ipv4/fib_rules.c @@ -44,8 +44,6 @@ struct fib4_rule __be32 srcmask; __be32 dst; __be32 dstmask; - u32 fwmark; - u32 fwmask; #ifdef CONFIG_NET_CLS_ROUTE u32 tclassid; #endif @@ -158,9 +156,6 @@ static int fib4_rule_match(struct fib_rule *rule, struct flowi *fl, int flags) if (r->tos && (r->tos != fl->fl4_tos)) return 0; - if ((r->fwmark ^ fl->mark) & r->fwmask) - return 0; - return 1; } @@ -216,18 +211,6 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb, if (tb[FRA_DST]) rule4->dst = nla_get_be32(tb[FRA_DST]); - if (tb[FRA_FWMARK]) { - rule4->fwmark = nla_get_u32(tb[FRA_FWMARK]); - if (rule4->fwmark) - /* compatibility: if the mark value is non-zero all bits - * are compared unless a mask is explicitly specified. - */ - rule4->fwmask = 0xFFFFFFFF; - } - - if (tb[FRA_FWMASK]) - rule4->fwmask = nla_get_u32(tb[FRA_FWMASK]); - #ifdef CONFIG_NET_CLS_ROUTE if (tb[FRA_FLOW]) rule4->tclassid = nla_get_u32(tb[FRA_FLOW]); @@ -258,12 +241,6 @@ static int fib4_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh, if (frh->tos && (rule4->tos != frh->tos)) return 0; - if (tb[FRA_FWMARK] && (rule4->fwmark != nla_get_u32(tb[FRA_FWMARK]))) - return 0; - - if (tb[FRA_FWMASK] && (rule4->fwmask != nla_get_u32(tb[FRA_FWMASK]))) - return 0; - #ifdef CONFIG_NET_CLS_ROUTE if (tb[FRA_FLOW] && (rule4->tclassid != nla_get_u32(tb[FRA_FLOW]))) return 0; @@ -288,12 +265,6 @@ static int fib4_rule_fill(struct fib_rule *rule, struct sk_buff *skb, frh->src_len = rule4->src_len; frh->tos = rule4->tos; - if (rule4->fwmark) - NLA_PUT_U32(skb, FRA_FWMARK, rule4->fwmark); - - if (rule4->fwmask || rule4->fwmark) - NLA_PUT_U32(skb, FRA_FWMASK, rule4->fwmask); - if (rule4->dst_len) NLA_PUT_BE32(skb, FRA_DST, rule4->dst); diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index 89bea64eee1..26374cbe2bc 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -25,8 +25,6 @@ struct fib6_rule struct fib_rule common; struct rt6key src; struct rt6key dst; - u32 fwmark; - u32 fwmask; u8 tclass; }; @@ -128,9 +126,6 @@ static int fib6_rule_match(struct fib_rule *rule, struct flowi *fl, int flags) if (r->tclass && r->tclass != ((ntohl(fl->fl6_flowlabel) >> 20) & 0xff)) return 0; - if ((r->fwmark ^ fl->mark) & r->fwmask) - return 0; - return 1; } @@ -173,21 +168,6 @@ static int fib6_rule_configure(struct fib_rule *rule, struct sk_buff *skb, nla_memcpy(&rule6->dst.addr, tb[FRA_DST], sizeof(struct in6_addr)); - if (tb[FRA_FWMARK]) { - rule6->fwmark = nla_get_u32(tb[FRA_FWMARK]); - if (rule6->fwmark) { - /* - * if the mark value is non-zero, - * all bits are compared by default - * unless a mask is explicitly specified. - */ - rule6->fwmask = 0xFFFFFFFF; - } - } - - if (tb[FRA_FWMASK]) - rule6->fwmask = nla_get_u32(tb[FRA_FWMASK]); - rule6->src.plen = frh->src_len; rule6->dst.plen = frh->dst_len; rule6->tclass = frh->tos; @@ -219,12 +199,6 @@ static int fib6_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh, nla_memcmp(tb[FRA_DST], &rule6->dst.addr, sizeof(struct in6_addr))) return 0; - if (tb[FRA_FWMARK] && (rule6->fwmark != nla_get_u32(tb[FRA_FWMARK]))) - return 0; - - if (tb[FRA_FWMASK] && (rule6->fwmask != nla_get_u32(tb[FRA_FWMASK]))) - return 0; - return 1; } @@ -246,12 +220,6 @@ static int fib6_rule_fill(struct fib_rule *rule, struct sk_buff *skb, NLA_PUT(skb, FRA_SRC, sizeof(struct in6_addr), &rule6->src.addr); - if (rule6->fwmark) - NLA_PUT_U32(skb, FRA_FWMARK, rule6->fwmark); - - if (rule6->fwmask || rule6->fwmark) - NLA_PUT_U32(skb, FRA_FWMASK, rule6->fwmask); - return 0; nla_put_failure: -- cgit v1.2.3-70-g09d2 From 1f6c9557e8206757c91b5737bb8dbd5b1ae3a773 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 9 Nov 2006 15:22:48 -0800 Subject: [NET] rules: Share common attribute validation policy Move the attribute policy for the non-specific attributes into net/fib_rules.h and include it in the respective protocols. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/net/fib_rules.h | 7 +++++++ net/decnet/dn_rules.c | 6 +----- net/ipv4/fib_rules.c | 6 +----- net/ipv6/fib6_rules.c | 6 +----- 4 files changed, 10 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h index 68542b565cf..e4ba781d289 100644 --- a/include/net/fib_rules.h +++ b/include/net/fib_rules.h @@ -59,6 +59,13 @@ struct fib_rules_ops struct module *owner; }; +#define FRA_GENERIC_POLICY \ + [FRA_IFNAME] = { .type = NLA_STRING, .len = IFNAMSIZ - 1 }, \ + [FRA_PRIORITY] = { .type = NLA_U32 }, \ + [FRA_FWMARK] = { .type = NLA_U32 }, \ + [FRA_FWMASK] = { .type = NLA_U32 }, \ + [FRA_TABLE] = { .type = NLA_U32 } + static inline void fib_rule_get(struct fib_rule *rule) { atomic_inc(&rule->refcnt); diff --git a/net/decnet/dn_rules.c b/net/decnet/dn_rules.c index 1f5d23c9668..e32d0c3d5a9 100644 --- a/net/decnet/dn_rules.c +++ b/net/decnet/dn_rules.c @@ -108,13 +108,9 @@ errout: } static struct nla_policy dn_fib_rule_policy[FRA_MAX+1] __read_mostly = { - [FRA_IFNAME] = { .type = NLA_STRING, .len = IFNAMSIZ - 1 }, - [FRA_PRIORITY] = { .type = NLA_U32 }, + FRA_GENERIC_POLICY, [FRA_SRC] = { .type = NLA_U16 }, [FRA_DST] = { .type = NLA_U16 }, - [FRA_FWMARK] = { .type = NLA_U32 }, - [FRA_FWMASK] = { .type = NLA_U32 }, - [FRA_TABLE] = { .type = NLA_U32 }, }; static int dn_fib_rule_match(struct fib_rule *rule, struct flowi *fl, int flags) diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c index d2a190a35d6..fd4a8cd4c06 100644 --- a/net/ipv4/fib_rules.c +++ b/net/ipv4/fib_rules.c @@ -170,14 +170,10 @@ static struct fib_table *fib_empty_table(void) } static struct nla_policy fib4_rule_policy[FRA_MAX+1] __read_mostly = { - [FRA_IFNAME] = { .type = NLA_STRING, .len = IFNAMSIZ - 1 }, - [FRA_PRIORITY] = { .type = NLA_U32 }, + FRA_GENERIC_POLICY, [FRA_SRC] = { .type = NLA_U32 }, [FRA_DST] = { .type = NLA_U32 }, - [FRA_FWMARK] = { .type = NLA_U32 }, - [FRA_FWMASK] = { .type = NLA_U32 }, [FRA_FLOW] = { .type = NLA_U32 }, - [FRA_TABLE] = { .type = NLA_U32 }, }; static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb, diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index 26374cbe2bc..8377477b2c1 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -130,13 +130,9 @@ static int fib6_rule_match(struct fib_rule *rule, struct flowi *fl, int flags) } static struct nla_policy fib6_rule_policy[FRA_MAX+1] __read_mostly = { - [FRA_IFNAME] = { .type = NLA_STRING, .len = IFNAMSIZ - 1 }, - [FRA_PRIORITY] = { .type = NLA_U32 }, + FRA_GENERIC_POLICY, [FRA_SRC] = { .len = sizeof(struct in6_addr) }, [FRA_DST] = { .len = sizeof(struct in6_addr) }, - [FRA_FWMARK] = { .type = NLA_U32 }, - [FRA_FWMASK] = { .type = NLA_U32 }, - [FRA_TABLE] = { .type = NLA_U32 }, }; static int fib6_rule_configure(struct fib_rule *rule, struct sk_buff *skb, -- cgit v1.2.3-70-g09d2 From 3dfbcc411e461db51a1ac1aa1c6ebe2c5a0275a0 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 9 Nov 2006 15:23:20 -0800 Subject: [NET] rules: Add support to invert selectors Introduces a new flag FIB_RULE_INVERT causing rules to apply if the specified selector doesn't match. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/fib_rules.h | 1 + net/core/fib_rules.c | 24 +++++++++++++++++------- 2 files changed, 18 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/include/linux/fib_rules.h b/include/linux/fib_rules.h index adcdfbdd14d..8270aac2aa5 100644 --- a/include/linux/fib_rules.h +++ b/include/linux/fib_rules.h @@ -6,6 +6,7 @@ /* rule is permanent, and cannot be deleted */ #define FIB_RULE_PERMANENT 1 +#define FIB_RULE_INVERT 2 struct fib_rule_hdr { diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index da91bf2e615..4148e274a20 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -107,6 +107,22 @@ out: EXPORT_SYMBOL_GPL(fib_rules_unregister); +static int fib_rule_match(struct fib_rule *rule, struct fib_rules_ops *ops, + struct flowi *fl, int flags) +{ + int ret = 0; + + if (rule->ifindex && (rule->ifindex != fl->iif)) + goto out; + + if ((rule->mark ^ fl->mark) & rule->mark_mask) + goto out; + + ret = ops->match(rule, fl, flags); +out: + return (rule->flags & FIB_RULE_INVERT) ? !ret : ret; +} + int fib_rules_lookup(struct fib_rules_ops *ops, struct flowi *fl, int flags, struct fib_lookup_arg *arg) { @@ -116,13 +132,7 @@ int fib_rules_lookup(struct fib_rules_ops *ops, struct flowi *fl, rcu_read_lock(); list_for_each_entry_rcu(rule, ops->rules_list, list) { - if (rule->ifindex && (rule->ifindex != fl->iif)) - continue; - - if ((rule->mark ^ fl->mark) & rule->mark_mask) - continue; - - if (!ops->match(rule, fl, flags)) + if (!fib_rule_match(rule, ops, fl, flags)) continue; err = ops->action(rule, fl, flags, arg); -- cgit v1.2.3-70-g09d2 From 72a3effaf633bcae9034b7e176bdbd78d64a71db Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 16 Nov 2006 02:30:37 -0800 Subject: [NET]: Size listen hash tables using backlog hint We currently allocate a fixed size (TCP_SYNQ_HSIZE=512) slots hash table for each LISTEN socket, regardless of various parameters (listen backlog for example) On x86_64, this means order-1 allocations (might fail), even for 'small' sockets, expecting few connections. On the contrary, a huge server wanting a backlog of 50000 is slowed down a bit because of this fixed limit. This patch makes the sizing of listen hash table a dynamic parameter, depending of : - net.core.somaxconn tunable (default is 128) - net.ipv4.tcp_max_syn_backlog tunable (default : 256, 1024 or 128) - backlog value given by user application (2nd parameter of listen()) For large allocations (bigger than PAGE_SIZE), we use vmalloc() instead of kmalloc(). We still limit memory allocation with the two existing tunables (somaxconn & tcp_max_syn_backlog). So for standard setups, this patch actually reduce RAM usage. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/request_sock.h | 8 ++++---- include/net/tcp.h | 1 - net/core/request_sock.c | 35 +++++++++++++++++++++++++---------- net/dccp/ipv4.c | 2 +- net/dccp/proto.c | 6 +++--- net/ipv4/af_inet.c | 2 +- net/ipv4/inet_connection_sock.c | 2 +- net/ipv4/tcp_ipv4.c | 6 +++--- net/ipv6/tcp_ipv6.c | 2 +- 9 files changed, 39 insertions(+), 25 deletions(-) (limited to 'include') diff --git a/include/net/request_sock.h b/include/net/request_sock.h index f743a941a4f..b5b023e79e5 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -28,8 +28,8 @@ struct proto; struct request_sock_ops { int family; - kmem_cache_t *slab; int obj_size; + kmem_cache_t *slab; int (*rtx_syn_ack)(struct sock *sk, struct request_sock *req, struct dst_entry *dst); @@ -51,13 +51,13 @@ struct request_sock { u32 rcv_wnd; /* rcv_wnd offered first time */ u32 ts_recent; unsigned long expires; - struct request_sock_ops *rsk_ops; + const struct request_sock_ops *rsk_ops; struct sock *sk; u32 secid; u32 peer_secid; }; -static inline struct request_sock *reqsk_alloc(struct request_sock_ops *ops) +static inline struct request_sock *reqsk_alloc(const struct request_sock_ops *ops) { struct request_sock *req = kmem_cache_alloc(ops->slab, SLAB_ATOMIC); @@ -121,7 +121,7 @@ struct request_sock_queue { }; extern int reqsk_queue_alloc(struct request_sock_queue *queue, - const int nr_table_entries); + unsigned int nr_table_entries); static inline struct listen_sock *reqsk_queue_yank_listen_sk(struct request_sock_queue *queue) { diff --git a/include/net/tcp.h b/include/net/tcp.h index 7a093d0aa0f..246916c2321 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -138,7 +138,6 @@ extern void tcp_time_wait(struct sock *sk, int state, int timeo); #define MAX_TCP_SYNCNT 127 #define TCP_SYNQ_INTERVAL (HZ/5) /* Period of SYNACK timer */ -#define TCP_SYNQ_HSIZE 512 /* Size of SYNACK hash table */ #define TCP_PAWS_24DAYS (60 * 60 * 24 * 24) #define TCP_PAWS_MSL 60 /* Per-host timestamps are invalidated diff --git a/net/core/request_sock.c b/net/core/request_sock.c index 79ebd75fbe4..5f0818d815e 100644 --- a/net/core/request_sock.c +++ b/net/core/request_sock.c @@ -15,6 +15,7 @@ #include #include #include +#include #include @@ -29,22 +30,31 @@ * it is absolutely not enough even at 100conn/sec. 256 cures most * of problems. This value is adjusted to 128 for very small machines * (<=32Mb of memory) and to 1024 on normal or better ones (>=256Mb). - * Further increasing requires to change hash table size. + * Note : Dont forget somaxconn that may limit backlog too. */ int sysctl_max_syn_backlog = 256; int reqsk_queue_alloc(struct request_sock_queue *queue, - const int nr_table_entries) + unsigned int nr_table_entries) { - const int lopt_size = sizeof(struct listen_sock) + - nr_table_entries * sizeof(struct request_sock *); - struct listen_sock *lopt = kzalloc(lopt_size, GFP_KERNEL); - + size_t lopt_size = sizeof(struct listen_sock); + struct listen_sock *lopt; + + nr_table_entries = min_t(u32, nr_table_entries, sysctl_max_syn_backlog); + nr_table_entries = max_t(u32, nr_table_entries, 8); + nr_table_entries = roundup_pow_of_two(nr_table_entries + 1); + lopt_size += nr_table_entries * sizeof(struct request_sock *); + if (lopt_size > PAGE_SIZE) + lopt = __vmalloc(lopt_size, + GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO, + PAGE_KERNEL); + else + lopt = kzalloc(lopt_size, GFP_KERNEL); if (lopt == NULL) return -ENOMEM; - for (lopt->max_qlen_log = 6; - (1 << lopt->max_qlen_log) < sysctl_max_syn_backlog; + for (lopt->max_qlen_log = 3; + (1 << lopt->max_qlen_log) < nr_table_entries; lopt->max_qlen_log++); get_random_bytes(&lopt->hash_rnd, sizeof(lopt->hash_rnd)); @@ -65,9 +75,11 @@ void reqsk_queue_destroy(struct request_sock_queue *queue) { /* make all the listen_opt local to us */ struct listen_sock *lopt = reqsk_queue_yank_listen_sk(queue); + size_t lopt_size = sizeof(struct listen_sock) + + lopt->nr_table_entries * sizeof(struct request_sock *); if (lopt->qlen != 0) { - int i; + unsigned int i; for (i = 0; i < lopt->nr_table_entries; i++) { struct request_sock *req; @@ -81,7 +93,10 @@ void reqsk_queue_destroy(struct request_sock_queue *queue) } BUG_TRAP(lopt->qlen == 0); - kfree(lopt); + if (lopt_size > PAGE_SIZE) + vfree(lopt); + else + kfree(lopt); } EXPORT_SYMBOL(reqsk_queue_destroy); diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index e08e7688a26..0a5d68dbb41 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -1022,7 +1022,7 @@ static void dccp_v4_reqsk_destructor(struct request_sock *req) kfree(inet_rsk(req)->opt); } -static struct request_sock_ops dccp_request_sock_ops = { +static struct request_sock_ops dccp_request_sock_ops _read_mostly = { .family = PF_INET, .obj_size = sizeof(struct dccp_request_sock), .rtx_syn_ack = dccp_v4_send_response, diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 72cbdcfc2c6..047d170a363 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -262,12 +262,12 @@ int dccp_destroy_sock(struct sock *sk) EXPORT_SYMBOL_GPL(dccp_destroy_sock); -static inline int dccp_listen_start(struct sock *sk) +static inline int dccp_listen_start(struct sock *sk, int backlog) { struct dccp_sock *dp = dccp_sk(sk); dp->dccps_role = DCCP_ROLE_LISTEN; - return inet_csk_listen_start(sk, TCP_SYNQ_HSIZE); + return inet_csk_listen_start(sk, backlog); } int dccp_disconnect(struct sock *sk, int flags) @@ -788,7 +788,7 @@ int inet_dccp_listen(struct socket *sock, int backlog) * FIXME: here it probably should be sk->sk_prot->listen_start * see tcp_listen_start */ - err = dccp_listen_start(sk); + err = dccp_listen_start(sk, backlog); if (err) goto out; } diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index edcf0932ac6..4a81d54a756 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -204,7 +204,7 @@ int inet_listen(struct socket *sock, int backlog) * we can only allow the backlog to be adjusted. */ if (old_state != TCP_LISTEN) { - err = inet_csk_listen_start(sk, TCP_SYNQ_HSIZE); + err = inet_csk_listen_start(sk, backlog); if (err) goto out; } diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 96bbe2a0aa1..9d68837888d 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -343,7 +343,7 @@ struct dst_entry* inet_csk_route_req(struct sock *sk, EXPORT_SYMBOL_GPL(inet_csk_route_req); static inline u32 inet_synq_hash(const __be32 raddr, const __be16 rport, - const u32 rnd, const u16 synq_hsize) + const u32 rnd, const u32 synq_hsize) { return jhash_2words((__force u32)raddr, (__force u32)rport, rnd) & (synq_hsize - 1); } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 22ef8bd2662..5fbf96552ca 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -715,7 +715,7 @@ static struct ip_options *tcp_v4_save_options(struct sock *sk, return dopt; } -struct request_sock_ops tcp_request_sock_ops = { +struct request_sock_ops tcp_request_sock_ops __read_mostly = { .family = PF_INET, .obj_size = sizeof(struct tcp_request_sock), .rtx_syn_ack = tcp_v4_send_synack, @@ -1385,7 +1385,7 @@ static void *listening_get_next(struct seq_file *seq, void *cur) if (st->state == TCP_SEQ_STATE_OPENREQ) { struct request_sock *req = cur; - icsk = inet_csk(st->syn_wait_sk); + icsk = inet_csk(st->syn_wait_sk); req = req->dl_next; while (1) { while (req) { @@ -1395,7 +1395,7 @@ static void *listening_get_next(struct seq_file *seq, void *cur) } req = req->dl_next; } - if (++st->sbucket >= TCP_SYNQ_HSIZE) + if (++st->sbucket >= icsk->icsk_accept_queue.listen_opt->nr_table_entries) break; get_req: req = icsk->icsk_accept_queue.listen_opt->syn_table[st->sbucket]; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index eb6d145ecfd..1a3c46c139f 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -526,7 +526,7 @@ static void tcp_v6_reqsk_destructor(struct request_sock *req) kfree_skb(inet6_rsk(req)->pktopts); } -static struct request_sock_ops tcp6_request_sock_ops = { +static struct request_sock_ops tcp6_request_sock_ops _read_mostly = { .family = AF_INET6, .obj_size = sizeof(struct tcp6_request_sock), .rtx_syn_ack = tcp_v6_send_synack, -- cgit v1.2.3-70-g09d2 From b68dbcab1dc70938fa5516d0ee82c0bf94e9a768 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Thu, 9 Nov 2006 16:29:57 -0800 Subject: [SCTP]: Fix warning An alternate solution would be to make the digest a pointer, allocate it in sctp_endpoint_init() and free it in sctp_endpoint_destroy(). I guess I should have originally done it this way... CC [M] net/sctp/sm_make_chunk.o net/sctp/sm_make_chunk.c: In function 'sctp_unpack_cookie': net/sctp/sm_make_chunk.c:1358: warning: initialization discards qualifiers from pointer target type The reason is that sctp_unpack_cookie() takes a const struct sctp_endpoint and modifies the digest in it (digest being embedded in the struct, not a pointer). Make digest a pointer to fix this warning. Signed-off-by: Vlad Yasevich Acked-by: Sridhar Samudrala Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 2 +- net/sctp/endpointola.c | 7 +++++++ 2 files changed, 8 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index c6d93bb0dcd..5596f5d97ed 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1270,7 +1270,7 @@ struct sctp_endpoint { * this here so we pre-allocate this once and can re-use * on every receive. */ - __u8 digest[SCTP_SIGNATURE_SIZE]; + __u8 *digest; /* sendbuf acct. policy. */ __u32 sndbuf_policy; diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c index 9b6b394b66f..090f2b2a0ca 100644 --- a/net/sctp/endpointola.c +++ b/net/sctp/endpointola.c @@ -72,6 +72,10 @@ static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep, { memset(ep, 0, sizeof(struct sctp_endpoint)); + ep->digest = kzalloc(SCTP_SIGNATURE_SIZE, gfp); + if (!ep->digest) + return NULL; + /* Initialize the base structure. */ /* What type of endpoint are we? */ ep->base.type = SCTP_EP_TYPE_SOCKET; @@ -182,6 +186,9 @@ static void sctp_endpoint_destroy(struct sctp_endpoint *ep) /* Free up the HMAC transform. */ crypto_free_hash(sctp_sk(ep->base.sk)->hmac); + /* Free the digest buffer */ + kfree(ep->digest); + /* Cleanup. */ sctp_inq_free(&ep->base.inqueue); sctp_bind_addr_free(&ep->base.bind_addr); -- cgit v1.2.3-70-g09d2 From 3ff825b28d3345ef381eceae22bf9d92231f23dc Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 9 Nov 2006 16:32:06 -0800 Subject: [TCP]: Add tcp_available_congestion_control sysctl. Create /proc/sys/net/ipv4/tcp_available_congestion_control that reflects currently available TCP choices. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- Documentation/networking/ip-sysctl.txt | 6 ++++++ include/linux/sysctl.h | 1 + include/net/tcp.h | 4 ++++ net/ipv4/sysctl_net_ipv4.c | 24 ++++++++++++++++++++++++ net/ipv4/tcp_cong.c | 16 ++++++++++++++++ 5 files changed, 51 insertions(+) (limited to 'include') diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index fd3c0c01235..db428085658 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -351,10 +351,16 @@ tcp_frto - BOOLEAN where packet loss is typically due to random radio interference rather than intermediate router congestion. +tcp_available_congestion_control - STRING + Shows the available congestion control choices that are registered. + More congestion control algorithms may be available as modules, + but not loaded. + tcp_congestion_control - STRING Set the congestion control algorithm to be used for new connections. The algorithm "reno" is always available, but additional choices may be available based on kernel configuration. + Default is set as part of kernel configuration. somaxconn - INTEGER Limit of socket listen() backlog, known in userspace as SOMAXCONN. diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index d98562f1df7..28a48279654 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -426,6 +426,7 @@ enum NET_CIPSOV4_CACHE_BUCKET_SIZE=119, NET_CIPSOV4_RBM_OPTFMT=120, NET_CIPSOV4_RBM_STRICTVALID=121, + NET_TCP_AVAIL_CONG_CONTROL=122, }; enum { diff --git a/include/net/tcp.h b/include/net/tcp.h index 246916c2321..6af4baf5b76 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -620,6 +620,9 @@ enum tcp_ca_event { * Interface for adding new TCP congestion control handlers */ #define TCP_CA_NAME_MAX 16 +#define TCP_CA_MAX 128 +#define TCP_CA_BUF_MAX (TCP_CA_NAME_MAX*TCP_CA_MAX) + struct tcp_congestion_ops { struct list_head list; @@ -659,6 +662,7 @@ extern void tcp_init_congestion_control(struct sock *sk); extern void tcp_cleanup_congestion_control(struct sock *sk); extern int tcp_set_default_congestion_control(const char *name); extern void tcp_get_default_congestion_control(char *name); +extern void tcp_get_available_congestion_control(char *buf, size_t len); extern int tcp_set_congestion_control(struct sock *sk, const char *name); extern void tcp_slow_start(struct tcp_sock *tp); diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 15061b31441..2e770f45d82 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -129,6 +129,23 @@ static int sysctl_tcp_congestion_control(ctl_table *table, int __user *name, return ret; } +static int proc_tcp_available_congestion_control(ctl_table *ctl, + int write, struct file * filp, + void __user *buffer, size_t *lenp, + loff_t *ppos) +{ + ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX, }; + int ret; + + tbl.data = kmalloc(tbl.maxlen, GFP_USER); + if (!tbl.data) + return -ENOMEM; + tcp_get_available_congestion_control(tbl.data, TCP_CA_BUF_MAX); + ret = proc_dostring(&tbl, write, filp, buffer, lenp, ppos); + kfree(tbl.data); + return ret; +} + ctl_table ipv4_table[] = { { .ctl_name = NET_IPV4_TCP_TIMESTAMPS, @@ -731,6 +748,13 @@ ctl_table ipv4_table[] = { .proc_handler = &proc_dointvec, }, #endif /* CONFIG_NETLABEL */ + { + .ctl_name = NET_TCP_AVAIL_CONG_CONTROL, + .procname = "tcp_available_congestion_control", + .maxlen = TCP_CA_BUF_MAX, + .mode = 0444, + .proc_handler = &proc_tcp_available_congestion_control, + }, { .ctl_name = 0 } }; diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index 1e2982f4acd..d846d7b95e1 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -139,6 +139,22 @@ static int __init tcp_congestion_default(void) late_initcall(tcp_congestion_default); +/* Build string with list of available congestion control values */ +void tcp_get_available_congestion_control(char *buf, size_t maxlen) +{ + struct tcp_congestion_ops *ca; + size_t offs = 0; + + rcu_read_lock(); + list_for_each_entry_rcu(ca, &tcp_cong_list, list) { + offs += snprintf(buf + offs, maxlen - offs, + "%s%s", + offs == 0 ? "" : " ", ca->name); + + } + rcu_read_unlock(); +} + /* Get current default congestion control */ void tcp_get_default_congestion_control(char *name) { -- cgit v1.2.3-70-g09d2 From ce7bc3bf15cbf5dc5a5587ccb6b04c5b4dde4336 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 9 Nov 2006 16:35:15 -0800 Subject: [TCP]: Restrict congestion control choices. Allow normal users to only choose among a restricted set of congestion control choices. The default is reno and what ever has been configured as default. But the policy can be changed by administrator at any time. For example, to allow any choice: cp /proc/sys/net/ipv4/tcp_available_congestion_control \ /proc/sys/net/ipv4/tcp_allowed_congestion_control Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- Documentation/networking/ip-sysctl.txt | 6 ++++ include/linux/sysctl.h | 1 + include/net/tcp.h | 3 ++ net/ipv4/sysctl_net_ipv4.c | 52 ++++++++++++++++++++++++++++ net/ipv4/tcp_cong.c | 63 ++++++++++++++++++++++++++++++++++ 5 files changed, 125 insertions(+) (limited to 'include') diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index db428085658..bbcc8deda17 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -351,6 +351,12 @@ tcp_frto - BOOLEAN where packet loss is typically due to random radio interference rather than intermediate router congestion. +tcp_allowed_congestion_control - STRING + Show/set the congestion control choices available to non-privileged + processes. The list is a subset of those listed in + tcp_available_congestion_control. + Default is "reno" and the default setting (tcp_congestion_control). + tcp_available_congestion_control - STRING Shows the available congestion control choices that are registered. More congestion control algorithms may be available as modules, diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 28a48279654..0725441621d 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -427,6 +427,7 @@ enum NET_CIPSOV4_RBM_OPTFMT=120, NET_CIPSOV4_RBM_STRICTVALID=121, NET_TCP_AVAIL_CONG_CONTROL=122, + NET_TCP_ALLOWED_CONG_CONTROL=123, }; enum { diff --git a/include/net/tcp.h b/include/net/tcp.h index 6af4baf5b76..e1a5d29d0a1 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -625,6 +625,7 @@ enum tcp_ca_event { struct tcp_congestion_ops { struct list_head list; + int non_restricted; /* initialize private data (optional) */ void (*init)(struct sock *sk); @@ -663,6 +664,8 @@ extern void tcp_cleanup_congestion_control(struct sock *sk); extern int tcp_set_default_congestion_control(const char *name); extern void tcp_get_default_congestion_control(char *name); extern void tcp_get_available_congestion_control(char *buf, size_t len); +extern void tcp_get_allowed_congestion_control(char *buf, size_t len); +extern int tcp_set_allowed_congestion_control(char *allowed); extern int tcp_set_congestion_control(struct sock *sk, const char *name); extern void tcp_slow_start(struct tcp_sock *tp); diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 2e770f45d82..dfcf47f10f8 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -146,6 +146,50 @@ static int proc_tcp_available_congestion_control(ctl_table *ctl, return ret; } +static int proc_allowed_congestion_control(ctl_table *ctl, + int write, struct file * filp, + void __user *buffer, size_t *lenp, + loff_t *ppos) +{ + ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX }; + int ret; + + tbl.data = kmalloc(tbl.maxlen, GFP_USER); + if (!tbl.data) + return -ENOMEM; + + tcp_get_allowed_congestion_control(tbl.data, tbl.maxlen); + ret = proc_dostring(&tbl, write, filp, buffer, lenp, ppos); + if (write && ret == 0) + ret = tcp_set_allowed_congestion_control(tbl.data); + kfree(tbl.data); + return ret; +} + +static int strategy_allowed_congestion_control(ctl_table *table, int __user *name, + int nlen, void __user *oldval, + size_t __user *oldlenp, + void __user *newval, size_t newlen, + void **context) +{ + ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX }; + int ret; + + tbl.data = kmalloc(tbl.maxlen, GFP_USER); + if (!tbl.data) + return -ENOMEM; + + tcp_get_available_congestion_control(tbl.data, tbl.maxlen); + ret = sysctl_string(&tbl, name, nlen, oldval, oldlenp, newval, newlen, + context); + if (ret == 0 && newval && newlen) + ret = tcp_set_allowed_congestion_control(tbl.data); + kfree(tbl.data); + + return ret; + +} + ctl_table ipv4_table[] = { { .ctl_name = NET_IPV4_TCP_TIMESTAMPS, @@ -755,6 +799,14 @@ ctl_table ipv4_table[] = { .mode = 0444, .proc_handler = &proc_tcp_available_congestion_control, }, + { + .ctl_name = NET_TCP_ALLOWED_CONG_CONTROL, + .procname = "tcp_allowed_congestion_control", + .maxlen = TCP_CA_BUF_MAX, + .mode = 0644, + .proc_handler = &proc_allowed_congestion_control, + .strategy = &strategy_allowed_congestion_control, + }, { .ctl_name = 0 } }; diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index d846d7b95e1..343d6197c92 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -123,6 +123,7 @@ int tcp_set_default_congestion_control(const char *name) #endif if (ca) { + ca->non_restricted = 1; /* default is always allowed */ list_move(&ca->list, &tcp_cong_list); ret = 0; } @@ -168,6 +169,64 @@ void tcp_get_default_congestion_control(char *name) rcu_read_unlock(); } +/* Built list of non-restricted congestion control values */ +void tcp_get_allowed_congestion_control(char *buf, size_t maxlen) +{ + struct tcp_congestion_ops *ca; + size_t offs = 0; + + *buf = '\0'; + rcu_read_lock(); + list_for_each_entry_rcu(ca, &tcp_cong_list, list) { + if (!ca->non_restricted) + continue; + offs += snprintf(buf + offs, maxlen - offs, + "%s%s", + offs == 0 ? "" : " ", ca->name); + + } + rcu_read_unlock(); +} + +/* Change list of non-restricted congestion control */ +int tcp_set_allowed_congestion_control(char *val) +{ + struct tcp_congestion_ops *ca; + char *clone, *name; + int ret = 0; + + clone = kstrdup(val, GFP_USER); + if (!clone) + return -ENOMEM; + + spin_lock(&tcp_cong_list_lock); + /* pass 1 check for bad entries */ + while ((name = strsep(&clone, " ")) && *name) { + ca = tcp_ca_find(name); + if (!ca) { + ret = -ENOENT; + goto out; + } + } + + /* pass 2 clear */ + list_for_each_entry_rcu(ca, &tcp_cong_list, list) + ca->non_restricted = 0; + + /* pass 3 mark as allowed */ + while ((name = strsep(&val, " ")) && *name) { + ca = tcp_ca_find(name); + WARN_ON(!ca); + if (ca) + ca->non_restricted = 1; + } +out: + spin_unlock(&tcp_cong_list_lock); + + return ret; +} + + /* Change congestion control for socket */ int tcp_set_congestion_control(struct sock *sk, const char *name) { @@ -183,6 +242,9 @@ int tcp_set_congestion_control(struct sock *sk, const char *name) if (!ca) err = -ENOENT; + else if (!(ca->non_restricted || capable(CAP_NET_ADMIN))) + err = -EPERM; + else if (!try_module_get(ca->owner)) err = -EBUSY; @@ -284,6 +346,7 @@ EXPORT_SYMBOL_GPL(tcp_reno_min_cwnd); struct tcp_congestion_ops tcp_reno = { .name = "reno", + .non_restricted = 1, .owner = THIS_MODULE, .ssthresh = tcp_reno_ssthresh, .cong_avoid = tcp_reno_cong_avoid, -- cgit v1.2.3-70-g09d2 From 931731123a103cfb3f70ac4b7abfc71d94ba1f03 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 9 Nov 2006 19:58:25 -0800 Subject: [TCP]: Don't set SKB owner in tcp_transmit_skb(). The data itself is already charged to the SKB, doing the skb_set_owner_w() just generates a lot of noise and extra atomics we don't really need. Lmbench improvements on lat_tcp are minimal: before: TCP latency using localhost: 23.2701 microseconds TCP latency using localhost: 23.1994 microseconds TCP latency using localhost: 23.2257 microseconds after: TCP latency using localhost: 22.8380 microseconds TCP latency using localhost: 22.9465 microseconds TCP latency using localhost: 22.8462 microseconds Signed-off-by: David S. Miller --- include/net/inet6_connection_sock.h | 2 +- include/net/inet_connection_sock.h | 3 ++- include/net/ip.h | 2 +- net/dccp/output.c | 4 ++-- net/ipv4/ip_output.c | 3 +-- net/ipv4/tcp_output.c | 3 +-- net/ipv6/inet6_connection_sock.c | 3 +-- net/sctp/protocol.c | 2 +- 8 files changed, 10 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/include/net/inet6_connection_sock.h b/include/net/inet6_connection_sock.h index f13ddc2543b..16aa96a6a53 100644 --- a/include/net/inet6_connection_sock.h +++ b/include/net/inet6_connection_sock.h @@ -38,5 +38,5 @@ extern void inet6_csk_reqsk_queue_hash_add(struct sock *sk, extern void inet6_csk_addr2sockaddr(struct sock *sk, struct sockaddr *uaddr); -extern int inet6_csk_xmit(struct sk_buff *skb, int ipfragok); +extern int inet6_csk_xmit(struct sk_buff *skb, struct sock *sk, int ipfragok); #endif /* _INET6_CONNECTION_SOCK_H */ diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index 0bcf9f237e1..fd3f1d8cdc4 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -36,7 +36,8 @@ struct tcp_congestion_ops; * (i.e. things that depend on the address family) */ struct inet_connection_sock_af_ops { - int (*queue_xmit)(struct sk_buff *skb, int ipfragok); + int (*queue_xmit)(struct sk_buff *skb, struct sock *sk, + int ipfragok); void (*send_check)(struct sock *sk, int len, struct sk_buff *skb); int (*rebuild_header)(struct sock *sk); diff --git a/include/net/ip.h b/include/net/ip.h index b6d95e55340..949fa868362 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -97,7 +97,7 @@ extern int ip_mc_output(struct sk_buff *skb); extern int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)); extern int ip_do_nat(struct sk_buff *skb); extern void ip_send_check(struct iphdr *ip); -extern int ip_queue_xmit(struct sk_buff *skb, int ipfragok); +extern int ip_queue_xmit(struct sk_buff *skb, struct sock *sk, int ipfragok); extern void ip_init(void); extern int ip_append_data(struct sock *sk, int getfrag(void *from, char *to, int offset, int len, diff --git a/net/dccp/output.c b/net/dccp/output.c index 7102e3aed4c..2cc4f4b2a9d 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -125,7 +125,7 @@ static int dccp_transmit_skb(struct sock *sk, struct sk_buff *skb) DCCP_INC_STATS(DCCP_MIB_OUTSEGS); memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); - err = icsk->icsk_af_ops->queue_xmit(skb, 0); + err = icsk->icsk_af_ops->queue_xmit(skb, sk, 0); if (err <= 0) return err; @@ -426,7 +426,7 @@ int dccp_send_reset(struct sock *sk, enum dccp_reset_codes code) code); if (skb != NULL) { memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); - err = inet_csk(sk)->icsk_af_ops->queue_xmit(skb, 0); + err = inet_csk(sk)->icsk_af_ops->queue_xmit(skb, sk, 0); if (err == NET_XMIT_CN) err = 0; } diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 23633bf042b..def32d8d3b0 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -288,9 +288,8 @@ int ip_output(struct sk_buff *skb) !(IPCB(skb)->flags & IPSKB_REROUTED)); } -int ip_queue_xmit(struct sk_buff *skb, int ipfragok) +int ip_queue_xmit(struct sk_buff *skb, struct sock *sk, int ipfragok) { - struct sock *sk = skb->sk; struct inet_sock *inet = inet_sk(sk); struct ip_options *opt = inet->opt; struct rtable *rt; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index ca406157724..f63e99aac2d 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -426,7 +426,6 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, th = (struct tcphdr *) skb_push(skb, tcp_header_size); skb->h.th = th; - skb_set_owner_w(skb, sk); /* Build TCP header and checksum it. */ th->source = inet->sport; @@ -479,7 +478,7 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, if (after(tcb->end_seq, tp->snd_nxt) || tcb->seq == tcb->end_seq) TCP_INC_STATS(TCP_MIB_OUTSEGS); - err = icsk->icsk_af_ops->queue_xmit(skb, 0); + err = icsk->icsk_af_ops->queue_xmit(skb, sk, 0); if (likely(err <= 0)) return err; diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c index e48fb4c53c4..c598962eaa4 100644 --- a/net/ipv6/inet6_connection_sock.c +++ b/net/ipv6/inet6_connection_sock.c @@ -139,9 +139,8 @@ void inet6_csk_addr2sockaddr(struct sock *sk, struct sockaddr * uaddr) EXPORT_SYMBOL_GPL(inet6_csk_addr2sockaddr); -int inet6_csk_xmit(struct sk_buff *skb, int ipfragok) +int inet6_csk_xmit(struct sk_buff *skb, struct sock *sk, int ipfragok) { - struct sock *sk = skb->sk; struct inet_sock *inet = inet_sk(sk); struct ipv6_pinfo *np = inet6_sk(sk); struct flowi fl; diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 5b4f82fd98f..c107d45f6e6 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -808,7 +808,7 @@ static inline int sctp_v4_xmit(struct sk_buff *skb, NIPQUAD(((struct rtable *)skb->dst)->rt_dst)); SCTP_INC_STATS(SCTP_MIB_OUTSCTPPACKS); - return ip_queue_xmit(skb, ipfragok); + return ip_queue_xmit(skb, skb->sk, ipfragok); } static struct sctp_af sctp_ipv4_specific; -- cgit v1.2.3-70-g09d2 From 9b42078ed6edfe04e9dc9a59b946ad912aeef717 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Fri, 10 Nov 2006 11:22:32 -0200 Subject: [DCCP]: Combine allocating & zeroing header space on skb This is a code simplification: it combines three often recurring operations into one inline function, * allocate `len' bytes header space in skb * fill these `len' bytes with zeroes * cast the start of this header space as dccp_hdr Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/dccp.h | 7 +++++++ net/dccp/ipv4.c | 9 ++------- net/dccp/ipv6.c | 8 ++------ net/dccp/output.c | 14 +++----------- 4 files changed, 14 insertions(+), 24 deletions(-) (limited to 'include') diff --git a/include/linux/dccp.h b/include/linux/dccp.h index 53553c99cad..90d04ffddae 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -256,6 +256,13 @@ static inline struct dccp_hdr *dccp_hdr(const struct sk_buff *skb) return (struct dccp_hdr *)skb->h.raw; } +static inline struct dccp_hdr *dccp_zeroed_hdr(struct sk_buff *skb, int headlen) +{ + skb->h.raw = skb_push(skb, headlen); + memset(skb->h.raw, 0, headlen); + return dccp_hdr(skb); +} + static inline struct dccp_hdr_ext *dccp_hdrx(const struct sk_buff *skb) { return (struct dccp_hdr_ext *)(skb->h.raw + sizeof(struct dccp_hdr)); diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index de64e6c7f93..ce8eed32dbe 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -212,12 +212,9 @@ static void dccp_v4_reqsk_send_ack(struct sk_buff *rxskb, /* Reserve space for headers. */ skb_reserve(skb, dccp_v4_ctl_socket->sk->sk_prot->max_header); - skb->dst = dst_clone(rxskb->dst); - skb->h.raw = skb_push(skb, dccp_hdr_ack_len); - dh = dccp_hdr(skb); - memset(dh, 0, dccp_hdr_ack_len); + dh = dccp_zeroed_hdr(skb, dccp_hdr_ack_len); /* Build DCCP header and checksum it. */ dh->dccph_type = DCCP_PKT_ACK; @@ -720,9 +717,7 @@ static void dccp_v4_ctl_send_reset(struct sk_buff *rxskb) skb_reserve(skb, dccp_v4_ctl_socket->sk->sk_prot->max_header); skb->dst = dst_clone(dst); - skb->h.raw = skb_push(skb, dccp_hdr_reset_len); - dh = dccp_hdr(skb); - memset(dh, 0, dccp_hdr_reset_len); + dh = dccp_zeroed_hdr(skb, dccp_hdr_reset_len); /* Build DCCP header and checksum it. */ dh->dccph_type = DCCP_PKT_RESET; diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 6f1c2ad8860..116bddb64b8 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -537,9 +537,7 @@ static void dccp_v6_ctl_send_reset(struct sk_buff *rxskb) skb_reserve(skb, dccp_v6_ctl_socket->sk->sk_prot->max_header); - skb->h.raw = skb_push(skb, dccp_hdr_reset_len); - dh = dccp_hdr(skb); - memset(dh, 0, dccp_hdr_reset_len); + dh = dccp_zeroed_hdr(skb, dccp_hdr_reset_len); /* Swap the send and the receive. */ dh->dccph_type = DCCP_PKT_RESET; @@ -601,9 +599,7 @@ static void dccp_v6_reqsk_send_ack(struct sk_buff *rxskb, skb_reserve(skb, dccp_v6_ctl_socket->sk->sk_prot->max_header); - skb->h.raw = skb_push(skb, dccp_hdr_ack_len); - dh = dccp_hdr(skb); - memset(dh, 0, dccp_hdr_ack_len); + dh = dccp_zeroed_hdr(skb, dccp_hdr_ack_len); /* Build DCCP header and checksum it. */ dh->dccph_type = DCCP_PKT_ACK; diff --git a/net/dccp/output.c b/net/dccp/output.c index 2cc4f4b2a9d..1ae2248557c 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -88,11 +88,9 @@ static int dccp_transmit_skb(struct sock *sk, struct sk_buff *skb) return -EPROTO; } - skb->h.raw = skb_push(skb, dccp_header_size); - dh = dccp_hdr(skb); /* Build DCCP header and checksum it. */ - memset(dh, 0, dccp_header_size); + dh = dccp_zeroed_hdr(skb, dccp_header_size); dh->dccph_type = dcb->dccpd_type; dh->dccph_sport = inet->sport; dh->dccph_dport = inet->dport; @@ -340,10 +338,7 @@ struct sk_buff *dccp_make_response(struct sock *sk, struct dst_entry *dst, return NULL; } - skb->h.raw = skb_push(skb, dccp_header_size); - - dh = dccp_hdr(skb); - memset(dh, 0, dccp_header_size); + dh = dccp_zeroed_hdr(skb, dccp_header_size); dh->dccph_sport = inet_sk(sk)->sport; dh->dccph_dport = inet_rsk(req)->rmt_port; @@ -392,10 +387,7 @@ static struct sk_buff *dccp_make_reset(struct sock *sk, struct dst_entry *dst, return NULL; } - skb->h.raw = skb_push(skb, dccp_header_size); - - dh = dccp_hdr(skb); - memset(dh, 0, dccp_header_size); + dh = dccp_zeroed_hdr(skb, dccp_header_size); dh->dccph_sport = inet_sk(sk)->sport; dh->dccph_dport = inet_sk(sk)->dport; -- cgit v1.2.3-70-g09d2 From 7a3025b1b3a0173be5de6ced18754b909da27b38 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Fri, 13 Oct 2006 16:17:25 +0900 Subject: [IPV6]: Introduce ip6_dst_idev() to get inet6_dev{} stored in dst_entry{}. Otherwise, we will see a lot of casts... Signed-off-by: YOSHIFUJI Hideaki --- include/net/ip6_fib.h | 5 +++++ net/ipv6/addrconf.c | 2 +- net/ipv6/ip6_input.c | 2 +- 3 files changed, 7 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index e4438de3bd6..f9cde44f93b 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -107,6 +107,11 @@ struct rt6_info u8 rt6i_protocol; }; +static inline struct inet6_dev *ip6_dst_idev(struct dst_entry *dst) +{ + return ((struct rt6_info *)dst)->rt6i_idev; +} + struct fib6_walker_t { struct fib6_walker_t *prev, *next; diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index b312a5f7a75..6a98f68348c 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1164,7 +1164,7 @@ record_it: int ipv6_get_saddr(struct dst_entry *dst, struct in6_addr *daddr, struct in6_addr *saddr) { - return ipv6_dev_get_saddr(dst ? ((struct rt6_info *)dst)->rt6i_idev->dev : NULL, daddr, saddr); + return ipv6_dev_get_saddr(dst ? ip6_dst_idev(dst)->dev : NULL, daddr, saddr); } diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c index 6b8e6d76a58..ebf54ae90a0 100644 --- a/net/ipv6/ip6_input.c +++ b/net/ipv6/ip6_input.c @@ -84,7 +84,7 @@ int ipv6_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt * arrived via the sending interface (ethX), because of the * nature of scoping architecture. --yoshfuji */ - IP6CB(skb)->iif = skb->dst ? ((struct rt6_info *)skb->dst)->rt6i_idev->dev->ifindex : dev->ifindex; + IP6CB(skb)->iif = skb->dst ? ip6_dst_idev(skb->dst)->dev->ifindex : dev->ifindex; if (unlikely(!pskb_may_pull(skb, sizeof(*hdr)))) goto err; -- cgit v1.2.3-70-g09d2 From a11d206d0f88e092419877c7f706cafb5e1c2e57 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Sat, 4 Nov 2006 20:11:37 +0900 Subject: [IPV6]: Per-interface statistics support. For IP MIB (RFC4293). Signed-off-by: YOSHIFUJI Hideaki --- include/net/if_inet6.h | 1 + include/net/ipv6.h | 21 ++++++++++++--- net/ipv6/exthdrs.c | 57 ++++++++++++++++++++++++++-------------- net/ipv6/icmp.c | 3 ++- net/ipv6/ip6_input.c | 40 +++++++++++++++++++--------- net/ipv6/ip6_output.c | 71 +++++++++++++++++++++++++++++--------------------- net/ipv6/mcast.c | 20 +++++++++----- net/ipv6/ndisc.c | 8 +++--- net/ipv6/netfilter.c | 2 +- net/ipv6/proc.c | 7 +++++ net/ipv6/raw.c | 4 +-- net/ipv6/reassembly.c | 65 +++++++++++++++++++++++++++------------------ net/ipv6/route.c | 4 +-- 13 files changed, 195 insertions(+), 108 deletions(-) (limited to 'include') diff --git a/include/net/if_inet6.h b/include/net/if_inet6.h index 34489c13c11..3ec7d07346d 100644 --- a/include/net/if_inet6.h +++ b/include/net/if_inet6.h @@ -152,6 +152,7 @@ struct ifacaddr6 struct ipv6_devstat { struct proc_dir_entry *proc_dir_entry; + DEFINE_SNMP_STAT(struct ipstats_mib, ipv6); DEFINE_SNMP_STAT(struct icmpv6_mib, icmpv6); }; diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 0b8c9b990ac..3c266ad99a0 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -113,9 +113,24 @@ extern int sysctl_mld_max_msf; /* MIBs */ DECLARE_SNMP_STAT(struct ipstats_mib, ipv6_statistics); -#define IP6_INC_STATS(field) SNMP_INC_STATS(ipv6_statistics, field) -#define IP6_INC_STATS_BH(field) SNMP_INC_STATS_BH(ipv6_statistics, field) -#define IP6_INC_STATS_USER(field) SNMP_INC_STATS_USER(ipv6_statistics, field) +#define IP6_INC_STATS(idev,field) ({ \ + struct inet6_dev *_idev = (idev); \ + if (likely(_idev != NULL)) \ + SNMP_INC_STATS(_idev->stats.ipv6, field); \ + SNMP_INC_STATS(ipv6_statistics, field); \ +}) +#define IP6_INC_STATS_BH(idev,field) ({ \ + struct inet6_dev *_idev = (idev); \ + if (likely(_idev != NULL)) \ + SNMP_INC_STATS_BH(_idev->stats.ipv6, field); \ + SNMP_INC_STATS_BH(ipv6_statistics, field); \ +}) +#define IP6_INC_STATS_USER(idev,field) ({ \ + struct inet6_dev *_idev = (idev); \ + if (likely(_idev != NULL)) \ + SNMP_INC_STATS_USER(_idev->stats.ipv6, field); \ + SNMP_INC_STATS_USER(ipv6_statistics, field); \ +}) DECLARE_SNMP_STAT(struct icmpv6_mib, icmpv6_statistics); #define ICMP6_INC_STATS(idev, field) ({ \ struct inet6_dev *_idev = (idev); \ diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index 88c96b10684..27829cc4ce8 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -284,10 +284,12 @@ static int ipv6_destopt_rcv(struct sk_buff **skbp) #ifdef CONFIG_IPV6_MIP6 __u16 dstbuf; #endif + struct dst_entry *dst; if (!pskb_may_pull(skb, (skb->h.raw-skb->data)+8) || !pskb_may_pull(skb, (skb->h.raw-skb->data)+((skb->h.raw[1]+1)<<3))) { - IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), + IPSTATS_MIB_INHDRERRORS); kfree_skb(skb); return -1; } @@ -298,7 +300,9 @@ static int ipv6_destopt_rcv(struct sk_buff **skbp) dstbuf = opt->dst1; #endif + dst = dst_clone(skb->dst); if (ip6_parse_tlv(tlvprocdestopt_lst, skbp)) { + dst_release(dst); skb = *skbp; skb->h.raw += ((skb->h.raw[1]+1)<<3); opt = IP6CB(skb); @@ -310,7 +314,8 @@ static int ipv6_destopt_rcv(struct sk_buff **skbp) return 1; } - IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(dst), IPSTATS_MIB_INHDRERRORS); + dst_release(dst); return -1; } @@ -365,7 +370,8 @@ static int ipv6_rthdr_rcv(struct sk_buff **skbp) if (!pskb_may_pull(skb, (skb->h.raw-skb->data)+8) || !pskb_may_pull(skb, (skb->h.raw-skb->data)+((skb->h.raw[1]+1)<<3))) { - IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), + IPSTATS_MIB_INHDRERRORS); kfree_skb(skb); return -1; } @@ -374,7 +380,8 @@ static int ipv6_rthdr_rcv(struct sk_buff **skbp) if (ipv6_addr_is_multicast(&skb->nh.ipv6h->daddr) || skb->pkt_type != PACKET_HOST) { - IP6_INC_STATS_BH(IPSTATS_MIB_INADDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), + IPSTATS_MIB_INADDRERRORS); kfree_skb(skb); return -1; } @@ -388,7 +395,8 @@ looped_back: * processed by own */ if (!addr) { - IP6_INC_STATS_BH(IPSTATS_MIB_INADDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), + IPSTATS_MIB_INADDRERRORS); kfree_skb(skb); return -1; } @@ -410,7 +418,8 @@ looped_back: switch (hdr->type) { case IPV6_SRCRT_TYPE_0: if (hdr->hdrlen & 0x01) { - IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), + IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, (&hdr->hdrlen) - skb->nh.raw); return -1; } @@ -419,14 +428,16 @@ looped_back: case IPV6_SRCRT_TYPE_2: /* Silently discard invalid RTH type 2 */ if (hdr->hdrlen != 2 || hdr->segments_left != 1) { - IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), + IPSTATS_MIB_INHDRERRORS); kfree_skb(skb); return -1; } break; #endif default: - IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), + IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, (&hdr->type) - skb->nh.raw); return -1; } @@ -439,7 +450,8 @@ looped_back: n = hdr->hdrlen >> 1; if (hdr->segments_left > n) { - IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), + IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, (&hdr->segments_left) - skb->nh.raw); return -1; } @@ -449,12 +461,14 @@ looped_back: */ if (skb_cloned(skb)) { struct sk_buff *skb2 = skb_copy(skb, GFP_ATOMIC); - kfree_skb(skb); /* the copy is a forwarded packet */ if (skb2 == NULL) { - IP6_INC_STATS_BH(IPSTATS_MIB_OUTDISCARDS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), + IPSTATS_MIB_OUTDISCARDS); + kfree_skb(skb); return -1; } + kfree_skb(skb); *skbp = skb = skb2; opt = IP6CB(skb2); hdr = (struct ipv6_rt_hdr *) skb2->h.raw; @@ -475,12 +489,14 @@ looped_back: if (xfrm6_input_addr(skb, (xfrm_address_t *)addr, (xfrm_address_t *)&skb->nh.ipv6h->saddr, IPPROTO_ROUTING) < 0) { - IP6_INC_STATS_BH(IPSTATS_MIB_INADDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), + IPSTATS_MIB_INADDRERRORS); kfree_skb(skb); return -1; } if (!ipv6_chk_home_addr(addr)) { - IP6_INC_STATS_BH(IPSTATS_MIB_INADDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), + IPSTATS_MIB_INADDRERRORS); kfree_skb(skb); return -1; } @@ -491,7 +507,8 @@ looped_back: } if (ipv6_addr_is_multicast(addr)) { - IP6_INC_STATS_BH(IPSTATS_MIB_INADDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), + IPSTATS_MIB_INADDRERRORS); kfree_skb(skb); return -1; } @@ -510,7 +527,8 @@ looped_back: if (skb->dst->dev->flags&IFF_LOOPBACK) { if (skb->nh.ipv6h->hop_limit <= 1) { - IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), + IPSTATS_MIB_INHDRERRORS); icmpv6_send(skb, ICMPV6_TIME_EXCEED, ICMPV6_EXC_HOPLIMIT, 0, skb->dev); kfree_skb(skb); @@ -632,24 +650,25 @@ static int ipv6_hop_jumbo(struct sk_buff **skbp, int optoff) if (skb->nh.raw[optoff+1] != 4 || (optoff&3) != 2) { LIMIT_NETDEBUG(KERN_DEBUG "ipv6_hop_jumbo: wrong jumbo opt length/alignment %d\n", skb->nh.raw[optoff+1]); - IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), + IPSTATS_MIB_INHDRERRORS); goto drop; } pkt_len = ntohl(*(u32*)(skb->nh.raw+optoff+2)); if (pkt_len <= IPV6_MAXPLEN) { - IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, optoff+2); return 0; } if (skb->nh.ipv6h->payload_len) { - IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, optoff); return 0; } if (pkt_len > skb->len - sizeof(struct ipv6hdr)) { - IP6_INC_STATS_BH(IPSTATS_MIB_INTRUNCATEDPKTS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), IPSTATS_MIB_INTRUNCATEDPKTS); goto drop; } diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 81bd45b26c9..52cca93ff2f 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -177,7 +177,8 @@ static inline int icmpv6_xrlim_allow(struct sock *sk, int type, */ dst = ip6_route_output(sk, fl); if (dst->error) { - IP6_INC_STATS(IPSTATS_MIB_OUTNOROUTES); + IP6_INC_STATS(ip6_dst_idev(dst), + IPSTATS_MIB_OUTNOROUTES); } else if (dst->dev && (dst->dev->flags&IFF_LOOPBACK)) { res = 1; } else { diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c index ebf54ae90a0..ad0b8abcdf4 100644 --- a/net/ipv6/ip6_input.c +++ b/net/ipv6/ip6_input.c @@ -60,14 +60,22 @@ int ipv6_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt { struct ipv6hdr *hdr; u32 pkt_len; + struct inet6_dev *idev; - if (skb->pkt_type == PACKET_OTHERHOST) - goto drop; + if (skb->pkt_type == PACKET_OTHERHOST) { + kfree_skb(skb); + return 0; + } + + rcu_read_lock(); - IP6_INC_STATS_BH(IPSTATS_MIB_INRECEIVES); + idev = __in6_dev_get(skb->dev); + + IP6_INC_STATS_BH(idev, IPSTATS_MIB_INRECEIVES); if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) { - IP6_INC_STATS_BH(IPSTATS_MIB_INDISCARDS); + IP6_INC_STATS_BH(idev, IPSTATS_MIB_INDISCARDS); + rcu_read_unlock(); goto out; } @@ -104,7 +112,7 @@ int ipv6_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt if (pkt_len + sizeof(struct ipv6hdr) > skb->len) goto truncated; if (pskb_trim_rcsum(skb, pkt_len + sizeof(struct ipv6hdr))) { - IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS_BH(idev, IPSTATS_MIB_INHDRERRORS); goto drop; } hdr = skb->nh.ipv6h; @@ -112,17 +120,21 @@ int ipv6_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt if (hdr->nexthdr == NEXTHDR_HOP) { if (ipv6_parse_hopopts(&skb) < 0) { - IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS_BH(idev, IPSTATS_MIB_INHDRERRORS); + rcu_read_unlock(); return 0; } } + rcu_read_unlock(); + return NF_HOOK(PF_INET6,NF_IP6_PRE_ROUTING, skb, dev, NULL, ip6_rcv_finish); truncated: - IP6_INC_STATS_BH(IPSTATS_MIB_INTRUNCATEDPKTS); + IP6_INC_STATS_BH(idev, IPSTATS_MIB_INTRUNCATEDPKTS); err: - IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS_BH(idev, IPSTATS_MIB_INHDRERRORS); drop: + rcu_read_unlock(); kfree_skb(skb); out: return 0; @@ -140,6 +152,7 @@ static inline int ip6_input_finish(struct sk_buff *skb) unsigned int nhoff; int nexthdr; u8 hash; + struct inet6_dev *idev; /* * Parse extension headers @@ -147,6 +160,7 @@ static inline int ip6_input_finish(struct sk_buff *skb) rcu_read_lock(); resubmit: + idev = ip6_dst_idev(skb->dst); if (!pskb_pull(skb, skb->h.raw - skb->data)) goto discard; nhoff = IP6CB(skb)->nhoff; @@ -185,24 +199,24 @@ resubmit: if (ret > 0) goto resubmit; else if (ret == 0) - IP6_INC_STATS_BH(IPSTATS_MIB_INDELIVERS); + IP6_INC_STATS_BH(idev, IPSTATS_MIB_INDELIVERS); } else { if (!raw_sk) { if (xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) { - IP6_INC_STATS_BH(IPSTATS_MIB_INUNKNOWNPROTOS); + IP6_INC_STATS_BH(idev, IPSTATS_MIB_INUNKNOWNPROTOS); icmpv6_send(skb, ICMPV6_PARAMPROB, ICMPV6_UNK_NEXTHDR, nhoff, skb->dev); } } else - IP6_INC_STATS_BH(IPSTATS_MIB_INDELIVERS); + IP6_INC_STATS_BH(idev, IPSTATS_MIB_INDELIVERS); kfree_skb(skb); } rcu_read_unlock(); return 0; discard: - IP6_INC_STATS_BH(IPSTATS_MIB_INDISCARDS); + IP6_INC_STATS_BH(idev, IPSTATS_MIB_INDISCARDS); rcu_read_unlock(); kfree_skb(skb); return 0; @@ -219,7 +233,7 @@ int ip6_mc_input(struct sk_buff *skb) struct ipv6hdr *hdr; int deliver; - IP6_INC_STATS_BH(IPSTATS_MIB_INMCASTPKTS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), IPSTATS_MIB_INMCASTPKTS); hdr = skb->nh.ipv6h; deliver = likely(!(skb->dev->flags & (IFF_PROMISC|IFF_ALLMULTI))) || diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 1bde3aca346..85f88927049 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -88,7 +88,7 @@ static inline int ip6_output_finish(struct sk_buff *skb) } else if (dst->neighbour) return dst->neighbour->output(skb); - IP6_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES); + IP6_INC_STATS_BH(ip6_dst_idev(dst), IPSTATS_MIB_OUTNOROUTES); kfree_skb(skb); return -EINVAL; @@ -118,6 +118,7 @@ static int ip6_output2(struct sk_buff *skb) if (ipv6_addr_is_multicast(&skb->nh.ipv6h->daddr)) { struct ipv6_pinfo* np = skb->sk ? inet6_sk(skb->sk) : NULL; + struct inet6_dev *idev = ip6_dst_idev(skb->dst); if (!(dev->flags & IFF_LOOPBACK) && (!np || np->mc_loop) && ipv6_chk_mcast_addr(dev, &skb->nh.ipv6h->daddr, @@ -133,13 +134,13 @@ static int ip6_output2(struct sk_buff *skb) ip6_dev_loopback_xmit); if (skb->nh.ipv6h->hop_limit == 0) { - IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS); + IP6_INC_STATS(idev, IPSTATS_MIB_OUTDISCARDS); kfree_skb(skb); return 0; } } - IP6_INC_STATS(IPSTATS_MIB_OUTMCASTPKTS); + IP6_INC_STATS(idev, IPSTATS_MIB_OUTMCASTPKTS); } return NF_HOOK(PF_INET6, NF_IP6_POST_ROUTING, skb,NULL, skb->dev,ip6_output_finish); @@ -182,12 +183,14 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl, if (skb_headroom(skb) < head_room) { struct sk_buff *skb2 = skb_realloc_headroom(skb, head_room); - kfree_skb(skb); - skb = skb2; - if (skb == NULL) { - IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS); + if (skb2 == NULL) { + IP6_INC_STATS(ip6_dst_idev(skb->dst), + IPSTATS_MIB_OUTDISCARDS); + kfree_skb(skb); return -ENOBUFS; } + kfree_skb(skb); + skb = skb2; if (sk) skb_set_owner_w(skb, sk); } @@ -230,7 +233,8 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl, mtu = dst_mtu(dst); if ((skb->len <= mtu) || ipfragok || skb_is_gso(skb)) { - IP6_INC_STATS(IPSTATS_MIB_OUTREQUESTS); + IP6_INC_STATS(ip6_dst_idev(skb->dst), + IPSTATS_MIB_OUTREQUESTS); return NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, dst->dev, dst_output); } @@ -239,7 +243,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl, printk(KERN_DEBUG "IPv6: sending pkt_too_big to self\n"); skb->dev = dst->dev; icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev); - IP6_INC_STATS(IPSTATS_MIB_FRAGFAILS); + IP6_INC_STATS(ip6_dst_idev(skb->dst), IPSTATS_MIB_FRAGFAILS); kfree_skb(skb); return -EMSGSIZE; } @@ -373,7 +377,7 @@ int ip6_forward(struct sk_buff *skb) goto error; if (!xfrm6_policy_check(NULL, XFRM_POLICY_FWD, skb)) { - IP6_INC_STATS(IPSTATS_MIB_INDISCARDS); + IP6_INC_STATS(ip6_dst_idev(dst), IPSTATS_MIB_INDISCARDS); goto drop; } @@ -406,7 +410,7 @@ int ip6_forward(struct sk_buff *skb) skb->dev = dst->dev; icmpv6_send(skb, ICMPV6_TIME_EXCEED, ICMPV6_EXC_HOPLIMIT, 0, skb->dev); - IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(dst), IPSTATS_MIB_INHDRERRORS); kfree_skb(skb); return -ETIMEDOUT; @@ -419,13 +423,13 @@ int ip6_forward(struct sk_buff *skb) if (proxied > 0) return ip6_input(skb); else if (proxied < 0) { - IP6_INC_STATS(IPSTATS_MIB_INDISCARDS); + IP6_INC_STATS(ip6_dst_idev(dst), IPSTATS_MIB_INDISCARDS); goto drop; } } if (!xfrm6_route_forward(skb)) { - IP6_INC_STATS(IPSTATS_MIB_INDISCARDS); + IP6_INC_STATS(ip6_dst_idev(dst), IPSTATS_MIB_INDISCARDS); goto drop; } dst = skb->dst; @@ -464,14 +468,14 @@ int ip6_forward(struct sk_buff *skb) /* Again, force OUTPUT device used as source address */ skb->dev = dst->dev; icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, dst_mtu(dst), skb->dev); - IP6_INC_STATS_BH(IPSTATS_MIB_INTOOBIGERRORS); - IP6_INC_STATS_BH(IPSTATS_MIB_FRAGFAILS); + IP6_INC_STATS_BH(ip6_dst_idev(dst), IPSTATS_MIB_INTOOBIGERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(dst), IPSTATS_MIB_FRAGFAILS); kfree_skb(skb); return -EMSGSIZE; } if (skb_cow(skb, dst->dev->hard_header_len)) { - IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS); + IP6_INC_STATS(ip6_dst_idev(dst), IPSTATS_MIB_OUTDISCARDS); goto drop; } @@ -481,11 +485,11 @@ int ip6_forward(struct sk_buff *skb) hdr->hop_limit--; - IP6_INC_STATS_BH(IPSTATS_MIB_OUTFORWDATAGRAMS); + IP6_INC_STATS_BH(ip6_dst_idev(dst), IPSTATS_MIB_OUTFORWDATAGRAMS); return NF_HOOK(PF_INET6,NF_IP6_FORWARD, skb, skb->dev, dst->dev, ip6_forward_finish); error: - IP6_INC_STATS_BH(IPSTATS_MIB_INADDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(dst), IPSTATS_MIB_INADDRERRORS); drop: kfree_skb(skb); return -EINVAL; @@ -622,7 +626,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) tmp_hdr = kmalloc(hlen, GFP_ATOMIC); if (!tmp_hdr) { - IP6_INC_STATS(IPSTATS_MIB_FRAGFAILS); + IP6_INC_STATS(ip6_dst_idev(skb->dst), IPSTATS_MIB_FRAGFAILS); return -ENOMEM; } @@ -643,7 +647,8 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) skb->data_len = first_len - skb_headlen(skb); skb->len = first_len; skb->nh.ipv6h->payload_len = htons(first_len - sizeof(struct ipv6hdr)); - + + dst_hold(&rt->u.dst); for (;;) { /* Prepare header of the next frame, @@ -667,7 +672,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) err = output(skb); if(!err) - IP6_INC_STATS(IPSTATS_MIB_FRAGCREATES); + IP6_INC_STATS(ip6_dst_idev(&rt->u.dst), IPSTATS_MIB_FRAGCREATES); if (err || !frag) break; @@ -680,7 +685,8 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) kfree(tmp_hdr); if (err == 0) { - IP6_INC_STATS(IPSTATS_MIB_FRAGOKS); + IP6_INC_STATS(ip6_dst_idev(&rt->u.dst), IPSTATS_MIB_FRAGOKS); + dst_release(&rt->u.dst); return 0; } @@ -690,7 +696,8 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) frag = skb; } - IP6_INC_STATS(IPSTATS_MIB_FRAGFAILS); + IP6_INC_STATS(ip6_dst_idev(&rt->u.dst), IPSTATS_MIB_FRAGFAILS); + dst_release(&rt->u.dst); return err; } @@ -723,7 +730,8 @@ slow_path: if ((frag = alloc_skb(len+hlen+sizeof(struct frag_hdr)+LL_RESERVED_SPACE(rt->u.dst.dev), GFP_ATOMIC)) == NULL) { NETDEBUG(KERN_INFO "IPv6: frag: no memory for new fragment!\n"); - IP6_INC_STATS(IPSTATS_MIB_FRAGFAILS); + IP6_INC_STATS(ip6_dst_idev(skb->dst), + IPSTATS_MIB_FRAGFAILS); err = -ENOMEM; goto fail; } @@ -784,15 +792,17 @@ slow_path: if (err) goto fail; - IP6_INC_STATS(IPSTATS_MIB_FRAGCREATES); + IP6_INC_STATS(ip6_dst_idev(skb->dst), IPSTATS_MIB_FRAGCREATES); } + IP6_INC_STATS(ip6_dst_idev(skb->dst), + IPSTATS_MIB_FRAGOKS); kfree_skb(skb); - IP6_INC_STATS(IPSTATS_MIB_FRAGOKS); return err; fail: + IP6_INC_STATS(ip6_dst_idev(skb->dst), + IPSTATS_MIB_FRAGFAILS); kfree_skb(skb); - IP6_INC_STATS(IPSTATS_MIB_FRAGFAILS); return err; } @@ -1265,7 +1275,7 @@ alloc_new_skb: return 0; error: inet->cork.length -= length; - IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS); + IP6_INC_STATS(rt->rt6i_idev, IPSTATS_MIB_OUTDISCARDS); return err; } @@ -1326,7 +1336,7 @@ int ip6_push_pending_frames(struct sock *sk) skb->priority = sk->sk_priority; skb->dst = dst_clone(&rt->u.dst); - IP6_INC_STATS(IPSTATS_MIB_OUTREQUESTS); + IP6_INC_STATS(rt->rt6i_idev, IPSTATS_MIB_OUTREQUESTS); err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, skb->dst->dev, dst_output); if (err) { if (err > 0) @@ -1357,7 +1367,8 @@ void ip6_flush_pending_frames(struct sock *sk) struct sk_buff *skb; while ((skb = __skb_dequeue_tail(&sk->sk_write_queue)) != NULL) { - IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS); + IP6_INC_STATS(ip6_dst_idev(skb->dst), + IPSTATS_MIB_OUTDISCARDS); kfree_skb(skb); } diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index 9055979083b..c006d02be8b 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -1465,7 +1465,7 @@ static void mld_sendpack(struct sk_buff *skb) struct inet6_dev *idev = in6_dev_get(skb->dev); int err; - IP6_INC_STATS(IPSTATS_MIB_OUTREQUESTS); + IP6_INC_STATS(idev, IPSTATS_MIB_OUTREQUESTS); payload_len = skb->tail - (unsigned char *)skb->nh.ipv6h - sizeof(struct ipv6hdr); mldlen = skb->tail - skb->h.raw; @@ -1477,9 +1477,9 @@ static void mld_sendpack(struct sk_buff *skb) mld_dev_queue_xmit); if (!err) { ICMP6_INC_STATS(idev,ICMP6_MIB_OUTMSGS); - IP6_INC_STATS(IPSTATS_MIB_OUTMCASTPKTS); + IP6_INC_STATS(idev, IPSTATS_MIB_OUTMCASTPKTS); } else - IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS); + IP6_INC_STATS(idev, IPSTATS_MIB_OUTDISCARDS); if (likely(idev != NULL)) in6_dev_put(idev); @@ -1763,7 +1763,10 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type) IPV6_TLV_ROUTERALERT, 2, 0, 0, IPV6_TLV_PADN, 0 }; - IP6_INC_STATS(IPSTATS_MIB_OUTREQUESTS); + rcu_read_lock(); + IP6_INC_STATS(__in6_dev_get(dev), + IPSTATS_MIB_OUTREQUESTS); + rcu_read_unlock(); snd_addr = addr; if (type == ICMPV6_MGM_REDUCTION) { snd_addr = &all_routers; @@ -1777,7 +1780,10 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type) skb = sock_alloc_send_skb(sk, LL_RESERVED_SPACE(dev) + full_len, 1, &err); if (skb == NULL) { - IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS); + rcu_read_lock(); + IP6_INC_STATS(__in6_dev_get(dev), + IPSTATS_MIB_OUTDISCARDS); + rcu_read_unlock(); return; } @@ -1816,9 +1822,9 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type) else ICMP6_INC_STATS(idev, ICMP6_MIB_OUTGROUPMEMBRESPONSES); ICMP6_INC_STATS(idev, ICMP6_MIB_OUTMSGS); - IP6_INC_STATS(IPSTATS_MIB_OUTMCASTPKTS); + IP6_INC_STATS(idev, IPSTATS_MIB_OUTMCASTPKTS); } else - IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS); + IP6_INC_STATS(idev, IPSTATS_MIB_OUTDISCARDS); if (likely(idev != NULL)) in6_dev_put(idev); diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 89d527ebd7f..1342be8b4cd 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -515,7 +515,7 @@ static void ndisc_send_na(struct net_device *dev, struct neighbour *neigh, skb->dst = dst; idev = in6_dev_get(dst->dev); - IP6_INC_STATS(IPSTATS_MIB_OUTREQUESTS); + IP6_INC_STATS(idev, IPSTATS_MIB_OUTREQUESTS); err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, dst->dev, dst_output); if (!err) { ICMP6_INC_STATS(idev, ICMP6_MIB_OUTNEIGHBORADVERTISEMENTS); @@ -601,7 +601,7 @@ void ndisc_send_ns(struct net_device *dev, struct neighbour *neigh, /* send it! */ skb->dst = dst; idev = in6_dev_get(dst->dev); - IP6_INC_STATS(IPSTATS_MIB_OUTREQUESTS); + IP6_INC_STATS(idev, IPSTATS_MIB_OUTREQUESTS); err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, dst->dev, dst_output); if (!err) { ICMP6_INC_STATS(idev, ICMP6_MIB_OUTNEIGHBORSOLICITS); @@ -676,7 +676,7 @@ void ndisc_send_rs(struct net_device *dev, struct in6_addr *saddr, /* send it! */ skb->dst = dst; idev = in6_dev_get(dst->dev); - IP6_INC_STATS(IPSTATS_MIB_OUTREQUESTS); + IP6_INC_STATS(idev, IPSTATS_MIB_OUTREQUESTS); err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, dst->dev, dst_output); if (!err) { ICMP6_INC_STATS(idev, ICMP6_MIB_OUTROUTERSOLICITS); @@ -1512,7 +1512,7 @@ void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh, buff->dst = dst; idev = in6_dev_get(dst->dev); - IP6_INC_STATS(IPSTATS_MIB_OUTREQUESTS); + IP6_INC_STATS(idev, IPSTATS_MIB_OUTREQUESTS); err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, buff, NULL, dst->dev, dst_output); if (!err) { ICMP6_INC_STATS(idev, ICMP6_MIB_OUTREDIRECTS); diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c index 580b1aba672..646a47456fd 100644 --- a/net/ipv6/netfilter.c +++ b/net/ipv6/netfilter.c @@ -31,7 +31,7 @@ int ip6_route_me_harder(struct sk_buff *skb) #endif if (dst->error) { - IP6_INC_STATS(IPSTATS_MIB_OUTNOROUTES); + IP6_INC_STATS(ip6_dst_idev(dst), IPSTATS_MIB_OUTNOROUTES); LIMIT_NETDEBUG(KERN_DEBUG "ip6_route_me_harder: No more route.\n"); dst_release(dst); return -EINVAL; diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c index efee7a6301a..4158d386b0a 100644 --- a/net/ipv6/proc.c +++ b/net/ipv6/proc.c @@ -161,6 +161,7 @@ static int snmp6_seq_show(struct seq_file *seq, void *v) if (idev) { seq_printf(seq, "%-32s\t%u\n", "ifIndex", idev->dev->ifindex); + snmp6_seq_show_item(seq, (void **)idev->stats.ipv6, snmp6_ipstats_list); snmp6_seq_show_item(seq, (void **)idev->stats.icmpv6, snmp6_icmp6_list); } else { snmp6_seq_show_item(seq, (void **)ipv6_statistics, snmp6_ipstats_list); @@ -281,6 +282,9 @@ int snmp6_alloc_dev(struct inet6_dev *idev) if (!idev || !idev->dev) return -EINVAL; + if (snmp6_mib_init((void **)idev->stats.ipv6, sizeof(struct ipstats_mib), + __alignof__(struct ipstats_mib)) < 0) + goto err_ip; if (snmp6_mib_init((void **)idev->stats.icmpv6, sizeof(struct icmpv6_mib), __alignof__(struct icmpv6_mib)) < 0) goto err_icmp; @@ -288,12 +292,15 @@ int snmp6_alloc_dev(struct inet6_dev *idev) return 0; err_icmp: + snmp6_mib_free((void **)idev->stats.ipv6); +err_ip: return err; } int snmp6_free_dev(struct inet6_dev *idev) { snmp6_mib_free((void **)idev->stats.icmpv6); + snmp6_mib_free((void **)idev->stats.ipv6); return 0; } diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 6bc66552442..18a90075f94 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -586,7 +586,7 @@ static int rawv6_send_hdrinc(struct sock *sk, void *from, int length, if (err) goto error_fault; - IP6_INC_STATS(IPSTATS_MIB_OUTREQUESTS); + IP6_INC_STATS(rt->rt6i_idev, IPSTATS_MIB_OUTREQUESTS); err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, rt->u.dst.dev, dst_output); if (err > 0) @@ -600,7 +600,7 @@ error_fault: err = -EFAULT; kfree_skb(skb); error: - IP6_INC_STATS(IPSTATS_MIB_OUTDISCARDS); + IP6_INC_STATS(rt->rt6i_idev, IPSTATS_MIB_OUTDISCARDS); return err; } diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index f39bbedd132..3af0d5a6cee 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -47,6 +47,7 @@ #include #include +#include #include #include #include @@ -257,7 +258,7 @@ static __inline__ void fq_kill(struct frag_queue *fq) } } -static void ip6_evictor(void) +static void ip6_evictor(struct inet6_dev *idev) { struct frag_queue *fq; struct list_head *tmp; @@ -284,14 +285,14 @@ static void ip6_evictor(void) spin_unlock(&fq->lock); fq_put(fq, &work); - IP6_INC_STATS_BH(IPSTATS_MIB_REASMFAILS); + IP6_INC_STATS_BH(idev, IPSTATS_MIB_REASMFAILS); } } static void ip6_frag_expire(unsigned long data) { struct frag_queue *fq = (struct frag_queue *) data; - struct net_device *dev; + struct net_device *dev = NULL; spin_lock(&fq->lock); @@ -300,17 +301,19 @@ static void ip6_frag_expire(unsigned long data) fq_kill(fq); - IP6_INC_STATS_BH(IPSTATS_MIB_REASMTIMEOUT); - IP6_INC_STATS_BH(IPSTATS_MIB_REASMFAILS); + dev = dev_get_by_index(fq->iif); + if (!dev) + goto out; + + rcu_read_lock(); + IP6_INC_STATS_BH(__in6_dev_get(dev), IPSTATS_MIB_REASMTIMEOUT); + IP6_INC_STATS_BH(__in6_dev_get(dev), IPSTATS_MIB_REASMFAILS); + rcu_read_unlock(); /* Don't send error if the first segment did not arrive. */ if (!(fq->last_in&FIRST_IN) || !fq->fragments) goto out; - dev = dev_get_by_index(fq->iif); - if (!dev) - goto out; - /* But use as source device on which LAST ARRIVED segment was received. And do not use fq->dev @@ -318,8 +321,9 @@ static void ip6_frag_expire(unsigned long data) */ fq->fragments->dev = dev; icmpv6_send(fq->fragments, ICMPV6_TIME_EXCEED, ICMPV6_EXC_FRAGTIME, 0, dev); - dev_put(dev); out: + if (dev) + dev_put(dev); spin_unlock(&fq->lock); fq_put(fq, NULL); } @@ -366,7 +370,8 @@ static struct frag_queue *ip6_frag_intern(struct frag_queue *fq_in) static struct frag_queue * -ip6_frag_create(u32 id, struct in6_addr *src, struct in6_addr *dst) +ip6_frag_create(u32 id, struct in6_addr *src, struct in6_addr *dst, + struct inet6_dev *idev) { struct frag_queue *fq; @@ -386,12 +391,13 @@ ip6_frag_create(u32 id, struct in6_addr *src, struct in6_addr *dst) return ip6_frag_intern(fq); oom: - IP6_INC_STATS_BH(IPSTATS_MIB_REASMFAILS); + IP6_INC_STATS_BH(idev, IPSTATS_MIB_REASMFAILS); return NULL; } static __inline__ struct frag_queue * -fq_find(u32 id, struct in6_addr *src, struct in6_addr *dst) +fq_find(u32 id, struct in6_addr *src, struct in6_addr *dst, + struct inet6_dev *idev) { struct frag_queue *fq; struct hlist_node *n; @@ -410,7 +416,7 @@ fq_find(u32 id, struct in6_addr *src, struct in6_addr *dst) } read_unlock(&ip6_frag_lock); - return ip6_frag_create(id, src, dst); + return ip6_frag_create(id, src, dst, idev); } @@ -428,7 +434,8 @@ static void ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb, ((u8 *) (fhdr + 1) - (u8 *) (skb->nh.ipv6h + 1))); if ((unsigned int)end > IPV6_MAXPLEN) { - IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), + IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb,ICMPV6_HDR_FIELD, (u8*)&fhdr->frag_off - skb->nh.raw); return; } @@ -455,7 +462,8 @@ static void ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb, /* RFC2460 says always send parameter problem in * this case. -DaveM */ - IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), + IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, offsetof(struct ipv6hdr, payload_len)); return; @@ -571,7 +579,7 @@ static void ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb, return; err: - IP6_INC_STATS(IPSTATS_MIB_REASMFAILS); + IP6_INC_STATS(ip6_dst_idev(skb->dst), IPSTATS_MIB_REASMFAILS); kfree_skb(skb); } @@ -665,7 +673,9 @@ static int ip6_frag_reasm(struct frag_queue *fq, struct sk_buff **skb_in, if (head->ip_summed == CHECKSUM_COMPLETE) head->csum = csum_partial(head->nh.raw, head->h.raw-head->nh.raw, head->csum); - IP6_INC_STATS_BH(IPSTATS_MIB_REASMOKS); + rcu_read_lock(); + IP6_INC_STATS_BH(__in6_dev_get(dev), IPSTATS_MIB_REASMOKS); + rcu_read_unlock(); fq->fragments = NULL; return 1; @@ -677,7 +687,9 @@ out_oom: if (net_ratelimit()) printk(KERN_DEBUG "ip6_frag_reasm: no memory for reassembly\n"); out_fail: - IP6_INC_STATS_BH(IPSTATS_MIB_REASMFAILS); + rcu_read_lock(); + IP6_INC_STATS_BH(__in6_dev_get(dev), IPSTATS_MIB_REASMFAILS); + rcu_read_unlock(); return -1; } @@ -691,16 +703,16 @@ static int ipv6_frag_rcv(struct sk_buff **skbp) hdr = skb->nh.ipv6h; - IP6_INC_STATS_BH(IPSTATS_MIB_REASMREQDS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), IPSTATS_MIB_REASMREQDS); /* Jumbo payload inhibits frag. header */ if (hdr->payload_len==0) { - IP6_INC_STATS(IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS(ip6_dst_idev(skb->dst), IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, skb->h.raw-skb->nh.raw); return -1; } if (!pskb_may_pull(skb, (skb->h.raw-skb->data)+sizeof(struct frag_hdr))) { - IP6_INC_STATS(IPSTATS_MIB_INHDRERRORS); + IP6_INC_STATS(ip6_dst_idev(skb->dst), IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, skb->h.raw-skb->nh.raw); return -1; } @@ -711,16 +723,17 @@ static int ipv6_frag_rcv(struct sk_buff **skbp) if (!(fhdr->frag_off & htons(0xFFF9))) { /* It is not a fragmented frame */ skb->h.raw += sizeof(struct frag_hdr); - IP6_INC_STATS_BH(IPSTATS_MIB_REASMOKS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), IPSTATS_MIB_REASMOKS); IP6CB(skb)->nhoff = (u8*)fhdr - skb->nh.raw; return 1; } if (atomic_read(&ip6_frag_mem) > sysctl_ip6frag_high_thresh) - ip6_evictor(); + ip6_evictor(ip6_dst_idev(skb->dst)); - if ((fq = fq_find(fhdr->identification, &hdr->saddr, &hdr->daddr)) != NULL) { + if ((fq = fq_find(fhdr->identification, &hdr->saddr, &hdr->daddr, + ip6_dst_idev(skb->dst))) != NULL) { int ret = -1; spin_lock(&fq->lock); @@ -736,7 +749,7 @@ static int ipv6_frag_rcv(struct sk_buff **skbp) return ret; } - IP6_INC_STATS_BH(IPSTATS_MIB_REASMFAILS); + IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), IPSTATS_MIB_REASMFAILS); kfree_skb(skb); return -1; } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index eb78b525224..0ad07c9087a 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1749,9 +1749,9 @@ static inline int ip6_pkt_drop(struct sk_buff *skb, int code) { int type = ipv6_addr_type(&skb->nh.ipv6h->daddr); if (type == IPV6_ADDR_ANY || type == IPV6_ADDR_RESERVED) - IP6_INC_STATS(IPSTATS_MIB_INADDRERRORS); + IP6_INC_STATS(ip6_dst_idev(skb->dst), IPSTATS_MIB_INADDRERRORS); - IP6_INC_STATS(IPSTATS_MIB_OUTNOROUTES); + IP6_INC_STATS(ip6_dst_idev(skb->dst), IPSTATS_MIB_OUTNOROUTES); icmpv6_send(skb, ICMPV6_DEST_UNREACH, code, 0, skb->dev); kfree_skb(skb); return 0; -- cgit v1.2.3-70-g09d2 From 6f4e5fff1e4d46714ea554fd83e44eab534e8b11 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Fri, 10 Nov 2006 17:43:06 -0200 Subject: [DCCP]: Support for partial checksums (RFC 4340, sec. 9.2) This patch does the following: a) introduces variable-length checksums as specified in [RFC 4340, sec. 9.2] b) provides necessary socket options and documentation as to how to use them c) basic support and infrastructure for the Minimum Checksum Coverage feature [RFC 4340, sec. 9.2.1]: acceptability tests, user notification and user interface In addition, it (1) fixes two bugs in the DCCPv4 checksum computation: * pseudo-header used checksum_len instead of skb->len * incorrect checksum coverage calculation based on dccph_x (2) removes dccp_v4_verify_checksum() since it reduplicates code of the checksum computation; code calling this function is updated accordingly. (3) now uses skb_checksum(), which is safer than checksum_partial() if the sk_buff has is a non-linear buffer (has pages attached to it). (4) fixes an outstanding TODO item: * If P.CsCov is too large for the packet size, drop packet and return. The code has been tested with applications, the latest version of tcpdump now comes with support for partial DCCP checksums. Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo --- Documentation/networking/dccp.txt | 16 ++++++ include/linux/dccp.h | 7 +++ net/dccp/dccp.h | 29 +++++++++-- net/dccp/ipv4.c | 105 +++++++++++++++++++------------------- net/dccp/ipv6.c | 74 +++++++++++++++++---------- net/dccp/output.c | 13 ++--- net/dccp/proto.c | 26 ++++++++-- 7 files changed, 173 insertions(+), 97 deletions(-) (limited to 'include') diff --git a/Documentation/networking/dccp.txt b/Documentation/networking/dccp.txt index 74563b38ffd..a8142a81038 100644 --- a/Documentation/networking/dccp.txt +++ b/Documentation/networking/dccp.txt @@ -47,6 +47,22 @@ the socket will fall back to 0 (which means that no meaningful service code is present). Connecting sockets set at most one service option; for listening sockets, multiple service codes can be specified. +DCCP_SOCKOPT_SEND_CSCOV and DCCP_SOCKOPT_RECV_CSCOV are used for setting the +partial checksum coverage (RFC 4340, sec. 9.2). The default is that checksums +always cover the entire packet and that only fully covered application data is +accepted by the receiver. Hence, when using this feature on the sender, it must +be enabled at the receiver, too with suitable choice of CsCov. + +DCCP_SOCKOPT_SEND_CSCOV sets the sender checksum coverage. Values in the + range 0..15 are acceptable. The default setting is 0 (full coverage), + values between 1..15 indicate partial coverage. +DCCP_SOCKOPT_SEND_CSCOV is for the receiver and has a different meaning: it + sets a threshold, where again values 0..15 are acceptable. The default + of 0 means that all packets with a partial coverage will be discarded. + Values in the range 1..15 indicate that packets with minimally such a + coverage value are also acceptable. The higher the number, the more + restrictive this setting (see [RFC 4340, sec. 9.2.1]). + Notes ===== diff --git a/include/linux/dccp.h b/include/linux/dccp.h index 90d04ffddae..0502dfa7f32 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -183,6 +183,7 @@ enum { DCCPF_ACK_RATIO = 5, DCCPF_SEND_ACK_VECTOR = 6, DCCPF_SEND_NDP_COUNT = 7, + DCCPF_MIN_CSUM_COVER = 8, /* 10-127 reserved */ DCCPF_MIN_CCID_SPECIFIC = 128, DCCPF_MAX_CCID_SPECIFIC = 255, @@ -200,6 +201,8 @@ struct dccp_so_feat { #define DCCP_SOCKOPT_SERVICE 2 #define DCCP_SOCKOPT_CHANGE_L 3 #define DCCP_SOCKOPT_CHANGE_R 4 +#define DCCP_SOCKOPT_SEND_CSCOV 10 +#define DCCP_SOCKOPT_RECV_CSCOV 11 #define DCCP_SOCKOPT_CCID_RX_INFO 128 #define DCCP_SOCKOPT_CCID_TX_INFO 192 @@ -450,6 +453,8 @@ struct dccp_ackvec; * @dccps_timestamp_echo - latest timestamp received on a TIMESTAMP option * @dccps_packet_size - Set thru setsockopt * @dccps_role - Role of this sock, one of %dccp_role + * @dccps_pcslen - sender partial checksum coverage (via sockopt) + * @dccps_pcrlen - receiver partial checksum coverage (via sockopt) * @dccps_ndp_count - number of Non Data Packets since last data packet * @dccps_hc_rx_ackvec - rx half connection ack vector * @dccps_xmit_timer - timer for when CCID is not ready to send @@ -474,6 +479,8 @@ struct dccp_sock { __u32 dccps_packet_size; __u16 dccps_l_ack_ratio; __u16 dccps_r_ack_ratio; + __u16 dccps_pcslen; + __u16 dccps_pcrlen; unsigned long dccps_ndp_count; __u32 dccps_mss_cache; struct dccp_minisock dccps_minisock; diff --git a/net/dccp/dccp.h b/net/dccp/dccp.h index 2990bfb1258..d5c414bf781 100644 --- a/net/dccp/dccp.h +++ b/net/dccp/dccp.h @@ -129,6 +129,30 @@ DECLARE_SNMP_STAT(struct dccp_mib, dccp_statistics); #define DCCP_ADD_STATS_USER(field, val) \ SNMP_ADD_STATS_USER(dccp_statistics, field, val) +/* + * Checksumming routines + */ +static inline int dccp_csum_coverage(const struct sk_buff *skb) +{ + const struct dccp_hdr* dh = dccp_hdr(skb); + + if (dh->dccph_cscov == 0) + return skb->len; + return (dh->dccph_doff + dh->dccph_cscov - 1) * sizeof(u32); +} + +static inline void dccp_csum_outgoing(struct sk_buff *skb) +{ + int cov = dccp_csum_coverage(skb); + + if (cov >= skb->len) + dccp_hdr(skb)->dccph_cscov = 0; + + skb->csum = skb_checksum(skb, 0, (cov > skb->len)? skb->len : cov, 0); +} + +extern void dccp_v4_send_check(struct sock *sk, int len, struct sk_buff *skb); + extern int dccp_retransmit_skb(struct sock *sk, struct sk_buff *skb); extern void dccp_send_ack(struct sock *sk); @@ -214,14 +238,9 @@ extern void dccp_shutdown(struct sock *sk, int how); extern int inet_dccp_listen(struct socket *sock, int backlog); extern unsigned int dccp_poll(struct file *file, struct socket *sock, poll_table *wait); -extern void dccp_v4_send_check(struct sock *sk, int len, - struct sk_buff *skb); extern int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len); -extern int dccp_v4_checksum(const struct sk_buff *skb, - const __be32 saddr, const __be32 daddr); - extern int dccp_send_reset(struct sock *sk, enum dccp_reset_codes code); extern void dccp_send_close(struct sock *sk, const int active); extern int dccp_invalid_packet(struct sk_buff *skb); diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 91bffaa761a..496112080f3 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -349,13 +349,19 @@ out: sock_put(sk); } -/* This routine computes an IPv4 DCCP checksum. */ -void dccp_v4_send_check(struct sock *sk, int len, struct sk_buff *skb) +static inline u16 dccp_v4_csum_finish(struct sk_buff *skb, + __be32 src, __be32 dst) +{ + return csum_tcpudp_magic(src, dst, skb->len, IPPROTO_DCCP, skb->csum); +} + +void dccp_v4_send_check(struct sock *sk, int unused, struct sk_buff *skb) { const struct inet_sock *inet = inet_sk(sk); struct dccp_hdr *dh = dccp_hdr(skb); - dh->dccph_checksum = dccp_v4_checksum(skb, inet->saddr, inet->daddr); + dccp_csum_outgoing(skb); + dh->dccph_checksum = dccp_v4_csum_finish(skb, inet->saddr, inet->daddr); } EXPORT_SYMBOL_GPL(dccp_v4_send_check); @@ -454,47 +460,6 @@ static struct sock *dccp_v4_hnd_req(struct sock *sk, struct sk_buff *skb) return sk; } -int dccp_v4_checksum(const struct sk_buff *skb, const __be32 saddr, - const __be32 daddr) -{ - const struct dccp_hdr* dh = dccp_hdr(skb); - int checksum_len; - u32 tmp; - - if (dh->dccph_cscov == 0) - checksum_len = skb->len; - else { - checksum_len = (dh->dccph_cscov + dh->dccph_x) * sizeof(u32); - checksum_len = checksum_len < skb->len ? checksum_len : - skb->len; - } - - tmp = csum_partial((unsigned char *)dh, checksum_len, 0); - return csum_tcpudp_magic(saddr, daddr, checksum_len, - IPPROTO_DCCP, tmp); -} - -EXPORT_SYMBOL_GPL(dccp_v4_checksum); - -static int dccp_v4_verify_checksum(struct sk_buff *skb, - const __be32 saddr, const __be32 daddr) -{ - struct dccp_hdr *dh = dccp_hdr(skb); - int checksum_len; - u32 tmp; - - if (dh->dccph_cscov == 0) - checksum_len = skb->len; - else { - checksum_len = (dh->dccph_cscov + dh->dccph_x) * sizeof(u32); - checksum_len = checksum_len < skb->len ? checksum_len : - skb->len; - } - tmp = csum_partial((unsigned char *)dh, checksum_len, 0); - return csum_tcpudp_magic(saddr, daddr, checksum_len, - IPPROTO_DCCP, tmp) == 0 ? 0 : -1; -} - static struct dst_entry* dccp_v4_route_skb(struct sock *sk, struct sk_buff *skb) { @@ -536,8 +501,8 @@ static int dccp_v4_send_response(struct sock *sk, struct request_sock *req, const struct inet_request_sock *ireq = inet_rsk(req); struct dccp_hdr *dh = dccp_hdr(skb); - dh->dccph_checksum = dccp_v4_checksum(skb, ireq->loc_addr, - ireq->rmt_addr); + dh->dccph_checksum = dccp_v4_csum_finish(skb, ireq->loc_addr, + ireq->rmt_addr); memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); err = ip_build_and_send_pkt(skb, sk, ireq->loc_addr, ireq->rmt_addr, @@ -602,8 +567,9 @@ static void dccp_v4_ctl_send_reset(struct sk_buff *rxskb) dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), DCCP_SKB_CB(rxskb)->dccpd_seq); - dh->dccph_checksum = dccp_v4_checksum(skb, rxskb->nh.iph->saddr, - rxskb->nh.iph->daddr); + dccp_csum_outgoing(skb); + dh->dccph_checksum = dccp_v4_csum_finish(skb, rxskb->nh.iph->saddr, + rxskb->nh.iph->daddr); bh_lock_sock(dccp_v4_ctl_socket->sk); err = ip_build_and_send_pkt(skb, dccp_v4_ctl_socket->sk, @@ -779,6 +745,7 @@ EXPORT_SYMBOL_GPL(dccp_v4_do_rcv); int dccp_invalid_packet(struct sk_buff *skb) { const struct dccp_hdr *dh; + unsigned int cscov; if (skb->pkt_type != PACKET_HOST) return 1; @@ -830,6 +797,22 @@ int dccp_invalid_packet(struct sk_buff *skb) return 1; } + /* + * If P.CsCov is too large for the packet size, drop packet and return. + * This must come _before_ checksumming (not as RFC 4340 suggests). + */ + cscov = dccp_csum_coverage(skb); + if (cscov > skb->len) { + LIMIT_NETDEBUG(KERN_WARNING + "DCCP: P.CsCov %u exceeds packet length %d\n", + dh->dccph_cscov, skb->len); + return 1; + } + + /* If header checksum is incorrect, drop packet and return. + * (This step is completed in the AF-dependent functions.) */ + skb->csum = skb_checksum(skb, 0, cscov, 0); + return 0; } @@ -840,16 +823,17 @@ static int dccp_v4_rcv(struct sk_buff *skb) { const struct dccp_hdr *dh; struct sock *sk; + int min_cov; - /* Step 1: Check header basics: */ + /* Step 1: Check header basics */ if (dccp_invalid_packet(skb)) goto discard_it; - /* If the header checksum is incorrect, drop packet and return */ - if (dccp_v4_verify_checksum(skb, skb->nh.iph->saddr, - skb->nh.iph->daddr) < 0) { - LIMIT_NETDEBUG(KERN_WARNING "%s: incorrect header checksum\n", + /* Step 1: If header checksum is incorrect, drop packet and return */ + if (dccp_v4_csum_finish(skb, skb->nh.iph->saddr, skb->nh.iph->daddr)) { + LIMIT_NETDEBUG(KERN_WARNING + "%s: dropped packet with invalid checksum\n", __FUNCTION__); goto discard_it; } @@ -905,6 +889,21 @@ static int dccp_v4_rcv(struct sk_buff *skb) goto no_dccp_socket; } + /* + * RFC 4340, sec. 9.2.1: Minimum Checksum Coverage + * o if MinCsCov = 0, only packets with CsCov = 0 are accepted + * o if MinCsCov > 0, also accept packets with CsCov >= MinCsCov + */ + min_cov = dccp_sk(sk)->dccps_pcrlen; + if (dh->dccph_cscov && (min_cov == 0 || dh->dccph_cscov < min_cov)) { + dccp_pr_debug("Packet CsCov %d does not satisfy MinCsCov %d\n", + dh->dccph_cscov, min_cov); + /* FIXME: "Such packets SHOULD be reported using Data Dropped + * options (Section 11.7) with Drop Code 0, Protocol + * Constraints." */ + goto discard_and_relse; + } + if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb)) goto discard_and_relse; nf_reset(skb); diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 201801e1532..193b946fd03 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -58,12 +58,22 @@ static void dccp_v6_hash(struct sock *sk) } } -static inline u16 dccp_v6_check(struct dccp_hdr *dh, int len, - struct in6_addr *saddr, - struct in6_addr *daddr, - unsigned long base) +/* add pseudo-header to DCCP checksum stored in skb->csum */ +static inline u16 dccp_v6_csum_finish(struct sk_buff *skb, + struct in6_addr *saddr, + struct in6_addr *daddr) { - return csum_ipv6_magic(saddr, daddr, len, IPPROTO_DCCP, base); + return csum_ipv6_magic(saddr, daddr, skb->len, IPPROTO_DCCP, skb->csum); +} + +static inline void dccp_v6_send_check(struct sock *sk, int unused_value, + struct sk_buff *skb) +{ + struct ipv6_pinfo *np = inet6_sk(sk); + struct dccp_hdr *dh = dccp_hdr(skb); + + dccp_csum_outgoing(skb); + dh->dccph_checksum = dccp_v6_csum_finish(skb, &np->saddr, &np->daddr); } static __u32 dccp_v6_init_sequence(struct sock *sk, struct sk_buff *skb) @@ -280,12 +290,9 @@ static int dccp_v6_send_response(struct sock *sk, struct request_sock *req, if (skb != NULL) { struct dccp_hdr *dh = dccp_hdr(skb); - dh->dccph_checksum = dccp_v6_check(dh, skb->len, - &ireq6->loc_addr, - &ireq6->rmt_addr, - csum_partial((char *)dh, - skb->len, - skb->csum)); + dh->dccph_checksum = dccp_v6_csum_finish(skb, + &ireq6->loc_addr, + &ireq6->rmt_addr); ipv6_addr_copy(&fl.fl6_dst, &ireq6->rmt_addr); err = ip6_xmit(sk, skb, &fl, opt, 0); if (err == NET_XMIT_CN) @@ -305,18 +312,6 @@ static void dccp_v6_reqsk_destructor(struct request_sock *req) kfree_skb(inet6_rsk(req)->pktopts); } -static void dccp_v6_send_check(struct sock *sk, int len, struct sk_buff *skb) -{ - struct ipv6_pinfo *np = inet6_sk(sk); - struct dccp_hdr *dh = dccp_hdr(skb); - - dh->dccph_checksum = csum_ipv6_magic(&np->saddr, &np->daddr, - len, IPPROTO_DCCP, - csum_partial((char *)dh, - dh->dccph_doff << 2, - skb->csum)); -} - static void dccp_v6_ctl_send_reset(struct sk_buff *rxskb) { struct dccp_hdr *rxdh = dccp_hdr(rxskb), *dh; @@ -360,12 +355,14 @@ static void dccp_v6_ctl_send_reset(struct sk_buff *rxskb) dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), DCCP_SKB_CB(rxskb)->dccpd_seq); + dccp_csum_outgoing(skb); + dh->dccph_checksum = dccp_v6_csum_finish(skb, &rxskb->nh.ipv6h->saddr, + &rxskb->nh.ipv6h->daddr); + memset(&fl, 0, sizeof(fl)); ipv6_addr_copy(&fl.fl6_dst, &rxskb->nh.ipv6h->saddr); ipv6_addr_copy(&fl.fl6_src, &rxskb->nh.ipv6h->daddr); - dh->dccph_checksum = csum_ipv6_magic(&fl.fl6_src, &fl.fl6_dst, - sizeof(*dh), IPPROTO_DCCP, - skb->csum); + fl.proto = IPPROTO_DCCP; fl.oif = inet6_iif(rxskb); fl.fl_ip_dport = dh->dccph_dport; @@ -825,12 +822,22 @@ static int dccp_v6_rcv(struct sk_buff **pskb) const struct dccp_hdr *dh; struct sk_buff *skb = *pskb; struct sock *sk; + int min_cov; - /* Step 1: Check header basics: */ + /* Step 1: Check header basics */ if (dccp_invalid_packet(skb)) goto discard_it; + /* Step 1: If header checksum is incorrect, drop packet and return. */ + if (dccp_v6_csum_finish(skb, &skb->nh.ipv6h->saddr, + &skb->nh.ipv6h->daddr)) { + LIMIT_NETDEBUG(KERN_WARNING + "%s: dropped packet with invalid checksum\n", + __FUNCTION__); + goto discard_it; + } + dh = dccp_hdr(skb); DCCP_SKB_CB(skb)->dccpd_seq = dccp_hdr_seq(skb); @@ -869,6 +876,19 @@ static int dccp_v6_rcv(struct sk_buff **pskb) goto no_dccp_socket; } + /* + * RFC 4340, sec. 9.2.1: Minimum Checksum Coverage + * o if MinCsCov = 0, only packets with CsCov = 0 are accepted + * o if MinCsCov > 0, also accept packets with CsCov >= MinCsCov + */ + min_cov = dccp_sk(sk)->dccps_pcrlen; + if (dh->dccph_cscov && (min_cov == 0 || dh->dccph_cscov < min_cov)) { + dccp_pr_debug("Packet CsCov %d does not satisfy MinCsCov %d\n", + dh->dccph_cscov, min_cov); + /* FIXME: send Data Dropped option (see also dccp_v4_rcv) */ + goto discard_and_relse; + } + if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb)) goto discard_and_relse; diff --git a/net/dccp/output.c b/net/dccp/output.c index 51654975e8e..992caedd772 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -96,6 +96,7 @@ static int dccp_transmit_skb(struct sock *sk, struct sk_buff *skb) dh->dccph_dport = inet->dport; dh->dccph_doff = (dccp_header_size + dcb->dccpd_opt_len) / 4; dh->dccph_ccval = dcb->dccpd_ccval; + dh->dccph_cscov = dp->dccps_pcslen; /* XXX For now we're using only 48 bits sequence numbers */ dh->dccph_x = 1; @@ -115,7 +116,7 @@ static int dccp_transmit_skb(struct sock *sk, struct sk_buff *skb) break; } - icsk->icsk_af_ops->send_check(sk, skb->len, skb); + icsk->icsk_af_ops->send_check(sk, 0, skb); if (set_ack) dccp_event_ack_sent(sk); @@ -329,7 +330,6 @@ struct sk_buff *dccp_make_response(struct sock *sk, struct dst_entry *dst, skb_reserve(skb, sk->sk_prot->max_header); skb->dst = dst_clone(dst); - skb->csum = 0; dreq = dccp_rsk(req); DCCP_SKB_CB(skb)->dccpd_type = DCCP_PKT_RESPONSE; @@ -352,6 +352,8 @@ struct sk_buff *dccp_make_response(struct sock *sk, struct dst_entry *dst, dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), dreq->dreq_isr); dccp_hdr_response(skb)->dccph_resp_service = dreq->dreq_service; + dccp_csum_outgoing(skb); + DCCP_INC_STATS(DCCP_MIB_OUTSEGS); return skb; } @@ -376,7 +378,6 @@ static struct sk_buff *dccp_make_reset(struct sock *sk, struct dst_entry *dst, skb_reserve(skb, sk->sk_prot->max_header); skb->dst = dst_clone(dst); - skb->csum = 0; dccp_inc_seqno(&dp->dccps_gss); @@ -401,7 +402,7 @@ static struct sk_buff *dccp_make_reset(struct sock *sk, struct dst_entry *dst, dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), dp->dccps_gsr); dccp_hdr_reset(skb)->dccph_reset_code = code; - inet_csk(sk)->icsk_af_ops->send_check(sk, skb->len, skb); + inet_csk(sk)->icsk_af_ops->send_check(sk, 0, skb); DCCP_INC_STATS(DCCP_MIB_OUTSEGS); return skb; @@ -475,7 +476,6 @@ int dccp_connect(struct sock *sk) skb_reserve(skb, sk->sk_prot->max_header); DCCP_SKB_CB(skb)->dccpd_type = DCCP_PKT_REQUEST; - skb->csum = 0; dccp_skb_entail(sk, skb); dccp_transmit_skb(sk, skb_clone(skb, GFP_KERNEL)); @@ -507,7 +507,6 @@ void dccp_send_ack(struct sock *sk) /* Reserve space for headers */ skb_reserve(skb, sk->sk_prot->max_header); - skb->csum = 0; DCCP_SKB_CB(skb)->dccpd_type = DCCP_PKT_ACK; dccp_transmit_skb(sk, skb); } @@ -561,7 +560,6 @@ void dccp_send_sync(struct sock *sk, const u64 seq, /* Reserve space for headers and prepare control bits. */ skb_reserve(skb, sk->sk_prot->max_header); - skb->csum = 0; DCCP_SKB_CB(skb)->dccpd_type = pkt_type; DCCP_SKB_CB(skb)->dccpd_seq = seq; @@ -587,7 +585,6 @@ void dccp_send_close(struct sock *sk, const int active) /* Reserve space for headers and prepare control bits. */ skb_reserve(skb, sk->sk_prot->max_header); - skb->csum = 0; DCCP_SKB_CB(skb)->dccpd_type = dp->dccps_role == DCCP_ROLE_CLIENT ? DCCP_PKT_CLOSE : DCCP_PKT_CLOSEREQ; diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 047d170a363..db54e557eff 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -472,7 +472,6 @@ static int do_dccp_setsockopt(struct sock *sk, int level, int optname, case DCCP_SOCKOPT_PACKET_SIZE: dp->dccps_packet_size = val; break; - case DCCP_SOCKOPT_CHANGE_L: if (optlen != sizeof(struct dccp_so_feat)) err = -EINVAL; @@ -481,7 +480,6 @@ static int do_dccp_setsockopt(struct sock *sk, int level, int optname, (struct dccp_so_feat __user *) optval); break; - case DCCP_SOCKOPT_CHANGE_R: if (optlen != sizeof(struct dccp_so_feat)) err = -EINVAL; @@ -490,12 +488,26 @@ static int do_dccp_setsockopt(struct sock *sk, int level, int optname, (struct dccp_so_feat __user *) optval); break; - + case DCCP_SOCKOPT_SEND_CSCOV: /* sender side, RFC 4340, sec. 9.2 */ + if (val < 0 || val > 15) + err = -EINVAL; + else + dp->dccps_pcslen = val; + break; + case DCCP_SOCKOPT_RECV_CSCOV: /* receiver side, RFC 4340 sec. 9.2.1 */ + if (val < 0 || val > 15) + err = -EINVAL; + else { + dp->dccps_pcrlen = val; + /* FIXME: add feature negotiation, + * ChangeL(MinimumChecksumCoverage, val) */ + } + break; default: err = -ENOPROTOOPT; break; } - + release_sock(sk); return err; } @@ -575,6 +587,12 @@ static int do_dccp_getsockopt(struct sock *sk, int level, int optname, case DCCP_SOCKOPT_SERVICE: return dccp_getsockopt_service(sk, len, (__be32 __user *)optval, optlen); + case DCCP_SOCKOPT_SEND_CSCOV: + val = dp->dccps_pcslen; + break; + case DCCP_SOCKOPT_RECV_CSCOV: + val = dp->dccps_pcrlen; + break; case 128 ... 191: return ccid_hc_rx_getsockopt(dp->dccps_hc_rx_ccid, sk, optname, len, (u32 __user *)optval, optlen); -- cgit v1.2.3-70-g09d2 From 339bf98ffc6a8d8eb16fc532ac57ffbced2f8a68 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Fri, 10 Nov 2006 14:10:15 -0800 Subject: [NETLINK]: Do precise netlink message allocations where possible Account for the netlink message header size directly in nlmsg_new() instead of relying on the caller calculate it correctly. Replaces error handling of message construction functions when constructing notifications with bug traps since a failure implies a bug in calculating the size of the skb. Signed-off-by: Thomas Graf Acked-by: Paul Moore Signed-off-by: David S. Miller --- include/linux/netlink.h | 1 + include/net/fib_rules.h | 1 + include/net/netlink.h | 9 ++--- kernel/taskstats.c | 3 +- net/bridge/br_netlink.c | 21 ++++++++---- net/core/fib_rules.c | 24 +++++++++++--- net/core/neighbour.c | 17 +++++++--- net/core/rtnetlink.c | 39 ++++++++++++++-------- net/decnet/dn_rules.c | 6 ++++ net/decnet/dn_table.c | 34 ++++++++++++++++--- net/ipv4/devinet.c | 18 +++++++--- net/ipv4/fib_rules.c | 8 +++++ net/ipv4/fib_semantics.c | 36 ++++++++++++++++---- net/ipv6/addrconf.c | 70 ++++++++++++++++++--------------------- net/ipv6/fib6_rules.c | 7 ++++ net/ipv6/route.c | 23 +++++++++---- net/netlabel/netlabel_cipso_v4.c | 2 +- net/netlabel/netlabel_mgmt.c | 4 +-- net/netlabel/netlabel_unlabeled.c | 2 +- net/netlink/af_netlink.c | 13 ++++---- net/netlink/genetlink.c | 2 +- 21 files changed, 233 insertions(+), 107 deletions(-) (limited to 'include') diff --git a/include/linux/netlink.h b/include/linux/netlink.h index 66411622e06..e61e1e13842 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -174,6 +174,7 @@ int netlink_sendskb(struct sock *sk, struct sk_buff *skb, int protocol); */ #define NLMSG_GOODORDER 0 #define NLMSG_GOODSIZE (SKB_MAX_ORDER(0, NLMSG_GOODORDER)) +#define NLMSG_DEFAULT_SIZE (NLMSG_GOODSIZE - NLMSG_HDRLEN) struct netlink_callback diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h index e4ba781d289..bc3c26494c3 100644 --- a/include/net/fib_rules.h +++ b/include/net/fib_rules.h @@ -52,6 +52,7 @@ struct fib_rules_ops struct nlmsghdr *, struct fib_rule_hdr *); u32 (*default_pref)(void); + size_t (*nlmsg_payload)(struct fib_rule *); int nlgroup; struct nla_policy *policy; diff --git a/include/net/netlink.h b/include/net/netlink.h index ce5cba19c39..30021339157 100644 --- a/include/net/netlink.h +++ b/include/net/netlink.h @@ -500,14 +500,15 @@ static inline struct nlmsghdr *nlmsg_put_answer(struct sk_buff *skb, /** * nlmsg_new - Allocate a new netlink message - * @size: maximum size of message + * @payload: size of the message payload * @flags: the type of memory to allocate. * - * Use NLMSG_GOODSIZE if size isn't know and you need a good default size. + * Use NLMSG_DEFAULT_SIZE if the size of the payload isn't known + * and a good default is needed. */ -static inline struct sk_buff *nlmsg_new(int size, gfp_t flags) +static inline struct sk_buff *nlmsg_new(size_t payload, gfp_t flags) { - return alloc_skb(size, flags); + return alloc_skb(nlmsg_total_size(payload), flags); } /** diff --git a/kernel/taskstats.c b/kernel/taskstats.c index f45c5e70773..4f3f0e48c84 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -77,8 +77,7 @@ static int prepare_reply(struct genl_info *info, u8 cmd, struct sk_buff **skbp, /* * If new attributes are added, please revisit this allocation */ - size = nlmsg_total_size(genlmsg_total_size(size)); - skb = nlmsg_new(size, GFP_KERNEL); + skb = nlmsg_new(genlmsg_total_size(size), GFP_KERNEL); if (!skb) return -ENOMEM; diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index 8f661195d09..15d6efbe751 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -15,6 +15,18 @@ #include #include "br_private.h" +static inline size_t br_nlmsg_size(void) +{ + return NLMSG_ALIGN(sizeof(struct ifinfomsg)) + + nla_total_size(IFNAMSIZ) /* IFLA_IFNAME */ + + nla_total_size(MAX_ADDR_LEN) /* IFLA_ADDRESS */ + + nla_total_size(4) /* IFLA_MASTER */ + + nla_total_size(4) /* IFLA_MTU */ + + nla_total_size(4) /* IFLA_LINK */ + + nla_total_size(1) /* IFLA_OPERSTATE */ + + nla_total_size(1); /* IFLA_PROTINFO */ +} + /* * Create one netlink message for one interface * Contains port and master info as well as carrier and bridge state. @@ -77,19 +89,16 @@ rtattr_failure: void br_ifinfo_notify(int event, struct net_bridge_port *port) { struct sk_buff *skb; - int payload = sizeof(struct ifinfomsg) + 128; int err = -ENOBUFS; pr_debug("bridge notify event=%d\n", event); - skb = nlmsg_new(nlmsg_total_size(payload), GFP_ATOMIC); + skb = nlmsg_new(br_nlmsg_size(), GFP_ATOMIC); if (skb == NULL) goto errout; err = br_fill_ifinfo(skb, port, 0, 0, event, 0); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in br_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, 0, RTNLGRP_LINK, NULL, GFP_ATOMIC); errout: diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index 4148e274a20..1df6cd4568d 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -306,6 +306,22 @@ errout: return err; } +static inline size_t fib_rule_nlmsg_size(struct fib_rules_ops *ops, + struct fib_rule *rule) +{ + size_t payload = NLMSG_ALIGN(sizeof(struct fib_rule_hdr)) + + nla_total_size(IFNAMSIZ) /* FRA_IFNAME */ + + nla_total_size(4) /* FRA_PRIORITY */ + + nla_total_size(4) /* FRA_TABLE */ + + nla_total_size(4) /* FRA_FWMARK */ + + nla_total_size(4); /* FRA_FWMASK */ + + if (ops->nlmsg_payload) + payload += ops->nlmsg_payload(rule); + + return payload; +} + static int fib_nl_fill_rule(struct sk_buff *skb, struct fib_rule *rule, u32 pid, u32 seq, int type, int flags, struct fib_rules_ops *ops) @@ -384,15 +400,13 @@ static void notify_rule_change(int event, struct fib_rule *rule, struct sk_buff *skb; int err = -ENOBUFS; - skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + skb = nlmsg_new(fib_rule_nlmsg_size(ops, rule), GFP_KERNEL); if (skb == NULL) goto errout; err = fib_nl_fill_rule(skb, rule, pid, nlh->nlmsg_seq, event, 0, ops); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in fib_rule_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, pid, ops->nlgroup, nlh, GFP_KERNEL); errout: diff --git a/net/core/neighbour.c b/net/core/neighbour.c index b4b478353b2..0e097ba14d7 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -2410,20 +2410,27 @@ static struct file_operations neigh_stat_seq_fops = { #endif /* CONFIG_PROC_FS */ #ifdef CONFIG_ARPD +static inline size_t neigh_nlmsg_size(void) +{ + return NLMSG_ALIGN(sizeof(struct ndmsg)) + + nla_total_size(MAX_ADDR_LEN) /* NDA_DST */ + + nla_total_size(MAX_ADDR_LEN) /* NDA_LLADDR */ + + nla_total_size(sizeof(struct nda_cacheinfo)) + + nla_total_size(4); /* NDA_PROBES */ +} + static void __neigh_notify(struct neighbour *n, int type, int flags) { struct sk_buff *skb; int err = -ENOBUFS; - skb = nlmsg_new(NLMSG_GOODSIZE, GFP_ATOMIC); + skb = nlmsg_new(neigh_nlmsg_size(), GFP_ATOMIC); if (skb == NULL) goto errout; err = neigh_fill_info(skb, n, 0, 0, type, flags); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in neigh_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, 0, RTNLGRP_NEIGH, NULL, GFP_ATOMIC); errout: diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 02f3c794789..50d6cb40c6e 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -273,6 +273,25 @@ static void copy_rtnl_link_stats(struct rtnl_link_stats *a, a->tx_compressed = b->tx_compressed; }; +static inline size_t if_nlmsg_size(int iwbuflen) +{ + return NLMSG_ALIGN(sizeof(struct ifinfomsg)) + + nla_total_size(IFNAMSIZ) /* IFLA_IFNAME */ + + nla_total_size(IFNAMSIZ) /* IFLA_QDISC */ + + nla_total_size(sizeof(struct rtnl_link_ifmap)) + + nla_total_size(sizeof(struct rtnl_link_stats)) + + nla_total_size(MAX_ADDR_LEN) /* IFLA_ADDRESS */ + + nla_total_size(MAX_ADDR_LEN) /* IFLA_BROADCAST */ + + nla_total_size(4) /* IFLA_TXQLEN */ + + nla_total_size(4) /* IFLA_WEIGHT */ + + nla_total_size(4) /* IFLA_MTU */ + + nla_total_size(4) /* IFLA_LINK */ + + nla_total_size(4) /* IFLA_MASTER */ + + nla_total_size(1) /* IFLA_OPERSTATE */ + + nla_total_size(1) /* IFLA_LINKMODE */ + + nla_total_size(iwbuflen); +} + static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, void *iwbuf, int iwbuflen, int type, u32 pid, u32 seq, u32 change, unsigned int flags) @@ -558,7 +577,7 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) struct sk_buff *nskb; char *iw_buf = NULL, *iw = NULL; int iw_buf_len = 0; - int err, payload; + int err; err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy); if (err < 0) @@ -587,9 +606,7 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) } #endif /* CONFIG_NET_WIRELESS_RTNETLINK */ - payload = NLMSG_ALIGN(sizeof(struct ifinfomsg) + - nla_total_size(iw_buf_len)); - nskb = nlmsg_new(nlmsg_total_size(payload), GFP_KERNEL); + nskb = nlmsg_new(if_nlmsg_size(iw_buf_len), GFP_KERNEL); if (nskb == NULL) { err = -ENOBUFS; goto errout; @@ -597,10 +614,8 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) err = rtnl_fill_ifinfo(nskb, dev, iw, iw_buf_len, RTM_NEWLINK, NETLINK_CB(skb).pid, nlh->nlmsg_seq, 0, 0); - if (err <= 0) { - kfree_skb(nskb); - goto errout; - } + /* failure impilies BUG in if_nlmsg_size or wireless_rtnetlink_get */ + BUG_ON(err < 0); err = rtnl_unicast(nskb, NETLINK_CB(skb).pid); errout: @@ -639,15 +654,13 @@ void rtmsg_ifinfo(int type, struct net_device *dev, unsigned change) struct sk_buff *skb; int err = -ENOBUFS; - skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + skb = nlmsg_new(if_nlmsg_size(0), GFP_KERNEL); if (skb == NULL) goto errout; err = rtnl_fill_ifinfo(skb, dev, NULL, 0, type, 0, 0, change, 0); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in if_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, 0, RTNLGRP_LINK, NULL, GFP_KERNEL); errout: diff --git a/net/decnet/dn_rules.c b/net/decnet/dn_rules.c index e32d0c3d5a9..b7dfd04a963 100644 --- a/net/decnet/dn_rules.c +++ b/net/decnet/dn_rules.c @@ -241,6 +241,12 @@ static u32 dn_fib_rule_default_pref(void) return 0; } +static size_t dn_fib_rule_nlmsg_payload(struct fib_rule *rule) +{ + return nla_total_size(2) /* dst */ + + nla_total_size(2); /* src */ +} + int dn_fib_dump_rules(struct sk_buff *skb, struct netlink_callback *cb) { return fib_rules_dump(skb, cb, AF_DECnet); diff --git a/net/decnet/dn_table.c b/net/decnet/dn_table.c index 317904bb589..e74b744254a 100644 --- a/net/decnet/dn_table.c +++ b/net/decnet/dn_table.c @@ -263,6 +263,32 @@ static int dn_fib_nh_match(struct rtmsg *r, struct nlmsghdr *nlh, struct dn_kern return 0; } +static inline size_t dn_fib_nlmsg_size(struct dn_fib_info *fi) +{ + size_t payload = NLMSG_ALIGN(struct rtmsg) + + nla_total_size(4) /* RTA_TABLE */ + + nla_total_size(2) /* RTA_DST */ + + nla_total_size(4); /* RTA_PRIORITY */ + + /* space for nested metrics */ + payload += nla_total_size((RTAX_MAX * nla_total_size(4))); + + if (fi->fib_nhs) { + /* Also handles the special case fib_nhs == 1 */ + + /* each nexthop is packed in an attribute */ + size_t nhsize = nla_total_size(sizeof(struct rtnexthop)); + + /* may contain a gateway attribute */ + nhsize += nla_total_size(4); + + /* all nexthops are packed in a nested attribute */ + payload += nla_total_size(fi->fib_nhs * nhsize); + } + + return payload; +} + static int dn_fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event, u32 tb_id, u8 type, u8 scope, void *dst, int dst_len, struct dn_fib_info *fi, unsigned int flags) @@ -335,17 +361,15 @@ static void dn_rtmsg_fib(int event, struct dn_fib_node *f, int z, u32 tb_id, u32 pid = req ? req->pid : 0; int err = -ENOBUFS; - skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + skb = nlmsg_new(dn_fib_nlmsg_size(DN_FIB_INFO(f), GFP_KERNEL)); if (skb == NULL) goto errout; err = dn_fib_dump_info(skb, pid, nlh->nlmsg_seq, event, tb_id, f->fn_type, f->fn_scope, &f->fn_key, z, DN_FIB_INFO(f), 0); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in dn_fib_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, pid, RTNLGRP_DECnet_ROUTE, nlh, GFP_KERNEL); errout: diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 7602c79a389..f38cbbae0ae 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -1120,6 +1120,16 @@ static struct notifier_block ip_netdev_notifier = { .notifier_call =inetdev_event, }; +static inline size_t inet_nlmsg_size(void) +{ + return NLMSG_ALIGN(sizeof(struct ifaddrmsg)) + + nla_total_size(4) /* IFA_ADDRESS */ + + nla_total_size(4) /* IFA_LOCAL */ + + nla_total_size(4) /* IFA_BROADCAST */ + + nla_total_size(4) /* IFA_ANYCAST */ + + nla_total_size(IFNAMSIZ); /* IFA_LABEL */ +} + static int inet_fill_ifaddr(struct sk_buff *skb, struct in_ifaddr *ifa, u32 pid, u32 seq, int event, unsigned int flags) { @@ -1208,15 +1218,13 @@ static void rtmsg_ifa(int event, struct in_ifaddr* ifa, struct nlmsghdr *nlh, u32 seq = nlh ? nlh->nlmsg_seq : 0; int err = -ENOBUFS; - skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + skb = nlmsg_new(inet_nlmsg_size(), GFP_KERNEL); if (skb == NULL) goto errout; err = inet_fill_ifaddr(skb, ifa, pid, seq, event, 0); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in inet_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, pid, RTNLGRP_IPV4_IFADDR, nlh, GFP_KERNEL); errout: diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c index fd4a8cd4c06..b837c33e040 100644 --- a/net/ipv4/fib_rules.c +++ b/net/ipv4/fib_rules.c @@ -299,6 +299,13 @@ static u32 fib4_rule_default_pref(void) return 0; } +static size_t fib4_rule_nlmsg_payload(struct fib_rule *rule) +{ + return nla_total_size(4) /* dst */ + + nla_total_size(4) /* src */ + + nla_total_size(4); /* flow */ +} + static struct fib_rules_ops fib4_rules_ops = { .family = AF_INET, .rule_size = sizeof(struct fib4_rule), @@ -308,6 +315,7 @@ static struct fib_rules_ops fib4_rules_ops = { .compare = fib4_rule_compare, .fill = fib4_rule_fill, .default_pref = fib4_rule_default_pref, + .nlmsg_payload = fib4_rule_nlmsg_payload, .nlgroup = RTNLGRP_IPV4_RULE, .policy = fib4_rule_policy, .rules_list = &fib4_rules, diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index 884d176e008..e63b8a98fb4 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -273,25 +273,49 @@ int ip_fib_check_default(__be32 gw, struct net_device *dev) return -1; } +static inline size_t fib_nlmsg_size(struct fib_info *fi) +{ + size_t payload = NLMSG_ALIGN(sizeof(struct rtmsg)) + + nla_total_size(4) /* RTA_TABLE */ + + nla_total_size(4) /* RTA_DST */ + + nla_total_size(4) /* RTA_PRIORITY */ + + nla_total_size(4); /* RTA_PREFSRC */ + + /* space for nested metrics */ + payload += nla_total_size((RTAX_MAX * nla_total_size(4))); + + if (fi->fib_nhs) { + /* Also handles the special case fib_nhs == 1 */ + + /* each nexthop is packed in an attribute */ + size_t nhsize = nla_total_size(sizeof(struct rtnexthop)); + + /* may contain flow and gateway attribute */ + nhsize += 2 * nla_total_size(4); + + /* all nexthops are packed in a nested attribute */ + payload += nla_total_size(fi->fib_nhs * nhsize); + } + + return payload; +} + void rtmsg_fib(int event, __be32 key, struct fib_alias *fa, int dst_len, u32 tb_id, struct nl_info *info) { struct sk_buff *skb; - int payload = sizeof(struct rtmsg) + 256; u32 seq = info->nlh ? info->nlh->nlmsg_seq : 0; int err = -ENOBUFS; - skb = nlmsg_new(nlmsg_total_size(payload), GFP_KERNEL); + skb = nlmsg_new(fib_nlmsg_size(fa->fa_info), GFP_KERNEL); if (skb == NULL) goto errout; err = fib_dump_info(skb, info->pid, seq, event, tb_id, fa->fa_type, fa->fa_scope, key, dst_len, fa->fa_tos, fa->fa_info, 0); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in fib_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, info->pid, RTNLGRP_IPV4_ROUTE, info->nlh, GFP_KERNEL); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 6a98f68348c..967ea320a9c 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -3098,10 +3098,9 @@ static inline int rt_scope(int ifa_scope) static inline int inet6_ifaddr_msgsize(void) { - return nlmsg_total_size(sizeof(struct ifaddrmsg) + - nla_total_size(16) + - nla_total_size(sizeof(struct ifa_cacheinfo)) + - 128); + return NLMSG_ALIGN(sizeof(struct ifaddrmsg)) + + nla_total_size(16) /* IFA_ADDRESS */ + + nla_total_size(sizeof(struct ifa_cacheinfo)); } static int inet6_fill_ifaddr(struct sk_buff *skb, struct inet6_ifaddr *ifa, @@ -3329,10 +3328,8 @@ static int inet6_rtm_getaddr(struct sk_buff *in_skb, struct nlmsghdr* nlh, err = inet6_fill_ifaddr(skb, ifa, NETLINK_CB(in_skb).pid, nlh->nlmsg_seq, RTM_NEWADDR, 0); - if (err < 0) { - kfree_skb(skb); - goto errout_ifa; - } + /* failure implies BUG in inet6_ifaddr_msgsize() */ + BUG_ON(err < 0); err = rtnl_unicast(skb, NETLINK_CB(in_skb).pid); errout_ifa: @@ -3351,10 +3348,8 @@ static void inet6_ifa_notify(int event, struct inet6_ifaddr *ifa) goto errout; err = inet6_fill_ifaddr(skb, ifa, 0, 0, event, 0); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in inet6_ifaddr_msgsize() */ + BUG_ON(err < 0); err = rtnl_notify(skb, 0, RTNLGRP_IPV6_IFADDR, NULL, GFP_ATOMIC); errout: @@ -3397,16 +3392,19 @@ static void inline ipv6_store_devconf(struct ipv6_devconf *cnf, array[DEVCONF_PROXY_NDP] = cnf->proxy_ndp; } -/* Maximum length of ifinfomsg attributes */ -#define INET6_IFINFO_RTA_SPACE \ - RTA_SPACE(IFNAMSIZ) /* IFNAME */ + \ - RTA_SPACE(MAX_ADDR_LEN) /* ADDRESS */ + \ - RTA_SPACE(sizeof(u32)) /* MTU */ + \ - RTA_SPACE(sizeof(int)) /* LINK */ + \ - RTA_SPACE(0) /* PROTINFO */ + \ - RTA_SPACE(sizeof(u32)) /* FLAGS */ + \ - RTA_SPACE(sizeof(struct ifla_cacheinfo)) /* CACHEINFO */ + \ - RTA_SPACE(sizeof(__s32[DEVCONF_MAX])) /* CONF */ +static inline size_t inet6_if_nlmsg_size(void) +{ + return NLMSG_ALIGN(sizeof(struct ifinfomsg)) + + nla_total_size(IFNAMSIZ) /* IFLA_IFNAME */ + + nla_total_size(MAX_ADDR_LEN) /* IFLA_ADDRESS */ + + nla_total_size(4) /* IFLA_MTU */ + + nla_total_size(4) /* IFLA_LINK */ + + nla_total_size( /* IFLA_PROTINFO */ + nla_total_size(4) /* IFLA_INET6_FLAGS */ + + nla_total_size(sizeof(struct ifla_cacheinfo)) + + nla_total_size(DEVCONF_MAX * 4) /* IFLA_INET6_CONF */ + ); +} static int inet6_fill_ifinfo(struct sk_buff *skb, struct inet6_dev *idev, u32 pid, u32 seq, int event, unsigned int flags) @@ -3501,18 +3499,15 @@ static int inet6_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb) void inet6_ifinfo_notify(int event, struct inet6_dev *idev) { struct sk_buff *skb; - int payload = sizeof(struct ifinfomsg) + INET6_IFINFO_RTA_SPACE; int err = -ENOBUFS; - skb = nlmsg_new(nlmsg_total_size(payload), GFP_ATOMIC); + skb = nlmsg_new(inet6_if_nlmsg_size(), GFP_ATOMIC); if (skb == NULL) goto errout; err = inet6_fill_ifinfo(skb, idev, 0, 0, event, 0); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in inet6_if_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, 0, RTNLGRP_IPV6_IFADDR, NULL, GFP_ATOMIC); errout: @@ -3520,10 +3515,12 @@ errout: rtnl_set_sk_err(RTNLGRP_IPV6_IFADDR, err); } -/* Maximum length of prefix_cacheinfo attributes */ -#define INET6_PREFIX_RTA_SPACE \ - RTA_SPACE(sizeof(((struct prefix_info *)NULL)->prefix)) /* ADDRESS */ + \ - RTA_SPACE(sizeof(struct prefix_cacheinfo)) /* CACHEINFO */ +static inline size_t inet6_prefix_nlmsg_size(void) +{ + return NLMSG_ALIGN(sizeof(struct prefixmsg)) + + nla_total_size(sizeof(struct in6_addr)) + + nla_total_size(sizeof(struct prefix_cacheinfo)); +} static int inet6_fill_prefix(struct sk_buff *skb, struct inet6_dev *idev, struct prefix_info *pinfo, u32 pid, u32 seq, @@ -3569,18 +3566,15 @@ static void inet6_prefix_notify(int event, struct inet6_dev *idev, struct prefix_info *pinfo) { struct sk_buff *skb; - int payload = sizeof(struct prefixmsg) + INET6_PREFIX_RTA_SPACE; int err = -ENOBUFS; - skb = nlmsg_new(nlmsg_total_size(payload), GFP_ATOMIC); + skb = nlmsg_new(inet6_prefix_nlmsg_size(), GFP_ATOMIC); if (skb == NULL) goto errout; err = inet6_fill_prefix(skb, idev, pinfo, 0, 0, event, 0); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in inet6_prefix_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, 0, RTNLGRP_IPV6_PREFIX, NULL, GFP_ATOMIC); errout: diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index 25804cb69cf..d587dde5897 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -232,6 +232,12 @@ static u32 fib6_rule_default_pref(void) return 0x3FFF; } +static size_t fib6_rule_nlmsg_payload(struct fib_rule *rule) +{ + return nla_total_size(16) /* dst */ + + nla_total_size(16); /* src */ +} + static struct fib_rules_ops fib6_rules_ops = { .family = AF_INET6, .rule_size = sizeof(struct fib6_rule), @@ -241,6 +247,7 @@ static struct fib_rules_ops fib6_rules_ops = { .compare = fib6_rule_compare, .fill = fib6_rule_fill, .default_pref = fib6_rule_default_pref, + .nlmsg_payload = fib6_rule_nlmsg_payload, .nlgroup = RTNLGRP_IPV6_RULE, .policy = fib6_rule_policy, .rules_list = &fib6_rules, diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 0ad07c9087a..a6472cb9054 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2006,6 +2006,20 @@ int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) return ip6_route_add(&cfg); } +static inline size_t rt6_nlmsg_size(void) +{ + return NLMSG_ALIGN(sizeof(struct rtmsg)) + + nla_total_size(16) /* RTA_SRC */ + + nla_total_size(16) /* RTA_DST */ + + nla_total_size(16) /* RTA_GATEWAY */ + + nla_total_size(16) /* RTA_PREFSRC */ + + nla_total_size(4) /* RTA_TABLE */ + + nla_total_size(4) /* RTA_IIF */ + + nla_total_size(4) /* RTA_OIF */ + + nla_total_size(4) /* RTA_PRIORITY */ + + nla_total_size(sizeof(struct rta_cacheinfo)); +} + static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, struct in6_addr *dst, struct in6_addr *src, int iif, int type, u32 pid, u32 seq, @@ -2200,7 +2214,6 @@ void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info) struct sk_buff *skb; u32 pid = 0, seq = 0; struct nlmsghdr *nlh = NULL; - int payload = sizeof(struct rtmsg) + 256; int err = -ENOBUFS; if (info) { @@ -2210,15 +2223,13 @@ void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info) seq = nlh->nlmsg_seq; } - skb = nlmsg_new(nlmsg_total_size(payload), gfp_any()); + skb = nlmsg_new(rt6_nlmsg_size(), gfp_any()); if (skb == NULL) goto errout; err = rt6_fill_node(skb, rt, NULL, NULL, 0, event, pid, seq, 0, 0); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in rt6_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, pid, RTNLGRP_IPV6_ROUTE, nlh, gfp_any()); errout: diff --git a/net/netlabel/netlabel_cipso_v4.c b/net/netlabel/netlabel_cipso_v4.c index a6ce1d6d5c5..f1788bd290f 100644 --- a/net/netlabel/netlabel_cipso_v4.c +++ b/net/netlabel/netlabel_cipso_v4.c @@ -452,7 +452,7 @@ static int netlbl_cipsov4_list(struct sk_buff *skb, struct genl_info *info) } list_start: - ans_skb = nlmsg_new(NLMSG_GOODSIZE * nlsze_mult, GFP_KERNEL); + ans_skb = nlmsg_new(NLMSG_DEFAULT_SIZE * nlsze_mult, GFP_KERNEL); if (ans_skb == NULL) { ret_val = -ENOMEM; goto list_failure; diff --git a/net/netlabel/netlabel_mgmt.c b/net/netlabel/netlabel_mgmt.c index 53c9079ad2c..c529622ff0b 100644 --- a/net/netlabel/netlabel_mgmt.c +++ b/net/netlabel/netlabel_mgmt.c @@ -356,7 +356,7 @@ static int netlbl_mgmt_listdef(struct sk_buff *skb, struct genl_info *info) void *data; struct netlbl_dom_map *entry; - ans_skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + ans_skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (ans_skb == NULL) return -ENOMEM; data = netlbl_netlink_hdr_put(ans_skb, @@ -492,7 +492,7 @@ static int netlbl_mgmt_version(struct sk_buff *skb, struct genl_info *info) struct sk_buff *ans_skb = NULL; void *data; - ans_skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + ans_skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (ans_skb == NULL) return -ENOMEM; data = netlbl_netlink_hdr_put(ans_skb, diff --git a/net/netlabel/netlabel_unlabeled.c b/net/netlabel/netlabel_unlabeled.c index 1833ad233b3..219dccade4e 100644 --- a/net/netlabel/netlabel_unlabeled.c +++ b/net/netlabel/netlabel_unlabeled.c @@ -138,7 +138,7 @@ static int netlbl_unlabel_list(struct sk_buff *skb, struct genl_info *info) struct sk_buff *ans_skb; void *data; - ans_skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + ans_skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (ans_skb == NULL) goto list_failure; data = netlbl_netlink_hdr_put(ans_skb, diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index d527c8977b1..f61d81b3c61 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1148,7 +1148,7 @@ static int netlink_sendmsg(struct kiocb *kiocb, struct socket *sock, if (len > sk->sk_sndbuf - 32) goto out; err = -ENOBUFS; - skb = nlmsg_new(len, GFP_KERNEL); + skb = alloc_skb(len, GFP_KERNEL); if (skb==NULL) goto out; @@ -1435,14 +1435,13 @@ void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err) struct sk_buff *skb; struct nlmsghdr *rep; struct nlmsgerr *errmsg; - int size; + size_t payload = sizeof(*errmsg); - if (err == 0) - size = nlmsg_total_size(sizeof(*errmsg)); - else - size = nlmsg_total_size(sizeof(*errmsg) + nlmsg_len(nlh)); + /* error messages get the original request appened */ + if (err) + payload += nlmsg_len(nlh); - skb = nlmsg_new(size, GFP_KERNEL); + skb = nlmsg_new(payload, GFP_KERNEL); if (!skb) { struct sock *sk; diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c index 49bc2db7982..70d60c81889 100644 --- a/net/netlink/genetlink.c +++ b/net/netlink/genetlink.c @@ -480,7 +480,7 @@ static struct sk_buff *ctrl_build_msg(struct genl_family *family, u32 pid, struct sk_buff *skb; int err; - skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (skb == NULL) return ERR_PTR(-ENOBUFS); -- cgit v1.2.3-70-g09d2 From 2e2e9e92bd723244ea20fa488b1780111f2b05e1 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Mon, 13 Nov 2006 13:23:52 -0200 Subject: [DCCP]: Add sysctls to control retransmission behaviour This adds 3 sysctls which govern the retransmission behaviour of DCCP control packets (3way handshake, feature negotiation). It removes 4 FIXMEs from the code. The close resemblance of sysctl variables to their TCP analogues is emphasised not only by their name, but also by giving them the same initial values. This is useful since there is not much practical experience with DCCP yet. Furthermore, with regard to the previous patch, it is now possible to limit the number of keepalive-Responses by setting net.dccp.default.request_retries (also a bit like in TCP). Lastly, added documentation of all existing DCCP sysctls. Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo --- Documentation/networking/dccp.txt | 41 +++++++++++++++++++++++++++++++++++++++ include/linux/sysctl.h | 3 +++ net/dccp/dccp.h | 11 +++++++++++ net/dccp/feat.h | 7 ------- net/dccp/proto.c | 1 + net/dccp/sysctl.c | 25 ++++++++++++++++++++++++ net/dccp/timer.c | 16 +++++++++------ 7 files changed, 91 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/Documentation/networking/dccp.txt b/Documentation/networking/dccp.txt index a8142a81038..c2328c862e9 100644 --- a/Documentation/networking/dccp.txt +++ b/Documentation/networking/dccp.txt @@ -63,6 +63,47 @@ DCCP_SOCKOPT_SEND_CSCOV is for the receiver and has a different meaning: it coverage value are also acceptable. The higher the number, the more restrictive this setting (see [RFC 4340, sec. 9.2.1]). +Sysctl variables +================ +Several DCCP default parameters can be managed by the following sysctls +(sysctl net.dccp.default or /proc/sys/net/dccp/default): + +request_retries + The number of active connection initiation retries (the number of + Requests minus one) before timing out. In addition, it also governs + the behaviour of the other, passive side: this variable also sets + the number of times DCCP repeats sending a Response when the initial + handshake does not progress from RESPOND to OPEN (i.e. when no Ack + is received after the initial Request). This value should be greater + than 0, suggested is less than 10. Analogue of tcp_syn_retries. + +retries1 + How often a DCCP Response is retransmitted until the listening DCCP + side considers its connecting peer dead. Analogue of tcp_retries1. + +retries2 + The number of times a general DCCP packet is retransmitted. This has + importance for retransmitted acknowledgments and feature negotiation, + data packets are never retransmitted. Analogue of tcp_retries2. + +send_ndp = 1 + Whether or not to send NDP count options (sec. 7.7.2). + +send_ackvec = 1 + Whether or not to send Ack Vector options (sec. 11.5). + +ack_ratio = 2 + The default Ack Ratio (sec. 11.3) to use. + +tx_ccid = 2 + Default CCID for the sender-receiver half-connection. + +rx_ccid = 2 + Default CCID for the receiver-sender half-connection. + +seq_window = 100 + The initial sequence window (sec. 7.5.2). + Notes ===== diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 0725441621d..2e8c5ad8279 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -614,6 +614,9 @@ enum { NET_DCCP_DEFAULT_ACK_RATIO = 4, NET_DCCP_DEFAULT_SEND_ACKVEC = 5, NET_DCCP_DEFAULT_SEND_NDP = 6, + NET_DCCP_DEFAULT_REQ_RETRIES = 7, + NET_DCCP_DEFAULT_RETRIES1 = 8, + NET_DCCP_DEFAULT_RETRIES2 = 9, }; /* /proc/sys/net/ipx */ diff --git a/net/dccp/dccp.h b/net/dccp/dccp.h index d5c414bf781..e7b96074a1b 100644 --- a/net/dccp/dccp.h +++ b/net/dccp/dccp.h @@ -64,6 +64,17 @@ extern void dccp_time_wait(struct sock *sk, int state, int timeo); #define DCCP_XMIT_TIMEO 30000 /* Time/msecs for blocking transmit per packet */ +/* sysctl variables for DCCP */ +extern int sysctl_dccp_request_retries; +extern int sysctl_dccp_retries1; +extern int sysctl_dccp_retries2; +extern int dccp_feat_default_sequence_window; +extern int dccp_feat_default_rx_ccid; +extern int dccp_feat_default_tx_ccid; +extern int dccp_feat_default_ack_ratio; +extern int dccp_feat_default_send_ack_vector; +extern int dccp_feat_default_send_ndp_count; + /* is seq1 < seq2 ? */ static inline int before48(const u64 seq1, const u64 seq2) { diff --git a/net/dccp/feat.h b/net/dccp/feat.h index cee553d416c..6048373c718 100644 --- a/net/dccp/feat.h +++ b/net/dccp/feat.h @@ -26,11 +26,4 @@ extern void dccp_feat_clean(struct dccp_minisock *dmsk); extern int dccp_feat_clone(struct sock *oldsk, struct sock *newsk); extern int dccp_feat_init(struct dccp_minisock *dmsk); -extern int dccp_feat_default_sequence_window; -extern int dccp_feat_default_rx_ccid; -extern int dccp_feat_default_tx_ccid; -extern int dccp_feat_default_ack_ratio; -extern int dccp_feat_default_send_ack_vector; -extern int dccp_feat_default_send_ndp_count; - #endif /* _DCCP_FEAT_H */ diff --git a/net/dccp/proto.c b/net/dccp/proto.c index db54e557eff..9c9c08cffda 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -212,6 +212,7 @@ int dccp_init_sock(struct sock *sk, const __u8 ctl_sock_initialized) dccp_init_xmit_timers(sk); icsk->icsk_rto = DCCP_TIMEOUT_INIT; + icsk->icsk_syn_retries = sysctl_dccp_request_retries; sk->sk_state = DCCP_CLOSED; sk->sk_write_space = dccp_write_space; icsk->icsk_sync_mss = dccp_sync_mss; diff --git a/net/dccp/sysctl.c b/net/dccp/sysctl.c index 38bc157876f..7b09f217998 100644 --- a/net/dccp/sysctl.c +++ b/net/dccp/sysctl.c @@ -11,6 +11,7 @@ #include #include +#include "dccp.h" #include "feat.h" #ifndef CONFIG_SYSCTL @@ -66,6 +67,30 @@ static struct ctl_table dccp_default_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, + { + .ctl_name = NET_DCCP_DEFAULT_REQ_RETRIES, + .procname = "request_retries", + .data = &sysctl_dccp_request_retries, + .maxlen = sizeof(sysctl_dccp_request_retries), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { + .ctl_name = NET_DCCP_DEFAULT_RETRIES1, + .procname = "retries1", + .data = &sysctl_dccp_retries1, + .maxlen = sizeof(sysctl_dccp_retries1), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { + .ctl_name = NET_DCCP_DEFAULT_RETRIES2, + .procname = "retries2", + .data = &sysctl_dccp_retries2, + .maxlen = sizeof(sysctl_dccp_retries2), + .mode = 0644, + .proc_handler = proc_dointvec, + }, { .ctl_name = 0, } }; diff --git a/net/dccp/timer.c b/net/dccp/timer.c index bda0af639ae..7b3f16e29a9 100644 --- a/net/dccp/timer.c +++ b/net/dccp/timer.c @@ -15,6 +15,11 @@ #include "dccp.h" +/* sysctl variables governing numbers of retransmission attempts */ +int sysctl_dccp_request_retries __read_mostly = TCP_SYN_RETRIES; +int sysctl_dccp_retries1 __read_mostly = TCP_RETR1; +int sysctl_dccp_retries2 __read_mostly = TCP_RETR2; + static void dccp_write_timer(unsigned long data); static void dccp_keepalive_timer(unsigned long data); static void dccp_delack_timer(unsigned long data); @@ -44,11 +49,10 @@ static int dccp_write_timeout(struct sock *sk) if (sk->sk_state == DCCP_REQUESTING || sk->sk_state == DCCP_PARTOPEN) { if (icsk->icsk_retransmits != 0) dst_negative_advice(&sk->sk_dst_cache); - retry_until = icsk->icsk_syn_retries ? : - /* FIXME! */ 3 /* FIXME! sysctl_tcp_syn_retries */; + retry_until = icsk->icsk_syn_retries ? + : sysctl_dccp_request_retries; } else { - if (icsk->icsk_retransmits >= - /* FIXME! sysctl_tcp_retries1 */ 5 /* FIXME! */) { + if (icsk->icsk_retransmits >= sysctl_dccp_retries1) { /* NOTE. draft-ietf-tcpimpl-pmtud-01.txt requires pmtu black hole detection. :-( @@ -72,7 +76,7 @@ static int dccp_write_timeout(struct sock *sk) dst_negative_advice(&sk->sk_dst_cache); } - retry_until = /* FIXME! */ 15 /* FIXME! sysctl_tcp_retries2 */; + retry_until = sysctl_dccp_retries2; /* * FIXME: see tcp_write_timout and tcp_out_of_resources */ @@ -196,7 +200,7 @@ backoff: icsk->icsk_rto = min(icsk->icsk_rto << 1, DCCP_RTO_MAX); inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, icsk->icsk_rto, DCCP_RTO_MAX); - if (icsk->icsk_retransmits > 3 /* FIXME: sysctl_dccp_retries1 */) + if (icsk->icsk_retransmits > sysctl_dccp_retries1) __sk_dst_reset(sk); out:; } -- cgit v1.2.3-70-g09d2 From 90833aa4f496d69ca374af6acef7d1614c8693ff Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Mon, 13 Nov 2006 16:02:22 -0800 Subject: [NET]: The scheduled removal of the frame diverter. This patch contains the scheduled removal of the frame diverter. Signed-off-by: Adrian Bunk Signed-off-by: David S. Miller --- Documentation/feature-removal-schedule.txt | 15 - drivers/net/Space.c | 1 - include/linux/Kbuild | 1 - include/linux/divert.h | 132 ------- include/linux/netdevice.h | 6 - include/linux/sockios.h | 4 +- net/Kconfig | 27 -- net/core/Makefile | 1 - net/core/dev.c | 20 +- net/core/dv.c | 546 ----------------------------- net/core/sysctl_net_core.c | 14 - net/socket.c | 6 - 12 files changed, 6 insertions(+), 767 deletions(-) delete mode 100644 include/linux/divert.h delete mode 100644 net/core/dv.c (limited to 'include') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index d52c4aaaf17..b3949cd3d01 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -227,21 +227,6 @@ Who: Patrick McHardy --------------------------- -What: frame diverter -When: November 2006 -Why: The frame diverter is included in most distribution kernels, but is - broken. It does not correctly handle many things: - - IPV6 - - non-linear skb's - - network device RCU on removal - - input frames not correctly checked for protocol errors - It also adds allocation overhead even if not enabled. - It is not clear if anyone is still using it. -Who: Stephen Hemminger - ---------------------------- - - What: PHYSDEVPATH, PHYSDEVBUS, PHYSDEVDRIVER in the uevent environment When: October 2008 Why: The stacking of class devices makes these values misleading and diff --git a/drivers/net/Space.c b/drivers/net/Space.c index a67f5efc983..602ed31a5dd 100644 --- a/drivers/net/Space.c +++ b/drivers/net/Space.c @@ -33,7 +33,6 @@ #include #include #include -#include /* A unified ethernet device probe. This is the easiest way to have every ethernet adaptor have the name "eth[0123...]". diff --git a/include/linux/Kbuild b/include/linux/Kbuild index a1155a2beb3..d7e04689304 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -193,7 +193,6 @@ unifdef-y += cuda.h unifdef-y += cyclades.h unifdef-y += dccp.h unifdef-y += dirent.h -unifdef-y += divert.h unifdef-y += dlm.h unifdef-y += elfcore.h unifdef-y += errno.h diff --git a/include/linux/divert.h b/include/linux/divert.h deleted file mode 100644 index 8fb4e9de684..00000000000 --- a/include/linux/divert.h +++ /dev/null @@ -1,132 +0,0 @@ -/* - * Frame Diversion, Benoit Locher - * - * Changes: - * 06/09/2000 BL: initial version - * - */ - -#ifndef _LINUX_DIVERT_H -#define _LINUX_DIVERT_H - -#include - -#define MAX_DIVERT_PORTS 8 /* Max number of ports to divert (tcp, udp) */ - -/* Divertable protocols */ -#define DIVERT_PROTO_NONE 0x0000 -#define DIVERT_PROTO_IP 0x0001 -#define DIVERT_PROTO_ICMP 0x0002 -#define DIVERT_PROTO_TCP 0x0004 -#define DIVERT_PROTO_UDP 0x0008 - -/* - * This is an Ethernet Frame Diverter option block - */ -struct divert_blk -{ - int divert; /* are we active */ - unsigned int protos; /* protocols */ - __u16 tcp_dst[MAX_DIVERT_PORTS]; /* specific tcp dst ports to divert */ - __u16 tcp_src[MAX_DIVERT_PORTS]; /* specific tcp src ports to divert */ - __u16 udp_dst[MAX_DIVERT_PORTS]; /* specific udp dst ports to divert */ - __u16 udp_src[MAX_DIVERT_PORTS]; /* specific udp src ports to divert */ -}; - -/* - * Diversion control block, for configuration with the userspace tool - * divert - */ - -typedef union _divert_cf_arg -{ - __s16 int16; - __u16 uint16; - __s32 int32; - __u32 uint32; - __s64 int64; - __u64 uint64; - void __user *ptr; -} divert_cf_arg; - - -struct divert_cf -{ - int cmd; /* Command */ - divert_cf_arg arg1, - arg2, - arg3; - int dev_index; /* device index (eth0=0, etc...) */ -}; - - -/* Diversion commands */ -#define DIVCMD_DIVERT 1 /* ENABLE/DISABLE diversion */ -#define DIVCMD_IP 2 /* ENABLE/DISABLE whold IP diversion */ -#define DIVCMD_TCP 3 /* ENABLE/DISABLE whold TCP diversion */ -#define DIVCMD_TCPDST 4 /* ADD/REMOVE TCP DST port for diversion */ -#define DIVCMD_TCPSRC 5 /* ADD/REMOVE TCP SRC port for diversion */ -#define DIVCMD_UDP 6 /* ENABLE/DISABLE whole UDP diversion */ -#define DIVCMD_UDPDST 7 /* ADD/REMOVE UDP DST port for diversion */ -#define DIVCMD_UDPSRC 8 /* ADD/REMOVE UDP SRC port for diversion */ -#define DIVCMD_ICMP 9 /* ENABLE/DISABLE whole ICMP diversion */ -#define DIVCMD_GETSTATUS 10 /* GET the status of the diverter */ -#define DIVCMD_RESET 11 /* Reset the diverter on the specified dev */ -#define DIVCMD_GETVERSION 12 /* Retrieve the diverter code version (char[32]) */ - -/* General syntax of the commands: - * - * DIVCMD_xxxxxx(arg1, arg2, arg3, dev_index) - * - * SIOCSIFDIVERT: - * DIVCMD_DIVERT(DIVARG1_ENABLE|DIVARG1_DISABLE, , ,ifindex) - * DIVCMD_IP(DIVARG1_ENABLE|DIVARG1_DISABLE, , , ifindex) - * DIVCMD_TCP(DIVARG1_ENABLE|DIVARG1_DISABLE, , , ifindex) - * DIVCMD_TCPDST(DIVARG1_ADD|DIVARG1_REMOVE, port, , ifindex) - * DIVCMD_TCPSRC(DIVARG1_ADD|DIVARG1_REMOVE, port, , ifindex) - * DIVCMD_UDP(DIVARG1_ENABLE|DIVARG1_DISABLE, , , ifindex) - * DIVCMD_UDPDST(DIVARG1_ADD|DIVARG1_REMOVE, port, , ifindex) - * DIVCMD_UDPSRC(DIVARG1_ADD|DIVARG1_REMOVE, port, , ifindex) - * DIVCMD_ICMP(DIVARG1_ENABLE|DIVARG1_DISABLE, , , ifindex) - * DIVCMD_RESET(, , , ifindex) - * - * SIOGIFDIVERT: - * DIVCMD_GETSTATUS(divert_blk, , , ifindex) - * DIVCMD_GETVERSION(string[3]) - */ - - -/* Possible values for arg1 */ -#define DIVARG1_ENABLE 0 /* ENABLE something */ -#define DIVARG1_DISABLE 1 /* DISABLE something */ -#define DIVARG1_ADD 2 /* ADD something */ -#define DIVARG1_REMOVE 3 /* REMOVE something */ - - -#ifdef __KERNEL__ - -/* diverter functions */ -#include - -#ifdef CONFIG_NET_DIVERT -#include - -int alloc_divert_blk(struct net_device *); -void free_divert_blk(struct net_device *); -int divert_ioctl(unsigned int cmd, struct divert_cf __user *arg); -void divert_frame(struct sk_buff *skb); -static inline void handle_diverter(struct sk_buff *skb) -{ - /* if diversion is supported on device, then divert */ - if (skb->dev->divert && skb->dev->divert->divert) - divert_frame(skb); -} - -#else -# define alloc_divert_blk(dev) (0) -# define free_divert_blk(dev) do {} while (0) -# define divert_ioctl(cmd, arg) (-ENOPKG) -# define handle_diverter(skb) do {} while (0) -#endif -#endif -#endif /* _LINUX_DIVERT_H */ diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 83b8c4f1d69..4e967b2e22c 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -38,7 +38,6 @@ #include #include -struct divert_blk; struct vlan_group; struct ethtool_ops; struct netpoll_info; @@ -517,11 +516,6 @@ struct net_device /* bridge stuff */ struct net_bridge_port *br_port; -#ifdef CONFIG_NET_DIVERT - /* this will get initialized at each interface type init routine */ - struct divert_blk *divert; -#endif /* CONFIG_NET_DIVERT */ - /* class/net/name entry */ struct class_device class_dev; /* space for optional statistics and wireless sysfs groups */ diff --git a/include/linux/sockios.h b/include/linux/sockios.h index e6b9d1d36ea..abef7596655 100644 --- a/include/linux/sockios.h +++ b/include/linux/sockios.h @@ -72,8 +72,8 @@ #define SIOCGIFTXQLEN 0x8942 /* Get the tx queue length */ #define SIOCSIFTXQLEN 0x8943 /* Set the tx queue length */ -#define SIOCGIFDIVERT 0x8944 /* Frame diversion support */ -#define SIOCSIFDIVERT 0x8945 /* Set frame diversion options */ +/* SIOCGIFDIVERT was: 0x8944 Frame diversion support */ +/* SIOCSIFDIVERT was: 0x8945 Set frame diversion options */ #define SIOCETHTOOL 0x8946 /* Ethtool interface */ diff --git a/net/Kconfig b/net/Kconfig index 67e39ad8b8b..867f9503251 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -175,33 +175,6 @@ source "net/ipx/Kconfig" source "drivers/net/appletalk/Kconfig" source "net/x25/Kconfig" source "net/lapb/Kconfig" - -config NET_DIVERT - bool "Frame Diverter (EXPERIMENTAL)" - depends on EXPERIMENTAL && BROKEN - ---help--- - The Frame Diverter allows you to divert packets from the - network, that are not aimed at the interface receiving it (in - promisc. mode). Typically, a Linux box setup as an Ethernet bridge - with the Frames Diverter on, can do some *really* transparent www - caching using a Squid proxy for example. - - This is very useful when you don't want to change your router's - config (or if you simply don't have access to it). - - The other possible usages of diverting Ethernet Frames are - numberous: - - reroute smtp traffic to another interface - - traffic-shape certain network streams - - transparently proxy smtp connections - - etc... - - For more informations, please refer to: - - - - If unsure, say N. - source "net/econet/Kconfig" source "net/wanrouter/Kconfig" source "net/sched/Kconfig" diff --git a/net/core/Makefile b/net/core/Makefile index 119568077da..73272d506e9 100644 --- a/net/core/Makefile +++ b/net/core/Makefile @@ -12,7 +12,6 @@ obj-y += dev.o ethtool.o dev_mcast.o dst.o netevent.o \ obj-$(CONFIG_XFRM) += flow.o obj-$(CONFIG_SYSFS) += net-sysfs.o -obj-$(CONFIG_NET_DIVERT) += dv.o obj-$(CONFIG_NET_PKTGEN) += pktgen.o obj-$(CONFIG_WIRELESS_EXT) += wireless.o obj-$(CONFIG_NETPOLL) += netpoll.o diff --git a/net/core/dev.c b/net/core/dev.c index 411c2428d26..5bf13b132dd 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -98,7 +98,6 @@ #include #include #include -#include #include #include #include @@ -1827,8 +1826,6 @@ int netif_receive_skb(struct sk_buff *skb) ncls: #endif - handle_diverter(skb); - if (handle_bridge(&skb, &pt_prev, &ret, orig_dev)) goto out; @@ -2898,10 +2895,6 @@ int register_netdevice(struct net_device *dev) spin_lock_init(&dev->ingress_lock); #endif - ret = alloc_divert_blk(dev); - if (ret) - goto out; - dev->iflink = -1; /* Init, if this function is available */ @@ -2910,13 +2903,13 @@ int register_netdevice(struct net_device *dev) if (ret) { if (ret > 0) ret = -EIO; - goto out_err; + goto out; } } if (!dev_valid_name(dev->name)) { ret = -EINVAL; - goto out_err; + goto out; } dev->ifindex = dev_new_index(); @@ -2930,7 +2923,7 @@ int register_netdevice(struct net_device *dev) = hlist_entry(p, struct net_device, name_hlist); if (!strncmp(d->name, dev->name, IFNAMSIZ)) { ret = -EEXIST; - goto out_err; + goto out; } } @@ -2974,7 +2967,7 @@ int register_netdevice(struct net_device *dev) ret = netdev_register_sysfs(dev); if (ret) - goto out_err; + goto out; dev->reg_state = NETREG_REGISTERED; /* @@ -3001,9 +2994,6 @@ int register_netdevice(struct net_device *dev) out: return ret; -out_err: - free_divert_blk(dev); - goto out; } /** @@ -3320,8 +3310,6 @@ int unregister_netdevice(struct net_device *dev) /* Notifier chain MUST detach us from master device. */ BUG_TRAP(!dev->master); - free_divert_blk(dev); - /* Finish processing unregister after unlock */ net_set_todo(dev); diff --git a/net/core/dv.c b/net/core/dv.c deleted file mode 100644 index 29ee77f1593..00000000000 --- a/net/core/dv.c +++ /dev/null @@ -1,546 +0,0 @@ -/* - * INET An implementation of the TCP/IP protocol suite for the LINUX - * operating system. INET is implemented using the BSD Socket - * interface as the means of communication with the user level. - * - * Generic frame diversion - * - * Authors: - * Benoit LOCHER: initial integration within the kernel with support for ethernet - * Dave Miller: improvement on the code (correctness, performance and source files) - * - */ -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -const char sysctl_divert_version[32]="0.46"; /* Current version */ - -static int __init dv_init(void) -{ - return 0; -} -module_init(dv_init); - -/* - * Allocate a divert_blk for a device. This must be an ethernet nic. - */ -int alloc_divert_blk(struct net_device *dev) -{ - int alloc_size = (sizeof(struct divert_blk) + 3) & ~3; - - dev->divert = NULL; - if (dev->type == ARPHRD_ETHER) { - dev->divert = kzalloc(alloc_size, GFP_KERNEL); - if (dev->divert == NULL) { - printk(KERN_INFO "divert: unable to allocate divert_blk for %s\n", - dev->name); - return -ENOMEM; - } - dev_hold(dev); - } - - return 0; -} - -/* - * Free a divert_blk allocated by the above function, if it was - * allocated on that device. - */ -void free_divert_blk(struct net_device *dev) -{ - if (dev->divert) { - kfree(dev->divert); - dev->divert=NULL; - dev_put(dev); - } -} - -/* - * Adds a tcp/udp (source or dest) port to an array - */ -static int add_port(u16 ports[], u16 port) -{ - int i; - - if (port == 0) - return -EINVAL; - - /* Storing directly in network format for performance, - * thanks Dave :) - */ - port = htons(port); - - for (i = 0; i < MAX_DIVERT_PORTS; i++) { - if (ports[i] == port) - return -EALREADY; - } - - for (i = 0; i < MAX_DIVERT_PORTS; i++) { - if (ports[i] == 0) { - ports[i] = port; - return 0; - } - } - - return -ENOBUFS; -} - -/* - * Removes a port from an array tcp/udp (source or dest) - */ -static int remove_port(u16 ports[], u16 port) -{ - int i; - - if (port == 0) - return -EINVAL; - - /* Storing directly in network format for performance, - * thanks Dave ! - */ - port = htons(port); - - for (i = 0; i < MAX_DIVERT_PORTS; i++) { - if (ports[i] == port) { - ports[i] = 0; - return 0; - } - } - - return -EINVAL; -} - -/* Some basic sanity checks on the arguments passed to divert_ioctl() */ -static int check_args(struct divert_cf *div_cf, struct net_device **dev) -{ - char devname[32]; - int ret; - - if (dev == NULL) - return -EFAULT; - - /* GETVERSION: all other args are unused */ - if (div_cf->cmd == DIVCMD_GETVERSION) - return 0; - - /* Network device index should reasonably be between 0 and 1000 :) */ - if (div_cf->dev_index < 0 || div_cf->dev_index > 1000) - return -EINVAL; - - /* Let's try to find the ifname */ - sprintf(devname, "eth%d", div_cf->dev_index); - *dev = dev_get_by_name(devname); - - /* dev should NOT be null */ - if (*dev == NULL) - return -EINVAL; - - ret = 0; - - /* user issuing the ioctl must be a super one :) */ - if (!capable(CAP_SYS_ADMIN)) { - ret = -EPERM; - goto out; - } - - /* Device must have a divert_blk member NOT null */ - if ((*dev)->divert == NULL) - ret = -EINVAL; -out: - dev_put(*dev); - return ret; -} - -/* - * control function of the diverter - */ -#if 0 -#define DVDBG(a) \ - printk(KERN_DEBUG "divert_ioctl() line %d %s\n", __LINE__, (a)) -#else -#define DVDBG(a) -#endif - -int divert_ioctl(unsigned int cmd, struct divert_cf __user *arg) -{ - struct divert_cf div_cf; - struct divert_blk *div_blk; - struct net_device *dev; - int ret; - - switch (cmd) { - case SIOCGIFDIVERT: - DVDBG("SIOCGIFDIVERT, copy_from_user"); - if (copy_from_user(&div_cf, arg, sizeof(struct divert_cf))) - return -EFAULT; - DVDBG("before check_args"); - ret = check_args(&div_cf, &dev); - if (ret) - return ret; - DVDBG("after checkargs"); - div_blk = dev->divert; - - DVDBG("befre switch()"); - switch (div_cf.cmd) { - case DIVCMD_GETSTATUS: - /* Now, just give the user the raw divert block - * for him to play with :) - */ - if (copy_to_user(div_cf.arg1.ptr, dev->divert, - sizeof(struct divert_blk))) - return -EFAULT; - break; - - case DIVCMD_GETVERSION: - DVDBG("GETVERSION: checking ptr"); - if (div_cf.arg1.ptr == NULL) - return -EINVAL; - DVDBG("GETVERSION: copying data to userland"); - if (copy_to_user(div_cf.arg1.ptr, - sysctl_divert_version, 32)) - return -EFAULT; - DVDBG("GETVERSION: data copied"); - break; - - default: - return -EINVAL; - } - - break; - - case SIOCSIFDIVERT: - if (copy_from_user(&div_cf, arg, sizeof(struct divert_cf))) - return -EFAULT; - - ret = check_args(&div_cf, &dev); - if (ret) - return ret; - - div_blk = dev->divert; - - switch(div_cf.cmd) { - case DIVCMD_RESET: - div_blk->divert = 0; - div_blk->protos = DIVERT_PROTO_NONE; - memset(div_blk->tcp_dst, 0, - MAX_DIVERT_PORTS * sizeof(u16)); - memset(div_blk->tcp_src, 0, - MAX_DIVERT_PORTS * sizeof(u16)); - memset(div_blk->udp_dst, 0, - MAX_DIVERT_PORTS * sizeof(u16)); - memset(div_blk->udp_src, 0, - MAX_DIVERT_PORTS * sizeof(u16)); - return 0; - - case DIVCMD_DIVERT: - switch(div_cf.arg1.int32) { - case DIVARG1_ENABLE: - if (div_blk->divert) - return -EALREADY; - div_blk->divert = 1; - break; - - case DIVARG1_DISABLE: - if (!div_blk->divert) - return -EALREADY; - div_blk->divert = 0; - break; - - default: - return -EINVAL; - } - - break; - - case DIVCMD_IP: - switch(div_cf.arg1.int32) { - case DIVARG1_ENABLE: - if (div_blk->protos & DIVERT_PROTO_IP) - return -EALREADY; - div_blk->protos |= DIVERT_PROTO_IP; - break; - - case DIVARG1_DISABLE: - if (!(div_blk->protos & DIVERT_PROTO_IP)) - return -EALREADY; - div_blk->protos &= ~DIVERT_PROTO_IP; - break; - - default: - return -EINVAL; - } - - break; - - case DIVCMD_TCP: - switch(div_cf.arg1.int32) { - case DIVARG1_ENABLE: - if (div_blk->protos & DIVERT_PROTO_TCP) - return -EALREADY; - div_blk->protos |= DIVERT_PROTO_TCP; - break; - - case DIVARG1_DISABLE: - if (!(div_blk->protos & DIVERT_PROTO_TCP)) - return -EALREADY; - div_blk->protos &= ~DIVERT_PROTO_TCP; - break; - - default: - return -EINVAL; - } - - break; - - case DIVCMD_TCPDST: - switch(div_cf.arg1.int32) { - case DIVARG1_ADD: - return add_port(div_blk->tcp_dst, - div_cf.arg2.uint16); - - case DIVARG1_REMOVE: - return remove_port(div_blk->tcp_dst, - div_cf.arg2.uint16); - - default: - return -EINVAL; - } - - break; - - case DIVCMD_TCPSRC: - switch(div_cf.arg1.int32) { - case DIVARG1_ADD: - return add_port(div_blk->tcp_src, - div_cf.arg2.uint16); - - case DIVARG1_REMOVE: - return remove_port(div_blk->tcp_src, - div_cf.arg2.uint16); - - default: - return -EINVAL; - } - - break; - - case DIVCMD_UDP: - switch(div_cf.arg1.int32) { - case DIVARG1_ENABLE: - if (div_blk->protos & DIVERT_PROTO_UDP) - return -EALREADY; - div_blk->protos |= DIVERT_PROTO_UDP; - break; - - case DIVARG1_DISABLE: - if (!(div_blk->protos & DIVERT_PROTO_UDP)) - return -EALREADY; - div_blk->protos &= ~DIVERT_PROTO_UDP; - break; - - default: - return -EINVAL; - } - - break; - - case DIVCMD_UDPDST: - switch(div_cf.arg1.int32) { - case DIVARG1_ADD: - return add_port(div_blk->udp_dst, - div_cf.arg2.uint16); - - case DIVARG1_REMOVE: - return remove_port(div_blk->udp_dst, - div_cf.arg2.uint16); - - default: - return -EINVAL; - } - - break; - - case DIVCMD_UDPSRC: - switch(div_cf.arg1.int32) { - case DIVARG1_ADD: - return add_port(div_blk->udp_src, - div_cf.arg2.uint16); - - case DIVARG1_REMOVE: - return remove_port(div_blk->udp_src, - div_cf.arg2.uint16); - - default: - return -EINVAL; - } - - break; - - case DIVCMD_ICMP: - switch(div_cf.arg1.int32) { - case DIVARG1_ENABLE: - if (div_blk->protos & DIVERT_PROTO_ICMP) - return -EALREADY; - div_blk->protos |= DIVERT_PROTO_ICMP; - break; - - case DIVARG1_DISABLE: - if (!(div_blk->protos & DIVERT_PROTO_ICMP)) - return -EALREADY; - div_blk->protos &= ~DIVERT_PROTO_ICMP; - break; - - default: - return -EINVAL; - } - - break; - - default: - return -EINVAL; - } - - break; - - default: - return -EINVAL; - } - - return 0; -} - - -/* - * Check if packet should have its dest mac address set to the box itself - * for diversion - */ - -#define ETH_DIVERT_FRAME(skb) \ - memcpy(eth_hdr(skb), skb->dev->dev_addr, ETH_ALEN); \ - skb->pkt_type=PACKET_HOST - -void divert_frame(struct sk_buff *skb) -{ - struct ethhdr *eth = eth_hdr(skb); - struct iphdr *iph; - struct tcphdr *tcph; - struct udphdr *udph; - struct divert_blk *divert = skb->dev->divert; - int i, src, dst; - unsigned char *skb_data_end = skb->data + skb->len; - - /* Packet is already aimed at us, return */ - if (!compare_ether_addr(eth->h_dest, skb->dev->dev_addr)) - return; - - /* proto is not IP, do nothing */ - if (eth->h_proto != htons(ETH_P_IP)) - return; - - /* Divert all IP frames ? */ - if (divert->protos & DIVERT_PROTO_IP) { - ETH_DIVERT_FRAME(skb); - return; - } - - /* Check for possible (maliciously) malformed IP frame (thanks Dave) */ - iph = (struct iphdr *) skb->data; - if (((iph->ihl<<2)+(unsigned char*)(iph)) >= skb_data_end) { - printk(KERN_INFO "divert: malformed IP packet !\n"); - return; - } - - switch (iph->protocol) { - /* Divert all ICMP frames ? */ - case IPPROTO_ICMP: - if (divert->protos & DIVERT_PROTO_ICMP) { - ETH_DIVERT_FRAME(skb); - return; - } - break; - - /* Divert all TCP frames ? */ - case IPPROTO_TCP: - if (divert->protos & DIVERT_PROTO_TCP) { - ETH_DIVERT_FRAME(skb); - return; - } - - /* Check for possible (maliciously) malformed IP - * frame (thanx Dave) - */ - tcph = (struct tcphdr *) - (((unsigned char *)iph) + (iph->ihl<<2)); - if (((unsigned char *)(tcph+1)) >= skb_data_end) { - printk(KERN_INFO "divert: malformed TCP packet !\n"); - return; - } - - /* Divert some tcp dst/src ports only ?*/ - for (i = 0; i < MAX_DIVERT_PORTS; i++) { - dst = divert->tcp_dst[i]; - src = divert->tcp_src[i]; - if ((dst && dst == tcph->dest) || - (src && src == tcph->source)) { - ETH_DIVERT_FRAME(skb); - return; - } - } - break; - - /* Divert all UDP frames ? */ - case IPPROTO_UDP: - if (divert->protos & DIVERT_PROTO_UDP) { - ETH_DIVERT_FRAME(skb); - return; - } - - /* Check for possible (maliciously) malformed IP - * packet (thanks Dave) - */ - udph = (struct udphdr *) - (((unsigned char *)iph) + (iph->ihl<<2)); - if (((unsigned char *)(udph+1)) >= skb_data_end) { - printk(KERN_INFO - "divert: malformed UDP packet !\n"); - return; - } - - /* Divert some udp dst/src ports only ? */ - for (i = 0; i < MAX_DIVERT_PORTS; i++) { - dst = divert->udp_dst[i]; - src = divert->udp_src[i]; - if ((dst && dst == udph->dest) || - (src && src == udph->source)) { - ETH_DIVERT_FRAME(skb); - return; - } - } - break; - } -} diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index 02534131d88..1e75b158546 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -21,10 +21,6 @@ extern __u32 sysctl_rmem_max; extern int sysctl_core_destroy_delay; -#ifdef CONFIG_NET_DIVERT -extern char sysctl_divert_version[]; -#endif /* CONFIG_NET_DIVERT */ - #ifdef CONFIG_XFRM extern u32 sysctl_xfrm_aevent_etime; extern u32 sysctl_xfrm_aevent_rseqth; @@ -105,16 +101,6 @@ ctl_table core_table[] = { .mode = 0644, .proc_handler = &proc_dointvec }, -#ifdef CONFIG_NET_DIVERT - { - .ctl_name = NET_CORE_DIVERT_VERSION, - .procname = "divert_version", - .data = (void *)sysctl_divert_version, - .maxlen = 32, - .mode = 0444, - .proc_handler = &proc_dostring - }, -#endif /* CONFIG_NET_DIVERT */ #ifdef CONFIG_XFRM { .ctl_name = NET_CORE_AEVENT_ETIME, diff --git a/net/socket.c b/net/socket.c index 6c9b9b326d7..e8db54702a6 100644 --- a/net/socket.c +++ b/net/socket.c @@ -77,7 +77,6 @@ #include #include #include -#include #include #include #include @@ -852,11 +851,6 @@ static long sock_ioctl(struct file *file, unsigned cmd, unsigned long arg) err = vlan_ioctl_hook(argp); mutex_unlock(&vlan_ioctl_mutex); break; - case SIOCGIFDIVERT: - case SIOCSIFDIVERT: - /* Convert this to call through a hook */ - err = divert_ioctl(cmd, argp); - break; case SIOCADDDLCI: case SIOCDELDLCI: err = -ENOPKG; -- cgit v1.2.3-70-g09d2 From b9df3cb8cf9a96e63dfdcd3056a9cbc71f2459e7 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Tue, 14 Nov 2006 11:21:36 -0200 Subject: [TCP/DCCP]: Introduce net_xmit_eval Throughout the TCP/DCCP (and tunnelling) code, it often happens that the return code of a transmit function needs to be tested against NET_XMIT_CN which is a value that does not indicate a strict error condition. This patch uses a macro for these recurring situations which is consistent with the already existing macro net_xmit_errno, saving on duplicated code. Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/netdevice.h | 4 ++++ include/net/ipip.h | 2 +- net/dccp/ipv4.c | 5 ++--- net/dccp/ipv6.c | 3 +-- net/dccp/output.c | 14 ++------------ net/ipv4/tcp_ipv4.c | 3 +-- net/ipv4/tcp_output.c | 8 +------- net/ipv6/ip6_tunnel.c | 2 +- net/ipv6/tcp_ipv6.c | 3 +-- 9 files changed, 14 insertions(+), 30 deletions(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 4e967b2e22c..caa3c259371 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -66,6 +66,10 @@ struct netpoll_info; #define NET_RX_CN_HIGH 4 /* The storm is here */ #define NET_RX_BAD 5 /* packet dropped due to kernel error */ +/* NET_XMIT_CN is special. It does not guarantee that this packet is lost. It + * indicates that the device will soon be dropping packets, or already drops + * some packets of the same priority; prompting us to send less aggressively. */ +#define net_xmit_eval(e) ((e) == NET_XMIT_CN? 0 : (e)) #define net_xmit_errno(e) ((e) != NET_XMIT_CN ? -ENOBUFS : 0) #endif diff --git a/include/net/ipip.h b/include/net/ipip.h index f490c3cbe37..84058858eea 100644 --- a/include/net/ipip.h +++ b/include/net/ipip.h @@ -35,7 +35,7 @@ struct ip_tunnel ip_send_check(iph); \ \ err = NF_HOOK(PF_INET, NF_IP_LOCAL_OUT, skb, NULL, rt->u.dst.dev, dst_output);\ - if (err == NET_XMIT_SUCCESS || err == NET_XMIT_CN) { \ + if (net_xmit_eval(err) == 0) { \ stats->tx_bytes += pkt_len; \ stats->tx_packets++; \ } else { \ diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index bc400b2ba25..61c09014dad 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -501,8 +501,7 @@ static int dccp_v4_send_response(struct sock *sk, struct request_sock *req, err = ip_build_and_send_pkt(skb, sk, ireq->loc_addr, ireq->rmt_addr, ireq->opt); - if (err == NET_XMIT_CN) - err = 0; + err = net_xmit_eval(err); } out: @@ -571,7 +570,7 @@ static void dccp_v4_ctl_send_reset(struct sk_buff *rxskb) rxskb->nh.iph->saddr, NULL); bh_unlock_sock(dccp_v4_ctl_socket->sk); - if (err == NET_XMIT_CN || err == 0) { + if (net_xmit_eval(err) == 0) { DCCP_INC_STATS_BH(DCCP_MIB_OUTSEGS); DCCP_INC_STATS_BH(DCCP_MIB_OUTRSTS); } diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 8d6ddb6389a..2165b1740c7 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -294,8 +294,7 @@ static int dccp_v6_send_response(struct sock *sk, struct request_sock *req, &ireq6->rmt_addr); ipv6_addr_copy(&fl.fl6_dst, &ireq6->rmt_addr); err = ip6_xmit(sk, skb, &fl, opt, 0); - if (err == NET_XMIT_CN) - err = 0; + err = net_xmit_eval(err); } done: diff --git a/net/dccp/output.c b/net/dccp/output.c index 0994b13f0f1..ef22f3cc791 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -125,16 +125,7 @@ static int dccp_transmit_skb(struct sock *sk, struct sk_buff *skb) memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); err = icsk->icsk_af_ops->queue_xmit(skb, sk, 0); - if (err <= 0) - return err; - - /* NET_XMIT_CN is special. It does not guarantee, - * that this packet is lost. It tells that device - * is about to start to drop packets or already - * drops some packets of the same priority and - * invokes us to send less aggressively. - */ - return err == NET_XMIT_CN ? 0 : err; + return net_xmit_eval(err); } return -ENOBUFS; } @@ -426,8 +417,7 @@ int dccp_send_reset(struct sock *sk, enum dccp_reset_codes code) if (skb != NULL) { memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); err = inet_csk(sk)->icsk_af_ops->queue_xmit(skb, sk, 0); - if (err == NET_XMIT_CN) - err = 0; + return net_xmit_eval(err); } } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 2eb58844403..0ad0904bf56 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -662,8 +662,7 @@ static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req, err = ip_build_and_send_pkt(skb, sk, ireq->loc_addr, ireq->rmt_addr, ireq->opt); - if (err == NET_XMIT_CN) - err = 0; + err = net_xmit_eval(err); } out: diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index f63e99aac2d..6a8581ab9a2 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -484,13 +484,7 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, tcp_enter_cwr(sk); - /* NET_XMIT_CN is special. It does not guarantee, - * that this packet is lost. It tells that device - * is about to start to drop packets or already - * drops some packets of the same priority and - * invokes us to send less aggressively. - */ - return err == NET_XMIT_CN ? 0 : err; + return net_xmit_eval(err); #undef SYSCTL_FLAG_TSTAMPS #undef SYSCTL_FLAG_WSCALE diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 4919f9294e2..80a11909159 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -748,7 +748,7 @@ ip6ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, skb->dst->dev, dst_output); - if (err == NET_XMIT_SUCCESS || err == NET_XMIT_CN) { + if (net_xmit_eval(err) == 0) { stats->tx_bytes += pkt_len; stats->tx_packets++; } else { diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 9a8e690fdf7..9a88395a762 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -502,8 +502,7 @@ static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req, ipv6_addr_copy(&fl.fl6_dst, &treq->rmt_addr); err = ip6_xmit(sk, skb, &fl, opt, 0); - if (err == NET_XMIT_CN) - err = 0; + err = net_xmit_eval(err); } done: -- cgit v1.2.3-70-g09d2 From c02fdc0e81e9c735d8d895af1e201b235df326d8 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Tue, 14 Nov 2006 12:48:10 -0200 Subject: [DCCP]: Make feature negotiation more readable This patch replaces cryptic feature negotiation messages of type Oct 31 15:42:20 kernel: dccp_feat_change: feat change type=32 feat=1 Oct 31 15:42:21 kernel: dccp_feat_change: feat change type=34 feat=1 Oct 31 15:42:21 kernel: dccp_feat_change: feat change type=32 feat=5 into ones of type: Nov 2 13:54:45 kernel: dccp_feat_change: ChangeL(CCID (1), 3) Nov 2 13:54:45 kernel: dccp_feat_change: ChangeR(CCID (1), 3) Nov 2 13:54:45 kernel: dccp_feat_change: ChangeL(Ack Ratio (5), 2) Also, * completed the feature number list wrt RFC 4340 sec. 6.4 * annotating which ones have been implemented so far * implemented rudimentary sanity checking in feat.c (FIXMEs) * some minor fixes Commiter note: uninlined dccp_feat_name and dccp_feat_typename, for consistency with dccp_{state,packet}_name, that, BTW, should be compiled only if CONFIG_IP_DCCP_DEBUG is selected, leaving this to another cset tho. Also shortened dccp_feat_negotiation_debug to dccp_feat_debug. Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/dccp.h | 25 ++++++----- net/dccp/feat.c | 116 +++++++++++++++++++++++++++++++++++++++------------ net/dccp/feat.h | 41 +++++++++++++++++- net/dccp/options.c | 4 +- 4 files changed, 145 insertions(+), 41 deletions(-) (limited to 'include') diff --git a/include/linux/dccp.h b/include/linux/dccp.h index 0502dfa7f32..696319ee2d5 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -175,18 +175,21 @@ enum { DCCPC_CCID3 = 3, }; -/* DCCP features */ -enum { - DCCPF_RESERVED = 0, - DCCPF_CCID = 1, - DCCPF_SEQUENCE_WINDOW = 3, - DCCPF_ACK_RATIO = 5, - DCCPF_SEND_ACK_VECTOR = 6, - DCCPF_SEND_NDP_COUNT = 7, +/* DCCP features (RFC 4340 section 6.4) */ + enum { + DCCPF_RESERVED = 0, + DCCPF_CCID = 1, + DCCPF_SHORT_SEQNOS = 2, /* XXX: not yet implemented */ + DCCPF_SEQUENCE_WINDOW = 3, + DCCPF_ECN_INCAPABLE = 4, /* XXX: not yet implemented */ + DCCPF_ACK_RATIO = 5, + DCCPF_SEND_ACK_VECTOR = 6, + DCCPF_SEND_NDP_COUNT = 7, DCCPF_MIN_CSUM_COVER = 8, - /* 10-127 reserved */ - DCCPF_MIN_CCID_SPECIFIC = 128, - DCCPF_MAX_CCID_SPECIFIC = 255, + DCCPF_DATA_CHECKSUM = 9, /* XXX: not yet implemented */ + /* 10-127 reserved */ + DCCPF_MIN_CCID_SPECIFIC = 128, + DCCPF_MAX_CCID_SPECIFIC = 255, }; /* this structure is argument to DCCP_SOCKOPT_CHANGE_X */ diff --git a/net/dccp/feat.c b/net/dccp/feat.c index a1b0682ee77..12cde2f2f13 100644 --- a/net/dccp/feat.c +++ b/net/dccp/feat.c @@ -12,7 +12,6 @@ #include -#include "dccp.h" #include "ccid.h" #include "feat.h" @@ -23,9 +22,17 @@ int dccp_feat_change(struct dccp_minisock *dmsk, u8 type, u8 feature, { struct dccp_opt_pend *opt; - dccp_pr_debug("feat change type=%d feat=%d\n", type, feature); + dccp_feat_debug(type, feature, *val); - /* XXX sanity check feat change request */ + if (!dccp_feat_is_valid_type(type)) { + pr_info("option type %d invalid in negotiation\n", type); + return 1; + } + if (!dccp_feat_is_valid_length(type, feature, len)) { + pr_info("invalid length %d\n", len); + return 1; + } + /* XXX add further sanity checks */ /* check if that feature is already being negotiated */ list_for_each_entry(opt, &dmsk->dccpms_pending, dccpop_node) { @@ -95,14 +102,14 @@ static int dccp_feat_update_ccid(struct sock *sk, u8 type, u8 new_ccid_nr) /* XXX taking only u8 vals */ static int dccp_feat_update(struct sock *sk, u8 type, u8 feat, u8 val) { - dccp_pr_debug("changing [%d] feat %d to %d\n", type, feat, val); + dccp_feat_debug(type, feat, val); switch (feat) { case DCCPF_CCID: return dccp_feat_update_ccid(sk, type, val); default: - dccp_pr_debug("IMPLEMENT changing [%d] feat %d to %d\n", - type, feat, val); + dccp_pr_debug("UNIMPLEMENTED: %s(%d, ...)\n", + dccp_feat_typename(type), feat); break; } return 0; @@ -265,10 +272,10 @@ static int dccp_feat_nn(struct sock *sk, u8 type, u8 feature, u8 *val, u8 len) u8 *copy; int rc; - /* NN features must be change L */ - if (type == DCCPO_CHANGE_R) { - dccp_pr_debug("received CHANGE_R %d for NN feat %d\n", - type, feature); + /* NN features must be Change L (sec. 6.3.2) */ + if (type != DCCPO_CHANGE_L) { + dccp_pr_debug("received %s for NN feature %d\n", + dccp_feat_typename(type), feature); return -EFAULT; } @@ -299,7 +306,8 @@ static int dccp_feat_nn(struct sock *sk, u8 type, u8 feature, u8 *val, u8 len) return rc; } - dccp_pr_debug("Confirming NN feature %d (val=%d)\n", feature, *copy); + dccp_feat_debug(type, feature, *copy); + list_add_tail(&opt->dccpop_node, &dmsk->dccpms_conf); return 0; @@ -318,14 +326,19 @@ static void dccp_feat_empty_confirm(struct dccp_minisock *dmsk, return; } - opt->dccpop_type = type == DCCPO_CHANGE_L ? DCCPO_CONFIRM_R : - DCCPO_CONFIRM_L; + switch (type) { + case DCCPO_CHANGE_L: opt->dccpop_type = DCCPO_CONFIRM_R; break; + case DCCPO_CHANGE_R: opt->dccpop_type = DCCPO_CONFIRM_L; break; + default: pr_info("invalid type %d\n", type); return; + + } opt->dccpop_feat = feature; opt->dccpop_val = NULL; opt->dccpop_len = 0; /* change feature */ - dccp_pr_debug("Empty confirm feature %d type %d\n", feature, type); + dccp_pr_debug("Empty %s(%d)\n", dccp_feat_typename(type), feature); + list_add_tail(&opt->dccpop_node, &dmsk->dccpms_conf); } @@ -359,7 +372,7 @@ int dccp_feat_change_recv(struct sock *sk, u8 type, u8 feature, u8 *val, u8 len) { int rc; - dccp_pr_debug("got feat change type=%d feat=%d\n", type, feature); + dccp_feat_debug(type, feature, *val); /* figure out if it's SP or NN feature */ switch (feature) { @@ -375,6 +388,8 @@ int dccp_feat_change_recv(struct sock *sk, u8 type, u8 feature, u8 *val, u8 len) /* XXX implement other features */ default: + dccp_pr_debug("UNIMPLEMENTED: not handling %s(%d, ...)\n", + dccp_feat_typename(type), feature); rc = -EFAULT; break; } @@ -403,20 +418,27 @@ int dccp_feat_confirm_recv(struct sock *sk, u8 type, u8 feature, u8 t; struct dccp_opt_pend *opt; struct dccp_minisock *dmsk = dccp_msk(sk); - int rc = 1; + int found = 0; int all_confirmed = 1; - dccp_pr_debug("got feat confirm type=%d feat=%d\n", type, feature); - - /* XXX sanity check type & feat */ + dccp_feat_debug(type, feature, *val); /* locate our change request */ - t = type == DCCPO_CONFIRM_L ? DCCPO_CHANGE_R : DCCPO_CHANGE_L; + switch (type) { + case DCCPO_CONFIRM_L: t = DCCPO_CHANGE_R; break; + case DCCPO_CONFIRM_R: t = DCCPO_CHANGE_L; break; + default: pr_info("invalid type %d\n", type); + return 1; + + } + /* XXX sanity check feature value */ list_for_each_entry(opt, &dmsk->dccpms_pending, dccpop_node) { if (!opt->dccpop_conf && opt->dccpop_type == t && opt->dccpop_feat == feature) { - /* we found it */ + found = 1; + dccp_pr_debug("feature %d found\n", opt->dccpop_feat); + /* XXX do sanity check */ opt->dccpop_conf = 1; @@ -425,9 +447,7 @@ int dccp_feat_confirm_recv(struct sock *sk, u8 type, u8 feature, dccp_feat_update(sk, opt->dccpop_type, opt->dccpop_feat, *val); - dccp_pr_debug("feat %d type %d confirmed %d\n", - feature, type, *val); - rc = 0; + /* XXX check the return value of dccp_feat_update */ break; } @@ -446,9 +466,9 @@ int dccp_feat_confirm_recv(struct sock *sk, u8 type, u8 feature, inet_csk_clear_xmit_timer(sk, ICSK_TIME_RETRANS); } - if (rc) - dccp_pr_debug("feat %d type %d never requested\n", - feature, type); + if (!found) + dccp_pr_debug("%s(%d, ...) never requested\n", + dccp_feat_typename(type), feature); return 0; } @@ -583,3 +603,45 @@ out: } EXPORT_SYMBOL_GPL(dccp_feat_init); + +#ifdef CONFIG_IP_DCCP_DEBUG +const char *dccp_feat_typename(const u8 type) +{ + switch(type) { + case DCCPO_CHANGE_L: return("ChangeL"); + case DCCPO_CONFIRM_L: return("ConfirmL"); + case DCCPO_CHANGE_R: return("ChangeR"); + case DCCPO_CONFIRM_R: return("ConfirmR"); + /* the following case must not appear in feature negotation */ + default: dccp_pr_debug("unknown type %d [BUG!]\n", type); + } + return NULL; +} + +EXPORT_SYMBOL_GPL(dccp_feat_typename); + +const char *dccp_feat_name(const u8 feat) +{ + static const char *feature_names[] = { + [DCCPF_RESERVED] = "Reserved", + [DCCPF_CCID] = "CCID", + [DCCPF_SHORT_SEQNOS] = "Allow Short Seqnos", + [DCCPF_SEQUENCE_WINDOW] = "Sequence Window", + [DCCPF_ECN_INCAPABLE] = "ECN Incapable", + [DCCPF_ACK_RATIO] = "Ack Ratio", + [DCCPF_SEND_ACK_VECTOR] = "Send ACK Vector", + [DCCPF_SEND_NDP_COUNT] = "Send NDP Count", + [DCCPF_MIN_CSUM_COVER] = "Min. Csum Coverage", + [DCCPF_DATA_CHECKSUM] = "Send Data Checksum", + }; + if (feat >= DCCPF_MIN_CCID_SPECIFIC) + return "CCID-specific"; + + if (dccp_feat_is_reserved(feat)) + return feature_names[DCCPF_RESERVED]; + + return feature_names[feat]; +} + +EXPORT_SYMBOL_GPL(dccp_feat_name); +#endif /* CONFIG_IP_DCCP_DEBUG */ diff --git a/net/dccp/feat.h b/net/dccp/feat.h index 6048373c718..2c373ad7edc 100644 --- a/net/dccp/feat.h +++ b/net/dccp/feat.h @@ -12,9 +12,46 @@ */ #include +#include "dccp.h" -struct sock; -struct dccp_minisock; +static inline int dccp_feat_is_valid_length(u8 type, u8 feature, u8 len) +{ + /* sec. 6.1: Confirm has at least length 3, + * sec. 6.2: Change has at least length 4 */ + if (len < 3) + return 1; + if (len < 4 && (type == DCCPO_CHANGE_L || type == DCCPO_CHANGE_R)) + return 1; + /* XXX: add per-feature length validation (sec. 6.6.8) */ + return 0; +} + +static inline int dccp_feat_is_reserved(const u8 feat) +{ + return (feat > DCCPF_DATA_CHECKSUM && + feat < DCCPF_MIN_CCID_SPECIFIC) || + feat == DCCPF_RESERVED; +} + +/* feature negotiation knows only these four option types (RFC 4340, sec. 6) */ +static inline int dccp_feat_is_valid_type(const u8 optnum) +{ + return optnum >= DCCPO_CHANGE_L && optnum <= DCCPO_CONFIRM_R; + +} + +#ifdef CONFIG_IP_DCCP_DEBUG +extern const char *dccp_feat_typename(const u8 type); +extern const char *dccp_feat_name(const u8 feat); + +static inline void dccp_feat_debug(const u8 type, const u8 feat, const u8 val) +{ + dccp_pr_debug("%s(%s (%d), %d)\n", dccp_feat_typename(type), + dccp_feat_name(feat), feat, val); +} +#else +#define dccp_feat_debug(type, feat, val) +#endif /* CONFIG_IP_DCCP_DEBUG */ extern int dccp_feat_change(struct dccp_minisock *dmsk, u8 type, u8 feature, u8 *val, u8 len, gfp_t gfp); diff --git a/net/dccp/options.c b/net/dccp/options.c index 121e794fe45..2d0ef27f4ab 100644 --- a/net/dccp/options.c +++ b/net/dccp/options.c @@ -465,8 +465,10 @@ static int dccp_insert_feat_opt(struct sk_buff *skb, u8 type, u8 feat, if (len) memcpy(to, val, len); - dccp_pr_debug("option %d feat %d len %d\n", type, feat, len); + dccp_pr_debug("%s(%s (%d), ...), length %d\n", + dccp_feat_typename(type), + dccp_feat_name(feat), feat, len); return 0; } -- cgit v1.2.3-70-g09d2 From 09dbc3895e3242346bd434dae743c456fd28fc6a Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Tue, 14 Nov 2006 12:57:34 -0200 Subject: [DCCP]: Miscellaneous code tidy-ups This patch does not change code; it performs some trivial clean/tidy-ups: * removal of a `debug_prefix' string in favour of the already existing dccp_role(sk) * add documentation of structures and constants * separated out the cases for invalid packets (step 1 of the packet validation) * removing duplicate statements * combining declaration & initialisation Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/dccp.h | 18 +++++++++++++++++- net/dccp/ackvec.c | 27 ++++++--------------------- net/dccp/ipv4.c | 39 +++++++++++++++++++-------------------- net/dccp/ipv6.c | 6 ++---- net/dccp/options.c | 18 +++++++----------- net/dccp/output.c | 1 + net/dccp/proto.c | 8 ++------ 7 files changed, 54 insertions(+), 63 deletions(-) (limited to 'include') diff --git a/include/linux/dccp.h b/include/linux/dccp.h index 696319ee2d5..d308f1228b6 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -209,6 +209,7 @@ struct dccp_so_feat { #define DCCP_SOCKOPT_CCID_RX_INFO 128 #define DCCP_SOCKOPT_CCID_TX_INFO 192 +/* maximum number of services provided on the same listening port */ #define DCCP_SERVICE_LIST_MAX_LEN 32 #ifdef __KERNEL__ @@ -355,6 +356,9 @@ static inline unsigned int dccp_hdr_len(const struct sk_buff *skb) * @dccpms_ccid - Congestion Control Id (CCID) (section 10) * @dccpms_send_ack_vector - Send Ack Vector Feature (section 11.5) * @dccpms_send_ndp_count - Send NDP Count Feature (7.7.2) + * @dccpms_ack_ratio - Ack Ratio Feature (section 11.3) + * @dccpms_pending - List of features being negotiated + * @dccpms_conf - */ struct dccp_minisock { __u64 dccpms_sequence_window; @@ -452,14 +456,26 @@ struct dccp_ackvec; * @dccps_gss - greatest sequence number sent * @dccps_gsr - greatest valid sequence number received * @dccps_gar - greatest valid ack number received on a non-Sync; initialized to %dccps_iss + * @dccps_service - first (passive sock) or unique (active sock) service code + * @dccps_service_list - second .. last service code on passive socket * @dccps_timestamp_time - time of latest TIMESTAMP option * @dccps_timestamp_echo - latest timestamp received on a TIMESTAMP option * @dccps_packet_size - Set thru setsockopt - * @dccps_role - Role of this sock, one of %dccp_role + * @dccps_l_ack_ratio - + * @dccps_r_ack_ratio - * @dccps_pcslen - sender partial checksum coverage (via sockopt) * @dccps_pcrlen - receiver partial checksum coverage (via sockopt) * @dccps_ndp_count - number of Non Data Packets since last data packet + * @dccps_mss_cache - + * @dccps_minisock - * @dccps_hc_rx_ackvec - rx half connection ack vector + * @dccps_hc_rx_ccid - + * @dccps_hc_tx_ccid - + * @dccps_options_received - + * @dccps_epoch - + * @dccps_role - Role of this sock, one of %dccp_role + * @dccps_hc_rx_insert_options - + * @dccps_hc_tx_insert_options - * @dccps_xmit_timer - timer for when CCID is not ready to send */ struct dccp_sock { diff --git a/net/dccp/ackvec.c b/net/dccp/ackvec.c index f8208874ac7..0c54b89a4e9 100644 --- a/net/dccp/ackvec.c +++ b/net/dccp/ackvec.c @@ -67,10 +67,6 @@ static void dccp_ackvec_insert_avr(struct dccp_ackvec *av, int dccp_insert_option_ackvec(struct sock *sk, struct sk_buff *skb) { struct dccp_sock *dp = dccp_sk(sk); -#ifdef CONFIG_IP_DCCP_DEBUG - const char *debug_prefix = dp->dccps_role == DCCP_ROLE_CLIENT ? - "CLIENT tx: " : "server tx: "; -#endif struct dccp_ackvec *av = dp->dccps_hc_rx_ackvec; int len = av->dccpav_vec_len + 2; struct timeval now; @@ -129,9 +125,9 @@ int dccp_insert_option_ackvec(struct sock *sk, struct sk_buff *skb) dccp_ackvec_insert_avr(av, avr); - dccp_pr_debug("%sACK Vector 0, len=%d, ack_seqno=%llu, " + dccp_pr_debug("%s ACK Vector 0, len=%d, ack_seqno=%llu, " "ack_ackno=%llu\n", - debug_prefix, avr->dccpavr_sent_len, + dccp_role(sk), avr->dccpavr_sent_len, (unsigned long long)avr->dccpavr_ack_seqno, (unsigned long long)avr->dccpavr_ack_ackno); return 0; @@ -380,14 +376,9 @@ void dccp_ackvec_check_rcv_ackno(struct dccp_ackvec *av, struct sock *sk, */ list_for_each_entry_reverse(avr, &av->dccpav_records, dccpavr_node) { if (ackno == avr->dccpavr_ack_seqno) { -#ifdef CONFIG_IP_DCCP_DEBUG - struct dccp_sock *dp = dccp_sk(sk); - const char *debug_prefix = dp->dccps_role == DCCP_ROLE_CLIENT ? - "CLIENT rx ack: " : "server rx ack: "; -#endif - dccp_pr_debug("%sACK packet 0, len=%d, ack_seqno=%llu, " + dccp_pr_debug("%s ACK packet 0, len=%d, ack_seqno=%llu, " "ack_ackno=%llu, ACKED!\n", - debug_prefix, 1, + dccp_role(sk), 1, (unsigned long long)avr->dccpavr_ack_seqno, (unsigned long long)avr->dccpavr_ack_ackno); dccp_ackvec_throw_record(av, avr); @@ -437,16 +428,10 @@ found: if (between48(avr->dccpavr_ack_seqno, ackno_end_rl, ackno)) { const u8 state = *vector & DCCP_ACKVEC_STATE_MASK; if (state != DCCP_ACKVEC_STATE_NOT_RECEIVED) { -#ifdef CONFIG_IP_DCCP_DEBUG - struct dccp_sock *dp = dccp_sk(sk); - const char *debug_prefix = - dp->dccps_role == DCCP_ROLE_CLIENT ? - "CLIENT rx ack: " : "server rx ack: "; -#endif - dccp_pr_debug("%sACK vector 0, len=%d, " + dccp_pr_debug("%s ACK vector 0, len=%d, " "ack_seqno=%llu, ack_ackno=%llu, " "ACKED!\n", - debug_prefix, len, + dccp_role(sk), len, (unsigned long long) avr->dccpavr_ack_seqno, (unsigned long long) diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 61c09014dad..34d6d197c3b 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -518,7 +518,7 @@ static void dccp_v4_ctl_send_reset(struct sk_buff *rxskb) sizeof(struct dccp_hdr_reset); struct sk_buff *skb; struct dst_entry *dst; - u64 seqno; + u64 seqno = 0; /* Never send a reset in response to a reset. */ if (rxdh->dccph_type == DCCP_PKT_RESET) @@ -552,13 +552,11 @@ static void dccp_v4_ctl_send_reset(struct sk_buff *rxskb) DCCP_SKB_CB(rxskb)->dccpd_reset_code; /* See "8.3.1. Abnormal Termination" in RFC 4340 */ - seqno = 0; if (DCCP_SKB_CB(rxskb)->dccpd_ack_seq != DCCP_PKT_WITHOUT_ACK_SEQ) dccp_set_seqno(&seqno, DCCP_SKB_CB(rxskb)->dccpd_ack_seq + 1); dccp_hdr_set_seq(dh, seqno); - dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), - DCCP_SKB_CB(rxskb)->dccpd_seq); + dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), DCCP_SKB_CB(rxskb)->dccpd_seq); dccp_csum_outgoing(skb); dh->dccph_checksum = dccp_v4_csum_finish(skb, rxskb->nh.iph->saddr, @@ -734,6 +732,11 @@ discard: EXPORT_SYMBOL_GPL(dccp_v4_do_rcv); +/** + * dccp_invalid_packet - check for malformed packets + * Implements RFC 4340, 8.5: Step 1: Check header basics + * Packets that fail these checks are ignored and do not receive Resets. + */ int dccp_invalid_packet(struct sk_buff *skb) { const struct dccp_hdr *dh; @@ -742,6 +745,7 @@ int dccp_invalid_packet(struct sk_buff *skb) if (skb->pkt_type != PACKET_HOST) return 1; + /* If the packet is shorter than 12 bytes, drop packet and return */ if (!pskb_may_pull(skb, sizeof(struct dccp_hdr))) { LIMIT_NETDEBUG(KERN_WARNING "DCCP: pskb_may_pull failed\n"); return 1; @@ -749,42 +753,37 @@ int dccp_invalid_packet(struct sk_buff *skb) dh = dccp_hdr(skb); - /* If the packet type is not understood, drop packet and return */ + /* If P.type is not understood, drop packet and return */ if (dh->dccph_type >= DCCP_PKT_INVALID) { LIMIT_NETDEBUG(KERN_WARNING "DCCP: invalid packet type\n"); return 1; } /* - * If P.Data Offset is too small for packet type, or too large for - * packet, drop packet and return + * If P.Data Offset is too small for packet type, drop packet and return */ if (dh->dccph_doff < dccp_hdr_len(skb) / sizeof(u32)) { LIMIT_NETDEBUG(KERN_WARNING "DCCP: P.Data Offset(%u) " - "too small 1\n", - dh->dccph_doff); + "too small\n", dh->dccph_doff); return 1; } - + /* + * If P.Data Offset is too too large for packet, drop packet and return + */ if (!pskb_may_pull(skb, dh->dccph_doff * sizeof(u32))) { LIMIT_NETDEBUG(KERN_WARNING "DCCP: P.Data Offset(%u) " - "too small 2\n", - dh->dccph_doff); + "too large\n", dh->dccph_doff); return 1; } - dh = dccp_hdr(skb); - /* * If P.type is not Data, Ack, or DataAck and P.X == 0 (the packet * has short sequence numbers), drop packet and return */ - if (dh->dccph_x == 0 && - dh->dccph_type != DCCP_PKT_DATA && - dh->dccph_type != DCCP_PKT_ACK && - dh->dccph_type != DCCP_PKT_DATAACK) { - LIMIT_NETDEBUG(KERN_WARNING "DCCP: P.type (%s) not Data, Ack " - "nor DataAck and P.X == 0\n", + if (dh->dccph_type >= DCCP_PKT_DATA && + dh->dccph_type <= DCCP_PKT_DATAACK && dh->dccph_x == 0) { + LIMIT_NETDEBUG(KERN_WARNING "DCCP: P.type (%s) not Data||Ack||" + "DataAck, while P.X == 0\n", dccp_packet_name(dh->dccph_type)); return 1; } diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 2165b1740c7..fc326173c21 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -318,7 +318,7 @@ static void dccp_v6_ctl_send_reset(struct sk_buff *rxskb) sizeof(struct dccp_hdr_reset); struct sk_buff *skb; struct flowi fl; - u64 seqno; + u64 seqno = 0; if (rxdh->dccph_type == DCCP_PKT_RESET) return; @@ -345,13 +345,11 @@ static void dccp_v6_ctl_send_reset(struct sk_buff *rxskb) DCCP_SKB_CB(rxskb)->dccpd_reset_code; /* See "8.3.1. Abnormal Termination" in RFC 4340 */ - seqno = 0; if (DCCP_SKB_CB(rxskb)->dccpd_ack_seq != DCCP_PKT_WITHOUT_ACK_SEQ) dccp_set_seqno(&seqno, DCCP_SKB_CB(rxskb)->dccpd_ack_seq + 1); dccp_hdr_set_seq(dh, seqno); - dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), - DCCP_SKB_CB(rxskb)->dccpd_seq); + dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), DCCP_SKB_CB(rxskb)->dccpd_seq); dccp_csum_outgoing(skb); dh->dccph_checksum = dccp_v6_csum_finish(skb, &rxskb->nh.ipv6h->saddr, diff --git a/net/dccp/options.c b/net/dccp/options.c index 2d0ef27f4ab..7e50678e247 100644 --- a/net/dccp/options.c +++ b/net/dccp/options.c @@ -60,10 +60,6 @@ static u32 dccp_decode_value_var(const unsigned char *bf, const u8 len) int dccp_parse_options(struct sock *sk, struct sk_buff *skb) { struct dccp_sock *dp = dccp_sk(sk); -#ifdef CONFIG_IP_DCCP_DEBUG - const char *debug_prefix = dp->dccps_role == DCCP_ROLE_CLIENT ? - "CLIENT rx opt: " : "server rx opt: "; -#endif const struct dccp_hdr *dh = dccp_hdr(skb); const u8 pkt_type = DCCP_SKB_CB(skb)->dccpd_type; unsigned char *options = (unsigned char *)dh + dccp_hdr_len(skb); @@ -119,7 +115,7 @@ int dccp_parse_options(struct sock *sk, struct sk_buff *skb) goto out_invalid_option; opt_recv->dccpor_ndp = dccp_decode_value_var(value, len); - dccp_pr_debug("%sNDP count=%d\n", debug_prefix, + dccp_pr_debug("%s rx opt: NDP count=%d\n", dccp_role(sk), opt_recv->dccpor_ndp); break; case DCCPO_CHANGE_L: @@ -165,8 +161,8 @@ int dccp_parse_options(struct sock *sk, struct sk_buff *skb) dp->dccps_timestamp_echo = opt_recv->dccpor_timestamp; dccp_timestamp(sk, &dp->dccps_timestamp_time); - dccp_pr_debug("%sTIMESTAMP=%u, ackno=%llu\n", - debug_prefix, opt_recv->dccpor_timestamp, + dccp_pr_debug("%s rx opt: TIMESTAMP=%u, ackno=%llu\n", + dccp_role(sk), opt_recv->dccpor_timestamp, (unsigned long long) DCCP_SKB_CB(skb)->dccpd_ack_seq); break; @@ -176,8 +172,8 @@ int dccp_parse_options(struct sock *sk, struct sk_buff *skb) opt_recv->dccpor_timestamp_echo = ntohl(*(__be32 *)value); - dccp_pr_debug("%sTIMESTAMP_ECHO=%u, len=%d, ackno=%llu, ", - debug_prefix, + dccp_pr_debug("%s rx opt: TIMESTAMP_ECHO=%u, len=%d, " + "ackno=%llu, ", dccp_role(sk), opt_recv->dccpor_timestamp_echo, len + 2, (unsigned long long) @@ -211,8 +207,8 @@ int dccp_parse_options(struct sock *sk, struct sk_buff *skb) if (elapsed_time > opt_recv->dccpor_elapsed_time) opt_recv->dccpor_elapsed_time = elapsed_time; - dccp_pr_debug("%sELAPSED_TIME=%d\n", debug_prefix, - elapsed_time); + dccp_pr_debug("%s rx opt: ELAPSED_TIME=%d\n", + dccp_role(sk), elapsed_time); break; /* * From RFC 4340, sec. 10.3: diff --git a/net/dccp/output.c b/net/dccp/output.c index ef22f3cc791..c34eada7f02 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -333,6 +333,7 @@ struct sk_buff *dccp_make_response(struct sock *sk, struct dst_entry *dst, return NULL; } + /* Build and checksum header */ dh = dccp_zeroed_hdr(skb, dccp_header_size); dh->dccph_sport = inet_sk(sk)->sport; diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 9c9c08cffda..0225bdacd3b 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -452,9 +452,8 @@ out_free_val: static int do_dccp_setsockopt(struct sock *sk, int level, int optname, char __user *optval, int optlen) { - struct dccp_sock *dp; - int err; - int val; + struct dccp_sock *dp = dccp_sk(sk); + int val, err = 0; if (optlen < sizeof(int)) return -EINVAL; @@ -466,9 +465,6 @@ static int do_dccp_setsockopt(struct sock *sk, int level, int optname, return dccp_setsockopt_service(sk, val, optval, optlen); lock_sock(sk); - dp = dccp_sk(sk); - err = 0; - switch (optname) { case DCCP_SOCKOPT_PACKET_SIZE: dp->dccps_packet_size = val; -- cgit v1.2.3-70-g09d2 From 93ec2c723e3f8a216dde2899aeb85c648672bc6b Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 26 Oct 2006 15:46:50 -0700 Subject: netpoll info leak After looking harder, Steve noticed that the netpoll beast leaked a little every time it shutdown for a nap. Not a big leak, but a nuisance kind of thing. He took out his refcount duct tape and patched the leak. It was overkill since there was already other locking in that area, but it looked clean and wouldn't attract fleas. Signed-off-by: Stephen Hemminger --- include/linux/netpoll.h | 1 + net/core/netpoll.c | 25 +++++++++++++++++++------ 2 files changed, 20 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index 1efe60c5c00..39845fc975f 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -25,6 +25,7 @@ struct netpoll { }; struct netpoll_info { + atomic_t refcnt; spinlock_t poll_lock; int poll_owner; int tries; diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 4de62f1f413..c66df2f45d2 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -658,8 +658,11 @@ int netpoll_setup(struct netpoll *np) npinfo->tries = MAX_RETRIES; spin_lock_init(&npinfo->rx_lock); skb_queue_head_init(&npinfo->arp_tx); - } else + atomic_set(&npinfo->refcnt, 1); + } else { npinfo = ndev->npinfo; + atomic_inc(&npinfo->refcnt); + } if (!ndev->poll_controller) { printk(KERN_ERR "%s: %s doesn't support polling, aborting.\n", @@ -766,12 +769,22 @@ void netpoll_cleanup(struct netpoll *np) if (np->dev) { npinfo = np->dev->npinfo; - if (npinfo && npinfo->rx_np == np) { - spin_lock_irqsave(&npinfo->rx_lock, flags); - npinfo->rx_np = NULL; - npinfo->rx_flags &= ~NETPOLL_RX_ENABLED; - spin_unlock_irqrestore(&npinfo->rx_lock, flags); + if (npinfo) { + if (npinfo->rx_np == np) { + spin_lock_irqsave(&npinfo->rx_lock, flags); + npinfo->rx_np = NULL; + npinfo->rx_flags &= ~NETPOLL_RX_ENABLED; + spin_unlock_irqrestore(&npinfo->rx_lock, flags); + } + + np->dev->npinfo = NULL; + if (atomic_dec_and_test(&npinfo->refcnt)) { + skb_queue_purge(&npinfo->arp_tx); + + kfree(npinfo); + } } + dev_put(np->dev); } -- cgit v1.2.3-70-g09d2 From b6cd27ed33886a5ffaf0925a6d98e13e18e8a1af Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 26 Oct 2006 15:46:51 -0700 Subject: netpoll per device txq When the netpoll beast got really busy, it tended to clog things, so it stored them for later. But the beast was putting all it's skb's in one basket. This was bad because maybe some pipes were clogged and others were not. Signed-off-by: Stephen Hemminger --- include/linux/netpoll.h | 2 ++ net/core/netpoll.c | 50 +++++++++++++++---------------------------------- 2 files changed, 17 insertions(+), 35 deletions(-) (limited to 'include') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index 39845fc975f..93a8b766442 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -33,6 +33,8 @@ struct netpoll_info { spinlock_t rx_lock; struct netpoll *rx_np; /* netpoll that registered an rx_hook */ struct sk_buff_head arp_tx; /* list of arp requests to reply to */ + struct sk_buff_head txq; + struct work_struct tx_work; }; void netpoll_poll(struct netpoll *np); diff --git a/net/core/netpoll.c b/net/core/netpoll.c index c66df2f45d2..ac4e8b8f57d 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -38,10 +38,6 @@ static struct sk_buff_head skb_pool; -static DEFINE_SPINLOCK(queue_lock); -static int queue_depth; -static struct sk_buff *queue_head, *queue_tail; - static atomic_t trapped; #define NETPOLL_RX_ENABLED 1 @@ -56,46 +52,25 @@ static void arp_reply(struct sk_buff *skb); static void queue_process(void *p) { - unsigned long flags; + struct netpoll_info *npinfo = p; struct sk_buff *skb; - while (queue_head) { - spin_lock_irqsave(&queue_lock, flags); - - skb = queue_head; - queue_head = skb->next; - if (skb == queue_tail) - queue_head = NULL; - - queue_depth--; - - spin_unlock_irqrestore(&queue_lock, flags); - + while ((skb = skb_dequeue(&npinfo->txq))) dev_queue_xmit(skb); - } -} -static DECLARE_WORK(send_queue, queue_process, NULL); +} void netpoll_queue(struct sk_buff *skb) { - unsigned long flags; + struct net_device *dev = skb->dev; + struct netpoll_info *npinfo = dev->npinfo; - if (queue_depth == MAX_QUEUE_DEPTH) { - __kfree_skb(skb); - return; + if (!npinfo) + kfree_skb(skb); + else { + skb_queue_tail(&npinfo->txq, skb); + schedule_work(&npinfo->tx_work); } - - spin_lock_irqsave(&queue_lock, flags); - if (!queue_head) - queue_head = skb; - else - queue_tail->next = skb; - queue_tail = skb; - queue_depth++; - spin_unlock_irqrestore(&queue_lock, flags); - - schedule_work(&send_queue); } static int checksum_udp(struct sk_buff *skb, struct udphdr *uh, @@ -658,6 +633,9 @@ int netpoll_setup(struct netpoll *np) npinfo->tries = MAX_RETRIES; spin_lock_init(&npinfo->rx_lock); skb_queue_head_init(&npinfo->arp_tx); + skb_queue_head_init(&npinfo->txq); + INIT_WORK(&npinfo->tx_work, queue_process, npinfo); + atomic_set(&npinfo->refcnt, 1); } else { npinfo = ndev->npinfo; @@ -780,6 +758,8 @@ void netpoll_cleanup(struct netpoll *np) np->dev->npinfo = NULL; if (atomic_dec_and_test(&npinfo->refcnt)) { skb_queue_purge(&npinfo->arp_tx); + skb_queue_purge(&npinfo->txq); + flush_scheduled_work(); kfree(npinfo); } -- cgit v1.2.3-70-g09d2 From 2bdfe0baeca0e2750037b8fba71905c00ac3c515 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 26 Oct 2006 15:46:54 -0700 Subject: netpoll retry cleanup The netpoll beast was still not happy. If the beast got clogged pipes, it tended to stare blankly off in space for a long time. The problem couldn't be completely fixed because the beast talked with irq's disabled. But it could be made less painful and shorter. Signed-off-by: Stephen Hemminger --- include/linux/netpoll.h | 1 - net/core/netpoll.c | 71 +++++++++++++++++++++++-------------------------- 2 files changed, 33 insertions(+), 39 deletions(-) (limited to 'include') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index 93a8b766442..c65d12ec7bb 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -28,7 +28,6 @@ struct netpoll_info { atomic_t refcnt; spinlock_t poll_lock; int poll_owner; - int tries; int rx_flags; spinlock_t rx_lock; struct netpoll *rx_np; /* netpoll that registered an rx_hook */ diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 93cb828f3aa..6b34c394672 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -34,12 +34,12 @@ #define MAX_UDP_CHUNK 1460 #define MAX_SKBS 32 #define MAX_QUEUE_DEPTH (MAX_SKBS / 2) -#define MAX_RETRIES 20000 static struct sk_buff_head skb_pool; static atomic_t trapped; +#define USEC_PER_POLL 50 #define NETPOLL_RX_ENABLED 1 #define NETPOLL_RX_DROP 2 @@ -72,6 +72,7 @@ static void queue_process(void *p) schedule_delayed_work(&npinfo->tx_work, HZ/10); return; } + netif_tx_unlock_bh(dev); } } @@ -244,50 +245,44 @@ repeat: static void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) { - int status; - struct netpoll_info *npinfo; + int status = NETDEV_TX_BUSY; + unsigned long tries; + struct net_device *dev = np->dev; + struct netpoll_info *npinfo = np->dev->npinfo; + + if (!npinfo || !netif_running(dev) || !netif_device_present(dev)) { + __kfree_skb(skb); + return; + } + + /* don't get messages out of order, and no recursion */ + if ( !(np->drop == netpoll_queue && skb_queue_len(&npinfo->txq)) + && npinfo->poll_owner != smp_processor_id() + && netif_tx_trylock(dev)) { + + /* try until next clock tick */ + for(tries = jiffies_to_usecs(1)/USEC_PER_POLL; tries > 0; --tries) { + if (!netif_queue_stopped(dev)) + status = dev->hard_start_xmit(skb, dev); + + if (status == NETDEV_TX_OK) + break; - if (!np || !np->dev || !netif_running(np->dev)) { - __kfree_skb(skb); - return; - } + /* tickle device maybe there is some cleanup */ + netpoll_poll(np); - npinfo = np->dev->npinfo; + udelay(USEC_PER_POLL); + } + netif_tx_unlock(dev); + } - /* avoid recursion */ - if (npinfo->poll_owner == smp_processor_id() || - np->dev->xmit_lock_owner == smp_processor_id()) { + if (status != NETDEV_TX_OK) { + /* requeue for later */ if (np->drop) np->drop(skb); else __kfree_skb(skb); - return; } - - do { - npinfo->tries--; - netif_tx_lock(np->dev); - - /* - * network drivers do not expect to be called if the queue is - * stopped. - */ - status = NETDEV_TX_BUSY; - if (!netif_queue_stopped(np->dev)) - status = np->dev->hard_start_xmit(skb, np->dev); - - netif_tx_unlock(np->dev); - - /* success */ - if(!status) { - npinfo->tries = MAX_RETRIES; /* reset */ - return; - } - - /* transmit busy */ - netpoll_poll(np); - udelay(50); - } while (npinfo->tries > 0); } void netpoll_send_udp(struct netpoll *np, const char *msg, int len) @@ -649,7 +644,7 @@ int netpoll_setup(struct netpoll *np) npinfo->rx_np = NULL; spin_lock_init(&npinfo->poll_lock); npinfo->poll_owner = -1; - npinfo->tries = MAX_RETRIES; + spin_lock_init(&npinfo->rx_lock); skb_queue_head_init(&npinfo->arp_tx); skb_queue_head_init(&npinfo->txq); -- cgit v1.2.3-70-g09d2 From 5de4a473bda49554e4e9bd93b78f43c49a7ea69c Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 26 Oct 2006 15:46:55 -0700 Subject: netpoll queue cleanup The beast had a long and not very happy history. At one point, a friend (netdump) had asked that he open up a little. Well, the friend was long gone now, and the beast had this dangling piece hanging (netpoll_queue). It wasn't hard to stitch the netpoll_queue back in where it belonged and make everything tidy. Signed-off-by: Stephen Hemminger --- drivers/net/netconsole.c | 1 - include/linux/netpoll.h | 4 ++-- net/core/netpoll.c | 23 +++-------------------- 3 files changed, 5 insertions(+), 23 deletions(-) (limited to 'include') diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c index e6e8a9797b6..69233f6aa05 100644 --- a/drivers/net/netconsole.c +++ b/drivers/net/netconsole.c @@ -60,7 +60,6 @@ static struct netpoll np = { .local_port = 6665, .remote_port = 6666, .remote_mac = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff}, - .drop = netpoll_queue, }; static int configured = 0; diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index c65d12ec7bb..b7eb008c43d 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -18,7 +18,7 @@ struct netpoll { struct net_device *dev; char dev_name[16], *name; void (*rx_hook)(struct netpoll *, int, char *, int); - void (*drop)(struct sk_buff *skb); + u32 local_ip, remote_ip; u16 local_port, remote_port; unsigned char local_mac[6], remote_mac[6]; @@ -44,7 +44,7 @@ int netpoll_trap(void); void netpoll_set_trap(int trap); void netpoll_cleanup(struct netpoll *np); int __netpoll_rx(struct sk_buff *skb); -void netpoll_queue(struct sk_buff *skb); + #ifdef CONFIG_NETPOLL static inline int netpoll_rx(struct sk_buff *skb) diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 6b34c394672..0d1de3c47a0 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -77,19 +77,6 @@ static void queue_process(void *p) } } -void netpoll_queue(struct sk_buff *skb) -{ - struct net_device *dev = skb->dev; - struct netpoll_info *npinfo = dev->npinfo; - - if (!npinfo) - kfree_skb(skb); - else { - skb_queue_tail(&npinfo->txq, skb); - schedule_work(&npinfo->tx_work); - } -} - static int checksum_udp(struct sk_buff *skb, struct udphdr *uh, unsigned short ulen, u32 saddr, u32 daddr) { @@ -256,7 +243,7 @@ static void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) } /* don't get messages out of order, and no recursion */ - if ( !(np->drop == netpoll_queue && skb_queue_len(&npinfo->txq)) + if ( skb_queue_len(&npinfo->txq) == 0 && npinfo->poll_owner != smp_processor_id() && netif_tx_trylock(dev)) { @@ -277,11 +264,8 @@ static void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) } if (status != NETDEV_TX_OK) { - /* requeue for later */ - if (np->drop) - np->drop(skb); - else - __kfree_skb(skb); + skb_queue_tail(&npinfo->txq, skb); + schedule_work(&npinfo->tx_work); } } @@ -809,4 +793,3 @@ EXPORT_SYMBOL(netpoll_setup); EXPORT_SYMBOL(netpoll_cleanup); EXPORT_SYMBOL(netpoll_send_udp); EXPORT_SYMBOL(netpoll_poll); -EXPORT_SYMBOL(netpoll_queue); -- cgit v1.2.3-70-g09d2 From bf6bce71eae386dbc37f93af7e5ad173450d9945 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 26 Oct 2006 15:46:56 -0700 Subject: netpoll header cleanup As Steve left netpoll beast, hopefully not to return soon. He noticed that the header was messy. He straightened it up and polished it a little, then waved goodbye. Signed-off-by: Stephen Hemminger --- include/linux/netpoll.h | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index b7eb008c43d..2cc9867b162 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -12,16 +12,15 @@ #include #include -struct netpoll; - struct netpoll { struct net_device *dev; - char dev_name[16], *name; + char dev_name[IFNAMSIZ]; + const char *name; void (*rx_hook)(struct netpoll *, int, char *, int); u32 local_ip, remote_ip; u16 local_port, remote_port; - unsigned char local_mac[6], remote_mac[6]; + u8 local_mac[ETH_ALEN], remote_mac[ETH_ALEN]; }; struct netpoll_info { -- cgit v1.2.3-70-g09d2 From cfb6eeb4c860592edd123fdea908d23c6ad1c7dc Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Tue, 14 Nov 2006 19:07:45 -0800 Subject: [TCP]: MD5 Signature Option (RFC2385) support. Based on implementation by Rick Payne. Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- CREDITS | 3 + include/linux/tcp.h | 35 ++- include/net/request_sock.h | 3 +- include/net/tcp.h | 143 ++++++++++ include/net/timewait_sock.h | 3 + net/dccp/ipv4.c | 6 +- net/dccp/ipv6.c | 6 +- net/dccp/minisocks.c | 2 +- net/ipv4/Kconfig | 16 ++ net/ipv4/tcp.c | 137 +++++++++ net/ipv4/tcp_input.c | 8 + net/ipv4/tcp_ipv4.c | 673 ++++++++++++++++++++++++++++++++++++++++++-- net/ipv4/tcp_minisocks.c | 64 ++++- net/ipv4/tcp_output.c | 111 +++++++- net/ipv6/tcp_ipv6.c | 568 +++++++++++++++++++++++++++++++++++-- 15 files changed, 1714 insertions(+), 64 deletions(-) (limited to 'include') diff --git a/CREDITS b/CREDITS index ccd4f9f4dd7..d0880082c19 100644 --- a/CREDITS +++ b/CREDITS @@ -2598,6 +2598,9 @@ S: Ucitelska 1576 S: Prague 8 S: 182 00 Czech Republic +N: Rick Payne +D: RFC2385 Support for TCP + N: Barak A. Pearlmutter E: bap@cs.unm.edu W: http://www.cs.unm.edu/~bap/ diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 2d36f6db370..0aecfc95559 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -19,6 +19,7 @@ #include #include +#include struct tcphdr { __be16 source; @@ -94,6 +95,7 @@ enum { #define TCP_INFO 11 /* Information about this connection. */ #define TCP_QUICKACK 12 /* Block/reenable quick acks */ #define TCP_CONGESTION 13 /* Congestion control algorithm */ +#define TCP_MD5SIG 14 /* TCP MD5 Signature (RFC2385) */ #define TCPI_OPT_TIMESTAMPS 1 #define TCPI_OPT_SACK 2 @@ -157,6 +159,17 @@ struct tcp_info __u32 tcpi_total_retrans; }; +/* for TCP_MD5SIG socket option */ +#define TCP_MD5SIG_MAXKEYLEN 80 + +struct tcp_md5sig { + struct __kernel_sockaddr_storage tcpm_addr; /* address associated */ + __u16 __tcpm_pad1; /* zero */ + __u16 tcpm_keylen; /* key length */ + __u32 __tcpm_pad2; /* zero */ + __u8 tcpm_key[TCP_MD5SIG_MAXKEYLEN]; /* key (binary) */ +}; + #ifdef __KERNEL__ #include @@ -197,9 +210,13 @@ struct tcp_options_received { }; struct tcp_request_sock { - struct inet_request_sock req; - __u32 rcv_isn; - __u32 snt_isn; + struct inet_request_sock req; +#ifdef CONFIG_TCP_MD5SIG + /* Only used by TCP MD5 Signature so far. */ + struct tcp_request_sock_ops *af_specific; +#endif + __u32 rcv_isn; + __u32 snt_isn; }; static inline struct tcp_request_sock *tcp_rsk(const struct request_sock *req) @@ -363,6 +380,14 @@ struct tcp_sock { __u32 probe_seq_start; __u32 probe_seq_end; } mtu_probe; + +#ifdef CONFIG_TCP_MD5SIG +/* TCP AF-Specific parts; only used by MD5 Signature support so far */ + struct tcp_sock_af_ops *af_specific; + +/* TCP MD5 Signagure Option information */ + struct tcp_md5sig_info *md5sig_info; +#endif }; static inline struct tcp_sock *tcp_sk(const struct sock *sk) @@ -377,6 +402,10 @@ struct tcp_timewait_sock { __u32 tw_rcv_wnd; __u32 tw_ts_recent; long tw_ts_recent_stamp; +#ifdef CONFIG_TCP_MD5SIG + __u16 tw_md5_keylen; + __u8 tw_md5_key[TCP_MD5SIG_MAXKEYLEN]; +#endif }; static inline struct tcp_timewait_sock *tcp_twsk(const struct sock *sk) diff --git a/include/net/request_sock.h b/include/net/request_sock.h index b5b023e79e5..e37baaf2080 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -35,7 +35,8 @@ struct request_sock_ops { struct dst_entry *dst); void (*send_ack)(struct sk_buff *skb, struct request_sock *req); - void (*send_reset)(struct sk_buff *skb); + void (*send_reset)(struct sock *sk, + struct sk_buff *skb); void (*destructor)(struct request_sock *req); }; diff --git a/include/net/tcp.h b/include/net/tcp.h index e1a5d29d0a1..363960872de 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -28,6 +28,7 @@ #include #include #include +#include #include #include @@ -161,6 +162,7 @@ extern void tcp_time_wait(struct sock *sk, int state, int timeo); #define TCPOPT_SACK_PERM 4 /* SACK Permitted */ #define TCPOPT_SACK 5 /* SACK Block */ #define TCPOPT_TIMESTAMP 8 /* Better RTT estimations/PAWS */ +#define TCPOPT_MD5SIG 19 /* MD5 Signature (RFC2385) */ /* * TCP option lengths @@ -170,6 +172,7 @@ extern void tcp_time_wait(struct sock *sk, int state, int timeo); #define TCPOLEN_WINDOW 3 #define TCPOLEN_SACK_PERM 2 #define TCPOLEN_TIMESTAMP 10 +#define TCPOLEN_MD5SIG 18 /* But this is what stacks really send out. */ #define TCPOLEN_TSTAMP_ALIGNED 12 @@ -178,6 +181,7 @@ extern void tcp_time_wait(struct sock *sk, int state, int timeo); #define TCPOLEN_SACK_BASE 2 #define TCPOLEN_SACK_BASE_ALIGNED 4 #define TCPOLEN_SACK_PERBLOCK 8 +#define TCPOLEN_MD5SIG_ALIGNED 20 /* Flags in tp->nonagle */ #define TCP_NAGLE_OFF 1 /* Nagle's algo is disabled */ @@ -299,6 +303,8 @@ extern void tcp_cleanup_rbuf(struct sock *sk, int copied); extern int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp); +extern void tcp_twsk_destructor(struct sock *sk); + static inline void tcp_dec_quickack_mode(struct sock *sk, const unsigned int pkts) { @@ -1064,6 +1070,114 @@ static inline void clear_all_retrans_hints(struct tcp_sock *tp){ tp->fastpath_skb_hint = NULL; } +/* MD5 Signature */ +struct crypto_hash; + +/* - key database */ +struct tcp_md5sig_key { + u8 *key; + u8 keylen; +}; + +struct tcp4_md5sig_key { + u8 *key; + u16 keylen; + __be32 addr; +}; + +struct tcp6_md5sig_key { + u8 *key; + u16 keylen; +#if 0 + u32 scope_id; /* XXX */ +#endif + struct in6_addr addr; +}; + +/* - sock block */ +struct tcp_md5sig_info { + struct tcp4_md5sig_key *keys4; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + struct tcp6_md5sig_key *keys6; + u32 entries6; + u32 alloced6; +#endif + u32 entries4; + u32 alloced4; +}; + +/* - pseudo header */ +struct tcp4_pseudohdr { + __be32 saddr; + __be32 daddr; + __u8 pad; + __u8 protocol; + __be16 len; +}; + +struct tcp6_pseudohdr { + struct in6_addr saddr; + struct in6_addr daddr; + __be32 len; + __be32 protocol; /* including padding */ +}; + +union tcp_md5sum_block { + struct tcp4_pseudohdr ip4; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + struct tcp6_pseudohdr ip6; +#endif +}; + +/* - pool: digest algorithm, hash description and scratch buffer */ +struct tcp_md5sig_pool { + struct hash_desc md5_desc; + union tcp_md5sum_block md5_blk; +}; + +#define TCP_MD5SIG_MAXKEYS (~(u32)0) /* really?! */ + +/* - functions */ +extern int tcp_v4_calc_md5_hash(char *md5_hash, + struct tcp_md5sig_key *key, + struct sock *sk, + struct dst_entry *dst, + struct request_sock *req, + struct tcphdr *th, + int protocol, int tcplen); +extern struct tcp_md5sig_key *tcp_v4_md5_lookup(struct sock *sk, + struct sock *addr_sk); + +extern int tcp_v4_md5_do_add(struct sock *sk, + __be32 addr, + u8 *newkey, + u8 newkeylen); + +extern int tcp_v4_md5_do_del(struct sock *sk, + u32 addr); + +extern struct tcp_md5sig_pool **tcp_alloc_md5sig_pool(void); +extern void tcp_free_md5sig_pool(void); + +extern struct tcp_md5sig_pool *__tcp_get_md5sig_pool(int cpu); +extern void __tcp_put_md5sig_pool(void); + +static inline +struct tcp_md5sig_pool *tcp_get_md5sig_pool(void) +{ + int cpu = get_cpu(); + struct tcp_md5sig_pool *ret = __tcp_get_md5sig_pool(cpu); + if (!ret) + put_cpu(); + return ret; +} + +static inline void tcp_put_md5sig_pool(void) +{ + __tcp_put_md5sig_pool(); + put_cpu(); +} + /* /proc */ enum tcp_seq_states { TCP_SEQ_STATE_LISTENING, @@ -1103,6 +1217,35 @@ extern int tcp4_proc_init(void); extern void tcp4_proc_exit(void); #endif +/* TCP af-specific functions */ +struct tcp_sock_af_ops { +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *(*md5_lookup) (struct sock *sk, + struct sock *addr_sk); + int (*calc_md5_hash) (char *location, + struct tcp_md5sig_key *md5, + struct sock *sk, + struct dst_entry *dst, + struct request_sock *req, + struct tcphdr *th, + int protocol, int len); + int (*md5_add) (struct sock *sk, + struct sock *addr_sk, + u8 *newkey, + u8 len); + int (*md5_parse) (struct sock *sk, + char __user *optval, + int optlen); +#endif +}; + +struct tcp_request_sock_ops { +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *(*md5_lookup) (struct sock *sk, + struct request_sock *req); +#endif +}; + extern void tcp_v4_init(struct net_proto_family *ops); extern void tcp_init(void); diff --git a/include/net/timewait_sock.h b/include/net/timewait_sock.h index be293d795e3..d7a306ea560 100644 --- a/include/net/timewait_sock.h +++ b/include/net/timewait_sock.h @@ -31,6 +31,9 @@ static inline int twsk_unique(struct sock *sk, struct sock *sktw, void *twp) static inline void twsk_destructor(struct sock *sk) { + BUG_ON(sk == NULL); + BUG_ON(sk->sk_prot == NULL); + BUG_ON(sk->sk_prot->twsk_prot == NULL); if (sk->sk_prot->twsk_prot->twsk_destructor != NULL) sk->sk_prot->twsk_prot->twsk_destructor(sk); } diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 34d6d197c3b..35985334dae 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -509,7 +509,7 @@ out: return err; } -static void dccp_v4_ctl_send_reset(struct sk_buff *rxskb) +static void dccp_v4_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) { int err; struct dccp_hdr *rxdh = dccp_hdr(rxskb), *dh; @@ -724,7 +724,7 @@ int dccp_v4_do_rcv(struct sock *sk, struct sk_buff *skb) return 0; reset: - dccp_v4_ctl_send_reset(skb); + dccp_v4_ctl_send_reset(sk, skb); discard: kfree_skb(skb); return 0; @@ -913,7 +913,7 @@ no_dccp_socket: if (dh->dccph_type != DCCP_PKT_RESET) { DCCP_SKB_CB(skb)->dccpd_reset_code = DCCP_RESET_CODE_NO_CONNECTION; - dccp_v4_ctl_send_reset(skb); + dccp_v4_ctl_send_reset(sk, skb); } discard_it: diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index fc326173c21..e0a0607862e 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -310,7 +310,7 @@ static void dccp_v6_reqsk_destructor(struct request_sock *req) kfree_skb(inet6_rsk(req)->pktopts); } -static void dccp_v6_ctl_send_reset(struct sk_buff *rxskb) +static void dccp_v6_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) { struct dccp_hdr *rxdh = dccp_hdr(rxskb), *dh; const u32 dccp_hdr_reset_len = sizeof(struct dccp_hdr) + @@ -805,7 +805,7 @@ static int dccp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) return 0; reset: - dccp_v6_ctl_send_reset(skb); + dccp_v6_ctl_send_reset(sk, skb); discard: if (opt_skb != NULL) __kfree_skb(opt_skb); @@ -902,7 +902,7 @@ no_dccp_socket: if (dh->dccph_type != DCCP_PKT_RESET) { DCCP_SKB_CB(skb)->dccpd_reset_code = DCCP_RESET_CODE_NO_CONNECTION; - dccp_v6_ctl_send_reset(skb); + dccp_v6_ctl_send_reset(sk, skb); } discard_it: diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c index 0c49733f5be..3975048d809 100644 --- a/net/dccp/minisocks.c +++ b/net/dccp/minisocks.c @@ -246,7 +246,7 @@ listen_overflow: DCCP_SKB_CB(skb)->dccpd_reset_code = DCCP_RESET_CODE_TOO_BUSY; drop: if (dccp_hdr(skb)->dccph_type != DCCP_PKT_RESET) - req->rsk_ops->send_reset(skb); + req->rsk_ops->send_reset(sk, skb); inet_csk_reqsk_queue_drop(sk, req, prev); goto out; diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig index bc298bcc344..39e0cb76358 100644 --- a/net/ipv4/Kconfig +++ b/net/ipv4/Kconfig @@ -618,5 +618,21 @@ config DEFAULT_TCP_CONG default "reno" if DEFAULT_RENO default "cubic" +config TCP_MD5SIG + bool "TCP: MD5 Signature Option support (RFC2385) (EXPERIMENTAL)" + depends on EXPERIMENTAL + select CRYPTO + select CRYPTO_MD5 + ---help--- + RFC2385 specifices a method of giving MD5 protection to TCP sessions. + Its main (only?) use is to protect BGP sessions between core routers + on the Internet. + + If unsure, say N. + +config TCP_MD5SIG_DEBUG + bool "TCP: MD5 Signature Option debugging" + depends on TCP_MD5SIG + source "net/ipv4/ipvs/Kconfig" diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index c05e8edaf54..dadef867a3b 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -258,6 +258,7 @@ #include #include #include +#include #include #include @@ -1942,6 +1943,13 @@ static int do_tcp_setsockopt(struct sock *sk, int level, } break; +#ifdef CONFIG_TCP_MD5SIG + case TCP_MD5SIG: + /* Read the IP->Key mappings from userspace */ + err = tp->af_specific->md5_parse(sk, optval, optlen); + break; +#endif + default: err = -ENOPROTOOPT; break; @@ -2231,6 +2239,135 @@ out: } EXPORT_SYMBOL(tcp_tso_segment); +#ifdef CONFIG_TCP_MD5SIG +static unsigned long tcp_md5sig_users; +static struct tcp_md5sig_pool **tcp_md5sig_pool; +static DEFINE_SPINLOCK(tcp_md5sig_pool_lock); + +static void __tcp_free_md5sig_pool(struct tcp_md5sig_pool **pool) +{ + int cpu; + for_each_possible_cpu(cpu) { + struct tcp_md5sig_pool *p = *per_cpu_ptr(pool, cpu); + if (p) { + if (p->md5_desc.tfm) + crypto_free_hash(p->md5_desc.tfm); + kfree(p); + p = NULL; + } + } + free_percpu(pool); +} + +void tcp_free_md5sig_pool(void) +{ + struct tcp_md5sig_pool **pool = NULL; + + spin_lock(&tcp_md5sig_pool_lock); + if (--tcp_md5sig_users == 0) { + pool = tcp_md5sig_pool; + tcp_md5sig_pool = NULL; + } + spin_unlock(&tcp_md5sig_pool_lock); + if (pool) + __tcp_free_md5sig_pool(pool); +} + +EXPORT_SYMBOL(tcp_free_md5sig_pool); + +struct tcp_md5sig_pool **__tcp_alloc_md5sig_pool(void) +{ + int cpu; + struct tcp_md5sig_pool **pool; + + pool = alloc_percpu(struct tcp_md5sig_pool *); + if (!pool) + return NULL; + + for_each_possible_cpu(cpu) { + struct tcp_md5sig_pool *p; + struct crypto_hash *hash; + + p = kzalloc(sizeof(*p), GFP_KERNEL); + if (!p) + goto out_free; + *per_cpu_ptr(pool, cpu) = p; + + hash = crypto_alloc_hash("md5", 0, CRYPTO_ALG_ASYNC); + if (!hash || IS_ERR(hash)) + goto out_free; + + p->md5_desc.tfm = hash; + } + return pool; +out_free: + __tcp_free_md5sig_pool(pool); + return NULL; +} + +struct tcp_md5sig_pool **tcp_alloc_md5sig_pool(void) +{ + struct tcp_md5sig_pool **pool; + int alloc = 0; + +retry: + spin_lock(&tcp_md5sig_pool_lock); + pool = tcp_md5sig_pool; + if (tcp_md5sig_users++ == 0) { + alloc = 1; + spin_unlock(&tcp_md5sig_pool_lock); + } else if (!pool) { + tcp_md5sig_users--; + spin_unlock(&tcp_md5sig_pool_lock); + cpu_relax(); + goto retry; + } else + spin_unlock(&tcp_md5sig_pool_lock); + + if (alloc) { + /* we cannot hold spinlock here because this may sleep. */ + struct tcp_md5sig_pool **p = __tcp_alloc_md5sig_pool(); + spin_lock(&tcp_md5sig_pool_lock); + if (!p) { + tcp_md5sig_users--; + spin_unlock(&tcp_md5sig_pool_lock); + return NULL; + } + pool = tcp_md5sig_pool; + if (pool) { + /* oops, it has already been assigned. */ + spin_unlock(&tcp_md5sig_pool_lock); + __tcp_free_md5sig_pool(p); + } else { + tcp_md5sig_pool = pool = p; + spin_unlock(&tcp_md5sig_pool_lock); + } + } + return pool; +} + +EXPORT_SYMBOL(tcp_alloc_md5sig_pool); + +struct tcp_md5sig_pool *__tcp_get_md5sig_pool(int cpu) +{ + struct tcp_md5sig_pool **p; + spin_lock(&tcp_md5sig_pool_lock); + p = tcp_md5sig_pool; + if (p) + tcp_md5sig_users++; + spin_unlock(&tcp_md5sig_pool_lock); + return (p ? *per_cpu_ptr(p, cpu) : NULL); +} + +EXPORT_SYMBOL(__tcp_get_md5sig_pool); + +void __tcp_put_md5sig_pool(void) { + __tcp_free_md5sig_pool(tcp_md5sig_pool); +} + +EXPORT_SYMBOL(__tcp_put_md5sig_pool); +#endif + extern void __skb_cb_too_small_for_tcp(int, int); extern struct tcp_congestion_ops tcp_reno; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 4a8c96cdec7..6ab3423674b 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -2677,6 +2677,14 @@ void tcp_parse_options(struct sk_buff *skb, struct tcp_options_received *opt_rx, opt_rx->sack_ok) { TCP_SKB_CB(skb)->sacked = (ptr - 2) - (unsigned char *)th; } +#ifdef CONFIG_TCP_MD5SIG + case TCPOPT_MD5SIG: + /* + * The MD5 Hash has already been + * checked (see tcp_v{4,6}_do_rcv()). + */ + break; +#endif }; ptr+=opsize-2; length-=opsize; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 0ad0904bf56..8c8e8112f98 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -78,6 +78,9 @@ #include #include +#include +#include + int sysctl_tcp_tw_reuse __read_mostly; int sysctl_tcp_low_latency __read_mostly; @@ -89,6 +92,13 @@ static struct socket *tcp_socket; void tcp_v4_send_check(struct sock *sk, int len, struct sk_buff *skb); +#ifdef CONFIG_TCP_MD5SIG +static struct tcp_md5sig_key *tcp_v4_md5_do_lookup(struct sock *sk, __be32 addr); +static int tcp_v4_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key, + __be32 saddr, __be32 daddr, struct tcphdr *th, + int protocol, int tcplen); +#endif + struct inet_hashinfo __cacheline_aligned tcp_hashinfo = { .lhash_lock = __RW_LOCK_UNLOCKED(tcp_hashinfo.lhash_lock), .lhash_users = ATOMIC_INIT(0), @@ -526,11 +536,19 @@ int tcp_v4_gso_send_check(struct sk_buff *skb) * Exception: precedence violation. We do not implement it in any case. */ -static void tcp_v4_send_reset(struct sk_buff *skb) +static void tcp_v4_send_reset(struct sock *sk, struct sk_buff *skb) { struct tcphdr *th = skb->h.th; - struct tcphdr rth; + struct { + struct tcphdr th; +#ifdef CONFIG_TCP_MD5SIG + u32 opt[(TCPOLEN_MD5SIG_ALIGNED >> 2)]; +#endif + } rep; struct ip_reply_arg arg; +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *key; +#endif /* Never send a reset in response to a reset. */ if (th->rst) @@ -540,29 +558,50 @@ static void tcp_v4_send_reset(struct sk_buff *skb) return; /* Swap the send and the receive. */ - memset(&rth, 0, sizeof(struct tcphdr)); - rth.dest = th->source; - rth.source = th->dest; - rth.doff = sizeof(struct tcphdr) / 4; - rth.rst = 1; + memset(&rep, 0, sizeof(rep)); + rep.th.dest = th->source; + rep.th.source = th->dest; + rep.th.doff = sizeof(struct tcphdr) / 4; + rep.th.rst = 1; if (th->ack) { - rth.seq = th->ack_seq; + rep.th.seq = th->ack_seq; } else { - rth.ack = 1; - rth.ack_seq = htonl(ntohl(th->seq) + th->syn + th->fin + - skb->len - (th->doff << 2)); + rep.th.ack = 1; + rep.th.ack_seq = htonl(ntohl(th->seq) + th->syn + th->fin + + skb->len - (th->doff << 2)); } memset(&arg, 0, sizeof arg); - arg.iov[0].iov_base = (unsigned char *)&rth; - arg.iov[0].iov_len = sizeof rth; + arg.iov[0].iov_base = (unsigned char *)&rep; + arg.iov[0].iov_len = sizeof(rep.th); + +#ifdef CONFIG_TCP_MD5SIG + key = sk ? tcp_v4_md5_do_lookup(sk, skb->nh.iph->daddr) : NULL; + if (key) { + rep.opt[0] = htonl((TCPOPT_NOP << 24) | + (TCPOPT_NOP << 16) | + (TCPOPT_MD5SIG << 8) | + TCPOLEN_MD5SIG); + /* Update length and the length the header thinks exists */ + arg.iov[0].iov_len += TCPOLEN_MD5SIG_ALIGNED; + rep.th.doff = arg.iov[0].iov_len / 4; + + tcp_v4_do_calc_md5_hash((__u8 *)&rep.opt[1], + key, + skb->nh.iph->daddr, + skb->nh.iph->saddr, + &rep.th, IPPROTO_TCP, + arg.iov[0].iov_len); + } +#endif + arg.csum = csum_tcpudp_nofold(skb->nh.iph->daddr, skb->nh.iph->saddr, /*XXX*/ sizeof(struct tcphdr), IPPROTO_TCP, 0); arg.csumoffset = offsetof(struct tcphdr, check) / 2; - ip_send_reply(tcp_socket->sk, skb, &arg, sizeof rth); + ip_send_reply(tcp_socket->sk, skb, &arg, arg.iov[0].iov_len); TCP_INC_STATS_BH(TCP_MIB_OUTSEGS); TCP_INC_STATS_BH(TCP_MIB_OUTRSTS); @@ -572,15 +611,24 @@ static void tcp_v4_send_reset(struct sk_buff *skb) outside socket context is ugly, certainly. What can I do? */ -static void tcp_v4_send_ack(struct sk_buff *skb, u32 seq, u32 ack, +static void tcp_v4_send_ack(struct tcp_timewait_sock *twsk, + struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32 ts) { struct tcphdr *th = skb->h.th; struct { struct tcphdr th; - u32 tsopt[TCPOLEN_TSTAMP_ALIGNED >> 2]; + u32 opt[(TCPOLEN_TSTAMP_ALIGNED >> 2) +#ifdef CONFIG_TCP_MD5SIG + + (TCPOLEN_MD5SIG_ALIGNED >> 2) +#endif + ]; } rep; struct ip_reply_arg arg; +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *key; + struct tcp_md5sig_key tw_key; +#endif memset(&rep.th, 0, sizeof(struct tcphdr)); memset(&arg, 0, sizeof arg); @@ -588,12 +636,12 @@ static void tcp_v4_send_ack(struct sk_buff *skb, u32 seq, u32 ack, arg.iov[0].iov_base = (unsigned char *)&rep; arg.iov[0].iov_len = sizeof(rep.th); if (ts) { - rep.tsopt[0] = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | - (TCPOPT_TIMESTAMP << 8) | - TCPOLEN_TIMESTAMP); - rep.tsopt[1] = htonl(tcp_time_stamp); - rep.tsopt[2] = htonl(ts); - arg.iov[0].iov_len = sizeof(rep); + rep.opt[0] = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | + (TCPOPT_TIMESTAMP << 8) | + TCPOLEN_TIMESTAMP); + rep.opt[1] = htonl(tcp_time_stamp); + rep.opt[2] = htonl(ts); + arg.iov[0].iov_len = TCPOLEN_TSTAMP_ALIGNED; } /* Swap the send and the receive. */ @@ -605,6 +653,44 @@ static void tcp_v4_send_ack(struct sk_buff *skb, u32 seq, u32 ack, rep.th.ack = 1; rep.th.window = htons(win); +#ifdef CONFIG_TCP_MD5SIG + /* + * The SKB holds an imcoming packet, but may not have a valid ->sk + * pointer. This is especially the case when we're dealing with a + * TIME_WAIT ack, because the sk structure is long gone, and only + * the tcp_timewait_sock remains. So the md5 key is stashed in that + * structure, and we use it in preference. I believe that (twsk || + * skb->sk) holds true, but we program defensively. + */ + if (!twsk && skb->sk) { + key = tcp_v4_md5_do_lookup(skb->sk, skb->nh.iph->daddr); + } else if (twsk && twsk->tw_md5_keylen) { + tw_key.key = twsk->tw_md5_key; + tw_key.keylen = twsk->tw_md5_keylen; + key = &tw_key; + } else { + key = NULL; + } + + if (key) { + int offset = (ts) ? 3 : 0; + + rep.opt[offset++] = htonl((TCPOPT_NOP << 24) | + (TCPOPT_NOP << 16) | + (TCPOPT_MD5SIG << 8) | + TCPOLEN_MD5SIG); + arg.iov[0].iov_len += TCPOLEN_MD5SIG_ALIGNED; + rep.th.doff = arg.iov[0].iov_len/4; + + tcp_v4_do_calc_md5_hash((__u8 *)&rep.opt[offset], + key, + skb->nh.iph->daddr, + skb->nh.iph->saddr, + &rep.th, IPPROTO_TCP, + arg.iov[0].iov_len); + } +#endif + arg.csum = csum_tcpudp_nofold(skb->nh.iph->daddr, skb->nh.iph->saddr, /*XXX*/ arg.iov[0].iov_len, IPPROTO_TCP, 0); @@ -618,9 +704,9 @@ static void tcp_v4_send_ack(struct sk_buff *skb, u32 seq, u32 ack, static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb) { struct inet_timewait_sock *tw = inet_twsk(sk); - const struct tcp_timewait_sock *tcptw = tcp_twsk(sk); + struct tcp_timewait_sock *tcptw = tcp_twsk(sk); - tcp_v4_send_ack(skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt, + tcp_v4_send_ack(tcptw, skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt, tcptw->tw_rcv_wnd >> tw->tw_rcv_wscale, tcptw->tw_ts_recent); inet_twsk_put(tw); @@ -628,7 +714,8 @@ static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb) static void tcp_v4_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req) { - tcp_v4_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, + tcp_v4_send_ack(NULL, skb, tcp_rsk(req)->snt_isn + 1, + tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent); } @@ -714,6 +801,461 @@ static struct ip_options *tcp_v4_save_options(struct sock *sk, return dopt; } +#ifdef CONFIG_TCP_MD5SIG +/* + * RFC2385 MD5 checksumming requires a mapping of + * IP address->MD5 Key. + * We need to maintain these in the sk structure. + */ + +/* Find the Key structure for an address. */ +static struct tcp_md5sig_key *tcp_v4_md5_do_lookup(struct sock *sk, __be32 addr) +{ + struct tcp_sock *tp = tcp_sk(sk); + int i; + + if (!tp->md5sig_info || !tp->md5sig_info->entries4) + return NULL; + for (i = 0; i < tp->md5sig_info->entries4; i++) { + if (tp->md5sig_info->keys4[i].addr == addr) + return (struct tcp_md5sig_key *)&tp->md5sig_info->keys4[i]; + } + return NULL; +} + +struct tcp_md5sig_key *tcp_v4_md5_lookup(struct sock *sk, + struct sock *addr_sk) +{ + return tcp_v4_md5_do_lookup(sk, inet_sk(addr_sk)->daddr); +} + +EXPORT_SYMBOL(tcp_v4_md5_lookup); + +struct tcp_md5sig_key *tcp_v4_reqsk_md5_lookup(struct sock *sk, + struct request_sock *req) +{ + return tcp_v4_md5_do_lookup(sk, inet_rsk(req)->rmt_addr); +} + +/* This can be called on a newly created socket, from other files */ +int tcp_v4_md5_do_add(struct sock *sk, __be32 addr, + u8 *newkey, u8 newkeylen) +{ + /* Add Key to the list */ + struct tcp4_md5sig_key *key; + struct tcp_sock *tp = tcp_sk(sk); + struct tcp4_md5sig_key *keys; + + key = (struct tcp4_md5sig_key *) tcp_v4_md5_do_lookup(sk, addr); + if (key) { + /* Pre-existing entry - just update that one. */ + kfree (key->key); + key->key = newkey; + key->keylen = newkeylen; + } else { + if (!tp->md5sig_info) { + tp->md5sig_info = kzalloc(sizeof(*tp->md5sig_info), GFP_ATOMIC); + if (!tp->md5sig_info) { + kfree(newkey); + return -ENOMEM; + } + } + if (tcp_alloc_md5sig_pool() == NULL) { + kfree(newkey); + return -ENOMEM; + } + if (tp->md5sig_info->alloced4 == tp->md5sig_info->entries4) { + keys = kmalloc((sizeof(struct tcp4_md5sig_key) * + (tp->md5sig_info->entries4 + 1)), GFP_ATOMIC); + if (!keys) { + kfree(newkey); + tcp_free_md5sig_pool(); + return -ENOMEM; + } + + if (tp->md5sig_info->entries4) + memcpy(keys, tp->md5sig_info->keys4, + (sizeof (struct tcp4_md5sig_key) * + tp->md5sig_info->entries4)); + + /* Free old key list, and reference new one */ + if (tp->md5sig_info->keys4) + kfree(tp->md5sig_info->keys4); + tp->md5sig_info->keys4 = keys; + tp->md5sig_info->alloced4++; + } + tp->md5sig_info->entries4++; + tp->md5sig_info->keys4[tp->md5sig_info->entries4 - 1].addr = addr; + tp->md5sig_info->keys4[tp->md5sig_info->entries4 - 1].key = newkey; + tp->md5sig_info->keys4[tp->md5sig_info->entries4 - 1].keylen = newkeylen; + } + return 0; +} + +EXPORT_SYMBOL(tcp_v4_md5_do_add); + +static int tcp_v4_md5_add_func(struct sock *sk, struct sock *addr_sk, + u8 *newkey, u8 newkeylen) +{ + return tcp_v4_md5_do_add(sk, inet_sk(addr_sk)->daddr, + newkey, newkeylen); +} + +int tcp_v4_md5_do_del(struct sock *sk, __be32 addr) +{ + struct tcp_sock *tp = tcp_sk(sk); + int i; + + for (i = 0; i < tp->md5sig_info->entries4; i++) { + if (tp->md5sig_info->keys4[i].addr == addr) { + /* Free the key */ + kfree(tp->md5sig_info->keys4[i].key); + tp->md5sig_info->entries4--; + + if (tp->md5sig_info->entries4 == 0) { + kfree(tp->md5sig_info->keys4); + tp->md5sig_info->keys4 = NULL; + } else { + /* Need to do some manipulation */ + if (tp->md5sig_info->entries4 != i) + memcpy(&tp->md5sig_info->keys4[i], + &tp->md5sig_info->keys4[i+1], + (tp->md5sig_info->entries4 - i) + * sizeof (struct tcp4_md5sig_key)); + } + tcp_free_md5sig_pool(); + return 0; + } + } + return -ENOENT; +} + +EXPORT_SYMBOL(tcp_v4_md5_do_del); + +static void tcp_v4_clear_md5_list (struct sock *sk) +{ + struct tcp_sock *tp = tcp_sk(sk); + + /* Free each key, then the set of key keys, + * the crypto element, and then decrement our + * hold on the last resort crypto. + */ + if (tp->md5sig_info->entries4) { + int i; + for (i = 0; i < tp->md5sig_info->entries4; i++) + kfree(tp->md5sig_info->keys4[i].key); + tp->md5sig_info->entries4 = 0; + tcp_free_md5sig_pool(); + } + if (tp->md5sig_info->keys4) { + kfree(tp->md5sig_info->keys4); + tp->md5sig_info->keys4 = NULL; + tp->md5sig_info->alloced4 = 0; + } +} + +static int tcp_v4_parse_md5_keys (struct sock *sk, char __user *optval, + int optlen) +{ + struct tcp_md5sig cmd; + struct sockaddr_in *sin = (struct sockaddr_in *)&cmd.tcpm_addr; + u8 *newkey; + + if (optlen < sizeof(cmd)) + return -EINVAL; + + if (copy_from_user (&cmd, optval, sizeof(cmd))) + return -EFAULT; + + if (sin->sin_family != AF_INET) + return -EINVAL; + + if (!cmd.tcpm_key || !cmd.tcpm_keylen) { + if (!tcp_sk(sk)->md5sig_info) + return -ENOENT; + return tcp_v4_md5_do_del(sk, sin->sin_addr.s_addr); + } + + if (cmd.tcpm_keylen > TCP_MD5SIG_MAXKEYLEN) + return -EINVAL; + + if (!tcp_sk(sk)->md5sig_info) { + struct tcp_sock *tp = tcp_sk(sk); + struct tcp_md5sig_info *p; + + p = kzalloc(sizeof(struct tcp_md5sig_info), GFP_KERNEL); + if (!p) + return -EINVAL; + + tp->md5sig_info = p; + + } + + newkey = kmalloc(cmd.tcpm_keylen, GFP_KERNEL); + if (!newkey) + return -ENOMEM; + memcpy(newkey, cmd.tcpm_key, cmd.tcpm_keylen); + return tcp_v4_md5_do_add(sk, sin->sin_addr.s_addr, + newkey, cmd.tcpm_keylen); +} + +static int tcp_v4_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key, + __be32 saddr, __be32 daddr, + struct tcphdr *th, int protocol, + int tcplen) +{ + struct scatterlist sg[4]; + __u16 data_len; + int block = 0; +#ifdef CONFIG_TCP_MD5SIG_DEBUG + int i; +#endif + __u16 old_checksum; + struct tcp_md5sig_pool *hp; + struct tcp4_pseudohdr *bp; + struct hash_desc *desc; + int err; + unsigned int nbytes = 0; + + /* + * Okay, so RFC2385 is turned on for this connection, + * so we need to generate the MD5 hash for the packet now. + */ + + hp = tcp_get_md5sig_pool(); + if (!hp) + goto clear_hash_noput; + + bp = &hp->md5_blk.ip4; + desc = &hp->md5_desc; + + /* + * 1. the TCP pseudo-header (in the order: source IP address, + * destination IP address, zero-padded protocol number, and + * segment length) + */ + bp->saddr = saddr; + bp->daddr = daddr; + bp->pad = 0; + bp->protocol = protocol; + bp->len = htons(tcplen); + sg_set_buf(&sg[block++], bp, sizeof(*bp)); + nbytes += sizeof(*bp); + +#ifdef CONFIG_TCP_MD5SIG_DEBUG + printk("Calcuating hash for: "); + for (i = 0; i < sizeof (*bp); i++) + printk ("%02x ", (unsigned int)((unsigned char *)bp)[i]); + printk(" "); +#endif + + /* 2. the TCP header, excluding options, and assuming a + * checksum of zero/ + */ + old_checksum = th->check; + th->check = 0; + sg_set_buf(&sg[block++], th, sizeof(struct tcphdr)); + nbytes += sizeof(struct tcphdr); +#ifdef CONFIG_TCP_MD5SIG_DEBUG + for (i = 0; i < sizeof (struct tcphdr); i++) + printk (" %02x", (unsigned int)((unsigned char *)th)[i]); +#endif + /* 3. the TCP segment data (if any) */ + data_len = tcplen - (th->doff << 2); + if (data_len > 0) { + unsigned char *data = (unsigned char *)th + (th->doff << 2); + sg_set_buf(&sg[block++], data, data_len); + nbytes += data_len; + } + + /* 4. an independently-specified key or password, known to both + * TCPs and presumably connection-specific + */ + sg_set_buf(&sg[block++], key->key, key->keylen); + nbytes += key->keylen; + +#ifdef CONFIG_TCP_MD5SIG_DEBUG + printk (" and password: "); + for (i = 0; i < key->keylen; i++) + printk ("%02x ", (unsigned int)key->key[i]); +#endif + + /* Now store the Hash into the packet */ + err = crypto_hash_init(desc); + if (err) + goto clear_hash; + err = crypto_hash_update(desc, sg, nbytes); + if (err) + goto clear_hash; + err = crypto_hash_final(desc, md5_hash); + if (err) + goto clear_hash; + + /* Reset header, and free up the crypto */ + tcp_put_md5sig_pool(); + th->check = old_checksum; + +out: +#ifdef CONFIG_TCP_MD5SIG_DEBUG + printk(" result:"); + for (i = 0; i < 16; i++) + printk (" %02x", (unsigned int)(((u8*)md5_hash)[i])); + printk("\n"); +#endif + return 0; +clear_hash: + tcp_put_md5sig_pool(); +clear_hash_noput: + memset(md5_hash, 0, 16); + goto out; +} + +int tcp_v4_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key, + struct sock *sk, + struct dst_entry *dst, + struct request_sock *req, + struct tcphdr *th, int protocol, + int tcplen) +{ + __be32 saddr, daddr; + + if (sk) { + saddr = inet_sk(sk)->saddr; + daddr = inet_sk(sk)->daddr; + } else { + struct rtable *rt = (struct rtable *)dst; + BUG_ON(!rt); + saddr = rt->rt_src; + daddr = rt->rt_dst; + } + return tcp_v4_do_calc_md5_hash(md5_hash, key, + saddr, daddr, + th, protocol, tcplen); +} + +EXPORT_SYMBOL(tcp_v4_calc_md5_hash); + +static int tcp_v4_inbound_md5_hash (struct sock *sk, struct sk_buff *skb) +{ + /* + * This gets called for each TCP segment that arrives + * so we want to be efficient. + * We have 3 drop cases: + * o No MD5 hash and one expected. + * o MD5 hash and we're not expecting one. + * o MD5 hash and its wrong. + */ + __u8 *hash_location = NULL; + struct tcp_md5sig_key *hash_expected; + struct iphdr *iph = skb->nh.iph; + struct tcphdr *th = skb->h.th; + int length = (th->doff << 2) - sizeof (struct tcphdr); + int genhash; + unsigned char *ptr; + unsigned char newhash[16]; + + hash_expected = tcp_v4_md5_do_lookup(sk, iph->saddr); + + /* + * If the TCP option length is less than the TCP_MD5SIG + * option length, then we can shortcut + */ + if (length < TCPOLEN_MD5SIG) { + if (hash_expected) + return 1; + else + return 0; + } + + /* Okay, we can't shortcut - we have to grub through the options */ + ptr = (unsigned char *)(th + 1); + while (length > 0) { + int opcode = *ptr++; + int opsize; + + switch (opcode) { + case TCPOPT_EOL: + goto done_opts; + case TCPOPT_NOP: + length--; + continue; + default: + opsize = *ptr++; + if (opsize < 2) + goto done_opts; + if (opsize > length) + goto done_opts; + + if (opcode == TCPOPT_MD5SIG) { + hash_location = ptr; + goto done_opts; + } + } + ptr += opsize-2; + length -= opsize; + } +done_opts: + /* We've parsed the options - do we have a hash? */ + if (!hash_expected && !hash_location) + return 0; + + if (hash_expected && !hash_location) { + if (net_ratelimit()) { + printk(KERN_INFO "MD5 Hash NOT expected but found " + "(" NIPQUAD_FMT ", %d)->(" NIPQUAD_FMT ", %d)\n", + NIPQUAD (iph->saddr), ntohs(th->source), + NIPQUAD (iph->daddr), ntohs(th->dest)); + } + return 1; + } + + if (!hash_expected && hash_location) { + if (net_ratelimit()) { + printk(KERN_INFO "MD5 Hash NOT expected but found " + "(" NIPQUAD_FMT ", %d)->(" NIPQUAD_FMT ", %d)\n", + NIPQUAD (iph->saddr), ntohs(th->source), + NIPQUAD (iph->daddr), ntohs(th->dest)); + } + return 1; + } + + /* Okay, so this is hash_expected and hash_location - + * so we need to calculate the checksum. + */ + genhash = tcp_v4_do_calc_md5_hash(newhash, + hash_expected, + iph->saddr, iph->daddr, + th, sk->sk_protocol, + skb->len); + + if (genhash || memcmp(hash_location, newhash, 16) != 0) { + if (net_ratelimit()) { + printk(KERN_INFO "MD5 Hash failed for " + "(" NIPQUAD_FMT ", %d)->(" NIPQUAD_FMT ", %d)%s\n", + NIPQUAD (iph->saddr), ntohs(th->source), + NIPQUAD (iph->daddr), ntohs(th->dest), + genhash ? " tcp_v4_calc_md5_hash failed" : ""); +#ifdef CONFIG_TCP_MD5SIG_DEBUG + do { + int i; + printk("Received: "); + for (i = 0; i < 16; i++) + printk("%02x ", 0xff & (int)hash_location[i]); + printk("\n"); + printk("Calculated: "); + for (i = 0; i < 16; i++) + printk("%02x ", 0xff & (int)newhash[i]); + printk("\n"); + } while(0); +#endif + } + return 1; + } + return 0; +} + +#endif + struct request_sock_ops tcp_request_sock_ops __read_mostly = { .family = PF_INET, .obj_size = sizeof(struct tcp_request_sock), @@ -723,9 +1265,16 @@ struct request_sock_ops tcp_request_sock_ops __read_mostly = { .send_reset = tcp_v4_send_reset, }; +struct tcp_request_sock_ops tcp_request_sock_ipv4_ops = { +#ifdef CONFIG_TCP_MD5SIG + .md5_lookup = tcp_v4_reqsk_md5_lookup, +#endif +}; + static struct timewait_sock_ops tcp_timewait_sock_ops = { .twsk_obj_size = sizeof(struct tcp_timewait_sock), .twsk_unique = tcp_twsk_unique, + .twsk_destructor= tcp_twsk_destructor, }; int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) @@ -773,6 +1322,10 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) if (!req) goto drop; +#ifdef CONFIG_TCP_MD5SIG + tcp_rsk(req)->af_specific = &tcp_request_sock_ipv4_ops; +#endif + tcp_clear_options(&tmp_opt); tmp_opt.mss_clamp = 536; tmp_opt.user_mss = tcp_sk(sk)->rx_opt.user_mss; @@ -891,6 +1444,9 @@ struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb, struct inet_sock *newinet; struct tcp_sock *newtp; struct sock *newsk; +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *key; +#endif if (sk_acceptq_is_full(sk)) goto exit_overflow; @@ -925,6 +1481,24 @@ struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb, newtp->advmss = dst_metric(dst, RTAX_ADVMSS); tcp_initialize_rcv_mss(newsk); +#ifdef CONFIG_TCP_MD5SIG + /* Copy over the MD5 key from the original socket */ + if ((key = tcp_v4_md5_do_lookup(sk, newinet->daddr)) != NULL) { + /* + * We're using one, so create a matching key + * on the newsk structure. If we fail to get + * memory, then we end up not copying the key + * across. Shucks. + */ + char *newkey = kmalloc(key->keylen, GFP_ATOMIC); + if (newkey) { + memcpy(newkey, key->key, key->keylen); + tcp_v4_md5_do_add(newsk, inet_sk(sk)->daddr, + newkey, key->keylen); + } + } +#endif + __inet_hash(&tcp_hashinfo, newsk, 0); __inet_inherit_port(&tcp_hashinfo, sk, newsk); @@ -1000,10 +1574,24 @@ static int tcp_v4_checksum_init(struct sk_buff *skb) */ int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb) { + struct sock *rsk; +#ifdef CONFIG_TCP_MD5SIG + /* + * We really want to reject the packet as early as possible + * if: + * o We're expecting an MD5'd packet and this is no MD5 tcp option + * o There is an MD5 option and we're not expecting one + */ + if (tcp_v4_inbound_md5_hash (sk, skb)) + goto discard; +#endif + if (sk->sk_state == TCP_ESTABLISHED) { /* Fast path */ TCP_CHECK_TIMER(sk); - if (tcp_rcv_established(sk, skb, skb->h.th, skb->len)) + if (tcp_rcv_established(sk, skb, skb->h.th, skb->len)) { + rsk = sk; goto reset; + } TCP_CHECK_TIMER(sk); return 0; } @@ -1017,20 +1605,24 @@ int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb) goto discard; if (nsk != sk) { - if (tcp_child_process(sk, nsk, skb)) + if (tcp_child_process(sk, nsk, skb)) { + rsk = nsk; goto reset; + } return 0; } } TCP_CHECK_TIMER(sk); - if (tcp_rcv_state_process(sk, skb, skb->h.th, skb->len)) + if (tcp_rcv_state_process(sk, skb, skb->h.th, skb->len)) { + rsk = sk; goto reset; + } TCP_CHECK_TIMER(sk); return 0; reset: - tcp_v4_send_reset(skb); + tcp_v4_send_reset(rsk, skb); discard: kfree_skb(skb); /* Be careful here. If this function gets more complicated and @@ -1139,7 +1731,7 @@ no_tcp_socket: bad_packet: TCP_INC_STATS_BH(TCP_MIB_INERRS); } else { - tcp_v4_send_reset(skb); + tcp_v4_send_reset(NULL, skb); } discard_it: @@ -1262,6 +1854,15 @@ struct inet_connection_sock_af_ops ipv4_specific = { #endif }; +struct tcp_sock_af_ops tcp_sock_ipv4_specific = { +#ifdef CONFIG_TCP_MD5SIG + .md5_lookup = tcp_v4_md5_lookup, + .calc_md5_hash = tcp_v4_calc_md5_hash, + .md5_add = tcp_v4_md5_add_func, + .md5_parse = tcp_v4_parse_md5_keys, +#endif +}; + /* NOTE: A lot of things set to zero explicitly by call to * sk_alloc() so need not be done here. */ @@ -1301,6 +1902,9 @@ static int tcp_v4_init_sock(struct sock *sk) icsk->icsk_af_ops = &ipv4_specific; icsk->icsk_sync_mss = tcp_sync_mss; +#ifdef CONFIG_TCP_MD5SIG + tp->af_specific = &tcp_sock_ipv4_specific; +#endif sk->sk_sndbuf = sysctl_tcp_wmem[1]; sk->sk_rcvbuf = sysctl_tcp_rmem[1]; @@ -1324,6 +1928,15 @@ int tcp_v4_destroy_sock(struct sock *sk) /* Cleans up our, hopefully empty, out_of_order_queue. */ __skb_queue_purge(&tp->out_of_order_queue); +#ifdef CONFIG_TCP_MD5SIG + /* Clean up the MD5 key list, if any */ + if (tp->md5sig_info) { + tcp_v4_clear_md5_list(sk); + kfree(tp->md5sig_info); + tp->md5sig_info = NULL; + } +#endif + #ifdef CONFIG_NET_DMA /* Cleans up our sk_async_wait_queue */ __skb_queue_purge(&sk->sk_async_wait_queue); diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 0163d982690..ac55d8892cf 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -306,6 +306,28 @@ void tcp_time_wait(struct sock *sk, int state, int timeo) tw->tw_ipv6only = np->ipv6only; } #endif + +#ifdef CONFIG_TCP_MD5SIG + /* + * The timewait bucket does not have the key DB from the + * sock structure. We just make a quick copy of the + * md5 key being used (if indeed we are using one) + * so the timewait ack generating code has the key. + */ + do { + struct tcp_md5sig_key *key; + memset(tcptw->tw_md5_key, 0, sizeof(tcptw->tw_md5_key)); + tcptw->tw_md5_keylen = 0; + key = tp->af_specific->md5_lookup(sk, sk); + if (key != NULL) { + memcpy(&tcptw->tw_md5_key, key->key, key->keylen); + tcptw->tw_md5_keylen = key->keylen; + if (tcp_alloc_md5sig_pool() == NULL) + BUG(); + } + } while(0); +#endif + /* Linkage updates. */ __inet_twsk_hashdance(tw, sk, &tcp_hashinfo); @@ -337,6 +359,17 @@ void tcp_time_wait(struct sock *sk, int state, int timeo) tcp_done(sk); } +void tcp_twsk_destructor(struct sock *sk) +{ + struct tcp_timewait_sock *twsk = tcp_twsk(sk); +#ifdef CONFIG_TCP_MD5SIG + if (twsk->tw_md5_keylen) + tcp_put_md5sig_pool(); +#endif +} + +EXPORT_SYMBOL_GPL(tcp_twsk_destructor); + /* This is not only more efficient than what we used to do, it eliminates * a lot of code duplication between IPv4/IPv6 SYN recv processing. -DaveM * @@ -435,6 +468,11 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct request_sock *req, newtp->rx_opt.ts_recent_stamp = 0; newtp->tcp_header_len = sizeof(struct tcphdr); } +#ifdef CONFIG_TCP_MD5SIG + newtp->md5sig_info = NULL; /*XXX*/ + if (newtp->af_specific->md5_lookup(sk, newsk)) + newtp->tcp_header_len += TCPOLEN_MD5SIG_ALIGNED; +#endif if (skb->len >= TCP_MIN_RCVMSS+newtp->tcp_header_len) newicsk->icsk_ack.last_seg_size = skb->len - newtp->tcp_header_len; newtp->rx_opt.mss_clamp = req->mss; @@ -617,6 +655,30 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, req, NULL); if (child == NULL) goto listen_overflow; +#ifdef CONFIG_TCP_MD5SIG + else { + /* Copy over the MD5 key from the original socket */ + struct tcp_md5sig_key *key; + struct tcp_sock *tp = tcp_sk(sk); + key = tp->af_specific->md5_lookup(sk, child); + if (key != NULL) { + /* + * We're using one, so create a matching key on the + * newsk structure. If we fail to get memory then we + * end up not copying the key across. Shucks. + */ + char *newkey = kmalloc(key->keylen, GFP_ATOMIC); + if (newkey) { + if (!tcp_alloc_md5sig_pool()) + BUG(); + memcpy(newkey, key->key, key->keylen); + tp->af_specific->md5_add(child, child, + newkey, + key->keylen); + } + } + } +#endif inet_csk_reqsk_queue_unlink(sk, req, prev); inet_csk_reqsk_queue_removed(sk, req); @@ -633,7 +695,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, embryonic_reset: NET_INC_STATS_BH(LINUX_MIB_EMBRYONICRSTS); if (!(flg & TCP_FLAG_RST)) - req->rsk_ops->send_reset(skb); + req->rsk_ops->send_reset(sk, skb); inet_csk_reqsk_queue_drop(sk, req, prev); return NULL; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 6a8581ab9a2..32c1a972fa3 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -270,7 +270,7 @@ static u16 tcp_select_window(struct sock *sk) } static void tcp_build_and_update_options(__be32 *ptr, struct tcp_sock *tp, - __u32 tstamp) + __u32 tstamp, __u8 **md5_hash) { if (tp->rx_opt.tstamp_ok) { *ptr++ = htonl((TCPOPT_NOP << 24) | @@ -298,16 +298,29 @@ static void tcp_build_and_update_options(__be32 *ptr, struct tcp_sock *tp, tp->rx_opt.eff_sacks--; } } +#ifdef CONFIG_TCP_MD5SIG + if (md5_hash) { + *ptr++ = htonl((TCPOPT_NOP << 24) | + (TCPOPT_NOP << 16) | + (TCPOPT_MD5SIG << 8) | + TCPOLEN_MD5SIG); + *md5_hash = (__u8 *)ptr; + } +#endif } /* Construct a tcp options header for a SYN or SYN_ACK packet. * If this is every changed make sure to change the definition of * MAX_SYN_SIZE to match the new maximum number of options that you * can generate. + * + * Note - that with the RFC2385 TCP option, we make room for the + * 16 byte MD5 hash. This will be filled in later, so the pointer for the + * location to be filled is passed back up. */ static void tcp_syn_build_options(__be32 *ptr, int mss, int ts, int sack, int offer_wscale, int wscale, __u32 tstamp, - __u32 ts_recent) + __u32 ts_recent, __u8 **md5_hash) { /* We always get an MSS option. * The option bytes which will be seen in normal data @@ -346,6 +359,20 @@ static void tcp_syn_build_options(__be32 *ptr, int mss, int ts, int sack, (TCPOPT_WINDOW << 16) | (TCPOLEN_WINDOW << 8) | (wscale)); +#ifdef CONFIG_TCP_MD5SIG + /* + * If MD5 is enabled, then we set the option, and include the size + * (always 18). The actual MD5 hash is added just before the + * packet is sent. + */ + if (md5_hash) { + *ptr++ = htonl((TCPOPT_NOP << 24) | + (TCPOPT_NOP << 16) | + (TCPOPT_MD5SIG << 8) | + TCPOLEN_MD5SIG); + *md5_hash = (__u8 *) ptr; + } +#endif } /* This routine actually transmits TCP packets queued in by @@ -366,6 +393,10 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, struct tcp_sock *tp; struct tcp_skb_cb *tcb; int tcp_header_size; +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *md5; + __u8 *md5_hash_location; +#endif struct tcphdr *th; int sysctl_flags; int err; @@ -424,6 +455,16 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, if (tcp_packets_in_flight(tp) == 0) tcp_ca_event(sk, CA_EVENT_TX_START); +#ifdef CONFIG_TCP_MD5SIG + /* + * Are we doing MD5 on this segment? If so - make + * room for it. + */ + md5 = tp->af_specific->md5_lookup(sk, sk); + if (md5) + tcp_header_size += TCPOLEN_MD5SIG_ALIGNED; +#endif + th = (struct tcphdr *) skb_push(skb, tcp_header_size); skb->h.th = th; @@ -460,13 +501,34 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, (sysctl_flags & SYSCTL_FLAG_WSCALE), tp->rx_opt.rcv_wscale, tcb->when, - tp->rx_opt.ts_recent); + tp->rx_opt.ts_recent, + +#ifdef CONFIG_TCP_MD5SIG + md5 ? &md5_hash_location : +#endif + NULL); } else { tcp_build_and_update_options((__be32 *)(th + 1), - tp, tcb->when); + tp, tcb->when, +#ifdef CONFIG_TCP_MD5SIG + md5 ? &md5_hash_location : +#endif + NULL); TCP_ECN_send(sk, tp, skb, tcp_header_size); } +#ifdef CONFIG_TCP_MD5SIG + /* Calculate the MD5 hash, as we have all we need now */ + if (md5) { + tp->af_specific->calc_md5_hash(md5_hash_location, + md5, + sk, NULL, NULL, + skb->h.th, + sk->sk_protocol, + skb->len); + } +#endif + icsk->icsk_af_ops->send_check(sk, skb->len, skb); if (likely(tcb->flags & TCPCB_FLAG_ACK)) @@ -840,6 +902,11 @@ unsigned int tcp_current_mss(struct sock *sk, int large_allowed) mss_now -= (TCPOLEN_SACK_BASE_ALIGNED + (tp->rx_opt.eff_sacks * TCPOLEN_SACK_PERBLOCK)); +#ifdef CONFIG_TCP_MD5SIG + if (tp->af_specific->md5_lookup(sk, sk)) + mss_now -= TCPOLEN_MD5SIG_ALIGNED; +#endif + xmit_size_goal = mss_now; if (doing_tso) { @@ -2033,6 +2100,10 @@ struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst, struct tcphdr *th; int tcp_header_size; struct sk_buff *skb; +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *md5; + __u8 *md5_hash_location; +#endif skb = sock_wmalloc(sk, MAX_TCP_HEADER + 15, 1, GFP_ATOMIC); if (skb == NULL) @@ -2048,6 +2119,13 @@ struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst, (ireq->wscale_ok ? TCPOLEN_WSCALE_ALIGNED : 0) + /* SACK_PERM is in the place of NOP NOP of TS */ ((ireq->sack_ok && !ireq->tstamp_ok) ? TCPOLEN_SACKPERM_ALIGNED : 0)); + +#ifdef CONFIG_TCP_MD5SIG + /* Are we doing MD5 on this segment? If so - make room for it */ + md5 = tcp_rsk(req)->af_specific->md5_lookup(sk, req); + if (md5) + tcp_header_size += TCPOLEN_MD5SIG_ALIGNED; +#endif skb->h.th = th = (struct tcphdr *) skb_push(skb, tcp_header_size); memset(th, 0, sizeof(struct tcphdr)); @@ -2085,11 +2163,29 @@ struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst, tcp_syn_build_options((__be32 *)(th + 1), dst_metric(dst, RTAX_ADVMSS), ireq->tstamp_ok, ireq->sack_ok, ireq->wscale_ok, ireq->rcv_wscale, TCP_SKB_CB(skb)->when, - req->ts_recent); + req->ts_recent, + ( +#ifdef CONFIG_TCP_MD5SIG + md5 ? &md5_hash_location : +#endif + NULL) + ); skb->csum = 0; th->doff = (tcp_header_size >> 2); TCP_INC_STATS(TCP_MIB_OUTSEGS); + +#ifdef CONFIG_TCP_MD5SIG + /* Okay, we have all we need - do the md5 hash if needed */ + if (md5) { + tp->af_specific->calc_md5_hash(md5_hash_location, + md5, + NULL, dst, req, + skb->h.th, sk->sk_protocol, + skb->len); + } +#endif + return skb; } @@ -2108,6 +2204,11 @@ static void tcp_connect_init(struct sock *sk) tp->tcp_header_len = sizeof(struct tcphdr) + (sysctl_tcp_timestamps ? TCPOLEN_TSTAMP_ALIGNED : 0); +#ifdef CONFIG_TCP_MD5SIG + if (tp->af_specific->md5_lookup(sk, sk) != NULL) + tp->tcp_header_len += TCPOLEN_MD5SIG_ALIGNED; +#endif + /* If user gave his TCP_MAXSEG, record it to clamp */ if (tp->rx_opt.user_mss) tp->rx_opt.mss_clamp = tp->rx_opt.user_mss; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 9a88395a762..663d1d23801 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -66,10 +66,13 @@ #include #include +#include +#include + /* Socket used for sending RSTs and ACKs */ static struct socket *tcp6_socket; -static void tcp_v6_send_reset(struct sk_buff *skb); +static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb); static void tcp_v6_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req); static void tcp_v6_send_check(struct sock *sk, int len, struct sk_buff *skb); @@ -78,6 +81,8 @@ static int tcp_v6_do_rcv(struct sock *sk, struct sk_buff *skb); static struct inet_connection_sock_af_ops ipv6_mapped; static struct inet_connection_sock_af_ops ipv6_specific; +static struct tcp_sock_af_ops tcp_sock_ipv6_specific; +static struct tcp_sock_af_ops tcp_sock_ipv6_mapped_specific; static int tcp_v6_get_port(struct sock *sk, unsigned short snum) { @@ -208,6 +213,9 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr, icsk->icsk_af_ops = &ipv6_mapped; sk->sk_backlog_rcv = tcp_v4_do_rcv; +#ifdef CONFIG_TCP_MD5SIG + tp->af_specific = &tcp_sock_ipv6_mapped_specific; +#endif err = tcp_v4_connect(sk, (struct sockaddr *)&sin, sizeof(sin)); @@ -215,6 +223,9 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr, icsk->icsk_ext_hdr_len = exthdrlen; icsk->icsk_af_ops = &ipv6_specific; sk->sk_backlog_rcv = tcp_v6_do_rcv; +#ifdef CONFIG_TCP_MD5SIG + tp->af_specific = &tcp_sock_ipv6_specific; +#endif goto failure; } else { ipv6_addr_set(&np->saddr, 0, 0, htonl(0x0000FFFF), @@ -518,6 +529,396 @@ static void tcp_v6_reqsk_destructor(struct request_sock *req) kfree_skb(inet6_rsk(req)->pktopts); } +#ifdef CONFIG_TCP_MD5SIG +static struct tcp_md5sig_key *tcp_v6_md5_do_lookup(struct sock *sk, + struct in6_addr *addr) +{ + struct tcp_sock *tp = tcp_sk(sk); + int i; + + BUG_ON(tp == NULL); + + if (!tp->md5sig_info || !tp->md5sig_info->entries6) + return NULL; + + for (i = 0; i < tp->md5sig_info->entries6; i++) { + if (ipv6_addr_cmp(&tp->md5sig_info->keys6[i].addr, addr) == 0) + return (struct tcp_md5sig_key *)&tp->md5sig_info->keys6[i]; + } + return NULL; +} + +static struct tcp_md5sig_key *tcp_v6_md5_lookup(struct sock *sk, + struct sock *addr_sk) +{ + return tcp_v6_md5_do_lookup(sk, &inet6_sk(addr_sk)->daddr); +} + +static struct tcp_md5sig_key *tcp_v6_reqsk_md5_lookup(struct sock *sk, + struct request_sock *req) +{ + return tcp_v6_md5_do_lookup(sk, &inet6_rsk(req)->rmt_addr); +} + +static int tcp_v6_md5_do_add(struct sock *sk, struct in6_addr *peer, + char *newkey, u8 newkeylen) +{ + /* Add key to the list */ + struct tcp6_md5sig_key *key; + struct tcp_sock *tp = tcp_sk(sk); + struct tcp6_md5sig_key *keys; + + key = (struct tcp6_md5sig_key*) tcp_v6_md5_do_lookup(sk, peer); + if (key) { + /* modify existing entry - just update that one */ + kfree(key->key); + key->key = newkey; + key->keylen = newkeylen; + } else { + /* reallocate new list if current one is full. */ + if (!tp->md5sig_info) { + tp->md5sig_info = kzalloc(sizeof(*tp->md5sig_info), GFP_ATOMIC); + if (!tp->md5sig_info) { + kfree(newkey); + return -ENOMEM; + } + } + tcp_alloc_md5sig_pool(); + if (tp->md5sig_info->alloced6 == tp->md5sig_info->entries6) { + keys = kmalloc((sizeof (tp->md5sig_info->keys6[0]) * + (tp->md5sig_info->entries6 + 1)), GFP_ATOMIC); + + if (!keys) { + tcp_free_md5sig_pool(); + kfree(newkey); + return -ENOMEM; + } + + if (tp->md5sig_info->entries6) + memmove(keys, tp->md5sig_info->keys6, + (sizeof (tp->md5sig_info->keys6[0]) * + tp->md5sig_info->entries6)); + + kfree(tp->md5sig_info->keys6); + tp->md5sig_info->keys6 = keys; + tp->md5sig_info->alloced6++; + } + + ipv6_addr_copy(&tp->md5sig_info->keys6[tp->md5sig_info->entries6].addr, + peer); + tp->md5sig_info->keys6[tp->md5sig_info->entries6].key = newkey; + tp->md5sig_info->keys6[tp->md5sig_info->entries6].keylen = newkeylen; + + tp->md5sig_info->entries6++; + } + return 0; +} + +static int tcp_v6_md5_add_func(struct sock *sk, struct sock *addr_sk, + u8 *newkey, __u8 newkeylen) +{ + return tcp_v6_md5_do_add(sk, &inet6_sk(addr_sk)->daddr, + newkey, newkeylen); +} + +static int tcp_v6_md5_do_del(struct sock *sk, struct in6_addr *peer) +{ + struct tcp_sock *tp = tcp_sk(sk); + int i; + + for (i = 0; i < tp->md5sig_info->entries6; i++) { + if (ipv6_addr_cmp(&tp->md5sig_info->keys6[i].addr, peer) == 0) { + /* Free the key */ + kfree(tp->md5sig_info->keys6[i].key); + tp->md5sig_info->entries6--; + + if (tp->md5sig_info->entries6 == 0) { + kfree(tp->md5sig_info->keys6); + tp->md5sig_info->keys6 = NULL; + + tcp_free_md5sig_pool(); + + return 0; + } else { + /* shrink the database */ + if (tp->md5sig_info->entries6 != i) + memmove(&tp->md5sig_info->keys6[i], + &tp->md5sig_info->keys6[i+1], + (tp->md5sig_info->entries6 - i) + * sizeof (tp->md5sig_info->keys6[0])); + } + } + } + return -ENOENT; +} + +static void tcp_v6_clear_md5_list (struct sock *sk) +{ + struct tcp_sock *tp = tcp_sk(sk); + int i; + + if (tp->md5sig_info->entries6) { + for (i = 0; i < tp->md5sig_info->entries6; i++) + kfree(tp->md5sig_info->keys6[i].key); + tp->md5sig_info->entries6 = 0; + tcp_free_md5sig_pool(); + } + + kfree(tp->md5sig_info->keys6); + tp->md5sig_info->keys6 = NULL; + tp->md5sig_info->alloced6 = 0; + + if (tp->md5sig_info->entries4) { + for (i = 0; i < tp->md5sig_info->entries4; i++) + kfree(tp->md5sig_info->keys4[i].key); + tp->md5sig_info->entries4 = 0; + tcp_free_md5sig_pool(); + } + + kfree(tp->md5sig_info->keys4); + tp->md5sig_info->keys4 = NULL; + tp->md5sig_info->alloced4 = 0; +} + +static int tcp_v6_parse_md5_keys (struct sock *sk, char __user *optval, + int optlen) +{ + struct tcp_md5sig cmd; + struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)&cmd.tcpm_addr; + u8 *newkey; + + if (optlen < sizeof(cmd)) + return -EINVAL; + + if (copy_from_user(&cmd, optval, sizeof(cmd))) + return -EFAULT; + + if (sin6->sin6_family != AF_INET6) + return -EINVAL; + + if (!cmd.tcpm_keylen) { + if (!tcp_sk(sk)->md5sig_info) + return -ENOENT; + if (ipv6_addr_type(&sin6->sin6_addr) & IPV6_ADDR_MAPPED) + return tcp_v4_md5_do_del(sk, sin6->sin6_addr.s6_addr32[3]); + return tcp_v6_md5_do_del(sk, &sin6->sin6_addr); + } + + if (cmd.tcpm_keylen > TCP_MD5SIG_MAXKEYLEN) + return -EINVAL; + + if (!tcp_sk(sk)->md5sig_info) { + struct tcp_sock *tp = tcp_sk(sk); + struct tcp_md5sig_info *p; + + p = kzalloc(sizeof(struct tcp_md5sig_info), GFP_KERNEL); + if (!p) + return -ENOMEM; + + tp->md5sig_info = p; + } + + newkey = kmalloc(cmd.tcpm_keylen, GFP_KERNEL); + if (!newkey) + return -ENOMEM; + memcpy(newkey, cmd.tcpm_key, cmd.tcpm_keylen); + if (ipv6_addr_type(&sin6->sin6_addr) & IPV6_ADDR_MAPPED) { + return tcp_v4_md5_do_add(sk, sin6->sin6_addr.s6_addr32[3], + newkey, cmd.tcpm_keylen); + } + return tcp_v6_md5_do_add(sk, &sin6->sin6_addr, newkey, cmd.tcpm_keylen); +} + +static int tcp_v6_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key, + struct in6_addr *saddr, + struct in6_addr *daddr, + struct tcphdr *th, int protocol, + int tcplen) +{ + struct scatterlist sg[4]; + __u16 data_len; + int block = 0; + __u16 cksum; + struct tcp_md5sig_pool *hp; + struct tcp6_pseudohdr *bp; + struct hash_desc *desc; + int err; + unsigned int nbytes = 0; + + hp = tcp_get_md5sig_pool(); + if (!hp) { + printk(KERN_WARNING "%s(): hash pool not found...\n", __FUNCTION__); + goto clear_hash_noput; + } + bp = &hp->md5_blk.ip6; + desc = &hp->md5_desc; + + /* 1. TCP pseudo-header (RFC2460) */ + ipv6_addr_copy(&bp->saddr, saddr); + ipv6_addr_copy(&bp->daddr, daddr); + bp->len = htonl(tcplen); + bp->protocol = htonl(protocol); + + sg_set_buf(&sg[block++], bp, sizeof(*bp)); + nbytes += sizeof(*bp); + + /* 2. TCP header, excluding options */ + cksum = th->check; + th->check = 0; + sg_set_buf(&sg[block++], th, sizeof(*th)); + nbytes += sizeof(*th); + + /* 3. TCP segment data (if any) */ + data_len = tcplen - (th->doff << 2); + if (data_len > 0) { + u8 *data = (u8 *)th + (th->doff << 2); + sg_set_buf(&sg[block++], data, data_len); + nbytes += data_len; + } + + /* 4. shared key */ + sg_set_buf(&sg[block++], key->key, key->keylen); + nbytes += key->keylen; + + /* Now store the hash into the packet */ + err = crypto_hash_init(desc); + if (err) { + printk(KERN_WARNING "%s(): hash_init failed\n", __FUNCTION__); + goto clear_hash; + } + err = crypto_hash_update(desc, sg, nbytes); + if (err) { + printk(KERN_WARNING "%s(): hash_update failed\n", __FUNCTION__); + goto clear_hash; + } + err = crypto_hash_final(desc, md5_hash); + if (err) { + printk(KERN_WARNING "%s(): hash_final failed\n", __FUNCTION__); + goto clear_hash; + } + + /* Reset header, and free up the crypto */ + tcp_put_md5sig_pool(); + th->check = cksum; +out: + return 0; +clear_hash: + tcp_put_md5sig_pool(); +clear_hash_noput: + memset(md5_hash, 0, 16); + goto out; +} + +static int tcp_v6_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key, + struct sock *sk, + struct dst_entry *dst, + struct request_sock *req, + struct tcphdr *th, int protocol, + int tcplen) +{ + struct in6_addr *saddr, *daddr; + + if (sk) { + saddr = &inet6_sk(sk)->saddr; + daddr = &inet6_sk(sk)->daddr; + } else { + saddr = &inet6_rsk(req)->loc_addr; + daddr = &inet6_rsk(req)->rmt_addr; + } + return tcp_v6_do_calc_md5_hash(md5_hash, key, + saddr, daddr, + th, protocol, tcplen); +} + +static int tcp_v6_inbound_md5_hash (struct sock *sk, struct sk_buff *skb) +{ + __u8 *hash_location = NULL; + struct tcp_md5sig_key *hash_expected; + struct ipv6hdr *ip6h = skb->nh.ipv6h; + struct tcphdr *th = skb->h.th; + int length = (th->doff << 2) - sizeof (*th); + int genhash; + u8 *ptr; + u8 newhash[16]; + + hash_expected = tcp_v6_md5_do_lookup(sk, &ip6h->saddr); + + /* If the TCP option is too short, we can short cut */ + if (length < TCPOLEN_MD5SIG) + return hash_expected ? 1 : 0; + + /* parse options */ + ptr = (u8*)(th + 1); + while (length > 0) { + int opcode = *ptr++; + int opsize; + + switch(opcode) { + case TCPOPT_EOL: + goto done_opts; + case TCPOPT_NOP: + length--; + continue; + default: + opsize = *ptr++; + if (opsize < 2 || opsize > length) + goto done_opts; + if (opcode == TCPOPT_MD5SIG) { + hash_location = ptr; + goto done_opts; + } + } + ptr += opsize - 2; + length -= opsize; + } + +done_opts: + /* do we have a hash as expected? */ + if (!hash_expected) { + if (!hash_location) + return 0; + if (net_ratelimit()) { + printk(KERN_INFO "MD5 Hash NOT expected but found " + "(" NIP6_FMT ", %u)->" + "(" NIP6_FMT ", %u)\n", + NIP6(ip6h->saddr), ntohs(th->source), + NIP6(ip6h->daddr), ntohs(th->dest)); + } + return 1; + } + + if (!hash_location) { + if (net_ratelimit()) { + printk(KERN_INFO "MD5 Hash expected but NOT found " + "(" NIP6_FMT ", %u)->" + "(" NIP6_FMT ", %u)\n", + NIP6(ip6h->saddr), ntohs(th->source), + NIP6(ip6h->daddr), ntohs(th->dest)); + } + return 1; + } + + /* check the signature */ + genhash = tcp_v6_do_calc_md5_hash(newhash, + hash_expected, + &ip6h->saddr, &ip6h->daddr, + th, sk->sk_protocol, + skb->len); + if (genhash || memcmp(hash_location, newhash, 16) != 0) { + if (net_ratelimit()) { + printk(KERN_INFO "MD5 Hash %s for " + "(" NIP6_FMT ", %u)->" + "(" NIP6_FMT ", %u)\n", + genhash ? "failed" : "mismatch", + NIP6(ip6h->saddr), ntohs(th->source), + NIP6(ip6h->daddr), ntohs(th->dest)); + } + return 1; + } + return 0; +} +#endif + static struct request_sock_ops tcp6_request_sock_ops __read_mostly = { .family = AF_INET6, .obj_size = sizeof(struct tcp6_request_sock), @@ -527,9 +928,16 @@ static struct request_sock_ops tcp6_request_sock_ops __read_mostly = { .send_reset = tcp_v6_send_reset }; +struct tcp_request_sock_ops tcp_request_sock_ipv6_ops = { +#ifdef CONFIG_TCP_MD5SIG + .md5_lookup = tcp_v6_reqsk_md5_lookup, +#endif +}; + static struct timewait_sock_ops tcp6_timewait_sock_ops = { .twsk_obj_size = sizeof(struct tcp6_timewait_sock), .twsk_unique = tcp_twsk_unique, + .twsk_destructor= tcp_twsk_destructor, }; static void tcp_v6_send_check(struct sock *sk, int len, struct sk_buff *skb) @@ -566,11 +974,15 @@ static int tcp_v6_gso_send_check(struct sk_buff *skb) return 0; } -static void tcp_v6_send_reset(struct sk_buff *skb) +static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb) { struct tcphdr *th = skb->h.th, *t1; struct sk_buff *buff; struct flowi fl; + int tot_len = sizeof(*th); +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *key; +#endif if (th->rst) return; @@ -578,25 +990,35 @@ static void tcp_v6_send_reset(struct sk_buff *skb) if (!ipv6_unicast_destination(skb)) return; +#ifdef CONFIG_TCP_MD5SIG + if (sk) + key = tcp_v6_md5_do_lookup(sk, &skb->nh.ipv6h->daddr); + else + key = NULL; + + if (key) + tot_len += TCPOLEN_MD5SIG_ALIGNED; +#endif + /* * We need to grab some memory, and put together an RST, * and then put it into the queue to be sent. */ - buff = alloc_skb(MAX_HEADER + sizeof(struct ipv6hdr) + sizeof(struct tcphdr), + buff = alloc_skb(MAX_HEADER + sizeof(struct ipv6hdr) + tot_len, GFP_ATOMIC); if (buff == NULL) return; - skb_reserve(buff, MAX_HEADER + sizeof(struct ipv6hdr) + sizeof(struct tcphdr)); + skb_reserve(buff, MAX_HEADER + sizeof(struct ipv6hdr) + tot_len); - t1 = (struct tcphdr *) skb_push(buff,sizeof(struct tcphdr)); + t1 = (struct tcphdr *) skb_push(buff, tot_len); /* Swap the send and the receive. */ memset(t1, 0, sizeof(*t1)); t1->dest = th->source; t1->source = th->dest; - t1->doff = sizeof(*t1)/4; + t1->doff = tot_len / 4; t1->rst = 1; if(th->ack) { @@ -607,6 +1029,22 @@ static void tcp_v6_send_reset(struct sk_buff *skb) + skb->len - (th->doff<<2)); } +#ifdef CONFIG_TCP_MD5SIG + if (key) { + u32 *opt = (u32*)(t1 + 1); + opt[0] = htonl((TCPOPT_NOP << 24) | + (TCPOPT_NOP << 16) | + (TCPOPT_MD5SIG << 8) | + TCPOLEN_MD5SIG); + tcp_v6_do_calc_md5_hash((__u8*)&opt[1], + key, + &skb->nh.ipv6h->daddr, + &skb->nh.ipv6h->saddr, + t1, IPPROTO_TCP, + tot_len); + } +#endif + buff->csum = csum_partial((char *)t1, sizeof(*t1), 0); memset(&fl, 0, sizeof(fl)); @@ -637,15 +1075,37 @@ static void tcp_v6_send_reset(struct sk_buff *skb) kfree_skb(buff); } -static void tcp_v6_send_ack(struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32 ts) +static void tcp_v6_send_ack(struct tcp_timewait_sock *tw, + struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32 ts) { struct tcphdr *th = skb->h.th, *t1; struct sk_buff *buff; struct flowi fl; int tot_len = sizeof(struct tcphdr); + u32 *topt; +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *key; + struct tcp_md5sig_key tw_key; +#endif + +#ifdef CONFIG_TCP_MD5SIG + if (!tw && skb->sk) { + key = tcp_v6_md5_do_lookup(skb->sk, &skb->nh.ipv6h->daddr); + } else if (tw && tw->tw_md5_keylen) { + tw_key.key = tw->tw_md5_key; + tw_key.keylen = tw->tw_md5_keylen; + key = &tw_key; + } else { + key = NULL; + } +#endif if (ts) tot_len += TCPOLEN_TSTAMP_ALIGNED; +#ifdef CONFIG_TCP_MD5SIG + if (key) + tot_len += TCPOLEN_MD5SIG_ALIGNED; +#endif buff = alloc_skb(MAX_HEADER + sizeof(struct ipv6hdr) + tot_len, GFP_ATOMIC); @@ -665,15 +1125,29 @@ static void tcp_v6_send_ack(struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32 t1->ack_seq = htonl(ack); t1->ack = 1; t1->window = htons(win); + + topt = (u32*)(t1 + 1); if (ts) { - u32 *ptr = (u32*)(t1 + 1); - *ptr++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | - (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP); - *ptr++ = htonl(tcp_time_stamp); - *ptr = htonl(ts); + *topt++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | + (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP); + *topt++ = htonl(tcp_time_stamp); + *topt = htonl(ts); } +#ifdef CONFIG_TCP_MD5SIG + if (key) { + *topt++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | + (TCPOPT_MD5SIG << 8) | TCPOLEN_MD5SIG); + tcp_v6_do_calc_md5_hash((__u8 *)topt, + key, + &skb->nh.ipv6h->daddr, + &skb->nh.ipv6h->saddr, + t1, IPPROTO_TCP, + tot_len); + } +#endif + buff->csum = csum_partial((char *)t1, tot_len, 0); memset(&fl, 0, sizeof(fl)); @@ -704,9 +1178,9 @@ static void tcp_v6_send_ack(struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32 static void tcp_v6_timewait_ack(struct sock *sk, struct sk_buff *skb) { struct inet_timewait_sock *tw = inet_twsk(sk); - const struct tcp_timewait_sock *tcptw = tcp_twsk(sk); + struct tcp_timewait_sock *tcptw = tcp_twsk(sk); - tcp_v6_send_ack(skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt, + tcp_v6_send_ack(tcptw, skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt, tcptw->tw_rcv_wnd >> tw->tw_rcv_wscale, tcptw->tw_ts_recent); @@ -715,7 +1189,7 @@ static void tcp_v6_timewait_ack(struct sock *sk, struct sk_buff *skb) static void tcp_v6_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req) { - tcp_v6_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent); + tcp_v6_send_ack(NULL, skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent); } @@ -786,6 +1260,10 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) if (req == NULL) goto drop; +#ifdef CONFIG_TCP_MD5SIG + tcp_rsk(req)->af_specific = &tcp_request_sock_ipv6_ops; +#endif + tcp_clear_options(&tmp_opt); tmp_opt.mss_clamp = IPV6_MIN_MTU - sizeof(struct tcphdr) - sizeof(struct ipv6hdr); tmp_opt.user_mss = tp->rx_opt.user_mss; @@ -844,6 +1322,9 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, struct tcp_sock *newtp; struct sock *newsk; struct ipv6_txoptions *opt; +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *key; +#endif if (skb->protocol == htons(ETH_P_IP)) { /* @@ -874,6 +1355,10 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, inet_csk(newsk)->icsk_af_ops = &ipv6_mapped; newsk->sk_backlog_rcv = tcp_v4_do_rcv; +#ifdef CONFIG_TCP_MD5SIG + newtp->af_specific = &tcp_sock_ipv6_mapped_specific; +#endif + newnp->pktoptions = NULL; newnp->opt = NULL; newnp->mcast_oif = inet6_iif(skb); @@ -1008,6 +1493,23 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, newinet->daddr = newinet->saddr = newinet->rcv_saddr = LOOPBACK4_IPV6; +#ifdef CONFIG_TCP_MD5SIG + /* Copy over the MD5 key from the original socket */ + if ((key = tcp_v6_md5_do_lookup(sk, &newnp->daddr)) != NULL) { + /* We're using one, so create a matching key + * on the newsk structure. If we fail to get + * memory, then we end up not copying the key + * across. Shucks. + */ + char *newkey = kmalloc(key->keylen, GFP_ATOMIC); + if (newkey) { + memcpy(newkey, key->key, key->keylen); + tcp_v6_md5_do_add(newsk, &inet6_sk(sk)->daddr, + newkey, key->keylen); + } + } +#endif + __inet6_hash(&tcp_hashinfo, newsk); inet_inherit_port(&tcp_hashinfo, sk, newsk); @@ -1067,6 +1569,11 @@ static int tcp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) if (skb->protocol == htons(ETH_P_IP)) return tcp_v4_do_rcv(sk, skb); +#ifdef CONFIG_TCP_MD5SIG + if (tcp_v6_inbound_md5_hash (sk, skb)) + goto discard; +#endif + if (sk_filter(sk, skb)) goto discard; @@ -1132,7 +1639,7 @@ static int tcp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) return 0; reset: - tcp_v6_send_reset(skb); + tcp_v6_send_reset(sk, skb); discard: if (opt_skb) __kfree_skb(opt_skb); @@ -1257,7 +1764,7 @@ no_tcp_socket: bad_packet: TCP_INC_STATS_BH(TCP_MIB_INERRS); } else { - tcp_v6_send_reset(skb); + tcp_v6_send_reset(NULL, skb); } discard_it: @@ -1336,6 +1843,15 @@ static struct inet_connection_sock_af_ops ipv6_specific = { #endif }; +static struct tcp_sock_af_ops tcp_sock_ipv6_specific = { +#ifdef CONFIG_TCP_MD5SIG + .md5_lookup = tcp_v6_md5_lookup, + .calc_md5_hash = tcp_v6_calc_md5_hash, + .md5_add = tcp_v6_md5_add_func, + .md5_parse = tcp_v6_parse_md5_keys, +#endif +}; + /* * TCP over IPv4 via INET6 API */ @@ -1358,6 +1874,15 @@ static struct inet_connection_sock_af_ops ipv6_mapped = { #endif }; +static struct tcp_sock_af_ops tcp_sock_ipv6_mapped_specific = { +#ifdef CONFIG_TCP_MD5SIG + .md5_lookup = tcp_v4_md5_lookup, + .calc_md5_hash = tcp_v4_calc_md5_hash, + .md5_add = tcp_v6_md5_add_func, + .md5_parse = tcp_v6_parse_md5_keys, +#endif +}; + /* NOTE: A lot of things set to zero explicitly by call to * sk_alloc() so need not be done here. */ @@ -1397,6 +1922,10 @@ static int tcp_v6_init_sock(struct sock *sk) sk->sk_write_space = sk_stream_write_space; sock_set_flag(sk, SOCK_USE_WRITE_QUEUE); +#ifdef CONFIG_TCP_MD5SIG + tp->af_specific = &tcp_sock_ipv6_specific; +#endif + sk->sk_sndbuf = sysctl_tcp_wmem[1]; sk->sk_rcvbuf = sysctl_tcp_rmem[1]; @@ -1407,6 +1936,11 @@ static int tcp_v6_init_sock(struct sock *sk) static int tcp_v6_destroy_sock(struct sock *sk) { +#ifdef CONFIG_TCP_MD5SIG + /* Clean up the MD5 key list */ + if (tcp_sk(sk)->md5sig_info) + tcp_v6_clear_md5_list(sk); +#endif tcp_v4_destroy_sock(sk); return inet6_destroy_sock(sk); } -- cgit v1.2.3-70-g09d2 From 3dabc7157859e706770c825aa229f8943db4e0e1 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Tue, 14 Nov 2006 19:44:52 -0800 Subject: [GENL]: Add genlmsg_new() to allocate generic netlink messages Signed-off-by: Thomas Graf Acked-by: Paul Moore Signed-off-by: David S. Miller --- include/net/genetlink.h | 11 +++++++++++ kernel/taskstats.c | 2 +- 2 files changed, 12 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/genetlink.h b/include/net/genetlink.h index b619314218a..2010465fa7d 100644 --- a/include/net/genetlink.h +++ b/include/net/genetlink.h @@ -187,4 +187,15 @@ static inline int genlmsg_total_size(int payload) return NLMSG_ALIGN(genlmsg_msg_size(payload)); } +/** + * genlmsg_new - Allocate a new generic netlink message + * @payload: size of the message payload + * @flags: the type of memory to allocate. + */ +static inline struct sk_buff *genlmsg_new(size_t payload, gfp_t flags) +{ + return nlmsg_new(genlmsg_total_size(payload), flags); +} + + #endif /* __NET_GENERIC_NETLINK_H */ diff --git a/kernel/taskstats.c b/kernel/taskstats.c index 4f3f0e48c84..faa5239813c 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -77,7 +77,7 @@ static int prepare_reply(struct genl_info *info, u8 cmd, struct sk_buff **skbp, /* * If new attributes are added, please revisit this allocation */ - skb = nlmsg_new(genlmsg_total_size(size), GFP_KERNEL); + skb = genlmsg_new(size, GFP_KERNEL); if (!skb) return -ENOMEM; -- cgit v1.2.3-70-g09d2 From 81878d27fdd297a33f3cfcf29483fe1abaf26dec Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Tue, 14 Nov 2006 19:45:27 -0800 Subject: [GENL]: Add genlmsg_reply() to simply unicast replies to requests A generic netlink user has no interest in knowing how to address the source of the original request. Signed-off-by: Thomas Graf Acked-by: Paul Moore Signed-off-by: David S. Miller --- include/net/genetlink.h | 10 ++++++++++ net/netlabel/netlabel_cipso_v4.c | 2 +- net/netlabel/netlabel_mgmt.c | 4 ++-- net/netlabel/netlabel_unlabeled.c | 2 +- net/netlink/genetlink.c | 2 +- 5 files changed, 15 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/net/genetlink.h b/include/net/genetlink.h index 2010465fa7d..797c18b5041 100644 --- a/include/net/genetlink.h +++ b/include/net/genetlink.h @@ -149,6 +149,16 @@ static inline int genlmsg_unicast(struct sk_buff *skb, u32 pid) return nlmsg_unicast(genl_sock, skb, pid); } +/** + * genlmsg_reply - reply to a request + * @skb: netlink message to be sent back + * @info: receiver information + */ +static inline int genlmsg_reply(struct sk_buff *skb, struct genl_info *info) +{ + return genlmsg_unicast(skb, info->snd_pid); +} + /** * gennlmsg_data - head of message payload * @gnlh: genetlink messsage header diff --git a/net/netlabel/netlabel_cipso_v4.c b/net/netlabel/netlabel_cipso_v4.c index f1788bd290f..52628878524 100644 --- a/net/netlabel/netlabel_cipso_v4.c +++ b/net/netlabel/netlabel_cipso_v4.c @@ -568,7 +568,7 @@ list_start: genlmsg_end(ans_skb, data); - ret_val = genlmsg_unicast(ans_skb, info->snd_pid); + ret_val = genlmsg_reply(ans_skb, info); if (ret_val != 0) goto list_failure; diff --git a/net/netlabel/netlabel_mgmt.c b/net/netlabel/netlabel_mgmt.c index c529622ff0b..784693735e0 100644 --- a/net/netlabel/netlabel_mgmt.c +++ b/net/netlabel/netlabel_mgmt.c @@ -390,7 +390,7 @@ static int netlbl_mgmt_listdef(struct sk_buff *skb, struct genl_info *info) genlmsg_end(ans_skb, data); - ret_val = genlmsg_unicast(ans_skb, info->snd_pid); + ret_val = genlmsg_reply(ans_skb, info); if (ret_val != 0) goto listdef_failure; return 0; @@ -512,7 +512,7 @@ static int netlbl_mgmt_version(struct sk_buff *skb, struct genl_info *info) genlmsg_end(ans_skb, data); - ret_val = genlmsg_unicast(ans_skb, info->snd_pid); + ret_val = genlmsg_reply(ans_skb, info); if (ret_val != 0) goto version_failure; return 0; diff --git a/net/netlabel/netlabel_unlabeled.c b/net/netlabel/netlabel_unlabeled.c index 219dccade4e..57dd07b51be 100644 --- a/net/netlabel/netlabel_unlabeled.c +++ b/net/netlabel/netlabel_unlabeled.c @@ -160,7 +160,7 @@ static int netlbl_unlabel_list(struct sk_buff *skb, struct genl_info *info) genlmsg_end(ans_skb, data); - ret_val = genlmsg_unicast(ans_skb, info->snd_pid); + ret_val = genlmsg_reply(ans_skb, info); if (ret_val != 0) goto list_failure; return 0; diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c index 70d60c81889..0249a56a9aa 100644 --- a/net/netlink/genetlink.c +++ b/net/netlink/genetlink.c @@ -529,7 +529,7 @@ static int ctrl_getfamily(struct sk_buff *skb, struct genl_info *info) goto errout; } - err = genlmsg_unicast(msg, info->snd_pid); + err = genlmsg_reply(msg, info); errout: return err; } -- cgit v1.2.3-70-g09d2 From 17c157c889f4b07258af6bfec9e4e9dcf3c00178 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Tue, 14 Nov 2006 19:46:02 -0800 Subject: [GENL]: Add genlmsg_put_reply() to simplify building reply headers By modyfing genlmsg_put() to take a genl_family and by adding genlmsg_put_reply() the process of constructing the netlink and generic netlink headers is simplified. Signed-off-by: Thomas Graf Acked-by: Paul Moore Signed-off-by: David S. Miller --- include/net/genetlink.h | 31 ++++++++++++++++++++++++------- kernel/taskstats.c | 8 ++------ net/netlabel/netlabel_cipso_v4.c | 17 +++++------------ net/netlabel/netlabel_mgmt.c | 34 ++++++++++------------------------ net/netlabel/netlabel_unlabeled.c | 8 ++------ net/netlabel/netlabel_user.h | 31 ------------------------------- net/netlink/genetlink.c | 17 ++++++++--------- 7 files changed, 51 insertions(+), 95 deletions(-) (limited to 'include') diff --git a/include/net/genetlink.h b/include/net/genetlink.h index 797c18b5041..7fd131c9a8c 100644 --- a/include/net/genetlink.h +++ b/include/net/genetlink.h @@ -79,33 +79,50 @@ extern struct sock *genl_sock; * @skb: socket buffer holding the message * @pid: netlink pid the message is addressed to * @seq: sequence number (usually the one of the sender) - * @type: netlink message type - * @hdrlen: length of the user specific header + * @family: generic netlink family * @flags netlink message flags * @cmd: generic netlink command - * @version: version * * Returns pointer to user specific header */ static inline void *genlmsg_put(struct sk_buff *skb, u32 pid, u32 seq, - int type, int hdrlen, int flags, - u8 cmd, u8 version) + struct genl_family *family, int flags, u8 cmd) { struct nlmsghdr *nlh; struct genlmsghdr *hdr; - nlh = nlmsg_put(skb, pid, seq, type, GENL_HDRLEN + hdrlen, flags); + nlh = nlmsg_put(skb, pid, seq, family->id, GENL_HDRLEN + + family->hdrsize, flags); if (nlh == NULL) return NULL; hdr = nlmsg_data(nlh); hdr->cmd = cmd; - hdr->version = version; + hdr->version = family->version; hdr->reserved = 0; return (char *) hdr + GENL_HDRLEN; } +/** + * genlmsg_put_reply - Add generic netlink header to a reply message + * @skb: socket buffer holding the message + * @info: receiver info + * @family: generic netlink family + * @flags: netlink message flags + * @cmd: generic netlink command + * + * Returns pointer to user specific header + */ +static inline void *genlmsg_put_reply(struct sk_buff *skb, + struct genl_info *info, + struct genl_family *family, + int flags, u8 cmd) +{ + return genlmsg_put(skb, info->snd_pid, info->snd_seq, family, + flags, cmd); +} + /** * genlmsg_end - Finalize a generic netlink message * @skb: socket buffer the message is stored in diff --git a/kernel/taskstats.c b/kernel/taskstats.c index faa5239813c..d3d28919d4b 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -85,13 +85,9 @@ static int prepare_reply(struct genl_info *info, u8 cmd, struct sk_buff **skbp, int seq = get_cpu_var(taskstats_seqnum)++; put_cpu_var(taskstats_seqnum); - reply = genlmsg_put(skb, 0, seq, - family.id, 0, 0, - cmd, family.version); + reply = genlmsg_put(skb, 0, seq, &family, 0, cmd); } else - reply = genlmsg_put(skb, info->snd_pid, info->snd_seq, - family.id, 0, 0, - cmd, family.version); + reply = genlmsg_put_reply(skb, info, &family, 0, cmd); if (reply == NULL) { nlmsg_free(skb); return -EINVAL; diff --git a/net/netlabel/netlabel_cipso_v4.c b/net/netlabel/netlabel_cipso_v4.c index 52628878524..fe9851fac85 100644 --- a/net/netlabel/netlabel_cipso_v4.c +++ b/net/netlabel/netlabel_cipso_v4.c @@ -457,12 +457,8 @@ list_start: ret_val = -ENOMEM; goto list_failure; } - data = netlbl_netlink_hdr_put(ans_skb, - info->snd_pid, - info->snd_seq, - netlbl_cipsov4_gnl_family.id, - 0, - NLBL_CIPSOV4_C_LIST); + data = genlmsg_put_reply(ans_skb, info, &netlbl_cipsov4_gnl_family, + 0, NLBL_CIPSOV4_C_LIST); if (data == NULL) { ret_val = -ENOMEM; goto list_failure; @@ -607,12 +603,9 @@ static int netlbl_cipsov4_listall_cb(struct cipso_v4_doi *doi_def, void *arg) struct netlbl_cipsov4_doiwalk_arg *cb_arg = arg; void *data; - data = netlbl_netlink_hdr_put(cb_arg->skb, - NETLINK_CB(cb_arg->nl_cb->skb).pid, - cb_arg->seq, - netlbl_cipsov4_gnl_family.id, - NLM_F_MULTI, - NLBL_CIPSOV4_C_LISTALL); + data = genlmsg_put(cb_arg->skb, NETLINK_CB(cb_arg->nl_cb->skb).pid, + cb_arg->seq, &netlbl_cipsov4_gnl_family, + NLM_F_MULTI, NLBL_CIPSOV4_C_LISTALL); if (data == NULL) goto listall_cb_failure; diff --git a/net/netlabel/netlabel_mgmt.c b/net/netlabel/netlabel_mgmt.c index 784693735e0..e8c80f33f3d 100644 --- a/net/netlabel/netlabel_mgmt.c +++ b/net/netlabel/netlabel_mgmt.c @@ -188,12 +188,9 @@ static int netlbl_mgmt_listall_cb(struct netlbl_dom_map *entry, void *arg) struct netlbl_domhsh_walk_arg *cb_arg = arg; void *data; - data = netlbl_netlink_hdr_put(cb_arg->skb, - NETLINK_CB(cb_arg->nl_cb->skb).pid, - cb_arg->seq, - netlbl_mgmt_gnl_family.id, - NLM_F_MULTI, - NLBL_MGMT_C_LISTALL); + data = genlmsg_put(cb_arg->skb, NETLINK_CB(cb_arg->nl_cb->skb).pid, + cb_arg->seq, &netlbl_mgmt_gnl_family, + NLM_F_MULTI, NLBL_MGMT_C_LISTALL); if (data == NULL) goto listall_cb_failure; @@ -359,12 +356,8 @@ static int netlbl_mgmt_listdef(struct sk_buff *skb, struct genl_info *info) ans_skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (ans_skb == NULL) return -ENOMEM; - data = netlbl_netlink_hdr_put(ans_skb, - info->snd_pid, - info->snd_seq, - netlbl_mgmt_gnl_family.id, - 0, - NLBL_MGMT_C_LISTDEF); + data = genlmsg_put_reply(ans_skb, info, &netlbl_mgmt_gnl_family, + 0, NLBL_MGMT_C_LISTDEF); if (data == NULL) goto listdef_failure; @@ -422,12 +415,9 @@ static int netlbl_mgmt_protocols_cb(struct sk_buff *skb, int ret_val = -ENOMEM; void *data; - data = netlbl_netlink_hdr_put(skb, - NETLINK_CB(cb->skb).pid, - cb->nlh->nlmsg_seq, - netlbl_mgmt_gnl_family.id, - NLM_F_MULTI, - NLBL_MGMT_C_PROTOCOLS); + data = genlmsg_put(skb, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, + &netlbl_mgmt_gnl_family, NLM_F_MULTI, + NLBL_MGMT_C_PROTOCOLS); if (data == NULL) goto protocols_cb_failure; @@ -495,12 +485,8 @@ static int netlbl_mgmt_version(struct sk_buff *skb, struct genl_info *info) ans_skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (ans_skb == NULL) return -ENOMEM; - data = netlbl_netlink_hdr_put(ans_skb, - info->snd_pid, - info->snd_seq, - netlbl_mgmt_gnl_family.id, - 0, - NLBL_MGMT_C_VERSION); + data = genlmsg_put_reply(ans_skb, info, &netlbl_mgmt_gnl_family, + 0, NLBL_MGMT_C_VERSION); if (data == NULL) goto version_failure; diff --git a/net/netlabel/netlabel_unlabeled.c b/net/netlabel/netlabel_unlabeled.c index 57dd07b51be..a1d4ae51db0 100644 --- a/net/netlabel/netlabel_unlabeled.c +++ b/net/netlabel/netlabel_unlabeled.c @@ -141,12 +141,8 @@ static int netlbl_unlabel_list(struct sk_buff *skb, struct genl_info *info) ans_skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (ans_skb == NULL) goto list_failure; - data = netlbl_netlink_hdr_put(ans_skb, - info->snd_pid, - info->snd_seq, - netlbl_unlabel_gnl_family.id, - 0, - NLBL_UNLABEL_C_LIST); + data = genlmsg_put_reply(ans_skb, info, &netlbl_unlabel_gnl_family, + 0, NLBL_UNLABEL_C_LIST); if (data == NULL) { ret_val = -ENOMEM; goto list_failure; diff --git a/net/netlabel/netlabel_user.h b/net/netlabel/netlabel_user.h index 47967ef3296..6d7f4ab46c2 100644 --- a/net/netlabel/netlabel_user.h +++ b/net/netlabel/netlabel_user.h @@ -41,37 +41,6 @@ /* NetLabel NETLINK helper functions */ -/** - * netlbl_netlink_hdr_put - Write the NETLINK buffers into a sk_buff - * @skb: the packet - * @pid: the PID of the receipient - * @seq: the sequence number - * @type: the generic NETLINK message family type - * @cmd: command - * - * Description: - * Write both a NETLINK nlmsghdr structure and a Generic NETLINK genlmsghdr - * struct to the packet. Returns a pointer to the start of the payload buffer - * on success or NULL on failure. - * - */ -static inline void *netlbl_netlink_hdr_put(struct sk_buff *skb, - u32 pid, - u32 seq, - int type, - int flags, - u8 cmd) -{ - return genlmsg_put(skb, - pid, - seq, - type, - 0, - flags, - cmd, - NETLBL_PROTO_VERSION); -} - /** * netlbl_netlink_auditinfo - Fetch the audit information from a NETLINK msg * @skb: the packet diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c index 0249a56a9aa..8903ef93db5 100644 --- a/net/netlink/genetlink.c +++ b/net/netlink/genetlink.c @@ -384,6 +384,13 @@ static void genl_rcv(struct sock *sk, int len) * Controller **************************************************************************/ +static struct genl_family genl_ctrl = { + .id = GENL_ID_CTRL, + .name = "nlctrl", + .version = 0x1, + .maxattr = CTRL_ATTR_MAX, +}; + static int ctrl_fill_info(struct genl_family *family, u32 pid, u32 seq, u32 flags, struct sk_buff *skb, u8 cmd) { @@ -392,8 +399,7 @@ static int ctrl_fill_info(struct genl_family *family, u32 pid, u32 seq, void *hdr; int idx = 1; - hdr = genlmsg_put(skb, pid, seq, GENL_ID_CTRL, 0, flags, cmd, - family->version); + hdr = genlmsg_put(skb, pid, seq, &genl_ctrl, flags, cmd); if (hdr == NULL) return -1; @@ -562,13 +568,6 @@ static struct genl_ops genl_ctrl_ops = { .policy = ctrl_policy, }; -static struct genl_family genl_ctrl = { - .id = GENL_ID_CTRL, - .name = "nlctrl", - .version = 0x1, - .maxattr = CTRL_ATTR_MAX, -}; - static int __init genl_init(void) { int i, err; -- cgit v1.2.3-70-g09d2 From 6051e2f4fb68fc8e5343db58fa680ece376f405c Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Tue, 14 Nov 2006 19:54:19 -0800 Subject: [IPv6] prefix: Convert RTM_NEWPREFIX notifications to use the new netlink api RTM_GETPREFIX is completely unused and is thus removed. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/rtnetlink.h | 2 -- net/core/rtnetlink.c | 1 - net/ipv6/addrconf.c | 30 ++++++++++++++---------------- security/selinux/nlmsgtab.c | 1 - 4 files changed, 14 insertions(+), 20 deletions(-) (limited to 'include') diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index 3a18addaed4..33b3d0ab3a9 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -81,8 +81,6 @@ enum { RTM_NEWPREFIX = 52, #define RTM_NEWPREFIX RTM_NEWPREFIX - RTM_GETPREFIX = 54, -#define RTM_GETPREFIX RTM_GETPREFIX RTM_GETMULTICAST = 58, #define RTM_GETMULTICAST RTM_GETMULTICAST diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 50d6cb40c6e..0cb4d9e53a0 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -108,7 +108,6 @@ static const int rtm_min[RTM_NR_FAMILIES] = [RTM_FAM(RTM_NEWTCLASS)] = NLMSG_LENGTH(sizeof(struct tcmsg)), [RTM_FAM(RTM_NEWTFILTER)] = NLMSG_LENGTH(sizeof(struct tcmsg)), [RTM_FAM(RTM_NEWACTION)] = NLMSG_LENGTH(sizeof(struct tcamsg)), - [RTM_FAM(RTM_NEWPREFIX)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)), [RTM_FAM(RTM_GETMULTICAST)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)), [RTM_FAM(RTM_GETANYCAST)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)), }; diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 46cd941d296..052f99eaf2a 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -3518,16 +3518,18 @@ static inline size_t inet6_prefix_nlmsg_size(void) } static int inet6_fill_prefix(struct sk_buff *skb, struct inet6_dev *idev, - struct prefix_info *pinfo, u32 pid, u32 seq, - int event, unsigned int flags) + struct prefix_info *pinfo, u32 pid, u32 seq, + int event, unsigned int flags) { - struct prefixmsg *pmsg; - struct nlmsghdr *nlh; - unsigned char *b = skb->tail; + struct prefixmsg *pmsg; + struct nlmsghdr *nlh; struct prefix_cacheinfo ci; - nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*pmsg), flags); - pmsg = NLMSG_DATA(nlh); + nlh = nlmsg_put(skb, pid, seq, event, sizeof(*pmsg), flags); + if (nlh == NULL) + return -ENOBUFS; + + pmsg = nlmsg_data(nlh); pmsg->prefix_family = AF_INET6; pmsg->prefix_pad1 = 0; pmsg->prefix_pad2 = 0; @@ -3535,26 +3537,22 @@ static int inet6_fill_prefix(struct sk_buff *skb, struct inet6_dev *idev, pmsg->prefix_len = pinfo->prefix_len; pmsg->prefix_type = pinfo->type; pmsg->prefix_pad3 = 0; - pmsg->prefix_flags = 0; if (pinfo->onlink) pmsg->prefix_flags |= IF_PREFIX_ONLINK; if (pinfo->autoconf) pmsg->prefix_flags |= IF_PREFIX_AUTOCONF; - RTA_PUT(skb, PREFIX_ADDRESS, sizeof(pinfo->prefix), &pinfo->prefix); + NLA_PUT(skb, PREFIX_ADDRESS, sizeof(pinfo->prefix), &pinfo->prefix); ci.preferred_time = ntohl(pinfo->prefered); ci.valid_time = ntohl(pinfo->valid); - RTA_PUT(skb, PREFIX_CACHEINFO, sizeof(ci), &ci); + NLA_PUT(skb, PREFIX_CACHEINFO, sizeof(ci), &ci); - nlh->nlmsg_len = skb->tail - b; - return skb->len; + return nlmsg_end(skb, nlh); -nlmsg_failure: -rtattr_failure: - skb_trim(skb, b - skb->data); - return -1; +nla_put_failure: + return nlmsg_cancel(skb, nlh); } static void inet6_prefix_notify(int event, struct inet6_dev *idev, diff --git a/security/selinux/nlmsgtab.c b/security/selinux/nlmsgtab.c index b8f4d25cf33..ccfe8755735 100644 --- a/security/selinux/nlmsgtab.c +++ b/security/selinux/nlmsgtab.c @@ -60,7 +60,6 @@ static struct nlmsg_perm nlmsg_route_perms[] = { RTM_DELACTION, NETLINK_ROUTE_SOCKET__NLMSG_WRITE }, { RTM_GETACTION, NETLINK_ROUTE_SOCKET__NLMSG_READ }, { RTM_NEWPREFIX, NETLINK_ROUTE_SOCKET__NLMSG_WRITE }, - { RTM_GETPREFIX, NETLINK_ROUTE_SOCKET__NLMSG_READ }, { RTM_GETMULTICAST, NETLINK_ROUTE_SOCKET__NLMSG_READ }, { RTM_GETANYCAST, NETLINK_ROUTE_SOCKET__NLMSG_READ }, { RTM_GETNEIGHTBL, NETLINK_ROUTE_SOCKET__NLMSG_READ }, -- cgit v1.2.3-70-g09d2 From ba4e58eca8aa9473b44fdfd312f26c4a2e7798b3 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Mon, 27 Nov 2006 11:10:57 -0800 Subject: [NET]: Supporting UDP-Lite (RFC 3828) in Linux This is a revision of the previously submitted patch, which alters the way files are organized and compiled in the following manner: * UDP and UDP-Lite now use separate object files * source file dependencies resolved via header files net/ipv{4,6}/udp_impl.h * order of inclusion files in udp.c/udplite.c adapted accordingly [NET/IPv4]: Support for the UDP-Lite protocol (RFC 3828) This patch adds support for UDP-Lite to the IPv4 stack, provided as an extension to the existing UDPv4 code: * generic routines are all located in net/ipv4/udp.c * UDP-Lite specific routines are in net/ipv4/udplite.c * MIB/statistics support in /proc/net/snmp and /proc/net/udplite * shared API with extensions for partial checksum coverage [NET/IPv6]: Extension for UDP-Lite over IPv6 It extends the existing UDPv6 code base with support for UDP-Lite in the same manner as per UDPv4. In particular, * UDPv6 generic and shared code is in net/ipv6/udp.c * UDP-Litev6 specific extensions are in net/ipv6/udplite.c * MIB/statistics support in /proc/net/snmp6 and /proc/net/udplite6 * support for IPV6_ADDRFORM * aligned the coding style of protocol initialisation with af_inet6.c * made the error handling in udpv6_queue_rcv_skb consistent; to return `-1' on error on all error cases * consolidation of shared code [NET]: UDP-Lite Documentation and basic XFRM/Netfilter support The UDP-Lite patch further provides * API documentation for UDP-Lite * basic xfrm support * basic netfilter support for IPv4 and IPv6 (LOG target) Signed-off-by: Gerrit Renker Signed-off-by: David S. Miller --- Documentation/networking/udplite.txt | 281 +++++++++++++++++++ include/linux/in.h | 1 + include/linux/socket.h | 1 + include/linux/udp.h | 12 + include/net/ipv6.h | 12 +- include/net/transp_v6.h | 2 + include/net/udp.h | 91 +++++- include/net/udplite.h | 149 ++++++++++ include/net/xfrm.h | 2 + net/ipv4/Makefile | 3 +- net/ipv4/af_inet.c | 8 +- net/ipv4/netfilter/ipt_LOG.c | 11 +- net/ipv4/proc.c | 13 + net/ipv4/udp.c | 518 ++++++++++++++++++++--------------- net/ipv4/udp_impl.h | 38 +++ net/ipv4/udplite.c | 119 ++++++++ net/ipv4/xfrm4_policy.c | 1 + net/ipv6/Makefile | 4 +- net/ipv6/af_inet6.c | 21 +- net/ipv6/ipv6_sockglue.c | 11 +- net/ipv6/netfilter/ip6t_LOG.c | 10 +- net/ipv6/proc.c | 11 + net/ipv6/udp.c | 361 +++++++++++++----------- net/ipv6/udp_impl.h | 34 +++ net/ipv6/udplite.c | 105 +++++++ net/ipv6/xfrm6_policy.c | 1 + net/netfilter/xt_multiport.c | 5 +- net/netfilter/xt_tcpudp.c | 20 +- 28 files changed, 1442 insertions(+), 403 deletions(-) create mode 100644 Documentation/networking/udplite.txt create mode 100644 include/net/udplite.h create mode 100644 net/ipv4/udp_impl.h create mode 100644 net/ipv4/udplite.c create mode 100644 net/ipv6/udp_impl.h create mode 100644 net/ipv6/udplite.c (limited to 'include') diff --git a/Documentation/networking/udplite.txt b/Documentation/networking/udplite.txt new file mode 100644 index 00000000000..dd6f46b83da --- /dev/null +++ b/Documentation/networking/udplite.txt @@ -0,0 +1,281 @@ + =========================================================================== + The UDP-Lite protocol (RFC 3828) + =========================================================================== + + + UDP-Lite is a Standards-Track IETF transport protocol whose characteristic + is a variable-length checksum. This has advantages for transport of multimedia + (video, VoIP) over wireless networks, as partly damaged packets can still be + fed into the codec instead of being discarded due to a failed checksum test. + + This file briefly describes the existing kernel support and the socket API. + For in-depth information, you can consult: + + o The UDP-Lite Homepage: http://www.erg.abdn.ac.uk/users/gerrit/udp-lite/ + Fom here you can also download some example application source code. + + o The UDP-Lite HOWTO on + http://www.erg.abdn.ac.uk/users/gerrit/udp-lite/files/UDP-Lite-HOWTO.txt + + o The Wireshark UDP-Lite WiKi (with capture files): + http://wiki.wireshark.org/Lightweight_User_Datagram_Protocol + + o The Protocol Spec, RFC 3828, http://www.ietf.org/rfc/rfc3828.txt + + + I) APPLICATIONS + + Several applications have been ported successfully to UDP-Lite. Ethereal + (now called wireshark) has UDP-Litev4/v6 support by default. The tarball on + + http://www.erg.abdn.ac.uk/users/gerrit/udp-lite/files/udplite_linux.tar.gz + + has source code for several v4/v6 client-server and network testing examples. + + Porting applications to UDP-Lite is straightforward: only socket level and + IPPROTO need to be changed; senders additionally set the checksum coverage + length (default = header length = 8). Details are in the next section. + + + II) PROGRAMMING API + + UDP-Lite provides a connectionless, unreliable datagram service and hence + uses the same socket type as UDP. In fact, porting from UDP to UDP-Lite is + very easy: simply add `IPPROTO_UDPLITE' as the last argument of the socket(2) + call so that the statement looks like: + + s = socket(PF_INET, SOCK_DGRAM, IPPROTO_UDPLITE); + + or, respectively, + + s = socket(PF_INET6, SOCK_DGRAM, IPPROTO_UDPLITE); + + With just the above change you are able to run UDP-Lite services or connect + to UDP-Lite servers. The kernel will assume that you are not interested in + using partial checksum coverage and so emulate UDP mode (full coverage). + + To make use of the partial checksum coverage facilities requires setting a + single socket option, which takes an integer specifying the coverage length: + + * Sender checksum coverage: UDPLITE_SEND_CSCOV + + For example, + + int val = 20; + setsockopt(s, SOL_UDPLITE, UDPLITE_SEND_CSCOV, &val, sizeof(int)); + + sets the checksum coverage length to 20 bytes (12b data + 8b header). + Of each packet only the first 20 bytes (plus the pseudo-header) will be + checksummed. This is useful for RTP applications which have a 12-byte + base header. + + + * Receiver checksum coverage: UDPLITE_RECV_CSCOV + + This option is the receiver-side analogue. It is truly optional, i.e. not + required to enable traffic with partial checksum coverage. Its function is + that of a traffic filter: when enabled, it instructs the kernel to drop + all packets which have a coverage _less_ than this value. For example, if + RTP and UDP headers are to be protected, a receiver can enforce that only + packets with a minimum coverage of 20 are admitted: + + int min = 20; + setsockopt(s, SOL_UDPLITE, UDPLITE_RECV_CSCOV, &min, sizeof(int)); + + The calls to getsockopt(2) are analogous. Being an extension and not a stand- + alone protocol, all socket options known from UDP can be used in exactly the + same manner as before, e.g. UDP_CORK or UDP_ENCAP. + + A detailed discussion of UDP-Lite checksum coverage options is in section IV. + + + III) HEADER FILES + + The socket API requires support through header files in /usr/include: + + * /usr/include/netinet/in.h + to define IPPROTO_UDPLITE + + * /usr/include/netinet/udplite.h + for UDP-Lite header fields and protocol constants + + For testing purposes, the following can serve as a `mini' header file: + + #define IPPROTO_UDPLITE 136 + #define SOL_UDPLITE 136 + #define UDPLITE_SEND_CSCOV 10 + #define UDPLITE_RECV_CSCOV 11 + + Ready-made header files for various distros are in the UDP-Lite tarball. + + + IV) KERNEL BEHAVIOUR WITH REGARD TO THE VARIOUS SOCKET OPTIONS + + To enable debugging messages, the log level need to be set to 8, as most + messages use the KERN_DEBUG level (7). + + 1) Sender Socket Options + + If the sender specifies a value of 0 as coverage length, the module + assumes full coverage, transmits a packet with coverage length of 0 + and according checksum. If the sender specifies a coverage < 8 and + different from 0, the kernel assumes 8 as default value. Finally, + if the specified coverage length exceeds the packet length, the packet + length is used instead as coverage length. + + 2) Receiver Socket Options + + The receiver specifies the minimum value of the coverage length it + is willing to accept. A value of 0 here indicates that the receiver + always wants the whole of the packet covered. In this case, all + partially covered packets are dropped and an error is logged. + + It is not possible to specify illegal values (<0 and <8); in these + cases the default of 8 is assumed. + + All packets arriving with a coverage value less than the specified + threshold are discarded, these events are also logged. + + 3) Disabling the Checksum Computation + + On both sender and receiver, checksumming will always be performed + and can not be disabled using SO_NO_CHECK. Thus + + setsockopt(sockfd, SOL_SOCKET, SO_NO_CHECK, ... ); + + will always will be ignored, while the value of + + getsockopt(sockfd, SOL_SOCKET, SO_NO_CHECK, &value, ...); + + is meaningless (as in TCP). Packets with a zero checksum field are + illegal (cf. RFC 3828, sec. 3.1) will be silently discarded. + + 4) Fragmentation + + The checksum computation respects both buffersize and MTU. The size + of UDP-Lite packets is determined by the size of the send buffer. The + minimum size of the send buffer is 2048 (defined as SOCK_MIN_SNDBUF + in include/net/sock.h), the default value is configurable as + net.core.wmem_default or via setting the SO_SNDBUF socket(7) + option. The maximum upper bound for the send buffer is determined + by net.core.wmem_max. + + Given a payload size larger than the send buffer size, UDP-Lite will + split the payload into several individual packets, filling up the + send buffer size in each case. + + The precise value also depends on the interface MTU. The interface MTU, + in turn, may trigger IP fragmentation. In this case, the generated + UDP-Lite packet is split into several IP packets, of which only the + first one contains the L4 header. + + The send buffer size has implications on the checksum coverage length. + Consider the following example: + + Payload: 1536 bytes Send Buffer: 1024 bytes + MTU: 1500 bytes Coverage Length: 856 bytes + + UDP-Lite will ship the 1536 bytes in two separate packets: + + Packet 1: 1024 payload + 8 byte header + 20 byte IP header = 1052 bytes + Packet 2: 512 payload + 8 byte header + 20 byte IP header = 540 bytes + + The coverage packet covers the UDP-Lite header and 848 bytes of the + payload in the first packet, the second packet is fully covered. Note + that for the second packet, the coverage length exceeds the packet + length. The kernel always re-adjusts the coverage length to the packet + length in such cases. + + As an example of what happens when one UDP-Lite packet is split into + several tiny fragments, consider the following example. + + Payload: 1024 bytes Send buffer size: 1024 bytes + MTU: 300 bytes Coverage length: 575 bytes + + +-+-----------+--------------+--------------+--------------+ + |8| 272 | 280 | 280 | 280 | + +-+-----------+--------------+--------------+--------------+ + 280 560 840 1032 + ^ + *****checksum coverage************* + + The UDP-Lite module generates one 1032 byte packet (1024 + 8 byte + header). According to the interface MTU, these are split into 4 IP + packets (280 byte IP payload + 20 byte IP header). The kernel module + sums the contents of the entire first two packets, plus 15 bytes of + the last packet before releasing the fragments to the IP module. + + To see the analogous case for IPv6 fragmentation, consider a link + MTU of 1280 bytes and a write buffer of 3356 bytes. If the checksum + coverage is less than 1232 bytes (MTU minus IPv6/fragment header + lengths), only the first fragment needs to be considered. When using + larger checksum coverage lengths, each eligible fragment needs to be + checksummed. Suppose we have a checksum coverage of 3062. The buffer + of 3356 bytes will be split into the following fragments: + + Fragment 1: 1280 bytes carrying 1232 bytes of UDP-Lite data + Fragment 2: 1280 bytes carrying 1232 bytes of UDP-Lite data + Fragment 3: 948 bytes carrying 900 bytes of UDP-Lite data + + The first two fragments have to be checksummed in full, of the last + fragment only 598 (= 3062 - 2*1232) bytes are checksummed. + + While it is important that such cases are dealt with correctly, they + are (annoyingly) rare: UDP-Lite is designed for optimising multimedia + performance over wireless (or generally noisy) links and thus smaller + coverage lenghts are likely to be expected. + + + V) UDP-LITE RUNTIME STATISTICS AND THEIR MEANING + + Exceptional and error conditions are logged to syslog at the KERN_DEBUG + level. Live statistics about UDP-Lite are available in /proc/net/snmp + and can (with newer versions of netstat) be viewed using + + netstat -svu + + This displays UDP-Lite statistics variables, whose meaning is as follows. + + InDatagrams: Total number of received datagrams. + + NoPorts: Number of packets received to an unknown port. + These cases are counted separately (not as InErrors). + + InErrors: Number of erroneous UDP-Lite packets. Errors include: + * internal socket queue receive errors + * packet too short (less than 8 bytes or stated + coverage length exceeds received length) + * xfrm4_policy_check() returned with error + * application has specified larger min. coverage + length than that of incoming packet + * checksum coverage violated + * bad checksum + + OutDatagrams: Total number of sent datagrams. + + These statistics derive from the UDP MIB (RFC 2013). + + + VI) IPTABLES + + There is packet match support for UDP-Lite as well as support for the LOG target. + If you copy and paste the following line into /etc/protcols, + + udplite 136 UDP-Lite # UDP-Lite [RFC 3828] + + then + iptables -A INPUT -p udplite -j LOG + + will produce logging output to syslog. Dropping and rejecting packets also works. + + + VII) MAINTAINER ADDRESS + + The UDP-Lite patch was developed at + University of Aberdeen + Electronics Research Group + Department of Engineering + Fraser Noble Building + Aberdeen AB24 3UE; UK + The current maintainer is Gerrit Renker, . Initial + code was developed by William Stanislaus, . diff --git a/include/linux/in.h b/include/linux/in.h index 2619859f6e1..1912e7c0bc2 100644 --- a/include/linux/in.h +++ b/include/linux/in.h @@ -45,6 +45,7 @@ enum { IPPROTO_COMP = 108, /* Compression Header protocol */ IPPROTO_SCTP = 132, /* Stream Control Transport Protocol */ + IPPROTO_UDPLITE = 136, /* UDP-Lite (RFC 3828) */ IPPROTO_RAW = 255, /* Raw IP packets */ IPPROTO_MAX diff --git a/include/linux/socket.h b/include/linux/socket.h index 36140909464..592b6667982 100644 --- a/include/linux/socket.h +++ b/include/linux/socket.h @@ -264,6 +264,7 @@ struct ucred { #define SOL_IPV6 41 #define SOL_ICMPV6 58 #define SOL_SCTP 132 +#define SOL_UDPLITE 136 /* UDP-Lite (RFC 3828) */ #define SOL_RAW 255 #define SOL_IPX 256 #define SOL_AX25 257 diff --git a/include/linux/udp.h b/include/linux/udp.h index 014b41d1e30..564f3b05010 100644 --- a/include/linux/udp.h +++ b/include/linux/udp.h @@ -38,6 +38,7 @@ struct udphdr { #include #include +#define UDP_HTABLE_SIZE 128 struct udp_sock { /* inet_sock has to be the first member */ @@ -50,12 +51,23 @@ struct udp_sock { * when the socket is uncorked. */ __u16 len; /* total length of pending frames */ + /* + * Fields specific to UDP-Lite. + */ + __u16 pcslen; + __u16 pcrlen; +/* indicator bits used by pcflag: */ +#define UDPLITE_BIT 0x1 /* set by udplite proto init function */ +#define UDPLITE_SEND_CC 0x2 /* set via udplite setsockopt */ +#define UDPLITE_RECV_CC 0x4 /* set via udplite setsocktopt */ + __u8 pcflag; /* marks socket as UDP-Lite if > 0 */ }; static inline struct udp_sock *udp_sk(const struct sock *sk) { return (struct udp_sock *)sk; } +#define IS_UDPLITE(__sk) (udp_sk(__sk)->pcflag) #endif diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 3c266ad99a0..9390649bbfe 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -158,9 +158,13 @@ DECLARE_SNMP_STAT(struct icmpv6_mib, icmpv6_statistics); SNMP_INC_STATS_OFFSET_BH(icmpv6_statistics, field, _offset); \ }) DECLARE_SNMP_STAT(struct udp_mib, udp_stats_in6); -#define UDP6_INC_STATS(field) SNMP_INC_STATS(udp_stats_in6, field) -#define UDP6_INC_STATS_BH(field) SNMP_INC_STATS_BH(udp_stats_in6, field) -#define UDP6_INC_STATS_USER(field) SNMP_INC_STATS_USER(udp_stats_in6, field) +DECLARE_SNMP_STAT(struct udp_mib, udplite_stats_in6); +#define UDP6_INC_STATS_BH(field, is_udplite) do { \ + if (is_udplite) SNMP_INC_STATS_BH(udplite_stats_in6, field); \ + else SNMP_INC_STATS_BH(udp_stats_in6, field); } while(0) +#define UDP6_INC_STATS_USER(field, is_udplite) do { \ + if (is_udplite) SNMP_INC_STATS_USER(udplite_stats_in6, field); \ + else SNMP_INC_STATS_USER(udp_stats_in6, field); } while(0) int snmp6_register_dev(struct inet6_dev *idev); int snmp6_unregister_dev(struct inet6_dev *idev); @@ -604,6 +608,8 @@ extern int tcp6_proc_init(void); extern void tcp6_proc_exit(void); extern int udp6_proc_init(void); extern void udp6_proc_exit(void); +extern int udplite6_proc_init(void); +extern void udplite6_proc_exit(void); extern int ipv6_misc_proc_init(void); extern void ipv6_misc_proc_exit(void); diff --git a/include/net/transp_v6.h b/include/net/transp_v6.h index 61f724c1036..409da3a9a45 100644 --- a/include/net/transp_v6.h +++ b/include/net/transp_v6.h @@ -11,6 +11,7 @@ extern struct proto rawv6_prot; extern struct proto udpv6_prot; +extern struct proto udplitev6_prot; extern struct proto tcpv6_prot; struct flowi; @@ -24,6 +25,7 @@ extern void ipv6_destopt_init(void); /* transport protocols */ extern void rawv6_init(void); extern void udpv6_init(void); +extern void udplitev6_init(void); extern void tcpv6_init(void); extern int udpv6_connect(struct sock *sk, diff --git a/include/net/udp.h b/include/net/udp.h index db0c05f6754..4f0626735ed 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -26,9 +26,28 @@ #include #include #include +#include +#include #include -#define UDP_HTABLE_SIZE 128 +/** + * struct udp_skb_cb - UDP(-Lite) private variables + * + * @header: private variables used by IPv4/IPv6 + * @cscov: checksum coverage length (UDP-Lite only) + * @partial_cov: if set indicates partial csum coverage + */ +struct udp_skb_cb { + union { + struct inet_skb_parm h4; +#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE) + struct inet6_skb_parm h6; +#endif + } header; + __u16 cscov; + __u8 partial_cov; +}; +#define UDP_SKB_CB(__skb) ((struct udp_skb_cb *)((__skb)->cb)) extern struct hlist_head udp_hash[UDP_HTABLE_SIZE]; extern rwlock_t udp_hash_lock; @@ -47,6 +66,62 @@ extern struct proto udp_prot; struct sk_buff; +/* + * Generic checksumming routines for UDP(-Lite) v4 and v6 + */ +static inline u16 __udp_lib_checksum_complete(struct sk_buff *skb) +{ + if (! UDP_SKB_CB(skb)->partial_cov) + return __skb_checksum_complete(skb); + return csum_fold(skb_checksum(skb, 0, UDP_SKB_CB(skb)->cscov, + skb->csum)); +} + +static __inline__ int udp_lib_checksum_complete(struct sk_buff *skb) +{ + return skb->ip_summed != CHECKSUM_UNNECESSARY && + __udp_lib_checksum_complete(skb); +} + +/** + * udp_csum_outgoing - compute UDPv4/v6 checksum over fragments + * @sk: socket we are writing to + * @skb: sk_buff containing the filled-in UDP header + * (checksum field must be zeroed out) + */ +static inline u32 udp_csum_outgoing(struct sock *sk, struct sk_buff *skb) +{ + u32 csum = csum_partial(skb->h.raw, sizeof(struct udphdr), 0); + + skb_queue_walk(&sk->sk_write_queue, skb) { + csum = csum_add(csum, skb->csum); + } + return csum; +} + +/* hash routines shared between UDPv4/6 and UDP-Litev4/6 */ +static inline void udp_lib_hash(struct sock *sk) +{ + BUG(); +} + +static inline void udp_lib_unhash(struct sock *sk) +{ + write_lock_bh(&udp_hash_lock); + if (sk_del_node_init(sk)) { + inet_sk(sk)->num = 0; + sock_prot_dec_use(sk->sk_prot); + } + write_unlock_bh(&udp_hash_lock); +} + +static inline void udp_lib_close(struct sock *sk, long timeout) +{ + sk_common_release(sk); +} + + +/* net/ipv4/udp.c */ extern int udp_get_port(struct sock *sk, unsigned short snum, int (*saddr_cmp)(const struct sock *, const struct sock *)); extern void udp_err(struct sk_buff *, u32); @@ -61,21 +136,29 @@ extern unsigned int udp_poll(struct file *file, struct socket *sock, poll_table *wait); DECLARE_SNMP_STAT(struct udp_mib, udp_statistics); -#define UDP_INC_STATS(field) SNMP_INC_STATS(udp_statistics, field) -#define UDP_INC_STATS_BH(field) SNMP_INC_STATS_BH(udp_statistics, field) -#define UDP_INC_STATS_USER(field) SNMP_INC_STATS_USER(udp_statistics, field) +/* + * SNMP statistics for UDP and UDP-Lite + */ +#define UDP_INC_STATS_USER(field, is_udplite) do { \ + if (is_udplite) SNMP_INC_STATS_USER(udplite_statistics, field); \ + else SNMP_INC_STATS_USER(udp_statistics, field); } while(0) +#define UDP_INC_STATS_BH(field, is_udplite) do { \ + if (is_udplite) SNMP_INC_STATS_BH(udplite_statistics, field); \ + else SNMP_INC_STATS_BH(udp_statistics, field); } while(0) /* /proc */ struct udp_seq_afinfo { struct module *owner; char *name; sa_family_t family; + struct hlist_head *hashtable; int (*seq_show) (struct seq_file *m, void *v); struct file_operations *seq_fops; }; struct udp_iter_state { sa_family_t family; + struct hlist_head *hashtable; int bucket; struct seq_operations seq_ops; }; diff --git a/include/net/udplite.h b/include/net/udplite.h new file mode 100644 index 00000000000..1473b3e4904 --- /dev/null +++ b/include/net/udplite.h @@ -0,0 +1,149 @@ +/* + * Definitions for the UDP-Lite (RFC 3828) code. + */ +#ifndef _UDPLITE_H +#define _UDPLITE_H + +/* UDP-Lite socket options */ +#define UDPLITE_SEND_CSCOV 10 /* sender partial coverage (as sent) */ +#define UDPLITE_RECV_CSCOV 11 /* receiver partial coverage (threshold ) */ + +extern struct proto udplite_prot; +extern struct hlist_head udplite_hash[UDP_HTABLE_SIZE]; + +/* UDP-Lite does not have a standardized MIB yet, so we inherit from UDP */ +DECLARE_SNMP_STAT(struct udp_mib, udplite_statistics); + +/* + * Checksum computation is all in software, hence simpler getfrag. + */ +static __inline__ int udplite_getfrag(void *from, char *to, int offset, + int len, int odd, struct sk_buff *skb) +{ + return memcpy_fromiovecend(to, (struct iovec *) from, offset, len); +} + +/* Designate sk as UDP-Lite socket */ +static inline int udplite_sk_init(struct sock *sk) +{ + udp_sk(sk)->pcflag = UDPLITE_BIT; + return 0; +} + +/* + * Checksumming routines + */ +static inline int udplite_checksum_init(struct sk_buff *skb, struct udphdr *uh) +{ + u16 cscov; + + /* In UDPv4 a zero checksum means that the transmitter generated no + * checksum. UDP-Lite (like IPv6) mandates checksums, hence packets + * with a zero checksum field are illegal. */ + if (uh->check == 0) { + LIMIT_NETDEBUG(KERN_DEBUG "UDPLITE: zeroed checksum field\n"); + return 1; + } + + UDP_SKB_CB(skb)->partial_cov = 0; + cscov = ntohs(uh->len); + + if (cscov == 0) /* Indicates that full coverage is required. */ + cscov = skb->len; + else if (cscov < 8 || cscov > skb->len) { + /* + * Coverage length violates RFC 3828: log and discard silently. + */ + LIMIT_NETDEBUG(KERN_DEBUG "UDPLITE: bad csum coverage %d/%d\n", + cscov, skb->len); + return 1; + + } else if (cscov < skb->len) + UDP_SKB_CB(skb)->partial_cov = 1; + + UDP_SKB_CB(skb)->cscov = cscov; + + /* + * There is no known NIC manufacturer supporting UDP-Lite yet, + * hence ip_summed is always (re-)set to CHECKSUM_NONE. + */ + skb->ip_summed = CHECKSUM_NONE; + + return 0; +} + +static __inline__ int udplite4_csum_init(struct sk_buff *skb, struct udphdr *uh) +{ + int rc = udplite_checksum_init(skb, uh); + + if (!rc) + skb->csum = csum_tcpudp_nofold(skb->nh.iph->saddr, + skb->nh.iph->daddr, + skb->len, IPPROTO_UDPLITE, 0); + return rc; +} + +static __inline__ int udplite6_csum_init(struct sk_buff *skb, struct udphdr *uh) +{ + int rc = udplite_checksum_init(skb, uh); + + if (!rc) + skb->csum = ~csum_ipv6_magic(&skb->nh.ipv6h->saddr, + &skb->nh.ipv6h->daddr, + skb->len, IPPROTO_UDPLITE, 0); + return rc; +} + +static inline int udplite_sender_cscov(struct udp_sock *up, struct udphdr *uh) +{ + int cscov = up->len; + + /* + * Sender has set `partial coverage' option on UDP-Lite socket + */ + if (up->pcflag & UDPLITE_SEND_CC) { + if (up->pcslen < up->len) { + /* up->pcslen == 0 means that full coverage is required, + * partial coverage only if 0 < up->pcslen < up->len */ + if (0 < up->pcslen) { + cscov = up->pcslen; + } + uh->len = htons(up->pcslen); + } + /* + * NOTE: Causes for the error case `up->pcslen > up->len': + * (i) Application error (will not be penalized). + * (ii) Payload too big for send buffer: data is split + * into several packets, each with its own header. + * In this case (e.g. last segment), coverage may + * exceed packet length. + * Since packets with coverage length > packet length are + * illegal, we fall back to the defaults here. + */ + } + return cscov; +} + +static inline u32 udplite_csum_outgoing(struct sock *sk, struct sk_buff *skb) +{ + u32 csum = 0; + int off, len, cscov = udplite_sender_cscov(udp_sk(sk), skb->h.uh); + + skb->ip_summed = CHECKSUM_NONE; /* no HW support for checksumming */ + + skb_queue_walk(&sk->sk_write_queue, skb) { + off = skb->h.raw - skb->data; + len = skb->len - off; + + csum = skb_checksum(skb, off, (cscov > len)? len : cscov, csum); + + if ((cscov -= len) <= 0) + break; + } + return csum; +} + +extern void udplite4_register(void); +extern int udplite_get_port(struct sock *sk, unsigned short snum, + int (*scmp)(const struct sock *, const struct sock *)); +#endif /* _UDPLITE_H */ diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 81c91e8a328..3878a88ff61 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -468,6 +468,7 @@ __be16 xfrm_flowi_sport(struct flowi *fl) switch(fl->proto) { case IPPROTO_TCP: case IPPROTO_UDP: + case IPPROTO_UDPLITE: case IPPROTO_SCTP: port = fl->fl_ip_sport; break; @@ -493,6 +494,7 @@ __be16 xfrm_flowi_dport(struct flowi *fl) switch(fl->proto) { case IPPROTO_TCP: case IPPROTO_UDP: + case IPPROTO_UDPLITE: case IPPROTO_SCTP: port = fl->fl_ip_dport; break; diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile index 15645c51520..7a068626fee 100644 --- a/net/ipv4/Makefile +++ b/net/ipv4/Makefile @@ -8,7 +8,8 @@ obj-y := route.o inetpeer.o protocol.o \ inet_timewait_sock.o inet_connection_sock.o \ tcp.o tcp_input.o tcp_output.o tcp_timer.o tcp_ipv4.o \ tcp_minisocks.o tcp_cong.o \ - datagram.o raw.o udp.o arp.o icmp.o devinet.o af_inet.o igmp.o \ + datagram.o raw.o udp.o udplite.o \ + arp.o icmp.o devinet.o af_inet.o igmp.o \ sysctl_net_ipv4.o fib_frontend.o fib_semantics.o obj-$(CONFIG_IP_FIB_HASH) += fib_hash.o diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 4a81d54a756..8db39f7e3bf 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -104,6 +104,7 @@ #include #include #include +#include #include #include #include @@ -1223,10 +1224,13 @@ static int __init init_ipv4_mibs(void) tcp_statistics[1] = alloc_percpu(struct tcp_mib); udp_statistics[0] = alloc_percpu(struct udp_mib); udp_statistics[1] = alloc_percpu(struct udp_mib); + udplite_statistics[0] = alloc_percpu(struct udp_mib); + udplite_statistics[1] = alloc_percpu(struct udp_mib); if (! (net_statistics[0] && net_statistics[1] && ip_statistics[0] && ip_statistics[1] && tcp_statistics[0] && tcp_statistics[1] - && udp_statistics[0] && udp_statistics[1])) + && udp_statistics[0] && udp_statistics[1] + && udplite_statistics[0] && udplite_statistics[1] ) ) return -ENOMEM; (void) tcp_mib_init(); @@ -1313,6 +1317,8 @@ static int __init inet_init(void) /* Setup TCP slab cache for open requests. */ tcp_init(); + /* Add UDP-Lite (RFC 3828) */ + udplite4_register(); /* * Set the ICMP layer up diff --git a/net/ipv4/netfilter/ipt_LOG.c b/net/ipv4/netfilter/ipt_LOG.c index 7dc820df8bc..46eee64a11f 100644 --- a/net/ipv4/netfilter/ipt_LOG.c +++ b/net/ipv4/netfilter/ipt_LOG.c @@ -171,11 +171,15 @@ static void dump_packet(const struct nf_loginfo *info, } break; } - case IPPROTO_UDP: { + case IPPROTO_UDP: + case IPPROTO_UDPLITE: { struct udphdr _udph, *uh; - /* Max length: 10 "PROTO=UDP " */ - printk("PROTO=UDP "); + if (ih->protocol == IPPROTO_UDP) + /* Max length: 10 "PROTO=UDP " */ + printk("PROTO=UDP " ); + else /* Max length: 14 "PROTO=UDPLITE " */ + printk("PROTO=UDPLITE "); if (ntohs(ih->frag_off) & IP_OFFSET) break; @@ -341,6 +345,7 @@ static void dump_packet(const struct nf_loginfo *info, /* IP: 40+46+6+11+127 = 230 */ /* TCP: 10+max(25,20+30+13+9+32+11+127) = 252 */ /* UDP: 10+max(25,20) = 35 */ + /* UDPLITE: 14+max(25,20) = 39 */ /* ICMP: 11+max(25, 18+25+max(19,14,24+3+n+10,3+n+10)) = 91+n */ /* ESP: 10+max(25)+15 = 50 */ /* AH: 9+max(25)+15 = 49 */ diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c index 9c6cbe3d9fb..cd873da54cb 100644 --- a/net/ipv4/proc.c +++ b/net/ipv4/proc.c @@ -38,6 +38,7 @@ #include #include #include +#include #include #include #include @@ -66,6 +67,7 @@ static int sockstat_seq_show(struct seq_file *seq, void *v) tcp_death_row.tw_count, atomic_read(&tcp_sockets_allocated), atomic_read(&tcp_memory_allocated)); seq_printf(seq, "UDP: inuse %d\n", fold_prot_inuse(&udp_prot)); + seq_printf(seq, "UDPLITE: inuse %d\n", fold_prot_inuse(&udplite_prot)); seq_printf(seq, "RAW: inuse %d\n", fold_prot_inuse(&raw_prot)); seq_printf(seq, "FRAG: inuse %d memory %d\n", ip_frag_nqueues, atomic_read(&ip_frag_mem)); @@ -304,6 +306,17 @@ static int snmp_seq_show(struct seq_file *seq, void *v) fold_field((void **) udp_statistics, snmp4_udp_list[i].entry)); + /* the UDP and UDP-Lite MIBs are the same */ + seq_puts(seq, "\nUdpLite:"); + for (i = 0; snmp4_udp_list[i].name != NULL; i++) + seq_printf(seq, " %s", snmp4_udp_list[i].name); + + seq_puts(seq, "\nUdpLite:"); + for (i = 0; snmp4_udp_list[i].name != NULL; i++) + seq_printf(seq, " %lu", + fold_field((void **) udplite_statistics, + snmp4_udp_list[i].entry) ); + seq_putc(seq, '\n'); return 0; } diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 9e1bd374875..98ba7509617 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -92,22 +92,16 @@ #include #include #include -#include #include -#include -#include #include -#include #include #include #include -#include -#include #include #include -#include #include #include +#include "udp_impl.h" /* * Snmp MIB for the UDP layer @@ -120,26 +114,30 @@ DEFINE_RWLOCK(udp_hash_lock); static int udp_port_rover; -static inline int udp_lport_inuse(u16 num) +static inline int __udp_lib_lport_inuse(__be16 num, struct hlist_head udptable[]) { struct sock *sk; struct hlist_node *node; - sk_for_each(sk, node, &udp_hash[num & (UDP_HTABLE_SIZE - 1)]) + sk_for_each(sk, node, &udptable[num & (UDP_HTABLE_SIZE - 1)]) if (inet_sk(sk)->num == num) return 1; return 0; } /** - * udp_get_port - common port lookup for IPv4 and IPv6 + * __udp_lib_get_port - UDP/-Lite port lookup for IPv4 and IPv6 * * @sk: socket struct in question * @snum: port number to look up + * @udptable: hash list table, must be of UDP_HTABLE_SIZE + * @port_rover: pointer to record of last unallocated port * @saddr_comp: AF-dependent comparison of bound local IP addresses */ -int udp_get_port(struct sock *sk, unsigned short snum, - int (*saddr_cmp)(const struct sock *sk1, const struct sock *sk2)) +int __udp_lib_get_port(struct sock *sk, unsigned short snum, + struct hlist_head udptable[], int *port_rover, + int (*saddr_comp)(const struct sock *sk1, + const struct sock *sk2 ) ) { struct hlist_node *node; struct hlist_head *head; @@ -150,15 +148,15 @@ int udp_get_port(struct sock *sk, unsigned short snum, if (snum == 0) { int best_size_so_far, best, result, i; - if (udp_port_rover > sysctl_local_port_range[1] || - udp_port_rover < sysctl_local_port_range[0]) - udp_port_rover = sysctl_local_port_range[0]; + if (*port_rover > sysctl_local_port_range[1] || + *port_rover < sysctl_local_port_range[0]) + *port_rover = sysctl_local_port_range[0]; best_size_so_far = 32767; - best = result = udp_port_rover; + best = result = *port_rover; for (i = 0; i < UDP_HTABLE_SIZE; i++, result++) { int size; - head = &udp_hash[result & (UDP_HTABLE_SIZE - 1)]; + head = &udptable[result & (UDP_HTABLE_SIZE - 1)]; if (hlist_empty(head)) { if (result > sysctl_local_port_range[1]) result = sysctl_local_port_range[0] + @@ -179,15 +177,15 @@ int udp_get_port(struct sock *sk, unsigned short snum, result = sysctl_local_port_range[0] + ((result - sysctl_local_port_range[0]) & (UDP_HTABLE_SIZE - 1)); - if (!udp_lport_inuse(result)) + if (! __udp_lib_lport_inuse(result, udptable)) break; } if (i >= (1 << 16) / UDP_HTABLE_SIZE) goto fail; gotit: - udp_port_rover = snum = result; + *port_rover = snum = result; } else { - head = &udp_hash[snum & (UDP_HTABLE_SIZE - 1)]; + head = &udptable[snum & (UDP_HTABLE_SIZE - 1)]; sk_for_each(sk2, node, head) if (inet_sk(sk2)->num == snum && @@ -195,12 +193,12 @@ gotit: (!sk2->sk_reuse || !sk->sk_reuse) && (!sk2->sk_bound_dev_if || !sk->sk_bound_dev_if || sk2->sk_bound_dev_if == sk->sk_bound_dev_if) && - (*saddr_cmp)(sk, sk2) ) + (*saddr_comp)(sk, sk2) ) goto fail; } inet_sk(sk)->num = snum; if (sk_unhashed(sk)) { - head = &udp_hash[snum & (UDP_HTABLE_SIZE - 1)]; + head = &udptable[snum & (UDP_HTABLE_SIZE - 1)]; sk_add_node(sk, head); sock_prot_inc_use(sk->sk_prot); } @@ -210,7 +208,13 @@ fail: return error; } -static inline int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2) +__inline__ int udp_get_port(struct sock *sk, unsigned short snum, + int (*scmp)(const struct sock *, const struct sock *)) +{ + return __udp_lib_get_port(sk, snum, udp_hash, &udp_port_rover, scmp); +} + +inline int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2) { struct inet_sock *inet1 = inet_sk(sk1), *inet2 = inet_sk(sk2); @@ -224,34 +228,20 @@ static inline int udp_v4_get_port(struct sock *sk, unsigned short snum) return udp_get_port(sk, snum, ipv4_rcv_saddr_equal); } - -static void udp_v4_hash(struct sock *sk) -{ - BUG(); -} - -static void udp_v4_unhash(struct sock *sk) -{ - write_lock_bh(&udp_hash_lock); - if (sk_del_node_init(sk)) { - inet_sk(sk)->num = 0; - sock_prot_dec_use(sk->sk_prot); - } - write_unlock_bh(&udp_hash_lock); -} - /* UDP is nearly always wildcards out the wazoo, it makes no sense to try * harder than this. -DaveM */ -static struct sock *udp_v4_lookup_longway(__be32 saddr, __be16 sport, - __be32 daddr, __be16 dport, int dif) +static struct sock *__udp4_lib_lookup(__be32 saddr, __be16 sport, + __be32 daddr, __be16 dport, + int dif, struct hlist_head udptable[]) { struct sock *sk, *result = NULL; struct hlist_node *node; unsigned short hnum = ntohs(dport); int badness = -1; - sk_for_each(sk, node, &udp_hash[hnum & (UDP_HTABLE_SIZE - 1)]) { + read_lock(&udp_hash_lock); + sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) { struct inet_sock *inet = inet_sk(sk); if (inet->num == hnum && !ipv6_only_sock(sk)) { @@ -285,20 +275,10 @@ static struct sock *udp_v4_lookup_longway(__be32 saddr, __be16 sport, } } } - return result; -} - -static __inline__ struct sock *udp_v4_lookup(__be32 saddr, __be16 sport, - __be32 daddr, __be16 dport, int dif) -{ - struct sock *sk; - - read_lock(&udp_hash_lock); - sk = udp_v4_lookup_longway(saddr, sport, daddr, dport, dif); - if (sk) - sock_hold(sk); + if (result) + sock_hold(result); read_unlock(&udp_hash_lock); - return sk; + return result; } static inline struct sock *udp_v4_mcast_next(struct sock *sk, @@ -340,7 +320,7 @@ found: * to find the appropriate port. */ -void udp_err(struct sk_buff *skb, u32 info) +void __udp4_lib_err(struct sk_buff *skb, u32 info, struct hlist_head udptable[]) { struct inet_sock *inet; struct iphdr *iph = (struct iphdr*)skb->data; @@ -351,7 +331,8 @@ void udp_err(struct sk_buff *skb, u32 info) int harderr; int err; - sk = udp_v4_lookup(iph->daddr, uh->dest, iph->saddr, uh->source, skb->dev->ifindex); + sk = __udp4_lib_lookup(iph->daddr, uh->dest, iph->saddr, uh->source, + skb->dev->ifindex, udptable ); if (sk == NULL) { ICMP_INC_STATS_BH(ICMP_MIB_INERRORS); return; /* No socket for error */ @@ -405,6 +386,11 @@ out: sock_put(sk); } +__inline__ void udp_err(struct sk_buff *skb, u32 info) +{ + return __udp4_lib_err(skb, info, udp_hash); +} + /* * Throw away all pending data and cancel the corking. Socket is locked. */ @@ -419,16 +405,56 @@ static void udp_flush_pending_frames(struct sock *sk) } } +/** + * udp4_hwcsum_outgoing - handle outgoing HW checksumming + * @sk: socket we are sending on + * @skb: sk_buff containing the filled-in UDP header + * (checksum field must be zeroed out) + */ +static void udp4_hwcsum_outgoing(struct sock *sk, struct sk_buff *skb, + __be32 src, __be32 dst, int len ) +{ + unsigned int csum = 0, offset; + struct udphdr *uh = skb->h.uh; + + if (skb_queue_len(&sk->sk_write_queue) == 1) { + /* + * Only one fragment on the socket. + */ + skb->csum = offsetof(struct udphdr, check); + uh->check = ~csum_tcpudp_magic(src, dst, len, IPPROTO_UDP, 0); + } else { + /* + * HW-checksum won't work as there are two or more + * fragments on the socket so that all csums of sk_buffs + * should be together + */ + offset = skb->h.raw - skb->data; + skb->csum = skb_checksum(skb, offset, skb->len - offset, 0); + + skb->ip_summed = CHECKSUM_NONE; + + skb_queue_walk(&sk->sk_write_queue, skb) { + csum = csum_add(csum, skb->csum); + } + + uh->check = csum_tcpudp_magic(src, dst, len, IPPROTO_UDP, csum); + if (uh->check == 0) + uh->check = -1; + } +} + /* * Push out all pending data as one UDP datagram. Socket is locked. */ -static int udp_push_pending_frames(struct sock *sk, struct udp_sock *up) +int udp_push_pending_frames(struct sock *sk, struct udp_sock *up) { struct inet_sock *inet = inet_sk(sk); struct flowi *fl = &inet->cork.fl; struct sk_buff *skb; struct udphdr *uh; int err = 0; + u32 csum = 0; /* Grab the skbuff where UDP header space exists. */ if ((skb = skb_peek(&sk->sk_write_queue)) == NULL) @@ -443,52 +469,28 @@ static int udp_push_pending_frames(struct sock *sk, struct udp_sock *up) uh->len = htons(up->len); uh->check = 0; - if (sk->sk_no_check == UDP_CSUM_NOXMIT) { + if (up->pcflag) /* UDP-Lite */ + csum = udplite_csum_outgoing(sk, skb); + + else if (sk->sk_no_check == UDP_CSUM_NOXMIT) { /* UDP csum disabled */ + skb->ip_summed = CHECKSUM_NONE; goto send; - } - if (skb_queue_len(&sk->sk_write_queue) == 1) { - /* - * Only one fragment on the socket. - */ - if (skb->ip_summed == CHECKSUM_PARTIAL) { - skb->csum = offsetof(struct udphdr, check); - uh->check = ~csum_tcpudp_magic(fl->fl4_src, fl->fl4_dst, - up->len, IPPROTO_UDP, 0); - } else { - skb->csum = csum_partial((char *)uh, - sizeof(struct udphdr), skb->csum); - uh->check = csum_tcpudp_magic(fl->fl4_src, fl->fl4_dst, - up->len, IPPROTO_UDP, skb->csum); - if (uh->check == 0) - uh->check = -1; - } - } else { - unsigned int csum = 0; - /* - * HW-checksum won't work as there are two or more - * fragments on the socket so that all csums of sk_buffs - * should be together. - */ - if (skb->ip_summed == CHECKSUM_PARTIAL) { - int offset = (unsigned char *)uh - skb->data; - skb->csum = skb_checksum(skb, offset, skb->len - offset, 0); + } else if (skb->ip_summed == CHECKSUM_PARTIAL) { /* UDP hardware csum */ - skb->ip_summed = CHECKSUM_NONE; - } else { - skb->csum = csum_partial((char *)uh, - sizeof(struct udphdr), skb->csum); - } + udp4_hwcsum_outgoing(sk, skb, fl->fl4_src,fl->fl4_dst, up->len); + goto send; + + } else /* `normal' UDP */ + csum = udp_csum_outgoing(sk, skb); + + /* add protocol-dependent pseudo-header */ + uh->check = csum_tcpudp_magic(fl->fl4_src, fl->fl4_dst, up->len, + sk->sk_protocol, csum ); + if (uh->check == 0) + uh->check = -1; - skb_queue_walk(&sk->sk_write_queue, skb) { - csum = csum_add(csum, skb->csum); - } - uh->check = csum_tcpudp_magic(fl->fl4_src, fl->fl4_dst, - up->len, IPPROTO_UDP, csum); - if (uh->check == 0) - uh->check = -1; - } send: err = ip_push_pending_frames(sk); out: @@ -497,12 +499,6 @@ out: return err; } - -static unsigned short udp_check(struct udphdr *uh, int len, __be32 saddr, __be32 daddr, unsigned long base) -{ - return(csum_tcpudp_magic(saddr, daddr, len, IPPROTO_UDP, base)); -} - int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, size_t len) { @@ -516,8 +512,9 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, __be32 daddr, faddr, saddr; __be16 dport; u8 tos; - int err; + int err, is_udplite = up->pcflag; int corkreq = up->corkflag || msg->msg_flags&MSG_MORE; + int (*getfrag)(void *, char *, int, int, int, struct sk_buff *); if (len > 0xFFFF) return -EMSGSIZE; @@ -622,7 +619,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, { .daddr = faddr, .saddr = saddr, .tos = tos } }, - .proto = IPPROTO_UDP, + .proto = sk->sk_protocol, .uli_u = { .ports = { .sport = inet->sport, .dport = dport } } }; @@ -668,8 +665,9 @@ back_from_confirm: do_append_data: up->len += ulen; - err = ip_append_data(sk, ip_generic_getfrag, msg->msg_iov, ulen, - sizeof(struct udphdr), &ipc, rt, + getfrag = is_udplite ? udplite_getfrag : ip_generic_getfrag; + err = ip_append_data(sk, getfrag, msg->msg_iov, ulen, + sizeof(struct udphdr), &ipc, rt, corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags); if (err) udp_flush_pending_frames(sk); @@ -684,7 +682,7 @@ out: if (free) kfree(ipc.opt); if (!err) { - UDP_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS); + UDP_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS, is_udplite); return len; } /* @@ -695,7 +693,7 @@ out: * seems like overkill. */ if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) { - UDP_INC_STATS_USER(UDP_MIB_SNDBUFERRORS); + UDP_INC_STATS_USER(UDP_MIB_SNDBUFERRORS, is_udplite); } return err; @@ -707,8 +705,8 @@ do_confirm: goto out; } -static int udp_sendpage(struct sock *sk, struct page *page, int offset, - size_t size, int flags) +int udp_sendpage(struct sock *sk, struct page *page, int offset, + size_t size, int flags) { struct udp_sock *up = udp_sk(sk); int ret; @@ -795,29 +793,18 @@ int udp_ioctl(struct sock *sk, int cmd, unsigned long arg) return(0); } -static __inline__ int __udp_checksum_complete(struct sk_buff *skb) -{ - return __skb_checksum_complete(skb); -} - -static __inline__ int udp_checksum_complete(struct sk_buff *skb) -{ - return skb->ip_summed != CHECKSUM_UNNECESSARY && - __udp_checksum_complete(skb); -} - /* * This should be easy, if there is something there we * return it, otherwise we block. */ -static int udp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, - size_t len, int noblock, int flags, int *addr_len) +int udp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, + size_t len, int noblock, int flags, int *addr_len) { struct inet_sock *inet = inet_sk(sk); struct sockaddr_in *sin = (struct sockaddr_in *)msg->msg_name; struct sk_buff *skb; - int copied, err; + int copied, err, copy_only, is_udplite = IS_UDPLITE(sk); /* * Check any passed addresses @@ -839,15 +826,25 @@ try_again: msg->msg_flags |= MSG_TRUNC; } - if (skb->ip_summed==CHECKSUM_UNNECESSARY) { - err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov, - copied); - } else if (msg->msg_flags&MSG_TRUNC) { - if (__udp_checksum_complete(skb)) + /* + * Decide whether to checksum and/or copy data. + * + * UDP: checksum may have been computed in HW, + * (re-)compute it if message is truncated. + * UDP-Lite: always needs to checksum, no HW support. + */ + copy_only = (skb->ip_summed==CHECKSUM_UNNECESSARY); + + if (is_udplite || (!copy_only && msg->msg_flags&MSG_TRUNC)) { + if (__udp_lib_checksum_complete(skb)) goto csum_copy_err; - err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov, - copied); - } else { + copy_only = 1; + } + + if (copy_only) + err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), + msg->msg_iov, copied ); + else { err = skb_copy_and_csum_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov); if (err == -EINVAL) @@ -880,7 +877,7 @@ out: return err; csum_copy_err: - UDP_INC_STATS_BH(UDP_MIB_INERRORS); + UDP_INC_STATS_BH(UDP_MIB_INERRORS, is_udplite); skb_kill_datagram(sk, skb, flags); @@ -912,11 +909,6 @@ int udp_disconnect(struct sock *sk, int flags) return 0; } -static void udp_close(struct sock *sk, long timeout) -{ - sk_common_release(sk); -} - /* return: * 1 if the the UDP system should process it * 0 if we should drop this packet @@ -1022,7 +1014,7 @@ static int udp_encap_rcv(struct sock * sk, struct sk_buff *skb) * Note that in the success and error cases, the skb is assumed to * have either been requeued or freed. */ -static int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) +int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) { struct udp_sock *up = udp_sk(sk); int rc; @@ -1030,10 +1022,8 @@ static int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) /* * Charge it to the socket, dropping if the queue is full. */ - if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb)) { - kfree_skb(skb); - return -1; - } + if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb)) + goto drop; nf_reset(skb); if (up->encap_type) { @@ -1057,31 +1047,68 @@ static int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) if (ret < 0) { /* process the ESP packet */ ret = xfrm4_rcv_encap(skb, up->encap_type); - UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS); + UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS, up->pcflag); return -ret; } /* FALLTHROUGH -- it's a UDP Packet */ } - if (sk->sk_filter && skb->ip_summed != CHECKSUM_UNNECESSARY) { - if (__udp_checksum_complete(skb)) { - UDP_INC_STATS_BH(UDP_MIB_INERRORS); - kfree_skb(skb); - return -1; + /* + * UDP-Lite specific tests, ignored on UDP sockets + */ + if ((up->pcflag & UDPLITE_RECV_CC) && UDP_SKB_CB(skb)->partial_cov) { + + /* + * MIB statistics other than incrementing the error count are + * disabled for the following two types of errors: these depend + * on the application settings, not on the functioning of the + * protocol stack as such. + * + * RFC 3828 here recommends (sec 3.3): "There should also be a + * way ... to ... at least let the receiving application block + * delivery of packets with coverage values less than a value + * provided by the application." + */ + if (up->pcrlen == 0) { /* full coverage was set */ + LIMIT_NETDEBUG(KERN_WARNING "UDPLITE: partial coverage " + "%d while full coverage %d requested\n", + UDP_SKB_CB(skb)->cscov, skb->len); + goto drop; } + /* The next case involves violating the min. coverage requested + * by the receiver. This is subtle: if receiver wants x and x is + * greater than the buffersize/MTU then receiver will complain + * that it wants x while sender emits packets of smaller size y. + * Therefore the above ...()->partial_cov statement is essential. + */ + if (UDP_SKB_CB(skb)->cscov < up->pcrlen) { + LIMIT_NETDEBUG(KERN_WARNING + "UDPLITE: coverage %d too small, need min %d\n", + UDP_SKB_CB(skb)->cscov, up->pcrlen); + goto drop; + } + } + + if (sk->sk_filter && skb->ip_summed != CHECKSUM_UNNECESSARY) { + if (__udp_lib_checksum_complete(skb)) + goto drop; skb->ip_summed = CHECKSUM_UNNECESSARY; } if ((rc = sock_queue_rcv_skb(sk,skb)) < 0) { /* Note that an ENOMEM error is charged twice */ if (rc == -ENOMEM) - UDP_INC_STATS_BH(UDP_MIB_RCVBUFERRORS); - UDP_INC_STATS_BH(UDP_MIB_INERRORS); - kfree_skb(skb); - return -1; + UDP_INC_STATS_BH(UDP_MIB_RCVBUFERRORS, up->pcflag); + goto drop; } - UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS); + + UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS, up->pcflag); return 0; + +drop: + UDP_INC_STATS_BH(UDP_MIB_INERRORS, up->pcflag); + kfree_skb(skb); + return -1; } /* @@ -1090,14 +1117,16 @@ static int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) * Note: called only from the BH handler context, * so we don't need to lock the hashes. */ -static int udp_v4_mcast_deliver(struct sk_buff *skb, struct udphdr *uh, - __be32 saddr, __be32 daddr) +static int __udp4_lib_mcast_deliver(struct sk_buff *skb, + struct udphdr *uh, + __be32 saddr, __be32 daddr, + struct hlist_head udptable[]) { struct sock *sk; int dif; read_lock(&udp_hash_lock); - sk = sk_head(&udp_hash[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]); + sk = sk_head(&udptable[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]); dif = skb->dev->ifindex; sk = udp_v4_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif); if (sk) { @@ -1131,65 +1160,75 @@ static int udp_v4_mcast_deliver(struct sk_buff *skb, struct udphdr *uh, * Otherwise, csum completion requires chacksumming packet body, * including udp header and folding it to skb->csum. */ -static void udp_checksum_init(struct sk_buff *skb, struct udphdr *uh, - unsigned short ulen, __be32 saddr, __be32 daddr) +static inline void udp4_csum_init(struct sk_buff *skb, struct udphdr *uh) { if (uh->check == 0) { skb->ip_summed = CHECKSUM_UNNECESSARY; } else if (skb->ip_summed == CHECKSUM_COMPLETE) { - if (!udp_check(uh, ulen, saddr, daddr, skb->csum)) + if (!csum_tcpudp_magic(skb->nh.iph->saddr, skb->nh.iph->daddr, + skb->len, IPPROTO_UDP, skb->csum )) skb->ip_summed = CHECKSUM_UNNECESSARY; } if (skb->ip_summed != CHECKSUM_UNNECESSARY) - skb->csum = csum_tcpudp_nofold(saddr, daddr, ulen, IPPROTO_UDP, 0); + skb->csum = csum_tcpudp_nofold(skb->nh.iph->saddr, + skb->nh.iph->daddr, + skb->len, IPPROTO_UDP, 0); /* Probably, we should checksum udp header (it should be in cache * in any case) and data in tiny packets (< rx copybreak). */ + + /* UDP = UDP-Lite with a non-partial checksum coverage */ + UDP_SKB_CB(skb)->partial_cov = 0; } /* * All we need to do is get the socket, and then do a checksum. */ -int udp_rcv(struct sk_buff *skb) +int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[], + int is_udplite) { struct sock *sk; - struct udphdr *uh; + struct udphdr *uh = skb->h.uh; unsigned short ulen; struct rtable *rt = (struct rtable*)skb->dst; __be32 saddr = skb->nh.iph->saddr; __be32 daddr = skb->nh.iph->daddr; - int len = skb->len; /* - * Validate the packet and the UDP length. + * Validate the packet. */ if (!pskb_may_pull(skb, sizeof(struct udphdr))) - goto no_header; - - uh = skb->h.uh; + goto drop; /* No space for header. */ ulen = ntohs(uh->len); - - if (ulen > len || ulen < sizeof(*uh)) + if (ulen > skb->len) goto short_packet; - if (pskb_trim_rcsum(skb, ulen)) - goto short_packet; + if(! is_udplite ) { /* UDP validates ulen. */ + + if (ulen < sizeof(*uh) || pskb_trim_rcsum(skb, ulen)) + goto short_packet; - udp_checksum_init(skb, uh, ulen, saddr, daddr); + udp4_csum_init(skb, uh); + + } else { /* UDP-Lite validates cscov. */ + if (udplite4_csum_init(skb, uh)) + goto csum_error; + } if(rt->rt_flags & (RTCF_BROADCAST|RTCF_MULTICAST)) - return udp_v4_mcast_deliver(skb, uh, saddr, daddr); + return __udp4_lib_mcast_deliver(skb, uh, saddr, daddr, udptable); - sk = udp_v4_lookup(saddr, uh->source, daddr, uh->dest, skb->dev->ifindex); + sk = __udp4_lib_lookup(saddr, uh->source, daddr, uh->dest, + skb->dev->ifindex, udptable ); if (sk != NULL) { int ret = udp_queue_rcv_skb(sk, skb); sock_put(sk); /* a return value > 0 means to resubmit the input, but - * it it wants the return to be -protocol, or 0 + * it wants the return to be -protocol, or 0 */ if (ret > 0) return -ret; @@ -1201,10 +1240,10 @@ int udp_rcv(struct sk_buff *skb) nf_reset(skb); /* No socket. Drop packet silently, if checksum is wrong */ - if (udp_checksum_complete(skb)) + if (udp_lib_checksum_complete(skb)) goto csum_error; - UDP_INC_STATS_BH(UDP_MIB_NOPORTS); + UDP_INC_STATS_BH(UDP_MIB_NOPORTS, is_udplite); icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0); /* @@ -1215,36 +1254,40 @@ int udp_rcv(struct sk_buff *skb) return(0); short_packet: - LIMIT_NETDEBUG(KERN_DEBUG "UDP: short packet: From %u.%u.%u.%u:%u %d/%d to %u.%u.%u.%u:%u\n", + LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: short packet: From %u.%u.%u.%u:%u %d/%d to %u.%u.%u.%u:%u\n", + is_udplite? "-Lite" : "", NIPQUAD(saddr), ntohs(uh->source), ulen, - len, + skb->len, NIPQUAD(daddr), ntohs(uh->dest)); -no_header: - UDP_INC_STATS_BH(UDP_MIB_INERRORS); - kfree_skb(skb); - return(0); + goto drop; csum_error: /* * RFC1122: OK. Discards the bad packet silently (as far as * the network is concerned, anyway) as per 4.1.3.4 (MUST). */ - LIMIT_NETDEBUG(KERN_DEBUG "UDP: bad checksum. From %d.%d.%d.%d:%d to %d.%d.%d.%d:%d ulen %d\n", + LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: bad checksum. From %d.%d.%d.%d:%d to %d.%d.%d.%d:%d ulen %d\n", + is_udplite? "-Lite" : "", NIPQUAD(saddr), ntohs(uh->source), NIPQUAD(daddr), ntohs(uh->dest), ulen); drop: - UDP_INC_STATS_BH(UDP_MIB_INERRORS); + UDP_INC_STATS_BH(UDP_MIB_INERRORS, is_udplite); kfree_skb(skb); return(0); } -static int udp_destroy_sock(struct sock *sk) +__inline__ int udp_rcv(struct sk_buff *skb) +{ + return __udp4_lib_rcv(skb, udp_hash, 0); +} + +int udp_destroy_sock(struct sock *sk) { lock_sock(sk); udp_flush_pending_frames(sk); @@ -1293,6 +1336,32 @@ static int do_udp_setsockopt(struct sock *sk, int level, int optname, } break; + /* + * UDP-Lite's partial checksum coverage (RFC 3828). + */ + /* The sender sets actual checksum coverage length via this option. + * The case coverage > packet length is handled by send module. */ + case UDPLITE_SEND_CSCOV: + if (!up->pcflag) /* Disable the option on UDP sockets */ + return -ENOPROTOOPT; + if (val != 0 && val < 8) /* Illegal coverage: use default (8) */ + val = 8; + up->pcslen = val; + up->pcflag |= UDPLITE_SEND_CC; + break; + + /* The receiver specifies a minimum checksum coverage value. To make + * sense, this should be set to at least 8 (as done below). If zero is + * used, this again means full checksum coverage. */ + case UDPLITE_RECV_CSCOV: + if (!up->pcflag) /* Disable the option on UDP sockets */ + return -ENOPROTOOPT; + if (val != 0 && val < 8) /* Avoid silly minimal values. */ + val = 8; + up->pcrlen = val; + up->pcflag |= UDPLITE_RECV_CC; + break; + default: err = -ENOPROTOOPT; break; @@ -1301,21 +1370,21 @@ static int do_udp_setsockopt(struct sock *sk, int level, int optname, return err; } -static int udp_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, int optlen) +int udp_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen) { - if (level != SOL_UDP) - return ip_setsockopt(sk, level, optname, optval, optlen); - return do_udp_setsockopt(sk, level, optname, optval, optlen); + if (level == SOL_UDP || level == SOL_UDPLITE) + return do_udp_setsockopt(sk, level, optname, optval, optlen); + return ip_setsockopt(sk, level, optname, optval, optlen); } #ifdef CONFIG_COMPAT -static int compat_udp_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, int optlen) +int compat_udp_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen) { - if (level != SOL_UDP) - return compat_ip_setsockopt(sk, level, optname, optval, optlen); - return do_udp_setsockopt(sk, level, optname, optval, optlen); + if (level == SOL_UDP || level == SOL_UDPLITE) + return do_udp_setsockopt(sk, level, optname, optval, optlen); + return compat_ip_setsockopt(sk, level, optname, optval, optlen); } #endif @@ -1342,6 +1411,16 @@ static int do_udp_getsockopt(struct sock *sk, int level, int optname, val = up->encap_type; break; + /* The following two cannot be changed on UDP sockets, the return is + * always 0 (which corresponds to the full checksum coverage of UDP). */ + case UDPLITE_SEND_CSCOV: + val = up->pcslen; + break; + + case UDPLITE_RECV_CSCOV: + val = up->pcrlen; + break; + default: return -ENOPROTOOPT; }; @@ -1353,21 +1432,21 @@ static int do_udp_getsockopt(struct sock *sk, int level, int optname, return 0; } -static int udp_getsockopt(struct sock *sk, int level, int optname, - char __user *optval, int __user *optlen) +int udp_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen) { - if (level != SOL_UDP) - return ip_getsockopt(sk, level, optname, optval, optlen); - return do_udp_getsockopt(sk, level, optname, optval, optlen); + if (level == SOL_UDP || level == SOL_UDPLITE) + return do_udp_getsockopt(sk, level, optname, optval, optlen); + return ip_getsockopt(sk, level, optname, optval, optlen); } #ifdef CONFIG_COMPAT -static int compat_udp_getsockopt(struct sock *sk, int level, int optname, +int compat_udp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { - if (level != SOL_UDP) - return compat_ip_getsockopt(sk, level, optname, optval, optlen); - return do_udp_getsockopt(sk, level, optname, optval, optlen); + if (level == SOL_UDP || level == SOL_UDPLITE) + return do_udp_getsockopt(sk, level, optname, optval, optlen); + return compat_ip_getsockopt(sk, level, optname, optval, optlen); } #endif /** @@ -1387,7 +1466,8 @@ unsigned int udp_poll(struct file *file, struct socket *sock, poll_table *wait) { unsigned int mask = datagram_poll(file, sock, wait); struct sock *sk = sock->sk; - + int is_lite = IS_UDPLITE(sk); + /* Check for false positives due to checksum errors */ if ( (mask & POLLRDNORM) && !(file->f_flags & O_NONBLOCK) && @@ -1397,8 +1477,8 @@ unsigned int udp_poll(struct file *file, struct socket *sock, poll_table *wait) spin_lock_bh(&rcvq->lock); while ((skb = skb_peek(rcvq)) != NULL) { - if (udp_checksum_complete(skb)) { - UDP_INC_STATS_BH(UDP_MIB_INERRORS); + if (udp_lib_checksum_complete(skb)) { + UDP_INC_STATS_BH(UDP_MIB_INERRORS, is_lite); __skb_unlink(skb, rcvq); kfree_skb(skb); } else { @@ -1420,7 +1500,7 @@ unsigned int udp_poll(struct file *file, struct socket *sock, poll_table *wait) struct proto udp_prot = { .name = "UDP", .owner = THIS_MODULE, - .close = udp_close, + .close = udp_lib_close, .connect = ip4_datagram_connect, .disconnect = udp_disconnect, .ioctl = udp_ioctl, @@ -1431,8 +1511,8 @@ struct proto udp_prot = { .recvmsg = udp_recvmsg, .sendpage = udp_sendpage, .backlog_rcv = udp_queue_rcv_skb, - .hash = udp_v4_hash, - .unhash = udp_v4_unhash, + .hash = udp_lib_hash, + .unhash = udp_lib_unhash, .get_port = udp_v4_get_port, .obj_size = sizeof(struct udp_sock), #ifdef CONFIG_COMPAT @@ -1451,7 +1531,7 @@ static struct sock *udp_get_first(struct seq_file *seq) for (state->bucket = 0; state->bucket < UDP_HTABLE_SIZE; ++state->bucket) { struct hlist_node *node; - sk_for_each(sk, node, &udp_hash[state->bucket]) { + sk_for_each(sk, node, state->hashtable + state->bucket) { if (sk->sk_family == state->family) goto found; } @@ -1472,7 +1552,7 @@ try_again: } while (sk && sk->sk_family != state->family); if (!sk && ++state->bucket < UDP_HTABLE_SIZE) { - sk = sk_head(&udp_hash[state->bucket]); + sk = sk_head(state->hashtable + state->bucket); goto try_again; } return sk; @@ -1522,6 +1602,7 @@ static int udp_seq_open(struct inode *inode, struct file *file) if (!s) goto out; s->family = afinfo->family; + s->hashtable = afinfo->hashtable; s->seq_ops.start = udp_seq_start; s->seq_ops.next = udp_seq_next; s->seq_ops.show = afinfo->seq_show; @@ -1588,7 +1669,7 @@ static void udp4_format_sock(struct sock *sp, char *tmpbuf, int bucket) atomic_read(&sp->sk_refcnt), sp); } -static int udp4_seq_show(struct seq_file *seq, void *v) +int udp4_seq_show(struct seq_file *seq, void *v) { if (v == SEQ_START_TOKEN) seq_printf(seq, "%-127s\n", @@ -1611,6 +1692,7 @@ static struct udp_seq_afinfo udp4_seq_afinfo = { .owner = THIS_MODULE, .name = "udp", .family = AF_INET, + .hashtable = udp_hash, .seq_show = udp4_seq_show, .seq_fops = &udp4_seq_fops, }; diff --git a/net/ipv4/udp_impl.h b/net/ipv4/udp_impl.h new file mode 100644 index 00000000000..f6f4277ba6d --- /dev/null +++ b/net/ipv4/udp_impl.h @@ -0,0 +1,38 @@ +#ifndef _UDP4_IMPL_H +#define _UDP4_IMPL_H +#include +#include +#include +#include + +extern int __udp4_lib_rcv(struct sk_buff *, struct hlist_head [], int ); +extern void __udp4_lib_err(struct sk_buff *, u32, struct hlist_head []); + +extern int __udp_lib_get_port(struct sock *sk, unsigned short snum, + struct hlist_head udptable[], int *port_rover, + int (*)(const struct sock*,const struct sock*)); +extern int ipv4_rcv_saddr_equal(const struct sock *, const struct sock *); + + +extern int udp_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen); +extern int udp_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen); + +#ifdef CONFIG_COMPAT +extern int compat_udp_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen); +extern int compat_udp_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen); +#endif +extern int udp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, + size_t len, int noblock, int flags, int *addr_len); +extern int udp_sendpage(struct sock *sk, struct page *page, int offset, + size_t size, int flags); +extern int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb); +extern int udp_destroy_sock(struct sock *sk); + +#ifdef CONFIG_PROC_FS +extern int udp4_seq_show(struct seq_file *seq, void *v); +#endif +#endif /* _UDP4_IMPL_H */ diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c new file mode 100644 index 00000000000..561de6d8c73 --- /dev/null +++ b/net/ipv4/udplite.c @@ -0,0 +1,119 @@ +/* + * UDPLITE An implementation of the UDP-Lite protocol (RFC 3828). + * + * Version: $Id: udplite.c,v 1.25 2006/10/19 07:22:36 gerrit Exp $ + * + * Authors: Gerrit Renker + * + * Changes: + * Fixes: + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ +#include "udp_impl.h" +DEFINE_SNMP_STAT(struct udp_mib, udplite_statistics) __read_mostly; + +struct hlist_head udplite_hash[UDP_HTABLE_SIZE]; +static int udplite_port_rover; + +__inline__ int udplite_get_port(struct sock *sk, unsigned short p, + int (*c)(const struct sock *, const struct sock *)) +{ + return __udp_lib_get_port(sk, p, udplite_hash, &udplite_port_rover, c); +} + +static __inline__ int udplite_v4_get_port(struct sock *sk, unsigned short snum) +{ + return udplite_get_port(sk, snum, ipv4_rcv_saddr_equal); +} + +__inline__ int udplite_rcv(struct sk_buff *skb) +{ + return __udp4_lib_rcv(skb, udplite_hash, 1); +} + +__inline__ void udplite_err(struct sk_buff *skb, u32 info) +{ + return __udp4_lib_err(skb, info, udplite_hash); +} + +static struct net_protocol udplite_protocol = { + .handler = udplite_rcv, + .err_handler = udplite_err, + .no_policy = 1, +}; + +struct proto udplite_prot = { + .name = "UDP-Lite", + .owner = THIS_MODULE, + .close = udp_lib_close, + .connect = ip4_datagram_connect, + .disconnect = udp_disconnect, + .ioctl = udp_ioctl, + .init = udplite_sk_init, + .destroy = udp_destroy_sock, + .setsockopt = udp_setsockopt, + .getsockopt = udp_getsockopt, + .sendmsg = udp_sendmsg, + .recvmsg = udp_recvmsg, + .sendpage = udp_sendpage, + .backlog_rcv = udp_queue_rcv_skb, + .hash = udp_lib_hash, + .unhash = udp_lib_unhash, + .get_port = udplite_v4_get_port, + .obj_size = sizeof(struct udp_sock), +#ifdef CONFIG_COMPAT + .compat_setsockopt = compat_udp_setsockopt, + .compat_getsockopt = compat_udp_getsockopt, +#endif +}; + +static struct inet_protosw udplite4_protosw = { + .type = SOCK_DGRAM, + .protocol = IPPROTO_UDPLITE, + .prot = &udplite_prot, + .ops = &inet_dgram_ops, + .capability = -1, + .no_check = 0, /* must checksum (RFC 3828) */ + .flags = INET_PROTOSW_PERMANENT, +}; + +#ifdef CONFIG_PROC_FS +static struct file_operations udplite4_seq_fops; +static struct udp_seq_afinfo udplite4_seq_afinfo = { + .owner = THIS_MODULE, + .name = "udplite", + .family = AF_INET, + .hashtable = udplite_hash, + .seq_show = udp4_seq_show, + .seq_fops = &udplite4_seq_fops, +}; +#endif + +void __init udplite4_register(void) +{ + if (proto_register(&udplite_prot, 1)) + goto out_register_err; + + if (inet_add_protocol(&udplite_protocol, IPPROTO_UDPLITE) < 0) + goto out_unregister_proto; + + inet_register_protosw(&udplite4_protosw); + +#ifdef CONFIG_PROC_FS + if (udp_proc_register(&udplite4_seq_afinfo)) /* udplite4_proc_init() */ + printk(KERN_ERR "%s: Cannot register /proc!\n", __FUNCTION__); +#endif + return; + +out_unregister_proto: + proto_unregister(&udplite_prot); +out_register_err: + printk(KERN_CRIT "%s: Cannot add UDP-Lite protocol.\n", __FUNCTION__); +} + +EXPORT_SYMBOL(udplite_hash); +EXPORT_SYMBOL(udplite_prot); +EXPORT_SYMBOL(udplite_get_port); diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index da766234607..d4107bb701b 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -199,6 +199,7 @@ _decode_session4(struct sk_buff *skb, struct flowi *fl) if (!(iph->frag_off & htons(IP_MF | IP_OFFSET))) { switch (iph->protocol) { case IPPROTO_UDP: + case IPPROTO_UDPLITE: case IPPROTO_TCP: case IPPROTO_SCTP: case IPPROTO_DCCP: diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile index addcc011bc0..8bacda109b7 100644 --- a/net/ipv6/Makefile +++ b/net/ipv6/Makefile @@ -5,8 +5,8 @@ obj-$(CONFIG_IPV6) += ipv6.o ipv6-objs := af_inet6.o anycast.o ip6_output.o ip6_input.o addrconf.o \ - route.o ip6_fib.o ipv6_sockglue.o ndisc.o udp.o raw.o \ - protocol.o icmp.o mcast.o reassembly.o tcp_ipv6.o \ + route.o ip6_fib.o ipv6_sockglue.o ndisc.o udp.o udplite.o \ + raw.o protocol.o icmp.o mcast.o reassembly.o tcp_ipv6.o \ exthdrs.o sysctl_net_ipv6.o datagram.o proc.o \ ip6_flowlabel.o ipv6_syms.o inet6_connection_sock.o diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 92bfccf62cb..1eb1c7f261d 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -49,6 +49,7 @@ #include #include #include +#include #include #include #include @@ -737,8 +738,13 @@ static int __init init_ipv6_mibs(void) if (snmp6_mib_init((void **)udp_stats_in6, sizeof (struct udp_mib), __alignof__(struct udp_mib)) < 0) goto err_udp_mib; + if (snmp6_mib_init((void **)udplite_stats_in6, sizeof (struct udp_mib), + __alignof__(struct udp_mib)) < 0) + goto err_udplite_mib; return 0; +err_udplite_mib: + snmp6_mib_free((void **)udp_stats_in6); err_udp_mib: snmp6_mib_free((void **)icmpv6_statistics); err_icmp_mib: @@ -753,6 +759,7 @@ static void cleanup_ipv6_mibs(void) snmp6_mib_free((void **)ipv6_statistics); snmp6_mib_free((void **)icmpv6_statistics); snmp6_mib_free((void **)udp_stats_in6); + snmp6_mib_free((void **)udplite_stats_in6); } static int __init inet6_init(void) @@ -780,10 +787,14 @@ static int __init inet6_init(void) if (err) goto out_unregister_tcp_proto; - err = proto_register(&rawv6_prot, 1); + err = proto_register(&udplitev6_prot, 1); if (err) goto out_unregister_udp_proto; + err = proto_register(&rawv6_prot, 1); + if (err) + goto out_unregister_udplite_proto; + /* Register the socket-side information for inet6_create. */ for(r = &inetsw6[0]; r < &inetsw6[SOCK_MAX]; ++r) @@ -837,6 +848,8 @@ static int __init inet6_init(void) goto proc_tcp6_fail; if (udp6_proc_init()) goto proc_udp6_fail; + if (udplite6_proc_init()) + goto proc_udplite6_fail; if (ipv6_misc_proc_init()) goto proc_misc6_fail; @@ -862,6 +875,7 @@ static int __init inet6_init(void) /* Init v6 transport protocols. */ udpv6_init(); + udplitev6_init(); tcpv6_init(); ipv6_packet_init(); @@ -879,6 +893,8 @@ proc_if6_fail: proc_anycast6_fail: ipv6_misc_proc_exit(); proc_misc6_fail: + udplite6_proc_exit(); +proc_udplite6_fail: udp6_proc_exit(); proc_udp6_fail: tcp6_proc_exit(); @@ -902,6 +918,8 @@ out_unregister_sock: sock_unregister(PF_INET6); out_unregister_raw_proto: proto_unregister(&rawv6_prot); +out_unregister_udplite_proto: + proto_unregister(&udplitev6_prot); out_unregister_udp_proto: proto_unregister(&udpv6_prot); out_unregister_tcp_proto: @@ -919,6 +937,7 @@ static void __exit inet6_exit(void) ac6_proc_exit(); ipv6_misc_proc_exit(); udp6_proc_exit(); + udplite6_proc_exit(); tcp6_proc_exit(); raw6_proc_exit(); #endif diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index de6b91981b3..1eafcfc95e8 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -51,6 +51,7 @@ #include #include #include +#include #include #include @@ -239,6 +240,7 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname, struct sk_buff *pktopt; if (sk->sk_protocol != IPPROTO_UDP && + sk->sk_protocol != IPPROTO_UDPLITE && sk->sk_protocol != IPPROTO_TCP) break; @@ -276,11 +278,15 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname, sk->sk_family = PF_INET; tcp_sync_mss(sk, icsk->icsk_pmtu_cookie); } else { + struct proto *prot = &udp_prot; + + if (sk->sk_protocol == IPPROTO_UDPLITE) + prot = &udplite_prot; local_bh_disable(); sock_prot_dec_use(sk->sk_prot); - sock_prot_inc_use(&udp_prot); + sock_prot_inc_use(prot); local_bh_enable(); - sk->sk_prot = &udp_prot; + sk->sk_prot = prot; sk->sk_socket->ops = &inet_dgram_ops; sk->sk_family = PF_INET; } @@ -813,6 +819,7 @@ static int do_ipv6_getsockopt(struct sock *sk, int level, int optname, switch (optname) { case IPV6_ADDRFORM: if (sk->sk_protocol != IPPROTO_UDP && + sk->sk_protocol != IPPROTO_UDPLITE && sk->sk_protocol != IPPROTO_TCP) return -EINVAL; if (sk->sk_state != TCP_ESTABLISHED) diff --git a/net/ipv6/netfilter/ip6t_LOG.c b/net/ipv6/netfilter/ip6t_LOG.c index 0cf537d3018..3cb6bb79cc0 100644 --- a/net/ipv6/netfilter/ip6t_LOG.c +++ b/net/ipv6/netfilter/ip6t_LOG.c @@ -270,11 +270,15 @@ static void dump_packet(const struct nf_loginfo *info, } break; } - case IPPROTO_UDP: { + case IPPROTO_UDP: + case IPPROTO_UDPLITE: { struct udphdr _udph, *uh; - /* Max length: 10 "PROTO=UDP " */ - printk("PROTO=UDP "); + if (currenthdr == IPPROTO_UDP) + /* Max length: 10 "PROTO=UDP " */ + printk("PROTO=UDP " ); + else /* Max length: 14 "PROTO=UDPLITE " */ + printk("PROTO=UDPLITE "); if (fragment) break; diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c index 4158d386b0a..35249d8487b 100644 --- a/net/ipv6/proc.c +++ b/net/ipv6/proc.c @@ -49,6 +49,8 @@ static int sockstat6_seq_show(struct seq_file *seq, void *v) fold_prot_inuse(&tcpv6_prot)); seq_printf(seq, "UDP6: inuse %d\n", fold_prot_inuse(&udpv6_prot)); + seq_printf(seq, "UDPLITE6: inuse %d\n", + fold_prot_inuse(&udplitev6_prot)); seq_printf(seq, "RAW6: inuse %d\n", fold_prot_inuse(&rawv6_prot)); seq_printf(seq, "FRAG6: inuse %d memory %d\n", @@ -133,6 +135,14 @@ static struct snmp_mib snmp6_udp6_list[] = { SNMP_MIB_SENTINEL }; +static struct snmp_mib snmp6_udplite6_list[] = { + SNMP_MIB_ITEM("UdpLite6InDatagrams", UDP_MIB_INDATAGRAMS), + SNMP_MIB_ITEM("UdpLite6NoPorts", UDP_MIB_NOPORTS), + SNMP_MIB_ITEM("UdpLite6InErrors", UDP_MIB_INERRORS), + SNMP_MIB_ITEM("UdpLite6OutDatagrams", UDP_MIB_OUTDATAGRAMS), + SNMP_MIB_SENTINEL +}; + static unsigned long fold_field(void *mib[], int offt) { @@ -167,6 +177,7 @@ static int snmp6_seq_show(struct seq_file *seq, void *v) snmp6_seq_show_item(seq, (void **)ipv6_statistics, snmp6_ipstats_list); snmp6_seq_show_item(seq, (void **)icmpv6_statistics, snmp6_icmp6_list); snmp6_seq_show_item(seq, (void **)udp_stats_in6, snmp6_udp6_list); + snmp6_seq_show_item(seq, (void **)udplite_stats_in6, snmp6_udplite6_list); } return 0; } diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 886300d13a5..5a64027bf2f 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -38,26 +38,18 @@ #include #include -#include -#include - -#include #include #include #include #include -#include -#include -#include #include -#include #include - #include #include #include #include +#include "udp_impl.h" DEFINE_SNMP_STAT(struct udp_mib, udp_stats_in6) __read_mostly; @@ -66,23 +58,9 @@ static inline int udp_v6_get_port(struct sock *sk, unsigned short snum) return udp_get_port(sk, snum, ipv6_rcv_saddr_equal); } -static void udp_v6_hash(struct sock *sk) -{ - BUG(); -} - -static void udp_v6_unhash(struct sock *sk) -{ - write_lock_bh(&udp_hash_lock); - if (sk_del_node_init(sk)) { - inet_sk(sk)->num = 0; - sock_prot_dec_use(sk->sk_prot); - } - write_unlock_bh(&udp_hash_lock); -} - -static struct sock *udp_v6_lookup(struct in6_addr *saddr, u16 sport, - struct in6_addr *daddr, u16 dport, int dif) +static struct sock *__udp6_lib_lookup(struct in6_addr *saddr, __be16 sport, + struct in6_addr *daddr, __be16 dport, + int dif, struct hlist_head udptable[]) { struct sock *sk, *result = NULL; struct hlist_node *node; @@ -90,7 +68,7 @@ static struct sock *udp_v6_lookup(struct in6_addr *saddr, u16 sport, int badness = -1; read_lock(&udp_hash_lock); - sk_for_each(sk, node, &udp_hash[hnum & (UDP_HTABLE_SIZE - 1)]) { + sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) { struct inet_sock *inet = inet_sk(sk); if (inet->num == hnum && sk->sk_family == PF_INET6) { @@ -131,21 +109,12 @@ static struct sock *udp_v6_lookup(struct in6_addr *saddr, u16 sport, return result; } -/* - * - */ - -static void udpv6_close(struct sock *sk, long timeout) -{ - sk_common_release(sk); -} - /* * This should be easy, if there is something there we * return it, otherwise we block. */ -static int udpv6_recvmsg(struct kiocb *iocb, struct sock *sk, +int udpv6_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, size_t len, int noblock, int flags, int *addr_len) { @@ -153,7 +122,7 @@ static int udpv6_recvmsg(struct kiocb *iocb, struct sock *sk, struct inet_sock *inet = inet_sk(sk); struct sk_buff *skb; size_t copied; - int err; + int err, copy_only, is_udplite = IS_UDPLITE(sk); if (addr_len) *addr_len=sizeof(struct sockaddr_in6); @@ -172,15 +141,21 @@ try_again: msg->msg_flags |= MSG_TRUNC; } - if (skb->ip_summed==CHECKSUM_UNNECESSARY) { - err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov, - copied); - } else if (msg->msg_flags&MSG_TRUNC) { - if (__skb_checksum_complete(skb)) + /* + * Decide whether to checksum and/or copy data. + */ + copy_only = (skb->ip_summed==CHECKSUM_UNNECESSARY); + + if (is_udplite || (!copy_only && msg->msg_flags&MSG_TRUNC)) { + if (__udp_lib_checksum_complete(skb)) goto csum_copy_err; - err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov, - copied); - } else { + copy_only = 1; + } + + if (copy_only) + err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), + msg->msg_iov, copied ); + else { err = skb_copy_and_csum_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov); if (err == -EINVAL) goto csum_copy_err; @@ -231,14 +206,15 @@ csum_copy_err: skb_kill_datagram(sk, skb, flags); if (flags & MSG_DONTWAIT) { - UDP6_INC_STATS_USER(UDP_MIB_INERRORS); + UDP6_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite); return -EAGAIN; } goto try_again; } -static void udpv6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, - int type, int code, int offset, __be32 info) +void __udp6_lib_err(struct sk_buff *skb, struct inet6_skb_parm *opt, + int type, int code, int offset, __be32 info, + struct hlist_head udptable[] ) { struct ipv6_pinfo *np; struct ipv6hdr *hdr = (struct ipv6hdr*)skb->data; @@ -248,8 +224,8 @@ static void udpv6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, struct sock *sk; int err; - sk = udp_v6_lookup(daddr, uh->dest, saddr, uh->source, inet6_iif(skb)); - + sk = __udp6_lib_lookup(daddr, uh->dest, + saddr, uh->source, inet6_iif(skb), udptable); if (sk == NULL) return; @@ -270,31 +246,55 @@ out: sock_put(sk); } -static inline int udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) +static __inline__ void udpv6_err(struct sk_buff *skb, + struct inet6_skb_parm *opt, int type, + int code, int offset, __u32 info ) +{ + return __udp6_lib_err(skb, opt, type, code, offset, info, udp_hash); +} + +int udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) { + struct udp_sock *up = udp_sk(sk); int rc; - if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb)) { - kfree_skb(skb); - return -1; - } + if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb)) + goto drop; - if (skb_checksum_complete(skb)) { - UDP6_INC_STATS_BH(UDP_MIB_INERRORS); - kfree_skb(skb); - return 0; + /* + * UDP-Lite specific tests, ignored on UDP sockets (see net/ipv4/udp.c). + */ + if ((up->pcflag & UDPLITE_RECV_CC) && UDP_SKB_CB(skb)->partial_cov) { + + if (up->pcrlen == 0) { /* full coverage was set */ + LIMIT_NETDEBUG(KERN_WARNING "UDPLITE6: partial coverage" + " %d while full coverage %d requested\n", + UDP_SKB_CB(skb)->cscov, skb->len); + goto drop; + } + if (UDP_SKB_CB(skb)->cscov < up->pcrlen) { + LIMIT_NETDEBUG(KERN_WARNING "UDPLITE6: coverage %d " + "too small, need min %d\n", + UDP_SKB_CB(skb)->cscov, up->pcrlen); + goto drop; + } } + if (udp_lib_checksum_complete(skb)) + goto drop; + if ((rc = sock_queue_rcv_skb(sk,skb)) < 0) { /* Note that an ENOMEM error is charged twice */ if (rc == -ENOMEM) - UDP6_INC_STATS_BH(UDP_MIB_RCVBUFERRORS); - UDP6_INC_STATS_BH(UDP_MIB_INERRORS); - kfree_skb(skb); - return 0; + UDP6_INC_STATS_BH(UDP_MIB_RCVBUFERRORS, up->pcflag); + goto drop; } - UDP6_INC_STATS_BH(UDP_MIB_INDATAGRAMS); + UDP6_INC_STATS_BH(UDP_MIB_INDATAGRAMS, up->pcflag); return 0; +drop: + UDP6_INC_STATS_BH(UDP_MIB_INERRORS, up->pcflag); + kfree_skb(skb); + return -1; } static struct sock *udp_v6_mcast_next(struct sock *sk, @@ -338,15 +338,15 @@ static struct sock *udp_v6_mcast_next(struct sock *sk, * Note: called only from the BH handler context, * so we don't need to lock the hashes. */ -static void udpv6_mcast_deliver(struct udphdr *uh, - struct in6_addr *saddr, struct in6_addr *daddr, - struct sk_buff *skb) +static int __udp6_lib_mcast_deliver(struct sk_buff *skb, struct in6_addr *saddr, + struct in6_addr *daddr, struct hlist_head udptable[]) { struct sock *sk, *sk2; + const struct udphdr *uh = skb->h.uh; int dif; read_lock(&udp_hash_lock); - sk = sk_head(&udp_hash[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]); + sk = sk_head(&udptable[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]); dif = inet6_iif(skb); sk = udp_v6_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif); if (!sk) { @@ -364,9 +364,34 @@ static void udpv6_mcast_deliver(struct udphdr *uh, udpv6_queue_rcv_skb(sk, skb); out: read_unlock(&udp_hash_lock); + return 0; } -static int udpv6_rcv(struct sk_buff **pskb) +static inline int udp6_csum_init(struct sk_buff *skb, struct udphdr *uh) + +{ + if (uh->check == 0) { + /* RFC 2460 section 8.1 says that we SHOULD log + this error. Well, it is reasonable. + */ + LIMIT_NETDEBUG(KERN_INFO "IPv6: udp checksum is 0\n"); + return 1; + } + if (skb->ip_summed == CHECKSUM_COMPLETE && + !csum_ipv6_magic(&skb->nh.ipv6h->saddr, &skb->nh.ipv6h->daddr, + skb->len, IPPROTO_UDP, skb->csum )) + skb->ip_summed = CHECKSUM_UNNECESSARY; + + if (skb->ip_summed != CHECKSUM_UNNECESSARY) + skb->csum = ~csum_ipv6_magic(&skb->nh.ipv6h->saddr, + &skb->nh.ipv6h->daddr, + skb->len, IPPROTO_UDP, 0); + + return (UDP_SKB_CB(skb)->partial_cov = 0); +} + +int __udp6_lib_rcv(struct sk_buff **pskb, struct hlist_head udptable[], + int is_udplite) { struct sk_buff *skb = *pskb; struct sock *sk; @@ -383,44 +408,39 @@ static int udpv6_rcv(struct sk_buff **pskb) uh = skb->h.uh; ulen = ntohs(uh->len); + if (ulen > skb->len) + goto short_packet; - /* Check for jumbo payload */ - if (ulen == 0) - ulen = skb->len; + if(! is_udplite ) { /* UDP validates ulen. */ - if (ulen > skb->len || ulen < sizeof(*uh)) - goto short_packet; + /* Check for jumbo payload */ + if (ulen == 0) + ulen = skb->len; - if (uh->check == 0) { - /* RFC 2460 section 8.1 says that we SHOULD log - this error. Well, it is reasonable. - */ - LIMIT_NETDEBUG(KERN_INFO "IPv6: udp checksum is 0\n"); - goto discard; - } + if (ulen < sizeof(*uh)) + goto short_packet; - if (ulen < skb->len) { - if (pskb_trim_rcsum(skb, ulen)) - goto discard; - saddr = &skb->nh.ipv6h->saddr; - daddr = &skb->nh.ipv6h->daddr; - uh = skb->h.uh; - } + if (ulen < skb->len) { + if (pskb_trim_rcsum(skb, ulen)) + goto short_packet; + saddr = &skb->nh.ipv6h->saddr; + daddr = &skb->nh.ipv6h->daddr; + uh = skb->h.uh; + } - if (skb->ip_summed == CHECKSUM_COMPLETE && - !csum_ipv6_magic(saddr, daddr, ulen, IPPROTO_UDP, skb->csum)) - skb->ip_summed = CHECKSUM_UNNECESSARY; + if (udp6_csum_init(skb, uh)) + goto discard; - if (skb->ip_summed != CHECKSUM_UNNECESSARY) - skb->csum = ~csum_ipv6_magic(saddr, daddr, ulen, IPPROTO_UDP, 0); + } else { /* UDP-Lite validates cscov. */ + if (udplite6_csum_init(skb, uh)) + goto discard; + } /* * Multicast receive code */ - if (ipv6_addr_is_multicast(daddr)) { - udpv6_mcast_deliver(uh, saddr, daddr, skb); - return 0; - } + if (ipv6_addr_is_multicast(daddr)) + return __udp6_lib_mcast_deliver(skb, saddr, daddr, udptable); /* Unicast */ @@ -428,15 +448,16 @@ static int udpv6_rcv(struct sk_buff **pskb) * check socket cache ... must talk to Alan about his plans * for sock caches... i'll skip this for now. */ - sk = udp_v6_lookup(saddr, uh->source, daddr, uh->dest, inet6_iif(skb)); + sk = __udp6_lib_lookup(saddr, uh->source, + daddr, uh->dest, inet6_iif(skb), udptable); if (sk == NULL) { if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) goto discard; - if (skb_checksum_complete(skb)) + if (udp_lib_checksum_complete(skb)) goto discard; - UDP6_INC_STATS_BH(UDP_MIB_NOPORTS); + UDP6_INC_STATS_BH(UDP_MIB_NOPORTS, is_udplite); icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0, dev); @@ -451,14 +472,20 @@ static int udpv6_rcv(struct sk_buff **pskb) return(0); short_packet: - if (net_ratelimit()) - printk(KERN_DEBUG "UDP: short packet: %d/%u\n", ulen, skb->len); + LIMIT_NETDEBUG(KERN_DEBUG "UDP%sv6: short packet: %d/%u\n", + is_udplite? "-Lite" : "", ulen, skb->len); discard: - UDP6_INC_STATS_BH(UDP_MIB_INERRORS); + UDP6_INC_STATS_BH(UDP_MIB_INERRORS, is_udplite); kfree_skb(skb); return(0); } + +static __inline__ int udpv6_rcv(struct sk_buff **pskb) +{ + return __udp6_lib_rcv(pskb, udp_hash, 0); +} + /* * Throw away all pending data and cancel the corking. Socket is locked. */ @@ -484,6 +511,7 @@ static int udp_v6_push_pending_frames(struct sock *sk, struct udp_sock *up) struct inet_sock *inet = inet_sk(sk); struct flowi *fl = &inet->cork.fl; int err = 0; + u32 csum = 0; /* Grab the skbuff where UDP header space exists. */ if ((skb = skb_peek(&sk->sk_write_queue)) == NULL) @@ -498,35 +526,17 @@ static int udp_v6_push_pending_frames(struct sock *sk, struct udp_sock *up) uh->len = htons(up->len); uh->check = 0; - if (sk->sk_no_check == UDP_CSUM_NOXMIT) { - skb->ip_summed = CHECKSUM_NONE; - goto send; - } - - if (skb_queue_len(&sk->sk_write_queue) == 1) { - skb->csum = csum_partial((char *)uh, - sizeof(struct udphdr), skb->csum); - uh->check = csum_ipv6_magic(&fl->fl6_src, - &fl->fl6_dst, - up->len, fl->proto, skb->csum); - } else { - u32 tmp_csum = 0; - - skb_queue_walk(&sk->sk_write_queue, skb) { - tmp_csum = csum_add(tmp_csum, skb->csum); - } - tmp_csum = csum_partial((char *)uh, - sizeof(struct udphdr), tmp_csum); - tmp_csum = csum_ipv6_magic(&fl->fl6_src, - &fl->fl6_dst, - up->len, fl->proto, tmp_csum); - uh->check = tmp_csum; + if (up->pcflag) + csum = udplite_csum_outgoing(sk, skb); + else + csum = udp_csum_outgoing(sk, skb); - } + /* add protocol-dependent pseudo-header */ + uh->check = csum_ipv6_magic(&fl->fl6_src, &fl->fl6_dst, + up->len, fl->proto, csum ); if (uh->check == 0) uh->check = -1; -send: err = ip6_push_pending_frames(sk); out: up->len = 0; @@ -534,7 +544,7 @@ out: return err; } -static int udpv6_sendmsg(struct kiocb *iocb, struct sock *sk, +int udpv6_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, size_t len) { struct ipv6_txoptions opt_space; @@ -554,6 +564,8 @@ static int udpv6_sendmsg(struct kiocb *iocb, struct sock *sk, int corkreq = up->corkflag || msg->msg_flags&MSG_MORE; int err; int connected = 0; + int is_udplite = up->pcflag; + int (*getfrag)(void *, char *, int, int, int, struct sk_buff *); /* destination address check */ if (sin6) { @@ -694,7 +706,7 @@ do_udp_sendmsg: opt = fl6_merge_options(&opt_space, flowlabel, opt); opt = ipv6_fixup_options(&opt_space, opt); - fl.proto = IPPROTO_UDP; + fl.proto = sk->sk_protocol; ipv6_addr_copy(&fl.fl6_dst, daddr); if (ipv6_addr_any(&fl.fl6_src) && !ipv6_addr_any(&np->saddr)) ipv6_addr_copy(&fl.fl6_src, &np->saddr); @@ -761,7 +773,8 @@ back_from_confirm: do_append_data: up->len += ulen; - err = ip6_append_data(sk, ip_generic_getfrag, msg->msg_iov, ulen, + getfrag = is_udplite ? udplite_getfrag : ip_generic_getfrag; + err = ip6_append_data(sk, getfrag, msg->msg_iov, ulen, sizeof(struct udphdr), hlimit, tclass, opt, &fl, (struct rt6_info*)dst, corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags); @@ -793,7 +806,7 @@ do_append_data: out: fl6_sock_release(flowlabel); if (!err) { - UDP6_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS); + UDP6_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS, is_udplite); return len; } /* @@ -804,7 +817,7 @@ out: * seems like overkill. */ if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) { - UDP6_INC_STATS_USER(UDP_MIB_SNDBUFERRORS); + UDP6_INC_STATS_USER(UDP_MIB_SNDBUFERRORS, is_udplite); } return err; @@ -816,7 +829,7 @@ do_confirm: goto out; } -static int udpv6_destroy_sock(struct sock *sk) +int udpv6_destroy_sock(struct sock *sk) { lock_sock(sk); udp_v6_flush_pending_frames(sk); @@ -854,7 +867,6 @@ static int do_udpv6_setsockopt(struct sock *sk, int level, int optname, release_sock(sk); } break; - case UDP_ENCAP: switch (val) { case 0: @@ -866,6 +878,24 @@ static int do_udpv6_setsockopt(struct sock *sk, int level, int optname, } break; + case UDPLITE_SEND_CSCOV: + if (!up->pcflag) /* Disable the option on UDP sockets */ + return -ENOPROTOOPT; + if (val != 0 && val < 8) /* Illegal coverage: use default (8) */ + val = 8; + up->pcslen = val; + up->pcflag |= UDPLITE_SEND_CC; + break; + + case UDPLITE_RECV_CSCOV: + if (!up->pcflag) /* Disable the option on UDP sockets */ + return -ENOPROTOOPT; + if (val != 0 && val < 8) /* Avoid silly minimal values. */ + val = 8; + up->pcrlen = val; + up->pcflag |= UDPLITE_RECV_CC; + break; + default: err = -ENOPROTOOPT; break; @@ -874,22 +904,21 @@ static int do_udpv6_setsockopt(struct sock *sk, int level, int optname, return err; } -static int udpv6_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, int optlen) +int udpv6_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen) { - if (level != SOL_UDP) - return ipv6_setsockopt(sk, level, optname, optval, optlen); - return do_udpv6_setsockopt(sk, level, optname, optval, optlen); + if (level == SOL_UDP || level == SOL_UDPLITE) + return do_udpv6_setsockopt(sk, level, optname, optval, optlen); + return ipv6_setsockopt(sk, level, optname, optval, optlen); } #ifdef CONFIG_COMPAT -static int compat_udpv6_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, int optlen) +int compat_udpv6_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen) { - if (level != SOL_UDP) - return compat_ipv6_setsockopt(sk, level, optname, - optval, optlen); - return do_udpv6_setsockopt(sk, level, optname, optval, optlen); + if (level == SOL_UDP || level == SOL_UDPLITE) + return do_udpv6_setsockopt(sk, level, optname, optval, optlen); + return compat_ipv6_setsockopt(sk, level, optname, optval, optlen); } #endif @@ -916,6 +945,14 @@ static int do_udpv6_getsockopt(struct sock *sk, int level, int optname, val = up->encap_type; break; + case UDPLITE_SEND_CSCOV: + val = up->pcslen; + break; + + case UDPLITE_RECV_CSCOV: + val = up->pcrlen; + break; + default: return -ENOPROTOOPT; }; @@ -927,22 +964,21 @@ static int do_udpv6_getsockopt(struct sock *sk, int level, int optname, return 0; } -static int udpv6_getsockopt(struct sock *sk, int level, int optname, - char __user *optval, int __user *optlen) +int udpv6_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen) { - if (level != SOL_UDP) - return ipv6_getsockopt(sk, level, optname, optval, optlen); - return do_udpv6_getsockopt(sk, level, optname, optval, optlen); + if (level == SOL_UDP || level == SOL_UDPLITE) + return do_udpv6_getsockopt(sk, level, optname, optval, optlen); + return ipv6_getsockopt(sk, level, optname, optval, optlen); } #ifdef CONFIG_COMPAT -static int compat_udpv6_getsockopt(struct sock *sk, int level, int optname, - char __user *optval, int __user *optlen) +int compat_udpv6_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen) { - if (level != SOL_UDP) - return compat_ipv6_getsockopt(sk, level, optname, - optval, optlen); - return do_udpv6_getsockopt(sk, level, optname, optval, optlen); + if (level == SOL_UDP || level == SOL_UDPLITE) + return do_udpv6_getsockopt(sk, level, optname, optval, optlen); + return compat_ipv6_getsockopt(sk, level, optname, optval, optlen); } #endif @@ -983,7 +1019,7 @@ static void udp6_sock_seq_show(struct seq_file *seq, struct sock *sp, int bucket atomic_read(&sp->sk_refcnt), sp); } -static int udp6_seq_show(struct seq_file *seq, void *v) +int udp6_seq_show(struct seq_file *seq, void *v) { if (v == SEQ_START_TOKEN) seq_printf(seq, @@ -1002,6 +1038,7 @@ static struct udp_seq_afinfo udp6_seq_afinfo = { .owner = THIS_MODULE, .name = "udp6", .family = AF_INET6, + .hashtable = udp_hash, .seq_show = udp6_seq_show, .seq_fops = &udp6_seq_fops, }; @@ -1021,7 +1058,7 @@ void udp6_proc_exit(void) { struct proto udpv6_prot = { .name = "UDPv6", .owner = THIS_MODULE, - .close = udpv6_close, + .close = udp_lib_close, .connect = ip6_datagram_connect, .disconnect = udp_disconnect, .ioctl = udp_ioctl, @@ -1031,8 +1068,8 @@ struct proto udpv6_prot = { .sendmsg = udpv6_sendmsg, .recvmsg = udpv6_recvmsg, .backlog_rcv = udpv6_queue_rcv_skb, - .hash = udp_v6_hash, - .unhash = udp_v6_unhash, + .hash = udp_lib_hash, + .unhash = udp_lib_unhash, .get_port = udp_v6_get_port, .obj_size = sizeof(struct udp6_sock), #ifdef CONFIG_COMPAT diff --git a/net/ipv6/udp_impl.h b/net/ipv6/udp_impl.h new file mode 100644 index 00000000000..ec987889912 --- /dev/null +++ b/net/ipv6/udp_impl.h @@ -0,0 +1,34 @@ +#ifndef _UDP6_IMPL_H +#define _UDP6_IMPL_H +#include +#include +#include +#include +#include + +extern int __udp6_lib_rcv(struct sk_buff **, struct hlist_head [], int ); +extern void __udp6_lib_err(struct sk_buff *, struct inet6_skb_parm *, + int , int , int , __be32 , struct hlist_head []); + +extern int udpv6_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen); +extern int udpv6_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen); +#ifdef CONFIG_COMPAT +extern int compat_udpv6_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen); +extern int compat_udpv6_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen); +#endif +extern int udpv6_sendmsg(struct kiocb *iocb, struct sock *sk, + struct msghdr *msg, size_t len); +extern int udpv6_recvmsg(struct kiocb *iocb, struct sock *sk, + struct msghdr *msg, size_t len, + int noblock, int flags, int *addr_len); +extern int udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb); +extern int udpv6_destroy_sock(struct sock *sk); + +#ifdef CONFIG_PROC_FS +extern int udp6_seq_show(struct seq_file *seq, void *v); +#endif +#endif /* _UDP6_IMPL_H */ diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c new file mode 100644 index 00000000000..e0ec5e63004 --- /dev/null +++ b/net/ipv6/udplite.c @@ -0,0 +1,105 @@ +/* + * UDPLITEv6 An implementation of the UDP-Lite protocol over IPv6. + * See also net/ipv4/udplite.c + * + * Version: $Id: udplite.c,v 1.9 2006/10/19 08:28:10 gerrit Exp $ + * + * Authors: Gerrit Renker + * + * Changes: + * Fixes: + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ +#include "udp_impl.h" + +DEFINE_SNMP_STAT(struct udp_mib, udplite_stats_in6) __read_mostly; + +static __inline__ int udplitev6_rcv(struct sk_buff **pskb) +{ + return __udp6_lib_rcv(pskb, udplite_hash, 1); +} + +static __inline__ void udplitev6_err(struct sk_buff *skb, + struct inet6_skb_parm *opt, + int type, int code, int offset, __u32 info) +{ + return __udp6_lib_err(skb, opt, type, code, offset, info, udplite_hash); +} + +static struct inet6_protocol udplitev6_protocol = { + .handler = udplitev6_rcv, + .err_handler = udplitev6_err, + .flags = INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL, +}; + +static __inline__ int udplite_v6_get_port(struct sock *sk, unsigned short snum) +{ + return udplite_get_port(sk, snum, ipv6_rcv_saddr_equal); +} + +struct proto udplitev6_prot = { + .name = "UDPLITEv6", + .owner = THIS_MODULE, + .close = udp_lib_close, + .connect = ip6_datagram_connect, + .disconnect = udp_disconnect, + .ioctl = udp_ioctl, + .init = udplite_sk_init, + .destroy = udpv6_destroy_sock, + .setsockopt = udpv6_setsockopt, + .getsockopt = udpv6_getsockopt, + .sendmsg = udpv6_sendmsg, + .recvmsg = udpv6_recvmsg, + .backlog_rcv = udpv6_queue_rcv_skb, + .hash = udp_lib_hash, + .unhash = udp_lib_unhash, + .get_port = udplite_v6_get_port, + .obj_size = sizeof(struct udp6_sock), +#ifdef CONFIG_COMPAT + .compat_setsockopt = compat_udpv6_setsockopt, + .compat_getsockopt = compat_udpv6_getsockopt, +#endif +}; + +static struct inet_protosw udplite6_protosw = { + .type = SOCK_DGRAM, + .protocol = IPPROTO_UDPLITE, + .prot = &udplitev6_prot, + .ops = &inet6_dgram_ops, + .capability = -1, + .no_check = 0, + .flags = INET_PROTOSW_PERMANENT, +}; + +void __init udplitev6_init(void) +{ + if (inet6_add_protocol(&udplitev6_protocol, IPPROTO_UDPLITE) < 0) + printk(KERN_ERR "%s: Could not register.\n", __FUNCTION__); + + inet6_register_protosw(&udplite6_protosw); +} + +#ifdef CONFIG_PROC_FS +static struct file_operations udplite6_seq_fops; +static struct udp_seq_afinfo udplite6_seq_afinfo = { + .owner = THIS_MODULE, + .name = "udplite6", + .family = AF_INET6, + .hashtable = udplite_hash, + .seq_show = udp6_seq_show, + .seq_fops = &udplite6_seq_fops, +}; + +int __init udplite6_proc_init(void) +{ + return udp_proc_register(&udplite6_seq_afinfo); +} + +void udplite6_proc_exit(void) +{ + udp_proc_unregister(&udplite6_seq_afinfo); +} +#endif diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index 2fba1f0739a..8dffd4daae9 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -274,6 +274,7 @@ _decode_session6(struct sk_buff *skb, struct flowi *fl) break; case IPPROTO_UDP: + case IPPROTO_UDPLITE: case IPPROTO_TCP: case IPPROTO_SCTP: case IPPROTO_DCCP: diff --git a/net/netfilter/xt_multiport.c b/net/netfilter/xt_multiport.c index b4293058c6f..1602086c7fd 100644 --- a/net/netfilter/xt_multiport.c +++ b/net/netfilter/xt_multiport.c @@ -1,5 +1,5 @@ -/* Kernel module to match one of a list of TCP/UDP/SCTP/DCCP ports: ports are in - the same place so we can treat them as equal. */ +/* Kernel module to match one of a list of TCP/UDP(-Lite)/SCTP/DCCP ports: + ports are in the same place so we can treat them as equal. */ /* (C) 1999-2001 Paul `Rusty' Russell * (C) 2002-2004 Netfilter Core Team @@ -162,6 +162,7 @@ check(u_int16_t proto, { /* Must specify supported protocol, no unknown flags or bad count */ return (proto == IPPROTO_TCP || proto == IPPROTO_UDP + || proto == IPPROTO_UDPLITE || proto == IPPROTO_SCTP || proto == IPPROTO_DCCP) && !(ip_invflags & XT_INV_PROTO) && (match_flags == XT_MULTIPORT_SOURCE diff --git a/net/netfilter/xt_tcpudp.c b/net/netfilter/xt_tcpudp.c index e76a68e0bc6..46414b562a1 100644 --- a/net/netfilter/xt_tcpudp.c +++ b/net/netfilter/xt_tcpudp.c @@ -10,7 +10,7 @@ #include #include -MODULE_DESCRIPTION("x_tables match for TCP and UDP, supports IPv4 and IPv6"); +MODULE_DESCRIPTION("x_tables match for TCP and UDP(-Lite), supports IPv4 and IPv6"); MODULE_LICENSE("GPL"); MODULE_ALIAS("xt_tcp"); MODULE_ALIAS("xt_udp"); @@ -234,6 +234,24 @@ static struct xt_match xt_tcpudp_match[] = { .proto = IPPROTO_UDP, .me = THIS_MODULE, }, + { + .name = "udplite", + .family = AF_INET, + .checkentry = udp_checkentry, + .match = udp_match, + .matchsize = sizeof(struct xt_udp), + .proto = IPPROTO_UDPLITE, + .me = THIS_MODULE, + }, + { + .name = "udplite", + .family = AF_INET6, + .checkentry = udp_checkentry, + .match = udp_match, + .matchsize = sizeof(struct xt_udp), + .proto = IPPROTO_UDPLITE, + .me = THIS_MODULE, + }, }; static int __init xt_tcpudp_init(void) -- cgit v1.2.3-70-g09d2 From 448c31aa34b5ee947d322e8747c4cf801fc4c104 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 20:47:46 -0800 Subject: [IRDA]: Trivial annotations. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/irda/irlap_frame.h | 16 ++++++++-------- net/irda/iriap.c | 8 ++++---- net/irda/irlan/irlan_common.c | 2 +- net/irda/irttp.c | 4 ++-- 4 files changed, 15 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/include/net/irda/irlap_frame.h b/include/net/irda/irlap_frame.h index 9dd54a5002b..641f88e848b 100644 --- a/include/net/irda/irlap_frame.h +++ b/include/net/irda/irlap_frame.h @@ -91,8 +91,8 @@ struct xid_frame { __u8 caddr; /* Connection address */ __u8 control; __u8 ident; /* Should always be XID_FORMAT */ - __u32 saddr; /* Source device address */ - __u32 daddr; /* Destination device address */ + __le32 saddr; /* Source device address */ + __le32 daddr; /* Destination device address */ __u8 flags; /* Discovery flags */ __u8 slotnr; __u8 version; @@ -101,15 +101,15 @@ struct xid_frame { struct test_frame { __u8 caddr; /* Connection address */ __u8 control; - __u32 saddr; /* Source device address */ - __u32 daddr; /* Destination device address */ + __le32 saddr; /* Source device address */ + __le32 daddr; /* Destination device address */ } IRDA_PACK; struct ua_frame { __u8 caddr; __u8 control; - __u32 saddr; /* Source device address */ - __u32 daddr; /* Dest device address */ + __le32 saddr; /* Source device address */ + __le32 daddr; /* Dest device address */ } IRDA_PACK; struct dm_frame { @@ -135,8 +135,8 @@ struct i_frame { struct snrm_frame { __u8 caddr; __u8 control; - __u32 saddr; - __u32 daddr; + __le32 saddr; + __le32 daddr; __u8 ncaddr; } IRDA_PACK; diff --git a/net/irda/iriap.c b/net/irda/iriap.c index 415cf4eec23..ab86d70393e 100644 --- a/net/irda/iriap.c +++ b/net/irda/iriap.c @@ -451,12 +451,12 @@ static void iriap_getvaluebyclass_confirm(struct iriap_cb *self, n = 2; /* Get length, MSB first */ - len = be16_to_cpu(get_unaligned((__u16 *)(fp+n))); n += 2; + len = be16_to_cpu(get_unaligned((__be16 *)(fp+n))); n += 2; IRDA_DEBUG(4, "%s(), len=%d\n", __FUNCTION__, len); /* Get object ID, MSB first */ - obj_id = be16_to_cpu(get_unaligned((__u16 *)(fp+n))); n += 2; + obj_id = be16_to_cpu(get_unaligned((__be16 *)(fp+n))); n += 2; type = fp[n++]; IRDA_DEBUG(4, "%s(), Value type = %d\n", __FUNCTION__, type); @@ -506,7 +506,7 @@ static void iriap_getvaluebyclass_confirm(struct iriap_cb *self, value = irias_new_string_value(fp+n); break; case IAS_OCT_SEQ: - value_len = be16_to_cpu(get_unaligned((__u16 *)(fp+n))); + value_len = be16_to_cpu(get_unaligned((__be16 *)(fp+n))); n += 2; /* Will truncate to IAS_MAX_OCTET_STRING bytes */ @@ -544,7 +544,7 @@ static void iriap_getvaluebyclass_response(struct iriap_cb *self, { struct sk_buff *tx_skb; int n; - __u32 tmp_be32; + __be32 tmp_be32; __be16 tmp_be16; __u8 *fp; diff --git a/net/irda/irlan/irlan_common.c b/net/irda/irlan/irlan_common.c index 9b962f24771..2bb04ac0932 100644 --- a/net/irda/irlan/irlan_common.c +++ b/net/irda/irlan/irlan_common.c @@ -995,7 +995,7 @@ static int __irlan_insert_param(struct sk_buff *skb, char *param, int type, { __u8 *frame; __u8 param_len; - __u16 tmp_le; /* Temporary value in little endian format */ + __le16 tmp_le; /* Temporary value in little endian format */ int n=0; if (skb == NULL) { diff --git a/net/irda/irttp.c b/net/irda/irttp.c index 3c2e70b77df..9c446a72ff1 100644 --- a/net/irda/irttp.c +++ b/net/irda/irttp.c @@ -1147,7 +1147,7 @@ int irttp_connect_request(struct tsap_cb *self, __u8 dtsap_sel, frame[3] = 0x02; /* Value length */ put_unaligned(cpu_to_be16((__u16) max_sdu_size), - (__u16 *)(frame+4)); + (__be16 *)(frame+4)); } else { /* Insert plain TTP header */ frame = skb_push(tx_skb, TTP_HEADER); @@ -1394,7 +1394,7 @@ int irttp_connect_response(struct tsap_cb *self, __u32 max_sdu_size, frame[3] = 0x02; /* Value length */ put_unaligned(cpu_to_be16((__u16) max_sdu_size), - (__u16 *)(frame+4)); + (__be16 *)(frame+4)); } else { /* Insert TTP header */ frame = skb_push(tx_skb, TTP_HEADER); -- cgit v1.2.3-70-g09d2 From 714e85be3557222bc25f69c252326207c900a7db Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 20:51:49 -0800 Subject: [IPV6]: Assorted trivial endianness annotations. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/inetdevice.h | 14 ++++++++------ include/net/arp.h | 2 +- include/net/ip.h | 11 ++++++----- net/ipv4/af_inet.c | 6 +++--- net/ipv4/arp.c | 2 +- net/ipv4/cipso_ipv4.c | 8 ++++---- net/ipv4/devinet.c | 10 +++++----- net/ipv4/ip_output.c | 2 +- net/ipv4/ip_sockglue.c | 2 +- net/ipv4/raw.c | 4 ++-- net/ipv4/route.c | 4 ++-- net/ipv4/syncookies.c | 18 +++++++++--------- net/ipv4/tcp_ipv4.c | 10 +++++----- net/ipv4/tcp_minisocks.c | 2 +- net/ipv4/tcp_timer.c | 2 +- 15 files changed, 50 insertions(+), 47 deletions(-) (limited to 'include') diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h index 5a0ab04627b..c0f7aec331c 100644 --- a/include/linux/inetdevice.h +++ b/include/linux/inetdevice.h @@ -124,12 +124,13 @@ static __inline__ int inet_ifa_match(__be32 addr, struct in_ifaddr *ifa) * Check if a mask is acceptable. */ -static __inline__ int bad_mask(u32 mask, u32 addr) +static __inline__ int bad_mask(__be32 mask, __be32 addr) { + __u32 hmask; if (addr & (mask = ~mask)) return 1; - mask = ntohl(mask); - if (mask & (mask+1)) + hmask = ntohl(mask); + if (hmask & (hmask+1)) return 1; return 0; } @@ -190,11 +191,12 @@ static __inline__ __be32 inet_make_mask(int logmask) return 0; } -static __inline__ int inet_mask_len(__u32 mask) +static __inline__ int inet_mask_len(__be32 mask) { - if (!(mask = ntohl(mask))) + __u32 hmask = ntohl(mask); + if (!hmask) return 0; - return 32 - ffz(~mask); + return 32 - ffz(~hmask); } diff --git a/include/net/arp.h b/include/net/arp.h index 6a3d9a7d302..f0266456860 100644 --- a/include/net/arp.h +++ b/include/net/arp.h @@ -16,7 +16,7 @@ extern void arp_send(int type, int ptype, __be32 dest_ip, struct net_device *dev, __be32 src_ip, unsigned char *dest_hw, unsigned char *src_hw, unsigned char *th); extern int arp_bind_neighbour(struct dst_entry *dst); -extern int arp_mc_map(u32 addr, u8 *haddr, struct net_device *dev, int dir); +extern int arp_mc_map(__be32 addr, u8 *haddr, struct net_device *dev, int dir); extern void arp_ifdown(struct net_device *dev); extern struct sk_buff *arp_create(int type, int ptype, __be32 dest_ip, diff --git a/include/net/ip.h b/include/net/ip.h index 949fa868362..412e8114667 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -123,7 +123,7 @@ extern int ip4_datagram_connect(struct sock *sk, * multicast packets. */ -static inline void ip_tr_mc_map(u32 addr, char *buf) +static inline void ip_tr_mc_map(__be32 addr, char *buf) { buf[0]=0xC0; buf[1]=0x00; @@ -238,9 +238,9 @@ static inline void ip_select_ident_more(struct iphdr *iph, struct dst_entry *dst * Map a multicast IP onto multicast MAC for type ethernet. */ -static inline void ip_eth_mc_map(u32 addr, char *buf) +static inline void ip_eth_mc_map(__be32 naddr, char *buf) { - addr=ntohl(addr); + __u32 addr=ntohl(naddr); buf[0]=0x01; buf[1]=0x00; buf[2]=0x5e; @@ -256,13 +256,14 @@ static inline void ip_eth_mc_map(u32 addr, char *buf) * Leave P_Key as 0 to be filled in by driver. */ -static inline void ip_ib_mc_map(u32 addr, char *buf) +static inline void ip_ib_mc_map(__be32 naddr, char *buf) { + __u32 addr; buf[0] = 0; /* Reserved */ buf[1] = 0xff; /* Multicast QPN */ buf[2] = 0xff; buf[3] = 0xff; - addr = ntohl(addr); + addr = ntohl(naddr); buf[4] = 0xff; buf[5] = 0x12; /* link local scope */ buf[6] = 0x40; /* IPv4 signature */ diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 8db39f7e3bf..1144900d37f 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -644,7 +644,7 @@ int inet_getname(struct socket *sock, struct sockaddr *uaddr, sin->sin_port = inet->dport; sin->sin_addr.s_addr = inet->daddr; } else { - __u32 addr = inet->rcv_saddr; + __be32 addr = inet->rcv_saddr; if (!addr) addr = inet->saddr; sin->sin_port = inet->sport; @@ -995,8 +995,8 @@ static int inet_sk_reselect_saddr(struct sock *sk) struct inet_sock *inet = inet_sk(sk); int err; struct rtable *rt; - __u32 old_saddr = inet->saddr; - __u32 new_saddr; + __be32 old_saddr = inet->saddr; + __be32 new_saddr; __be32 daddr = inet->daddr; if (inet->opt && inet->opt->srr) diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index cfb5d3de9c8..3981e8be9ab 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -203,7 +203,7 @@ struct neigh_table arp_tbl = { .gc_thresh3 = 1024, }; -int arp_mc_map(u32 addr, u8 *haddr, struct net_device *dev, int dir) +int arp_mc_map(__be32 addr, u8 *haddr, struct net_device *dev, int dir) { switch (dev->type) { case ARPHRD_ETHER: diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c index 6460233407c..c3a61ebbade 100644 --- a/net/ipv4/cipso_ipv4.c +++ b/net/ipv4/cipso_ipv4.c @@ -966,7 +966,7 @@ static int cipso_v4_gentag_hdr(const struct cipso_v4_doi *doi_def, buf[0] = IPOPT_CIPSO; buf[1] = CIPSO_V4_HDR_LEN + len; - *(u32 *)&buf[2] = htonl(doi_def->doi); + *(__be32 *)&buf[2] = htonl(doi_def->doi); return 0; } @@ -1140,7 +1140,7 @@ int cipso_v4_validate(unsigned char **option) } rcu_read_lock(); - doi_def = cipso_v4_doi_getdef(ntohl(*((u32 *)&opt[2]))); + doi_def = cipso_v4_doi_getdef(ntohl(*((__be32 *)&opt[2]))); if (doi_def == NULL) { err_offset = 2; goto validate_return_locked; @@ -1370,7 +1370,7 @@ int cipso_v4_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr) if (ret_val == 0) return ret_val; - doi = ntohl(*(u32 *)&cipso_ptr[2]); + doi = ntohl(*(__be32 *)&cipso_ptr[2]); rcu_read_lock(); doi_def = cipso_v4_doi_getdef(doi); if (doi_def == NULL) { @@ -1436,7 +1436,7 @@ int cipso_v4_skbuff_getattr(const struct sk_buff *skb, if (cipso_v4_cache_check(cipso_ptr, cipso_ptr[1], secattr) == 0) return 0; - doi = ntohl(*(u32 *)&cipso_ptr[2]); + doi = ntohl(*(__be32 *)&cipso_ptr[2]); rcu_read_lock(); doi_def = cipso_v4_doi_getdef(doi); if (doi_def == NULL) diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index f38cbbae0ae..a1b356c8aa5 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -577,20 +577,20 @@ static int inet_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg * Determine a default network mask, based on the IP address. */ -static __inline__ int inet_abc_len(u32 addr) +static __inline__ int inet_abc_len(__be32 addr) { int rc = -1; /* Something else, probably a multicast. */ if (ZERONET(addr)) rc = 0; else { - addr = ntohl(addr); + __u32 haddr = ntohl(addr); - if (IN_CLASSA(addr)) + if (IN_CLASSA(haddr)) rc = 8; - else if (IN_CLASSB(addr)) + else if (IN_CLASSB(haddr)) rc = 16; - else if (IN_CLASSC(addr)) + else if (IN_CLASSC(haddr)) rc = 24; } diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index def32d8d3b0..90942a384a4 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -341,7 +341,7 @@ packet_routed: /* OK, we know where to send it, allocate and build IP header. */ iph = (struct iphdr *) skb_push(skb, sizeof(struct iphdr) + (opt ? opt->optlen : 0)); - *((__u16 *)iph) = htons((4 << 12) | (5 << 8) | (inet->tos & 0xff)); + *((__be16 *)iph) = htons((4 << 12) | (5 << 8) | (inet->tos & 0xff)); iph->tot_len = htons(skb->len); if (ip_dont_fragment(sk, &rt->u.dst) && !ipfragok) iph->frag_off = htons(IP_DF); diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c index 4b132953bcc..57d4bae6f08 100644 --- a/net/ipv4/ip_sockglue.c +++ b/net/ipv4/ip_sockglue.c @@ -355,7 +355,7 @@ int ip_recv_error(struct sock *sk, struct msghdr *msg, int len) sin = (struct sockaddr_in *)msg->msg_name; if (sin) { sin->sin_family = AF_INET; - sin->sin_addr.s_addr = *(u32*)(skb->nh.raw + serr->addr_offset); + sin->sin_addr.s_addr = *(__be32*)(skb->nh.raw + serr->addr_offset); sin->sin_port = serr->port; memset(&sin->sin_zero, 0, sizeof(sin->sin_zero)); } diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 5c31dead2bd..a6c63bbd9dd 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -854,8 +854,8 @@ static void raw_seq_stop(struct seq_file *seq, void *v) static __inline__ char *get_raw_sock(struct sock *sp, char *tmpbuf, int i) { struct inet_sock *inet = inet_sk(sp); - unsigned int dest = inet->daddr, - src = inet->rcv_saddr; + __be32 dest = inet->daddr, + src = inet->rcv_saddr; __u16 destp = 0, srcp = inet->num; diff --git a/net/ipv4/route.c b/net/ipv4/route.c index d7152b2b2c6..ee00b6506ab 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -566,8 +566,8 @@ static inline u32 rt_score(struct rtable *rt) static inline int compare_keys(struct flowi *fl1, struct flowi *fl2) { - return ((fl1->nl_u.ip4_u.daddr ^ fl2->nl_u.ip4_u.daddr) | - (fl1->nl_u.ip4_u.saddr ^ fl2->nl_u.ip4_u.saddr) | + return ((__force u32)((fl1->nl_u.ip4_u.daddr ^ fl2->nl_u.ip4_u.daddr) | + (fl1->nl_u.ip4_u.saddr ^ fl2->nl_u.ip4_u.saddr)) | (fl1->mark ^ fl2->mark) | (*(u16 *)&fl1->nl_u.ip4_u.tos ^ *(u16 *)&fl2->nl_u.ip4_u.tos) | diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index 661e0a4bca7..6b19530905a 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -35,23 +35,23 @@ module_init(init_syncookies); #define COOKIEBITS 24 /* Upper bits store count */ #define COOKIEMASK (((__u32)1 << COOKIEBITS) - 1) -static u32 cookie_hash(u32 saddr, u32 daddr, u32 sport, u32 dport, +static u32 cookie_hash(__be32 saddr, __be32 daddr, __be16 sport, __be16 dport, u32 count, int c) { __u32 tmp[16 + 5 + SHA_WORKSPACE_WORDS]; memcpy(tmp + 3, syncookie_secret[c], sizeof(syncookie_secret[c])); - tmp[0] = saddr; - tmp[1] = daddr; - tmp[2] = (sport << 16) + dport; + tmp[0] = (__force u32)saddr; + tmp[1] = (__force u32)daddr; + tmp[2] = ((__force u32)sport << 16) + (__force u32)dport; tmp[3] = count; sha_transform(tmp + 16, (__u8 *)tmp, tmp + 16 + 5); return tmp[17]; } -static __u32 secure_tcp_syn_cookie(__u32 saddr, __u32 daddr, __u16 sport, - __u16 dport, __u32 sseq, __u32 count, +static __u32 secure_tcp_syn_cookie(__be32 saddr, __be32 daddr, __be16 sport, + __be16 dport, __u32 sseq, __u32 count, __u32 data) { /* @@ -80,8 +80,8 @@ static __u32 secure_tcp_syn_cookie(__u32 saddr, __u32 daddr, __u16 sport, * "maxdiff" if the current (passed-in) "count". The return value * is (__u32)-1 if this test fails. */ -static __u32 check_tcp_syn_cookie(__u32 cookie, __u32 saddr, __u32 daddr, - __u16 sport, __u16 dport, __u32 sseq, +static __u32 check_tcp_syn_cookie(__u32 cookie, __be32 saddr, __be32 daddr, + __be16 sport, __be16 dport, __u32 sseq, __u32 count, __u32 maxdiff) { __u32 diff; @@ -220,7 +220,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, } ireq = inet_rsk(req); treq = tcp_rsk(req); - treq->rcv_isn = htonl(skb->h.th->seq) - 1; + treq->rcv_isn = ntohl(skb->h.th->seq) - 1; treq->snt_isn = cookie; req->mss = mss; ireq->rmt_port = skb->h.th->source; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 8c8e8112f98..0ca8dead03b 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -542,7 +542,7 @@ static void tcp_v4_send_reset(struct sock *sk, struct sk_buff *skb) struct { struct tcphdr th; #ifdef CONFIG_TCP_MD5SIG - u32 opt[(TCPOLEN_MD5SIG_ALIGNED >> 2)]; + __be32 opt[(TCPOLEN_MD5SIG_ALIGNED >> 2)]; #endif } rep; struct ip_reply_arg arg; @@ -618,9 +618,9 @@ static void tcp_v4_send_ack(struct tcp_timewait_sock *twsk, struct tcphdr *th = skb->h.th; struct { struct tcphdr th; - u32 opt[(TCPOLEN_TSTAMP_ALIGNED >> 2) + __be32 opt[(TCPOLEN_TSTAMP_ALIGNED >> 2) #ifdef CONFIG_TCP_MD5SIG - + (TCPOLEN_MD5SIG_ALIGNED >> 2) + + (TCPOLEN_MD5SIG_ALIGNED >> 2) #endif ]; } rep; @@ -2333,8 +2333,8 @@ static void get_tcp4_sock(struct sock *sp, char *tmpbuf, int i) struct tcp_sock *tp = tcp_sk(sp); const struct inet_connection_sock *icsk = inet_csk(sp); struct inet_sock *inet = inet_sk(sp); - unsigned int dest = inet->daddr; - unsigned int src = inet->rcv_saddr; + __be32 dest = inet->daddr; + __be32 src = inet->rcv_saddr; __u16 destp = ntohs(inet->dport); __u16 srcp = ntohs(inet->sport); diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 4a0ee901a88..383cb38461c 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -493,7 +493,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, struct request_sock **prev) { struct tcphdr *th = skb->h.th; - u32 flg = tcp_flag_word(th) & (TCP_FLAG_RST|TCP_FLAG_SYN|TCP_FLAG_ACK); + __be32 flg = tcp_flag_word(th) & (TCP_FLAG_RST|TCP_FLAG_SYN|TCP_FLAG_ACK); int paws_reject = 0; struct tcp_options_received tmp_opt; struct sock *child; diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index fb09ade5897..3355c276b61 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -297,7 +297,7 @@ static void tcp_retransmit_timer(struct sock *sk) if (net_ratelimit()) { struct inet_sock *inet = inet_sk(sk); printk(KERN_DEBUG "TCP: Treason uncloaked! Peer %u.%u.%u.%u:%u/%u shrinks window %u:%u. Repaired.\n", - NIPQUAD(inet->daddr), htons(inet->dport), + NIPQUAD(inet->daddr), ntohs(inet->dport), inet->num, tp->snd_una, tp->snd_nxt); } #endif -- cgit v1.2.3-70-g09d2 From b09b845ca6724c3bbdc00c0cb2313258c7189ca9 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 20:52:19 -0800 Subject: [RANDOM]: Annotate random.h IP helpers. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- drivers/char/random.c | 48 ++++++++++++++++++++++++------------------------ include/linux/random.h | 20 ++++++++++---------- 2 files changed, 34 insertions(+), 34 deletions(-) (limited to 'include') diff --git a/drivers/char/random.c b/drivers/char/random.c index eb6b13f4211..d40df30c2b1 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -1466,8 +1466,8 @@ static __init int seqgen_init(void) late_initcall(seqgen_init); #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) -__u32 secure_tcpv6_sequence_number(__u32 *saddr, __u32 *daddr, - __u16 sport, __u16 dport) +__u32 secure_tcpv6_sequence_number(__be32 *saddr, __be32 *daddr, + __be16 sport, __be16 dport) { struct timeval tv; __u32 seq; @@ -1479,10 +1479,10 @@ __u32 secure_tcpv6_sequence_number(__u32 *saddr, __u32 *daddr, */ memcpy(hash, saddr, 16); - hash[4]=(sport << 16) + dport; + hash[4]=((__force u16)sport << 16) + (__force u16)dport; memcpy(&hash[5],keyptr->secret,sizeof(__u32) * 7); - seq = twothirdsMD4Transform(daddr, hash) & HASH_MASK; + seq = twothirdsMD4Transform((const __u32 *)daddr, hash) & HASH_MASK; seq += keyptr->count; do_gettimeofday(&tv); @@ -1496,7 +1496,7 @@ EXPORT_SYMBOL(secure_tcpv6_sequence_number); /* The code below is shamelessly stolen from secure_tcp_sequence_number(). * All blames to Andrey V. Savochkin . */ -__u32 secure_ip_id(__u32 daddr) +__u32 secure_ip_id(__be32 daddr) { struct keydata *keyptr; __u32 hash[4]; @@ -1508,7 +1508,7 @@ __u32 secure_ip_id(__u32 daddr) * The dest ip address is placed in the starting vector, * which is then hashed with random data. */ - hash[0] = daddr; + hash[0] = (__force __u32)daddr; hash[1] = keyptr->secret[9]; hash[2] = keyptr->secret[10]; hash[3] = keyptr->secret[11]; @@ -1518,8 +1518,8 @@ __u32 secure_ip_id(__u32 daddr) #ifdef CONFIG_INET -__u32 secure_tcp_sequence_number(__u32 saddr, __u32 daddr, - __u16 sport, __u16 dport) +__u32 secure_tcp_sequence_number(__be32 saddr, __be32 daddr, + __be16 sport, __be16 dport) { struct timeval tv; __u32 seq; @@ -1532,9 +1532,9 @@ __u32 secure_tcp_sequence_number(__u32 saddr, __u32 daddr, * Note that the words are placed into the starting vector, which is * then mixed with a partial MD4 over random data. */ - hash[0]=saddr; - hash[1]=daddr; - hash[2]=(sport << 16) + dport; + hash[0]=(__force u32)saddr; + hash[1]=(__force u32)daddr; + hash[2]=((__force u16)sport << 16) + (__force u16)dport; hash[3]=keyptr->secret[11]; seq = half_md4_transform(hash, keyptr->secret) & HASH_MASK; @@ -1559,7 +1559,7 @@ __u32 secure_tcp_sequence_number(__u32 saddr, __u32 daddr, EXPORT_SYMBOL(secure_tcp_sequence_number); /* Generate secure starting point for ephemeral IPV4 transport port search */ -u32 secure_ipv4_port_ephemeral(__u32 saddr, __u32 daddr, __u16 dport) +u32 secure_ipv4_port_ephemeral(__be32 saddr, __be32 daddr, __be16 dport) { struct keydata *keyptr = get_keyptr(); u32 hash[4]; @@ -1568,25 +1568,25 @@ u32 secure_ipv4_port_ephemeral(__u32 saddr, __u32 daddr, __u16 dport) * Pick a unique starting offset for each ephemeral port search * (saddr, daddr, dport) and 48bits of random data. */ - hash[0] = saddr; - hash[1] = daddr; - hash[2] = dport ^ keyptr->secret[10]; + hash[0] = (__force u32)saddr; + hash[1] = (__force u32)daddr; + hash[2] = (__force u32)dport ^ keyptr->secret[10]; hash[3] = keyptr->secret[11]; return half_md4_transform(hash, keyptr->secret); } #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) -u32 secure_ipv6_port_ephemeral(const __u32 *saddr, const __u32 *daddr, __u16 dport) +u32 secure_ipv6_port_ephemeral(const __be32 *saddr, const __be32 *daddr, __be16 dport) { struct keydata *keyptr = get_keyptr(); u32 hash[12]; memcpy(hash, saddr, 16); - hash[4] = dport; + hash[4] = (__force u32)dport; memcpy(&hash[5],keyptr->secret,sizeof(__u32) * 7); - return twothirdsMD4Transform(daddr, hash); + return twothirdsMD4Transform((const __u32 *)daddr, hash); } #endif @@ -1595,17 +1595,17 @@ u32 secure_ipv6_port_ephemeral(const __u32 *saddr, const __u32 *daddr, __u16 dpo * bit's 32-47 increase every key exchange * 0-31 hash(source, dest) */ -u64 secure_dccp_sequence_number(__u32 saddr, __u32 daddr, - __u16 sport, __u16 dport) +u64 secure_dccp_sequence_number(__be32 saddr, __be32 daddr, + __be16 sport, __be16 dport) { struct timeval tv; u64 seq; __u32 hash[4]; struct keydata *keyptr = get_keyptr(); - hash[0] = saddr; - hash[1] = daddr; - hash[2] = (sport << 16) + dport; + hash[0] = (__force u32)saddr; + hash[1] = (__force u32)daddr; + hash[2] = ((__force u16)sport << 16) + (__force u16)dport; hash[3] = keyptr->secret[11]; seq = half_md4_transform(hash, keyptr->secret); @@ -1641,7 +1641,7 @@ unsigned int get_random_int(void) * drain on it), and uses halfMD4Transform within the second. We * also mix it with jiffies and the PID: */ - return secure_ip_id(current->pid + jiffies); + return secure_ip_id((__force __be32)(current->pid + jiffies)); } /* diff --git a/include/linux/random.h b/include/linux/random.h index 0248b30e306..01ad71033d6 100644 --- a/include/linux/random.h +++ b/include/linux/random.h @@ -51,16 +51,16 @@ extern void add_interrupt_randomness(int irq); extern void get_random_bytes(void *buf, int nbytes); void generate_random_uuid(unsigned char uuid_out[16]); -extern __u32 secure_ip_id(__u32 daddr); -extern u32 secure_ipv4_port_ephemeral(__u32 saddr, __u32 daddr, __u16 dport); -extern u32 secure_ipv6_port_ephemeral(const __u32 *saddr, const __u32 *daddr, - __u16 dport); -extern __u32 secure_tcp_sequence_number(__u32 saddr, __u32 daddr, - __u16 sport, __u16 dport); -extern __u32 secure_tcpv6_sequence_number(__u32 *saddr, __u32 *daddr, - __u16 sport, __u16 dport); -extern u64 secure_dccp_sequence_number(__u32 saddr, __u32 daddr, - __u16 sport, __u16 dport); +extern __u32 secure_ip_id(__be32 daddr); +extern u32 secure_ipv4_port_ephemeral(__be32 saddr, __be32 daddr, __be16 dport); +extern u32 secure_ipv6_port_ephemeral(const __be32 *saddr, const __be32 *daddr, + __be16 dport); +extern __u32 secure_tcp_sequence_number(__be32 saddr, __be32 daddr, + __be16 sport, __be16 dport); +extern __u32 secure_tcpv6_sequence_number(__be32 *saddr, __be32 *daddr, + __be16 sport, __be16 dport); +extern u64 secure_dccp_sequence_number(__be32 saddr, __be32 daddr, + __be16 sport, __be16 dport); #ifndef MODULE extern struct file_operations random_fops, urandom_fops; -- cgit v1.2.3-70-g09d2 From e69a4adc669fe210817ec50ae3f9a7a5ad62d4e8 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 20:56:00 -0800 Subject: [IPV6]: Misc endianness annotations. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/addrconf.h | 8 ++++---- include/net/ip6_route.h | 2 +- include/net/ipv6.h | 2 +- include/net/ndisc.h | 4 ++-- net/ipv6/addrconf.c | 2 +- net/ipv6/af_inet6.c | 4 ++-- net/ipv6/datagram.c | 6 +++--- net/ipv6/exthdrs.c | 2 +- net/ipv6/exthdrs_core.c | 2 +- net/ipv6/inet6_connection_sock.c | 8 ++++---- net/ipv6/ip6_fib.c | 8 ++++---- net/ipv6/ip6_tunnel.c | 2 +- net/ipv6/mip6.c | 4 ++-- net/ipv6/ndisc.c | 5 +++-- net/ipv6/netfilter/ip6_tables.c | 3 ++- net/ipv6/netfilter/ip6t_LOG.c | 4 ++-- net/ipv6/raw.c | 2 +- net/ipv6/reassembly.c | 26 +++++++++++++------------- net/ipv6/route.c | 2 +- net/ipv6/sit.c | 18 +++++++++--------- net/ipv6/tcp_ipv6.c | 4 ++-- net/ipv6/udp.c | 4 ++-- 22 files changed, 62 insertions(+), 60 deletions(-) (limited to 'include') diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 44f1b673f91..88df8fc814e 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -35,9 +35,9 @@ struct prefix_info { #else #error "Please fix " #endif - __u32 valid; - __u32 prefered; - __u32 reserved2; + __be32 valid; + __be32 prefered; + __be32 reserved2; struct in6_addr prefix; }; @@ -183,7 +183,7 @@ static __inline__ u8 ipv6_addr_hash(const struct in6_addr *addr) * This will include the IEEE address token on links that support it. */ - word = addr->s6_addr32[2] ^ addr->s6_addr32[3]; + word = (__force u32)(addr->s6_addr32[2] ^ addr->s6_addr32[3]); word ^= (word >> 16); word ^= (word >> 8); diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index c14b70ed4c5..4e927ebd1cb 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -20,7 +20,7 @@ struct route_info { route_pref:2, reserved_h:3; #endif - __u32 lifetime; + __be32 lifetime; __u8 prefix[0]; /* 0,8 or 16 */ }; diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 9390649bbfe..4ca9e93decc 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -563,7 +563,7 @@ extern int ip6_datagram_connect(struct sock *sk, struct sockaddr *addr, int addr_len); extern int ipv6_recv_error(struct sock *sk, struct msghdr *msg, int len); -extern void ipv6_icmp_error(struct sock *sk, struct sk_buff *skb, int err, u16 port, +extern void ipv6_icmp_error(struct sock *sk, struct sk_buff *skb, int err, __be16 port, u32 info, u8 *payload); extern void ipv6_local_error(struct sock *sk, int err, struct flowi *fl, u32 info); diff --git a/include/net/ndisc.h b/include/net/ndisc.h index d3915dabe6d..475b10c575b 100644 --- a/include/net/ndisc.h +++ b/include/net/ndisc.h @@ -66,8 +66,8 @@ struct rs_msg { struct ra_msg { struct icmp6hdr icmph; - __u32 reachable_time; - __u32 retrans_timer; + __be32 reachable_time; + __be32 retrans_timer; }; struct nd_opt_hdr { diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 052f99eaf2a..b41c5d8b278 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -232,7 +232,7 @@ static inline unsigned ipv6_addr_scope2type(unsigned scope) int __ipv6_addr_type(const struct in6_addr *addr) { - u32 st; + __be32 st; st = addr->s6_addr32[0]; diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 1eb1c7f261d..87c8f54872b 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -222,7 +222,7 @@ lookup_protocol: * the user to assign a number at socket * creation time automatically shares. */ - inet->sport = ntohs(inet->num); + inet->sport = htons(inet->num); sk->sk_prot->hash(sk); } if (sk->sk_prot->init) { @@ -342,7 +342,7 @@ int inet6_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) sk->sk_userlocks |= SOCK_BINDADDR_LOCK; if (snum) sk->sk_userlocks |= SOCK_BINDPORT_LOCK; - inet->sport = ntohs(inet->num); + inet->sport = htons(inet->num); inet->dport = 0; inet->daddr = 0; out: diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c index a67434af691..5c94fea90e9 100644 --- a/net/ipv6/datagram.c +++ b/net/ipv6/datagram.c @@ -207,7 +207,7 @@ out: } void ipv6_icmp_error(struct sock *sk, struct sk_buff *skb, int err, - u16 port, u32 info, u8 *payload) + __be16 port, u32 info, u8 *payload) { struct ipv6_pinfo *np = inet6_sk(sk); struct icmp6hdr *icmph = (struct icmp6hdr *)skb->h.raw; @@ -324,7 +324,7 @@ int ipv6_recv_error(struct sock *sk, struct msghdr *msg, int len) } else { ipv6_addr_set(&sin->sin6_addr, 0, 0, htonl(0xffff), - *(u32*)(skb->nh.raw + serr->addr_offset)); + *(__be32*)(skb->nh.raw + serr->addr_offset)); } } @@ -397,7 +397,7 @@ int datagram_recv_ctl(struct sock *sk, struct msghdr *msg, struct sk_buff *skb) } if (np->rxopt.bits.rxtclass) { - int tclass = (ntohl(*(u32 *)skb->nh.ipv6h) >> 20) & 0xff; + int tclass = (ntohl(*(__be32 *)skb->nh.ipv6h) >> 20) & 0xff; put_cmsg(msg, SOL_IPV6, IPV6_TCLASS, sizeof(tclass), &tclass); } diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index 27829cc4ce8..0711f92d6a1 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -655,7 +655,7 @@ static int ipv6_hop_jumbo(struct sk_buff **skbp, int optoff) goto drop; } - pkt_len = ntohl(*(u32*)(skb->nh.raw+optoff+2)); + pkt_len = ntohl(*(__be32*)(skb->nh.raw+optoff+2)); if (pkt_len <= IPV6_MAXPLEN) { IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, optoff+2); diff --git a/net/ipv6/exthdrs_core.c b/net/ipv6/exthdrs_core.c index 315bc1fbec3..21cbbbddaf4 100644 --- a/net/ipv6/exthdrs_core.c +++ b/net/ipv6/exthdrs_core.c @@ -77,7 +77,7 @@ int ipv6_skip_exthdr(const struct sk_buff *skb, int start, u8 *nexthdrp) if (hp == NULL) return -1; if (nexthdr == NEXTHDR_FRAGMENT) { - unsigned short _frag_off, *fp; + __be16 _frag_off, *fp; fp = skb_header_pointer(skb, start+offsetof(struct frag_hdr, frag_off), diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c index c598962eaa4..c700302ad51 100644 --- a/net/ipv6/inet6_connection_sock.c +++ b/net/ipv6/inet6_connection_sock.c @@ -55,16 +55,16 @@ EXPORT_SYMBOL_GPL(inet6_csk_bind_conflict); static u32 inet6_synq_hash(const struct in6_addr *raddr, const __be16 rport, const u32 rnd, const u16 synq_hsize) { - u32 a = raddr->s6_addr32[0]; - u32 b = raddr->s6_addr32[1]; - u32 c = raddr->s6_addr32[2]; + u32 a = (__force u32)raddr->s6_addr32[0]; + u32 b = (__force u32)raddr->s6_addr32[1]; + u32 c = (__force u32)raddr->s6_addr32[2]; a += JHASH_GOLDEN_RATIO; b += JHASH_GOLDEN_RATIO; c += rnd; __jhash_mix(a, b, c); - a += raddr->s6_addr32[3]; + a += (__force u32)raddr->s6_addr32[3]; b += (__force u32)rport; __jhash_mix(a, b, c); diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index f98ca30d7c1..bf526115e51 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -139,9 +139,9 @@ static __inline__ u32 fib6_new_sernum(void) * test bit */ -static __inline__ int addr_bit_set(void *token, int fn_bit) +static __inline__ __be32 addr_bit_set(void *token, int fn_bit) { - __u32 *addr = token; + __be32 *addr = token; return htonl(1 << ((~fn_bit)&0x1F)) & addr[fn_bit>>5]; } @@ -434,7 +434,7 @@ static struct fib6_node * fib6_add_1(struct fib6_node *root, void *addr, struct fib6_node *pn = NULL; struct rt6key *key; int bit; - int dir = 0; + __be32 dir = 0; __u32 sernum = fib6_new_sernum(); RT6_TRACE("fib6_add_1\n"); @@ -829,7 +829,7 @@ static struct fib6_node * fib6_lookup_1(struct fib6_node *root, struct lookup_args *args) { struct fib6_node *fn; - int dir; + __be32 dir; if (unlikely(args->offset == 0)) return NULL; diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 80a11909159..25bc5ed4910 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -66,7 +66,7 @@ MODULE_LICENSE("GPL"); #define HASH_SIZE 32 -#define HASH(addr) (((addr)->s6_addr32[0] ^ (addr)->s6_addr32[1] ^ \ +#define HASH(addr) ((__force u32)((addr)->s6_addr32[0] ^ (addr)->s6_addr32[1] ^ \ (addr)->s6_addr32[2] ^ (addr)->s6_addr32[3]) & \ (HASH_SIZE - 1)) diff --git a/net/ipv6/mip6.c b/net/ipv6/mip6.c index 7ccdc8fc5a3..be7dd7db65d 100644 --- a/net/ipv6/mip6.c +++ b/net/ipv6/mip6.c @@ -262,10 +262,10 @@ static int mip6_destopt_reject(struct xfrm_state *x, struct sk_buff *skb, struct sel.proto = fl->proto; sel.dport = xfrm_flowi_dport(fl); if (sel.dport) - sel.dport_mask = ~((__u16)0); + sel.dport_mask = htons(~0); sel.sport = xfrm_flowi_sport(fl); if (sel.sport) - sel.sport_mask = ~((__u16)0); + sel.sport_mask = htons(~0); sel.ifindex = fl->oif; err = km_report(IPPROTO_DSTOPTS, &sel, diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 1342be8b4cd..56ea9283730 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -1267,10 +1267,11 @@ skip_defrtr: } if (ndopts.nd_opts_mtu) { + __be32 n; u32 mtu; - memcpy(&mtu, ((u8*)(ndopts.nd_opts_mtu+1))+2, sizeof(mtu)); - mtu = ntohl(mtu); + memcpy(&n, ((u8*)(ndopts.nd_opts_mtu+1))+2, sizeof(mtu)); + mtu = ntohl(n); if (mtu < IPV6_MIN_MTU || mtu > skb->dev->mtu) { ND_PRINTK2(KERN_WARNING diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 204e02162d4..f63fb86d7c7 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -1481,7 +1481,8 @@ int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset, if (hp == NULL) return -EBADMSG; if (nexthdr == NEXTHDR_FRAGMENT) { - unsigned short _frag_off, *fp; + unsigned short _frag_off; + __be16 *fp; fp = skb_header_pointer(skb, start+offsetof(struct frag_hdr, frag_off), diff --git a/net/ipv6/netfilter/ip6t_LOG.c b/net/ipv6/netfilter/ip6t_LOG.c index 3cb6bb79cc0..f4857cf97f0 100644 --- a/net/ipv6/netfilter/ip6t_LOG.c +++ b/net/ipv6/netfilter/ip6t_LOG.c @@ -69,9 +69,9 @@ static void dump_packet(const struct nf_loginfo *info, /* Max length: 44 "LEN=65535 TC=255 HOPLIMIT=255 FLOWLBL=FFFFF " */ printk("LEN=%Zu TC=%u HOPLIMIT=%u FLOWLBL=%u ", ntohs(ih->payload_len) + sizeof(struct ipv6hdr), - (ntohl(*(u_int32_t *)ih) & 0x0ff00000) >> 20, + (ntohl(*(__be32 *)ih) & 0x0ff00000) >> 20, ih->hop_limit, - (ntohl(*(u_int32_t *)ih) & 0x000fffff)); + (ntohl(*(__be32 *)ih) & 0x000fffff)); fragment = 0; ptr = ip6hoff + sizeof(struct ipv6hdr); diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 813fa1d9686..b03040a2081 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -220,7 +220,7 @@ static int rawv6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) struct inet_sock *inet = inet_sk(sk); struct ipv6_pinfo *np = inet6_sk(sk); struct sockaddr_in6 *addr = (struct sockaddr_in6 *) uaddr; - __u32 v4addr = 0; + __be32 v4addr = 0; int addr_type; int err; diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index 3af0d5a6cee..6f9a9046510 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -77,7 +77,7 @@ struct frag_queue struct hlist_node list; struct list_head lru_list; /* lru list member */ - __u32 id; /* fragment id */ + __be32 id; /* fragment id */ struct in6_addr saddr; struct in6_addr daddr; @@ -125,28 +125,28 @@ static __inline__ void fq_unlink(struct frag_queue *fq) * callers should be careful not to use the hash value outside the ipfrag_lock * as doing so could race with ipfrag_hash_rnd being recalculated. */ -static unsigned int ip6qhashfn(u32 id, struct in6_addr *saddr, +static unsigned int ip6qhashfn(__be32 id, struct in6_addr *saddr, struct in6_addr *daddr) { u32 a, b, c; - a = saddr->s6_addr32[0]; - b = saddr->s6_addr32[1]; - c = saddr->s6_addr32[2]; + a = (__force u32)saddr->s6_addr32[0]; + b = (__force u32)saddr->s6_addr32[1]; + c = (__force u32)saddr->s6_addr32[2]; a += JHASH_GOLDEN_RATIO; b += JHASH_GOLDEN_RATIO; c += ip6_frag_hash_rnd; __jhash_mix(a, b, c); - a += saddr->s6_addr32[3]; - b += daddr->s6_addr32[0]; - c += daddr->s6_addr32[1]; + a += (__force u32)saddr->s6_addr32[3]; + b += (__force u32)daddr->s6_addr32[0]; + c += (__force u32)daddr->s6_addr32[1]; __jhash_mix(a, b, c); - a += daddr->s6_addr32[2]; - b += daddr->s6_addr32[3]; - c += id; + a += (__force u32)daddr->s6_addr32[2]; + b += (__force u32)daddr->s6_addr32[3]; + c += (__force u32)id; __jhash_mix(a, b, c); return c & (IP6Q_HASHSZ - 1); @@ -370,7 +370,7 @@ static struct frag_queue *ip6_frag_intern(struct frag_queue *fq_in) static struct frag_queue * -ip6_frag_create(u32 id, struct in6_addr *src, struct in6_addr *dst, +ip6_frag_create(__be32 id, struct in6_addr *src, struct in6_addr *dst, struct inet6_dev *idev) { struct frag_queue *fq; @@ -396,7 +396,7 @@ oom: } static __inline__ struct frag_queue * -fq_find(u32 id, struct in6_addr *src, struct in6_addr *dst, +fq_find(__be32 id, struct in6_addr *src, struct in6_addr *dst, struct inet6_dev *idev) { struct frag_queue *fq; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index a6472cb9054..0bf17a3cf08 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -440,7 +440,7 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len, if (pref == ICMPV6_ROUTER_PREF_INVALID) pref = ICMPV6_ROUTER_PREF_MEDIUM; - lifetime = htonl(rinfo->lifetime); + lifetime = ntohl(rinfo->lifetime); if (lifetime == 0xffffffff) { /* infinity */ } else if (lifetime > 0x7fffffff/HZ) { diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index be699f85b2c..85ff3dc4514 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -60,7 +60,7 @@ */ #define HASH_SIZE 16 -#define HASH(addr) ((addr^(addr>>4))&0xF) +#define HASH(addr) (((__force u32)addr^((__force u32)addr>>4))&0xF) static int ipip6_fb_tunnel_init(struct net_device *dev); static int ipip6_tunnel_init(struct net_device *dev); @@ -76,7 +76,7 @@ static struct ip_tunnel **tunnels[4] = { tunnels_wc, tunnels_l, tunnels_r, tunne static DEFINE_RWLOCK(ipip6_lock); -static struct ip_tunnel * ipip6_tunnel_lookup(u32 remote, u32 local) +static struct ip_tunnel * ipip6_tunnel_lookup(__be32 remote, __be32 local) { unsigned h0 = HASH(remote); unsigned h1 = HASH(local); @@ -102,8 +102,8 @@ static struct ip_tunnel * ipip6_tunnel_lookup(u32 remote, u32 local) static struct ip_tunnel ** ipip6_bucket(struct ip_tunnel *t) { - u32 remote = t->parms.iph.daddr; - u32 local = t->parms.iph.saddr; + __be32 remote = t->parms.iph.daddr; + __be32 local = t->parms.iph.saddr; unsigned h = 0; int prio = 0; @@ -144,8 +144,8 @@ static void ipip6_tunnel_link(struct ip_tunnel *t) static struct ip_tunnel * ipip6_tunnel_locate(struct ip_tunnel_parm *parms, int create) { - u32 remote = parms->iph.daddr; - u32 local = parms->iph.saddr; + __be32 remote = parms->iph.daddr; + __be32 local = parms->iph.saddr; struct ip_tunnel *t, **tp, *nt; struct net_device *dev; unsigned h = 0; @@ -405,9 +405,9 @@ out: /* Returns the embedded IPv4 address if the IPv6 address comes from 6to4 (RFC 3056) addr space */ -static inline u32 try_6to4(struct in6_addr *v6dst) +static inline __be32 try_6to4(struct in6_addr *v6dst) { - u32 dst = 0; + __be32 dst = 0; if (v6dst->s6_addr16[0] == htons(0x2002)) { /* 6to4 v6 addr has 16 bits prefix, 32 v4addr, 16 SLA, ... */ @@ -432,7 +432,7 @@ static int ipip6_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) struct net_device *tdev; /* Device to other host */ struct iphdr *iph; /* Our new IP header */ int max_headroom; /* The extra header space needed */ - u32 dst = tiph->daddr; + __be32 dst = tiph->daddr; int mtu; struct in6_addr *addr6; int addr_type; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index edef7eef834..53f270995d8 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1084,7 +1084,7 @@ static void tcp_v6_send_ack(struct tcp_timewait_sock *tw, struct sk_buff *buff; struct flowi fl; int tot_len = sizeof(struct tcphdr); - u32 *topt; + __be32 *topt; #ifdef CONFIG_TCP_MD5SIG struct tcp_md5sig_key *key; struct tcp_md5sig_key tw_key; @@ -1128,7 +1128,7 @@ static void tcp_v6_send_ack(struct tcp_timewait_sock *tw, t1->ack = 1; t1->window = htons(win); - topt = (u32*)(t1 + 1); + topt = (__be32 *)(t1 + 1); if (ts) { *topt++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 5a64027bf2f..e6e1f85f1bb 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -298,8 +298,8 @@ drop: } static struct sock *udp_v6_mcast_next(struct sock *sk, - u16 loc_port, struct in6_addr *loc_addr, - u16 rmt_port, struct in6_addr *rmt_addr, + __be16 loc_port, struct in6_addr *loc_addr, + __be16 rmt_port, struct in6_addr *rmt_addr, int dif) { struct hlist_node *node; -- cgit v1.2.3-70-g09d2 From ef296f56f8501f3ac68b8d44f63205f544126733 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 20:56:33 -0800 Subject: [IPV6]: __ipv6_addr_diff() annotations and cleanup. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/ipv6.h | 15 ++++----------- 1 file changed, 4 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 4ca9e93decc..00328b71a08 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -394,22 +394,15 @@ static inline int ipv6_addr_any(const struct in6_addr *a) */ static inline int __ipv6_addr_diff(const void *token1, const void *token2, int addrlen) { - const __u32 *a1 = token1, *a2 = token2; + const __be32 *a1 = token1, *a2 = token2; int i; addrlen >>= 2; for (i = 0; i < addrlen; i++) { - __u32 xb = a1[i] ^ a2[i]; - if (xb) { - int j = 31; - - xb = ntohl(xb); - while ((xb & (1 << j)) == 0) - j--; - - return (i * 32 + 31 - j); - } + __be32 xb = a1[i] ^ a2[i]; + if (xb) + return i * 32 + 32 - fls(ntohl(xb)); } /* -- cgit v1.2.3-70-g09d2 From 42d224aa170a4f7446cea6c972d9302d524545e5 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:11:08 -0800 Subject: [NETFILTER]: More trivial annotations. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/netfilter_ipv4/ip_conntrack_tftp.h | 2 +- net/ipv4/netfilter/ip_conntrack_amanda.c | 2 +- net/ipv4/netfilter/ip_conntrack_helper_h323.c | 4 ++-- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter_ipv4/ip_conntrack_tftp.h b/include/linux/netfilter_ipv4/ip_conntrack_tftp.h index cde9729aa17..a404fc0abf0 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_tftp.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_tftp.h @@ -4,7 +4,7 @@ #define TFTP_PORT 69 struct tftphdr { - u_int16_t opcode; + __be16 opcode; }; #define TFTP_OPCODE_READ 1 diff --git a/net/ipv4/netfilter/ip_conntrack_amanda.c b/net/ipv4/netfilter/ip_conntrack_amanda.c index 6c7383a8e42..7fca246615d 100644 --- a/net/ipv4/netfilter/ip_conntrack_amanda.c +++ b/net/ipv4/netfilter/ip_conntrack_amanda.c @@ -180,7 +180,7 @@ static struct ip_conntrack_helper amanda_helper = { .help = help, .name = "amanda", - .tuple = { .src = { .u = { __constant_htons(10080) } }, + .tuple = { .src = { .u = { .udp = {.port = __constant_htons(10080) } } }, .dst = { .protonum = IPPROTO_UDP }, }, .mask = { .src = { .u = { 0xFFFF } }, diff --git a/net/ipv4/netfilter/ip_conntrack_helper_h323.c b/net/ipv4/netfilter/ip_conntrack_helper_h323.c index 6cb9070cd0b..a06b340ea41 100644 --- a/net/ipv4/netfilter/ip_conntrack_helper_h323.c +++ b/net/ipv4/netfilter/ip_conntrack_helper_h323.c @@ -1153,7 +1153,7 @@ static struct ip_conntrack_helper ip_conntrack_helper_q931 = { .me = THIS_MODULE, .max_expected = H323_RTP_CHANNEL_MAX * 4 + 4 /* T.120 and H.245 */ , .timeout = 240, - .tuple = {.src = {.u = {__constant_htons(Q931_PORT)}}, + .tuple = {.src = {.u = {.tcp = {.port = __constant_htons(Q931_PORT)}}}, .dst = {.protonum = IPPROTO_TCP}}, .mask = {.src = {.u = {0xFFFF}}, .dst = {.protonum = 0xFF}}, @@ -1746,7 +1746,7 @@ static struct ip_conntrack_helper ip_conntrack_helper_ras = { .me = THIS_MODULE, .max_expected = 32, .timeout = 240, - .tuple = {.src = {.u = {__constant_htons(RAS_PORT)}}, + .tuple = {.src = {.u = {.tcp = {.port = __constant_htons(RAS_PORT)}}}, .dst = {.protonum = IPPROTO_UDP}}, .mask = {.src = {.u = {0xFFFE}}, .dst = {.protonum = 0xFF}}, -- cgit v1.2.3-70-g09d2 From 30d492da738a8d5f4ec884b3e1a13eef97714994 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:11:29 -0800 Subject: [ATM]: Annotations. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/atmarp.h | 2 +- include/linux/atmbr2684.h | 4 ++-- include/linux/atmmpc.h | 16 ++++++++-------- include/net/atmclip.h | 2 +- net/atm/br2684.c | 2 +- net/atm/clip.c | 10 +++++----- net/atm/lec.c | 6 +++--- net/atm/lec.h | 6 +++--- net/atm/mpc.c | 26 +++++++++++++------------- net/atm/mpc.h | 6 +++--- net/atm/mpoa_caches.c | 14 +++++++------- net/atm/mpoa_caches.h | 16 ++++++++-------- net/atm/mpoa_proc.c | 6 +++--- 13 files changed, 58 insertions(+), 58 deletions(-) (limited to 'include') diff --git a/include/linux/atmarp.h b/include/linux/atmarp.h index 24f82338f59..ee108f9e9cb 100644 --- a/include/linux/atmarp.h +++ b/include/linux/atmarp.h @@ -37,7 +37,7 @@ enum atmarp_ctrl_type { struct atmarp_ctrl { enum atmarp_ctrl_type type; /* message type */ int itf_num;/* interface number (if present) */ - uint32_t ip; /* IP address (act_need only) */ + __be32 ip; /* IP address (act_need only) */ }; #endif diff --git a/include/linux/atmbr2684.h b/include/linux/atmbr2684.h index 7981b733f1e..969fb6c9e1c 100644 --- a/include/linux/atmbr2684.h +++ b/include/linux/atmbr2684.h @@ -86,8 +86,8 @@ struct atm_backend_br2684 { * efficient per-if in/out filters, this support will be removed */ struct br2684_filter { - __u32 prefix; /* network byte order */ - __u32 netmask; /* 0 = disable filter */ + __be32 prefix; /* network byte order */ + __be32 netmask; /* 0 = disable filter */ }; struct br2684_filter_set { diff --git a/include/linux/atmmpc.h b/include/linux/atmmpc.h index 5fbfa68136d..ea1650425a1 100644 --- a/include/linux/atmmpc.h +++ b/include/linux/atmmpc.h @@ -13,7 +13,7 @@ struct atmmpc_ioc { int dev_num; - uint32_t ipaddr; /* the IP address of the shortcut */ + __be32 ipaddr; /* the IP address of the shortcut */ int type; /* ingress or egress */ }; @@ -21,8 +21,8 @@ typedef struct in_ctrl_info { uint8_t Last_NHRP_CIE_code; uint8_t Last_Q2931_cause_value; uint8_t eg_MPC_ATM_addr[ATM_ESA_LEN]; - uint32_t tag; - uint32_t in_dst_ip; /* IP address this ingress MPC sends packets to */ + __be32 tag; + __be32 in_dst_ip; /* IP address this ingress MPC sends packets to */ uint16_t holding_time; uint32_t request_id; } in_ctrl_info; @@ -30,10 +30,10 @@ typedef struct in_ctrl_info { typedef struct eg_ctrl_info { uint8_t DLL_header[256]; uint8_t DH_length; - uint32_t cache_id; - uint32_t tag; - uint32_t mps_ip; - uint32_t eg_dst_ip; /* IP address to which ingress MPC sends packets */ + __be32 cache_id; + __be32 tag; + __be32 mps_ip; + __be32 eg_dst_ip; /* IP address to which ingress MPC sends packets */ uint8_t in_MPC_data_ATM_addr[ATM_ESA_LEN]; uint16_t holding_time; } eg_ctrl_info; @@ -49,7 +49,7 @@ struct mpc_parameters { struct k_message { uint16_t type; - uint32_t ip_mask; + __be32 ip_mask; uint8_t MPS_ctrl[ATM_ESA_LEN]; union { in_ctrl_info in_info; diff --git a/include/net/atmclip.h b/include/net/atmclip.h index 90fcc98e676..b5a51a7bb36 100644 --- a/include/net/atmclip.h +++ b/include/net/atmclip.h @@ -36,7 +36,7 @@ struct clip_vcc { struct atmarp_entry { - u32 ip; /* IP address */ + __be32 ip; /* IP address */ struct clip_vcc *vccs; /* active VCCs; NULL if resolution is pending */ unsigned long expires; /* entry expiration time */ diff --git a/net/atm/br2684.c b/net/atm/br2684.c index d00cca97eb3..b04162f10d8 100644 --- a/net/atm/br2684.c +++ b/net/atm/br2684.c @@ -372,7 +372,7 @@ static int br2684_setfilt(struct atm_vcc *atmvcc, void __user *arg) /* Returns 1 if packet should be dropped */ static inline int -packet_fails_filter(u16 type, struct br2684_vcc *brvcc, struct sk_buff *skb) +packet_fails_filter(__be16 type, struct br2684_vcc *brvcc, struct sk_buff *skb) { if (brvcc->filter.netmask == 0) return 0; /* no filter in place */ diff --git a/net/atm/clip.c b/net/atm/clip.c index 7af2c411da8..1c416934b7c 100644 --- a/net/atm/clip.c +++ b/net/atm/clip.c @@ -54,7 +54,7 @@ static struct atm_vcc *atmarpd; static struct neigh_table clip_tbl; static struct timer_list idle_timer; -static int to_atmarpd(enum atmarp_ctrl_type type, int itf, unsigned long ip) +static int to_atmarpd(enum atmarp_ctrl_type type, int itf, __be32 ip) { struct sock *sk; struct atmarp_ctrl *ctrl; @@ -220,7 +220,7 @@ static void clip_push(struct atm_vcc *vcc, struct sk_buff *skb) || memcmp(skb->data, llc_oui, sizeof (llc_oui))) skb->protocol = htons(ETH_P_IP); else { - skb->protocol = ((u16 *) skb->data)[3]; + skb->protocol = ((__be16 *) skb->data)[3]; skb_pull(skb, RFC1483LLC_LEN); if (skb->protocol == htons(ETH_P_ARP)) { PRIV(skb->dev)->stats.rx_packets++; @@ -430,7 +430,7 @@ static int clip_start_xmit(struct sk_buff *skb, struct net_device *dev) here = skb_push(skb, RFC1483LLC_LEN); memcpy(here, llc_oui, sizeof(llc_oui)); - ((u16 *) here)[3] = skb->protocol; + ((__be16 *) here)[3] = skb->protocol; } atomic_add(skb->truesize, &sk_atm(vcc)->sk_wmem_alloc); ATM_SKB(skb)->atm_options = vcc->atm_options; @@ -509,7 +509,7 @@ static int clip_mkip(struct atm_vcc *vcc, int timeout) return 0; } -static int clip_setentry(struct atm_vcc *vcc, u32 ip) +static int clip_setentry(struct atm_vcc *vcc, __be32 ip) { struct neighbour *neigh; struct atmarp_entry *entry; @@ -752,7 +752,7 @@ static int clip_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) err = clip_mkip(vcc, arg); break; case ATMARP_SETENTRY: - err = clip_setentry(vcc, arg); + err = clip_setentry(vcc, (__force __be32)arg); break; case ATMARP_ENCAP: err = clip_encap(vcc, arg); diff --git a/net/atm/lec.c b/net/atm/lec.c index 66c57c1091a..9f5f931743b 100644 --- a/net/atm/lec.c +++ b/net/atm/lec.c @@ -204,9 +204,9 @@ static unsigned char *get_tr_dst(unsigned char *packet, unsigned char *rdesc) memset(rdesc, 0, ETH_ALEN); /* offset 4 comes from LAN destination field in LE control frames */ if (trh->rcf & htons((uint16_t) TR_RCF_DIR_BIT)) - memcpy(&rdesc[4], &trh->rseg[num_rdsc - 2], sizeof(uint16_t)); + memcpy(&rdesc[4], &trh->rseg[num_rdsc - 2], sizeof(__be16)); else { - memcpy(&rdesc[4], &trh->rseg[1], sizeof(uint16_t)); + memcpy(&rdesc[4], &trh->rseg[1], sizeof(__be16)); rdesc[5] = ((ntohs(trh->rseg[0]) & 0x000f) | (rdesc[5] & 0xf0)); } @@ -775,7 +775,7 @@ static void lec_push(struct atm_vcc *vcc, struct sk_buff *skb) unsigned char *src, *dst; atm_return(vcc, skb->truesize); - if (*(uint16_t *) skb->data == htons(priv->lecid) || + if (*(__be16 *) skb->data == htons(priv->lecid) || !priv->lecd || !(dev->flags & IFF_UP)) { /* * Probably looping back, or if lecd is missing, diff --git a/net/atm/lec.h b/net/atm/lec.h index 877f5093969..24cc95f8674 100644 --- a/net/atm/lec.h +++ b/net/atm/lec.h @@ -14,14 +14,14 @@ #define LEC_HEADER_LEN 16 struct lecdatahdr_8023 { - unsigned short le_header; + __be16 le_header; unsigned char h_dest[ETH_ALEN]; unsigned char h_source[ETH_ALEN]; - unsigned short h_type; + __be16 h_type; }; struct lecdatahdr_8025 { - unsigned short le_header; + __be16 le_header; unsigned char ac_pad; unsigned char fc; unsigned char h_dest[ETH_ALEN]; diff --git a/net/atm/mpc.c b/net/atm/mpc.c index 0d2b994af51..f15f5d84786 100644 --- a/net/atm/mpc.c +++ b/net/atm/mpc.c @@ -152,7 +152,7 @@ static struct mpoa_client *find_mpc_by_lec(struct net_device *dev) /* * Overwrites the old entry or makes a new one. */ -struct atm_mpoa_qos *atm_mpoa_add_qos(uint32_t dst_ip, struct atm_qos *qos) +struct atm_mpoa_qos *atm_mpoa_add_qos(__be32 dst_ip, struct atm_qos *qos) { struct atm_mpoa_qos *entry; @@ -177,7 +177,7 @@ struct atm_mpoa_qos *atm_mpoa_add_qos(uint32_t dst_ip, struct atm_qos *qos) return entry; } -struct atm_mpoa_qos *atm_mpoa_search_qos(uint32_t dst_ip) +struct atm_mpoa_qos *atm_mpoa_search_qos(__be32 dst_ip) { struct atm_mpoa_qos *qos; @@ -460,11 +460,11 @@ static int send_via_shortcut(struct sk_buff *skb, struct mpoa_client *mpc) in_cache_entry *entry; struct iphdr *iph; char *buff; - uint32_t ipaddr = 0; + __be32 ipaddr = 0; static struct { struct llc_snap_hdr hdr; - uint32_t tag; + __be32 tag; } tagged_llc_snap_hdr = { {0xaa, 0xaa, 0x03, {0x00, 0x00, 0x00}, {0x88, 0x4c}}, 0 @@ -559,7 +559,7 @@ static int atm_mpoa_vcc_attach(struct atm_vcc *vcc, void __user *arg) struct mpoa_client *mpc; struct atmmpc_ioc ioc_data; in_cache_entry *in_entry; - uint32_t ipaddr; + __be32 ipaddr; bytes_left = copy_from_user(&ioc_data, arg, sizeof(struct atmmpc_ioc)); if (bytes_left != 0) { @@ -638,7 +638,7 @@ static void mpc_push(struct atm_vcc *vcc, struct sk_buff *skb) struct sk_buff *new_skb; eg_cache_entry *eg; struct mpoa_client *mpc; - uint32_t tag; + __be32 tag; char *tmp; ddprintk("mpoa: (%s) mpc_push:\n", dev->name); @@ -683,7 +683,7 @@ static void mpc_push(struct atm_vcc *vcc, struct sk_buff *skb) } tmp = skb->data + sizeof(struct llc_snap_hdr); - tag = *(uint32_t *)tmp; + tag = *(__be32 *)tmp; eg = mpc->eg_ops->get_by_tag(tag, mpc); if (eg == NULL) { @@ -1029,7 +1029,7 @@ static int mpoa_event_listener(struct notifier_block *mpoa_notifier, unsigned lo static void MPOA_trigger_rcvd(struct k_message *msg, struct mpoa_client *mpc) { - uint32_t dst_ip = msg->content.in_info.in_dst_ip; + __be32 dst_ip = msg->content.in_info.in_dst_ip; in_cache_entry *entry; entry = mpc->in_ops->get(dst_ip, mpc); @@ -1066,7 +1066,7 @@ static void MPOA_trigger_rcvd(struct k_message *msg, struct mpoa_client *mpc) */ static void check_qos_and_open_shortcut(struct k_message *msg, struct mpoa_client *client, in_cache_entry *entry) { - uint32_t dst_ip = msg->content.in_info.in_dst_ip; + __be32 dst_ip = msg->content.in_info.in_dst_ip; struct atm_mpoa_qos *qos = atm_mpoa_search_qos(dst_ip); eg_cache_entry *eg_entry = client->eg_ops->get_by_src_ip(dst_ip, client); @@ -1102,7 +1102,7 @@ static void check_qos_and_open_shortcut(struct k_message *msg, struct mpoa_clien static void MPOA_res_reply_rcvd(struct k_message *msg, struct mpoa_client *mpc) { - uint32_t dst_ip = msg->content.in_info.in_dst_ip; + __be32 dst_ip = msg->content.in_info.in_dst_ip; in_cache_entry *entry = mpc->in_ops->get(dst_ip, mpc); dprintk("mpoa: (%s) MPOA_res_reply_rcvd: ip %u.%u.%u.%u\n", mpc->dev->name, NIPQUAD(dst_ip)); @@ -1148,8 +1148,8 @@ static void MPOA_res_reply_rcvd(struct k_message *msg, struct mpoa_client *mpc) static void ingress_purge_rcvd(struct k_message *msg, struct mpoa_client *mpc) { - uint32_t dst_ip = msg->content.in_info.in_dst_ip; - uint32_t mask = msg->ip_mask; + __be32 dst_ip = msg->content.in_info.in_dst_ip; + __be32 mask = msg->ip_mask; in_cache_entry *entry = mpc->in_ops->get_with_mask(dst_ip, mpc, mask); if(entry == NULL){ @@ -1173,7 +1173,7 @@ static void ingress_purge_rcvd(struct k_message *msg, struct mpoa_client *mpc) static void egress_purge_rcvd(struct k_message *msg, struct mpoa_client *mpc) { - uint32_t cache_id = msg->content.eg_info.cache_id; + __be32 cache_id = msg->content.eg_info.cache_id; eg_cache_entry *entry = mpc->eg_ops->get_by_cache_id(cache_id, mpc); if (entry == NULL) { diff --git a/net/atm/mpc.h b/net/atm/mpc.h index 3c7981a229e..51f460d005c 100644 --- a/net/atm/mpc.h +++ b/net/atm/mpc.h @@ -36,14 +36,14 @@ struct mpoa_client { struct atm_mpoa_qos { struct atm_mpoa_qos *next; - uint32_t ipaddr; + __be32 ipaddr; struct atm_qos qos; }; /* MPOA QoS operations */ -struct atm_mpoa_qos *atm_mpoa_add_qos(uint32_t dst_ip, struct atm_qos *qos); -struct atm_mpoa_qos *atm_mpoa_search_qos(uint32_t dst_ip); +struct atm_mpoa_qos *atm_mpoa_add_qos(__be32 dst_ip, struct atm_qos *qos); +struct atm_mpoa_qos *atm_mpoa_search_qos(__be32 dst_ip); int atm_mpoa_delete_qos(struct atm_mpoa_qos *qos); /* Display QoS entries. This is for the procfs */ diff --git a/net/atm/mpoa_caches.c b/net/atm/mpoa_caches.c index fbf13cdcf46..f3b99b38c91 100644 --- a/net/atm/mpoa_caches.c +++ b/net/atm/mpoa_caches.c @@ -22,7 +22,7 @@ #define ddprintk(format,args...) #endif -static in_cache_entry *in_cache_get(uint32_t dst_ip, +static in_cache_entry *in_cache_get(__be32 dst_ip, struct mpoa_client *client) { in_cache_entry *entry; @@ -42,9 +42,9 @@ static in_cache_entry *in_cache_get(uint32_t dst_ip, return NULL; } -static in_cache_entry *in_cache_get_with_mask(uint32_t dst_ip, +static in_cache_entry *in_cache_get_with_mask(__be32 dst_ip, struct mpoa_client *client, - uint32_t mask) + __be32 mask) { in_cache_entry *entry; @@ -84,7 +84,7 @@ static in_cache_entry *in_cache_get_by_vcc(struct atm_vcc *vcc, return NULL; } -static in_cache_entry *in_cache_add_entry(uint32_t dst_ip, +static in_cache_entry *in_cache_add_entry(__be32 dst_ip, struct mpoa_client *client) { in_cache_entry* entry = kmalloc(sizeof(in_cache_entry), GFP_KERNEL); @@ -319,7 +319,7 @@ static void in_destroy_cache(struct mpoa_client *mpc) return; } -static eg_cache_entry *eg_cache_get_by_cache_id(uint32_t cache_id, struct mpoa_client *mpc) +static eg_cache_entry *eg_cache_get_by_cache_id(__be32 cache_id, struct mpoa_client *mpc) { eg_cache_entry *entry; @@ -339,7 +339,7 @@ static eg_cache_entry *eg_cache_get_by_cache_id(uint32_t cache_id, struct mpoa_c } /* This can be called from any context since it saves CPU flags */ -static eg_cache_entry *eg_cache_get_by_tag(uint32_t tag, struct mpoa_client *mpc) +static eg_cache_entry *eg_cache_get_by_tag(__be32 tag, struct mpoa_client *mpc) { unsigned long flags; eg_cache_entry *entry; @@ -380,7 +380,7 @@ static eg_cache_entry *eg_cache_get_by_vcc(struct atm_vcc *vcc, struct mpoa_clie return NULL; } -static eg_cache_entry *eg_cache_get_by_src_ip(uint32_t ipaddr, struct mpoa_client *mpc) +static eg_cache_entry *eg_cache_get_by_src_ip(__be32 ipaddr, struct mpoa_client *mpc) { eg_cache_entry *entry; diff --git a/net/atm/mpoa_caches.h b/net/atm/mpoa_caches.h index 6c9886a03d0..84de977def2 100644 --- a/net/atm/mpoa_caches.h +++ b/net/atm/mpoa_caches.h @@ -29,12 +29,12 @@ typedef struct in_cache_entry { } in_cache_entry; struct in_cache_ops{ - in_cache_entry *(*add_entry)(uint32_t dst_ip, + in_cache_entry *(*add_entry)(__be32 dst_ip, struct mpoa_client *client); - in_cache_entry *(*get)(uint32_t dst_ip, struct mpoa_client *client); - in_cache_entry *(*get_with_mask)(uint32_t dst_ip, + in_cache_entry *(*get)(__be32 dst_ip, struct mpoa_client *client); + in_cache_entry *(*get_with_mask)(__be32 dst_ip, struct mpoa_client *client, - uint32_t mask); + __be32 mask); in_cache_entry *(*get_by_vcc)(struct atm_vcc *vcc, struct mpoa_client *client); void (*put)(in_cache_entry *entry); @@ -56,17 +56,17 @@ typedef struct eg_cache_entry{ struct atm_vcc *shortcut; uint32_t packets_rcvd; uint16_t entry_state; - uint32_t latest_ip_addr; /* The src IP address of the last packet */ + __be32 latest_ip_addr; /* The src IP address of the last packet */ struct eg_ctrl_info ctrl_info; atomic_t use; } eg_cache_entry; struct eg_cache_ops{ eg_cache_entry *(*add_entry)(struct k_message *msg, struct mpoa_client *client); - eg_cache_entry *(*get_by_cache_id)(uint32_t cache_id, struct mpoa_client *client); - eg_cache_entry *(*get_by_tag)(uint32_t cache_id, struct mpoa_client *client); + eg_cache_entry *(*get_by_cache_id)(__be32 cache_id, struct mpoa_client *client); + eg_cache_entry *(*get_by_tag)(__be32 cache_id, struct mpoa_client *client); eg_cache_entry *(*get_by_vcc)(struct atm_vcc *vcc, struct mpoa_client *client); - eg_cache_entry *(*get_by_src_ip)(uint32_t ipaddr, struct mpoa_client *client); + eg_cache_entry *(*get_by_src_ip)(__be32 ipaddr, struct mpoa_client *client); void (*put)(eg_cache_entry *entry); void (*remove_entry)(eg_cache_entry *entry, struct mpoa_client *client); void (*update)(eg_cache_entry *entry, uint16_t holding_time); diff --git a/net/atm/mpoa_proc.c b/net/atm/mpoa_proc.c index d37b8911b3a..3844c85d602 100644 --- a/net/atm/mpoa_proc.c +++ b/net/atm/mpoa_proc.c @@ -231,14 +231,14 @@ static int parse_qos(const char *buff) */ unsigned char ip[4]; int tx_pcr, tx_sdu, rx_pcr, rx_sdu; - uint32_t ipaddr; + __be32 ipaddr; struct atm_qos qos; memset(&qos, 0, sizeof(struct atm_qos)); if (sscanf(buff, "del %hhu.%hhu.%hhu.%hhu", ip, ip+1, ip+2, ip+3) == 4) { - ipaddr = *(uint32_t *)ip; + ipaddr = *(__be32 *)ip; return atm_mpoa_delete_qos(atm_mpoa_search_qos(ipaddr)); } @@ -250,7 +250,7 @@ static int parse_qos(const char *buff) ip, ip+1, ip+2, ip+3, &tx_pcr, &tx_sdu, &rx_pcr, &rx_sdu) != 8) return 0; - ipaddr = *(uint32_t *)ip; + ipaddr = *(__be32 *)ip; qos.txtp.traffic_class = ATM_CBR; qos.txtp.max_pcr = tx_pcr; qos.txtp.max_sdu = tx_sdu; -- cgit v1.2.3-70-g09d2 From 47c183fa5ea7feebc356da8ccbd9105a41f8e534 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:11:51 -0800 Subject: [BRIDGE]: Annotations. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/netfilter_bridge.h | 2 +- include/linux/netfilter_bridge/ebt_802_3.h | 10 +++++----- include/linux/netfilter_bridge/ebt_among.h | 2 +- include/linux/netfilter_bridge/ebt_arp.h | 14 +++++++------- include/linux/netfilter_bridge/ebt_ip.h | 8 ++++---- include/linux/netfilter_bridge/ebt_vlan.h | 2 +- include/linux/netfilter_bridge/ebtables.h | 2 +- net/bridge/br_netfilter.c | 2 +- net/bridge/netfilter/ebt_802_3.c | 2 +- net/bridge/netfilter/ebt_among.c | 22 +++++++++++----------- net/bridge/netfilter/ebt_arp.c | 6 +++--- net/bridge/netfilter/ebt_ip.c | 4 ++-- net/bridge/netfilter/ebt_log.c | 6 +++--- net/bridge/netfilter/ebt_vlan.c | 2 +- 14 files changed, 42 insertions(+), 42 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter_bridge.h b/include/linux/netfilter_bridge.h index 9a4dd11af86..6c4613f8ad7 100644 --- a/include/linux/netfilter_bridge.h +++ b/include/linux/netfilter_bridge.h @@ -64,7 +64,7 @@ static inline int nf_bridge_pad(const struct sk_buff *skb) struct bridge_skb_cb { union { - __u32 ipv4; + __be32 ipv4; } daddr; }; diff --git a/include/linux/netfilter_bridge/ebt_802_3.h b/include/linux/netfilter_bridge/ebt_802_3.h index b9f712c14a0..07f044ff1a6 100644 --- a/include/linux/netfilter_bridge/ebt_802_3.h +++ b/include/linux/netfilter_bridge/ebt_802_3.h @@ -28,21 +28,21 @@ struct hdr_ui { uint8_t ssap; uint8_t ctrl; uint8_t orig[3]; - uint16_t type; + __be16 type; }; struct hdr_ni { uint8_t dsap; uint8_t ssap; - uint16_t ctrl; + __be16 ctrl; uint8_t orig[3]; - uint16_t type; + __be16 type; }; struct ebt_802_3_hdr { uint8_t daddr[6]; uint8_t saddr[6]; - uint16_t len; + __be16 len; union { struct hdr_ui ui; struct hdr_ni ni; @@ -61,7 +61,7 @@ static inline struct ebt_802_3_hdr *ebt_802_3_hdr(const struct sk_buff *skb) struct ebt_802_3_info { uint8_t sap; - uint16_t type; + __be16 type; uint8_t bitmask; uint8_t invflags; }; diff --git a/include/linux/netfilter_bridge/ebt_among.h b/include/linux/netfilter_bridge/ebt_among.h index 307c1fed851..7654069233c 100644 --- a/include/linux/netfilter_bridge/ebt_among.h +++ b/include/linux/netfilter_bridge/ebt_among.h @@ -32,7 +32,7 @@ struct ebt_mac_wormhash_tuple { uint32_t cmp[2]; - uint32_t ip; + __be32 ip; }; struct ebt_mac_wormhash diff --git a/include/linux/netfilter_bridge/ebt_arp.h b/include/linux/netfilter_bridge/ebt_arp.h index 537ec6b487a..97e4dbde1f8 100644 --- a/include/linux/netfilter_bridge/ebt_arp.h +++ b/include/linux/netfilter_bridge/ebt_arp.h @@ -14,13 +14,13 @@ struct ebt_arp_info { - uint16_t htype; - uint16_t ptype; - uint16_t opcode; - uint32_t saddr; - uint32_t smsk; - uint32_t daddr; - uint32_t dmsk; + __be16 htype; + __be16 ptype; + __be16 opcode; + __be32 saddr; + __be32 smsk; + __be32 daddr; + __be32 dmsk; unsigned char smaddr[ETH_ALEN]; unsigned char smmsk[ETH_ALEN]; unsigned char dmaddr[ETH_ALEN]; diff --git a/include/linux/netfilter_bridge/ebt_ip.h b/include/linux/netfilter_bridge/ebt_ip.h index 7247385cdcb..d6847475bf2 100644 --- a/include/linux/netfilter_bridge/ebt_ip.h +++ b/include/linux/netfilter_bridge/ebt_ip.h @@ -28,10 +28,10 @@ /* the same values are used for the invflags */ struct ebt_ip_info { - uint32_t saddr; - uint32_t daddr; - uint32_t smsk; - uint32_t dmsk; + __be32 saddr; + __be32 daddr; + __be32 smsk; + __be32 dmsk; uint8_t tos; uint8_t protocol; uint8_t bitmask; diff --git a/include/linux/netfilter_bridge/ebt_vlan.h b/include/linux/netfilter_bridge/ebt_vlan.h index cb1fcc41565..1d98be4031e 100644 --- a/include/linux/netfilter_bridge/ebt_vlan.h +++ b/include/linux/netfilter_bridge/ebt_vlan.h @@ -10,7 +10,7 @@ struct ebt_vlan_info { uint16_t id; /* VLAN ID {1-4095} */ uint8_t prio; /* VLAN User Priority {0-7} */ - uint16_t encap; /* VLAN Encapsulated frame code {0-65535} */ + __be16 encap; /* VLAN Encapsulated frame code {0-65535} */ uint8_t bitmask; /* Args bitmask bit 1=1 - ID arg, bit 2=1 User-Priority arg, bit 3=1 encap*/ uint8_t invflags; /* Inverse bitmask bit 1=1 - inversed ID arg, diff --git a/include/linux/netfilter_bridge/ebtables.h b/include/linux/netfilter_bridge/ebtables.h index b1a7cc90877..e6ea70de24d 100644 --- a/include/linux/netfilter_bridge/ebtables.h +++ b/include/linux/netfilter_bridge/ebtables.h @@ -141,7 +141,7 @@ struct ebt_entry { /* this needs to be the first field */ unsigned int bitmask; unsigned int invflags; - uint16_t ethproto; + __be16 ethproto; /* the physical in-dev */ char in[IFNAMSIZ]; /* the logical in-dev */ diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c index ac181be13d8..2a5d31b1a19 100644 --- a/net/bridge/br_netfilter.c +++ b/net/bridge/br_netfilter.c @@ -381,7 +381,7 @@ static int check_hbh_len(struct sk_buff *skb) case IPV6_TLV_JUMBO: if (skb->nh.raw[off + 1] != 4 || (off & 3) != 2) goto bad; - pkt_len = ntohl(*(u32 *) (skb->nh.raw + off + 2)); + pkt_len = ntohl(*(__be32 *) (skb->nh.raw + off + 2)); if (pkt_len <= IPV6_MAXPLEN || skb->nh.ipv6h->payload_len) goto bad; diff --git a/net/bridge/netfilter/ebt_802_3.c b/net/bridge/netfilter/ebt_802_3.c index d42f63f5e9f..9abbc09ccdc 100644 --- a/net/bridge/netfilter/ebt_802_3.c +++ b/net/bridge/netfilter/ebt_802_3.c @@ -17,7 +17,7 @@ static int ebt_filter_802_3(const struct sk_buff *skb, const struct net_device * { struct ebt_802_3_info *info = (struct ebt_802_3_info *)data; struct ebt_802_3_hdr *hdr = ebt_802_3_hdr(skb); - uint16_t type = hdr->llc.ui.ctrl & IS_UI ? hdr->llc.ui.type : hdr->llc.ni.type; + __be16 type = hdr->llc.ui.ctrl & IS_UI ? hdr->llc.ui.type : hdr->llc.ni.type; if (info->bitmask & EBT_802_3_SAP) { if (FWINV(info->sap != hdr->llc.ui.ssap, EBT_802_3_SAP)) diff --git a/net/bridge/netfilter/ebt_among.c b/net/bridge/netfilter/ebt_among.c index a614485828a..ce97c4285f9 100644 --- a/net/bridge/netfilter/ebt_among.c +++ b/net/bridge/netfilter/ebt_among.c @@ -15,7 +15,7 @@ #include static int ebt_mac_wormhash_contains(const struct ebt_mac_wormhash *wh, - const char *mac, uint32_t ip) + const char *mac, __be32 ip) { /* You may be puzzled as to how this code works. * Some tricks were used, refer to @@ -70,7 +70,7 @@ static int ebt_mac_wormhash_check_integrity(const struct ebt_mac_wormhash return 0; } -static int get_ip_dst(const struct sk_buff *skb, uint32_t *addr) +static int get_ip_dst(const struct sk_buff *skb, __be32 *addr) { if (eth_hdr(skb)->h_proto == htons(ETH_P_IP)) { struct iphdr _iph, *ih; @@ -81,16 +81,16 @@ static int get_ip_dst(const struct sk_buff *skb, uint32_t *addr) *addr = ih->daddr; } else if (eth_hdr(skb)->h_proto == htons(ETH_P_ARP)) { struct arphdr _arph, *ah; - uint32_t buf, *bp; + __be32 buf, *bp; ah = skb_header_pointer(skb, 0, sizeof(_arph), &_arph); if (ah == NULL || - ah->ar_pln != sizeof(uint32_t) || + ah->ar_pln != sizeof(__be32) || ah->ar_hln != ETH_ALEN) return -1; bp = skb_header_pointer(skb, sizeof(struct arphdr) + - 2 * ETH_ALEN + sizeof(uint32_t), - sizeof(uint32_t), &buf); + 2 * ETH_ALEN + sizeof(__be32), + sizeof(__be32), &buf); if (bp == NULL) return -1; *addr = *bp; @@ -98,7 +98,7 @@ static int get_ip_dst(const struct sk_buff *skb, uint32_t *addr) return 0; } -static int get_ip_src(const struct sk_buff *skb, uint32_t *addr) +static int get_ip_src(const struct sk_buff *skb, __be32 *addr) { if (eth_hdr(skb)->h_proto == htons(ETH_P_IP)) { struct iphdr _iph, *ih; @@ -109,15 +109,15 @@ static int get_ip_src(const struct sk_buff *skb, uint32_t *addr) *addr = ih->saddr; } else if (eth_hdr(skb)->h_proto == htons(ETH_P_ARP)) { struct arphdr _arph, *ah; - uint32_t buf, *bp; + __be32 buf, *bp; ah = skb_header_pointer(skb, 0, sizeof(_arph), &_arph); if (ah == NULL || - ah->ar_pln != sizeof(uint32_t) || + ah->ar_pln != sizeof(__be32) || ah->ar_hln != ETH_ALEN) return -1; bp = skb_header_pointer(skb, sizeof(struct arphdr) + - ETH_ALEN, sizeof(uint32_t), &buf); + ETH_ALEN, sizeof(__be32), &buf); if (bp == NULL) return -1; *addr = *bp; @@ -133,7 +133,7 @@ static int ebt_filter_among(const struct sk_buff *skb, struct ebt_among_info *info = (struct ebt_among_info *) data; const char *dmac, *smac; const struct ebt_mac_wormhash *wh_dst, *wh_src; - uint32_t dip = 0, sip = 0; + __be32 dip = 0, sip = 0; wh_dst = ebt_among_wh_dst(info); wh_src = ebt_among_wh_src(info); diff --git a/net/bridge/netfilter/ebt_arp.c b/net/bridge/netfilter/ebt_arp.c index a6c81d9f73b..9c599800a90 100644 --- a/net/bridge/netfilter/ebt_arp.c +++ b/net/bridge/netfilter/ebt_arp.c @@ -35,10 +35,10 @@ static int ebt_filter_arp(const struct sk_buff *skb, const struct net_device *in return EBT_NOMATCH; if (info->bitmask & (EBT_ARP_SRC_IP | EBT_ARP_DST_IP)) { - uint32_t _addr, *ap; + __be32 _addr, *ap; /* IPv4 addresses are always 4 bytes */ - if (ah->ar_pln != sizeof(uint32_t)) + if (ah->ar_pln != sizeof(__be32)) return EBT_NOMATCH; if (info->bitmask & EBT_ARP_SRC_IP) { ap = skb_header_pointer(skb, sizeof(struct arphdr) + @@ -53,7 +53,7 @@ static int ebt_filter_arp(const struct sk_buff *skb, const struct net_device *in if (info->bitmask & EBT_ARP_DST_IP) { ap = skb_header_pointer(skb, sizeof(struct arphdr) + - 2*ah->ar_hln+sizeof(uint32_t), + 2*ah->ar_hln+sizeof(__be32), sizeof(_addr), &_addr); if (ap == NULL) return EBT_NOMATCH; diff --git a/net/bridge/netfilter/ebt_ip.c b/net/bridge/netfilter/ebt_ip.c index 65b665ce57b..e4c642448e1 100644 --- a/net/bridge/netfilter/ebt_ip.c +++ b/net/bridge/netfilter/ebt_ip.c @@ -20,8 +20,8 @@ #include struct tcpudphdr { - uint16_t src; - uint16_t dst; + __be16 src; + __be16 dst; }; static int ebt_filter_ip(const struct sk_buff *skb, const struct net_device *in, diff --git a/net/bridge/netfilter/ebt_log.c b/net/bridge/netfilter/ebt_log.c index 466ed3440b7..a184f879f25 100644 --- a/net/bridge/netfilter/ebt_log.c +++ b/net/bridge/netfilter/ebt_log.c @@ -38,8 +38,8 @@ static int ebt_log_check(const char *tablename, unsigned int hookmask, struct tcpudphdr { - uint16_t src; - uint16_t dst; + __be16 src; + __be16 dst; }; struct arppayload @@ -130,7 +130,7 @@ ebt_log_packet(unsigned int pf, unsigned int hooknum, * then log the ARP payload */ if (ah->ar_hrd == htons(1) && ah->ar_hln == ETH_ALEN && - ah->ar_pln == sizeof(uint32_t)) { + ah->ar_pln == sizeof(__be32)) { struct arppayload _arpp, *ap; ap = skb_header_pointer(skb, sizeof(_arph), diff --git a/net/bridge/netfilter/ebt_vlan.c b/net/bridge/netfilter/ebt_vlan.c index a2b452862b7..7ee37762296 100644 --- a/net/bridge/netfilter/ebt_vlan.c +++ b/net/bridge/netfilter/ebt_vlan.c @@ -55,7 +55,7 @@ ebt_filter_vlan(const struct sk_buff *skb, unsigned short id; /* VLAN ID, given from frame TCI */ unsigned char prio; /* user_priority, given from frame TCI */ /* VLAN encapsulated Type/Length field, given from orig frame */ - unsigned short encap; + __be16 encap; fp = skb_header_pointer(skb, 0, sizeof(_frame), &_frame); if (fp == NULL) -- cgit v1.2.3-70-g09d2 From a64b78a077a71c9b9c0c1b0be699083379783c3d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:12:29 -0800 Subject: [NET]: Annotate net_srandom(). Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/net.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/net.h b/include/linux/net.h index 15c733b816f..6f0dfeba509 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -196,7 +196,7 @@ extern struct socket *sockfd_lookup(int fd, int *err); extern int net_ratelimit(void); #define net_random() random32() -#define net_srandom(seed) srandom32(seed) +#define net_srandom(seed) srandom32((__force u32)seed) extern int kernel_sendmsg(struct socket *sock, struct msghdr *msg, struct kvec *vec, size_t num, size_t len); -- cgit v1.2.3-70-g09d2 From 2bc357987a6510e61d33f3b20fa989fb2b6a10b8 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:14:18 -0800 Subject: [NET]: Introduce types for checksums. New types - for 16bit checksums and "unfolded" 32bit variant. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/types.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/linux/types.h b/include/linux/types.h index 9f11fdd2bd7..745c409ebbb 100644 --- a/include/linux/types.h +++ b/include/linux/types.h @@ -182,6 +182,8 @@ typedef __u32 __bitwise __be32; typedef __u64 __bitwise __le64; typedef __u64 __bitwise __be64; #endif +typedef __u16 __bitwise __sum16; +typedef __u32 __bitwise __wsum; #ifdef __KERNEL__ typedef unsigned __bitwise__ gfp_t; -- cgit v1.2.3-70-g09d2 From 9be259aae5264511fe0a8b5e3d6711e0fd1d55df Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:14:53 -0800 Subject: [NET]: Alpha checksum annotations and cleanups. * sanitize prototypes and annotate * kill useless access_ok() in csum_partial_copy_from_user() (the only caller checks it already). * do_csum_partial_copy_from_user() is not needed now * replace htons(len) with len << 8 - they are the same wrt checksums on little-endian. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- arch/alpha/lib/checksum.c | 37 +++++++++++++++++-------------------- arch/alpha/lib/csum_partial_copy.c | 31 +++++++++---------------------- include/asm-alpha/checksum.h | 34 ++++++++++++++++------------------ 3 files changed, 42 insertions(+), 60 deletions(-) (limited to 'include') diff --git a/arch/alpha/lib/checksum.c b/arch/alpha/lib/checksum.c index 89044e6385f..ab3761c437a 100644 --- a/arch/alpha/lib/checksum.c +++ b/arch/alpha/lib/checksum.c @@ -41,28 +41,25 @@ static inline unsigned short from64to16(unsigned long x) * computes the checksum of the TCP/UDP pseudo-header * returns a 16-bit checksum, already complemented. */ -unsigned short int csum_tcpudp_magic(unsigned long saddr, - unsigned long daddr, +__sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { - return ~from64to16(saddr + daddr + sum + - ((unsigned long) ntohs(len) << 16) + - ((unsigned long) proto << 8)); + return (__force __sum16)~from64to16( + (__force u64)saddr + (__force u64)daddr + + (__force u64)sum + ((len + proto) << 8)); } -unsigned int csum_tcpudp_nofold(unsigned long saddr, - unsigned long daddr, +__wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { unsigned long result; - result = (saddr + daddr + sum + - ((unsigned long) ntohs(len) << 16) + - ((unsigned long) proto << 8)); + result = (__force u64)saddr + (__force u64)daddr + + (__force u64)sum + ((len + proto) << 8); /* Fold down to 32-bits so we don't lose in the typedef-less network stack. */ @@ -70,7 +67,7 @@ unsigned int csum_tcpudp_nofold(unsigned long saddr, result = (result & 0xffffffff) + (result >> 32); /* 33 to 32 */ result = (result & 0xffffffff) + (result >> 32); - return result; + return (__force __wsum)result; } /* @@ -146,9 +143,9 @@ out: * This is a version of ip_compute_csum() optimized for IP headers, * which always checksum on 4 octet boundaries. */ -unsigned short ip_fast_csum(unsigned char * iph, unsigned int ihl) +__sum16 ip_fast_csum(const void *iph, unsigned int ihl) { - return ~do_csum(iph,ihl*4); + return (__force __sum16)~do_csum(iph,ihl*4); } /* @@ -163,15 +160,15 @@ unsigned short ip_fast_csum(unsigned char * iph, unsigned int ihl) * * it's best to have buff aligned on a 32-bit boundary */ -unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum) +__wsum csum_partial(const void *buff, int len, __wsum sum) { unsigned long result = do_csum(buff, len); /* add in old sum, and carry.. */ - result += sum; + result += (__force u32)sum; /* 32+c bits -> 32 bits */ result = (result & 0xffffffff) + (result >> 32); - return result; + return (__force __wsum)result; } EXPORT_SYMBOL(csum_partial); @@ -180,7 +177,7 @@ EXPORT_SYMBOL(csum_partial); * this routine is used for miscellaneous IP-like checksums, mainly * in icmp.c */ -unsigned short ip_compute_csum(unsigned char * buff, int len) +__sum16 ip_compute_csum(const void *buff, int len) { - return ~from64to16(do_csum(buff,len)); + return (__force __sum16)~from64to16(do_csum(buff,len)); } diff --git a/arch/alpha/lib/csum_partial_copy.c b/arch/alpha/lib/csum_partial_copy.c index a37948f3037..4ca75c74ce9 100644 --- a/arch/alpha/lib/csum_partial_copy.c +++ b/arch/alpha/lib/csum_partial_copy.c @@ -329,11 +329,11 @@ csum_partial_cfu_unaligned(const unsigned long __user * src, return checksum; } -static unsigned int -do_csum_partial_copy_from_user(const char __user *src, char *dst, int len, - unsigned int sum, int *errp) +__wsum +csum_partial_copy_from_user(const void __user *src, void *dst, int len, + __wsum sum, int *errp) { - unsigned long checksum = (unsigned) sum; + unsigned long checksum = (__force u32) sum; unsigned long soff = 7 & (unsigned long) src; unsigned long doff = 7 & (unsigned long) dst; @@ -367,25 +367,12 @@ do_csum_partial_copy_from_user(const char __user *src, char *dst, int len, } checksum = from64to16 (checksum); } - return checksum; -} - -unsigned int -csum_partial_copy_from_user(const char __user *src, char *dst, int len, - unsigned int sum, int *errp) -{ - if (!access_ok(VERIFY_READ, src, len)) { - *errp = -EFAULT; - memset(dst, 0, len); - return sum; - } - - return do_csum_partial_copy_from_user(src, dst, len, sum, errp); + return (__force __wsum)checksum; } -unsigned int -csum_partial_copy_nocheck(const char __user *src, char *dst, int len, - unsigned int sum) +__wsum +csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum) { - return do_csum_partial_copy_from_user(src, dst, len, sum, NULL); + return csum_partial_copy_from_user((__force const void __user *)src, + dst, len, sum, NULL); } diff --git a/include/asm-alpha/checksum.h b/include/asm-alpha/checksum.h index a5c9f08447f..d3854bbf0a9 100644 --- a/include/asm-alpha/checksum.h +++ b/include/asm-alpha/checksum.h @@ -7,21 +7,20 @@ * This is a version of ip_compute_csum() optimized for IP headers, * which always checksum on 4 octet boundaries. */ -extern unsigned short ip_fast_csum(unsigned char * iph, unsigned int ihl); +extern __sum16 ip_fast_csum(const void *iph, unsigned int ihl); /* * computes the checksum of the TCP/UDP pseudo-header * returns a 16-bit checksum, already complemented */ -extern unsigned short int csum_tcpudp_magic(unsigned long saddr, - unsigned long daddr, +extern __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum); + __wsum sum); -unsigned int csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, +__wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum); + __wsum sum); /* * computes the checksum of a memory block at buff, length len, @@ -35,7 +34,7 @@ unsigned int csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, * * it's best to have buff aligned on a 32-bit boundary */ -extern unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum); +extern __wsum csum_partial(const void *buff, int len, __wsum sum); /* * the same as csum_partial, but copies from src while it @@ -44,9 +43,9 @@ extern unsigned int csum_partial(const unsigned char * buff, int len, unsigned i * here even more important to align src and dst on a 32-bit (or even * better 64-bit) boundary */ -unsigned int csum_partial_copy_from_user(const char __user *src, char *dst, int len, unsigned int sum, int *errp); +__wsum csum_partial_copy_from_user(const void __user *src, void *dst, int len, __wsum sum, int *errp); -unsigned int csum_partial_copy_nocheck(const char *src, char *dst, int len, unsigned int sum); +__wsum csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum); /* @@ -54,24 +53,23 @@ unsigned int csum_partial_copy_nocheck(const char *src, char *dst, int len, unsi * in icmp.c */ -extern unsigned short ip_compute_csum(unsigned char * buff, int len); +extern __sum16 ip_compute_csum(const void *buff, int len); /* * Fold a partial checksum without adding pseudo headers */ -static inline unsigned short csum_fold(unsigned int sum) +static inline __sum16 csum_fold(__wsum csum) { + u32 sum = (__force u32)csum; sum = (sum & 0xffff) + (sum >> 16); sum = (sum & 0xffff) + (sum >> 16); - return ~sum; + return (__force __sum16)~sum; } #define _HAVE_ARCH_IPV6_CSUM -extern unsigned short int csum_ipv6_magic(struct in6_addr *saddr, - struct in6_addr *daddr, - __u32 len, - unsigned short proto, - unsigned int sum); - +extern __sum16 csum_ipv6_magic(const struct in6_addr *saddr, + const struct in6_addr *daddr, + __u32 len, unsigned short proto, + __wsum sum); #endif -- cgit v1.2.3-70-g09d2 From 3532010bcf7699f2ce9a2baab58b4b9a5426d97e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:15:19 -0800 Subject: [NET]: Cris checksum annotations and cleanups. * sanitize prototypes and annotate * kill cast-as-lvalue abuses in csum_partial() * usual ntohs-equals-shift for checksum purposes Signed-off-by: Al Viro Signed-off-by: David S. Miller --- arch/cris/arch-v10/lib/old_checksum.c | 62 ++++++++++++++++++----------------- include/asm-cris/arch-v10/checksum.h | 10 +++--- include/asm-cris/arch-v32/checksum.h | 10 +++--- include/asm-cris/checksum.h | 34 +++++++++---------- 4 files changed, 57 insertions(+), 59 deletions(-) (limited to 'include') diff --git a/arch/cris/arch-v10/lib/old_checksum.c b/arch/cris/arch-v10/lib/old_checksum.c index 22a6f0aa9ce..497634a6482 100644 --- a/arch/cris/arch-v10/lib/old_checksum.c +++ b/arch/cris/arch-v10/lib/old_checksum.c @@ -47,39 +47,41 @@ #include -unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum) +__wsum csum_partial(const void *p, int len, __wsum __sum) { - /* - * Experiments with ethernet and slip connections show that buff - * is aligned on either a 2-byte or 4-byte boundary. - */ - const unsigned char *endMarker = buff + len; - const unsigned char *marker = endMarker - (len % 16); + u32 sum = (__force u32)__sum; + const u16 *buff = p; + /* + * Experiments with ethernet and slip connections show that buff + * is aligned on either a 2-byte or 4-byte boundary. + */ + const void *endMarker = p + len; + const void *marker = endMarker - (len % 16); #if 0 - if((int)buff & 0x3) - printk("unaligned buff %p\n", buff); - __delay(900); /* extra delay of 90 us to test performance hit */ + if((int)buff & 0x3) + printk("unaligned buff %p\n", buff); + __delay(900); /* extra delay of 90 us to test performance hit */ #endif - BITON; - while (buff < marker) { - sum += *((unsigned short *)buff)++; - sum += *((unsigned short *)buff)++; - sum += *((unsigned short *)buff)++; - sum += *((unsigned short *)buff)++; - sum += *((unsigned short *)buff)++; - sum += *((unsigned short *)buff)++; - sum += *((unsigned short *)buff)++; - sum += *((unsigned short *)buff)++; - } - marker = endMarker - (len % 2); - while(buff < marker) { - sum += *((unsigned short *)buff)++; - } - if(endMarker - buff > 0) { - sum += *buff; /* add extra byte seperately */ - } - BITOFF; - return(sum); + BITON; + while (buff < marker) { + sum += *buff++; + sum += *buff++; + sum += *buff++; + sum += *buff++; + sum += *buff++; + sum += *buff++; + sum += *buff++; + sum += *buff++; + } + marker = endMarker - (len % 2); + while (buff < marker) + sum += *buff++; + + if (endMarker > buff) + sum += *(const u8 *)buff; /* add extra byte seperately */ + + BITOFF; + return (__force __wsum)sum; } EXPORT_SYMBOL(csum_partial); diff --git a/include/asm-cris/arch-v10/checksum.h b/include/asm-cris/arch-v10/checksum.h index 633f234f336..b8000c5d7fe 100644 --- a/include/asm-cris/arch-v10/checksum.h +++ b/include/asm-cris/arch-v10/checksum.h @@ -8,11 +8,11 @@ * to split all of those into 16-bit components, then add. */ -static inline unsigned int -csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, unsigned short len, - unsigned short proto, unsigned int sum) +static inline __wsum +csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, + unsigned short proto, __wsum sum) { - int res; + __wsum res; __asm__ ("add.d %2, %0\n\t" "ax\n\t" "add.d %3, %0\n\t" @@ -21,7 +21,7 @@ csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, unsigned short len, "ax\n\t" "addq 0, %0\n" : "=r" (res) - : "0" (sum), "r" (daddr), "r" (saddr), "r" ((ntohs(len) << 16) + (proto << 8))); + : "0" (sum), "r" (daddr), "r" (saddr), "r" ((len + proto) << 8)); return res; } diff --git a/include/asm-cris/arch-v32/checksum.h b/include/asm-cris/arch-v32/checksum.h index 97ef89efea6..e5dcfce6e0d 100644 --- a/include/asm-cris/arch-v32/checksum.h +++ b/include/asm-cris/arch-v32/checksum.h @@ -9,11 +9,11 @@ * checksum. Which means it would be necessary to split all those into * 16-bit components and then add. */ -static inline unsigned int -csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, - unsigned short len, unsigned short proto, unsigned int sum) +static inline __wsum +csum_tcpudp_nofold(__be32 saddr, __be32 daddr, + unsigned short len, unsigned short proto, __wsum sum) { - int res; + __wsum res; __asm__ __volatile__ ("add.d %2, %0\n\t" "addc %3, %0\n\t" @@ -21,7 +21,7 @@ csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, "addc 0, %0\n\t" : "=r" (res) : "0" (sum), "r" (daddr), "r" (saddr), \ - "r" ((ntohs(len) << 16) + (proto << 8))); + "r" ((len + proto) << 8)); return res; } diff --git a/include/asm-cris/checksum.h b/include/asm-cris/checksum.h index 26a7719bbb8..180dbf2757b 100644 --- a/include/asm-cris/checksum.h +++ b/include/asm-cris/checksum.h @@ -17,7 +17,7 @@ * * it's best to have buff aligned on a 32-bit boundary */ -unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum); +__wsum csum_partial(const void *buff, int len, __wsum sum); /* * the same as csum_partial, but copies from src while it @@ -27,26 +27,23 @@ unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum) * better 64-bit) boundary */ -unsigned int csum_partial_copy_nocheck(const char *src, char *dst, - int len, unsigned int sum); +__wsum csum_partial_copy_nocheck(const void *src, void *dst, + int len, __wsum sum); /* * Fold a partial checksum into a word */ -static inline unsigned int csum_fold(unsigned int sum) +static inline __sum16 csum_fold(__wsum csum) { - /* the while loop is unnecessary really, it's always enough with two - iterations */ - - while(sum >> 16) - sum = (sum & 0xffff) + (sum >> 16); /* add in end-around carry */ - - return ~sum; + u32 sum = (__force u32)csum; + sum = (sum & 0xffff) + (sum >> 16); /* add in end-around carry */ + sum = (sum & 0xffff) + (sum >> 16); /* add in end-around carry */ + return (__force __sum16)~sum; } -extern unsigned int csum_partial_copy_from_user(const char *src, char *dst, - int len, unsigned int sum, +extern __wsum csum_partial_copy_from_user(const void __user *src, void *dst, + int len, __wsum sum, int *errptr); /* @@ -55,8 +52,7 @@ extern unsigned int csum_partial_copy_from_user(const char *src, char *dst, * */ -static inline unsigned short ip_fast_csum(unsigned char * iph, - unsigned int ihl) +static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl) { return csum_fold(csum_partial(iph, ihl * 4, 0)); } @@ -66,11 +62,10 @@ static inline unsigned short ip_fast_csum(unsigned char * iph, * returns a 16-bit checksum, already complemented */ -static inline unsigned short int csum_tcpudp_magic(unsigned long saddr, - unsigned long daddr, +static inline __sum16 int csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); } @@ -80,7 +75,8 @@ static inline unsigned short int csum_tcpudp_magic(unsigned long saddr, * in icmp.c */ -static inline unsigned short ip_compute_csum(unsigned char * buff, int len) { +static inline __sum16 ip_compute_csum(const void *buff, int len) +{ return csum_fold (csum_partial(buff, len, 0)); } -- cgit v1.2.3-70-g09d2 From 8042c44b8a6171ed75b7dd6a224df18d993f6094 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:15:40 -0800 Subject: [NET]: FRV checksum annotations. * sanitize prototypes and annotate * collapse csum_partial_copy Signed-off-by: Al Viro Signed-off-by: David S. Miller --- arch/frv/lib/checksum.c | 24 ++++++++++++------------ include/asm-frv/checksum.h | 41 +++++++++++++++++++---------------------- 2 files changed, 31 insertions(+), 34 deletions(-) (limited to 'include') diff --git a/arch/frv/lib/checksum.c b/arch/frv/lib/checksum.c index 2581a960d58..44e16d59bc1 100644 --- a/arch/frv/lib/checksum.c +++ b/arch/frv/lib/checksum.c @@ -104,15 +104,15 @@ out: * * it's best to have buff aligned on a 32-bit boundary */ -unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum) +__wsum csum_partial(const void *buff, int len, __wsum sum) { unsigned int result = do_csum(buff, len); /* add in old sum, and carry.. */ - result += sum; - if (sum > result) + result += (__force u32)sum; + if ((__force u32)sum > result) result += 1; - return result; + return (__force __wsum)result; } EXPORT_SYMBOL(csum_partial); @@ -121,9 +121,9 @@ EXPORT_SYMBOL(csum_partial); * this routine is used for miscellaneous IP-like checksums, mainly * in icmp.c */ -unsigned short ip_compute_csum(const unsigned char * buff, int len) +__sum16 ip_compute_csum(const void *buff, int len) { - return ~do_csum(buff, len); + return (__force __sum16)~do_csum(buff, len); } EXPORT_SYMBOL(ip_compute_csum); @@ -131,9 +131,9 @@ EXPORT_SYMBOL(ip_compute_csum); /* * copy from fs while checksumming, otherwise like csum_partial */ -unsigned int -csum_partial_copy_from_user(const char __user *src, char *dst, - int len, int sum, int *csum_err) +__wsum +csum_partial_copy_from_user(const void __user *src, void *dst, + int len, __wsum sum, int *csum_err) { int rem; @@ -156,11 +156,11 @@ EXPORT_SYMBOL(csum_partial_copy_from_user); /* * copy from ds while checksumming, otherwise like csum_partial */ -unsigned int -csum_partial_copy(const char *src, char *dst, int len, int sum) +__wsum +csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum) { memcpy(dst, src, len); return csum_partial(dst, len, sum); } -EXPORT_SYMBOL(csum_partial_copy); +EXPORT_SYMBOL(csum_partial_copy_nocheck); diff --git a/include/asm-frv/checksum.h b/include/asm-frv/checksum.h index 42bf0db2287..9b168985018 100644 --- a/include/asm-frv/checksum.h +++ b/include/asm-frv/checksum.h @@ -26,7 +26,7 @@ * * it's best to have buff aligned on a 32-bit boundary */ -unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum); +__wsum csum_partial(const void *buff, int len, __wsum sum); /* * the same as csum_partial, but copies from src while it @@ -35,7 +35,7 @@ unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum) * here even more important to align src and dst on a 32-bit (or even * better 64-bit) boundary */ -unsigned int csum_partial_copy(const char *src, char *dst, int len, int sum); +__wsum csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum); /* * the same as csum_partial_copy, but copies from user space. @@ -43,11 +43,8 @@ unsigned int csum_partial_copy(const char *src, char *dst, int len, int sum); * here even more important to align src and dst on a 32-bit (or even * better 64-bit) boundary */ -extern unsigned int csum_partial_copy_from_user(const char __user *src, char *dst, - int len, int sum, int *csum_err); - -#define csum_partial_copy_nocheck(src, dst, len, sum) \ - csum_partial_copy((src), (dst), (len), (sum)) +extern __wsum csum_partial_copy_from_user(const void __user *src, void *dst, + int len, __wsum sum, int *csum_err); /* * This is a version of ip_compute_csum() optimized for IP headers, @@ -55,7 +52,7 @@ extern unsigned int csum_partial_copy_from_user(const char __user *src, char *ds * */ static inline -unsigned short ip_fast_csum(unsigned char *iph, unsigned int ihl) +__sum16 ip_fast_csum(const void *iph, unsigned int ihl) { unsigned int tmp, inc, sum = 0; @@ -81,13 +78,13 @@ unsigned short ip_fast_csum(unsigned char *iph, unsigned int ihl) : "icc0", "icc1" ); - return ~sum; + return (__force __sum16)~sum; } /* * Fold a partial checksum */ -static inline unsigned int csum_fold(unsigned int sum) +static inline __sum16 csum_fold(__wsum sum) { unsigned int tmp; @@ -100,16 +97,16 @@ static inline unsigned int csum_fold(unsigned int sum) : "0"(sum) ); - return ~sum; + return (__force __sum16)~sum; } /* * computes the checksum of the TCP/UDP pseudo-header * returns a 16-bit checksum, already complemented */ -static inline unsigned int -csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, unsigned short len, - unsigned short proto, unsigned int sum) +static inline __wsum +csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, + unsigned short proto, __wsum sum) { asm(" addcc %1,%0,%0,icc0 \n" " addxcc %2,%0,%0,icc0 \n" @@ -122,9 +119,9 @@ csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, unsigned short len, return sum; } -static inline unsigned short int -csum_tcpudp_magic(unsigned long saddr, unsigned long daddr, unsigned short len, - unsigned short proto, unsigned int sum) +static inline __sum16 +csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, + unsigned short proto, __wsum sum) { return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); } @@ -133,12 +130,12 @@ csum_tcpudp_magic(unsigned long saddr, unsigned long daddr, unsigned short len, * this routine is used for miscellaneous IP-like checksums, mainly * in icmp.c */ -extern unsigned short ip_compute_csum(const unsigned char * buff, int len); +extern __sum16 ip_compute_csum(const void *buff, int len); #define _HAVE_ARCH_IPV6_CSUM -static inline unsigned short int -csum_ipv6_magic(struct in6_addr *saddr, struct in6_addr *daddr, - __u32 len, unsigned short proto, unsigned int sum) +static inline __sum16 +csum_ipv6_magic(const struct in6_addr *saddr, const struct in6_addr *daddr, + __u32 len, unsigned short proto, __wsum sum) { unsigned long tmp, tmp2; @@ -177,7 +174,7 @@ csum_ipv6_magic(struct in6_addr *saddr, struct in6_addr *daddr, : "icc0" ); - return ~sum; + return (__force __sum16)~sum; } #endif /* _ASM_CHECKSUM_H */ -- cgit v1.2.3-70-g09d2 From db521083bcb75505e9c3e21cbabe8274ee0daea6 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:16:07 -0800 Subject: [NET]: H8300 checksum annotations and cleanups. * sanitize prototypes and annotate * collapse csum_partial_copy NB: csum_partial() is almost certainly still buggy. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- arch/h8300/kernel/h8300_ksyms.c | 2 +- arch/h8300/lib/checksum.c | 29 +++++++++++++++++------------ include/asm-h8300/checksum.h | 31 ++++++++++++++----------------- 3 files changed, 32 insertions(+), 30 deletions(-) (limited to 'include') diff --git a/arch/h8300/kernel/h8300_ksyms.c b/arch/h8300/kernel/h8300_ksyms.c index 9b4be053de3..d1b15267ac8 100644 --- a/arch/h8300/kernel/h8300_ksyms.c +++ b/arch/h8300/kernel/h8300_ksyms.c @@ -39,7 +39,7 @@ EXPORT_SYMBOL(enable_irq); EXPORT_SYMBOL(disable_irq); /* Networking helper routines. */ -EXPORT_SYMBOL(csum_partial_copy); +EXPORT_SYMBOL(csum_partial_copy_nocheck); /* The following are special because they're not called explicitly (the C compiler generates them). Fortunately, diff --git a/arch/h8300/lib/checksum.c b/arch/h8300/lib/checksum.c index 5aa688d9242..bdc5b032acd 100644 --- a/arch/h8300/lib/checksum.c +++ b/arch/h8300/lib/checksum.c @@ -96,9 +96,9 @@ out: * This is a version of ip_compute_csum() optimized for IP headers, * which always checksum on 4 octet boundaries. */ -unsigned short ip_fast_csum(unsigned char * iph, unsigned int ihl) +__sum16 ip_fast_csum(const void *iph, unsigned int ihl) { - return ~do_csum(iph,ihl*4); + return (__force __sum16)~do_csum(iph,ihl*4); } /* @@ -113,15 +113,19 @@ unsigned short ip_fast_csum(unsigned char * iph, unsigned int ihl) * * it's best to have buff aligned on a 32-bit boundary */ -unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum) +/* + * Egads... That thing apparently assumes that *all* checksums it ever sees will + * be folded. Very likely a bug. + */ +__wsum csum_partial(const void *buff, int len, __wsum sum) { unsigned int result = do_csum(buff, len); /* add in old sum, and carry.. */ - result += sum; + result += (__force u32)sum; /* 16+c bits -> 16 bits */ result = (result & 0xffff) + (result >> 16); - return result; + return (__force __wsum)result; } EXPORT_SYMBOL(csum_partial); @@ -130,20 +134,21 @@ EXPORT_SYMBOL(csum_partial); * this routine is used for miscellaneous IP-like checksums, mainly * in icmp.c */ -unsigned short ip_compute_csum(const unsigned char * buff, int len) +__sum16 ip_compute_csum(const void *buff, int len) { - return ~do_csum(buff,len); + return (__force __sum16)~do_csum(buff,len); } /* * copy from fs while checksumming, otherwise like csum_partial */ -unsigned int -csum_partial_copy_from_user(const char *src, char *dst, int len, int sum, int *csum_err) +__wsum +csum_partial_copy_from_user(const void __user *src, void *dst, int len, + __wsum sum, int *csum_err) { if (csum_err) *csum_err = 0; - memcpy(dst, src, len); + memcpy(dst, (__force const void *)src, len); return csum_partial(dst, len, sum); } @@ -151,8 +156,8 @@ csum_partial_copy_from_user(const char *src, char *dst, int len, int sum, int *c * copy from ds while checksumming, otherwise like csum_partial */ -unsigned int -csum_partial_copy(const char *src, char *dst, int len, int sum) +__wsum +csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum) { memcpy(dst, src, len); return csum_partial(dst, len, sum); diff --git a/include/asm-h8300/checksum.h b/include/asm-h8300/checksum.h index 3051931dd30..98724e12508 100644 --- a/include/asm-h8300/checksum.h +++ b/include/asm-h8300/checksum.h @@ -13,7 +13,7 @@ * * it's best to have buff aligned on a 32-bit boundary */ -unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum); +__wsum csum_partial(const void *buff, int len, __wsum sum); /* * the same as csum_partial, but copies from src while it @@ -23,7 +23,7 @@ unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum) * better 64-bit) boundary */ -unsigned int csum_partial_copy(const char *src, char *dst, int len, int sum); +__wsum csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum); /* @@ -33,20 +33,17 @@ unsigned int csum_partial_copy(const char *src, char *dst, int len, int sum); * better 64-bit) boundary */ -extern unsigned int csum_partial_copy_from_user(const char *src, char *dst, - int len, int sum, int *csum_err); +extern __wsum csum_partial_copy_from_user(const void __user *src, void *dst, + int len, __wsum sum, int *csum_err); -#define csum_partial_copy_nocheck(src, dst, len, sum) \ - csum_partial_copy((src), (dst), (len), (sum)) - -unsigned short ip_fast_csum(unsigned char *iph, unsigned int ihl); +__sum16 ip_fast_csum(const void *iph, unsigned int ihl); /* * Fold a partial checksum */ -static inline unsigned int csum_fold(unsigned int sum) +static inline __sum16 csum_fold(__wsum sum) { __asm__("mov.l %0,er0\n\t" "add.w e0,r0\n\t" @@ -58,7 +55,7 @@ static inline unsigned int csum_fold(unsigned int sum) : "=r"(sum) : "0"(sum) : "er0"); - return ~sum; + return (__force __sum16)~sum; } @@ -67,9 +64,9 @@ static inline unsigned int csum_fold(unsigned int sum) * returns a 16-bit checksum, already complemented */ -static inline unsigned int -csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, unsigned short len, - unsigned short proto, unsigned int sum) +static inline __wsum +csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, + unsigned short proto, __wsum sum) { __asm__ ("sub.l er0,er0\n\t" "add.l %2,%0\n\t" @@ -88,9 +85,9 @@ csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, unsigned short len, return sum; } -static inline unsigned short int -csum_tcpudp_magic(unsigned long saddr, unsigned long daddr, unsigned short len, - unsigned short proto, unsigned int sum) +static inline __sum16 +csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, + unsigned short proto, __wsum sum) { return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); } @@ -100,6 +97,6 @@ csum_tcpudp_magic(unsigned long saddr, unsigned long daddr, unsigned short len, * in icmp.c */ -extern unsigned short ip_compute_csum(const unsigned char * buff, int len); +extern __sum16 ip_compute_csum(const void *buff, int len); #endif /* _H8300_CHECKSUM_H */ -- cgit v1.2.3-70-g09d2 From 322529961e3b3e64fdf1a3e46a45294456c91acf Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:16:30 -0800 Subject: [NET]: IA64 checksum annotations and cleanups. * sanitize prototypes, annotate * ntohs -> shift in checksum calculations * kill access_ok() in csum_partial_copy_from_user * collapse do_csum_partial_copy_from_user Signed-off-by: Al Viro Signed-off-by: David S. Miller --- arch/ia64/lib/checksum.c | 38 ++++++++++++++++++-------------------- arch/ia64/lib/csum_partial_copy.c | 31 +++++++++---------------------- include/asm-ia64/checksum.h | 31 ++++++++++++++----------------- 3 files changed, 41 insertions(+), 59 deletions(-) (limited to 'include') diff --git a/arch/ia64/lib/checksum.c b/arch/ia64/lib/checksum.c index beb11721d9f..4411d9baeb2 100644 --- a/arch/ia64/lib/checksum.c +++ b/arch/ia64/lib/checksum.c @@ -33,32 +33,32 @@ from64to16 (unsigned long x) * computes the checksum of the TCP/UDP pseudo-header * returns a 16-bit checksum, already complemented. */ -unsigned short int -csum_tcpudp_magic (unsigned long saddr, unsigned long daddr, unsigned short len, - unsigned short proto, unsigned int sum) +__sum16 +csum_tcpudp_magic (__be32 saddr, __be32 daddr, unsigned short len, + unsigned short proto, __wsum sum) { - return ~from64to16(saddr + daddr + sum + ((unsigned long) ntohs(len) << 16) + - ((unsigned long) proto << 8)); + return (__force __sum16)~from64to16( + (__force u64)saddr + (__force u64)daddr + + (__force u64)sum + ((len + proto) << 8)); } EXPORT_SYMBOL(csum_tcpudp_magic); -unsigned int -csum_tcpudp_nofold (unsigned long saddr, unsigned long daddr, unsigned short len, - unsigned short proto, unsigned int sum) +__wsum +csum_tcpudp_nofold (__be32 saddr, __be32 daddr, unsigned short len, + unsigned short proto, __wsum sum) { unsigned long result; - result = (saddr + daddr + sum + - ((unsigned long) ntohs(len) << 16) + - ((unsigned long) proto << 8)); + result = (__force u64)saddr + (__force u64)daddr + + (__force u64)sum + ((len + proto) << 8); /* Fold down to 32-bits so we don't lose in the typedef-less network stack. */ /* 64 to 33 */ result = (result & 0xffffffff) + (result >> 32); /* 33 to 32 */ result = (result & 0xffffffff) + (result >> 32); - return result; + return (__force __wsum)result; } extern unsigned long do_csum (const unsigned char *, long); @@ -75,16 +75,15 @@ extern unsigned long do_csum (const unsigned char *, long); * * it's best to have buff aligned on a 32-bit boundary */ -unsigned int -csum_partial (const unsigned char * buff, int len, unsigned int sum) +__wsum csum_partial(const void *buff, int len, __wsum sum) { - unsigned long result = do_csum(buff, len); + u64 result = do_csum(buff, len); /* add in old sum, and carry.. */ - result += sum; + result += (__force u32)sum; /* 32+c bits -> 32 bits */ result = (result & 0xffffffff) + (result >> 32); - return result; + return (__force __wsum)result; } EXPORT_SYMBOL(csum_partial); @@ -93,10 +92,9 @@ EXPORT_SYMBOL(csum_partial); * this routine is used for miscellaneous IP-like checksums, mainly * in icmp.c */ -unsigned short -ip_compute_csum (unsigned char * buff, int len) +__sum16 ip_compute_csum (const void *buff, int len) { - return ~do_csum(buff,len); + return (__force __sum16)~do_csum(buff,len); } EXPORT_SYMBOL(ip_compute_csum); diff --git a/arch/ia64/lib/csum_partial_copy.c b/arch/ia64/lib/csum_partial_copy.c index 36866e8a5d2..503dfe6d145 100644 --- a/arch/ia64/lib/csum_partial_copy.c +++ b/arch/ia64/lib/csum_partial_copy.c @@ -104,9 +104,9 @@ out: */ extern unsigned long do_csum(const unsigned char *, long); -static unsigned int -do_csum_partial_copy_from_user (const unsigned char __user *src, unsigned char *dst, - int len, unsigned int psum, int *errp) +__wsum +csum_partial_copy_from_user(const void __user *src, void *dst, + int len, __wsum psum, int *errp) { unsigned long result; @@ -122,30 +122,17 @@ do_csum_partial_copy_from_user (const unsigned char __user *src, unsigned char * result = do_csum(dst, len); /* add in old sum, and carry.. */ - result += psum; + result += (__force u32)psum; /* 32+c bits -> 32 bits */ result = (result & 0xffffffff) + (result >> 32); - return result; -} - -unsigned int -csum_partial_copy_from_user (const unsigned char __user *src, unsigned char *dst, - int len, unsigned int sum, int *errp) -{ - if (!access_ok(VERIFY_READ, src, len)) { - *errp = -EFAULT; - memset(dst, 0, len); - return sum; - } - - return do_csum_partial_copy_from_user(src, dst, len, sum, errp); + return (__force __wsum)result; } -unsigned int -csum_partial_copy_nocheck(const unsigned char __user *src, unsigned char *dst, - int len, unsigned int sum) +__wsum +csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum) { - return do_csum_partial_copy_from_user(src, dst, len, sum, NULL); + return csum_partial_copy_from_user((__force const void __user *)src, + dst, len, sum, NULL); } EXPORT_SYMBOL(csum_partial_copy_nocheck); diff --git a/include/asm-ia64/checksum.h b/include/asm-ia64/checksum.h index 1f230ff8ea8..bd40f4756ce 100644 --- a/include/asm-ia64/checksum.h +++ b/include/asm-ia64/checksum.h @@ -10,23 +10,21 @@ * This is a version of ip_compute_csum() optimized for IP headers, * which always checksum on 4 octet boundaries. */ -extern unsigned short ip_fast_csum (unsigned char * iph, unsigned int ihl); +extern __sum16 ip_fast_csum(const void *iph, unsigned int ihl); /* * Computes the checksum of the TCP/UDP pseudo-header returns a 16-bit * checksum, already complemented */ -extern unsigned short int csum_tcpudp_magic (unsigned long saddr, - unsigned long daddr, +extern __sum16 csum_tcpudp_magic (__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum); + __wsum sum); -extern unsigned int csum_tcpudp_nofold (unsigned long saddr, - unsigned long daddr, +extern __wsum csum_tcpudp_nofold (__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum); + __wsum sum); /* * Computes the checksum of a memory block at buff, length len, @@ -40,8 +38,7 @@ extern unsigned int csum_tcpudp_nofold (unsigned long saddr, * * it's best to have buff aligned on a 32-bit boundary */ -extern unsigned int csum_partial (const unsigned char * buff, int len, - unsigned int sum); +extern __wsum csum_partial(const void *buff, int len, __wsum sum); /* * Same as csum_partial, but copies from src while it checksums. @@ -49,28 +46,28 @@ extern unsigned int csum_partial (const unsigned char * buff, int len, * Here it is even more important to align src and dst on a 32-bit (or * even better 64-bit) boundary. */ -extern unsigned int csum_partial_copy_from_user (const char *src, char *dst, - int len, unsigned int sum, +extern __wsum csum_partial_copy_from_user(const void __user *src, void *dst, + int len, __wsum sum, int *errp); -extern unsigned int csum_partial_copy_nocheck (const char *src, char *dst, - int len, unsigned int sum); +extern __wsum csum_partial_copy_nocheck(const void *src, void *dst, + int len, __wsum sum); /* * This routine is used for miscellaneous IP-like checksums, mainly in * icmp.c */ -extern unsigned short ip_compute_csum (unsigned char *buff, int len); +extern __sum16 ip_compute_csum(const void *buff, int len); /* * Fold a partial checksum without adding pseudo headers. */ -static inline unsigned short -csum_fold (unsigned int sum) +static inline __sum16 csum_fold(__wsum csum) { + u32 sum = (__force u32)csum; sum = (sum & 0xffff) + (sum >> 16); sum = (sum & 0xffff) + (sum >> 16); - return ~sum; + return (__force __sum16)~sum; } #endif /* _ASM_IA64_CHECKSUM_H */ -- cgit v1.2.3-70-g09d2 From 85d20dee20f0958df1615e73698f6b0c525812f7 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:16:55 -0800 Subject: [NET]: M32R checksum annotations and cleanups. * sanitize prototypes, annotate * ntohs -> shift in checksum calculations in l-e case * kill shift-by-16 in checksum calculations Signed-off-by: Al Viro Signed-off-by: David S. Miller --- arch/m32r/lib/csum_partial_copy.c | 12 ++++----- include/asm-m32r/checksum.h | 52 ++++++++++++++++++--------------------- 2 files changed, 29 insertions(+), 35 deletions(-) (limited to 'include') diff --git a/arch/m32r/lib/csum_partial_copy.c b/arch/m32r/lib/csum_partial_copy.c index 3d5f0614585..5596f3df833 100644 --- a/arch/m32r/lib/csum_partial_copy.c +++ b/arch/m32r/lib/csum_partial_copy.c @@ -27,9 +27,8 @@ /* * Copy while checksumming, otherwise like csum_partial */ -unsigned int -csum_partial_copy_nocheck (const unsigned char *src, unsigned char *dst, - int len, unsigned int sum) +__wsum +csum_partial_copy_nocheck (const void *src, void *dst, int len, __wsum sum) { sum = csum_partial(src, len, sum); memcpy(dst, src, len); @@ -42,10 +41,9 @@ EXPORT_SYMBOL(csum_partial_copy_nocheck); * Copy from userspace and compute checksum. If we catch an exception * then zero the rest of the buffer. */ -unsigned int -csum_partial_copy_from_user (const unsigned char __user *src, - unsigned char *dst, - int len, unsigned int sum, int *err_ptr) +__wsum +csum_partial_copy_from_user (const void __user *src, void *dst, + int len, __wsum sum, int *err_ptr) { int missing; diff --git a/include/asm-m32r/checksum.h b/include/asm-m32r/checksum.h index 877ebf46e9f..a7a7c4f44ab 100644 --- a/include/asm-m32r/checksum.h +++ b/include/asm-m32r/checksum.h @@ -31,8 +31,7 @@ * * it's best to have buff aligned on a 32-bit boundary */ -asmlinkage unsigned int csum_partial(const unsigned char *buff, - int len, unsigned int sum); +asmlinkage __wsum csum_partial(const void *buff, int len, __wsum sum); /* * The same as csum_partial, but copies from src while it checksums. @@ -40,24 +39,22 @@ asmlinkage unsigned int csum_partial(const unsigned char *buff, * Here even more important to align src and dst on a 32-bit (or even * better 64-bit) boundary */ -extern unsigned int csum_partial_copy_nocheck(const unsigned char *src, - unsigned char *dst, - int len, unsigned int sum); +extern __wsum csum_partial_copy_nocheck(const void *src, void *dst, + int len, __wsum sum); /* * This is a new version of the above that records errors it finds in *errp, * but continues and zeros thre rest of the buffer. */ -extern unsigned int csum_partial_copy_from_user(const unsigned char __user *src, - unsigned char *dst, - int len, unsigned int sum, +extern __wsum csum_partial_copy_from_user(const void __user *src, void *dst, + int len, __wsum sum, int *err_ptr); /* * Fold a partial checksum */ -static inline unsigned int csum_fold(unsigned int sum) +static inline __sum16 csum_fold(__wsum sum) { unsigned long tmpreg; __asm__( @@ -72,16 +69,17 @@ static inline unsigned int csum_fold(unsigned int sum) : "0" (sum) : "cbit" ); - return sum; + return (__force __sum16)sum; } /* * This is a version of ip_compute_csum() optimized for IP headers, * which always checksum on 4 octet boundaries. */ -static inline unsigned short ip_fast_csum(unsigned char * iph, - unsigned int ihl) { - unsigned long sum, tmpreg0, tmpreg1; +static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl) +{ + unsigned long tmpreg0, tmpreg1; + __wsum sum; __asm__ __volatile__( " ld %0, @%1+ \n" @@ -115,16 +113,15 @@ static inline unsigned short ip_fast_csum(unsigned char * iph, return csum_fold(sum); } -static inline unsigned long csum_tcpudp_nofold(unsigned long saddr, - unsigned long daddr, +static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { #if defined(__LITTLE_ENDIAN) - unsigned long len_proto = (ntohs(len)<<16)+proto*256; + unsigned long len_proto = (proto + len) << 8; #else - unsigned long len_proto = (proto<<16)+len; + unsigned long len_proto = proto + len; #endif unsigned long tmpreg; @@ -147,11 +144,10 @@ static inline unsigned long csum_tcpudp_nofold(unsigned long saddr, * computes the checksum of the TCP/UDP pseudo-header * returns a 16-bit checksum, already complemented */ -static inline unsigned short int csum_tcpudp_magic(unsigned long saddr, - unsigned long daddr, +static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); } @@ -161,16 +157,16 @@ static inline unsigned short int csum_tcpudp_magic(unsigned long saddr, * in icmp.c */ -static inline unsigned short ip_compute_csum(unsigned char * buff, int len) { +static inline __sum16 ip_compute_csum(const void *buff, int len) +{ return csum_fold (csum_partial(buff, len, 0)); } #define _HAVE_ARCH_IPV6_CSUM -static inline unsigned short int csum_ipv6_magic(struct in6_addr *saddr, - struct in6_addr *daddr, - __u16 len, - unsigned short proto, - unsigned int sum) +static inline __sum16 csum_ipv6_magic(const struct in6_addr *saddr, + const struct in6_addr *daddr, + __u32 len, unsigned short proto, + __wsum sum) { unsigned long tmpreg0, tmpreg1, tmpreg2, tmpreg3; __asm__( @@ -197,7 +193,7 @@ static inline unsigned short int csum_ipv6_magic(struct in6_addr *saddr, : "=&r" (sum), "=&r" (tmpreg0), "=&r" (tmpreg1), "=&r" (tmpreg2), "=&r" (tmpreg3) : "r" (saddr), "r" (daddr), - "r" (htonl((__u32) (len))), "r" (htonl(proto)), "0" (sum) + "r" (htonl(len)), "r" (htonl(proto)), "0" (sum) : "cbit" ); -- cgit v1.2.3-70-g09d2 From 2061acaaae0e165f0104ec9d327a02addbcabd62 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:17:19 -0800 Subject: [NET]: M68K checksum annotations and cleanups. * sanitize prototypes, annotate Signed-off-by: Al Viro Signed-off-by: David S. Miller --- arch/m68k/lib/checksum.c | 13 ++++++------- include/asm-m68k/checksum.h | 46 ++++++++++++++++++++++----------------------- 2 files changed, 28 insertions(+), 31 deletions(-) (limited to 'include') diff --git a/arch/m68k/lib/checksum.c b/arch/m68k/lib/checksum.c index cb13c6e3cca..aed3be29e06 100644 --- a/arch/m68k/lib/checksum.c +++ b/arch/m68k/lib/checksum.c @@ -39,8 +39,7 @@ * computes a partial checksum, e.g. for TCP/UDP fragments */ -unsigned int -csum_partial (const unsigned char *buff, int len, unsigned int sum) +__wsum csum_partial(const void *buff, int len, __wsum sum) { unsigned long tmp1, tmp2; /* @@ -133,9 +132,9 @@ EXPORT_SYMBOL(csum_partial); * copy from user space while checksumming, with exception handling. */ -unsigned int -csum_partial_copy_from_user(const unsigned char __user *src, unsigned char *dst, - int len, int sum, int *csum_err) +__wsum +csum_partial_copy_from_user(const void __user *src, void *dst, + int len, __wsum sum, int *csum_err) { /* * GCC doesn't like more than 10 operands for the asm @@ -325,8 +324,8 @@ csum_partial_copy_from_user(const unsigned char __user *src, unsigned char *dst, * copy from kernel space while checksumming, otherwise like csum_partial */ -unsigned int -csum_partial_copy_nocheck(const unsigned char *src, unsigned char *dst, int len, int sum) +__wsum +csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum) { unsigned long tmp1, tmp2; __asm__("movel %2,%4\n\t" diff --git a/include/asm-m68k/checksum.h b/include/asm-m68k/checksum.h index 17280ef719f..494f9aec37e 100644 --- a/include/asm-m68k/checksum.h +++ b/include/asm-m68k/checksum.h @@ -15,7 +15,7 @@ * * it's best to have buff aligned on a 32-bit boundary */ -unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum); +__wsum csum_partial(const void *buff, int len, __wsum sum); /* * the same as csum_partial, but copies from src while it @@ -25,22 +25,21 @@ unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum) * better 64-bit) boundary */ -extern unsigned int csum_partial_copy_from_user(const unsigned char __user *src, - unsigned char *dst, - int len, int sum, +extern __wsum csum_partial_copy_from_user(const void __user *src, + void *dst, + int len, __wsum sum, int *csum_err); -extern unsigned int csum_partial_copy_nocheck(const unsigned char *src, - unsigned char *dst, int len, - int sum); +extern __wsum csum_partial_copy_nocheck(const void *src, + void *dst, int len, + __wsum sum); /* * This is a version of ip_compute_csum() optimized for IP headers, * which always checksum on 4 octet boundaries. * */ -static inline unsigned short -ip_fast_csum(unsigned char *iph, unsigned int ihl) +static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl) { unsigned int sum = 0; unsigned long tmp; @@ -58,29 +57,29 @@ ip_fast_csum(unsigned char *iph, unsigned int ihl) : "=d" (sum), "=&a" (iph), "=&d" (ihl), "=&d" (tmp) : "0" (sum), "1" (iph), "2" (ihl) : "memory"); - return ~sum; + return (__force __sum16)~sum; } /* * Fold a partial checksum */ -static inline unsigned int csum_fold(unsigned int sum) +static inline __sum16 csum_fold(__wsum sum) { - unsigned int tmp = sum; + unsigned int tmp = (__force u32)sum; __asm__("swap %1\n\t" "addw %1, %0\n\t" "clrw %1\n\t" "addxw %1, %0" : "=&d" (sum), "=&d" (tmp) : "0" (sum), "1" (tmp)); - return ~sum; + return (__force __sum16)~sum; } -static inline unsigned int -csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, unsigned short len, - unsigned short proto, unsigned int sum) +static inline __wsum +csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, + unsigned short proto, __wsum sum) { __asm__ ("addl %2,%0\n\t" "addxl %3,%0\n\t" @@ -98,9 +97,9 @@ csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, unsigned short len, * computes the checksum of the TCP/UDP pseudo-header * returns a 16-bit checksum, already complemented */ -static inline unsigned short int -csum_tcpudp_magic(unsigned long saddr, unsigned long daddr, unsigned short len, - unsigned short proto, unsigned int sum) +static inline __sum16 +csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, + unsigned short proto, __wsum sum) { return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); } @@ -110,16 +109,15 @@ csum_tcpudp_magic(unsigned long saddr, unsigned long daddr, unsigned short len, * in icmp.c */ -static inline unsigned short -ip_compute_csum(unsigned char * buff, int len) +static inline __sum16 ip_compute_csum(const void *buff, int len) { return csum_fold (csum_partial(buff, len, 0)); } #define _HAVE_ARCH_IPV6_CSUM -static __inline__ unsigned short int -csum_ipv6_magic(struct in6_addr *saddr, struct in6_addr *daddr, - __u32 len, unsigned short proto, unsigned int sum) +static __inline__ __sum16 +csum_ipv6_magic(const struct in6_addr *saddr, const struct in6_addr *daddr, + __u32 len, unsigned short proto, __wsum sum) { register unsigned long tmp; __asm__("addl %2@,%0\n\t" -- cgit v1.2.3-70-g09d2 From 59ed05a7e891d694a43df96ac613f7e8e164eb95 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:17:56 -0800 Subject: [NET]: M68Knommu checksum annotations and cleanups. * sanitize prototypes, annotated * collapsed csum_partial_copy() Signed-off-by: Al Viro Signed-off-by: David S. Miller --- arch/m68knommu/kernel/m68k_ksyms.c | 2 +- arch/m68knommu/lib/checksum.c | 28 +++++++++++------------ include/asm-m68knommu/checksum.h | 46 ++++++++++++++++++-------------------- 3 files changed, 37 insertions(+), 39 deletions(-) (limited to 'include') diff --git a/arch/m68knommu/kernel/m68k_ksyms.c b/arch/m68knommu/kernel/m68k_ksyms.c index 1e62150f358..25327c9eadd 100644 --- a/arch/m68knommu/kernel/m68k_ksyms.c +++ b/arch/m68knommu/kernel/m68k_ksyms.c @@ -38,7 +38,7 @@ EXPORT_SYMBOL(ip_fast_csum); EXPORT_SYMBOL(kernel_thread); /* Networking helper routines. */ -EXPORT_SYMBOL(csum_partial_copy); +EXPORT_SYMBOL(csum_partial_copy_nocheck); /* The following are special because they're not called explicitly (the C compiler generates them). Fortunately, diff --git a/arch/m68knommu/lib/checksum.c b/arch/m68knommu/lib/checksum.c index 7bec6fdee34..269d83bfbbe 100644 --- a/arch/m68knommu/lib/checksum.c +++ b/arch/m68knommu/lib/checksum.c @@ -96,9 +96,9 @@ out: * This is a version of ip_compute_csum() optimized for IP headers, * which always checksum on 4 octet boundaries. */ -unsigned short ip_fast_csum(unsigned char * iph, unsigned int ihl) +__sum16 ip_fast_csum(const void *iph, unsigned int ihl) { - return ~do_csum(iph,ihl*4); + return (__force __sum16)~do_csum(iph,ihl*4); } /* @@ -113,15 +113,15 @@ unsigned short ip_fast_csum(unsigned char * iph, unsigned int ihl) * * it's best to have buff aligned on a 32-bit boundary */ -unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum) +__wsum csum_partial(const void *buff, int len, __wsum sum) { unsigned int result = do_csum(buff, len); /* add in old sum, and carry.. */ - result += sum; - if (sum > result) + result += (__force u32)sum; + if ((__force u32)sum > result) result += 1; - return result; + return (__force __wsum)result; } EXPORT_SYMBOL(csum_partial); @@ -130,21 +130,21 @@ EXPORT_SYMBOL(csum_partial); * this routine is used for miscellaneous IP-like checksums, mainly * in icmp.c */ -unsigned short ip_compute_csum(const unsigned char * buff, int len) +__sum16 ip_compute_csum(const void *buff, int len) { - return ~do_csum(buff,len); + return (__force __sum16)~do_csum(buff,len); } /* * copy from fs while checksumming, otherwise like csum_partial */ -unsigned int -csum_partial_copy_from_user(const unsigned char *src, unsigned char *dst, - int len, int sum, int *csum_err) +__wsum +csum_partial_copy_from_user(const void __user *src, void *dst, + int len, __wsum sum, int *csum_err) { if (csum_err) *csum_err = 0; - memcpy(dst, src, len); + memcpy(dst, (__force const void *)src, len); return csum_partial(dst, len, sum); } @@ -152,8 +152,8 @@ csum_partial_copy_from_user(const unsigned char *src, unsigned char *dst, * copy from ds while checksumming, otherwise like csum_partial */ -unsigned int -csum_partial_copy(const unsigned char *src, unsigned char *dst, int len, int sum) +__wsum +csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum) { memcpy(dst, src, len); return csum_partial(dst, len, sum); diff --git a/include/asm-m68knommu/checksum.h b/include/asm-m68knommu/checksum.h index 294ec7583ac..81883482ffb 100644 --- a/include/asm-m68knommu/checksum.h +++ b/include/asm-m68knommu/checksum.h @@ -15,7 +15,7 @@ * * it's best to have buff aligned on a 32-bit boundary */ -unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum); +__wsum csum_partial(const void *buff, int len, __wsum sum); /* * the same as csum_partial, but copies from src while it @@ -25,8 +25,8 @@ unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum) * better 64-bit) boundary */ -unsigned int csum_partial_copy(const unsigned char *src, unsigned char *dst, - int len, int sum); +__wsum csum_partial_copy_nocheck(const void *src, void *dst, + int len, __wsum sum); /* @@ -36,33 +36,31 @@ unsigned int csum_partial_copy(const unsigned char *src, unsigned char *dst, * better 64-bit) boundary */ -extern unsigned int csum_partial_copy_from_user(const unsigned char *src, - unsigned char *dst, int len, int sum, int *csum_err); +extern __wsum csum_partial_copy_from_user(const void __user *src, + void *dst, int len, __wsum sum, int *csum_err); -#define csum_partial_copy_nocheck(src, dst, len, sum) \ - csum_partial_copy((src), (dst), (len), (sum)) - -unsigned short ip_fast_csum(unsigned char *iph, unsigned int ihl); +__sum16 ip_fast_csum(const void *iph, unsigned int ihl); /* * Fold a partial checksum */ -static inline unsigned int csum_fold(unsigned int sum) +static inline __sum16 csum_fold(__wsum sum) { + unsigned int tmp = (__force u32)sum; #ifdef CONFIG_COLDFIRE - sum = (sum & 0xffff) + (sum >> 16); - sum = (sum & 0xffff) + (sum >> 16); + tmp = (tmp & 0xffff) + (tmp >> 16); + tmp = (tmp & 0xffff) + (tmp >> 16); + return (__force __sum16)~tmp; #else - unsigned int tmp = sum; __asm__("swap %1\n\t" "addw %1, %0\n\t" "clrw %1\n\t" "addxw %1, %0" : "=&d" (sum), "=&d" (tmp) : "0" (sum), "1" (sum)); + return (__force __sum16)~sum; #endif - return ~sum; } @@ -71,9 +69,9 @@ static inline unsigned int csum_fold(unsigned int sum) * returns a 16-bit checksum, already complemented */ -static inline unsigned int -csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, unsigned short len, - unsigned short proto, unsigned int sum) +static inline __wsum +csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, + unsigned short proto, __wsum sum) { __asm__ ("addl %1,%0\n\t" "addxl %4,%0\n\t" @@ -86,9 +84,9 @@ csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, unsigned short len, return sum; } -static inline unsigned short int -csum_tcpudp_magic(unsigned long saddr, unsigned long daddr, unsigned short len, - unsigned short proto, unsigned int sum) +static inline __sum16 +csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, + unsigned short proto, __wsum sum) { return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); } @@ -98,12 +96,12 @@ csum_tcpudp_magic(unsigned long saddr, unsigned long daddr, unsigned short len, * in icmp.c */ -extern unsigned short ip_compute_csum(const unsigned char * buff, int len); +extern __sum16 ip_compute_csum(const void *buff, int len); #define _HAVE_ARCH_IPV6_CSUM -static __inline__ unsigned short int -csum_ipv6_magic(struct in6_addr *saddr, struct in6_addr *daddr, - __u32 len, unsigned short proto, unsigned int sum) +static __inline__ __sum16 +csum_ipv6_magic(const struct in6_addr *saddr, const struct in6_addr *daddr, + __u32 len, unsigned short proto, __wsum sum) { register unsigned long tmp; __asm__("addl %2@,%0\n\t" -- cgit v1.2.3-70-g09d2 From 8e3d8433d8c22ca6c42cba4a67d300c39aae7822 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:18:18 -0800 Subject: [NET]: MIPS checksum annotations and cleanups. * sanitize prototypes, annotate * kill shift-by-16 in checksum calculations * htons->shift in l-e checksum calculations Signed-off-by: Al Viro Signed-off-by: David S. Miller --- arch/mips/lib/csum_partial_copy.c | 8 +++--- include/asm-mips/checksum.h | 55 ++++++++++++++++++--------------------- 2 files changed, 30 insertions(+), 33 deletions(-) (limited to 'include') diff --git a/arch/mips/lib/csum_partial_copy.c b/arch/mips/lib/csum_partial_copy.c index 6e9f366f961..1720f2ceeea 100644 --- a/arch/mips/lib/csum_partial_copy.c +++ b/arch/mips/lib/csum_partial_copy.c @@ -16,8 +16,8 @@ /* * copy while checksumming, otherwise like csum_partial */ -unsigned int csum_partial_copy_nocheck(const unsigned char *src, - unsigned char *dst, int len, unsigned int sum) +__wsum csum_partial_copy_nocheck(const void *src, + void *dst, int len, __wsum sum) { /* * It's 2:30 am and I don't feel like doing it real ... @@ -33,8 +33,8 @@ unsigned int csum_partial_copy_nocheck(const unsigned char *src, * Copy from userspace and compute checksum. If we catch an exception * then zero the rest of the buffer. */ -unsigned int csum_partial_copy_from_user (const unsigned char __user *src, - unsigned char *dst, int len, unsigned int sum, int *err_ptr) +__wsum csum_partial_copy_from_user (const void __user *src, + void *dst, int len, __wsum sum, int *err_ptr) { int missing; diff --git a/include/asm-mips/checksum.h b/include/asm-mips/checksum.h index a5e6050ec0f..9b768c3b96b 100644 --- a/include/asm-mips/checksum.h +++ b/include/asm-mips/checksum.h @@ -27,23 +27,22 @@ * * it's best to have buff aligned on a 32-bit boundary */ -unsigned int csum_partial(const unsigned char *buff, int len, unsigned int sum); +__wsum csum_partial(const void *buff, int len, __wsum sum); /* * this is a new version of the above that records errors it finds in *errp, * but continues and zeros the rest of the buffer. */ -unsigned int csum_partial_copy_from_user(const unsigned char __user *src, - unsigned char *dst, int len, - unsigned int sum, int *errp); +__wsum csum_partial_copy_from_user(const void __user *src, + void *dst, int len, + __wsum sum, int *errp); /* * Copy and checksum to user */ #define HAVE_CSUM_COPY_USER -static inline unsigned int csum_and_copy_to_user (const unsigned char *src, - unsigned char __user *dst, - int len, int sum, +static inline __wsum csum_and_copy_to_user (const void *src, void __user *dst, + int len, __wsum sum, int *err_ptr) { might_sleep(); @@ -51,7 +50,7 @@ static inline unsigned int csum_and_copy_to_user (const unsigned char *src, if (copy_to_user(dst, src, len)) { *err_ptr = -EFAULT; - return -1; + return (__force __wsum)-1; } return sum; @@ -61,13 +60,13 @@ static inline unsigned int csum_and_copy_to_user (const unsigned char *src, * the same as csum_partial, but copies from user space (but on MIPS * we have just one address space, so this is identical to the above) */ -unsigned int csum_partial_copy_nocheck(const unsigned char *src, unsigned char *dst, - int len, unsigned int sum); +__wsum csum_partial_copy_nocheck(const void *src, void *dst, + int len, __wsum sum); /* * Fold a partial checksum without adding pseudo headers */ -static inline unsigned short int csum_fold(unsigned int sum) +static inline __sum16 csum_fold(__wsum sum) { __asm__( " .set push # csum_fold\n" @@ -82,7 +81,7 @@ static inline unsigned short int csum_fold(unsigned int sum) : "=r" (sum) : "0" (sum)); - return sum; + return (__force __sum16)sum; } /* @@ -92,10 +91,10 @@ static inline unsigned short int csum_fold(unsigned int sum) * By Jorge Cwik , adapted for linux by * Arnt Gulbrandsen. */ -static inline unsigned short ip_fast_csum(unsigned char *iph, unsigned int ihl) +static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl) { - unsigned int *word = (unsigned int *) iph; - unsigned int *stop = word + ihl; + const unsigned int *word = iph; + const unsigned int *stop = word + ihl; unsigned int csum; int carry; @@ -123,9 +122,9 @@ static inline unsigned short ip_fast_csum(unsigned char *iph, unsigned int ihl) return csum_fold(csum); } -static inline unsigned int csum_tcpudp_nofold(unsigned long saddr, - unsigned long daddr, unsigned short len, unsigned short proto, - unsigned int sum) +static inline __wsum csum_tcpudp_nofold(__be32 saddr, + __be32 daddr, unsigned short len, unsigned short proto, + __wsum sum) { __asm__( " .set push # csum_tcpudp_nofold\n" @@ -155,9 +154,9 @@ static inline unsigned int csum_tcpudp_nofold(unsigned long saddr, : "=r" (sum) : "0" (daddr), "r"(saddr), #ifdef __MIPSEL__ - "r" (((unsigned long)htons(len)<<16) + proto*256), + "r" ((proto + len) << 8), #else - "r" (((unsigned long)(proto)<<16) + len), + "r" (proto + len), #endif "r" (sum)); @@ -168,11 +167,10 @@ static inline unsigned int csum_tcpudp_nofold(unsigned long saddr, * computes the checksum of the TCP/UDP pseudo-header * returns a 16-bit checksum, already complemented */ -static inline unsigned short int csum_tcpudp_magic(unsigned long saddr, - unsigned long daddr, +static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { return csum_fold(csum_tcpudp_nofold(saddr, daddr, len, proto, sum)); } @@ -181,17 +179,16 @@ static inline unsigned short int csum_tcpudp_magic(unsigned long saddr, * this routine is used for miscellaneous IP-like checksums, mainly * in icmp.c */ -static inline unsigned short ip_compute_csum(unsigned char * buff, int len) +static inline __sum16 ip_compute_csum(const void *buff, int len) { return csum_fold(csum_partial(buff, len, 0)); } #define _HAVE_ARCH_IPV6_CSUM -static __inline__ unsigned short int csum_ipv6_magic(struct in6_addr *saddr, - struct in6_addr *daddr, - __u32 len, - unsigned short proto, - unsigned int sum) +static __inline__ __sum16 csum_ipv6_magic(const struct in6_addr *saddr, + const struct in6_addr *daddr, + __u32 len, unsigned short proto, + __wsum sum) { __asm__( " .set push # csum_ipv6_magic\n" -- cgit v1.2.3-70-g09d2 From 7814e4b6d6ce59071887600a8659641ba3d30a43 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:18:39 -0800 Subject: [NET]: PARISC checksum annotations and cleanups. * sanitized prototypes, annotated * kill shift-by-16 in checksum calculation Signed-off-by: Al Viro Signed-off-by: David S. Miller --- arch/parisc/lib/checksum.c | 17 +++++++------ include/asm-parisc/checksum.h | 55 ++++++++++++++++++++----------------------- 2 files changed, 36 insertions(+), 36 deletions(-) (limited to 'include') diff --git a/arch/parisc/lib/checksum.c b/arch/parisc/lib/checksum.c index 8a1e08068e7..462696d30d3 100644 --- a/arch/parisc/lib/checksum.c +++ b/arch/parisc/lib/checksum.c @@ -101,11 +101,14 @@ out: /* * computes a partial checksum, e.g. for TCP/UDP fragments */ -unsigned int csum_partial(const unsigned char *buff, int len, unsigned int sum) +/* + * why bother folding? + */ +__wsum csum_partial(const void *buff, int len, __wsum sum) { unsigned int result = do_csum(buff, len); addc(result, sum); - return from32to16(result); + return (__force __wsum)from32to16(result); } EXPORT_SYMBOL(csum_partial); @@ -113,8 +116,8 @@ EXPORT_SYMBOL(csum_partial); /* * copy while checksumming, otherwise like csum_partial */ -unsigned int csum_partial_copy_nocheck(const unsigned char *src, unsigned char *dst, - int len, unsigned int sum) +__wsum csum_partial_copy_nocheck(const void *src, void *dst, + int len, __wsum sum) { /* * It's 2:30 am and I don't feel like doing it real ... @@ -131,9 +134,9 @@ EXPORT_SYMBOL(csum_partial_copy_nocheck); * Copy from userspace and compute checksum. If we catch an exception * then zero the rest of the buffer. */ -unsigned int csum_partial_copy_from_user(const unsigned char __user *src, - unsigned char *dst, int len, - unsigned int sum, int *err_ptr) +__wsum csum_partial_copy_from_user(const void __user *src, + void *dst, int len, + __wsum sum, int *err_ptr) { int missing; diff --git a/include/asm-parisc/checksum.h b/include/asm-parisc/checksum.h index 229cb56fdb7..cc3ec1bd891 100644 --- a/include/asm-parisc/checksum.h +++ b/include/asm-parisc/checksum.h @@ -15,7 +15,7 @@ * * it's best to have buff aligned on a 32-bit boundary */ -extern unsigned int csum_partial(const unsigned char *, int, unsigned int); +extern __wsum csum_partial(const void *, int, __wsum); /* * The same as csum_partial, but copies from src while it checksums. @@ -23,15 +23,14 @@ extern unsigned int csum_partial(const unsigned char *, int, unsigned int); * Here even more important to align src and dst on a 32-bit (or even * better 64-bit) boundary */ -extern unsigned int csum_partial_copy_nocheck(const unsigned char *, unsigned char *, - int, unsigned int); +extern __wsum csum_partial_copy_nocheck(const void *, void *, int, __wsum); /* * this is a new version of the above that records errors it finds in *errp, * but continues and zeros the rest of the buffer. */ -extern unsigned int csum_partial_copy_from_user(const unsigned char __user *src, - unsigned char *dst, int len, unsigned int sum, int *errp); +extern __wsum csum_partial_copy_from_user(const void __user *src, + void *dst, int len, __wsum sum, int *errp); /* * Optimized for IP headers, which always checksum on 4 octet boundaries. @@ -39,11 +38,10 @@ extern unsigned int csum_partial_copy_from_user(const unsigned char __user *src, * Written by Randolph Chung , and then mucked with by * LaMont Jones */ -static inline unsigned short ip_fast_csum(unsigned char * iph, - unsigned int ihl) { +static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl) +{ unsigned int sum; - __asm__ __volatile__ ( " ldws,ma 4(%1), %0\n" " addib,<= -4, %2, 2f\n" @@ -69,27 +67,27 @@ static inline unsigned short ip_fast_csum(unsigned char * iph, : "1" (iph), "2" (ihl) : "r19", "r20", "r21" ); - return(sum); + return (__force __sum16)sum; } /* * Fold a partial checksum */ -static inline unsigned int csum_fold(unsigned int sum) +static inline __sum16 csum_fold(__wsum csum) { + u32 sum = (__force u32)csum; /* add the swapped two 16-bit halves of sum, a possible carry from adding the two 16-bit halves, will carry from the lower half into the upper half, giving us the correct sum in the upper half. */ sum += (sum << 16) + (sum >> 16); - return (~sum) >> 16; + return (__force __sum16)(~sum >> 16); } -static inline unsigned long csum_tcpudp_nofold(unsigned long saddr, - unsigned long daddr, +static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { __asm__( " add %1, %0, %0\n" @@ -97,19 +95,18 @@ static inline unsigned long csum_tcpudp_nofold(unsigned long saddr, " addc %3, %0, %0\n" " addc %%r0, %0, %0\n" : "=r" (sum) - : "r" (daddr), "r"(saddr), "r"((proto<<16)+len), "0"(sum)); - return sum; + : "r" (daddr), "r"(saddr), "r"(proto+len), "0"(sum)); + return sum; } /* * computes the checksum of the TCP/UDP pseudo-header * returns a 16-bit checksum, already complemented */ -static inline unsigned short int csum_tcpudp_magic(unsigned long saddr, - unsigned long daddr, +static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); } @@ -118,17 +115,17 @@ static inline unsigned short int csum_tcpudp_magic(unsigned long saddr, * this routine is used for miscellaneous IP-like checksums, mainly * in icmp.c */ -static inline unsigned short ip_compute_csum(unsigned char * buf, int len) { +static inline __sum16 ip_compute_csum(const void *buf, int len) +{ return csum_fold (csum_partial(buf, len, 0)); } #define _HAVE_ARCH_IPV6_CSUM -static __inline__ unsigned short int csum_ipv6_magic(struct in6_addr *saddr, - struct in6_addr *daddr, - __u16 len, - unsigned short proto, - unsigned int sum) +static __inline__ __sum16 csum_ipv6_magic(const struct in6_addr *saddr, + const struct in6_addr *daddr, + __u32 len, unsigned short proto, + __wsum sum) { __asm__ __volatile__ ( @@ -193,9 +190,9 @@ static __inline__ unsigned short int csum_ipv6_magic(struct in6_addr *saddr, * Copy and checksum to user */ #define HAVE_CSUM_COPY_USER -static __inline__ unsigned int csum_and_copy_to_user (const unsigned char *src, - unsigned char __user *dst, - int len, int sum, +static __inline__ __wsum csum_and_copy_to_user(const void *src, + void __user *dst, + int len, __wsum sum, int *err_ptr) { /* code stolen from include/asm-mips64 */ @@ -203,7 +200,7 @@ static __inline__ unsigned int csum_and_copy_to_user (const unsigned char *src, if (copy_to_user(dst, src, len)) { *err_ptr = -EFAULT; - return -1; + return (__force __wsum)-1; } return sum; -- cgit v1.2.3-70-g09d2 From c459dd90f0de00db1ca1328482214019f6ca292f Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:19:01 -0800 Subject: [NET]: SH64 checksum annotations and cleanups. * sanitize prototypes, annotate * collapse csum_partial_copy * kill csum_partial_copy_fromuser * ntohs->shift in checksum calculation Signed-off-by: Al Viro Signed-off-by: David S. Miller --- arch/sh64/kernel/sh_ksyms.c | 2 +- arch/sh64/lib/c-checksum.c | 49 ++++++++++++++++----------------------------- include/asm-sh64/checksum.h | 41 +++++++++++++------------------------ 3 files changed, 32 insertions(+), 60 deletions(-) (limited to 'include') diff --git a/arch/sh64/kernel/sh_ksyms.c b/arch/sh64/kernel/sh_ksyms.c index 4b2df7247b5..7aa4b4f7bc5 100644 --- a/arch/sh64/kernel/sh_ksyms.c +++ b/arch/sh64/kernel/sh_ksyms.c @@ -38,7 +38,7 @@ EXPORT_SYMBOL(disable_irq); EXPORT_SYMBOL(kernel_thread); /* Networking helper routines. */ -EXPORT_SYMBOL(csum_partial_copy); +EXPORT_SYMBOL(csum_partial_copy_nocheck); EXPORT_SYMBOL(strstr); diff --git a/arch/sh64/lib/c-checksum.c b/arch/sh64/lib/c-checksum.c index 0e8a742abf8..4b2676380de 100644 --- a/arch/sh64/lib/c-checksum.c +++ b/arch/sh64/lib/c-checksum.c @@ -118,24 +118,24 @@ static unsigned long do_csum(const unsigned char *buff, int len) /* computes the checksum of a memory block at buff, length len, and adds in "sum" (32-bit) */ -unsigned int csum_partial(const unsigned char *buff, int len, unsigned int sum) +__wsum csum_partial(const void *buff, int len, __wsum sum) { unsigned long long result = do_csum(buff, len); /* add in old sum, and carry.. */ - result += sum; + result += (__force u32)sum; /* 32+c bits -> 32 bits */ result = (result & 0xffffffff) + (result >> 32); pr_debug("csum_partial, buff %p len %d sum 0x%x result=0x%016Lx\n", buff, len, sum, result); - return result; + return (__force __wsum)result; } /* Copy while checksumming, otherwise like csum_partial. */ -unsigned int -csum_partial_copy(const unsigned char *src, unsigned char *dst, int len, unsigned int sum) +__wsum +csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum) { sum = csum_partial(src, len, sum); memcpy(dst, src, len); @@ -145,9 +145,9 @@ csum_partial_copy(const unsigned char *src, unsigned char *dst, int len, unsigne /* Copy from userspace and compute checksum. If we catch an exception then zero the rest of the buffer. */ -unsigned int -csum_partial_copy_from_user(const unsigned char *src, unsigned char *dst, int len, - unsigned int sum, int *err_ptr) +__wsum +csum_partial_copy_from_user(const void __user *src, void *dst, int len, + __wsum sum, int *err_ptr) { int missing; @@ -166,9 +166,9 @@ csum_partial_copy_from_user(const unsigned char *src, unsigned char *dst, int le } /* Copy to userspace and compute checksum. */ -unsigned int +__wsum csum_partial_copy_to_user(const unsigned char *src, unsigned char *dst, int len, - unsigned int sum, int *err_ptr) + __wsum sum, int *err_ptr) { sum = csum_partial(src, len, sum); @@ -182,28 +182,24 @@ csum_partial_copy_to_user(const unsigned char *src, unsigned char *dst, int len, * This is a version of ip_compute_csum() optimized for IP headers, * which always checksum on 4 octet boundaries. */ -unsigned short ip_fast_csum(unsigned char *iph, unsigned int ihl) +__sum16 ip_fast_csum(const void *iph, unsigned int ihl) { pr_debug("ip_fast_csum %p,%d\n", iph, ihl); - return ~do_csum(iph, ihl * 4); + return (__force __sum16)~do_csum(iph, ihl * 4); } -unsigned int csum_tcpudp_nofold(unsigned long saddr, - unsigned long daddr, +__wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, - unsigned short proto, unsigned int sum) + unsigned short proto, __wsum sum) { unsigned long long result; pr_debug("ntohs(0x%x)=0x%x\n", 0xdead, ntohs(0xdead)); pr_debug("htons(0x%x)=0x%x\n", 0xdead, htons(0xdead)); - result = ((unsigned long long) saddr + - (unsigned long long) daddr + - (unsigned long long) sum + - ((unsigned long long) ntohs(len) << 16) + - ((unsigned long long) proto << 8)); + result = (__force u64) saddr + (__force u64) daddr + + (__force u64) sum + ((len + proto) << 8); /* Fold down to 32-bits so we don't loose in the typedef-less network stack. */ @@ -215,16 +211,5 @@ unsigned int csum_tcpudp_nofold(unsigned long saddr, pr_debug("%s saddr %x daddr %x len %x proto %x sum %x result %08Lx\n", __FUNCTION__, saddr, daddr, len, proto, sum, result); - return result; -} - -// Post SIM: -unsigned int -csum_partial_copy_nocheck(const unsigned char *src, unsigned char *dst, int len, unsigned int sum) -{ - // unsigned dummy; - pr_debug("csum_partial_copy_nocheck src %p dst %p len %d\n", src, dst, - len); - - return csum_partial_copy(src, dst, len, sum); + return (__wsum)result; } diff --git a/include/asm-sh64/checksum.h b/include/asm-sh64/checksum.h index fd034e9ae6e..ba594ccb42e 100644 --- a/include/asm-sh64/checksum.h +++ b/include/asm-sh64/checksum.h @@ -26,8 +26,7 @@ * * it's best to have buff aligned on a 32-bit boundary */ -asmlinkage unsigned int csum_partial(const unsigned char *buff, int len, - unsigned int sum); +asmlinkage __wsum csum_partial(const void *buff, int len, __wsum sum); /* * Note: when you get a NULL pointer exception here this means someone @@ -38,46 +37,34 @@ asmlinkage unsigned int csum_partial(const unsigned char *buff, int len, */ -unsigned int csum_partial_copy_nocheck(const char *src, char *dst, int len, - unsigned int sum); +__wsum csum_partial_copy_nocheck(const void *src, void *dst, int len, + __wsum sum); -unsigned int csum_partial_copy_from_user(const char *src, char *dst, - int len, int sum, int *err_ptr); +__wsum csum_partial_copy_from_user(const void __user *src, void *dst, + int len, __wsum sum, int *err_ptr); -/* - * These are the old (and unsafe) way of doing checksums, a warning message will be - * printed if they are used and an exeption occurs. - * - * these functions should go away after some time. - */ - -#define csum_partial_copy_fromuser csum_partial_copy - -unsigned int csum_partial_copy(const char *src, char *dst, int len, - unsigned int sum); - -static inline unsigned short csum_fold(unsigned int sum) +static inline __sum16 csum_fold(__wsum csum) { + u32 sum = (__force u32)csum; sum = (sum & 0xffff) + (sum >> 16); sum = (sum & 0xffff) + (sum >> 16); - return ~(sum); + return (__force __sum16)~sum; } -unsigned short ip_fast_csum(unsigned char * iph, unsigned int ihl); +__sum16 ip_fast_csum(const void *iph, unsigned int ihl); -unsigned long csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, +__wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum); + __wsum sum); /* * computes the checksum of the TCP/UDP pseudo-header * returns a 16-bit checksum, already complemented */ -static inline unsigned short int csum_tcpudp_magic(unsigned long saddr, - unsigned long daddr, +static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); } @@ -86,7 +73,7 @@ static inline unsigned short int csum_tcpudp_magic(unsigned long saddr, * this routine is used for miscellaneous IP-like checksums, mainly * in icmp.c */ -static inline unsigned short ip_compute_csum(unsigned char * buff, int len) +static inline __sum16 ip_compute_csum(const void *buff, int len) { return csum_fold(csum_partial(buff, len, 0)); } -- cgit v1.2.3-70-g09d2 From 9d3d41955845939cb41b87affb039db0bae03b65 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:19:44 -0800 Subject: [NET]: V850 checksum annotations and cleanups. * sanitize prototypes, annotate * collapse csum_partial_copy * usual ntohs->shift Signed-off-by: Al Viro Signed-off-by: David S. Miller --- arch/v850/kernel/v850_ksyms.c | 2 +- arch/v850/lib/checksum.c | 26 +++++++++++++------------- include/asm-v850/checksum.h | 38 +++++++++++++++++--------------------- 3 files changed, 31 insertions(+), 35 deletions(-) (limited to 'include') diff --git a/arch/v850/kernel/v850_ksyms.c b/arch/v850/kernel/v850_ksyms.c index 67bc48e57c6..93575fdc874 100644 --- a/arch/v850/kernel/v850_ksyms.c +++ b/arch/v850/kernel/v850_ksyms.c @@ -24,7 +24,7 @@ EXPORT_SYMBOL (kernel_thread); EXPORT_SYMBOL (__bug); /* Networking helper routines. */ -EXPORT_SYMBOL (csum_partial_copy); +EXPORT_SYMBOL (csum_partial_copy_nocheck); EXPORT_SYMBOL (csum_partial_copy_from_user); EXPORT_SYMBOL (ip_compute_csum); EXPORT_SYMBOL (ip_fast_csum); diff --git a/arch/v850/lib/checksum.c b/arch/v850/lib/checksum.c index fa587263307..042158dfe17 100644 --- a/arch/v850/lib/checksum.c +++ b/arch/v850/lib/checksum.c @@ -88,32 +88,32 @@ out: * This is a version of ip_compute_csum() optimized for IP headers, * which always checksum on 4 octet boundaries. */ -unsigned short ip_fast_csum(unsigned char * iph, unsigned int ihl) +__sum16 ip_fast_csum(const void *iph, unsigned int ihl) { - return ~do_csum(iph,ihl*4); + return (__force __sum16)~do_csum(iph,ihl*4); } /* * this routine is used for miscellaneous IP-like checksums, mainly * in icmp.c */ -unsigned short ip_compute_csum(const unsigned char * buff, int len) +__sum16 ip_compute_csum(const void *buff, int len) { - return ~do_csum(buff,len); + return (__force __sum16)~do_csum(buff,len); } /* * computes a partial checksum, e.g. for TCP/UDP fragments */ -unsigned int csum_partial(const unsigned char *buff, int len, unsigned int sum) +__wsum csum_partial(const void *buff, int len, __wsum sum) { unsigned int result = do_csum(buff, len); /* add in old sum, and carry.. */ - result += sum; - if(sum > result) + result += (__force u32)sum; + if ((__force u32)sum > result) result += 1; - return result; + return (__force __wsum)result; } EXPORT_SYMBOL(csum_partial); @@ -121,8 +121,8 @@ EXPORT_SYMBOL(csum_partial); /* * copy while checksumming, otherwise like csum_partial */ -unsigned int csum_partial_copy(const unsigned char *src, unsigned char *dst, - int len, unsigned int sum) +__wsum csum_partial_copy_nocheck(const void *src, void *dst, + int len, __wsum sum) { /* * It's 2:30 am and I don't feel like doing it real ... @@ -138,9 +138,9 @@ unsigned int csum_partial_copy(const unsigned char *src, unsigned char *dst, * Copy from userspace and compute checksum. If we catch an exception * then zero the rest of the buffer. */ -unsigned int csum_partial_copy_from_user (const unsigned char *src, - unsigned char *dst, - int len, unsigned int sum, +__wsum csum_partial_copy_from_user (const void *src, + void *dst, + int len, __wsum sum, int *err_ptr) { int missing; diff --git a/include/asm-v850/checksum.h b/include/asm-v850/checksum.h index 4df5e71098f..d1dddd93826 100644 --- a/include/asm-v850/checksum.h +++ b/include/asm-v850/checksum.h @@ -26,8 +26,7 @@ * * it's best to have buff aligned on a 32-bit boundary */ -extern unsigned int csum_partial (const unsigned char * buff, int len, - unsigned int sum); +extern __wsum csum_partial(const void *buff, int len, __wsum sum); /* * the same as csum_partial, but copies from src while it @@ -36,8 +35,8 @@ extern unsigned int csum_partial (const unsigned char * buff, int len, * here even more important to align src and dst on a 32-bit (or even * better 64-bit) boundary */ -extern unsigned csum_partial_copy (const unsigned char *src, - unsigned char *dst, int len, unsigned sum); +extern __wsum csum_partial_copy_nocheck(const void *src, + void *dst, int len, __wsum sum); /* @@ -46,20 +45,17 @@ extern unsigned csum_partial_copy (const unsigned char *src, * here even more important to align src and dst on a 32-bit (or even * better 64-bit) boundary */ -extern unsigned csum_partial_copy_from_user (const unsigned char *src, - unsigned char *dst, - int len, unsigned sum, +extern __wsum csum_partial_copy_from_user (const void *src, + void *dst, + int len, __wsum sum, int *csum_err); -#define csum_partial_copy_nocheck(src, dst, len, sum) \ - csum_partial_copy ((src), (dst), (len), (sum)) - -unsigned short ip_fast_csum (unsigned char *iph, unsigned int ihl); +__sum16 ip_fast_csum(const void *iph, unsigned int ihl); /* * Fold a partial checksum */ -static inline unsigned int csum_fold (unsigned long sum) +static inline __sum16 csum_fold (__wsum sum) { unsigned int result; /* @@ -68,7 +64,7 @@ static inline unsigned int csum_fold (unsigned long sum) add %1, %0 H L H+L+C H+L */ asm ("hsw %1, %0; add %1, %0" : "=&r" (result) : "r" (sum)); - return (~result) >> 16; + return (__force __sum16)(~result >> 16); } @@ -76,10 +72,10 @@ static inline unsigned int csum_fold (unsigned long sum) * computes the checksum of the TCP/UDP pseudo-header * returns a 16-bit checksum, already complemented */ -static inline unsigned int -csum_tcpudp_nofold (unsigned long saddr, unsigned long daddr, +static inline __wsum +csum_tcpudp_nofold (__be32 saddr, __be32 daddr, unsigned short len, - unsigned short proto, unsigned int sum) + unsigned short proto, __wsum sum) { int __carry; __asm__ ("add %2, %0;" @@ -93,15 +89,15 @@ csum_tcpudp_nofold (unsigned long saddr, unsigned long daddr, "add %1, %0" : "=&r" (sum), "=&r" (__carry) : "r" (daddr), "r" (saddr), - "r" (ntohs (len) + (proto << 8)), + "r" ((len + proto) << 8), "0" (sum)); return sum; } -static inline unsigned short int -csum_tcpudp_magic (unsigned long saddr, unsigned long daddr, +static inline __sum16 +csum_tcpudp_magic (__be32 saddr, __be32 daddr, unsigned short len, - unsigned short proto, unsigned int sum) + unsigned short proto, __wsum sum) { return csum_fold (csum_tcpudp_nofold (saddr, daddr, len, proto, sum)); } @@ -110,7 +106,7 @@ csum_tcpudp_magic (unsigned long saddr, unsigned long daddr, * this routine is used for miscellaneous IP-like checksums, mainly * in icmp.c */ -extern unsigned short ip_compute_csum (const unsigned char * buff, int len); +extern __sum16 ip_compute_csum(const void *buff, int len); #endif /* __V850_CHECKSUM_H__ */ -- cgit v1.2.3-70-g09d2 From a4f89fb7c072b8592b296c2ba216269c0c96db43 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:20:08 -0800 Subject: [NET]: X86_64 checksum annotations and cleanups. * sanitize prototypes, annotate * usual ntohs->shift Signed-off-by: Al Viro Signed-off-by: David S. Miller --- arch/x86_64/lib/csum-partial.c | 7 +++--- arch/x86_64/lib/csum-wrappers.c | 37 ++++++++++++++++------------ include/asm-x86_64/checksum.h | 54 +++++++++++++++++++++-------------------- 3 files changed, 53 insertions(+), 45 deletions(-) (limited to 'include') diff --git a/arch/x86_64/lib/csum-partial.c b/arch/x86_64/lib/csum-partial.c index c493735218d..06ae630de82 100644 --- a/arch/x86_64/lib/csum-partial.c +++ b/arch/x86_64/lib/csum-partial.c @@ -132,9 +132,10 @@ static __force_inline unsigned do_csum(const unsigned char *buff, unsigned len) * * it's best to have buff aligned on a 64-bit boundary */ -unsigned csum_partial(const unsigned char *buff, unsigned len, unsigned sum) +__wsum csum_partial(const void *buff, int len, __wsum sum) { - return add32_with_carry(do_csum(buff, len), sum); + return (__force __wsum)add32_with_carry(do_csum(buff, len), + (__force u32)sum); } EXPORT_SYMBOL(csum_partial); @@ -143,7 +144,7 @@ EXPORT_SYMBOL(csum_partial); * this routine is used for miscellaneous IP-like checksums, mainly * in icmp.c */ -unsigned short ip_compute_csum(unsigned char * buff, int len) +__sum16 ip_compute_csum(const void *buff, int len) { return csum_fold(csum_partial(buff,len,0)); } diff --git a/arch/x86_64/lib/csum-wrappers.c b/arch/x86_64/lib/csum-wrappers.c index b1320ec5842..fd42a4a095f 100644 --- a/arch/x86_64/lib/csum-wrappers.c +++ b/arch/x86_64/lib/csum-wrappers.c @@ -18,9 +18,9 @@ * Returns an 32bit unfolded checksum of the buffer. * src and dst are best aligned to 64bits. */ -unsigned int -csum_partial_copy_from_user(const unsigned char __user *src, unsigned char *dst, - int len, unsigned int isum, int *errp) +__wsum +csum_partial_copy_from_user(const void __user *src, void *dst, + int len, __wsum isum, int *errp) { might_sleep(); *errp = 0; @@ -34,17 +34,19 @@ csum_partial_copy_from_user(const unsigned char __user *src, unsigned char *dst, if (unlikely((unsigned long)src & 6)) { while (((unsigned long)src & 6) && len >= 2) { __u16 val16; - *errp = __get_user(val16, (__u16 __user *)src); + *errp = __get_user(val16, (const __u16 __user *)src); if (*errp) return isum; *(__u16 *)dst = val16; - isum = add32_with_carry(isum, val16); + isum = (__force __wsum)add32_with_carry( + (__force unsigned)isum, val16); src += 2; dst += 2; len -= 2; } } - isum = csum_partial_copy_generic((__force void *)src,dst,len,isum,errp,NULL); + isum = csum_partial_copy_generic((__force const void *)src, + dst, len, isum, errp, NULL); if (likely(*errp == 0)) return isum; } @@ -66,9 +68,9 @@ EXPORT_SYMBOL(csum_partial_copy_from_user); * Returns an 32bit unfolded checksum of the buffer. * src and dst are best aligned to 64bits. */ -unsigned int -csum_partial_copy_to_user(unsigned const char *src, unsigned char __user *dst, - int len, unsigned int isum, int *errp) +__wsum +csum_partial_copy_to_user(const void *src, void __user *dst, + int len, __wsum isum, int *errp) { might_sleep(); if (unlikely(!access_ok(VERIFY_WRITE, dst, len))) { @@ -79,7 +81,8 @@ csum_partial_copy_to_user(unsigned const char *src, unsigned char __user *dst, if (unlikely((unsigned long)dst & 6)) { while (((unsigned long)dst & 6) && len >= 2) { __u16 val16 = *(__u16 *)src; - isum = add32_with_carry(isum, val16); + isum = (__force __wsum)add32_with_carry( + (__force unsigned)isum, val16); *errp = __put_user(val16, (__u16 __user *)dst); if (*errp) return isum; @@ -104,19 +107,21 @@ EXPORT_SYMBOL(csum_partial_copy_to_user); * * Returns an 32bit unfolded checksum of the buffer. */ -unsigned int -csum_partial_copy_nocheck(const unsigned char *src, unsigned char *dst, int len, unsigned int sum) +__wsum +csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum) { return csum_partial_copy_generic(src,dst,len,sum,NULL,NULL); } EXPORT_SYMBOL(csum_partial_copy_nocheck); -unsigned short csum_ipv6_magic(struct in6_addr *saddr, struct in6_addr *daddr, - __u32 len, unsigned short proto, unsigned int sum) +__sum16 csum_ipv6_magic(const struct in6_addr *saddr, + const struct in6_addr *daddr, + __u32 len, unsigned short proto, __wsum sum) { __u64 rest, sum64; - rest = (__u64)htonl(len) + (__u64)htons(proto) + (__u64)sum; + rest = (__force __u64)htonl(len) + (__force __u64)htons(proto) + + (__force __u64)sum; asm(" addq (%[saddr]),%[sum]\n" " adcq 8(%[saddr]),%[sum]\n" " adcq (%[daddr]),%[sum]\n" @@ -124,7 +129,7 @@ unsigned short csum_ipv6_magic(struct in6_addr *saddr, struct in6_addr *daddr, " adcq $0,%[sum]\n" : [sum] "=r" (sum64) : "[sum]" (rest),[saddr] "r" (saddr), [daddr] "r" (daddr)); - return csum_fold(add32_with_carry(sum64 & 0xffffffff, sum64>>32)); + return csum_fold((__force __wsum)add32_with_carry(sum64 & 0xffffffff, sum64>>32)); } EXPORT_SYMBOL(csum_ipv6_magic); diff --git a/include/asm-x86_64/checksum.h b/include/asm-x86_64/checksum.h index 989469e8e0b..419fe88a034 100644 --- a/include/asm-x86_64/checksum.h +++ b/include/asm-x86_64/checksum.h @@ -19,15 +19,16 @@ * the last step before putting a checksum into a packet. * Make sure not to mix with 64bit checksums. */ -static inline unsigned int csum_fold(unsigned int sum) +static inline __sum16 csum_fold(__wsum sum) { __asm__( " addl %1,%0\n" " adcl $0xffff,%0" : "=r" (sum) - : "r" (sum << 16), "0" (sum & 0xffff0000) + : "r" ((__force u32)sum << 16), + "0" ((__force u32)sum & 0xffff0000) ); - return (~sum) >> 16; + return (__force __sum16)(~(__force u32)sum >> 16); } /* @@ -43,7 +44,7 @@ static inline unsigned int csum_fold(unsigned int sum) * iph: ipv4 header * ihl: length of header / 4 */ -static inline unsigned short ip_fast_csum(unsigned char *iph, unsigned int ihl) +static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl) { unsigned int sum; @@ -70,7 +71,7 @@ static inline unsigned short ip_fast_csum(unsigned char *iph, unsigned int ihl) : "=r" (sum), "=r" (iph), "=r" (ihl) : "1" (iph), "2" (ihl) : "memory"); - return(sum); + return (__force __sum16)sum; } /** @@ -84,16 +85,17 @@ static inline unsigned short ip_fast_csum(unsigned char *iph, unsigned int ihl) * Returns the pseudo header checksum the input data. Result is * 32bit unfolded. */ -static inline unsigned long -csum_tcpudp_nofold(unsigned saddr, unsigned daddr, unsigned short len, - unsigned short proto, unsigned int sum) +static inline __wsum +csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, + unsigned short proto, __wsum sum) { asm(" addl %1, %0\n" " adcl %2, %0\n" " adcl %3, %0\n" " adcl $0, %0\n" : "=r" (sum) - : "g" (daddr), "g" (saddr), "g" ((ntohs(len)<<16)+proto*256), "0" (sum)); + : "g" (daddr), "g" (saddr), + "g" ((len + proto)<<8), "0" (sum)); return sum; } @@ -109,9 +111,9 @@ csum_tcpudp_nofold(unsigned saddr, unsigned daddr, unsigned short len, * Returns the 16bit pseudo header checksum the input data already * complemented and ready to be filled in. */ -static inline unsigned short int -csum_tcpudp_magic(unsigned long saddr, unsigned long daddr, - unsigned short len, unsigned short proto, unsigned int sum) +static inline __sum16 +csum_tcpudp_magic(__be32 saddr, __be32 daddr, + unsigned short len, unsigned short proto, __wsum sum) { return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); } @@ -126,25 +128,25 @@ csum_tcpudp_magic(unsigned long saddr, unsigned long daddr, * Before filling it in it needs to be csum_fold()'ed. * buff should be aligned to a 64bit boundary if possible. */ -extern unsigned int csum_partial(const unsigned char *buff, unsigned len, unsigned int sum); +extern __wsum csum_partial(const void *buff, int len, __wsum sum); #define _HAVE_ARCH_COPY_AND_CSUM_FROM_USER 1 #define HAVE_CSUM_COPY_USER 1 /* Do not call this directly. Use the wrappers below */ -extern unsigned long csum_partial_copy_generic(const unsigned char *src, const unsigned char *dst, - unsigned len, - unsigned sum, +extern __wsum csum_partial_copy_generic(const void *src, const void *dst, + int len, + __wsum sum, int *src_err_ptr, int *dst_err_ptr); -extern unsigned int csum_partial_copy_from_user(const unsigned char __user *src, unsigned char *dst, - int len, unsigned int isum, int *errp); -extern unsigned int csum_partial_copy_to_user(const unsigned char *src, unsigned char __user *dst, - int len, unsigned int isum, int *errp); -extern unsigned int csum_partial_copy_nocheck(const unsigned char *src, unsigned char *dst, int len, - unsigned int sum); +extern __wsum csum_partial_copy_from_user(const void __user *src, void *dst, + int len, __wsum isum, int *errp); +extern __wsum csum_partial_copy_to_user(const void *src, void __user *dst, + int len, __wsum isum, int *errp); +extern __wsum csum_partial_copy_nocheck(const void *src, void *dst, int len, + __wsum sum); /* Old names. To be removed. */ #define csum_and_copy_to_user csum_partial_copy_to_user @@ -158,7 +160,7 @@ extern unsigned int csum_partial_copy_nocheck(const unsigned char *src, unsigned * Returns the 16bit folded/inverted checksum of the passed buffer. * Ready to fill in. */ -extern unsigned short ip_compute_csum(unsigned char * buff, int len); +extern __sum16 ip_compute_csum(const void *buff, int len); /** * csum_ipv6_magic - Compute checksum of an IPv6 pseudo header. @@ -176,9 +178,9 @@ extern unsigned short ip_compute_csum(unsigned char * buff, int len); struct in6_addr; #define _HAVE_ARCH_IPV6_CSUM 1 -extern unsigned short -csum_ipv6_magic(struct in6_addr *saddr, struct in6_addr *daddr, - __u32 len, unsigned short proto, unsigned int sum); +extern __sum16 +csum_ipv6_magic(const struct in6_addr *saddr, const struct in6_addr *daddr, + __u32 len, unsigned short proto, __wsum sum); static inline unsigned add32_with_carry(unsigned a, unsigned b) { -- cgit v1.2.3-70-g09d2 From eb5a9658656c3d633cc973ec90ccfd6c439dabb6 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:20:28 -0800 Subject: [NET]: ARM checksum annotations and cleanups. * sanitize prototypes, annotate * kill csum_partial_copy * usual ntohs->shift, this time in assembler part Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/asm-arm/checksum.h | 83 +++++++++++++++++++++++----------------------- 1 file changed, 41 insertions(+), 42 deletions(-) (limited to 'include') diff --git a/include/asm-arm/checksum.h b/include/asm-arm/checksum.h index 747bdd31a74..8c0bb5bb14e 100644 --- a/include/asm-arm/checksum.h +++ b/include/asm-arm/checksum.h @@ -23,7 +23,7 @@ * * it's best to have buff aligned on a 32-bit boundary */ -unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum); +__wsum csum_partial(const void *buff, int len, __wsum sum); /* * the same as csum_partial, but copies from src while it @@ -33,26 +33,18 @@ unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum) * better 64-bit) boundary */ -unsigned int -csum_partial_copy_nocheck(const char *src, char *dst, int len, int sum); +__wsum +csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum); -unsigned int -csum_partial_copy_from_user(const char __user *src, char *dst, int len, int sum, int *err_ptr); - -/* - * This is the old (and unsafe) way of doing checksums, a warning message will - * be printed if it is used and an exception occurs. - * - * this functions should go away after some time. - */ -#define csum_partial_copy(src,dst,len,sum) csum_partial_copy_nocheck(src,dst,len,sum) +__wsum +csum_partial_copy_from_user(const void __user *src, void *dst, int len, __wsum sum, int *err_ptr); /* * This is a version of ip_compute_csum() optimized for IP headers, * which always checksum on 4 octet boundaries. */ -static inline unsigned short -ip_fast_csum(unsigned char * iph, unsigned int ihl) +static inline __sum16 +ip_fast_csum(const void *iph, unsigned int ihl) { unsigned int sum, tmp1; @@ -78,14 +70,13 @@ ip_fast_csum(unsigned char * iph, unsigned int ihl) : "=r" (sum), "=r" (iph), "=r" (ihl), "=r" (tmp1) : "1" (iph), "2" (ihl) : "cc", "memory"); - return sum; + return (__force __sum16)sum; } /* * Fold a partial checksum without adding pseudo headers */ -static inline unsigned int -csum_fold(unsigned int sum) +static inline __sum16 csum_fold(__wsum sum) { __asm__( "adds %0, %1, %1, lsl #16 @ csum_fold \n\ @@ -93,21 +84,25 @@ csum_fold(unsigned int sum) : "=r" (sum) : "r" (sum) : "cc"); - return (~sum) >> 16; + return (__force __sum16)(~(__force u32)sum >> 16); } -static inline unsigned int -csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, unsigned short len, - unsigned int proto, unsigned int sum) +static inline __wsum +csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, + unsigned short proto, __wsum sum) { __asm__( "adds %0, %1, %2 @ csum_tcpudp_nofold \n\ - adcs %0, %0, %3 \n\ - adcs %0, %0, %4 \n\ - adcs %0, %0, %5 \n\ + adcs %0, %0, %3 \n" +#ifdef __ARMEB__ + "adcs %0, %0, %4 \n" +#else + "adcs %0, %0, %4, lsl #8 \n" +#endif + "adcs %0, %0, %5 \n\ adc %0, %0, #0" : "=&r"(sum) - : "r" (sum), "r" (daddr), "r" (saddr), "r" (ntohs(len)), "Ir" (ntohs(proto)) + : "r" (sum), "r" (daddr), "r" (saddr), "r" (len), "Ir" (htons(proto)) : "cc"); return sum; } @@ -115,23 +110,27 @@ csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, unsigned short len, * computes the checksum of the TCP/UDP pseudo-header * returns a 16-bit checksum, already complemented */ -static inline unsigned short int -csum_tcpudp_magic(unsigned long saddr, unsigned long daddr, unsigned short len, - unsigned int proto, unsigned int sum) +static inline __sum16 +csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, + unsigned short proto, __wsum sum) { __asm__( "adds %0, %1, %2 @ csum_tcpudp_magic \n\ - adcs %0, %0, %3 \n\ - adcs %0, %0, %4 \n\ - adcs %0, %0, %5 \n\ + adcs %0, %0, %3 \n" +#ifdef __ARMEB__ + "adcs %0, %0, %4 \n" +#else + "adcs %0, %0, %4, lsl #8 \n" +#endif + "adcs %0, %0, %5 \n\ adc %0, %0, #0 \n\ adds %0, %0, %0, lsl #16 \n\ addcs %0, %0, #0x10000 \n\ mvn %0, %0" : "=&r"(sum) - : "r" (sum), "r" (daddr), "r" (saddr), "r" (ntohs(len)), "Ir" (ntohs(proto)) + : "r" (sum), "r" (daddr), "r" (saddr), "r" (len), "Ir" (htons(proto)) : "cc"); - return sum >> 16; + return (__force __sum16)((__force u32)sum >> 16); } @@ -139,20 +138,20 @@ csum_tcpudp_magic(unsigned long saddr, unsigned long daddr, unsigned short len, * this routine is used for miscellaneous IP-like checksums, mainly * in icmp.c */ -static inline unsigned short -ip_compute_csum(unsigned char * buff, int len) +static inline __sum16 +ip_compute_csum(const void *buff, int len) { return csum_fold(csum_partial(buff, len, 0)); } #define _HAVE_ARCH_IPV6_CSUM -extern unsigned long -__csum_ipv6_magic(struct in6_addr *saddr, struct in6_addr *daddr, __u32 len, - __u32 proto, unsigned int sum); +extern __wsum +__csum_ipv6_magic(const struct in6_addr *saddr, const struct in6_addr *daddr, __be32 len, + __be32 proto, __wsum sum); -static inline unsigned short int -csum_ipv6_magic(struct in6_addr *saddr, struct in6_addr *daddr, __u32 len, - unsigned short proto, unsigned int sum) +static inline __sum16 +csum_ipv6_magic(const struct in6_addr *saddr, const struct in6_addr *daddr, __u32 len, + unsigned short proto, __wsum sum) { return csum_fold(__csum_ipv6_magic(saddr, daddr, htonl(len), htonl(proto), sum)); -- cgit v1.2.3-70-g09d2 From 0ffb7122bc988ce328e84d14c81d029bc62c47c5 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:20:51 -0800 Subject: [NET]: ARM26 checksum annotations and cleanups. * sanitize prototypes, annotate * kill csum_partial_copy Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/asm-arm26/checksum.h | 63 +++++++++++++++++++------------------------- 1 file changed, 27 insertions(+), 36 deletions(-) (limited to 'include') diff --git a/include/asm-arm26/checksum.h b/include/asm-arm26/checksum.h index d4256d5f3a7..f2b4b0a403b 100644 --- a/include/asm-arm26/checksum.h +++ b/include/asm-arm26/checksum.h @@ -23,7 +23,7 @@ * * it's best to have buff aligned on a 32-bit boundary */ -unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum); +__wsum csum_partial(const void *buff, int len, __wsum sum); /* * the same as csum_partial, but copies from src while it @@ -33,26 +33,18 @@ unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum) * better 64-bit) boundary */ -unsigned int -csum_partial_copy_nocheck(const char *src, char *dst, int len, int sum); +__wsum +csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum); -unsigned int -csum_partial_copy_from_user(const char __user *src, char *dst, int len, int sum, int *err_ptr); - -/* - * This is the old (and unsafe) way of doing checksums, a warning message will - * be printed if it is used and an exception occurs. - * - * this functions should go away after some time. - */ -#define csum_partial_copy(src,dst,len,sum) csum_partial_copy_nocheck(src,dst,len,sum) +__wsum +csum_partial_copy_from_user(const void __user *src, void *dst, int len, __wsum sum, int *err_ptr); /* * This is a version of ip_compute_csum() optimized for IP headers, * which always checksum on 4 octet boundaries. */ -static inline unsigned short -ip_fast_csum(unsigned char * iph, unsigned int ihl) +static inline __sum16 +ip_fast_csum(const void *iph, unsigned int ihl) { unsigned int sum, tmp1; @@ -78,14 +70,13 @@ ip_fast_csum(unsigned char * iph, unsigned int ihl) : "=r" (sum), "=r" (iph), "=r" (ihl), "=r" (tmp1) : "1" (iph), "2" (ihl) : "cc"); - return sum; + return (__force __sum16)sum; } /* * Fold a partial checksum without adding pseudo headers */ -static inline unsigned int -csum_fold(unsigned int sum) +static inline __sum16 csum_fold(__wsum sum) { __asm__( "adds %0, %1, %1, lsl #16 @ csum_fold \n\ @@ -93,12 +84,12 @@ csum_fold(unsigned int sum) : "=r" (sum) : "r" (sum) : "cc"); - return (~sum) >> 16; + return (__force __sum16)(~(__force u32)sum >> 16); } -static inline unsigned int -csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, unsigned short len, - unsigned int proto, unsigned int sum) +static inline __wsum +csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, + unsigned short proto, __wsum sum) { __asm__( "adds %0, %1, %2 @ csum_tcpudp_nofold \n\ @@ -107,7 +98,7 @@ csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, unsigned short len, adcs %0, %0, %5 \n\ adc %0, %0, #0" : "=&r"(sum) - : "r" (sum), "r" (daddr), "r" (saddr), "r" (ntohs(len)), "Ir" (ntohs(proto)) + : "r" (sum), "r" (daddr), "r" (saddr), "r" (htons(len)), "Ir" (htons(proto)) : "cc"); return sum; } @@ -115,9 +106,9 @@ csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, unsigned short len, * computes the checksum of the TCP/UDP pseudo-header * returns a 16-bit checksum, already complemented */ -static inline unsigned short int -csum_tcpudp_magic(unsigned long saddr, unsigned long daddr, unsigned short len, - unsigned int proto, unsigned int sum) +static inline __sum16 +csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, + unsigned short proto, __wsum sum) { __asm__( "adds %0, %1, %2 @ csum_tcpudp_magic \n\ @@ -129,9 +120,9 @@ csum_tcpudp_magic(unsigned long saddr, unsigned long daddr, unsigned short len, addcs %0, %0, #0x10000 \n\ mvn %0, %0" : "=&r"(sum) - : "r" (sum), "r" (daddr), "r" (saddr), "r" (ntohs(len)), "Ir" (ntohs(proto)) + : "r" (sum), "r" (daddr), "r" (saddr), "r" (htons(len)), "Ir" (htons(proto)) : "cc"); - return sum >> 16; + return (__force __sum16)((__force u32)sum >> 16); } @@ -139,20 +130,20 @@ csum_tcpudp_magic(unsigned long saddr, unsigned long daddr, unsigned short len, * this routine is used for miscellaneous IP-like checksums, mainly * in icmp.c */ -static inline unsigned short -ip_compute_csum(unsigned char * buff, int len) +static inline __sum16 +ip_compute_csum(const void *buff, int len) { return csum_fold(csum_partial(buff, len, 0)); } #define _HAVE_ARCH_IPV6_CSUM -extern unsigned long -__csum_ipv6_magic(struct in6_addr *saddr, struct in6_addr *daddr, __u32 len, - __u32 proto, unsigned int sum); +extern __wsum +__csum_ipv6_magic(const struct in6_addr *saddr, const struct in6_addr *daddr, __be32 len, + __be32 proto, __wsum sum); -static inline unsigned short int -csum_ipv6_magic(struct in6_addr *saddr, struct in6_addr *daddr, __u32 len, - unsigned short proto, unsigned int sum) +static inline __sum16 +csum_ipv6_magic(const struct in6_addr *saddr, const struct in6_addr *daddr, __u32 len, + unsigned short proto, __wsum sum) { return csum_fold(__csum_ipv6_magic(saddr, daddr, htonl(len), htonl(proto), sum)); -- cgit v1.2.3-70-g09d2 From 475b8f311bf3e9b5a024b779435f408395a76890 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:21:15 -0800 Subject: [NET]: AVR32 checksum annotations and cleanups. * sanitize prototypes, annotate * kill useless shifts * kill useless ntohs (it's big-endian, for fsck sake!) Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/asm-avr32/checksum.h | 40 ++++++++++++++++++---------------------- 1 file changed, 18 insertions(+), 22 deletions(-) (limited to 'include') diff --git a/include/asm-avr32/checksum.h b/include/asm-avr32/checksum.h index 41b7af09edc..af9d53f0f5d 100644 --- a/include/asm-avr32/checksum.h +++ b/include/asm-avr32/checksum.h @@ -20,8 +20,7 @@ * * it's best to have buff aligned on a 32-bit boundary */ -unsigned int csum_partial(const unsigned char * buff, int len, - unsigned int sum); +__wsum csum_partial(const void *buff, int len, __wsum sum); /* * the same as csum_partial, but copies from src while it @@ -30,8 +29,8 @@ unsigned int csum_partial(const unsigned char * buff, int len, * here even more important to align src and dst on a 32-bit (or even * better 64-bit) boundary */ -unsigned int csum_partial_copy_generic(const char *src, char *dst, int len, - int sum, int *src_err_ptr, +__wsum csum_partial_copy_generic(const void *src, void *dst, int len, + __wsum sum, int *src_err_ptr, int *dst_err_ptr); /* @@ -42,17 +41,17 @@ unsigned int csum_partial_copy_generic(const char *src, char *dst, int len, * verify_area(). */ static inline -unsigned int csum_partial_copy_nocheck(const char *src, char *dst, - int len, int sum) +__wsum csum_partial_copy_nocheck(const void *src, void *dst, + int len, __wsum sum) { return csum_partial_copy_generic(src, dst, len, sum, NULL, NULL); } static inline -unsigned int csum_partial_copy_from_user (const char __user *src, char *dst, - int len, int sum, int *err_ptr) +__wsum csum_partial_copy_from_user(const void __user *src, void *dst, + int len, __wsum sum, int *err_ptr) { - return csum_partial_copy_generic((const char __force *)src, dst, len, + return csum_partial_copy_generic((const void __force *)src, dst, len, sum, err_ptr, NULL); } @@ -60,8 +59,7 @@ unsigned int csum_partial_copy_from_user (const char __user *src, char *dst, * This is a version of ip_compute_csum() optimized for IP headers, * which always checksum on 4 octet boundaries. */ -static inline unsigned short ip_fast_csum(unsigned char *iph, - unsigned int ihl) +static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl) { unsigned int sum, tmp; @@ -90,14 +88,14 @@ static inline unsigned short ip_fast_csum(unsigned char *iph, : "=r"(sum), "=r"(iph), "=r"(ihl), "=r"(tmp) : "1"(iph), "2"(ihl) : "memory", "cc"); - return sum; + return (__force __sum16)sum; } /* * Fold a partial checksum */ -static inline unsigned int csum_fold(unsigned int sum) +static inline __sum16 csum_fold(__wsum sum) { unsigned int tmp; @@ -109,21 +107,20 @@ static inline unsigned int csum_fold(unsigned int sum) : "=&r"(sum), "=&r"(tmp) : "0"(sum)); - return ~sum; + return (__force __sum16)~sum; } -static inline unsigned long csum_tcpudp_nofold(unsigned long saddr, - unsigned long daddr, +static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { asm(" add %0, %1\n" " adc %0, %0, %2\n" " adc %0, %0, %3\n" " acr %0" : "=r"(sum) - : "r"(daddr), "r"(saddr), "r"(ntohs(len) | (proto << 16)), + : "r"(daddr), "r"(saddr), "r"(len + proto), "0"(sum) : "cc"); @@ -134,11 +131,10 @@ static inline unsigned long csum_tcpudp_nofold(unsigned long saddr, * computes the checksum of the TCP/UDP pseudo-header * returns a 16-bit checksum, already complemented */ -static inline unsigned short int csum_tcpudp_magic(unsigned long saddr, - unsigned long daddr, +static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); } @@ -148,7 +144,7 @@ static inline unsigned short int csum_tcpudp_magic(unsigned long saddr, * in icmp.c */ -static inline unsigned short ip_compute_csum(unsigned char * buff, int len) +static inline __sum16 ip_compute_csum(const void *buff, int len) { return csum_fold(csum_partial(buff, len, 0)); } -- cgit v1.2.3-70-g09d2 From 72685fcd286e94fef0b692f634d304b7240cef04 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:21:37 -0800 Subject: [NET]: I386 checksum annotations and cleanups. * sanitize prototypes, annotate * usual ntohs->shift Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/asm-i386/checksum.h | 65 +++++++++++++++++++++------------------------ 1 file changed, 31 insertions(+), 34 deletions(-) (limited to 'include') diff --git a/include/asm-i386/checksum.h b/include/asm-i386/checksum.h index 67d3630c4e8..75194abbe8e 100644 --- a/include/asm-i386/checksum.h +++ b/include/asm-i386/checksum.h @@ -17,7 +17,7 @@ * * it's best to have buff aligned on a 32-bit boundary */ -asmlinkage unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum); +asmlinkage __wsum csum_partial(const void *buff, int len, __wsum sum); /* * the same as csum_partial, but copies from src while it @@ -27,8 +27,8 @@ asmlinkage unsigned int csum_partial(const unsigned char * buff, int len, unsign * better 64-bit) boundary */ -asmlinkage unsigned int csum_partial_copy_generic(const unsigned char *src, unsigned char *dst, - int len, int sum, int *src_err_ptr, int *dst_err_ptr); +asmlinkage __wsum csum_partial_copy_generic(const void *src, void *dst, + int len, __wsum sum, int *src_err_ptr, int *dst_err_ptr); /* * Note: when you get a NULL pointer exception here this means someone @@ -38,18 +38,18 @@ asmlinkage unsigned int csum_partial_copy_generic(const unsigned char *src, unsi * access_ok(). */ static __inline__ -unsigned int csum_partial_copy_nocheck (const unsigned char *src, unsigned char *dst, - int len, int sum) +__wsum csum_partial_copy_nocheck (const void *src, void *dst, + int len, __wsum sum) { return csum_partial_copy_generic ( src, dst, len, sum, NULL, NULL); } static __inline__ -unsigned int csum_partial_copy_from_user(const unsigned char __user *src, unsigned char *dst, - int len, int sum, int *err_ptr) +__wsum csum_partial_copy_from_user(const void __user *src, void *dst, + int len, __wsum sum, int *err_ptr) { might_sleep(); - return csum_partial_copy_generic((__force unsigned char *)src, dst, + return csum_partial_copy_generic((__force void *)src, dst, len, sum, err_ptr, NULL); } @@ -60,8 +60,7 @@ unsigned int csum_partial_copy_from_user(const unsigned char __user *src, unsign * By Jorge Cwik , adapted for linux by * Arnt Gulbrandsen. */ -static inline unsigned short ip_fast_csum(unsigned char * iph, - unsigned int ihl) +static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl) { unsigned int sum; @@ -89,29 +88,29 @@ static inline unsigned short ip_fast_csum(unsigned char * iph, : "=r" (sum), "=r" (iph), "=r" (ihl) : "1" (iph), "2" (ihl) : "memory"); - return(sum); + return (__force __sum16)sum; } /* * Fold a partial checksum */ -static inline unsigned int csum_fold(unsigned int sum) +static inline __sum16 csum_fold(__wsum sum) { __asm__( "addl %1, %0 ;\n" "adcl $0xffff, %0 ;\n" : "=r" (sum) - : "r" (sum << 16), "0" (sum & 0xffff0000) + : "r" ((__force u32)sum << 16), + "0" ((__force u32)sum & 0xffff0000) ); - return (~sum) >> 16; + return (__force __sum16)(~(__force u32)sum >> 16); } -static inline unsigned long csum_tcpudp_nofold(unsigned long saddr, - unsigned long daddr, - unsigned short len, - unsigned short proto, - unsigned int sum) +static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr, + unsigned short len, + unsigned short proto, + __wsum sum) { __asm__( "addl %1, %0 ;\n" @@ -119,7 +118,7 @@ static inline unsigned long csum_tcpudp_nofold(unsigned long saddr, "adcl %3, %0 ;\n" "adcl $0, %0 ;\n" : "=r" (sum) - : "g" (daddr), "g"(saddr), "g"((ntohs(len)<<16)+proto*256), "0"(sum)); + : "g" (daddr), "g"(saddr), "g"((len + proto) << 8), "0"(sum)); return sum; } @@ -127,11 +126,10 @@ static inline unsigned long csum_tcpudp_nofold(unsigned long saddr, * computes the checksum of the TCP/UDP pseudo-header * returns a 16-bit checksum, already complemented */ -static inline unsigned short int csum_tcpudp_magic(unsigned long saddr, - unsigned long daddr, +static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); } @@ -141,17 +139,16 @@ static inline unsigned short int csum_tcpudp_magic(unsigned long saddr, * in icmp.c */ -static inline unsigned short ip_compute_csum(unsigned char * buff, int len) +static inline __sum16 ip_compute_csum(const void *buff, int len) { return csum_fold (csum_partial(buff, len, 0)); } #define _HAVE_ARCH_IPV6_CSUM -static __inline__ unsigned short int csum_ipv6_magic(struct in6_addr *saddr, - struct in6_addr *daddr, - __u32 len, - unsigned short proto, - unsigned int sum) +static __inline__ __sum16 csum_ipv6_magic(const struct in6_addr *saddr, + const struct in6_addr *daddr, + __u32 len, unsigned short proto, + __wsum sum) { __asm__( "addl 0(%1), %0 ;\n" @@ -176,19 +173,19 @@ static __inline__ unsigned short int csum_ipv6_magic(struct in6_addr *saddr, * Copy and checksum to user */ #define HAVE_CSUM_COPY_USER -static __inline__ unsigned int csum_and_copy_to_user(const unsigned char *src, - unsigned char __user *dst, - int len, int sum, +static __inline__ __wsum csum_and_copy_to_user(const void *src, + void __user *dst, + int len, __wsum sum, int *err_ptr) { might_sleep(); if (access_ok(VERIFY_WRITE, dst, len)) - return csum_partial_copy_generic(src, (__force unsigned char *)dst, len, sum, NULL, err_ptr); + return csum_partial_copy_generic(src, (__force void *)dst, len, sum, NULL, err_ptr); if (len) *err_ptr = -EFAULT; - return -1; /* invalid checksum */ + return (__force __wsum)-1; /* invalid checksum */ } #endif -- cgit v1.2.3-70-g09d2 From 879178cfbe56baa42ee73b9668816872c97d8ccd Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:21:58 -0800 Subject: [NET]: POWERPC checksum annotations and cleanups. * sanitize prototypes, annotate * kill useless shifts Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/asm-powerpc/checksum.h | 59 +++++++++++++++--------------------------- 1 file changed, 21 insertions(+), 38 deletions(-) (limited to 'include') diff --git a/include/asm-powerpc/checksum.h b/include/asm-powerpc/checksum.h index 609ecbbd721..7cdf358337c 100644 --- a/include/asm-powerpc/checksum.h +++ b/include/asm-powerpc/checksum.h @@ -14,17 +14,16 @@ * which always checksum on 4 octet boundaries. ihl is the number * of 32-bit words and is always >= 5. */ -extern unsigned short ip_fast_csum(unsigned char * iph, unsigned int ihl); +extern __sum16 ip_fast_csum(const void *iph, unsigned int ihl); /* * computes the checksum of the TCP/UDP pseudo-header * returns a 16-bit checksum, already complemented */ -extern unsigned short csum_tcpudp_magic(unsigned long saddr, - unsigned long daddr, +extern __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum); + __wsum sum); /* * computes the checksum of a memory block at buff, length len, @@ -38,8 +37,7 @@ extern unsigned short csum_tcpudp_magic(unsigned long saddr, * * it's best to have buff aligned on a 32-bit boundary */ -extern unsigned int csum_partial(const unsigned char * buff, int len, - unsigned int sum); +extern __wsum csum_partial(const void *buff, int len, __wsum sum); /* * Computes the checksum of a memory block at src, length len, @@ -51,20 +49,15 @@ extern unsigned int csum_partial(const unsigned char * buff, int len, * Like csum_partial, this must be called with even lengths, * except for the last fragment. */ -extern unsigned int csum_partial_copy_generic(const char *src, char *dst, - int len, unsigned int sum, +extern __wsum csum_partial_copy_generic(const void *src, void *dst, + int len, __wsum sum, int *src_err, int *dst_err); /* * the same as csum_partial, but copies from src to dst while it * checksums. */ -unsigned int csum_partial_copy_nocheck(const char *src, - char *dst, - int len, - unsigned int sum); - #define csum_partial_copy_from_user(src, dst, len, sum, errp) \ - csum_partial_copy_generic((src), (dst), (len), (sum), (errp), NULL) + csum_partial_copy_generic((__force const void *)(src), (dst), (len), (sum), (errp), NULL) #define csum_partial_copy_nocheck(src, dst, len, sum) \ csum_partial_copy_generic((src), (dst), (len), (sum), NULL, NULL) @@ -74,7 +67,7 @@ unsigned int csum_partial_copy_nocheck(const char *src, * turns a 32-bit partial checksum (e.g. from csum_partial) into a * 1's complement 16-bit checksum. */ -static inline unsigned int csum_fold(unsigned int sum) +static inline __sum16 csum_fold(__wsum sum) { unsigned int tmp; @@ -83,41 +76,32 @@ static inline unsigned int csum_fold(unsigned int sum) /* if there is a carry from adding the two 16-bit halves, it will carry from the lower half into the upper half, giving us the correct sum in the upper half. */ - sum = ~(sum + tmp) >> 16; - return sum; + return (__force __sum16)(~((__force u32)sum + tmp) >> 16); } /* * this routine is used for miscellaneous IP-like checksums, mainly * in icmp.c */ -static inline unsigned short ip_compute_csum(unsigned char * buff, int len) +static inline __sum16 ip_compute_csum(const void *buff, int len) { return csum_fold(csum_partial(buff, len, 0)); } -#ifdef __powerpc64__ -static inline u32 csum_tcpudp_nofold(u32 saddr, - u32 daddr, +static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { - unsigned long s = sum; +#ifdef __powerpc64__ + unsigned long s = (__force u32)sum; - s += saddr; - s += daddr; - s += (proto << 16) + len; + s += (__force u32)saddr; + s += (__force u32)daddr; + s += proto + len; s += (s >> 32); - return (u32) s; -} + return (__force __wsum) s; #else -static inline unsigned long csum_tcpudp_nofold(unsigned long saddr, - unsigned long daddr, - unsigned short len, - unsigned short proto, - unsigned int sum) -{ __asm__("\n\ addc %0,%0,%1 \n\ adde %0,%0,%2 \n\ @@ -125,10 +109,9 @@ static inline unsigned long csum_tcpudp_nofold(unsigned long saddr, addze %0,%0 \n\ " : "=r" (sum) - : "r" (daddr), "r"(saddr), "r"((proto<<16)+len), "0"(sum)); - return sum; -} - + : "r" (daddr), "r"(saddr), "r"(proto + len), "0"(sum)); + return sum; #endif +} #endif /* __KERNEL__ */ #endif -- cgit v1.2.3-70-g09d2 From f994aae1bd8e4813d59a2ed64d17585fe42d03fc Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:22:18 -0800 Subject: [NET]: S390 checksum annotations and cleanups. * sanitize prototypes, annotate * kill useless shifts Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/asm-s390/checksum.h | 45 +++++++++++++++++++++------------------------ 1 file changed, 21 insertions(+), 24 deletions(-) (limited to 'include') diff --git a/include/asm-s390/checksum.h b/include/asm-s390/checksum.h index 37c362d89fa..0a3cd7ec845 100644 --- a/include/asm-s390/checksum.h +++ b/include/asm-s390/checksum.h @@ -27,8 +27,8 @@ * * it's best to have buff aligned on a 32-bit boundary */ -static inline unsigned int -csum_partial(const unsigned char * buff, int len, unsigned int sum) +static inline __wsum +csum_partial(const void *buff, int len, __wsum sum) { register unsigned long reg2 asm("2") = (unsigned long) buff; register unsigned long reg3 asm("3") = (unsigned long) len; @@ -49,9 +49,9 @@ csum_partial(const unsigned char * buff, int len, unsigned int sum) * Copy from userspace and compute checksum. If we catch an exception * then zero the rest of the buffer. */ -static inline unsigned int -csum_partial_copy_from_user(const char __user *src, char *dst, - int len, unsigned int sum, +static inline __wsum +csum_partial_copy_from_user(const void __user *src, void *dst, + int len, __wsum sum, int *err_ptr) { int missing; @@ -66,8 +66,8 @@ csum_partial_copy_from_user(const char __user *src, char *dst, } -static inline unsigned int -csum_partial_copy_nocheck (const char *src, char *dst, int len, unsigned int sum) +static inline __wsum +csum_partial_copy_nocheck (const void *src, void *dst, int len, __wsum sum) { memcpy(dst,src,len); return csum_partial(dst, len, sum); @@ -76,8 +76,7 @@ csum_partial_copy_nocheck (const char *src, char *dst, int len, unsigned int sum /* * Fold a partial checksum without adding pseudo headers */ -static inline unsigned short -csum_fold(unsigned int sum) +static inline __sum16 csum_fold(__wsum sum) { #ifndef __s390x__ register_pair rp; @@ -100,7 +99,7 @@ csum_fold(unsigned int sum) " srl %0,16\n" /* %0 = H+L+C */ : "+&d" (sum) : : "cc", "2", "3"); #endif /* __s390x__ */ - return ((unsigned short) ~sum); + return (__force __sum16) ~sum; } /* @@ -108,8 +107,7 @@ csum_fold(unsigned int sum) * which always checksum on 4 octet boundaries. * */ -static inline unsigned short -ip_fast_csum(unsigned char *iph, unsigned int ihl) +static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl) { return csum_fold(csum_partial(iph, ihl*4, 0)); } @@ -118,10 +116,10 @@ ip_fast_csum(unsigned char *iph, unsigned int ihl) * computes the checksum of the TCP/UDP pseudo-header * returns a 32-bit checksum */ -static inline unsigned int -csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, +static inline __wsum +csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { #ifndef __s390x__ asm volatile( @@ -137,12 +135,12 @@ csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, "1:" : "+&d" (sum) : "d" (daddr) : "cc"); asm volatile( - " alr %0,%1\n" /* sum += (len<<16) + (proto<<8) */ + " alr %0,%1\n" /* sum += len + proto */ " brc 12,2f\n" " ahi %0,1\n" /* add carry */ "2:" : "+&d" (sum) - : "d" (((unsigned int) len<<16) + (unsigned int) proto) + : "d" (len + proto) : "cc"); #else /* __s390x__ */ asm volatile( @@ -153,7 +151,7 @@ csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, "0: algr %0,%2\n" /* sum += daddr */ " brc 12,1f\n" " aghi %0,1\n" /* add carry */ - "1: algfr %0,%3\n" /* sum += (len<<16) + proto */ + "1: algfr %0,%3\n" /* sum += len + proto */ " brc 12,2f\n" " aghi %0,1\n" /* add carry */ "2: srlg 0,%0,32\n" @@ -163,7 +161,7 @@ csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, "3: llgfr %0,%0" : "+&d" (sum) : "d" (saddr), "d" (daddr), - "d" (((unsigned int) len<<16) + (unsigned int) proto) + "d" (len + proto) : "cc", "0"); #endif /* __s390x__ */ return sum; @@ -174,10 +172,10 @@ csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr, * returns a 16-bit checksum, already complemented */ -static inline unsigned short int -csum_tcpudp_magic(unsigned long saddr, unsigned long daddr, +static inline __sum16 +csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); } @@ -187,8 +185,7 @@ csum_tcpudp_magic(unsigned long saddr, unsigned long daddr, * in icmp.c */ -static inline unsigned short -ip_compute_csum(unsigned char * buff, int len) +static inline __sum16 ip_compute_csum(const void *buff, int len) { return csum_fold(csum_partial(buff, len, 0)); } -- cgit v1.2.3-70-g09d2 From 7c73a746ba726faa5ab12981b4b192768eb439de Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:22:35 -0800 Subject: [NET]: SH checksum annotations and cleanups. * sanitize prototypes, annotate * kill useless shifts * usual ntohs->shift Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/asm-sh/checksum.h | 59 +++++++++++++++++++++++------------------------ 1 file changed, 29 insertions(+), 30 deletions(-) (limited to 'include') diff --git a/include/asm-sh/checksum.h b/include/asm-sh/checksum.h index 08168afe674..d44344c88e7 100644 --- a/include/asm-sh/checksum.h +++ b/include/asm-sh/checksum.h @@ -23,7 +23,7 @@ * * it's best to have buff aligned on a 32-bit boundary */ -asmlinkage unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum); +asmlinkage __wsum csum_partial(const void *buff, int len, __wsum sum); /* * the same as csum_partial, but copies from src while it @@ -33,8 +33,8 @@ asmlinkage unsigned int csum_partial(const unsigned char * buff, int len, unsign * better 64-bit) boundary */ -asmlinkage unsigned int csum_partial_copy_generic(const unsigned char *src, unsigned char *dst, - int len, int sum, int *src_err_ptr, int *dst_err_ptr); +asmlinkage __wsum csum_partial_copy_generic(const void *src, void *dst, + int len, __wsum sum, int *src_err_ptr, int *dst_err_ptr); /* * Note: when you get a NULL pointer exception here this means someone @@ -44,24 +44,25 @@ asmlinkage unsigned int csum_partial_copy_generic(const unsigned char *src, unsi * access_ok(). */ static __inline__ -unsigned int csum_partial_copy_nocheck (const unsigned char *src, unsigned char *dst, - int len, int sum) +__wsum csum_partial_copy_nocheck(const void *src, void *dst, + int len, __wsum sum) { return csum_partial_copy_generic ( src, dst, len, sum, NULL, NULL); } static __inline__ -unsigned int csum_partial_copy_from_user (const unsigned char *src, unsigned char *dst, - int len, int sum, int *err_ptr) +__wsum csum_partial_copy_from_user(const void __user *src, void *dst, + int len, __wsum sum, int *err_ptr) { - return csum_partial_copy_generic ( src, dst, len, sum, err_ptr, NULL); + return csum_partial_copy_generic((__force const void *)src, dst, + len, sum, err_ptr, NULL); } /* * Fold a partial checksum */ -static __inline__ unsigned int csum_fold(unsigned int sum) +static __inline__ __sum16 csum_fold(__wsum sum) { unsigned int __dummy; __asm__("swap.w %0, %1\n\t" @@ -74,7 +75,7 @@ static __inline__ unsigned int csum_fold(unsigned int sum) : "=r" (sum), "=&r" (__dummy) : "0" (sum) : "t"); - return sum; + return (__force __sum16)sum; } /* @@ -84,7 +85,7 @@ static __inline__ unsigned int csum_fold(unsigned int sum) * i386 version by Jorge Cwik , adapted * for linux by * Arnt Gulbrandsen. */ -static __inline__ unsigned short ip_fast_csum(unsigned char * iph, unsigned int ihl) +static __inline__ __sum16 ip_fast_csum(const void *iph, unsigned int ihl) { unsigned int sum, __dummy0, __dummy1; @@ -112,16 +113,15 @@ static __inline__ unsigned short ip_fast_csum(unsigned char * iph, unsigned int return csum_fold(sum); } -static __inline__ unsigned long csum_tcpudp_nofold(unsigned long saddr, - unsigned long daddr, +static __inline__ __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { #ifdef __LITTLE_ENDIAN__ - unsigned long len_proto = (ntohs(len)<<16)+proto*256; + unsigned long len_proto = (proto + len) << 8; #else - unsigned long len_proto = (proto<<16)+len; + unsigned long len_proto = proto + len; #endif __asm__("clrt\n\t" "addc %0, %1\n\t" @@ -139,11 +139,10 @@ static __inline__ unsigned long csum_tcpudp_nofold(unsigned long saddr, * computes the checksum of the TCP/UDP pseudo-header * returns a 16-bit checksum, already complemented */ -static __inline__ unsigned short int csum_tcpudp_magic(unsigned long saddr, - unsigned long daddr, +static __inline__ __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); } @@ -153,18 +152,17 @@ static __inline__ unsigned short int csum_tcpudp_magic(unsigned long saddr, * in icmp.c */ -static __inline__ unsigned short ip_compute_csum(unsigned char * buff, int len) +static __inline__ __sum16 ip_compute_csum(const void *buff, int len) { return csum_fold (csum_partial(buff, len, 0)); } #define _HAVE_ARCH_IPV6_CSUM #ifdef CONFIG_IPV6 -static __inline__ unsigned short int csum_ipv6_magic(struct in6_addr *saddr, - struct in6_addr *daddr, - __u32 len, - unsigned short proto, - unsigned int sum) +static __inline__ __sum16 csum_ipv6_magic(const struct in6_addr *saddr, + const struct in6_addr *daddr, + __u32 len, unsigned short proto, + __wsum sum) { unsigned int __dummy; __asm__("clrt\n\t" @@ -201,17 +199,18 @@ static __inline__ unsigned short int csum_ipv6_magic(struct in6_addr *saddr, * Copy and checksum to user */ #define HAVE_CSUM_COPY_USER -static __inline__ unsigned int csum_and_copy_to_user (const unsigned char *src, - unsigned char __user *dst, - int len, int sum, +static __inline__ __wsum csum_and_copy_to_user (const void *src, + void __user *dst, + int len, __wsum sum, int *err_ptr) { if (access_ok(VERIFY_WRITE, dst, len)) - return csum_partial_copy_generic(src, dst, len, sum, NULL, err_ptr); + return csum_partial_copy_generic((__force const void *)src, + dst, len, sum, NULL, err_ptr); if (len) *err_ptr = -EFAULT; - return -1; /* invalid checksum */ + return (__force __wsum)-1; /* invalid checksum */ } #endif /* __ASM_SH_CHECKSUM_H */ -- cgit v1.2.3-70-g09d2 From 16ec7e168db6747e69b101b86ddf86f001fdcd60 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:22:56 -0800 Subject: [NET]: SPARC checksum annotations and cleanups. * sanitize prototypes, annotate * kill bogus access_ok() in csum_partial_copy_from_user (the only caller checks) * kill useless shift Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/asm-sparc/checksum.h | 103 +++++++++++++++++++------------------------ 1 file changed, 46 insertions(+), 57 deletions(-) (limited to 'include') diff --git a/include/asm-sparc/checksum.h b/include/asm-sparc/checksum.h index 28615810897..267e631e9bb 100644 --- a/include/asm-sparc/checksum.h +++ b/include/asm-sparc/checksum.h @@ -30,7 +30,7 @@ * * it's best to have buff aligned on a 32-bit boundary */ -extern unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum); +extern __wsum csum_partial(const void *buff, int len, __wsum sum); /* the same as csum_partial, but copies from fs:src while it * checksums @@ -41,9 +41,8 @@ extern unsigned int csum_partial(const unsigned char * buff, int len, unsigned i extern unsigned int __csum_partial_copy_sparc_generic (const unsigned char *, unsigned char *); -static inline unsigned int -csum_partial_copy_nocheck (const unsigned char *src, unsigned char *dst, int len, - unsigned int sum) +static inline __wsum +csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum) { register unsigned int ret asm("o0") = (unsigned int)src; register char *d asm("o1") = dst; @@ -57,42 +56,36 @@ csum_partial_copy_nocheck (const unsigned char *src, unsigned char *dst, int len : "o2", "o3", "o4", "o5", "o7", "g2", "g3", "g4", "g5", "g7", "memory", "cc"); - return ret; + return (__force __wsum)ret; } -static inline unsigned int -csum_partial_copy_from_user(const unsigned char __user *src, unsigned char *dst, int len, - unsigned int sum, int *err) +static inline __wsum +csum_partial_copy_from_user(const void __user *src, void *dst, int len, + __wsum sum, int *err) { - if (!access_ok (VERIFY_READ, src, len)) { - *err = -EFAULT; - memset (dst, 0, len); - return sum; - } else { - register unsigned long ret asm("o0") = (unsigned long)src; - register char *d asm("o1") = dst; - register int l asm("g1") = len; - register unsigned int s asm("g7") = sum; + register unsigned long ret asm("o0") = (unsigned long)src; + register char *d asm("o1") = dst; + register int l asm("g1") = len; + register __wsum s asm("g7") = sum; - __asm__ __volatile__ ( - ".section __ex_table,#alloc\n\t" - ".align 4\n\t" - ".word 1f,2\n\t" - ".previous\n" - "1:\n\t" - "call __csum_partial_copy_sparc_generic\n\t" - " st %8, [%%sp + 64]\n" - : "=&r" (ret), "=&r" (d), "=&r" (l), "=&r" (s) - : "0" (ret), "1" (d), "2" (l), "3" (s), "r" (err) - : "o2", "o3", "o4", "o5", "o7", "g2", "g3", "g4", "g5", - "cc", "memory"); - return ret; - } - } + __asm__ __volatile__ ( + ".section __ex_table,#alloc\n\t" + ".align 4\n\t" + ".word 1f,2\n\t" + ".previous\n" + "1:\n\t" + "call __csum_partial_copy_sparc_generic\n\t" + " st %8, [%%sp + 64]\n" + : "=&r" (ret), "=&r" (d), "=&r" (l), "=&r" (s) + : "0" (ret), "1" (d), "2" (l), "3" (s), "r" (err) + : "o2", "o3", "o4", "o5", "o7", "g2", "g3", "g4", "g5", + "cc", "memory"); + return (__force __wsum)ret; +} -static inline unsigned int -csum_partial_copy_to_user(const unsigned char *src, unsigned char __user *dst, int len, - unsigned int sum, int *err) +static inline __wsum +csum_partial_copy_to_user(const void *src, void __user *dst, int len, + __wsum sum, int *err) { if (!access_ok (VERIFY_WRITE, dst, len)) { *err = -EFAULT; @@ -101,7 +94,7 @@ csum_partial_copy_to_user(const unsigned char *src, unsigned char __user *dst, i register unsigned long ret asm("o0") = (unsigned long)src; register char __user *d asm("o1") = dst; register int l asm("g1") = len; - register unsigned int s asm("g7") = sum; + register __wsum s asm("g7") = sum; __asm__ __volatile__ ( ".section __ex_table,#alloc\n\t" @@ -116,7 +109,7 @@ csum_partial_copy_to_user(const unsigned char *src, unsigned char __user *dst, i : "o2", "o3", "o4", "o5", "o7", "g2", "g3", "g4", "g5", "cc", "memory"); - return ret; + return (__force __wsum)ret; } } @@ -126,10 +119,9 @@ csum_partial_copy_to_user(const unsigned char *src, unsigned char __user *dst, i /* ihl is always 5 or greater, almost always is 5, and iph is word aligned * the majority of the time. */ -static inline unsigned short ip_fast_csum(const unsigned char *iph, - unsigned int ihl) +static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl) { - unsigned short sum; + __sum16 sum; /* Note: We must read %2 before we touch %0 for the first time, * because GCC can legitimately use the same register for @@ -164,7 +156,7 @@ static inline unsigned short ip_fast_csum(const unsigned char *iph, } /* Fold a partial checksum without adding pseudo headers. */ -static inline unsigned int csum_fold(unsigned int sum) +static inline __sum16 csum_fold(__wsum sum) { unsigned int tmp; @@ -173,23 +165,22 @@ static inline unsigned int csum_fold(unsigned int sum) "addx\t%1, %%g0, %1\n\t" "xnor\t%%g0, %1, %0" : "=&r" (sum), "=r" (tmp) - : "0" (sum), "1" (sum<<16) + : "0" (sum), "1" ((__force u32)sum<<16) : "cc"); - return sum; + return (__force __sum16)sum; } -static inline unsigned long csum_tcpudp_nofold(unsigned long saddr, - unsigned long daddr, - unsigned int len, +static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr, + unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { __asm__ __volatile__("addcc\t%1, %0, %0\n\t" "addxcc\t%2, %0, %0\n\t" "addxcc\t%3, %0, %0\n\t" "addx\t%0, %%g0, %0\n\t" : "=r" (sum), "=r" (saddr) - : "r" (daddr), "r" ((proto<<16)+len), "0" (sum), + : "r" (daddr), "r" (proto + len), "0" (sum), "1" (saddr) : "cc"); return sum; @@ -199,22 +190,20 @@ static inline unsigned long csum_tcpudp_nofold(unsigned long saddr, * computes the checksum of the TCP/UDP pseudo-header * returns a 16-bit checksum, already complemented */ -static inline unsigned short int csum_tcpudp_magic(unsigned long saddr, - unsigned long daddr, +static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); } #define _HAVE_ARCH_IPV6_CSUM -static inline unsigned short int csum_ipv6_magic(struct in6_addr *saddr, - struct in6_addr *daddr, - __u32 len, - unsigned short proto, - unsigned int sum) +static inline __sum16 csum_ipv6_magic(const struct in6_addr *saddr, + const struct in6_addr *daddr, + __u32 len, unsigned short proto, + __wsum sum) { __asm__ __volatile__ ( "addcc %3, %4, %%g4\n\t" @@ -245,7 +234,7 @@ static inline unsigned short int csum_ipv6_magic(struct in6_addr *saddr, } /* this routine is used for miscellaneous IP-like checksums, mainly in icmp.c */ -static inline unsigned short ip_compute_csum(unsigned char * buff, int len) +static inline __sum16 ip_compute_csum(const void *buff, int len) { return csum_fold(csum_partial(buff, len, 0)); } -- cgit v1.2.3-70-g09d2 From d5c63936410fbcabd92df1ac12f3f63ca23c7a86 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:23:20 -0800 Subject: [NET]: SPARC64 checksum annotations and cleanups. * sanitize prototypes, annotate * kill useless shift Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/asm-sparc64/checksum.h | 77 ++++++++++++++++++++---------------------- 1 file changed, 36 insertions(+), 41 deletions(-) (limited to 'include') diff --git a/include/asm-sparc64/checksum.h b/include/asm-sparc64/checksum.h index dc8bed246fc..70a006da763 100644 --- a/include/asm-sparc64/checksum.h +++ b/include/asm-sparc64/checksum.h @@ -30,7 +30,7 @@ * * it's best to have buff aligned on a 32-bit boundary */ -extern unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum); +extern __wsum csum_partial(const void * buff, int len, __wsum sum); /* the same as csum_partial, but copies from user space while it * checksums @@ -38,52 +38,50 @@ extern unsigned int csum_partial(const unsigned char * buff, int len, unsigned i * here even more important to align src and dst on a 32-bit (or even * better 64-bit) boundary */ -extern unsigned int csum_partial_copy_nocheck(const unsigned char *src, - unsigned char *dst, - int len, unsigned int sum); - -extern long __csum_partial_copy_from_user(const unsigned char __user *src, - unsigned char *dst, int len, - unsigned int sum); - -static inline unsigned int -csum_partial_copy_from_user(const unsigned char __user *src, - unsigned char *dst, int len, - unsigned int sum, int *err) +extern __wsum csum_partial_copy_nocheck(const void *src, void *dst, + int len, __wsum sum); + +extern long __csum_partial_copy_from_user(const void __user *src, + void *dst, int len, + __wsum sum); + +static inline __wsum +csum_partial_copy_from_user(const void __user *src, + void *dst, int len, + __wsum sum, int *err) { long ret = __csum_partial_copy_from_user(src, dst, len, sum); if (ret < 0) *err = -EFAULT; - return (unsigned int) ret; + return (__force __wsum) ret; } /* * Copy and checksum to user */ #define HAVE_CSUM_COPY_USER -extern long __csum_partial_copy_to_user(const unsigned char *src, - unsigned char __user *dst, int len, - unsigned int sum); - -static inline unsigned int -csum_and_copy_to_user(const unsigned char *src, - unsigned char __user *dst, int len, - unsigned int sum, int *err) +extern long __csum_partial_copy_to_user(const void *src, + void __user *dst, int len, + __wsum sum); + +static inline __wsum +csum_and_copy_to_user(const void *src, + void __user *dst, int len, + __wsum sum, int *err) { long ret = __csum_partial_copy_to_user(src, dst, len, sum); if (ret < 0) *err = -EFAULT; - return (unsigned int) ret; + return (__force __wsum) ret; } /* ihl is always 5 or greater, almost always is 5, and iph is word aligned * the majority of the time. */ -extern unsigned short ip_fast_csum(__const__ unsigned char *iph, - unsigned int ihl); +extern __sum16 ip_fast_csum(const void *iph, unsigned int ihl); /* Fold a partial checksum without adding pseudo headers. */ -static inline unsigned short csum_fold(unsigned int sum) +static inline __sum16 csum_fold(__wsum sum) { unsigned int tmp; @@ -93,16 +91,15 @@ static inline unsigned short csum_fold(unsigned int sum) " addc %1, %%g0, %1\n" " xnor %%g0, %1, %0\n" : "=&r" (sum), "=r" (tmp) - : "0" (sum), "1" (sum<<16) + : "0" (sum), "1" ((__force u32)sum<<16) : "cc"); - return (sum & 0xffff); + return (__force __sum16)sum; } -static inline unsigned long csum_tcpudp_nofold(unsigned long saddr, - unsigned long daddr, +static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned int len, unsigned short proto, - unsigned int sum) + __wsum sum) { __asm__ __volatile__( " addcc %1, %0, %0\n" @@ -110,7 +107,7 @@ static inline unsigned long csum_tcpudp_nofold(unsigned long saddr, " addccc %3, %0, %0\n" " addc %0, %%g0, %0\n" : "=r" (sum), "=r" (saddr) - : "r" (daddr), "r" ((proto<<16)+len), "0" (sum), "1" (saddr) + : "r" (daddr), "r" (proto + len), "0" (sum), "1" (saddr) : "cc"); return sum; } @@ -119,22 +116,20 @@ static inline unsigned long csum_tcpudp_nofold(unsigned long saddr, * computes the checksum of the TCP/UDP pseudo-header * returns a 16-bit checksum, already complemented */ -static inline unsigned short int csum_tcpudp_magic(unsigned long saddr, - unsigned long daddr, +static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); } #define _HAVE_ARCH_IPV6_CSUM -static inline unsigned short int csum_ipv6_magic(struct in6_addr *saddr, - struct in6_addr *daddr, - __u32 len, - unsigned short proto, - unsigned int sum) +static inline __sum16 csum_ipv6_magic(const struct in6_addr *saddr, + const struct in6_addr *daddr, + __u32 len, unsigned short proto, + __wsum sum) { __asm__ __volatile__ ( " addcc %3, %4, %%g7\n" @@ -165,7 +160,7 @@ static inline unsigned short int csum_ipv6_magic(struct in6_addr *saddr, } /* this routine is used for miscellaneous IP-like checksums, mainly in icmp.c */ -static inline unsigned short ip_compute_csum(unsigned char * buff, int len) +static inline __sum16 ip_compute_csum(const void *buff, int len) { return csum_fold(csum_partial(buff, len, 0)); } -- cgit v1.2.3-70-g09d2 From b8e4e01dd5576a14602a49829f9f9d997bb60b6b Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:23:40 -0800 Subject: [NET]: XTENSA checksum annotations and cleanups. * sanitize prototypes, annotate * kill csum_partial_copy_fromuser * kill csum_partial_copy * kill useless shifts * usual ntohs->shift Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/asm-xtensa/checksum.h | 64 ++++++++++++++++++------------------------- 1 file changed, 26 insertions(+), 38 deletions(-) (limited to 'include') diff --git a/include/asm-xtensa/checksum.h b/include/asm-xtensa/checksum.h index 03114f8d1e1..5435aff9a4b 100644 --- a/include/asm-xtensa/checksum.h +++ b/include/asm-xtensa/checksum.h @@ -26,7 +26,7 @@ * * it's best to have buff aligned on a 32-bit boundary */ -asmlinkage unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum); +asmlinkage __wsum csum_partial(const void *buff, int len, __wsum sum); /* * the same as csum_partial, but copies from src while it @@ -36,7 +36,7 @@ asmlinkage unsigned int csum_partial(const unsigned char * buff, int len, unsign * better 64-bit) boundary */ -asmlinkage unsigned int csum_partial_copy_generic( const char *src, char *dst, int len, int sum, +asmlinkage __wsum csum_partial_copy_generic(const void *src, void *dst, int len, __wsum sum, int *src_err_ptr, int *dst_err_ptr); /* @@ -46,34 +46,25 @@ asmlinkage unsigned int csum_partial_copy_generic( const char *src, char *dst, i * If you use these functions directly please don't forget the access_ok(). */ static inline -unsigned int csum_partial_copy_nocheck ( const char *src, char *dst, - int len, int sum) +__wsum csum_partial_copy_nocheck(const void *src, void *dst, + int len, __wsum sum) { - return csum_partial_copy_generic ( src, dst, len, sum, NULL, NULL); + return csum_partial_copy_generic(src, dst, len, sum, NULL, NULL); } static inline -unsigned int csum_partial_copy_from_user ( const char *src, char *dst, - int len, int sum, int *err_ptr) +__wsum csum_partial_copy_from_user(const void __user *src, void *dst, + int len, __wsum sum, int *err_ptr) { - return csum_partial_copy_generic ( src, dst, len, sum, err_ptr, NULL); + return csum_partial_copy_generic((__force const void *)src, dst, + len, sum, err_ptr, NULL); } -/* - * These are the old (and unsafe) way of doing checksums, a warning message will be - * printed if they are used and an exeption occurs. - * - * these functions should go away after some time. - */ - -#define csum_partial_copy_fromuser csum_partial_copy -unsigned int csum_partial_copy( const char *src, char *dst, int len, int sum); - /* * Fold a partial checksum */ -static __inline__ unsigned int csum_fold(unsigned int sum) +static __inline__ __sum16 csum_fold(__wsum sum) { unsigned int __dummy; __asm__("extui %1, %0, 16, 16\n\t" @@ -87,14 +78,14 @@ static __inline__ unsigned int csum_fold(unsigned int sum) "extui %0, %0, 0, 16\n\t" : "=r" (sum), "=&r" (__dummy) : "0" (sum)); - return sum; + return (__force __sum16)sum; } /* * This is a version of ip_compute_csum() optimized for IP headers, * which always checksum on 4 octet boundaries. */ -static __inline__ unsigned short ip_fast_csum(unsigned char * iph, unsigned int ihl) +static __inline__ __sum16 ip_fast_csum(const void *iph, unsigned int ihl) { unsigned int sum, tmp, endaddr; @@ -127,17 +118,16 @@ static __inline__ unsigned short ip_fast_csum(unsigned char * iph, unsigned int return csum_fold(sum); } -static __inline__ unsigned long csum_tcpudp_nofold(unsigned long saddr, - unsigned long daddr, +static __inline__ __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { #ifdef __XTENSA_EL__ - unsigned long len_proto = (ntohs(len)<<16)+proto*256; + unsigned long len_proto = (len + proto) << 8; #elif defined(__XTENSA_EB__) - unsigned long len_proto = (proto<<16)+len; + unsigned long len_proto = len + proto; #else # error processor byte order undefined! #endif @@ -162,11 +152,10 @@ static __inline__ unsigned long csum_tcpudp_nofold(unsigned long saddr, * computes the checksum of the TCP/UDP pseudo-header * returns a 16-bit checksum, already complemented */ -static __inline__ unsigned short int csum_tcpudp_magic(unsigned long saddr, - unsigned long daddr, +static __inline__ __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len, unsigned short proto, - unsigned int sum) + __wsum sum) { return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); } @@ -176,17 +165,16 @@ static __inline__ unsigned short int csum_tcpudp_magic(unsigned long saddr, * in icmp.c */ -static __inline__ unsigned short ip_compute_csum(unsigned char * buff, int len) +static __inline__ __sum16 ip_compute_csum(const void *buff, int len) { return csum_fold (csum_partial(buff, len, 0)); } #define _HAVE_ARCH_IPV6_CSUM -static __inline__ unsigned short int csum_ipv6_magic(struct in6_addr *saddr, - struct in6_addr *daddr, - __u32 len, - unsigned short proto, - unsigned int sum) +static __inline__ __sum16 csum_ipv6_magic(const struct in6_addr *saddr, + const struct in6_addr *daddr, + __u32 len, unsigned short proto, + __wsum sum) { unsigned int __dummy; __asm__("l32i %1, %2, 0\n\t" @@ -248,8 +236,8 @@ static __inline__ unsigned short int csum_ipv6_magic(struct in6_addr *saddr, * Copy and checksum to user */ #define HAVE_CSUM_COPY_USER -static __inline__ unsigned int csum_and_copy_to_user (const char *src, char *dst, - int len, int sum, int *err_ptr) +static __inline__ __wsum csum_and_copy_to_user(const void *src, void __user *dst, + int len, __wsum sum, int *err_ptr) { if (access_ok(VERIFY_WRITE, dst, len)) return csum_partial_copy_generic(src, dst, len, sum, NULL, err_ptr); @@ -257,6 +245,6 @@ static __inline__ unsigned int csum_and_copy_to_user (const char *src, char *dst if (len) *err_ptr = -EFAULT; - return -1; /* invalid checksum */ + return (__force __wsum)-1; /* invalid checksum */ } #endif -- cgit v1.2.3-70-g09d2 From 56649d5d3c4cb0fe6dd34808ca9f9208d84130ab Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:23:59 -0800 Subject: [NET]: Generic checksum annotations and cleanups. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/checksum.h | 42 ++++++++++++++----------- include/net/ip6_checksum.h | 76 +++++++++++++++++++++++----------------------- 2 files changed, 63 insertions(+), 55 deletions(-) (limited to 'include') diff --git a/include/net/checksum.h b/include/net/checksum.h index e3ea7cc2c72..2b3c8dc6c11 100644 --- a/include/net/checksum.h +++ b/include/net/checksum.h @@ -27,8 +27,8 @@ #ifndef _HAVE_ARCH_COPY_AND_CSUM_FROM_USER static inline -unsigned int csum_and_copy_from_user (const unsigned char __user *src, unsigned char *dst, - int len, int sum, int *err_ptr) +__wsum csum_and_copy_from_user (const void __user *src, void *dst, + int len, __wsum sum, int *err_ptr) { if (access_ok(VERIFY_READ, src, len)) return csum_partial_copy_from_user(src, dst, len, sum, err_ptr); @@ -41,8 +41,8 @@ unsigned int csum_and_copy_from_user (const unsigned char __user *src, unsigned #endif #ifndef HAVE_CSUM_COPY_USER -static __inline__ unsigned int csum_and_copy_to_user -(const unsigned char *src, unsigned char __user *dst, int len, unsigned int sum, int *err_ptr) +static __inline__ __wsum csum_and_copy_to_user +(const void *src, void __user *dst, int len, __wsum sum, int *err_ptr) { sum = csum_partial(src, len, sum); @@ -53,35 +53,43 @@ static __inline__ unsigned int csum_and_copy_to_user if (len) *err_ptr = -EFAULT; - return -1; /* invalid checksum */ + return (__force __wsum)-1; /* invalid checksum */ } #endif -static inline unsigned int csum_add(unsigned int csum, unsigned int addend) +static inline __wsum csum_add(__wsum csum, __wsum addend) { - csum += addend; - return csum + (csum < addend); + u32 res = (__force u32)csum; + res += (__force u32)addend; + return (__force __wsum)(res + (res < (__force u32)addend)); } -static inline unsigned int csum_sub(unsigned int csum, unsigned int addend) +static inline __wsum csum_sub(__wsum csum, __wsum addend) { return csum_add(csum, ~addend); } -static inline unsigned int -csum_block_add(unsigned int csum, unsigned int csum2, int offset) +static inline __wsum +csum_block_add(__wsum csum, __wsum csum2, int offset) { + u32 sum = (__force u32)csum2; if (offset&1) - csum2 = ((csum2&0xFF00FF)<<8)+((csum2>>8)&0xFF00FF); - return csum_add(csum, csum2); + sum = ((sum&0xFF00FF)<<8)+((sum>>8)&0xFF00FF); + return csum_add(csum, (__force __wsum)sum); } -static inline unsigned int -csum_block_sub(unsigned int csum, unsigned int csum2, int offset) +static inline __wsum +csum_block_sub(__wsum csum, __wsum csum2, int offset) { + u32 sum = (__force u32)csum2; if (offset&1) - csum2 = ((csum2&0xFF00FF)<<8)+((csum2>>8)&0xFF00FF); - return csum_sub(csum, csum2); + sum = ((sum&0xFF00FF)<<8)+((sum>>8)&0xFF00FF); + return csum_sub(csum, (__force __wsum)sum); +} + +static inline __wsum csum_unfold(__sum16 n) +{ + return (__force __wsum)n; } #endif diff --git a/include/net/ip6_checksum.h b/include/net/ip6_checksum.h index 3dfc885bdf2..68e2b32cf1d 100644 --- a/include/net/ip6_checksum.h +++ b/include/net/ip6_checksum.h @@ -34,60 +34,60 @@ #ifndef _HAVE_ARCH_IPV6_CSUM -static __inline__ unsigned short int csum_ipv6_magic(struct in6_addr *saddr, - struct in6_addr *daddr, - __u16 len, - unsigned short proto, - unsigned int csum) +static __inline__ __sum16 csum_ipv6_magic(const struct in6_addr *saddr, + const struct in6_addr *daddr, + __u32 len, unsigned short proto, + __wsum csum) { int carry; __u32 ulen; __u32 uproto; + __u32 sum = (__force u32)csum; - csum += saddr->s6_addr32[0]; - carry = (csum < saddr->s6_addr32[0]); - csum += carry; + sum += (__force u32)saddr->s6_addr32[0]; + carry = (sum < (__force u32)saddr->s6_addr32[0]); + sum += carry; - csum += saddr->s6_addr32[1]; - carry = (csum < saddr->s6_addr32[1]); - csum += carry; + sum += (__force u32)saddr->s6_addr32[1]; + carry = (sum < (__force u32)saddr->s6_addr32[1]); + sum += carry; - csum += saddr->s6_addr32[2]; - carry = (csum < saddr->s6_addr32[2]); - csum += carry; + sum += (__force u32)saddr->s6_addr32[2]; + carry = (sum < (__force u32)saddr->s6_addr32[2]); + sum += carry; - csum += saddr->s6_addr32[3]; - carry = (csum < saddr->s6_addr32[3]); - csum += carry; + sum += (__force u32)saddr->s6_addr32[3]; + carry = (sum < (__force u32)saddr->s6_addr32[3]); + sum += carry; - csum += daddr->s6_addr32[0]; - carry = (csum < daddr->s6_addr32[0]); - csum += carry; + sum += (__force u32)daddr->s6_addr32[0]; + carry = (sum < (__force u32)daddr->s6_addr32[0]); + sum += carry; - csum += daddr->s6_addr32[1]; - carry = (csum < daddr->s6_addr32[1]); - csum += carry; + sum += (__force u32)daddr->s6_addr32[1]; + carry = (sum < (__force u32)daddr->s6_addr32[1]); + sum += carry; - csum += daddr->s6_addr32[2]; - carry = (csum < daddr->s6_addr32[2]); - csum += carry; + sum += (__force u32)daddr->s6_addr32[2]; + carry = (sum < (__force u32)daddr->s6_addr32[2]); + sum += carry; - csum += daddr->s6_addr32[3]; - carry = (csum < daddr->s6_addr32[3]); - csum += carry; + sum += (__force u32)daddr->s6_addr32[3]; + carry = (sum < (__force u32)daddr->s6_addr32[3]); + sum += carry; - ulen = htonl((__u32) len); - csum += ulen; - carry = (csum < ulen); - csum += carry; + ulen = (__force u32)htonl((__u32) len); + sum += ulen; + carry = (sum < ulen); + sum += carry; - uproto = htonl(proto); - csum += uproto; - carry = (csum < uproto); - csum += carry; + uproto = (__force u32)htonl(proto); + sum += uproto; + carry = (sum < uproto); + sum += carry; - return csum_fold(csum); + return csum_fold((__force __wsum)csum); } #endif -- cgit v1.2.3-70-g09d2 From 9981a0e36a572e9fcf84bfab915fdc93bed0e3c9 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:24:30 -0800 Subject: [NET]: Annotate checksums in on-the-wire packets. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/dccp.h | 2 +- include/linux/icmp.h | 2 +- include/linux/icmpv6.h | 2 +- include/linux/igmp.h | 2 +- include/linux/ip.h | 2 +- include/linux/tcp.h | 2 +- include/linux/udp.h | 2 +- 7 files changed, 7 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/include/linux/dccp.h b/include/linux/dccp.h index d308f1228b6..72cc355d7a0 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -30,7 +30,7 @@ struct dccp_hdr { #else #error "Adjust your defines" #endif - __u16 dccph_checksum; + __sum16 dccph_checksum; #if defined(__LITTLE_ENDIAN_BITFIELD) __u8 dccph_x:1, dccph_type:4, diff --git a/include/linux/icmp.h b/include/linux/icmp.h index 878cfe4e587..24da4fbc1a2 100644 --- a/include/linux/icmp.h +++ b/include/linux/icmp.h @@ -68,7 +68,7 @@ struct icmphdr { __u8 type; __u8 code; - __be16 checksum; + __sum16 checksum; union { struct { __be16 id; diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h index dc79396aac2..68d3526c3a0 100644 --- a/include/linux/icmpv6.h +++ b/include/linux/icmpv6.h @@ -7,7 +7,7 @@ struct icmp6hdr { __u8 icmp6_type; __u8 icmp6_code; - __be16 icmp6_cksum; + __sum16 icmp6_cksum; union { diff --git a/include/linux/igmp.h b/include/linux/igmp.h index 21dd5690527..6e7ea2f0a57 100644 --- a/include/linux/igmp.h +++ b/include/linux/igmp.h @@ -30,7 +30,7 @@ struct igmphdr { __u8 type; __u8 code; /* For newer IGMP */ - __be16 csum; + __sum16 csum; __be32 group; }; diff --git a/include/linux/ip.h b/include/linux/ip.h index ecee9bb27d0..1d36b971a8b 100644 --- a/include/linux/ip.h +++ b/include/linux/ip.h @@ -98,7 +98,7 @@ struct iphdr { __be16 frag_off; __u8 ttl; __u8 protocol; - __be16 check; + __sum16 check; __be32 saddr; __be32 daddr; /*The options start here. */ diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 0aecfc95559..dd61b172ac6 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -52,7 +52,7 @@ struct tcphdr { #error "Adjust your defines" #endif __be16 window; - __be16 check; + __sum16 check; __be16 urg_ptr; }; diff --git a/include/linux/udp.h b/include/linux/udp.h index 564f3b05010..7e08c07efe0 100644 --- a/include/linux/udp.h +++ b/include/linux/udp.h @@ -23,7 +23,7 @@ struct udphdr { __be16 source; __be16 dest; __be16 len; - __be16 check; + __sum16 check; }; /* UDP socket options */ -- cgit v1.2.3-70-g09d2 From d6f5493c1a42b1203e5b0cb0507afd00bb2927bf Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:26:08 -0800 Subject: [NET]: Annotate callers of csum_tcpudp_nofold() in net/* Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/ip.h | 2 +- net/core/netpoll.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/net/ip.h b/include/net/ip.h index 412e8114667..eafbcaf728f 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -135,7 +135,7 @@ static inline void ip_tr_mc_map(__be32 addr, char *buf) struct ip_reply_arg { struct kvec iov[1]; - u32 csum; + __wsum csum; int csumoffset; /* u16 offset of csum in iov[0].iov_base */ /* -1 if not needed */ }; diff --git a/net/core/netpoll.c b/net/core/netpoll.c index edd3246873e..26ee1791aa0 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -78,9 +78,9 @@ static void queue_process(void *p) } static int checksum_udp(struct sk_buff *skb, struct udphdr *uh, - unsigned short ulen, u32 saddr, u32 daddr) + unsigned short ulen, __be32 saddr, __be32 daddr) { - unsigned int psum; + __wsum psum; if (uh->check == 0 || skb->ip_summed == CHECKSUM_UNNECESSARY) return 0; -- cgit v1.2.3-70-g09d2 From 6b11687ef003ed595033da89643c8995676f929d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:28:24 -0800 Subject: [NET]: Annotate csum_tcpudp_magic() callers in net/* Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/tcp.h | 6 +++--- net/ipv4/udp.c | 3 ++- 2 files changed, 5 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/net/tcp.h b/include/net/tcp.h index 363960872de..826aaecdb99 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -807,9 +807,9 @@ static inline void tcp_update_wl(struct tcp_sock *tp, u32 ack, u32 seq) /* * Calculate(/check) TCP checksum */ -static inline u16 tcp_v4_check(struct tcphdr *th, int len, - unsigned long saddr, unsigned long daddr, - unsigned long base) +static inline __sum16 tcp_v4_check(struct tcphdr *th, int len, + __be32 saddr, __be32 daddr, + __wsum base) { return csum_tcpudp_magic(saddr,daddr,len,IPPROTO_TCP,base); } diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 98ba7509617..f9c4ed7207b 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -414,8 +414,9 @@ static void udp_flush_pending_frames(struct sock *sk) static void udp4_hwcsum_outgoing(struct sock *sk, struct sk_buff *skb, __be32 src, __be32 dst, int len ) { - unsigned int csum = 0, offset; + unsigned int offset; struct udphdr *uh = skb->h.uh; + __wsum csum = 0; if (skb_queue_len(&sk->sk_write_queue) == 1) { /* -- cgit v1.2.3-70-g09d2 From 868c86bcb5bdea7ed8d45979b17bb919af9254db Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:35:48 -0800 Subject: [NET]: annotate csum_ipv6_magic() callers in net/* Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/udp.h | 4 ++-- include/net/udplite.h | 4 ++-- net/dccp/ipv6.c | 2 +- net/ipv6/icmp.c | 14 +++++++------- net/ipv6/mcast.c | 4 ++-- net/ipv6/netfilter.c | 5 +++-- net/ipv6/raw.c | 11 +++++------ net/ipv6/tcp_ipv6.c | 8 ++++---- net/ipv6/udp.c | 9 +++++---- 9 files changed, 31 insertions(+), 30 deletions(-) (limited to 'include') diff --git a/include/net/udp.h b/include/net/udp.h index 4f0626735ed..39e825a6909 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -89,9 +89,9 @@ static __inline__ int udp_lib_checksum_complete(struct sk_buff *skb) * @skb: sk_buff containing the filled-in UDP header * (checksum field must be zeroed out) */ -static inline u32 udp_csum_outgoing(struct sock *sk, struct sk_buff *skb) +static inline __wsum udp_csum_outgoing(struct sock *sk, struct sk_buff *skb) { - u32 csum = csum_partial(skb->h.raw, sizeof(struct udphdr), 0); + __wsum csum = csum_partial(skb->h.raw, sizeof(struct udphdr), 0); skb_queue_walk(&sk->sk_write_queue, skb) { csum = csum_add(csum, skb->csum); diff --git a/include/net/udplite.h b/include/net/udplite.h index 1473b3e4904..406eb755b34 100644 --- a/include/net/udplite.h +++ b/include/net/udplite.h @@ -124,10 +124,10 @@ static inline int udplite_sender_cscov(struct udp_sock *up, struct udphdr *uh) return cscov; } -static inline u32 udplite_csum_outgoing(struct sock *sk, struct sk_buff *skb) +static inline __wsum udplite_csum_outgoing(struct sock *sk, struct sk_buff *skb) { - u32 csum = 0; int off, len, cscov = udplite_sender_cscov(udp_sk(sk), skb->h.uh); + __wsum csum = 0; skb->ip_summed = CHECKSUM_NONE; /* no HW support for checksumming */ diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index e0a0607862e..f28e406a4a1 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -59,7 +59,7 @@ static void dccp_v6_hash(struct sock *sk) } /* add pseudo-header to DCCP checksum stored in skb->csum */ -static inline u16 dccp_v6_csum_finish(struct sk_buff *skb, +static inline __sum16 dccp_v6_csum_finish(struct sk_buff *skb, struct in6_addr *saddr, struct in6_addr *daddr) { diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index bd51847acd5..4ab8acf37b5 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -234,7 +234,7 @@ static int icmpv6_push_pending_frames(struct sock *sk, struct flowi *fl, struct len, fl->proto, skb->csum); } else { - u32 tmp_csum = 0; + __wsum tmp_csum = 0; skb_queue_walk(&sk->sk_write_queue, skb) { tmp_csum = csum_add(tmp_csum, skb->csum); @@ -242,10 +242,10 @@ static int icmpv6_push_pending_frames(struct sock *sk, struct flowi *fl, struct tmp_csum = csum_partial((char *)icmp6h, sizeof(struct icmp6hdr), tmp_csum); - tmp_csum = csum_ipv6_magic(&fl->fl6_src, - &fl->fl6_dst, - len, fl->proto, tmp_csum); - icmp6h->icmp6_cksum = tmp_csum; + icmp6h->icmp6_cksum = csum_ipv6_magic(&fl->fl6_src, + &fl->fl6_dst, + len, fl->proto, + tmp_csum); } ip6_push_pending_frames(sk); out: @@ -636,8 +636,8 @@ static int icmpv6_rcv(struct sk_buff **pskb) break; /* fall through */ case CHECKSUM_NONE: - skb->csum = ~csum_ipv6_magic(saddr, daddr, skb->len, - IPPROTO_ICMPV6, 0); + skb->csum = ~csum_unfold(csum_ipv6_magic(saddr, daddr, skb->len, + IPPROTO_ICMPV6, 0)); if (__skb_checksum_complete(skb)) { LIMIT_NETDEBUG(KERN_DEBUG "ICMPv6 checksum failed [" NIP6_FMT " > " NIP6_FMT "]\n", NIP6(*saddr), NIP6(*daddr)); diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index c006d02be8b..a1c231a04ac 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -91,7 +91,7 @@ struct mld2_grec { struct mld2_report { __u8 type; __u8 resv1; - __u16 csum; + __sum16 csum; __be16 resv2; __be16 ngrec; struct mld2_grec grec[0]; @@ -100,7 +100,7 @@ struct mld2_report { struct mld2_query { __u8 type; __u8 code; - __u16 csum; + __sum16 csum; __be16 mrc; __be16 resv1; struct in6_addr mca; diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c index 646a47456fd..8d1b542806c 100644 --- a/net/ipv6/netfilter.c +++ b/net/ipv6/netfilter.c @@ -100,12 +100,13 @@ unsigned int nf_ip6_checksum(struct sk_buff *skb, unsigned int hook, } /* fall through */ case CHECKSUM_NONE: - skb->csum = ~csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr, + skb->csum = ~csum_unfold( + csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr, skb->len - dataoff, protocol, csum_sub(0, skb_checksum(skb, 0, - dataoff, 0))); + dataoff, 0)))); csum = __skb_checksum_complete(skb); } return csum; diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index b03040a2081..cee5db27e8b 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -370,9 +370,9 @@ int rawv6_rcv(struct sock *sk, struct sk_buff *skb) skb->ip_summed = CHECKSUM_UNNECESSARY; } if (skb->ip_summed != CHECKSUM_UNNECESSARY) - skb->csum = ~csum_ipv6_magic(&skb->nh.ipv6h->saddr, + skb->csum = ~csum_unfold(csum_ipv6_magic(&skb->nh.ipv6h->saddr, &skb->nh.ipv6h->daddr, - skb->len, inet->num, 0); + skb->len, inet->num, 0)); if (inet->hdrincl) { if (skb_checksum_complete(skb)) { @@ -479,8 +479,8 @@ static int rawv6_push_pending_frames(struct sock *sk, struct flowi *fl, int offset; int len; int total_len; - u32 tmp_csum; - u16 csum; + __wsum tmp_csum; + __sum16 csum; if (!rp->checksum) goto send; @@ -532,14 +532,13 @@ static int rawv6_push_pending_frames(struct sock *sk, struct flowi *fl, if (unlikely(csum)) tmp_csum = csum_sub(tmp_csum, csum); - tmp_csum = csum_ipv6_magic(&fl->fl6_src, + csum = csum_ipv6_magic(&fl->fl6_src, &fl->fl6_dst, total_len, fl->proto, tmp_csum); if (tmp_csum == 0 && fl->proto == IPPROTO_UDP) tmp_csum = -1; - csum = tmp_csum; if (skb_store_bits(skb, offset, &csum, 2)) BUG(); diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 53f270995d8..394bc54c5c2 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -105,10 +105,10 @@ static void tcp_v6_hash(struct sock *sk) } } -static __inline__ u16 tcp_v6_check(struct tcphdr *th, int len, +static __inline__ __sum16 tcp_v6_check(struct tcphdr *th, int len, struct in6_addr *saddr, struct in6_addr *daddr, - unsigned long base) + __wsum base) { return csum_ipv6_magic(saddr, daddr, len, IPPROTO_TCP, base); } @@ -1537,8 +1537,8 @@ static int tcp_v6_checksum_init(struct sk_buff *skb) } } - skb->csum = ~tcp_v6_check(skb->h.th,skb->len,&skb->nh.ipv6h->saddr, - &skb->nh.ipv6h->daddr, 0); + skb->csum = ~csum_unfold(tcp_v6_check(skb->h.th,skb->len,&skb->nh.ipv6h->saddr, + &skb->nh.ipv6h->daddr, 0)); if (skb->len <= 76) { return __skb_checksum_complete(skb); diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index e6e1f85f1bb..0d22008d522 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -383,9 +383,10 @@ static inline int udp6_csum_init(struct sk_buff *skb, struct udphdr *uh) skb->ip_summed = CHECKSUM_UNNECESSARY; if (skb->ip_summed != CHECKSUM_UNNECESSARY) - skb->csum = ~csum_ipv6_magic(&skb->nh.ipv6h->saddr, - &skb->nh.ipv6h->daddr, - skb->len, IPPROTO_UDP, 0); + skb->csum = ~csum_unfold(csum_ipv6_magic(&skb->nh.ipv6h->saddr, + &skb->nh.ipv6h->daddr, + ulen, IPPROTO_UDP, + 0)); return (UDP_SKB_CB(skb)->partial_cov = 0); } @@ -511,7 +512,7 @@ static int udp_v6_push_pending_frames(struct sock *sk, struct udp_sock *up) struct inet_sock *inet = inet_sk(sk); struct flowi *fl = &inet->cork.fl; int err = 0; - u32 csum = 0; + __wsum csum = 0; /* Grab the skbuff where UDP header space exists. */ if ((skb = skb_peek(&sk->sk_write_queue)) == NULL) -- cgit v1.2.3-70-g09d2 From 44bb93633f57a55979f3c2589b10fd6a2bfc7c08 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:36:14 -0800 Subject: [NET]: Annotate csum_partial() callers in net/* Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/socket.h | 2 +- include/linux/sunrpc/xdr.h | 2 +- net/core/iovec.c | 4 ++-- net/core/skbuff.c | 2 +- net/ipv4/ip_output.c | 8 ++++---- net/unix/af_unix.c | 3 ++- 6 files changed, 11 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/linux/socket.h b/include/linux/socket.h index 592b6667982..92cd38efad7 100644 --- a/include/linux/socket.h +++ b/include/linux/socket.h @@ -293,7 +293,7 @@ extern int memcpy_fromiovecend(unsigned char *kdata, struct iovec *iov, extern int csum_partial_copy_fromiovecend(unsigned char *kdata, struct iovec *iov, int offset, - unsigned int len, int *csump); + unsigned int len, __wsum *csump); extern int verify_iovec(struct msghdr *m, struct iovec *iov, char *address, int mode); extern int memcpy_toiovec(struct iovec *v, unsigned char *kdata, int len); diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h index ac69e551160..9a527c36439 100644 --- a/include/linux/sunrpc/xdr.h +++ b/include/linux/sunrpc/xdr.h @@ -151,7 +151,7 @@ typedef struct { struct sk_buff *skb; unsigned int offset; size_t count; - unsigned int csum; + __wsum csum; } skb_reader_t; typedef size_t (*skb_read_actor_t)(skb_reader_t *desc, void *to, size_t len); diff --git a/net/core/iovec.c b/net/core/iovec.c index 65e4b56fbc7..04b249c40b5 100644 --- a/net/core/iovec.c +++ b/net/core/iovec.c @@ -158,9 +158,9 @@ int memcpy_fromiovecend(unsigned char *kdata, struct iovec *iov, int offset, * call to this function will be unaligned also. */ int csum_partial_copy_fromiovecend(unsigned char *kdata, struct iovec *iov, - int offset, unsigned int len, int *csump) + int offset, unsigned int len, __wsum *csump) { - int csum = *csump; + __wsum csum = *csump; int partial_cnt = 0, err = 0; /* Skip over the finished iovecs */ diff --git a/net/core/skbuff.c b/net/core/skbuff.c index dfa02cc8d68..c0e3427057f 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1265,7 +1265,7 @@ unsigned int skb_checksum(const struct sk_buff *skb, int offset, end = start + skb_shinfo(skb)->frags[i].size; if ((copy = end - offset) > 0) { - unsigned int csum2; + __wsum csum2; u8 *vaddr; skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 5f3e35c0363..f9194f7e39d 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -682,7 +682,7 @@ ip_generic_getfrag(void *from, char *to, int offset, int len, int odd, struct sk if (memcpy_fromiovecend(to, iov, offset, len) < 0) return -EFAULT; } else { - unsigned int csum = 0; + __wsum csum = 0; if (csum_partial_copy_fromiovecend(to, iov, offset, len, &csum) < 0) return -EFAULT; skb->csum = csum_block_add(skb->csum, csum, odd); @@ -690,11 +690,11 @@ ip_generic_getfrag(void *from, char *to, int offset, int len, int odd, struct sk return 0; } -static inline unsigned int +static inline __wsum csum_page(struct page *page, int offset, int copy) { char *kaddr; - unsigned int csum; + __wsum csum; kaddr = kmap(page); csum = csum_partial(kaddr + offset, copy, 0); kunmap(page); @@ -1166,7 +1166,7 @@ ssize_t ip_append_page(struct sock *sk, struct page *page, } if (skb->ip_summed == CHECKSUM_NONE) { - unsigned int csum; + __wsum csum; csum = csum_page(page, offset, len); skb->csum = csum_block_add(skb->csum, csum, skb->len); } diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index b43a27828df..2f208c7f4d4 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -151,8 +151,9 @@ static inline void unix_set_secdata(struct scm_cookie *scm, struct sk_buff *skb) * each socket state is protected by separate rwlock. */ -static inline unsigned unix_hash_fold(unsigned hash) +static inline unsigned unix_hash_fold(__wsum n) { + unsigned hash = (__force unsigned)n; hash ^= hash>>16; hash ^= hash>>8; return hash&(UNIX_HASH_SIZE-1); -- cgit v1.2.3-70-g09d2 From 5084205faf45384fff25c4cf77dd5c96279283ad Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:36:34 -0800 Subject: [NET]: Annotate callers of csum_partial_copy_...() and csum_and_copy...() in net/* Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/skbuff.h | 3 +-- include/net/sock.h | 2 +- net/core/datagram.c | 6 +++--- net/core/skbuff.c | 2 +- net/ipv4/ip_output.c | 2 +- 5 files changed, 7 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index e3ae544b395..64fa7f4c702 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1199,8 +1199,7 @@ static inline int skb_add_data(struct sk_buff *skb, if (skb->ip_summed == CHECKSUM_NONE) { int err = 0; - unsigned int csum = csum_and_copy_from_user(from, - skb_put(skb, copy), + __wsum csum = csum_and_copy_from_user(from, skb_put(skb, copy), copy, 0, &err); if (!err) { skb->csum = csum_block_add(skb->csum, csum, off); diff --git a/include/net/sock.h b/include/net/sock.h index 35ffbdd35d3..dc4b92b8abe 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1088,7 +1088,7 @@ static inline int skb_copy_to_page(struct sock *sk, char __user *from, { if (skb->ip_summed == CHECKSUM_NONE) { int err = 0; - unsigned int csum = csum_and_copy_from_user(from, + __wsum csum = csum_and_copy_from_user(from, page_address(page) + off, copy, 0, &err); if (err) diff --git a/net/core/datagram.c b/net/core/datagram.c index e5a05a046fe..0d9c9bac400 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -321,7 +321,7 @@ fault: static int skb_copy_and_csum_datagram(const struct sk_buff *skb, int offset, u8 __user *to, int len, - unsigned int *csump) + __wsum *csump) { int start = skb_headlen(skb); int pos = 0; @@ -350,7 +350,7 @@ static int skb_copy_and_csum_datagram(const struct sk_buff *skb, int offset, end = start + skb_shinfo(skb)->frags[i].size; if ((copy = end - offset) > 0) { - unsigned int csum2; + __wsum csum2; int err = 0; u8 *vaddr; skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; @@ -386,7 +386,7 @@ static int skb_copy_and_csum_datagram(const struct sk_buff *skb, int offset, end = start + list->len; if ((copy = end - offset) > 0) { - unsigned int csum2 = 0; + __wsum csum2 = 0; if (copy > len) copy = len; if (skb_copy_and_csum_datagram(list, diff --git a/net/core/skbuff.c b/net/core/skbuff.c index c0e3427057f..da6683f4b31 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1342,7 +1342,7 @@ unsigned int skb_copy_and_csum_bits(const struct sk_buff *skb, int offset, end = start + skb_shinfo(skb)->frags[i].size; if ((copy = end - offset) > 0) { - unsigned int csum2; + __wsum csum2; u8 *vaddr; skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index f9194f7e39d..1da3d32f828 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -1314,7 +1314,7 @@ void ip_flush_pending_frames(struct sock *sk) static int ip_reply_glue_bits(void *dptr, char *to, int offset, int len, int odd, struct sk_buff *skb) { - unsigned int csum; + __wsum csum; csum = csum_partial_copy_nocheck(dptr+offset, to, len, 0); skb->csum = csum_block_add(skb->csum, csum, odd); -- cgit v1.2.3-70-g09d2 From 2bbbc86890ac4c911c5057f69af93853e52a42a8 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:37:14 -0800 Subject: [NET]: Annotate skb_checksum() and callers. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/skbuff.h | 4 ++-- net/core/skbuff.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 64fa7f4c702..784129fb61d 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1334,8 +1334,8 @@ extern int skb_copy_and_csum_datagram_iovec(struct sk_buff *skb, extern void skb_free_datagram(struct sock *sk, struct sk_buff *skb); extern void skb_kill_datagram(struct sock *sk, struct sk_buff *skb, unsigned int flags); -extern unsigned int skb_checksum(const struct sk_buff *skb, int offset, - int len, unsigned int csum); +extern __wsum skb_checksum(const struct sk_buff *skb, int offset, + int len, __wsum csum); extern int skb_copy_bits(const struct sk_buff *skb, int offset, void *to, int len); extern int skb_store_bits(const struct sk_buff *skb, int offset, diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 6e8c15b39eb..be9b541f536 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1240,8 +1240,8 @@ EXPORT_SYMBOL(skb_store_bits); /* Checksum skb data. */ -unsigned int skb_checksum(const struct sk_buff *skb, int offset, - int len, unsigned int csum) +__wsum skb_checksum(const struct sk_buff *skb, int offset, + int len, __wsum csum) { int start = skb_headlen(skb); int i, copy = start - offset; -- cgit v1.2.3-70-g09d2 From 81d77662763ae527ba3a9b9275467901aaab7dfd Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:37:33 -0800 Subject: [NET]: Annotate skb_copy_and_csum_bits() and callers. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/skbuff.h | 4 ++-- net/core/skbuff.c | 6 +++--- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 784129fb61d..874ca029fbb 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1340,9 +1340,9 @@ extern int skb_copy_bits(const struct sk_buff *skb, int offset, void *to, int len); extern int skb_store_bits(const struct sk_buff *skb, int offset, void *from, int len); -extern unsigned int skb_copy_and_csum_bits(const struct sk_buff *skb, +extern __wsum skb_copy_and_csum_bits(const struct sk_buff *skb, int offset, u8 *to, int len, - unsigned int csum); + __wsum csum); extern void skb_copy_and_csum_dev(const struct sk_buff *skb, u8 *to); extern void skb_split(struct sk_buff *skb, struct sk_buff *skb1, const u32 len); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index be9b541f536..07c25d60192 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1315,8 +1315,8 @@ __wsum skb_checksum(const struct sk_buff *skb, int offset, /* Both of above in one bottle. */ -unsigned int skb_copy_and_csum_bits(const struct sk_buff *skb, int offset, - u8 *to, int len, unsigned int csum) +__wsum skb_copy_and_csum_bits(const struct sk_buff *skb, int offset, + u8 *to, int len, __wsum csum) { int start = skb_headlen(skb); int i, copy = start - offset; @@ -1368,7 +1368,7 @@ unsigned int skb_copy_and_csum_bits(const struct sk_buff *skb, int offset, struct sk_buff *list = skb_shinfo(skb)->frag_list; for (; list; list = list->next) { - unsigned int csum2; + __wsum csum2; int end; BUG_TRAP(start <= offset + len); -- cgit v1.2.3-70-g09d2 From b1550f221255f7dd06048d832bf0f3d166840e1c Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:37:50 -0800 Subject: [NET]: Annotate ip_vs_checksum_complete() and callers. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/ip_vs.h | 2 +- net/ipv4/ipvs/ip_vs_core.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h index 903108e583f..fedea8db156 100644 --- a/include/net/ip_vs.h +++ b/include/net/ip_vs.h @@ -988,7 +988,7 @@ extern int ip_vs_make_skb_writable(struct sk_buff **pskb, int len); extern void ip_vs_nat_icmp(struct sk_buff *skb, struct ip_vs_protocol *pp, struct ip_vs_conn *cp, int dir); -extern u16 ip_vs_checksum_complete(struct sk_buff *skb, int offset); +extern __sum16 ip_vs_checksum_complete(struct sk_buff *skb, int offset); static inline u16 ip_vs_check_diff(u32 old, u32 new, u16 oldsum) { diff --git a/net/ipv4/ipvs/ip_vs_core.c b/net/ipv4/ipvs/ip_vs_core.c index fac2dffd667..34257520a3a 100644 --- a/net/ipv4/ipvs/ip_vs_core.c +++ b/net/ipv4/ipvs/ip_vs_core.c @@ -536,7 +536,7 @@ static unsigned int ip_vs_post_routing(unsigned int hooknum, return NF_STOP; } -u16 ip_vs_checksum_complete(struct sk_buff *skb, int offset) +__sum16 ip_vs_checksum_complete(struct sk_buff *skb, int offset) { return csum_fold(skb_checksum(skb, offset, skb->len - offset, 0)); } -- cgit v1.2.3-70-g09d2 From b51655b958dfb1176bfcf99466231fdbef8751ff Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:40:42 -0800 Subject: [NET]: Annotate __skb_checksum_complete() and friends. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/netfilter.h | 6 +++--- include/linux/netfilter_ipv4.h | 2 +- include/linux/netfilter_ipv6.h | 2 +- include/linux/skbuff.h | 2 +- include/net/tcp.h | 2 +- include/net/udp.h | 8 ++++---- net/core/datagram.c | 2 +- net/core/netpoll.c | 4 ++-- net/ipv4/netfilter.c | 4 ++-- net/ipv4/tcp_input.c | 4 ++-- net/ipv4/tcp_ipv4.c | 2 +- net/ipv6/netfilter.c | 4 ++-- net/ipv6/tcp_ipv6.c | 2 +- 13 files changed, 22 insertions(+), 22 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index b7e67d1d438..707bb2e53c4 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -290,7 +290,7 @@ extern u_int16_t nf_proto_csum_update(struct sk_buff *skb, struct nf_afinfo { unsigned short family; - unsigned int (*checksum)(struct sk_buff *skb, unsigned int hook, + __sum16 (*checksum)(struct sk_buff *skb, unsigned int hook, unsigned int dataoff, u_int8_t protocol); void (*saveroute)(const struct sk_buff *skb, struct nf_info *info); @@ -305,12 +305,12 @@ static inline struct nf_afinfo *nf_get_afinfo(unsigned short family) return rcu_dereference(nf_afinfo[family]); } -static inline unsigned int +static inline __sum16 nf_checksum(struct sk_buff *skb, unsigned int hook, unsigned int dataoff, u_int8_t protocol, unsigned short family) { struct nf_afinfo *afinfo; - unsigned int csum = 0; + __sum16 csum = 0; rcu_read_lock(); afinfo = nf_get_afinfo(family); diff --git a/include/linux/netfilter_ipv4.h b/include/linux/netfilter_ipv4.h index 5b63a231a76..5821eb5a0a3 100644 --- a/include/linux/netfilter_ipv4.h +++ b/include/linux/netfilter_ipv4.h @@ -79,7 +79,7 @@ enum nf_ip_hook_priorities { #ifdef __KERNEL__ extern int ip_route_me_harder(struct sk_buff **pskb, unsigned addr_type); extern int ip_xfrm_me_harder(struct sk_buff **pskb); -extern unsigned int nf_ip_checksum(struct sk_buff *skb, unsigned int hook, +extern __sum16 nf_ip_checksum(struct sk_buff *skb, unsigned int hook, unsigned int dataoff, u_int8_t protocol); #endif /*__KERNEL__*/ diff --git a/include/linux/netfilter_ipv6.h b/include/linux/netfilter_ipv6.h index d97e268cdfe..ab81a6dc94e 100644 --- a/include/linux/netfilter_ipv6.h +++ b/include/linux/netfilter_ipv6.h @@ -74,7 +74,7 @@ enum nf_ip6_hook_priorities { #ifdef CONFIG_NETFILTER extern int ip6_route_me_harder(struct sk_buff *skb); -extern unsigned int nf_ip6_checksum(struct sk_buff *skb, unsigned int hook, +extern __sum16 nf_ip6_checksum(struct sk_buff *skb, unsigned int hook, unsigned int dataoff, u_int8_t protocol); extern int ipv6_netfilter_init(void); diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 874ca029fbb..41753667541 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1398,7 +1398,7 @@ static inline void skb_set_timestamp(struct sk_buff *skb, const struct timeval * extern void __net_timestamp(struct sk_buff *skb); -extern unsigned int __skb_checksum_complete(struct sk_buff *skb); +extern __sum16 __skb_checksum_complete(struct sk_buff *skb); /** * skb_checksum_complete - Calculate checksum of an entire packet diff --git a/include/net/tcp.h b/include/net/tcp.h index 826aaecdb99..aa7989c5379 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -814,7 +814,7 @@ static inline __sum16 tcp_v4_check(struct tcphdr *th, int len, return csum_tcpudp_magic(saddr,daddr,len,IPPROTO_TCP,base); } -static inline int __tcp_checksum_complete(struct sk_buff *skb) +static inline __sum16 __tcp_checksum_complete(struct sk_buff *skb) { return __skb_checksum_complete(skb); } diff --git a/include/net/udp.h b/include/net/udp.h index 39e825a6909..c5ccd9a3387 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -69,15 +69,15 @@ struct sk_buff; /* * Generic checksumming routines for UDP(-Lite) v4 and v6 */ -static inline u16 __udp_lib_checksum_complete(struct sk_buff *skb) +static inline __sum16 __udp_lib_checksum_complete(struct sk_buff *skb) { if (! UDP_SKB_CB(skb)->partial_cov) return __skb_checksum_complete(skb); - return csum_fold(skb_checksum(skb, 0, UDP_SKB_CB(skb)->cscov, - skb->csum)); + return csum_fold(skb_checksum(skb, 0, UDP_SKB_CB(skb)->cscov, + skb->csum)); } -static __inline__ int udp_lib_checksum_complete(struct sk_buff *skb) +static inline __sum16 udp_lib_checksum_complete(struct sk_buff *skb) { return skb->ip_summed != CHECKSUM_UNNECESSARY && __udp_lib_checksum_complete(skb); diff --git a/net/core/datagram.c b/net/core/datagram.c index 0d9c9bac400..797fdd4352c 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -411,7 +411,7 @@ fault: return -EFAULT; } -unsigned int __skb_checksum_complete(struct sk_buff *skb) +__sum16 __skb_checksum_complete(struct sk_buff *skb) { __sum16 sum; diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 26ee1791aa0..8be3681d3d8 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -77,8 +77,8 @@ static void queue_process(void *p) } } -static int checksum_udp(struct sk_buff *skb, struct udphdr *uh, - unsigned short ulen, __be32 saddr, __be32 daddr) +static __sum16 checksum_udp(struct sk_buff *skb, struct udphdr *uh, + unsigned short ulen, __be32 saddr, __be32 daddr) { __wsum psum; diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c index b797a37c01c..a68966059b5 100644 --- a/net/ipv4/netfilter.c +++ b/net/ipv4/netfilter.c @@ -162,11 +162,11 @@ static int nf_ip_reroute(struct sk_buff **pskb, const struct nf_info *info) return 0; } -unsigned int nf_ip_checksum(struct sk_buff *skb, unsigned int hook, +__sum16 nf_ip_checksum(struct sk_buff *skb, unsigned int hook, unsigned int dataoff, u_int8_t protocol) { struct iphdr *iph = skb->nh.iph; - unsigned int csum = 0; + __sum16 csum = 0; switch (skb->ip_summed) { case CHECKSUM_COMPLETE: diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 6ab3423674b..9304034c0c4 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -3790,9 +3790,9 @@ static int tcp_copy_to_iovec(struct sock *sk, struct sk_buff *skb, int hlen) return err; } -static int __tcp_checksum_complete_user(struct sock *sk, struct sk_buff *skb) +static __sum16 __tcp_checksum_complete_user(struct sock *sk, struct sk_buff *skb) { - int result; + __sum16 result; if (sock_owned_by_user(sk)) { local_bh_enable(); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 0ca8dead03b..dadf8027241 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1544,7 +1544,7 @@ static struct sock *tcp_v4_hnd_req(struct sock *sk, struct sk_buff *skb) return sk; } -static int tcp_v4_checksum_init(struct sk_buff *skb) +static __sum16 tcp_v4_checksum_init(struct sk_buff *skb) { if (skb->ip_summed == CHECKSUM_COMPLETE) { if (!tcp_v4_check(skb->h.th, skb->len, skb->nh.iph->saddr, diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c index 8d1b542806c..f6294e5bcb3 100644 --- a/net/ipv6/netfilter.c +++ b/net/ipv6/netfilter.c @@ -80,11 +80,11 @@ static int nf_ip6_reroute(struct sk_buff **pskb, const struct nf_info *info) return 0; } -unsigned int nf_ip6_checksum(struct sk_buff *skb, unsigned int hook, +__sum16 nf_ip6_checksum(struct sk_buff *skb, unsigned int hook, unsigned int dataoff, u_int8_t protocol) { struct ipv6hdr *ip6h = skb->nh.ipv6h; - unsigned int csum = 0; + __sum16 csum = 0; switch (skb->ip_summed) { case CHECKSUM_COMPLETE: diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 394bc54c5c2..147ce499f50 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1527,7 +1527,7 @@ out: return NULL; } -static int tcp_v6_checksum_init(struct sk_buff *skb) +static __sum16 tcp_v6_checksum_init(struct sk_buff *skb) { if (skb->ip_summed == CHECKSUM_COMPLETE) { if (!tcp_v6_check(skb->h.th,skb->len,&skb->nh.ipv6h->saddr, -- cgit v1.2.3-70-g09d2 From f6ab028804bdc580fe0915494dbf31f5ea473ca7 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 16 Nov 2006 02:36:50 -0800 Subject: [NET]: Make mangling a checksum (0 -> 0xffff on the wire) explicit. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/checksum.h | 1 + net/ipv4/ipvs/ip_vs_proto_udp.c | 6 +++--- net/ipv4/netfilter/ip_nat_helper.c | 2 +- net/ipv4/netfilter/ip_nat_proto_udp.c | 2 +- net/ipv4/udp.c | 4 ++-- net/ipv6/raw.c | 4 ++-- net/ipv6/udp.c | 2 +- 7 files changed, 11 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/net/checksum.h b/include/net/checksum.h index 2b3c8dc6c11..124246172a8 100644 --- a/include/net/checksum.h +++ b/include/net/checksum.h @@ -92,4 +92,5 @@ static inline __wsum csum_unfold(__sum16 n) return (__force __wsum)n; } +#define CSUM_MANGLED_0 ((__force __sum16)0xffff) #endif diff --git a/net/ipv4/ipvs/ip_vs_proto_udp.c b/net/ipv4/ipvs/ip_vs_proto_udp.c index 261581e6a81..3647397e416 100644 --- a/net/ipv4/ipvs/ip_vs_proto_udp.c +++ b/net/ipv4/ipvs/ip_vs_proto_udp.c @@ -125,7 +125,7 @@ udp_fast_csum_update(struct udphdr *uhdr, __be32 oldip, __be32 newip, ip_vs_check_diff(oldport ^ htons(0xFFFF), newport, uhdr->check)); if (!uhdr->check) - uhdr->check = -1; + uhdr->check = CSUM_MANGLED_0; } static int @@ -173,7 +173,7 @@ udp_snat_handler(struct sk_buff **pskb, cp->protocol, (*pskb)->csum); if (udph->check == 0) - udph->check = -1; + udph->check = CSUM_MANGLED_0; IP_VS_DBG(11, "O-pkt: %s O-csum=%d (+%zd)\n", pp->name, udph->check, (char*)&(udph->check) - (char*)udph); @@ -228,7 +228,7 @@ udp_dnat_handler(struct sk_buff **pskb, cp->protocol, (*pskb)->csum); if (udph->check == 0) - udph->check = -1; + udph->check = CSUM_MANGLED_0; (*pskb)->ip_summed = CHECKSUM_UNNECESSARY; } return 1; diff --git a/net/ipv4/netfilter/ip_nat_helper.c b/net/ipv4/netfilter/ip_nat_helper.c index 3bf85848055..3e7fd64c216 100644 --- a/net/ipv4/netfilter/ip_nat_helper.c +++ b/net/ipv4/netfilter/ip_nat_helper.c @@ -264,7 +264,7 @@ ip_nat_mangle_udp_packet(struct sk_buff **pskb, csum_partial((char *)udph, datalen, 0)); if (!udph->check) - udph->check = -1; + udph->check = CSUM_MANGLED_0; } else udph->check = nf_proto_csum_update(*pskb, htons(oldlen) ^ htons(0xFFFF), diff --git a/net/ipv4/netfilter/ip_nat_proto_udp.c b/net/ipv4/netfilter/ip_nat_proto_udp.c index 4bbec7730d1..82f8a6ab07e 100644 --- a/net/ipv4/netfilter/ip_nat_proto_udp.c +++ b/net/ipv4/netfilter/ip_nat_proto_udp.c @@ -121,7 +121,7 @@ udp_manip_pkt(struct sk_buff **pskb, *portptr ^ htons(0xFFFF), newport, hdr->check, 0); if (!hdr->check) - hdr->check = -1; + hdr->check = CSUM_MANGLED_0; } *portptr = newport; return 1; diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index f9c4ed7207b..dc19ba1e73b 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -441,7 +441,7 @@ static void udp4_hwcsum_outgoing(struct sock *sk, struct sk_buff *skb, uh->check = csum_tcpudp_magic(src, dst, len, IPPROTO_UDP, csum); if (uh->check == 0) - uh->check = -1; + uh->check = CSUM_MANGLED_0; } } @@ -490,7 +490,7 @@ int udp_push_pending_frames(struct sock *sk, struct udp_sock *up) uh->check = csum_tcpudp_magic(fl->fl4_src, fl->fl4_dst, up->len, sk->sk_protocol, csum ); if (uh->check == 0) - uh->check = -1; + uh->check = CSUM_MANGLED_0; send: err = ip_push_pending_frames(sk); diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 9b21d67ca48..c2e629d6aea 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -536,8 +536,8 @@ static int rawv6_push_pending_frames(struct sock *sk, struct flowi *fl, &fl->fl6_dst, total_len, fl->proto, tmp_csum); - if (tmp_csum == 0 && fl->proto == IPPROTO_UDP) - tmp_csum = -1; + if (csum == 0 && fl->proto == IPPROTO_UDP) + csum = CSUM_MANGLED_0; if (skb_store_bits(skb, offset, &csum, 2)) BUG(); diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 0d22008d522..bb45b9b7cbd 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -536,7 +536,7 @@ static int udp_v6_push_pending_frames(struct sock *sk, struct udp_sock *up) uh->check = csum_ipv6_magic(&fl->fl6_src, &fl->fl6_dst, up->len, fl->proto, csum ); if (uh->check == 0) - uh->check = -1; + uh->check = CSUM_MANGLED_0; err = ip6_push_pending_frames(sk); out: -- cgit v1.2.3-70-g09d2 From 5c78f275e696a25a9919671aa8aa3d0a3d967978 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:42:26 -0800 Subject: [NET]: IP header modifier helpers annotations. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/dsfield.h | 4 ++-- include/net/inet_ecn.h | 6 +++--- include/net/ip.h | 6 +++--- 3 files changed, 8 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/net/dsfield.h b/include/net/dsfield.h index dae8453385b..eb65bf2e250 100644 --- a/include/net/dsfield.h +++ b/include/net/dsfield.h @@ -27,7 +27,7 @@ static inline __u8 ipv6_get_dsfield(struct ipv6hdr *ipv6h) static inline void ipv4_change_dsfield(struct iphdr *iph,__u8 mask, __u8 value) { - __u32 check = ntohs(iph->check); + __u32 check = ntohs((__force __be16)iph->check); __u8 dsfield; dsfield = (iph->tos & mask) | value; @@ -35,7 +35,7 @@ static inline void ipv4_change_dsfield(struct iphdr *iph,__u8 mask, if ((check+1) >> 16) check = (check+1) & 0xffff; check -= dsfield; check += check >> 16; /* adjust carry */ - iph->check = htons(check); + iph->check = (__force __sum16)htons(check); iph->tos = dsfield; } diff --git a/include/net/inet_ecn.h b/include/net/inet_ecn.h index cadc0eab54f..10117c8503e 100644 --- a/include/net/inet_ecn.h +++ b/include/net/inet_ecn.h @@ -53,7 +53,7 @@ static inline __u8 INET_ECN_encapsulate(__u8 outer, __u8 inner) static inline int IP_ECN_set_ce(struct iphdr *iph) { - u32 check = iph->check; + u32 check = (__force u32)iph->check; u32 ecn = (iph->tos + 1) & INET_ECN_MASK; /* @@ -71,9 +71,9 @@ static inline int IP_ECN_set_ce(struct iphdr *iph) * INET_ECN_ECT_1 => check += htons(0xFFFD) * INET_ECN_ECT_0 => check += htons(0xFFFE) */ - check += htons(0xFFFB) + htons(ecn); + check += (__force u16)htons(0xFFFB) + (__force u16)htons(ecn); - iph->check = check + (check>=0xFFFF); + iph->check = (__force __sum16)(check + (check>=0xFFFF)); iph->tos |= INET_ECN_CE; return 1; } diff --git a/include/net/ip.h b/include/net/ip.h index eafbcaf728f..83cb9ac5554 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -192,9 +192,9 @@ extern void ipfrag_init(void); static inline int ip_decrease_ttl(struct iphdr *iph) { - u32 check = iph->check; - check += htons(0x0100); - iph->check = check + (check>=0xFFFF); + u32 check = (__force u32)iph->check; + check += (__force u32)htons(0x0100); + iph->check = (__force __sum16)(check + (check>=0xFFFF)); return --iph->ttl; } -- cgit v1.2.3-70-g09d2 From f9214b2627fb8582af1986afc4aed5942086a535 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 16 Nov 2006 02:41:18 -0800 Subject: [NET]: ipvs checksum annotations. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/ip_vs.h | 14 ++++++++++---- net/ipv4/ipvs/ip_vs_proto_tcp.c | 6 +++--- net/ipv4/ipvs/ip_vs_proto_udp.c | 6 +++--- 3 files changed, 16 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h index fedea8db156..672564e5a81 100644 --- a/include/net/ip_vs.h +++ b/include/net/ip_vs.h @@ -990,12 +990,18 @@ extern void ip_vs_nat_icmp(struct sk_buff *skb, struct ip_vs_protocol *pp, extern __sum16 ip_vs_checksum_complete(struct sk_buff *skb, int offset); -static inline u16 ip_vs_check_diff(u32 old, u32 new, u16 oldsum) +static inline __wsum ip_vs_check_diff4(__be32 old, __be32 new, __wsum oldsum) { - u32 diff[2] = { old, new }; + __be32 diff[2] = { ~old, new }; - return csum_fold(csum_partial((char *) diff, sizeof(diff), - oldsum ^ 0xFFFF)); + return csum_partial((char *) diff, sizeof(diff), oldsum); +} + +static inline __wsum ip_vs_check_diff2(__be16 old, __be16 new, __wsum oldsum) +{ + __be16 diff[2] = { ~old, new }; + + return csum_partial((char *) diff, sizeof(diff), oldsum); } #endif /* __KERNEL__ */ diff --git a/net/ipv4/ipvs/ip_vs_proto_tcp.c b/net/ipv4/ipvs/ip_vs_proto_tcp.c index b60a07910df..16a9ebee2fe 100644 --- a/net/ipv4/ipvs/ip_vs_proto_tcp.c +++ b/net/ipv4/ipvs/ip_vs_proto_tcp.c @@ -116,9 +116,9 @@ tcp_fast_csum_update(struct tcphdr *tcph, __be32 oldip, __be32 newip, __be16 oldport, __be16 newport) { tcph->check = - ip_vs_check_diff(~oldip, newip, - ip_vs_check_diff(oldport ^ htons(0xFFFF), - newport, tcph->check)); + csum_fold(ip_vs_check_diff4(oldip, newip, + ip_vs_check_diff2(oldport, newport, + ~csum_unfold(tcph->check)))); } diff --git a/net/ipv4/ipvs/ip_vs_proto_udp.c b/net/ipv4/ipvs/ip_vs_proto_udp.c index 3647397e416..03f0a414cfa 100644 --- a/net/ipv4/ipvs/ip_vs_proto_udp.c +++ b/net/ipv4/ipvs/ip_vs_proto_udp.c @@ -121,9 +121,9 @@ udp_fast_csum_update(struct udphdr *uhdr, __be32 oldip, __be32 newip, __be16 oldport, __be16 newport) { uhdr->check = - ip_vs_check_diff(~oldip, newip, - ip_vs_check_diff(oldport ^ htons(0xFFFF), - newport, uhdr->check)); + csum_fold(ip_vs_check_diff4(oldip, newip, + ip_vs_check_diff2(oldport, newport, + ~csum_unfold(uhdr->check)))); if (!uhdr->check) uhdr->check = CSUM_MANGLED_0; } -- cgit v1.2.3-70-g09d2 From 43bc0ca7eadc024e9e5b935fa5e0892df4fec9eb Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:43:23 -0800 Subject: [NET]: netfilter checksum annotations Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/netfilter.h | 26 +++++++++++++++---- .../linux/netfilter_ipv4/ip_conntrack_proto_gre.h | 4 +-- net/ipv4/netfilter/ip_nat_core.c | 6 ++--- net/ipv4/netfilter/ip_nat_helper.c | 30 ++++++++-------------- net/ipv4/netfilter/ip_nat_proto_gre.c | 8 +++--- net/ipv4/netfilter/ip_nat_proto_icmp.c | 6 ++--- net/ipv4/netfilter/ip_nat_proto_tcp.c | 5 ++-- net/ipv4/netfilter/ip_nat_proto_udp.c | 7 ++--- net/ipv4/netfilter/ipt_ECN.c | 11 +++----- net/ipv4/netfilter/ipt_TCPMSS.c | 24 ++++++----------- net/ipv4/netfilter/ipt_TOS.c | 5 ++-- net/ipv4/netfilter/ipt_TTL.c | 5 ++-- net/netfilter/core.c | 27 ++++++++----------- 13 files changed, 70 insertions(+), 94 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index 707bb2e53c4..6ab5e2d6133 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -282,11 +282,27 @@ extern void nf_invalidate_cache(int pf); Returns true or false. */ extern int skb_make_writable(struct sk_buff **pskb, unsigned int writable_len); -extern u_int16_t nf_csum_update(u_int32_t oldval, u_int32_t newval, - u_int32_t csum); -extern u_int16_t nf_proto_csum_update(struct sk_buff *skb, - u_int32_t oldval, u_int32_t newval, - u_int16_t csum, int pseudohdr); +static inline void nf_csum_replace4(__sum16 *sum, __be32 from, __be32 to) +{ + __be32 diff[] = { ~from, to }; + + *sum = csum_fold(csum_partial((char *)diff, sizeof(diff), ~csum_unfold(*sum))); +} + +static inline void nf_csum_replace2(__sum16 *sum, __be16 from, __be16 to) +{ + nf_csum_replace4(sum, (__force __be32)from, (__force __be32)to); +} + +extern void nf_proto_csum_replace4(__sum16 *sum, struct sk_buff *skb, + __be32 from, __be32 to, int pseudohdr); + +static inline void nf_proto_csum_replace2(__sum16 *sum, struct sk_buff *skb, + __be16 from, __be16 to, int pseudohdr) +{ + nf_proto_csum_replace4(sum, skb, (__force __be32)from, + (__force __be32)to, pseudohdr); +} struct nf_afinfo { unsigned short family; diff --git a/include/linux/netfilter_ipv4/ip_conntrack_proto_gre.h b/include/linux/netfilter_ipv4/ip_conntrack_proto_gre.h index 1d853aa873e..e371e0fc167 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_proto_gre.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_proto_gre.h @@ -102,11 +102,11 @@ static inline __be32 *gre_key(struct gre_hdr *greh) } /* get pointer ot gre csum, if present */ -static inline u_int16_t *gre_csum(struct gre_hdr *greh) +static inline __sum16 *gre_csum(struct gre_hdr *greh) { if (!greh->csum) return NULL; - return (u_int16_t *) (greh+sizeof(*greh)); + return (__sum16 *) (greh+sizeof(*greh)); } #endif /* __KERNEL__ */ diff --git a/net/ipv4/netfilter/ip_nat_core.c b/net/ipv4/netfilter/ip_nat_core.c index 4b6260a9740..9d1a5175dcd 100644 --- a/net/ipv4/netfilter/ip_nat_core.c +++ b/net/ipv4/netfilter/ip_nat_core.c @@ -362,12 +362,10 @@ manip_pkt(u_int16_t proto, iph = (void *)(*pskb)->data + iphdroff; if (maniptype == IP_NAT_MANIP_SRC) { - iph->check = nf_csum_update(~iph->saddr, target->src.ip, - iph->check); + nf_csum_replace4(&iph->check, iph->saddr, target->src.ip); iph->saddr = target->src.ip; } else { - iph->check = nf_csum_update(~iph->daddr, target->dst.ip, - iph->check); + nf_csum_replace4(&iph->check, iph->daddr, target->dst.ip); iph->daddr = target->dst.ip; } return 1; diff --git a/net/ipv4/netfilter/ip_nat_helper.c b/net/ipv4/netfilter/ip_nat_helper.c index 3e7fd64c216..ee80feb4b2a 100644 --- a/net/ipv4/netfilter/ip_nat_helper.c +++ b/net/ipv4/netfilter/ip_nat_helper.c @@ -188,10 +188,8 @@ ip_nat_mangle_tcp_packet(struct sk_buff **pskb, csum_partial((char *)tcph, datalen, 0)); } else - tcph->check = nf_proto_csum_update(*pskb, - htons(oldlen) ^ htons(0xFFFF), - htons(datalen), - tcph->check, 1); + nf_proto_csum_replace2(&tcph->check, *pskb, + htons(oldlen), htons(datalen), 1); if (rep_len != match_len) { set_bit(IPS_SEQ_ADJUST_BIT, &ct->status); @@ -266,10 +264,8 @@ ip_nat_mangle_udp_packet(struct sk_buff **pskb, if (!udph->check) udph->check = CSUM_MANGLED_0; } else - udph->check = nf_proto_csum_update(*pskb, - htons(oldlen) ^ htons(0xFFFF), - htons(datalen), - udph->check, 1); + nf_proto_csum_replace2(&udph->check, *pskb, + htons(oldlen), htons(datalen), 1); return 1; } EXPORT_SYMBOL(ip_nat_mangle_udp_packet); @@ -307,14 +303,10 @@ sack_adjust(struct sk_buff *skb, ntohl(sack->start_seq), new_start_seq, ntohl(sack->end_seq), new_end_seq); - tcph->check = nf_proto_csum_update(skb, - ~sack->start_seq, - new_start_seq, - tcph->check, 0); - tcph->check = nf_proto_csum_update(skb, - ~sack->end_seq, - new_end_seq, - tcph->check, 0); + nf_proto_csum_replace4(&tcph->check, skb, + sack->start_seq, new_start_seq, 0); + nf_proto_csum_replace4(&tcph->check, skb, + sack->end_seq, new_end_seq, 0); sack->start_seq = new_start_seq; sack->end_seq = new_end_seq; sackoff += sizeof(*sack); @@ -397,10 +389,8 @@ ip_nat_seq_adjust(struct sk_buff **pskb, else newack = htonl(ntohl(tcph->ack_seq) - other_way->offset_before); - tcph->check = nf_proto_csum_update(*pskb, ~tcph->seq, newseq, - tcph->check, 0); - tcph->check = nf_proto_csum_update(*pskb, ~tcph->ack_seq, newack, - tcph->check, 0); + nf_proto_csum_replace4(&tcph->check, *pskb, tcph->seq, newseq, 0); + nf_proto_csum_replace4(&tcph->check, *pskb, tcph->ack_seq, newack, 0); DEBUGP("Adjusting sequence number from %u->%u, ack from %u->%u\n", ntohl(tcph->seq), ntohl(newseq), ntohl(tcph->ack_seq), diff --git a/net/ipv4/netfilter/ip_nat_proto_gre.c b/net/ipv4/netfilter/ip_nat_proto_gre.c index bf91f9312b3..95810202d84 100644 --- a/net/ipv4/netfilter/ip_nat_proto_gre.c +++ b/net/ipv4/netfilter/ip_nat_proto_gre.c @@ -129,11 +129,9 @@ gre_manip_pkt(struct sk_buff **pskb, } if (greh->csum) { /* FIXME: Never tested this code... */ - *(gre_csum(greh)) = - nf_proto_csum_update(*pskb, - ~*(gre_key(greh)), - tuple->dst.u.gre.key, - *(gre_csum(greh)), 0); + nf_proto_csum_replace4(gre_csum(greh), *pskb, + *(gre_key(greh)), + tuple->dst.u.gre.key, 0); } *(gre_key(greh)) = tuple->dst.u.gre.key; break; diff --git a/net/ipv4/netfilter/ip_nat_proto_icmp.c b/net/ipv4/netfilter/ip_nat_proto_icmp.c index 3f6efc13ac7..75266fe3e0f 100644 --- a/net/ipv4/netfilter/ip_nat_proto_icmp.c +++ b/net/ipv4/netfilter/ip_nat_proto_icmp.c @@ -66,10 +66,8 @@ icmp_manip_pkt(struct sk_buff **pskb, return 0; hdr = (struct icmphdr *)((*pskb)->data + hdroff); - hdr->checksum = nf_proto_csum_update(*pskb, - hdr->un.echo.id ^ htons(0xFFFF), - tuple->src.u.icmp.id, - hdr->checksum, 0); + nf_proto_csum_replace2(&hdr->checksum, *pskb, + hdr->un.echo.id, tuple->src.u.icmp.id, 0); hdr->un.echo.id = tuple->src.u.icmp.id; return 1; } diff --git a/net/ipv4/netfilter/ip_nat_proto_tcp.c b/net/ipv4/netfilter/ip_nat_proto_tcp.c index 12deb13b93b..b586d18b3fb 100644 --- a/net/ipv4/netfilter/ip_nat_proto_tcp.c +++ b/net/ipv4/netfilter/ip_nat_proto_tcp.c @@ -129,9 +129,8 @@ tcp_manip_pkt(struct sk_buff **pskb, if (hdrsize < sizeof(*hdr)) return 1; - hdr->check = nf_proto_csum_update(*pskb, ~oldip, newip, hdr->check, 1); - hdr->check = nf_proto_csum_update(*pskb, oldport ^ htons(0xFFFF), newport, - hdr->check, 0); + nf_proto_csum_replace4(&hdr->check, *pskb, oldip, newip, 1); + nf_proto_csum_replace2(&hdr->check, *pskb, oldport, newport, 0); return 1; } diff --git a/net/ipv4/netfilter/ip_nat_proto_udp.c b/net/ipv4/netfilter/ip_nat_proto_udp.c index 82f8a6ab07e..5ced0877b32 100644 --- a/net/ipv4/netfilter/ip_nat_proto_udp.c +++ b/net/ipv4/netfilter/ip_nat_proto_udp.c @@ -115,11 +115,8 @@ udp_manip_pkt(struct sk_buff **pskb, } if (hdr->check || (*pskb)->ip_summed == CHECKSUM_PARTIAL) { - hdr->check = nf_proto_csum_update(*pskb, ~oldip, newip, - hdr->check, 1); - hdr->check = nf_proto_csum_update(*pskb, - *portptr ^ htons(0xFFFF), newport, - hdr->check, 0); + nf_proto_csum_replace4(&hdr->check, *pskb, oldip, newip, 1); + nf_proto_csum_replace2(&hdr->check, *pskb, *portptr, newport, 0); if (!hdr->check) hdr->check = CSUM_MANGLED_0; } diff --git a/net/ipv4/netfilter/ipt_ECN.c b/net/ipv4/netfilter/ipt_ECN.c index 1aa4517fbcd..b55d670a24d 100644 --- a/net/ipv4/netfilter/ipt_ECN.c +++ b/net/ipv4/netfilter/ipt_ECN.c @@ -28,17 +28,16 @@ static inline int set_ect_ip(struct sk_buff **pskb, const struct ipt_ECN_info *einfo) { struct iphdr *iph = (*pskb)->nh.iph; - u_int16_t oldtos; if ((iph->tos & IPT_ECN_IP_MASK) != (einfo->ip_ect & IPT_ECN_IP_MASK)) { + __u8 oldtos; if (!skb_make_writable(pskb, sizeof(struct iphdr))) return 0; iph = (*pskb)->nh.iph; oldtos = iph->tos; iph->tos &= ~IPT_ECN_IP_MASK; iph->tos |= (einfo->ip_ect & IPT_ECN_IP_MASK); - iph->check = nf_csum_update(htons(oldtos) ^ htons(0xFFFF), - htons(iph->tos), iph->check); + nf_csum_replace2(&iph->check, htons(oldtos), htons(iph->tos)); } return 1; } @@ -72,10 +71,8 @@ set_ect_tcp(struct sk_buff **pskb, const struct ipt_ECN_info *einfo) if (einfo->operation & IPT_ECN_OP_SET_CWR) tcph->cwr = einfo->proto.tcp.cwr; - tcph->check = nf_proto_csum_update((*pskb), - oldval ^ htons(0xFFFF), - ((__be16 *)tcph)[6], - tcph->check, 0); + nf_proto_csum_replace2(&tcph->check, *pskb, + oldval, ((__be16 *)tcph)[6], 0); return 1; } diff --git a/net/ipv4/netfilter/ipt_TCPMSS.c b/net/ipv4/netfilter/ipt_TCPMSS.c index 108b6b76311..93eb5c3c188 100644 --- a/net/ipv4/netfilter/ipt_TCPMSS.c +++ b/net/ipv4/netfilter/ipt_TCPMSS.c @@ -97,10 +97,8 @@ ipt_tcpmss_target(struct sk_buff **pskb, opt[i+2] = (newmss & 0xff00) >> 8; opt[i+3] = (newmss & 0x00ff); - tcph->check = nf_proto_csum_update(*pskb, - htons(oldmss)^htons(0xFFFF), - htons(newmss), - tcph->check, 0); + nf_proto_csum_replace2(&tcph->check, *pskb, + htons(oldmss), htons(newmss), 0); return IPT_CONTINUE; } } @@ -126,28 +124,22 @@ ipt_tcpmss_target(struct sk_buff **pskb, opt = (u_int8_t *)tcph + sizeof(struct tcphdr); memmove(opt + TCPOLEN_MSS, opt, tcplen - sizeof(struct tcphdr)); - tcph->check = nf_proto_csum_update(*pskb, - htons(tcplen) ^ htons(0xFFFF), - htons(tcplen + TCPOLEN_MSS), - tcph->check, 1); + nf_proto_csum_replace2(&tcph->check, *pskb, + htons(tcplen), htons(tcplen + TCPOLEN_MSS), 1); opt[0] = TCPOPT_MSS; opt[1] = TCPOLEN_MSS; opt[2] = (newmss & 0xff00) >> 8; opt[3] = (newmss & 0x00ff); - tcph->check = nf_proto_csum_update(*pskb, htonl(~0), *((__be32 *)opt), - tcph->check, 0); + nf_proto_csum_replace4(&tcph->check, *pskb, 0, *((__be32 *)opt), 0); oldval = ((__be16 *)tcph)[6]; tcph->doff += TCPOLEN_MSS/4; - tcph->check = nf_proto_csum_update(*pskb, - oldval ^ htons(0xFFFF), - ((__be16 *)tcph)[6], - tcph->check, 0); + nf_proto_csum_replace2(&tcph->check, *pskb, + oldval, ((__be16 *)tcph)[6], 0); newtotlen = htons(ntohs(iph->tot_len) + TCPOLEN_MSS); - iph->check = nf_csum_update(iph->tot_len ^ htons(0xFFFF), - newtotlen, iph->check); + nf_csum_replace2(&iph->check, iph->tot_len, newtotlen); iph->tot_len = newtotlen; return IPT_CONTINUE; } diff --git a/net/ipv4/netfilter/ipt_TOS.c b/net/ipv4/netfilter/ipt_TOS.c index 83b80b3a5d2..18e74ac4d42 100644 --- a/net/ipv4/netfilter/ipt_TOS.c +++ b/net/ipv4/netfilter/ipt_TOS.c @@ -30,16 +30,15 @@ target(struct sk_buff **pskb, { const struct ipt_tos_target_info *tosinfo = targinfo; struct iphdr *iph = (*pskb)->nh.iph; - u_int16_t oldtos; if ((iph->tos & IPTOS_TOS_MASK) != tosinfo->tos) { + __u8 oldtos; if (!skb_make_writable(pskb, sizeof(struct iphdr))) return NF_DROP; iph = (*pskb)->nh.iph; oldtos = iph->tos; iph->tos = (iph->tos & IPTOS_PREC_MASK) | tosinfo->tos; - iph->check = nf_csum_update(htons(oldtos) ^ htons(0xFFFF), - htons(iph->tos), iph->check); + nf_csum_replace2(&iph->check, htons(oldtos), htons(iph->tos)); } return IPT_CONTINUE; } diff --git a/net/ipv4/netfilter/ipt_TTL.c b/net/ipv4/netfilter/ipt_TTL.c index ac9517d62af..fffe5ca82e9 100644 --- a/net/ipv4/netfilter/ipt_TTL.c +++ b/net/ipv4/netfilter/ipt_TTL.c @@ -54,9 +54,8 @@ ipt_ttl_target(struct sk_buff **pskb, } if (new_ttl != iph->ttl) { - iph->check = nf_csum_update(htons((iph->ttl << 8)) ^ htons(0xFFFF), - htons(new_ttl << 8), - iph->check); + nf_csum_replace2(&iph->check, htons(iph->ttl << 8), + htons(new_ttl << 8)); iph->ttl = new_ttl; } diff --git a/net/netfilter/core.c b/net/netfilter/core.c index d80b935b3a9..17f9e1cbc73 100644 --- a/net/netfilter/core.c +++ b/net/netfilter/core.c @@ -222,28 +222,21 @@ copy_skb: } EXPORT_SYMBOL(skb_make_writable); -u_int16_t nf_csum_update(u_int32_t oldval, u_int32_t newval, u_int32_t csum) -{ - u_int32_t diff[] = { oldval, newval }; - - return csum_fold(csum_partial((char *)diff, sizeof(diff), ~csum)); -} -EXPORT_SYMBOL(nf_csum_update); - -u_int16_t nf_proto_csum_update(struct sk_buff *skb, - u_int32_t oldval, u_int32_t newval, - u_int16_t csum, int pseudohdr) +void nf_proto_csum_replace4(__sum16 *sum, struct sk_buff *skb, + __be32 from, __be32 to, int pseudohdr) { + __be32 diff[] = { ~from, to }; if (skb->ip_summed != CHECKSUM_PARTIAL) { - csum = nf_csum_update(oldval, newval, csum); + *sum = csum_fold(csum_partial((char *)diff, sizeof(diff), + ~csum_unfold(*sum))); if (skb->ip_summed == CHECKSUM_COMPLETE && pseudohdr) - skb->csum = nf_csum_update(oldval, newval, skb->csum); + skb->csum = ~csum_partial((char *)diff, sizeof(diff), + ~skb->csum); } else if (pseudohdr) - csum = ~nf_csum_update(oldval, newval, ~csum); - - return csum; + *sum = ~csum_fold(csum_partial((char *)diff, sizeof(diff), + csum_unfold(*sum))); } -EXPORT_SYMBOL(nf_proto_csum_update); +EXPORT_SYMBOL(nf_proto_csum_replace4); /* This does not belong here, but locally generated errors need it if connection tracking in use: without this, connection may not be in hash table, and hence -- cgit v1.2.3-70-g09d2 From 1f61ab5ca5cca939a6509892d84b34849e155036 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:44:08 -0800 Subject: [NET]: Preliminaty annotation of skb->csum. It's still not completely right; we need to split it into anon unions of __wsum and unsigned - for cases when we use it for partial checksum and for offset of checksum in skb Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/skbuff.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 41753667541..fcab543d79a 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -273,8 +273,8 @@ struct sk_buff { unsigned int len, data_len, - mac_len, - csum; + mac_len; + __wsum csum; __u32 priority; __u8 local_df:1, cloned:1, -- cgit v1.2.3-70-g09d2 From 6bb100b9fc8f1ce330231b360028ab705a9f0378 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 15 Nov 2006 01:09:32 -0800 Subject: [UDPLite]: udplite.h needs ip6_checksum.h Signed-off-by: David S. Miller --- include/net/udplite.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/net/udplite.h b/include/net/udplite.h index 406eb755b34..3abaab7b78c 100644 --- a/include/net/udplite.h +++ b/include/net/udplite.h @@ -4,6 +4,8 @@ #ifndef _UDPLITE_H #define _UDPLITE_H +#include + /* UDP-Lite socket options */ #define UDPLITE_SEND_CSCOV 10 /* sender partial coverage (as sent) */ #define UDPLITE_RECV_CSCOV 11 /* receiver partial coverage (threshold ) */ -- cgit v1.2.3-70-g09d2 From 58a5a7b9555ea231b557ebef5cabeaf8e951df0b Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Thu, 16 Nov 2006 14:06:06 -0200 Subject: [NET]: Conditionally use bh_lock_sock_nested in sk_receive_skb Spotted by Ian McDonald, tentatively fixed by Gerrit Renker: http://www.mail-archive.com/dccp%40vger.kernel.org/msg00599.html Rewritten not to unroll sk_receive_skb, in the common case, i.e. no lock debugging, its optimized away. Signed-off-by: Arnaldo Carvalho de Melo --- drivers/net/pppoe.c | 2 +- include/net/sock.h | 3 ++- net/core/sock.c | 7 +++++-- net/dccp/ipv4.c | 2 +- net/dccp/ipv6.c | 2 +- net/decnet/dn_nsp_in.c | 2 +- 6 files changed, 11 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/drivers/net/pppoe.c b/drivers/net/pppoe.c index 0adee733b76..315d5c3fc66 100644 --- a/drivers/net/pppoe.c +++ b/drivers/net/pppoe.c @@ -393,7 +393,7 @@ static int pppoe_rcv(struct sk_buff *skb, po = get_item((unsigned long) ph->sid, eth_hdr(skb)->h_source); if (po != NULL) - return sk_receive_skb(sk_pppox(po), skb); + return sk_receive_skb(sk_pppox(po), skb, 0); drop: kfree_skb(skb); out: diff --git a/include/net/sock.h b/include/net/sock.h index dc4b92b8abe..26fc0b16bc0 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -954,7 +954,8 @@ static inline void sock_put(struct sock *sk) sk_free(sk); } -extern int sk_receive_skb(struct sock *sk, struct sk_buff *skb); +extern int sk_receive_skb(struct sock *sk, struct sk_buff *skb, + const int nested); /* Detach socket from process context. * Announce socket dead, detach it from wait queue and inode. diff --git a/net/core/sock.c b/net/core/sock.c index 32ff1c551d6..ab8fafadb4b 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -270,7 +270,7 @@ out: } EXPORT_SYMBOL(sock_queue_rcv_skb); -int sk_receive_skb(struct sock *sk, struct sk_buff *skb) +int sk_receive_skb(struct sock *sk, struct sk_buff *skb, const int nested) { int rc = NET_RX_SUCCESS; @@ -279,7 +279,10 @@ int sk_receive_skb(struct sock *sk, struct sk_buff *skb) skb->dev = NULL; - bh_lock_sock(sk); + if (nested) + bh_lock_sock_nested(sk); + else + bh_lock_sock(sk); if (!sock_owned_by_user(sk)) { /* * trylock + unlock semantics: diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index a20eb71d45d..7114befe7d5 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -899,7 +899,7 @@ static int dccp_v4_rcv(struct sk_buff *skb) goto discard_and_relse; nf_reset(skb); - return sk_receive_skb(sk, skb); + return sk_receive_skb(sk, skb, 1); no_dccp_socket: if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 6eda430ae92..03bb8298250 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -888,7 +888,7 @@ static int dccp_v6_rcv(struct sk_buff **pskb) if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb)) goto discard_and_relse; - return sk_receive_skb(sk, skb) ? -1 : 0; + return sk_receive_skb(sk, skb, 1) ? -1 : 0; no_dccp_socket: if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) diff --git a/net/decnet/dn_nsp_in.c b/net/decnet/dn_nsp_in.c index 7683d4f754d..39a6cf7fb56 100644 --- a/net/decnet/dn_nsp_in.c +++ b/net/decnet/dn_nsp_in.c @@ -804,7 +804,7 @@ got_it: goto free_out; } - return sk_receive_skb(sk, skb); + return sk_receive_skb(sk, skb, 0); } return dn_nsp_no_socket(skb, reason); -- cgit v1.2.3-70-g09d2 From 1f758d93548fb3c6297c05a351a4ba532de6a497 Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Fri, 17 Nov 2006 17:38:43 -0500 Subject: NetLabel: use gfp_t instead of int where it makes sense There were a few places in the NetLabel code where the int type was being used instead of the gfp_t type, this patch corrects this mistake. Signed-off-by: Paul Moore Signed-off-by: James Morris --- include/net/netlabel.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/netlabel.h b/include/net/netlabel.h index 12c214b9ead..ba2f6823805 100644 --- a/include/net/netlabel.h +++ b/include/net/netlabel.h @@ -205,7 +205,7 @@ static inline void netlbl_secattr_destroy(struct netlbl_lsm_secattr *secattr) * pointer on success, or NULL on failure. * */ -static inline struct netlbl_lsm_secattr *netlbl_secattr_alloc(int flags) +static inline struct netlbl_lsm_secattr *netlbl_secattr_alloc(gfp_t flags) { return kzalloc(sizeof(struct netlbl_lsm_secattr), flags); } -- cgit v1.2.3-70-g09d2 From c6fa82a9dd6160e0bc980cb0401c16bf62f2fe66 Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Fri, 17 Nov 2006 17:38:45 -0500 Subject: NetLabel: change netlbl_secattr_init() to return void The netlbl_secattr_init() function would always return 0 making it pointless to have a return value. This patch changes the function to return void. Signed-off-by: Paul Moore Signed-off-by: James Morris --- include/net/netlabel.h | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/net/netlabel.h b/include/net/netlabel.h index ba2f6823805..4e223aa25e5 100644 --- a/include/net/netlabel.h +++ b/include/net/netlabel.h @@ -169,14 +169,12 @@ static inline void netlbl_secattr_cache_free(struct netlbl_lsm_cache *cache) * @secattr: the struct to initialize * * Description: - * Initialize an already allocated netlbl_lsm_secattr struct. Returns zero on - * success, negative values on error. + * Initialize an already allocated netlbl_lsm_secattr struct. * */ -static inline int netlbl_secattr_init(struct netlbl_lsm_secattr *secattr) +static inline void netlbl_secattr_init(struct netlbl_lsm_secattr *secattr) { memset(secattr, 0, sizeof(*secattr)); - return 0; } /** -- cgit v1.2.3-70-g09d2 From 701a90bad99b8081a824cca52c178c8fc8f46bb2 Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Fri, 17 Nov 2006 17:38:46 -0500 Subject: NetLabel: make netlbl_lsm_secattr struct easier/quicker to understand The existing netlbl_lsm_secattr struct required the LSM to check all of the fields to determine if any security attributes were present resulting in a lot of work in the common case of no attributes. This patch adds a 'flags' field which is used to indicate which attributes are present in the structure; this should allow the LSM to do a quick comparison to determine if the structure holds any security attributes. Example: if (netlbl_lsm_secattr->flags) /* security attributes present */ else /* NO security attributes present */ Signed-off-by: Paul Moore Signed-off-by: James Morris --- include/net/netlabel.h | 13 +++++++++++-- net/ipv4/cipso_ipv4.c | 22 ++++++++++++++-------- net/netlabel/netlabel_kapi.c | 5 ++++- security/selinux/ss/services.c | 24 ++++++++++++++++-------- 4 files changed, 45 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/include/net/netlabel.h b/include/net/netlabel.h index 4e223aa25e5..d605d795401 100644 --- a/include/net/netlabel.h +++ b/include/net/netlabel.h @@ -111,11 +111,17 @@ struct netlbl_lsm_cache { void (*free) (const void *data); void *data; }; +#define NETLBL_SECATTR_NONE 0x00000000 +#define NETLBL_SECATTR_DOMAIN 0x00000001 +#define NETLBL_SECATTR_CACHE 0x00000002 +#define NETLBL_SECATTR_MLS_LVL 0x00000004 +#define NETLBL_SECATTR_MLS_CAT 0x00000008 struct netlbl_lsm_secattr { + u32 flags; + char *domain; u32 mls_lvl; - u32 mls_lvl_vld; unsigned char *mls_cat; size_t mls_cat_len; @@ -174,7 +180,10 @@ static inline void netlbl_secattr_cache_free(struct netlbl_lsm_cache *cache) */ static inline void netlbl_secattr_init(struct netlbl_lsm_secattr *secattr) { - memset(secattr, 0, sizeof(*secattr)); + secattr->flags = 0; + secattr->domain = NULL; + secattr->mls_cat = NULL; + secattr->cache = NULL; } /** diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c index 095038ad72a..f0a0785047f 100644 --- a/net/ipv4/cipso_ipv4.c +++ b/net/ipv4/cipso_ipv4.c @@ -319,6 +319,7 @@ static int cipso_v4_cache_check(const unsigned char *key, entry->activity += 1; atomic_inc(&entry->lsm_data->refcount); secattr->cache = entry->lsm_data; + secattr->flags |= NETLBL_SECATTR_CACHE; if (prev_entry == NULL) { spin_unlock_bh(&cipso_v4_cache[bkt].lock); return 0; @@ -991,12 +992,15 @@ static int cipso_v4_gentag_rbm(const struct cipso_v4_doi *doi_def, unsigned char **buffer, u32 *buffer_len) { - int ret_val = -EPERM; + int ret_val; unsigned char *buf = NULL; u32 buf_len; u32 level; - if (secattr->mls_cat) { + if ((secattr->flags & NETLBL_SECATTR_MLS_LVL) == 0) + return -EPERM; + + if (secattr->flags & NETLBL_SECATTR_MLS_CAT) { buf = kzalloc(CIPSO_V4_HDR_LEN + 4 + CIPSO_V4_TAG1_CAT_LEN, GFP_ATOMIC); if (buf == NULL) @@ -1013,10 +1017,10 @@ static int cipso_v4_gentag_rbm(const struct cipso_v4_doi *doi_def, /* This will send packets using the "optimized" format when * possibile as specified in section 3.4.2.6 of the * CIPSO draft. */ - if (cipso_v4_rbm_optfmt && (ret_val > 0 && ret_val < 10)) - ret_val = 10; - - buf_len = 4 + ret_val; + if (cipso_v4_rbm_optfmt && ret_val > 0 && ret_val <= 10) + buf_len = 14; + else + buf_len = 4 + ret_val; } else { buf = kzalloc(CIPSO_V4_HDR_LEN + 4, GFP_ATOMIC); if (buf == NULL) @@ -1070,7 +1074,7 @@ static int cipso_v4_parsetag_rbm(const struct cipso_v4_doi *doi_def, if (ret_val != 0) return ret_val; secattr->mls_lvl = level; - secattr->mls_lvl_vld = 1; + secattr->flags |= NETLBL_SECATTR_MLS_LVL; if (tag_len > 4) { switch (doi_def->type) { @@ -1094,8 +1098,10 @@ static int cipso_v4_parsetag_rbm(const struct cipso_v4_doi *doi_def, if (ret_val < 0) { kfree(secattr->mls_cat); return ret_val; + } else if (ret_val > 0) { + secattr->mls_cat_len = ret_val; + secattr->flags |= NETLBL_SECATTR_MLS_CAT; } - secattr->mls_cat_len = ret_val; } return 0; diff --git a/net/netlabel/netlabel_kapi.c b/net/netlabel/netlabel_kapi.c index ff971103fd0..da2f1975a04 100644 --- a/net/netlabel/netlabel_kapi.c +++ b/net/netlabel/netlabel_kapi.c @@ -62,6 +62,9 @@ int netlbl_socket_setattr(const struct socket *sock, int ret_val = -ENOENT; struct netlbl_dom_map *dom_entry; + if ((secattr->flags & NETLBL_SECATTR_DOMAIN) == 0) + return -ENOENT; + rcu_read_lock(); dom_entry = netlbl_domhsh_getentry(secattr->domain); if (dom_entry == NULL) @@ -200,7 +203,7 @@ void netlbl_cache_invalidate(void) int netlbl_cache_add(const struct sk_buff *skb, const struct netlbl_lsm_secattr *secattr) { - if (secattr->cache == NULL) + if ((secattr->flags & NETLBL_SECATTR_CACHE) == 0) return -ENOMSG; if (CIPSO_V4_OPTEXIST(skb)) diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c index 408820486af..1f5bbb246d2 100644 --- a/security/selinux/ss/services.c +++ b/security/selinux/ss/services.c @@ -2254,8 +2254,6 @@ static void selinux_netlbl_cache_add(struct sk_buff *skb, struct context *ctx) cache = kzalloc(sizeof(*cache), GFP_ATOMIC); if (cache == NULL) goto netlbl_cache_add_return; - secattr.cache->free = selinux_netlbl_cache_free; - secattr.cache->data = (void *)cache; cache->type = NETLBL_CACHE_T_MLS; if (ebitmap_cpy(&cache->data.mls_label.level[0].cat, @@ -2268,6 +2266,10 @@ static void selinux_netlbl_cache_add(struct sk_buff *skb, struct context *ctx) cache->data.mls_label.level[0].sens = ctx->range.level[0].sens; cache->data.mls_label.level[1].sens = ctx->range.level[0].sens; + secattr.cache->free = selinux_netlbl_cache_free; + secattr.cache->data = (void *)cache; + secattr.flags = NETLBL_SECATTR_CACHE; + netlbl_cache_add(skb, &secattr); netlbl_cache_add_return: @@ -2313,7 +2315,7 @@ static int selinux_netlbl_secattr_to_sid(struct sk_buff *skb, POLICY_RDLOCK; - if (secattr->cache) { + if (secattr->flags & NETLBL_SECATTR_CACHE) { cache = NETLBL_CACHE(secattr->cache->data); switch (cache->type) { case NETLBL_CACHE_T_SID: @@ -2346,7 +2348,7 @@ static int selinux_netlbl_secattr_to_sid(struct sk_buff *skb, default: goto netlbl_secattr_to_sid_return; } - } else if (secattr->mls_lvl_vld) { + } else if (secattr->flags & NETLBL_SECATTR_MLS_LVL) { ctx = sidtab_search(&sidtab, base_sid); if (ctx == NULL) goto netlbl_secattr_to_sid_return; @@ -2355,7 +2357,7 @@ static int selinux_netlbl_secattr_to_sid(struct sk_buff *skb, ctx_new.role = ctx->role; ctx_new.type = ctx->type; mls_import_lvl(&ctx_new, secattr->mls_lvl, secattr->mls_lvl); - if (secattr->mls_cat) { + if (secattr->flags & NETLBL_SECATTR_MLS_CAT) { if (mls_import_cat(&ctx_new, secattr->mls_cat, secattr->mls_cat_len, @@ -2414,11 +2416,13 @@ static int selinux_netlbl_skbuff_getsid(struct sk_buff *skb, netlbl_secattr_init(&secattr); rc = netlbl_skbuff_getattr(skb, &secattr); - if (rc == 0) + if (rc == 0 && secattr.flags != NETLBL_SECATTR_NONE) rc = selinux_netlbl_secattr_to_sid(skb, &secattr, base_sid, sid); + else + *sid = SECSID_NULL; netlbl_secattr_destroy(&secattr); return rc; @@ -2455,7 +2459,6 @@ static int selinux_netlbl_socket_setsid(struct socket *sock, u32 sid) secattr.domain = kstrdup(policydb.p_type_val_to_name[ctx->type - 1], GFP_ATOMIC); mls_export_lvl(ctx, &secattr.mls_lvl, NULL); - secattr.mls_lvl_vld = 1; rc = mls_export_cat(ctx, &secattr.mls_cat, &secattr.mls_cat_len, @@ -2464,6 +2467,10 @@ static int selinux_netlbl_socket_setsid(struct socket *sock, u32 sid) if (rc != 0) goto netlbl_socket_setsid_return; + secattr.flags |= NETLBL_SECATTR_DOMAIN | NETLBL_SECATTR_MLS_LVL; + if (secattr.mls_cat) + secattr.flags |= NETLBL_SECATTR_MLS_CAT; + rc = netlbl_socket_setattr(sock, &secattr); if (rc == 0) sksec->nlbl_state = NLBL_LABELED; @@ -2564,6 +2571,7 @@ void selinux_netlbl_sock_graft(struct sock *sk, struct socket *sock) netlbl_secattr_init(&secattr); if (netlbl_sock_getattr(sk, &secattr) == 0 && + secattr.flags != NETLBL_SECATTR_NONE && selinux_netlbl_secattr_to_sid(NULL, &secattr, SECINITSID_UNLABELED, @@ -2756,7 +2764,7 @@ int selinux_netlbl_socket_setsockopt(struct socket *sock, sksec->nlbl_state == NLBL_LABELED) { netlbl_secattr_init(&secattr); rc = netlbl_socket_getattr(sock, &secattr); - if (rc == 0 && (secattr.cache || secattr.mls_lvl_vld)) + if (rc == 0 && secattr.flags != NETLBL_SECATTR_NONE) rc = -EACCES; netlbl_secattr_destroy(&secattr); } -- cgit v1.2.3-70-g09d2 From c6b1677a54330b72d961270c9a66bb99c62c4a8e Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Fri, 17 Nov 2006 17:38:51 -0500 Subject: NetLabel: use the correct CIPSOv4 MLS label limits The CIPSOv4 engine currently has MLS label limits which are slightly larger than what the draft allows. This is not a major problem due to the current implementation but we should fix this so it doesn't bite us later. Signed-off-by: Paul Moore Signed-off-by: James Morris --- include/net/cipso_ipv4.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/cipso_ipv4.h b/include/net/cipso_ipv4.h index 718b4d9c891..4c9522c5178 100644 --- a/include/net/cipso_ipv4.h +++ b/include/net/cipso_ipv4.h @@ -58,10 +58,10 @@ #define CIPSO_V4_MAP_PASS 2 /* limits */ -#define CIPSO_V4_MAX_REM_LVLS 256 +#define CIPSO_V4_MAX_REM_LVLS 255 #define CIPSO_V4_INV_LVL 0x80000000 #define CIPSO_V4_MAX_LOC_LVLS (CIPSO_V4_INV_LVL - 1) -#define CIPSO_V4_MAX_REM_CATS 65536 +#define CIPSO_V4_MAX_REM_CATS 65534 #define CIPSO_V4_INV_CAT 0x80000000 #define CIPSO_V4_MAX_LOC_CATS (CIPSO_V4_INV_CAT - 1) -- cgit v1.2.3-70-g09d2 From bac0dff6cd194f7a28f7e840c9b6a7aa71c6ef97 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Sun, 19 Nov 2006 14:15:05 -0800 Subject: [BNX2]: Add 5709 PCI ID. Add PCI ID and detection for 5709 copper and SerDes chips. Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/bnx2.c | 13 +++++++++++-- include/linux/pci_ids.h | 1 + 2 files changed, 12 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c index 92897efbc26..157463b4fa9 100644 --- a/drivers/net/bnx2.c +++ b/drivers/net/bnx2.c @@ -86,6 +86,7 @@ typedef enum { NC370F, BCM5708, BCM5708S, + BCM5709, } board_t; /* indexed by board_t, above */ @@ -99,6 +100,7 @@ static const struct { { "HP NC370F Multifunction Gigabit Server Adapter" }, { "Broadcom NetXtreme II BCM5708 1000Base-T" }, { "Broadcom NetXtreme II BCM5708 1000Base-SX" }, + { "Broadcom NetXtreme II BCM5709 1000Base-T" }, }; static struct pci_device_id bnx2_pci_tbl[] = { @@ -116,6 +118,8 @@ static struct pci_device_id bnx2_pci_tbl[] = { PCI_ANY_ID, PCI_ANY_ID, 0, 0, BCM5706S }, { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_5708S, PCI_ANY_ID, PCI_ANY_ID, 0, 0, BCM5708S }, + { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_5709, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, BCM5709 }, { 0, } }; @@ -5854,10 +5858,15 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev) bp->phy_addr = 1; /* Disable WOL support if we are running on a SERDES chip. */ - if (CHIP_BOND_ID(bp) & CHIP_BOND_ID_SERDES_BIT) { + if (CHIP_NUM(bp) == CHIP_NUM_5709) { + if (CHIP_BOND_ID(bp) != BNX2_MISC_DUAL_MEDIA_CTRL_BOND_ID_C) + bp->phy_flags |= PHY_SERDES_FLAG; + } else if (CHIP_BOND_ID(bp) & CHIP_BOND_ID_SERDES_BIT) bp->phy_flags |= PHY_SERDES_FLAG; + + if (bp->phy_flags & PHY_SERDES_FLAG) { bp->flags |= NO_WOL_FLAG; - if (CHIP_NUM(bp) == CHIP_NUM_5708) { + if (CHIP_NUM(bp) != CHIP_NUM_5706) { bp->phy_addr = 2; reg = REG_RD_IND(bp, bp->shmem_base + BNX2_SHARED_HW_CFG_CONFIG); diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index fd5033b8a92..c09da1e30c5 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1897,6 +1897,7 @@ #define PCI_VENDOR_ID_BROADCOM 0x14e4 #define PCI_DEVICE_ID_TIGON3_5752 0x1600 #define PCI_DEVICE_ID_TIGON3_5752M 0x1601 +#define PCI_DEVICE_ID_NX2_5709 0x1639 #define PCI_DEVICE_ID_TIGON3_5700 0x1644 #define PCI_DEVICE_ID_TIGON3_5701 0x1645 #define PCI_DEVICE_ID_TIGON3_5702 0x1646 -- cgit v1.2.3-70-g09d2 From 82e3ab9dbeebd5c8d5402ad1607d22086271a56d Mon Sep 17 00:00:00 2001 From: Ian McDonald Date: Mon, 20 Nov 2006 19:19:32 -0200 Subject: [DCCP]: Adds the tx buffer sysctls This one got lost on the way from Ian to Gerrit to me, fix it. Signed-off-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- Documentation/networking/dccp.txt | 4 ++++ include/linux/sysctl.h | 1 + net/dccp/sysctl.c | 9 +++++++++ 3 files changed, 14 insertions(+) (limited to 'include') diff --git a/Documentation/networking/dccp.txt b/Documentation/networking/dccp.txt index 1910d097a0a..dda15886bcb 100644 --- a/Documentation/networking/dccp.txt +++ b/Documentation/networking/dccp.txt @@ -100,6 +100,10 @@ rx_ccid = 2 seq_window = 100 The initial sequence window (sec. 7.5.2). +tx_qlen = 5 + The size of the transmit buffer in packets. A value of 0 corresponds + to an unbounded transmit buffer. + Notes ===== diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 2e8c5ad8279..61dd99c6c2a 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -617,6 +617,7 @@ enum { NET_DCCP_DEFAULT_REQ_RETRIES = 7, NET_DCCP_DEFAULT_RETRIES1 = 8, NET_DCCP_DEFAULT_RETRIES2 = 9, + NET_DCCP_DEFAULT_TX_QLEN = 10, }; /* /proc/sys/net/ipx */ diff --git a/net/dccp/sysctl.c b/net/dccp/sysctl.c index 8b62061e570..4775ba3faa0 100644 --- a/net/dccp/sysctl.c +++ b/net/dccp/sysctl.c @@ -91,6 +91,15 @@ static struct ctl_table dccp_default_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, + { + .ctl_name = NET_DCCP_DEFAULT_TX_QLEN, + .procname = "tx_qlen", + .data = &sysctl_dccp_tx_qlen, + .maxlen = sizeof(sysctl_dccp_tx_qlen), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { .ctl_name = 0, } }; -- cgit v1.2.3-70-g09d2 From 89c89458152c4d387eeca6532b6e50780fc59f8b Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Mon, 20 Nov 2006 16:56:48 -0800 Subject: [IPV6] net/ipv6/sit.c: make 2 functions static This patch makes two needlessly global functions static. Signed-off-by: Adrian Bunk Signed-off-by: David S. Miller --- include/net/ipip.h | 4 ---- net/ipv6/sit.c | 4 ++-- 2 files changed, 2 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/net/ipip.h b/include/net/ipip.h index 84058858eea..7cdc914322f 100644 --- a/include/net/ipip.h +++ b/include/net/ipip.h @@ -44,8 +44,4 @@ struct ip_tunnel } \ } while (0) - -extern int sit_init(void); -extern void sit_cleanup(void); - #endif diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 85ff3dc4514..77b7b091143 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -809,7 +809,7 @@ static void __exit sit_destroy_tunnels(void) } } -void __exit sit_cleanup(void) +static void __exit sit_cleanup(void) { inet_del_protocol(&sit_protocol, IPPROTO_IPV6); @@ -819,7 +819,7 @@ void __exit sit_cleanup(void) rtnl_unlock(); } -int __init sit_init(void) +static int __init sit_init(void) { int err; -- cgit v1.2.3-70-g09d2 From f3ffaf14681e3cad61006873be8656ab41b793e0 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 16:59:12 -0800 Subject: [SCTP]: Annotate SCTP headers. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/sctp.h | 72 ++++++++++++++++++++++++++-------------------------- 1 file changed, 36 insertions(+), 36 deletions(-) (limited to 'include') diff --git a/include/linux/sctp.h b/include/linux/sctp.h index 6ec66dec29f..35108fe7a68 100644 --- a/include/linux/sctp.h +++ b/include/linux/sctp.h @@ -57,17 +57,17 @@ /* Section 3.1. SCTP Common Header Format */ typedef struct sctphdr { - __u16 source; - __u16 dest; - __u32 vtag; - __u32 checksum; + __be16 source; + __be16 dest; + __be32 vtag; + __be32 checksum; } __attribute__((packed)) sctp_sctphdr_t; /* Section 3.2. Chunk Field Descriptions. */ typedef struct sctp_chunkhdr { __u8 type; __u8 flags; - __u16 length; + __be16 length; } __attribute__((packed)) sctp_chunkhdr_t; @@ -153,8 +153,8 @@ enum { SCTP_CHUNK_FLAG_T = 0x01 }; */ typedef struct sctp_paramhdr { - __u16 type; - __u16 length; + __be16 type; + __be16 length; } __attribute__((packed)) sctp_paramhdr_t; typedef enum { @@ -203,10 +203,10 @@ enum { SCTP_PARAM_ACTION_MASK = __constant_htons(0xc000), }; /* RFC 2960 Section 3.3.1 Payload Data (DATA) (0) */ typedef struct sctp_datahdr { - __u32 tsn; - __u16 stream; - __u16 ssn; - __u32 ppid; + __be32 tsn; + __be16 stream; + __be16 ssn; + __be32 ppid; __u8 payload[0]; } __attribute__((packed)) sctp_datahdr_t; @@ -232,11 +232,11 @@ enum { SCTP_DATA_FRAG_MASK = 0x03, }; * endpoints. */ typedef struct sctp_inithdr { - __u32 init_tag; - __u32 a_rwnd; - __u16 num_outbound_streams; - __u16 num_inbound_streams; - __u32 initial_tsn; + __be32 init_tag; + __be32 a_rwnd; + __be16 num_outbound_streams; + __be16 num_inbound_streams; + __be32 initial_tsn; __u8 params[0]; } __attribute__((packed)) sctp_inithdr_t; @@ -261,7 +261,7 @@ typedef struct sctp_ipv6addr_param { /* Section 3.3.2.1 Cookie Preservative (9) */ typedef struct sctp_cookie_preserve_param { sctp_paramhdr_t param_hdr; - uint32_t lifespan_increment; + __be32 lifespan_increment; } __attribute__((packed)) sctp_cookie_preserve_param_t; /* Section 3.3.2.1 Host Name Address (11) */ @@ -273,7 +273,7 @@ typedef struct sctp_hostname_param { /* Section 3.3.2.1 Supported Address Types (12) */ typedef struct sctp_supported_addrs_param { sctp_paramhdr_t param_hdr; - uint16_t types[0]; + __be16 types[0]; } __attribute__((packed)) sctp_supported_addrs_param_t; /* Appendix A. ECN Capable (32768) */ @@ -284,7 +284,7 @@ typedef struct sctp_ecn_capable_param { /* ADDIP Section 3.2.6 Adaption Layer Indication */ typedef struct sctp_adaption_ind_param { struct sctp_paramhdr param_hdr; - __u32 adaption_ind; + __be32 adaption_ind; } __attribute__((packed)) sctp_adaption_ind_param_t; /* RFC 2960. Section 3.3.3 Initiation Acknowledgement (INIT ACK) (2): @@ -316,11 +316,11 @@ typedef struct sctp_unrecognized_param { */ typedef struct sctp_gap_ack_block { - __u16 start; - __u16 end; + __be16 start; + __be16 end; } __attribute__((packed)) sctp_gap_ack_block_t; -typedef uint32_t sctp_dup_tsn_t; +typedef __be32 sctp_dup_tsn_t; typedef union { sctp_gap_ack_block_t gab; @@ -328,10 +328,10 @@ typedef union { } sctp_sack_variable_t; typedef struct sctp_sackhdr { - __u32 cum_tsn_ack; - __u32 a_rwnd; - __u16 num_gap_ack_blocks; - __u16 num_dup_tsns; + __be32 cum_tsn_ack; + __be32 a_rwnd; + __be16 num_gap_ack_blocks; + __be16 num_dup_tsns; sctp_sack_variable_t variable[0]; } __attribute__((packed)) sctp_sackhdr_t; @@ -371,7 +371,7 @@ typedef struct sctp_abort_chunk { * and the highest consecutive acking value. */ typedef struct sctp_shutdownhdr { - __u32 cum_tsn_ack; + __be32 cum_tsn_ack; } __attribute__((packed)) sctp_shutdownhdr_t; struct sctp_shutdown_chunk_t { @@ -382,8 +382,8 @@ struct sctp_shutdown_chunk_t { /* RFC 2960. Section 3.3.10 Operation Error (ERROR) (9) */ typedef struct sctp_errhdr { - __u16 cause; - __u16 length; + __be16 cause; + __be16 length; __u8 variable[0]; } __attribute__((packed)) sctp_errhdr_t; @@ -462,7 +462,7 @@ typedef enum { * Explicit Congestion Notification Echo (ECNE) (12) */ typedef struct sctp_ecnehdr { - __u32 lowest_tsn; + __be32 lowest_tsn; } sctp_ecnehdr_t; typedef struct sctp_ecne_chunk { @@ -474,7 +474,7 @@ typedef struct sctp_ecne_chunk { * Congestion Window Reduced (CWR) (13) */ typedef struct sctp_cwrhdr { - __u32 lowest_tsn; + __be32 lowest_tsn; } sctp_cwrhdr_t; typedef struct sctp_cwr_chunk { @@ -529,12 +529,12 @@ typedef struct sctp_cwr_chunk { * chunks this field MUST be filled in. */ struct sctp_fwdtsn_skip { - __u16 stream; - __u16 ssn; + __be16 stream; + __be16 ssn; } __attribute__((packed)); struct sctp_fwdtsn_hdr { - __u32 new_cum_tsn; + __be32 new_cum_tsn; struct sctp_fwdtsn_skip skip[0]; } __attribute((packed)); @@ -578,11 +578,11 @@ struct sctp_fwdtsn_chunk { */ typedef struct sctp_addip_param { sctp_paramhdr_t param_hdr; - __u32 crr_id; + __be32 crr_id; } __attribute__((packed)) sctp_addip_param_t; typedef struct sctp_addiphdr { - __u32 serial; + __be32 serial; __u8 params[0]; } __attribute__((packed)) sctp_addiphdr_t; -- cgit v1.2.3-70-g09d2 From 5bf2db03908b9121805af3c76e3ac2d0759e199f Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 16:59:45 -0800 Subject: [SCTP]: Annotate sctp_init_cause(). Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sctp/sm.h | 2 +- net/sctp/sm_make_chunk.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h index de313de4fef..a3b0e1e896f 100644 --- a/include/net/sctp/sm.h +++ b/include/net/sctp/sm.h @@ -213,7 +213,7 @@ struct sctp_chunk *sctp_make_shutdown_ack(const struct sctp_association *asoc, const struct sctp_chunk *); struct sctp_chunk *sctp_make_shutdown_complete(const struct sctp_association *, const struct sctp_chunk *); -void sctp_init_cause(struct sctp_chunk *, __u16 cause, const void *, size_t); +void sctp_init_cause(struct sctp_chunk *, __be16 cause, const void *, size_t); struct sctp_chunk *sctp_make_abort(const struct sctp_association *, const struct sctp_chunk *, const size_t hint); diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 507dff72c58..37472be1601 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -111,7 +111,7 @@ static const struct sctp_paramhdr prsctp_param = { * provided chunk, as most cause codes will be embedded inside an * abort chunk. */ -void sctp_init_cause(struct sctp_chunk *chunk, __u16 cause_code, +void sctp_init_cause(struct sctp_chunk *chunk, __be16 cause_code, const void *payload, size_t paylen) { sctp_errhdr_t err; -- cgit v1.2.3-70-g09d2 From 63706c5c6fd07f58bed85d0aa031ffbce3a0385f Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:00:05 -0800 Subject: [SCTP]: sctp_make_op_error() annotations. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sctp/sm.h | 2 +- net/sctp/sm_make_chunk.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h index a3b0e1e896f..a7b935a76a6 100644 --- a/include/net/sctp/sm.h +++ b/include/net/sctp/sm.h @@ -236,7 +236,7 @@ struct sctp_chunk *sctp_make_heartbeat_ack(const struct sctp_association *, const size_t paylen); struct sctp_chunk *sctp_make_op_error(const struct sctp_association *, const struct sctp_chunk *chunk, - __u16 cause_code, + __be16 cause_code, const void *payload, size_t paylen); diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 37472be1601..a3bde93bd8a 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -951,7 +951,7 @@ nodata: /* Create an Operation Error chunk. */ struct sctp_chunk *sctp_make_op_error(const struct sctp_association *asoc, const struct sctp_chunk *chunk, - __u16 cause_code, const void *payload, + __be16 cause_code, const void *payload, size_t paylen) { struct sctp_chunk *retval; -- cgit v1.2.3-70-g09d2 From dc251b2b1c4bfea51903cb9fbc141a5b33f6aca7 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:00:44 -0800 Subject: [SCTP]: SCTP_CMD_INIT_FAILED annotations. argument stored for SCTP_CMD_INIT_FAILED is always __be16 (protocol error). Introduced new field and accessor for it (SCTP_PERR()); switched to their use (from SCTP_U32() and .u32) Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sctp/command.h | 2 ++ net/sctp/sm_sideeffect.c | 2 +- net/sctp/sm_statefuns.c | 12 ++++++------ 3 files changed, 9 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/include/net/sctp/command.h b/include/net/sctp/command.h index 807d6f1ef4b..dfe5d44b61a 100644 --- a/include/net/sctp/command.h +++ b/include/net/sctp/command.h @@ -119,6 +119,7 @@ typedef union { __u16 u16; __u8 u8; int error; + __be16 err; sctp_state_t state; sctp_event_timeout_t to; unsigned long zero; @@ -167,6 +168,7 @@ SCTP_ARG_CONSTRUCTOR(U32, __u32, u32) SCTP_ARG_CONSTRUCTOR(U16, __u16, u16) SCTP_ARG_CONSTRUCTOR(U8, __u8, u8) SCTP_ARG_CONSTRUCTOR(ERROR, int, error) +SCTP_ARG_CONSTRUCTOR(PERR, __be16, err) /* protocol error */ SCTP_ARG_CONSTRUCTOR(STATE, sctp_state_t, state) SCTP_ARG_CONSTRUCTOR(TO, sctp_event_timeout_t, to) SCTP_ARG_CONSTRUCTOR(PTR, void *, ptr) diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index 9c10bdec1af..e4c43a3438d 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -1360,7 +1360,7 @@ static int sctp_cmd_interpreter(sctp_event_t event_type, break; case SCTP_CMD_INIT_FAILED: - sctp_cmd_init_failed(commands, asoc, cmd->obj.u32); + sctp_cmd_init_failed(commands, asoc, cmd->obj.err); break; case SCTP_CMD_ASSOC_FAILED: diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index b9063ed2a65..f462a227c7a 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -2138,7 +2138,7 @@ static sctp_disposition_t sctp_sf_do_5_2_6_stale(const struct sctp_endpoint *ep, sctp_add_cmd_sf(commands, SCTP_CMD_SET_SK_ERR, SCTP_ERROR(ETIMEDOUT)); sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED, - SCTP_U32(SCTP_ERROR_STALE_COOKIE)); + SCTP_PERR(SCTP_ERROR_STALE_COOKIE)); return SCTP_DISPOSITION_DELETE_TCB; } @@ -2370,7 +2370,7 @@ static sctp_disposition_t sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands, sctp_add_cmd_sf(commands, SCTP_CMD_SET_SK_ERR, SCTP_ERROR(sk_err)); /* CMD_INIT_FAILED will DELETE_TCB. */ sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED, - SCTP_U32(error)); + SCTP_PERR(error)); return SCTP_DISPOSITION_ABORT; } @@ -3743,7 +3743,7 @@ static sctp_disposition_t sctp_sf_violation_chunklen( sctp_add_cmd_sf(commands, SCTP_CMD_SET_SK_ERR, SCTP_ERROR(ECONNREFUSED)); sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED, - SCTP_U32(SCTP_ERROR_PROTO_VIOLATION)); + SCTP_PERR(SCTP_ERROR_PROTO_VIOLATION)); } else { sctp_add_cmd_sf(commands, SCTP_CMD_SET_SK_ERR, SCTP_ERROR(ECONNABORTED)); @@ -4199,7 +4199,7 @@ sctp_disposition_t sctp_sf_cookie_wait_prm_abort( SCTP_ERROR(ECONNREFUSED)); /* Delete the established association. */ sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED, - SCTP_U32(SCTP_ERROR_USER_ABORT)); + SCTP_PERR(SCTP_ERROR_USER_ABORT)); return retval; } @@ -4693,7 +4693,7 @@ sctp_disposition_t sctp_sf_t1_init_timer_expire(const struct sctp_endpoint *ep, sctp_add_cmd_sf(commands, SCTP_CMD_SET_SK_ERR, SCTP_ERROR(ETIMEDOUT)); sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED, - SCTP_U32(SCTP_ERROR_NO_ERROR)); + SCTP_PERR(SCTP_ERROR_NO_ERROR)); return SCTP_DISPOSITION_DELETE_TCB; } @@ -4745,7 +4745,7 @@ sctp_disposition_t sctp_sf_t1_cookie_timer_expire(const struct sctp_endpoint *ep sctp_add_cmd_sf(commands, SCTP_CMD_SET_SK_ERR, SCTP_ERROR(ETIMEDOUT)); sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED, - SCTP_U32(SCTP_ERROR_NO_ERROR)); + SCTP_PERR(SCTP_ERROR_NO_ERROR)); return SCTP_DISPOSITION_DELETE_TCB; } -- cgit v1.2.3-70-g09d2 From 72f17e1c0984fbdb60abf245d81b947393910100 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:01:23 -0800 Subject: [SCTP]: Annotate tsn_dups. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sctp/tsnmap.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/sctp/tsnmap.h b/include/net/sctp/tsnmap.h index 021947da70e..70a824df6f6 100644 --- a/include/net/sctp/tsnmap.h +++ b/include/net/sctp/tsnmap.h @@ -105,7 +105,7 @@ struct sctp_tsnmap { * every SACK. Store up to SCTP_MAX_DUP_TSNS worth of * information. */ - __u32 dup_tsns[SCTP_MAX_DUP_TSNS]; + __be32 dup_tsns[SCTP_MAX_DUP_TSNS]; __u16 num_dup_tsns; /* Record gap ack block information here. */ @@ -162,7 +162,7 @@ static inline __u16 sctp_tsnmap_num_dups(struct sctp_tsnmap *map) } /* Return pointer to duplicate tsn array as needed by SACK. */ -static inline __u32 *sctp_tsnmap_get_dups(struct sctp_tsnmap *map) +static inline __be32 *sctp_tsnmap_get_dups(struct sctp_tsnmap *map) { map->num_dup_tsns = 0; return map->dup_tsns; -- cgit v1.2.3-70-g09d2 From dbc16db1e58da6c346ca3e63870c17b93fbed0f0 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:01:42 -0800 Subject: [SCTP]: Trivial sctp endianness annotations. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sctp/constants.h | 2 +- include/net/sctp/sctp.h | 2 +- include/net/sctp/sm.h | 2 +- include/net/sctp/structs.h | 2 +- net/sctp/associola.c | 2 +- net/sctp/ipv6.c | 4 ++-- net/sctp/protocol.c | 2 +- net/sctp/sm_make_chunk.c | 18 +++++++++--------- 8 files changed, 17 insertions(+), 17 deletions(-) (limited to 'include') diff --git a/include/net/sctp/constants.h b/include/net/sctp/constants.h index 6c632e26f72..5ddb8559986 100644 --- a/include/net/sctp/constants.h +++ b/include/net/sctp/constants.h @@ -356,7 +356,7 @@ typedef enum { * addresses. */ #define IS_IPV4_UNUSABLE_ADDRESS(a) \ - ((INADDR_BROADCAST == *a) || \ + ((htonl(INADDR_BROADCAST) == *a) || \ (MULTICAST(*a)) || \ (((unsigned char *)(a))[0] == 0) || \ ((((unsigned char *)(a))[0] == 198) && \ diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index 764e3af5be9..215461f18db 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -585,7 +585,7 @@ static inline int ipver2af(__u8 ipver) } /* Convert from an address parameter type to an address family. */ -static inline int param_type2af(__u16 type) +static inline int param_type2af(__be16 type) { switch (type) { case SCTP_PARAM_IPV4_ADDRESS: diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h index a7b935a76a6..3269ed1cc22 100644 --- a/include/net/sctp/sm.h +++ b/include/net/sctp/sm.h @@ -243,7 +243,7 @@ struct sctp_chunk *sctp_make_op_error(const struct sctp_association *, struct sctp_chunk *sctp_make_asconf_update_ip(struct sctp_association *, union sctp_addr *, struct sockaddr *, - int, __u16); + int, __be16); struct sctp_chunk *sctp_make_asconf_set_prim(struct sctp_association *asoc, union sctp_addr *addr); struct sctp_chunk *sctp_process_asconf(struct sctp_association *asoc, diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 5596f5d97ed..ca77bc5b7ff 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1482,7 +1482,7 @@ struct sctp_association { /* This mask is used to disable sending the ASCONF chunk * with specified parameter to peer. */ - __u16 addip_disabled_mask; + __be16 addip_disabled_mask; struct sctp_inithdr i; int cookie_len; diff --git a/net/sctp/associola.c b/net/sctp/associola.c index ed0445fe85e..a5d7abba89e 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -868,7 +868,7 @@ struct sctp_transport *sctp_assoc_lookup_tsn(struct sctp_association *asoc, struct list_head *entry, *pos; struct sctp_transport *transport; struct sctp_chunk *chunk; - __u32 key = htonl(tsn); + __be32 key = htonl(tsn); match = NULL; diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 3c316c688ad..1f9f5ec0759 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -239,7 +239,7 @@ static inline int sctp_v6_addr_match_len(union sctp_addr *s1, int i, j; for (i = 0; i < 4 ; i++) { - __u32 a1xora2; + __be32 a1xora2; a1xora2 = a1->s6_addr32[i] ^ a2->s6_addr32[i]; @@ -425,7 +425,7 @@ static int sctp_v6_to_addr_param(const union sctp_addr *addr, int length = sizeof(sctp_ipv6addr_param_t); param->v6.param_hdr.type = SCTP_PARAM_IPV6_ADDRESS; - param->v6.param_hdr.length = ntohs(length); + param->v6.param_hdr.length = htons(length); ipv6_addr_copy(¶m->v6.addr, &addr->v6.sin6_addr); return length; diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index c107d45f6e6..4310e7fac44 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -307,7 +307,7 @@ static int sctp_v4_to_addr_param(const union sctp_addr *addr, int length = sizeof(sctp_ipv4addr_param_t); param->v4.param_hdr.type = SCTP_PARAM_IPV4_ADDRESS; - param->v4.param_hdr.length = ntohs(length); + param->v4.param_hdr.length = htons(length); param->v4.addr.s_addr = addr->v4.sin_addr.s_addr; return length; diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index a3bde93bd8a..ccbba6b24d1 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -1190,7 +1190,7 @@ void sctp_chunk_assign_ssn(struct sctp_chunk *chunk) if (chunk->chunk_hdr->flags & SCTP_DATA_UNORDERED) { ssn = 0; } else { - sid = htons(chunk->subh.data_hdr->stream); + sid = ntohs(chunk->subh.data_hdr->stream); if (chunk->chunk_hdr->flags & SCTP_DATA_LAST_FRAG) ssn = sctp_ssn_next(&chunk->asoc->ssnmap->out, sid); else @@ -2230,7 +2230,7 @@ struct sctp_chunk *sctp_make_asconf_update_ip(struct sctp_association *asoc, union sctp_addr *laddr, struct sockaddr *addrs, int addrcnt, - __u16 flags) + __be16 flags) { sctp_addip_param_t param; struct sctp_chunk *retval; @@ -2364,13 +2364,13 @@ static struct sctp_chunk *sctp_make_asconf_ack(const struct sctp_association *as /* Add response parameters to an ASCONF_ACK chunk. */ static void sctp_add_asconf_response(struct sctp_chunk *chunk, __u32 crr_id, - __u16 err_code, sctp_addip_param_t *asconf_param) + __be16 err_code, sctp_addip_param_t *asconf_param) { sctp_addip_param_t ack_param; sctp_errhdr_t err_param; int asconf_param_len = 0; int err_param_len = 0; - __u16 response_type; + __be16 response_type; if (SCTP_ERROR_NO_ERROR == err_code) { response_type = SCTP_PARAM_SUCCESS_REPORT; @@ -2404,7 +2404,7 @@ static void sctp_add_asconf_response(struct sctp_chunk *chunk, __u32 crr_id, } /* Process a asconf parameter. */ -static __u16 sctp_process_asconf_param(struct sctp_association *asoc, +static __be16 sctp_process_asconf_param(struct sctp_association *asoc, struct sctp_chunk *asconf, sctp_addip_param_t *asconf_param) { @@ -2487,7 +2487,7 @@ struct sctp_chunk *sctp_process_asconf(struct sctp_association *asoc, sctp_addip_param_t *asconf_param; struct sctp_chunk *asconf_ack; - __u16 err_code; + __be16 err_code; int length = 0; int chunk_len = asconf->skb->len; __u32 serial; @@ -2630,7 +2630,7 @@ static int sctp_asconf_param_success(struct sctp_association *asoc, * All TLVs after the failed response are considered unsuccessful unless a * specific success indication is present for the parameter. */ -static __u16 sctp_get_asconf_response(struct sctp_chunk *asconf_ack, +static __be16 sctp_get_asconf_response(struct sctp_chunk *asconf_ack, sctp_addip_param_t *asconf_param, int no_err) { @@ -2638,7 +2638,7 @@ static __u16 sctp_get_asconf_response(struct sctp_chunk *asconf_ack, sctp_errhdr_t *err_param; int length; int asconf_ack_len = asconf_ack->skb->len; - __u16 err_code; + __be16 err_code; if (no_err) err_code = SCTP_ERROR_NO_ERROR; @@ -2694,7 +2694,7 @@ int sctp_process_asconf_ack(struct sctp_association *asoc, int all_param_pass = 0; int no_err = 1; int retval = 0; - __u16 err_code = SCTP_ERROR_NO_ERROR; + __be16 err_code = SCTP_ERROR_NO_ERROR; /* Skip the chunkhdr and addiphdr from the last asconf sent and store * a pointer to address parameter. -- cgit v1.2.3-70-g09d2 From 04afd8b282d702bc122051751466000e9513ef96 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:02:01 -0800 Subject: [SCTP]: Beginning of sin_port fixes. That's going to be a long series. Introduced temporary helpers doing copy-and-convert for sctp_addr; they are used to kill flip-in-place in global data structures and will be used to gradually push host-endian uses of sctp_addr out of existence. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 22 ++++++++++++++++++++++ net/sctp/socket.c | 29 ++++++++++++----------------- 2 files changed, 34 insertions(+), 17 deletions(-) (limited to 'include') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index ca77bc5b7ff..2aa61ac9a9f 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -74,6 +74,28 @@ union sctp_addr { struct sockaddr sa; }; +static inline void flip_to_n(union sctp_addr *to, const union sctp_addr *from) +{ + size_t len; + if (from->sa.sa_family == AF_INET6) + len = sizeof(struct sockaddr_in6); + else + len = sizeof(struct sockaddr); + memcpy(to, from, len); + to->v4.sin_port = htons(from->v4.sin_port); +} + +static inline void flip_to_h(union sctp_addr *to, const union sctp_addr *from) +{ + size_t len; + if (from->sa.sa_family == AF_INET6) + len = sizeof(struct sockaddr_in6); + else + len = sizeof(struct sockaddr); + memcpy(to, from, len); + to->v4.sin_port = ntohs(from->v4.sin_port); +} + /* Forward declarations for data structures. */ struct sctp_globals; struct sctp_endpoint; diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 935bc9187fd..f4d13ab79f4 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -228,12 +228,12 @@ static struct sctp_transport *sctp_addr_id2transport(struct sock *sk, struct sctp_association *addr_asoc = NULL, *id_asoc = NULL; struct sctp_transport *transport; union sctp_addr *laddr = (union sctp_addr *)addr; + union sctp_addr tmp; - laddr->v4.sin_port = ntohs(laddr->v4.sin_port); + flip_to_h(&tmp, laddr); addr_asoc = sctp_endpoint_lookup_assoc(sctp_sk(sk)->ep, - (union sctp_addr *)addr, + &tmp, &transport); - laddr->v4.sin_port = htons(laddr->v4.sin_port); if (!addr_asoc) return NULL; @@ -313,6 +313,7 @@ SCTP_STATIC int sctp_do_bind(struct sock *sk, union sctp_addr *addr, int len) struct sctp_af *af; unsigned short snum; int ret = 0; + union sctp_addr tmp; /* Common sockaddr verification. */ af = sctp_sockaddr_af(sp, addr, len); @@ -368,9 +369,8 @@ SCTP_STATIC int sctp_do_bind(struct sock *sk, union sctp_addr *addr, int len) sctp_write_lock(&ep->base.addr_lock); /* Use GFP_ATOMIC since BHs are disabled. */ - addr->v4.sin_port = ntohs(addr->v4.sin_port); - ret = sctp_add_bind_addr(bp, addr, 1, GFP_ATOMIC); - addr->v4.sin_port = htons(addr->v4.sin_port); + flip_to_h(&tmp, addr); + ret = sctp_add_bind_addr(bp, &tmp, 1, GFP_ATOMIC); sctp_write_unlock(&ep->base.addr_lock); sctp_local_bh_enable(); @@ -4194,12 +4194,8 @@ static int sctp_getsockopt_primary_addr(struct sock *sk, int len, if (!asoc->peer.primary_path) return -ENOTCONN; - asoc->peer.primary_path->ipaddr.v4.sin_port = - htons(asoc->peer.primary_path->ipaddr.v4.sin_port); - memcpy(&prim.ssp_addr, &asoc->peer.primary_path->ipaddr, - sizeof(union sctp_addr)); - asoc->peer.primary_path->ipaddr.v4.sin_port = - ntohs(asoc->peer.primary_path->ipaddr.v4.sin_port); + flip_to_n((union sctp_addr *)&prim.ssp_addr, + &asoc->peer.primary_path->ipaddr); sctp_get_pf_specific(sk->sk_family)->addr_v4map(sp, (union sctp_addr *)&prim.ssp_addr); @@ -4642,12 +4638,12 @@ static long sctp_get_port_local(struct sock *sk, union sctp_addr *addr) { struct sctp_bind_hashbucket *head; /* hash list */ struct sctp_bind_bucket *pp; /* hash list port iterator */ + union sctp_addr tmp; unsigned short snum; int ret; - /* NOTE: Remember to put this back to net order. */ - addr->v4.sin_port = ntohs(addr->v4.sin_port); - snum = addr->v4.sin_port; + flip_to_h(&tmp, addr); + snum = ntohs(addr->v4.sin_port); SCTP_DEBUG_PRINTK("sctp_get_port() begins, snum=%d\n", snum); sctp_local_bh_disable(); @@ -4744,7 +4740,7 @@ pp_found: if (reuse && sk2->sk_reuse) continue; - if (sctp_bind_addr_match(&ep2->base.bind_addr, addr, + if (sctp_bind_addr_match(&ep2->base.bind_addr, &tmp, sctp_sk(sk))) { ret = (long)sk2; goto fail_unlock; @@ -4784,7 +4780,6 @@ fail_unlock: fail: sctp_local_bh_enable(); - addr->v4.sin_port = htons(addr->v4.sin_port); return ret; } -- cgit v1.2.3-70-g09d2 From 09ef7fecea40c5e4c0dfe35bed3f0ed8da554cf5 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:04:10 -0800 Subject: [SCTP]: Beginning of conversion to net-endian for embedded sctp_addr. Part 1: rename sctp_chunk->source, sctp_sockaddr_entry->a, sctp_transport->ipaddr and sctp_transport->saddr (to ..._h) The next patch will reintroduce these fields and keep them as net-endian mirrors of the original (renamed) ones. Split in two patches to make sure that we hadn't forgotten any instanes. Later in the series we'll eliminate uses of host-endian variants (basically switching users to net-endian counterparts as we progress through that mess). Then host-endian ones will die. Other embedded host-endian sctp_addr will be easier to switch directly, so we leave them alone for now. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 8 ++++---- net/sctp/associola.c | 30 +++++++++++++++--------------- net/sctp/bind_addr.c | 22 +++++++++++----------- net/sctp/endpointola.c | 2 +- net/sctp/ipv6.c | 24 ++++++++++++------------ net/sctp/proc.c | 6 +++--- net/sctp/protocol.c | 20 ++++++++++---------- net/sctp/sm_make_chunk.c | 8 ++++---- net/sctp/sm_sideeffect.c | 8 ++++---- net/sctp/sm_statefuns.c | 8 ++++---- net/sctp/socket.c | 32 ++++++++++++++++---------------- net/sctp/transport.c | 14 +++++++------- 12 files changed, 91 insertions(+), 91 deletions(-) (limited to 'include') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 2aa61ac9a9f..953a73f9689 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -713,7 +713,7 @@ struct sctp_chunk { unsigned long sent_at; /* What is the origin IP address for this chunk? */ - union sctp_addr source; + union sctp_addr source_h; /* Destination address for this chunk. */ union sctp_addr dest; @@ -756,7 +756,7 @@ const union sctp_addr *sctp_source(const struct sctp_chunk *chunk); /* This is a structure for holding either an IPv6 or an IPv4 address. */ struct sctp_sockaddr_entry { struct list_head list; - union sctp_addr a; + union sctp_addr a_h; __u8 use_as_src; }; @@ -842,7 +842,7 @@ struct sctp_transport { int dead; /* This is the peer's IP address and port. */ - union sctp_addr ipaddr; + union sctp_addr ipaddr_h; /* These are the functions we call to handle LLP stuff. */ struct sctp_af *af_specific; @@ -900,7 +900,7 @@ struct sctp_transport { /* Destination */ struct dst_entry *dst; /* Source address. */ - union sctp_addr saddr; + union sctp_addr saddr_h; /* When was the last time(in jiffies) that a data packet was sent on * this transport? This is used to adjust the cwnd when the transport diff --git a/net/sctp/associola.c b/net/sctp/associola.c index 746b0b0f5ac..83318e72790 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -442,7 +442,7 @@ void sctp_assoc_set_primary(struct sctp_association *asoc, asoc->peer.primary_path = transport; /* Set a default msg_name for events. */ - memcpy(&asoc->peer.primary_addr, &transport->ipaddr, + memcpy(&asoc->peer.primary_addr, &transport->ipaddr_h, sizeof(union sctp_addr)); /* If the primary path is changing, assume that the @@ -487,8 +487,8 @@ void sctp_assoc_rm_peer(struct sctp_association *asoc, SCTP_DEBUG_PRINTK_IPADDR("sctp_assoc_rm_peer:association %p addr: ", " port: %d\n", asoc, - (&peer->ipaddr), - peer->ipaddr.v4.sin_port); + (&peer->ipaddr_h), + peer->ipaddr_h.v4.sin_port); /* If we are to remove the current retran_path, update it * to the next peer before removing this peer from the list. @@ -669,7 +669,7 @@ void sctp_assoc_del_peer(struct sctp_association *asoc, list_for_each_safe(pos, temp, &asoc->peer.transport_addr_list) { transport = list_entry(pos, struct sctp_transport, transports); - if (sctp_cmp_addr_exact(addr, &transport->ipaddr)) { + if (sctp_cmp_addr_exact(addr, &transport->ipaddr_h)) { /* Do book keeping for removing the peer and free it. */ sctp_assoc_rm_peer(asoc, transport); break; @@ -689,7 +689,7 @@ struct sctp_transport *sctp_assoc_lookup_paddr( list_for_each(pos, &asoc->peer.transport_addr_list) { t = list_entry(pos, struct sctp_transport, transports); - if (sctp_cmp_addr_exact(address, &t->ipaddr)) + if (sctp_cmp_addr_exact(address, &t->ipaddr_h)) return t; } @@ -733,7 +733,7 @@ void sctp_assoc_control_transport(struct sctp_association *asoc, * user. */ memset(&addr, 0, sizeof(struct sockaddr_storage)); - flip_to_n((union sctp_addr *)&addr, &transport->ipaddr); + flip_to_n((union sctp_addr *)&addr, &transport->ipaddr_h); event = sctp_ulpevent_make_peer_addr_change(asoc, &addr, 0, spc_state, error, GFP_ATOMIC); if (event) @@ -1043,8 +1043,8 @@ void sctp_assoc_update(struct sctp_association *asoc, /* Remove any peer addresses not present in the new association. */ list_for_each_safe(pos, temp, &asoc->peer.transport_addr_list) { trans = list_entry(pos, struct sctp_transport, transports); - if (!sctp_assoc_lookup_paddr(new, &trans->ipaddr)) - sctp_assoc_del_peer(asoc, &trans->ipaddr); + if (!sctp_assoc_lookup_paddr(new, &trans->ipaddr_h)) + sctp_assoc_del_peer(asoc, &trans->ipaddr_h); } /* If the case is A (association restart), use @@ -1067,8 +1067,8 @@ void sctp_assoc_update(struct sctp_association *asoc, list_for_each(pos, &new->peer.transport_addr_list) { trans = list_entry(pos, struct sctp_transport, transports); - if (!sctp_assoc_lookup_paddr(asoc, &trans->ipaddr)) - sctp_assoc_add_peer(asoc, &trans->ipaddr, + if (!sctp_assoc_lookup_paddr(asoc, &trans->ipaddr_h)) + sctp_assoc_add_peer(asoc, &trans->ipaddr_h, GFP_ATOMIC, trans->state); } @@ -1136,8 +1136,8 @@ void sctp_assoc_update_retran_path(struct sctp_association *asoc) " %p addr: ", " port: %d\n", asoc, - (&t->ipaddr), - t->ipaddr.v4.sin_port); + (&t->ipaddr_h), + t->ipaddr_h.v4.sin_port); } /* Choose the transport for sending a INIT packet. */ @@ -1161,8 +1161,8 @@ struct sctp_transport *sctp_assoc_choose_init_transport( " %p addr: ", " port: %d\n", asoc, - (&t->ipaddr), - t->ipaddr.v4.sin_port); + (&t->ipaddr_h), + t->ipaddr_h.v4.sin_port); return t; } @@ -1307,7 +1307,7 @@ int sctp_assoc_set_bind_addr_from_ep(struct sctp_association *asoc, /* Use scoping rules to determine the subset of addresses from * the endpoint. */ - scope = sctp_scope(&asoc->peer.active_path->ipaddr); + scope = sctp_scope(&asoc->peer.active_path->ipaddr_h); flags = (PF_INET6 == asoc->base.sk->sk_family) ? SCTP_ADDR6_ALLOWED : 0; if (asoc->peer.ipv4_address) flags |= SCTP_ADDR4_PEERSUPP; diff --git a/net/sctp/bind_addr.c b/net/sctp/bind_addr.c index 23b5b664a8a..b8115cb31ae 100644 --- a/net/sctp/bind_addr.c +++ b/net/sctp/bind_addr.c @@ -77,7 +77,7 @@ int sctp_bind_addr_copy(struct sctp_bind_addr *dest, /* Extract the addresses which are relevant for this scope. */ list_for_each(pos, &src->address_list) { addr = list_entry(pos, struct sctp_sockaddr_entry, list); - error = sctp_copy_one_addr(dest, &addr->a, scope, + error = sctp_copy_one_addr(dest, &addr->a_h, scope, gfp, flags); if (error < 0) goto out; @@ -91,7 +91,7 @@ int sctp_bind_addr_copy(struct sctp_bind_addr *dest, list_for_each(pos, &src->address_list) { addr = list_entry(pos, struct sctp_sockaddr_entry, list); - error = sctp_copy_one_addr(dest, &addr->a, + error = sctp_copy_one_addr(dest, &addr->a_h, SCTP_SCOPE_LINK, gfp, flags); if (error < 0) @@ -155,13 +155,13 @@ int sctp_add_bind_addr(struct sctp_bind_addr *bp, union sctp_addr *new, if (!addr) return -ENOMEM; - memcpy(&addr->a, new, sizeof(*new)); + memcpy(&addr->a_h, new, sizeof(*new)); /* Fix up the port if it has not yet been set. * Both v4 and v6 have the port at the same offset. */ - if (!addr->a.v4.sin_port) - addr->a.v4.sin_port = bp->port; + if (!addr->a_h.v4.sin_port) + addr->a_h.v4.sin_port = bp->port; addr->use_as_src = use_as_src; @@ -182,7 +182,7 @@ int sctp_del_bind_addr(struct sctp_bind_addr *bp, union sctp_addr *del_addr) list_for_each_safe(pos, temp, &bp->address_list) { addr = list_entry(pos, struct sctp_sockaddr_entry, list); - if (sctp_cmp_addr_exact(&addr->a, del_addr)) { + if (sctp_cmp_addr_exact(&addr->a_h, del_addr)) { /* Found the exact match. */ list_del(pos); kfree(addr); @@ -237,8 +237,8 @@ union sctp_params sctp_bind_addrs_to_raw(const struct sctp_bind_addr *bp, list_for_each(pos, &bp->address_list) { addr = list_entry(pos, struct sctp_sockaddr_entry, list); - af = sctp_get_af_specific(addr->a.v4.sin_family); - len = af->to_addr_param(&addr->a, &rawaddr); + af = sctp_get_af_specific(addr->a_h.v4.sin_family); + len = af->to_addr_param(&addr->a_h, &rawaddr); memcpy(addrparms.v, &rawaddr, len); addrparms.v += len; addrparms_len += len; @@ -305,7 +305,7 @@ int sctp_bind_addr_match(struct sctp_bind_addr *bp, list_for_each(pos, &bp->address_list) { laddr = list_entry(pos, struct sctp_sockaddr_entry, list); - if (opt->pf->cmp_addr(&laddr->a, addr, opt)) + if (opt->pf->cmp_addr(&laddr->a_h, addr, opt)) return 1; } @@ -339,13 +339,13 @@ union sctp_addr *sctp_find_unmatch_addr(struct sctp_bind_addr *bp, return NULL; flip_to_h(&tmp, addr); - if (opt->pf->cmp_addr(&laddr->a, &tmp, opt)) + if (opt->pf->cmp_addr(&laddr->a_h, &tmp, opt)) break; addr_buf += af->sockaddr_len; } if (i == addrcnt) - return &laddr->a; + return &laddr->a_h; } return NULL; diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c index 090f2b2a0ca..75904b037ec 100644 --- a/net/sctp/endpointola.c +++ b/net/sctp/endpointola.c @@ -305,7 +305,7 @@ int sctp_endpoint_is_peeled_off(struct sctp_endpoint *ep, bp = &ep->base.bind_addr; list_for_each(pos, &bp->address_list) { addr = list_entry(pos, struct sctp_sockaddr_entry, list); - if (sctp_has_association(&addr->a, paddr)) { + if (sctp_has_association(&addr->a_h, paddr)) { sctp_read_unlock(&ep->base.addr_lock); return 1; } diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 1f9f5ec0759..e533970fa33 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -161,17 +161,17 @@ static int sctp_v6_xmit(struct sk_buff *skb, struct sctp_transport *transport, /* Fill in the dest address from the route entry passed with the skb * and the source address from the transport. */ - ipv6_addr_copy(&fl.fl6_dst, &transport->ipaddr.v6.sin6_addr); - ipv6_addr_copy(&fl.fl6_src, &transport->saddr.v6.sin6_addr); + ipv6_addr_copy(&fl.fl6_dst, &transport->ipaddr_h.v6.sin6_addr); + ipv6_addr_copy(&fl.fl6_src, &transport->saddr_h.v6.sin6_addr); fl.fl6_flowlabel = np->flow_label; IP6_ECN_flow_xmit(sk, fl.fl6_flowlabel); if (ipv6_addr_type(&fl.fl6_src) & IPV6_ADDR_LINKLOCAL) - fl.oif = transport->saddr.v6.sin6_scope_id; + fl.oif = transport->saddr_h.v6.sin6_scope_id; else fl.oif = sk->sk_bound_dev_if; fl.fl_ip_sport = inet_sk(sk)->sport; - fl.fl_ip_dport = transport->ipaddr.v6.sin6_port; + fl.fl_ip_dport = transport->ipaddr_h.v6.sin6_port; if (np->opt && np->opt->srcrt) { struct rt0_hdr *rt0 = (struct rt0_hdr *) np->opt->srcrt; @@ -290,11 +290,11 @@ static void sctp_v6_get_saddr(struct sctp_association *asoc, list_for_each(pos, &bp->address_list) { laddr = list_entry(pos, struct sctp_sockaddr_entry, list); if ((laddr->use_as_src) && - (laddr->a.sa.sa_family == AF_INET6) && - (scope <= sctp_scope(&laddr->a))) { - bmatchlen = sctp_v6_addr_match_len(daddr, &laddr->a); + (laddr->a_h.sa.sa_family == AF_INET6) && + (scope <= sctp_scope(&laddr->a_h))) { + bmatchlen = sctp_v6_addr_match_len(daddr, &laddr->a_h); if (!baddr || (matchlen < bmatchlen)) { - baddr = &laddr->a; + baddr = &laddr->a_h; matchlen = bmatchlen; } } @@ -332,10 +332,10 @@ static void sctp_v6_copy_addrlist(struct list_head *addrlist, /* Add the address to the local list. */ addr = t_new(struct sctp_sockaddr_entry, GFP_ATOMIC); if (addr) { - addr->a.v6.sin6_family = AF_INET6; - addr->a.v6.sin6_port = 0; - addr->a.v6.sin6_addr = ifp->addr; - addr->a.v6.sin6_scope_id = dev->ifindex; + addr->a_h.v6.sin6_family = AF_INET6; + addr->a_h.v6.sin6_port = 0; + addr->a_h.v6.sin6_addr = ifp->addr; + addr->a_h.v6.sin6_scope_id = dev->ifindex; INIT_LIST_HEAD(&addr->list); list_add_tail(&addr->list, addrlist); } diff --git a/net/sctp/proc.c b/net/sctp/proc.c index 7f49e769080..47ccec565a2 100644 --- a/net/sctp/proc.c +++ b/net/sctp/proc.c @@ -155,12 +155,12 @@ static void sctp_seq_dump_local_addrs(struct seq_file *seq, struct sctp_ep_commo if (epb->type == SCTP_EP_TYPE_ASSOCIATION) { asoc = sctp_assoc(epb); peer = asoc->peer.primary_path; - primary = &peer->saddr; + primary = &peer->saddr_h; } list_for_each(pos, &epb->bind_addr.address_list) { laddr = list_entry(pos, struct sctp_sockaddr_entry, list); - addr = (union sctp_addr *)&laddr->a; + addr = (union sctp_addr *)&laddr->a_h; af = sctp_get_af_specific(addr->sa.sa_family); if (primary && af->cmp_addr(addr, primary)) { seq_printf(seq, "*"); @@ -180,7 +180,7 @@ static void sctp_seq_dump_remote_addrs(struct seq_file *seq, struct sctp_associa primary = &(assoc->peer.primary_addr); list_for_each(pos, &assoc->peer.transport_addr_list) { transport = list_entry(pos, struct sctp_transport, transports); - addr = (union sctp_addr *)&transport->ipaddr; + addr = (union sctp_addr *)&transport->ipaddr_h; af = sctp_get_af_specific(addr->sa.sa_family); if (af->cmp_addr(addr, primary)) { seq_printf(seq, "*"); diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 4310e7fac44..20883ffd51d 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -150,9 +150,9 @@ static void sctp_v4_copy_addrlist(struct list_head *addrlist, /* Add the address to the local list. */ addr = t_new(struct sctp_sockaddr_entry, GFP_ATOMIC); if (addr) { - addr->a.v4.sin_family = AF_INET; - addr->a.v4.sin_port = 0; - addr->a.v4.sin_addr.s_addr = ifa->ifa_local; + addr->a_h.v4.sin_family = AF_INET; + addr->a_h.v4.sin_port = 0; + addr->a_h.v4.sin_addr.s_addr = ifa->ifa_local; list_add_tail(&addr->list, addrlist); } } @@ -223,17 +223,17 @@ int sctp_copy_local_addr_list(struct sctp_bind_addr *bp, sctp_scope_t scope, sctp_spin_lock_irqsave(&sctp_local_addr_lock, flags); list_for_each(pos, &sctp_local_addr_list) { addr = list_entry(pos, struct sctp_sockaddr_entry, list); - if (sctp_in_scope(&addr->a, scope)) { + if (sctp_in_scope(&addr->a_h, scope)) { /* Now that the address is in scope, check to see if * the address type is really supported by the local * sock as well as the remote peer. */ - if ((((AF_INET == addr->a.sa.sa_family) && + if ((((AF_INET == addr->a_h.sa.sa_family) && (copy_flags & SCTP_ADDR4_PEERSUPP))) || - (((AF_INET6 == addr->a.sa.sa_family) && + (((AF_INET6 == addr->a_h.sa.sa_family) && (copy_flags & SCTP_ADDR6_ALLOWED) && (copy_flags & SCTP_ADDR6_PEERSUPP)))) { - error = sctp_add_bind_addr(bp, &addr->a, 1, + error = sctp_add_bind_addr(bp, &addr->a_h, 1, GFP_ATOMIC); if (error) goto end_copy; @@ -482,7 +482,7 @@ static struct dst_entry *sctp_v4_get_dst(struct sctp_association *asoc, if (!laddr->use_as_src) continue; sctp_v4_dst_saddr(&dst_saddr, dst, bp->port); - if (sctp_v4_cmp_addr(&dst_saddr, &laddr->a)) + if (sctp_v4_cmp_addr(&dst_saddr, &laddr->a_h)) goto out_unlock; } sctp_read_unlock(addr_lock); @@ -502,8 +502,8 @@ static struct dst_entry *sctp_v4_get_dst(struct sctp_association *asoc, laddr = list_entry(pos, struct sctp_sockaddr_entry, list); if ((laddr->use_as_src) && - (AF_INET == laddr->a.sa.sa_family)) { - fl.fl4_src = laddr->a.v4.sin_addr.s_addr; + (AF_INET == laddr->a_h.sa.sa_family)) { + fl.fl4_src = laddr->a_h.v4.sin_addr.s_addr; if (!ip_route_output_key(&rt, &fl)) { dst = &rt->u.dst; goto out_unlock; diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index ccbba6b24d1..13109cfc3ae 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -1031,7 +1031,7 @@ nodata: void sctp_init_addrs(struct sctp_chunk *chunk, union sctp_addr *src, union sctp_addr *dest) { - memcpy(&chunk->source, src, sizeof(union sctp_addr)); + memcpy(&chunk->source_h, src, sizeof(union sctp_addr)); memcpy(&chunk->dest, dest, sizeof(union sctp_addr)); } @@ -1040,10 +1040,10 @@ const union sctp_addr *sctp_source(const struct sctp_chunk *chunk) { /* If we have a known transport, use that. */ if (chunk->transport) { - return &chunk->transport->ipaddr; + return &chunk->transport->ipaddr_h; } else { /* Otherwise, extract it from the IP header. */ - return &chunk->source; + return &chunk->source_h; } } @@ -2594,7 +2594,7 @@ static int sctp_asconf_param_success(struct sctp_association *asoc, sctp_write_lock(&asoc->base.addr_lock); list_for_each(pos, &bp->address_list) { saddr = list_entry(pos, struct sctp_sockaddr_entry, list); - if (sctp_cmp_addr_exact(&saddr->a, &addr)) + if (sctp_cmp_addr_exact(&saddr->a_h, &addr)) saddr->use_as_src = 1; } sctp_write_unlock(&asoc->base.addr_lock); diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index 2423b8ef9ce..e1c5cddf32d 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -441,8 +441,8 @@ static void sctp_do_8_2_transport_strike(struct sctp_association *asoc, SCTP_DEBUG_PRINTK_IPADDR("transport_strike:association %p", " transport IP: port:%d failed.\n", asoc, - (&transport->ipaddr), - transport->ipaddr.v4.sin_port); + (&transport->ipaddr_h), + transport->ipaddr_h.v4.sin_port); sctp_assoc_control_transport(asoc, transport, SCTP_TRANSPORT_DOWN, SCTP_FAILED_THRESHOLD); @@ -833,9 +833,9 @@ static void sctp_cmd_del_non_primary(struct sctp_association *asoc) list_for_each_safe(pos, temp, &asoc->peer.transport_addr_list) { t = list_entry(pos, struct sctp_transport, transports); - if (!sctp_cmp_addr_exact(&t->ipaddr, + if (!sctp_cmp_addr_exact(&t->ipaddr_h, &asoc->peer.primary_addr)) { - sctp_assoc_del_peer(asoc, &t->ipaddr); + sctp_assoc_del_peer(asoc, &t->ipaddr_h); } } diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index b70e3258b6d..f4607276754 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -852,7 +852,7 @@ static sctp_disposition_t sctp_sf_heartbeat(const struct sctp_endpoint *ep, hbinfo.param_hdr.type = SCTP_PARAM_HEARTBEAT_INFO; hbinfo.param_hdr.length = htons(sizeof(sctp_sender_hb_info_t)); - hbinfo.daddr = transport->ipaddr; + hbinfo.daddr = transport->ipaddr_h; hbinfo.sent_at = jiffies; hbinfo.hb_nonce = transport->hb_nonce; @@ -1167,8 +1167,8 @@ static int sctp_sf_check_restart_addrs(const struct sctp_association *new_asoc, list_for_each(pos2, &asoc->peer.transport_addr_list) { addr = list_entry(pos2, struct sctp_transport, transports); - if (sctp_cmp_addr_exact(&new_addr->ipaddr, - &addr->ipaddr)) { + if (sctp_cmp_addr_exact(&new_addr->ipaddr_h, + &addr->ipaddr_h)) { found = 1; break; } @@ -1179,7 +1179,7 @@ static int sctp_sf_check_restart_addrs(const struct sctp_association *new_asoc, /* If a new address was added, ABORT the sender. */ if (!found && new_addr) { - sctp_sf_send_restart_abort(&new_addr->ipaddr, init, commands); + sctp_sf_send_restart_abort(&new_addr->ipaddr_h, init, commands); } /* Return success if all addresses were found. */ diff --git a/net/sctp/socket.c b/net/sctp/socket.c index b6e038a387c..459f32c5c3a 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -551,7 +551,7 @@ static int sctp_send_asconf_add_ip(struct sock *sk, laddr = list_entry(p, struct sctp_sockaddr_entry, list); sctp_read_unlock(&asoc->base.addr_lock); - chunk = sctp_make_asconf_update_ip(asoc, &laddr->a, addrs, + chunk = sctp_make_asconf_update_ip(asoc, &laddr->a_h, addrs, addrcnt, SCTP_PARAM_ADD_IP); if (!chunk) { retval = -ENOMEM; @@ -779,7 +779,7 @@ static int sctp_send_asconf_del_ip(struct sock *sk, saddr = list_entry(pos1, struct sctp_sockaddr_entry, list); - if (sctp_cmp_addr_exact(&saddr->a, &saveaddr)) + if (sctp_cmp_addr_exact(&saddr->a_h, &saveaddr)) saddr->use_as_src = 0; } addr_buf += af->sockaddr_len; @@ -3218,7 +3218,7 @@ static int sctp_getsockopt_sctp_status(struct sock *sk, int len, status.sstat_fragmentation_point = asoc->frag_point; status.sstat_primary.spinfo_assoc_id = sctp_assoc2id(transport->asoc); flip_to_n((union sctp_addr *)&status.sstat_primary.spinfo_address, - &transport->ipaddr); + &transport->ipaddr_h); /* Map ipv4 address into v4-mapped-on-v6 address. */ sctp_get_pf_specific(sk->sk_family)->addr_v4map(sctp_sk(sk), (union sctp_addr *)&status.sstat_primary.spinfo_address); @@ -3767,7 +3767,7 @@ static int sctp_getsockopt_peer_addrs_old(struct sock *sk, int len, to = (void __user *)getaddrs.addrs; list_for_each(pos, &asoc->peer.transport_addr_list) { from = list_entry(pos, struct sctp_transport, transports); - memcpy(&temp, &from->ipaddr, sizeof(temp)); + memcpy(&temp, &from->ipaddr_h, sizeof(temp)); sctp_get_pf_specific(sk->sk_family)->addr_v4map(sp, &temp); addrlen = sctp_get_af_specific(sk->sk_family)->sockaddr_len; temp.v4.sin_port = htons(temp.v4.sin_port); @@ -3816,7 +3816,7 @@ static int sctp_getsockopt_peer_addrs(struct sock *sk, int len, list_for_each(pos, &asoc->peer.transport_addr_list) { from = list_entry(pos, struct sctp_transport, transports); - memcpy(&temp, &from->ipaddr, sizeof(temp)); + memcpy(&temp, &from->ipaddr_h, sizeof(temp)); sctp_get_pf_specific(sk->sk_family)->addr_v4map(sp, &temp); addrlen = sctp_get_af_specific(sk->sk_family)->sockaddr_len; if(space_left < addrlen) @@ -3882,14 +3882,14 @@ static int sctp_getsockopt_local_addrs_num_old(struct sock *sk, int len, if (sctp_list_single_entry(&bp->address_list)) { addr = list_entry(bp->address_list.next, struct sctp_sockaddr_entry, list); - if (sctp_is_any(&addr->a)) { + if (sctp_is_any(&addr->a_h)) { sctp_spin_lock_irqsave(&sctp_local_addr_lock, flags); list_for_each(pos, &sctp_local_addr_list) { addr = list_entry(pos, struct sctp_sockaddr_entry, list); if ((PF_INET == sk->sk_family) && - (AF_INET6 == addr->a.sa.sa_family)) + (AF_INET6 == addr->a_h.sa.sa_family)) continue; cnt++; } @@ -3927,9 +3927,9 @@ static int sctp_copy_laddrs_to_user_old(struct sock *sk, __u16 port, int max_add list_for_each(pos, &sctp_local_addr_list) { addr = list_entry(pos, struct sctp_sockaddr_entry, list); if ((PF_INET == sk->sk_family) && - (AF_INET6 == addr->a.sa.sa_family)) + (AF_INET6 == addr->a_h.sa.sa_family)) continue; - memcpy(&temp, &addr->a, sizeof(temp)); + memcpy(&temp, &addr->a_h, sizeof(temp)); sctp_get_pf_specific(sk->sk_family)->addr_v4map(sctp_sk(sk), &temp); addrlen = sctp_get_af_specific(temp.sa.sa_family)->sockaddr_len; @@ -3962,9 +3962,9 @@ static int sctp_copy_laddrs_to_user(struct sock *sk, __u16 port, list_for_each(pos, &sctp_local_addr_list) { addr = list_entry(pos, struct sctp_sockaddr_entry, list); if ((PF_INET == sk->sk_family) && - (AF_INET6 == addr->a.sa.sa_family)) + (AF_INET6 == addr->a_h.sa.sa_family)) continue; - memcpy(&temp, &addr->a, sizeof(temp)); + memcpy(&temp, &addr->a_h, sizeof(temp)); sctp_get_pf_specific(sk->sk_family)->addr_v4map(sctp_sk(sk), &temp); addrlen = sctp_get_af_specific(temp.sa.sa_family)->sockaddr_len; @@ -4038,7 +4038,7 @@ static int sctp_getsockopt_local_addrs_old(struct sock *sk, int len, if (sctp_list_single_entry(&bp->address_list)) { addr = list_entry(bp->address_list.next, struct sctp_sockaddr_entry, list); - if (sctp_is_any(&addr->a)) { + if (sctp_is_any(&addr->a_h)) { cnt = sctp_copy_laddrs_to_user_old(sk, bp->port, getaddrs.addr_num, to); @@ -4052,7 +4052,7 @@ static int sctp_getsockopt_local_addrs_old(struct sock *sk, int len, list_for_each(pos, &bp->address_list) { addr = list_entry(pos, struct sctp_sockaddr_entry, list); - memcpy(&temp, &addr->a, sizeof(temp)); + memcpy(&temp, &addr->a_h, sizeof(temp)); sctp_get_pf_specific(sk->sk_family)->addr_v4map(sp, &temp); addrlen = sctp_get_af_specific(temp.sa.sa_family)->sockaddr_len; temp.v4.sin_port = htons(temp.v4.sin_port); @@ -4128,7 +4128,7 @@ static int sctp_getsockopt_local_addrs(struct sock *sk, int len, if (sctp_list_single_entry(&bp->address_list)) { addr = list_entry(bp->address_list.next, struct sctp_sockaddr_entry, list); - if (sctp_is_any(&addr->a)) { + if (sctp_is_any(&addr->a_h)) { cnt = sctp_copy_laddrs_to_user(sk, bp->port, &to, space_left); if (cnt < 0) { @@ -4141,7 +4141,7 @@ static int sctp_getsockopt_local_addrs(struct sock *sk, int len, list_for_each(pos, &bp->address_list) { addr = list_entry(pos, struct sctp_sockaddr_entry, list); - memcpy(&temp, &addr->a, sizeof(temp)); + memcpy(&temp, &addr->a_h, sizeof(temp)); sctp_get_pf_specific(sk->sk_family)->addr_v4map(sp, &temp); addrlen = sctp_get_af_specific(temp.sa.sa_family)->sockaddr_len; if(space_left < addrlen) @@ -4195,7 +4195,7 @@ static int sctp_getsockopt_primary_addr(struct sock *sk, int len, return -ENOTCONN; flip_to_n((union sctp_addr *)&prim.ssp_addr, - &asoc->peer.primary_path->ipaddr); + &asoc->peer.primary_path->ipaddr_h); sctp_get_pf_specific(sk->sk_family)->addr_v4map(sp, (union sctp_addr *)&prim.ssp_addr); diff --git a/net/sctp/transport.c b/net/sctp/transport.c index 3e5936a5f67..1f56c9d9573 100644 --- a/net/sctp/transport.c +++ b/net/sctp/transport.c @@ -61,12 +61,12 @@ static struct sctp_transport *sctp_transport_init(struct sctp_transport *peer, gfp_t gfp) { /* Copy in the address. */ - peer->ipaddr = *addr; + peer->ipaddr_h = *addr; peer->af_specific = sctp_get_af_specific(addr->sa.sa_family); peer->asoc = NULL; peer->dst = NULL; - memset(&peer->saddr, 0, sizeof(union sctp_addr)); + memset(&peer->saddr_h, 0, sizeof(union sctp_addr)); /* From 6.3.1 RTO Calculation: * @@ -232,7 +232,7 @@ void sctp_transport_pmtu(struct sctp_transport *transport) { struct dst_entry *dst; - dst = transport->af_specific->get_dst(NULL, &transport->ipaddr, NULL); + dst = transport->af_specific->get_dst(NULL, &transport->ipaddr_h, NULL); if (dst) { transport->pathmtu = dst_mtu(dst); @@ -249,15 +249,15 @@ void sctp_transport_route(struct sctp_transport *transport, { struct sctp_association *asoc = transport->asoc; struct sctp_af *af = transport->af_specific; - union sctp_addr *daddr = &transport->ipaddr; + union sctp_addr *daddr = &transport->ipaddr_h; struct dst_entry *dst; dst = af->get_dst(asoc, daddr, saddr); if (saddr) - memcpy(&transport->saddr, saddr, sizeof(union sctp_addr)); + memcpy(&transport->saddr_h, saddr, sizeof(union sctp_addr)); else - af->get_saddr(asoc, dst, daddr, &transport->saddr); + af->get_saddr(asoc, dst, daddr, &transport->saddr_h); transport->dst = dst; if ((transport->param_flags & SPP_PMTUD_DISABLE) && transport->pathmtu) { @@ -270,7 +270,7 @@ void sctp_transport_route(struct sctp_transport *transport, * association's active path for getsockname(). */ if (asoc && (transport == asoc->peer.active_path)) - opt->pf->af->to_sk_saddr(&transport->saddr, + opt->pf->af->to_sk_saddr(&transport->saddr_h, asoc->base.sk); } else transport->pathmtu = SCTP_DEFAULT_MAXSEGMENT; -- cgit v1.2.3-70-g09d2 From 2a6fd78adec062f16f8662563115679e669efaca Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:04:42 -0800 Subject: [SCTP] embedded sctp_addr: net-endian mirrors Add sctp_chunk->source, sctp_sockaddr_entry->a, sctp_transport->ipaddr and sctp_transport->saddr, maintain them as net-endian mirrors of their host-endian counterparts. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 4 ++++ net/sctp/bind_addr.c | 2 ++ net/sctp/ipv6.c | 9 +++++---- net/sctp/protocol.c | 7 ++++--- net/sctp/sm_make_chunk.c | 1 + net/sctp/transport.c | 3 +++ 6 files changed, 19 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 953a73f9689..2c35f2fd560 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -713,6 +713,7 @@ struct sctp_chunk { unsigned long sent_at; /* What is the origin IP address for this chunk? */ + union sctp_addr source; union sctp_addr source_h; /* Destination address for this chunk. */ union sctp_addr dest; @@ -756,6 +757,7 @@ const union sctp_addr *sctp_source(const struct sctp_chunk *chunk); /* This is a structure for holding either an IPv6 or an IPv4 address. */ struct sctp_sockaddr_entry { struct list_head list; + union sctp_addr a; union sctp_addr a_h; __u8 use_as_src; }; @@ -842,6 +844,7 @@ struct sctp_transport { int dead; /* This is the peer's IP address and port. */ + union sctp_addr ipaddr; union sctp_addr ipaddr_h; /* These are the functions we call to handle LLP stuff. */ @@ -900,6 +903,7 @@ struct sctp_transport { /* Destination */ struct dst_entry *dst; /* Source address. */ + union sctp_addr saddr; union sctp_addr saddr_h; /* When was the last time(in jiffies) that a data packet was sent on diff --git a/net/sctp/bind_addr.c b/net/sctp/bind_addr.c index b8115cb31ae..2b36e423817 100644 --- a/net/sctp/bind_addr.c +++ b/net/sctp/bind_addr.c @@ -163,6 +163,8 @@ int sctp_add_bind_addr(struct sctp_bind_addr *bp, union sctp_addr *new, if (!addr->a_h.v4.sin_port) addr->a_h.v4.sin_port = bp->port; + flip_to_n(&addr->a, &addr->a_h); + addr->use_as_src = use_as_src; INIT_LIST_HEAD(&addr->list); diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index e533970fa33..8ebd177296b 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -332,10 +332,11 @@ static void sctp_v6_copy_addrlist(struct list_head *addrlist, /* Add the address to the local list. */ addr = t_new(struct sctp_sockaddr_entry, GFP_ATOMIC); if (addr) { - addr->a_h.v6.sin6_family = AF_INET6; - addr->a_h.v6.sin6_port = 0; - addr->a_h.v6.sin6_addr = ifp->addr; - addr->a_h.v6.sin6_scope_id = dev->ifindex; + addr->a.v6.sin6_family = AF_INET6; + addr->a.v6.sin6_port = 0; + addr->a.v6.sin6_addr = ifp->addr; + addr->a.v6.sin6_scope_id = dev->ifindex; + addr->a_h = addr->a; INIT_LIST_HEAD(&addr->list); list_add_tail(&addr->list, addrlist); } diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 20883ffd51d..36327404503 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -150,9 +150,10 @@ static void sctp_v4_copy_addrlist(struct list_head *addrlist, /* Add the address to the local list. */ addr = t_new(struct sctp_sockaddr_entry, GFP_ATOMIC); if (addr) { - addr->a_h.v4.sin_family = AF_INET; - addr->a_h.v4.sin_port = 0; - addr->a_h.v4.sin_addr.s_addr = ifa->ifa_local; + addr->a.v4.sin_family = AF_INET; + addr->a.v4.sin_port = 0; + addr->a.v4.sin_addr.s_addr = ifa->ifa_local; + addr->a_h = addr->a; list_add_tail(&addr->list, addrlist); } } diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 13109cfc3ae..b8e0f72b319 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -1032,6 +1032,7 @@ void sctp_init_addrs(struct sctp_chunk *chunk, union sctp_addr *src, union sctp_addr *dest) { memcpy(&chunk->source_h, src, sizeof(union sctp_addr)); + flip_to_n(&chunk->source, &chunk->source_h); memcpy(&chunk->dest, dest, sizeof(union sctp_addr)); } diff --git a/net/sctp/transport.c b/net/sctp/transport.c index 1f56c9d9573..bacd09448b8 100644 --- a/net/sctp/transport.c +++ b/net/sctp/transport.c @@ -62,10 +62,12 @@ static struct sctp_transport *sctp_transport_init(struct sctp_transport *peer, { /* Copy in the address. */ peer->ipaddr_h = *addr; + flip_to_n(&peer->ipaddr, &peer->ipaddr_h); peer->af_specific = sctp_get_af_specific(addr->sa.sa_family); peer->asoc = NULL; peer->dst = NULL; + memset(&peer->saddr, 0, sizeof(union sctp_addr)); memset(&peer->saddr_h, 0, sizeof(union sctp_addr)); /* From 6.3.1 RTO Calculation: @@ -258,6 +260,7 @@ void sctp_transport_route(struct sctp_transport *transport, memcpy(&transport->saddr_h, saddr, sizeof(union sctp_addr)); else af->get_saddr(asoc, dst, daddr, &transport->saddr_h); + flip_to_n(&transport->saddr, &transport->saddr_h); transport->dst = dst; if ((transport->param_flags & SPP_PMTUD_DISABLE) && transport->pathmtu) { -- cgit v1.2.3-70-g09d2 From 854d43a465cc8ba8e501320b3bc27359d909da2f Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:06:24 -0800 Subject: [SCTP]: Annotate ->dst_saddr() switched to taking a pointer to net-endian sctp_addr and a net-endian port number. Instances and callers adjusted; interestingly enough, the only calls are direct calls of specific instances - the method is not used at all. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 2 +- net/sctp/ipv6.c | 2 +- net/sctp/protocol.c | 8 +++----- 3 files changed, 5 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 2c35f2fd560..1616a38a0be 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -559,7 +559,7 @@ struct sctp_af { struct net_device *); void (*dst_saddr) (union sctp_addr *saddr, struct dst_entry *dst, - unsigned short port); + __be16 port); int (*cmp_addr) (const union sctp_addr *addr1, const union sctp_addr *addr2); void (*addr_copy) (union sctp_addr *dst, diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 8ebd177296b..6139d1d964e 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -434,7 +434,7 @@ static int sctp_v6_to_addr_param(const union sctp_addr *addr, /* Initialize a sctp_addr from a dst_entry. */ static void sctp_v6_dst_saddr(union sctp_addr *addr, struct dst_entry *dst, - unsigned short port) + __be16 port) { struct rt6_info *rt = (struct rt6_info *)dst; addr->sa.sa_family = AF_INET6; diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 2db140e901d..d1fbd1f5c6d 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -316,7 +316,7 @@ static int sctp_v4_to_addr_param(const union sctp_addr *addr, /* Initialize a sctp_addr from a dst_entry. */ static void sctp_v4_dst_saddr(union sctp_addr *saddr, struct dst_entry *dst, - unsigned short port) + __be16 port) { struct rtable *rt = (struct rtable *)dst; saddr->v4.sin_family = AF_INET; @@ -478,14 +478,12 @@ static struct dst_entry *sctp_v4_get_dst(struct sctp_association *asoc, */ sctp_read_lock(addr_lock); list_for_each(pos, &bp->address_list) { - union sctp_addr tmp; laddr = list_entry(pos, struct sctp_sockaddr_entry, list); if (!laddr->use_as_src) continue; - sctp_v4_dst_saddr(&dst_saddr, dst, bp->port); - flip_to_n(&tmp, &dst_saddr); - if (sctp_v4_cmp_addr(&tmp, &laddr->a)) + sctp_v4_dst_saddr(&dst_saddr, dst, htons(bp->port)); + if (sctp_v4_cmp_addr(&dst_saddr, &laddr->a)) goto out_unlock; } sctp_read_unlock(addr_lock); -- cgit v1.2.3-70-g09d2 From dd86d136f9feb72c52a5b07707affe80edbc8dda Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:11:13 -0800 Subject: [SCTP]: Switch ->from_addr_param() to net-endian. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 2 +- net/sctp/bind_addr.c | 6 ++---- net/sctp/input.c | 6 ++---- net/sctp/ipv6.c | 2 +- net/sctp/protocol.c | 2 +- net/sctp/sm_make_chunk.c | 29 ++++++++++++----------------- 6 files changed, 19 insertions(+), 28 deletions(-) (limited to 'include') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 1616a38a0be..8f6619dbb66 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -575,7 +575,7 @@ struct sctp_af { struct sock *sk); void (*from_addr_param) (union sctp_addr *, union sctp_addr_param *, - __u16 port, int iif); + __be16 port, int iif); int (*to_addr_param) (const union sctp_addr *, union sctp_addr_param *); int (*addr_valid) (union sctp_addr *, diff --git a/net/sctp/bind_addr.c b/net/sctp/bind_addr.c index eb3a5e25777..fd0f812f8a4 100644 --- a/net/sctp/bind_addr.c +++ b/net/sctp/bind_addr.c @@ -264,7 +264,6 @@ int sctp_raw_to_bind_addrs(struct sctp_bind_addr *bp, __u8 *raw_addr_list, int retval = 0; int len; struct sctp_af *af; - union sctp_addr tmp; /* Convert the raw address to standard address format */ while (addrs_len) { @@ -278,9 +277,8 @@ int sctp_raw_to_bind_addrs(struct sctp_bind_addr *bp, __u8 *raw_addr_list, break; } - af->from_addr_param(&addr, rawaddr, port, 0); - flip_to_n(&tmp, &addr); - retval = sctp_add_bind_addr(bp, &tmp, 1, gfp); + af->from_addr_param(&addr, rawaddr, htons(port), 0); + retval = sctp_add_bind_addr(bp, &addr, 1, gfp); if (retval) { /* Can't finish building the list, clean up. */ sctp_bind_addr_clean(bp); diff --git a/net/sctp/input.c b/net/sctp/input.c index 3b4a552c9f1..2d660f8d94f 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -914,7 +914,6 @@ static struct sctp_association *__sctp_rcv_init_lookup(struct sk_buff *skb, sctp_init_chunk_t *init; struct sctp_transport *transport; struct sctp_af *af; - union sctp_addr tmp2; ch = (sctp_chunkhdr_t *) skb->data; @@ -961,10 +960,9 @@ static struct sctp_association *__sctp_rcv_init_lookup(struct sk_buff *skb, if (!af) continue; - af->from_addr_param(paddr, params.addr, ntohs(sh->source), 0); - flip_to_n(&tmp2, paddr); + af->from_addr_param(paddr, params.addr, sh->source, 0); - asoc = __sctp_lookup_association(laddr, &tmp2, &transport); + asoc = __sctp_lookup_association(laddr, paddr, &transport); if (asoc) return asoc; } diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index d6b585b85cb..566cdc9977c 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -408,7 +408,7 @@ static void sctp_v6_to_sk_daddr(union sctp_addr *addr, struct sock *sk) /* Initialize a sctp_addr from an address parameter. */ static void sctp_v6_from_addr_param(union sctp_addr *addr, union sctp_addr_param *param, - __u16 port, int iif) + __be16 port, int iif) { addr->v6.sin6_family = AF_INET6; addr->v6.sin6_port = port; diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index a6bcbf560e3..87d36f440d1 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -292,7 +292,7 @@ static void sctp_v4_to_sk_daddr(union sctp_addr *addr, struct sock *sk) /* Initialize a sctp_addr from an address parameter. */ static void sctp_v4_from_addr_param(union sctp_addr *addr, union sctp_addr_param *param, - __u16 port, int iif) + __be16 port, int iif) { addr->v4.sin_family = AF_INET; addr->v4.sin_port = port; diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 46cfcca5e41..b367ad59ed4 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -2021,7 +2021,6 @@ static int sctp_process_param(struct sctp_association *asoc, sctp_scope_t scope; time_t stale; struct sctp_af *af; - union sctp_addr tmp; /* We maintain all INIT parameters in network byte order all the * time. This allows us to not worry about whether the parameters @@ -2034,11 +2033,10 @@ static int sctp_process_param(struct sctp_association *asoc, /* Fall through. */ case SCTP_PARAM_IPV4_ADDRESS: af = sctp_get_af_specific(param_type2af(param.p->type)); - af->from_addr_param(&addr, param.addr, asoc->peer.port, 0); - flip_to_n(&tmp, &addr); + af->from_addr_param(&addr, param.addr, htons(asoc->peer.port), 0); scope = sctp_scope(peer_addr); - if (sctp_in_scope(&tmp, scope)) - if (!sctp_assoc_add_peer(asoc, &tmp, gfp, SCTP_UNCONFIRMED)) + if (sctp_in_scope(&addr, scope)) + if (!sctp_assoc_add_peer(asoc, &addr, gfp, SCTP_UNCONFIRMED)) return 0; break; @@ -2421,7 +2419,7 @@ static __be16 sctp_process_asconf_param(struct sctp_association *asoc, union sctp_addr addr; struct list_head *pos; union sctp_addr_param *addr_param; - union sctp_addr tmp, tmp_addr; + union sctp_addr tmp; addr_param = (union sctp_addr_param *) ((void *)asconf_param + sizeof(sctp_addip_param_t)); @@ -2430,8 +2428,7 @@ static __be16 sctp_process_asconf_param(struct sctp_association *asoc, if (unlikely(!af)) return SCTP_ERROR_INV_PARAM; - af->from_addr_param(&addr, addr_param, asoc->peer.port, 0); - flip_to_n(&tmp_addr, &addr); + af->from_addr_param(&addr, addr_param, htons(asoc->peer.port), 0); switch (asconf_param->param_hdr.type) { case SCTP_PARAM_ADD_IP: /* ADDIP 4.3 D9) If an endpoint receives an ADD IP address @@ -2441,7 +2438,7 @@ static __be16 sctp_process_asconf_param(struct sctp_association *asoc, * Due to Resource Shortage'. */ - peer = sctp_assoc_add_peer(asoc, &tmp_addr, GFP_ATOMIC, SCTP_UNCONFIRMED); + peer = sctp_assoc_add_peer(asoc, &addr, GFP_ATOMIC, SCTP_UNCONFIRMED); if (!peer) return SCTP_ERROR_RSRC_LOW; @@ -2467,13 +2464,13 @@ static __be16 sctp_process_asconf_param(struct sctp_association *asoc, * Delete Source IP Address' */ flip_to_n(&tmp, sctp_source(asconf)); - if (sctp_cmp_addr_exact(&tmp, &tmp_addr)) + if (sctp_cmp_addr_exact(&tmp, &addr)) return SCTP_ERROR_DEL_SRC_IP; - sctp_assoc_del_peer(asoc, &tmp_addr); + sctp_assoc_del_peer(asoc, &addr); break; case SCTP_PARAM_SET_PRIMARY: - peer = sctp_assoc_lookup_paddr(asoc, &tmp_addr); + peer = sctp_assoc_lookup_paddr(asoc, &addr); if (!peer) return SCTP_ERROR_INV_PARAM; @@ -2591,15 +2588,13 @@ static int sctp_asconf_param_success(struct sctp_association *asoc, struct sctp_transport *transport; struct sctp_sockaddr_entry *saddr; int retval = 0; - union sctp_addr tmp; addr_param = (union sctp_addr_param *) ((void *)asconf_param + sizeof(sctp_addip_param_t)); /* We have checked the packet before, so we do not check again. */ af = sctp_get_af_specific(param_type2af(addr_param->v4.param_hdr.type)); - af->from_addr_param(&addr, addr_param, bp->port, 0); - flip_to_n(&tmp, &addr); + af->from_addr_param(&addr, addr_param, htons(bp->port), 0); switch (asconf_param->param_hdr.type) { case SCTP_PARAM_ADD_IP: @@ -2607,7 +2602,7 @@ static int sctp_asconf_param_success(struct sctp_association *asoc, sctp_write_lock(&asoc->base.addr_lock); list_for_each(pos, &bp->address_list) { saddr = list_entry(pos, struct sctp_sockaddr_entry, list); - if (sctp_cmp_addr_exact(&saddr->a, &tmp)) + if (sctp_cmp_addr_exact(&saddr->a, &addr)) saddr->use_as_src = 1; } sctp_write_unlock(&asoc->base.addr_lock); @@ -2616,7 +2611,7 @@ static int sctp_asconf_param_success(struct sctp_association *asoc, case SCTP_PARAM_DEL_IP: sctp_local_bh_disable(); sctp_write_lock(&asoc->base.addr_lock); - retval = sctp_del_bind_addr(bp, &tmp); + retval = sctp_del_bind_addr(bp, &addr); sctp_write_unlock(&asoc->base.addr_lock); sctp_local_bh_enable(); list_for_each(pos, &asoc->peer.transport_addr_list) { -- cgit v1.2.3-70-g09d2 From a926626893aca20567f27af1c5edc830e1c51b2b Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:13:58 -0800 Subject: [SCTP]: Switch all remaining users of ->saddr_h to ->saddr. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 1 - net/sctp/ipv6.c | 4 ++-- net/sctp/transport.c | 4 +--- 3 files changed, 3 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 8f6619dbb66..27519562014 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -904,7 +904,6 @@ struct sctp_transport { struct dst_entry *dst; /* Source address. */ union sctp_addr saddr; - union sctp_addr saddr_h; /* When was the last time(in jiffies) that a data packet was sent on * this transport? This is used to adjust the cwnd when the transport diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 7f4c5bafea1..3dc8ed14352 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -162,12 +162,12 @@ static int sctp_v6_xmit(struct sk_buff *skb, struct sctp_transport *transport, * and the source address from the transport. */ ipv6_addr_copy(&fl.fl6_dst, &transport->ipaddr_h.v6.sin6_addr); - ipv6_addr_copy(&fl.fl6_src, &transport->saddr_h.v6.sin6_addr); + ipv6_addr_copy(&fl.fl6_src, &transport->saddr.v6.sin6_addr); fl.fl6_flowlabel = np->flow_label; IP6_ECN_flow_xmit(sk, fl.fl6_flowlabel); if (ipv6_addr_type(&fl.fl6_src) & IPV6_ADDR_LINKLOCAL) - fl.oif = transport->saddr_h.v6.sin6_scope_id; + fl.oif = transport->saddr.v6.sin6_scope_id; else fl.oif = sk->sk_bound_dev_if; fl.fl_ip_sport = inet_sk(sk)->sport; diff --git a/net/sctp/transport.c b/net/sctp/transport.c index d1b6834a3a4..05ab5e9c7c4 100644 --- a/net/sctp/transport.c +++ b/net/sctp/transport.c @@ -68,7 +68,6 @@ static struct sctp_transport *sctp_transport_init(struct sctp_transport *peer, peer->dst = NULL; memset(&peer->saddr, 0, sizeof(union sctp_addr)); - memset(&peer->saddr_h, 0, sizeof(union sctp_addr)); /* From 6.3.1 RTO Calculation: * @@ -261,7 +260,6 @@ void sctp_transport_route(struct sctp_transport *transport, else af->get_saddr(asoc, dst, daddr, &transport->saddr); - flip_to_h(&transport->saddr_h, &transport->saddr); transport->dst = dst; if ((transport->param_flags & SPP_PMTUD_DISABLE) && transport->pathmtu) { return; @@ -273,7 +271,7 @@ void sctp_transport_route(struct sctp_transport *transport, * association's active path for getsockname(). */ if (asoc && (transport == asoc->peer.active_path)) - opt->pf->af->to_sk_saddr(&transport->saddr_h, + opt->pf->af->to_sk_saddr(&transport->saddr, asoc->base.sk); } else transport->pathmtu = SCTP_DEFAULT_MAXSEGMENT; -- cgit v1.2.3-70-g09d2 From 80f15d62418040e78849f5fc3a4a5af9c9d1fec7 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:14:23 -0800 Subject: [SCTP]: ->source_h is not used anymore. kill it Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 1 - net/sctp/sm_make_chunk.c | 1 - 2 files changed, 2 deletions(-) (limited to 'include') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 27519562014..9bb5c038b95 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -714,7 +714,6 @@ struct sctp_chunk { /* What is the origin IP address for this chunk? */ union sctp_addr source; - union sctp_addr source_h; /* Destination address for this chunk. */ union sctp_addr dest; diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index beee1aa09e7..9d55497627f 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -1032,7 +1032,6 @@ void sctp_init_addrs(struct sctp_chunk *chunk, union sctp_addr *src, union sctp_addr *dest) { memcpy(&chunk->source, src, sizeof(union sctp_addr)); - flip_to_h(&chunk->source_h, &chunk->source); memcpy(&chunk->dest, dest, sizeof(union sctp_addr)); } -- cgit v1.2.3-70-g09d2 From 74af924ab6562717ef9aab1061ec05bbbf31d979 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:22:26 -0800 Subject: [SCTP]: ->a_h is gone now. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 1 - net/sctp/bind_addr.c | 2 -- net/sctp/ipv6.c | 1 - net/sctp/protocol.c | 1 - 4 files changed, 5 deletions(-) (limited to 'include') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 9bb5c038b95..bb857957a88 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -757,7 +757,6 @@ const union sctp_addr *sctp_source(const struct sctp_chunk *chunk); struct sctp_sockaddr_entry { struct list_head list; union sctp_addr a; - union sctp_addr a_h; __u8 use_as_src; }; diff --git a/net/sctp/bind_addr.c b/net/sctp/bind_addr.c index d6664dd30e5..00994158e49 100644 --- a/net/sctp/bind_addr.c +++ b/net/sctp/bind_addr.c @@ -163,8 +163,6 @@ int sctp_add_bind_addr(struct sctp_bind_addr *bp, union sctp_addr *new, if (!addr->a.v4.sin_port) addr->a.v4.sin_port = htons(bp->port); - flip_to_h(&addr->a_h, &addr->a); - addr->use_as_src = use_as_src; INIT_LIST_HEAD(&addr->list); diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index b3fa426bd1d..04d5e13e918 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -336,7 +336,6 @@ static void sctp_v6_copy_addrlist(struct list_head *addrlist, addr->a.v6.sin6_port = 0; addr->a.v6.sin6_addr = ifp->addr; addr->a.v6.sin6_scope_id = dev->ifindex; - addr->a_h = addr->a; INIT_LIST_HEAD(&addr->list); list_add_tail(&addr->list, addrlist); } diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index c20f05b3308..807c97da33a 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -153,7 +153,6 @@ static void sctp_v4_copy_addrlist(struct list_head *addrlist, addr->a.v4.sin_family = AF_INET; addr->a.v4.sin_port = 0; addr->a.v4.sin_addr.s_addr = ifa->ifa_local; - addr->a_h = addr->a; list_add_tail(&addr->list, addrlist); } } -- cgit v1.2.3-70-g09d2 From 516b20ee2d4df76e7f76332e161a25c70e8f7bea Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:24:02 -0800 Subject: [SCTP]: ->a_h is gone now. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 1 - net/sctp/transport.c | 1 - 2 files changed, 2 deletions(-) (limited to 'include') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index bb857957a88..02f1fd5de4b 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -843,7 +843,6 @@ struct sctp_transport { /* This is the peer's IP address and port. */ union sctp_addr ipaddr; - union sctp_addr ipaddr_h; /* These are the functions we call to handle LLP stuff. */ struct sctp_af *af_specific; diff --git a/net/sctp/transport.c b/net/sctp/transport.c index 05ab5e9c7c4..3e5936a5f67 100644 --- a/net/sctp/transport.c +++ b/net/sctp/transport.c @@ -62,7 +62,6 @@ static struct sctp_transport *sctp_transport_init(struct sctp_transport *peer, { /* Copy in the address. */ peer->ipaddr = *addr; - flip_to_h(&peer->ipaddr_h, &peer->ipaddr); peer->af_specific = sctp_get_af_specific(addr->sa.sa_family); peer->asoc = NULL; -- cgit v1.2.3-70-g09d2 From c9c938cb050e6d45088c5c39e4097742e875c496 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:24:36 -0800 Subject: [SCTP]: flip_to_{h,n}() are not needed anymore. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 22 ---------------------- 1 file changed, 22 deletions(-) (limited to 'include') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 02f1fd5de4b..d44202b0487 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -74,28 +74,6 @@ union sctp_addr { struct sockaddr sa; }; -static inline void flip_to_n(union sctp_addr *to, const union sctp_addr *from) -{ - size_t len; - if (from->sa.sa_family == AF_INET6) - len = sizeof(struct sockaddr_in6); - else - len = sizeof(struct sockaddr); - memcpy(to, from, len); - to->v4.sin_port = htons(from->v4.sin_port); -} - -static inline void flip_to_h(union sctp_addr *to, const union sctp_addr *from) -{ - size_t len; - if (from->sa.sa_family == AF_INET6) - len = sizeof(struct sockaddr_in6); - else - len = sizeof(struct sockaddr); - memcpy(to, from, len); - to->v4.sin_port = ntohs(from->v4.sin_port); -} - /* Forward declarations for data structures. */ struct sctp_globals; struct sctp_endpoint; -- cgit v1.2.3-70-g09d2 From 6fbfa9f951878ab489147d9e459191d4aacfa819 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:24:53 -0800 Subject: [SCTP]: Annotate ->inaddr_any(). Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 2 +- net/sctp/ipv6.c | 2 +- net/sctp/protocol.c | 2 +- net/sctp/socket.c | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index d44202b0487..3a6b8745538 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -560,7 +560,7 @@ struct sctp_af { struct sctp_sock *, const struct sk_buff *); sctp_scope_t (*scope) (union sctp_addr *); - void (*inaddr_any) (union sctp_addr *, unsigned short); + void (*inaddr_any) (union sctp_addr *, __be16); int (*is_any) (const union sctp_addr *); int (*available) (union sctp_addr *, struct sctp_sock *); diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 8e2b1e4ec66..268d32c8959 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -478,7 +478,7 @@ static int sctp_v6_cmp_addr(const union sctp_addr *addr1, } /* Initialize addr struct to INADDR_ANY. */ -static void sctp_v6_inaddr_any(union sctp_addr *addr, unsigned short port) +static void sctp_v6_inaddr_any(union sctp_addr *addr, __be16 port) { memset(addr, 0x00, sizeof(union sctp_addr)); addr->v6.sin6_family = AF_INET6; diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index c6297744a4e..5da887528bf 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -338,7 +338,7 @@ static int sctp_v4_cmp_addr(const union sctp_addr *addr1, } /* Initialize addr struct to INADDR_ANY. */ -static void sctp_v4_inaddr_any(union sctp_addr *addr, unsigned short port) +static void sctp_v4_inaddr_any(union sctp_addr *addr, __be16 port) { addr->v4.sin_family = AF_INET; addr->v4.sin_addr.s_addr = INADDR_ANY; diff --git a/net/sctp/socket.c b/net/sctp/socket.c index adbe531fded..56f77d35a8e 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -5055,7 +5055,7 @@ static int sctp_autobind(struct sock *sk) { union sctp_addr autoaddr; struct sctp_af *af; - unsigned short port; + __be16 port; /* Initialize a local sockaddr structure to INADDR_ANY. */ af = sctp_sk(sk)->pf->af; -- cgit v1.2.3-70-g09d2 From e1857ea28dc76e2a929d1fff4d5d5cf712f12f4e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:25:32 -0800 Subject: [SCTP]: sctp_association ->peer.i is a host-endian analog of sctp_inthdr. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 3a6b8745538..2db4a1bdd1a 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1314,6 +1314,13 @@ int sctp_process_init(struct sctp_association *, sctp_cid_t cid, __u32 sctp_generate_tag(const struct sctp_endpoint *); __u32 sctp_generate_tsn(const struct sctp_endpoint *); +struct sctp_inithdr_host { + __u32 init_tag; + __u32 a_rwnd; + __u16 num_outbound_streams; + __u16 num_inbound_streams; + __u32 initial_tsn; +}; /* RFC2960 * @@ -1484,7 +1491,7 @@ struct sctp_association { */ __be16 addip_disabled_mask; - struct sctp_inithdr i; + struct sctp_inithdr_host i; int cookie_len; void *cookie; -- cgit v1.2.3-70-g09d2 From 3dbe86566ed262dae3b5472b9360cb5b65d42716 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:25:49 -0800 Subject: [SCTP]: Annotate ->supported_addrs(). Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 2 +- net/sctp/ipv6.c | 2 +- net/sctp/protocol.c | 2 +- net/sctp/sm_make_chunk.c | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 2db4a1bdd1a..f8cbe40f52c 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -587,7 +587,7 @@ struct sctp_pf { struct sctp_sock *); int (*bind_verify) (struct sctp_sock *, union sctp_addr *); int (*send_verify) (struct sctp_sock *, union sctp_addr *); - int (*supported_addrs)(const struct sctp_sock *, __u16 *); + int (*supported_addrs)(const struct sctp_sock *, __be16 *); struct sock *(*create_accept_sk) (struct sock *sk, struct sctp_association *asoc); void (*addr_v4map) (struct sctp_sock *, union sctp_addr *); diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 268d32c8959..3c3e560087c 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -853,7 +853,7 @@ static int sctp_inet6_send_verify(struct sctp_sock *opt, union sctp_addr *addr) * Returns number of addresses supported. */ static int sctp_inet6_supported_addrs(const struct sctp_sock *opt, - __u16 *types) + __be16 *types) { types[0] = SCTP_PARAM_IPV4_ADDRESS; types[1] = SCTP_PARAM_IPV6_ADDRESS; diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 5da887528bf..11f3b549f4a 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -791,7 +791,7 @@ static int sctp_inet_send_verify(struct sctp_sock *opt, union sctp_addr *addr) * chunks. Returns number of addresses supported. */ static int sctp_inet_supported_addrs(const struct sctp_sock *opt, - __u16 *types) + __be16 *types) { types[0] = SCTP_PARAM_IPV4_ADDRESS; return 1; diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 9d55497627f..9674e6e3205 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -183,7 +183,7 @@ struct sctp_chunk *sctp_make_init(const struct sctp_association *asoc, int num_types, addrs_len = 0; struct sctp_sock *sp; sctp_supported_addrs_param_t sat; - __u16 types[2]; + __be16 types[2]; sctp_adaption_ind_param_t aiparam; /* RFC 2960 3.3.2 Initiation (INIT) (1) -- cgit v1.2.3-70-g09d2 From 962c837275e8a8c1df41f1882e971636093cdee4 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:26:08 -0800 Subject: [SCTP]: Netfilter sctp annotations. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_sctp.h | 2 +- net/netfilter/xt_sctp.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sctp.h b/include/linux/netfilter/nf_conntrack_sctp.h index b8994d9fd1a..5cf2c115cce 100644 --- a/include/linux/netfilter/nf_conntrack_sctp.h +++ b/include/linux/netfilter/nf_conntrack_sctp.h @@ -20,7 +20,7 @@ struct ip_ct_sctp { enum sctp_conntrack state; - u_int32_t vtag[IP_CT_DIR_MAX]; + __be32 vtag[IP_CT_DIR_MAX]; u_int32_t ttag[IP_CT_DIR_MAX]; }; diff --git a/net/netfilter/xt_sctp.c b/net/netfilter/xt_sctp.c index 7956acaaa24..71bf036f833 100644 --- a/net/netfilter/xt_sctp.c +++ b/net/netfilter/xt_sctp.c @@ -71,7 +71,7 @@ match_packet(const struct sk_buff *skb, duprintf("Chunk num: %d\toffset: %d\ttype: %d\tlength: %d\tflags: %x\n", ++i, offset, sch->type, htons(sch->length), sch->flags); - offset += (htons(sch->length) + 3) & ~3; + offset += (ntohs(sch->length) + 3) & ~3; duprintf("skb->len: %d\toffset: %d\n", skb->len, offset); -- cgit v1.2.3-70-g09d2 From 2178eda82616566b7397791afa6e5487990bac8e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:26:53 -0800 Subject: [SCTP]: SCTP_CMD_PROCESS_CTSN annotations. argument passed as __be32 Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sctp/command.h | 2 ++ net/sctp/sm_sideeffect.c | 2 +- net/sctp/sm_statefuns.c | 2 +- 3 files changed, 4 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/sctp/command.h b/include/net/sctp/command.h index dfe5d44b61a..6114c4f54b0 100644 --- a/include/net/sctp/command.h +++ b/include/net/sctp/command.h @@ -116,6 +116,7 @@ typedef enum { typedef union { __s32 i32; __u32 u32; + __be32 be32; __u16 u16; __u8 u8; int error; @@ -165,6 +166,7 @@ SCTP_## name (type arg) \ SCTP_ARG_CONSTRUCTOR(I32, __s32, i32) SCTP_ARG_CONSTRUCTOR(U32, __u32, u32) +SCTP_ARG_CONSTRUCTOR(BE32, __be32, be32) SCTP_ARG_CONSTRUCTOR(U16, __u16, u16) SCTP_ARG_CONSTRUCTOR(U8, __u8, u8) SCTP_ARG_CONSTRUCTOR(ERROR, int, error) diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index c512ce49183..7bbc6156e45 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -1420,7 +1420,7 @@ static int sctp_cmd_interpreter(sctp_event_t event_type, case SCTP_CMD_PROCESS_CTSN: /* Dummy up a SACK for processing. */ - sackh.cum_tsn_ack = cmd->obj.u32; + sackh.cum_tsn_ack = cmd->obj.be32; sackh.a_rwnd = 0; sackh.num_gap_ack_blocks = 0; sackh.num_dup_tsns = 0; diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index b70e3258b6d..c136d991da6 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -2466,7 +2466,7 @@ sctp_disposition_t sctp_sf_do_9_2_shutdown(const struct sctp_endpoint *ep, * received by the SHUTDOWN sender. */ sctp_add_cmd_sf(commands, SCTP_CMD_PROCESS_CTSN, - SCTP_U32(chunk->subh.shutdown_hdr->cum_tsn_ack)); + SCTP_BE32(chunk->subh.shutdown_hdr->cum_tsn_ack)); out: return disposition; -- cgit v1.2.3-70-g09d2 From 8e5200f54062b8af0ed1d186ea0f113854786d89 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 18:06:37 -0800 Subject: [NET]: Fix assorted misannotations (from md5 and udplite merges). Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/tcp.h | 2 +- include/net/udp.h | 2 +- include/net/udplite.h | 4 ++-- net/ipv4/tcp_ipv4.c | 2 +- net/ipv4/udp.c | 4 ++-- net/ipv6/tcp_ipv6.c | 4 ++-- net/ipv6/udp.c | 2 +- net/ipv6/udplite.c | 2 +- 8 files changed, 11 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/net/tcp.h b/include/net/tcp.h index aa7989c5379..c99774f15eb 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1154,7 +1154,7 @@ extern int tcp_v4_md5_do_add(struct sock *sk, u8 newkeylen); extern int tcp_v4_md5_do_del(struct sock *sk, - u32 addr); + __be32 addr); extern struct tcp_md5sig_pool **tcp_alloc_md5sig_pool(void); extern void tcp_free_md5sig_pool(void); diff --git a/include/net/udp.h b/include/net/udp.h index c5ccd9a3387..eac69ff0582 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -77,7 +77,7 @@ static inline __sum16 __udp_lib_checksum_complete(struct sk_buff *skb) skb->csum)); } -static inline __sum16 udp_lib_checksum_complete(struct sk_buff *skb) +static inline int udp_lib_checksum_complete(struct sk_buff *skb) { return skb->ip_summed != CHECKSUM_UNNECESSARY && __udp_lib_checksum_complete(skb); diff --git a/include/net/udplite.h b/include/net/udplite.h index 3abaab7b78c..67ac5142430 100644 --- a/include/net/udplite.h +++ b/include/net/udplite.h @@ -90,9 +90,9 @@ static __inline__ int udplite6_csum_init(struct sk_buff *skb, struct udphdr *uh) int rc = udplite_checksum_init(skb, uh); if (!rc) - skb->csum = ~csum_ipv6_magic(&skb->nh.ipv6h->saddr, + skb->csum = ~csum_unfold(csum_ipv6_magic(&skb->nh.ipv6h->saddr, &skb->nh.ipv6h->daddr, - skb->len, IPPROTO_UDPLITE, 0); + skb->len, IPPROTO_UDPLITE, 0)); return rc; } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index b7d5522092e..e9d467124c4 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1020,7 +1020,7 @@ static int tcp_v4_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key, #ifdef CONFIG_TCP_MD5SIG_DEBUG int i; #endif - __u16 old_checksum; + __sum16 old_checksum; struct tcp_md5sig_pool *hp; struct tcp4_pseudohdr *bp; struct hash_desc *desc; diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 7eb76fbf1b4..28e4cf662ce 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -114,7 +114,7 @@ DEFINE_RWLOCK(udp_hash_lock); static int udp_port_rover; -static inline int __udp_lib_lport_inuse(__be16 num, struct hlist_head udptable[]) +static inline int __udp_lib_lport_inuse(__u16 num, struct hlist_head udptable[]) { struct sock *sk; struct hlist_node *node; @@ -455,7 +455,7 @@ static int udp_push_pending_frames(struct sock *sk, struct udp_sock *up) struct sk_buff *skb; struct udphdr *uh; int err = 0; - u32 csum = 0; + __wsum csum = 0; /* Grab the skbuff where UDP header space exists. */ if ((skb = skb_peek(&sk->sk_write_queue)) == NULL) diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index d2170da77e5..0adb337c4b7 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -739,7 +739,7 @@ static int tcp_v6_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key, struct scatterlist sg[4]; __u16 data_len; int block = 0; - __u16 cksum; + __sum16 cksum; struct tcp_md5sig_pool *hp; struct tcp6_pseudohdr *bp; struct hash_desc *desc; @@ -1032,7 +1032,7 @@ static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb) #ifdef CONFIG_TCP_MD5SIG if (key) { - u32 *opt = (u32*)(t1 + 1); + __be32 *opt = (__be32*)(t1 + 1); opt[0] = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | (TCPOPT_MD5SIG << 8) | diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index efa8950ddd3..b3ea8af50a9 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -248,7 +248,7 @@ out: static __inline__ void udpv6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, int type, - int code, int offset, __u32 info ) + int code, int offset, __be32 info ) { return __udp6_lib_err(skb, opt, type, code, offset, info, udp_hash); } diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c index e0ec5e63004..d4cafacc235 100644 --- a/net/ipv6/udplite.c +++ b/net/ipv6/udplite.c @@ -24,7 +24,7 @@ static __inline__ int udplitev6_rcv(struct sk_buff **pskb) static __inline__ void udplitev6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, - int type, int code, int offset, __u32 info) + int type, int code, int offset, __be32 info) { return __udp6_lib_err(skb, opt, type, code, offset, info, udplite_hash); } -- cgit v1.2.3-70-g09d2 From ff1dcadb1b55dbf471c5ed109dbbdf06bd19ef3b Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 18:07:29 -0800 Subject: [NET]: Split skb->csum ... into anonymous union of __wsum and __u32 (csum and csum_offset resp.) Signed-off-by: Al Viro Signed-off-by: David S. Miller --- drivers/net/cassini.c | 2 +- drivers/net/e1000/e1000_main.c | 2 +- drivers/net/ixgb/ixgb_main.c | 2 +- drivers/net/myri10ge/myri10ge.c | 2 +- drivers/net/sk98lin/skge.c | 4 ++-- drivers/net/skge.c | 2 +- drivers/net/sky2.c | 2 +- drivers/net/sungem.c | 2 +- drivers/net/sunhme.c | 2 +- include/linux/skbuff.h | 5 ++++- net/core/dev.c | 4 ++-- net/core/skbuff.c | 2 +- net/ipv4/tcp_ipv4.c | 4 ++-- net/ipv4/udp.c | 2 +- net/ipv6/tcp_ipv6.c | 4 ++-- 15 files changed, 22 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/drivers/net/cassini.c b/drivers/net/cassini.c index 521c5b71023..fd2cc13f7d9 100644 --- a/drivers/net/cassini.c +++ b/drivers/net/cassini.c @@ -2825,7 +2825,7 @@ static inline int cas_xmit_tx_ringN(struct cas *cp, int ring, u64 csum_start_off, csum_stuff_off; csum_start_off = (u64) (skb->h.raw - skb->data); - csum_stuff_off = (u64) ((skb->h.raw + skb->csum) - skb->data); + csum_stuff_off = csum_start_off + skb->csum_offset; ctrl = TX_DESC_CSUM_EN | CAS_BASE(TX_DESC_CSUM_START, csum_start_off) | diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c index 7a0828869ec..32dde0adb68 100644 --- a/drivers/net/e1000/e1000_main.c +++ b/drivers/net/e1000/e1000_main.c @@ -2826,7 +2826,7 @@ e1000_tx_csum(struct e1000_adapter *adapter, struct e1000_tx_ring *tx_ring, context_desc = E1000_CONTEXT_DESC(*tx_ring, i); context_desc->upper_setup.tcp_fields.tucss = css; - context_desc->upper_setup.tcp_fields.tucso = css + skb->csum; + context_desc->upper_setup.tcp_fields.tucso = css + skb->csum_offset; context_desc->upper_setup.tcp_fields.tucse = 0; context_desc->tcp_seg_setup.data = 0; context_desc->cmd_and_length = cpu_to_le32(E1000_TXD_CMD_DEXT); diff --git a/drivers/net/ixgb/ixgb_main.c b/drivers/net/ixgb/ixgb_main.c index e09f575a3a3..7b127212e62 100644 --- a/drivers/net/ixgb/ixgb_main.c +++ b/drivers/net/ixgb/ixgb_main.c @@ -1249,7 +1249,7 @@ ixgb_tx_csum(struct ixgb_adapter *adapter, struct sk_buff *skb) if(likely(skb->ip_summed == CHECKSUM_PARTIAL)) { struct ixgb_buffer *buffer_info; css = skb->h.raw - skb->data; - cso = (skb->h.raw + skb->csum) - skb->data; + cso = css + skb->csum_offset; i = adapter->tx_ring.next_to_use; context_desc = IXGB_CONTEXT_DESC(adapter->tx_ring, i); diff --git a/drivers/net/myri10ge/myri10ge.c b/drivers/net/myri10ge/myri10ge.c index 806081b5973..36350e6db1c 100644 --- a/drivers/net/myri10ge/myri10ge.c +++ b/drivers/net/myri10ge/myri10ge.c @@ -1955,7 +1955,7 @@ again: flags = (MXGEFW_FLAGS_NO_TSO | MXGEFW_FLAGS_FIRST); if (likely(skb->ip_summed == CHECKSUM_PARTIAL)) { cksum_offset = (skb->h.raw - skb->data); - pseudo_hdr_offset = (skb->h.raw + skb->csum) - skb->data; + pseudo_hdr_offset = cksum_offset + skb->csum_offset; /* If the headers are excessively large, then we must * fall back to a software checksum */ if (unlikely(cksum_offset > 255 || pseudo_hdr_offset > 127)) { diff --git a/drivers/net/sk98lin/skge.c b/drivers/net/sk98lin/skge.c index a5d41ebc9fb..12cbfd190dd 100644 --- a/drivers/net/sk98lin/skge.c +++ b/drivers/net/sk98lin/skge.c @@ -1562,7 +1562,7 @@ struct sk_buff *pMessage) /* pointer to send-message */ if (pMessage->ip_summed == CHECKSUM_PARTIAL) { u16 hdrlen = pMessage->h.raw - pMessage->data; - u16 offset = hdrlen + pMessage->csum; + u16 offset = hdrlen + pMessage->csum_offset; if ((pMessage->h.ipiph->protocol == IPPROTO_UDP ) && (pAC->GIni.GIChipRev == 0) && @@ -1681,7 +1681,7 @@ struct sk_buff *pMessage) /* pointer to send-message */ */ if (pMessage->ip_summed == CHECKSUM_PARTIAL) { u16 hdrlen = pMessage->h.raw - pMessage->data; - u16 offset = hdrlen + pMessage->csum; + u16 offset = hdrlen + pMessage->csum_offset; Control = BMU_STFWD; diff --git a/drivers/net/skge.c b/drivers/net/skge.c index 27b537c8d5e..5513907e839 100644 --- a/drivers/net/skge.c +++ b/drivers/net/skge.c @@ -2565,7 +2565,7 @@ static int skge_xmit_frame(struct sk_buff *skb, struct net_device *dev) td->csum_offs = 0; td->csum_start = offset; - td->csum_write = offset + skb->csum; + td->csum_write = offset + skb->csum_offset; } else control = BMU_CHECK; diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index 0ef1848b976..842abd9396c 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -1350,7 +1350,7 @@ static int sky2_xmit_frame(struct sk_buff *skb, struct net_device *dev) u32 tcpsum; tcpsum = offset << 16; /* sum start */ - tcpsum |= offset + skb->csum; /* sum write */ + tcpsum |= offset + skb->csum_offset; /* sum write */ ctrl = CALSUM | WR_SUM | INIT_SUM | LOCK_SUM; if (skb->nh.iph->protocol == IPPROTO_UDP) diff --git a/drivers/net/sungem.c b/drivers/net/sungem.c index 253e96e7ad2..334c6cfd659 100644 --- a/drivers/net/sungem.c +++ b/drivers/net/sungem.c @@ -1030,7 +1030,7 @@ static int gem_start_xmit(struct sk_buff *skb, struct net_device *dev) u64 csum_start_off, csum_stuff_off; csum_start_off = (u64) (skb->h.raw - skb->data); - csum_stuff_off = (u64) ((skb->h.raw + skb->csum) - skb->data); + csum_stuff_off = csum_start_off + skb->csum_offset; ctrl = (TXDCTRL_CENAB | (csum_start_off << 15) | diff --git a/drivers/net/sunhme.c b/drivers/net/sunhme.c index 9d7cd130c19..ec432ea879f 100644 --- a/drivers/net/sunhme.c +++ b/drivers/net/sunhme.c @@ -2272,7 +2272,7 @@ static int happy_meal_start_xmit(struct sk_buff *skb, struct net_device *dev) u32 csum_start_off, csum_stuff_off; csum_start_off = (u32) (skb->h.raw - skb->data); - csum_stuff_off = (u32) ((skb->h.raw + skb->csum) - skb->data); + csum_stuff_off = csum_start_off + skb->csum_offset; tx_flags = (TXFLAG_OWN | TXFLAG_CSENABLE | ((csum_start_off << 14) & TXFLAG_CSBUFBEGIN) | diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index fcab543d79a..14ec16d2d9b 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -274,7 +274,10 @@ struct sk_buff { unsigned int len, data_len, mac_len; - __wsum csum; + union { + __wsum csum; + __u32 csum_offset; + }; __u32 priority; __u8 local_df:1, cloned:1, diff --git a/net/core/dev.c b/net/core/dev.c index 1a36b17f4b5..59d058a3b50 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1191,9 +1191,9 @@ int skb_checksum_help(struct sk_buff *skb) offset = skb->tail - skb->h.raw; BUG_ON(offset <= 0); - BUG_ON(skb->csum + 2 > offset); + BUG_ON(skb->csum_offset + 2 > offset); - *(__sum16*)(skb->h.raw + skb->csum) = csum_fold(csum); + *(__sum16*)(skb->h.raw + skb->csum_offset) = csum_fold(csum); out_set_summed: skb->ip_summed = CHECKSUM_NONE; diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 07c25d60192..a90bc439488 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1414,7 +1414,7 @@ void skb_copy_and_csum_dev(const struct sk_buff *skb, u8 *to) skb->len - csstart, 0); if (skb->ip_summed == CHECKSUM_PARTIAL) { - long csstuff = csstart + skb->csum; + long csstuff = csstart + skb->csum_offset; *((__sum16 *)(to + csstuff)) = csum_fold(csum); } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index e9d467124c4..4913f25e5ad 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -504,7 +504,7 @@ void tcp_v4_send_check(struct sock *sk, int len, struct sk_buff *skb) if (skb->ip_summed == CHECKSUM_PARTIAL) { th->check = ~tcp_v4_check(th, len, inet->saddr, inet->daddr, 0); - skb->csum = offsetof(struct tcphdr, check); + skb->csum_offset = offsetof(struct tcphdr, check); } else { th->check = tcp_v4_check(th, len, inet->saddr, inet->daddr, csum_partial((char *)th, @@ -526,7 +526,7 @@ int tcp_v4_gso_send_check(struct sk_buff *skb) th->check = 0; th->check = ~tcp_v4_check(th, skb->len, iph->saddr, iph->daddr, 0); - skb->csum = offsetof(struct tcphdr, check); + skb->csum_offset = offsetof(struct tcphdr, check); skb->ip_summed = CHECKSUM_PARTIAL; return 0; } diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 28e4cf662ce..1807a30694d 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -422,7 +422,7 @@ static void udp4_hwcsum_outgoing(struct sock *sk, struct sk_buff *skb, /* * Only one fragment on the socket. */ - skb->csum = offsetof(struct udphdr, check); + skb->csum_offset = offsetof(struct udphdr, check); uh->check = ~csum_tcpudp_magic(src, dst, len, IPPROTO_UDP, 0); } else { /* diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 0adb337c4b7..517c50024bf 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -948,7 +948,7 @@ static void tcp_v6_send_check(struct sock *sk, int len, struct sk_buff *skb) if (skb->ip_summed == CHECKSUM_PARTIAL) { th->check = ~csum_ipv6_magic(&np->saddr, &np->daddr, len, IPPROTO_TCP, 0); - skb->csum = offsetof(struct tcphdr, check); + skb->csum_offset = offsetof(struct tcphdr, check); } else { th->check = csum_ipv6_magic(&np->saddr, &np->daddr, len, IPPROTO_TCP, csum_partial((char *)th, th->doff<<2, @@ -970,7 +970,7 @@ static int tcp_v6_gso_send_check(struct sk_buff *skb) th->check = 0; th->check = ~csum_ipv6_magic(&ipv6h->saddr, &ipv6h->daddr, skb->len, IPPROTO_TCP, 0); - skb->csum = offsetof(struct tcphdr, check); + skb->csum_offset = offsetof(struct tcphdr, check); skb->ip_summed = CHECKSUM_PARTIAL; return 0; } -- cgit v1.2.3-70-g09d2 From 66c6f529c31e2886536aad4b2320d566deb1f150 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 18:07:51 -0800 Subject: [NET]: net/sched annotations. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/net/sch_generic.h | 2 +- net/sched/cls_rsvp.h | 16 ++++++++-------- net/sched/sch_api.c | 2 +- net/sched/sch_dsmark.c | 2 +- 4 files changed, 11 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index b0e9108a4e1..6b3829480b3 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -144,7 +144,7 @@ struct tcf_proto void *root; int (*classify)(struct sk_buff*, struct tcf_proto*, struct tcf_result *); - u32 protocol; + __be16 protocol; /* All the rest */ u32 prio; diff --git a/net/sched/cls_rsvp.h b/net/sched/cls_rsvp.h index 6e230ecfba0..587b9adab38 100644 --- a/net/sched/cls_rsvp.h +++ b/net/sched/cls_rsvp.h @@ -77,7 +77,7 @@ struct rsvp_head struct rsvp_session { struct rsvp_session *next; - u32 dst[RSVP_DST_LEN]; + __be32 dst[RSVP_DST_LEN]; struct tc_rsvp_gpi dpi; u8 protocol; u8 tunnelid; @@ -89,7 +89,7 @@ struct rsvp_session struct rsvp_filter { struct rsvp_filter *next; - u32 src[RSVP_DST_LEN]; + __be32 src[RSVP_DST_LEN]; struct tc_rsvp_gpi spi; u8 tunnelhdr; @@ -100,17 +100,17 @@ struct rsvp_filter struct rsvp_session *sess; }; -static __inline__ unsigned hash_dst(u32 *dst, u8 protocol, u8 tunnelid) +static __inline__ unsigned hash_dst(__be32 *dst, u8 protocol, u8 tunnelid) { - unsigned h = dst[RSVP_DST_LEN-1]; + unsigned h = (__force __u32)dst[RSVP_DST_LEN-1]; h ^= h>>16; h ^= h>>8; return (h ^ protocol ^ tunnelid) & 0xFF; } -static __inline__ unsigned hash_src(u32 *src) +static __inline__ unsigned hash_src(__be32 *src) { - unsigned h = src[RSVP_DST_LEN-1]; + unsigned h = (__force __u32)src[RSVP_DST_LEN-1]; h ^= h>>16; h ^= h>>8; h ^= h>>4; @@ -138,7 +138,7 @@ static int rsvp_classify(struct sk_buff *skb, struct tcf_proto *tp, struct rsvp_session *s; struct rsvp_filter *f; unsigned h1, h2; - u32 *dst, *src; + __be32 *dst, *src; u8 protocol; u8 tunnelid = 0; u8 *xprt; @@ -410,7 +410,7 @@ static int rsvp_change(struct tcf_proto *tp, unsigned long base, struct rtattr *tb[TCA_RSVP_MAX]; struct tcf_exts e; unsigned h1, h2; - u32 *dst; + __be32 *dst; int err; if (opt == NULL) diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 0b648929114..d99802f3023 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -1112,7 +1112,7 @@ int tc_classify(struct sk_buff *skb, struct tcf_proto *tp, struct tcf_result *res) { int err = 0; - u32 protocol = skb->protocol; + __be16 protocol = skb->protocol; #ifdef CONFIG_NET_CLS_ACT struct tcf_proto *otp = tp; reclassify: diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c index 11c8a2119b9..0b7c0e82d60 100644 --- a/net/sched/sch_dsmark.c +++ b/net/sched/sch_dsmark.c @@ -307,7 +307,7 @@ static struct sk_buff *dsmark_dequeue(struct Qdisc *sch) if (p->mask[index] != 0xff || p->value[index]) printk(KERN_WARNING "dsmark_dequeue: " "unsupported protocol %d\n", - htons(skb->protocol)); + ntohs(skb->protocol)); break; }; -- cgit v1.2.3-70-g09d2 From 4a89c2562cb81c24c515b9de041aced4d21cb6d1 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Fri, 24 Nov 2006 17:14:51 -0800 Subject: [DECNET] address: Convert to new netlink interface Extends the netlink interface to support the __le16 type and converts address addition, deletion and, dumping to use the new netlink interface. Fixes multiple occasions of possible illegal memory references due to not validated netlink attributes. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/net/netlink.h | 12 +++++ net/decnet/dn_dev.c | 142 ++++++++++++++++++++++++++++++-------------------- 2 files changed, 97 insertions(+), 57 deletions(-) (limited to 'include') diff --git a/include/net/netlink.h b/include/net/netlink.h index 30021339157..fd75fd65d59 100644 --- a/include/net/netlink.h +++ b/include/net/netlink.h @@ -829,6 +829,9 @@ static inline int nla_put_msecs(struct sk_buff *skb, int attrtype, #define NLA_PUT_U16(skb, attrtype, value) \ NLA_PUT_TYPE(skb, u16, attrtype, value) +#define NLA_PUT_LE16(skb, attrtype, value) \ + NLA_PUT_TYPE(skb, __le16, attrtype, value) + #define NLA_PUT_U32(skb, attrtype, value) \ NLA_PUT_TYPE(skb, u32, attrtype, value) @@ -874,6 +877,15 @@ static inline u16 nla_get_u16(struct nlattr *nla) return *(u16 *) nla_data(nla); } +/** + * nla_get_le16 - return payload of __le16 attribute + * @nla: __le16 netlink attribute + */ +static inline __le16 nla_get_le16(struct nlattr *nla) +{ + return *(__le16 *) nla_data(nla); +} + /** * nla_get_u8 - return payload of u8 attribute * @nla: u8 netlink attribute diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c index 309b2c7a4c6..0b9d4c95515 100644 --- a/net/decnet/dn_dev.c +++ b/net/decnet/dn_dev.c @@ -645,41 +645,62 @@ static struct dn_dev *dn_dev_by_index(int ifindex) return dn_dev; } -static int dn_dev_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) +static struct nla_policy dn_ifa_policy[IFA_MAX+1] __read_mostly = { + [IFA_ADDRESS] = { .type = NLA_U16 }, + [IFA_LOCAL] = { .type = NLA_U16 }, + [IFA_LABEL] = { .type = NLA_STRING, + .len = IFNAMSIZ - 1 }, +}; + +static int dn_nl_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) { - struct rtattr **rta = arg; + struct nlattr *tb[IFA_MAX+1]; struct dn_dev *dn_db; - struct ifaddrmsg *ifm = NLMSG_DATA(nlh); + struct ifaddrmsg *ifm; struct dn_ifaddr *ifa, **ifap; + int err = -EADDRNOTAVAIL; + + err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFA_MAX, dn_ifa_policy); + if (err < 0) + goto errout; + ifm = nlmsg_data(nlh); if ((dn_db = dn_dev_by_index(ifm->ifa_index)) == NULL) - return -EADDRNOTAVAIL; + goto errout; + + for (ifap = &dn_db->ifa_list; (ifa = *ifap); ifap = &ifa->ifa_next) { + if (tb[IFA_LOCAL] && + nla_memcmp(tb[IFA_LOCAL], &ifa->ifa_local, 2)) + continue; - for(ifap = &dn_db->ifa_list; (ifa=*ifap) != NULL; ifap = &ifa->ifa_next) { - void *tmp = rta[IFA_LOCAL-1]; - if ((tmp && memcmp(RTA_DATA(tmp), &ifa->ifa_local, 2)) || - (rta[IFA_LABEL-1] && rtattr_strcmp(rta[IFA_LABEL-1], ifa->ifa_label))) + if (tb[IFA_LABEL] && nla_strcmp(tb[IFA_LABEL], ifa->ifa_label)) continue; dn_dev_del_ifa(dn_db, ifap, 1); return 0; } - return -EADDRNOTAVAIL; +errout: + return err; } -static int dn_dev_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) +static int dn_nl_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) { - struct rtattr **rta = arg; + struct nlattr *tb[IFA_MAX+1]; struct net_device *dev; struct dn_dev *dn_db; - struct ifaddrmsg *ifm = NLMSG_DATA(nlh); + struct ifaddrmsg *ifm; struct dn_ifaddr *ifa; - int rv; + int err; + + err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFA_MAX, dn_ifa_policy); + if (err < 0) + return err; - if (rta[IFA_LOCAL-1] == NULL) + if (tb[IFA_LOCAL] == NULL) return -EINVAL; + ifm = nlmsg_data(nlh); if ((dev = __dev_get_by_index(ifm->ifa_index)) == NULL) return -ENODEV; @@ -693,22 +714,25 @@ static int dn_dev_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *a if ((ifa = dn_dev_alloc_ifa()) == NULL) return -ENOBUFS; - if (!rta[IFA_ADDRESS - 1]) - rta[IFA_ADDRESS - 1] = rta[IFA_LOCAL - 1]; - memcpy(&ifa->ifa_local, RTA_DATA(rta[IFA_LOCAL-1]), 2); - memcpy(&ifa->ifa_address, RTA_DATA(rta[IFA_ADDRESS-1]), 2); + if (tb[IFA_ADDRESS] == NULL) + tb[IFA_ADDRESS] = tb[IFA_LOCAL]; + + ifa->ifa_local = nla_get_le16(tb[IFA_LOCAL]); + ifa->ifa_address = nla_get_le16(tb[IFA_ADDRESS]); ifa->ifa_flags = ifm->ifa_flags; ifa->ifa_scope = ifm->ifa_scope; ifa->ifa_dev = dn_db; - if (rta[IFA_LABEL-1]) - rtattr_strlcpy(ifa->ifa_label, rta[IFA_LABEL-1], IFNAMSIZ); + + if (tb[IFA_LABEL]) + nla_strlcpy(ifa->ifa_label, tb[IFA_LABEL], IFNAMSIZ); else memcpy(ifa->ifa_label, dev->name, IFNAMSIZ); - rv = dn_dev_insert_ifa(dn_db, ifa); - if (rv) + err = dn_dev_insert_ifa(dn_db, ifa); + if (err) dn_dev_free_ifa(ifa); - return rv; + + return err; } static inline size_t dn_ifaddr_nlmsg_size(void) @@ -719,34 +743,34 @@ static inline size_t dn_ifaddr_nlmsg_size(void) + nla_total_size(2); /* IFA_LOCAL */ } -static int dn_dev_fill_ifaddr(struct sk_buff *skb, struct dn_ifaddr *ifa, - u32 pid, u32 seq, int event, unsigned int flags) +static int dn_nl_fill_ifaddr(struct sk_buff *skb, struct dn_ifaddr *ifa, + u32 pid, u32 seq, int event, unsigned int flags) { struct ifaddrmsg *ifm; struct nlmsghdr *nlh; - unsigned char *b = skb->tail; - nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*ifm), flags); - ifm = NLMSG_DATA(nlh); + nlh = nlmsg_put(skb, pid, seq, event, sizeof(*ifm), flags); + if (nlh == NULL) + return -ENOBUFS; + ifm = nlmsg_data(nlh); ifm->ifa_family = AF_DECnet; ifm->ifa_prefixlen = 16; ifm->ifa_flags = ifa->ifa_flags | IFA_F_PERMANENT; ifm->ifa_scope = ifa->ifa_scope; ifm->ifa_index = ifa->ifa_dev->dev->ifindex; + if (ifa->ifa_address) - RTA_PUT(skb, IFA_ADDRESS, 2, &ifa->ifa_address); + NLA_PUT_LE16(skb, IFA_ADDRESS, ifa->ifa_address); if (ifa->ifa_local) - RTA_PUT(skb, IFA_LOCAL, 2, &ifa->ifa_local); + NLA_PUT_LE16(skb, IFA_LOCAL, ifa->ifa_local); if (ifa->ifa_label[0]) - RTA_PUT(skb, IFA_LABEL, IFNAMSIZ, &ifa->ifa_label); - nlh->nlmsg_len = skb->tail - b; - return skb->len; + NLA_PUT_STRING(skb, IFA_LABEL, ifa->ifa_label); -nlmsg_failure: -rtattr_failure: - skb_trim(skb, b - skb->data); - return -1; + return nlmsg_end(skb, nlh); + +nla_put_failure: + return nlmsg_cancel(skb, nlh); } static void dn_ifaddr_notify(int event, struct dn_ifaddr *ifa) @@ -758,7 +782,7 @@ static void dn_ifaddr_notify(int event, struct dn_ifaddr *ifa) if (skb == NULL) goto errout; - err = dn_dev_fill_ifaddr(skb, ifa, 0, 0, event, 0); + err = dn_nl_fill_ifaddr(skb, ifa, 0, 0, event, 0); /* failure implies BUG in dn_ifaddr_nlmsg_size() */ BUG_ON(err < 0); @@ -768,39 +792,43 @@ errout: rtnl_set_sk_err(RTNLGRP_DECnet_IFADDR, err); } -static int dn_dev_dump_ifaddr(struct sk_buff *skb, struct netlink_callback *cb) +static int dn_nl_dump_ifaddr(struct sk_buff *skb, struct netlink_callback *cb) { - int idx, dn_idx; - int s_idx, s_dn_idx; + int idx, dn_idx = 0, skip_ndevs, skip_naddr; struct net_device *dev; struct dn_dev *dn_db; struct dn_ifaddr *ifa; - s_idx = cb->args[0]; - s_dn_idx = dn_idx = cb->args[1]; + skip_ndevs = cb->args[0]; + skip_naddr = cb->args[1]; + read_lock(&dev_base_lock); - for(dev = dev_base, idx = 0; dev; dev = dev->next, idx++) { - if (idx < s_idx) + for (dev = dev_base, idx = 0; dev; dev = dev->next, idx++) { + if (idx < skip_ndevs) continue; - if (idx > s_idx) - s_dn_idx = 0; + else if (idx > skip_ndevs) { + /* Only skip over addresses for first dev dumped + * in this iteration (idx == skip_ndevs) */ + skip_naddr = 0; + } + if ((dn_db = dev->dn_ptr) == NULL) continue; - for(ifa = dn_db->ifa_list, dn_idx = 0; ifa; ifa = ifa->ifa_next, dn_idx++) { - if (dn_idx < s_dn_idx) + for (ifa = dn_db->ifa_list, dn_idx = 0; ifa; + ifa = ifa->ifa_next, dn_idx++) { + if (dn_idx < skip_naddr) continue; - if (dn_dev_fill_ifaddr(skb, ifa, - NETLINK_CB(cb->skb).pid, - cb->nlh->nlmsg_seq, - RTM_NEWADDR, - NLM_F_MULTI) <= 0) + if (dn_nl_fill_ifaddr(skb, ifa, NETLINK_CB(cb->skb).pid, + cb->nlh->nlmsg_seq, RTM_NEWADDR, + NLM_F_MULTI) < 0) goto done; } } done: read_unlock(&dev_base_lock); + cb->args[0] = idx; cb->args[1] = dn_idx; @@ -1417,9 +1445,9 @@ static struct file_operations dn_dev_seq_fops = { static struct rtnetlink_link dnet_rtnetlink_table[RTM_NR_MSGTYPES] = { - [RTM_NEWADDR - RTM_BASE] = { .doit = dn_dev_rtm_newaddr, }, - [RTM_DELADDR - RTM_BASE] = { .doit = dn_dev_rtm_deladdr, }, - [RTM_GETADDR - RTM_BASE] = { .dumpit = dn_dev_dump_ifaddr, }, + [RTM_NEWADDR - RTM_BASE] = { .doit = dn_nl_newaddr, }, + [RTM_DELADDR - RTM_BASE] = { .doit = dn_nl_deladdr, }, + [RTM_GETADDR - RTM_BASE] = { .dumpit = dn_nl_dump_ifaddr, }, #ifdef CONFIG_DECNET_ROUTER [RTM_NEWROUTE - RTM_BASE] = { .doit = dn_fib_rtm_newroute, }, [RTM_DELROUTE - RTM_BASE] = { .doit = dn_fib_rtm_delroute, }, -- cgit v1.2.3-70-g09d2 From d61c167dd0797a16584f7a922dd5d50efad1d28a Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Mon, 27 Nov 2006 12:31:45 -0200 Subject: [NET]: Add documentation for TFRC structures This adds documentation for the TFRC structure fields. Signed-off-by: Gerrit Renker Signed-off-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/tfrc.h | 20 ++++++++++++++++++-- 1 file changed, 18 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/tfrc.h b/include/linux/tfrc.h index 7dab7831c3c..31a9b25276f 100644 --- a/include/linux/tfrc.h +++ b/include/linux/tfrc.h @@ -1,7 +1,8 @@ #ifndef _LINUX_TFRC_H_ #define _LINUX_TFRC_H_ /* - * include/linux/tfrc.h + * TFRC - Data Structures for the TCP-Friendly Rate Control congestion + * control mechanism as specified in RFC 3448. * * Copyright (c) 2005 The University of Waikato, Hamilton, New Zealand. * Copyright (c) 2005 Ian McDonald @@ -13,15 +14,30 @@ * the Free Software Foundation; either version 2 of the License, or * (at your option) any later version. */ - #include +/** tfrc_rx_info - TFRC Receiver Data Structure + * + * @tfrcrx_x_recv: receiver estimate of sending rate (3.2.2) + * @tfrcrx_rtt: round-trip-time (communicated by sender) + * @tfrcrx_p: current estimate of loss event rate (3.2.2) + */ struct tfrc_rx_info { __u32 tfrcrx_x_recv; __u32 tfrcrx_rtt; __u32 tfrcrx_p; }; +/** tfrc_tx_info - TFRC Sender Data Structure + * + * @tfrctx_x: computed transmit rate (4.3 (4)) + * @tfrctx_x_recv: receiver estimate of send rate (4.3) + * @tfrctx_x_calc: return value of throughput equation (3.1) + * @tfrctx_rtt: (moving average) estimate of RTT (4.3) + * @tfrctx_p: current loss event rate (5.4) + * @tfrctx_rto: estimate of RTO, equals 4*RTT (4.3) + * @tfrctx_ipi: inter-packet interval (4.6) + */ struct tfrc_tx_info { __u32 tfrctx_x; __u32 tfrctx_x_recv; -- cgit v1.2.3-70-g09d2 From 4e9b82693542003b028c8494e9e3c49615b91ce7 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Mon, 27 Nov 2006 09:25:58 -0800 Subject: [NETLINK]: Remove unused dst_pid field in netlink_skb_parms The destination PID is passed directly to netlink_unicast() respectively netlink_multicast(). Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/netlink.h | 1 - net/decnet/dn_route.c | 2 -- net/ipv4/fib_frontend.c | 1 - net/ipv4/netfilter/ip_conntrack_netlink.c | 4 +--- net/netfilter/nf_conntrack_netlink.c | 4 +--- net/netlink/af_netlink.c | 1 - net/xfrm/xfrm_user.c | 2 -- 7 files changed, 2 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/include/linux/netlink.h b/include/linux/netlink.h index e61e1e13842..b3b9b609ee8 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -141,7 +141,6 @@ struct netlink_skb_parms { struct ucred creds; /* Skb credentials */ __u32 pid; - __u32 dst_pid; __u32 dst_group; kernel_cap_t eff_cap; __u32 loginuid; /* Login (audit) uid */ diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index f759d6f422e..4eb985236ae 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -1592,8 +1592,6 @@ int dn_cache_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, void *arg) if (rtm->rtm_flags & RTM_F_NOTIFY) rt->rt_flags |= RTCF_NOTIFY; - NETLINK_CB(skb).dst_pid = NETLINK_CB(in_skb).pid; - err = dn_rt_fill_info(skb, NETLINK_CB(in_skb).pid, nlh->nlmsg_seq, RTM_NEWROUTE, 0, 0); if (err == 0) diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index 14025345cc5..d47b72af89e 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -811,7 +811,6 @@ static void nl_fib_input(struct sock *sk, int len) pid = nlh->nlmsg_pid; /*pid of sending process */ NETLINK_CB(skb).pid = 0; /* from kernel */ - NETLINK_CB(skb).dst_pid = pid; NETLINK_CB(skb).dst_group = 0; /* unicast */ netlink_unicast(sk, skb, pid, MSG_DONTWAIT); } diff --git a/net/ipv4/netfilter/ip_conntrack_netlink.c b/net/ipv4/netfilter/ip_conntrack_netlink.c index 55f0ae64108..1bb8ed33c5b 100644 --- a/net/ipv4/netfilter/ip_conntrack_netlink.c +++ b/net/ipv4/netfilter/ip_conntrack_netlink.c @@ -743,7 +743,6 @@ ctnetlink_get_conntrack(struct sock *ctnl, struct sk_buff *skb, ip_conntrack_put(ct); return -ENOMEM; } - NETLINK_CB(skb2).dst_pid = NETLINK_CB(skb).pid; err = ctnetlink_fill_info(skb2, NETLINK_CB(skb).pid, nlh->nlmsg_seq, IPCTNL_MSG_CT_NEW, 1, ct); @@ -1273,8 +1272,7 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb2) goto out; - NETLINK_CB(skb2).dst_pid = NETLINK_CB(skb).pid; - + err = ctnetlink_exp_fill_info(skb2, NETLINK_CB(skb).pid, nlh->nlmsg_seq, IPCTNL_MSG_EXP_NEW, 1, exp); diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index ab67c2be2b5..cfb35fd8d9b 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -752,7 +752,6 @@ ctnetlink_get_conntrack(struct sock *ctnl, struct sk_buff *skb, nf_ct_put(ct); return -ENOMEM; } - NETLINK_CB(skb2).dst_pid = NETLINK_CB(skb).pid; err = ctnetlink_fill_info(skb2, NETLINK_CB(skb).pid, nlh->nlmsg_seq, IPCTNL_MSG_CT_NEW, 1, ct); @@ -1300,8 +1299,7 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb2) goto out; - NETLINK_CB(skb2).dst_pid = NETLINK_CB(skb).pid; - + err = ctnetlink_exp_fill_info(skb2, NETLINK_CB(skb).pid, nlh->nlmsg_seq, IPCTNL_MSG_EXP_NEW, 1, exp); diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index f61d81b3c61..3baafb10f8f 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1153,7 +1153,6 @@ static int netlink_sendmsg(struct kiocb *kiocb, struct socket *sock, goto out; NETLINK_CB(skb).pid = nlk->pid; - NETLINK_CB(skb).dst_pid = dst_pid; NETLINK_CB(skb).dst_group = dst_group; NETLINK_CB(skb).loginuid = audit_get_loginuid(current->audit_context); selinux_get_task_sid(current, &(NETLINK_CB(skb).sid)); diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 6c4d2f49d9e..8dbb38b91de 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -650,7 +650,6 @@ static struct sk_buff *xfrm_state_netlink(struct sk_buff *in_skb, if (!skb) return ERR_PTR(-ENOMEM); - NETLINK_CB(skb).dst_pid = NETLINK_CB(in_skb).pid; info.in_skb = in_skb; info.out_skb = skb; info.nlmsg_seq = seq; @@ -1168,7 +1167,6 @@ static struct sk_buff *xfrm_policy_netlink(struct sk_buff *in_skb, if (!skb) return ERR_PTR(-ENOMEM); - NETLINK_CB(skb).dst_pid = NETLINK_CB(in_skb).pid; info.in_skb = in_skb; info.out_skb = skb; info.nlmsg_seq = seq; -- cgit v1.2.3-70-g09d2 From e3703b3de1f049b38733ba520e5038f23063068e Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Mon, 27 Nov 2006 09:27:07 -0800 Subject: [RTNETLINK]: Add rtnl_put_cacheinfo() to unify some code IPv4, IPv6, and DECNet all use struct rta_cacheinfo in a similiar way, therefore rtnl_put_cacheinfo() is added to reuse code. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/rtnetlink.h | 3 +++ net/core/rtnetlink.c | 20 ++++++++++++++++++++ net/decnet/dn_route.c | 16 +++++----------- net/ipv4/route.c | 26 +++++++++++--------------- net/ipv6/route.c | 19 ++++++------------- 5 files changed, 45 insertions(+), 39 deletions(-) (limited to 'include') diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index 33b3d0ab3a9..493297acdae 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -585,6 +585,9 @@ extern int rtnl_notify(struct sk_buff *skb, u32 pid, u32 group, struct nlmsghdr *nlh, gfp_t flags); extern void rtnl_set_sk_err(u32 group, int error); extern int rtnetlink_put_metrics(struct sk_buff *skb, u32 *metrics); +extern int rtnl_put_cacheinfo(struct sk_buff *skb, struct dst_entry *dst, + u32 id, u32 ts, u32 tsage, long expires, + u32 error); extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const void *data); diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 0cb4d9e53a0..e76539a5eb5 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -212,6 +212,26 @@ nla_put_failure: return nla_nest_cancel(skb, mx); } +int rtnl_put_cacheinfo(struct sk_buff *skb, struct dst_entry *dst, u32 id, + u32 ts, u32 tsage, long expires, u32 error) +{ + struct rta_cacheinfo ci = { + .rta_lastuse = jiffies_to_clock_t(jiffies - dst->lastuse), + .rta_used = dst->__use, + .rta_clntref = atomic_read(&(dst->__refcnt)), + .rta_error = error, + .rta_id = id, + .rta_ts = ts, + .rta_tsage = tsage, + }; + + if (expires) + ci.rta_expires = jiffies_to_clock_t(expires); + + return nla_put(skb, RTA_CACHEINFO, sizeof(ci), &ci); +} + +EXPORT_SYMBOL_GPL(rtnl_put_cacheinfo); static void set_operstate(struct net_device *dev, unsigned char transition) { diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index 4eb985236ae..9881933167b 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -1469,7 +1469,7 @@ static int dn_rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, struct rtmsg *r; struct nlmsghdr *nlh; unsigned char *b = skb->tail; - struct rta_cacheinfo ci; + long expires; nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*r), flags); r = NLMSG_DATA(nlh); @@ -1502,16 +1502,10 @@ static int dn_rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, RTA_PUT(skb, RTA_GATEWAY, 2, &rt->rt_gateway); if (rtnetlink_put_metrics(skb, rt->u.dst.metrics) < 0) goto rtattr_failure; - ci.rta_lastuse = jiffies_to_clock_t(jiffies - rt->u.dst.lastuse); - ci.rta_used = rt->u.dst.__use; - ci.rta_clntref = atomic_read(&rt->u.dst.__refcnt); - if (rt->u.dst.expires) - ci.rta_expires = jiffies_to_clock_t(rt->u.dst.expires - jiffies); - else - ci.rta_expires = 0; - ci.rta_error = rt->u.dst.error; - ci.rta_id = ci.rta_ts = ci.rta_tsage = 0; - RTA_PUT(skb, RTA_CACHEINFO, sizeof(ci), &ci); + expires = rt->u.dst.expires ? rt->u.dst.expires - jiffies : 0; + if (rtnl_put_cacheinfo(skb, &rt->u.dst, 0, 0, 0, expires, + rt->u.dst.error) < 0) + goto rtattr_failure; if (rt->fl.iif) RTA_PUT(skb, RTA_IIF, sizeof(int), &rt->fl.iif); diff --git a/net/ipv4/route.c b/net/ipv4/route.c index ee00b6506ab..9f3924c4905 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -2629,7 +2629,8 @@ static int rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event, struct rtable *rt = (struct rtable*)skb->dst; struct rtmsg *r; struct nlmsghdr *nlh; - struct rta_cacheinfo ci; + long expires; + u32 id = 0, ts = 0, tsage = 0, error; nlh = nlmsg_put(skb, pid, seq, event, sizeof(*r), flags); if (nlh == NULL) @@ -2676,20 +2677,13 @@ static int rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event, if (rtnetlink_put_metrics(skb, rt->u.dst.metrics) < 0) goto nla_put_failure; - ci.rta_lastuse = jiffies_to_clock_t(jiffies - rt->u.dst.lastuse); - ci.rta_used = rt->u.dst.__use; - ci.rta_clntref = atomic_read(&rt->u.dst.__refcnt); - if (rt->u.dst.expires) - ci.rta_expires = jiffies_to_clock_t(rt->u.dst.expires - jiffies); - else - ci.rta_expires = 0; - ci.rta_error = rt->u.dst.error; - ci.rta_id = ci.rta_ts = ci.rta_tsage = 0; + error = rt->u.dst.error; + expires = rt->u.dst.expires ? rt->u.dst.expires - jiffies : 0; if (rt->peer) { - ci.rta_id = rt->peer->ip_id_count; + id = rt->peer->ip_id_count; if (rt->peer->tcp_ts_stamp) { - ci.rta_ts = rt->peer->tcp_ts; - ci.rta_tsage = xtime.tv_sec - rt->peer->tcp_ts_stamp; + ts = rt->peer->tcp_ts; + tsage = xtime.tv_sec - rt->peer->tcp_ts_stamp; } } @@ -2708,7 +2702,7 @@ static int rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event, } else { if (err == -EMSGSIZE) goto nla_put_failure; - ci.rta_error = err; + error = err; } } } else @@ -2716,7 +2710,9 @@ static int rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event, NLA_PUT_U32(skb, RTA_IIF, rt->fl.iif); } - NLA_PUT(skb, RTA_CACHEINFO, sizeof(ci), &ci); + if (rtnl_put_cacheinfo(skb, &rt->u.dst, id, ts, tsage, + expires, error) < 0) + goto nla_put_failure; return nlmsg_end(skb, nlh); diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 0bf17a3cf08..9f80518aacb 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2027,7 +2027,7 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, { struct rtmsg *rtm; struct nlmsghdr *nlh; - struct rta_cacheinfo ci; + long expires; u32 table; if (prefix) { /* user wants prefix routes only */ @@ -2101,18 +2101,11 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, NLA_PUT_U32(skb, RTA_OIF, rt->rt6i_dev->ifindex); NLA_PUT_U32(skb, RTA_PRIORITY, rt->rt6i_metric); - ci.rta_lastuse = jiffies_to_clock_t(jiffies - rt->u.dst.lastuse); - if (rt->rt6i_expires) - ci.rta_expires = jiffies_to_clock_t(rt->rt6i_expires - jiffies); - else - ci.rta_expires = 0; - ci.rta_used = rt->u.dst.__use; - ci.rta_clntref = atomic_read(&rt->u.dst.__refcnt); - ci.rta_error = rt->u.dst.error; - ci.rta_id = 0; - ci.rta_ts = 0; - ci.rta_tsage = 0; - NLA_PUT(skb, RTA_CACHEINFO, sizeof(ci), &ci); + + expires = rt->rt6i_expires ? rt->rt6i_expires - jiffies : 0; + if (rtnl_put_cacheinfo(skb, &rt->u.dst, 0, 0, 0, + expires, rt->u.dst.error) < 0) + goto nla_put_failure; return nlmsg_end(skb, nlh); -- cgit v1.2.3-70-g09d2 From 4c0a6cb0db19de411c4bf7fcdc79d4c7c4ccafb1 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Mon, 27 Nov 2006 09:29:59 -0800 Subject: [UDP(-Lite)]: consolidate v4 and v6 get|setsockopt code This patch consolidates set/getsockopt code between UDP(-Lite) v4 and 6. The justification is that UDP(-Lite) is a transport-layer protocol and therefore the socket option code (at least in theory) should be AF-independent. Furthermore, there is the following code reduplication: * do_udp{,v6}_getsockopt is 100% identical between v4 and v6 * do_udp{,v6}_setsockopt is identical up to the following differerence --v4 in contrast to v4 additionally allows the experimental encapsulation types UDP_ENCAP_ESPINUDP and UDP_ENCAP_ESPINUDP_NON_IKE --the remainder is identical between v4 and v6 I believe that this difference is of little relevance. The advantages in not duplicating twice almost completely identical code. The patch further simplifies the interface of udp{,v6}_push_pending_frames, since for the second argument (struct udp_sock *up) it always holds that up = udp_sk(sk); where sk is the first function argument. Signed-off-by: Gerrit Renker Signed-off-by: David S. Miller --- include/net/udp.h | 5 +++ net/ipv4/udp.c | 30 ++++++++------ net/ipv6/udp.c | 118 +++++------------------------------------------------- 3 files changed, 32 insertions(+), 121 deletions(-) (limited to 'include') diff --git a/include/net/udp.h b/include/net/udp.h index eac69ff0582..1548d68d45d 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -134,6 +134,11 @@ extern int udp_ioctl(struct sock *sk, int cmd, unsigned long arg); extern int udp_disconnect(struct sock *sk, int flags); extern unsigned int udp_poll(struct file *file, struct socket *sock, poll_table *wait); +extern int udp_lib_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen); +extern int udp_lib_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen, + int (*push_pending_frames)(struct sock *)); DECLARE_SNMP_STAT(struct udp_mib, udp_statistics); /* diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 1807a30694d..035915fc9ed 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -448,8 +448,9 @@ static void udp4_hwcsum_outgoing(struct sock *sk, struct sk_buff *skb, /* * Push out all pending data as one UDP datagram. Socket is locked. */ -static int udp_push_pending_frames(struct sock *sk, struct udp_sock *up) +static int udp_push_pending_frames(struct sock *sk) { + struct udp_sock *up = udp_sk(sk); struct inet_sock *inet = inet_sk(sk); struct flowi *fl = &inet->cork.fl; struct sk_buff *skb; @@ -673,7 +674,7 @@ do_append_data: if (err) udp_flush_pending_frames(sk); else if (!corkreq) - err = udp_push_pending_frames(sk, up); + err = udp_push_pending_frames(sk); else if (unlikely(skb_queue_empty(&sk->sk_write_queue))) up->pending = 0; release_sock(sk); @@ -746,7 +747,7 @@ int udp_sendpage(struct sock *sk, struct page *page, int offset, up->len += size; if (!(up->corkflag || (flags&MSG_MORE))) - ret = udp_push_pending_frames(sk, up); + ret = udp_push_pending_frames(sk); if (!ret) ret = size; out: @@ -1299,8 +1300,9 @@ int udp_destroy_sock(struct sock *sk) /* * Socket option code for UDP */ -static int do_udp_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, int optlen) +int udp_lib_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen, + int (*push_pending_frames)(struct sock *)) { struct udp_sock *up = udp_sk(sk); int val; @@ -1319,7 +1321,7 @@ static int do_udp_setsockopt(struct sock *sk, int level, int optname, } else { up->corkflag = 0; lock_sock(sk); - udp_push_pending_frames(sk, up); + (*push_pending_frames)(sk); release_sock(sk); } break; @@ -1375,7 +1377,8 @@ int udp_setsockopt(struct sock *sk, int level, int optname, char __user *optval, int optlen) { if (level == SOL_UDP || level == SOL_UDPLITE) - return do_udp_setsockopt(sk, level, optname, optval, optlen); + return udp_lib_setsockopt(sk, level, optname, optval, optlen, + udp_push_pending_frames); return ip_setsockopt(sk, level, optname, optval, optlen); } @@ -1384,13 +1387,14 @@ int compat_udp_setsockopt(struct sock *sk, int level, int optname, char __user *optval, int optlen) { if (level == SOL_UDP || level == SOL_UDPLITE) - return do_udp_setsockopt(sk, level, optname, optval, optlen); + return udp_lib_setsockopt(sk, level, optname, optval, optlen, + udp_push_pending_frames); return compat_ip_setsockopt(sk, level, optname, optval, optlen); } #endif -static int do_udp_getsockopt(struct sock *sk, int level, int optname, - char __user *optval, int __user *optlen) +int udp_lib_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen) { struct udp_sock *up = udp_sk(sk); int val, len; @@ -1437,7 +1441,7 @@ int udp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { if (level == SOL_UDP || level == SOL_UDPLITE) - return do_udp_getsockopt(sk, level, optname, optval, optlen); + return udp_lib_getsockopt(sk, level, optname, optval, optlen); return ip_getsockopt(sk, level, optname, optval, optlen); } @@ -1446,7 +1450,7 @@ int compat_udp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { if (level == SOL_UDP || level == SOL_UDPLITE) - return do_udp_getsockopt(sk, level, optname, optval, optlen); + return udp_lib_getsockopt(sk, level, optname, optval, optlen); return compat_ip_getsockopt(sk, level, optname, optval, optlen); } #endif @@ -1716,6 +1720,8 @@ EXPORT_SYMBOL(udp_ioctl); EXPORT_SYMBOL(udp_get_port); EXPORT_SYMBOL(udp_prot); EXPORT_SYMBOL(udp_sendmsg); +EXPORT_SYMBOL(udp_lib_getsockopt); +EXPORT_SYMBOL(udp_lib_setsockopt); EXPORT_SYMBOL(udp_poll); #ifdef CONFIG_PROC_FS diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index b3ea8af50a9..f52a5c3cc0a 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -505,10 +505,11 @@ static void udp_v6_flush_pending_frames(struct sock *sk) * Sending */ -static int udp_v6_push_pending_frames(struct sock *sk, struct udp_sock *up) +static int udp_v6_push_pending_frames(struct sock *sk) { struct sk_buff *skb; struct udphdr *uh; + struct udp_sock *up = udp_sk(sk); struct inet_sock *inet = inet_sk(sk); struct flowi *fl = &inet->cork.fl; int err = 0; @@ -782,7 +783,7 @@ do_append_data: if (err) udp_v6_flush_pending_frames(sk); else if (!corkreq) - err = udp_v6_push_pending_frames(sk, up); + err = udp_v6_push_pending_frames(sk); else if (unlikely(skb_queue_empty(&sk->sk_write_queue))) up->pending = 0; @@ -844,72 +845,12 @@ int udpv6_destroy_sock(struct sock *sk) /* * Socket option code for UDP */ -static int do_udpv6_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, int optlen) -{ - struct udp_sock *up = udp_sk(sk); - int val; - int err = 0; - - if(optlencorkflag = 1; - } else { - up->corkflag = 0; - lock_sock(sk); - udp_v6_push_pending_frames(sk, up); - release_sock(sk); - } - break; - case UDP_ENCAP: - switch (val) { - case 0: - up->encap_type = val; - break; - default: - err = -ENOPROTOOPT; - break; - } - break; - - case UDPLITE_SEND_CSCOV: - if (!up->pcflag) /* Disable the option on UDP sockets */ - return -ENOPROTOOPT; - if (val != 0 && val < 8) /* Illegal coverage: use default (8) */ - val = 8; - up->pcslen = val; - up->pcflag |= UDPLITE_SEND_CC; - break; - - case UDPLITE_RECV_CSCOV: - if (!up->pcflag) /* Disable the option on UDP sockets */ - return -ENOPROTOOPT; - if (val != 0 && val < 8) /* Avoid silly minimal values. */ - val = 8; - up->pcrlen = val; - up->pcflag |= UDPLITE_RECV_CC; - break; - - default: - err = -ENOPROTOOPT; - break; - }; - - return err; -} - int udpv6_setsockopt(struct sock *sk, int level, int optname, char __user *optval, int optlen) { if (level == SOL_UDP || level == SOL_UDPLITE) - return do_udpv6_setsockopt(sk, level, optname, optval, optlen); + return udp_lib_setsockopt(sk, level, optname, optval, optlen, + udp_v6_push_pending_frames); return ipv6_setsockopt(sk, level, optname, optval, optlen); } @@ -918,58 +859,17 @@ int compat_udpv6_setsockopt(struct sock *sk, int level, int optname, char __user *optval, int optlen) { if (level == SOL_UDP || level == SOL_UDPLITE) - return do_udpv6_setsockopt(sk, level, optname, optval, optlen); + return udp_lib_setsockopt(sk, level, optname, optval, optlen, + udp_v6_push_pending_frames); return compat_ipv6_setsockopt(sk, level, optname, optval, optlen); } #endif -static int do_udpv6_getsockopt(struct sock *sk, int level, int optname, - char __user *optval, int __user *optlen) -{ - struct udp_sock *up = udp_sk(sk); - int val, len; - - if(get_user(len,optlen)) - return -EFAULT; - - len = min_t(unsigned int, len, sizeof(int)); - - if(len < 0) - return -EINVAL; - - switch(optname) { - case UDP_CORK: - val = up->corkflag; - break; - - case UDP_ENCAP: - val = up->encap_type; - break; - - case UDPLITE_SEND_CSCOV: - val = up->pcslen; - break; - - case UDPLITE_RECV_CSCOV: - val = up->pcrlen; - break; - - default: - return -ENOPROTOOPT; - }; - - if(put_user(len, optlen)) - return -EFAULT; - if(copy_to_user(optval, &val,len)) - return -EFAULT; - return 0; -} - int udpv6_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { if (level == SOL_UDP || level == SOL_UDPLITE) - return do_udpv6_getsockopt(sk, level, optname, optval, optlen); + return udp_lib_getsockopt(sk, level, optname, optval, optlen); return ipv6_getsockopt(sk, level, optname, optval, optlen); } @@ -978,7 +878,7 @@ int compat_udpv6_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { if (level == SOL_UDP || level == SOL_UDPLITE) - return do_udpv6_getsockopt(sk, level, optname, optval, optlen); + return udp_lib_getsockopt(sk, level, optname, optval, optlen); return compat_ipv6_getsockopt(sk, level, optname, optval, optlen); } #endif -- cgit v1.2.3-70-g09d2 From 850db6b8c53ff12eadee9ef2aa69360c06bce057 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Mon, 27 Nov 2006 17:56:43 -0200 Subject: [INET_CONNECTION_SOCK]: Pack struct inet_connection_sock_af_ops We have a hole in: [acme@newtoy net-2.6.20]$ pahole net/ipv6/tcp_ipv6.o inet_connection_sock_af_ops /* /pub/scm/linux/kernel/git/acme/net-2.6.20/include/net/inet_connection_sock.h:38 */ struct inet_connection_sock_af_ops { int (*queue_xmit)(); /* 0 4 */ void (*send_check)(); /* 4 4 */ int (*rebuild_header)(); /* 8 4 */ int (*conn_request)(); /* 12 4 */ struct sock * (*syn_recv_sock)(); /* 16 4 */ int (*remember_stamp)(); /* 20 4 */ __u16 net_header_len; /* 24 2 */ /* XXX 2 bytes hole, try to pack */ int (*setsockopt)(); /* 28 4 */ int (*getsockopt)(); /* 32 4 */ int (*compat_setsockopt)(); /* 36 4 */ int (*compat_getsockopt)(); /* 40 4 */ void (*addr2sockaddr)(); /* 44 4 */ int sockaddr_len; /* 48 4 */ }; /* size: 52, sum members: 50, holes: 1, sum holes: 2 */ But we don't need sockaddr_len to be an int: [acme@newtoy net-2.6.20]$ find net -name "*.[ch]" | xargs grep '\.sockaddr_len.\+=' | sort -u net/dccp/ipv4.c: .sockaddr_len = sizeof(struct sockaddr_in), net/dccp/ipv6.c: .sockaddr_len = sizeof(struct sockaddr_in6), net/ipv4/tcp_ipv4.c: .sockaddr_len = sizeof(struct sockaddr_in), net/ipv6/tcp_ipv6.c: .sockaddr_len = sizeof(struct sockaddr_in6), net/sctp/ipv6.c: .sockaddr_len = sizeof(struct sockaddr_in6), net/sctp/protocol.c: .sockaddr_len = sizeof(struct sockaddr_in), [acme@newtoy net-2.6.20]$ pahole --sizes net/ipv6/tcp_ipv6.o | grep sockaddr_in struct sockaddr_in: 16 0 struct sockaddr_in6: 28 0 [acme@newtoy net-2.6.20]$ So I turned sockaddr_len a 'u16', and now: [acme@newtoy net-2.6.20]$ pahole net/ipv6/tcp_ipv6.o inet_connection_sock_af_ops /* /pub/scm/linux/kernel/git/acme/net-2.6.20/include/net/inet_connection_sock.h:38 */ struct inet_connection_sock_af_ops { int (*queue_xmit)(); /* 0 4 */ void (*send_check)(); /* 4 4 */ int (*rebuild_header)(); /* 8 4 */ int (*conn_request)(); /* 12 4 */ struct sock * (*syn_recv_sock)(); /* 16 4 */ int (*remember_stamp)(); /* 20 4 */ u16 net_header_len; /* 24 2 */ u16 sockaddr_len; /* 26 2 */ int (*setsockopt)(); /* 28 4 */ int (*getsockopt)(); /* 32 4 */ int (*compat_setsockopt)(); /* 36 4 */ int (*compat_getsockopt)(); /* 40 4 */ void (*addr2sockaddr)(); /* 44 4 */ }; /* size: 48 */ So we've saved 4 bytes: [acme@newtoy net-2.6.20]$ codiff -sV /tmp/tcp_ipv6.o.before net/ipv6/tcp_ipv6.o /pub/scm/linux/kernel/git/acme/net-2.6.20/net/ipv6/tcp_ipv6.c: struct inet_connection_sock_af_ops | -4 net_header_len; from: __u16 /* 24(0) 2(0) */ to: u16 /* 24(0) 2(0) */ sockaddr_len; from: int /* 48(0) 4(0) */ to: u16 /* 26(0) 2(0) */ 1 struct changed [acme@newtoy net-2.6.20]$ Signed-off-by: Arnaldo Carvalho de Melo --- include/net/inet_connection_sock.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index fd3f1d8cdc4..cccea051e92 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -46,7 +46,8 @@ struct inet_connection_sock_af_ops { struct request_sock *req, struct dst_entry *dst); int (*remember_stamp)(struct sock *sk); - __u16 net_header_len; + u16 net_header_len; + u16 sockaddr_len; int (*setsockopt)(struct sock *sk, int level, int optname, char __user *optval, int optlen); int (*getsockopt)(struct sock *sk, int level, int optname, @@ -58,7 +59,6 @@ struct inet_connection_sock_af_ops { int level, int optname, char __user *optval, int __user *optlen); void (*addr2sockaddr)(struct sock *sk, struct sockaddr *); - int sockaddr_len; }; /** inet_connection_sock - INET connection oriented sock -- cgit v1.2.3-70-g09d2 From d5c42c0ec4f7fd5a4e19e33a2d561758b67c55c8 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Mon, 27 Nov 2006 17:58:02 -0200 Subject: [NET]: Pack struct hh_cache [acme@newtoy net-2.6.20]$ pahole net/ipv4/tcp.o hh_cache /* /pub/scm/linux/kernel/git/acme/net-2.6.20/include/linux/netdevice.h:190 */ struct hh_cache { struct hh_cache * hh_next; /* 0 4 */ atomic_t hh_refcnt; /* 4 4 */ __be16 hh_type; /* 8 2 */ /* XXX 2 bytes hole, try to pack */ int hh_len; /* 12 4 */ int (*hh_output)(); /* 16 4 */ rwlock_t hh_lock; /* 20 36 */ long unsigned int hh_data[24]; /* 56 96 */ }; /* size: 152, sum members: 150, holes: 1, sum holes: 2 */ [acme@newtoy net-2.6.20]$ find net -name "*.[ch]" | xargs grep 'hh_len.\+=' | sort -u net/atm/br2684.c: hh->hh_len = PADLEN + ETH_HLEN; net/ethernet/eth.c: hh->hh_len = ETH_HLEN; net/ipv4/ipconfig.c: int hh_len = LL_RESERVED_SPACE(dev); net/ipv4/ip_output.c: hh_len = LL_RESERVED_SPACE(rt->u.dst.dev); net/ipv4/ip_output.c: int hh_len = LL_RESERVED_SPACE(dev); net/ipv4/netfilter.c: hh_len = (*pskb)->dst->dev->hard_header_len; net/ipv4/raw.c: hh_len = LL_RESERVED_SPACE(rt->u.dst.dev); net/ipv6/ip6_output.c: hh_len = LL_RESERVED_SPACE(rt->u.dst.dev); net/ipv6/netfilter/ip6t_REJECT.c: hh_len = (dst->dev->hard_header_len + 15)&~15; net/ipv6/raw.c: hh_len = LL_RESERVED_SPACE(rt->u.dst.dev); [acme@newtoy net-2.6.20]$ [acme@newtoy net-2.6.20]$ find include -name "*.h" | xargs grep 'define ETH_HLEN' include/linux/if_ether.h:#define ETH_HLEN 14 /* Total octets in header. */ (((dev)->hard_header_len&~(HH_DATA_MOD - 1)) + HH_DATA_MOD) [acme@newtoy net-2.6.20]$ pahole net/ipv4/tcp.o net_device | grep hard_header_len short unsigned int hard_header_len; /* 106 2 */ [acme@newtoy net-2.6.20]$ So I think we're safe in turning hh_len an u16, end result: [acme@newtoy net-2.6.20]$ codiff -sV /tmp/tcp.o.before net/ipv4/tcp.o /pub/scm/linux/kernel/git/acme/net-2.6.20/net/ipv4/tcp.c: struct hh_cache | -4 hh_len; from: int /* 12(0) 4(0) */ to: u16 /* 10(0) 2(0) */ 1 struct changed [acme@newtoy net-2.6.20]$ Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/netdevice.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index caa3c259371..949eada46ce 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -196,7 +196,7 @@ struct hh_cache * NOTE: For VLANs, this will be the * encapuslated type. --BLG */ - int hh_len; /* length of header */ + u16 hh_len; /* length of header */ int (*hh_output)(struct sk_buff *skb); rwlock_t hh_lock; -- cgit v1.2.3-70-g09d2 From 46ca5f5dc4f1c9b5ac02c0090ae8ff4ac8560446 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Mon, 27 Nov 2006 17:58:59 -0200 Subject: [XFRM]: Pack struct xfrm_policy [acme@newtoy net-2.6.20]$ pahole net/ipv4/tcp.o xfrm_policy /* /pub/scm/linux/kernel/git/acme/net-2.6.20/include/linux/security.h:67 */ struct xfrm_policy { struct xfrm_policy * next; /* 0 4 */ struct hlist_node bydst; /* 4 8 */ struct hlist_node byidx; /* 12 8 */ rwlock_t lock; /* 20 36 */ atomic_t refcnt; /* 56 4 */ struct timer_list timer; /* 60 24 */ u8 type; /* 84 1 */ /* XXX 3 bytes hole, try to pack */ u32 priority; /* 88 4 */ u32 index; /* 92 4 */ struct xfrm_selector selector; /* 96 56 */ struct xfrm_lifetime_cfg lft; /* 152 64 */ struct xfrm_lifetime_cur curlft; /* 216 32 */ struct dst_entry * bundles; /* 248 4 */ __u16 family; /* 252 2 */ __u8 action; /* 254 1 */ __u8 flags; /* 255 1 */ __u8 dead; /* 256 1 */ __u8 xfrm_nr; /* 257 1 */ /* XXX 2 bytes hole, try to pack */ struct xfrm_sec_ctx * security; /* 260 4 */ struct xfrm_tmpl xfrm_vec[6]; /* 264 360 */ }; /* size: 624, sum members: 619, holes: 2, sum holes: 5 */ So lets have just one hole instead of two, by moving 'type' to just before 'action', end result: [acme@newtoy net-2.6.20]$ codiff -s /tmp/tcp.o.before net/ipv4/tcp.o /pub/scm/linux/kernel/git/acme/net-2.6.20/net/ipv4/tcp.c: struct xfrm_policy | -4 1 struct changed [acme@newtoy net-2.6.20]$ [acme@newtoy net-2.6.20]$ pahole -c 64 net/ipv4/tcp.o xfrm_policy /* /pub/scm/linux/kernel/git/acme/net-2.6.20/include/linux/security.h:67 */ struct xfrm_policy { struct xfrm_policy * next; /* 0 4 */ struct hlist_node bydst; /* 4 8 */ struct hlist_node byidx; /* 12 8 */ rwlock_t lock; /* 20 36 */ atomic_t refcnt; /* 56 4 */ struct timer_list timer; /* 60 24 */ u32 priority; /* 84 4 */ u32 index; /* 88 4 */ struct xfrm_selector selector; /* 92 56 */ struct xfrm_lifetime_cfg lft; /* 148 64 */ struct xfrm_lifetime_cur curlft; /* 212 32 */ struct dst_entry * bundles; /* 244 4 */ u16 family; /* 248 2 */ u8 type; /* 250 1 */ u8 action; /* 251 1 */ u8 flags; /* 252 1 */ u8 dead; /* 253 1 */ u8 xfrm_nr; /* 254 1 */ /* XXX 1 byte hole, try to pack */ struct xfrm_sec_ctx * security; /* 256 4 */ struct xfrm_tmpl xfrm_vec[6]; /* 260 360 */ }; /* size: 620, sum members: 619, holes: 1, sum holes: 1 */ Are there any fugly data dependencies here? None that I know. In the process changed the removed the __ prefixed types, that are just for userspace visible headers. Signed-off-by: Arnaldo Carvalho de Melo --- include/net/xfrm.h | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 3878a88ff61..984e5c45265 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -340,18 +340,19 @@ struct xfrm_policy atomic_t refcnt; struct timer_list timer; - u8 type; u32 priority; u32 index; struct xfrm_selector selector; struct xfrm_lifetime_cfg lft; struct xfrm_lifetime_cur curlft; struct dst_entry *bundles; - __u16 family; - __u8 action; - __u8 flags; - __u8 dead; - __u8 xfrm_nr; + u16 family; + u8 type; + u8 action; + u8 flags; + u8 dead; + u8 xfrm_nr; + /* XXX 1 byte hole, try to pack */ struct xfrm_sec_ctx *security; struct xfrm_tmpl xfrm_vec[XFRM_MAX_DEPTH]; }; -- cgit v1.2.3-70-g09d2 From 2ff52f282cf287d60e9eda1f3b5ec83e00a86130 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Tue, 28 Nov 2006 00:48:32 -0200 Subject: [TCP]: Change tcp_header_len member in tcp_sock to u16 With this we eliminate the last hole in struct tcp_sock. End result: [acme@newtoy net-2.6.20]$ codiff -sV /tmp/tcp.o.before net/ipv4/tcp.o /pub/scm/linux/kernel/git/acme/net-2.6.20/net/ipv4/tcp.c: struct tcp_sock | -4 tcp_header_len; from: int /* 1000(0) 4(0) */ to: u16 /* 1000(0) 2(0) */ 1 struct changed [acme@newtoy net-2.6.20]$ Now sizeof(tcp_sock) is just... [acme@newtoy net-2.6.20]$ pahole --sizes ../OUTPUT/qemu/net-2.6.20/net/ipv4/tcp.o | grep -w tcp_sock struct tcp_sock: 1500 0 1500 bytes ;-) Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/tcp.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index dd61b172ac6..b42ff0efc2d 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -227,7 +227,8 @@ static inline struct tcp_request_sock *tcp_rsk(const struct request_sock *req) struct tcp_sock { /* inet_connection_sock has to be the first member of tcp_sock */ struct inet_connection_sock inet_conn; - int tcp_header_len; /* Bytes of tcp header to send */ + u16 tcp_header_len; /* Bytes of tcp header to send */ + u16 xmit_size_goal; /* Goal for segmenting output packets */ /* * Header prediction flags @@ -268,8 +269,6 @@ struct tcp_sock { __u32 snd_wnd; /* The window we expect to receive */ __u32 max_window; /* Maximal window ever seen from peer */ __u32 mss_cache; /* Cached effective mss, not including SACKS */ - __u16 xmit_size_goal; /* Goal for segmenting output packets */ - /* XXX Two bytes hole, try to pack */ __u32 window_clamp; /* Maximal window to advertise */ __u32 rcv_ssthresh; /* Current window clamp */ -- cgit v1.2.3-70-g09d2 From 3a137d2065571864be0301b9ebd72ddb01060997 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Tue, 28 Nov 2006 01:12:38 -0200 Subject: [TCP]: Renove the __ prefix on the struct tcp_sock members As this struct is not userland visible at all. Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/tcp.h | 156 ++++++++++++++++++++++++++-------------------------- 1 file changed, 78 insertions(+), 78 deletions(-) (limited to 'include') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index b42ff0efc2d..3cc70d1a350 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -185,17 +185,17 @@ struct tcp_sack_block_wire { }; struct tcp_sack_block { - __u32 start_seq; - __u32 end_seq; + u32 start_seq; + u32 end_seq; }; struct tcp_options_received { /* PAWS/RTTM data */ long ts_recent_stamp;/* Time we stored ts_recent (for aging) */ - __u32 ts_recent; /* Time stamp to echo next */ - __u32 rcv_tsval; /* Time stamp value */ - __u32 rcv_tsecr; /* Time stamp echo reply */ - __u16 saw_tstamp : 1, /* Saw TIMESTAMP on last packet */ + u32 ts_recent; /* Time stamp to echo next */ + u32 rcv_tsval; /* Time stamp value */ + u32 rcv_tsecr; /* Time stamp echo reply */ + u16 saw_tstamp : 1, /* Saw TIMESTAMP on last packet */ tstamp_ok : 1, /* TIMESTAMP seen on SYN packet */ dsack : 1, /* D-SACK is scheduled */ wscale_ok : 1, /* Wscale seen on SYN packet */ @@ -203,10 +203,10 @@ struct tcp_options_received { snd_wscale : 4, /* Window scaling received from sender */ rcv_wscale : 4; /* Window scaling to send to receiver */ /* SACKs data */ - __u8 eff_sacks; /* Size of SACK array to send with next packet */ - __u8 num_sacks; /* Number of SACK blocks */ - __u16 user_mss; /* mss requested by user in ioctl */ - __u16 mss_clamp; /* Maximal mss, negotiated at connection setup */ + u8 eff_sacks; /* Size of SACK array to send with next packet */ + u8 num_sacks; /* Number of SACK blocks */ + u16 user_mss; /* mss requested by user in ioctl */ + u16 mss_clamp; /* Maximal mss, negotiated at connection setup */ }; struct tcp_request_sock { @@ -215,8 +215,8 @@ struct tcp_request_sock { /* Only used by TCP MD5 Signature so far. */ struct tcp_request_sock_ops *af_specific; #endif - __u32 rcv_isn; - __u32 snt_isn; + u32 rcv_isn; + u32 snt_isn; }; static inline struct tcp_request_sock *tcp_rsk(const struct request_sock *req) @@ -241,13 +241,13 @@ struct tcp_sock { * read the code and the spec side by side (and laugh ...) * See RFC793 and RFC1122. The RFC writes these in capitals. */ - __u32 rcv_nxt; /* What we want to receive next */ - __u32 snd_nxt; /* Next sequence we send */ + u32 rcv_nxt; /* What we want to receive next */ + u32 snd_nxt; /* Next sequence we send */ - __u32 snd_una; /* First byte we want an ack for */ - __u32 snd_sml; /* Last byte of the most recently transmitted small packet */ - __u32 rcv_tstamp; /* timestamp of last received ACK (for keepalives) */ - __u32 lsndtime; /* timestamp of last sent data packet (for restart window) */ + u32 snd_una; /* First byte we want an ack for */ + u32 snd_sml; /* Last byte of the most recently transmitted small packet */ + u32 rcv_tstamp; /* timestamp of last received ACK (for keepalives) */ + u32 lsndtime; /* timestamp of last sent data packet (for restart window) */ /* Data for direct copy to user */ struct { @@ -265,30 +265,30 @@ struct tcp_sock { #endif } ucopy; - __u32 snd_wl1; /* Sequence for window update */ - __u32 snd_wnd; /* The window we expect to receive */ - __u32 max_window; /* Maximal window ever seen from peer */ - __u32 mss_cache; /* Cached effective mss, not including SACKS */ + u32 snd_wl1; /* Sequence for window update */ + u32 snd_wnd; /* The window we expect to receive */ + u32 max_window; /* Maximal window ever seen from peer */ + u32 mss_cache; /* Cached effective mss, not including SACKS */ - __u32 window_clamp; /* Maximal window to advertise */ - __u32 rcv_ssthresh; /* Current window clamp */ + u32 window_clamp; /* Maximal window to advertise */ + u32 rcv_ssthresh; /* Current window clamp */ - __u32 frto_highmark; /* snd_nxt when RTO occurred */ - __u8 reordering; /* Packet reordering metric. */ - __u8 frto_counter; /* Number of new acks after RTO */ - __u8 nonagle; /* Disable Nagle algorithm? */ - __u8 keepalive_probes; /* num of allowed keep alive probes */ + u32 frto_highmark; /* snd_nxt when RTO occurred */ + u8 reordering; /* Packet reordering metric. */ + u8 frto_counter; /* Number of new acks after RTO */ + u8 nonagle; /* Disable Nagle algorithm? */ + u8 keepalive_probes; /* num of allowed keep alive probes */ /* RTT measurement */ - __u32 srtt; /* smoothed round trip time << 3 */ - __u32 mdev; /* medium deviation */ - __u32 mdev_max; /* maximal mdev for the last rtt period */ - __u32 rttvar; /* smoothed mdev_max */ - __u32 rtt_seq; /* sequence number to update rttvar */ - - __u32 packets_out; /* Packets which are "in flight" */ - __u32 left_out; /* Packets which leaved network */ - __u32 retrans_out; /* Retransmitted packets out */ + u32 srtt; /* smoothed round trip time << 3 */ + u32 mdev; /* medium deviation */ + u32 mdev_max; /* maximal mdev for the last rtt period */ + u32 rttvar; /* smoothed mdev_max */ + u32 rtt_seq; /* sequence number to update rttvar */ + + u32 packets_out; /* Packets which are "in flight" */ + u32 left_out; /* Packets which leaved network */ + u32 retrans_out; /* Retransmitted packets out */ /* * Options received (usually on last packet, some only on SYN packets). */ @@ -297,20 +297,20 @@ struct tcp_sock { /* * Slow start and congestion control (see also Nagle, and Karn & Partridge) */ - __u32 snd_ssthresh; /* Slow start size threshold */ - __u32 snd_cwnd; /* Sending congestion window */ - __u16 snd_cwnd_cnt; /* Linear increase counter */ - __u16 snd_cwnd_clamp; /* Do not allow snd_cwnd to grow above this */ - __u32 snd_cwnd_used; - __u32 snd_cwnd_stamp; + u32 snd_ssthresh; /* Slow start size threshold */ + u32 snd_cwnd; /* Sending congestion window */ + u16 snd_cwnd_cnt; /* Linear increase counter */ + u16 snd_cwnd_clamp; /* Do not allow snd_cwnd to grow above this */ + u32 snd_cwnd_used; + u32 snd_cwnd_stamp; struct sk_buff_head out_of_order_queue; /* Out of order segments go here */ - __u32 rcv_wnd; /* Current receiver window */ - __u32 rcv_wup; /* rcv_nxt on last window update sent */ - __u32 write_seq; /* Tail(+1) of data held in tcp send buffer */ - __u32 pushed_seq; /* Last pushed seq, required to talk to windows */ - __u32 copied_seq; /* Head of yet unread data */ + u32 rcv_wnd; /* Current receiver window */ + u32 rcv_wup; /* rcv_nxt on last window update sent */ + u32 write_seq; /* Tail(+1) of data held in tcp send buffer */ + u32 pushed_seq; /* Last pushed seq, required to talk to windows */ + u32 copied_seq; /* Head of yet unread data */ /* SACKs data */ struct tcp_sack_block duplicate_sack[1]; /* D-SACK block */ @@ -331,26 +331,26 @@ struct tcp_sock { int retransmit_cnt_hint; int forward_cnt_hint; - __u16 advmss; /* Advertised MSS */ - __u16 prior_ssthresh; /* ssthresh saved at recovery start */ - __u32 lost_out; /* Lost packets */ - __u32 sacked_out; /* SACK'd packets */ - __u32 fackets_out; /* FACK'd packets */ - __u32 high_seq; /* snd_nxt at onset of congestion */ + u16 advmss; /* Advertised MSS */ + u16 prior_ssthresh; /* ssthresh saved at recovery start */ + u32 lost_out; /* Lost packets */ + u32 sacked_out; /* SACK'd packets */ + u32 fackets_out; /* FACK'd packets */ + u32 high_seq; /* snd_nxt at onset of congestion */ - __u32 retrans_stamp; /* Timestamp of the last retransmit, + u32 retrans_stamp; /* Timestamp of the last retransmit, * also used in SYN-SENT to remember stamp of * the first SYN. */ - __u32 undo_marker; /* tracking retrans started here. */ + u32 undo_marker; /* tracking retrans started here. */ int undo_retrans; /* number of undoable retransmissions. */ - __u32 urg_seq; /* Seq of received urgent pointer */ - __u16 urg_data; /* Saved octet of OOB data and control flags */ - __u8 urg_mode; /* In urgent mode */ - __u8 ecn_flags; /* ECN status bits. */ - __u32 snd_up; /* Urgent pointer */ + u32 urg_seq; /* Seq of received urgent pointer */ + u16 urg_data; /* Saved octet of OOB data and control flags */ + u8 urg_mode; /* In urgent mode */ + u8 ecn_flags; /* ECN status bits. */ + u32 snd_up; /* Urgent pointer */ - __u32 total_retrans; /* Total retransmits for entire connection */ - __u32 bytes_acked; /* Appropriate Byte Counting - RFC3465 */ + u32 total_retrans; /* Total retransmits for entire connection */ + u32 bytes_acked; /* Appropriate Byte Counting - RFC3465 */ unsigned int keepalive_time; /* time before keep alive takes place */ unsigned int keepalive_intvl; /* time interval between keep alive probes */ @@ -358,26 +358,26 @@ struct tcp_sock { unsigned long last_synq_overflow; - __u32 tso_deferred; + u32 tso_deferred; /* Receiver side RTT estimation */ struct { - __u32 rtt; - __u32 seq; - __u32 time; + u32 rtt; + u32 seq; + u32 time; } rcv_rtt_est; /* Receiver queue space */ struct { int space; - __u32 seq; - __u32 time; + u32 seq; + u32 time; } rcvq_space; /* TCP-specific MTU probe information. */ struct { - __u32 probe_seq_start; - __u32 probe_seq_end; + u32 probe_seq_start; + u32 probe_seq_end; } mtu_probe; #ifdef CONFIG_TCP_MD5SIG @@ -396,14 +396,14 @@ static inline struct tcp_sock *tcp_sk(const struct sock *sk) struct tcp_timewait_sock { struct inet_timewait_sock tw_sk; - __u32 tw_rcv_nxt; - __u32 tw_snd_nxt; - __u32 tw_rcv_wnd; - __u32 tw_ts_recent; + u32 tw_rcv_nxt; + u32 tw_snd_nxt; + u32 tw_rcv_wnd; + u32 tw_ts_recent; long tw_ts_recent_stamp; #ifdef CONFIG_TCP_MD5SIG - __u16 tw_md5_keylen; - __u8 tw_md5_key[TCP_MD5SIG_MAXKEYLEN]; + u16 tw_md5_keylen; + u8 tw_md5_key[TCP_MD5SIG_MAXKEYLEN]; #endif }; -- cgit v1.2.3-70-g09d2 From ee41e2dff1a0ac548f12871f4bb23fe9e69e13eb Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Tue, 28 Nov 2006 03:11:33 -0200 Subject: [INET]: Change protocol field in struct inet_protosw to u16 [acme@newtoy net-2.6.20]$ pahole /tmp/tcp_ipv6.o inet_protosw /* /pub/scm/linux/kernel/git/acme/net-2.6.20/include/net/protocol.h:69 */ struct inet_protosw { struct list_head list; /* 0 8 */ short unsigned int type; /* 8 2 */ /* XXX 2 bytes hole, try to pack */ int protocol; /* 12 4 */ struct proto * prot; /* 16 4 */ const struct proto_ops * ops; /* 20 4 */ int capability; /* 24 4 */ char no_check; /* 28 1 */ unsigned char flags; /* 29 1 */ }; /* size: 32, sum members: 28, holes: 1, sum holes: 2, padding: 2 */ So that we can kill that hole, protocol can only go all the way to 255 (RAW). Signed-off-by: Arnaldo Carvalho de Melo --- include/net/protocol.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/protocol.h b/include/net/protocol.h index 28c4cac3f8a..105bf12b0c7 100644 --- a/include/net/protocol.h +++ b/include/net/protocol.h @@ -71,7 +71,7 @@ struct inet_protosw { /* These two fields form the lookup key. */ unsigned short type; /* This is the 2nd argument to socket(2). */ - int protocol; /* This is the L4 protocol number. */ + unsigned short protocol; /* This is the L4 protocol number. */ struct proto *prot; const struct proto_ops *ops; -- cgit v1.2.3-70-g09d2 From 4384260443efe90a2ec0d907568dbc58ae792cd0 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Tue, 28 Nov 2006 18:14:10 -0200 Subject: [DCCP]: Remove allocation of sysctl numbers This is in response to a request sent earlier by Eric W. Biederman and replaces all sysctl numbers for net.dccp.default with CTL_UNNUMBERED. It has been tested to compile and to work. Commiter note: I've removed the use of CTL_UNNUMBERED, not setting .ctl_name sets it to 0, that is the what CTL_UNNUMBERED is, reason is to avoid unneeded source code cluttering. Signed-off-by: Gerrit Renker Signed-off-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/sysctl.h | 14 -------------- net/dccp/sysctl.c | 10 ---------- 2 files changed, 24 deletions(-) (limited to 'include') diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 61dd99c6c2a..94316a98e0d 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -606,20 +606,6 @@ enum { NET_DCCP_DEFAULT=1, }; -/* /proc/sys/net/dccp/default */ -enum { - NET_DCCP_DEFAULT_SEQ_WINDOW = 1, - NET_DCCP_DEFAULT_RX_CCID = 2, - NET_DCCP_DEFAULT_TX_CCID = 3, - NET_DCCP_DEFAULT_ACK_RATIO = 4, - NET_DCCP_DEFAULT_SEND_ACKVEC = 5, - NET_DCCP_DEFAULT_SEND_NDP = 6, - NET_DCCP_DEFAULT_REQ_RETRIES = 7, - NET_DCCP_DEFAULT_RETRIES1 = 8, - NET_DCCP_DEFAULT_RETRIES2 = 9, - NET_DCCP_DEFAULT_TX_QLEN = 10, -}; - /* /proc/sys/net/ipx */ enum { NET_IPX_PPROP_BROADCASTING=1, diff --git a/net/dccp/sysctl.c b/net/dccp/sysctl.c index 4775ba3faa0..fdcfca3e920 100644 --- a/net/dccp/sysctl.c +++ b/net/dccp/sysctl.c @@ -20,7 +20,6 @@ static struct ctl_table dccp_default_table[] = { { - .ctl_name = NET_DCCP_DEFAULT_SEQ_WINDOW, .procname = "seq_window", .data = &sysctl_dccp_feat_sequence_window, .maxlen = sizeof(sysctl_dccp_feat_sequence_window), @@ -28,7 +27,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_RX_CCID, .procname = "rx_ccid", .data = &sysctl_dccp_feat_rx_ccid, .maxlen = sizeof(sysctl_dccp_feat_rx_ccid), @@ -36,7 +34,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_TX_CCID, .procname = "tx_ccid", .data = &sysctl_dccp_feat_tx_ccid, .maxlen = sizeof(sysctl_dccp_feat_tx_ccid), @@ -44,7 +41,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_ACK_RATIO, .procname = "ack_ratio", .data = &sysctl_dccp_feat_ack_ratio, .maxlen = sizeof(sysctl_dccp_feat_ack_ratio), @@ -52,7 +48,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_SEND_ACKVEC, .procname = "send_ackvec", .data = &sysctl_dccp_feat_send_ack_vector, .maxlen = sizeof(sysctl_dccp_feat_send_ack_vector), @@ -60,7 +55,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_SEND_NDP, .procname = "send_ndp", .data = &sysctl_dccp_feat_send_ndp_count, .maxlen = sizeof(sysctl_dccp_feat_send_ndp_count), @@ -68,7 +62,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_REQ_RETRIES, .procname = "request_retries", .data = &sysctl_dccp_request_retries, .maxlen = sizeof(sysctl_dccp_request_retries), @@ -76,7 +69,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_RETRIES1, .procname = "retries1", .data = &sysctl_dccp_retries1, .maxlen = sizeof(sysctl_dccp_retries1), @@ -84,7 +76,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_RETRIES2, .procname = "retries2", .data = &sysctl_dccp_retries2, .maxlen = sizeof(sysctl_dccp_retries2), @@ -92,7 +83,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_TX_QLEN, .procname = "tx_qlen", .data = &sysctl_dccp_tx_qlen, .maxlen = sizeof(sysctl_dccp_tx_qlen), -- cgit v1.2.3-70-g09d2 From 5aed324369c94a2c38469c8288e42eb1a9fac400 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Tue, 28 Nov 2006 19:33:36 -0200 Subject: [DCCP]: Tidy up unused structures This removes and cleans up unused variables and structures which have become unnecessary following the introduction of the EWMA patch to automatically track the CCID 3 receiver/sender packet sizes `s'. It deprecates the PACKET_SIZE socket option by returning an error code and printing a deprecation warning if an application tries to read or write this socket option. Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/dccp.h | 4 +--- net/dccp/ccids/ccid3.h | 4 ---- net/dccp/proto.c | 8 ++++---- 3 files changed, 5 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/linux/dccp.h b/include/linux/dccp.h index 72cc355d7a0..ed6cc8962d8 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -200,7 +200,7 @@ struct dccp_so_feat { }; /* DCCP socket options */ -#define DCCP_SOCKOPT_PACKET_SIZE 1 +#define DCCP_SOCKOPT_PACKET_SIZE 1 /* XXX deprecated, without effect */ #define DCCP_SOCKOPT_SERVICE 2 #define DCCP_SOCKOPT_CHANGE_L 3 #define DCCP_SOCKOPT_CHANGE_R 4 @@ -460,7 +460,6 @@ struct dccp_ackvec; * @dccps_service_list - second .. last service code on passive socket * @dccps_timestamp_time - time of latest TIMESTAMP option * @dccps_timestamp_echo - latest timestamp received on a TIMESTAMP option - * @dccps_packet_size - Set thru setsockopt * @dccps_l_ack_ratio - * @dccps_r_ack_ratio - * @dccps_pcslen - sender partial checksum coverage (via sockopt) @@ -495,7 +494,6 @@ struct dccp_sock { struct dccp_service_list *dccps_service_list; struct timeval dccps_timestamp_time; __u32 dccps_timestamp_echo; - __u32 dccps_packet_size; __u16 dccps_l_ack_ratio; __u16 dccps_r_ack_ratio; __u16 dccps_pcslen; diff --git a/net/dccp/ccids/ccid3.h b/net/dccp/ccids/ccid3.h index 970921700ce..dbb884426df 100644 --- a/net/dccp/ccids/ccid3.h +++ b/net/dccp/ccids/ccid3.h @@ -42,10 +42,6 @@ #include #include "../ccid.h" -#define TFRC_MIN_PACKET_SIZE 16 -#define TFRC_STD_PACKET_SIZE 256 -#define TFRC_MAX_PACKET_SIZE 65535 - /* Two seconds as per RFC 3448 4.2 */ #define TFRC_INITIAL_TIMEOUT (2 * USEC_PER_SEC) diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 3c44d502e5c..2604e34d8f3 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -470,7 +470,8 @@ static int do_dccp_setsockopt(struct sock *sk, int level, int optname, lock_sock(sk); switch (optname) { case DCCP_SOCKOPT_PACKET_SIZE: - dp->dccps_packet_size = val; + DCCP_WARN("sockopt(PACKET_SIZE) is deprecated: fix your app\n"); + err = -EINVAL; break; case DCCP_SOCKOPT_CHANGE_L: if (optlen != sizeof(struct dccp_so_feat)) @@ -581,9 +582,8 @@ static int do_dccp_getsockopt(struct sock *sk, int level, int optname, switch (optname) { case DCCP_SOCKOPT_PACKET_SIZE: - val = dp->dccps_packet_size; - len = sizeof(dp->dccps_packet_size); - break; + DCCP_WARN("sockopt(PACKET_SIZE) is deprecated: fix your app\n"); + return -EINVAL; case DCCP_SOCKOPT_SERVICE: return dccp_getsockopt_service(sk, len, (__be32 __user *)optval, optlen); -- cgit v1.2.3-70-g09d2 From 77ab9cff0f4112703df3ef7903c1a15adb967114 Mon Sep 17 00:00:00 2001 From: Martin Josefsson Date: Wed, 29 Nov 2006 02:34:58 +0100 Subject: [NETFILTER]: nf_conntrack: split out expectation handling This patch splits out expectation handling into its own file nf_conntrack_expect.c Signed-off-by: Martin Josefsson Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack.h | 49 +--- include/net/netfilter/nf_conntrack_core.h | 7 + include/net/netfilter/nf_conntrack_expect.h | 72 ++++++ include/net/netfilter/nf_conntrack_helper.h | 10 - net/netfilter/Makefile | 2 +- net/netfilter/nf_conntrack_core.c | 265 +------------------- net/netfilter/nf_conntrack_expect.c | 365 ++++++++++++++++++++++++++++ net/netfilter/nf_conntrack_ftp.c | 1 + net/netfilter/nf_conntrack_netlink.c | 1 + net/netfilter/nf_conntrack_standalone.c | 81 +----- 10 files changed, 452 insertions(+), 401 deletions(-) create mode 100644 include/net/netfilter/nf_conntrack_expect.h create mode 100644 net/netfilter/nf_conntrack_expect.c (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 1fbd8193d5f..9d2581fc04b 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -124,44 +124,6 @@ struct nf_conn char data[0]; }; -struct nf_conntrack_expect -{ - /* Internal linked list (global expectation list) */ - struct list_head list; - - /* We expect this tuple, with the following mask */ - struct nf_conntrack_tuple tuple, mask; - - /* Function to call after setup and insertion */ - void (*expectfn)(struct nf_conn *new, - struct nf_conntrack_expect *this); - - /* The conntrack of the master connection */ - struct nf_conn *master; - - /* Timer function; deletes the expectation. */ - struct timer_list timeout; - - /* Usage count. */ - atomic_t use; - - /* Unique ID */ - unsigned int id; - - /* Flags */ - unsigned int flags; - -#ifdef CONFIG_NF_NAT_NEEDED - /* This is the original per-proto part, used to map the - * expected connection the way the recipient expects. */ - union nf_conntrack_manip_proto saved_proto; - /* Direction relative to the master connection. */ - enum ip_conntrack_dir dir; -#endif -}; - -#define NF_CT_EXPECT_PERMANENT 0x1 - static inline struct nf_conn * nf_ct_tuplehash_to_ctrack(const struct nf_conntrack_tuple_hash *hash) { @@ -208,16 +170,6 @@ __nf_conntrack_find(const struct nf_conntrack_tuple *tuple, extern void nf_conntrack_hash_insert(struct nf_conn *ct); -extern struct nf_conntrack_expect * -__nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple); - -extern struct nf_conntrack_expect * -nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple); - -extern void nf_ct_unlink_expect(struct nf_conntrack_expect *exp); - -extern void nf_ct_remove_expectations(struct nf_conn *ct); - extern void nf_conntrack_flush(void); extern struct nf_conntrack_helper * @@ -295,6 +247,7 @@ extern int nf_conntrack_checksum; #ifdef CONFIG_NF_CONNTRACK_EVENTS #include #include +#include struct nf_conntrack_ecache { struct nf_conn *ct; diff --git a/include/net/netfilter/nf_conntrack_core.h b/include/net/netfilter/nf_conntrack_core.h index da254525a4c..84a8e01941f 100644 --- a/include/net/netfilter/nf_conntrack_core.h +++ b/include/net/netfilter/nf_conntrack_core.h @@ -13,6 +13,8 @@ #define _NF_CONNTRACK_CORE_H #include +#include +#include /* This header is used to share core functionality between the standalone connection tracking module, and the compatibility layer's use @@ -70,6 +72,11 @@ static inline int nf_conntrack_confirm(struct sk_buff **pskb) extern void __nf_conntrack_attach(struct sk_buff *nskb, struct sk_buff *skb); +int +print_tuple(struct seq_file *s, const struct nf_conntrack_tuple *tuple, + struct nf_conntrack_l3proto *l3proto, + struct nf_conntrack_protocol *proto); + extern struct list_head *nf_conntrack_hash; extern struct list_head nf_conntrack_expect_list; extern rwlock_t nf_conntrack_lock ; diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h new file mode 100644 index 00000000000..5aa483e0345 --- /dev/null +++ b/include/net/netfilter/nf_conntrack_expect.h @@ -0,0 +1,72 @@ +/* + * connection tracking expectations. + */ + +#ifndef _NF_CONNTRACK_EXPECT_H +#define _NF_CONNTRACK_EXPECT_H +#include + +extern struct list_head nf_conntrack_expect_list; +extern kmem_cache_t *nf_conntrack_expect_cachep; +extern struct file_operations exp_file_ops; + +struct nf_conntrack_expect +{ + /* Internal linked list (global expectation list) */ + struct list_head list; + + /* We expect this tuple, with the following mask */ + struct nf_conntrack_tuple tuple, mask; + + /* Function to call after setup and insertion */ + void (*expectfn)(struct nf_conn *new, + struct nf_conntrack_expect *this); + + /* The conntrack of the master connection */ + struct nf_conn *master; + + /* Timer function; deletes the expectation. */ + struct timer_list timeout; + + /* Usage count. */ + atomic_t use; + + /* Unique ID */ + unsigned int id; + + /* Flags */ + unsigned int flags; + +#ifdef CONFIG_NF_NAT_NEEDED + /* This is the original per-proto part, used to map the + * expected connection the way the recipient expects. */ + union nf_conntrack_manip_proto saved_proto; + /* Direction relative to the master connection. */ + enum ip_conntrack_dir dir; +#endif +}; + +#define NF_CT_EXPECT_PERMANENT 0x1 + + +struct nf_conntrack_expect * +__nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple); + +struct nf_conntrack_expect * +nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple); + +struct nf_conntrack_expect * +find_expectation(const struct nf_conntrack_tuple *tuple); + +void nf_ct_unlink_expect(struct nf_conntrack_expect *exp); +void nf_ct_remove_expectations(struct nf_conn *ct); +void nf_conntrack_unexpect_related(struct nf_conntrack_expect *exp); + +/* Allocate space for an expectation: this is mandatory before calling + nf_conntrack_expect_related. You will have to call put afterwards. */ +struct nf_conntrack_expect *nf_conntrack_expect_alloc(struct nf_conn *me); +void nf_conntrack_expect_put(struct nf_conntrack_expect *exp); +int nf_conntrack_expect_related(struct nf_conntrack_expect *expect); + +#endif /*_NF_CONNTRACK_EXPECT_H*/ + diff --git a/include/net/netfilter/nf_conntrack_helper.h b/include/net/netfilter/nf_conntrack_helper.h index 86ec8174ad0..3cbd13e2216 100644 --- a/include/net/netfilter/nf_conntrack_helper.h +++ b/include/net/netfilter/nf_conntrack_helper.h @@ -40,14 +40,4 @@ struct nf_conntrack_helper extern int nf_conntrack_helper_register(struct nf_conntrack_helper *); extern void nf_conntrack_helper_unregister(struct nf_conntrack_helper *); -/* Allocate space for an expectation: this is mandatory before calling - nf_conntrack_expect_related. You will have to call put afterwards. */ -extern struct nf_conntrack_expect * -nf_conntrack_expect_alloc(struct nf_conn *master); -extern void nf_conntrack_expect_put(struct nf_conntrack_expect *exp); - -/* Add an expected connection: can have more than one per connection */ -extern int nf_conntrack_expect_related(struct nf_conntrack_expect *exp); -extern void nf_conntrack_unexpect_related(struct nf_conntrack_expect *exp); - #endif /*_NF_CONNTRACK_HELPER_H*/ diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index a74be492fd0..d33b9ac01aa 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -1,5 +1,5 @@ netfilter-objs := core.o nf_log.o nf_queue.o nf_sockopt.o -nf_conntrack-objs := nf_conntrack_core.o nf_conntrack_standalone.o nf_conntrack_l3proto_generic.o nf_conntrack_proto_generic.o nf_conntrack_proto_tcp.o nf_conntrack_proto_udp.o +nf_conntrack-objs := nf_conntrack_core.o nf_conntrack_standalone.o nf_conntrack_expect.o nf_conntrack_l3proto_generic.o nf_conntrack_proto_generic.o nf_conntrack_proto_tcp.o nf_conntrack_proto_udp.o obj-$(CONFIG_NETFILTER) = netfilter.o diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index de0567b1f42..72fbae4e594 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -55,6 +55,7 @@ #include #include #include +#include #include #include @@ -72,21 +73,19 @@ DEFINE_RWLOCK(nf_conntrack_lock); atomic_t nf_conntrack_count = ATOMIC_INIT(0); void (*nf_conntrack_destroyed)(struct nf_conn *conntrack) = NULL; -LIST_HEAD(nf_conntrack_expect_list); struct nf_conntrack_protocol **nf_ct_protos[PF_MAX] __read_mostly; struct nf_conntrack_l3proto *nf_ct_l3protos[PF_MAX] __read_mostly; static LIST_HEAD(helpers); unsigned int nf_conntrack_htable_size __read_mostly = 0; int nf_conntrack_max __read_mostly; struct list_head *nf_conntrack_hash __read_mostly; -static kmem_cache_t *nf_conntrack_expect_cachep __read_mostly; struct nf_conn nf_conntrack_untracked; unsigned int nf_ct_log_invalid __read_mostly; static LIST_HEAD(unconfirmed); static int nf_conntrack_vmalloc __read_mostly; static unsigned int nf_conntrack_next_id; -static unsigned int nf_conntrack_expect_next_id; + #ifdef CONFIG_NF_CONNTRACK_EVENTS ATOMIC_NOTIFIER_HEAD(nf_conntrack_chain); ATOMIC_NOTIFIER_HEAD(nf_conntrack_expect_chain); @@ -438,103 +437,6 @@ nf_ct_invert_tuple(struct nf_conntrack_tuple *inverse, return protocol->invert_tuple(inverse, orig); } -/* nf_conntrack_expect helper functions */ -void nf_ct_unlink_expect(struct nf_conntrack_expect *exp) -{ - struct nf_conn_help *master_help = nfct_help(exp->master); - - NF_CT_ASSERT(master_help); - ASSERT_WRITE_LOCK(&nf_conntrack_lock); - NF_CT_ASSERT(!timer_pending(&exp->timeout)); - - list_del(&exp->list); - NF_CT_STAT_INC(expect_delete); - master_help->expecting--; - nf_conntrack_expect_put(exp); -} - -static void expectation_timed_out(unsigned long ul_expect) -{ - struct nf_conntrack_expect *exp = (void *)ul_expect; - - write_lock_bh(&nf_conntrack_lock); - nf_ct_unlink_expect(exp); - write_unlock_bh(&nf_conntrack_lock); - nf_conntrack_expect_put(exp); -} - -struct nf_conntrack_expect * -__nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple) -{ - struct nf_conntrack_expect *i; - - list_for_each_entry(i, &nf_conntrack_expect_list, list) { - if (nf_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask)) - return i; - } - return NULL; -} - -/* Just find a expectation corresponding to a tuple. */ -struct nf_conntrack_expect * -nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple) -{ - struct nf_conntrack_expect *i; - - read_lock_bh(&nf_conntrack_lock); - i = __nf_conntrack_expect_find(tuple); - if (i) - atomic_inc(&i->use); - read_unlock_bh(&nf_conntrack_lock); - - return i; -} - -/* If an expectation for this connection is found, it gets delete from - * global list then returned. */ -static struct nf_conntrack_expect * -find_expectation(const struct nf_conntrack_tuple *tuple) -{ - struct nf_conntrack_expect *i; - - list_for_each_entry(i, &nf_conntrack_expect_list, list) { - /* If master is not in hash table yet (ie. packet hasn't left - this machine yet), how can other end know about expected? - Hence these are not the droids you are looking for (if - master ct never got confirmed, we'd hold a reference to it - and weird things would happen to future packets). */ - if (nf_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask) - && nf_ct_is_confirmed(i->master)) { - if (i->flags & NF_CT_EXPECT_PERMANENT) { - atomic_inc(&i->use); - return i; - } else if (del_timer(&i->timeout)) { - nf_ct_unlink_expect(i); - return i; - } - } - } - return NULL; -} - -/* delete all expectations for this conntrack */ -void nf_ct_remove_expectations(struct nf_conn *ct) -{ - struct nf_conntrack_expect *i, *tmp; - struct nf_conn_help *help = nfct_help(ct); - - /* Optimization: most connection never expect any others. */ - if (!help || help->expecting == 0) - return; - - list_for_each_entry_safe(i, tmp, &nf_conntrack_expect_list, list) { - if (i->master == ct && del_timer(&i->timeout)) { - nf_ct_unlink_expect(i); - nf_conntrack_expect_put(i); - } - } -} - static void clean_from_lists(struct nf_conn *ct) { @@ -1133,169 +1035,6 @@ int nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse, orig->dst.protonum)); } -/* Would two expected things clash? */ -static inline int expect_clash(const struct nf_conntrack_expect *a, - const struct nf_conntrack_expect *b) -{ - /* Part covered by intersection of masks must be unequal, - otherwise they clash */ - struct nf_conntrack_tuple intersect_mask; - int count; - - intersect_mask.src.l3num = a->mask.src.l3num & b->mask.src.l3num; - intersect_mask.src.u.all = a->mask.src.u.all & b->mask.src.u.all; - intersect_mask.dst.u.all = a->mask.dst.u.all & b->mask.dst.u.all; - intersect_mask.dst.protonum = a->mask.dst.protonum - & b->mask.dst.protonum; - - for (count = 0; count < NF_CT_TUPLE_L3SIZE; count++){ - intersect_mask.src.u3.all[count] = - a->mask.src.u3.all[count] & b->mask.src.u3.all[count]; - } - - for (count = 0; count < NF_CT_TUPLE_L3SIZE; count++){ - intersect_mask.dst.u3.all[count] = - a->mask.dst.u3.all[count] & b->mask.dst.u3.all[count]; - } - - return nf_ct_tuple_mask_cmp(&a->tuple, &b->tuple, &intersect_mask); -} - -static inline int expect_matches(const struct nf_conntrack_expect *a, - const struct nf_conntrack_expect *b) -{ - return a->master == b->master - && nf_ct_tuple_equal(&a->tuple, &b->tuple) - && nf_ct_tuple_equal(&a->mask, &b->mask); -} - -/* Generally a bad idea to call this: could have matched already. */ -void nf_conntrack_unexpect_related(struct nf_conntrack_expect *exp) -{ - struct nf_conntrack_expect *i; - - write_lock_bh(&nf_conntrack_lock); - /* choose the the oldest expectation to evict */ - list_for_each_entry_reverse(i, &nf_conntrack_expect_list, list) { - if (expect_matches(i, exp) && del_timer(&i->timeout)) { - nf_ct_unlink_expect(i); - write_unlock_bh(&nf_conntrack_lock); - nf_conntrack_expect_put(i); - return; - } - } - write_unlock_bh(&nf_conntrack_lock); -} - -/* We don't increase the master conntrack refcount for non-fulfilled - * conntracks. During the conntrack destruction, the expectations are - * always killed before the conntrack itself */ -struct nf_conntrack_expect *nf_conntrack_expect_alloc(struct nf_conn *me) -{ - struct nf_conntrack_expect *new; - - new = kmem_cache_alloc(nf_conntrack_expect_cachep, GFP_ATOMIC); - if (!new) { - DEBUGP("expect_related: OOM allocating expect\n"); - return NULL; - } - new->master = me; - atomic_set(&new->use, 1); - return new; -} - -void nf_conntrack_expect_put(struct nf_conntrack_expect *exp) -{ - if (atomic_dec_and_test(&exp->use)) - kmem_cache_free(nf_conntrack_expect_cachep, exp); -} - -static void nf_conntrack_expect_insert(struct nf_conntrack_expect *exp) -{ - struct nf_conn_help *master_help = nfct_help(exp->master); - - atomic_inc(&exp->use); - master_help->expecting++; - list_add(&exp->list, &nf_conntrack_expect_list); - - init_timer(&exp->timeout); - exp->timeout.data = (unsigned long)exp; - exp->timeout.function = expectation_timed_out; - exp->timeout.expires = jiffies + master_help->helper->timeout * HZ; - add_timer(&exp->timeout); - - exp->id = ++nf_conntrack_expect_next_id; - atomic_inc(&exp->use); - NF_CT_STAT_INC(expect_create); -} - -/* Race with expectations being used means we could have none to find; OK. */ -static void evict_oldest_expect(struct nf_conn *master) -{ - struct nf_conntrack_expect *i; - - list_for_each_entry_reverse(i, &nf_conntrack_expect_list, list) { - if (i->master == master) { - if (del_timer(&i->timeout)) { - nf_ct_unlink_expect(i); - nf_conntrack_expect_put(i); - } - break; - } - } -} - -static inline int refresh_timer(struct nf_conntrack_expect *i) -{ - struct nf_conn_help *master_help = nfct_help(i->master); - - if (!del_timer(&i->timeout)) - return 0; - - i->timeout.expires = jiffies + master_help->helper->timeout*HZ; - add_timer(&i->timeout); - return 1; -} - -int nf_conntrack_expect_related(struct nf_conntrack_expect *expect) -{ - struct nf_conntrack_expect *i; - struct nf_conn *master = expect->master; - struct nf_conn_help *master_help = nfct_help(master); - int ret; - - NF_CT_ASSERT(master_help); - - DEBUGP("nf_conntrack_expect_related %p\n", related_to); - DEBUGP("tuple: "); NF_CT_DUMP_TUPLE(&expect->tuple); - DEBUGP("mask: "); NF_CT_DUMP_TUPLE(&expect->mask); - - write_lock_bh(&nf_conntrack_lock); - list_for_each_entry(i, &nf_conntrack_expect_list, list) { - if (expect_matches(i, expect)) { - /* Refresh timer: if it's dying, ignore.. */ - if (refresh_timer(i)) { - ret = 0; - goto out; - } - } else if (expect_clash(i, expect)) { - ret = -EBUSY; - goto out; - } - } - /* Will be over limit? */ - if (master_help->helper->max_expected && - master_help->expecting >= master_help->helper->max_expected) - evict_oldest_expect(master); - - nf_conntrack_expect_insert(expect); - nf_conntrack_expect_event(IPEXP_NEW, expect); - ret = 0; -out: - write_unlock_bh(&nf_conntrack_lock); - return ret; -} - int nf_conntrack_helper_register(struct nf_conntrack_helper *me) { int ret; diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c new file mode 100644 index 00000000000..076e678b318 --- /dev/null +++ b/net/netfilter/nf_conntrack_expect.c @@ -0,0 +1,365 @@ +/* Expectation handling for nf_conntrack. */ + +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * (C) 2003,2004 USAGI/WIDE Project + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +LIST_HEAD(nf_conntrack_expect_list); +kmem_cache_t *nf_conntrack_expect_cachep __read_mostly; +DECLARE_PER_CPU(struct ip_conntrack_stat, nf_conntrack_stat); +static unsigned int nf_conntrack_expect_next_id; + +/* nf_conntrack_expect helper functions */ +void nf_ct_unlink_expect(struct nf_conntrack_expect *exp) +{ + struct nf_conn_help *master_help = nfct_help(exp->master); + + NF_CT_ASSERT(master_help); + NF_CT_ASSERT(!timer_pending(&exp->timeout)); + + list_del(&exp->list); + NF_CT_STAT_INC(expect_delete); + master_help->expecting--; + nf_conntrack_expect_put(exp); +} + +static void expectation_timed_out(unsigned long ul_expect) +{ + struct nf_conntrack_expect *exp = (void *)ul_expect; + + write_lock_bh(&nf_conntrack_lock); + nf_ct_unlink_expect(exp); + write_unlock_bh(&nf_conntrack_lock); + nf_conntrack_expect_put(exp); +} + +struct nf_conntrack_expect * +__nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple) +{ + struct nf_conntrack_expect *i; + + list_for_each_entry(i, &nf_conntrack_expect_list, list) { + if (nf_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask)) + return i; + } + return NULL; +} + +/* Just find a expectation corresponding to a tuple. */ +struct nf_conntrack_expect * +nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple) +{ + struct nf_conntrack_expect *i; + + read_lock_bh(&nf_conntrack_lock); + i = __nf_conntrack_expect_find(tuple); + if (i) + atomic_inc(&i->use); + read_unlock_bh(&nf_conntrack_lock); + + return i; +} + +/* If an expectation for this connection is found, it gets delete from + * global list then returned. */ +struct nf_conntrack_expect * +find_expectation(const struct nf_conntrack_tuple *tuple) +{ + struct nf_conntrack_expect *i; + + list_for_each_entry(i, &nf_conntrack_expect_list, list) { + /* If master is not in hash table yet (ie. packet hasn't left + this machine yet), how can other end know about expected? + Hence these are not the droids you are looking for (if + master ct never got confirmed, we'd hold a reference to it + and weird things would happen to future packets). */ + if (nf_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask) + && nf_ct_is_confirmed(i->master)) { + if (i->flags & NF_CT_EXPECT_PERMANENT) { + atomic_inc(&i->use); + return i; + } else if (del_timer(&i->timeout)) { + nf_ct_unlink_expect(i); + return i; + } + } + } + return NULL; +} + +/* delete all expectations for this conntrack */ +void nf_ct_remove_expectations(struct nf_conn *ct) +{ + struct nf_conntrack_expect *i, *tmp; + struct nf_conn_help *help = nfct_help(ct); + + /* Optimization: most connection never expect any others. */ + if (!help || help->expecting == 0) + return; + + list_for_each_entry_safe(i, tmp, &nf_conntrack_expect_list, list) { + if (i->master == ct && del_timer(&i->timeout)) { + nf_ct_unlink_expect(i); + nf_conntrack_expect_put(i); + } + } +} + +/* Would two expected things clash? */ +static inline int expect_clash(const struct nf_conntrack_expect *a, + const struct nf_conntrack_expect *b) +{ + /* Part covered by intersection of masks must be unequal, + otherwise they clash */ + struct nf_conntrack_tuple intersect_mask; + int count; + + intersect_mask.src.l3num = a->mask.src.l3num & b->mask.src.l3num; + intersect_mask.src.u.all = a->mask.src.u.all & b->mask.src.u.all; + intersect_mask.dst.u.all = a->mask.dst.u.all & b->mask.dst.u.all; + intersect_mask.dst.protonum = a->mask.dst.protonum + & b->mask.dst.protonum; + + for (count = 0; count < NF_CT_TUPLE_L3SIZE; count++){ + intersect_mask.src.u3.all[count] = + a->mask.src.u3.all[count] & b->mask.src.u3.all[count]; + } + + for (count = 0; count < NF_CT_TUPLE_L3SIZE; count++){ + intersect_mask.dst.u3.all[count] = + a->mask.dst.u3.all[count] & b->mask.dst.u3.all[count]; + } + + return nf_ct_tuple_mask_cmp(&a->tuple, &b->tuple, &intersect_mask); +} + +static inline int expect_matches(const struct nf_conntrack_expect *a, + const struct nf_conntrack_expect *b) +{ + return a->master == b->master + && nf_ct_tuple_equal(&a->tuple, &b->tuple) + && nf_ct_tuple_equal(&a->mask, &b->mask); +} + +/* Generally a bad idea to call this: could have matched already. */ +void nf_conntrack_unexpect_related(struct nf_conntrack_expect *exp) +{ + struct nf_conntrack_expect *i; + + write_lock_bh(&nf_conntrack_lock); + /* choose the the oldest expectation to evict */ + list_for_each_entry_reverse(i, &nf_conntrack_expect_list, list) { + if (expect_matches(i, exp) && del_timer(&i->timeout)) { + nf_ct_unlink_expect(i); + write_unlock_bh(&nf_conntrack_lock); + nf_conntrack_expect_put(i); + return; + } + } + write_unlock_bh(&nf_conntrack_lock); +} + +/* We don't increase the master conntrack refcount for non-fulfilled + * conntracks. During the conntrack destruction, the expectations are + * always killed before the conntrack itself */ +struct nf_conntrack_expect *nf_conntrack_expect_alloc(struct nf_conn *me) +{ + struct nf_conntrack_expect *new; + + new = kmem_cache_alloc(nf_conntrack_expect_cachep, GFP_ATOMIC); + if (!new) + return NULL; + + new->master = me; + atomic_set(&new->use, 1); + return new; +} + +void nf_conntrack_expect_put(struct nf_conntrack_expect *exp) +{ + if (atomic_dec_and_test(&exp->use)) + kmem_cache_free(nf_conntrack_expect_cachep, exp); +} + +static void nf_conntrack_expect_insert(struct nf_conntrack_expect *exp) +{ + struct nf_conn_help *master_help = nfct_help(exp->master); + + atomic_inc(&exp->use); + master_help->expecting++; + list_add(&exp->list, &nf_conntrack_expect_list); + + init_timer(&exp->timeout); + exp->timeout.data = (unsigned long)exp; + exp->timeout.function = expectation_timed_out; + exp->timeout.expires = jiffies + master_help->helper->timeout * HZ; + add_timer(&exp->timeout); + + exp->id = ++nf_conntrack_expect_next_id; + atomic_inc(&exp->use); + NF_CT_STAT_INC(expect_create); +} + +/* Race with expectations being used means we could have none to find; OK. */ +static void evict_oldest_expect(struct nf_conn *master) +{ + struct nf_conntrack_expect *i; + + list_for_each_entry_reverse(i, &nf_conntrack_expect_list, list) { + if (i->master == master) { + if (del_timer(&i->timeout)) { + nf_ct_unlink_expect(i); + nf_conntrack_expect_put(i); + } + break; + } + } +} + +static inline int refresh_timer(struct nf_conntrack_expect *i) +{ + struct nf_conn_help *master_help = nfct_help(i->master); + + if (!del_timer(&i->timeout)) + return 0; + + i->timeout.expires = jiffies + master_help->helper->timeout*HZ; + add_timer(&i->timeout); + return 1; +} + +int nf_conntrack_expect_related(struct nf_conntrack_expect *expect) +{ + struct nf_conntrack_expect *i; + struct nf_conn *master = expect->master; + struct nf_conn_help *master_help = nfct_help(master); + int ret; + + NF_CT_ASSERT(master_help); + + write_lock_bh(&nf_conntrack_lock); + list_for_each_entry(i, &nf_conntrack_expect_list, list) { + if (expect_matches(i, expect)) { + /* Refresh timer: if it's dying, ignore.. */ + if (refresh_timer(i)) { + ret = 0; + goto out; + } + } else if (expect_clash(i, expect)) { + ret = -EBUSY; + goto out; + } + } + /* Will be over limit? */ + if (master_help->helper->max_expected && + master_help->expecting >= master_help->helper->max_expected) + evict_oldest_expect(master); + + nf_conntrack_expect_insert(expect); + nf_conntrack_expect_event(IPEXP_NEW, expect); + ret = 0; +out: + write_unlock_bh(&nf_conntrack_lock); + return ret; +} + +#ifdef CONFIG_PROC_FS +static void *exp_seq_start(struct seq_file *s, loff_t *pos) +{ + struct list_head *e = &nf_conntrack_expect_list; + loff_t i; + + /* strange seq_file api calls stop even if we fail, + * thus we need to grab lock since stop unlocks */ + read_lock_bh(&nf_conntrack_lock); + + if (list_empty(e)) + return NULL; + + for (i = 0; i <= *pos; i++) { + e = e->next; + if (e == &nf_conntrack_expect_list) + return NULL; + } + return e; +} + +static void *exp_seq_next(struct seq_file *s, void *v, loff_t *pos) +{ + struct list_head *e = v; + + ++*pos; + e = e->next; + + if (e == &nf_conntrack_expect_list) + return NULL; + + return e; +} + +static void exp_seq_stop(struct seq_file *s, void *v) +{ + read_unlock_bh(&nf_conntrack_lock); +} + +static int exp_seq_show(struct seq_file *s, void *v) +{ + struct nf_conntrack_expect *expect = v; + + if (expect->timeout.function) + seq_printf(s, "%ld ", timer_pending(&expect->timeout) + ? (long)(expect->timeout.expires - jiffies)/HZ : 0); + else + seq_printf(s, "- "); + seq_printf(s, "l3proto = %u proto=%u ", + expect->tuple.src.l3num, + expect->tuple.dst.protonum); + print_tuple(s, &expect->tuple, + __nf_ct_l3proto_find(expect->tuple.src.l3num), + __nf_ct_proto_find(expect->tuple.src.l3num, + expect->tuple.dst.protonum)); + return seq_putc(s, '\n'); +} + +static struct seq_operations exp_seq_ops = { + .start = exp_seq_start, + .next = exp_seq_next, + .stop = exp_seq_stop, + .show = exp_seq_show +}; + +static int exp_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &exp_seq_ops); +} + +struct file_operations exp_file_ops = { + .owner = THIS_MODULE, + .open = exp_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release +}; +#endif /* CONFIG_PROC_FS */ diff --git a/net/netfilter/nf_conntrack_ftp.c b/net/netfilter/nf_conntrack_ftp.c index 0c17a5bd112..503fabf1884 100644 --- a/net/netfilter/nf_conntrack_ftp.c +++ b/net/netfilter/nf_conntrack_ftp.c @@ -26,6 +26,7 @@ #include #include +#include #include #include diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index cfb35fd8d9b..c3d53d91929 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -35,6 +35,7 @@ #include #include #include +#include #include #include #include diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index 5954f677381..72643d32cab 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -36,6 +36,7 @@ #include #include #include +#include #include #if 0 @@ -66,7 +67,7 @@ static int kill_proto(struct nf_conn *i, void *data) } #ifdef CONFIG_PROC_FS -static int +int print_tuple(struct seq_file *s, const struct nf_conntrack_tuple *tuple, struct nf_conntrack_l3proto *l3proto, struct nf_conntrack_protocol *proto) @@ -258,84 +259,6 @@ static struct file_operations ct_file_ops = { .release = seq_release_private, }; -/* expects */ -static void *exp_seq_start(struct seq_file *s, loff_t *pos) -{ - struct list_head *e = &nf_conntrack_expect_list; - loff_t i; - - /* strange seq_file api calls stop even if we fail, - * thus we need to grab lock since stop unlocks */ - read_lock_bh(&nf_conntrack_lock); - - if (list_empty(e)) - return NULL; - - for (i = 0; i <= *pos; i++) { - e = e->next; - if (e == &nf_conntrack_expect_list) - return NULL; - } - return e; -} - -static void *exp_seq_next(struct seq_file *s, void *v, loff_t *pos) -{ - struct list_head *e = v; - - ++*pos; - e = e->next; - - if (e == &nf_conntrack_expect_list) - return NULL; - - return e; -} - -static void exp_seq_stop(struct seq_file *s, void *v) -{ - read_unlock_bh(&nf_conntrack_lock); -} - -static int exp_seq_show(struct seq_file *s, void *v) -{ - struct nf_conntrack_expect *expect = v; - - if (expect->timeout.function) - seq_printf(s, "%ld ", timer_pending(&expect->timeout) - ? (long)(expect->timeout.expires - jiffies)/HZ : 0); - else - seq_printf(s, "- "); - seq_printf(s, "l3proto = %u proto=%u ", - expect->tuple.src.l3num, - expect->tuple.dst.protonum); - print_tuple(s, &expect->tuple, - __nf_ct_l3proto_find(expect->tuple.src.l3num), - __nf_ct_proto_find(expect->tuple.src.l3num, - expect->tuple.dst.protonum)); - return seq_putc(s, '\n'); -} - -static struct seq_operations exp_seq_ops = { - .start = exp_seq_start, - .next = exp_seq_next, - .stop = exp_seq_stop, - .show = exp_seq_show -}; - -static int exp_open(struct inode *inode, struct file *file) -{ - return seq_open(file, &exp_seq_ops); -} - -static struct file_operations exp_file_ops = { - .owner = THIS_MODULE, - .open = exp_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release -}; - static void *ct_cpu_seq_start(struct seq_file *seq, loff_t *pos) { int cpu; -- cgit v1.2.3-70-g09d2 From 7e5d03bb9d2b96fdeab0cb0c98b93e6cf7130c96 Mon Sep 17 00:00:00 2001 From: Martin Josefsson Date: Wed, 29 Nov 2006 02:34:59 +0100 Subject: [NETFILTER]: nf_conntrack: split out helper handling This patch splits out handling of helpers into its own file nf_conntrack_helper.c Signed-off-by: Martin Josefsson Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack_core.h | 2 + include/net/netfilter/nf_conntrack_helper.h | 10 ++ net/netfilter/Makefile | 2 +- net/netfilter/nf_conntrack_core.c | 120 +--------------------- net/netfilter/nf_conntrack_helper.c | 151 ++++++++++++++++++++++++++++ 5 files changed, 165 insertions(+), 120 deletions(-) create mode 100644 net/netfilter/nf_conntrack_helper.c (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_core.h b/include/net/netfilter/nf_conntrack_core.h index 84a8e01941f..bf8d3008a38 100644 --- a/include/net/netfilter/nf_conntrack_core.h +++ b/include/net/netfilter/nf_conntrack_core.h @@ -80,4 +80,6 @@ print_tuple(struct seq_file *s, const struct nf_conntrack_tuple *tuple, extern struct list_head *nf_conntrack_hash; extern struct list_head nf_conntrack_expect_list; extern rwlock_t nf_conntrack_lock ; +extern struct list_head unconfirmed; + #endif /* _NF_CONNTRACK_CORE_H */ diff --git a/include/net/netfilter/nf_conntrack_helper.h b/include/net/netfilter/nf_conntrack_helper.h index 3cbd13e2216..fbba9e8b95f 100644 --- a/include/net/netfilter/nf_conntrack_helper.h +++ b/include/net/netfilter/nf_conntrack_helper.h @@ -37,6 +37,16 @@ struct nf_conntrack_helper int (*to_nfattr)(struct sk_buff *skb, const struct nf_conn *ct); }; +extern struct nf_conntrack_helper * +__nf_ct_helper_find(const struct nf_conntrack_tuple *tuple); + +extern struct nf_conntrack_helper * +nf_ct_helper_find_get( const struct nf_conntrack_tuple *tuple); + +extern struct nf_conntrack_helper * +__nf_conntrack_helper_find_byname(const char *name); + +extern void nf_ct_helper_put(struct nf_conntrack_helper *helper); extern int nf_conntrack_helper_register(struct nf_conntrack_helper *); extern void nf_conntrack_helper_unregister(struct nf_conntrack_helper *); diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index d33b9ac01aa..236376483ef 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -1,5 +1,5 @@ netfilter-objs := core.o nf_log.o nf_queue.o nf_sockopt.o -nf_conntrack-objs := nf_conntrack_core.o nf_conntrack_standalone.o nf_conntrack_expect.o nf_conntrack_l3proto_generic.o nf_conntrack_proto_generic.o nf_conntrack_proto_tcp.o nf_conntrack_proto_udp.o +nf_conntrack-objs := nf_conntrack_core.o nf_conntrack_standalone.o nf_conntrack_expect.o nf_conntrack_helper.o nf_conntrack_l3proto_generic.o nf_conntrack_proto_generic.o nf_conntrack_proto_tcp.o nf_conntrack_proto_udp.o obj-$(CONFIG_NETFILTER) = netfilter.o diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 72fbae4e594..cf6face67af 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -75,13 +75,12 @@ atomic_t nf_conntrack_count = ATOMIC_INIT(0); void (*nf_conntrack_destroyed)(struct nf_conn *conntrack) = NULL; struct nf_conntrack_protocol **nf_ct_protos[PF_MAX] __read_mostly; struct nf_conntrack_l3proto *nf_ct_l3protos[PF_MAX] __read_mostly; -static LIST_HEAD(helpers); unsigned int nf_conntrack_htable_size __read_mostly = 0; int nf_conntrack_max __read_mostly; struct list_head *nf_conntrack_hash __read_mostly; struct nf_conn nf_conntrack_untracked; unsigned int nf_ct_log_invalid __read_mostly; -static LIST_HEAD(unconfirmed); +LIST_HEAD(unconfirmed); static int nf_conntrack_vmalloc __read_mostly; static unsigned int nf_conntrack_next_id; @@ -696,46 +695,6 @@ static int early_drop(struct list_head *chain) return dropped; } -static struct nf_conntrack_helper * -__nf_ct_helper_find(const struct nf_conntrack_tuple *tuple) -{ - struct nf_conntrack_helper *h; - - list_for_each_entry(h, &helpers, list) { - if (nf_ct_tuple_mask_cmp(tuple, &h->tuple, &h->mask)) - return h; - } - return NULL; -} - -struct nf_conntrack_helper * -nf_ct_helper_find_get( const struct nf_conntrack_tuple *tuple) -{ - struct nf_conntrack_helper *helper; - - /* need nf_conntrack_lock to assure that helper exists until - * try_module_get() is called */ - read_lock_bh(&nf_conntrack_lock); - - helper = __nf_ct_helper_find(tuple); - if (helper) { - /* need to increase module usage count to assure helper will - * not go away while the caller is e.g. busy putting a - * conntrack in the hash that uses the helper */ - if (!try_module_get(helper->me)) - helper = NULL; - } - - read_unlock_bh(&nf_conntrack_lock); - - return helper; -} - -void nf_ct_helper_put(struct nf_conntrack_helper *helper) -{ - module_put(helper->me); -} - static struct nf_conn * __nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, const struct nf_conntrack_tuple *repl, @@ -1035,83 +994,6 @@ int nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse, orig->dst.protonum)); } -int nf_conntrack_helper_register(struct nf_conntrack_helper *me) -{ - int ret; - BUG_ON(me->timeout == 0); - - ret = nf_conntrack_register_cache(NF_CT_F_HELP, "nf_conntrack:help", - sizeof(struct nf_conn) - + sizeof(struct nf_conn_help) - + __alignof__(struct nf_conn_help)); - if (ret < 0) { - printk(KERN_ERR "nf_conntrack_helper_reigster: Unable to create slab cache for conntracks\n"); - return ret; - } - write_lock_bh(&nf_conntrack_lock); - list_add(&me->list, &helpers); - write_unlock_bh(&nf_conntrack_lock); - - return 0; -} - -struct nf_conntrack_helper * -__nf_conntrack_helper_find_byname(const char *name) -{ - struct nf_conntrack_helper *h; - - list_for_each_entry(h, &helpers, list) { - if (!strcmp(h->name, name)) - return h; - } - - return NULL; -} - -static inline void unhelp(struct nf_conntrack_tuple_hash *i, - const struct nf_conntrack_helper *me) -{ - struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(i); - struct nf_conn_help *help = nfct_help(ct); - - if (help && help->helper == me) { - nf_conntrack_event(IPCT_HELPER, ct); - help->helper = NULL; - } -} - -void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me) -{ - unsigned int i; - struct nf_conntrack_tuple_hash *h; - struct nf_conntrack_expect *exp, *tmp; - - /* Need write lock here, to delete helper. */ - write_lock_bh(&nf_conntrack_lock); - list_del(&me->list); - - /* Get rid of expectations */ - list_for_each_entry_safe(exp, tmp, &nf_conntrack_expect_list, list) { - struct nf_conn_help *help = nfct_help(exp->master); - if (help->helper == me && del_timer(&exp->timeout)) { - nf_ct_unlink_expect(exp); - nf_conntrack_expect_put(exp); - } - } - - /* Get rid of expecteds, set helpers to NULL. */ - list_for_each_entry(h, &unconfirmed, list) - unhelp(h, me); - for (i = 0; i < nf_conntrack_htable_size; i++) { - list_for_each_entry(h, &nf_conntrack_hash[i], list) - unhelp(h, me); - } - write_unlock_bh(&nf_conntrack_lock); - - /* Someone could be still looking at the helper in a bh. */ - synchronize_net(); -} - /* Refresh conntrack for this many jiffies and do accounting if do_acct is 1 */ void __nf_ct_refresh_acct(struct nf_conn *ct, enum ip_conntrack_info ctinfo, diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c new file mode 100644 index 00000000000..ca044875a56 --- /dev/null +++ b/net/netfilter/nf_conntrack_helper.c @@ -0,0 +1,151 @@ +/* Helper handling for netfilter. */ + +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * (C) 2003,2004 USAGI/WIDE Project + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define ASSERT_READ_LOCK(x) +#define ASSERT_WRITE_LOCK(x) + +#include +#include +#include +#include +#include + +static LIST_HEAD(helpers); + +struct nf_conntrack_helper * +__nf_ct_helper_find(const struct nf_conntrack_tuple *tuple) +{ + struct nf_conntrack_helper *h; + + list_for_each_entry(h, &helpers, list) { + if (nf_ct_tuple_mask_cmp(tuple, &h->tuple, &h->mask)) + return h; + } + return NULL; +} + +struct nf_conntrack_helper * +nf_ct_helper_find_get( const struct nf_conntrack_tuple *tuple) +{ + struct nf_conntrack_helper *helper; + + /* need nf_conntrack_lock to assure that helper exists until + * try_module_get() is called */ + read_lock_bh(&nf_conntrack_lock); + + helper = __nf_ct_helper_find(tuple); + if (helper) { + /* need to increase module usage count to assure helper will + * not go away while the caller is e.g. busy putting a + * conntrack in the hash that uses the helper */ + if (!try_module_get(helper->me)) + helper = NULL; + } + + read_unlock_bh(&nf_conntrack_lock); + + return helper; +} + +void nf_ct_helper_put(struct nf_conntrack_helper *helper) +{ + module_put(helper->me); +} + +struct nf_conntrack_helper * +__nf_conntrack_helper_find_byname(const char *name) +{ + struct nf_conntrack_helper *h; + + list_for_each_entry(h, &helpers, list) { + if (!strcmp(h->name, name)) + return h; + } + + return NULL; +} + +static inline int unhelp(struct nf_conntrack_tuple_hash *i, + const struct nf_conntrack_helper *me) +{ + struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(i); + struct nf_conn_help *help = nfct_help(ct); + + if (help && help->helper == me) { + nf_conntrack_event(IPCT_HELPER, ct); + help->helper = NULL; + } + return 0; +} + +int nf_conntrack_helper_register(struct nf_conntrack_helper *me) +{ + int ret; + BUG_ON(me->timeout == 0); + + ret = nf_conntrack_register_cache(NF_CT_F_HELP, "nf_conntrack:help", + sizeof(struct nf_conn) + + sizeof(struct nf_conn_help) + + __alignof__(struct nf_conn_help)); + if (ret < 0) { + printk(KERN_ERR "nf_conntrack_helper_register: Unable to create slab cache for conntracks\n"); + return ret; + } + write_lock_bh(&nf_conntrack_lock); + list_add(&me->list, &helpers); + write_unlock_bh(&nf_conntrack_lock); + + return 0; +} + +void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me) +{ + unsigned int i; + struct nf_conntrack_tuple_hash *h; + struct nf_conntrack_expect *exp, *tmp; + + /* Need write lock here, to delete helper. */ + write_lock_bh(&nf_conntrack_lock); + list_del(&me->list); + + /* Get rid of expectations */ + list_for_each_entry_safe(exp, tmp, &nf_conntrack_expect_list, list) { + struct nf_conn_help *help = nfct_help(exp->master); + if (help->helper == me && del_timer(&exp->timeout)) { + nf_ct_unlink_expect(exp); + nf_conntrack_expect_put(exp); + } + } + + /* Get rid of expecteds, set helpers to NULL. */ + list_for_each_entry(h, &unconfirmed, list) + unhelp(h, me); + for (i = 0; i < nf_conntrack_htable_size; i++) { + list_for_each_entry(h, &nf_conntrack_hash[i], list) + unhelp(h, me); + } + write_unlock_bh(&nf_conntrack_lock); + + /* Someone could be still looking at the helper in a bh. */ + synchronize_net(); +} -- cgit v1.2.3-70-g09d2 From f61801218a58381f498ae5c38ae3eae0bc73e976 Mon Sep 17 00:00:00 2001 From: Martin Josefsson Date: Wed, 29 Nov 2006 02:35:01 +0100 Subject: [NETFILTER]: nf_conntrack: split out the event cache This patch splits out the event cache into its own file nf_conntrack_ecache.c Signed-off-by: Martin Josefsson Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack.h | 81 ------------------------ include/net/netfilter/nf_conntrack_core.h | 1 + include/net/netfilter/nf_conntrack_ecache.h | 95 +++++++++++++++++++++++++++++ net/netfilter/Makefile | 4 +- net/netfilter/nf_conntrack_core.c | 67 -------------------- net/netfilter/nf_conntrack_ecache.c | 91 +++++++++++++++++++++++++++ net/netfilter/nf_conntrack_ftp.c | 1 + net/netfilter/nf_conntrack_proto_sctp.c | 1 + net/netfilter/nf_conntrack_proto_tcp.c | 1 + net/netfilter/nf_conntrack_proto_udp.c | 2 + net/netfilter/nf_conntrack_standalone.c | 2 +- net/netfilter/xt_CONNMARK.c | 3 + 12 files changed, 199 insertions(+), 150 deletions(-) create mode 100644 include/net/netfilter/nf_conntrack_ecache.h create mode 100644 net/netfilter/nf_conntrack_ecache.c (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 9d2581fc04b..9a2950551cd 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -244,87 +244,6 @@ extern int nf_conntrack_checksum; #define NF_CT_STAT_INC(count) (__get_cpu_var(nf_conntrack_stat).count++) -#ifdef CONFIG_NF_CONNTRACK_EVENTS -#include -#include -#include - -struct nf_conntrack_ecache { - struct nf_conn *ct; - unsigned int events; -}; -DECLARE_PER_CPU(struct nf_conntrack_ecache, nf_conntrack_ecache); - -#define CONNTRACK_ECACHE(x) (__get_cpu_var(nf_conntrack_ecache).x) - -extern struct atomic_notifier_head nf_conntrack_chain; -extern struct atomic_notifier_head nf_conntrack_expect_chain; - -static inline int nf_conntrack_register_notifier(struct notifier_block *nb) -{ - return atomic_notifier_chain_register(&nf_conntrack_chain, nb); -} - -static inline int nf_conntrack_unregister_notifier(struct notifier_block *nb) -{ - return atomic_notifier_chain_unregister(&nf_conntrack_chain, nb); -} - -static inline int -nf_conntrack_expect_register_notifier(struct notifier_block *nb) -{ - return atomic_notifier_chain_register(&nf_conntrack_expect_chain, nb); -} - -static inline int -nf_conntrack_expect_unregister_notifier(struct notifier_block *nb) -{ - return atomic_notifier_chain_unregister(&nf_conntrack_expect_chain, - nb); -} - -extern void nf_ct_deliver_cached_events(const struct nf_conn *ct); -extern void __nf_ct_event_cache_init(struct nf_conn *ct); - -static inline void -nf_conntrack_event_cache(enum ip_conntrack_events event, - const struct sk_buff *skb) -{ - struct nf_conn *ct = (struct nf_conn *)skb->nfct; - struct nf_conntrack_ecache *ecache; - - local_bh_disable(); - ecache = &__get_cpu_var(nf_conntrack_ecache); - if (ct != ecache->ct) - __nf_ct_event_cache_init(ct); - ecache->events |= event; - local_bh_enable(); -} - -static inline void nf_conntrack_event(enum ip_conntrack_events event, - struct nf_conn *ct) -{ - if (nf_ct_is_confirmed(ct) && !nf_ct_is_dying(ct)) - atomic_notifier_call_chain(&nf_conntrack_chain, event, ct); -} - -static inline void -nf_conntrack_expect_event(enum ip_conntrack_expect_events event, - struct nf_conntrack_expect *exp) -{ - atomic_notifier_call_chain(&nf_conntrack_expect_chain, event, exp); -} -#else /* CONFIG_NF_CONNTRACK_EVENTS */ -static inline void nf_conntrack_event_cache(enum ip_conntrack_events event, - const struct sk_buff *skb) {} -static inline void nf_conntrack_event(enum ip_conntrack_events event, - struct nf_conn *ct) {} -static inline void nf_ct_deliver_cached_events(const struct nf_conn *ct) {} -static inline void -nf_conntrack_expect_event(enum ip_conntrack_expect_events event, - struct nf_conntrack_expect *exp) {} -#endif /* CONFIG_NF_CONNTRACK_EVENTS */ - /* no helper, no nat */ #define NF_CT_F_BASIC 0 /* for helper */ diff --git a/include/net/netfilter/nf_conntrack_core.h b/include/net/netfilter/nf_conntrack_core.h index bf8d3008a38..4676b492d63 100644 --- a/include/net/netfilter/nf_conntrack_core.h +++ b/include/net/netfilter/nf_conntrack_core.h @@ -15,6 +15,7 @@ #include #include #include +#include /* This header is used to share core functionality between the standalone connection tracking module, and the compatibility layer's use diff --git a/include/net/netfilter/nf_conntrack_ecache.h b/include/net/netfilter/nf_conntrack_ecache.h new file mode 100644 index 00000000000..b62a8a9ec9d --- /dev/null +++ b/include/net/netfilter/nf_conntrack_ecache.h @@ -0,0 +1,95 @@ +/* + * connection tracking event cache. + */ + +#ifndef _NF_CONNTRACK_ECACHE_H +#define _NF_CONNTRACK_ECACHE_H +#include + +#include +#include +#include + +#ifdef CONFIG_NF_CONNTRACK_EVENTS +struct nf_conntrack_ecache { + struct nf_conn *ct; + unsigned int events; +}; +DECLARE_PER_CPU(struct nf_conntrack_ecache, nf_conntrack_ecache); + +#define CONNTRACK_ECACHE(x) (__get_cpu_var(nf_conntrack_ecache).x) + +extern struct atomic_notifier_head nf_conntrack_chain; +extern struct atomic_notifier_head nf_conntrack_expect_chain; + +static inline int nf_conntrack_register_notifier(struct notifier_block *nb) +{ + return atomic_notifier_chain_register(&nf_conntrack_chain, nb); +} + +static inline int nf_conntrack_unregister_notifier(struct notifier_block *nb) +{ + return atomic_notifier_chain_unregister(&nf_conntrack_chain, nb); +} + +static inline int +nf_conntrack_expect_register_notifier(struct notifier_block *nb) +{ + return atomic_notifier_chain_register(&nf_conntrack_expect_chain, nb); +} + +static inline int +nf_conntrack_expect_unregister_notifier(struct notifier_block *nb) +{ + return atomic_notifier_chain_unregister(&nf_conntrack_expect_chain, + nb); +} + +extern void nf_ct_deliver_cached_events(const struct nf_conn *ct); +extern void __nf_ct_event_cache_init(struct nf_conn *ct); +extern void nf_ct_event_cache_flush(void); + +static inline void +nf_conntrack_event_cache(enum ip_conntrack_events event, + const struct sk_buff *skb) +{ + struct nf_conn *ct = (struct nf_conn *)skb->nfct; + struct nf_conntrack_ecache *ecache; + + local_bh_disable(); + ecache = &__get_cpu_var(nf_conntrack_ecache); + if (ct != ecache->ct) + __nf_ct_event_cache_init(ct); + ecache->events |= event; + local_bh_enable(); +} + +static inline void nf_conntrack_event(enum ip_conntrack_events event, + struct nf_conn *ct) +{ + if (nf_ct_is_confirmed(ct) && !nf_ct_is_dying(ct)) + atomic_notifier_call_chain(&nf_conntrack_chain, event, ct); +} + +static inline void +nf_conntrack_expect_event(enum ip_conntrack_expect_events event, + struct nf_conntrack_expect *exp) +{ + atomic_notifier_call_chain(&nf_conntrack_expect_chain, event, exp); +} + +#else /* CONFIG_NF_CONNTRACK_EVENTS */ + +static inline void nf_conntrack_event_cache(enum ip_conntrack_events event, + const struct sk_buff *skb) {} +static inline void nf_conntrack_event(enum ip_conntrack_events event, + struct nf_conn *ct) {} +static inline void nf_ct_deliver_cached_events(const struct nf_conn *ct) {} +static inline void +nf_conntrack_expect_event(enum ip_conntrack_expect_events event, + struct nf_conntrack_expect *exp) {} +static inline void nf_ct_event_cache_flush(void) {} +#endif /* CONFIG_NF_CONNTRACK_EVENTS */ + +#endif /*_NF_CONNTRACK_ECACHE_H*/ + diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 236376483ef..fd0641d6679 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -1,5 +1,7 @@ netfilter-objs := core.o nf_log.o nf_queue.o nf_sockopt.o -nf_conntrack-objs := nf_conntrack_core.o nf_conntrack_standalone.o nf_conntrack_expect.o nf_conntrack_helper.o nf_conntrack_l3proto_generic.o nf_conntrack_proto_generic.o nf_conntrack_proto_tcp.o nf_conntrack_proto_udp.o + +nf_conntrack-y := nf_conntrack_core.o nf_conntrack_standalone.o nf_conntrack_expect.o nf_conntrack_helper.o nf_conntrack_l3proto_generic.o nf_conntrack_proto_generic.o nf_conntrack_proto_tcp.o nf_conntrack_proto_udp.o +nf_conntrack-$(CONFIG_NF_CONNTRACK_EVENTS) += nf_conntrack_ecache.o obj-$(CONFIG_NETFILTER) = netfilter.o diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index cf6face67af..029acc4f964 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -85,73 +85,6 @@ static int nf_conntrack_vmalloc __read_mostly; static unsigned int nf_conntrack_next_id; -#ifdef CONFIG_NF_CONNTRACK_EVENTS -ATOMIC_NOTIFIER_HEAD(nf_conntrack_chain); -ATOMIC_NOTIFIER_HEAD(nf_conntrack_expect_chain); - -DEFINE_PER_CPU(struct nf_conntrack_ecache, nf_conntrack_ecache); - -/* deliver cached events and clear cache entry - must be called with locally - * disabled softirqs */ -static inline void -__nf_ct_deliver_cached_events(struct nf_conntrack_ecache *ecache) -{ - DEBUGP("ecache: delivering events for %p\n", ecache->ct); - if (nf_ct_is_confirmed(ecache->ct) && !nf_ct_is_dying(ecache->ct) - && ecache->events) - atomic_notifier_call_chain(&nf_conntrack_chain, ecache->events, - ecache->ct); - - ecache->events = 0; - nf_ct_put(ecache->ct); - ecache->ct = NULL; -} - -/* Deliver all cached events for a particular conntrack. This is called - * by code prior to async packet handling for freeing the skb */ -void nf_ct_deliver_cached_events(const struct nf_conn *ct) -{ - struct nf_conntrack_ecache *ecache; - - local_bh_disable(); - ecache = &__get_cpu_var(nf_conntrack_ecache); - if (ecache->ct == ct) - __nf_ct_deliver_cached_events(ecache); - local_bh_enable(); -} - -/* Deliver cached events for old pending events, if current conntrack != old */ -void __nf_ct_event_cache_init(struct nf_conn *ct) -{ - struct nf_conntrack_ecache *ecache; - - /* take care of delivering potentially old events */ - ecache = &__get_cpu_var(nf_conntrack_ecache); - BUG_ON(ecache->ct == ct); - if (ecache->ct) - __nf_ct_deliver_cached_events(ecache); - /* initialize for this conntrack/packet */ - ecache->ct = ct; - nf_conntrack_get(&ct->ct_general); -} - -/* flush the event cache - touches other CPU's data and must not be called - * while packets are still passing through the code */ -static void nf_ct_event_cache_flush(void) -{ - struct nf_conntrack_ecache *ecache; - int cpu; - - for_each_possible_cpu(cpu) { - ecache = &per_cpu(nf_conntrack_ecache, cpu); - if (ecache->ct) - nf_ct_put(ecache->ct); - } -} -#else -static inline void nf_ct_event_cache_flush(void) {} -#endif /* CONFIG_NF_CONNTRACK_EVENTS */ - DEFINE_PER_CPU(struct ip_conntrack_stat, nf_conntrack_stat); EXPORT_PER_CPU_SYMBOL(nf_conntrack_stat); diff --git a/net/netfilter/nf_conntrack_ecache.c b/net/netfilter/nf_conntrack_ecache.c new file mode 100644 index 00000000000..35763e96fdd --- /dev/null +++ b/net/netfilter/nf_conntrack_ecache.c @@ -0,0 +1,91 @@ +/* Event cache for netfilter. */ + +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * (C) 2003,2004 USAGI/WIDE Project + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +ATOMIC_NOTIFIER_HEAD(nf_conntrack_chain); +ATOMIC_NOTIFIER_HEAD(nf_conntrack_expect_chain); + +DEFINE_PER_CPU(struct nf_conntrack_ecache, nf_conntrack_ecache); + +/* deliver cached events and clear cache entry - must be called with locally + * disabled softirqs */ +static inline void +__nf_ct_deliver_cached_events(struct nf_conntrack_ecache *ecache) +{ + if (nf_ct_is_confirmed(ecache->ct) && !nf_ct_is_dying(ecache->ct) + && ecache->events) + atomic_notifier_call_chain(&nf_conntrack_chain, ecache->events, + ecache->ct); + + ecache->events = 0; + nf_ct_put(ecache->ct); + ecache->ct = NULL; +} + +/* Deliver all cached events for a particular conntrack. This is called + * by code prior to async packet handling for freeing the skb */ +void nf_ct_deliver_cached_events(const struct nf_conn *ct) +{ + struct nf_conntrack_ecache *ecache; + + local_bh_disable(); + ecache = &__get_cpu_var(nf_conntrack_ecache); + if (ecache->ct == ct) + __nf_ct_deliver_cached_events(ecache); + local_bh_enable(); +} + +/* Deliver cached events for old pending events, if current conntrack != old */ +void __nf_ct_event_cache_init(struct nf_conn *ct) +{ + struct nf_conntrack_ecache *ecache; + + /* take care of delivering potentially old events */ + ecache = &__get_cpu_var(nf_conntrack_ecache); + BUG_ON(ecache->ct == ct); + if (ecache->ct) + __nf_ct_deliver_cached_events(ecache); + /* initialize for this conntrack/packet */ + ecache->ct = ct; + nf_conntrack_get(&ct->ct_general); +} + +/* flush the event cache - touches other CPU's data and must not be called + * while packets are still passing through the code */ +void nf_ct_event_cache_flush(void) +{ + struct nf_conntrack_ecache *ecache; + int cpu; + + for_each_possible_cpu(cpu) { + ecache = &per_cpu(nf_conntrack_ecache, cpu); + if (ecache->ct) + nf_ct_put(ecache->ct); + } +} + diff --git a/net/netfilter/nf_conntrack_ftp.c b/net/netfilter/nf_conntrack_ftp.c index 503fabf1884..10836fc22a4 100644 --- a/net/netfilter/nf_conntrack_ftp.c +++ b/net/netfilter/nf_conntrack_ftp.c @@ -27,6 +27,7 @@ #include #include +#include #include #include diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c index af568777372..ba3e31f2acd 100644 --- a/net/netfilter/nf_conntrack_proto_sctp.c +++ b/net/netfilter/nf_conntrack_proto_sctp.c @@ -33,6 +33,7 @@ #include #include +#include #if 0 #define DEBUGP(format, ...) printk(format, ## __VA_ARGS__) diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index 238bbb5b72e..e6b6f23ecdd 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -43,6 +43,7 @@ #include #include #include +#include #if 0 #define DEBUGP printk diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c index d28981cf9af..cf9db035d39 100644 --- a/net/netfilter/nf_conntrack_proto_udp.c +++ b/net/netfilter/nf_conntrack_proto_udp.c @@ -22,10 +22,12 @@ #include #include #include + #include #include #include #include +#include unsigned int nf_ct_udp_timeout __read_mostly = 30*HZ; unsigned int nf_ct_udp_timeout_stream __read_mostly = 180*HZ; diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index 72643d32cab..73ca156bb09 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -33,9 +33,9 @@ #define ASSERT_WRITE_LOCK(x) #include +#include #include #include -#include #include #include diff --git a/net/netfilter/xt_CONNMARK.c b/net/netfilter/xt_CONNMARK.c index 67ed5315299..b5548239d41 100644 --- a/net/netfilter/xt_CONNMARK.c +++ b/net/netfilter/xt_CONNMARK.c @@ -31,6 +31,9 @@ MODULE_ALIAS("ipt_CONNMARK"); #include #include #include +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) +#include +#endif static unsigned int target(struct sk_buff **pskb, -- cgit v1.2.3-70-g09d2 From 605dcad6c85226e6d43387917b329d65b95cef39 Mon Sep 17 00:00:00 2001 From: Martin Josefsson Date: Wed, 29 Nov 2006 02:35:06 +0100 Subject: [NETFILTER]: nf_conntrack: rename struct nf_conntrack_protocol Rename 'struct nf_conntrack_protocol' to 'struct nf_conntrack_l4proto' in order to help distinguish it from 'struct nf_conntrack_l3proto'. It gets rather confusing with 'nf_conntrack_protocol'. Signed-off-by: Martin Josefsson Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack_core.h | 10 +- include/net/netfilter/nf_conntrack_l3proto.h | 4 +- include/net/netfilter/nf_conntrack_l4proto.h | 129 +++++++++++++++++++++++++ include/net/netfilter/nf_conntrack_protocol.h | 129 ------------------------- net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 26 ++--- net/ipv4/netfilter/nf_conntrack_proto_icmp.c | 12 +-- net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 26 ++--- net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c | 12 +-- net/netfilter/nf_conntrack_core.c | 46 ++++----- net/netfilter/nf_conntrack_ecache.c | 2 +- net/netfilter/nf_conntrack_expect.c | 2 +- net/netfilter/nf_conntrack_helper.c | 2 +- net/netfilter/nf_conntrack_l3proto_generic.c | 4 +- net/netfilter/nf_conntrack_netlink.c | 56 +++++------ net/netfilter/nf_conntrack_proto.c | 72 +++++++------- net/netfilter/nf_conntrack_proto_generic.c | 6 +- net/netfilter/nf_conntrack_proto_sctp.c | 26 ++--- net/netfilter/nf_conntrack_proto_tcp.c | 14 +-- net/netfilter/nf_conntrack_proto_udp.c | 14 +-- net/netfilter/nf_conntrack_standalone.c | 30 +++--- 20 files changed, 311 insertions(+), 311 deletions(-) create mode 100644 include/net/netfilter/nf_conntrack_l4proto.h delete mode 100644 include/net/netfilter/nf_conntrack_protocol.h (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_core.h b/include/net/netfilter/nf_conntrack_core.h index 4676b492d63..7fdc72c0135 100644 --- a/include/net/netfilter/nf_conntrack_core.h +++ b/include/net/netfilter/nf_conntrack_core.h @@ -14,7 +14,7 @@ #include #include -#include +#include #include /* This header is used to share core functionality between the @@ -32,7 +32,7 @@ extern struct nf_conntrack_l3proto *nf_ct_find_l3proto(u_int16_t pf); /* Like above, but you already have conntrack read lock. */ extern struct nf_conntrack_l3proto *__nf_ct_find_l3proto(u_int16_t l3proto); -struct nf_conntrack_protocol; +struct nf_conntrack_l4proto; extern int nf_ct_get_tuple(const struct sk_buff *skb, @@ -42,13 +42,13 @@ nf_ct_get_tuple(const struct sk_buff *skb, u_int8_t protonum, struct nf_conntrack_tuple *tuple, const struct nf_conntrack_l3proto *l3proto, - const struct nf_conntrack_protocol *protocol); + const struct nf_conntrack_l4proto *l4proto); extern int nf_ct_invert_tuple(struct nf_conntrack_tuple *inverse, const struct nf_conntrack_tuple *orig, const struct nf_conntrack_l3proto *l3proto, - const struct nf_conntrack_protocol *protocol); + const struct nf_conntrack_l4proto *l4proto); /* Find a connection corresponding to a tuple. */ extern struct nf_conntrack_tuple_hash * @@ -76,7 +76,7 @@ extern void __nf_conntrack_attach(struct sk_buff *nskb, struct sk_buff *skb); int print_tuple(struct seq_file *s, const struct nf_conntrack_tuple *tuple, struct nf_conntrack_l3proto *l3proto, - struct nf_conntrack_protocol *proto); + struct nf_conntrack_l4proto *proto); extern struct list_head *nf_conntrack_hash; extern struct list_head nf_conntrack_expect_list; diff --git a/include/net/netfilter/nf_conntrack_l3proto.h b/include/net/netfilter/nf_conntrack_l3proto.h index dac43b15a5b..ffe23a1e0b9 100644 --- a/include/net/netfilter/nf_conntrack_l3proto.h +++ b/include/net/netfilter/nf_conntrack_l3proto.h @@ -96,13 +96,13 @@ extern void nf_ct_l3proto_put(struct nf_conntrack_l3proto *p); /* Existing built-in protocols */ extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4; extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6; -extern struct nf_conntrack_l3proto nf_conntrack_generic_l3proto; +extern struct nf_conntrack_l3proto nf_conntrack_l3proto_generic; static inline struct nf_conntrack_l3proto * __nf_ct_l3proto_find(u_int16_t l3proto) { if (unlikely(l3proto >= AF_MAX)) - return &nf_conntrack_generic_l3proto; + return &nf_conntrack_l3proto_generic; return nf_ct_l3protos[l3proto]; } diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h new file mode 100644 index 00000000000..0c6babd5439 --- /dev/null +++ b/include/net/netfilter/nf_conntrack_l4proto.h @@ -0,0 +1,129 @@ +/* + * Header for use in defining a given L4 protocol for connection tracking. + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI + * - generalized L3 protocol dependent part. + * + * Derived from include/linux/netfiter_ipv4/ip_conntrack_protcol.h + */ + +#ifndef _NF_CONNTRACK_L4PROTO_H +#define _NF_CONNTRACK_L4PROTO_H +#include + +struct seq_file; +struct nfattr; + +struct nf_conntrack_l4proto +{ + /* Next pointer. */ + struct list_head list; + + /* L3 Protocol number. */ + u_int16_t l3proto; + + /* L4 Protocol number. */ + u_int8_t l4proto; + + /* Protocol name */ + const char *name; + + /* Try to fill in the third arg: dataoff is offset past network protocol + hdr. Return true if possible. */ + int (*pkt_to_tuple)(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple); + + /* Invert the per-proto part of the tuple: ie. turn xmit into reply. + * Some packets can't be inverted: return 0 in that case. + */ + int (*invert_tuple)(struct nf_conntrack_tuple *inverse, + const struct nf_conntrack_tuple *orig); + + /* Print out the per-protocol part of the tuple. Return like seq_* */ + int (*print_tuple)(struct seq_file *s, + const struct nf_conntrack_tuple *); + + /* Print out the private part of the conntrack. */ + int (*print_conntrack)(struct seq_file *s, const struct nf_conn *); + + /* Returns verdict for packet, or -1 for invalid. */ + int (*packet)(struct nf_conn *conntrack, + const struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info ctinfo, + int pf, + unsigned int hooknum); + + /* Called when a new connection for this protocol found; + * returns TRUE if it's OK. If so, packet() called next. */ + int (*new)(struct nf_conn *conntrack, const struct sk_buff *skb, + unsigned int dataoff); + + /* Called when a conntrack entry is destroyed */ + void (*destroy)(struct nf_conn *conntrack); + + int (*error)(struct sk_buff *skb, unsigned int dataoff, + enum ip_conntrack_info *ctinfo, + int pf, unsigned int hooknum); + + /* convert protoinfo to nfnetink attributes */ + int (*to_nfattr)(struct sk_buff *skb, struct nfattr *nfa, + const struct nf_conn *ct); + + /* convert nfnetlink attributes to protoinfo */ + int (*from_nfattr)(struct nfattr *tb[], struct nf_conn *ct); + + int (*tuple_to_nfattr)(struct sk_buff *skb, + const struct nf_conntrack_tuple *t); + int (*nfattr_to_tuple)(struct nfattr *tb[], + struct nf_conntrack_tuple *t); + + /* Module (if any) which this is connected to. */ + struct module *me; +}; + +/* Existing built-in protocols */ +extern struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp6; +extern struct nf_conntrack_l4proto nf_conntrack_l4proto_udp4; +extern struct nf_conntrack_l4proto nf_conntrack_l4proto_udp6; +extern struct nf_conntrack_l4proto nf_conntrack_l4proto_generic; + +#define MAX_NF_CT_PROTO 256 +extern struct nf_conntrack_l4proto **nf_ct_protos[PF_MAX]; + +extern struct nf_conntrack_l4proto * +__nf_ct_l4proto_find(u_int16_t l3proto, u_int8_t l4proto); + +extern struct nf_conntrack_l4proto * +nf_ct_l4proto_find_get(u_int16_t l3proto, u_int8_t protocol); + +extern void nf_ct_l4proto_put(struct nf_conntrack_l4proto *p); + +/* Protocol registration. */ +extern int nf_conntrack_l4proto_register(struct nf_conntrack_l4proto *proto); +extern void nf_conntrack_l4proto_unregister(struct nf_conntrack_l4proto *proto); + +/* Generic netlink helpers */ +extern int nf_ct_port_tuple_to_nfattr(struct sk_buff *skb, + const struct nf_conntrack_tuple *tuple); +extern int nf_ct_port_nfattr_to_tuple(struct nfattr *tb[], + struct nf_conntrack_tuple *t); + +/* Log invalid packets */ +extern unsigned int nf_ct_log_invalid; + +#ifdef CONFIG_SYSCTL +#ifdef DEBUG_INVALID_PACKETS +#define LOG_INVALID(proto) \ + (nf_ct_log_invalid == (proto) || nf_ct_log_invalid == IPPROTO_RAW) +#else +#define LOG_INVALID(proto) \ + ((nf_ct_log_invalid == (proto) || nf_ct_log_invalid == IPPROTO_RAW) \ + && net_ratelimit()) +#endif +#else +#define LOG_INVALID(proto) 0 +#endif /* CONFIG_SYSCTL */ + +#endif /*_NF_CONNTRACK_PROTOCOL_H*/ diff --git a/include/net/netfilter/nf_conntrack_protocol.h b/include/net/netfilter/nf_conntrack_protocol.h deleted file mode 100644 index 1f33737fcea..00000000000 --- a/include/net/netfilter/nf_conntrack_protocol.h +++ /dev/null @@ -1,129 +0,0 @@ -/* - * Header for use in defining a given protocol for connection tracking. - * - * 16 Dec 2003: Yasuyuki Kozakai @USAGI - * - generalized L3 protocol dependent part. - * - * Derived from include/linux/netfiter_ipv4/ip_conntrack_protcol.h - */ - -#ifndef _NF_CONNTRACK_PROTOCOL_H -#define _NF_CONNTRACK_PROTOCOL_H -#include - -struct seq_file; -struct nfattr; - -struct nf_conntrack_protocol -{ - /* Next pointer. */ - struct list_head list; - - /* L3 Protocol number. */ - u_int16_t l3proto; - - /* Protocol number. */ - u_int8_t proto; - - /* Protocol name */ - const char *name; - - /* Try to fill in the third arg: dataoff is offset past network protocol - hdr. Return true if possible. */ - int (*pkt_to_tuple)(const struct sk_buff *skb, - unsigned int dataoff, - struct nf_conntrack_tuple *tuple); - - /* Invert the per-proto part of the tuple: ie. turn xmit into reply. - * Some packets can't be inverted: return 0 in that case. - */ - int (*invert_tuple)(struct nf_conntrack_tuple *inverse, - const struct nf_conntrack_tuple *orig); - - /* Print out the per-protocol part of the tuple. Return like seq_* */ - int (*print_tuple)(struct seq_file *s, - const struct nf_conntrack_tuple *); - - /* Print out the private part of the conntrack. */ - int (*print_conntrack)(struct seq_file *s, const struct nf_conn *); - - /* Returns verdict for packet, or -1 for invalid. */ - int (*packet)(struct nf_conn *conntrack, - const struct sk_buff *skb, - unsigned int dataoff, - enum ip_conntrack_info ctinfo, - int pf, - unsigned int hooknum); - - /* Called when a new connection for this protocol found; - * returns TRUE if it's OK. If so, packet() called next. */ - int (*new)(struct nf_conn *conntrack, const struct sk_buff *skb, - unsigned int dataoff); - - /* Called when a conntrack entry is destroyed */ - void (*destroy)(struct nf_conn *conntrack); - - int (*error)(struct sk_buff *skb, unsigned int dataoff, - enum ip_conntrack_info *ctinfo, - int pf, unsigned int hooknum); - - /* convert protoinfo to nfnetink attributes */ - int (*to_nfattr)(struct sk_buff *skb, struct nfattr *nfa, - const struct nf_conn *ct); - - /* convert nfnetlink attributes to protoinfo */ - int (*from_nfattr)(struct nfattr *tb[], struct nf_conn *ct); - - int (*tuple_to_nfattr)(struct sk_buff *skb, - const struct nf_conntrack_tuple *t); - int (*nfattr_to_tuple)(struct nfattr *tb[], - struct nf_conntrack_tuple *t); - - /* Module (if any) which this is connected to. */ - struct module *me; -}; - -/* Existing built-in protocols */ -extern struct nf_conntrack_protocol nf_conntrack_protocol_tcp6; -extern struct nf_conntrack_protocol nf_conntrack_protocol_udp4; -extern struct nf_conntrack_protocol nf_conntrack_protocol_udp6; -extern struct nf_conntrack_protocol nf_conntrack_generic_protocol; - -#define MAX_NF_CT_PROTO 256 -extern struct nf_conntrack_protocol **nf_ct_protos[PF_MAX]; - -extern struct nf_conntrack_protocol * -__nf_ct_proto_find(u_int16_t l3proto, u_int8_t protocol); - -extern struct nf_conntrack_protocol * -nf_ct_proto_find_get(u_int16_t l3proto, u_int8_t protocol); - -extern void nf_ct_proto_put(struct nf_conntrack_protocol *p); - -/* Protocol registration. */ -extern int nf_conntrack_protocol_register(struct nf_conntrack_protocol *proto); -extern void nf_conntrack_protocol_unregister(struct nf_conntrack_protocol *proto); - -/* Generic netlink helpers */ -extern int nf_ct_port_tuple_to_nfattr(struct sk_buff *skb, - const struct nf_conntrack_tuple *tuple); -extern int nf_ct_port_nfattr_to_tuple(struct nfattr *tb[], - struct nf_conntrack_tuple *t); - -/* Log invalid packets */ -extern unsigned int nf_ct_log_invalid; - -#ifdef CONFIG_SYSCTL -#ifdef DEBUG_INVALID_PACKETS -#define LOG_INVALID(proto) \ - (nf_ct_log_invalid == (proto) || nf_ct_log_invalid == IPPROTO_RAW) -#else -#define LOG_INVALID(proto) \ - ((nf_ct_log_invalid == (proto) || nf_ct_log_invalid == IPPROTO_RAW) \ - && net_ratelimit()) -#endif -#else -#define LOG_INVALID(proto) 0 -#endif /* CONFIG_SYSCTL */ - -#endif /*_NF_CONNTRACK_PROTOCOL_H*/ diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c index 0af803df82b..370df0fdb22 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c @@ -27,7 +27,7 @@ #include #include #include -#include +#include #include #include #include @@ -429,9 +429,9 @@ struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4 = { .me = THIS_MODULE, }; -extern struct nf_conntrack_protocol nf_conntrack_protocol_tcp4; -extern struct nf_conntrack_protocol nf_conntrack_protocol_udp4; -extern struct nf_conntrack_protocol nf_conntrack_protocol_icmp; +extern struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp4; +extern struct nf_conntrack_l4proto nf_conntrack_l4proto_udp4; +extern struct nf_conntrack_l4proto nf_conntrack_l4proto_icmp; MODULE_ALIAS("nf_conntrack-" __stringify(AF_INET)); MODULE_LICENSE("GPL"); @@ -448,19 +448,19 @@ static int __init nf_conntrack_l3proto_ipv4_init(void) return ret; } - ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_tcp4); + ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_tcp4); if (ret < 0) { printk("nf_conntrack_ipv4: can't register tcp.\n"); goto cleanup_sockopt; } - ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_udp4); + ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_udp4); if (ret < 0) { printk("nf_conntrack_ipv4: can't register udp.\n"); goto cleanup_tcp; } - ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_icmp); + ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_icmp); if (ret < 0) { printk("nf_conntrack_ipv4: can't register icmp.\n"); goto cleanup_udp; @@ -495,11 +495,11 @@ static int __init nf_conntrack_l3proto_ipv4_init(void) cleanup_ipv4: nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv4); cleanup_icmp: - nf_conntrack_protocol_unregister(&nf_conntrack_protocol_icmp); + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_icmp); cleanup_udp: - nf_conntrack_protocol_unregister(&nf_conntrack_protocol_udp4); + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_udp4); cleanup_tcp: - nf_conntrack_protocol_unregister(&nf_conntrack_protocol_tcp4); + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_tcp4); cleanup_sockopt: nf_unregister_sockopt(&so_getorigdst); return ret; @@ -513,9 +513,9 @@ static void __exit nf_conntrack_l3proto_ipv4_fini(void) #endif nf_unregister_hooks(ipv4_conntrack_ops, ARRAY_SIZE(ipv4_conntrack_ops)); nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv4); - nf_conntrack_protocol_unregister(&nf_conntrack_protocol_icmp); - nf_conntrack_protocol_unregister(&nf_conntrack_protocol_udp4); - nf_conntrack_protocol_unregister(&nf_conntrack_protocol_tcp4); + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_icmp); + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_udp4); + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_tcp4); nf_unregister_sockopt(&so_getorigdst); } diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c index 790f00d500c..08223a57a44 100644 --- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c +++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c @@ -22,7 +22,7 @@ #include #include #include -#include +#include #include unsigned long nf_ct_icmp_timeout __read_mostly = 30*HZ; @@ -152,7 +152,7 @@ icmp_error_message(struct sk_buff *skb, struct icmphdr icmp; struct iphdr ip; } _in, *inside; - struct nf_conntrack_protocol *innerproto; + struct nf_conntrack_l4proto *innerproto; struct nf_conntrack_tuple_hash *h; int dataoff; @@ -170,7 +170,7 @@ icmp_error_message(struct sk_buff *skb, return -NF_ACCEPT; } - innerproto = __nf_ct_proto_find(PF_INET, inside->ip.protocol); + innerproto = __nf_ct_l4proto_find(PF_INET, inside->ip.protocol); dataoff = skb->nh.iph->ihl*4 + sizeof(inside->icmp); /* Are they talking about one of our connections? */ if (!nf_ct_get_tuple(skb, dataoff, dataoff + inside->ip.ihl*4, PF_INET, @@ -321,11 +321,11 @@ static int icmp_nfattr_to_tuple(struct nfattr *tb[], } #endif -struct nf_conntrack_protocol nf_conntrack_protocol_icmp = +struct nf_conntrack_l4proto nf_conntrack_l4proto_icmp = { .list = { NULL, NULL }, .l3proto = PF_INET, - .proto = IPPROTO_ICMP, + .l4proto = IPPROTO_ICMP, .name = "icmp", .pkt_to_tuple = icmp_pkt_to_tuple, .invert_tuple = icmp_invert_tuple, @@ -343,4 +343,4 @@ struct nf_conntrack_protocol nf_conntrack_protocol_icmp = #endif }; -EXPORT_SYMBOL(nf_conntrack_protocol_icmp); +EXPORT_SYMBOL(nf_conntrack_l4proto_icmp); diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c index e5e53fff9e3..cdbba44017d 100644 --- a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c +++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c @@ -33,7 +33,7 @@ #include #include #include -#include +#include #include #include @@ -458,9 +458,9 @@ struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6 = { .me = THIS_MODULE, }; -extern struct nf_conntrack_protocol nf_conntrack_protocol_tcp6; -extern struct nf_conntrack_protocol nf_conntrack_protocol_udp6; -extern struct nf_conntrack_protocol nf_conntrack_protocol_icmpv6; +extern struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp6; +extern struct nf_conntrack_l4proto nf_conntrack_l4proto_udp6; +extern struct nf_conntrack_l4proto nf_conntrack_l4proto_icmpv6; extern int nf_ct_frag6_init(void); extern void nf_ct_frag6_cleanup(void); @@ -479,19 +479,19 @@ static int __init nf_conntrack_l3proto_ipv6_init(void) printk("nf_conntrack_ipv6: can't initialize frag6.\n"); return ret; } - ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_tcp6); + ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_tcp6); if (ret < 0) { printk("nf_conntrack_ipv6: can't register tcp.\n"); goto cleanup_frag6; } - ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_udp6); + ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_udp6); if (ret < 0) { printk("nf_conntrack_ipv6: can't register udp.\n"); goto cleanup_tcp; } - ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_icmpv6); + ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_icmpv6); if (ret < 0) { printk("nf_conntrack_ipv6: can't register icmpv6.\n"); goto cleanup_udp; @@ -527,11 +527,11 @@ static int __init nf_conntrack_l3proto_ipv6_init(void) cleanup_ipv6: nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv6); cleanup_icmpv6: - nf_conntrack_protocol_unregister(&nf_conntrack_protocol_icmpv6); + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_icmpv6); cleanup_udp: - nf_conntrack_protocol_unregister(&nf_conntrack_protocol_udp6); + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_udp6); cleanup_tcp: - nf_conntrack_protocol_unregister(&nf_conntrack_protocol_tcp6); + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_tcp6); cleanup_frag6: nf_ct_frag6_cleanup(); return ret; @@ -545,9 +545,9 @@ static void __exit nf_conntrack_l3proto_ipv6_fini(void) #endif nf_unregister_hooks(ipv6_conntrack_ops, ARRAY_SIZE(ipv6_conntrack_ops)); nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv6); - nf_conntrack_protocol_unregister(&nf_conntrack_protocol_icmpv6); - nf_conntrack_protocol_unregister(&nf_conntrack_protocol_udp6); - nf_conntrack_protocol_unregister(&nf_conntrack_protocol_tcp6); + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_icmpv6); + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_udp6); + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_tcp6); nf_ct_frag6_cleanup(); } diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c index 34d447208ff..2cdf225f162 100644 --- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c +++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c @@ -29,7 +29,7 @@ #include #include #include -#include +#include #include #include @@ -155,7 +155,7 @@ icmpv6_error_message(struct sk_buff *skb, struct nf_conntrack_tuple_hash *h; struct icmp6hdr _hdr, *hp; unsigned int inip6off; - struct nf_conntrack_protocol *inproto; + struct nf_conntrack_l4proto *inproto; u_int8_t inprotonum; unsigned int inprotoff; @@ -185,7 +185,7 @@ icmpv6_error_message(struct sk_buff *skb, return -NF_ACCEPT; } - inproto = __nf_ct_proto_find(PF_INET6, inprotonum); + inproto = __nf_ct_l4proto_find(PF_INET6, inprotonum); /* Are they talking about one of our connections? */ if (!nf_ct_get_tuple(skb, inip6off, inprotoff, PF_INET6, inprotonum, @@ -301,10 +301,10 @@ static int icmpv6_nfattr_to_tuple(struct nfattr *tb[], } #endif -struct nf_conntrack_protocol nf_conntrack_protocol_icmpv6 = +struct nf_conntrack_l4proto nf_conntrack_l4proto_icmpv6 = { .l3proto = PF_INET6, - .proto = IPPROTO_ICMPV6, + .l4proto = IPPROTO_ICMPV6, .name = "icmpv6", .pkt_to_tuple = icmpv6_pkt_to_tuple, .invert_tuple = icmpv6_invert_tuple, @@ -320,4 +320,4 @@ struct nf_conntrack_protocol nf_conntrack_protocol_icmpv6 = #endif }; -EXPORT_SYMBOL(nf_conntrack_protocol_icmpv6); +EXPORT_SYMBOL(nf_conntrack_l4proto_icmpv6); diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 6607ed41586..441a647b784 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -54,7 +54,7 @@ #include #include -#include +#include #include #include #include @@ -256,7 +256,7 @@ nf_ct_get_tuple(const struct sk_buff *skb, u_int8_t protonum, struct nf_conntrack_tuple *tuple, const struct nf_conntrack_l3proto *l3proto, - const struct nf_conntrack_protocol *protocol) + const struct nf_conntrack_l4proto *l4proto) { NF_CT_TUPLE_U_BLANK(tuple); @@ -267,14 +267,14 @@ nf_ct_get_tuple(const struct sk_buff *skb, tuple->dst.protonum = protonum; tuple->dst.dir = IP_CT_DIR_ORIGINAL; - return protocol->pkt_to_tuple(skb, dataoff, tuple); + return l4proto->pkt_to_tuple(skb, dataoff, tuple); } int nf_ct_invert_tuple(struct nf_conntrack_tuple *inverse, const struct nf_conntrack_tuple *orig, const struct nf_conntrack_l3proto *l3proto, - const struct nf_conntrack_protocol *protocol) + const struct nf_conntrack_l4proto *l4proto) { NF_CT_TUPLE_U_BLANK(inverse); @@ -285,7 +285,7 @@ nf_ct_invert_tuple(struct nf_conntrack_tuple *inverse, inverse->dst.dir = !orig->dst.dir; inverse->dst.protonum = orig->dst.protonum; - return protocol->invert_tuple(inverse, orig); + return l4proto->invert_tuple(inverse, orig); } static void @@ -305,7 +305,7 @@ destroy_conntrack(struct nf_conntrack *nfct) { struct nf_conn *ct = (struct nf_conn *)nfct; struct nf_conntrack_l3proto *l3proto; - struct nf_conntrack_protocol *proto; + struct nf_conntrack_l4proto *l4proto; DEBUGP("destroy_conntrack(%p)\n", ct); NF_CT_ASSERT(atomic_read(&nfct->use) == 0); @@ -321,9 +321,9 @@ destroy_conntrack(struct nf_conntrack *nfct) if (l3proto && l3proto->destroy) l3proto->destroy(ct); - proto = __nf_ct_proto_find(ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.l3num, ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.protonum); - if (proto && proto->destroy) - proto->destroy(ct); + l4proto = __nf_ct_l4proto_find(ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.l3num, ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.protonum); + if (l4proto && l4proto->destroy) + l4proto->destroy(ct); if (nf_conntrack_destroyed) nf_conntrack_destroyed(ct); @@ -647,7 +647,7 @@ void nf_conntrack_free(struct nf_conn *conntrack) static struct nf_conntrack_tuple_hash * init_conntrack(const struct nf_conntrack_tuple *tuple, struct nf_conntrack_l3proto *l3proto, - struct nf_conntrack_protocol *protocol, + struct nf_conntrack_l4proto *l4proto, struct sk_buff *skb, unsigned int dataoff) { @@ -655,7 +655,7 @@ init_conntrack(const struct nf_conntrack_tuple *tuple, struct nf_conntrack_tuple repl_tuple; struct nf_conntrack_expect *exp; - if (!nf_ct_invert_tuple(&repl_tuple, tuple, l3proto, protocol)) { + if (!nf_ct_invert_tuple(&repl_tuple, tuple, l3proto, l4proto)) { DEBUGP("Can't invert tuple.\n"); return NULL; } @@ -666,7 +666,7 @@ init_conntrack(const struct nf_conntrack_tuple *tuple, return (struct nf_conntrack_tuple_hash *)conntrack; } - if (!protocol->new(conntrack, skb, dataoff)) { + if (!l4proto->new(conntrack, skb, dataoff)) { nf_conntrack_free(conntrack); DEBUGP("init conntrack: can't track with proto module\n"); return NULL; @@ -718,7 +718,7 @@ resolve_normal_ct(struct sk_buff *skb, u_int16_t l3num, u_int8_t protonum, struct nf_conntrack_l3proto *l3proto, - struct nf_conntrack_protocol *proto, + struct nf_conntrack_l4proto *l4proto, int *set_reply, enum ip_conntrack_info *ctinfo) { @@ -728,7 +728,7 @@ resolve_normal_ct(struct sk_buff *skb, if (!nf_ct_get_tuple(skb, (unsigned int)(skb->nh.raw - skb->data), dataoff, l3num, protonum, &tuple, l3proto, - proto)) { + l4proto)) { DEBUGP("resolve_normal_ct: Can't get tuple\n"); return NULL; } @@ -736,7 +736,7 @@ resolve_normal_ct(struct sk_buff *skb, /* look for tuple match */ h = nf_conntrack_find_get(&tuple, NULL); if (!h) { - h = init_conntrack(&tuple, l3proto, proto, skb, dataoff); + h = init_conntrack(&tuple, l3proto, l4proto, skb, dataoff); if (!h) return NULL; if (IS_ERR(h)) @@ -774,7 +774,7 @@ nf_conntrack_in(int pf, unsigned int hooknum, struct sk_buff **pskb) struct nf_conn *ct; enum ip_conntrack_info ctinfo; struct nf_conntrack_l3proto *l3proto; - struct nf_conntrack_protocol *proto; + struct nf_conntrack_l4proto *l4proto; unsigned int dataoff; u_int8_t protonum; int set_reply = 0; @@ -792,19 +792,19 @@ nf_conntrack_in(int pf, unsigned int hooknum, struct sk_buff **pskb) return -ret; } - proto = __nf_ct_proto_find((u_int16_t)pf, protonum); + l4proto = __nf_ct_l4proto_find((u_int16_t)pf, protonum); /* It may be an special packet, error, unclean... * inverse of the return code tells to the netfilter * core what to do with the packet. */ - if (proto->error != NULL && - (ret = proto->error(*pskb, dataoff, &ctinfo, pf, hooknum)) <= 0) { + if (l4proto->error != NULL && + (ret = l4proto->error(*pskb, dataoff, &ctinfo, pf, hooknum)) <= 0) { NF_CT_STAT_INC(error); NF_CT_STAT_INC(invalid); return -ret; } - ct = resolve_normal_ct(*pskb, dataoff, pf, protonum, l3proto, proto, + ct = resolve_normal_ct(*pskb, dataoff, pf, protonum, l3proto, l4proto, &set_reply, &ctinfo); if (!ct) { /* Not valid part of a connection */ @@ -820,7 +820,7 @@ nf_conntrack_in(int pf, unsigned int hooknum, struct sk_buff **pskb) NF_CT_ASSERT((*pskb)->nfct); - ret = proto->packet(ct, *pskb, dataoff, ctinfo, pf, hooknum); + ret = l4proto->packet(ct, *pskb, dataoff, ctinfo, pf, hooknum); if (ret < 0) { /* Invalid: inverse of the return code tells * the netfilter core what to do */ @@ -842,7 +842,7 @@ int nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse, { return nf_ct_invert_tuple(inverse, orig, __nf_ct_l3proto_find(orig->src.l3num), - __nf_ct_proto_find(orig->src.l3num, + __nf_ct_l4proto_find(orig->src.l3num, orig->dst.protonum)); } @@ -1199,7 +1199,7 @@ int __init nf_conntrack_init(void) /* Don't NEED lock here, but good form anyway. */ write_lock_bh(&nf_conntrack_lock); for (i = 0; i < PF_MAX; i++) - nf_ct_l3protos[i] = &nf_conntrack_generic_l3proto; + nf_ct_l3protos[i] = &nf_conntrack_l3proto_generic; write_unlock_bh(&nf_conntrack_lock); /* For use by REJECT target */ diff --git a/net/netfilter/nf_conntrack_ecache.c b/net/netfilter/nf_conntrack_ecache.c index 35763e96fdd..5152efd511d 100644 --- a/net/netfilter/nf_conntrack_ecache.c +++ b/net/netfilter/nf_conntrack_ecache.c @@ -22,7 +22,7 @@ #include #include -#include +#include #include #include #include diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index 076e678b318..7269bffd8e4 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -338,7 +338,7 @@ static int exp_seq_show(struct seq_file *s, void *v) expect->tuple.dst.protonum); print_tuple(s, &expect->tuple, __nf_ct_l3proto_find(expect->tuple.src.l3num), - __nf_ct_proto_find(expect->tuple.src.l3num, + __nf_ct_l4proto_find(expect->tuple.src.l3num, expect->tuple.dst.protonum)); return seq_putc(s, '\n'); } diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index 2af92204722..5cbe55f66c1 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -26,7 +26,7 @@ #include #include -#include +#include #include #include diff --git a/net/netfilter/nf_conntrack_l3proto_generic.c b/net/netfilter/nf_conntrack_l3proto_generic.c index 21e0bc91cf2..1852c9d927d 100644 --- a/net/netfilter/nf_conntrack_l3proto_generic.c +++ b/net/netfilter/nf_conntrack_l3proto_generic.c @@ -26,7 +26,7 @@ #include #include -#include +#include #include #include #include @@ -84,7 +84,7 @@ static u_int32_t generic_get_features(const struct nf_conntrack_tuple *tuple) return NF_CT_F_BASIC; } -struct nf_conntrack_l3proto nf_conntrack_generic_l3proto = { +struct nf_conntrack_l3proto nf_conntrack_l3proto_generic = { .l3proto = PF_UNSPEC, .name = "unknown", .pkt_to_tuple = generic_pkt_to_tuple, diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index c3d53d91929..dc083022013 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -38,7 +38,7 @@ #include #include #include -#include +#include #include #include @@ -51,15 +51,15 @@ static char __initdata version[] = "0.93"; static inline int ctnetlink_dump_tuples_proto(struct sk_buff *skb, const struct nf_conntrack_tuple *tuple, - struct nf_conntrack_protocol *proto) + struct nf_conntrack_l4proto *l4proto) { int ret = 0; struct nfattr *nest_parms = NFA_NEST(skb, CTA_TUPLE_PROTO); NFA_PUT(skb, CTA_PROTO_NUM, sizeof(u_int8_t), &tuple->dst.protonum); - if (likely(proto->tuple_to_nfattr)) - ret = proto->tuple_to_nfattr(skb, tuple); + if (likely(l4proto->tuple_to_nfattr)) + ret = l4proto->tuple_to_nfattr(skb, tuple); NFA_NEST_END(skb, nest_parms); @@ -94,7 +94,7 @@ ctnetlink_dump_tuples(struct sk_buff *skb, { int ret; struct nf_conntrack_l3proto *l3proto; - struct nf_conntrack_protocol *proto; + struct nf_conntrack_l4proto *l4proto; l3proto = nf_ct_l3proto_find_get(tuple->src.l3num); ret = ctnetlink_dump_tuples_ip(skb, tuple, l3proto); @@ -103,9 +103,9 @@ ctnetlink_dump_tuples(struct sk_buff *skb, if (unlikely(ret < 0)) return ret; - proto = nf_ct_proto_find_get(tuple->src.l3num, tuple->dst.protonum); - ret = ctnetlink_dump_tuples_proto(skb, tuple, proto); - nf_ct_proto_put(proto); + l4proto = nf_ct_l4proto_find_get(tuple->src.l3num, tuple->dst.protonum); + ret = ctnetlink_dump_tuples_proto(skb, tuple, l4proto); + nf_ct_l4proto_put(l4proto); return ret; } @@ -142,27 +142,27 @@ nfattr_failure: static inline int ctnetlink_dump_protoinfo(struct sk_buff *skb, const struct nf_conn *ct) { - struct nf_conntrack_protocol *proto = nf_ct_proto_find_get(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num, ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum); + struct nf_conntrack_l4proto *l4proto = nf_ct_l4proto_find_get(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num, ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum); struct nfattr *nest_proto; int ret; - if (!proto->to_nfattr) { - nf_ct_proto_put(proto); + if (!l4proto->to_nfattr) { + nf_ct_l4proto_put(l4proto); return 0; } nest_proto = NFA_NEST(skb, CTA_PROTOINFO); - ret = proto->to_nfattr(skb, nest_proto, ct); + ret = l4proto->to_nfattr(skb, nest_proto, ct); - nf_ct_proto_put(proto); + nf_ct_l4proto_put(l4proto); NFA_NEST_END(skb, nest_proto); return ret; nfattr_failure: - nf_ct_proto_put(proto); + nf_ct_l4proto_put(l4proto); return -1; } @@ -493,7 +493,7 @@ ctnetlink_parse_tuple_proto(struct nfattr *attr, struct nf_conntrack_tuple *tuple) { struct nfattr *tb[CTA_PROTO_MAX]; - struct nf_conntrack_protocol *proto; + struct nf_conntrack_l4proto *l4proto; int ret = 0; nfattr_parse_nested(tb, CTA_PROTO_MAX, attr); @@ -505,12 +505,12 @@ ctnetlink_parse_tuple_proto(struct nfattr *attr, return -EINVAL; tuple->dst.protonum = *(u_int8_t *)NFA_DATA(tb[CTA_PROTO_NUM-1]); - proto = nf_ct_proto_find_get(tuple->src.l3num, tuple->dst.protonum); + l4proto = nf_ct_l4proto_find_get(tuple->src.l3num, tuple->dst.protonum); - if (likely(proto->nfattr_to_tuple)) - ret = proto->nfattr_to_tuple(tb, tuple); + if (likely(l4proto->nfattr_to_tuple)) + ret = l4proto->nfattr_to_tuple(tb, tuple); - nf_ct_proto_put(proto); + nf_ct_l4proto_put(l4proto); return ret; } @@ -890,18 +890,18 @@ static inline int ctnetlink_change_protoinfo(struct nf_conn *ct, struct nfattr *cda[]) { struct nfattr *tb[CTA_PROTOINFO_MAX], *attr = cda[CTA_PROTOINFO-1]; - struct nf_conntrack_protocol *proto; + struct nf_conntrack_l4proto *l4proto; u_int16_t npt = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum; u_int16_t l3num = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; int err = 0; nfattr_parse_nested(tb, CTA_PROTOINFO_MAX, attr); - proto = nf_ct_proto_find_get(l3num, npt); + l4proto = nf_ct_l4proto_find_get(l3num, npt); - if (proto->from_nfattr) - err = proto->from_nfattr(tb, ct); - nf_ct_proto_put(proto); + if (l4proto->from_nfattr) + err = l4proto->from_nfattr(tb, ct); + nf_ct_l4proto_put(l4proto); return err; } @@ -1081,7 +1081,7 @@ ctnetlink_exp_dump_mask(struct sk_buff *skb, { int ret; struct nf_conntrack_l3proto *l3proto; - struct nf_conntrack_protocol *proto; + struct nf_conntrack_l4proto *l4proto; struct nfattr *nest_parms = NFA_NEST(skb, CTA_EXPECT_MASK); l3proto = nf_ct_l3proto_find_get(tuple->src.l3num); @@ -1091,9 +1091,9 @@ ctnetlink_exp_dump_mask(struct sk_buff *skb, if (unlikely(ret < 0)) goto nfattr_failure; - proto = nf_ct_proto_find_get(tuple->src.l3num, tuple->dst.protonum); - ret = ctnetlink_dump_tuples_proto(skb, mask, proto); - nf_ct_proto_put(proto); + l4proto = nf_ct_l4proto_find_get(tuple->src.l3num, tuple->dst.protonum); + ret = ctnetlink_dump_tuples_proto(skb, mask, l4proto); + nf_ct_l4proto_put(l4proto); if (unlikely(ret < 0)) goto nfattr_failure; diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c index 24c61a38471..330b9acc62d 100644 --- a/net/netfilter/nf_conntrack_proto.c +++ b/net/netfilter/nf_conntrack_proto.c @@ -24,38 +24,38 @@ #include #include -#include +#include #include -struct nf_conntrack_protocol **nf_ct_protos[PF_MAX] __read_mostly; +struct nf_conntrack_l4proto **nf_ct_protos[PF_MAX] __read_mostly; struct nf_conntrack_l3proto *nf_ct_l3protos[PF_MAX] __read_mostly; -struct nf_conntrack_protocol * -__nf_ct_proto_find(u_int16_t l3proto, u_int8_t protocol) +struct nf_conntrack_l4proto * +__nf_ct_l4proto_find(u_int16_t l3proto, u_int8_t l4proto) { if (unlikely(l3proto >= AF_MAX || nf_ct_protos[l3proto] == NULL)) - return &nf_conntrack_generic_protocol; + return &nf_conntrack_l4proto_generic; - return nf_ct_protos[l3proto][protocol]; + return nf_ct_protos[l3proto][l4proto]; } /* this is guaranteed to always return a valid protocol helper, since * it falls back to generic_protocol */ -struct nf_conntrack_protocol * -nf_ct_proto_find_get(u_int16_t l3proto, u_int8_t protocol) +struct nf_conntrack_l4proto * +nf_ct_l4proto_find_get(u_int16_t l3proto, u_int8_t l4proto) { - struct nf_conntrack_protocol *p; + struct nf_conntrack_l4proto *p; preempt_disable(); - p = __nf_ct_proto_find(l3proto, protocol); + p = __nf_ct_l4proto_find(l3proto, l4proto); if (!try_module_get(p->me)) - p = &nf_conntrack_generic_protocol; + p = &nf_conntrack_l4proto_generic; preempt_enable(); return p; } -void nf_ct_proto_put(struct nf_conntrack_protocol *p) +void nf_ct_l4proto_put(struct nf_conntrack_l4proto *p) { module_put(p->me); } @@ -68,7 +68,7 @@ nf_ct_l3proto_find_get(u_int16_t l3proto) preempt_disable(); p = __nf_ct_l3proto_find(l3proto); if (!try_module_get(p->me)) - p = &nf_conntrack_generic_l3proto; + p = &nf_conntrack_l3proto_generic; preempt_enable(); return p; @@ -86,7 +86,7 @@ nf_ct_l3proto_try_module_get(unsigned short l3proto) struct nf_conntrack_l3proto *p; retry: p = nf_ct_l3proto_find_get(l3proto); - if (p == &nf_conntrack_generic_l3proto) { + if (p == &nf_conntrack_l3proto_generic) { ret = request_module("nf_conntrack-%d", l3proto); if (!ret) goto retry; @@ -114,14 +114,14 @@ static int kill_l3proto(struct nf_conn *i, void *data) ((struct nf_conntrack_l3proto *)data)->l3proto); } -static int kill_proto(struct nf_conn *i, void *data) +static int kill_l4proto(struct nf_conn *i, void *data) { - struct nf_conntrack_protocol *proto; - proto = (struct nf_conntrack_protocol *)data; + struct nf_conntrack_l4proto *l4proto; + l4proto = (struct nf_conntrack_l4proto *)data; return (i->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum == - proto->proto) && + l4proto->l4proto) && (i->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num == - proto->l3proto); + l4proto->l3proto); } int nf_conntrack_l3proto_register(struct nf_conntrack_l3proto *proto) @@ -129,7 +129,7 @@ int nf_conntrack_l3proto_register(struct nf_conntrack_l3proto *proto) int ret = 0; write_lock_bh(&nf_conntrack_lock); - if (nf_ct_l3protos[proto->l3proto] != &nf_conntrack_generic_l3proto) { + if (nf_ct_l3protos[proto->l3proto] != &nf_conntrack_l3proto_generic) { ret = -EBUSY; goto out; } @@ -143,7 +143,7 @@ out: void nf_conntrack_l3proto_unregister(struct nf_conntrack_l3proto *proto) { write_lock_bh(&nf_conntrack_lock); - nf_ct_l3protos[proto->l3proto] = &nf_conntrack_generic_l3proto; + nf_ct_l3protos[proto->l3proto] = &nf_conntrack_l3proto_generic; write_unlock_bh(&nf_conntrack_lock); /* Somebody could be still looking at the proto in bh. */ @@ -155,43 +155,43 @@ void nf_conntrack_l3proto_unregister(struct nf_conntrack_l3proto *proto) /* FIXME: Allow NULL functions and sub in pointers to generic for them. --RR */ -int nf_conntrack_protocol_register(struct nf_conntrack_protocol *proto) +int nf_conntrack_l4proto_register(struct nf_conntrack_l4proto *l4proto) { int ret = 0; retry: write_lock_bh(&nf_conntrack_lock); - if (nf_ct_protos[proto->l3proto]) { - if (nf_ct_protos[proto->l3proto][proto->proto] - != &nf_conntrack_generic_protocol) { + if (nf_ct_protos[l4proto->l3proto]) { + if (nf_ct_protos[l4proto->l3proto][l4proto->l4proto] + != &nf_conntrack_l4proto_generic) { ret = -EBUSY; goto out_unlock; } } else { /* l3proto may be loaded latter. */ - struct nf_conntrack_protocol **proto_array; + struct nf_conntrack_l4proto **proto_array; int i; write_unlock_bh(&nf_conntrack_lock); - proto_array = (struct nf_conntrack_protocol **) + proto_array = (struct nf_conntrack_l4proto **) kmalloc(MAX_NF_CT_PROTO * - sizeof(struct nf_conntrack_protocol *), + sizeof(struct nf_conntrack_l4proto *), GFP_KERNEL); if (proto_array == NULL) { ret = -ENOMEM; goto out; } for (i = 0; i < MAX_NF_CT_PROTO; i++) - proto_array[i] = &nf_conntrack_generic_protocol; + proto_array[i] = &nf_conntrack_l4proto_generic; write_lock_bh(&nf_conntrack_lock); - if (nf_ct_protos[proto->l3proto]) { + if (nf_ct_protos[l4proto->l3proto]) { /* bad timing, but no problem */ write_unlock_bh(&nf_conntrack_lock); kfree(proto_array); } else { - nf_ct_protos[proto->l3proto] = proto_array; + nf_ct_protos[l4proto->l3proto] = proto_array; write_unlock_bh(&nf_conntrack_lock); } @@ -202,7 +202,7 @@ retry: goto retry; } - nf_ct_protos[proto->l3proto][proto->proto] = proto; + nf_ct_protos[l4proto->l3proto][l4proto->l4proto] = l4proto; out_unlock: write_unlock_bh(&nf_conntrack_lock); @@ -210,16 +210,16 @@ out: return ret; } -void nf_conntrack_protocol_unregister(struct nf_conntrack_protocol *proto) +void nf_conntrack_l4proto_unregister(struct nf_conntrack_l4proto *l4proto) { write_lock_bh(&nf_conntrack_lock); - nf_ct_protos[proto->l3proto][proto->proto] - = &nf_conntrack_generic_protocol; + nf_ct_protos[l4proto->l3proto][l4proto->l4proto] + = &nf_conntrack_l4proto_generic; write_unlock_bh(&nf_conntrack_lock); /* Somebody could be still looking at the proto in bh. */ synchronize_net(); /* Remove all contrack entries for this protocol */ - nf_ct_iterate_cleanup(kill_proto, proto); + nf_ct_iterate_cleanup(kill_l4proto, l4proto); } diff --git a/net/netfilter/nf_conntrack_proto_generic.c b/net/netfilter/nf_conntrack_proto_generic.c index 26408bb0955..f44811e0b47 100644 --- a/net/netfilter/nf_conntrack_proto_generic.c +++ b/net/netfilter/nf_conntrack_proto_generic.c @@ -15,7 +15,7 @@ #include #include #include -#include +#include unsigned int nf_ct_generic_timeout __read_mostly = 600*HZ; @@ -71,10 +71,10 @@ static int new(struct nf_conn *conntrack, const struct sk_buff *skb, return 1; } -struct nf_conntrack_protocol nf_conntrack_generic_protocol = +struct nf_conntrack_l4proto nf_conntrack_l4proto_generic = { .l3proto = PF_UNSPEC, - .proto = 0, + .l4proto = 0, .name = "unknown", .pkt_to_tuple = generic_pkt_to_tuple, .invert_tuple = generic_invert_tuple, diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c index ba3e31f2acd..5115619138e 100644 --- a/net/netfilter/nf_conntrack_proto_sctp.c +++ b/net/netfilter/nf_conntrack_proto_sctp.c @@ -32,7 +32,7 @@ #include #include -#include +#include #include #if 0 @@ -509,9 +509,9 @@ static int sctp_new(struct nf_conn *conntrack, const struct sk_buff *skb, return 1; } -struct nf_conntrack_protocol nf_conntrack_protocol_sctp4 = { +struct nf_conntrack_l4proto nf_conntrack_l4proto_sctp4 = { .l3proto = PF_INET, - .proto = IPPROTO_SCTP, + .l4proto = IPPROTO_SCTP, .name = "sctp", .pkt_to_tuple = sctp_pkt_to_tuple, .invert_tuple = sctp_invert_tuple, @@ -523,9 +523,9 @@ struct nf_conntrack_protocol nf_conntrack_protocol_sctp4 = { .me = THIS_MODULE }; -struct nf_conntrack_protocol nf_conntrack_protocol_sctp6 = { +struct nf_conntrack_l4proto nf_conntrack_l4proto_sctp6 = { .l3proto = PF_INET6, - .proto = IPPROTO_SCTP, + .l4proto = IPPROTO_SCTP, .name = "sctp", .pkt_to_tuple = sctp_pkt_to_tuple, .invert_tuple = sctp_invert_tuple, @@ -625,14 +625,14 @@ int __init nf_conntrack_proto_sctp_init(void) { int ret; - ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_sctp4); + ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_sctp4); if (ret) { - printk("nf_conntrack_proto_sctp4: protocol register failed\n"); + printk("nf_conntrack_l4proto_sctp4: protocol register failed\n"); goto out; } - ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_sctp6); + ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_sctp6); if (ret) { - printk("nf_conntrack_proto_sctp6: protocol register failed\n"); + printk("nf_conntrack_l4proto_sctp6: protocol register failed\n"); goto cleanup_sctp4; } @@ -648,10 +648,10 @@ int __init nf_conntrack_proto_sctp_init(void) #ifdef CONFIG_SYSCTL cleanup: - nf_conntrack_protocol_unregister(&nf_conntrack_protocol_sctp6); + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_sctp6); #endif cleanup_sctp4: - nf_conntrack_protocol_unregister(&nf_conntrack_protocol_sctp4); + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_sctp4); out: DEBUGP("SCTP conntrack module loading %s\n", ret ? "failed": "succeeded"); @@ -660,8 +660,8 @@ int __init nf_conntrack_proto_sctp_init(void) void __exit nf_conntrack_proto_sctp_fini(void) { - nf_conntrack_protocol_unregister(&nf_conntrack_protocol_sctp6); - nf_conntrack_protocol_unregister(&nf_conntrack_protocol_sctp4); + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_sctp6); + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_sctp4); #ifdef CONFIG_SYSCTL unregister_sysctl_table(nf_ct_sysctl_header); #endif diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index e6b6f23ecdd..32b88b7efc4 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -42,7 +42,7 @@ #include #include #include -#include +#include #include #if 0 @@ -1169,10 +1169,10 @@ static int nfattr_to_tcp(struct nfattr *cda[], struct nf_conn *ct) } #endif -struct nf_conntrack_protocol nf_conntrack_protocol_tcp4 = +struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp4 = { .l3proto = PF_INET, - .proto = IPPROTO_TCP, + .l4proto = IPPROTO_TCP, .name = "tcp", .pkt_to_tuple = tcp_pkt_to_tuple, .invert_tuple = tcp_invert_tuple, @@ -1190,10 +1190,10 @@ struct nf_conntrack_protocol nf_conntrack_protocol_tcp4 = #endif }; -struct nf_conntrack_protocol nf_conntrack_protocol_tcp6 = +struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp6 = { .l3proto = PF_INET6, - .proto = IPPROTO_TCP, + .l4proto = IPPROTO_TCP, .name = "tcp", .pkt_to_tuple = tcp_pkt_to_tuple, .invert_tuple = tcp_invert_tuple, @@ -1211,5 +1211,5 @@ struct nf_conntrack_protocol nf_conntrack_protocol_tcp6 = #endif }; -EXPORT_SYMBOL(nf_conntrack_protocol_tcp4); -EXPORT_SYMBOL(nf_conntrack_protocol_tcp6); +EXPORT_SYMBOL(nf_conntrack_l4proto_tcp4); +EXPORT_SYMBOL(nf_conntrack_l4proto_tcp6); diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c index cf9db035d39..200d7d32436 100644 --- a/net/netfilter/nf_conntrack_proto_udp.c +++ b/net/netfilter/nf_conntrack_proto_udp.c @@ -26,7 +26,7 @@ #include #include #include -#include +#include #include unsigned int nf_ct_udp_timeout __read_mostly = 30*HZ; @@ -148,10 +148,10 @@ static int udp_error(struct sk_buff *skb, unsigned int dataoff, return NF_ACCEPT; } -struct nf_conntrack_protocol nf_conntrack_protocol_udp4 = +struct nf_conntrack_l4proto nf_conntrack_l4proto_udp4 = { .l3proto = PF_INET, - .proto = IPPROTO_UDP, + .l4proto = IPPROTO_UDP, .name = "udp", .pkt_to_tuple = udp_pkt_to_tuple, .invert_tuple = udp_invert_tuple, @@ -167,10 +167,10 @@ struct nf_conntrack_protocol nf_conntrack_protocol_udp4 = #endif }; -struct nf_conntrack_protocol nf_conntrack_protocol_udp6 = +struct nf_conntrack_l4proto nf_conntrack_l4proto_udp6 = { .l3proto = PF_INET6, - .proto = IPPROTO_UDP, + .l4proto = IPPROTO_UDP, .name = "udp", .pkt_to_tuple = udp_pkt_to_tuple, .invert_tuple = udp_invert_tuple, @@ -186,5 +186,5 @@ struct nf_conntrack_protocol nf_conntrack_protocol_udp6 = #endif }; -EXPORT_SYMBOL(nf_conntrack_protocol_udp4); -EXPORT_SYMBOL(nf_conntrack_protocol_udp6); +EXPORT_SYMBOL(nf_conntrack_l4proto_udp4); +EXPORT_SYMBOL(nf_conntrack_l4proto_udp6); diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index eb031d20e51..6ebac7d9ee6 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -35,7 +35,7 @@ #include #include #include -#include +#include #include #include @@ -54,9 +54,9 @@ DECLARE_PER_CPU(struct ip_conntrack_stat, nf_conntrack_stat); int print_tuple(struct seq_file *s, const struct nf_conntrack_tuple *tuple, struct nf_conntrack_l3proto *l3proto, - struct nf_conntrack_protocol *proto) + struct nf_conntrack_l4proto *l4proto) { - return l3proto->print_tuple(s, tuple) || proto->print_tuple(s, tuple); + return l3proto->print_tuple(s, tuple) || l4proto->print_tuple(s, tuple); } #ifdef CONFIG_NF_CT_ACCT @@ -135,7 +135,7 @@ static int ct_seq_show(struct seq_file *s, void *v) const struct nf_conntrack_tuple_hash *hash = v; const struct nf_conn *conntrack = nf_ct_tuplehash_to_ctrack(hash); struct nf_conntrack_l3proto *l3proto; - struct nf_conntrack_protocol *proto; + struct nf_conntrack_l4proto *l4proto; ASSERT_READ_LOCK(&nf_conntrack_lock); NF_CT_ASSERT(conntrack); @@ -148,16 +148,16 @@ static int ct_seq_show(struct seq_file *s, void *v) .tuple.src.l3num); NF_CT_ASSERT(l3proto); - proto = __nf_ct_proto_find(conntrack->tuplehash[IP_CT_DIR_ORIGINAL] + l4proto = __nf_ct_l4proto_find(conntrack->tuplehash[IP_CT_DIR_ORIGINAL] .tuple.src.l3num, conntrack->tuplehash[IP_CT_DIR_ORIGINAL] .tuple.dst.protonum); - NF_CT_ASSERT(proto); + NF_CT_ASSERT(l4proto); if (seq_printf(s, "%-8s %u %-8s %u %ld ", l3proto->name, conntrack->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num, - proto->name, + l4proto->name, conntrack->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum, timer_pending(&conntrack->timeout) ? (long)(conntrack->timeout.expires - jiffies)/HZ : 0) != 0) @@ -166,11 +166,11 @@ static int ct_seq_show(struct seq_file *s, void *v) if (l3proto->print_conntrack(s, conntrack)) return -ENOSPC; - if (proto->print_conntrack(s, conntrack)) + if (l4proto->print_conntrack(s, conntrack)) return -ENOSPC; if (print_tuple(s, &conntrack->tuplehash[IP_CT_DIR_ORIGINAL].tuple, - l3proto, proto)) + l3proto, l4proto)) return -ENOSPC; if (seq_print_counters(s, &conntrack->counters[IP_CT_DIR_ORIGINAL])) @@ -181,7 +181,7 @@ static int ct_seq_show(struct seq_file *s, void *v) return -ENOSPC; if (print_tuple(s, &conntrack->tuplehash[IP_CT_DIR_REPLY].tuple, - l3proto, proto)) + l3proto, l4proto)) return -ENOSPC; if (seq_print_counters(s, &conntrack->counters[IP_CT_DIR_REPLY])) @@ -655,8 +655,8 @@ EXPORT_SYMBOL(nf_ct_l3proto_try_module_get); EXPORT_SYMBOL(nf_ct_l3proto_module_put); EXPORT_SYMBOL(nf_conntrack_l3proto_register); EXPORT_SYMBOL(nf_conntrack_l3proto_unregister); -EXPORT_SYMBOL(nf_conntrack_protocol_register); -EXPORT_SYMBOL(nf_conntrack_protocol_unregister); +EXPORT_SYMBOL(nf_conntrack_l4proto_register); +EXPORT_SYMBOL(nf_conntrack_l4proto_unregister); EXPORT_SYMBOL(nf_ct_invert_tuplepr); EXPORT_SYMBOL(nf_conntrack_destroyed); EXPORT_SYMBOL(need_conntrack); @@ -665,9 +665,9 @@ EXPORT_SYMBOL(nf_conntrack_helper_unregister); EXPORT_SYMBOL(nf_ct_iterate_cleanup); EXPORT_SYMBOL(__nf_ct_refresh_acct); EXPORT_SYMBOL(nf_ct_protos); -EXPORT_SYMBOL(__nf_ct_proto_find); -EXPORT_SYMBOL(nf_ct_proto_find_get); -EXPORT_SYMBOL(nf_ct_proto_put); +EXPORT_SYMBOL(__nf_ct_l4proto_find); +EXPORT_SYMBOL(nf_ct_l4proto_find_get); +EXPORT_SYMBOL(nf_ct_l4proto_put); EXPORT_SYMBOL(nf_ct_l3proto_find_get); EXPORT_SYMBOL(nf_ct_l3proto_put); EXPORT_SYMBOL(nf_ct_l3protos); -- cgit v1.2.3-70-g09d2 From ae5718fb3dd0a11a4c9a061bf86417d52d58a6b3 Mon Sep 17 00:00:00 2001 From: Martin Josefsson Date: Wed, 29 Nov 2006 02:35:08 +0100 Subject: [NETFILTER]: nf_conntrack: more sanity checks in protocol registration/unregistration Add some more sanity checks when registering/unregistering l3/l4 protocols. Signed-off-by: Martin Josefsson Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack_l3proto.h | 2 +- include/net/netfilter/nf_conntrack_l4proto.h | 2 +- net/netfilter/nf_conntrack_core.c | 2 +- net/netfilter/nf_conntrack_proto.c | 55 +++++++++++++++++++++++++--- 4 files changed, 52 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_l3proto.h b/include/net/netfilter/nf_conntrack_l3proto.h index ffe23a1e0b9..11b4b29bfab 100644 --- a/include/net/netfilter/nf_conntrack_l3proto.h +++ b/include/net/netfilter/nf_conntrack_l3proto.h @@ -86,7 +86,7 @@ extern struct nf_conntrack_l3proto *nf_ct_l3protos[AF_MAX]; /* Protocol registration. */ extern int nf_conntrack_l3proto_register(struct nf_conntrack_l3proto *proto); -extern void nf_conntrack_l3proto_unregister(struct nf_conntrack_l3proto *proto); +extern int nf_conntrack_l3proto_unregister(struct nf_conntrack_l3proto *proto); extern struct nf_conntrack_l3proto * nf_ct_l3proto_find_get(u_int16_t l3proto); diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h index 0c6babd5439..5193e4857b1 100644 --- a/include/net/netfilter/nf_conntrack_l4proto.h +++ b/include/net/netfilter/nf_conntrack_l4proto.h @@ -102,7 +102,7 @@ extern void nf_ct_l4proto_put(struct nf_conntrack_l4proto *p); /* Protocol registration. */ extern int nf_conntrack_l4proto_register(struct nf_conntrack_l4proto *proto); -extern void nf_conntrack_l4proto_unregister(struct nf_conntrack_l4proto *proto); +extern int nf_conntrack_l4proto_unregister(struct nf_conntrack_l4proto *proto); /* Generic netlink helpers */ extern int nf_ct_port_tuple_to_nfattr(struct sk_buff *skb, diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 441a647b784..73615d32a80 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -1198,7 +1198,7 @@ int __init nf_conntrack_init(void) /* Don't NEED lock here, but good form anyway. */ write_lock_bh(&nf_conntrack_lock); - for (i = 0; i < PF_MAX; i++) + for (i = 0; i < AF_MAX; i++) nf_ct_l3protos[i] = &nf_conntrack_l3proto_generic; write_unlock_bh(&nf_conntrack_lock); diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c index 330b9acc62d..a6a3b1ddd00 100644 --- a/net/netfilter/nf_conntrack_proto.c +++ b/net/netfilter/nf_conntrack_proto.c @@ -28,7 +28,7 @@ #include struct nf_conntrack_l4proto **nf_ct_protos[PF_MAX] __read_mostly; -struct nf_conntrack_l3proto *nf_ct_l3protos[PF_MAX] __read_mostly; +struct nf_conntrack_l3proto *nf_ct_l3protos[AF_MAX] __read_mostly; struct nf_conntrack_l4proto * __nf_ct_l4proto_find(u_int16_t l3proto, u_int8_t l4proto) @@ -128,21 +128,40 @@ int nf_conntrack_l3proto_register(struct nf_conntrack_l3proto *proto) { int ret = 0; + if (proto->l3proto >= AF_MAX) { + ret = -EBUSY; + goto out; + } + write_lock_bh(&nf_conntrack_lock); if (nf_ct_l3protos[proto->l3proto] != &nf_conntrack_l3proto_generic) { ret = -EBUSY; - goto out; + goto out_unlock; } nf_ct_l3protos[proto->l3proto] = proto; -out: - write_unlock_bh(&nf_conntrack_lock); +out_unlock: + write_unlock_bh(&nf_conntrack_lock); +out: return ret; } -void nf_conntrack_l3proto_unregister(struct nf_conntrack_l3proto *proto) +int nf_conntrack_l3proto_unregister(struct nf_conntrack_l3proto *proto) { + int ret = 0; + + if (proto->l3proto >= AF_MAX) { + ret = -EBUSY; + goto out; + } + write_lock_bh(&nf_conntrack_lock); + if (nf_ct_l3protos[proto->l3proto] != proto) { + write_unlock_bh(&nf_conntrack_lock); + ret = -EBUSY; + goto out; + } + nf_ct_l3protos[proto->l3proto] = &nf_conntrack_l3proto_generic; write_unlock_bh(&nf_conntrack_lock); @@ -151,6 +170,9 @@ void nf_conntrack_l3proto_unregister(struct nf_conntrack_l3proto *proto) /* Remove all contrack entries for this protocol */ nf_ct_iterate_cleanup(kill_l3proto, proto); + +out: + return ret; } /* FIXME: Allow NULL functions and sub in pointers to generic for @@ -159,6 +181,11 @@ int nf_conntrack_l4proto_register(struct nf_conntrack_l4proto *l4proto) { int ret = 0; + if (l4proto->l3proto >= PF_MAX) { + ret = -EBUSY; + goto out; + } + retry: write_lock_bh(&nf_conntrack_lock); if (nf_ct_protos[l4proto->l3proto]) { @@ -210,9 +237,22 @@ out: return ret; } -void nf_conntrack_l4proto_unregister(struct nf_conntrack_l4proto *l4proto) +int nf_conntrack_l4proto_unregister(struct nf_conntrack_l4proto *l4proto) { + int ret = 0; + + if (l4proto->l3proto >= PF_MAX) { + ret = -EBUSY; + goto out; + } + write_lock_bh(&nf_conntrack_lock); + if (nf_ct_protos[l4proto->l3proto][l4proto->l4proto] + != l4proto) { + write_unlock_bh(&nf_conntrack_lock); + ret = -EBUSY; + goto out; + } nf_ct_protos[l4proto->l3proto][l4proto->l4proto] = &nf_conntrack_l4proto_generic; write_unlock_bh(&nf_conntrack_lock); @@ -222,4 +262,7 @@ void nf_conntrack_l4proto_unregister(struct nf_conntrack_l4proto *l4proto) /* Remove all contrack entries for this protocol */ nf_ct_iterate_cleanup(kill_l4proto, l4proto); + +out: + return ret; } -- cgit v1.2.3-70-g09d2 From 824621eddd2087b0e6f0b2e0da91e8d77e4919bf Mon Sep 17 00:00:00 2001 From: Martin Josefsson Date: Wed, 29 Nov 2006 02:35:11 +0100 Subject: [NETFILTER]: nf_conntrack: remove unused struct list_head from protocols Remove unused struct list_head from struct nf_conntrack_l3proto and nf_conntrack_l4proto as all protocols are kept in arrays, not linked lists. Signed-off-by: Martin Josefsson Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack_l3proto.h | 3 --- include/net/netfilter/nf_conntrack_l4proto.h | 3 --- net/ipv4/netfilter/nf_conntrack_proto_icmp.c | 1 - 3 files changed, 7 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_l3proto.h b/include/net/netfilter/nf_conntrack_l3proto.h index 11b4b29bfab..6364df05942 100644 --- a/include/net/netfilter/nf_conntrack_l3proto.h +++ b/include/net/netfilter/nf_conntrack_l3proto.h @@ -18,9 +18,6 @@ struct nfattr; struct nf_conntrack_l3proto { - /* Next pointer. */ - struct list_head list; - /* L3 Protocol Family number. ex) PF_INET */ u_int16_t l3proto; diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h index 5193e4857b1..c22804aa227 100644 --- a/include/net/netfilter/nf_conntrack_l4proto.h +++ b/include/net/netfilter/nf_conntrack_l4proto.h @@ -16,9 +16,6 @@ struct nfattr; struct nf_conntrack_l4proto { - /* Next pointer. */ - struct list_head list; - /* L3 Protocol number. */ u_int16_t l3proto; diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c index 08223a57a44..95fc22b4969 100644 --- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c +++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c @@ -323,7 +323,6 @@ static int icmp_nfattr_to_tuple(struct nfattr *tb[], struct nf_conntrack_l4proto nf_conntrack_l4proto_icmp = { - .list = { NULL, NULL }, .l3proto = PF_INET, .l4proto = IPPROTO_ICMP, .name = "icmp", -- cgit v1.2.3-70-g09d2 From f8eb24a89afa12b48fa7e39775faea6d64b8e538 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 29 Nov 2006 02:35:15 +0100 Subject: [NETFILTER]: nf_conntrack: move extern declaration to header files Using extern in a C file is a bad idea because the compiler can't catch type errors. Signed-off-by: Patrick McHardy --- include/net/netfilter/ipv4/nf_conntrack_ipv4.h | 4 ++++ include/net/netfilter/ipv6/nf_conntrack_ipv6.h | 25 +++++++++++++++++++++++++ include/net/netfilter/nf_conntrack.h | 5 +++++ net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 6 ------ net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 18 ------------------ net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c | 3 --- net/netfilter/nf_conntrack_expect.c | 1 - net/netfilter/nf_conntrack_l3proto_generic.c | 2 -- net/netfilter/nf_conntrack_standalone.c | 7 ------- 9 files changed, 34 insertions(+), 37 deletions(-) create mode 100644 include/net/netfilter/ipv6/nf_conntrack_ipv6.h (limited to 'include') diff --git a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h index 91684436af8..8c4f7e89ce5 100644 --- a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h +++ b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h @@ -37,4 +37,8 @@ struct nf_conntrack_ipv4 { struct sk_buff * nf_ct_ipv4_ct_gather_frags(struct sk_buff *skb); +extern struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp4; +extern struct nf_conntrack_l4proto nf_conntrack_l4proto_udp4; +extern struct nf_conntrack_l4proto nf_conntrack_l4proto_icmp; + #endif /*_NF_CONNTRACK_IPV4_H*/ diff --git a/include/net/netfilter/ipv6/nf_conntrack_ipv6.h b/include/net/netfilter/ipv6/nf_conntrack_ipv6.h new file mode 100644 index 00000000000..b4b6049e01f --- /dev/null +++ b/include/net/netfilter/ipv6/nf_conntrack_ipv6.h @@ -0,0 +1,25 @@ +#ifndef _NF_CONNTRACK_IPV6_H +#define _NF_CONNTRACK_IPV6_H + +extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6; + +extern struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp6; +extern struct nf_conntrack_l4proto nf_conntrack_l4proto_udp6; +extern struct nf_conntrack_l4proto nf_conntrack_l4proto_icmpv6; + +extern int nf_ct_ipv6_skip_exthdr(struct sk_buff *skb, int start, + u8 *nexthdrp, int len); + +extern int nf_ct_frag6_init(void); +extern void nf_ct_frag6_cleanup(void); +extern struct sk_buff *nf_ct_frag6_gather(struct sk_buff *skb); +extern void nf_ct_frag6_output(unsigned int hooknum, struct sk_buff *skb, + struct net_device *in, + struct net_device *out, + int (*okfn)(struct sk_buff *)); + +extern unsigned int nf_ct_frag6_timeout; +extern unsigned int nf_ct_frag6_low_thresh; +extern unsigned int nf_ct_frag6_high_thresh; + +#endif /* _NF_CONNTRACK_IPV6_H*/ diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 9a2950551cd..d0d0e649144 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -79,6 +79,8 @@ struct nf_conn_help { #include +#include + struct nf_conn { /* Usage count in here is 1 for hash table/destruct timer, 1 per skb, @@ -241,7 +243,10 @@ static inline int nf_ct_is_dying(struct nf_conn *ct) extern unsigned int nf_conntrack_htable_size; extern int nf_conntrack_checksum; +extern atomic_t nf_conntrack_count; +extern int nf_conntrack_max; +DECLARE_PER_CPU(struct ip_conntrack_stat, nf_conntrack_stat); #define NF_CT_STAT_INC(count) (__get_cpu_var(nf_conntrack_stat).count++) /* no helper, no nat */ diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c index 370df0fdb22..d1907082d7d 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c @@ -38,8 +38,6 @@ #define DEBUGP(format, args...) #endif -DECLARE_PER_CPU(struct nf_conntrack_stat, nf_conntrack_stat); - static int ipv4_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, struct nf_conntrack_tuple *tuple) { @@ -429,10 +427,6 @@ struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4 = { .me = THIS_MODULE, }; -extern struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp4; -extern struct nf_conntrack_l4proto nf_conntrack_l4proto_udp4; -extern struct nf_conntrack_l4proto nf_conntrack_l4proto_icmp; - MODULE_ALIAS("nf_conntrack-" __stringify(AF_INET)); MODULE_LICENSE("GPL"); diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c index cdbba44017d..9108ecc22be 100644 --- a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c +++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c @@ -43,8 +43,6 @@ #define DEBUGP(format, args...) #endif -DECLARE_PER_CPU(struct ip_conntrack_stat, nf_conntrack_stat); - static int ipv6_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, struct nf_conntrack_tuple *tuple) { @@ -211,11 +209,6 @@ out: return nf_conntrack_confirm(pskb); } -extern struct sk_buff *nf_ct_frag6_gather(struct sk_buff *skb); -extern void nf_ct_frag6_output(unsigned int hooknum, struct sk_buff *skb, - struct net_device *in, - struct net_device *out, - int (*okfn)(struct sk_buff *)); static unsigned int ipv6_defrag(unsigned int hooknum, struct sk_buff **pskb, const struct net_device *in, @@ -335,11 +328,6 @@ static struct nf_hook_ops ipv6_conntrack_ops[] = { /* From nf_conntrack_proto_icmpv6.c */ extern unsigned int nf_ct_icmpv6_timeout; -/* From nf_conntrack_reasm.c */ -extern unsigned int nf_ct_frag6_timeout; -extern unsigned int nf_ct_frag6_low_thresh; -extern unsigned int nf_ct_frag6_high_thresh; - static struct ctl_table_header *nf_ct_ipv6_sysctl_header; static ctl_table nf_ct_sysctl_table[] = { @@ -458,12 +446,6 @@ struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6 = { .me = THIS_MODULE, }; -extern struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp6; -extern struct nf_conntrack_l4proto nf_conntrack_l4proto_udp6; -extern struct nf_conntrack_l4proto nf_conntrack_l4proto_icmpv6; -extern int nf_ct_frag6_init(void); -extern void nf_ct_frag6_cleanup(void); - MODULE_ALIAS("nf_conntrack-" __stringify(AF_INET6)); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Yasuyuki KOZAKAI @USAGI "); diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c index 2cdf225f162..b3b468c0ef7 100644 --- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c +++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c @@ -142,9 +142,6 @@ static int icmpv6_new(struct nf_conn *conntrack, return 1; } -extern int -nf_ct_ipv6_skip_exthdr(struct sk_buff *skb, int start, u8 *nexthdrp, int len); -extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6; static int icmpv6_error_message(struct sk_buff *skb, unsigned int icmp6off, diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index 7269bffd8e4..79cfd79a42f 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -28,7 +28,6 @@ LIST_HEAD(nf_conntrack_expect_list); kmem_cache_t *nf_conntrack_expect_cachep __read_mostly; -DECLARE_PER_CPU(struct ip_conntrack_stat, nf_conntrack_stat); static unsigned int nf_conntrack_expect_next_id; /* nf_conntrack_expect helper functions */ diff --git a/net/netfilter/nf_conntrack_l3proto_generic.c b/net/netfilter/nf_conntrack_l3proto_generic.c index 1852c9d927d..3124b3a3010 100644 --- a/net/netfilter/nf_conntrack_l3proto_generic.c +++ b/net/netfilter/nf_conntrack_l3proto_generic.c @@ -37,8 +37,6 @@ #define DEBUGP(format, args...) #endif -DECLARE_PER_CPU(struct nf_conntrack_stat, nf_conntrack_stat); - static int generic_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, struct nf_conntrack_tuple *tuple) { diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index f87d333b0c0..2283a268694 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -44,9 +44,6 @@ MODULE_LICENSE("GPL"); -extern atomic_t nf_conntrack_count; -DECLARE_PER_CPU(struct ip_conntrack_stat, nf_conntrack_stat); - #ifdef CONFIG_PROC_FS int print_tuple(struct seq_file *s, const struct nf_conntrack_tuple *tuple, @@ -334,10 +331,6 @@ int nf_conntrack_checksum __read_mostly = 1; #ifdef CONFIG_SYSCTL -/* From nf_conntrack_core.c */ -extern int nf_conntrack_max; -extern unsigned int nf_conntrack_htable_size; - /* From nf_conntrack_proto_tcp.c */ extern unsigned int nf_ct_tcp_timeout_syn_sent; extern unsigned int nf_ct_tcp_timeout_syn_recv; -- cgit v1.2.3-70-g09d2 From d62f9ed4a490309bd9e5df0b42ba5d096e7b5902 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 29 Nov 2006 02:35:17 +0100 Subject: [NETFILTER]: nf_conntrack: automatic sysctl registation for conntrack protocols Add helper functions for sysctl registration with optional instantiating of common path elements (like net/netfilter) and use it for support for automatic registation of conntrack protocol sysctls. Signed-off-by: Patrick McHardy --- include/linux/netfilter.h | 10 ++ include/net/netfilter/nf_conntrack_l3proto.h | 6 ++ include/net/netfilter/nf_conntrack_l4proto.h | 6 ++ net/netfilter/Makefile | 1 + net/netfilter/nf_conntrack_proto.c | 102 ++++++++++++++++++++ net/netfilter/nf_sysctl.c | 134 +++++++++++++++++++++++++++ 6 files changed, 259 insertions(+) create mode 100644 net/netfilter/nf_sysctl.c (limited to 'include') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index 6ab5e2d6133..f6f3fcbd70e 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -117,6 +117,16 @@ void nf_unregister_hooks(struct nf_hook_ops *reg, unsigned int n); int nf_register_sockopt(struct nf_sockopt_ops *reg); void nf_unregister_sockopt(struct nf_sockopt_ops *reg); +#ifdef CONFIG_SYSCTL +/* Sysctl registration */ +struct ctl_table_header *nf_register_sysctl_table(struct ctl_table *path, + struct ctl_table *table); +void nf_unregister_sysctl_table(struct ctl_table_header *header, + struct ctl_table *table); +extern struct ctl_table nf_net_netfilter_sysctl_path[]; +extern struct ctl_table nf_net_ipv4_netfilter_sysctl_path[]; +#endif /* CONFIG_SYSCTL */ + extern struct list_head nf_hooks[NPROTO][NF_MAX_HOOKS]; /* those NF_LOG_* defines and struct nf_loginfo are legacy definitios that will diff --git a/include/net/netfilter/nf_conntrack_l3proto.h b/include/net/netfilter/nf_conntrack_l3proto.h index 6364df05942..664ddcffe00 100644 --- a/include/net/netfilter/nf_conntrack_l3proto.h +++ b/include/net/netfilter/nf_conntrack_l3proto.h @@ -75,6 +75,12 @@ struct nf_conntrack_l3proto int (*nfattr_to_tuple)(struct nfattr *tb[], struct nf_conntrack_tuple *t); +#ifdef CONFIG_SYSCTL + struct ctl_table_header *ctl_table_header; + struct ctl_table *ctl_table_path; + struct ctl_table *ctl_table; +#endif /* CONFIG_SYSCTL */ + /* Module (if any) which this is connected to. */ struct module *me; }; diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h index c22804aa227..fe1e8fa30d2 100644 --- a/include/net/netfilter/nf_conntrack_l4proto.h +++ b/include/net/netfilter/nf_conntrack_l4proto.h @@ -76,6 +76,12 @@ struct nf_conntrack_l4proto int (*nfattr_to_tuple)(struct nfattr *tb[], struct nf_conntrack_tuple *t); +#ifdef CONFIG_SYSCTL + struct ctl_table_header **ctl_table_header; + struct ctl_table *ctl_table; + unsigned int *ctl_table_users; +#endif /* CONFIG_SYSCTL */ + /* Module (if any) which this is connected to. */ struct module *me; }; diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 627105df104..84d529ded95 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -4,6 +4,7 @@ nf_conntrack-y := nf_conntrack_core.o nf_conntrack_standalone.o nf_conntrack_exp nf_conntrack-$(CONFIG_NF_CONNTRACK_EVENTS) += nf_conntrack_ecache.o obj-$(CONFIG_NETFILTER) = netfilter.o +obj-$(CONFIG_SYSCTL) += nf_sysctl.o obj-$(CONFIG_NETFILTER_NETLINK) += nfnetlink.o obj-$(CONFIG_NETFILTER_NETLINK_QUEUE) += nfnetlink_queue.o diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c index a6a3b1ddd00..941b5c3754a 100644 --- a/net/netfilter/nf_conntrack_proto.c +++ b/net/netfilter/nf_conntrack_proto.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include @@ -30,6 +31,34 @@ struct nf_conntrack_l4proto **nf_ct_protos[PF_MAX] __read_mostly; struct nf_conntrack_l3proto *nf_ct_l3protos[AF_MAX] __read_mostly; +#ifdef CONFIG_SYSCTL +static DEFINE_MUTEX(nf_ct_proto_sysctl_mutex); + +static int +nf_ct_register_sysctl(struct ctl_table_header **header, struct ctl_table *path, + struct ctl_table *table, unsigned int *users) +{ + if (*header == NULL) { + *header = nf_register_sysctl_table(path, table); + if (*header == NULL) + return -ENOMEM; + } + if (users != NULL) + (*users)++; + return 0; +} + +static void +nf_ct_unregister_sysctl(struct ctl_table_header **header, + struct ctl_table *table, unsigned int *users) +{ + if (users != NULL && --*users > 0) + return; + nf_unregister_sysctl_table(*header, table); + *header = NULL; +} +#endif + struct nf_conntrack_l4proto * __nf_ct_l4proto_find(u_int16_t l3proto, u_int8_t l4proto) { @@ -124,6 +153,33 @@ static int kill_l4proto(struct nf_conn *i, void *data) l4proto->l3proto); } +static int nf_ct_l3proto_register_sysctl(struct nf_conntrack_l3proto *l3proto) +{ + int err = 0; + +#ifdef CONFIG_SYSCTL + mutex_lock(&nf_ct_proto_sysctl_mutex); + if (l3proto->ctl_table != NULL) { + err = nf_ct_register_sysctl(&l3proto->ctl_table_header, + l3proto->ctl_table_path, + l3proto->ctl_table, NULL); + } + mutex_unlock(&nf_ct_proto_sysctl_mutex); +#endif + return err; +} + +static void nf_ct_l3proto_unregister_sysctl(struct nf_conntrack_l3proto *l3proto) +{ +#ifdef CONFIG_SYSCTL + mutex_lock(&nf_ct_proto_sysctl_mutex); + if (l3proto->ctl_table_header != NULL) + nf_ct_unregister_sysctl(&l3proto->ctl_table_header, + l3proto->ctl_table, NULL); + mutex_unlock(&nf_ct_proto_sysctl_mutex); +#endif +} + int nf_conntrack_l3proto_register(struct nf_conntrack_l3proto *proto) { int ret = 0; @@ -139,6 +195,12 @@ int nf_conntrack_l3proto_register(struct nf_conntrack_l3proto *proto) goto out_unlock; } nf_ct_l3protos[proto->l3proto] = proto; + write_unlock_bh(&nf_conntrack_lock); + + ret = nf_ct_l3proto_register_sysctl(proto); + if (ret < 0) + nf_conntrack_l3proto_unregister(proto); + return ret; out_unlock: write_unlock_bh(&nf_conntrack_lock); @@ -165,6 +227,8 @@ int nf_conntrack_l3proto_unregister(struct nf_conntrack_l3proto *proto) nf_ct_l3protos[proto->l3proto] = &nf_conntrack_l3proto_generic; write_unlock_bh(&nf_conntrack_lock); + nf_ct_l3proto_unregister_sysctl(proto); + /* Somebody could be still looking at the proto in bh. */ synchronize_net(); @@ -175,6 +239,36 @@ out: return ret; } +static int nf_ct_l4proto_register_sysctl(struct nf_conntrack_l4proto *l4proto) +{ + int err = 0; + +#ifdef CONFIG_SYSCTL + mutex_lock(&nf_ct_proto_sysctl_mutex); + if (l4proto->ctl_table != NULL) { + err = nf_ct_register_sysctl(l4proto->ctl_table_header, + nf_net_netfilter_sysctl_path, + l4proto->ctl_table, + l4proto->ctl_table_users); + } + mutex_unlock(&nf_ct_proto_sysctl_mutex); +#endif + return err; +} + +static void nf_ct_l4proto_unregister_sysctl(struct nf_conntrack_l4proto *l4proto) +{ +#ifdef CONFIG_SYSCTL + mutex_lock(&nf_ct_proto_sysctl_mutex); + if (l4proto->ctl_table_header != NULL && + *l4proto->ctl_table_header != NULL) + nf_ct_unregister_sysctl(l4proto->ctl_table_header, + l4proto->ctl_table, + l4proto->ctl_table_users); + mutex_unlock(&nf_ct_proto_sysctl_mutex); +#endif +} + /* FIXME: Allow NULL functions and sub in pointers to generic for them. --RR */ int nf_conntrack_l4proto_register(struct nf_conntrack_l4proto *l4proto) @@ -230,6 +324,12 @@ retry: } nf_ct_protos[l4proto->l3proto][l4proto->l4proto] = l4proto; + write_unlock_bh(&nf_conntrack_lock); + + ret = nf_ct_l4proto_register_sysctl(l4proto); + if (ret < 0) + nf_conntrack_l4proto_unregister(l4proto); + return ret; out_unlock: write_unlock_bh(&nf_conntrack_lock); @@ -257,6 +357,8 @@ int nf_conntrack_l4proto_unregister(struct nf_conntrack_l4proto *l4proto) = &nf_conntrack_l4proto_generic; write_unlock_bh(&nf_conntrack_lock); + nf_ct_l4proto_unregister_sysctl(l4proto); + /* Somebody could be still looking at the proto in bh. */ synchronize_net(); diff --git a/net/netfilter/nf_sysctl.c b/net/netfilter/nf_sysctl.c new file mode 100644 index 00000000000..06ddddb2911 --- /dev/null +++ b/net/netfilter/nf_sysctl.c @@ -0,0 +1,134 @@ +/* nf_sysctl.c netfilter sysctl registration/unregistation + * + * Copyright (c) 2006 Patrick McHardy + */ +#include +#include +#include +#include + +static void +path_free(struct ctl_table *path, struct ctl_table *table) +{ + struct ctl_table *t, *next; + + for (t = path; t != NULL && t != table; t = next) { + next = t->child; + kfree(t); + } +} + +static struct ctl_table * +path_dup(struct ctl_table *path, struct ctl_table *table) +{ + struct ctl_table *t, *last = NULL, *tmp; + + for (t = path; t != NULL; t = t->child) { + /* twice the size since path elements are terminated by an + * empty element */ + tmp = kmemdup(t, 2 * sizeof(*t), GFP_KERNEL); + if (tmp == NULL) { + if (last != NULL) + path_free(path, table); + return NULL; + } + + if (last != NULL) + last->child = tmp; + else + path = tmp; + last = tmp; + } + + if (last != NULL) + last->child = table; + else + path = table; + + return path; +} + +struct ctl_table_header * +nf_register_sysctl_table(struct ctl_table *path, struct ctl_table *table) +{ + struct ctl_table_header *header; + + path = path_dup(path, table); + if (path == NULL) + return NULL; + header = register_sysctl_table(path, 0); + if (header == NULL) + path_free(path, table); + return header; +} +EXPORT_SYMBOL_GPL(nf_register_sysctl_table); + +void +nf_unregister_sysctl_table(struct ctl_table_header *header, + struct ctl_table *table) +{ + struct ctl_table *path = header->ctl_table; + + unregister_sysctl_table(header); + path_free(path, table); +} +EXPORT_SYMBOL_GPL(nf_unregister_sysctl_table); + +/* net/netfilter */ +static struct ctl_table nf_net_netfilter_table[] = { + { + .ctl_name = NET_NETFILTER, + .procname = "netfilter", + .mode = 0555, + }, + { + .ctl_name = 0 + } +}; +struct ctl_table nf_net_netfilter_sysctl_path[] = { + { + .ctl_name = CTL_NET, + .procname = "net", + .mode = 0555, + .child = nf_net_netfilter_table, + }, + { + .ctl_name = 0 + } +}; +EXPORT_SYMBOL_GPL(nf_net_netfilter_sysctl_path); + +/* net/ipv4/netfilter */ +static struct ctl_table nf_net_ipv4_netfilter_table[] = { + { + .ctl_name = NET_IPV4_NETFILTER, + .procname = "netfilter", + .mode = 0555, + }, + { + .ctl_name = 0 + } +}; +static struct ctl_table nf_net_ipv4_table[] = { + { + .ctl_name = NET_IPV4, + .procname = "ipv4", + .mode = 0555, + .child = nf_net_ipv4_netfilter_table, + }, + { + .ctl_name = 0 + } +}; +struct ctl_table nf_net_ipv4_netfilter_sysctl_path[] = { + { + .ctl_name = CTL_NET, + .procname = "net", + .mode = 0555, + .child = nf_net_ipv4_table, + }, + { + .ctl_name = 0 + } +}; +EXPORT_SYMBOL_GPL(nf_net_ipv4_netfilter_sysctl_path); -- cgit v1.2.3-70-g09d2 From a999e6837603e4b5a164333c93918a1292f074c8 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 29 Nov 2006 02:35:20 +0100 Subject: [NETFILTER]: nf_conntrack: sysctl compatibility with old connection tracking This patch adds an option to keep the connection tracking sysctls visible under their old names. Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack_l4proto.h | 6 +- net/ipv4/netfilter/Kconfig | 11 +++ net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 58 ++++++++++++++ net/ipv4/netfilter/nf_conntrack_proto_icmp.c | 18 +++++ net/netfilter/nf_conntrack_core.c | 2 + net/netfilter/nf_conntrack_proto.c | 20 +++++ net/netfilter/nf_conntrack_proto_generic.c | 18 +++++ net/netfilter/nf_conntrack_proto_sctp.c | 67 ++++++++++++++++ net/netfilter/nf_conntrack_proto_tcp.c | 107 +++++++++++++++++++++++++ net/netfilter/nf_conntrack_proto_udp.c | 26 ++++++ 10 files changed, 332 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h index fe1e8fa30d2..fc8af08ff54 100644 --- a/include/net/netfilter/nf_conntrack_l4proto.h +++ b/include/net/netfilter/nf_conntrack_l4proto.h @@ -80,7 +80,11 @@ struct nf_conntrack_l4proto struct ctl_table_header **ctl_table_header; struct ctl_table *ctl_table; unsigned int *ctl_table_users; -#endif /* CONFIG_SYSCTL */ +#ifdef CONFIG_NF_CONNTRACK_PROC_COMPAT + struct ctl_table_header *ctl_compat_table_header; + struct ctl_table *ctl_compat_table; +#endif +#endif /* Module (if any) which this is connected to. */ struct module *me; diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index d88c292f118..4ac5b5c4678 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -19,6 +19,17 @@ config NF_CONNTRACK_IPV4 To compile it as a module, choose M here. If unsure, say N. +config NF_CONNTRACK_PROC_COMPAT + bool "proc/sysctl compatibility with old connection tracking" + depends on NF_CONNTRACK + default y + help + This option enables /proc and sysctl compatibility with the old + layer 3 dependant connection tracking. This is needed to keep + old programs that have not been adapted to the new names working. + + If unsure, say Y. + # connection tracking, helpers and protocols config IP_NF_CONNTRACK tristate "Connection tracking (required for masq/NAT)" diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c index 786c4ce96cd..bcec6822f2e 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c @@ -266,6 +266,60 @@ static struct nf_hook_ops ipv4_conntrack_ops[] = { }, }; +#if defined(CONFIG_SYSCTL) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT) +static int log_invalid_proto_min = 0; +static int log_invalid_proto_max = 255; + +static ctl_table ip_ct_sysctl_table[] = { + { + .ctl_name = NET_IPV4_NF_CONNTRACK_MAX, + .procname = "ip_conntrack_max", + .data = &nf_conntrack_max, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_COUNT, + .procname = "ip_conntrack_count", + .data = &nf_conntrack_count, + .maxlen = sizeof(int), + .mode = 0444, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_BUCKETS, + .procname = "ip_conntrack_buckets", + .data = &nf_conntrack_htable_size, + .maxlen = sizeof(unsigned int), + .mode = 0444, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_CHECKSUM, + .procname = "ip_conntrack_checksum", + .data = &nf_conntrack_checksum, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_LOG_INVALID, + .procname = "ip_conntrack_log_invalid", + .data = &nf_ct_log_invalid, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_minmax, + .strategy = &sysctl_intvec, + .extra1 = &log_invalid_proto_min, + .extra2 = &log_invalid_proto_max, + }, + { + .ctl_name = 0 + } +}; +#endif /* CONFIG_SYSCTL && CONFIG_NF_CONNTRACK_PROC_COMPAT */ + /* Fast function for those who don't want to parse /proc (and I don't blame them). */ /* Reversing the socket's dst/src point of view gives us the reply @@ -385,6 +439,10 @@ struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4 = { defined(CONFIG_NF_CT_NETLINK_MODULE) .tuple_to_nfattr = ipv4_tuple_to_nfattr, .nfattr_to_tuple = ipv4_nfattr_to_tuple, +#endif +#if defined(CONFIG_SYSCTL) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT) + .ctl_table_path = nf_net_ipv4_netfilter_sysctl_path, + .ctl_table = ip_ct_sysctl_table, #endif .me = THIS_MODULE, }; diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c index c59f28193a3..46aa44abc07 100644 --- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c +++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c @@ -336,6 +336,21 @@ static struct ctl_table icmp_sysctl_table[] = { .ctl_name = 0 } }; +#ifdef CONFIG_NF_CONNTRACK_PROC_COMPAT +static struct ctl_table icmp_compat_sysctl_table[] = { + { + .ctl_name = NET_IPV4_NF_CONNTRACK_ICMP_TIMEOUT, + .procname = "ip_conntrack_icmp_timeout", + .data = &nf_ct_icmp_timeout, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = 0 + } +}; +#endif /* CONFIG_NF_CONNTRACK_PROC_COMPAT */ #endif /* CONFIG_SYSCTL */ struct nf_conntrack_l4proto nf_conntrack_l4proto_icmp = @@ -360,6 +375,9 @@ struct nf_conntrack_l4proto nf_conntrack_l4proto_icmp = #ifdef CONFIG_SYSCTL .ctl_table_header = &icmp_sysctl_header, .ctl_table = icmp_sysctl_table, +#ifdef CONFIG_NF_CONNTRACK_PROC_COMPAT + .ctl_compat_table = icmp_compat_sysctl_table, +#endif #endif }; diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 4b972791149..a6728067780 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -66,10 +66,12 @@ DEFINE_RWLOCK(nf_conntrack_lock); /* nf_conntrack_standalone needs this */ atomic_t nf_conntrack_count = ATOMIC_INIT(0); +EXPORT_SYMBOL_GPL(nf_conntrack_count); void (*nf_conntrack_destroyed)(struct nf_conn *conntrack) = NULL; unsigned int nf_conntrack_htable_size __read_mostly; int nf_conntrack_max __read_mostly; +EXPORT_SYMBOL_GPL(nf_conntrack_max); struct list_head *nf_conntrack_hash __read_mostly; struct nf_conn nf_conntrack_untracked __read_mostly; unsigned int nf_ct_log_invalid __read_mostly; diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c index 891c9c56c31..4798afcbbb0 100644 --- a/net/netfilter/nf_conntrack_proto.c +++ b/net/netfilter/nf_conntrack_proto.c @@ -250,7 +250,22 @@ static int nf_ct_l4proto_register_sysctl(struct nf_conntrack_l4proto *l4proto) nf_net_netfilter_sysctl_path, l4proto->ctl_table, l4proto->ctl_table_users); + if (err < 0) + goto out; } +#ifdef CONFIG_NF_CONNTRACK_PROC_COMPAT + if (l4proto->ctl_compat_table != NULL) { + err = nf_ct_register_sysctl(&l4proto->ctl_compat_table_header, + nf_net_ipv4_netfilter_sysctl_path, + l4proto->ctl_compat_table, NULL); + if (err == 0) + goto out; + nf_ct_unregister_sysctl(l4proto->ctl_table_header, + l4proto->ctl_table, + l4proto->ctl_table_users); + } +#endif /* CONFIG_NF_CONNTRACK_PROC_COMPAT */ +out: mutex_unlock(&nf_ct_proto_sysctl_mutex); #endif /* CONFIG_SYSCTL */ return err; @@ -265,6 +280,11 @@ static void nf_ct_l4proto_unregister_sysctl(struct nf_conntrack_l4proto *l4proto nf_ct_unregister_sysctl(l4proto->ctl_table_header, l4proto->ctl_table, l4proto->ctl_table_users); +#ifdef CONFIG_NF_CONNTRACK_PROC_COMPAT + if (l4proto->ctl_compat_table_header != NULL) + nf_ct_unregister_sysctl(&l4proto->ctl_compat_table_header, + l4proto->ctl_compat_table, NULL); +#endif /* CONFIG_NF_CONNTRACK_PROC_COMPAT */ mutex_unlock(&nf_ct_proto_sysctl_mutex); #endif /* CONFIG_SYSCTL */ } diff --git a/net/netfilter/nf_conntrack_proto_generic.c b/net/netfilter/nf_conntrack_proto_generic.c index 15306b95251..69902531c23 100644 --- a/net/netfilter/nf_conntrack_proto_generic.c +++ b/net/netfilter/nf_conntrack_proto_generic.c @@ -86,6 +86,21 @@ static struct ctl_table generic_sysctl_table[] = { .ctl_name = 0 } }; +#ifdef CONFIG_NF_CONNTRACK_PROC_COMPAT +static struct ctl_table generic_compat_sysctl_table[] = { + { + .ctl_name = NET_IPV4_NF_CONNTRACK_GENERIC_TIMEOUT, + .procname = "ip_conntrack_generic_timeout", + .data = &nf_ct_generic_timeout, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = 0 + } +}; +#endif /* CONFIG_NF_CONNTRACK_PROC_COMPAT */ #endif /* CONFIG_SYSCTL */ struct nf_conntrack_l4proto nf_conntrack_l4proto_generic = @@ -102,5 +117,8 @@ struct nf_conntrack_l4proto nf_conntrack_l4proto_generic = #ifdef CONFIG_SYSCTL .ctl_table_header = &generic_sysctl_header, .ctl_table = generic_sysctl_table, +#ifdef CONFIG_NF_CONNTRACK_PROC_COMPAT + .ctl_compat_table = generic_compat_sysctl_table, +#endif #endif }; diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c index cc693308878..7c64b9c1cf4 100644 --- a/net/netfilter/nf_conntrack_proto_sctp.c +++ b/net/netfilter/nf_conntrack_proto_sctp.c @@ -573,6 +573,70 @@ static struct ctl_table sctp_sysctl_table[] = { .ctl_name = 0 } }; + +#ifdef CONFIG_NF_CONNTRACK_PROC_COMPAT +static struct ctl_table sctp_compat_sysctl_table[] = { + { + .ctl_name = NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_CLOSED, + .procname = "ip_conntrack_sctp_timeout_closed", + .data = &nf_ct_sctp_timeout_closed, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_COOKIE_WAIT, + .procname = "ip_conntrack_sctp_timeout_cookie_wait", + .data = &nf_ct_sctp_timeout_cookie_wait, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_COOKIE_ECHOED, + .procname = "ip_conntrack_sctp_timeout_cookie_echoed", + .data = &nf_ct_sctp_timeout_cookie_echoed, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_ESTABLISHED, + .procname = "ip_conntrack_sctp_timeout_established", + .data = &nf_ct_sctp_timeout_established, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_SENT, + .procname = "ip_conntrack_sctp_timeout_shutdown_sent", + .data = &nf_ct_sctp_timeout_shutdown_sent, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_RECD, + .procname = "ip_conntrack_sctp_timeout_shutdown_recd", + .data = &nf_ct_sctp_timeout_shutdown_recd, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_ACK_SENT, + .procname = "ip_conntrack_sctp_timeout_shutdown_ack_sent", + .data = &nf_ct_sctp_timeout_shutdown_ack_sent, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = 0 + } +}; +#endif /* CONFIG_NF_CONNTRACK_PROC_COMPAT */ #endif struct nf_conntrack_l4proto nf_conntrack_l4proto_sctp4 = { @@ -590,6 +654,9 @@ struct nf_conntrack_l4proto nf_conntrack_l4proto_sctp4 = { .ctl_table_users = &sctp_sysctl_table_users, .ctl_table_header = &sctp_sysctl_header, .ctl_table = sctp_sysctl_table, +#ifdef CONFIG_NF_CONNTRACK_PROC_COMPAT + .ctl_compat_table = sctp_compat_sysctl_table, +#endif #endif }; diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index 6f6f9a061e7..d99c7c4176d 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -1273,6 +1273,110 @@ static struct ctl_table tcp_sysctl_table[] = { .ctl_name = 0 } }; + +#ifdef CONFIG_NF_CONNTRACK_PROC_COMPAT +static struct ctl_table tcp_compat_sysctl_table[] = { + { + .ctl_name = NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_SYN_SENT, + .procname = "ip_conntrack_tcp_timeout_syn_sent", + .data = &nf_ct_tcp_timeout_syn_sent, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_SYN_RECV, + .procname = "ip_conntrack_tcp_timeout_syn_recv", + .data = &nf_ct_tcp_timeout_syn_recv, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_ESTABLISHED, + .procname = "ip_conntrack_tcp_timeout_established", + .data = &nf_ct_tcp_timeout_established, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_FIN_WAIT, + .procname = "ip_conntrack_tcp_timeout_fin_wait", + .data = &nf_ct_tcp_timeout_fin_wait, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_CLOSE_WAIT, + .procname = "ip_conntrack_tcp_timeout_close_wait", + .data = &nf_ct_tcp_timeout_close_wait, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_LAST_ACK, + .procname = "ip_conntrack_tcp_timeout_last_ack", + .data = &nf_ct_tcp_timeout_last_ack, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_TIME_WAIT, + .procname = "ip_conntrack_tcp_timeout_time_wait", + .data = &nf_ct_tcp_timeout_time_wait, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_CLOSE, + .procname = "ip_conntrack_tcp_timeout_close", + .data = &nf_ct_tcp_timeout_close, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_MAX_RETRANS, + .procname = "ip_conntrack_tcp_timeout_max_retrans", + .data = &nf_ct_tcp_timeout_max_retrans, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_TCP_LOOSE, + .procname = "ip_conntrack_tcp_loose", + .data = &nf_ct_tcp_loose, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_TCP_BE_LIBERAL, + .procname = "ip_conntrack_tcp_be_liberal", + .data = &nf_ct_tcp_be_liberal, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_TCP_MAX_RETRANS, + .procname = "ip_conntrack_tcp_max_retrans", + .data = &nf_ct_tcp_max_retrans, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = 0 + } +}; +#endif /* CONFIG_NF_CONNTRACK_PROC_COMPAT */ #endif /* CONFIG_SYSCTL */ struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp4 = @@ -1298,6 +1402,9 @@ struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp4 = .ctl_table_users = &tcp_sysctl_table_users, .ctl_table_header = &tcp_sysctl_header, .ctl_table = tcp_sysctl_table, +#ifdef CONFIG_NF_CONNTRACK_PROC_COMPAT + .ctl_compat_table = tcp_compat_sysctl_table, +#endif #endif }; diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c index d86749cb4a4..908fde7719f 100644 --- a/net/netfilter/nf_conntrack_proto_udp.c +++ b/net/netfilter/nf_conntrack_proto_udp.c @@ -172,6 +172,29 @@ static struct ctl_table udp_sysctl_table[] = { .ctl_name = 0 } }; +#ifdef CONFIG_NF_CONNTRACK_PROC_COMPAT +static struct ctl_table udp_compat_sysctl_table[] = { + { + .ctl_name = NET_IPV4_NF_CONNTRACK_UDP_TIMEOUT, + .procname = "ip_conntrack_udp_timeout", + .data = &nf_ct_udp_timeout, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_IPV4_NF_CONNTRACK_UDP_TIMEOUT_STREAM, + .procname = "ip_conntrack_udp_timeout_stream", + .data = &nf_ct_udp_timeout_stream, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = 0 + } +}; +#endif /* CONFIG_NF_CONNTRACK_PROC_COMPAT */ #endif /* CONFIG_SYSCTL */ struct nf_conntrack_l4proto nf_conntrack_l4proto_udp4 = @@ -195,6 +218,9 @@ struct nf_conntrack_l4proto nf_conntrack_l4proto_udp4 = .ctl_table_users = &udp_sysctl_table_users, .ctl_table_header = &udp_sysctl_header, .ctl_table = udp_sysctl_table, +#ifdef CONFIG_NF_CONNTRACK_PROC_COMPAT + .ctl_compat_table = udp_compat_sysctl_table, +#endif #endif }; -- cgit v1.2.3-70-g09d2 From e4bd8bce3e8b53e2c0a0d5c9afbc29731e517f8d Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 29 Nov 2006 02:35:22 +0100 Subject: [NETFILTER]: nf_conntrack: /proc compatibility with old connection tracking This patch adds /proc/net/ip_conntrack, /proc/net/ip_conntrack_expect and /proc/net/stat/ip_conntrack files to keep old programs using them working. The /proc/net/ip_conntrack and /proc/net/ip_conntrack_expect files show only IPv4 entries, the /proc/net/stat/ip_conntrack shows global statistics. Signed-off-by: Patrick McHardy --- include/net/netfilter/ipv4/nf_conntrack_ipv4.h | 3 + net/ipv4/netfilter/Makefile | 5 + net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 13 +- .../netfilter/nf_conntrack_l3proto_ipv4_compat.c | 412 +++++++++++++++++++++ net/netfilter/nf_conntrack_l3proto_generic.c | 1 + net/netfilter/nf_conntrack_standalone.c | 1 + 6 files changed, 434 insertions(+), 1 deletion(-) create mode 100644 net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c (limited to 'include') diff --git a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h index 8c4f7e89ce5..34b453a81a6 100644 --- a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h +++ b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h @@ -41,4 +41,7 @@ extern struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp4; extern struct nf_conntrack_l4proto nf_conntrack_l4proto_udp4; extern struct nf_conntrack_l4proto nf_conntrack_l4proto_icmp; +extern int nf_conntrack_ipv4_compat_init(void); +extern void nf_conntrack_ipv4_compat_fini(void); + #endif /*_NF_CONNTRACK_IPV4_H*/ diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index 09aaed1a806..4ce20ebc4d6 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -89,6 +89,11 @@ obj-$(CONFIG_IP_NF_QUEUE) += ip_queue.o # objects for l3 independent conntrack nf_conntrack_ipv4-objs := nf_conntrack_l3proto_ipv4.o nf_conntrack_proto_icmp.o +ifeq ($(CONFIG_NF_CONNTRACK_PROC_COMPAT),y) +ifeq ($(CONFIG_PROC_FS),y) +nf_conntrack_ipv4-objs += nf_conntrack_l3proto_ipv4_compat.o +endif +endif # l3 independent conntrack obj-$(CONFIG_NF_CONNTRACK_IPV4) += nf_conntrack_ipv4.o diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c index bcec6822f2e..f24e872d4b9 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c @@ -492,8 +492,16 @@ static int __init nf_conntrack_l3proto_ipv4_init(void) printk("nf_conntrack_ipv4: can't register hooks.\n"); goto cleanup_ipv4; } +#if defined(CONFIG_PROC_FS) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT) + ret = nf_conntrack_ipv4_compat_init(); + if (ret < 0) + goto cleanup_hooks; +#endif return ret; - +#if defined(CONFIG_PROC_FS) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT) + cleanup_hooks: + nf_unregister_hooks(ipv4_conntrack_ops, ARRAY_SIZE(ipv4_conntrack_ops)); +#endif cleanup_ipv4: nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv4); cleanup_icmp: @@ -510,6 +518,9 @@ static int __init nf_conntrack_l3proto_ipv4_init(void) static void __exit nf_conntrack_l3proto_ipv4_fini(void) { synchronize_net(); +#if defined(CONFIG_PROC_FS) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT) + nf_conntrack_ipv4_compat_fini(); +#endif nf_unregister_hooks(ipv4_conntrack_ops, ARRAY_SIZE(ipv4_conntrack_ops)); nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv4); nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_icmp); diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c new file mode 100644 index 00000000000..3b31bc64960 --- /dev/null +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c @@ -0,0 +1,412 @@ +/* ip_conntrack proc compat - based on ip_conntrack_standalone.c + * + * (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +#ifdef CONFIG_NF_CT_ACCT +static unsigned int +seq_print_counters(struct seq_file *s, + const struct ip_conntrack_counter *counter) +{ + return seq_printf(s, "packets=%llu bytes=%llu ", + (unsigned long long)counter->packets, + (unsigned long long)counter->bytes); +} +#else +#define seq_print_counters(x, y) 0 +#endif + +struct ct_iter_state { + unsigned int bucket; +}; + +static struct list_head *ct_get_first(struct seq_file *seq) +{ + struct ct_iter_state *st = seq->private; + + for (st->bucket = 0; + st->bucket < nf_conntrack_htable_size; + st->bucket++) { + if (!list_empty(&nf_conntrack_hash[st->bucket])) + return nf_conntrack_hash[st->bucket].next; + } + return NULL; +} + +static struct list_head *ct_get_next(struct seq_file *seq, struct list_head *head) +{ + struct ct_iter_state *st = seq->private; + + head = head->next; + while (head == &nf_conntrack_hash[st->bucket]) { + if (++st->bucket >= nf_conntrack_htable_size) + return NULL; + head = nf_conntrack_hash[st->bucket].next; + } + return head; +} + +static struct list_head *ct_get_idx(struct seq_file *seq, loff_t pos) +{ + struct list_head *head = ct_get_first(seq); + + if (head) + while (pos && (head = ct_get_next(seq, head))) + pos--; + return pos ? NULL : head; +} + +static void *ct_seq_start(struct seq_file *seq, loff_t *pos) +{ + read_lock_bh(&nf_conntrack_lock); + return ct_get_idx(seq, *pos); +} + +static void *ct_seq_next(struct seq_file *s, void *v, loff_t *pos) +{ + (*pos)++; + return ct_get_next(s, v); +} + +static void ct_seq_stop(struct seq_file *s, void *v) +{ + read_unlock_bh(&nf_conntrack_lock); +} + +static int ct_seq_show(struct seq_file *s, void *v) +{ + const struct nf_conntrack_tuple_hash *hash = v; + const struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(hash); + struct nf_conntrack_l3proto *l3proto; + struct nf_conntrack_l4proto *l4proto; + + NF_CT_ASSERT(ct); + + /* we only want to print DIR_ORIGINAL */ + if (NF_CT_DIRECTION(hash)) + return 0; + if (ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num != AF_INET) + return 0; + + l3proto = __nf_ct_l3proto_find(ct->tuplehash[IP_CT_DIR_ORIGINAL] + .tuple.src.l3num); + NF_CT_ASSERT(l3proto); + l4proto = __nf_ct_l4proto_find(ct->tuplehash[IP_CT_DIR_ORIGINAL] + .tuple.src.l3num, + ct->tuplehash[IP_CT_DIR_ORIGINAL] + .tuple.dst.protonum); + NF_CT_ASSERT(l4proto); + + if (seq_printf(s, "%-8s %u %ld ", + l4proto->name, + ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum, + timer_pending(&ct->timeout) + ? (long)(ct->timeout.expires - jiffies)/HZ : 0) != 0) + return -ENOSPC; + + if (l3proto->print_conntrack(s, ct)) + return -ENOSPC; + + if (l4proto->print_conntrack(s, ct)) + return -ENOSPC; + + if (print_tuple(s, &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple, + l3proto, l4proto)) + return -ENOSPC; + + if (seq_print_counters(s, &ct->counters[IP_CT_DIR_ORIGINAL])) + return -ENOSPC; + + if (!(test_bit(IPS_SEEN_REPLY_BIT, &ct->status))) + if (seq_printf(s, "[UNREPLIED] ")) + return -ENOSPC; + + if (print_tuple(s, &ct->tuplehash[IP_CT_DIR_REPLY].tuple, + l3proto, l4proto)) + return -ENOSPC; + + if (seq_print_counters(s, &ct->counters[IP_CT_DIR_REPLY])) + return -ENOSPC; + + if (test_bit(IPS_ASSURED_BIT, &ct->status)) + if (seq_printf(s, "[ASSURED] ")) + return -ENOSPC; + +#ifdef CONFIG_NF_CONNTRACK_MARK + if (seq_printf(s, "mark=%u ", ct->mark)) + return -ENOSPC; +#endif + +#ifdef CONFIG_NF_CONNTRACK_SECMARK + if (seq_printf(s, "secmark=%u ", ct->secmark)) + return -ENOSPC; +#endif + + if (seq_printf(s, "use=%u\n", atomic_read(&ct->ct_general.use))) + return -ENOSPC; + + return 0; +} + +static struct seq_operations ct_seq_ops = { + .start = ct_seq_start, + .next = ct_seq_next, + .stop = ct_seq_stop, + .show = ct_seq_show +}; + +static int ct_open(struct inode *inode, struct file *file) +{ + struct seq_file *seq; + struct ct_iter_state *st; + int ret; + + st = kmalloc(sizeof(struct ct_iter_state), GFP_KERNEL); + if (st == NULL) + return -ENOMEM; + ret = seq_open(file, &ct_seq_ops); + if (ret) + goto out_free; + seq = file->private_data; + seq->private = st; + memset(st, 0, sizeof(struct ct_iter_state)); + return ret; +out_free: + kfree(st); + return ret; +} + +static struct file_operations ct_file_ops = { + .owner = THIS_MODULE, + .open = ct_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_private, +}; + +/* expects */ +static void *exp_seq_start(struct seq_file *s, loff_t *pos) +{ + struct list_head *e = &nf_conntrack_expect_list; + loff_t i; + + /* strange seq_file api calls stop even if we fail, + * thus we need to grab lock since stop unlocks */ + read_lock_bh(&nf_conntrack_lock); + + if (list_empty(e)) + return NULL; + + for (i = 0; i <= *pos; i++) { + e = e->next; + if (e == &nf_conntrack_expect_list) + return NULL; + } + return e; +} + +static void *exp_seq_next(struct seq_file *s, void *v, loff_t *pos) +{ + struct list_head *e = v; + + ++*pos; + e = e->next; + + if (e == &nf_conntrack_expect_list) + return NULL; + + return e; +} + +static void exp_seq_stop(struct seq_file *s, void *v) +{ + read_unlock_bh(&nf_conntrack_lock); +} + +static int exp_seq_show(struct seq_file *s, void *v) +{ + struct nf_conntrack_expect *exp = v; + + if (exp->tuple.src.l3num != AF_INET) + return 0; + + if (exp->timeout.function) + seq_printf(s, "%ld ", timer_pending(&exp->timeout) + ? (long)(exp->timeout.expires - jiffies)/HZ : 0); + else + seq_printf(s, "- "); + + seq_printf(s, "proto=%u ", exp->tuple.dst.protonum); + + print_tuple(s, &exp->tuple, + __nf_ct_l3proto_find(exp->tuple.src.l3num), + __nf_ct_l4proto_find(exp->tuple.src.l3num, + exp->tuple.dst.protonum)); + return seq_putc(s, '\n'); +} + +static struct seq_operations exp_seq_ops = { + .start = exp_seq_start, + .next = exp_seq_next, + .stop = exp_seq_stop, + .show = exp_seq_show +}; + +static int exp_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &exp_seq_ops); +} + +static struct file_operations ip_exp_file_ops = { + .owner = THIS_MODULE, + .open = exp_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release +}; + +static void *ct_cpu_seq_start(struct seq_file *seq, loff_t *pos) +{ + int cpu; + + if (*pos == 0) + return SEQ_START_TOKEN; + + for (cpu = *pos-1; cpu < NR_CPUS; ++cpu) { + if (!cpu_possible(cpu)) + continue; + *pos = cpu+1; + return &per_cpu(nf_conntrack_stat, cpu); + } + + return NULL; +} + +static void *ct_cpu_seq_next(struct seq_file *seq, void *v, loff_t *pos) +{ + int cpu; + + for (cpu = *pos; cpu < NR_CPUS; ++cpu) { + if (!cpu_possible(cpu)) + continue; + *pos = cpu+1; + return &per_cpu(nf_conntrack_stat, cpu); + } + + return NULL; +} + +static void ct_cpu_seq_stop(struct seq_file *seq, void *v) +{ +} + +static int ct_cpu_seq_show(struct seq_file *seq, void *v) +{ + unsigned int nr_conntracks = atomic_read(&nf_conntrack_count); + struct ip_conntrack_stat *st = v; + + if (v == SEQ_START_TOKEN) { + seq_printf(seq, "entries searched found new invalid ignore delete delete_list insert insert_failed drop early_drop icmp_error expect_new expect_create expect_delete\n"); + return 0; + } + + seq_printf(seq, "%08x %08x %08x %08x %08x %08x %08x %08x " + "%08x %08x %08x %08x %08x %08x %08x %08x \n", + nr_conntracks, + st->searched, + st->found, + st->new, + st->invalid, + st->ignore, + st->delete, + st->delete_list, + st->insert, + st->insert_failed, + st->drop, + st->early_drop, + st->error, + + st->expect_new, + st->expect_create, + st->expect_delete + ); + return 0; +} + +static struct seq_operations ct_cpu_seq_ops = { + .start = ct_cpu_seq_start, + .next = ct_cpu_seq_next, + .stop = ct_cpu_seq_stop, + .show = ct_cpu_seq_show, +}; + +static int ct_cpu_seq_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &ct_cpu_seq_ops); +} + +static struct file_operations ct_cpu_seq_fops = { + .owner = THIS_MODULE, + .open = ct_cpu_seq_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_private, +}; + +int __init nf_conntrack_ipv4_compat_init(void) +{ + struct proc_dir_entry *proc, *proc_exp, *proc_stat; + + proc = proc_net_fops_create("ip_conntrack", 0440, &ct_file_ops); + if (!proc) + goto err1; + + proc_exp = proc_net_fops_create("ip_conntrack_expect", 0440, + &ip_exp_file_ops); + if (!proc_exp) + goto err2; + + proc_stat = create_proc_entry("ip_conntrack", S_IRUGO, proc_net_stat); + if (!proc_stat) + goto err3; + + proc_stat->proc_fops = &ct_cpu_seq_fops; + proc_stat->owner = THIS_MODULE; + + return 0; + +err3: + proc_net_remove("ip_conntrack_expect"); +err2: + proc_net_remove("ip_conntrack"); +err1: + return -ENOMEM; +} + +void __exit nf_conntrack_ipv4_compat_fini(void) +{ + remove_proc_entry("ip_conntrack", proc_net_stat); + proc_net_remove("ip_conntrack_expect"); + proc_net_remove("ip_conntrack"); +} diff --git a/net/netfilter/nf_conntrack_l3proto_generic.c b/net/netfilter/nf_conntrack_l3proto_generic.c index 3124b3a3010..a3d31c3ac8e 100644 --- a/net/netfilter/nf_conntrack_l3proto_generic.c +++ b/net/netfilter/nf_conntrack_l3proto_generic.c @@ -92,3 +92,4 @@ struct nf_conntrack_l3proto nf_conntrack_l3proto_generic = { .prepare = generic_prepare, .get_features = generic_get_features, }; +EXPORT_SYMBOL_GPL(nf_conntrack_l3proto_generic); diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index 9d8906848b9..3db24f4f0f5 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -52,6 +52,7 @@ print_tuple(struct seq_file *s, const struct nf_conntrack_tuple *tuple, { return l3proto->print_tuple(s, tuple) || l4proto->print_tuple(s, tuple); } +EXPORT_SYMBOL_GPL(print_tuple); #ifdef CONFIG_NF_CT_ACCT static unsigned int -- cgit v1.2.3-70-g09d2 From 468ec44bd5a863736d955f78b8c38896f26864a1 Mon Sep 17 00:00:00 2001 From: Yasuyuki Kozakai Date: Wed, 29 Nov 2006 02:35:23 +0100 Subject: [NETFILTER]: conntrack: add '_get' to {ip, nf}_conntrack_expect_find We usually uses 'xxx_find_get' for function which increments reference count. Signed-off-by: Yasuyuki Kozakai Signed-off-by: Patrick McHardy --- include/linux/netfilter_ipv4/ip_conntrack.h | 2 +- include/net/netfilter/nf_conntrack_expect.h | 2 +- net/ipv4/netfilter/ip_conntrack_core.c | 2 +- net/ipv4/netfilter/ip_conntrack_helper_pptp.c | 4 ++-- net/ipv4/netfilter/ip_conntrack_netlink.c | 4 ++-- net/ipv4/netfilter/ip_conntrack_standalone.c | 2 +- net/ipv4/netfilter/ip_nat_helper_pptp.c | 2 +- net/netfilter/nf_conntrack_expect.c | 2 +- net/netfilter/nf_conntrack_netlink.c | 4 ++-- net/netfilter/nf_conntrack_standalone.c | 2 +- 10 files changed, 13 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter_ipv4/ip_conntrack.h b/include/linux/netfilter_ipv4/ip_conntrack.h index 64e868034c4..61da56941dc 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack.h +++ b/include/linux/netfilter_ipv4/ip_conntrack.h @@ -277,7 +277,7 @@ extern struct ip_conntrack_expect * __ip_conntrack_expect_find(const struct ip_conntrack_tuple *tuple); extern struct ip_conntrack_expect * -ip_conntrack_expect_find(const struct ip_conntrack_tuple *tuple); +ip_conntrack_expect_find_get(const struct ip_conntrack_tuple *tuple); extern struct ip_conntrack_tuple_hash * __ip_conntrack_find(const struct ip_conntrack_tuple *tuple, diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h index 5aa483e0345..2d335f024c8 100644 --- a/include/net/netfilter/nf_conntrack_expect.h +++ b/include/net/netfilter/nf_conntrack_expect.h @@ -53,7 +53,7 @@ struct nf_conntrack_expect * __nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple); struct nf_conntrack_expect * -nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple); +nf_conntrack_expect_find_get(const struct nf_conntrack_tuple *tuple); struct nf_conntrack_expect * find_expectation(const struct nf_conntrack_tuple *tuple); diff --git a/net/ipv4/netfilter/ip_conntrack_core.c b/net/ipv4/netfilter/ip_conntrack_core.c index 8b848aa77bf..4d1f954d459 100644 --- a/net/ipv4/netfilter/ip_conntrack_core.c +++ b/net/ipv4/netfilter/ip_conntrack_core.c @@ -233,7 +233,7 @@ __ip_conntrack_expect_find(const struct ip_conntrack_tuple *tuple) /* Just find a expectation corresponding to a tuple. */ struct ip_conntrack_expect * -ip_conntrack_expect_find(const struct ip_conntrack_tuple *tuple) +ip_conntrack_expect_find_get(const struct ip_conntrack_tuple *tuple) { struct ip_conntrack_expect *i; diff --git a/net/ipv4/netfilter/ip_conntrack_helper_pptp.c b/net/ipv4/netfilter/ip_conntrack_helper_pptp.c index a2af5e0c7f9..a5c057bcecf 100644 --- a/net/ipv4/netfilter/ip_conntrack_helper_pptp.c +++ b/net/ipv4/netfilter/ip_conntrack_helper_pptp.c @@ -142,7 +142,7 @@ static void pptp_expectfn(struct ip_conntrack *ct, DEBUGP("trying to unexpect other dir: "); DUMP_TUPLE(&inv_t); - exp_other = ip_conntrack_expect_find(&inv_t); + exp_other = ip_conntrack_expect_find_get(&inv_t); if (exp_other) { /* delete other expectation. */ DEBUGP("found\n"); @@ -176,7 +176,7 @@ static int destroy_sibling_or_exp(const struct ip_conntrack_tuple *t) ip_conntrack_put(sibling); return 1; } else { - exp = ip_conntrack_expect_find(t); + exp = ip_conntrack_expect_find_get(t); if (exp) { DEBUGP("unexpect_related of expect %p\n", exp); ip_conntrack_unexpect_related(exp); diff --git a/net/ipv4/netfilter/ip_conntrack_netlink.c b/net/ipv4/netfilter/ip_conntrack_netlink.c index 1bb8ed33c5b..3d277aa869d 100644 --- a/net/ipv4/netfilter/ip_conntrack_netlink.c +++ b/net/ipv4/netfilter/ip_conntrack_netlink.c @@ -1256,7 +1256,7 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, if (err < 0) return err; - exp = ip_conntrack_expect_find(&tuple); + exp = ip_conntrack_expect_find_get(&tuple); if (!exp) return -ENOENT; @@ -1309,7 +1309,7 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, return err; /* bump usage count to 2 */ - exp = ip_conntrack_expect_find(&tuple); + exp = ip_conntrack_expect_find_get(&tuple); if (!exp) return -ENOENT; diff --git a/net/ipv4/netfilter/ip_conntrack_standalone.c b/net/ipv4/netfilter/ip_conntrack_standalone.c index 02135756562..2df67538ffb 100644 --- a/net/ipv4/netfilter/ip_conntrack_standalone.c +++ b/net/ipv4/netfilter/ip_conntrack_standalone.c @@ -926,7 +926,7 @@ EXPORT_SYMBOL(__ip_ct_refresh_acct); EXPORT_SYMBOL(ip_conntrack_expect_alloc); EXPORT_SYMBOL(ip_conntrack_expect_put); EXPORT_SYMBOL_GPL(__ip_conntrack_expect_find); -EXPORT_SYMBOL_GPL(ip_conntrack_expect_find); +EXPORT_SYMBOL_GPL(ip_conntrack_expect_find_get); EXPORT_SYMBOL(ip_conntrack_expect_related); EXPORT_SYMBOL(ip_conntrack_unexpect_related); EXPORT_SYMBOL_GPL(ip_conntrack_expect_list); diff --git a/net/ipv4/netfilter/ip_nat_helper_pptp.c b/net/ipv4/netfilter/ip_nat_helper_pptp.c index 329fdcd7d70..acf55d86310 100644 --- a/net/ipv4/netfilter/ip_nat_helper_pptp.c +++ b/net/ipv4/netfilter/ip_nat_helper_pptp.c @@ -101,7 +101,7 @@ static void pptp_nat_expected(struct ip_conntrack *ct, DEBUGP("trying to unexpect other dir: "); DUMP_TUPLE(&t); - other_exp = ip_conntrack_expect_find(&t); + other_exp = ip_conntrack_expect_find_get(&t); if (other_exp) { ip_conntrack_unexpect_related(other_exp); ip_conntrack_expect_put(other_exp); diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index 79cfd79a42f..aa5903e4da1 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -68,7 +68,7 @@ __nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple) /* Just find a expectation corresponding to a tuple. */ struct nf_conntrack_expect * -nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple) +nf_conntrack_expect_find_get(const struct nf_conntrack_tuple *tuple) { struct nf_conntrack_expect *i; diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index dc083022013..7357b8f47ac 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -1284,7 +1284,7 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, if (err < 0) return err; - exp = nf_conntrack_expect_find(&tuple); + exp = nf_conntrack_expect_find_get(&tuple); if (!exp) return -ENOENT; @@ -1339,7 +1339,7 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, return err; /* bump usage count to 2 */ - exp = nf_conntrack_expect_find(&tuple); + exp = nf_conntrack_expect_find_get(&tuple); if (!exp) return -ENOENT; diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index 3db24f4f0f5..be94b635972 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -549,7 +549,7 @@ EXPORT_SYMBOL(__nf_conntrack_find); EXPORT_SYMBOL(nf_ct_unlink_expect); EXPORT_SYMBOL(nf_conntrack_hash_insert); EXPORT_SYMBOL(__nf_conntrack_expect_find); -EXPORT_SYMBOL(nf_conntrack_expect_find); +EXPORT_SYMBOL(nf_conntrack_expect_find_get); EXPORT_SYMBOL(nf_conntrack_expect_list); #if defined(CONFIG_NF_CT_NETLINK) || \ defined(CONFIG_NF_CT_NETLINK_MODULE) -- cgit v1.2.3-70-g09d2 From 9d5b8baa4e9ace9be113c6151aaeeb3c07a26fc8 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 29 Nov 2006 02:35:26 +0100 Subject: [NETFILTER]: sip conntrack: minor cleanup - Use enum for header field enumeration - Use numerical value instead of pointer to header info structure to identify headers, unexport ct_sip_hdrs - group SIP and SDP entries in header info structure - remove double forward declaration of ct_sip_get_info Signed-off-by: Patrick McHardy --- include/linux/netfilter_ipv4/ip_conntrack_sip.h | 28 +++++------- net/ipv4/netfilter/ip_conntrack_sip.c | 60 +++++++++++++------------ net/ipv4/netfilter/ip_nat_sip.c | 30 ++++++------- 3 files changed, 55 insertions(+), 63 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter_ipv4/ip_conntrack_sip.h b/include/linux/netfilter_ipv4/ip_conntrack_sip.h index 913dad66c0f..2a15eb51fd6 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_sip.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_sip.h @@ -5,23 +5,15 @@ #define SIP_PORT 5060 #define SIP_TIMEOUT 3600 -#define POS_VIA 0 -#define POS_CONTACT 1 -#define POS_CONTENT 2 -#define POS_MEDIA 3 -#define POS_OWNER 4 -#define POS_CONNECTION 5 -#define POS_REQ_HEADER 6 -#define POS_SDP_HEADER 7 - -struct sip_header_nfo { - const char *lname; - const char *sname; - const char *ln_str; - size_t lnlen; - size_t snlen; - size_t ln_strlen; - int (*match_len)(const char *, const char *, int *); +enum sip_header_pos { + POS_REQ_HEADER, + POS_VIA, + POS_CONTACT, + POS_CONTENT, + POS_MEDIA, + POS_OWNER, + POS_CONNECTION, + POS_SDP_HEADER, }; extern unsigned int (*ip_nat_sip_hook)(struct sk_buff **pskb, @@ -36,7 +28,7 @@ extern unsigned int (*ip_nat_sdp_hook)(struct sk_buff **pskb, extern int ct_sip_get_info(const char *dptr, size_t dlen, unsigned int *matchoff, unsigned int *matchlen, - struct sip_header_nfo *hnfo); + enum sip_header_pos pos); extern int ct_sip_lnlen(const char *line, const char *limit); extern const char *ct_sip_search(const char *needle, const char *haystack, size_t needle_len, size_t haystack_len); diff --git a/net/ipv4/netfilter/ip_conntrack_sip.c b/net/ipv4/netfilter/ip_conntrack_sip.c index bf423db4ed4..cc317658166 100644 --- a/net/ipv4/netfilter/ip_conntrack_sip.c +++ b/net/ipv4/netfilter/ip_conntrack_sip.c @@ -52,20 +52,32 @@ unsigned int (*ip_nat_sdp_hook)(struct sk_buff **pskb, const char *dptr); EXPORT_SYMBOL_GPL(ip_nat_sdp_hook); -int ct_sip_get_info(const char *dptr, size_t dlen, - unsigned int *matchoff, - unsigned int *matchlen, - struct sip_header_nfo *hnfo); -EXPORT_SYMBOL_GPL(ct_sip_get_info); - - static int digits_len(const char *dptr, const char *limit, int *shift); static int epaddr_len(const char *dptr, const char *limit, int *shift); static int skp_digits_len(const char *dptr, const char *limit, int *shift); static int skp_epaddr_len(const char *dptr, const char *limit, int *shift); -struct sip_header_nfo ct_sip_hdrs[] = { - { /* Via header */ +struct sip_header_nfo { + const char *lname; + const char *sname; + const char *ln_str; + size_t lnlen; + size_t snlen; + size_t ln_strlen; + int (*match_len)(const char *, const char *, int *); +}; + +static struct sip_header_nfo ct_sip_hdrs[] = { + [POS_REQ_HEADER] = { /* SIP Requests headers */ + .lname = "sip:", + .lnlen = sizeof("sip:") - 1, + .sname = "sip:", + .snlen = sizeof("sip:") - 1, /* yes, i know.. ;) */ + .ln_str = "@", + .ln_strlen = sizeof("@") - 1, + .match_len = epaddr_len + }, + [POS_VIA] = { /* SIP Via header */ .lname = "Via:", .lnlen = sizeof("Via:") - 1, .sname = "\r\nv:", @@ -74,7 +86,7 @@ struct sip_header_nfo ct_sip_hdrs[] = { .ln_strlen = sizeof("UDP ") - 1, .match_len = epaddr_len, }, - { /* Contact header */ + [POS_CONTACT] = { /* SIP Contact header */ .lname = "Contact:", .lnlen = sizeof("Contact:") - 1, .sname = "\r\nm:", @@ -83,7 +95,7 @@ struct sip_header_nfo ct_sip_hdrs[] = { .ln_strlen = sizeof("sip:") - 1, .match_len = skp_epaddr_len }, - { /* Content length header */ + [POS_CONTENT] = { /* SIP Content length header */ .lname = "Content-Length:", .lnlen = sizeof("Content-Length:") - 1, .sname = "\r\nl:", @@ -92,7 +104,7 @@ struct sip_header_nfo ct_sip_hdrs[] = { .ln_strlen = sizeof(":") - 1, .match_len = skp_digits_len }, - { /* SDP media info */ + [POS_MEDIA] = { /* SDP media info */ .lname = "\nm=", .lnlen = sizeof("\nm=") - 1, .sname = "\rm=", @@ -101,7 +113,7 @@ struct sip_header_nfo ct_sip_hdrs[] = { .ln_strlen = sizeof("audio ") - 1, .match_len = digits_len }, - { /* SDP owner address*/ + [POS_OWNER] = { /* SDP owner address*/ .lname = "\no=", .lnlen = sizeof("\no=") - 1, .sname = "\ro=", @@ -110,7 +122,7 @@ struct sip_header_nfo ct_sip_hdrs[] = { .ln_strlen = sizeof("IN IP4 ") - 1, .match_len = epaddr_len }, - { /* SDP connection info */ + [POS_CONNECTION] = { /* SDP connection info */ .lname = "\nc=", .lnlen = sizeof("\nc=") - 1, .sname = "\rc=", @@ -119,16 +131,7 @@ struct sip_header_nfo ct_sip_hdrs[] = { .ln_strlen = sizeof("IN IP4 ") - 1, .match_len = epaddr_len }, - { /* Requests headers */ - .lname = "sip:", - .lnlen = sizeof("sip:") - 1, - .sname = "sip:", - .snlen = sizeof("sip:") - 1, /* yes, i know.. ;) */ - .ln_str = "@", - .ln_strlen = sizeof("@") - 1, - .match_len = epaddr_len - }, - { /* SDP version header */ + [POS_SDP_HEADER] = { /* SDP version header */ .lname = "\nv=", .lnlen = sizeof("\nv=") - 1, .sname = "\rv=", @@ -138,7 +141,6 @@ struct sip_header_nfo ct_sip_hdrs[] = { .match_len = digits_len } }; -EXPORT_SYMBOL_GPL(ct_sip_hdrs); /* get line lenght until first CR or LF seen. */ int ct_sip_lnlen(const char *line, const char *limit) @@ -263,8 +265,9 @@ static int skp_epaddr_len(const char *dptr, const char *limit, int *shift) int ct_sip_get_info(const char *dptr, size_t dlen, unsigned int *matchoff, unsigned int *matchlen, - struct sip_header_nfo *hnfo) + enum sip_header_pos pos) { + struct sip_header_nfo *hnfo = &ct_sip_hdrs[pos]; const char *limit, *aux, *k = dptr; int shift = 0; @@ -298,6 +301,7 @@ int ct_sip_get_info(const char *dptr, size_t dlen, DEBUGP("%s header not found.\n", hnfo->lname); return 0; } +EXPORT_SYMBOL_GPL(ct_sip_get_info); static int set_expected_rtp(struct sk_buff **pskb, struct ip_conntrack *ct, @@ -393,7 +397,7 @@ static int sip_help(struct sk_buff **pskb, } /* Get ip and port address from SDP packet. */ if (ct_sip_get_info(dptr, datalen, &matchoff, &matchlen, - &ct_sip_hdrs[POS_CONNECTION]) > 0) { + POS_CONNECTION) > 0) { /* We'll drop only if there are parse problems. */ if (parse_ipaddr(dptr + matchoff, NULL, &ipaddr, @@ -402,7 +406,7 @@ static int sip_help(struct sk_buff **pskb, goto out; } if (ct_sip_get_info(dptr, datalen, &matchoff, &matchlen, - &ct_sip_hdrs[POS_MEDIA]) > 0) { + POS_MEDIA) > 0) { port = simple_strtoul(dptr + matchoff, NULL, 10); if (port < 1024) { diff --git a/net/ipv4/netfilter/ip_nat_sip.c b/net/ipv4/netfilter/ip_nat_sip.c index 9fad9809953..47097aac63d 100644 --- a/net/ipv4/netfilter/ip_nat_sip.c +++ b/net/ipv4/netfilter/ip_nat_sip.c @@ -29,18 +29,16 @@ MODULE_DESCRIPTION("SIP NAT helper"); #define DEBUGP(format, args...) #endif -extern struct sip_header_nfo ct_sip_hdrs[]; - static unsigned int mangle_sip_packet(struct sk_buff **pskb, enum ip_conntrack_info ctinfo, struct ip_conntrack *ct, const char **dptr, size_t dlen, char *buffer, int bufflen, - struct sip_header_nfo *hnfo) + enum sip_header_pos pos) { unsigned int matchlen, matchoff; - if (ct_sip_get_info(*dptr, dlen, &matchoff, &matchlen, hnfo) <= 0) + if (ct_sip_get_info(*dptr, dlen, &matchoff, &matchlen, pos) <= 0) return 0; if (!ip_nat_mangle_udp_packet(pskb, ct, ctinfo, @@ -80,14 +78,13 @@ static unsigned int ip_nat_sip(struct sk_buff **pskb, if ((ctinfo) < IP_CT_IS_REPLY) { mangle_sip_packet(pskb, ctinfo, ct, dptr, (*pskb)->len - dataoff, - buffer, bufflen, - &ct_sip_hdrs[POS_CONTACT]); + buffer, bufflen, POS_CONTACT); return 1; } if (!mangle_sip_packet(pskb, ctinfo, ct, dptr, (*pskb)->len - dataoff, - buffer, bufflen, &ct_sip_hdrs[POS_VIA])) + buffer, bufflen, POS_VIA)) return 0; /* This search should ignore case, but later.. */ @@ -102,25 +99,24 @@ static unsigned int ip_nat_sip(struct sk_buff **pskb, return mangle_sip_packet(pskb, ctinfo, ct, dptr, (*pskb)->len - dataoff, - buffer, bufflen, - &ct_sip_hdrs[POS_CONTACT]); + buffer, bufflen, POS_CONTACT); } if ((ctinfo) < IP_CT_IS_REPLY) { if (!mangle_sip_packet(pskb, ctinfo, ct, dptr, (*pskb)->len - dataoff, - buffer, bufflen, &ct_sip_hdrs[POS_VIA])) + buffer, bufflen, POS_VIA)) return 0; /* Mangle Contact if exists only. - watch udp_nat_mangle()! */ mangle_sip_packet(pskb, ctinfo, ct, dptr, (*pskb)->len - dataoff, - buffer, bufflen, &ct_sip_hdrs[POS_CONTACT]); + buffer, bufflen, POS_CONTACT); return 1; } /* This mangle requests headers. */ return mangle_sip_packet(pskb, ctinfo, ct, dptr, ct_sip_lnlen(*dptr, *dptr + (*pskb)->len - dataoff), - buffer, bufflen, &ct_sip_hdrs[POS_REQ_HEADER]); + buffer, bufflen, POS_REQ_HEADER); } static int mangle_content_len(struct sk_buff **pskb, @@ -136,7 +132,7 @@ static int mangle_content_len(struct sk_buff **pskb, /* Get actual SDP lenght */ if (ct_sip_get_info(dptr, (*pskb)->len - dataoff, &matchoff, - &matchlen, &ct_sip_hdrs[POS_SDP_HEADER]) > 0) { + &matchlen, POS_SDP_HEADER) > 0) { /* since ct_sip_get_info() give us a pointer passing 'v=' we need to add 2 bytes in this count. */ @@ -144,7 +140,7 @@ static int mangle_content_len(struct sk_buff **pskb, /* Now, update SDP lenght */ if (ct_sip_get_info(dptr, (*pskb)->len - dataoff, &matchoff, - &matchlen, &ct_sip_hdrs[POS_CONTENT]) > 0) { + &matchlen, POS_CONTENT) > 0) { bufflen = sprintf(buffer, "%u", c_len); @@ -170,17 +166,17 @@ static unsigned int mangle_sdp(struct sk_buff **pskb, /* Mangle owner and contact info. */ bufflen = sprintf(buffer, "%u.%u.%u.%u", NIPQUAD(newip)); if (!mangle_sip_packet(pskb, ctinfo, ct, &dptr, (*pskb)->len - dataoff, - buffer, bufflen, &ct_sip_hdrs[POS_OWNER])) + buffer, bufflen, POS_OWNER)) return 0; if (!mangle_sip_packet(pskb, ctinfo, ct, &dptr, (*pskb)->len - dataoff, - buffer, bufflen, &ct_sip_hdrs[POS_CONNECTION])) + buffer, bufflen, POS_CONNECTION)) return 0; /* Mangle media port. */ bufflen = sprintf(buffer, "%u", port); if (!mangle_sip_packet(pskb, ctinfo, ct, &dptr, (*pskb)->len - dataoff, - buffer, bufflen, &ct_sip_hdrs[POS_MEDIA])) + buffer, bufflen, POS_MEDIA)) return 0; return mangle_content_len(pskb, ctinfo, ct, dptr); -- cgit v1.2.3-70-g09d2 From 40883e8184947879f135605a05c0764c60656cc5 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 29 Nov 2006 02:35:27 +0100 Subject: [NETFILTER]: sip conntrack: do case insensitive SIP header search SIP headers are generally case-insensitive, only SDP headers are case sensitive. Signed-off-by: Patrick McHardy --- include/linux/netfilter_ipv4/ip_conntrack_sip.h | 3 ++- net/ipv4/netfilter/ip_conntrack_sip.c | 20 ++++++++++++++++---- net/ipv4/netfilter/ip_nat_sip.c | 7 ++++--- 3 files changed, 22 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter_ipv4/ip_conntrack_sip.h b/include/linux/netfilter_ipv4/ip_conntrack_sip.h index 2a15eb51fd6..51c65ac18c5 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_sip.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_sip.h @@ -31,6 +31,7 @@ extern int ct_sip_get_info(const char *dptr, size_t dlen, enum sip_header_pos pos); extern int ct_sip_lnlen(const char *line, const char *limit); extern const char *ct_sip_search(const char *needle, const char *haystack, - size_t needle_len, size_t haystack_len); + size_t needle_len, size_t haystack_len, + int case_sensitive); #endif /* __KERNEL__ */ #endif /* __IP_CONNTRACK_SIP_H__ */ diff --git a/net/ipv4/netfilter/ip_conntrack_sip.c b/net/ipv4/netfilter/ip_conntrack_sip.c index cc317658166..a9c0d1d3e56 100644 --- a/net/ipv4/netfilter/ip_conntrack_sip.c +++ b/net/ipv4/netfilter/ip_conntrack_sip.c @@ -64,6 +64,7 @@ struct sip_header_nfo { size_t lnlen; size_t snlen; size_t ln_strlen; + int case_sensitive; int (*match_len)(const char *, const char *, int *); }; @@ -105,6 +106,7 @@ static struct sip_header_nfo ct_sip_hdrs[] = { .match_len = skp_digits_len }, [POS_MEDIA] = { /* SDP media info */ + .case_sensitive = 1, .lname = "\nm=", .lnlen = sizeof("\nm=") - 1, .sname = "\rm=", @@ -114,6 +116,7 @@ static struct sip_header_nfo ct_sip_hdrs[] = { .match_len = digits_len }, [POS_OWNER] = { /* SDP owner address*/ + .case_sensitive = 1, .lname = "\no=", .lnlen = sizeof("\no=") - 1, .sname = "\ro=", @@ -123,6 +126,7 @@ static struct sip_header_nfo ct_sip_hdrs[] = { .match_len = epaddr_len }, [POS_CONNECTION] = { /* SDP connection info */ + .case_sensitive = 1, .lname = "\nc=", .lnlen = sizeof("\nc=") - 1, .sname = "\rc=", @@ -132,6 +136,7 @@ static struct sip_header_nfo ct_sip_hdrs[] = { .match_len = epaddr_len }, [POS_SDP_HEADER] = { /* SDP version header */ + .case_sensitive = 1, .lname = "\nv=", .lnlen = sizeof("\nv=") - 1, .sname = "\rv=", @@ -161,13 +166,19 @@ EXPORT_SYMBOL_GPL(ct_sip_lnlen); /* Linear string search, case sensitive. */ const char *ct_sip_search(const char *needle, const char *haystack, - size_t needle_len, size_t haystack_len) + size_t needle_len, size_t haystack_len, + int case_sensitive) { const char *limit = haystack + (haystack_len - needle_len); while (haystack <= limit) { - if (memcmp(haystack, needle, needle_len) == 0) - return haystack; + if (case_sensitive) { + if (strncmp(haystack, needle, needle_len) == 0) + return haystack; + } else { + if (strnicmp(haystack, needle, needle_len) == 0) + return haystack; + } haystack++; } return NULL; @@ -280,7 +291,8 @@ int ct_sip_get_info(const char *dptr, size_t dlen, continue; } aux = ct_sip_search(hnfo->ln_str, dptr, hnfo->ln_strlen, - ct_sip_lnlen(dptr, limit)); + ct_sip_lnlen(dptr, limit), + hnfo->case_sensitive); if (!aux) { DEBUGP("'%s' not found in '%s'.\n", hnfo->ln_str, hnfo->lname); diff --git a/net/ipv4/netfilter/ip_nat_sip.c b/net/ipv4/netfilter/ip_nat_sip.c index 47097aac63d..e16604c4339 100644 --- a/net/ipv4/netfilter/ip_nat_sip.c +++ b/net/ipv4/netfilter/ip_nat_sip.c @@ -87,14 +87,15 @@ static unsigned int ip_nat_sip(struct sk_buff **pskb, buffer, bufflen, POS_VIA)) return 0; - /* This search should ignore case, but later.. */ aux = ct_sip_search("CSeq:", *dptr, sizeof("CSeq:") - 1, - (*pskb)->len - dataoff); + (*pskb)->len - dataoff, 0); if (!aux) return 0; if (!ct_sip_search("REGISTER", aux, sizeof("REGISTER"), - ct_sip_lnlen(aux, *dptr + (*pskb)->len - dataoff))) + ct_sip_lnlen(aux, + *dptr + (*pskb)->len - dataoff), + 1)) return 1; return mangle_sip_packet(pskb, ctinfo, ct, dptr, -- cgit v1.2.3-70-g09d2 From 1b683b551209ca46ae59b29572018001db5af078 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 29 Nov 2006 02:35:30 +0100 Subject: [NETFILTER]: sip conntrack: better NAT handling The NAT handling of the SIP helper has a few problems: - Request headers are only mangled in the reply direction, From/To headers not at all, which can lead to authentication failures with DNAT in case the authentication domain is the IP address - Contact headers in responses are only mangled for REGISTER responses - Headers may be mangled even though they contain addresses not participating in the connection, like alternative addresses - Packets are droppen when domain names are used where the helper expects IP addresses This patch takes a different approach, instead of fixed rules what field to mangle to what content, it adds symetric mapping of From/To/Via/Contact headers, which allows to deal properly with echoed addresses in responses and foreign addresses not belonging to the connection. Signed-off-by: Patrick McHardy --- include/linux/netfilter_ipv4/ip_conntrack_sip.h | 5 +- net/ipv4/netfilter/ip_conntrack_sip.c | 29 +++- net/ipv4/netfilter/ip_nat_sip.c | 171 ++++++++++++++---------- 3 files changed, 135 insertions(+), 70 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter_ipv4/ip_conntrack_sip.h b/include/linux/netfilter_ipv4/ip_conntrack_sip.h index 51c65ac18c5..bef6c646def 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_sip.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_sip.h @@ -6,7 +6,10 @@ #define SIP_TIMEOUT 3600 enum sip_header_pos { - POS_REQ_HEADER, + POS_REG_REQ_URI, + POS_REQ_URI, + POS_FROM, + POS_TO, POS_VIA, POS_CONTACT, POS_CONTENT, diff --git a/net/ipv4/netfilter/ip_conntrack_sip.c b/net/ipv4/netfilter/ip_conntrack_sip.c index 0a6a13c45b0..3a26d63eed8 100644 --- a/net/ipv4/netfilter/ip_conntrack_sip.c +++ b/net/ipv4/netfilter/ip_conntrack_sip.c @@ -69,13 +69,38 @@ struct sip_header_nfo { }; static struct sip_header_nfo ct_sip_hdrs[] = { - [POS_REQ_HEADER] = { /* SIP Requests headers */ + [POS_REG_REQ_URI] = { /* SIP REGISTER request URI */ + .lname = "sip:", + .lnlen = sizeof("sip:") - 1, + .ln_str = ":", + .ln_strlen = sizeof(":") - 1, + .match_len = epaddr_len + }, + [POS_REQ_URI] = { /* SIP request URI */ .lname = "sip:", .lnlen = sizeof("sip:") - 1, .ln_str = "@", .ln_strlen = sizeof("@") - 1, .match_len = epaddr_len }, + [POS_FROM] = { /* SIP From header */ + .lname = "From:", + .lnlen = sizeof("From:") - 1, + .sname = "\r\nf:", + .snlen = sizeof("\r\nf:") - 1, + .ln_str = "sip:", + .ln_strlen = sizeof("sip:") - 1, + .match_len = skp_epaddr_len, + }, + [POS_TO] = { /* SIP To header */ + .lname = "To:", + .lnlen = sizeof("To:") - 1, + .sname = "\r\nt:", + .snlen = sizeof("\r\nt:") - 1, + .ln_str = "sip:", + .ln_strlen = sizeof("sip:") - 1, + .match_len = skp_epaddr_len, + }, [POS_VIA] = { /* SIP Via header */ .lname = "Via:", .lnlen = sizeof("Via:") - 1, @@ -284,7 +309,7 @@ int ct_sip_get_info(const char *dptr, size_t dlen, while (dptr <= limit) { if ((strncmp(dptr, hnfo->lname, hnfo->lnlen) != 0) && - (hinfo->sname == NULL || + (hnfo->sname == NULL || strncmp(dptr, hnfo->sname, hnfo->snlen) != 0)) { dptr++; continue; diff --git a/net/ipv4/netfilter/ip_nat_sip.c b/net/ipv4/netfilter/ip_nat_sip.c index e16604c4339..6223abc924f 100644 --- a/net/ipv4/netfilter/ip_nat_sip.c +++ b/net/ipv4/netfilter/ip_nat_sip.c @@ -29,25 +29,70 @@ MODULE_DESCRIPTION("SIP NAT helper"); #define DEBUGP(format, args...) #endif -static unsigned int mangle_sip_packet(struct sk_buff **pskb, - enum ip_conntrack_info ctinfo, - struct ip_conntrack *ct, - const char **dptr, size_t dlen, - char *buffer, int bufflen, - enum sip_header_pos pos) +struct addr_map { + struct { + char src[sizeof("nnn.nnn.nnn.nnn:nnnnn")]; + char dst[sizeof("nnn.nnn.nnn.nnn:nnnnn")]; + unsigned int srclen, srciplen; + unsigned int dstlen, dstiplen; + } addr[IP_CT_DIR_MAX]; +}; + +static void addr_map_init(struct ip_conntrack *ct, struct addr_map *map) { - unsigned int matchlen, matchoff; + struct ip_conntrack_tuple *t; + enum ip_conntrack_dir dir; + unsigned int n; + + for (dir = 0; dir < IP_CT_DIR_MAX; dir++) { + t = &ct->tuplehash[dir].tuple; + + n = sprintf(map->addr[dir].src, "%u.%u.%u.%u", + NIPQUAD(t->src.ip)); + map->addr[dir].srciplen = n; + n += sprintf(map->addr[dir].src + n, ":%u", + ntohs(t->src.u.udp.port)); + map->addr[dir].srclen = n; + + n = sprintf(map->addr[dir].dst, "%u.%u.%u.%u", + NIPQUAD(t->dst.ip)); + map->addr[dir].dstiplen = n; + n += sprintf(map->addr[dir].dst + n, ":%u", + ntohs(t->dst.u.udp.port)); + map->addr[dir].dstlen = n; + } +} + +static int map_sip_addr(struct sk_buff **pskb, enum ip_conntrack_info ctinfo, + struct ip_conntrack *ct, const char **dptr, size_t dlen, + enum sip_header_pos pos, struct addr_map *map) +{ + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + unsigned int matchlen, matchoff, addrlen; + char *addr; if (ct_sip_get_info(*dptr, dlen, &matchoff, &matchlen, pos) <= 0) - return 0; + return 1; + + if ((matchlen == map->addr[dir].srciplen || + matchlen == map->addr[dir].srclen) && + memcmp(*dptr + matchoff, map->addr[dir].src, matchlen) == 0) { + addr = map->addr[!dir].dst; + addrlen = map->addr[!dir].dstlen; + } else if ((matchlen == map->addr[dir].dstiplen || + matchlen == map->addr[dir].dstlen) && + memcmp(*dptr + matchoff, map->addr[dir].dst, matchlen) == 0) { + addr = map->addr[!dir].src; + addrlen = map->addr[!dir].srclen; + } else + return 1; if (!ip_nat_mangle_udp_packet(pskb, ct, ctinfo, - matchoff, matchlen, buffer, bufflen)) + matchoff, matchlen, addr, addrlen)) return 0; - - /* We need to reload this. Thanks Patrick. */ *dptr = (*pskb)->data + (*pskb)->nh.iph->ihl*4 + sizeof(struct udphdr); return 1; + } static unsigned int ip_nat_sip(struct sk_buff **pskb, @@ -55,69 +100,61 @@ static unsigned int ip_nat_sip(struct sk_buff **pskb, struct ip_conntrack *ct, const char **dptr) { - enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); - char buffer[sizeof("nnn.nnn.nnn.nnn:nnnnn")]; - unsigned int bufflen, dataoff; - __be32 ip; - __be16 port; + enum sip_header_pos pos; + struct addr_map map; + int dataoff, datalen; dataoff = (*pskb)->nh.iph->ihl*4 + sizeof(struct udphdr); + datalen = (*pskb)->len - dataoff; + if (datalen < sizeof("SIP/2.0") - 1) + return NF_DROP; + + addr_map_init(ct, &map); + + /* Basic rules: requests and responses. */ + if (strncmp(*dptr, "SIP/2.0", sizeof("SIP/2.0") - 1) != 0) { + /* 10.2: Constructing the REGISTER Request: + * + * The "userinfo" and "@" components of the SIP URI MUST NOT + * be present. + */ + if (datalen >= sizeof("REGISTER") - 1 && + strncmp(*dptr, "REGISTER", sizeof("REGISTER") - 1) == 0) + pos = POS_REG_REQ_URI; + else + pos = POS_REQ_URI; + + if (!map_sip_addr(pskb, ctinfo, ct, dptr, datalen, pos, &map)) + return NF_DROP; + } - ip = ct->tuplehash[!dir].tuple.dst.ip; - port = ct->tuplehash[!dir].tuple.dst.u.udp.port; - bufflen = sprintf(buffer, "%u.%u.%u.%u:%u", NIPQUAD(ip), ntohs(port)); + if (!map_sip_addr(pskb, ctinfo, ct, dptr, datalen, POS_FROM, &map) || + !map_sip_addr(pskb, ctinfo, ct, dptr, datalen, POS_TO, &map) || + !map_sip_addr(pskb, ctinfo, ct, dptr, datalen, POS_VIA, &map) || + !map_sip_addr(pskb, ctinfo, ct, dptr, datalen, POS_CONTACT, &map)) + return NF_DROP; + return NF_ACCEPT; +} + +static unsigned int mangle_sip_packet(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct ip_conntrack *ct, + const char **dptr, size_t dlen, + char *buffer, int bufflen, + enum sip_header_pos pos) +{ + unsigned int matchlen, matchoff; - /* short packet ? */ - if (((*pskb)->len - dataoff) < (sizeof("SIP/2.0") - 1)) + if (ct_sip_get_info(*dptr, dlen, &matchoff, &matchlen, pos) <= 0) return 0; - /* Basic rules: requests and responses. */ - if (memcmp(*dptr, "SIP/2.0", sizeof("SIP/2.0") - 1) == 0) { - const char *aux; - - if ((ctinfo) < IP_CT_IS_REPLY) { - mangle_sip_packet(pskb, ctinfo, ct, dptr, - (*pskb)->len - dataoff, - buffer, bufflen, POS_CONTACT); - return 1; - } + if (!ip_nat_mangle_udp_packet(pskb, ct, ctinfo, + matchoff, matchlen, buffer, bufflen)) + return 0; - if (!mangle_sip_packet(pskb, ctinfo, ct, dptr, - (*pskb)->len - dataoff, - buffer, bufflen, POS_VIA)) - return 0; - - aux = ct_sip_search("CSeq:", *dptr, sizeof("CSeq:") - 1, - (*pskb)->len - dataoff, 0); - if (!aux) - return 0; - - if (!ct_sip_search("REGISTER", aux, sizeof("REGISTER"), - ct_sip_lnlen(aux, - *dptr + (*pskb)->len - dataoff), - 1)) - return 1; - - return mangle_sip_packet(pskb, ctinfo, ct, dptr, - (*pskb)->len - dataoff, - buffer, bufflen, POS_CONTACT); - } - if ((ctinfo) < IP_CT_IS_REPLY) { - if (!mangle_sip_packet(pskb, ctinfo, ct, dptr, - (*pskb)->len - dataoff, - buffer, bufflen, POS_VIA)) - return 0; - - /* Mangle Contact if exists only. - watch udp_nat_mangle()! */ - mangle_sip_packet(pskb, ctinfo, ct, dptr, (*pskb)->len - dataoff, - buffer, bufflen, POS_CONTACT); - return 1; - } - /* This mangle requests headers. */ - return mangle_sip_packet(pskb, ctinfo, ct, dptr, - ct_sip_lnlen(*dptr, - *dptr + (*pskb)->len - dataoff), - buffer, bufflen, POS_REQ_HEADER); + /* We need to reload this. Thanks Patrick. */ + *dptr = (*pskb)->data + (*pskb)->nh.iph->ihl*4 + sizeof(struct udphdr); + return 1; } static int mangle_content_len(struct sk_buff **pskb, -- cgit v1.2.3-70-g09d2 From 829e17a1a602572ffa3beefe582dc103ee9fb9c7 Mon Sep 17 00:00:00 2001 From: Eric Leblond Date: Wed, 29 Nov 2006 02:35:33 +0100 Subject: [NETFILTER]: nfnetlink_queue: allow changing queue length through netlink Signed-off-by: Eric Leblond Signed-off-by: Patrick McHardy --- include/linux/netfilter/nfnetlink_queue.h | 1 + net/netfilter/nfnetlink_queue.c | 8 ++++++++ 2 files changed, 9 insertions(+) (limited to 'include') diff --git a/include/linux/netfilter/nfnetlink_queue.h b/include/linux/netfilter/nfnetlink_queue.h index 4beea3d6fda..83e789633e3 100644 --- a/include/linux/netfilter/nfnetlink_queue.h +++ b/include/linux/netfilter/nfnetlink_queue.h @@ -82,6 +82,7 @@ enum nfqnl_attr_config { NFQA_CFG_UNSPEC, NFQA_CFG_CMD, /* nfqnl_msg_config_cmd */ NFQA_CFG_PARAMS, /* nfqnl_msg_config_params */ + NFQA_CFG_QUEUE_MAXLEN, /* u_int32_t */ __NFQA_CFG_MAX }; #define NFQA_CFG_MAX (__NFQA_CFG_MAX-1) diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index 82e4454659b..a88a017da22 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -947,6 +947,14 @@ nfqnl_recv_config(struct sock *ctnl, struct sk_buff *skb, ntohl(params->copy_range)); } + if (nfqa[NFQA_CFG_QUEUE_MAXLEN-1]) { + __be32 *queue_maxlen; + queue_maxlen = NFA_DATA(nfqa[NFQA_CFG_QUEUE_MAXLEN-1]); + spin_lock_bh(&queue->lock); + queue->queue_maxlen = ntohl(*queue_maxlen); + spin_unlock_bh(&queue->lock); + } + out_put: instance_put(queue); return ret; -- cgit v1.2.3-70-g09d2 From d7a5c32442ed3d528b9ddfd3d5b837bad0ffa9da Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 29 Nov 2006 02:35:34 +0100 Subject: [NETFILTER]: nfnetlink_log: remove useless prefix length limitation There is no reason for limiting netlink attributes in size. Signed-off-by: Patrick McHardy --- include/linux/netfilter/nfnetlink_log.h | 2 -- net/netfilter/nfnetlink_log.c | 19 ++++++++++--------- 2 files changed, 10 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nfnetlink_log.h b/include/linux/netfilter/nfnetlink_log.h index 55a2a2b814e..5966afa026e 100644 --- a/include/linux/netfilter/nfnetlink_log.h +++ b/include/linux/netfilter/nfnetlink_log.h @@ -32,8 +32,6 @@ struct nfulnl_msg_packet_timestamp { aligned_be64 usec; }; -#define NFULNL_PREFIXLEN 30 /* just like old log target */ - enum nfulnl_attr_type { NFULA_UNSPEC, NFULA_PACKET_HDR, diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index bd3ffa6f1a6..d1505dd25c6 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -408,7 +408,7 @@ __build_packet_message(struct nfulnl_instance *inst, const struct net_device *indev, const struct net_device *outdev, const struct nf_loginfo *li, - const char *prefix) + const char *prefix, unsigned int plen) { unsigned char *old_tail; struct nfulnl_msg_packet_hdr pmsg; @@ -432,12 +432,8 @@ __build_packet_message(struct nfulnl_instance *inst, NFA_PUT(inst->skb, NFULA_PACKET_HDR, sizeof(pmsg), &pmsg); - if (prefix) { - int slen = strlen(prefix); - if (slen > NFULNL_PREFIXLEN) - slen = NFULNL_PREFIXLEN; - NFA_PUT(inst->skb, NFULA_PREFIX, slen, prefix); - } + if (prefix) + NFA_PUT(inst->skb, NFULA_PREFIX, plen, prefix); if (indev) { tmp_uint = htonl(indev->ifindex); @@ -601,6 +597,7 @@ nfulnl_log_packet(unsigned int pf, const struct nf_loginfo *li; unsigned int qthreshold; unsigned int nlbufsiz; + unsigned int plen; if (li_user && li_user->type == NF_LOG_TYPE_ULOG) li = li_user; @@ -616,6 +613,10 @@ nfulnl_log_packet(unsigned int pf, return; } + plen = 0; + if (prefix) + plen = strlen(prefix); + /* all macros expand to constant values at compile time */ /* FIXME: do we want to make the size calculation conditional based on * what is actually present? way more branches and checks, but more @@ -630,7 +631,7 @@ nfulnl_log_packet(unsigned int pf, #endif + NFA_SPACE(sizeof(u_int32_t)) /* mark */ + NFA_SPACE(sizeof(u_int32_t)) /* uid */ - + NFA_SPACE(NFULNL_PREFIXLEN) /* prefix */ + + NFA_SPACE(plen) /* prefix */ + NFA_SPACE(sizeof(struct nfulnl_msg_packet_hw)) + NFA_SPACE(sizeof(struct nfulnl_msg_packet_timestamp)); @@ -701,7 +702,7 @@ nfulnl_log_packet(unsigned int pf, inst->qlen++; __build_packet_message(inst, skb, data_len, pf, - hooknum, in, out, li, prefix); + hooknum, in, out, li, prefix, plen); /* timer_pending always called within inst->lock, so there * is no chance of a race here */ -- cgit v1.2.3-70-g09d2 From 39b46fc6f0d1161a5585cd8af7b3a05e8118ab7e Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 29 Nov 2006 02:35:36 +0100 Subject: [NETFILTER]: x_tables: add port of hashlimit match for IPv4 and IPv6 Signed-off-by: Patrick McHardy --- include/linux/netfilter/Kbuild | 1 + include/linux/netfilter/xt_hashlimit.h | 40 ++ include/linux/netfilter_ipv4/ipt_hashlimit.h | 42 +- net/ipv4/netfilter/Kconfig | 14 - net/ipv4/netfilter/Makefile | 1 - net/ipv4/netfilter/ipt_hashlimit.c | 733 ------------------------- net/netfilter/Kconfig | 14 + net/netfilter/Makefile | 1 + net/netfilter/xt_hashlimit.c | 772 +++++++++++++++++++++++++++ 9 files changed, 836 insertions(+), 782 deletions(-) create mode 100644 include/linux/netfilter/xt_hashlimit.h delete mode 100644 net/ipv4/netfilter/ipt_hashlimit.c create mode 100644 net/netfilter/xt_hashlimit.c (limited to 'include') diff --git a/include/linux/netfilter/Kbuild b/include/linux/netfilter/Kbuild index 312bd2ffee3..e379a2d89ea 100644 --- a/include/linux/netfilter/Kbuild +++ b/include/linux/netfilter/Kbuild @@ -14,6 +14,7 @@ header-y += xt_dscp.h header-y += xt_DSCP.h header-y += xt_esp.h header-y += xt_helper.h +header-y += xt_hashlimit.h header-y += xt_length.h header-y += xt_limit.h header-y += xt_mac.h diff --git a/include/linux/netfilter/xt_hashlimit.h b/include/linux/netfilter/xt_hashlimit.h new file mode 100644 index 00000000000..b4556b8edbf --- /dev/null +++ b/include/linux/netfilter/xt_hashlimit.h @@ -0,0 +1,40 @@ +#ifndef _XT_HASHLIMIT_H +#define _XT_HASHLIMIT_H + +/* timings are in milliseconds. */ +#define XT_HASHLIMIT_SCALE 10000 +/* 1/10,000 sec period => max of 10,000/sec. Min rate is then 429490 + seconds, or one every 59 hours. */ + +/* details of this structure hidden by the implementation */ +struct xt_hashlimit_htable; + +#define XT_HASHLIMIT_HASH_DIP 0x0001 +#define XT_HASHLIMIT_HASH_DPT 0x0002 +#define XT_HASHLIMIT_HASH_SIP 0x0004 +#define XT_HASHLIMIT_HASH_SPT 0x0008 + +struct hashlimit_cfg { + u_int32_t mode; /* bitmask of IPT_HASHLIMIT_HASH_* */ + u_int32_t avg; /* Average secs between packets * scale */ + u_int32_t burst; /* Period multiplier for upper limit. */ + + /* user specified */ + u_int32_t size; /* how many buckets */ + u_int32_t max; /* max number of entries */ + u_int32_t gc_interval; /* gc interval */ + u_int32_t expire; /* when do entries expire? */ +}; + +struct xt_hashlimit_info { + char name [IFNAMSIZ]; /* name */ + struct hashlimit_cfg cfg; + struct xt_hashlimit_htable *hinfo; + + /* Used internally by the kernel */ + union { + void *ptr; + struct xt_hashlimit_info *master; + } u; +}; +#endif /*_XT_HASHLIMIT_H*/ diff --git a/include/linux/netfilter_ipv4/ipt_hashlimit.h b/include/linux/netfilter_ipv4/ipt_hashlimit.h index ac2cb64ecd7..5662120a3d7 100644 --- a/include/linux/netfilter_ipv4/ipt_hashlimit.h +++ b/include/linux/netfilter_ipv4/ipt_hashlimit.h @@ -1,40 +1,14 @@ #ifndef _IPT_HASHLIMIT_H #define _IPT_HASHLIMIT_H -/* timings are in milliseconds. */ -#define IPT_HASHLIMIT_SCALE 10000 -/* 1/10,000 sec period => max of 10,000/sec. Min rate is then 429490 - seconds, or one every 59 hours. */ +#include -/* details of this structure hidden by the implementation */ -struct ipt_hashlimit_htable; +#define IPT_HASHLIMIT_SCALE XT_HASHLIMIT_SCALE +#define IPT_HASHLIMIT_HASH_DIP XT_HASHLIMIT_HASH_DIP +#define IPT_HASHLIMIT_HASH_DPT XT_HASHLIMIT_HASH_DPT +#define IPT_HASHLIMIT_HASH_SIP XT_HASHLIMIT_HASH_SIP +#define IPT_HASHLIMIT_HASH_SPT XT_HASHLIMIT_HASH_SPT -#define IPT_HASHLIMIT_HASH_DIP 0x0001 -#define IPT_HASHLIMIT_HASH_DPT 0x0002 -#define IPT_HASHLIMIT_HASH_SIP 0x0004 -#define IPT_HASHLIMIT_HASH_SPT 0x0008 +#define ipt_hashlimit_info xt_hashlimit_info -struct hashlimit_cfg { - u_int32_t mode; /* bitmask of IPT_HASHLIMIT_HASH_* */ - u_int32_t avg; /* Average secs between packets * scale */ - u_int32_t burst; /* Period multiplier for upper limit. */ - - /* user specified */ - u_int32_t size; /* how many buckets */ - u_int32_t max; /* max number of entries */ - u_int32_t gc_interval; /* gc interval */ - u_int32_t expire; /* when do entries expire? */ -}; - -struct ipt_hashlimit_info { - char name [IFNAMSIZ]; /* name */ - struct hashlimit_cfg cfg; - struct ipt_hashlimit_htable *hinfo; - - /* Used internally by the kernel */ - union { - void *ptr; - struct ipt_hashlimit_info *master; - } u; -}; -#endif /*_IPT_HASHLIMIT_H*/ +#endif /* _IPT_HASHLIMIT_H */ diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 4ac5b5c4678..bc298a3f236 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -326,20 +326,6 @@ config IP_NF_MATCH_ADDRTYPE If you want to compile it as a module, say M here and read . If unsure, say `N'. -config IP_NF_MATCH_HASHLIMIT - tristate 'hashlimit match support' - depends on IP_NF_IPTABLES - help - This option adds a new iptables `hashlimit' match. - - As opposed to `limit', this match dynamically creates a hash table - of limit buckets, based on your selection of source/destination - ip addresses and/or ports. - - It enables you to express policies like `10kpps for any given - destination IP' or `500pps from any given source IP' with a single - IPtables rule. - # `filter', generic and specific targets config IP_NF_FILTER tristate "Packet filtering" diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index 4ce20ebc4d6..21359d83f0c 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -53,7 +53,6 @@ obj-$(CONFIG_IP_NF_NAT) += iptable_nat.o obj-$(CONFIG_IP_NF_RAW) += iptable_raw.o # matches -obj-$(CONFIG_IP_NF_MATCH_HASHLIMIT) += ipt_hashlimit.o obj-$(CONFIG_IP_NF_MATCH_IPRANGE) += ipt_iprange.o obj-$(CONFIG_IP_NF_MATCH_OWNER) += ipt_owner.o obj-$(CONFIG_IP_NF_MATCH_TOS) += ipt_tos.o diff --git a/net/ipv4/netfilter/ipt_hashlimit.c b/net/ipv4/netfilter/ipt_hashlimit.c deleted file mode 100644 index 33ccdbf8e79..00000000000 --- a/net/ipv4/netfilter/ipt_hashlimit.c +++ /dev/null @@ -1,733 +0,0 @@ -/* iptables match extension to limit the number of packets per second - * seperately for each hashbucket (sourceip/sourceport/dstip/dstport) - * - * (C) 2003-2004 by Harald Welte - * - * $Id: ipt_hashlimit.c 3244 2004-10-20 16:24:29Z laforge@netfilter.org $ - * - * Development of this code was funded by Astaro AG, http://www.astaro.com/ - * - * based on ipt_limit.c by: - * Jérôme de Vivie - * Hervé Eychenne - * Rusty Russell - * - * The general idea is to create a hash table for every dstip and have a - * seperate limit counter per tuple. This way you can do something like 'limit - * the number of syn packets for each of my internal addresses. - * - * Ideally this would just be implemented as a general 'hash' match, which would - * allow us to attach any iptables target to it's hash buckets. But this is - * not possible in the current iptables architecture. As always, pkttables for - * 2.7.x will help ;) - */ -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include - -/* FIXME: this is just for IP_NF_ASSERRT */ -#include -#include - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Harald Welte "); -MODULE_DESCRIPTION("iptables match for limiting per hash-bucket"); - -/* need to declare this at the top */ -static struct proc_dir_entry *hashlimit_procdir; -static struct file_operations dl_file_ops; - -/* hash table crap */ - -struct dsthash_dst { - __be32 src_ip; - __be32 dst_ip; - /* ports have to be consecutive !!! */ - __be16 src_port; - __be16 dst_port; -}; - -struct dsthash_ent { - /* static / read-only parts in the beginning */ - struct hlist_node node; - struct dsthash_dst dst; - - /* modified structure members in the end */ - unsigned long expires; /* precalculated expiry time */ - struct { - unsigned long prev; /* last modification */ - u_int32_t credit; - u_int32_t credit_cap, cost; - } rateinfo; -}; - -struct ipt_hashlimit_htable { - struct hlist_node node; /* global list of all htables */ - atomic_t use; - - struct hashlimit_cfg cfg; /* config */ - - /* used internally */ - spinlock_t lock; /* lock for list_head */ - u_int32_t rnd; /* random seed for hash */ - int rnd_initialized; - struct timer_list timer; /* timer for gc */ - atomic_t count; /* number entries in table */ - - /* seq_file stuff */ - struct proc_dir_entry *pde; - - struct hlist_head hash[0]; /* hashtable itself */ -}; - -static DEFINE_SPINLOCK(hashlimit_lock); /* protects htables list */ -static DEFINE_MUTEX(hlimit_mutex); /* additional checkentry protection */ -static HLIST_HEAD(hashlimit_htables); -static kmem_cache_t *hashlimit_cachep __read_mostly; - -static inline int dst_cmp(const struct dsthash_ent *ent, struct dsthash_dst *b) -{ - return (ent->dst.dst_ip == b->dst_ip - && ent->dst.dst_port == b->dst_port - && ent->dst.src_port == b->src_port - && ent->dst.src_ip == b->src_ip); -} - -static inline u_int32_t -hash_dst(const struct ipt_hashlimit_htable *ht, const struct dsthash_dst *dst) -{ - return (jhash_3words((__force u32)dst->dst_ip, - ((__force u32)dst->dst_port<<16 | - (__force u32)dst->src_port), - (__force u32)dst->src_ip, ht->rnd) % ht->cfg.size); -} - -static inline struct dsthash_ent * -__dsthash_find(const struct ipt_hashlimit_htable *ht, struct dsthash_dst *dst) -{ - struct dsthash_ent *ent; - struct hlist_node *pos; - u_int32_t hash = hash_dst(ht, dst); - - if (!hlist_empty(&ht->hash[hash])) - hlist_for_each_entry(ent, pos, &ht->hash[hash], node) { - if (dst_cmp(ent, dst)) { - return ent; - } - } - - return NULL; -} - -/* allocate dsthash_ent, initialize dst, put in htable and lock it */ -static struct dsthash_ent * -__dsthash_alloc_init(struct ipt_hashlimit_htable *ht, struct dsthash_dst *dst) -{ - struct dsthash_ent *ent; - - /* initialize hash with random val at the time we allocate - * the first hashtable entry */ - if (!ht->rnd_initialized) { - get_random_bytes(&ht->rnd, 4); - ht->rnd_initialized = 1; - } - - if (ht->cfg.max && - atomic_read(&ht->count) >= ht->cfg.max) { - /* FIXME: do something. question is what.. */ - if (net_ratelimit()) - printk(KERN_WARNING - "ipt_hashlimit: max count of %u reached\n", - ht->cfg.max); - return NULL; - } - - ent = kmem_cache_alloc(hashlimit_cachep, GFP_ATOMIC); - if (!ent) { - if (net_ratelimit()) - printk(KERN_ERR - "ipt_hashlimit: can't allocate dsthash_ent\n"); - return NULL; - } - - atomic_inc(&ht->count); - - ent->dst.dst_ip = dst->dst_ip; - ent->dst.dst_port = dst->dst_port; - ent->dst.src_ip = dst->src_ip; - ent->dst.src_port = dst->src_port; - - hlist_add_head(&ent->node, &ht->hash[hash_dst(ht, dst)]); - - return ent; -} - -static inline void -__dsthash_free(struct ipt_hashlimit_htable *ht, struct dsthash_ent *ent) -{ - hlist_del(&ent->node); - kmem_cache_free(hashlimit_cachep, ent); - atomic_dec(&ht->count); -} -static void htable_gc(unsigned long htlong); - -static int htable_create(struct ipt_hashlimit_info *minfo) -{ - int i; - unsigned int size; - struct ipt_hashlimit_htable *hinfo; - - if (minfo->cfg.size) - size = minfo->cfg.size; - else { - size = (((num_physpages << PAGE_SHIFT) / 16384) - / sizeof(struct list_head)); - if (num_physpages > (1024 * 1024 * 1024 / PAGE_SIZE)) - size = 8192; - if (size < 16) - size = 16; - } - /* FIXME: don't use vmalloc() here or anywhere else -HW */ - hinfo = vmalloc(sizeof(struct ipt_hashlimit_htable) - + (sizeof(struct list_head) * size)); - if (!hinfo) { - printk(KERN_ERR "ipt_hashlimit: Unable to create hashtable\n"); - return -1; - } - minfo->hinfo = hinfo; - - /* copy match config into hashtable config */ - memcpy(&hinfo->cfg, &minfo->cfg, sizeof(hinfo->cfg)); - hinfo->cfg.size = size; - if (!hinfo->cfg.max) - hinfo->cfg.max = 8 * hinfo->cfg.size; - else if (hinfo->cfg.max < hinfo->cfg.size) - hinfo->cfg.max = hinfo->cfg.size; - - for (i = 0; i < hinfo->cfg.size; i++) - INIT_HLIST_HEAD(&hinfo->hash[i]); - - atomic_set(&hinfo->count, 0); - atomic_set(&hinfo->use, 1); - hinfo->rnd_initialized = 0; - spin_lock_init(&hinfo->lock); - hinfo->pde = create_proc_entry(minfo->name, 0, hashlimit_procdir); - if (!hinfo->pde) { - vfree(hinfo); - return -1; - } - hinfo->pde->proc_fops = &dl_file_ops; - hinfo->pde->data = hinfo; - - init_timer(&hinfo->timer); - hinfo->timer.expires = jiffies + msecs_to_jiffies(hinfo->cfg.gc_interval); - hinfo->timer.data = (unsigned long )hinfo; - hinfo->timer.function = htable_gc; - add_timer(&hinfo->timer); - - spin_lock_bh(&hashlimit_lock); - hlist_add_head(&hinfo->node, &hashlimit_htables); - spin_unlock_bh(&hashlimit_lock); - - return 0; -} - -static int select_all(struct ipt_hashlimit_htable *ht, struct dsthash_ent *he) -{ - return 1; -} - -static int select_gc(struct ipt_hashlimit_htable *ht, struct dsthash_ent *he) -{ - return (jiffies >= he->expires); -} - -static void htable_selective_cleanup(struct ipt_hashlimit_htable *ht, - int (*select)(struct ipt_hashlimit_htable *ht, - struct dsthash_ent *he)) -{ - int i; - - IP_NF_ASSERT(ht->cfg.size && ht->cfg.max); - - /* lock hash table and iterate over it */ - spin_lock_bh(&ht->lock); - for (i = 0; i < ht->cfg.size; i++) { - struct dsthash_ent *dh; - struct hlist_node *pos, *n; - hlist_for_each_entry_safe(dh, pos, n, &ht->hash[i], node) { - if ((*select)(ht, dh)) - __dsthash_free(ht, dh); - } - } - spin_unlock_bh(&ht->lock); -} - -/* hash table garbage collector, run by timer */ -static void htable_gc(unsigned long htlong) -{ - struct ipt_hashlimit_htable *ht = (struct ipt_hashlimit_htable *)htlong; - - htable_selective_cleanup(ht, select_gc); - - /* re-add the timer accordingly */ - ht->timer.expires = jiffies + msecs_to_jiffies(ht->cfg.gc_interval); - add_timer(&ht->timer); -} - -static void htable_destroy(struct ipt_hashlimit_htable *hinfo) -{ - /* remove timer, if it is pending */ - if (timer_pending(&hinfo->timer)) - del_timer(&hinfo->timer); - - /* remove proc entry */ - remove_proc_entry(hinfo->pde->name, hashlimit_procdir); - - htable_selective_cleanup(hinfo, select_all); - vfree(hinfo); -} - -static struct ipt_hashlimit_htable *htable_find_get(char *name) -{ - struct ipt_hashlimit_htable *hinfo; - struct hlist_node *pos; - - spin_lock_bh(&hashlimit_lock); - hlist_for_each_entry(hinfo, pos, &hashlimit_htables, node) { - if (!strcmp(name, hinfo->pde->name)) { - atomic_inc(&hinfo->use); - spin_unlock_bh(&hashlimit_lock); - return hinfo; - } - } - spin_unlock_bh(&hashlimit_lock); - - return NULL; -} - -static void htable_put(struct ipt_hashlimit_htable *hinfo) -{ - if (atomic_dec_and_test(&hinfo->use)) { - spin_lock_bh(&hashlimit_lock); - hlist_del(&hinfo->node); - spin_unlock_bh(&hashlimit_lock); - htable_destroy(hinfo); - } -} - - -/* The algorithm used is the Simple Token Bucket Filter (TBF) - * see net/sched/sch_tbf.c in the linux source tree - */ - -/* Rusty: This is my (non-mathematically-inclined) understanding of - this algorithm. The `average rate' in jiffies becomes your initial - amount of credit `credit' and the most credit you can ever have - `credit_cap'. The `peak rate' becomes the cost of passing the - test, `cost'. - - `prev' tracks the last packet hit: you gain one credit per jiffy. - If you get credit balance more than this, the extra credit is - discarded. Every time the match passes, you lose `cost' credits; - if you don't have that many, the test fails. - - See Alexey's formal explanation in net/sched/sch_tbf.c. - - To get the maximum range, we multiply by this factor (ie. you get N - credits per jiffy). We want to allow a rate as low as 1 per day - (slowest userspace tool allows), which means - CREDITS_PER_JIFFY*HZ*60*60*24 < 2^32 ie. -*/ -#define MAX_CPJ (0xFFFFFFFF / (HZ*60*60*24)) - -/* Repeated shift and or gives us all 1s, final shift and add 1 gives - * us the power of 2 below the theoretical max, so GCC simply does a - * shift. */ -#define _POW2_BELOW2(x) ((x)|((x)>>1)) -#define _POW2_BELOW4(x) (_POW2_BELOW2(x)|_POW2_BELOW2((x)>>2)) -#define _POW2_BELOW8(x) (_POW2_BELOW4(x)|_POW2_BELOW4((x)>>4)) -#define _POW2_BELOW16(x) (_POW2_BELOW8(x)|_POW2_BELOW8((x)>>8)) -#define _POW2_BELOW32(x) (_POW2_BELOW16(x)|_POW2_BELOW16((x)>>16)) -#define POW2_BELOW32(x) ((_POW2_BELOW32(x)>>1) + 1) - -#define CREDITS_PER_JIFFY POW2_BELOW32(MAX_CPJ) - -/* Precision saver. */ -static inline u_int32_t -user2credits(u_int32_t user) -{ - /* If multiplying would overflow... */ - if (user > 0xFFFFFFFF / (HZ*CREDITS_PER_JIFFY)) - /* Divide first. */ - return (user / IPT_HASHLIMIT_SCALE) * HZ * CREDITS_PER_JIFFY; - - return (user * HZ * CREDITS_PER_JIFFY) / IPT_HASHLIMIT_SCALE; -} - -static inline void rateinfo_recalc(struct dsthash_ent *dh, unsigned long now) -{ - dh->rateinfo.credit += (now - xchg(&dh->rateinfo.prev, now)) - * CREDITS_PER_JIFFY; - if (dh->rateinfo.credit > dh->rateinfo.credit_cap) - dh->rateinfo.credit = dh->rateinfo.credit_cap; -} - -static int -hashlimit_match(const struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - const struct xt_match *match, - const void *matchinfo, - int offset, - unsigned int protoff, - int *hotdrop) -{ - struct ipt_hashlimit_info *r = - ((struct ipt_hashlimit_info *)matchinfo)->u.master; - struct ipt_hashlimit_htable *hinfo = r->hinfo; - unsigned long now = jiffies; - struct dsthash_ent *dh; - struct dsthash_dst dst; - - /* build 'dst' according to hinfo->cfg and current packet */ - memset(&dst, 0, sizeof(dst)); - if (hinfo->cfg.mode & IPT_HASHLIMIT_HASH_DIP) - dst.dst_ip = skb->nh.iph->daddr; - if (hinfo->cfg.mode & IPT_HASHLIMIT_HASH_SIP) - dst.src_ip = skb->nh.iph->saddr; - if (hinfo->cfg.mode & IPT_HASHLIMIT_HASH_DPT - ||hinfo->cfg.mode & IPT_HASHLIMIT_HASH_SPT) { - __be16 _ports[2], *ports; - - switch (skb->nh.iph->protocol) { - case IPPROTO_TCP: - case IPPROTO_UDP: - case IPPROTO_SCTP: - case IPPROTO_DCCP: - ports = skb_header_pointer(skb, skb->nh.iph->ihl*4, - sizeof(_ports), &_ports); - break; - default: - _ports[0] = _ports[1] = 0; - ports = _ports; - break; - } - if (!ports) { - /* We've been asked to examine this packet, and we - can't. Hence, no choice but to drop. */ - *hotdrop = 1; - return 0; - } - if (hinfo->cfg.mode & IPT_HASHLIMIT_HASH_SPT) - dst.src_port = ports[0]; - if (hinfo->cfg.mode & IPT_HASHLIMIT_HASH_DPT) - dst.dst_port = ports[1]; - } - - spin_lock_bh(&hinfo->lock); - dh = __dsthash_find(hinfo, &dst); - if (!dh) { - dh = __dsthash_alloc_init(hinfo, &dst); - - if (!dh) { - /* enomem... don't match == DROP */ - if (net_ratelimit()) - printk(KERN_ERR "%s: ENOMEM\n", __FUNCTION__); - spin_unlock_bh(&hinfo->lock); - return 0; - } - - dh->expires = jiffies + msecs_to_jiffies(hinfo->cfg.expire); - - dh->rateinfo.prev = jiffies; - dh->rateinfo.credit = user2credits(hinfo->cfg.avg * - hinfo->cfg.burst); - dh->rateinfo.credit_cap = user2credits(hinfo->cfg.avg * - hinfo->cfg.burst); - dh->rateinfo.cost = user2credits(hinfo->cfg.avg); - } else { - /* update expiration timeout */ - dh->expires = now + msecs_to_jiffies(hinfo->cfg.expire); - rateinfo_recalc(dh, now); - } - - if (dh->rateinfo.credit >= dh->rateinfo.cost) { - /* We're underlimit. */ - dh->rateinfo.credit -= dh->rateinfo.cost; - spin_unlock_bh(&hinfo->lock); - return 1; - } - - spin_unlock_bh(&hinfo->lock); - - /* default case: we're overlimit, thus don't match */ - return 0; -} - -static int -hashlimit_checkentry(const char *tablename, - const void *inf, - const struct xt_match *match, - void *matchinfo, - unsigned int hook_mask) -{ - struct ipt_hashlimit_info *r = matchinfo; - - /* Check for overflow. */ - if (r->cfg.burst == 0 - || user2credits(r->cfg.avg * r->cfg.burst) < - user2credits(r->cfg.avg)) { - printk(KERN_ERR "ipt_hashlimit: Overflow, try lower: %u/%u\n", - r->cfg.avg, r->cfg.burst); - return 0; - } - - if (r->cfg.mode == 0 - || r->cfg.mode > (IPT_HASHLIMIT_HASH_DPT - |IPT_HASHLIMIT_HASH_DIP - |IPT_HASHLIMIT_HASH_SIP - |IPT_HASHLIMIT_HASH_SPT)) - return 0; - - if (!r->cfg.gc_interval) - return 0; - - if (!r->cfg.expire) - return 0; - - if (r->name[sizeof(r->name) - 1] != '\0') - return 0; - - /* This is the best we've got: We cannot release and re-grab lock, - * since checkentry() is called before ip_tables.c grabs ipt_mutex. - * We also cannot grab the hashtable spinlock, since htable_create will - * call vmalloc, and that can sleep. And we cannot just re-search - * the list of htable's in htable_create(), since then we would - * create duplicate proc files. -HW */ - mutex_lock(&hlimit_mutex); - r->hinfo = htable_find_get(r->name); - if (!r->hinfo && (htable_create(r) != 0)) { - mutex_unlock(&hlimit_mutex); - return 0; - } - mutex_unlock(&hlimit_mutex); - - /* Ugly hack: For SMP, we only want to use one set */ - r->u.master = r; - - return 1; -} - -static void -hashlimit_destroy(const struct xt_match *match, void *matchinfo) -{ - struct ipt_hashlimit_info *r = matchinfo; - - htable_put(r->hinfo); -} - -#ifdef CONFIG_COMPAT -struct compat_ipt_hashlimit_info { - char name[IFNAMSIZ]; - struct hashlimit_cfg cfg; - compat_uptr_t hinfo; - compat_uptr_t master; -}; - -static void compat_from_user(void *dst, void *src) -{ - int off = offsetof(struct compat_ipt_hashlimit_info, hinfo); - - memcpy(dst, src, off); - memset(dst + off, 0, sizeof(struct compat_ipt_hashlimit_info) - off); -} - -static int compat_to_user(void __user *dst, void *src) -{ - int off = offsetof(struct compat_ipt_hashlimit_info, hinfo); - - return copy_to_user(dst, src, off) ? -EFAULT : 0; -} -#endif - -static struct ipt_match ipt_hashlimit = { - .name = "hashlimit", - .match = hashlimit_match, - .matchsize = sizeof(struct ipt_hashlimit_info), -#ifdef CONFIG_COMPAT - .compatsize = sizeof(struct compat_ipt_hashlimit_info), - .compat_from_user = compat_from_user, - .compat_to_user = compat_to_user, -#endif - .checkentry = hashlimit_checkentry, - .destroy = hashlimit_destroy, - .me = THIS_MODULE -}; - -/* PROC stuff */ - -static void *dl_seq_start(struct seq_file *s, loff_t *pos) -{ - struct proc_dir_entry *pde = s->private; - struct ipt_hashlimit_htable *htable = pde->data; - unsigned int *bucket; - - spin_lock_bh(&htable->lock); - if (*pos >= htable->cfg.size) - return NULL; - - bucket = kmalloc(sizeof(unsigned int), GFP_ATOMIC); - if (!bucket) - return ERR_PTR(-ENOMEM); - - *bucket = *pos; - return bucket; -} - -static void *dl_seq_next(struct seq_file *s, void *v, loff_t *pos) -{ - struct proc_dir_entry *pde = s->private; - struct ipt_hashlimit_htable *htable = pde->data; - unsigned int *bucket = (unsigned int *)v; - - *pos = ++(*bucket); - if (*pos >= htable->cfg.size) { - kfree(v); - return NULL; - } - return bucket; -} - -static void dl_seq_stop(struct seq_file *s, void *v) -{ - struct proc_dir_entry *pde = s->private; - struct ipt_hashlimit_htable *htable = pde->data; - unsigned int *bucket = (unsigned int *)v; - - kfree(bucket); - - spin_unlock_bh(&htable->lock); -} - -static inline int dl_seq_real_show(struct dsthash_ent *ent, struct seq_file *s) -{ - /* recalculate to show accurate numbers */ - rateinfo_recalc(ent, jiffies); - - return seq_printf(s, "%ld %u.%u.%u.%u:%u->%u.%u.%u.%u:%u %u %u %u\n", - (long)(ent->expires - jiffies)/HZ, - NIPQUAD(ent->dst.src_ip), ntohs(ent->dst.src_port), - NIPQUAD(ent->dst.dst_ip), ntohs(ent->dst.dst_port), - ent->rateinfo.credit, ent->rateinfo.credit_cap, - ent->rateinfo.cost); -} - -static int dl_seq_show(struct seq_file *s, void *v) -{ - struct proc_dir_entry *pde = s->private; - struct ipt_hashlimit_htable *htable = pde->data; - unsigned int *bucket = (unsigned int *)v; - struct dsthash_ent *ent; - struct hlist_node *pos; - - if (!hlist_empty(&htable->hash[*bucket])) - hlist_for_each_entry(ent, pos, &htable->hash[*bucket], node) { - if (dl_seq_real_show(ent, s)) { - /* buffer was filled and unable to print that tuple */ - return 1; - } - } - - return 0; -} - -static struct seq_operations dl_seq_ops = { - .start = dl_seq_start, - .next = dl_seq_next, - .stop = dl_seq_stop, - .show = dl_seq_show -}; - -static int dl_proc_open(struct inode *inode, struct file *file) -{ - int ret = seq_open(file, &dl_seq_ops); - - if (!ret) { - struct seq_file *sf = file->private_data; - sf->private = PDE(inode); - } - return ret; -} - -static struct file_operations dl_file_ops = { - .owner = THIS_MODULE, - .open = dl_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release -}; - -static int init_or_fini(int fini) -{ - int ret = 0; - - if (fini) - goto cleanup; - - if (ipt_register_match(&ipt_hashlimit)) { - ret = -EINVAL; - goto cleanup_nothing; - } - - hashlimit_cachep = kmem_cache_create("ipt_hashlimit", - sizeof(struct dsthash_ent), 0, - 0, NULL, NULL); - if (!hashlimit_cachep) { - printk(KERN_ERR "Unable to create ipt_hashlimit slab cache\n"); - ret = -ENOMEM; - goto cleanup_unreg_match; - } - - hashlimit_procdir = proc_mkdir("ipt_hashlimit", proc_net); - if (!hashlimit_procdir) { - printk(KERN_ERR "Unable to create proc dir entry\n"); - ret = -ENOMEM; - goto cleanup_free_slab; - } - - return ret; - -cleanup: - remove_proc_entry("ipt_hashlimit", proc_net); -cleanup_free_slab: - kmem_cache_destroy(hashlimit_cachep); -cleanup_unreg_match: - ipt_unregister_match(&ipt_hashlimit); -cleanup_nothing: - return ret; - -} - -static int __init ipt_hashlimit_init(void) -{ - return init_or_fini(0); -} - -static void __exit ipt_hashlimit_fini(void) -{ - init_or_fini(1); -} - -module_init(ipt_hashlimit_init); -module_exit(ipt_hashlimit_fini); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index f619c652726..7e6125467c1 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -464,5 +464,19 @@ config NETFILTER_XT_MATCH_TCPMSS To compile it as a module, choose M here. If unsure, say N. +config NETFILTER_XT_MATCH_HASHLIMIT + tristate '"hashlimit" match support' + depends on NETFILTER_XTABLES + help + This option adds a `hashlimit' match. + + As opposed to `limit', this match dynamically creates a hash table + of limit buckets, based on your selection of source/destination + addresses and/or ports. + + It enables you to express policies like `10kpps for any given + destination address' or `500pps from any given source address' + with a single rule. + endmenu diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 84d529ded95..f85811bfcfe 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -59,3 +59,4 @@ obj-$(CONFIG_NETFILTER_XT_MATCH_STATISTIC) += xt_statistic.o obj-$(CONFIG_NETFILTER_XT_MATCH_STRING) += xt_string.o obj-$(CONFIG_NETFILTER_XT_MATCH_TCPMSS) += xt_tcpmss.o obj-$(CONFIG_NETFILTER_XT_MATCH_PHYSDEV) += xt_physdev.o +obj-$(CONFIG_NETFILTER_XT_MATCH_HASHLIMIT) += xt_hashlimit.o diff --git a/net/netfilter/xt_hashlimit.c b/net/netfilter/xt_hashlimit.c new file mode 100644 index 00000000000..46de566d372 --- /dev/null +++ b/net/netfilter/xt_hashlimit.c @@ -0,0 +1,772 @@ +/* iptables match extension to limit the number of packets per second + * seperately for each hashbucket (sourceip/sourceport/dstip/dstport) + * + * (C) 2003-2004 by Harald Welte + * + * $Id: ipt_hashlimit.c 3244 2004-10-20 16:24:29Z laforge@netfilter.org $ + * + * Development of this code was funded by Astaro AG, http://www.astaro.com/ + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Harald Welte "); +MODULE_DESCRIPTION("iptables match for limiting per hash-bucket"); +MODULE_ALIAS("ipt_hashlimit"); +MODULE_ALIAS("ip6t_hashlimit"); + +/* need to declare this at the top */ +static struct proc_dir_entry *hashlimit_procdir4; +static struct proc_dir_entry *hashlimit_procdir6; +static struct file_operations dl_file_ops; + +/* hash table crap */ +struct dsthash_dst { + union { + struct { + __be32 src; + __be32 dst; + } ip; + struct { + __be32 src[4]; + __be32 dst[4]; + } ip6; + } addr; + __be16 src_port; + __be16 dst_port; +}; + +struct dsthash_ent { + /* static / read-only parts in the beginning */ + struct hlist_node node; + struct dsthash_dst dst; + + /* modified structure members in the end */ + unsigned long expires; /* precalculated expiry time */ + struct { + unsigned long prev; /* last modification */ + u_int32_t credit; + u_int32_t credit_cap, cost; + } rateinfo; +}; + +struct xt_hashlimit_htable { + struct hlist_node node; /* global list of all htables */ + atomic_t use; + int family; + + struct hashlimit_cfg cfg; /* config */ + + /* used internally */ + spinlock_t lock; /* lock for list_head */ + u_int32_t rnd; /* random seed for hash */ + int rnd_initialized; + unsigned int count; /* number entries in table */ + struct timer_list timer; /* timer for gc */ + + /* seq_file stuff */ + struct proc_dir_entry *pde; + + struct hlist_head hash[0]; /* hashtable itself */ +}; + +static DEFINE_SPINLOCK(hashlimit_lock); /* protects htables list */ +static DEFINE_MUTEX(hlimit_mutex); /* additional checkentry protection */ +static HLIST_HEAD(hashlimit_htables); +static kmem_cache_t *hashlimit_cachep __read_mostly; + +static inline int dst_cmp(const struct dsthash_ent *ent, struct dsthash_dst *b) +{ + return !memcmp(&ent->dst, b, sizeof(ent->dst)); +} + +static u_int32_t +hash_dst(const struct xt_hashlimit_htable *ht, const struct dsthash_dst *dst) +{ + return jhash(dst, sizeof(*dst), ht->rnd) % ht->cfg.size; +} + +static struct dsthash_ent * +dsthash_find(const struct xt_hashlimit_htable *ht, struct dsthash_dst *dst) +{ + struct dsthash_ent *ent; + struct hlist_node *pos; + u_int32_t hash = hash_dst(ht, dst); + + if (!hlist_empty(&ht->hash[hash])) { + hlist_for_each_entry(ent, pos, &ht->hash[hash], node) + if (dst_cmp(ent, dst)) + return ent; + } + return NULL; +} + +/* allocate dsthash_ent, initialize dst, put in htable and lock it */ +static struct dsthash_ent * +dsthash_alloc_init(struct xt_hashlimit_htable *ht, struct dsthash_dst *dst) +{ + struct dsthash_ent *ent; + + /* initialize hash with random val at the time we allocate + * the first hashtable entry */ + if (!ht->rnd_initialized) { + get_random_bytes(&ht->rnd, 4); + ht->rnd_initialized = 1; + } + + if (ht->cfg.max && ht->count >= ht->cfg.max) { + /* FIXME: do something. question is what.. */ + if (net_ratelimit()) + printk(KERN_WARNING + "xt_hashlimit: max count of %u reached\n", + ht->cfg.max); + return NULL; + } + + ent = kmem_cache_alloc(hashlimit_cachep, GFP_ATOMIC); + if (!ent) { + if (net_ratelimit()) + printk(KERN_ERR + "xt_hashlimit: can't allocate dsthash_ent\n"); + return NULL; + } + memcpy(&ent->dst, dst, sizeof(ent->dst)); + + hlist_add_head(&ent->node, &ht->hash[hash_dst(ht, dst)]); + ht->count++; + return ent; +} + +static inline void +dsthash_free(struct xt_hashlimit_htable *ht, struct dsthash_ent *ent) +{ + hlist_del(&ent->node); + kmem_cache_free(hashlimit_cachep, ent); + ht->count--; +} +static void htable_gc(unsigned long htlong); + +static int htable_create(struct xt_hashlimit_info *minfo, int family) +{ + struct xt_hashlimit_htable *hinfo; + unsigned int size; + unsigned int i; + + if (minfo->cfg.size) + size = minfo->cfg.size; + else { + size = ((num_physpages << PAGE_SHIFT) / 16384) / + sizeof(struct list_head); + if (num_physpages > (1024 * 1024 * 1024 / PAGE_SIZE)) + size = 8192; + if (size < 16) + size = 16; + } + /* FIXME: don't use vmalloc() here or anywhere else -HW */ + hinfo = vmalloc(sizeof(struct xt_hashlimit_htable) + + sizeof(struct list_head) * size); + if (!hinfo) { + printk(KERN_ERR "xt_hashlimit: unable to create hashtable\n"); + return -1; + } + minfo->hinfo = hinfo; + + /* copy match config into hashtable config */ + memcpy(&hinfo->cfg, &minfo->cfg, sizeof(hinfo->cfg)); + hinfo->cfg.size = size; + if (!hinfo->cfg.max) + hinfo->cfg.max = 8 * hinfo->cfg.size; + else if (hinfo->cfg.max < hinfo->cfg.size) + hinfo->cfg.max = hinfo->cfg.size; + + for (i = 0; i < hinfo->cfg.size; i++) + INIT_HLIST_HEAD(&hinfo->hash[i]); + + atomic_set(&hinfo->use, 1); + hinfo->count = 0; + hinfo->family = family; + hinfo->rnd_initialized = 0; + spin_lock_init(&hinfo->lock); + hinfo->pde = create_proc_entry(minfo->name, 0, + family == AF_INET ? hashlimit_procdir4 : + hashlimit_procdir6); + if (!hinfo->pde) { + vfree(hinfo); + return -1; + } + hinfo->pde->proc_fops = &dl_file_ops; + hinfo->pde->data = hinfo; + + init_timer(&hinfo->timer); + hinfo->timer.expires = jiffies + msecs_to_jiffies(hinfo->cfg.gc_interval); + hinfo->timer.data = (unsigned long )hinfo; + hinfo->timer.function = htable_gc; + add_timer(&hinfo->timer); + + spin_lock_bh(&hashlimit_lock); + hlist_add_head(&hinfo->node, &hashlimit_htables); + spin_unlock_bh(&hashlimit_lock); + + return 0; +} + +static int select_all(struct xt_hashlimit_htable *ht, struct dsthash_ent *he) +{ + return 1; +} + +static int select_gc(struct xt_hashlimit_htable *ht, struct dsthash_ent *he) +{ + return (jiffies >= he->expires); +} + +static void htable_selective_cleanup(struct xt_hashlimit_htable *ht, + int (*select)(struct xt_hashlimit_htable *ht, + struct dsthash_ent *he)) +{ + unsigned int i; + + /* lock hash table and iterate over it */ + spin_lock_bh(&ht->lock); + for (i = 0; i < ht->cfg.size; i++) { + struct dsthash_ent *dh; + struct hlist_node *pos, *n; + hlist_for_each_entry_safe(dh, pos, n, &ht->hash[i], node) { + if ((*select)(ht, dh)) + dsthash_free(ht, dh); + } + } + spin_unlock_bh(&ht->lock); +} + +/* hash table garbage collector, run by timer */ +static void htable_gc(unsigned long htlong) +{ + struct xt_hashlimit_htable *ht = (struct xt_hashlimit_htable *)htlong; + + htable_selective_cleanup(ht, select_gc); + + /* re-add the timer accordingly */ + ht->timer.expires = jiffies + msecs_to_jiffies(ht->cfg.gc_interval); + add_timer(&ht->timer); +} + +static void htable_destroy(struct xt_hashlimit_htable *hinfo) +{ + /* remove timer, if it is pending */ + if (timer_pending(&hinfo->timer)) + del_timer(&hinfo->timer); + + /* remove proc entry */ + remove_proc_entry(hinfo->pde->name, + hinfo->family == AF_INET ? hashlimit_procdir4 : + hashlimit_procdir6); + htable_selective_cleanup(hinfo, select_all); + vfree(hinfo); +} + +static struct xt_hashlimit_htable *htable_find_get(char *name, int family) +{ + struct xt_hashlimit_htable *hinfo; + struct hlist_node *pos; + + spin_lock_bh(&hashlimit_lock); + hlist_for_each_entry(hinfo, pos, &hashlimit_htables, node) { + if (!strcmp(name, hinfo->pde->name) && + hinfo->family == family) { + atomic_inc(&hinfo->use); + spin_unlock_bh(&hashlimit_lock); + return hinfo; + } + } + spin_unlock_bh(&hashlimit_lock); + return NULL; +} + +static void htable_put(struct xt_hashlimit_htable *hinfo) +{ + if (atomic_dec_and_test(&hinfo->use)) { + spin_lock_bh(&hashlimit_lock); + hlist_del(&hinfo->node); + spin_unlock_bh(&hashlimit_lock); + htable_destroy(hinfo); + } +} + +/* The algorithm used is the Simple Token Bucket Filter (TBF) + * see net/sched/sch_tbf.c in the linux source tree + */ + +/* Rusty: This is my (non-mathematically-inclined) understanding of + this algorithm. The `average rate' in jiffies becomes your initial + amount of credit `credit' and the most credit you can ever have + `credit_cap'. The `peak rate' becomes the cost of passing the + test, `cost'. + + `prev' tracks the last packet hit: you gain one credit per jiffy. + If you get credit balance more than this, the extra credit is + discarded. Every time the match passes, you lose `cost' credits; + if you don't have that many, the test fails. + + See Alexey's formal explanation in net/sched/sch_tbf.c. + + To get the maximum range, we multiply by this factor (ie. you get N + credits per jiffy). We want to allow a rate as low as 1 per day + (slowest userspace tool allows), which means + CREDITS_PER_JIFFY*HZ*60*60*24 < 2^32 ie. +*/ +#define MAX_CPJ (0xFFFFFFFF / (HZ*60*60*24)) + +/* Repeated shift and or gives us all 1s, final shift and add 1 gives + * us the power of 2 below the theoretical max, so GCC simply does a + * shift. */ +#define _POW2_BELOW2(x) ((x)|((x)>>1)) +#define _POW2_BELOW4(x) (_POW2_BELOW2(x)|_POW2_BELOW2((x)>>2)) +#define _POW2_BELOW8(x) (_POW2_BELOW4(x)|_POW2_BELOW4((x)>>4)) +#define _POW2_BELOW16(x) (_POW2_BELOW8(x)|_POW2_BELOW8((x)>>8)) +#define _POW2_BELOW32(x) (_POW2_BELOW16(x)|_POW2_BELOW16((x)>>16)) +#define POW2_BELOW32(x) ((_POW2_BELOW32(x)>>1) + 1) + +#define CREDITS_PER_JIFFY POW2_BELOW32(MAX_CPJ) + +/* Precision saver. */ +static inline u_int32_t +user2credits(u_int32_t user) +{ + /* If multiplying would overflow... */ + if (user > 0xFFFFFFFF / (HZ*CREDITS_PER_JIFFY)) + /* Divide first. */ + return (user / XT_HASHLIMIT_SCALE) * HZ * CREDITS_PER_JIFFY; + + return (user * HZ * CREDITS_PER_JIFFY) / XT_HASHLIMIT_SCALE; +} + +static inline void rateinfo_recalc(struct dsthash_ent *dh, unsigned long now) +{ + dh->rateinfo.credit += (now - dh->rateinfo.prev) * CREDITS_PER_JIFFY; + if (dh->rateinfo.credit > dh->rateinfo.credit_cap) + dh->rateinfo.credit = dh->rateinfo.credit_cap; + dh->rateinfo.prev = now; +} + +static int +hashlimit_init_dst(struct xt_hashlimit_htable *hinfo, struct dsthash_dst *dst, + const struct sk_buff *skb, unsigned int protoff) +{ + __be16 _ports[2], *ports; + int nexthdr; + + memset(dst, 0, sizeof(*dst)); + + switch (hinfo->family) { + case AF_INET: + if (hinfo->cfg.mode & XT_HASHLIMIT_HASH_DIP) + dst->addr.ip.dst = skb->nh.iph->daddr; + if (hinfo->cfg.mode & XT_HASHLIMIT_HASH_SIP) + dst->addr.ip.src = skb->nh.iph->saddr; + + if (!(hinfo->cfg.mode & + (XT_HASHLIMIT_HASH_DPT | XT_HASHLIMIT_HASH_SPT))) + return 0; + nexthdr = skb->nh.iph->protocol; + break; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + case AF_INET6: + if (hinfo->cfg.mode & XT_HASHLIMIT_HASH_DIP) + memcpy(&dst->addr.ip6.dst, &skb->nh.ipv6h->daddr, + sizeof(dst->addr.ip6.dst)); + if (hinfo->cfg.mode & XT_HASHLIMIT_HASH_SIP) + memcpy(&dst->addr.ip6.src, &skb->nh.ipv6h->saddr, + sizeof(dst->addr.ip6.src)); + + if (!(hinfo->cfg.mode & + (XT_HASHLIMIT_HASH_DPT | XT_HASHLIMIT_HASH_SPT))) + return 0; + nexthdr = ipv6_find_hdr(skb, &protoff, -1, NULL); + if (nexthdr < 0) + return -1; + break; +#endif + default: + BUG(); + return 0; + } + + switch (nexthdr) { + case IPPROTO_TCP: + case IPPROTO_UDP: + case IPPROTO_SCTP: + case IPPROTO_DCCP: + ports = skb_header_pointer(skb, protoff, sizeof(_ports), + &_ports); + break; + default: + _ports[0] = _ports[1] = 0; + ports = _ports; + break; + } + if (!ports) + return -1; + if (hinfo->cfg.mode & XT_HASHLIMIT_HASH_SPT) + dst->src_port = ports[0]; + if (hinfo->cfg.mode & XT_HASHLIMIT_HASH_DPT) + dst->dst_port = ports[1]; + return 0; +} + +static int +hashlimit_match(const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const struct xt_match *match, + const void *matchinfo, + int offset, + unsigned int protoff, + int *hotdrop) +{ + struct xt_hashlimit_info *r = + ((struct xt_hashlimit_info *)matchinfo)->u.master; + struct xt_hashlimit_htable *hinfo = r->hinfo; + unsigned long now = jiffies; + struct dsthash_ent *dh; + struct dsthash_dst dst; + + if (hashlimit_init_dst(hinfo, &dst, skb, protoff) < 0) + goto hotdrop; + + spin_lock_bh(&hinfo->lock); + dh = dsthash_find(hinfo, &dst); + if (!dh) { + dh = dsthash_alloc_init(hinfo, &dst); + if (!dh) { + spin_unlock_bh(&hinfo->lock); + goto hotdrop; + } + + dh->expires = jiffies + msecs_to_jiffies(hinfo->cfg.expire); + dh->rateinfo.prev = jiffies; + dh->rateinfo.credit = user2credits(hinfo->cfg.avg * + hinfo->cfg.burst); + dh->rateinfo.credit_cap = user2credits(hinfo->cfg.avg * + hinfo->cfg.burst); + dh->rateinfo.cost = user2credits(hinfo->cfg.avg); + } else { + /* update expiration timeout */ + dh->expires = now + msecs_to_jiffies(hinfo->cfg.expire); + rateinfo_recalc(dh, now); + } + + if (dh->rateinfo.credit >= dh->rateinfo.cost) { + /* We're underlimit. */ + dh->rateinfo.credit -= dh->rateinfo.cost; + spin_unlock_bh(&hinfo->lock); + return 1; + } + + spin_unlock_bh(&hinfo->lock); + + /* default case: we're overlimit, thus don't match */ + return 0; + +hotdrop: + *hotdrop = 1; + return 0; +} + +static int +hashlimit_checkentry(const char *tablename, + const void *inf, + const struct xt_match *match, + void *matchinfo, + unsigned int hook_mask) +{ + struct xt_hashlimit_info *r = matchinfo; + + /* Check for overflow. */ + if (r->cfg.burst == 0 || + user2credits(r->cfg.avg * r->cfg.burst) < user2credits(r->cfg.avg)) { + printk(KERN_ERR "xt_hashlimit: overflow, try lower: %u/%u\n", + r->cfg.avg, r->cfg.burst); + return 0; + } + if (r->cfg.mode == 0 || + r->cfg.mode > (XT_HASHLIMIT_HASH_DPT | + XT_HASHLIMIT_HASH_DIP | + XT_HASHLIMIT_HASH_SIP | + XT_HASHLIMIT_HASH_SPT)) + return 0; + if (!r->cfg.gc_interval) + return 0; + if (!r->cfg.expire) + return 0; + if (r->name[sizeof(r->name) - 1] != '\0') + return 0; + + /* This is the best we've got: We cannot release and re-grab lock, + * since checkentry() is called before x_tables.c grabs xt_mutex. + * We also cannot grab the hashtable spinlock, since htable_create will + * call vmalloc, and that can sleep. And we cannot just re-search + * the list of htable's in htable_create(), since then we would + * create duplicate proc files. -HW */ + mutex_lock(&hlimit_mutex); + r->hinfo = htable_find_get(r->name, match->family); + if (!r->hinfo && htable_create(r, match->family) != 0) { + mutex_unlock(&hlimit_mutex); + return 0; + } + mutex_unlock(&hlimit_mutex); + + /* Ugly hack: For SMP, we only want to use one set */ + r->u.master = r; + return 1; +} + +static void +hashlimit_destroy(const struct xt_match *match, void *matchinfo) +{ + struct xt_hashlimit_info *r = matchinfo; + + htable_put(r->hinfo); +} + +#ifdef CONFIG_COMPAT +struct compat_xt_hashlimit_info { + char name[IFNAMSIZ]; + struct hashlimit_cfg cfg; + compat_uptr_t hinfo; + compat_uptr_t master; +}; + +static void compat_from_user(void *dst, void *src) +{ + int off = offsetof(struct compat_xt_hashlimit_info, hinfo); + + memcpy(dst, src, off); + memset(dst + off, 0, sizeof(struct compat_xt_hashlimit_info) - off); +} + +static int compat_to_user(void __user *dst, void *src) +{ + int off = offsetof(struct compat_xt_hashlimit_info, hinfo); + + return copy_to_user(dst, src, off) ? -EFAULT : 0; +} +#endif + +static struct xt_match xt_hashlimit[] = { + { + .name = "hashlimit", + .family = AF_INET, + .match = hashlimit_match, + .matchsize = sizeof(struct xt_hashlimit_info), +#ifdef CONFIG_COMPAT + .compatsize = sizeof(struct compat_xt_hashlimit_info), + .compat_from_user = compat_from_user, + .compat_to_user = compat_to_user, +#endif + .checkentry = hashlimit_checkentry, + .destroy = hashlimit_destroy, + .me = THIS_MODULE + }, + { + .name = "hashlimit", + .family = AF_INET6, + .match = hashlimit_match, + .matchsize = sizeof(struct xt_hashlimit_info), +#ifdef CONFIG_COMPAT + .compatsize = sizeof(struct compat_xt_hashlimit_info), + .compat_from_user = compat_from_user, + .compat_to_user = compat_to_user, +#endif + .checkentry = hashlimit_checkentry, + .destroy = hashlimit_destroy, + .me = THIS_MODULE + }, +}; + +/* PROC stuff */ +static void *dl_seq_start(struct seq_file *s, loff_t *pos) +{ + struct proc_dir_entry *pde = s->private; + struct xt_hashlimit_htable *htable = pde->data; + unsigned int *bucket; + + spin_lock_bh(&htable->lock); + if (*pos >= htable->cfg.size) + return NULL; + + bucket = kmalloc(sizeof(unsigned int), GFP_ATOMIC); + if (!bucket) + return ERR_PTR(-ENOMEM); + + *bucket = *pos; + return bucket; +} + +static void *dl_seq_next(struct seq_file *s, void *v, loff_t *pos) +{ + struct proc_dir_entry *pde = s->private; + struct xt_hashlimit_htable *htable = pde->data; + unsigned int *bucket = (unsigned int *)v; + + *pos = ++(*bucket); + if (*pos >= htable->cfg.size) { + kfree(v); + return NULL; + } + return bucket; +} + +static void dl_seq_stop(struct seq_file *s, void *v) +{ + struct proc_dir_entry *pde = s->private; + struct xt_hashlimit_htable *htable = pde->data; + unsigned int *bucket = (unsigned int *)v; + + kfree(bucket); + spin_unlock_bh(&htable->lock); +} + +static int dl_seq_real_show(struct dsthash_ent *ent, int family, + struct seq_file *s) +{ + /* recalculate to show accurate numbers */ + rateinfo_recalc(ent, jiffies); + + switch (family) { + case AF_INET: + return seq_printf(s, "%ld %u.%u.%u.%u:%u->" + "%u.%u.%u.%u:%u %u %u %u\n", + (long)(ent->expires - jiffies)/HZ, + NIPQUAD(ent->dst.addr.ip.src), + ntohs(ent->dst.src_port), + NIPQUAD(ent->dst.addr.ip.dst), + ntohs(ent->dst.dst_port), + ent->rateinfo.credit, ent->rateinfo.credit_cap, + ent->rateinfo.cost); + case AF_INET6: + return seq_printf(s, "%ld " NIP6_FMT ":%u->" + NIP6_FMT ":%u %u %u %u\n", + (long)(ent->expires - jiffies)/HZ, + NIP6(*(struct in6_addr *)&ent->dst.addr.ip6.src), + ntohs(ent->dst.src_port), + NIP6(*(struct in6_addr *)&ent->dst.addr.ip6.dst), + ntohs(ent->dst.dst_port), + ent->rateinfo.credit, ent->rateinfo.credit_cap, + ent->rateinfo.cost); + default: + BUG(); + return 0; + } +} + +static int dl_seq_show(struct seq_file *s, void *v) +{ + struct proc_dir_entry *pde = s->private; + struct xt_hashlimit_htable *htable = pde->data; + unsigned int *bucket = (unsigned int *)v; + struct dsthash_ent *ent; + struct hlist_node *pos; + + if (!hlist_empty(&htable->hash[*bucket])) { + hlist_for_each_entry(ent, pos, &htable->hash[*bucket], node) + if (dl_seq_real_show(ent, htable->family, s)) + return 1; + } + return 0; +} + +static struct seq_operations dl_seq_ops = { + .start = dl_seq_start, + .next = dl_seq_next, + .stop = dl_seq_stop, + .show = dl_seq_show +}; + +static int dl_proc_open(struct inode *inode, struct file *file) +{ + int ret = seq_open(file, &dl_seq_ops); + + if (!ret) { + struct seq_file *sf = file->private_data; + sf->private = PDE(inode); + } + return ret; +} + +static struct file_operations dl_file_ops = { + .owner = THIS_MODULE, + .open = dl_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release +}; + +static int __init xt_hashlimit_init(void) +{ + int err; + + err = xt_register_matches(xt_hashlimit, ARRAY_SIZE(xt_hashlimit)); + if (err < 0) + goto err1; + + err = -ENOMEM; + hashlimit_cachep = kmem_cache_create("xt_hashlimit", + sizeof(struct dsthash_ent), 0, 0, + NULL, NULL); + if (!hashlimit_cachep) { + printk(KERN_ERR "xt_hashlimit: unable to create slab cache\n"); + goto err2; + } + hashlimit_procdir4 = proc_mkdir("ipt_hashlimit", proc_net); + if (!hashlimit_procdir4) { + printk(KERN_ERR "xt_hashlimit: unable to create proc dir " + "entry\n"); + goto err3; + } + hashlimit_procdir6 = proc_mkdir("ip6t_hashlimit", proc_net); + if (!hashlimit_procdir6) { + printk(KERN_ERR "xt_hashlimit: tnable to create proc dir " + "entry\n"); + goto err4; + } + return 0; +err4: + remove_proc_entry("ipt_hashlimit", proc_net); +err3: + kmem_cache_destroy(hashlimit_cachep); +err2: + xt_unregister_matches(xt_hashlimit, ARRAY_SIZE(xt_hashlimit)); +err1: + return err; + +} + +static void __exit xt_hashlimit_fini(void) +{ + remove_proc_entry("ipt_hashlimit", proc_net); + remove_proc_entry("ip6t_hashlimit", proc_net); + kmem_cache_destroy(hashlimit_cachep); + xt_unregister_matches(xt_hashlimit, ARRAY_SIZE(xt_hashlimit)); +} + +module_init(xt_hashlimit_init); +module_exit(xt_hashlimit_fini); -- cgit v1.2.3-70-g09d2 From baf7b1e11282127e068d149825cccec002091d61 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 29 Nov 2006 02:35:38 +0100 Subject: [NETFILTER]: x_tables: add NFLOG target Add new NFLOG target to allow use of nfnetlink_log for both IPv4 and IPv6. Currently we have two (unsupported by userspace) hacks in the LOG and ULOG targets to optionally call to the nflog API. They lack a few features, namely the IPv4 and IPv6 LOG targets can not specify a number of arguments related to nfnetlink_log, while the ULOG target is only available for IPv4. Remove those hacks and add a clean way to use nfnetlink_log. Signed-off-by: Patrick McHardy --- include/linux/netfilter/Kbuild | 1 + include/linux/netfilter/xt_NFLOG.h | 18 +++++++ include/linux/netfilter_ipv4/ipt_LOG.h | 2 +- include/linux/netfilter_ipv6/ip6t_LOG.h | 2 +- net/ipv4/netfilter/ipt_LOG.c | 9 +--- net/ipv6/netfilter/ip6t_LOG.c | 9 +--- net/netfilter/Kconfig | 11 +++++ net/netfilter/Makefile | 1 + net/netfilter/xt_NFLOG.c | 86 +++++++++++++++++++++++++++++++++ 9 files changed, 123 insertions(+), 16 deletions(-) create mode 100644 include/linux/netfilter/xt_NFLOG.h create mode 100644 net/netfilter/xt_NFLOG.c (limited to 'include') diff --git a/include/linux/netfilter/Kbuild b/include/linux/netfilter/Kbuild index e379a2d89ea..6328175a1c3 100644 --- a/include/linux/netfilter/Kbuild +++ b/include/linux/netfilter/Kbuild @@ -22,6 +22,7 @@ header-y += xt_mark.h header-y += xt_MARK.h header-y += xt_multiport.h header-y += xt_NFQUEUE.h +header-y += xt_NFLOG.h header-y += xt_pkttype.h header-y += xt_policy.h header-y += xt_realm.h diff --git a/include/linux/netfilter/xt_NFLOG.h b/include/linux/netfilter/xt_NFLOG.h new file mode 100644 index 00000000000..cdcd0ed58f7 --- /dev/null +++ b/include/linux/netfilter/xt_NFLOG.h @@ -0,0 +1,18 @@ +#ifndef _XT_NFLOG_TARGET +#define _XT_NFLOG_TARGET + +#define XT_NFLOG_DEFAULT_GROUP 0x1 +#define XT_NFLOG_DEFAULT_THRESHOLD 1 + +#define XT_NFLOG_MASK 0x0 + +struct xt_nflog_info { + u_int32_t len; + u_int16_t group; + u_int16_t threshold; + u_int16_t flags; + u_int16_t pad; + char prefix[64]; +}; + +#endif /* _XT_NFLOG_TARGET */ diff --git a/include/linux/netfilter_ipv4/ipt_LOG.h b/include/linux/netfilter_ipv4/ipt_LOG.h index 892f9a33fea..90fa6525ef9 100644 --- a/include/linux/netfilter_ipv4/ipt_LOG.h +++ b/include/linux/netfilter_ipv4/ipt_LOG.h @@ -6,7 +6,7 @@ #define IPT_LOG_TCPOPT 0x02 /* Log TCP options */ #define IPT_LOG_IPOPT 0x04 /* Log IP options */ #define IPT_LOG_UID 0x08 /* Log UID owning local socket */ -#define IPT_LOG_NFLOG 0x10 /* Log using nf_log backend */ +#define IPT_LOG_NFLOG 0x10 /* Unsupported, don't reuse */ #define IPT_LOG_MASK 0x1f struct ipt_log_info { diff --git a/include/linux/netfilter_ipv6/ip6t_LOG.h b/include/linux/netfilter_ipv6/ip6t_LOG.h index 060c1a1c6c6..0d0119b0458 100644 --- a/include/linux/netfilter_ipv6/ip6t_LOG.h +++ b/include/linux/netfilter_ipv6/ip6t_LOG.h @@ -6,7 +6,7 @@ #define IP6T_LOG_TCPOPT 0x02 /* Log TCP options */ #define IP6T_LOG_IPOPT 0x04 /* Log IP options */ #define IP6T_LOG_UID 0x08 /* Log UID owning local socket */ -#define IP6T_LOG_NFLOG 0x10 /* Log using nf_log backend */ +#define IP6T_LOG_NFLOG 0x10 /* Unsupported, don't use */ #define IP6T_LOG_MASK 0x1f struct ip6t_log_info { diff --git a/net/ipv4/netfilter/ipt_LOG.c b/net/ipv4/netfilter/ipt_LOG.c index 46eee64a11f..c96de16fefa 100644 --- a/net/ipv4/netfilter/ipt_LOG.c +++ b/net/ipv4/netfilter/ipt_LOG.c @@ -430,13 +430,8 @@ ipt_log_target(struct sk_buff **pskb, li.u.log.level = loginfo->level; li.u.log.logflags = loginfo->logflags; - if (loginfo->logflags & IPT_LOG_NFLOG) - nf_log_packet(PF_INET, hooknum, *pskb, in, out, &li, - "%s", loginfo->prefix); - else - ipt_log_packet(PF_INET, hooknum, *pskb, in, out, &li, - loginfo->prefix); - + ipt_log_packet(PF_INET, hooknum, *pskb, in, out, &li, + loginfo->prefix); return IPT_CONTINUE; } diff --git a/net/ipv6/netfilter/ip6t_LOG.c b/net/ipv6/netfilter/ip6t_LOG.c index f4857cf97f0..33b1faa90d7 100644 --- a/net/ipv6/netfilter/ip6t_LOG.c +++ b/net/ipv6/netfilter/ip6t_LOG.c @@ -440,13 +440,8 @@ ip6t_log_target(struct sk_buff **pskb, li.u.log.level = loginfo->level; li.u.log.logflags = loginfo->logflags; - if (loginfo->logflags & IP6T_LOG_NFLOG) - nf_log_packet(PF_INET6, hooknum, *pskb, in, out, &li, - "%s", loginfo->prefix); - else - ip6t_log_packet(PF_INET6, hooknum, *pskb, in, out, &li, - loginfo->prefix); - + ip6t_log_packet(PF_INET6, hooknum, *pskb, in, out, &li, + loginfo->prefix); return IP6T_CONTINUE; } diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 7e6125467c1..d191dacead5 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -184,6 +184,17 @@ config NETFILTER_XT_TARGET_NFQUEUE To compile it as a module, choose M here. If unsure, say N. +config NETFILTER_XT_TARGET_NFLOG + tristate '"NFLOG" target support' + depends on NETFILTER_XTABLES + help + This option enables the NFLOG target, which allows to LOG + messages through the netfilter logging API, which can use + either the old LOG target, the old ULOG target or nfnetlink_log + as backend. + + To compile it as a module, choose M here. If unsure, say N. + config NETFILTER_XT_TARGET_NOTRACK tristate '"NOTRACK" target support' depends on NETFILTER_XTABLES diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index f85811bfcfe..7f0089c584b 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -31,6 +31,7 @@ obj-$(CONFIG_NETFILTER_XT_TARGET_CONNMARK) += xt_CONNMARK.o obj-$(CONFIG_NETFILTER_XT_TARGET_DSCP) += xt_DSCP.o obj-$(CONFIG_NETFILTER_XT_TARGET_MARK) += xt_MARK.o obj-$(CONFIG_NETFILTER_XT_TARGET_NFQUEUE) += xt_NFQUEUE.o +obj-$(CONFIG_NETFILTER_XT_TARGET_NFLOG) += xt_NFLOG.o obj-$(CONFIG_NETFILTER_XT_TARGET_NOTRACK) += xt_NOTRACK.o obj-$(CONFIG_NETFILTER_XT_TARGET_SECMARK) += xt_SECMARK.o obj-$(CONFIG_NETFILTER_XT_TARGET_CONNSECMARK) += xt_CONNSECMARK.o diff --git a/net/netfilter/xt_NFLOG.c b/net/netfilter/xt_NFLOG.c new file mode 100644 index 00000000000..901ed7abaa1 --- /dev/null +++ b/net/netfilter/xt_NFLOG.c @@ -0,0 +1,86 @@ +/* + * Copyright (c) 2006 Patrick McHardy + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include + +#include +#include + +MODULE_AUTHOR("Patrick McHardy "); +MODULE_DESCRIPTION("x_tables NFLOG target"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ipt_NFLOG"); +MODULE_ALIAS("ip6t_NFLOG"); + +static unsigned int +nflog_target(struct sk_buff **pskb, + const struct net_device *in, const struct net_device *out, + unsigned int hooknum, const struct xt_target *target, + const void *targinfo) +{ + const struct xt_nflog_info *info = targinfo; + struct nf_loginfo li; + + li.type = NF_LOG_TYPE_ULOG; + li.u.ulog.copy_len = info->len; + li.u.ulog.group = info->group; + li.u.ulog.qthreshold = info->threshold; + + nf_log_packet(target->family, hooknum, *pskb, in, out, &li, + "%s", info->prefix); + return XT_CONTINUE; +} + +static int +nflog_checkentry(const char *tablename, const void *entry, + const struct xt_target *target, void *targetinfo, + unsigned int hookmask) +{ + struct xt_nflog_info *info = targetinfo; + + if (info->flags & ~XT_NFLOG_MASK) + return 0; + if (info->prefix[sizeof(info->prefix) - 1] != '\0') + return 0; + return 1; +} + +static struct xt_target xt_nflog_target[] = { + { + .name = "NFLOG", + .family = AF_INET, + .checkentry = nflog_checkentry, + .target = nflog_target, + .targetsize = sizeof(struct xt_nflog_info), + .me = THIS_MODULE, + }, + { + .name = "NFLOG", + .family = AF_INET6, + .checkentry = nflog_checkentry, + .target = nflog_target, + .targetsize = sizeof(struct xt_nflog_info), + .me = THIS_MODULE, + }, +}; + +static int __init xt_nflog_init(void) +{ + return xt_register_targets(xt_nflog_target, + ARRAY_SIZE(xt_nflog_target)); +} + +static void __exit xt_nflog_fini(void) +{ + xt_unregister_targets(xt_nflog_target, ARRAY_SIZE(xt_nflog_target)); +} + +module_init(xt_nflog_init); +module_exit(xt_nflog_fini); -- cgit v1.2.3-70-g09d2 From d12cdc3ccf140bd2febef1c1be92284571da983f Mon Sep 17 00:00:00 2001 From: Bart De Schuymer Date: Wed, 29 Nov 2006 02:35:40 +0100 Subject: [NETFILTER]: ebtables: add --snap-arp option The attached patch adds --snat-arp support, which makes it possible to change the source mac address in both the mac header and the arp header with one rule. Signed-off-by: Bart De Schuymer Signed-off-by: Patrick McHardy --- include/linux/netfilter_bridge/ebt_nat.h | 1 + include/linux/netfilter_bridge/ebtables.h | 4 ++++ net/bridge/netfilter/ebt_mark.c | 6 +++--- net/bridge/netfilter/ebt_snat.c | 27 ++++++++++++++++++++++++--- 4 files changed, 32 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter_bridge/ebt_nat.h b/include/linux/netfilter_bridge/ebt_nat.h index 26fd90da4cd..435b886a51a 100644 --- a/include/linux/netfilter_bridge/ebt_nat.h +++ b/include/linux/netfilter_bridge/ebt_nat.h @@ -1,6 +1,7 @@ #ifndef __LINUX_BRIDGE_EBT_NAT_H #define __LINUX_BRIDGE_EBT_NAT_H +#define NAT_ARP_BIT (0x00000010) struct ebt_nat_info { unsigned char mac[ETH_ALEN]; diff --git a/include/linux/netfilter_bridge/ebtables.h b/include/linux/netfilter_bridge/ebtables.h index e6ea70de24d..87775264ff0 100644 --- a/include/linux/netfilter_bridge/ebtables.h +++ b/include/linux/netfilter_bridge/ebtables.h @@ -26,6 +26,10 @@ #define EBT_CONTINUE -3 #define EBT_RETURN -4 #define NUM_STANDARD_TARGETS 4 +/* ebtables target modules store the verdict inside an int. We can + * reclaim a part of this int for backwards compatible extensions. + * The 4 lsb are more than enough to store the verdict. */ +#define EBT_VERDICT_BITS 0x0000000F struct ebt_counter { diff --git a/net/bridge/netfilter/ebt_mark.c b/net/bridge/netfilter/ebt_mark.c index 2458638561c..62d23c7b25e 100644 --- a/net/bridge/netfilter/ebt_mark.c +++ b/net/bridge/netfilter/ebt_mark.c @@ -33,7 +33,7 @@ static int ebt_target_mark(struct sk_buff **pskb, unsigned int hooknr, else (*pskb)->mark ^= info->mark; - return info->target | -16; + return info->target | ~EBT_VERDICT_BITS; } static int ebt_target_mark_check(const char *tablename, unsigned int hookmask, @@ -44,13 +44,13 @@ static int ebt_target_mark_check(const char *tablename, unsigned int hookmask, if (datalen != EBT_ALIGN(sizeof(struct ebt_mark_t_info))) return -EINVAL; - tmp = info->target | -16; + tmp = info->target | ~EBT_VERDICT_BITS; if (BASE_CHAIN && tmp == EBT_RETURN) return -EINVAL; CLEAR_BASE_CHAIN_BIT; if (tmp < -NUM_STANDARD_TARGETS || tmp >= 0) return -EINVAL; - tmp = info->target & -16; + tmp = info->target & ~EBT_VERDICT_BITS; if (tmp != MARK_SET_VALUE && tmp != MARK_OR_VALUE && tmp != MARK_AND_VALUE && tmp != MARK_XOR_VALUE) return -EINVAL; diff --git a/net/bridge/netfilter/ebt_snat.c b/net/bridge/netfilter/ebt_snat.c index cbb33e24ca8..a50722182bf 100644 --- a/net/bridge/netfilter/ebt_snat.c +++ b/net/bridge/netfilter/ebt_snat.c @@ -12,6 +12,8 @@ #include #include #include +#include +#include static int ebt_target_snat(struct sk_buff **pskb, unsigned int hooknr, const struct net_device *in, const struct net_device *out, @@ -31,24 +33,43 @@ static int ebt_target_snat(struct sk_buff **pskb, unsigned int hooknr, *pskb = nskb; } memcpy(eth_hdr(*pskb)->h_source, info->mac, ETH_ALEN); - return info->target; + if (!(info->target & NAT_ARP_BIT) && + eth_hdr(*pskb)->h_proto == htons(ETH_P_ARP)) { + struct arphdr _ah, *ap; + + ap = skb_header_pointer(*pskb, 0, sizeof(_ah), &_ah); + if (ap == NULL) + return EBT_DROP; + if (ap->ar_hln != ETH_ALEN) + goto out; + if (skb_store_bits(*pskb, sizeof(_ah), info->mac,ETH_ALEN)) + return EBT_DROP; + } +out: + return info->target | ~EBT_VERDICT_BITS; } static int ebt_target_snat_check(const char *tablename, unsigned int hookmask, const struct ebt_entry *e, void *data, unsigned int datalen) { struct ebt_nat_info *info = (struct ebt_nat_info *) data; + int tmp; if (datalen != EBT_ALIGN(sizeof(struct ebt_nat_info))) return -EINVAL; - if (BASE_CHAIN && info->target == EBT_RETURN) + tmp = info->target | ~EBT_VERDICT_BITS; + if (BASE_CHAIN && tmp == EBT_RETURN) return -EINVAL; CLEAR_BASE_CHAIN_BIT; if (strcmp(tablename, "nat")) return -EINVAL; if (hookmask & ~(1 << NF_BR_POST_ROUTING)) return -EINVAL; - if (INVALID_TARGET) + + if (tmp < -NUM_STANDARD_TARGETS || tmp >= 0) + return -EINVAL; + tmp = info->target | EBT_VERDICT_BITS; + if ((tmp & ~NAT_ARP_BIT) != ~NAT_ARP_BIT) return -EINVAL; return 0; } -- cgit v1.2.3-70-g09d2 From 02752760359db6b00a3ffb1acfc13ef8d9eb1e3f Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Wed, 29 Nov 2006 13:18:18 -0500 Subject: NetLabel: convert to an extensibile/sparse category bitmap The original NetLabel category bitmap was a straight char bitmap which worked fine for the initial release as it only supported 240 bits due to limitations in the CIPSO restricted bitmap tag (tag type 0x01). This patch converts that straight char bitmap into an extensibile/sparse bitmap in order to lay the foundation for other CIPSO tag types and protocols. This patch also has a nice side effect in that all of the security attributes passed by NetLabel into the LSM are now in a format which is in the host's native byte/bit ordering which makes the LSM specific code much simpler; look at the changes in security/selinux/ss/ebitmap.c as an example. Signed-off-by: Paul Moore Signed-off-by: James Morris --- include/net/netlabel.h | 102 ++++++++++++++++++++- net/ipv4/cipso_ipv4.c | 168 ++++++++++++++-------------------- net/netlabel/netlabel_kapi.c | 201 +++++++++++++++++++++++++++++++++++++++++ security/selinux/ss/ebitmap.c | 198 ++++++++++++++++++---------------------- security/selinux/ss/ebitmap.h | 26 ++++-- security/selinux/ss/mls.c | 156 +++++++++++--------------------- security/selinux/ss/mls.h | 46 +++++++--- security/selinux/ss/services.c | 23 ++--- 8 files changed, 568 insertions(+), 352 deletions(-) (limited to 'include') diff --git a/include/net/netlabel.h b/include/net/netlabel.h index d605d795401..83da7e1f0d3 100644 --- a/include/net/netlabel.h +++ b/include/net/netlabel.h @@ -111,6 +111,22 @@ struct netlbl_lsm_cache { void (*free) (const void *data); void *data; }; +/* The catmap bitmap field MUST be a power of two in length and large + * enough to hold at least 240 bits. Special care (i.e. check the code!) + * should be used when changing these values as the LSM implementation + * probably has functions which rely on the sizes of these types to speed + * processing. */ +#define NETLBL_CATMAP_MAPTYPE u64 +#define NETLBL_CATMAP_MAPCNT 4 +#define NETLBL_CATMAP_MAPSIZE (sizeof(NETLBL_CATMAP_MAPTYPE) * 8) +#define NETLBL_CATMAP_SIZE (NETLBL_CATMAP_MAPSIZE * \ + NETLBL_CATMAP_MAPCNT) +#define NETLBL_CATMAP_BIT (NETLBL_CATMAP_MAPTYPE)0x01 +struct netlbl_lsm_secattr_catmap { + u32 startbit; + NETLBL_CATMAP_MAPTYPE bitmap[NETLBL_CATMAP_MAPCNT]; + struct netlbl_lsm_secattr_catmap *next; +}; #define NETLBL_SECATTR_NONE 0x00000000 #define NETLBL_SECATTR_DOMAIN 0x00000001 #define NETLBL_SECATTR_CACHE 0x00000002 @@ -122,8 +138,7 @@ struct netlbl_lsm_secattr { char *domain; u32 mls_lvl; - unsigned char *mls_cat; - size_t mls_cat_len; + struct netlbl_lsm_secattr_catmap *mls_cat; struct netlbl_lsm_cache *cache; }; @@ -170,6 +185,41 @@ static inline void netlbl_secattr_cache_free(struct netlbl_lsm_cache *cache) kfree(cache); } +/** + * netlbl_secattr_catmap_alloc - Allocate a LSM secattr catmap + * @flags: memory allocation flags + * + * Description: + * Allocate memory for a LSM secattr catmap, returns a pointer on success, NULL + * on failure. + * + */ +static inline struct netlbl_lsm_secattr_catmap *netlbl_secattr_catmap_alloc( + gfp_t flags) +{ + return kzalloc(sizeof(struct netlbl_lsm_secattr_catmap), flags); +} + +/** + * netlbl_secattr_catmap_free - Free a LSM secattr catmap + * @catmap: the category bitmap + * + * Description: + * Free a LSM secattr catmap. + * + */ +static inline void netlbl_secattr_catmap_free( + struct netlbl_lsm_secattr_catmap *catmap) +{ + struct netlbl_lsm_secattr_catmap *iter; + + do { + iter = catmap; + catmap = catmap->next; + kfree(iter); + } while (catmap); +} + /** * netlbl_secattr_init - Initialize a netlbl_lsm_secattr struct * @secattr: the struct to initialize @@ -200,7 +250,8 @@ static inline void netlbl_secattr_destroy(struct netlbl_lsm_secattr *secattr) if (secattr->cache) netlbl_secattr_cache_free(secattr->cache); kfree(secattr->domain); - kfree(secattr->mls_cat); + if (secattr->mls_cat) + netlbl_secattr_catmap_free(secattr->mls_cat); } /** @@ -231,6 +282,51 @@ static inline void netlbl_secattr_free(struct netlbl_lsm_secattr *secattr) kfree(secattr); } +#ifdef CONFIG_NETLABEL +int netlbl_secattr_catmap_walk(struct netlbl_lsm_secattr_catmap *catmap, + u32 offset); +int netlbl_secattr_catmap_walk_rng(struct netlbl_lsm_secattr_catmap *catmap, + u32 offset); +int netlbl_secattr_catmap_setbit(struct netlbl_lsm_secattr_catmap *catmap, + u32 bit, + gfp_t flags); +int netlbl_secattr_catmap_setrng(struct netlbl_lsm_secattr_catmap *catmap, + u32 start, + u32 end, + gfp_t flags); +#else +static inline int netlbl_secattr_catmap_walk( + struct netlbl_lsm_secattr_catmap *catmap, + u32 offset) +{ + return -ENOENT; +} + +static inline int netlbl_secattr_catmap_walk_rng( + struct netlbl_lsm_secattr_catmap *catmap, + u32 offset) +{ + return -ENOENT; +} + +static inline int netlbl_secattr_catmap_setbit( + struct netlbl_lsm_secattr_catmap *catmap, + u32 bit, + gfp_t flags) +{ + return 0; +} + +static inline int netlbl_secattr_catmap_setrng( + struct netlbl_lsm_secattr_catmap *catmap, + u32 start, + u32 end, + gfp_t flags) +{ + return 0; +} +#endif + /* * LSM protocol operations */ diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c index f3957cfaed4..08144f8fc91 100644 --- a/net/ipv4/cipso_ipv4.c +++ b/net/ipv4/cipso_ipv4.c @@ -819,8 +819,7 @@ static int cipso_v4_map_cat_rbm_valid(const struct cipso_v4_doi *doi_def, /** * cipso_v4_map_cat_rbm_hton - Perform a category mapping from host to network * @doi_def: the DOI definition - * @host_cat: the category bitmap in host format - * @host_cat_len: the length of the host's category bitmap in bytes + * @secattr: the security attributes * @net_cat: the zero'd out category bitmap in network/CIPSO format * @net_cat_len: the length of the CIPSO bitmap in bytes * @@ -831,61 +830,51 @@ static int cipso_v4_map_cat_rbm_valid(const struct cipso_v4_doi *doi_def, * */ static int cipso_v4_map_cat_rbm_hton(const struct cipso_v4_doi *doi_def, - const unsigned char *host_cat, - u32 host_cat_len, + const struct netlbl_lsm_secattr *secattr, unsigned char *net_cat, u32 net_cat_len) { int host_spot = -1; - u32 net_spot; + u32 net_spot = CIPSO_V4_INV_CAT; u32 net_spot_max = 0; - u32 host_clen_bits = host_cat_len * 8; u32 net_clen_bits = net_cat_len * 8; - u32 host_cat_size; - u32 *host_cat_array; + u32 host_cat_size = 0; + u32 *host_cat_array = NULL; - switch (doi_def->type) { - case CIPSO_V4_MAP_PASS: - net_spot_max = host_cat_len; - while (net_spot_max > 0 && host_cat[net_spot_max - 1] == 0) - net_spot_max--; - if (net_spot_max > net_cat_len) - return -EINVAL; - memcpy(net_cat, host_cat, net_spot_max); - return net_spot_max; - case CIPSO_V4_MAP_STD: + if (doi_def->type == CIPSO_V4_MAP_STD) { host_cat_size = doi_def->map.std->cat.local_size; host_cat_array = doi_def->map.std->cat.local; - for (;;) { - host_spot = cipso_v4_bitmap_walk(host_cat, - host_clen_bits, - host_spot + 1, - 1); - if (host_spot < 0) - break; + } + + for (;;) { + host_spot = netlbl_secattr_catmap_walk(secattr->mls_cat, + host_spot + 1); + if (host_spot < 0) + break; + + switch (doi_def->type) { + case CIPSO_V4_MAP_PASS: + net_spot = host_spot; + break; + case CIPSO_V4_MAP_STD: if (host_spot >= host_cat_size) return -EPERM; - net_spot = host_cat_array[host_spot]; if (net_spot >= CIPSO_V4_INV_CAT) return -EPERM; - if (net_spot >= net_clen_bits) - return -ENOSPC; - cipso_v4_bitmap_setbit(net_cat, net_spot, 1); - - if (net_spot > net_spot_max) - net_spot_max = net_spot; + break; } + if (net_spot >= net_clen_bits) + return -ENOSPC; + cipso_v4_bitmap_setbit(net_cat, net_spot, 1); - if (host_spot == -2) - return -EFAULT; - - if (++net_spot_max % 8) - return net_spot_max / 8 + 1; - return net_spot_max / 8; + if (net_spot > net_spot_max) + net_spot_max = net_spot; } - return -EINVAL; + if (++net_spot_max % 8) + return net_spot_max / 8 + 1; + return net_spot_max / 8; } /** @@ -893,66 +882,59 @@ static int cipso_v4_map_cat_rbm_hton(const struct cipso_v4_doi *doi_def, * @doi_def: the DOI definition * @net_cat: the category bitmap in network/CIPSO format * @net_cat_len: the length of the CIPSO bitmap in bytes - * @host_cat: the zero'd out category bitmap in host format - * @host_cat_len: the length of the host's category bitmap in bytes + * @secattr: the security attributes * * Description: * Perform a label mapping to translate a CIPSO bitmap to the correct local - * MLS category bitmap using the given DOI definition. Returns the minimum - * size in bytes of the host bitmap on success, negative values otherwise. + * MLS category bitmap using the given DOI definition. Returns zero on + * success, negative values on failure. * */ static int cipso_v4_map_cat_rbm_ntoh(const struct cipso_v4_doi *doi_def, const unsigned char *net_cat, u32 net_cat_len, - unsigned char *host_cat, - u32 host_cat_len) + struct netlbl_lsm_secattr *secattr) { - u32 host_spot; - u32 host_spot_max = 0; + int ret_val; int net_spot = -1; + u32 host_spot = CIPSO_V4_INV_CAT; u32 net_clen_bits = net_cat_len * 8; - u32 host_clen_bits = host_cat_len * 8; - u32 net_cat_size; - u32 *net_cat_array; + u32 net_cat_size = 0; + u32 *net_cat_array = NULL; - switch (doi_def->type) { - case CIPSO_V4_MAP_PASS: - if (net_cat_len > host_cat_len) - return -EINVAL; - memcpy(host_cat, net_cat, net_cat_len); - return net_cat_len; - case CIPSO_V4_MAP_STD: + if (doi_def->type == CIPSO_V4_MAP_STD) { net_cat_size = doi_def->map.std->cat.cipso_size; net_cat_array = doi_def->map.std->cat.cipso; - for (;;) { - net_spot = cipso_v4_bitmap_walk(net_cat, - net_clen_bits, - net_spot + 1, - 1); - if (net_spot < 0) - break; - if (net_spot >= net_cat_size || - net_cat_array[net_spot] >= CIPSO_V4_INV_CAT) - return -EPERM; + } + for (;;) { + net_spot = cipso_v4_bitmap_walk(net_cat, + net_clen_bits, + net_spot + 1, + 1); + if (net_spot < 0) { + if (net_spot == -2) + return -EFAULT; + return 0; + } + + switch (doi_def->type) { + case CIPSO_V4_MAP_PASS: + host_spot = net_spot; + break; + case CIPSO_V4_MAP_STD: + if (net_spot >= net_cat_size) + return -EPERM; host_spot = net_cat_array[net_spot]; if (host_spot >= CIPSO_V4_INV_CAT) return -EPERM; - if (host_spot >= host_clen_bits) - return -ENOSPC; - cipso_v4_bitmap_setbit(host_cat, host_spot, 1); - - if (host_spot > host_spot_max) - host_spot_max = host_spot; + break; } - - if (net_spot == -2) - return -EFAULT; - - if (++host_spot_max % 8) - return host_spot_max / 8 + 1; - return host_spot_max / 8; + ret_val = netlbl_secattr_catmap_setbit(secattr->mls_cat, + host_spot, + GFP_ATOMIC); + if (ret_val != 0) + return ret_val; } return -EINVAL; @@ -1016,8 +998,7 @@ static int cipso_v4_gentag_rbm(const struct cipso_v4_doi *doi_def, if (secattr->flags & NETLBL_SECATTR_MLS_CAT) { ret_val = cipso_v4_map_cat_rbm_hton(doi_def, - secattr->mls_cat, - secattr->mls_cat_len, + secattr, &buffer[4], buffer_len - 4); if (ret_val < 0) @@ -1067,31 +1048,20 @@ static int cipso_v4_parsetag_rbm(const struct cipso_v4_doi *doi_def, secattr->flags |= NETLBL_SECATTR_MLS_LVL; if (tag_len > 4) { - switch (doi_def->type) { - case CIPSO_V4_MAP_PASS: - secattr->mls_cat_len = tag_len - 4; - break; - case CIPSO_V4_MAP_STD: - secattr->mls_cat_len = - doi_def->map.std->cat.local_size; - break; - } - secattr->mls_cat = kzalloc(secattr->mls_cat_len, GFP_ATOMIC); + secattr->mls_cat = netlbl_secattr_catmap_alloc(GFP_ATOMIC); if (secattr->mls_cat == NULL) return -ENOMEM; ret_val = cipso_v4_map_cat_rbm_ntoh(doi_def, &tag[4], tag_len - 4, - secattr->mls_cat, - secattr->mls_cat_len); - if (ret_val < 0) { - kfree(secattr->mls_cat); + secattr); + if (ret_val != 0) { + netlbl_secattr_catmap_free(secattr->mls_cat); return ret_val; - } else if (ret_val > 0) { - secattr->mls_cat_len = ret_val; - secattr->flags |= NETLBL_SECATTR_MLS_CAT; } + + secattr->flags |= NETLBL_SECATTR_MLS_CAT; } return 0; diff --git a/net/netlabel/netlabel_kapi.c b/net/netlabel/netlabel_kapi.c index b35ebf94918..e03a3282c55 100644 --- a/net/netlabel/netlabel_kapi.c +++ b/net/netlabel/netlabel_kapi.c @@ -39,6 +39,207 @@ #include "netlabel_unlabeled.h" #include "netlabel_user.h" +/* + * Security Attribute Functions + */ + +/** + * netlbl_secattr_catmap_walk - Walk a LSM secattr catmap looking for a bit + * @catmap: the category bitmap + * @offset: the offset to start searching at, in bits + * + * Description: + * This function walks a LSM secattr category bitmap starting at @offset and + * returns the spot of the first set bit or -ENOENT if no bits are set. + * + */ +int netlbl_secattr_catmap_walk(struct netlbl_lsm_secattr_catmap *catmap, + u32 offset) +{ + struct netlbl_lsm_secattr_catmap *iter = catmap; + u32 node_idx; + u32 node_bit; + NETLBL_CATMAP_MAPTYPE bitmap; + + if (offset > iter->startbit) { + while (offset >= (iter->startbit + NETLBL_CATMAP_SIZE)) { + iter = iter->next; + if (iter == NULL) + return -ENOENT; + } + node_idx = (offset - iter->startbit) / NETLBL_CATMAP_MAPSIZE; + node_bit = offset - iter->startbit - + (NETLBL_CATMAP_MAPSIZE * node_idx); + } else { + node_idx = 0; + node_bit = 0; + } + bitmap = iter->bitmap[node_idx] >> node_bit; + + for (;;) { + if (bitmap != 0) { + while ((bitmap & NETLBL_CATMAP_BIT) == 0) { + bitmap >>= 1; + node_bit++; + } + return iter->startbit + + (NETLBL_CATMAP_MAPSIZE * node_idx) + node_bit; + } + if (++node_idx >= NETLBL_CATMAP_MAPCNT) { + if (iter->next != NULL) { + iter = iter->next; + node_idx = 0; + } else + return -ENOENT; + } + bitmap = iter->bitmap[node_idx]; + node_bit = 0; + } + + return -ENOENT; +} + +/** + * netlbl_secattr_catmap_walk_rng - Find the end of a string of set bits + * @catmap: the category bitmap + * @offset: the offset to start searching at, in bits + * + * Description: + * This function walks a LSM secattr category bitmap starting at @offset and + * returns the spot of the first cleared bit or -ENOENT if the offset is past + * the end of the bitmap. + * + */ +int netlbl_secattr_catmap_walk_rng(struct netlbl_lsm_secattr_catmap *catmap, + u32 offset) +{ + struct netlbl_lsm_secattr_catmap *iter = catmap; + u32 node_idx; + u32 node_bit; + NETLBL_CATMAP_MAPTYPE bitmask; + NETLBL_CATMAP_MAPTYPE bitmap; + + if (offset > iter->startbit) { + while (offset >= (iter->startbit + NETLBL_CATMAP_SIZE)) { + iter = iter->next; + if (iter == NULL) + return -ENOENT; + } + node_idx = (offset - iter->startbit) / NETLBL_CATMAP_MAPSIZE; + node_bit = offset - iter->startbit - + (NETLBL_CATMAP_MAPSIZE * node_idx); + } else { + node_idx = 0; + node_bit = 0; + } + bitmask = NETLBL_CATMAP_BIT << node_bit; + + for (;;) { + bitmap = iter->bitmap[node_idx]; + while (bitmask != 0 && (bitmap & bitmask) != 0) { + bitmask <<= 1; + node_bit++; + } + + if (bitmask != 0) + return iter->startbit + + (NETLBL_CATMAP_MAPSIZE * node_idx) + + node_bit - 1; + else if (++node_idx >= NETLBL_CATMAP_MAPCNT) { + if (iter->next == NULL) + return iter->startbit + NETLBL_CATMAP_SIZE - 1; + iter = iter->next; + node_idx = 0; + } + bitmask = NETLBL_CATMAP_BIT; + node_bit = 0; + } + + return -ENOENT; +} + +/** + * netlbl_secattr_catmap_setbit - Set a bit in a LSM secattr catmap + * @catmap: the category bitmap + * @bit: the bit to set + * @flags: memory allocation flags + * + * Description: + * Set the bit specified by @bit in @catmap. Returns zero on success, + * negative values on failure. + * + */ +int netlbl_secattr_catmap_setbit(struct netlbl_lsm_secattr_catmap *catmap, + u32 bit, + gfp_t flags) +{ + struct netlbl_lsm_secattr_catmap *iter = catmap; + u32 node_bit; + u32 node_idx; + + while (iter->next != NULL && + bit >= (iter->startbit + NETLBL_CATMAP_SIZE)) + iter = iter->next; + if (bit >= (iter->startbit + NETLBL_CATMAP_SIZE)) { + iter->next = netlbl_secattr_catmap_alloc(flags); + if (iter->next == NULL) + return -ENOMEM; + iter = iter->next; + iter->startbit = bit & ~(NETLBL_CATMAP_SIZE - 1); + } + + /* gcc always rounds to zero when doing integer division */ + node_idx = (bit - iter->startbit) / NETLBL_CATMAP_MAPSIZE; + node_bit = bit - iter->startbit - (NETLBL_CATMAP_MAPSIZE * node_idx); + iter->bitmap[node_idx] |= NETLBL_CATMAP_BIT << node_bit; + + return 0; +} + +/** + * netlbl_secattr_catmap_setrng - Set a range of bits in a LSM secattr catmap + * @catmap: the category bitmap + * @start: the starting bit + * @end: the last bit in the string + * @flags: memory allocation flags + * + * Description: + * Set a range of bits, starting at @start and ending with @end. Returns zero + * on success, negative values on failure. + * + */ +int netlbl_secattr_catmap_setrng(struct netlbl_lsm_secattr_catmap *catmap, + u32 start, + u32 end, + gfp_t flags) +{ + int ret_val = 0; + struct netlbl_lsm_secattr_catmap *iter = catmap; + u32 iter_max_spot; + u32 spot; + + /* XXX - This could probably be made a bit faster by combining writes + * to the catmap instead of setting a single bit each time, but for + * right now skipping to the start of the range in the catmap should + * be a nice improvement over calling the individual setbit function + * repeatedly from a loop. */ + + while (iter->next != NULL && + start >= (iter->startbit + NETLBL_CATMAP_SIZE)) + iter = iter->next; + iter_max_spot = iter->startbit + NETLBL_CATMAP_SIZE; + + for (spot = start; spot <= end && ret_val == 0; spot++) { + if (spot >= iter_max_spot && iter->next != NULL) { + iter = iter->next; + iter_max_spot = iter->startbit + NETLBL_CATMAP_SIZE; + } + ret_val = netlbl_secattr_catmap_setbit(iter, spot, GFP_ATOMIC); + } + + return ret_val; +} + /* * LSM Functions */ diff --git a/security/selinux/ss/ebitmap.c b/security/selinux/ss/ebitmap.c index d539346ab3a..ce492a6b38e 100644 --- a/security/selinux/ss/ebitmap.c +++ b/security/selinux/ss/ebitmap.c @@ -6,7 +6,7 @@ /* * Updated: Hewlett-Packard * - * Added ebitmap_export() and ebitmap_import() + * Added support to import/export the NetLabel category bitmap * * (c) Copyright Hewlett-Packard Development Company, L.P., 2006 */ @@ -14,6 +14,7 @@ #include #include #include +#include #include "ebitmap.h" #include "policydb.h" @@ -67,141 +68,120 @@ int ebitmap_cpy(struct ebitmap *dst, struct ebitmap *src) return 0; } +#ifdef CONFIG_NETLABEL /** - * ebitmap_export - Export an ebitmap to a unsigned char bitmap string - * @src: the ebitmap to export - * @dst: the resulting bitmap string - * @dst_len: length of dst in bytes + * ebitmap_netlbl_export - Export an ebitmap into a NetLabel category bitmap + * @ebmap: the ebitmap to export + * @catmap: the NetLabel category bitmap * * Description: - * Allocate a buffer at least src->highbit bits long and export the extensible - * bitmap into the buffer. The bitmap string will be in little endian format, - * i.e. LSB first. The value returned in dst_len may not the true size of the - * buffer as the length of the buffer is rounded up to a multiple of MAPTYPE. - * The caller must free the buffer when finished. Returns zero on success, - * negative values on failure. + * Export a SELinux extensibile bitmap into a NetLabel category bitmap. + * Returns zero on success, negative values on error. * */ -int ebitmap_export(const struct ebitmap *src, - unsigned char **dst, - size_t *dst_len) +int ebitmap_netlbl_export(struct ebitmap *ebmap, + struct netlbl_lsm_secattr_catmap **catmap) { - size_t bitmap_len; - unsigned char *bitmap; - struct ebitmap_node *iter_node; - MAPTYPE node_val; - size_t bitmap_byte; - unsigned char bitmask; - - if (src->highbit == 0) { - *dst = NULL; - *dst_len = 0; + struct ebitmap_node *e_iter = ebmap->node; + struct netlbl_lsm_secattr_catmap *c_iter; + u32 cmap_idx; + + /* This function is a much simpler because SELinux's MAPTYPE happens + * to be the same as NetLabel's NETLBL_CATMAP_MAPTYPE, if MAPTYPE is + * changed from a u64 this function will most likely need to be changed + * as well. It's not ideal but I think the tradeoff in terms of + * neatness and speed is worth it. */ + + if (e_iter == NULL) { + *catmap = NULL; return 0; } - bitmap_len = src->highbit / 8; - if (src->highbit % 7) - bitmap_len += 1; - - bitmap = kzalloc((bitmap_len & ~(sizeof(MAPTYPE) - 1)) + - sizeof(MAPTYPE), - GFP_ATOMIC); - if (bitmap == NULL) + c_iter = netlbl_secattr_catmap_alloc(GFP_ATOMIC); + if (c_iter == NULL) return -ENOMEM; + *catmap = c_iter; + c_iter->startbit = e_iter->startbit & ~(NETLBL_CATMAP_SIZE - 1); + + while (e_iter != NULL) { + if (e_iter->startbit >= + (c_iter->startbit + NETLBL_CATMAP_SIZE)) { + c_iter->next = netlbl_secattr_catmap_alloc(GFP_ATOMIC); + if (c_iter->next == NULL) + goto netlbl_export_failure; + c_iter = c_iter->next; + c_iter->startbit = e_iter->startbit & + ~(NETLBL_CATMAP_SIZE - 1); + } + cmap_idx = (e_iter->startbit - c_iter->startbit) / + NETLBL_CATMAP_MAPSIZE; + c_iter->bitmap[cmap_idx] = e_iter->map; + e_iter = e_iter->next; + } - iter_node = src->node; - do { - bitmap_byte = iter_node->startbit / 8; - bitmask = 0x80; - node_val = iter_node->map; - do { - if (bitmask == 0) { - bitmap_byte++; - bitmask = 0x80; - } - if (node_val & (MAPTYPE)0x01) - bitmap[bitmap_byte] |= bitmask; - node_val >>= 1; - bitmask >>= 1; - } while (node_val > 0); - iter_node = iter_node->next; - } while (iter_node); - - *dst = bitmap; - *dst_len = bitmap_len; return 0; + +netlbl_export_failure: + netlbl_secattr_catmap_free(*catmap); + return -ENOMEM; } /** - * ebitmap_import - Import an unsigned char bitmap string into an ebitmap - * @src: the bitmap string - * @src_len: the bitmap length in bytes - * @dst: the empty ebitmap + * ebitmap_netlbl_import - Import a NetLabel category bitmap into an ebitmap + * @ebmap: the ebitmap to export + * @catmap: the NetLabel category bitmap * * Description: - * This function takes a little endian bitmap string in src and imports it into - * the ebitmap pointed to by dst. Returns zero on success, negative values on - * failure. + * Import a NetLabel category bitmap into a SELinux extensibile bitmap. + * Returns zero on success, negative values on error. * */ -int ebitmap_import(const unsigned char *src, - size_t src_len, - struct ebitmap *dst) +int ebitmap_netlbl_import(struct ebitmap *ebmap, + struct netlbl_lsm_secattr_catmap *catmap) { - size_t src_off = 0; - size_t node_limit; - struct ebitmap_node *node_new; - struct ebitmap_node *node_last = NULL; - u32 i_byte; - u32 i_bit; - unsigned char src_byte; - - while (src_off < src_len) { - if (src_len - src_off >= sizeof(MAPTYPE)) { - if (*(MAPTYPE *)&src[src_off] == 0) { - src_off += sizeof(MAPTYPE); - continue; - } - node_limit = sizeof(MAPTYPE); - } else { - for (src_byte = 0, i_byte = src_off; - i_byte < src_len && src_byte == 0; - i_byte++) - src_byte |= src[i_byte]; - if (src_byte == 0) - break; - node_limit = src_len - src_off; - } + struct ebitmap_node *e_iter = NULL; + struct ebitmap_node *emap_prev = NULL; + struct netlbl_lsm_secattr_catmap *c_iter = catmap; + u32 c_idx; - node_new = kzalloc(sizeof(*node_new), GFP_ATOMIC); - if (unlikely(node_new == NULL)) { - ebitmap_destroy(dst); - return -ENOMEM; - } - node_new->startbit = src_off * 8; - for (i_byte = 0; i_byte < node_limit; i_byte++) { - src_byte = src[src_off++]; - for (i_bit = i_byte * 8; src_byte != 0; i_bit++) { - if (src_byte & 0x80) - node_new->map |= MAPBIT << i_bit; - src_byte <<= 1; - } - } + /* This function is a much simpler because SELinux's MAPTYPE happens + * to be the same as NetLabel's NETLBL_CATMAP_MAPTYPE, if MAPTYPE is + * changed from a u64 this function will most likely need to be changed + * as well. It's not ideal but I think the tradeoff in terms of + * neatness and speed is worth it. */ - if (node_last != NULL) - node_last->next = node_new; - else - dst->node = node_new; - node_last = node_new; - } + do { + for (c_idx = 0; c_idx < NETLBL_CATMAP_MAPCNT; c_idx++) { + if (c_iter->bitmap[c_idx] == 0) + continue; + + e_iter = kzalloc(sizeof(*e_iter), GFP_ATOMIC); + if (e_iter == NULL) + goto netlbl_import_failure; + if (emap_prev == NULL) + ebmap->node = e_iter; + else + emap_prev->next = e_iter; + emap_prev = e_iter; - if (likely(node_last != NULL)) - dst->highbit = node_last->startbit + MAPSIZE; + e_iter->startbit = c_iter->startbit + + NETLBL_CATMAP_MAPSIZE * c_idx; + e_iter->map = c_iter->bitmap[c_idx]; + } + c_iter = c_iter->next; + } while (c_iter != NULL); + if (e_iter != NULL) + ebmap->highbit = e_iter->startbit + MAPSIZE; else - ebitmap_init(dst); + ebitmap_destroy(ebmap); return 0; + +netlbl_import_failure: + ebitmap_destroy(ebmap); + return -ENOMEM; } +#endif /* CONFIG_NETLABEL */ int ebitmap_contains(struct ebitmap *e1, struct ebitmap *e2) { diff --git a/security/selinux/ss/ebitmap.h b/security/selinux/ss/ebitmap.h index da2d4651b10..1270e34b61c 100644 --- a/security/selinux/ss/ebitmap.h +++ b/security/selinux/ss/ebitmap.h @@ -14,6 +14,8 @@ #ifndef _SS_EBITMAP_H_ #define _SS_EBITMAP_H_ +#include + #define MAPTYPE u64 /* portion of bitmap in each node */ #define MAPSIZE (sizeof(MAPTYPE) * 8) /* number of bits in node bitmap */ #define MAPBIT 1ULL /* a bit in the node bitmap */ @@ -69,16 +71,28 @@ static inline int ebitmap_node_get_bit(struct ebitmap_node * n, int ebitmap_cmp(struct ebitmap *e1, struct ebitmap *e2); int ebitmap_cpy(struct ebitmap *dst, struct ebitmap *src); -int ebitmap_export(const struct ebitmap *src, - unsigned char **dst, - size_t *dst_len); -int ebitmap_import(const unsigned char *src, - size_t src_len, - struct ebitmap *dst); int ebitmap_contains(struct ebitmap *e1, struct ebitmap *e2); int ebitmap_get_bit(struct ebitmap *e, unsigned long bit); int ebitmap_set_bit(struct ebitmap *e, unsigned long bit, int value); void ebitmap_destroy(struct ebitmap *e); int ebitmap_read(struct ebitmap *e, void *fp); +#ifdef CONFIG_NETLABEL +int ebitmap_netlbl_export(struct ebitmap *ebmap, + struct netlbl_lsm_secattr_catmap **catmap); +int ebitmap_netlbl_import(struct ebitmap *ebmap, + struct netlbl_lsm_secattr_catmap *catmap); +#else +static inline int ebitmap_netlbl_export(struct ebitmap *ebmap, + struct netlbl_lsm_secattr_catmap **catmap) +{ + return -ENOMEM; +} +static inline int ebitmap_netlbl_import(struct ebitmap *ebmap, + struct netlbl_lsm_secattr_catmap *catmap) +{ + return -ENOMEM; +} +#endif + #endif /* _SS_EBITMAP_H_ */ diff --git a/security/selinux/ss/mls.c b/security/selinux/ss/mls.c index 2cca8e25162..b4f682dc13f 100644 --- a/security/selinux/ss/mls.c +++ b/security/selinux/ss/mls.c @@ -13,7 +13,7 @@ /* * Updated: Hewlett-Packard * - * Added support to import/export the MLS label + * Added support to import/export the MLS label from NetLabel * * (c) Copyright Hewlett-Packard Development Company, L.P., 2006 */ @@ -22,6 +22,7 @@ #include #include #include +#include #include "sidtab.h" #include "mls.h" #include "policydb.h" @@ -571,163 +572,108 @@ int mls_compute_sid(struct context *scontext, return -EINVAL; } +#ifdef CONFIG_NETLABEL /** - * mls_export_lvl - Export the MLS sensitivity levels + * mls_export_netlbl_lvl - Export the MLS sensitivity levels to NetLabel * @context: the security context - * @low: the low sensitivity level - * @high: the high sensitivity level + * @secattr: the NetLabel security attributes * * Description: - * Given the security context copy the low MLS sensitivity level into lvl_low - * and the high sensitivity level in lvl_high. The MLS levels are only - * exported if the pointers are not NULL, if they are NULL then that level is - * not exported. + * Given the security context copy the low MLS sensitivity level into the + * NetLabel MLS sensitivity level field. * */ -void mls_export_lvl(const struct context *context, u32 *low, u32 *high) +void mls_export_netlbl_lvl(struct context *context, + struct netlbl_lsm_secattr *secattr) { if (!selinux_mls_enabled) return; - if (low != NULL) - *low = context->range.level[0].sens - 1; - if (high != NULL) - *high = context->range.level[1].sens - 1; + secattr->mls_lvl = context->range.level[0].sens - 1; + secattr->flags |= NETLBL_SECATTR_MLS_LVL; } /** - * mls_import_lvl - Import the MLS sensitivity levels + * mls_import_netlbl_lvl - Import the NetLabel MLS sensitivity levels * @context: the security context - * @low: the low sensitivity level - * @high: the high sensitivity level + * @secattr: the NetLabel security attributes * * Description: - * Given the security context and the two sensitivty levels, set the MLS levels - * in the context according the two given as parameters. Returns zero on - * success, negative values on failure. + * Given the security context and the NetLabel security attributes, copy the + * NetLabel MLS sensitivity level into the context. * */ -void mls_import_lvl(struct context *context, u32 low, u32 high) +void mls_import_netlbl_lvl(struct context *context, + struct netlbl_lsm_secattr *secattr) { if (!selinux_mls_enabled) return; - context->range.level[0].sens = low + 1; - context->range.level[1].sens = high + 1; + context->range.level[0].sens = secattr->mls_lvl + 1; + context->range.level[1].sens = context->range.level[0].sens; } /** - * mls_export_cat - Export the MLS categories + * mls_export_netlbl_cat - Export the MLS categories to NetLabel * @context: the security context - * @low: the low category - * @low_len: length of the cat_low bitmap in bytes - * @high: the high category - * @high_len: length of the cat_high bitmap in bytes + * @secattr: the NetLabel security attributes * * Description: - * Given the security context export the low MLS category bitmap into cat_low - * and the high category bitmap into cat_high. The MLS categories are only - * exported if the pointers are not NULL, if they are NULL then that level is - * not exported. The caller is responsibile for freeing the memory when - * finished. Returns zero on success, negative values on failure. + * Given the security context copy the low MLS categories into the NetLabel + * MLS category field. Returns zero on success, negative values on failure. * */ -int mls_export_cat(const struct context *context, - unsigned char **low, - size_t *low_len, - unsigned char **high, - size_t *high_len) +int mls_export_netlbl_cat(struct context *context, + struct netlbl_lsm_secattr *secattr) { - int rc = -EPERM; + int rc; - if (!selinux_mls_enabled) { - *low = NULL; - *low_len = 0; - *high = NULL; - *high_len = 0; + if (!selinux_mls_enabled) return 0; - } - if (low != NULL) { - rc = ebitmap_export(&context->range.level[0].cat, - low, - low_len); - if (rc != 0) - goto export_cat_failure; - } - if (high != NULL) { - rc = ebitmap_export(&context->range.level[1].cat, - high, - high_len); - if (rc != 0) - goto export_cat_failure; - } - - return 0; + rc = ebitmap_netlbl_export(&context->range.level[0].cat, + &secattr->mls_cat); + if (rc == 0 && secattr->mls_cat != NULL) + secattr->flags |= NETLBL_SECATTR_MLS_CAT; -export_cat_failure: - if (low != NULL) { - kfree(*low); - *low = NULL; - *low_len = 0; - } - if (high != NULL) { - kfree(*high); - *high = NULL; - *high_len = 0; - } return rc; } /** - * mls_import_cat - Import the MLS categories + * mls_import_netlbl_cat - Import the MLS categories from NetLabel * @context: the security context - * @low: the low category - * @low_len: length of the cat_low bitmap in bytes - * @high: the high category - * @high_len: length of the cat_high bitmap in bytes + * @secattr: the NetLabel security attributes * * Description: - * Given the security context and the two category bitmap strings import the - * categories into the security context. The MLS categories are only imported - * if the pointers are not NULL, if they are NULL they are skipped. Returns - * zero on success, negative values on failure. + * Copy the NetLabel security attributes into the SELinux context; since the + * NetLabel security attribute only contains a single MLS category use it for + * both the low and high categories of the context. Returns zero on success, + * negative values on failure. * */ -int mls_import_cat(struct context *context, - const unsigned char *low, - size_t low_len, - const unsigned char *high, - size_t high_len) +int mls_import_netlbl_cat(struct context *context, + struct netlbl_lsm_secattr *secattr) { - int rc = -EPERM; + int rc; if (!selinux_mls_enabled) return 0; - if (low != NULL) { - rc = ebitmap_import(low, - low_len, - &context->range.level[0].cat); - if (rc != 0) - goto import_cat_failure; - } - if (high != NULL) { - if (high == low) - rc = ebitmap_cpy(&context->range.level[1].cat, - &context->range.level[0].cat); - else - rc = ebitmap_import(high, - high_len, - &context->range.level[1].cat); - if (rc != 0) - goto import_cat_failure; - } + rc = ebitmap_netlbl_import(&context->range.level[0].cat, + secattr->mls_cat); + if (rc != 0) + goto import_netlbl_cat_failure; + + rc = ebitmap_cpy(&context->range.level[1].cat, + &context->range.level[0].cat); + if (rc != 0) + goto import_netlbl_cat_failure; return 0; -import_cat_failure: +import_netlbl_cat_failure: ebitmap_destroy(&context->range.level[0].cat); ebitmap_destroy(&context->range.level[1].cat); return rc; } +#endif /* CONFIG_NETLABEL */ diff --git a/security/selinux/ss/mls.h b/security/selinux/ss/mls.h index df6032c6d49..661d6fc7696 100644 --- a/security/selinux/ss/mls.h +++ b/security/selinux/ss/mls.h @@ -13,7 +13,7 @@ /* * Updated: Hewlett-Packard * - * Added support to import/export the MLS label + * Added support to import/export the MLS label from NetLabel * * (c) Copyright Hewlett-Packard Development Company, L.P., 2006 */ @@ -69,19 +69,37 @@ int mls_compute_sid(struct context *scontext, int mls_setup_user_range(struct context *fromcon, struct user_datum *user, struct context *usercon); -void mls_export_lvl(const struct context *context, u32 *low, u32 *high); -void mls_import_lvl(struct context *context, u32 low, u32 high); - -int mls_export_cat(const struct context *context, - unsigned char **low, - size_t *low_len, - unsigned char **high, - size_t *high_len); -int mls_import_cat(struct context *context, - const unsigned char *low, - size_t low_len, - const unsigned char *high, - size_t high_len); +#ifdef CONFIG_NETLABEL +void mls_export_netlbl_lvl(struct context *context, + struct netlbl_lsm_secattr *secattr); +void mls_import_netlbl_lvl(struct context *context, + struct netlbl_lsm_secattr *secattr); +int mls_export_netlbl_cat(struct context *context, + struct netlbl_lsm_secattr *secattr); +int mls_import_netlbl_cat(struct context *context, + struct netlbl_lsm_secattr *secattr); +#else +static inline void mls_export_netlbl_lvl(struct context *context, + struct netlbl_lsm_secattr *secattr) +{ + return; +} +static inline void mls_import_netlbl_lvl(struct context *context, + struct netlbl_lsm_secattr *secattr) +{ + return; +} +static inline int mls_export_netlbl_cat(struct context *context, + struct netlbl_lsm_secattr *secattr) +{ + return -ENOMEM; +} +static inline int mls_import_netlbl_cat(struct context *context, + struct netlbl_lsm_secattr *secattr) +{ + return -ENOMEM; +} +#endif #endif /* _SS_MLS_H */ diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c index b43dd803fd5..bdb7070dd3d 100644 --- a/security/selinux/ss/services.c +++ b/security/selinux/ss/services.c @@ -55,6 +55,7 @@ #include "objsec.h" #include "selinux_netlabel.h" #include "xfrm.h" +#include "ebitmap.h" extern void selnl_notify_policyload(u32 seqno); unsigned int policydb_loaded_version; @@ -2384,13 +2385,10 @@ static int selinux_netlbl_secattr_to_sid(struct sk_buff *skb, ctx_new.user = ctx->user; ctx_new.role = ctx->role; ctx_new.type = ctx->type; - mls_import_lvl(&ctx_new, secattr->mls_lvl, secattr->mls_lvl); + mls_import_netlbl_lvl(&ctx_new, secattr); if (secattr->flags & NETLBL_SECATTR_MLS_CAT) { - if (mls_import_cat(&ctx_new, - secattr->mls_cat, - secattr->mls_cat_len, - NULL, - 0) != 0) + if (ebitmap_netlbl_import(&ctx_new.range.level[0].cat, + secattr->mls_cat) != 0) goto netlbl_secattr_to_sid_return; ctx_new.range.level[1].cat.highbit = ctx_new.range.level[0].cat.highbit; @@ -2486,19 +2484,12 @@ static int selinux_netlbl_socket_setsid(struct socket *sock, u32 sid) secattr.domain = kstrdup(policydb.p_type_val_to_name[ctx->type - 1], GFP_ATOMIC); - mls_export_lvl(ctx, &secattr.mls_lvl, NULL); - rc = mls_export_cat(ctx, - &secattr.mls_cat, - &secattr.mls_cat_len, - NULL, - NULL); + secattr.flags |= NETLBL_SECATTR_DOMAIN; + mls_export_netlbl_lvl(ctx, &secattr); + rc = mls_export_netlbl_cat(ctx, &secattr); if (rc != 0) goto netlbl_socket_setsid_return; - secattr.flags |= NETLBL_SECATTR_DOMAIN | NETLBL_SECATTR_MLS_LVL; - if (secattr.mls_cat) - secattr.flags |= NETLBL_SECATTR_MLS_CAT; - rc = netlbl_socket_setattr(sock, &secattr); if (rc == 0) { spin_lock(&sksec->nlbl_lock); -- cgit v1.2.3-70-g09d2 From 9f9afec48221fe4a19f84a9341f5b304bf7d7783 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 29 Nov 2006 17:35:18 -0800 Subject: [NET_SCHED]: Set parent classid in default qdiscs Set parent classids in default qdiscs to allow walking up the tree from outside the qdiscs. This is needed by the next patch. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/net/sch_generic.h | 2 +- net/sched/sch_atm.c | 5 +++-- net/sched/sch_cbq.c | 8 +++++--- net/sched/sch_dsmark.c | 5 +++-- net/sched/sch_generic.c | 7 +++++-- net/sched/sch_hfsc.c | 8 +++++--- net/sched/sch_htb.c | 7 ++++--- net/sched/sch_netem.c | 3 ++- net/sched/sch_prio.c | 3 ++- net/sched/sch_red.c | 8 +++++--- net/sched/sch_tbf.c | 8 +++++--- 11 files changed, 40 insertions(+), 24 deletions(-) (limited to 'include') diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 6b3829480b3..b37572db12a 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -174,7 +174,7 @@ extern void qdisc_reset(struct Qdisc *qdisc); extern void qdisc_destroy(struct Qdisc *qdisc); extern struct Qdisc *qdisc_alloc(struct net_device *dev, struct Qdisc_ops *ops); extern struct Qdisc *qdisc_create_dflt(struct net_device *dev, - struct Qdisc_ops *ops); + struct Qdisc_ops *ops, u32 parentid); static inline void tcf_destroy(struct tcf_proto *tp) diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c index dbf44da0912..edc7bb0b9c8 100644 --- a/net/sched/sch_atm.c +++ b/net/sched/sch_atm.c @@ -316,7 +316,7 @@ static int atm_tc_change(struct Qdisc *sch, u32 classid, u32 parent, } memset(flow,0,sizeof(*flow)); flow->filter_list = NULL; - if (!(flow->q = qdisc_create_dflt(sch->dev,&pfifo_qdisc_ops))) + if (!(flow->q = qdisc_create_dflt(sch->dev,&pfifo_qdisc_ops,classid))) flow->q = &noop_qdisc; DPRINTK("atm_tc_change: qdisc %p\n",flow->q); flow->sock = sock; @@ -576,7 +576,8 @@ static int atm_tc_init(struct Qdisc *sch,struct rtattr *opt) DPRINTK("atm_tc_init(sch %p,[qdisc %p],opt %p)\n",sch,p,opt); p->flows = &p->link; - if(!(p->link.q = qdisc_create_dflt(sch->dev,&pfifo_qdisc_ops))) + if(!(p->link.q = qdisc_create_dflt(sch->dev,&pfifo_qdisc_ops, + sch->handle))) p->link.q = &noop_qdisc; DPRINTK("atm_tc_init: link (%p) qdisc %p\n",&p->link,p->link.q); p->link.filter_list = NULL; diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c index bac881bfe36..908b10d0d61 100644 --- a/net/sched/sch_cbq.c +++ b/net/sched/sch_cbq.c @@ -1429,7 +1429,8 @@ static int cbq_init(struct Qdisc *sch, struct rtattr *opt) q->link.sibling = &q->link; q->link.classid = sch->handle; q->link.qdisc = sch; - if (!(q->link.q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops))) + if (!(q->link.q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, + sch->handle))) q->link.q = &noop_qdisc; q->link.priority = TC_CBQ_MAXPRIO-1; @@ -1674,7 +1675,8 @@ static int cbq_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, if (cl) { if (new == NULL) { - if ((new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops)) == NULL) + if ((new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, + cl->classid)) == NULL) return -ENOBUFS; } else { #ifdef CONFIG_NET_CLS_POLICE @@ -1932,7 +1934,7 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct rtattr **t cl->R_tab = rtab; rtab = NULL; cl->refcnt = 1; - if (!(cl->q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops))) + if (!(cl->q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, classid))) cl->q = &noop_qdisc; cl->classid = classid; cl->tparent = parent; diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c index 0b7c0e82d60..afedd92b749 100644 --- a/net/sched/sch_dsmark.c +++ b/net/sched/sch_dsmark.c @@ -88,7 +88,8 @@ static int dsmark_graft(struct Qdisc *sch, unsigned long arg, sch, p, new, old); if (new == NULL) { - new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops); + new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, + sch->handle); if (new == NULL) new = &noop_qdisc; } @@ -387,7 +388,7 @@ static int dsmark_init(struct Qdisc *sch, struct rtattr *opt) p->default_index = default_index; p->set_tc_index = RTA_GET_FLAG(tb[TCA_DSMARK_SET_TC_INDEX-1]); - p->q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops); + p->q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, sch->handle); if (p->q == NULL) p->q = &noop_qdisc; diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 88c6a99ce53..deafeb9cb70 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -450,13 +450,15 @@ errout: return ERR_PTR(-err); } -struct Qdisc * qdisc_create_dflt(struct net_device *dev, struct Qdisc_ops *ops) +struct Qdisc * qdisc_create_dflt(struct net_device *dev, struct Qdisc_ops *ops, + unsigned int parentid) { struct Qdisc *sch; sch = qdisc_alloc(dev, ops); if (IS_ERR(sch)) goto errout; + sch->parent = parentid; if (!ops->init || ops->init(sch, NULL) == 0) return sch; @@ -520,7 +522,8 @@ void dev_activate(struct net_device *dev) if (dev->qdisc_sleeping == &noop_qdisc) { struct Qdisc *qdisc; if (dev->tx_queue_len) { - qdisc = qdisc_create_dflt(dev, &pfifo_fast_ops); + qdisc = qdisc_create_dflt(dev, &pfifo_fast_ops, + TC_H_ROOT); if (qdisc == NULL) { printk(KERN_INFO "%s: activation failed\n", dev->name); return; diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index 6a6735a2ed3..1142d298ecf 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -1138,7 +1138,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid, cl->classid = classid; cl->sched = q; cl->cl_parent = parent; - cl->qdisc = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops); + cl->qdisc = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, classid); if (cl->qdisc == NULL) cl->qdisc = &noop_qdisc; cl->stats_lock = &sch->dev->queue_lock; @@ -1271,7 +1271,8 @@ hfsc_graft_class(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, if (cl->level > 0) return -EINVAL; if (new == NULL) { - new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops); + new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, + cl->classid); if (new == NULL) new = &noop_qdisc; } @@ -1514,7 +1515,8 @@ hfsc_init_qdisc(struct Qdisc *sch, struct rtattr *opt) q->root.refcnt = 1; q->root.classid = sch->handle; q->root.sched = q; - q->root.qdisc = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops); + q->root.qdisc = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, + sch->handle); if (q->root.qdisc == NULL) q->root.qdisc = &noop_qdisc; q->root.stats_lock = &sch->dev->queue_lock; diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index 08fa4d08361..3b36e9d60c2 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -1223,8 +1223,9 @@ static int htb_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, struct htb_class *cl = (struct htb_class *)arg; if (cl && !cl->level) { - if (new == NULL && (new = qdisc_create_dflt(sch->dev, - &pfifo_qdisc_ops)) + if (new == NULL && + (new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, + cl->classid)) == NULL) return -ENOBUFS; sch_tree_lock(sch); @@ -1415,7 +1416,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, /* create leaf qdisc early because it uses kmalloc(GFP_KERNEL) so that can't be used inside of sch_tree_lock -- thanks to Karlis Peisenieks */ - new_q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops); + new_q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, classid); sch_tree_lock(sch); if (parent && !parent->level) { /* turn parent into inner node */ diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 0441876aa1e..90aeeb7b716 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -574,7 +574,8 @@ static int netem_init(struct Qdisc *sch, struct rtattr *opt) q->timer.function = netem_watchdog; q->timer.data = (unsigned long) sch; - q->qdisc = qdisc_create_dflt(sch->dev, &tfifo_qdisc_ops); + q->qdisc = qdisc_create_dflt(sch->dev, &tfifo_qdisc_ops, + TC_H_MAKE(sch->handle, 1)); if (!q->qdisc) { pr_debug("netem: qdisc create failed\n"); return -ENOMEM; diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c index a5fa03c0c19..3fc0c0fa5bc 100644 --- a/net/sched/sch_prio.c +++ b/net/sched/sch_prio.c @@ -230,7 +230,8 @@ static int prio_tune(struct Qdisc *sch, struct rtattr *opt) for (i=0; ibands; i++) { if (q->queues[i] == &noop_qdisc) { struct Qdisc *child; - child = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops); + child = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, + TC_H_MAKE(sch->handle, i + 1)); if (child) { sch_tree_lock(sch); child = xchg(&q->queues[i], child); diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c index d65cadddea6..ee66c5ca80c 100644 --- a/net/sched/sch_red.c +++ b/net/sched/sch_red.c @@ -175,12 +175,14 @@ static void red_destroy(struct Qdisc *sch) qdisc_destroy(q->qdisc); } -static struct Qdisc *red_create_dflt(struct net_device *dev, u32 limit) +static struct Qdisc *red_create_dflt(struct Qdisc *sch, u32 limit) { - struct Qdisc *q = qdisc_create_dflt(dev, &bfifo_qdisc_ops); + struct Qdisc *q; struct rtattr *rta; int ret; + q = qdisc_create_dflt(sch->dev, &bfifo_qdisc_ops, + TC_H_MAKE(sch->handle, 1)); if (q) { rta = kmalloc(RTA_LENGTH(sizeof(struct tc_fifo_qopt)), GFP_KERNEL); @@ -219,7 +221,7 @@ static int red_change(struct Qdisc *sch, struct rtattr *opt) ctl = RTA_DATA(tb[TCA_RED_PARMS-1]); if (ctl->limit > 0) { - child = red_create_dflt(sch->dev, ctl->limit); + child = red_create_dflt(sch, ctl->limit); if (child == NULL) return -ENOMEM; } diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c index d9a5d298d75..2562a6040b7 100644 --- a/net/sched/sch_tbf.c +++ b/net/sched/sch_tbf.c @@ -273,12 +273,14 @@ static void tbf_reset(struct Qdisc* sch) del_timer(&q->wd_timer); } -static struct Qdisc *tbf_create_dflt_qdisc(struct net_device *dev, u32 limit) +static struct Qdisc *tbf_create_dflt_qdisc(struct Qdisc *sch, u32 limit) { - struct Qdisc *q = qdisc_create_dflt(dev, &bfifo_qdisc_ops); + struct Qdisc *q; struct rtattr *rta; int ret; + q = qdisc_create_dflt(sch->dev, &bfifo_qdisc_ops, + TC_H_MAKE(sch->handle, 1)); if (q) { rta = kmalloc(RTA_LENGTH(sizeof(struct tc_fifo_qopt)), GFP_KERNEL); if (rta) { @@ -341,7 +343,7 @@ static int tbf_change(struct Qdisc* sch, struct rtattr *opt) goto done; if (qopt->limit > 0) { - if ((child = tbf_create_dflt_qdisc(sch->dev, qopt->limit)) == NULL) + if ((child = tbf_create_dflt_qdisc(sch, qopt->limit)) == NULL) goto done; } -- cgit v1.2.3-70-g09d2 From 43effa1e57fc4635e0301b27d78f9d83afe78974 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 29 Nov 2006 17:35:48 -0800 Subject: [NET_SCHED]: Fix endless loops caused by inaccurate qlen counters (part 1) There are multiple problems related to qlen adjustment that can lead to an upper qdisc getting out of sync with the real number of packets queued, leading to endless dequeueing attempts by the upper layer code. All qdiscs must maintain an accurate q.qlen counter. There are basically two groups of operations affecting the qlen: operations that propagate down the tree (enqueue, dequeue, requeue, drop, reset) beginning at the root qdisc and operations only affecting a subtree or single qdisc (change, graft, delete class). Since qlen changes during operations from the second group don't propagate to ancestor qdiscs, their qlen values become desynchronized. This patch adds a function to propagate qlen changes up the qdisc tree, optionally calling a callback function to perform qdisc-internal maintenance when the child qdisc becomes empty. The follow-up patches will convert all qdiscs to use this function where necessary. Noticed by Timo Steinbach . Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/net/sch_generic.h | 2 ++ net/sched/sch_api.c | 38 ++++++++++++++++++++++++++++++++------ 2 files changed, 34 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index b37572db12a..82086392735 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -60,6 +60,7 @@ struct Qdisc_class_ops int (*graft)(struct Qdisc *, unsigned long cl, struct Qdisc *, struct Qdisc **); struct Qdisc * (*leaf)(struct Qdisc *, unsigned long cl); + void (*qlen_notify)(struct Qdisc *, unsigned long); /* Class manipulation routines */ unsigned long (*get)(struct Qdisc *, u32 classid); @@ -172,6 +173,7 @@ extern void dev_activate(struct net_device *dev); extern void dev_deactivate(struct net_device *dev); extern void qdisc_reset(struct Qdisc *qdisc); extern void qdisc_destroy(struct Qdisc *qdisc); +extern void qdisc_tree_decrease_qlen(struct Qdisc *qdisc, unsigned int n); extern struct Qdisc *qdisc_alloc(struct net_device *dev, struct Qdisc_ops *ops); extern struct Qdisc *qdisc_create_dflt(struct net_device *dev, struct Qdisc_ops *ops, u32 parentid); diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index d99802f3023..05a93da6f05 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -191,21 +191,27 @@ int unregister_qdisc(struct Qdisc_ops *qops) (root qdisc, all its children, children of children etc.) */ -struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle) +static struct Qdisc *__qdisc_lookup(struct net_device *dev, u32 handle) { struct Qdisc *q; - read_lock(&qdisc_tree_lock); list_for_each_entry(q, &dev->qdisc_list, list) { - if (q->handle == handle) { - read_unlock(&qdisc_tree_lock); + if (q->handle == handle) return q; - } } - read_unlock(&qdisc_tree_lock); return NULL; } +struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle) +{ + struct Qdisc *q; + + read_lock(&qdisc_tree_lock); + q = __qdisc_lookup(dev, handle); + read_unlock(&qdisc_tree_lock); + return q; +} + static struct Qdisc *qdisc_leaf(struct Qdisc *p, u32 classid) { unsigned long cl; @@ -348,6 +354,26 @@ dev_graft_qdisc(struct net_device *dev, struct Qdisc *qdisc) return oqdisc; } +void qdisc_tree_decrease_qlen(struct Qdisc *sch, unsigned int n) +{ + struct Qdisc_class_ops *cops; + unsigned long cl; + u32 parentid; + + if (n == 0) + return; + while ((parentid = sch->parent)) { + sch = __qdisc_lookup(sch->dev, TC_H_MAJ(parentid)); + cops = sch->ops->cl_ops; + if (cops->qlen_notify) { + cl = cops->get(sch, parentid); + cops->qlen_notify(sch, cl); + cops->put(sch, cl); + } + sch->q.qlen -= n; + } +} +EXPORT_SYMBOL(qdisc_tree_decrease_qlen); /* Graft qdisc "new" to class "classid" of qdisc "parent" or to device "dev". -- cgit v1.2.3-70-g09d2 From 76b3f055f38954c67dab13844eb92203580038f8 Mon Sep 17 00:00:00 2001 From: Miika Komu Date: Thu, 30 Nov 2006 16:40:43 -0800 Subject: [IPSEC]: Add encapsulation family. Signed-off-by: Miika Komu Signed-off-by: Diego Beltrami Signed-off-by: Kazunori Miyazawa Signed-off-by: David S. Miller --- include/net/xfrm.h | 2 ++ net/xfrm/xfrm_policy.c | 1 + 2 files changed, 3 insertions(+) (limited to 'include') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 984e5c45265..15ec19dcf9c 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -310,6 +310,8 @@ struct xfrm_tmpl /* Source address of tunnel. Ignored, if it is not a tunnel. */ xfrm_address_t saddr; + unsigned short encap_family; + __u32 reqid; /* Mode: transport, tunnel etc. */ diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index e5ea1347a4f..64d3938f74c 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -1211,6 +1211,7 @@ xfrm_tmpl_resolve_one(struct xfrm_policy *policy, struct flowi *fl, if (tmpl->mode == XFRM_MODE_TUNNEL) { remote = &tmpl->id.daddr; local = &tmpl->saddr; + family = tmpl->encap_family; if (xfrm_addr_any(local, family)) { error = xfrm_get_saddr(&tmp, remote, family); if (error) -- cgit v1.2.3-70-g09d2 From 1e419cd9953f59d06d7b88d0e2911a68a0044f33 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 30 Nov 2006 19:28:48 -0800 Subject: [EBTABLES]: Split ebt_replace into user and kernel variants, annotate. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/netfilter_bridge/ebtables.h | 19 ++++++++++++++++++- net/bridge/netfilter/ebtable_broute.c | 2 +- net/bridge/netfilter/ebtable_filter.c | 2 +- net/bridge/netfilter/ebtable_nat.c | 2 +- net/bridge/netfilter/ebtables.c | 19 ++++++++++--------- 5 files changed, 31 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter_bridge/ebtables.h b/include/linux/netfilter_bridge/ebtables.h index 87775264ff0..94e0a7dc0cb 100644 --- a/include/linux/netfilter_bridge/ebtables.h +++ b/include/linux/netfilter_bridge/ebtables.h @@ -38,6 +38,23 @@ struct ebt_counter }; struct ebt_replace +{ + char name[EBT_TABLE_MAXNAMELEN]; + unsigned int valid_hooks; + /* nr of rules in the table */ + unsigned int nentries; + /* total size of the entries */ + unsigned int entries_size; + /* start of the chains */ + struct ebt_entries __user *hook_entry[NF_BR_NUMHOOKS]; + /* nr of counters userspace expects back */ + unsigned int num_counters; + /* where the kernel will put the old counters */ + struct ebt_counter __user *counters; + char __user *entries; +}; + +struct ebt_replace_kernel { char name[EBT_TABLE_MAXNAMELEN]; unsigned int valid_hooks; @@ -255,7 +272,7 @@ struct ebt_table { struct list_head list; char name[EBT_TABLE_MAXNAMELEN]; - struct ebt_replace *table; + struct ebt_replace_kernel *table; unsigned int valid_hooks; rwlock_t lock; /* e.g. could be the table explicitly only allows certain diff --git a/net/bridge/netfilter/ebtable_broute.c b/net/bridge/netfilter/ebtable_broute.c index 9a6e548e148..d37ce047893 100644 --- a/net/bridge/netfilter/ebtable_broute.c +++ b/net/bridge/netfilter/ebtable_broute.c @@ -23,7 +23,7 @@ static struct ebt_entries initial_chain = { .policy = EBT_ACCEPT, }; -static struct ebt_replace initial_table = +static struct ebt_replace_kernel initial_table = { .name = "broute", .valid_hooks = 1 << NF_BR_BROUTING, diff --git a/net/bridge/netfilter/ebtable_filter.c b/net/bridge/netfilter/ebtable_filter.c index 3d5bd44f239..127135ead2d 100644 --- a/net/bridge/netfilter/ebtable_filter.c +++ b/net/bridge/netfilter/ebtable_filter.c @@ -30,7 +30,7 @@ static struct ebt_entries initial_chains[] = }, }; -static struct ebt_replace initial_table = +static struct ebt_replace_kernel initial_table = { .name = "filter", .valid_hooks = FILTER_VALID_HOOKS, diff --git a/net/bridge/netfilter/ebtable_nat.c b/net/bridge/netfilter/ebtable_nat.c index 04dd42efda1..9c50488b62e 100644 --- a/net/bridge/netfilter/ebtable_nat.c +++ b/net/bridge/netfilter/ebtable_nat.c @@ -30,7 +30,7 @@ static struct ebt_entries initial_chains[] = } }; -static struct ebt_replace initial_table = +static struct ebt_replace_kernel initial_table = { .name = "nat", .valid_hooks = NAT_VALID_HOOKS, diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index 00a89705c1c..bee558a4180 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -417,7 +417,8 @@ static int ebt_verify_pointers(struct ebt_replace *repl, for (i = 0; i < NF_BR_NUMHOOKS; i++) { if ((valid_hooks & (1 << i)) == 0) continue; - if ((char *)repl->hook_entry[i] == repl->entries + offset) + if ((char __user *)repl->hook_entry[i] == + repl->entries + offset) break; } @@ -1156,7 +1157,7 @@ int ebt_register_table(struct ebt_table *table) { struct ebt_table_info *newinfo; struct ebt_table *t; - struct ebt_replace *repl; + struct ebt_replace_kernel *repl; int ret, i, countersize; void *p; @@ -1320,33 +1321,33 @@ free_tmp: } static inline int ebt_make_matchname(struct ebt_entry_match *m, - char *base, char *ubase) + char *base, char __user *ubase) { - char *hlp = ubase - base + (char *)m; + char __user *hlp = ubase + ((char *)m - base); if (copy_to_user(hlp, m->u.match->name, EBT_FUNCTION_MAXNAMELEN)) return -EFAULT; return 0; } static inline int ebt_make_watchername(struct ebt_entry_watcher *w, - char *base, char *ubase) + char *base, char __user *ubase) { - char *hlp = ubase - base + (char *)w; + char __user *hlp = ubase + ((char *)w - base); if (copy_to_user(hlp , w->u.watcher->name, EBT_FUNCTION_MAXNAMELEN)) return -EFAULT; return 0; } -static inline int ebt_make_names(struct ebt_entry *e, char *base, char *ubase) +static inline int ebt_make_names(struct ebt_entry *e, char *base, char __user *ubase) { int ret; - char *hlp; + char __user *hlp; struct ebt_entry_target *t; if (e->bitmask == 0) return 0; - hlp = ubase - base + (char *)e + e->target_offset; + hlp = ubase + (((char *)e + e->target_offset) - base); t = (struct ebt_entry_target *)(((char *)e) + e->target_offset); ret = EBT_MATCH_ITERATE(e, ebt_make_matchname, base, ubase); -- cgit v1.2.3-70-g09d2 From a4d1366d5080fcb4514db58c320b9f60d78918e6 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Fri, 1 Dec 2006 20:07:42 -0800 Subject: [GENETLINK]: Add cmd dump completion. Remove assumption that generic netlink commands cannot have dump completion callbacks. Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller --- include/net/genetlink.h | 2 ++ net/netlink/genetlink.c | 2 +- 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/genetlink.h b/include/net/genetlink.h index 7fd131c9a8c..adff4c898d5 100644 --- a/include/net/genetlink.h +++ b/include/net/genetlink.h @@ -53,6 +53,7 @@ struct genl_info * @policy: attribute validation policy * @doit: standard command callback * @dumpit: callback for dumpers + * @done: completion callback for dumps * @ops_list: operations list */ struct genl_ops @@ -64,6 +65,7 @@ struct genl_ops struct genl_info *info); int (*dumpit)(struct sk_buff *skb, struct netlink_callback *cb); + int (*done)(struct netlink_callback *cb); struct list_head ops_list; }; diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c index cc874f0fcbd..b9b03747c1f 100644 --- a/net/netlink/genetlink.c +++ b/net/netlink/genetlink.c @@ -331,7 +331,7 @@ static int genl_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, } *errp = err = netlink_dump_start(genl_sock, skb, nlh, - ops->dumpit, NULL); + ops->dumpit, ops->done); if (err == 0) skb_pull(skb, min(NLMSG_ALIGN(nlh->nlmsg_len), skb->len)); -- cgit v1.2.3-70-g09d2 From 9a7c9337a09b1ef07e55f95a4309957a2328a01f Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 2 Dec 2006 22:04:04 -0800 Subject: [NET]: Accept wildcard delimiters in in[46]_pton Accept -1 as delimiter to abort parsing without an error at the first unknown character. This is needed by the upcoming nf_conntrack SIP helper, where addresses are delimited by either '\r' or '\n' characters. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/inet.h | 4 ++-- net/core/utils.c | 10 ++++++---- 2 files changed, 8 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/inet.h b/include/linux/inet.h index b7c6da7d6d3..675a7dbe86f 100644 --- a/include/linux/inet.h +++ b/include/linux/inet.h @@ -46,7 +46,7 @@ #include extern __be32 in_aton(const char *str); -extern int in4_pton(const char *src, int srclen, u8 *dst, char delim, const char **end); -extern int in6_pton(const char *src, int srclen, u8 *dst, char delim, const char **end); +extern int in4_pton(const char *src, int srclen, u8 *dst, int delim, const char **end); +extern int in6_pton(const char *src, int srclen, u8 *dst, int delim, const char **end); #endif #endif /* _LINUX_INET_H */ diff --git a/net/core/utils.c b/net/core/utils.c index d93fe64f669..61556065f07 100644 --- a/net/core/utils.c +++ b/net/core/utils.c @@ -88,7 +88,7 @@ EXPORT_SYMBOL(in_aton); #define IN6PTON_NULL 0x20000000 /* first/tail */ #define IN6PTON_UNKNOWN 0x40000000 -static inline int digit2bin(char c, char delim) +static inline int digit2bin(char c, int delim) { if (c == delim || c == '\0') return IN6PTON_DELIM; @@ -99,7 +99,7 @@ static inline int digit2bin(char c, char delim) return IN6PTON_UNKNOWN; } -static inline int xdigit2bin(char c, char delim) +static inline int xdigit2bin(char c, int delim) { if (c == delim || c == '\0') return IN6PTON_DELIM; @@ -113,12 +113,14 @@ static inline int xdigit2bin(char c, char delim) return (IN6PTON_XDIGIT | (c - 'a' + 10)); if (c >= 'A' && c <= 'F') return (IN6PTON_XDIGIT | (c - 'A' + 10)); + if (delim == -1) + return IN6PTON_DELIM; return IN6PTON_UNKNOWN; } int in4_pton(const char *src, int srclen, u8 *dst, - char delim, const char **end) + int delim, const char **end) { const char *s; u8 *d; @@ -173,7 +175,7 @@ EXPORT_SYMBOL(in4_pton); int in6_pton(const char *src, int srclen, u8 *dst, - char delim, const char **end) + int delim, const char **end) { const char *s, *tok = NULL; u8 *d, *dc = NULL; -- cgit v1.2.3-70-g09d2 From f9aae95828d3478520f4bd73221bcb450ec1a5c0 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 2 Dec 2006 22:04:50 -0800 Subject: [NETFILTER]: nf_conntrack: fix helper structure alignment Adding the alignment to the size doesn't make any sense, what it should do is align the size of the conntrack structure to the alignment requirements of the helper structure and return an aligned pointer in nfct_help(). Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/net/netfilter/nf_conntrack.h | 1 + net/netfilter/nf_conntrack_helper.c | 9 +++++---- 2 files changed, 6 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index d0d0e649144..b4beb8c799e 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -273,6 +273,7 @@ static inline struct nf_conn_help *nfct_help(const struct nf_conn *ct) if (!(ct->features & NF_CT_F_HELP)) return NULL; + offset = ALIGN(offset, __alignof__(struct nf_conn_help)); return (struct nf_conn_help *) ((void *)ct + offset); } diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index 03b3ed8fca9..2628f4ba35e 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -97,13 +97,14 @@ static inline int unhelp(struct nf_conntrack_tuple_hash *i, int nf_conntrack_helper_register(struct nf_conntrack_helper *me) { - int ret; + int size, ret; + BUG_ON(me->timeout == 0); + size = ALIGN(sizeof(struct nf_conn), __alignof__(struct nf_conn_help)) + + sizeof(struct nf_conn_help); ret = nf_conntrack_register_cache(NF_CT_F_HELP, "nf_conntrack:help", - sizeof(struct nf_conn) - + sizeof(struct nf_conn_help) - + __alignof__(struct nf_conn_help)); + size); if (ret < 0) { printk(KERN_ERR "nf_conntrack_helper_register: Unable to create slab cache for conntracks\n"); return ret; -- cgit v1.2.3-70-g09d2 From bff9a89bcac5b68ac0a1ea856b1726a35ae1eabb Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 2 Dec 2006 22:05:08 -0800 Subject: [NETFILTER]: nf_conntrack: endian annotations Resync with Al Viro's ip_conntrack annotations and fix a missed spot in ip_nat_proto_icmp.c. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/net/netfilter/nf_conntrack_tuple.h | 18 ++++++------- net/ipv4/netfilter/ip_nat_proto_icmp.c | 4 +-- net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 8 +++--- net/ipv4/netfilter/nf_conntrack_proto_icmp.c | 2 +- net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c | 2 +- net/ipv6/netfilter/nf_conntrack_reasm.c | 26 +++++++++--------- net/netfilter/nf_conntrack_core.c | 6 ++--- net/netfilter/nf_conntrack_ftp.c | 10 +++---- net/netfilter/nf_conntrack_netlink.c | 37 +++++++++++++------------- net/netfilter/nf_conntrack_proto_sctp.c | 2 +- net/netfilter/nf_conntrack_proto_tcp.c | 8 +++--- 11 files changed, 58 insertions(+), 65 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_tuple.h b/include/net/netfilter/nf_conntrack_tuple.h index 530ef1f7528..be9dc9a0eb7 100644 --- a/include/net/netfilter/nf_conntrack_tuple.h +++ b/include/net/netfilter/nf_conntrack_tuple.h @@ -26,8 +26,8 @@ network order! */ union nf_conntrack_man_l3proto { u_int32_t all[NF_CT_TUPLE_L3SIZE]; - u_int32_t ip; - u_int32_t ip6[4]; + __be32 ip; + __be32 ip6[4]; }; /* The protocol-specific manipulable parts of the tuple: always in @@ -38,16 +38,16 @@ union nf_conntrack_man_proto u_int16_t all; struct { - u_int16_t port; + __be16 port; } tcp; struct { - u_int16_t port; + __be16 port; } udp; struct { - u_int16_t id; + __be16 id; } icmp; struct { - u_int16_t port; + __be16 port; } sctp; }; @@ -77,16 +77,16 @@ struct nf_conntrack_tuple u_int16_t all; struct { - u_int16_t port; + __be16 port; } tcp; struct { - u_int16_t port; + __be16 port; } udp; struct { u_int8_t type, code; } icmp; struct { - u_int16_t port; + __be16 port; } sctp; } u; diff --git a/net/ipv4/netfilter/ip_nat_proto_icmp.c b/net/ipv4/netfilter/ip_nat_proto_icmp.c index 75266fe3e0f..fb716edd5bc 100644 --- a/net/ipv4/netfilter/ip_nat_proto_icmp.c +++ b/net/ipv4/netfilter/ip_nat_proto_icmp.c @@ -24,8 +24,8 @@ icmp_in_range(const struct ip_conntrack_tuple *tuple, const union ip_conntrack_manip_proto *min, const union ip_conntrack_manip_proto *max) { - return (tuple->src.u.icmp.id >= min->icmp.id - && tuple->src.u.icmp.id <= max->icmp.id); + return ntohs(tuple->src.u.icmp.id) >= ntohs(min->icmp.id) && + ntohs(tuple->src.u.icmp.id) <= ntohs(max->icmp.id); } static int diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c index f24e872d4b9..2b73e3edd94 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c @@ -41,7 +41,7 @@ static int ipv4_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, struct nf_conntrack_tuple *tuple) { - u_int32_t _addrs[2], *ap; + __be32 _addrs[2], *ap; ap = skb_header_pointer(skb, nhoff + offsetof(struct iphdr, saddr), sizeof(u_int32_t) * 2, _addrs); if (ap == NULL) @@ -410,10 +410,8 @@ static int ipv4_nfattr_to_tuple(struct nfattr *tb[], if (nfattr_bad_size(tb, CTA_IP_MAX, cta_min_ip)) return -EINVAL; - t->src.u3.ip = - *(u_int32_t *)NFA_DATA(tb[CTA_IP_V4_SRC-1]); - t->dst.u3.ip = - *(u_int32_t *)NFA_DATA(tb[CTA_IP_V4_DST-1]); + t->src.u3.ip = *(__be32 *)NFA_DATA(tb[CTA_IP_V4_SRC-1]); + t->dst.u3.ip = *(__be32 *)NFA_DATA(tb[CTA_IP_V4_DST-1]); return 0; } diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c index 46aa44abc07..d25002fb817 100644 --- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c +++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c @@ -311,7 +311,7 @@ static int icmp_nfattr_to_tuple(struct nfattr *tb[], tuple->dst.u.icmp.code = *(u_int8_t *)NFA_DATA(tb[CTA_PROTO_ICMP_CODE-1]); tuple->src.u.icmp.id = - *(u_int16_t *)NFA_DATA(tb[CTA_PROTO_ICMP_ID-1]); + *(__be16 *)NFA_DATA(tb[CTA_PROTO_ICMP_ID-1]); if (tuple->dst.u.icmp.type >= sizeof(invmap) || !invmap[tuple->dst.u.icmp.type]) diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c index 1e8e700f613..3905cacc69a 100644 --- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c +++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c @@ -287,7 +287,7 @@ static int icmpv6_nfattr_to_tuple(struct nfattr *tb[], tuple->dst.u.icmp.code = *(u_int8_t *)NFA_DATA(tb[CTA_PROTO_ICMPV6_CODE-1]); tuple->src.u.icmp.id = - *(u_int16_t *)NFA_DATA(tb[CTA_PROTO_ICMPV6_ID-1]); + *(__be16 *)NFA_DATA(tb[CTA_PROTO_ICMPV6_ID-1]); if (tuple->dst.u.icmp.type < 128 || tuple->dst.u.icmp.type - 128 >= sizeof(invmap) diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c index bf93c1ea6be..37e5fca923a 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -72,7 +72,7 @@ struct nf_ct_frag6_queue struct hlist_node list; struct list_head lru_list; /* lru list member */ - __u32 id; /* fragment id */ + __be32 id; /* fragment id */ struct in6_addr saddr; struct in6_addr daddr; @@ -115,28 +115,28 @@ static __inline__ void fq_unlink(struct nf_ct_frag6_queue *fq) write_unlock(&nf_ct_frag6_lock); } -static unsigned int ip6qhashfn(u32 id, struct in6_addr *saddr, +static unsigned int ip6qhashfn(__be32 id, struct in6_addr *saddr, struct in6_addr *daddr) { u32 a, b, c; - a = saddr->s6_addr32[0]; - b = saddr->s6_addr32[1]; - c = saddr->s6_addr32[2]; + a = (__force u32)saddr->s6_addr32[0]; + b = (__force u32)saddr->s6_addr32[1]; + c = (__force u32)saddr->s6_addr32[2]; a += JHASH_GOLDEN_RATIO; b += JHASH_GOLDEN_RATIO; c += nf_ct_frag6_hash_rnd; __jhash_mix(a, b, c); - a += saddr->s6_addr32[3]; - b += daddr->s6_addr32[0]; - c += daddr->s6_addr32[1]; + a += (__force u32)saddr->s6_addr32[3]; + b += (__force u32)daddr->s6_addr32[0]; + c += (__force u32)daddr->s6_addr32[1]; __jhash_mix(a, b, c); - a += daddr->s6_addr32[2]; - b += daddr->s6_addr32[3]; - c += id; + a += (__force u32)daddr->s6_addr32[2]; + b += (__force u32)daddr->s6_addr32[3]; + c += (__force u32)id; __jhash_mix(a, b, c); return c & (FRAG6Q_HASHSZ - 1); @@ -338,7 +338,7 @@ static struct nf_ct_frag6_queue *nf_ct_frag6_intern(unsigned int hash, static struct nf_ct_frag6_queue * -nf_ct_frag6_create(unsigned int hash, u32 id, struct in6_addr *src, struct in6_addr *dst) +nf_ct_frag6_create(unsigned int hash, __be32 id, struct in6_addr *src, struct in6_addr *dst) { struct nf_ct_frag6_queue *fq; @@ -366,7 +366,7 @@ oom: } static __inline__ struct nf_ct_frag6_queue * -fq_find(u32 id, struct in6_addr *src, struct in6_addr *dst) +fq_find(__be32 id, struct in6_addr *src, struct in6_addr *dst) { struct nf_ct_frag6_queue *fq; struct hlist_node *n; diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index a6728067780..a401b1e3102 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -936,10 +936,8 @@ int nf_ct_port_nfattr_to_tuple(struct nfattr *tb[], if (nfattr_bad_size(tb, CTA_PROTO_MAX, cta_min_proto)) return -EINVAL; - t->src.u.tcp.port = - *(u_int16_t *)NFA_DATA(tb[CTA_PROTO_SRC_PORT-1]); - t->dst.u.tcp.port = - *(u_int16_t *)NFA_DATA(tb[CTA_PROTO_DST_PORT-1]); + t->src.u.tcp.port = *(__be16 *)NFA_DATA(tb[CTA_PROTO_SRC_PORT-1]); + t->dst.u.tcp.port = *(__be16 *)NFA_DATA(tb[CTA_PROTO_DST_PORT-1]); return 0; } diff --git a/net/netfilter/nf_conntrack_ftp.c b/net/netfilter/nf_conntrack_ftp.c index 2d2350152b9..fdac52beeb8 100644 --- a/net/netfilter/nf_conntrack_ftp.c +++ b/net/netfilter/nf_conntrack_ftp.c @@ -173,7 +173,7 @@ static int try_rfc959(const char *data, size_t dlen, /* Grab port: number up to delimiter */ static int get_port(const char *data, int start, size_t dlen, char delim, - u_int16_t *port) + __be16 *port) { u_int16_t tmp_port = 0; int i; @@ -502,12 +502,12 @@ static int help(struct sk_buff **pskb, .u = { .tcp = { 0 }}, }, .dst = { .protonum = 0xFF, - .u = { .tcp = { 0xFFFF }}, + .u = { .tcp = { __constant_htons(0xFFFF) }}, }, }; if (cmd.l3num == PF_INET) { - exp->mask.src.u3.ip = 0xFFFFFFFF; - exp->mask.dst.u3.ip = 0xFFFFFFFF; + exp->mask.src.u3.ip = htonl(0xFFFFFFFF); + exp->mask.dst.u3.ip = htonl(0xFFFFFFFF); } else { memset(exp->mask.src.u3.ip6, 0xFF, sizeof(exp->mask.src.u3.ip6)); @@ -588,7 +588,7 @@ static int __init nf_conntrack_ftp_init(void) ftp[i][j].tuple.src.u.tcp.port = htons(ports[i]); ftp[i][j].tuple.dst.protonum = IPPROTO_TCP; ftp[i][j].mask.src.l3num = 0xFFFF; - ftp[i][j].mask.src.u.tcp.port = 0xFFFF; + ftp[i][j].mask.src.u.tcp.port = htons(0xFFFF); ftp[i][j].mask.dst.protonum = 0xFF; ftp[i][j].max_expected = 1; ftp[i][j].timeout = 5 * 60; /* 5 Minutes */ diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index e3a72047212..a693d3bd4c1 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -113,7 +113,7 @@ ctnetlink_dump_tuples(struct sk_buff *skb, static inline int ctnetlink_dump_status(struct sk_buff *skb, const struct nf_conn *ct) { - u_int32_t status = htonl((u_int32_t) ct->status); + __be32 status = htonl((u_int32_t) ct->status); NFA_PUT(skb, CTA_STATUS, sizeof(status), &status); return 0; @@ -125,7 +125,7 @@ static inline int ctnetlink_dump_timeout(struct sk_buff *skb, const struct nf_conn *ct) { long timeout_l = ct->timeout.expires - jiffies; - u_int32_t timeout; + __be32 timeout; if (timeout_l < 0) timeout = 0; @@ -196,7 +196,7 @@ ctnetlink_dump_counters(struct sk_buff *skb, const struct nf_conn *ct, { enum ctattr_type type = dir ? CTA_COUNTERS_REPLY: CTA_COUNTERS_ORIG; struct nfattr *nest_count = NFA_NEST(skb, type); - u_int32_t tmp; + __be32 tmp; tmp = htonl(ct->counters[dir].packets); NFA_PUT(skb, CTA_COUNTERS32_PACKETS, sizeof(u_int32_t), &tmp); @@ -219,7 +219,7 @@ nfattr_failure: static inline int ctnetlink_dump_mark(struct sk_buff *skb, const struct nf_conn *ct) { - u_int32_t mark = htonl(ct->mark); + __be32 mark = htonl(ct->mark); NFA_PUT(skb, CTA_MARK, sizeof(u_int32_t), &mark); return 0; @@ -234,7 +234,7 @@ nfattr_failure: static inline int ctnetlink_dump_id(struct sk_buff *skb, const struct nf_conn *ct) { - u_int32_t id = htonl(ct->id); + __be32 id = htonl(ct->id); NFA_PUT(skb, CTA_ID, sizeof(u_int32_t), &id); return 0; @@ -245,7 +245,7 @@ nfattr_failure: static inline int ctnetlink_dump_use(struct sk_buff *skb, const struct nf_conn *ct) { - u_int32_t use = htonl(atomic_read(&ct->ct_general.use)); + __be32 use = htonl(atomic_read(&ct->ct_general.use)); NFA_PUT(skb, CTA_USE, sizeof(u_int32_t), &use); return 0; @@ -610,12 +610,12 @@ ctnetlink_parse_nat(struct nfattr *nat, return -EINVAL; if (tb[CTA_NAT_MINIP-1]) - range->min_ip = *(u_int32_t *)NFA_DATA(tb[CTA_NAT_MINIP-1]); + range->min_ip = *(__be32 *)NFA_DATA(tb[CTA_NAT_MINIP-1]); if (!tb[CTA_NAT_MAXIP-1]) range->max_ip = range->min_ip; else - range->max_ip = *(u_int32_t *)NFA_DATA(tb[CTA_NAT_MAXIP-1]); + range->max_ip = *(__be32 *)NFA_DATA(tb[CTA_NAT_MAXIP-1]); if (range->min_ip) range->flags |= IP_NAT_RANGE_MAP_IPS; @@ -688,7 +688,7 @@ ctnetlink_del_conntrack(struct sock *ctnl, struct sk_buff *skb, ct = nf_ct_tuplehash_to_ctrack(h); if (cda[CTA_ID-1]) { - u_int32_t id = ntohl(*(u_int32_t *)NFA_DATA(cda[CTA_ID-1])); + u_int32_t id = ntohl(*(__be32 *)NFA_DATA(cda[CTA_ID-1])); if (ct->id != id) { nf_ct_put(ct); return -ENOENT; @@ -781,7 +781,7 @@ static inline int ctnetlink_change_status(struct nf_conn *ct, struct nfattr *cda[]) { unsigned long d; - unsigned status = ntohl(*(u_int32_t *)NFA_DATA(cda[CTA_STATUS-1])); + unsigned int status = ntohl(*(__be32 *)NFA_DATA(cda[CTA_STATUS-1])); d = ct->status ^ status; if (d & (IPS_EXPECTED|IPS_CONFIRMED|IPS_DYING)) @@ -880,7 +880,7 @@ ctnetlink_change_helper(struct nf_conn *ct, struct nfattr *cda[]) static inline int ctnetlink_change_timeout(struct nf_conn *ct, struct nfattr *cda[]) { - u_int32_t timeout = ntohl(*(u_int32_t *)NFA_DATA(cda[CTA_TIMEOUT-1])); + u_int32_t timeout = ntohl(*(__be32 *)NFA_DATA(cda[CTA_TIMEOUT-1])); if (!del_timer(&ct->timeout)) return -ETIME; @@ -942,7 +942,7 @@ ctnetlink_change_conntrack(struct nf_conn *ct, struct nfattr *cda[]) #if defined(CONFIG_NF_CONNTRACK_MARK) if (cda[CTA_MARK-1]) - ct->mark = ntohl(*(u_int32_t *)NFA_DATA(cda[CTA_MARK-1])); + ct->mark = ntohl(*(__be32 *)NFA_DATA(cda[CTA_MARK-1])); #endif return 0; @@ -963,7 +963,7 @@ ctnetlink_create_conntrack(struct nfattr *cda[], if (!cda[CTA_TIMEOUT-1]) goto err; - ct->timeout.expires = ntohl(*(u_int32_t *)NFA_DATA(cda[CTA_TIMEOUT-1])); + ct->timeout.expires = ntohl(*(__be32 *)NFA_DATA(cda[CTA_TIMEOUT-1])); ct->timeout.expires = jiffies + ct->timeout.expires * HZ; ct->status |= IPS_CONFIRMED; @@ -982,7 +982,7 @@ ctnetlink_create_conntrack(struct nfattr *cda[], #if defined(CONFIG_NF_CONNTRACK_MARK) if (cda[CTA_MARK-1]) - ct->mark = ntohl(*(u_int32_t *)NFA_DATA(cda[CTA_MARK-1])); + ct->mark = ntohl(*(__be32 *)NFA_DATA(cda[CTA_MARK-1])); #endif help = nfct_help(ct); @@ -1117,8 +1117,8 @@ ctnetlink_exp_dump_expect(struct sk_buff *skb, const struct nf_conntrack_expect *exp) { struct nf_conn *master = exp->master; - u_int32_t timeout = htonl((exp->timeout.expires - jiffies) / HZ); - u_int32_t id = htonl(exp->id); + __be32 timeout = htonl((exp->timeout.expires - jiffies) / HZ); + __be32 id = htonl(exp->id); if (ctnetlink_exp_dump_tuple(skb, &exp->tuple, CTA_EXPECT_TUPLE) < 0) goto nfattr_failure; @@ -1296,7 +1296,7 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, return -ENOENT; if (cda[CTA_EXPECT_ID-1]) { - u_int32_t id = *(u_int32_t *)NFA_DATA(cda[CTA_EXPECT_ID-1]); + __be32 id = *(__be32 *)NFA_DATA(cda[CTA_EXPECT_ID-1]); if (exp->id != ntohl(id)) { nf_conntrack_expect_put(exp); return -ENOENT; @@ -1351,8 +1351,7 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, return -ENOENT; if (cda[CTA_EXPECT_ID-1]) { - u_int32_t id = - *(u_int32_t *)NFA_DATA(cda[CTA_EXPECT_ID-1]); + __be32 id = *(__be32 *)NFA_DATA(cda[CTA_EXPECT_ID-1]); if (exp->id != ntohl(id)) { nf_conntrack_expect_put(exp); return -ENOENT; diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c index 7c64b9c1cf4..02b9323e18b 100644 --- a/net/netfilter/nf_conntrack_proto_sctp.c +++ b/net/netfilter/nf_conntrack_proto_sctp.c @@ -217,7 +217,7 @@ static int sctp_print_conntrack(struct seq_file *s, for (offset = dataoff + sizeof(sctp_sctphdr_t), count = 0; \ offset < skb->len && \ (sch = skb_header_pointer(skb, offset, sizeof(_sch), &_sch)); \ - offset += (htons(sch->length) + 3) & ~3, count++) + offset += (ntohs(sch->length) + 3) & ~3, count++) /* Some validity checks to make sure the chunks are fine */ static int do_basic_checks(struct nf_conn *conntrack, diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index d99c7c4176d..71f492fc641 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -474,8 +474,8 @@ static void tcp_sack(const struct sk_buff *skb, unsigned int dataoff, /* Fast path for timestamp-only option */ if (length == TCPOLEN_TSTAMP_ALIGNED*4 - && *(__u32 *)ptr == - __constant_ntohl((TCPOPT_NOP << 24) + && *(__be32 *)ptr == + __constant_htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP)) @@ -506,9 +506,7 @@ static void tcp_sack(const struct sk_buff *skb, unsigned int dataoff, for (i = 0; i < (opsize - TCPOLEN_SACK_BASE); i += TCPOLEN_SACK_PERBLOCK) { - memcpy(&tmp, (__u32 *)(ptr + i) + 1, - sizeof(__u32)); - tmp = ntohl(tmp); + tmp = ntohl(*((__be32 *)(ptr+i)+1)); if (after(tmp, *sack)) *sack = tmp; -- cgit v1.2.3-70-g09d2 From 9457d851fc5df54522d733f72cbb1f02ab59272e Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 2 Dec 2006 22:05:25 -0800 Subject: [NETFILTER]: nf_conntrack: automatic helper assignment for expectations Some helpers (namely H.323) manually assign further helpers to expected connections. This is not possible with nf_conntrack anymore since we need to know whether a helper is used at allocation time. Handle the helper assignment centrally, which allows to perform the correct allocation and as a nice side effect eliminates the need for the H.323 helper to fiddle with nf_conntrack_lock. Mid term the allocation scheme really needs to be redesigned since we do both the helper and expectation lookup _twice_ for every new connection. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/net/netfilter/nf_conntrack_expect.h | 3 +++ net/netfilter/nf_conntrack_core.c | 19 ++++++++++++++----- net/netfilter/nf_conntrack_ftp.c | 1 + net/netfilter/nf_conntrack_helper.c | 3 ++- net/netfilter/nf_conntrack_netlink.c | 1 + 5 files changed, 21 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h index 2d335f024c8..5d853e826d1 100644 --- a/include/net/netfilter/nf_conntrack_expect.h +++ b/include/net/netfilter/nf_conntrack_expect.h @@ -22,6 +22,9 @@ struct nf_conntrack_expect void (*expectfn)(struct nf_conn *new, struct nf_conntrack_expect *this); + /* Helper to assign to new connection */ + struct nf_conntrack_helper *helper; + /* The conntrack of the master connection */ struct nf_conn *master; diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index a401b1e3102..f952a7fb6ae 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -545,10 +545,10 @@ static int early_drop(struct list_head *chain) static struct nf_conn * __nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, const struct nf_conntrack_tuple *repl, - const struct nf_conntrack_l3proto *l3proto) + const struct nf_conntrack_l3proto *l3proto, + u_int32_t features) { struct nf_conn *conntrack = NULL; - u_int32_t features = 0; struct nf_conntrack_helper *helper; if (unlikely(!nf_conntrack_hash_rnd_initted)) { @@ -574,7 +574,7 @@ __nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, } /* find features needed by this conntrack. */ - features = l3proto->get_features(orig); + features |= l3proto->get_features(orig); /* FIXME: protect helper list per RCU */ read_lock_bh(&nf_conntrack_lock); @@ -624,7 +624,7 @@ struct nf_conn *nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, struct nf_conntrack_l3proto *l3proto; l3proto = __nf_ct_l3proto_find(orig->src.l3num); - return __nf_conntrack_alloc(orig, repl, l3proto); + return __nf_conntrack_alloc(orig, repl, l3proto, 0); } void nf_conntrack_free(struct nf_conn *conntrack) @@ -649,13 +649,20 @@ init_conntrack(const struct nf_conntrack_tuple *tuple, struct nf_conn *conntrack; struct nf_conntrack_tuple repl_tuple; struct nf_conntrack_expect *exp; + u_int32_t features = 0; if (!nf_ct_invert_tuple(&repl_tuple, tuple, l3proto, l4proto)) { DEBUGP("Can't invert tuple.\n"); return NULL; } - conntrack = __nf_conntrack_alloc(tuple, &repl_tuple, l3proto); + read_lock_bh(&nf_conntrack_lock); + exp = __nf_conntrack_expect_find(tuple); + if (exp && exp->helper) + features = NF_CT_F_HELP; + read_unlock_bh(&nf_conntrack_lock); + + conntrack = __nf_conntrack_alloc(tuple, &repl_tuple, l3proto, features); if (conntrack == NULL || IS_ERR(conntrack)) { DEBUGP("Can't allocate conntrack.\n"); return (struct nf_conntrack_tuple_hash *)conntrack; @@ -676,6 +683,8 @@ init_conntrack(const struct nf_conntrack_tuple *tuple, /* Welcome, Mr. Bond. We've been expecting you... */ __set_bit(IPS_EXPECTED_BIT, &conntrack->status); conntrack->master = exp->master; + if (exp->helper) + nfct_help(conntrack)->helper = exp->helper; #ifdef CONFIG_NF_CONNTRACK_MARK conntrack->mark = exp->master->mark; #endif diff --git a/net/netfilter/nf_conntrack_ftp.c b/net/netfilter/nf_conntrack_ftp.c index fdac52beeb8..e96c41d17ee 100644 --- a/net/netfilter/nf_conntrack_ftp.c +++ b/net/netfilter/nf_conntrack_ftp.c @@ -516,6 +516,7 @@ static int help(struct sk_buff **pskb, } exp->expectfn = NULL; + exp->helper = NULL; exp->flags = 0; /* Now, NAT might want to mangle the packet, and register the diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index 2628f4ba35e..81542dc7506 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -129,7 +129,8 @@ void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me) /* Get rid of expectations */ list_for_each_entry_safe(exp, tmp, &nf_conntrack_expect_list, list) { struct nf_conn_help *help = nfct_help(exp->master); - if (help->helper == me && del_timer(&exp->timeout)) { + if ((help->helper == me || exp->helper == me) && + del_timer(&exp->timeout)) { nf_ct_unlink_expect(exp); nf_conntrack_expect_put(exp); } diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index a693d3bd4c1..acaef402595 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -1447,6 +1447,7 @@ ctnetlink_create_expect(struct nfattr *cda[], u_int8_t u3) exp->expectfn = NULL; exp->flags = 0; exp->master = ct; + exp->helper = NULL; memcpy(&exp->tuple, &tuple, sizeof(struct nf_conntrack_tuple)); memcpy(&exp->mask, &mask, sizeof(struct nf_conntrack_tuple)); -- cgit v1.2.3-70-g09d2 From 5b1158e909ecbe1a052203e0d8df15633f829930 Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Sat, 2 Dec 2006 22:07:13 -0800 Subject: [NETFILTER]: Add NAT support for nf_conntrack Add NAT support for nf_conntrack. Joint work of Jozsef Kadlecsik, Yasuyuki Kozakai, Martin Josefsson and myself. Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter.h | 2 +- include/net/netfilter/ipv4/nf_conntrack_ipv4.h | 20 +- include/net/netfilter/nf_conntrack.h | 29 +- include/net/netfilter/nf_conntrack_expect.h | 2 +- include/net/netfilter/nf_nat.h | 77 +++ include/net/netfilter/nf_nat_core.h | 27 ++ include/net/netfilter/nf_nat_helper.h | 32 ++ include/net/netfilter/nf_nat_protocol.h | 70 +++ include/net/netfilter/nf_nat_rule.h | 35 ++ net/ipv4/netfilter/Kconfig | 30 +- net/ipv4/netfilter/Makefile | 7 + net/ipv4/netfilter/ip_nat_standalone.c | 6 - net/ipv4/netfilter/ipt_MASQUERADE.c | 29 +- net/ipv4/netfilter/ipt_NETMAP.c | 4 + net/ipv4/netfilter/ipt_REDIRECT.c | 6 +- net/ipv4/netfilter/ipt_SAME.c | 12 +- net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 7 +- net/ipv4/netfilter/nf_nat_core.c | 647 +++++++++++++++++++++++++ net/ipv4/netfilter/nf_nat_helper.c | 433 +++++++++++++++++ net/ipv4/netfilter/nf_nat_proto_icmp.c | 86 ++++ net/ipv4/netfilter/nf_nat_proto_tcp.c | 148 ++++++ net/ipv4/netfilter/nf_nat_proto_udp.c | 138 ++++++ net/ipv4/netfilter/nf_nat_proto_unknown.c | 54 +++ net/ipv4/netfilter/nf_nat_rule.c | 343 +++++++++++++ net/ipv4/netfilter/nf_nat_standalone.c | 406 ++++++++++++++++ net/netfilter/nf_conntrack_core.c | 23 +- net/netfilter/nf_conntrack_netlink.c | 46 +- net/netfilter/nf_conntrack_proto_tcp.c | 2 +- net/netfilter/nf_conntrack_standalone.c | 5 +- 29 files changed, 2667 insertions(+), 59 deletions(-) create mode 100644 include/net/netfilter/nf_nat.h create mode 100644 include/net/netfilter/nf_nat_core.h create mode 100644 include/net/netfilter/nf_nat_helper.h create mode 100644 include/net/netfilter/nf_nat_protocol.h create mode 100644 include/net/netfilter/nf_nat_rule.h create mode 100644 net/ipv4/netfilter/nf_nat_core.c create mode 100644 net/ipv4/netfilter/nf_nat_helper.c create mode 100644 net/ipv4/netfilter/nf_nat_proto_icmp.c create mode 100644 net/ipv4/netfilter/nf_nat_proto_tcp.c create mode 100644 net/ipv4/netfilter/nf_nat_proto_udp.c create mode 100644 net/ipv4/netfilter/nf_nat_proto_unknown.c create mode 100644 net/ipv4/netfilter/nf_nat_rule.c create mode 100644 net/ipv4/netfilter/nf_nat_standalone.c (limited to 'include') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index f6f3fcbd70e..d4c4c5120bc 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -357,7 +357,7 @@ extern void (*ip_nat_decode_session)(struct sk_buff *, struct flowi *); static inline void nf_nat_decode_session(struct sk_buff *skb, struct flowi *fl, int family) { -#ifdef CONFIG_IP_NF_NAT_NEEDED +#if defined(CONFIG_IP_NF_NAT_NEEDED) || defined(CONFIG_NF_NAT_NEEDED) void (*decodefn)(struct sk_buff *, struct flowi *); if (family == AF_INET && (decodefn = ip_nat_decode_session) != NULL) diff --git a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h index 34b453a81a6..a1c57ee0a4f 100644 --- a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h +++ b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h @@ -9,29 +9,23 @@ #ifndef _NF_CONNTRACK_IPV4_H #define _NF_CONNTRACK_IPV4_H -#ifdef CONFIG_IP_NF_NAT_NEEDED -#include +#ifdef CONFIG_NF_NAT_NEEDED +#include /* per conntrack: nat application helper private data */ -union ip_conntrack_nat_help { +union nf_conntrack_nat_help { /* insert nat helper private data here */ }; -struct nf_conntrack_ipv4_nat { - struct ip_nat_info info; - union ip_conntrack_nat_help help; +struct nf_conn_nat { + struct nf_nat_info info; + union nf_conntrack_nat_help help; #if defined(CONFIG_IP_NF_TARGET_MASQUERADE) || \ defined(CONFIG_IP_NF_TARGET_MASQUERADE_MODULE) int masq_index; #endif }; -#endif /* CONFIG_IP_NF_NAT_NEEDED */ - -struct nf_conntrack_ipv4 { -#ifdef CONFIG_IP_NF_NAT_NEEDED - struct nf_conntrack_ipv4_nat *nat; -#endif -}; +#endif /* CONFIG_NF_NAT_NEEDED */ /* Returns new sk_buff, or NULL */ struct sk_buff * diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index b4beb8c799e..9948af06868 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -264,18 +264,45 @@ nf_conntrack_unregister_cache(u_int32_t features); /* valid combinations: * basic: nf_conn, nf_conn .. nf_conn_help - * nat: nf_conn .. nf_conn_nat, nf_conn .. nf_conn_nat, nf_conn help + * nat: nf_conn .. nf_conn_nat, nf_conn .. nf_conn_nat .. nf_conn help */ +#ifdef CONFIG_NF_NAT_NEEDED +static inline struct nf_conn_nat *nfct_nat(const struct nf_conn *ct) +{ + unsigned int offset = sizeof(struct nf_conn); + + if (!(ct->features & NF_CT_F_NAT)) + return NULL; + + offset = ALIGN(offset, __alignof__(struct nf_conn_nat)); + return (struct nf_conn_nat *) ((void *)ct + offset); +} + static inline struct nf_conn_help *nfct_help(const struct nf_conn *ct) { unsigned int offset = sizeof(struct nf_conn); if (!(ct->features & NF_CT_F_HELP)) return NULL; + if (ct->features & NF_CT_F_NAT) { + offset = ALIGN(offset, __alignof__(struct nf_conn_nat)); + offset += sizeof(struct nf_conn_nat); + } offset = ALIGN(offset, __alignof__(struct nf_conn_help)); return (struct nf_conn_help *) ((void *)ct + offset); } +#else /* No NAT */ +static inline struct nf_conn_help *nfct_help(const struct nf_conn *ct) +{ + unsigned int offset = sizeof(struct nf_conn); + + if (!(ct->features & NF_CT_F_HELP)) + return NULL; + offset = ALIGN(offset, __alignof__(struct nf_conn_help)); + return (struct nf_conn_help *) ((void *)ct + offset); +} +#endif /* CONFIG_NF_NAT_NEEDED */ #endif /* __KERNEL__ */ #endif /* _NF_CONNTRACK_H */ diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h index 5d853e826d1..b969c430b36 100644 --- a/include/net/netfilter/nf_conntrack_expect.h +++ b/include/net/netfilter/nf_conntrack_expect.h @@ -43,7 +43,7 @@ struct nf_conntrack_expect #ifdef CONFIG_NF_NAT_NEEDED /* This is the original per-proto part, used to map the * expected connection the way the recipient expects. */ - union nf_conntrack_manip_proto saved_proto; + union nf_conntrack_man_proto saved_proto; /* Direction relative to the master connection. */ enum ip_conntrack_dir dir; #endif diff --git a/include/net/netfilter/nf_nat.h b/include/net/netfilter/nf_nat.h new file mode 100644 index 00000000000..61c62068ca6 --- /dev/null +++ b/include/net/netfilter/nf_nat.h @@ -0,0 +1,77 @@ +#ifndef _NF_NAT_H +#define _NF_NAT_H +#include +#include + +#define NF_NAT_MAPPING_TYPE_MAX_NAMELEN 16 + +enum nf_nat_manip_type +{ + IP_NAT_MANIP_SRC, + IP_NAT_MANIP_DST +}; + +/* SRC manip occurs POST_ROUTING or LOCAL_IN */ +#define HOOK2MANIP(hooknum) ((hooknum) != NF_IP_POST_ROUTING && (hooknum) != NF_IP_LOCAL_IN) + +#define IP_NAT_RANGE_MAP_IPS 1 +#define IP_NAT_RANGE_PROTO_SPECIFIED 2 + +/* NAT sequence number modifications */ +struct nf_nat_seq { + /* position of the last TCP sequence number modification (if any) */ + u_int32_t correction_pos; + + /* sequence number offset before and after last modification */ + int16_t offset_before, offset_after; +}; + +/* Single range specification. */ +struct nf_nat_range +{ + /* Set to OR of flags above. */ + unsigned int flags; + + /* Inclusive: network order. */ + __be32 min_ip, max_ip; + + /* Inclusive: network order */ + union nf_conntrack_man_proto min, max; +}; + +/* For backwards compat: don't use in modern code. */ +struct nf_nat_multi_range_compat +{ + unsigned int rangesize; /* Must be 1. */ + + /* hangs off end. */ + struct nf_nat_range range[1]; +}; + +#ifdef __KERNEL__ +#include + +/* The structure embedded in the conntrack structure. */ +struct nf_nat_info +{ + struct list_head bysource; + struct nf_nat_seq seq[IP_CT_DIR_MAX]; +}; + +struct nf_conn; + +/* Set up the info structure to map into this range. */ +extern unsigned int nf_nat_setup_info(struct nf_conn *ct, + const struct nf_nat_range *range, + unsigned int hooknum); + +/* Is this tuple already taken? (not by us)*/ +extern int nf_nat_used_tuple(const struct nf_conntrack_tuple *tuple, + const struct nf_conn *ignored_conntrack); + +extern int nf_nat_module_is_loaded; + +#else /* !__KERNEL__: iptables wants this to compile. */ +#define nf_nat_multi_range nf_nat_multi_range_compat +#endif /*__KERNEL__*/ +#endif diff --git a/include/net/netfilter/nf_nat_core.h b/include/net/netfilter/nf_nat_core.h new file mode 100644 index 00000000000..9778ffa9344 --- /dev/null +++ b/include/net/netfilter/nf_nat_core.h @@ -0,0 +1,27 @@ +#ifndef _NF_NAT_CORE_H +#define _NF_NAT_CORE_H +#include +#include + +/* This header used to share core functionality between the standalone + NAT module, and the compatibility layer's use of NAT for masquerading. */ + +extern unsigned int nf_nat_packet(struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int hooknum, + struct sk_buff **pskb); + +extern int nf_nat_icmp_reply_translation(struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int hooknum, + struct sk_buff **pskb); + +static inline int nf_nat_initialized(struct nf_conn *ct, + enum nf_nat_manip_type manip) +{ + if (manip == IP_NAT_MANIP_SRC) + return test_bit(IPS_SRC_NAT_DONE_BIT, &ct->status); + else + return test_bit(IPS_DST_NAT_DONE_BIT, &ct->status); +} +#endif /* _NF_NAT_CORE_H */ diff --git a/include/net/netfilter/nf_nat_helper.h b/include/net/netfilter/nf_nat_helper.h new file mode 100644 index 00000000000..ec98ecf95fc --- /dev/null +++ b/include/net/netfilter/nf_nat_helper.h @@ -0,0 +1,32 @@ +#ifndef _NF_NAT_HELPER_H +#define _NF_NAT_HELPER_H +/* NAT protocol helper routines. */ + +#include + +struct sk_buff; + +/* These return true or false. */ +extern int nf_nat_mangle_tcp_packet(struct sk_buff **skb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int match_offset, + unsigned int match_len, + const char *rep_buffer, + unsigned int rep_len); +extern int nf_nat_mangle_udp_packet(struct sk_buff **skb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int match_offset, + unsigned int match_len, + const char *rep_buffer, + unsigned int rep_len); +extern int nf_nat_seq_adjust(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo); + +/* Setup NAT on this expected conntrack so it follows master, but goes + * to port ct->master->saved_proto. */ +extern void nf_nat_follow_master(struct nf_conn *ct, + struct nf_conntrack_expect *this); +#endif diff --git a/include/net/netfilter/nf_nat_protocol.h b/include/net/netfilter/nf_nat_protocol.h new file mode 100644 index 00000000000..a9ec5ef6146 --- /dev/null +++ b/include/net/netfilter/nf_nat_protocol.h @@ -0,0 +1,70 @@ +/* Header for use in defining a given protocol. */ +#ifndef _NF_NAT_PROTOCOL_H +#define _NF_NAT_PROTOCOL_H +#include +#include + +struct nf_nat_range; + +struct nf_nat_protocol +{ + /* Protocol name */ + const char *name; + + /* Protocol number. */ + unsigned int protonum; + + struct module *me; + + /* Translate a packet to the target according to manip type. + Return true if succeeded. */ + int (*manip_pkt)(struct sk_buff **pskb, + unsigned int iphdroff, + const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype); + + /* Is the manipable part of the tuple between min and max incl? */ + int (*in_range)(const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype, + const union nf_conntrack_man_proto *min, + const union nf_conntrack_man_proto *max); + + /* Alter the per-proto part of the tuple (depending on + maniptype), to give a unique tuple in the given range if + possible; return false if not. Per-protocol part of tuple + is initialized to the incoming packet. */ + int (*unique_tuple)(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + enum nf_nat_manip_type maniptype, + const struct nf_conn *ct); + + int (*range_to_nfattr)(struct sk_buff *skb, + const struct nf_nat_range *range); + + int (*nfattr_to_range)(struct nfattr *tb[], + struct nf_nat_range *range); +}; + +/* Protocol registration. */ +extern int nf_nat_protocol_register(struct nf_nat_protocol *proto); +extern void nf_nat_protocol_unregister(struct nf_nat_protocol *proto); + +extern struct nf_nat_protocol *nf_nat_proto_find_get(u_int8_t protocol); +extern void nf_nat_proto_put(struct nf_nat_protocol *proto); + +/* Built-in protocols. */ +extern struct nf_nat_protocol nf_nat_protocol_tcp; +extern struct nf_nat_protocol nf_nat_protocol_udp; +extern struct nf_nat_protocol nf_nat_protocol_icmp; +extern struct nf_nat_protocol nf_nat_unknown_protocol; + +extern int init_protocols(void) __init; +extern void cleanup_protocols(void); +extern struct nf_nat_protocol *find_nat_proto(u_int16_t protonum); + +extern int nf_nat_port_range_to_nfattr(struct sk_buff *skb, + const struct nf_nat_range *range); +extern int nf_nat_port_nfattr_to_range(struct nfattr *tb[], + struct nf_nat_range *range); + +#endif /*_NF_NAT_PROTO_H*/ diff --git a/include/net/netfilter/nf_nat_rule.h b/include/net/netfilter/nf_nat_rule.h new file mode 100644 index 00000000000..f191c672bcc --- /dev/null +++ b/include/net/netfilter/nf_nat_rule.h @@ -0,0 +1,35 @@ +#ifndef _NF_NAT_RULE_H +#define _NF_NAT_RULE_H +#include +#include +#include + +/* Compatibility definitions for ipt_FOO modules */ +#define ip_nat_range nf_nat_range +#define ip_conntrack_tuple nf_conntrack_tuple +#define ip_conntrack_get nf_ct_get +#define ip_conntrack nf_conn +#define ip_nat_setup_info nf_nat_setup_info +#define ip_nat_multi_range_compat nf_nat_multi_range_compat +#define ip_ct_iterate_cleanup nf_ct_iterate_cleanup +#define IP_NF_ASSERT NF_CT_ASSERT + +extern int nf_nat_rule_init(void) __init; +extern void nf_nat_rule_cleanup(void); +extern int nf_nat_rule_find(struct sk_buff **pskb, + unsigned int hooknum, + const struct net_device *in, + const struct net_device *out, + struct nf_conn *ct, + struct nf_nat_info *info); + +extern unsigned int +alloc_null_binding(struct nf_conn *ct, + struct nf_nat_info *info, + unsigned int hooknum); + +extern unsigned int +alloc_null_binding_confirmed(struct nf_conn *ct, + struct nf_nat_info *info, + unsigned int hooknum); +#endif /* _NF_NAT_RULE_H */ diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 71485276b81..01789aeaeb5 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -6,7 +6,7 @@ menu "IP: Netfilter Configuration" depends on INET && NETFILTER config NF_CONNTRACK_IPV4 - tristate "IPv4 support for new connection tracking (EXPERIMENTAL)" + tristate "IPv4 support for new connection tracking (required for NAT) (EXPERIMENTAL)" depends on EXPERIMENTAL && NF_CONNTRACK ---help--- Connection tracking keeps a record of what packets have passed @@ -387,7 +387,7 @@ config IP_NF_TARGET_TCPMSS To compile it as a module, choose M here. If unsure, say N. -# NAT + specific targets +# NAT + specific targets: ip_conntrack config IP_NF_NAT tristate "Full NAT" depends on IP_NF_IPTABLES && IP_NF_CONNTRACK @@ -398,14 +398,30 @@ config IP_NF_NAT To compile it as a module, choose M here. If unsure, say N. +# NAT + specific targets: nf_conntrack +config NF_NAT + tristate "Full NAT" + depends on IP_NF_IPTABLES && NF_CONNTRACK + help + The Full NAT option allows masquerading, port forwarding and other + forms of full Network Address Port Translation. It is controlled by + the `nat' table in iptables: see the man page for iptables(8). + + To compile it as a module, choose M here. If unsure, say N. + config IP_NF_NAT_NEEDED bool - depends on IP_NF_NAT != n + depends on IP_NF_NAT + default y + +config NF_NAT_NEEDED + bool + depends on NF_NAT default y config IP_NF_TARGET_MASQUERADE tristate "MASQUERADE target support" - depends on IP_NF_NAT + depends on (NF_NAT || IP_NF_NAT) help Masquerading is a special case of NAT: all outgoing connections are changed to seem to come from a particular interface's address, and @@ -417,7 +433,7 @@ config IP_NF_TARGET_MASQUERADE config IP_NF_TARGET_REDIRECT tristate "REDIRECT target support" - depends on IP_NF_NAT + depends on (NF_NAT || IP_NF_NAT) help REDIRECT is a special case of NAT: all incoming connections are mapped onto the incoming interface's address, causing the packets to @@ -428,7 +444,7 @@ config IP_NF_TARGET_REDIRECT config IP_NF_TARGET_NETMAP tristate "NETMAP target support" - depends on IP_NF_NAT + depends on (NF_NAT || IP_NF_NAT) help NETMAP is an implementation of static 1:1 NAT mapping of network addresses. It maps the network address part, while keeping the host @@ -439,7 +455,7 @@ config IP_NF_TARGET_NETMAP config IP_NF_TARGET_SAME tristate "SAME target support" - depends on IP_NF_NAT + depends on (NF_NAT || IP_NF_NAT) help This option adds a `SAME' target, which works like the standard SNAT target, but attempts to give clients the same IP for all connections. diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index 21359d83f0c..ec31690764a 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -5,7 +5,12 @@ # objects for the standalone - connection tracking / NAT ip_conntrack-objs := ip_conntrack_standalone.o ip_conntrack_core.o ip_conntrack_proto_generic.o ip_conntrack_proto_tcp.o ip_conntrack_proto_udp.o ip_conntrack_proto_icmp.o ip_nat-objs := ip_nat_core.o ip_nat_helper.o ip_nat_proto_unknown.o ip_nat_proto_tcp.o ip_nat_proto_udp.o ip_nat_proto_icmp.o +nf_nat-objs := nf_nat_core.o nf_nat_helper.o nf_nat_proto_unknown.o nf_nat_proto_tcp.o nf_nat_proto_udp.o nf_nat_proto_icmp.o +ifneq ($(CONFIG_NF_NAT),) +iptable_nat-objs := nf_nat_rule.o nf_nat_standalone.o +else iptable_nat-objs := ip_nat_rule.o ip_nat_standalone.o +endif ip_conntrack_pptp-objs := ip_conntrack_helper_pptp.o ip_conntrack_proto_gre.o ip_nat_pptp-objs := ip_nat_helper_pptp.o ip_nat_proto_gre.o @@ -16,6 +21,7 @@ ip_nat_h323-objs := ip_nat_helper_h323.o # connection tracking obj-$(CONFIG_IP_NF_CONNTRACK) += ip_conntrack.o obj-$(CONFIG_IP_NF_NAT) += ip_nat.o +obj-$(CONFIG_NF_NAT) += nf_nat.o # conntrack netlink interface obj-$(CONFIG_IP_NF_CONNTRACK_NETLINK) += ip_conntrack_netlink.o @@ -50,6 +56,7 @@ obj-$(CONFIG_IP_NF_IPTABLES) += ip_tables.o obj-$(CONFIG_IP_NF_FILTER) += iptable_filter.o obj-$(CONFIG_IP_NF_MANGLE) += iptable_mangle.o obj-$(CONFIG_IP_NF_NAT) += iptable_nat.o +obj-$(CONFIG_NF_NAT) += iptable_nat.o obj-$(CONFIG_IP_NF_RAW) += iptable_raw.o # matches diff --git a/net/ipv4/netfilter/ip_nat_standalone.c b/net/ipv4/netfilter/ip_nat_standalone.c index d85d2de5044..ad66328baa5 100644 --- a/net/ipv4/netfilter/ip_nat_standalone.c +++ b/net/ipv4/netfilter/ip_nat_standalone.c @@ -44,12 +44,6 @@ #define DEBUGP(format, args...) #endif -#define HOOKNAME(hooknum) ((hooknum) == NF_IP_POST_ROUTING ? "POST_ROUTING" \ - : ((hooknum) == NF_IP_PRE_ROUTING ? "PRE_ROUTING" \ - : ((hooknum) == NF_IP_LOCAL_OUT ? "LOCAL_OUT" \ - : ((hooknum) == NF_IP_LOCAL_IN ? "LOCAL_IN" \ - : "*ERROR*"))) - #ifdef CONFIG_XFRM static void nat_decode_session(struct sk_buff *skb, struct flowi *fl) { diff --git a/net/ipv4/netfilter/ipt_MASQUERADE.c b/net/ipv4/netfilter/ipt_MASQUERADE.c index 3dbfcfac8a8..28b9233956b 100644 --- a/net/ipv4/netfilter/ipt_MASQUERADE.c +++ b/net/ipv4/netfilter/ipt_MASQUERADE.c @@ -2,7 +2,7 @@ (depending on route). */ /* (C) 1999-2001 Paul `Rusty' Russell - * (C) 2002-2004 Netfilter Core Team + * (C) 2002-2006 Netfilter Core Team * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -20,7 +20,11 @@ #include #include #include +#ifdef CONFIG_NF_NAT_NEEDED +#include +#else #include +#endif #include MODULE_LICENSE("GPL"); @@ -65,23 +69,33 @@ masquerade_target(struct sk_buff **pskb, const struct xt_target *target, const void *targinfo) { +#ifdef CONFIG_NF_NAT_NEEDED + struct nf_conn_nat *nat; +#endif struct ip_conntrack *ct; enum ip_conntrack_info ctinfo; - const struct ip_nat_multi_range_compat *mr; struct ip_nat_range newrange; + const struct ip_nat_multi_range_compat *mr; struct rtable *rt; __be32 newsrc; IP_NF_ASSERT(hooknum == NF_IP_POST_ROUTING); ct = ip_conntrack_get(*pskb, &ctinfo); +#ifdef CONFIG_NF_NAT_NEEDED + nat = nfct_nat(ct); +#endif IP_NF_ASSERT(ct && (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED || ctinfo == IP_CT_RELATED + IP_CT_IS_REPLY)); /* Source address is 0.0.0.0 - locally generated packet that is * probably not supposed to be masqueraded. */ +#ifdef CONFIG_NF_NAT_NEEDED + if (ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip == 0) +#else if (ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.ip == 0) +#endif return NF_ACCEPT; mr = targinfo; @@ -93,7 +107,11 @@ masquerade_target(struct sk_buff **pskb, } write_lock_bh(&masq_lock); +#ifdef CONFIG_NF_NAT_NEEDED + nat->masq_index = out->ifindex; +#else ct->nat.masq_index = out->ifindex; +#endif write_unlock_bh(&masq_lock); /* Transfer from original range. */ @@ -109,10 +127,17 @@ masquerade_target(struct sk_buff **pskb, static inline int device_cmp(struct ip_conntrack *i, void *ifindex) { +#ifdef CONFIG_NF_NAT_NEEDED + struct nf_conn_nat *nat = nfct_nat(i); +#endif int ret; read_lock_bh(&masq_lock); +#ifdef CONFIG_NF_NAT_NEEDED + ret = (nat->masq_index == (int)(long)ifindex); +#else ret = (i->nat.masq_index == (int)(long)ifindex); +#endif read_unlock_bh(&masq_lock); return ret; diff --git a/net/ipv4/netfilter/ipt_NETMAP.c b/net/ipv4/netfilter/ipt_NETMAP.c index 58a88f22710..9390e90f2b2 100644 --- a/net/ipv4/netfilter/ipt_NETMAP.c +++ b/net/ipv4/netfilter/ipt_NETMAP.c @@ -15,7 +15,11 @@ #include #include #include +#ifdef CONFIG_NF_NAT_NEEDED +#include +#else #include +#endif #define MODULENAME "NETMAP" MODULE_LICENSE("GPL"); diff --git a/net/ipv4/netfilter/ipt_REDIRECT.c b/net/ipv4/netfilter/ipt_REDIRECT.c index c0dcfe9d610..462eceb3a1b 100644 --- a/net/ipv4/netfilter/ipt_REDIRECT.c +++ b/net/ipv4/netfilter/ipt_REDIRECT.c @@ -1,6 +1,6 @@ /* Redirect. Simple mapping which alters dst to a local IP address. */ /* (C) 1999-2001 Paul `Rusty' Russell - * (C) 2002-2004 Netfilter Core Team + * (C) 2002-2006 Netfilter Core Team * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -18,7 +18,11 @@ #include #include #include +#ifdef CONFIG_NF_NAT_NEEDED +#include +#else #include +#endif MODULE_LICENSE("GPL"); MODULE_AUTHOR("Netfilter Core Team "); diff --git a/net/ipv4/netfilter/ipt_SAME.c b/net/ipv4/netfilter/ipt_SAME.c index b38b13328d7..3dcf2941133 100644 --- a/net/ipv4/netfilter/ipt_SAME.c +++ b/net/ipv4/netfilter/ipt_SAME.c @@ -34,7 +34,11 @@ #include #include #include +#ifdef CONFIG_NF_NAT_NEEDED +#include +#else #include +#endif #include MODULE_LICENSE("GPL"); @@ -152,11 +156,17 @@ same_target(struct sk_buff **pskb, Here we calculate the index in same->iparray which holds the ipaddress we should use */ +#ifdef CONFIG_NF_NAT_NEEDED + tmpip = ntohl(t->src.u3.ip); + + if (!(same->info & IPT_SAME_NODST)) + tmpip += ntohl(t->dst.u3.ip); +#else tmpip = ntohl(t->src.ip); if (!(same->info & IPT_SAME_NODST)) tmpip += ntohl(t->dst.ip); - +#endif aindex = tmpip % same->ipnum; new_ip = htonl(same->iparray[aindex]); diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c index 653f57b8a10..5655109dcaf 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c @@ -111,10 +111,10 @@ ipv4_prepare(struct sk_buff **pskb, unsigned int hooknum, unsigned int *dataoff, return NF_ACCEPT; } -int nat_module_is_loaded = 0; +int nf_nat_module_is_loaded = 0; static u_int32_t ipv4_get_features(const struct nf_conntrack_tuple *tuple) { - if (nat_module_is_loaded) + if (nf_nat_module_is_loaded) return NF_CT_F_NAT; return NF_CT_F_BASIC; @@ -532,3 +532,6 @@ module_init(nf_conntrack_l3proto_ipv4_init); module_exit(nf_conntrack_l3proto_ipv4_fini); EXPORT_SYMBOL(nf_ct_ipv4_gather_frags); +#ifdef CONFIG_NF_NAT_NEEDED +EXPORT_SYMBOL(nf_nat_module_is_loaded); +#endif diff --git a/net/ipv4/netfilter/nf_nat_core.c b/net/ipv4/netfilter/nf_nat_core.c new file mode 100644 index 00000000000..86a92272b05 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_core.c @@ -0,0 +1,647 @@ +/* NAT for netfilter; shared with compatibility layer. */ + +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include /* For tcp_prot in getorigdst */ +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +static DEFINE_RWLOCK(nf_nat_lock); + +static struct nf_conntrack_l3proto *l3proto = NULL; + +/* Calculated at init based on memory size */ +static unsigned int nf_nat_htable_size; + +static struct list_head *bysource; + +#define MAX_IP_NAT_PROTO 256 +static struct nf_nat_protocol *nf_nat_protos[MAX_IP_NAT_PROTO]; + +static inline struct nf_nat_protocol * +__nf_nat_proto_find(u_int8_t protonum) +{ + return nf_nat_protos[protonum]; +} + +struct nf_nat_protocol * +nf_nat_proto_find_get(u_int8_t protonum) +{ + struct nf_nat_protocol *p; + + /* we need to disable preemption to make sure 'p' doesn't get + * removed until we've grabbed the reference */ + preempt_disable(); + p = __nf_nat_proto_find(protonum); + if (!try_module_get(p->me)) + p = &nf_nat_unknown_protocol; + preempt_enable(); + + return p; +} +EXPORT_SYMBOL_GPL(nf_nat_proto_find_get); + +void +nf_nat_proto_put(struct nf_nat_protocol *p) +{ + module_put(p->me); +} +EXPORT_SYMBOL_GPL(nf_nat_proto_put); + +/* We keep an extra hash for each conntrack, for fast searching. */ +static inline unsigned int +hash_by_src(const struct nf_conntrack_tuple *tuple) +{ + /* Original src, to ensure we map it consistently if poss. */ + return jhash_3words((__force u32)tuple->src.u3.ip, tuple->src.u.all, + tuple->dst.protonum, 0) % nf_nat_htable_size; +} + +/* Noone using conntrack by the time this called. */ +static void nf_nat_cleanup_conntrack(struct nf_conn *conn) +{ + struct nf_conn_nat *nat; + if (!(conn->status & IPS_NAT_DONE_MASK)) + return; + + nat = nfct_nat(conn); + write_lock_bh(&nf_nat_lock); + list_del(&nat->info.bysource); + write_unlock_bh(&nf_nat_lock); +} + +/* Is this tuple already taken? (not by us) */ +int +nf_nat_used_tuple(const struct nf_conntrack_tuple *tuple, + const struct nf_conn *ignored_conntrack) +{ + /* Conntrack tracking doesn't keep track of outgoing tuples; only + incoming ones. NAT means they don't have a fixed mapping, + so we invert the tuple and look for the incoming reply. + + We could keep a separate hash if this proves too slow. */ + struct nf_conntrack_tuple reply; + + nf_ct_invert_tuplepr(&reply, tuple); + return nf_conntrack_tuple_taken(&reply, ignored_conntrack); +} +EXPORT_SYMBOL(nf_nat_used_tuple); + +/* If we source map this tuple so reply looks like reply_tuple, will + * that meet the constraints of range. */ +static int +in_range(const struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range) +{ + struct nf_nat_protocol *proto; + + proto = __nf_nat_proto_find(tuple->dst.protonum); + /* If we are supposed to map IPs, then we must be in the + range specified, otherwise let this drag us onto a new src IP. */ + if (range->flags & IP_NAT_RANGE_MAP_IPS) { + if (ntohl(tuple->src.u3.ip) < ntohl(range->min_ip) || + ntohl(tuple->src.u3.ip) > ntohl(range->max_ip)) + return 0; + } + + if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED) || + proto->in_range(tuple, IP_NAT_MANIP_SRC, + &range->min, &range->max)) + return 1; + + return 0; +} + +static inline int +same_src(const struct nf_conn *ct, + const struct nf_conntrack_tuple *tuple) +{ + const struct nf_conntrack_tuple *t; + + t = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; + return (t->dst.protonum == tuple->dst.protonum && + t->src.u3.ip == tuple->src.u3.ip && + t->src.u.all == tuple->src.u.all); +} + +/* Only called for SRC manip */ +static int +find_appropriate_src(const struct nf_conntrack_tuple *tuple, + struct nf_conntrack_tuple *result, + const struct nf_nat_range *range) +{ + unsigned int h = hash_by_src(tuple); + struct nf_conn_nat *nat; + struct nf_conn *ct; + + read_lock_bh(&nf_nat_lock); + list_for_each_entry(nat, &bysource[h], info.bysource) { + ct = (struct nf_conn *)((char *)nat - offsetof(struct nf_conn, data)); + if (same_src(ct, tuple)) { + /* Copy source part from reply tuple. */ + nf_ct_invert_tuplepr(result, + &ct->tuplehash[IP_CT_DIR_REPLY].tuple); + result->dst = tuple->dst; + + if (in_range(result, range)) { + read_unlock_bh(&nf_nat_lock); + return 1; + } + } + } + read_unlock_bh(&nf_nat_lock); + return 0; +} + +/* For [FUTURE] fragmentation handling, we want the least-used + src-ip/dst-ip/proto triple. Fairness doesn't come into it. Thus + if the range specifies 1.2.3.4 ports 10000-10005 and 1.2.3.5 ports + 1-65535, we don't do pro-rata allocation based on ports; we choose + the ip with the lowest src-ip/dst-ip/proto usage. +*/ +static void +find_best_ips_proto(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + const struct nf_conn *ct, + enum nf_nat_manip_type maniptype) +{ + __be32 *var_ipp; + /* Host order */ + u_int32_t minip, maxip, j; + + /* No IP mapping? Do nothing. */ + if (!(range->flags & IP_NAT_RANGE_MAP_IPS)) + return; + + if (maniptype == IP_NAT_MANIP_SRC) + var_ipp = &tuple->src.u3.ip; + else + var_ipp = &tuple->dst.u3.ip; + + /* Fast path: only one choice. */ + if (range->min_ip == range->max_ip) { + *var_ipp = range->min_ip; + return; + } + + /* Hashing source and destination IPs gives a fairly even + * spread in practice (if there are a small number of IPs + * involved, there usually aren't that many connections + * anyway). The consistency means that servers see the same + * client coming from the same IP (some Internet Banking sites + * like this), even across reboots. */ + minip = ntohl(range->min_ip); + maxip = ntohl(range->max_ip); + j = jhash_2words((__force u32)tuple->src.u3.ip, + (__force u32)tuple->dst.u3.ip, 0); + *var_ipp = htonl(minip + j % (maxip - minip + 1)); +} + +/* Manipulate the tuple into the range given. For NF_IP_POST_ROUTING, + * we change the source to map into the range. For NF_IP_PRE_ROUTING + * and NF_IP_LOCAL_OUT, we change the destination to map into the + * range. It might not be possible to get a unique tuple, but we try. + * At worst (or if we race), we will end up with a final duplicate in + * __ip_conntrack_confirm and drop the packet. */ +static void +get_unique_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig_tuple, + const struct nf_nat_range *range, + struct nf_conn *ct, + enum nf_nat_manip_type maniptype) +{ + struct nf_nat_protocol *proto; + + /* 1) If this srcip/proto/src-proto-part is currently mapped, + and that same mapping gives a unique tuple within the given + range, use that. + + This is only required for source (ie. NAT/masq) mappings. + So far, we don't do local source mappings, so multiple + manips not an issue. */ + if (maniptype == IP_NAT_MANIP_SRC) { + if (find_appropriate_src(orig_tuple, tuple, range)) { + DEBUGP("get_unique_tuple: Found current src map\n"); + if (!nf_nat_used_tuple(tuple, ct)) + return; + } + } + + /* 2) Select the least-used IP/proto combination in the given + range. */ + *tuple = *orig_tuple; + find_best_ips_proto(tuple, range, ct, maniptype); + + /* 3) The per-protocol part of the manip is made to map into + the range to make a unique tuple. */ + + proto = nf_nat_proto_find_get(orig_tuple->dst.protonum); + + /* Only bother mapping if it's not already in range and unique */ + if ((!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED) || + proto->in_range(tuple, maniptype, &range->min, &range->max)) && + !nf_nat_used_tuple(tuple, ct)) { + nf_nat_proto_put(proto); + return; + } + + /* Last change: get protocol to try to obtain unique tuple. */ + proto->unique_tuple(tuple, range, maniptype, ct); + + nf_nat_proto_put(proto); +} + +unsigned int +nf_nat_setup_info(struct nf_conn *ct, + const struct nf_nat_range *range, + unsigned int hooknum) +{ + struct nf_conntrack_tuple curr_tuple, new_tuple; + struct nf_conn_nat *nat = nfct_nat(ct); + struct nf_nat_info *info = &nat->info; + int have_to_hash = !(ct->status & IPS_NAT_DONE_MASK); + enum nf_nat_manip_type maniptype = HOOK2MANIP(hooknum); + + NF_CT_ASSERT(hooknum == NF_IP_PRE_ROUTING || + hooknum == NF_IP_POST_ROUTING || + hooknum == NF_IP_LOCAL_IN || + hooknum == NF_IP_LOCAL_OUT); + BUG_ON(nf_nat_initialized(ct, maniptype)); + + /* What we've got will look like inverse of reply. Normally + this is what is in the conntrack, except for prior + manipulations (future optimization: if num_manips == 0, + orig_tp = + conntrack->tuplehash[IP_CT_DIR_ORIGINAL].tuple) */ + nf_ct_invert_tuplepr(&curr_tuple, + &ct->tuplehash[IP_CT_DIR_REPLY].tuple); + + get_unique_tuple(&new_tuple, &curr_tuple, range, ct, maniptype); + + if (!nf_ct_tuple_equal(&new_tuple, &curr_tuple)) { + struct nf_conntrack_tuple reply; + + /* Alter conntrack table so will recognize replies. */ + nf_ct_invert_tuplepr(&reply, &new_tuple); + nf_conntrack_alter_reply(ct, &reply); + + /* Non-atomic: we own this at the moment. */ + if (maniptype == IP_NAT_MANIP_SRC) + ct->status |= IPS_SRC_NAT; + else + ct->status |= IPS_DST_NAT; + } + + /* Place in source hash if this is the first time. */ + if (have_to_hash) { + unsigned int srchash; + + srchash = hash_by_src(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + write_lock_bh(&nf_nat_lock); + list_add(&info->bysource, &bysource[srchash]); + write_unlock_bh(&nf_nat_lock); + } + + /* It's done. */ + if (maniptype == IP_NAT_MANIP_DST) + set_bit(IPS_DST_NAT_DONE_BIT, &ct->status); + else + set_bit(IPS_SRC_NAT_DONE_BIT, &ct->status); + + return NF_ACCEPT; +} +EXPORT_SYMBOL(nf_nat_setup_info); + +/* Returns true if succeeded. */ +static int +manip_pkt(u_int16_t proto, + struct sk_buff **pskb, + unsigned int iphdroff, + const struct nf_conntrack_tuple *target, + enum nf_nat_manip_type maniptype) +{ + struct iphdr *iph; + struct nf_nat_protocol *p; + + if (!skb_make_writable(pskb, iphdroff + sizeof(*iph))) + return 0; + + iph = (void *)(*pskb)->data + iphdroff; + + /* Manipulate protcol part. */ + p = nf_nat_proto_find_get(proto); + if (!p->manip_pkt(pskb, iphdroff, target, maniptype)) { + nf_nat_proto_put(p); + return 0; + } + nf_nat_proto_put(p); + + iph = (void *)(*pskb)->data + iphdroff; + + if (maniptype == IP_NAT_MANIP_SRC) { + nf_csum_replace4(&iph->check, iph->saddr, target->src.u3.ip); + iph->saddr = target->src.u3.ip; + } else { + nf_csum_replace4(&iph->check, iph->daddr, target->dst.u3.ip); + iph->daddr = target->dst.u3.ip; + } + return 1; +} + +/* Do packet manipulations according to nf_nat_setup_info. */ +unsigned int nf_nat_packet(struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int hooknum, + struct sk_buff **pskb) +{ + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + unsigned long statusbit; + enum nf_nat_manip_type mtype = HOOK2MANIP(hooknum); + + if (mtype == IP_NAT_MANIP_SRC) + statusbit = IPS_SRC_NAT; + else + statusbit = IPS_DST_NAT; + + /* Invert if this is reply dir. */ + if (dir == IP_CT_DIR_REPLY) + statusbit ^= IPS_NAT_MASK; + + /* Non-atomic: these bits don't change. */ + if (ct->status & statusbit) { + struct nf_conntrack_tuple target; + + /* We are aiming to look like inverse of other direction. */ + nf_ct_invert_tuplepr(&target, &ct->tuplehash[!dir].tuple); + + if (!manip_pkt(target.dst.protonum, pskb, 0, &target, mtype)) + return NF_DROP; + } + return NF_ACCEPT; +} +EXPORT_SYMBOL_GPL(nf_nat_packet); + +/* Dir is direction ICMP is coming from (opposite to packet it contains) */ +int nf_nat_icmp_reply_translation(struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int hooknum, + struct sk_buff **pskb) +{ + struct { + struct icmphdr icmp; + struct iphdr ip; + } *inside; + struct nf_conntrack_tuple inner, target; + int hdrlen = (*pskb)->nh.iph->ihl * 4; + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + unsigned long statusbit; + enum nf_nat_manip_type manip = HOOK2MANIP(hooknum); + + if (!skb_make_writable(pskb, hdrlen + sizeof(*inside))) + return 0; + + inside = (void *)(*pskb)->data + (*pskb)->nh.iph->ihl*4; + + /* We're actually going to mangle it beyond trivial checksum + adjustment, so make sure the current checksum is correct. */ + if (nf_ip_checksum(*pskb, hooknum, hdrlen, 0)) + return 0; + + /* Must be RELATED */ + NF_CT_ASSERT((*pskb)->nfctinfo == IP_CT_RELATED || + (*pskb)->nfctinfo == IP_CT_RELATED+IP_CT_IS_REPLY); + + /* Redirects on non-null nats must be dropped, else they'll + start talking to each other without our translation, and be + confused... --RR */ + if (inside->icmp.type == ICMP_REDIRECT) { + /* If NAT isn't finished, assume it and drop. */ + if ((ct->status & IPS_NAT_DONE_MASK) != IPS_NAT_DONE_MASK) + return 0; + + if (ct->status & IPS_NAT_MASK) + return 0; + } + + DEBUGP("icmp_reply_translation: translating error %p manp %u dir %s\n", + *pskb, manip, dir == IP_CT_DIR_ORIGINAL ? "ORIG" : "REPLY"); + + if (!nf_ct_get_tuple(*pskb, + (*pskb)->nh.iph->ihl*4 + sizeof(struct icmphdr), + (*pskb)->nh.iph->ihl*4 + + sizeof(struct icmphdr) + inside->ip.ihl*4, + (u_int16_t)AF_INET, + inside->ip.protocol, + &inner, + l3proto, + __nf_ct_l4proto_find((u_int16_t)PF_INET, + inside->ip.protocol))) + return 0; + + /* Change inner back to look like incoming packet. We do the + opposite manip on this hook to normal, because it might not + pass all hooks (locally-generated ICMP). Consider incoming + packet: PREROUTING (DST manip), routing produces ICMP, goes + through POSTROUTING (which must correct the DST manip). */ + if (!manip_pkt(inside->ip.protocol, pskb, + (*pskb)->nh.iph->ihl*4 + sizeof(inside->icmp), + &ct->tuplehash[!dir].tuple, + !manip)) + return 0; + + if ((*pskb)->ip_summed != CHECKSUM_PARTIAL) { + /* Reloading "inside" here since manip_pkt inner. */ + inside = (void *)(*pskb)->data + (*pskb)->nh.iph->ihl*4; + inside->icmp.checksum = 0; + inside->icmp.checksum = + csum_fold(skb_checksum(*pskb, hdrlen, + (*pskb)->len - hdrlen, 0)); + } + + /* Change outer to look the reply to an incoming packet + * (proto 0 means don't invert per-proto part). */ + if (manip == IP_NAT_MANIP_SRC) + statusbit = IPS_SRC_NAT; + else + statusbit = IPS_DST_NAT; + + /* Invert if this is reply dir. */ + if (dir == IP_CT_DIR_REPLY) + statusbit ^= IPS_NAT_MASK; + + if (ct->status & statusbit) { + nf_ct_invert_tuplepr(&target, &ct->tuplehash[!dir].tuple); + if (!manip_pkt(0, pskb, 0, &target, manip)) + return 0; + } + + return 1; +} +EXPORT_SYMBOL_GPL(nf_nat_icmp_reply_translation); + +/* Protocol registration. */ +int nf_nat_protocol_register(struct nf_nat_protocol *proto) +{ + int ret = 0; + + write_lock_bh(&nf_nat_lock); + if (nf_nat_protos[proto->protonum] != &nf_nat_unknown_protocol) { + ret = -EBUSY; + goto out; + } + nf_nat_protos[proto->protonum] = proto; + out: + write_unlock_bh(&nf_nat_lock); + return ret; +} +EXPORT_SYMBOL(nf_nat_protocol_register); + +/* Noone stores the protocol anywhere; simply delete it. */ +void nf_nat_protocol_unregister(struct nf_nat_protocol *proto) +{ + write_lock_bh(&nf_nat_lock); + nf_nat_protos[proto->protonum] = &nf_nat_unknown_protocol; + write_unlock_bh(&nf_nat_lock); + + /* Someone could be still looking at the proto in a bh. */ + synchronize_net(); +} +EXPORT_SYMBOL(nf_nat_protocol_unregister); + +#if defined(CONFIG_IP_NF_CONNTRACK_NETLINK) || \ + defined(CONFIG_IP_NF_CONNTRACK_NETLINK_MODULE) +int +nf_nat_port_range_to_nfattr(struct sk_buff *skb, + const struct nf_nat_range *range) +{ + NFA_PUT(skb, CTA_PROTONAT_PORT_MIN, sizeof(__be16), + &range->min.tcp.port); + NFA_PUT(skb, CTA_PROTONAT_PORT_MAX, sizeof(__be16), + &range->max.tcp.port); + + return 0; + +nfattr_failure: + return -1; +} +EXPORT_SYMBOL_GPL(nf_nat_port_nfattr_to_range); + +int +nf_nat_port_nfattr_to_range(struct nfattr *tb[], struct nf_nat_range *range) +{ + int ret = 0; + + /* we have to return whether we actually parsed something or not */ + + if (tb[CTA_PROTONAT_PORT_MIN-1]) { + ret = 1; + range->min.tcp.port = + *(__be16 *)NFA_DATA(tb[CTA_PROTONAT_PORT_MIN-1]); + } + + if (!tb[CTA_PROTONAT_PORT_MAX-1]) { + if (ret) + range->max.tcp.port = range->min.tcp.port; + } else { + ret = 1; + range->max.tcp.port = + *(__be16 *)NFA_DATA(tb[CTA_PROTONAT_PORT_MAX-1]); + } + + return ret; +} +EXPORT_SYMBOL_GPL(nf_nat_port_range_to_nfattr); +#endif + +static int __init nf_nat_init(void) +{ + size_t i; + + /* Leave them the same for the moment. */ + nf_nat_htable_size = nf_conntrack_htable_size; + + /* One vmalloc for both hash tables */ + bysource = vmalloc(sizeof(struct list_head) * nf_nat_htable_size); + if (!bysource) + return -ENOMEM; + + /* Sew in builtin protocols. */ + write_lock_bh(&nf_nat_lock); + for (i = 0; i < MAX_IP_NAT_PROTO; i++) + nf_nat_protos[i] = &nf_nat_unknown_protocol; + nf_nat_protos[IPPROTO_TCP] = &nf_nat_protocol_tcp; + nf_nat_protos[IPPROTO_UDP] = &nf_nat_protocol_udp; + nf_nat_protos[IPPROTO_ICMP] = &nf_nat_protocol_icmp; + write_unlock_bh(&nf_nat_lock); + + for (i = 0; i < nf_nat_htable_size; i++) { + INIT_LIST_HEAD(&bysource[i]); + } + + /* FIXME: Man, this is a hack. */ + NF_CT_ASSERT(nf_conntrack_destroyed == NULL); + nf_conntrack_destroyed = &nf_nat_cleanup_conntrack; + + /* Initialize fake conntrack so that NAT will skip it */ + nf_conntrack_untracked.status |= IPS_NAT_DONE_MASK; + + l3proto = nf_ct_l3proto_find_get((u_int16_t)AF_INET); + return 0; +} + +/* Clear NAT section of all conntracks, in case we're loaded again. */ +static int clean_nat(struct nf_conn *i, void *data) +{ + struct nf_conn_nat *nat = nfct_nat(i); + + if (!nat) + return 0; + memset(nat, 0, sizeof(nat)); + i->status &= ~(IPS_NAT_MASK | IPS_NAT_DONE_MASK | IPS_SEQ_ADJUST); + return 0; +} + +static void __exit nf_nat_cleanup(void) +{ + nf_ct_iterate_cleanup(&clean_nat, NULL); + nf_conntrack_destroyed = NULL; + vfree(bysource); + nf_ct_l3proto_put(l3proto); +} + +MODULE_LICENSE("GPL"); + +module_init(nf_nat_init); +module_exit(nf_nat_cleanup); diff --git a/net/ipv4/netfilter/nf_nat_helper.c b/net/ipv4/netfilter/nf_nat_helper.c new file mode 100644 index 00000000000..98fbfc84d18 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_helper.c @@ -0,0 +1,433 @@ +/* ip_nat_helper.c - generic support functions for NAT helpers + * + * (C) 2000-2002 Harald Welte + * (C) 2003-2006 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#define DUMP_OFFSET(x) printk("offset_before=%d, offset_after=%d, correction_pos=%u\n", x->offset_before, x->offset_after, x->correction_pos); +#else +#define DEBUGP(format, args...) +#define DUMP_OFFSET(x) +#endif + +static DEFINE_SPINLOCK(nf_nat_seqofs_lock); + +/* Setup TCP sequence correction given this change at this sequence */ +static inline void +adjust_tcp_sequence(u32 seq, + int sizediff, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo) +{ + int dir; + struct nf_nat_seq *this_way, *other_way; + struct nf_conn_nat *nat = nfct_nat(ct); + + DEBUGP("nf_nat_resize_packet: old_size = %u, new_size = %u\n", + (*skb)->len, new_size); + + dir = CTINFO2DIR(ctinfo); + + this_way = &nat->info.seq[dir]; + other_way = &nat->info.seq[!dir]; + + DEBUGP("nf_nat_resize_packet: Seq_offset before: "); + DUMP_OFFSET(this_way); + + spin_lock_bh(&nf_nat_seqofs_lock); + + /* SYN adjust. If it's uninitialized, or this is after last + * correction, record it: we don't handle more than one + * adjustment in the window, but do deal with common case of a + * retransmit */ + if (this_way->offset_before == this_way->offset_after || + before(this_way->correction_pos, seq)) { + this_way->correction_pos = seq; + this_way->offset_before = this_way->offset_after; + this_way->offset_after += sizediff; + } + spin_unlock_bh(&nf_nat_seqofs_lock); + + DEBUGP("nf_nat_resize_packet: Seq_offset after: "); + DUMP_OFFSET(this_way); +} + +/* Frobs data inside this packet, which is linear. */ +static void mangle_contents(struct sk_buff *skb, + unsigned int dataoff, + unsigned int match_offset, + unsigned int match_len, + const char *rep_buffer, + unsigned int rep_len) +{ + unsigned char *data; + + BUG_ON(skb_is_nonlinear(skb)); + data = (unsigned char *)skb->nh.iph + dataoff; + + /* move post-replacement */ + memmove(data + match_offset + rep_len, + data + match_offset + match_len, + skb->tail - (data + match_offset + match_len)); + + /* insert data from buffer */ + memcpy(data + match_offset, rep_buffer, rep_len); + + /* update skb info */ + if (rep_len > match_len) { + DEBUGP("nf_nat_mangle_packet: Extending packet by " + "%u from %u bytes\n", rep_len - match_len, + skb->len); + skb_put(skb, rep_len - match_len); + } else { + DEBUGP("nf_nat_mangle_packet: Shrinking packet from " + "%u from %u bytes\n", match_len - rep_len, + skb->len); + __skb_trim(skb, skb->len + rep_len - match_len); + } + + /* fix IP hdr checksum information */ + skb->nh.iph->tot_len = htons(skb->len); + ip_send_check(skb->nh.iph); +} + +/* Unusual, but possible case. */ +static int enlarge_skb(struct sk_buff **pskb, unsigned int extra) +{ + struct sk_buff *nskb; + + if ((*pskb)->len + extra > 65535) + return 0; + + nskb = skb_copy_expand(*pskb, skb_headroom(*pskb), extra, GFP_ATOMIC); + if (!nskb) + return 0; + + /* Transfer socket to new skb. */ + if ((*pskb)->sk) + skb_set_owner_w(nskb, (*pskb)->sk); + kfree_skb(*pskb); + *pskb = nskb; + return 1; +} + +/* Generic function for mangling variable-length address changes inside + * NATed TCP connections (like the PORT XXX,XXX,XXX,XXX,XXX,XXX + * command in FTP). + * + * Takes care about all the nasty sequence number changes, checksumming, + * skb enlargement, ... + * + * */ +int +nf_nat_mangle_tcp_packet(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int match_offset, + unsigned int match_len, + const char *rep_buffer, + unsigned int rep_len) +{ + struct iphdr *iph; + struct tcphdr *tcph; + int oldlen, datalen; + + if (!skb_make_writable(pskb, (*pskb)->len)) + return 0; + + if (rep_len > match_len && + rep_len - match_len > skb_tailroom(*pskb) && + !enlarge_skb(pskb, rep_len - match_len)) + return 0; + + SKB_LINEAR_ASSERT(*pskb); + + iph = (*pskb)->nh.iph; + tcph = (void *)iph + iph->ihl*4; + + oldlen = (*pskb)->len - iph->ihl*4; + mangle_contents(*pskb, iph->ihl*4 + tcph->doff*4, + match_offset, match_len, rep_buffer, rep_len); + + datalen = (*pskb)->len - iph->ihl*4; + if ((*pskb)->ip_summed != CHECKSUM_PARTIAL) { + tcph->check = 0; + tcph->check = tcp_v4_check(tcph, datalen, + iph->saddr, iph->daddr, + csum_partial((char *)tcph, + datalen, 0)); + } else + nf_proto_csum_replace2(&tcph->check, *pskb, + htons(oldlen), htons(datalen), 1); + + if (rep_len != match_len) { + set_bit(IPS_SEQ_ADJUST_BIT, &ct->status); + adjust_tcp_sequence(ntohl(tcph->seq), + (int)rep_len - (int)match_len, + ct, ctinfo); + /* Tell TCP window tracking about seq change */ + nf_conntrack_tcp_update(*pskb, (*pskb)->nh.iph->ihl*4, + ct, CTINFO2DIR(ctinfo)); + } + return 1; +} +EXPORT_SYMBOL(nf_nat_mangle_tcp_packet); + +/* Generic function for mangling variable-length address changes inside + * NATed UDP connections (like the CONNECT DATA XXXXX MESG XXXXX INDEX XXXXX + * command in the Amanda protocol) + * + * Takes care about all the nasty sequence number changes, checksumming, + * skb enlargement, ... + * + * XXX - This function could be merged with nf_nat_mangle_tcp_packet which + * should be fairly easy to do. + */ +int +nf_nat_mangle_udp_packet(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int match_offset, + unsigned int match_len, + const char *rep_buffer, + unsigned int rep_len) +{ + struct iphdr *iph; + struct udphdr *udph; + int datalen, oldlen; + + /* UDP helpers might accidentally mangle the wrong packet */ + iph = (*pskb)->nh.iph; + if ((*pskb)->len < iph->ihl*4 + sizeof(*udph) + + match_offset + match_len) + return 0; + + if (!skb_make_writable(pskb, (*pskb)->len)) + return 0; + + if (rep_len > match_len && + rep_len - match_len > skb_tailroom(*pskb) && + !enlarge_skb(pskb, rep_len - match_len)) + return 0; + + iph = (*pskb)->nh.iph; + udph = (void *)iph + iph->ihl*4; + + oldlen = (*pskb)->len - iph->ihl*4; + mangle_contents(*pskb, iph->ihl*4 + sizeof(*udph), + match_offset, match_len, rep_buffer, rep_len); + + /* update the length of the UDP packet */ + datalen = (*pskb)->len - iph->ihl*4; + udph->len = htons(datalen); + + /* fix udp checksum if udp checksum was previously calculated */ + if (!udph->check && (*pskb)->ip_summed != CHECKSUM_PARTIAL) + return 1; + + if ((*pskb)->ip_summed != CHECKSUM_PARTIAL) { + udph->check = 0; + udph->check = csum_tcpudp_magic(iph->saddr, iph->daddr, + datalen, IPPROTO_UDP, + csum_partial((char *)udph, + datalen, 0)); + if (!udph->check) + udph->check = CSUM_MANGLED_0; + } else + nf_proto_csum_replace2(&udph->check, *pskb, + htons(oldlen), htons(datalen), 1); + + return 1; +} +EXPORT_SYMBOL(nf_nat_mangle_udp_packet); + +/* Adjust one found SACK option including checksum correction */ +static void +sack_adjust(struct sk_buff *skb, + struct tcphdr *tcph, + unsigned int sackoff, + unsigned int sackend, + struct nf_nat_seq *natseq) +{ + while (sackoff < sackend) { + struct tcp_sack_block_wire *sack; + __be32 new_start_seq, new_end_seq; + + sack = (void *)skb->data + sackoff; + if (after(ntohl(sack->start_seq) - natseq->offset_before, + natseq->correction_pos)) + new_start_seq = htonl(ntohl(sack->start_seq) + - natseq->offset_after); + else + new_start_seq = htonl(ntohl(sack->start_seq) + - natseq->offset_before); + + if (after(ntohl(sack->end_seq) - natseq->offset_before, + natseq->correction_pos)) + new_end_seq = htonl(ntohl(sack->end_seq) + - natseq->offset_after); + else + new_end_seq = htonl(ntohl(sack->end_seq) + - natseq->offset_before); + + DEBUGP("sack_adjust: start_seq: %d->%d, end_seq: %d->%d\n", + ntohl(sack->start_seq), new_start_seq, + ntohl(sack->end_seq), new_end_seq); + + nf_proto_csum_replace4(&tcph->check, skb, + sack->start_seq, new_start_seq, 0); + nf_proto_csum_replace4(&tcph->check, skb, + sack->end_seq, new_end_seq, 0); + sack->start_seq = new_start_seq; + sack->end_seq = new_end_seq; + sackoff += sizeof(*sack); + } +} + +/* TCP SACK sequence number adjustment */ +static inline unsigned int +nf_nat_sack_adjust(struct sk_buff **pskb, + struct tcphdr *tcph, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo) +{ + unsigned int dir, optoff, optend; + struct nf_conn_nat *nat = nfct_nat(ct); + + optoff = (*pskb)->nh.iph->ihl*4 + sizeof(struct tcphdr); + optend = (*pskb)->nh.iph->ihl*4 + tcph->doff*4; + + if (!skb_make_writable(pskb, optend)) + return 0; + + dir = CTINFO2DIR(ctinfo); + + while (optoff < optend) { + /* Usually: option, length. */ + unsigned char *op = (*pskb)->data + optoff; + + switch (op[0]) { + case TCPOPT_EOL: + return 1; + case TCPOPT_NOP: + optoff++; + continue; + default: + /* no partial options */ + if (optoff + 1 == optend || + optoff + op[1] > optend || + op[1] < 2) + return 0; + if (op[0] == TCPOPT_SACK && + op[1] >= 2+TCPOLEN_SACK_PERBLOCK && + ((op[1] - 2) % TCPOLEN_SACK_PERBLOCK) == 0) + sack_adjust(*pskb, tcph, optoff+2, + optoff+op[1], + &nat->info.seq[!dir]); + optoff += op[1]; + } + } + return 1; +} + +/* TCP sequence number adjustment. Returns 1 on success, 0 on failure */ +int +nf_nat_seq_adjust(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo) +{ + struct tcphdr *tcph; + int dir; + __be32 newseq, newack; + struct nf_conn_nat *nat = nfct_nat(ct); + struct nf_nat_seq *this_way, *other_way; + + dir = CTINFO2DIR(ctinfo); + + this_way = &nat->info.seq[dir]; + other_way = &nat->info.seq[!dir]; + + if (!skb_make_writable(pskb, (*pskb)->nh.iph->ihl*4+sizeof(*tcph))) + return 0; + + tcph = (void *)(*pskb)->data + (*pskb)->nh.iph->ihl*4; + if (after(ntohl(tcph->seq), this_way->correction_pos)) + newseq = htonl(ntohl(tcph->seq) + this_way->offset_after); + else + newseq = htonl(ntohl(tcph->seq) + this_way->offset_before); + + if (after(ntohl(tcph->ack_seq) - other_way->offset_before, + other_way->correction_pos)) + newack = htonl(ntohl(tcph->ack_seq) - other_way->offset_after); + else + newack = htonl(ntohl(tcph->ack_seq) - other_way->offset_before); + + nf_proto_csum_replace4(&tcph->check, *pskb, tcph->seq, newseq, 0); + nf_proto_csum_replace4(&tcph->check, *pskb, tcph->ack_seq, newack, 0); + + DEBUGP("Adjusting sequence number from %u->%u, ack from %u->%u\n", + ntohl(tcph->seq), ntohl(newseq), ntohl(tcph->ack_seq), + ntohl(newack)); + + tcph->seq = newseq; + tcph->ack_seq = newack; + + if (!nf_nat_sack_adjust(pskb, tcph, ct, ctinfo)) + return 0; + + nf_conntrack_tcp_update(*pskb, (*pskb)->nh.iph->ihl*4, ct, dir); + + return 1; +} +EXPORT_SYMBOL(nf_nat_seq_adjust); + +/* Setup NAT on this expected conntrack so it follows master. */ +/* If we fail to get a free NAT slot, we'll get dropped on confirm */ +void nf_nat_follow_master(struct nf_conn *ct, + struct nf_conntrack_expect *exp) +{ + struct nf_nat_range range; + + /* This must be a fresh one. */ + BUG_ON(ct->status & IPS_NAT_DONE_MASK); + + /* Change src to where master sends to */ + range.flags = IP_NAT_RANGE_MAP_IPS; + range.min_ip = range.max_ip + = ct->master->tuplehash[!exp->dir].tuple.dst.u3.ip; + /* hook doesn't matter, but it has to do source manip */ + nf_nat_setup_info(ct, &range, NF_IP_POST_ROUTING); + + /* For DST manip, map port here to where it's expected. */ + range.flags = (IP_NAT_RANGE_MAP_IPS | IP_NAT_RANGE_PROTO_SPECIFIED); + range.min = range.max = exp->saved_proto; + range.min_ip = range.max_ip + = ct->master->tuplehash[!exp->dir].tuple.src.u3.ip; + /* hook doesn't matter, but it has to do destination manip */ + nf_nat_setup_info(ct, &range, NF_IP_PRE_ROUTING); +} +EXPORT_SYMBOL(nf_nat_follow_master); diff --git a/net/ipv4/netfilter/nf_nat_proto_icmp.c b/net/ipv4/netfilter/nf_nat_proto_icmp.c new file mode 100644 index 00000000000..dcfd772972d --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_proto_icmp.c @@ -0,0 +1,86 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +static int +icmp_in_range(const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype, + const union nf_conntrack_man_proto *min, + const union nf_conntrack_man_proto *max) +{ + return ntohs(tuple->src.u.icmp.id) >= ntohs(min->icmp.id) && + ntohs(tuple->src.u.icmp.id) <= ntohs(max->icmp.id); +} + +static int +icmp_unique_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + enum nf_nat_manip_type maniptype, + const struct nf_conn *ct) +{ + static u_int16_t id; + unsigned int range_size; + unsigned int i; + + range_size = ntohs(range->max.icmp.id) - ntohs(range->min.icmp.id) + 1; + /* If no range specified... */ + if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED)) + range_size = 0xFFFF; + + for (i = 0; i < range_size; i++, id++) { + tuple->src.u.icmp.id = htons(ntohs(range->min.icmp.id) + + (id % range_size)); + if (!nf_nat_used_tuple(tuple, ct)) + return 1; + } + return 0; +} + +static int +icmp_manip_pkt(struct sk_buff **pskb, + unsigned int iphdroff, + const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype) +{ + struct iphdr *iph = (struct iphdr *)((*pskb)->data + iphdroff); + struct icmphdr *hdr; + unsigned int hdroff = iphdroff + iph->ihl*4; + + if (!skb_make_writable(pskb, hdroff + sizeof(*hdr))) + return 0; + + hdr = (struct icmphdr *)((*pskb)->data + hdroff); + nf_proto_csum_replace2(&hdr->checksum, *pskb, + hdr->un.echo.id, tuple->src.u.icmp.id, 0); + hdr->un.echo.id = tuple->src.u.icmp.id; + return 1; +} + +struct nf_nat_protocol nf_nat_protocol_icmp = { + .name = "ICMP", + .protonum = IPPROTO_ICMP, + .me = THIS_MODULE, + .manip_pkt = icmp_manip_pkt, + .in_range = icmp_in_range, + .unique_tuple = icmp_unique_tuple, +#if defined(CONFIG_IP_NF_CONNTRACK_NETLINK) || \ + defined(CONFIG_IP_NF_CONNTRACK_NETLINK_MODULE) + .range_to_nfattr = nf_nat_port_range_to_nfattr, + .nfattr_to_range = nf_nat_port_nfattr_to_range, +#endif +}; diff --git a/net/ipv4/netfilter/nf_nat_proto_tcp.c b/net/ipv4/netfilter/nf_nat_proto_tcp.c new file mode 100644 index 00000000000..7e26a7e9bee --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_proto_tcp.c @@ -0,0 +1,148 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +static int +tcp_in_range(const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype, + const union nf_conntrack_man_proto *min, + const union nf_conntrack_man_proto *max) +{ + __be16 port; + + if (maniptype == IP_NAT_MANIP_SRC) + port = tuple->src.u.tcp.port; + else + port = tuple->dst.u.tcp.port; + + return ntohs(port) >= ntohs(min->tcp.port) && + ntohs(port) <= ntohs(max->tcp.port); +} + +static int +tcp_unique_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + enum nf_nat_manip_type maniptype, + const struct nf_conn *ct) +{ + static u_int16_t port; + __be16 *portptr; + unsigned int range_size, min, i; + + if (maniptype == IP_NAT_MANIP_SRC) + portptr = &tuple->src.u.tcp.port; + else + portptr = &tuple->dst.u.tcp.port; + + /* If no range specified... */ + if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED)) { + /* If it's dst rewrite, can't change port */ + if (maniptype == IP_NAT_MANIP_DST) + return 0; + + /* Map privileged onto privileged. */ + if (ntohs(*portptr) < 1024) { + /* Loose convention: >> 512 is credential passing */ + if (ntohs(*portptr)<512) { + min = 1; + range_size = 511 - min + 1; + } else { + min = 600; + range_size = 1023 - min + 1; + } + } else { + min = 1024; + range_size = 65535 - 1024 + 1; + } + } else { + min = ntohs(range->min.tcp.port); + range_size = ntohs(range->max.tcp.port) - min + 1; + } + + for (i = 0; i < range_size; i++, port++) { + *portptr = htons(min + port % range_size); + if (!nf_nat_used_tuple(tuple, ct)) + return 1; + } + return 0; +} + +static int +tcp_manip_pkt(struct sk_buff **pskb, + unsigned int iphdroff, + const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype) +{ + struct iphdr *iph = (struct iphdr *)((*pskb)->data + iphdroff); + struct tcphdr *hdr; + unsigned int hdroff = iphdroff + iph->ihl*4; + __be32 oldip, newip; + __be16 *portptr, newport, oldport; + int hdrsize = 8; /* TCP connection tracking guarantees this much */ + + /* this could be a inner header returned in icmp packet; in such + cases we cannot update the checksum field since it is outside of + the 8 bytes of transport layer headers we are guaranteed */ + if ((*pskb)->len >= hdroff + sizeof(struct tcphdr)) + hdrsize = sizeof(struct tcphdr); + + if (!skb_make_writable(pskb, hdroff + hdrsize)) + return 0; + + iph = (struct iphdr *)((*pskb)->data + iphdroff); + hdr = (struct tcphdr *)((*pskb)->data + hdroff); + + if (maniptype == IP_NAT_MANIP_SRC) { + /* Get rid of src ip and src pt */ + oldip = iph->saddr; + newip = tuple->src.u3.ip; + newport = tuple->src.u.tcp.port; + portptr = &hdr->source; + } else { + /* Get rid of dst ip and dst pt */ + oldip = iph->daddr; + newip = tuple->dst.u3.ip; + newport = tuple->dst.u.tcp.port; + portptr = &hdr->dest; + } + + oldport = *portptr; + *portptr = newport; + + if (hdrsize < sizeof(*hdr)) + return 1; + + nf_proto_csum_replace4(&hdr->check, *pskb, oldip, newip, 1); + nf_proto_csum_replace2(&hdr->check, *pskb, oldport, newport, 0); + return 1; +} + +struct nf_nat_protocol nf_nat_protocol_tcp = { + .name = "TCP", + .protonum = IPPROTO_TCP, + .me = THIS_MODULE, + .manip_pkt = tcp_manip_pkt, + .in_range = tcp_in_range, + .unique_tuple = tcp_unique_tuple, +#if defined(CONFIG_IP_NF_CONNTRACK_NETLINK) || \ + defined(CONFIG_IP_NF_CONNTRACK_NETLINK_MODULE) + .range_to_nfattr = nf_nat_port_range_to_nfattr, + .nfattr_to_range = nf_nat_port_nfattr_to_range, +#endif +}; diff --git a/net/ipv4/netfilter/nf_nat_proto_udp.c b/net/ipv4/netfilter/nf_nat_proto_udp.c new file mode 100644 index 00000000000..ab0ce4c8699 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_proto_udp.c @@ -0,0 +1,138 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +static int +udp_in_range(const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype, + const union nf_conntrack_man_proto *min, + const union nf_conntrack_man_proto *max) +{ + __be16 port; + + if (maniptype == IP_NAT_MANIP_SRC) + port = tuple->src.u.udp.port; + else + port = tuple->dst.u.udp.port; + + return ntohs(port) >= ntohs(min->udp.port) && + ntohs(port) <= ntohs(max->udp.port); +} + +static int +udp_unique_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + enum nf_nat_manip_type maniptype, + const struct nf_conn *ct) +{ + static u_int16_t port; + __be16 *portptr; + unsigned int range_size, min, i; + + if (maniptype == IP_NAT_MANIP_SRC) + portptr = &tuple->src.u.udp.port; + else + portptr = &tuple->dst.u.udp.port; + + /* If no range specified... */ + if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED)) { + /* If it's dst rewrite, can't change port */ + if (maniptype == IP_NAT_MANIP_DST) + return 0; + + if (ntohs(*portptr) < 1024) { + /* Loose convention: >> 512 is credential passing */ + if (ntohs(*portptr)<512) { + min = 1; + range_size = 511 - min + 1; + } else { + min = 600; + range_size = 1023 - min + 1; + } + } else { + min = 1024; + range_size = 65535 - 1024 + 1; + } + } else { + min = ntohs(range->min.udp.port); + range_size = ntohs(range->max.udp.port) - min + 1; + } + + for (i = 0; i < range_size; i++, port++) { + *portptr = htons(min + port % range_size); + if (!nf_nat_used_tuple(tuple, ct)) + return 1; + } + return 0; +} + +static int +udp_manip_pkt(struct sk_buff **pskb, + unsigned int iphdroff, + const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype) +{ + struct iphdr *iph = (struct iphdr *)((*pskb)->data + iphdroff); + struct udphdr *hdr; + unsigned int hdroff = iphdroff + iph->ihl*4; + __be32 oldip, newip; + __be16 *portptr, newport; + + if (!skb_make_writable(pskb, hdroff + sizeof(*hdr))) + return 0; + + iph = (struct iphdr *)((*pskb)->data + iphdroff); + hdr = (struct udphdr *)((*pskb)->data + hdroff); + + if (maniptype == IP_NAT_MANIP_SRC) { + /* Get rid of src ip and src pt */ + oldip = iph->saddr; + newip = tuple->src.u3.ip; + newport = tuple->src.u.udp.port; + portptr = &hdr->source; + } else { + /* Get rid of dst ip and dst pt */ + oldip = iph->daddr; + newip = tuple->dst.u3.ip; + newport = tuple->dst.u.udp.port; + portptr = &hdr->dest; + } + if (hdr->check || (*pskb)->ip_summed == CHECKSUM_PARTIAL) { + nf_proto_csum_replace4(&hdr->check, *pskb, oldip, newip, 1); + nf_proto_csum_replace2(&hdr->check, *pskb, *portptr, newport, + 0); + if (!hdr->check) + hdr->check = CSUM_MANGLED_0; + } + *portptr = newport; + return 1; +} + +struct nf_nat_protocol nf_nat_protocol_udp = { + .name = "UDP", + .protonum = IPPROTO_UDP, + .me = THIS_MODULE, + .manip_pkt = udp_manip_pkt, + .in_range = udp_in_range, + .unique_tuple = udp_unique_tuple, +#if defined(CONFIG_IP_NF_CONNTRACK_NETLINK) || \ + defined(CONFIG_IP_NF_CONNTRACK_NETLINK_MODULE) + .range_to_nfattr = nf_nat_port_range_to_nfattr, + .nfattr_to_range = nf_nat_port_nfattr_to_range, +#endif +}; diff --git a/net/ipv4/netfilter/nf_nat_proto_unknown.c b/net/ipv4/netfilter/nf_nat_proto_unknown.c new file mode 100644 index 00000000000..f50d0203f9c --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_proto_unknown.c @@ -0,0 +1,54 @@ +/* The "unknown" protocol. This is what is used for protocols we + * don't understand. It's returned by ip_ct_find_proto(). + */ + +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include + +#include +#include +#include +#include + +static int unknown_in_range(const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type manip_type, + const union nf_conntrack_man_proto *min, + const union nf_conntrack_man_proto *max) +{ + return 1; +} + +static int unknown_unique_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + enum nf_nat_manip_type maniptype, + const struct nf_conn *ct) +{ + /* Sorry: we can't help you; if it's not unique, we can't frob + anything. */ + return 0; +} + +static int +unknown_manip_pkt(struct sk_buff **pskb, + unsigned int iphdroff, + const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype) +{ + return 1; +} + +struct nf_nat_protocol nf_nat_unknown_protocol = { + .name = "unknown", + /* .me isn't set: getting a ref to this cannot fail. */ + .manip_pkt = unknown_manip_pkt, + .in_range = unknown_in_range, + .unique_tuple = unknown_unique_tuple, +}; diff --git a/net/ipv4/netfilter/nf_nat_rule.c b/net/ipv4/netfilter/nf_nat_rule.c new file mode 100644 index 00000000000..b868ee0195d --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_rule.c @@ -0,0 +1,343 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +/* Everything about the rules for NAT. */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +#define NAT_VALID_HOOKS ((1<range[0], hooknum); +} + +/* Before 2.6.11 we did implicit source NAT if required. Warn about change. */ +static void warn_if_extra_mangle(__be32 dstip, __be32 srcip) +{ + static int warned = 0; + struct flowi fl = { .nl_u = { .ip4_u = { .daddr = dstip } } }; + struct rtable *rt; + + if (ip_route_output_key(&rt, &fl) != 0) + return; + + if (rt->rt_src != srcip && !warned) { + printk("NAT: no longer support implicit source local NAT\n"); + printk("NAT: packet src %u.%u.%u.%u -> dst %u.%u.%u.%u\n", + NIPQUAD(srcip), NIPQUAD(dstip)); + warned = 1; + } + ip_rt_put(rt); +} + +static unsigned int ipt_dnat_target(struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + unsigned int hooknum, + const struct xt_target *target, + const void *targinfo) +{ + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + const struct nf_nat_multi_range_compat *mr = targinfo; + + NF_CT_ASSERT(hooknum == NF_IP_PRE_ROUTING || + hooknum == NF_IP_LOCAL_OUT); + + ct = nf_ct_get(*pskb, &ctinfo); + + /* Connection must be valid and new. */ + NF_CT_ASSERT(ct && (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED)); + + if (hooknum == NF_IP_LOCAL_OUT && + mr->range[0].flags & IP_NAT_RANGE_MAP_IPS) + warn_if_extra_mangle((*pskb)->nh.iph->daddr, + mr->range[0].min_ip); + + return nf_nat_setup_info(ct, &mr->range[0], hooknum); +} + +static int ipt_snat_checkentry(const char *tablename, + const void *entry, + const struct xt_target *target, + void *targinfo, + unsigned int hook_mask) +{ + struct nf_nat_multi_range_compat *mr = targinfo; + + /* Must be a valid range */ + if (mr->rangesize != 1) { + printk("SNAT: multiple ranges no longer supported\n"); + return 0; + } + return 1; +} + +static int ipt_dnat_checkentry(const char *tablename, + const void *entry, + const struct xt_target *target, + void *targinfo, + unsigned int hook_mask) +{ + struct nf_nat_multi_range_compat *mr = targinfo; + + /* Must be a valid range */ + if (mr->rangesize != 1) { + printk("DNAT: multiple ranges no longer supported\n"); + return 0; + } + return 1; +} + +inline unsigned int +alloc_null_binding(struct nf_conn *ct, + struct nf_nat_info *info, + unsigned int hooknum) +{ + /* Force range to this IP; let proto decide mapping for + per-proto parts (hence not IP_NAT_RANGE_PROTO_SPECIFIED). + Use reply in case it's already been mangled (eg local packet). + */ + __be32 ip + = (HOOK2MANIP(hooknum) == IP_NAT_MANIP_SRC + ? ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip + : ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3.ip); + struct nf_nat_range range + = { IP_NAT_RANGE_MAP_IPS, ip, ip, { 0 }, { 0 } }; + + DEBUGP("Allocating NULL binding for %p (%u.%u.%u.%u)\n", + ct, NIPQUAD(ip)); + return nf_nat_setup_info(ct, &range, hooknum); +} + +unsigned int +alloc_null_binding_confirmed(struct nf_conn *ct, + struct nf_nat_info *info, + unsigned int hooknum) +{ + __be32 ip + = (HOOK2MANIP(hooknum) == IP_NAT_MANIP_SRC + ? ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip + : ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3.ip); + u_int16_t all + = (HOOK2MANIP(hooknum) == IP_NAT_MANIP_SRC + ? ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u.all + : ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u.all); + struct nf_nat_range range + = { IP_NAT_RANGE_MAP_IPS, ip, ip, { all }, { all } }; + + DEBUGP("Allocating NULL binding for confirmed %p (%u.%u.%u.%u)\n", + ct, NIPQUAD(ip)); + return nf_nat_setup_info(ct, &range, hooknum); +} + +int nf_nat_rule_find(struct sk_buff **pskb, + unsigned int hooknum, + const struct net_device *in, + const struct net_device *out, + struct nf_conn *ct, + struct nf_nat_info *info) +{ + int ret; + + ret = ipt_do_table(pskb, hooknum, in, out, &nat_table); + + if (ret == NF_ACCEPT) { + if (!nf_nat_initialized(ct, HOOK2MANIP(hooknum))) + /* NUL mapping */ + ret = alloc_null_binding(ct, info, hooknum); + } + return ret; +} + +static struct ipt_target ipt_snat_reg = { + .name = "SNAT", + .target = ipt_snat_target, + .targetsize = sizeof(struct nf_nat_multi_range_compat), + .table = "nat", + .hooks = 1 << NF_IP_POST_ROUTING, + .checkentry = ipt_snat_checkentry, + .family = AF_INET, +}; + +static struct xt_target ipt_dnat_reg = { + .name = "DNAT", + .target = ipt_dnat_target, + .targetsize = sizeof(struct nf_nat_multi_range_compat), + .table = "nat", + .hooks = (1 << NF_IP_PRE_ROUTING) | (1 << NF_IP_LOCAL_OUT), + .checkentry = ipt_dnat_checkentry, + .family = AF_INET, +}; + +int __init nf_nat_rule_init(void) +{ + int ret; + + ret = ipt_register_table(&nat_table, &nat_initial_table.repl); + if (ret != 0) + return ret; + ret = xt_register_target(&ipt_snat_reg); + if (ret != 0) + goto unregister_table; + + ret = xt_register_target(&ipt_dnat_reg); + if (ret != 0) + goto unregister_snat; + + return ret; + + unregister_snat: + xt_unregister_target(&ipt_snat_reg); + unregister_table: + ipt_unregister_table(&nat_table); + + return ret; +} + +void nf_nat_rule_cleanup(void) +{ + xt_unregister_target(&ipt_dnat_reg); + xt_unregister_target(&ipt_snat_reg); + ipt_unregister_table(&nat_table); +} diff --git a/net/ipv4/netfilter/nf_nat_standalone.c b/net/ipv4/netfilter/nf_nat_standalone.c new file mode 100644 index 00000000000..730a7a44c88 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_standalone.c @@ -0,0 +1,406 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +#define HOOKNAME(hooknum) ((hooknum) == NF_IP_POST_ROUTING ? "POST_ROUTING" \ + : ((hooknum) == NF_IP_PRE_ROUTING ? "PRE_ROUTING" \ + : ((hooknum) == NF_IP_LOCAL_OUT ? "LOCAL_OUT" \ + : ((hooknum) == NF_IP_LOCAL_IN ? "LOCAL_IN" \ + : "*ERROR*"))) + +#ifdef CONFIG_XFRM +static void nat_decode_session(struct sk_buff *skb, struct flowi *fl) +{ + struct nf_conn *ct; + struct nf_conntrack_tuple *t; + enum ip_conntrack_info ctinfo; + enum ip_conntrack_dir dir; + unsigned long statusbit; + + ct = nf_ct_get(skb, &ctinfo); + if (ct == NULL) + return; + dir = CTINFO2DIR(ctinfo); + t = &ct->tuplehash[dir].tuple; + + if (dir == IP_CT_DIR_ORIGINAL) + statusbit = IPS_DST_NAT; + else + statusbit = IPS_SRC_NAT; + + if (ct->status & statusbit) { + fl->fl4_dst = t->dst.u3.ip; + if (t->dst.protonum == IPPROTO_TCP || + t->dst.protonum == IPPROTO_UDP) + fl->fl_ip_dport = t->dst.u.tcp.port; + } + + statusbit ^= IPS_NAT_MASK; + + if (ct->status & statusbit) { + fl->fl4_src = t->src.u3.ip; + if (t->dst.protonum == IPPROTO_TCP || + t->dst.protonum == IPPROTO_UDP) + fl->fl_ip_sport = t->src.u.tcp.port; + } +} +#endif + +static unsigned int +nf_nat_fn(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + struct nf_conn_nat *nat; + struct nf_nat_info *info; + /* maniptype == SRC for postrouting. */ + enum nf_nat_manip_type maniptype = HOOK2MANIP(hooknum); + + /* We never see fragments: conntrack defrags on pre-routing + and local-out, and nf_nat_out protects post-routing. */ + NF_CT_ASSERT(!((*pskb)->nh.iph->frag_off + & htons(IP_MF|IP_OFFSET))); + + ct = nf_ct_get(*pskb, &ctinfo); + /* Can't track? It's not due to stress, or conntrack would + have dropped it. Hence it's the user's responsibilty to + packet filter it out, or implement conntrack/NAT for that + protocol. 8) --RR */ + if (!ct) { + /* Exception: ICMP redirect to new connection (not in + hash table yet). We must not let this through, in + case we're doing NAT to the same network. */ + if ((*pskb)->nh.iph->protocol == IPPROTO_ICMP) { + struct icmphdr _hdr, *hp; + + hp = skb_header_pointer(*pskb, + (*pskb)->nh.iph->ihl*4, + sizeof(_hdr), &_hdr); + if (hp != NULL && + hp->type == ICMP_REDIRECT) + return NF_DROP; + } + return NF_ACCEPT; + } + + /* Don't try to NAT if this packet is not conntracked */ + if (ct == &nf_conntrack_untracked) + return NF_ACCEPT; + + nat = nfct_nat(ct); + if (!nat) + return NF_DROP; + + switch (ctinfo) { + case IP_CT_RELATED: + case IP_CT_RELATED+IP_CT_IS_REPLY: + if ((*pskb)->nh.iph->protocol == IPPROTO_ICMP) { + if (!nf_nat_icmp_reply_translation(ct, ctinfo, + hooknum, pskb)) + return NF_DROP; + else + return NF_ACCEPT; + } + /* Fall thru... (Only ICMPs can be IP_CT_IS_REPLY) */ + case IP_CT_NEW: + info = &nat->info; + + /* Seen it before? This can happen for loopback, retrans, + or local packets.. */ + if (!nf_nat_initialized(ct, maniptype)) { + unsigned int ret; + + if (unlikely(nf_ct_is_confirmed(ct))) + /* NAT module was loaded late */ + ret = alloc_null_binding_confirmed(ct, info, + hooknum); + else if (hooknum == NF_IP_LOCAL_IN) + /* LOCAL_IN hook doesn't have a chain! */ + ret = alloc_null_binding(ct, info, hooknum); + else + ret = nf_nat_rule_find(pskb, hooknum, in, out, + ct, info); + + if (ret != NF_ACCEPT) { + return ret; + } + } else + DEBUGP("Already setup manip %s for ct %p\n", + maniptype == IP_NAT_MANIP_SRC ? "SRC" : "DST", + ct); + break; + + default: + /* ESTABLISHED */ + NF_CT_ASSERT(ctinfo == IP_CT_ESTABLISHED || + ctinfo == (IP_CT_ESTABLISHED+IP_CT_IS_REPLY)); + info = &nat->info; + } + + NF_CT_ASSERT(info); + return nf_nat_packet(ct, ctinfo, hooknum, pskb); +} + +static unsigned int +nf_nat_in(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + unsigned int ret; + __be32 daddr = (*pskb)->nh.iph->daddr; + + ret = nf_nat_fn(hooknum, pskb, in, out, okfn); + if (ret != NF_DROP && ret != NF_STOLEN && + daddr != (*pskb)->nh.iph->daddr) { + dst_release((*pskb)->dst); + (*pskb)->dst = NULL; + } + return ret; +} + +static unsigned int +nf_nat_out(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ +#ifdef CONFIG_XFRM + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; +#endif + unsigned int ret; + + /* root is playing with raw sockets. */ + if ((*pskb)->len < sizeof(struct iphdr) || + (*pskb)->nh.iph->ihl * 4 < sizeof(struct iphdr)) + return NF_ACCEPT; + + ret = nf_nat_fn(hooknum, pskb, in, out, okfn); +#ifdef CONFIG_XFRM + if (ret != NF_DROP && ret != NF_STOLEN && + (ct = nf_ct_get(*pskb, &ctinfo)) != NULL) { + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + + if (ct->tuplehash[dir].tuple.src.u3.ip != + ct->tuplehash[!dir].tuple.dst.u3.ip + || ct->tuplehash[dir].tuple.src.u.all != + ct->tuplehash[!dir].tuple.dst.u.all + ) + return ip_xfrm_me_harder(pskb) == 0 ? ret : NF_DROP; + } +#endif + return ret; +} + +static unsigned int +nf_nat_local_fn(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + unsigned int ret; + + /* root is playing with raw sockets. */ + if ((*pskb)->len < sizeof(struct iphdr) || + (*pskb)->nh.iph->ihl * 4 < sizeof(struct iphdr)) + return NF_ACCEPT; + + ret = nf_nat_fn(hooknum, pskb, in, out, okfn); + if (ret != NF_DROP && ret != NF_STOLEN && + (ct = nf_ct_get(*pskb, &ctinfo)) != NULL) { + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + + if (ct->tuplehash[dir].tuple.dst.u3.ip != + ct->tuplehash[!dir].tuple.src.u3.ip +#ifdef CONFIG_XFRM + || ct->tuplehash[dir].tuple.dst.u.all != + ct->tuplehash[!dir].tuple.src.u.all +#endif + ) + if (ip_route_me_harder(pskb, RTN_UNSPEC)) + ret = NF_DROP; + } + return ret; +} + +static unsigned int +nf_nat_adjust(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + + ct = nf_ct_get(*pskb, &ctinfo); + if (ct && test_bit(IPS_SEQ_ADJUST_BIT, &ct->status)) { + DEBUGP("nf_nat_standalone: adjusting sequence number\n"); + if (!nf_nat_seq_adjust(pskb, ct, ctinfo)) + return NF_DROP; + } + return NF_ACCEPT; +} + +/* We must be after connection tracking and before packet filtering. */ + +static struct nf_hook_ops nf_nat_ops[] = { + /* Before packet filtering, change destination */ + { + .hook = nf_nat_in, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_PRE_ROUTING, + .priority = NF_IP_PRI_NAT_DST, + }, + /* After packet filtering, change source */ + { + .hook = nf_nat_out, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_POST_ROUTING, + .priority = NF_IP_PRI_NAT_SRC, + }, + /* After conntrack, adjust sequence number */ + { + .hook = nf_nat_adjust, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_POST_ROUTING, + .priority = NF_IP_PRI_NAT_SEQ_ADJUST, + }, + /* Before packet filtering, change destination */ + { + .hook = nf_nat_local_fn, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_LOCAL_OUT, + .priority = NF_IP_PRI_NAT_DST, + }, + /* After packet filtering, change source */ + { + .hook = nf_nat_fn, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_LOCAL_IN, + .priority = NF_IP_PRI_NAT_SRC, + }, + /* After conntrack, adjust sequence number */ + { + .hook = nf_nat_adjust, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_LOCAL_IN, + .priority = NF_IP_PRI_NAT_SEQ_ADJUST, + }, +}; + +static int __init nf_nat_standalone_init(void) +{ + int size, ret = 0; + + need_conntrack(); + + size = ALIGN(sizeof(struct nf_conn), __alignof__(struct nf_conn_nat)) + + sizeof(struct nf_conn_nat); + ret = nf_conntrack_register_cache(NF_CT_F_NAT, "nf_nat:base", size); + if (ret < 0) { + printk(KERN_ERR "nf_nat_init: Unable to create slab cache\n"); + return ret; + } + + size = ALIGN(size, __alignof__(struct nf_conn_help)) + + sizeof(struct nf_conn_help); + ret = nf_conntrack_register_cache(NF_CT_F_NAT|NF_CT_F_HELP, + "nf_nat:help", size); + if (ret < 0) { + printk(KERN_ERR "nf_nat_init: Unable to create slab cache\n"); + goto cleanup_register_cache; + } +#ifdef CONFIG_XFRM + BUG_ON(ip_nat_decode_session != NULL); + ip_nat_decode_session = nat_decode_session; +#endif + ret = nf_nat_rule_init(); + if (ret < 0) { + printk("nf_nat_init: can't setup rules.\n"); + goto cleanup_decode_session; + } + ret = nf_register_hooks(nf_nat_ops, ARRAY_SIZE(nf_nat_ops)); + if (ret < 0) { + printk("nf_nat_init: can't register hooks.\n"); + goto cleanup_rule_init; + } + nf_nat_module_is_loaded = 1; + return ret; + + cleanup_rule_init: + nf_nat_rule_cleanup(); + cleanup_decode_session: +#ifdef CONFIG_XFRM + ip_nat_decode_session = NULL; + synchronize_net(); +#endif + nf_conntrack_unregister_cache(NF_CT_F_NAT|NF_CT_F_HELP); + cleanup_register_cache: + nf_conntrack_unregister_cache(NF_CT_F_NAT); + return ret; +} + +static void __exit nf_nat_standalone_fini(void) +{ + nf_unregister_hooks(nf_nat_ops, ARRAY_SIZE(nf_nat_ops)); + nf_nat_rule_cleanup(); + nf_nat_module_is_loaded = 0; +#ifdef CONFIG_XFRM + ip_nat_decode_session = NULL; + synchronize_net(); +#endif + /* Conntrack caches are unregistered in nf_conntrack_cleanup */ +} + +module_init(nf_nat_standalone_init); +module_exit(nf_nat_standalone_fini); + +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ip_nat"); diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index f952a7fb6ae..aa8beabfeeb 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -579,7 +579,8 @@ __nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, /* FIXME: protect helper list per RCU */ read_lock_bh(&nf_conntrack_lock); helper = __nf_ct_helper_find(repl); - if (helper) + /* NAT might want to assign a helper later */ + if (helper || features & NF_CT_F_NAT) features |= NF_CT_F_HELP; read_unlock_bh(&nf_conntrack_lock); @@ -850,6 +851,26 @@ int nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse, orig->dst.protonum)); } +/* Alter reply tuple (maybe alter helper). This is for NAT, and is + implicitly racy: see __nf_conntrack_confirm */ +void nf_conntrack_alter_reply(struct nf_conn *ct, + const struct nf_conntrack_tuple *newreply) +{ + struct nf_conn_help *help = nfct_help(ct); + + write_lock_bh(&nf_conntrack_lock); + /* Should be unconfirmed, so not in hash table yet */ + NF_CT_ASSERT(!nf_ct_is_confirmed(ct)); + + DEBUGP("Altering reply tuple of %p to ", ct); + NF_CT_DUMP_TUPLE(newreply); + + ct->tuplehash[IP_CT_DIR_REPLY].tuple = *newreply; + if (!ct->master && help && help->expecting == 0) + help->helper = __nf_ct_helper_find(newreply); + write_unlock_bh(&nf_conntrack_lock); +} + /* Refresh conntrack for this many jiffies and do accounting if do_acct is 1 */ void __nf_ct_refresh_acct(struct nf_conn *ct, enum ip_conntrack_info ctinfo, diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index f9c8ddd5973..bd1d2de75e4 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -39,7 +39,11 @@ #include #include #include -#include +#include +#ifdef CONFIG_NF_NAT_NEEDED +#include +#include +#endif #include #include @@ -430,7 +434,7 @@ ctnetlink_dump_table(struct sk_buff *skb, struct netlink_callback *cb) restart: list_for_each_prev(i, &nf_conntrack_hash[cb->args[0]]) { h = (struct nf_conntrack_tuple_hash *) i; - if (DIRECTION(h) != IP_CT_DIR_ORIGINAL) + if (NF_CT_DIRECTION(h) != IP_CT_DIR_ORIGINAL) continue; ct = nf_ct_tuplehash_to_ctrack(h); /* Dump entries of a given L3 protocol number. @@ -556,28 +560,28 @@ ctnetlink_parse_tuple(struct nfattr *cda[], struct nf_conntrack_tuple *tuple, return 0; } -#ifdef CONFIG_IP_NF_NAT_NEEDED +#ifdef CONFIG_NF_NAT_NEEDED static const size_t cta_min_protonat[CTA_PROTONAT_MAX] = { [CTA_PROTONAT_PORT_MIN-1] = sizeof(u_int16_t), [CTA_PROTONAT_PORT_MAX-1] = sizeof(u_int16_t), }; -static int ctnetlink_parse_nat_proto(struct nfattr *attr, +static int nfnetlink_parse_nat_proto(struct nfattr *attr, const struct nf_conn *ct, - struct ip_nat_range *range) + struct nf_nat_range *range) { struct nfattr *tb[CTA_PROTONAT_MAX]; - struct ip_nat_protocol *npt; + struct nf_nat_protocol *npt; nfattr_parse_nested(tb, CTA_PROTONAT_MAX, attr); if (nfattr_bad_size(tb, CTA_PROTONAT_MAX, cta_min_protonat)) return -EINVAL; - npt = ip_nat_proto_find_get(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum); + npt = nf_nat_proto_find_get(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum); if (!npt->nfattr_to_range) { - ip_nat_proto_put(npt); + nf_nat_proto_put(npt); return 0; } @@ -585,7 +589,7 @@ static int ctnetlink_parse_nat_proto(struct nfattr *attr, if (npt->nfattr_to_range(tb, range) > 0) range->flags |= IP_NAT_RANGE_PROTO_SPECIFIED; - ip_nat_proto_put(npt); + nf_nat_proto_put(npt); return 0; } @@ -596,8 +600,8 @@ static const size_t cta_min_nat[CTA_NAT_MAX] = { }; static inline int -ctnetlink_parse_nat(struct nfattr *nat, - const struct nf_conn *ct, struct ip_nat_range *range) +nfnetlink_parse_nat(struct nfattr *nat, + const struct nf_conn *ct, struct nf_nat_range *range) { struct nfattr *tb[CTA_NAT_MAX]; int err; @@ -623,7 +627,7 @@ ctnetlink_parse_nat(struct nfattr *nat, if (!tb[CTA_NAT_PROTO-1]) return 0; - err = ctnetlink_parse_nat_proto(tb[CTA_NAT_PROTO-1], ct, range); + err = nfnetlink_parse_nat_proto(tb[CTA_NAT_PROTO-1], ct, range); if (err < 0) return err; @@ -798,35 +802,35 @@ ctnetlink_change_status(struct nf_conn *ct, struct nfattr *cda[]) return -EINVAL; if (cda[CTA_NAT_SRC-1] || cda[CTA_NAT_DST-1]) { -#ifndef CONFIG_IP_NF_NAT_NEEDED +#ifndef CONFIG_NF_NAT_NEEDED return -EINVAL; #else - struct ip_nat_range range; + struct nf_nat_range range; if (cda[CTA_NAT_DST-1]) { - if (ctnetlink_parse_nat(cda[CTA_NAT_DST-1], ct, + if (nfnetlink_parse_nat(cda[CTA_NAT_DST-1], ct, &range) < 0) return -EINVAL; - if (ip_nat_initialized(ct, + if (nf_nat_initialized(ct, HOOK2MANIP(NF_IP_PRE_ROUTING))) return -EEXIST; - ip_nat_setup_info(ct, &range, hooknum); + nf_nat_setup_info(ct, &range, NF_IP_PRE_ROUTING); } if (cda[CTA_NAT_SRC-1]) { - if (ctnetlink_parse_nat(cda[CTA_NAT_SRC-1], ct, + if (nfnetlink_parse_nat(cda[CTA_NAT_SRC-1], ct, &range) < 0) return -EINVAL; - if (ip_nat_initialized(ct, + if (nf_nat_initialized(ct, HOOK2MANIP(NF_IP_POST_ROUTING))) return -EEXIST; - ip_nat_setup_info(ct, &range, hooknum); + nf_nat_setup_info(ct, &range, NF_IP_POST_ROUTING); } #endif } /* Be careful here, modifying NAT bits can screw up things, * so don't let users modify them directly if they don't pass - * ip_nat_range. */ + * nf_nat_range. */ ct->status |= status & ~(IPS_NAT_DONE_MASK | IPS_NAT_MASK); return 0; } diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index 71f492fc641..8156e429b88 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -730,7 +730,7 @@ static int tcp_in_window(struct ip_ct_tcp *state, return res; } -#ifdef CONFIG_IP_NF_NAT_NEEDED +#ifdef CONFIG_NF_NAT_NEEDED /* Update sender->td_end after NAT successfully mangled the packet */ /* Caller must linearize skb at tcp header. */ void nf_conntrack_tcp_update(struct sk_buff *skb, diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index be94b635972..3f56a3a6c39 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -530,8 +530,11 @@ EXPORT_SYMBOL(nf_conntrack_lock); EXPORT_SYMBOL(nf_conntrack_hash); EXPORT_SYMBOL(nf_conntrack_untracked); EXPORT_SYMBOL_GPL(nf_conntrack_find_get); -#ifdef CONFIG_IP_NF_NAT_NEEDED +#ifdef CONFIG_NF_NAT_NEEDED EXPORT_SYMBOL(nf_conntrack_tcp_update); +EXPORT_SYMBOL(nf_conntrack_register_cache); +EXPORT_SYMBOL(nf_conntrack_unregister_cache); +EXPORT_SYMBOL(nf_conntrack_alter_reply); #endif EXPORT_SYMBOL(__nf_conntrack_confirm); EXPORT_SYMBOL(nf_ct_get_tuple); -- cgit v1.2.3-70-g09d2 From 55a733247d6d2883d9bb77825fafac3dfca13fc2 Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Sat, 2 Dec 2006 22:07:44 -0800 Subject: [NETFILTER]: nf_nat: add FTP NAT helper port Add FTP NAT helper. Split out from Jozsef's big nf_nat patch with a few small fixes by myself. Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_ftp.h | 20 +-- include/linux/netfilter_ipv4/ip_conntrack_ftp.h | 40 +++++- include/net/netfilter/nf_conntrack.h | 2 +- net/ipv4/netfilter/Kconfig | 25 ++-- net/ipv4/netfilter/Makefile | 5 +- net/ipv4/netfilter/nf_nat_ftp.c | 179 ++++++++++++++++++++++++ net/netfilter/nf_conntrack_ftp.c | 20 +-- 7 files changed, 260 insertions(+), 31 deletions(-) create mode 100644 net/ipv4/netfilter/nf_nat_ftp.c (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_ftp.h b/include/linux/netfilter/nf_conntrack_ftp.h index ad4a41c9ce9..81453ea7e4c 100644 --- a/include/linux/netfilter/nf_conntrack_ftp.h +++ b/include/linux/netfilter/nf_conntrack_ftp.h @@ -3,16 +3,16 @@ /* FTP tracking. */ /* This enum is exposed to userspace */ -enum ip_ct_ftp_type +enum nf_ct_ftp_type { /* PORT command from client */ - IP_CT_FTP_PORT, + NF_CT_FTP_PORT, /* PASV response from server */ - IP_CT_FTP_PASV, + NF_CT_FTP_PASV, /* EPRT command from client */ - IP_CT_FTP_EPRT, + NF_CT_FTP_EPRT, /* EPSV response from server */ - IP_CT_FTP_EPSV, + NF_CT_FTP_EPSV, }; #ifdef __KERNEL__ @@ -21,23 +21,23 @@ enum ip_ct_ftp_type #define NUM_SEQ_TO_REMEMBER 2 /* This structure exists only once per master */ -struct ip_ct_ftp_master { +struct nf_ct_ftp_master { /* Valid seq positions for cmd matching after newline */ u_int32_t seq_aft_nl[IP_CT_DIR_MAX][NUM_SEQ_TO_REMEMBER]; /* 0 means seq_match_aft_nl not set */ int seq_aft_nl_num[IP_CT_DIR_MAX]; }; -struct ip_conntrack_expect; +struct nf_conntrack_expect; /* For NAT to hook in when we find a packet which describes what other * connection we should expect. */ -extern unsigned int (*ip_nat_ftp_hook)(struct sk_buff **pskb, +extern unsigned int (*nf_nat_ftp_hook)(struct sk_buff **pskb, enum ip_conntrack_info ctinfo, - enum ip_ct_ftp_type type, + enum nf_ct_ftp_type type, unsigned int matchoff, unsigned int matchlen, - struct ip_conntrack_expect *exp, + struct nf_conntrack_expect *exp, u32 *seq); #endif /* __KERNEL__ */ diff --git a/include/linux/netfilter_ipv4/ip_conntrack_ftp.h b/include/linux/netfilter_ipv4/ip_conntrack_ftp.h index 63811934de4..2129fc3972a 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_ftp.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_ftp.h @@ -1,6 +1,44 @@ #ifndef _IP_CONNTRACK_FTP_H #define _IP_CONNTRACK_FTP_H +/* FTP tracking. */ -#include +/* This enum is exposed to userspace */ +enum ip_ct_ftp_type +{ + /* PORT command from client */ + IP_CT_FTP_PORT, + /* PASV response from server */ + IP_CT_FTP_PASV, + /* EPRT command from client */ + IP_CT_FTP_EPRT, + /* EPSV response from server */ + IP_CT_FTP_EPSV, +}; + +#ifdef __KERNEL__ + +#define FTP_PORT 21 + +#define NUM_SEQ_TO_REMEMBER 2 +/* This structure exists only once per master */ +struct ip_ct_ftp_master { + /* Valid seq positions for cmd matching after newline */ + u_int32_t seq_aft_nl[IP_CT_DIR_MAX][NUM_SEQ_TO_REMEMBER]; + /* 0 means seq_match_aft_nl not set */ + int seq_aft_nl_num[IP_CT_DIR_MAX]; +}; + +struct ip_conntrack_expect; + +/* For NAT to hook in when we find a packet which describes what other + * connection we should expect. */ +extern unsigned int (*ip_nat_ftp_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + enum ip_ct_ftp_type type, + unsigned int matchoff, + unsigned int matchlen, + struct ip_conntrack_expect *exp, + u32 *seq); +#endif /* __KERNEL__ */ #endif /* _IP_CONNTRACK_FTP_H */ diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 9948af06868..83694cfdfa8 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -45,7 +45,7 @@ union nf_conntrack_expect_proto { /* per conntrack: application helper private data */ union nf_conntrack_help { /* insert conntrack helper private data (master) here */ - struct ip_ct_ftp_master ct_ftp_info; + struct nf_ct_ftp_master ct_ftp_info; }; #include diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 01789aeaeb5..52f876db68f 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -477,20 +477,29 @@ config IP_NF_NAT_SNMP_BASIC To compile it as a module, choose M here. If unsure, say N. +# If they want FTP, set to $CONFIG_IP_NF_NAT (m or y), +# or $CONFIG_IP_NF_FTP (m or y), whichever is weaker. +# From kconfig-language.txt: +# +# '&&' (6) +# +# (6) Returns the result of min(/expr/, /expr/). +config IP_NF_NAT_FTP + tristate + depends on IP_NF_IPTABLES && IP_NF_CONNTRACK && IP_NF_NAT + default IP_NF_NAT && IP_NF_FTP + +config NF_NAT_FTP + tristate + depends on IP_NF_IPTABLES && NF_CONNTRACK && NF_NAT + default NF_NAT && NF_CONNTRACK_FTP + config IP_NF_NAT_IRC tristate depends on IP_NF_IPTABLES!=n && IP_NF_CONNTRACK!=n && IP_NF_NAT!=n default IP_NF_NAT if IP_NF_IRC=y default m if IP_NF_IRC=m -# If they want FTP, set to $CONFIG_IP_NF_NAT (m or y), -# or $CONFIG_IP_NF_FTP (m or y), whichever is weaker. Argh. -config IP_NF_NAT_FTP - tristate - depends on IP_NF_IPTABLES!=n && IP_NF_CONNTRACK!=n && IP_NF_NAT!=n - default IP_NF_NAT if IP_NF_FTP=y - default m if IP_NF_FTP=m - config IP_NF_NAT_TFTP tristate depends on IP_NF_IPTABLES!=n && IP_NF_CONNTRACK!=n && IP_NF_NAT!=n diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index ec31690764a..c0c6194bb27 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -40,7 +40,7 @@ obj-$(CONFIG_IP_NF_IRC) += ip_conntrack_irc.o obj-$(CONFIG_IP_NF_SIP) += ip_conntrack_sip.o obj-$(CONFIG_IP_NF_NETBIOS_NS) += ip_conntrack_netbios_ns.o -# NAT helpers +# NAT helpers (ip_conntrack) obj-$(CONFIG_IP_NF_NAT_H323) += ip_nat_h323.o obj-$(CONFIG_IP_NF_NAT_PPTP) += ip_nat_pptp.o obj-$(CONFIG_IP_NF_NAT_AMANDA) += ip_nat_amanda.o @@ -49,6 +49,9 @@ obj-$(CONFIG_IP_NF_NAT_FTP) += ip_nat_ftp.o obj-$(CONFIG_IP_NF_NAT_IRC) += ip_nat_irc.o obj-$(CONFIG_IP_NF_NAT_SIP) += ip_nat_sip.o +# NAT helpers (nf_conntrack) +obj-$(CONFIG_NF_NAT_FTP) += nf_nat_ftp.o + # generic IP tables obj-$(CONFIG_IP_NF_IPTABLES) += ip_tables.o diff --git a/net/ipv4/netfilter/nf_nat_ftp.c b/net/ipv4/netfilter/nf_nat_ftp.c new file mode 100644 index 00000000000..751b5980175 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_ftp.c @@ -0,0 +1,179 @@ +/* FTP extension for TCP NAT alteration. */ + +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Rusty Russell "); +MODULE_DESCRIPTION("ftp NAT helper"); +MODULE_ALIAS("ip_nat_ftp"); + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +/* FIXME: Time out? --RR */ + +static int +mangle_rfc959_packet(struct sk_buff **pskb, + __be32 newip, + u_int16_t port, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + u32 *seq) +{ + char buffer[sizeof("nnn,nnn,nnn,nnn,nnn,nnn")]; + + sprintf(buffer, "%u,%u,%u,%u,%u,%u", + NIPQUAD(newip), port>>8, port&0xFF); + + DEBUGP("calling nf_nat_mangle_tcp_packet\n"); + + *seq += strlen(buffer) - matchlen; + return nf_nat_mangle_tcp_packet(pskb, ct, ctinfo, matchoff, + matchlen, buffer, strlen(buffer)); +} + +/* |1|132.235.1.2|6275| */ +static int +mangle_eprt_packet(struct sk_buff **pskb, + __be32 newip, + u_int16_t port, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + u32 *seq) +{ + char buffer[sizeof("|1|255.255.255.255|65535|")]; + + sprintf(buffer, "|1|%u.%u.%u.%u|%u|", NIPQUAD(newip), port); + + DEBUGP("calling nf_nat_mangle_tcp_packet\n"); + + *seq += strlen(buffer) - matchlen; + return nf_nat_mangle_tcp_packet(pskb, ct, ctinfo, matchoff, + matchlen, buffer, strlen(buffer)); +} + +/* |1|132.235.1.2|6275| */ +static int +mangle_epsv_packet(struct sk_buff **pskb, + __be32 newip, + u_int16_t port, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + u32 *seq) +{ + char buffer[sizeof("|||65535|")]; + + sprintf(buffer, "|||%u|", port); + + DEBUGP("calling nf_nat_mangle_tcp_packet\n"); + + *seq += strlen(buffer) - matchlen; + return nf_nat_mangle_tcp_packet(pskb, ct, ctinfo, matchoff, + matchlen, buffer, strlen(buffer)); +} + +static int (*mangle[])(struct sk_buff **, __be32, u_int16_t, + unsigned int, unsigned int, struct nf_conn *, + enum ip_conntrack_info, u32 *seq) += { + [NF_CT_FTP_PORT] = mangle_rfc959_packet, + [NF_CT_FTP_PASV] = mangle_rfc959_packet, + [NF_CT_FTP_EPRT] = mangle_eprt_packet, + [NF_CT_FTP_EPSV] = mangle_epsv_packet +}; + +/* So, this packet has hit the connection tracking matching code. + Mangle it, and change the expectation to match the new version. */ +static unsigned int nf_nat_ftp(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + enum nf_ct_ftp_type type, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conntrack_expect *exp, + u32 *seq) +{ + __be32 newip; + u_int16_t port; + int dir = CTINFO2DIR(ctinfo); + struct nf_conn *ct = exp->master; + + DEBUGP("FTP_NAT: type %i, off %u len %u\n", type, matchoff, matchlen); + + /* Connection will come from wherever this packet goes, hence !dir */ + newip = ct->tuplehash[!dir].tuple.dst.u3.ip; + exp->saved_proto.tcp.port = exp->tuple.dst.u.tcp.port; + exp->dir = !dir; + + /* When you see the packet, we need to NAT it the same as the + * this one. */ + exp->expectfn = nf_nat_follow_master; + + /* Try to get same port: if not, try to change it. */ + for (port = ntohs(exp->saved_proto.tcp.port); port != 0; port++) { + exp->tuple.dst.u.tcp.port = htons(port); + if (nf_conntrack_expect_related(exp) == 0) + break; + } + + if (port == 0) + return NF_DROP; + + if (!mangle[type](pskb, newip, port, matchoff, matchlen, ct, ctinfo, + seq)) { + nf_conntrack_unexpect_related(exp); + return NF_DROP; + } + return NF_ACCEPT; +} + +static void __exit nf_nat_ftp_fini(void) +{ + rcu_assign_pointer(nf_nat_ftp_hook, NULL); + synchronize_rcu(); +} + +static int __init nf_nat_ftp_init(void) +{ + BUG_ON(rcu_dereference(nf_nat_ftp_hook)); + rcu_assign_pointer(nf_nat_ftp_hook, nf_nat_ftp); + return 0; +} + +/* Prior to 2.6.11, we had a ports param. No longer, but don't break users. */ +static int warn_set(const char *val, struct kernel_param *kp) +{ + printk(KERN_INFO KBUILD_MODNAME + ": kernel >= 2.6.10 only uses 'ports' for conntrack modules\n"); + return 0; +} +module_param_call(ports, warn_set, NULL, NULL, 0); + +module_init(nf_nat_ftp_init); +module_exit(nf_nat_ftp_fini); diff --git a/net/netfilter/nf_conntrack_ftp.c b/net/netfilter/nf_conntrack_ftp.c index e299d657e4f..92a94716876 100644 --- a/net/netfilter/nf_conntrack_ftp.c +++ b/net/netfilter/nf_conntrack_ftp.c @@ -51,7 +51,7 @@ module_param(loose, bool, 0600); unsigned int (*nf_nat_ftp_hook)(struct sk_buff **pskb, enum ip_conntrack_info ctinfo, - enum ip_ct_ftp_type type, + enum nf_ct_ftp_type type, unsigned int matchoff, unsigned int matchlen, struct nf_conntrack_expect *exp, @@ -74,7 +74,7 @@ static struct ftp_search { size_t plen; char skip; char term; - enum ip_ct_ftp_type ftptype; + enum nf_ct_ftp_type ftptype; int (*getnum)(const char *, size_t, struct nf_conntrack_man *, char); } search[IP_CT_DIR_MAX][2] = { [IP_CT_DIR_ORIGINAL] = { @@ -83,7 +83,7 @@ static struct ftp_search { .plen = sizeof("PORT") - 1, .skip = ' ', .term = '\r', - .ftptype = IP_CT_FTP_PORT, + .ftptype = NF_CT_FTP_PORT, .getnum = try_rfc959, }, { @@ -91,7 +91,7 @@ static struct ftp_search { .plen = sizeof("EPRT") - 1, .skip = ' ', .term = '\r', - .ftptype = IP_CT_FTP_EPRT, + .ftptype = NF_CT_FTP_EPRT, .getnum = try_eprt, }, }, @@ -101,7 +101,7 @@ static struct ftp_search { .plen = sizeof("227 ") - 1, .skip = '(', .term = ')', - .ftptype = IP_CT_FTP_PASV, + .ftptype = NF_CT_FTP_PASV, .getnum = try_rfc959, }, { @@ -109,7 +109,7 @@ static struct ftp_search { .plen = sizeof("229 ") - 1, .skip = '(', .term = ')', - .ftptype = IP_CT_FTP_EPSV, + .ftptype = NF_CT_FTP_EPSV, .getnum = try_epsv_response, }, }, @@ -320,7 +320,7 @@ static int find_pattern(const char *data, size_t dlen, } /* Look up to see if we're just after a \n. */ -static int find_nl_seq(u32 seq, const struct ip_ct_ftp_master *info, int dir) +static int find_nl_seq(u32 seq, const struct nf_ct_ftp_master *info, int dir) { unsigned int i; @@ -331,7 +331,7 @@ static int find_nl_seq(u32 seq, const struct ip_ct_ftp_master *info, int dir) } /* We don't update if it's older than what we have. */ -static void update_nl_seq(u32 nl_seq, struct ip_ct_ftp_master *info, int dir, +static void update_nl_seq(u32 nl_seq, struct nf_ct_ftp_master *info, int dir, struct sk_buff *skb) { unsigned int i, oldest = NUM_SEQ_TO_REMEMBER; @@ -367,7 +367,7 @@ static int help(struct sk_buff **pskb, u32 seq; int dir = CTINFO2DIR(ctinfo); unsigned int matchlen, matchoff; - struct ip_ct_ftp_master *ct_ftp_info = &nfct_help(ct)->help.ct_ftp_info; + struct nf_ct_ftp_master *ct_ftp_info = &nfct_help(ct)->help.ct_ftp_info; struct nf_conntrack_expect *exp; struct nf_conntrack_man cmd = {}; unsigned int i; @@ -523,7 +523,7 @@ static int help(struct sk_buff **pskb, /* Now, NAT might want to mangle the packet, and register the * (possibly changed) expectation itself. */ nf_nat_ftp = rcu_dereference(nf_nat_ftp_hook); - if (nf_nat_ftp) + if (nf_nat_ftp && ct->status & IPS_NAT_MASK) ret = nf_nat_ftp(pskb, ctinfo, search[dir][i].ftptype, matchoff, matchlen, exp, &seq); else { -- cgit v1.2.3-70-g09d2 From d6a9b6500a8941599bcef98e7de49e1260d104ed Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 2 Dec 2006 22:08:01 -0800 Subject: [NETFILTER]: nf_conntrack: add helper function for expectation initialization Expectation address masks need to be differently initialized depending on the address family, create helper function to avoid cluttering up the code too much. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/net/netfilter/nf_conntrack_expect.h | 4 ++ include/net/netfilter/nf_conntrack_tuple.h | 10 ++--- net/netfilter/nf_conntrack_expect.c | 68 +++++++++++++++++++++++++++++ 3 files changed, 75 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h index b969c430b36..54a3d038bea 100644 --- a/include/net/netfilter/nf_conntrack_expect.h +++ b/include/net/netfilter/nf_conntrack_expect.h @@ -68,6 +68,10 @@ void nf_conntrack_unexpect_related(struct nf_conntrack_expect *exp); /* Allocate space for an expectation: this is mandatory before calling nf_conntrack_expect_related. You will have to call put afterwards. */ struct nf_conntrack_expect *nf_conntrack_expect_alloc(struct nf_conn *me); +void nf_conntrack_expect_init(struct nf_conntrack_expect *, int, + union nf_conntrack_address *, + union nf_conntrack_address *, + u_int8_t, __be16 *, __be16 *); void nf_conntrack_expect_put(struct nf_conntrack_expect *exp); int nf_conntrack_expect_related(struct nf_conntrack_expect *expect); diff --git a/include/net/netfilter/nf_conntrack_tuple.h b/include/net/netfilter/nf_conntrack_tuple.h index be9dc9a0eb7..c96a9c57673 100644 --- a/include/net/netfilter/nf_conntrack_tuple.h +++ b/include/net/netfilter/nf_conntrack_tuple.h @@ -24,7 +24,7 @@ /* The l3 protocol-specific manipulable parts of the tuple: always in network order! */ -union nf_conntrack_man_l3proto { +union nf_conntrack_address { u_int32_t all[NF_CT_TUPLE_L3SIZE]; __be32 ip; __be32 ip6[4]; @@ -54,7 +54,7 @@ union nf_conntrack_man_proto /* The manipulable part of the tuple. */ struct nf_conntrack_man { - union nf_conntrack_man_l3proto u3; + union nf_conntrack_address u3; union nf_conntrack_man_proto u; /* Layer 3 protocol */ u_int16_t l3num; @@ -67,11 +67,7 @@ struct nf_conntrack_tuple /* These are the parts of the tuple which are fixed. */ struct { - union { - u_int32_t all[NF_CT_TUPLE_L3SIZE]; - u_int32_t ip; - u_int32_t ip6[4]; - } u3; + union nf_conntrack_address u3; union { /* Add other protocols here. */ u_int16_t all; diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index aa5903e4da1..68623ae778c 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -196,6 +196,74 @@ struct nf_conntrack_expect *nf_conntrack_expect_alloc(struct nf_conn *me) return new; } +void nf_conntrack_expect_init(struct nf_conntrack_expect *exp, int family, + union nf_conntrack_address *saddr, + union nf_conntrack_address *daddr, + u_int8_t proto, __be16 *src, __be16 *dst) +{ + int len; + + if (family == AF_INET) + len = 4; + else + len = 16; + + exp->flags = 0; + exp->expectfn = NULL; + exp->helper = NULL; + exp->tuple.src.l3num = family; + exp->tuple.dst.protonum = proto; + exp->mask.src.l3num = 0xFFFF; + exp->mask.dst.protonum = 0xFF; + + if (saddr) { + memcpy(&exp->tuple.src.u3, saddr, len); + if (sizeof(exp->tuple.src.u3) > len) + /* address needs to be cleared for nf_ct_tuple_equal */ + memset((void *)&exp->tuple.src.u3 + len, 0x00, + sizeof(exp->tuple.src.u3) - len); + memset(&exp->mask.src.u3, 0xFF, len); + if (sizeof(exp->mask.src.u3) > len) + memset((void *)&exp->mask.src.u3 + len, 0x00, + sizeof(exp->mask.src.u3) - len); + } else { + memset(&exp->tuple.src.u3, 0x00, sizeof(exp->tuple.src.u3)); + memset(&exp->mask.src.u3, 0x00, sizeof(exp->mask.src.u3)); + } + + if (daddr) { + memcpy(&exp->tuple.dst.u3, daddr, len); + if (sizeof(exp->tuple.dst.u3) > len) + /* address needs to be cleared for nf_ct_tuple_equal */ + memset((void *)&exp->tuple.dst.u3 + len, 0x00, + sizeof(exp->tuple.dst.u3) - len); + memset(&exp->mask.dst.u3, 0xFF, len); + if (sizeof(exp->mask.dst.u3) > len) + memset((void *)&exp->mask.dst.u3 + len, 0x00, + sizeof(exp->mask.dst.u3) - len); + } else { + memset(&exp->tuple.dst.u3, 0x00, sizeof(exp->tuple.dst.u3)); + memset(&exp->mask.dst.u3, 0x00, sizeof(exp->mask.dst.u3)); + } + + if (src) { + exp->tuple.src.u.all = (__force u16)*src; + exp->mask.src.u.all = 0xFFFF; + } else { + exp->tuple.src.u.all = 0; + exp->mask.src.u.all = 0; + } + + if (dst) { + exp->tuple.dst.u.all = (__force u16)*dst; + exp->mask.dst.u.all = 0xFFFF; + } else { + exp->tuple.dst.u.all = 0; + exp->mask.dst.u.all = 0; + } +} +EXPORT_SYMBOL_GPL(nf_conntrack_expect_init); + void nf_conntrack_expect_put(struct nf_conntrack_expect *exp) { if (atomic_dec_and_test(&exp->use)) -- cgit v1.2.3-70-g09d2 From 16958900578b94585c2ab9a2d20d837b4d5e3ba6 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 2 Dec 2006 22:08:26 -0800 Subject: [NETFILTER]: nf_conntrack/nf_nat: add amanda helper port Add IPv4 and IPv6 capable nf_conntrack port of the Amanda conntrack/NAT helper. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_amanda.h | 10 ++ net/ipv4/netfilter/Kconfig | 5 + net/ipv4/netfilter/Makefile | 1 + net/ipv4/netfilter/nf_nat_amanda.c | 78 +++++++++ net/netfilter/Kconfig | 15 ++ net/netfilter/Makefile | 1 + net/netfilter/nf_conntrack_amanda.c | 237 ++++++++++++++++++++++++++ 7 files changed, 347 insertions(+) create mode 100644 include/linux/netfilter/nf_conntrack_amanda.h create mode 100644 net/ipv4/netfilter/nf_nat_amanda.c create mode 100644 net/netfilter/nf_conntrack_amanda.c (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_amanda.h b/include/linux/netfilter/nf_conntrack_amanda.h new file mode 100644 index 00000000000..26c223544ae --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_amanda.h @@ -0,0 +1,10 @@ +#ifndef _NF_CONNTRACK_AMANDA_H +#define _NF_CONNTRACK_AMANDA_H +/* AMANDA tracking. */ + +extern unsigned int (*nf_nat_amanda_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conntrack_expect *exp); +#endif /* _NF_CONNTRACK_AMANDA_H */ diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 52f876db68f..6993ec53dc0 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -512,6 +512,11 @@ config IP_NF_NAT_AMANDA default IP_NF_NAT if IP_NF_AMANDA=y default m if IP_NF_AMANDA=m +config NF_NAT_AMANDA + tristate + depends on IP_NF_IPTABLES && NF_CONNTRACK && NF_NAT + default NF_NAT && NF_CONNTRACK_AMANDA + config IP_NF_NAT_PPTP tristate depends on IP_NF_NAT!=n && IP_NF_PPTP!=n diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index c0c6194bb27..8893249bbe9 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -50,6 +50,7 @@ obj-$(CONFIG_IP_NF_NAT_IRC) += ip_nat_irc.o obj-$(CONFIG_IP_NF_NAT_SIP) += ip_nat_sip.o # NAT helpers (nf_conntrack) +obj-$(CONFIG_NF_NAT_AMANDA) += nf_nat_amanda.o obj-$(CONFIG_NF_NAT_FTP) += nf_nat_ftp.o # generic IP tables diff --git a/net/ipv4/netfilter/nf_nat_amanda.c b/net/ipv4/netfilter/nf_nat_amanda.c new file mode 100644 index 00000000000..0f17098917b --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_amanda.c @@ -0,0 +1,78 @@ +/* Amanda extension for TCP NAT alteration. + * (C) 2002 by Brian J. Murrell + * based on a copy of HW's ip_nat_irc.c as well as other modules + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +MODULE_AUTHOR("Brian J. Murrell "); +MODULE_DESCRIPTION("Amanda NAT helper"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ip_nat_amanda"); + +static unsigned int help(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conntrack_expect *exp) +{ + char buffer[sizeof("65535")]; + u_int16_t port; + unsigned int ret; + + /* Connection comes from client. */ + exp->saved_proto.tcp.port = exp->tuple.dst.u.tcp.port; + exp->dir = IP_CT_DIR_ORIGINAL; + + /* When you see the packet, we need to NAT it the same as the + * this one (ie. same IP: it will be TCP and master is UDP). */ + exp->expectfn = nf_nat_follow_master; + + /* Try to get same port: if not, try to change it. */ + for (port = ntohs(exp->saved_proto.tcp.port); port != 0; port++) { + exp->tuple.dst.u.tcp.port = htons(port); + if (nf_conntrack_expect_related(exp) == 0) + break; + } + + if (port == 0) + return NF_DROP; + + sprintf(buffer, "%u", port); + ret = nf_nat_mangle_udp_packet(pskb, exp->master, ctinfo, + matchoff, matchlen, + buffer, strlen(buffer)); + if (ret != NF_ACCEPT) + nf_conntrack_unexpect_related(exp); + return ret; +} + +static void __exit nf_nat_amanda_fini(void) +{ + rcu_assign_pointer(nf_nat_amanda_hook, NULL); + synchronize_rcu(); +} + +static int __init nf_nat_amanda_init(void) +{ + BUG_ON(rcu_dereference(nf_nat_amanda_hook)); + rcu_assign_pointer(nf_nat_amanda_hook, help); + return 0; +} + +module_init(nf_nat_amanda_init); +module_exit(nf_nat_amanda_fini); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index efe56f768f3..f85fd43b344 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -131,6 +131,21 @@ config NF_CT_PROTO_SCTP If you want to compile it as a module, say M here and read Documentation/modules.txt. If unsure, say `N'. +config NF_CONNTRACK_AMANDA + tristate "Amanda backup protocol support (EXPERIMENTAL)" + depends on EXPERIMENTAL && NF_CONNTRACK + select TEXTSEARCH + select TEXTSEARCH_KMP + help + If you are running the Amanda backup package + on this machine or machines that will be MASQUERADED through this + machine, then you may want to enable this feature. This allows the + connection tracking and natting code to allow the sub-channels that + Amanda requires for communication of the backup data, messages and + index. + + To compile it as a module, choose M here. If unsure, say N. + config NF_CONNTRACK_FTP tristate "FTP support on new connection tracking (EXPERIMENTAL)" depends on EXPERIMENTAL && NF_CONNTRACK diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 7f0089c584b..a5ee9381742 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -20,6 +20,7 @@ obj-$(CONFIG_NF_CT_PROTO_SCTP) += nf_conntrack_proto_sctp.o obj-$(CONFIG_NF_CT_NETLINK) += nf_conntrack_netlink.o # connection tracking helpers +obj-$(CONFIG_NF_CONNTRACK_AMANDA) += nf_conntrack_amanda.o obj-$(CONFIG_NF_CONNTRACK_FTP) += nf_conntrack_ftp.o # generic X tables diff --git a/net/netfilter/nf_conntrack_amanda.c b/net/netfilter/nf_conntrack_amanda.c new file mode 100644 index 00000000000..5c495dc9d94 --- /dev/null +++ b/net/netfilter/nf_conntrack_amanda.c @@ -0,0 +1,237 @@ +/* Amanda extension for IP connection tracking + * + * (C) 2002 by Brian J. Murrell + * based on HW's ip_conntrack_irc.c as well as other modules + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +static unsigned int master_timeout __read_mostly = 300; +static char *ts_algo = "kmp"; + +MODULE_AUTHOR("Brian J. Murrell "); +MODULE_DESCRIPTION("Amanda connection tracking module"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ip_conntrack_amanda"); + +module_param(master_timeout, uint, 0600); +MODULE_PARM_DESC(master_timeout, "timeout for the master connection"); +module_param(ts_algo, charp, 0400); +MODULE_PARM_DESC(ts_algo, "textsearch algorithm to use (default kmp)"); + +unsigned int (*nf_nat_amanda_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conntrack_expect *exp) + __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_amanda_hook); + +enum amanda_strings { + SEARCH_CONNECT, + SEARCH_NEWLINE, + SEARCH_DATA, + SEARCH_MESG, + SEARCH_INDEX, +}; + +static struct { + char *string; + size_t len; + struct ts_config *ts; +} search[] __read_mostly = { + [SEARCH_CONNECT] = { + .string = "CONNECT ", + .len = 8, + }, + [SEARCH_NEWLINE] = { + .string = "\n", + .len = 1, + }, + [SEARCH_DATA] = { + .string = "DATA ", + .len = 5, + }, + [SEARCH_MESG] = { + .string = "MESG ", + .len = 5, + }, + [SEARCH_INDEX] = { + .string = "INDEX ", + .len = 6, + }, +}; + +static int amanda_help(struct sk_buff **pskb, + unsigned int protoff, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo) +{ + struct ts_state ts; + struct nf_conntrack_expect *exp; + struct nf_conntrack_tuple *tuple; + unsigned int dataoff, start, stop, off, i; + char pbuf[sizeof("65535")], *tmp; + u_int16_t len; + __be16 port; + int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; + int ret = NF_ACCEPT; + typeof(nf_nat_amanda_hook) nf_nat_amanda; + + /* Only look at packets from the Amanda server */ + if (CTINFO2DIR(ctinfo) == IP_CT_DIR_ORIGINAL) + return NF_ACCEPT; + + /* increase the UDP timeout of the master connection as replies from + * Amanda clients to the server can be quite delayed */ + nf_ct_refresh(ct, *pskb, master_timeout * HZ); + + /* No data? */ + dataoff = protoff + sizeof(struct udphdr); + if (dataoff >= (*pskb)->len) { + if (net_ratelimit()) + printk("amanda_help: skblen = %u\n", (*pskb)->len); + return NF_ACCEPT; + } + + memset(&ts, 0, sizeof(ts)); + start = skb_find_text(*pskb, dataoff, (*pskb)->len, + search[SEARCH_CONNECT].ts, &ts); + if (start == UINT_MAX) + goto out; + start += dataoff + search[SEARCH_CONNECT].len; + + memset(&ts, 0, sizeof(ts)); + stop = skb_find_text(*pskb, start, (*pskb)->len, + search[SEARCH_NEWLINE].ts, &ts); + if (stop == UINT_MAX) + goto out; + stop += start; + + for (i = SEARCH_DATA; i <= SEARCH_INDEX; i++) { + memset(&ts, 0, sizeof(ts)); + off = skb_find_text(*pskb, start, stop, search[i].ts, &ts); + if (off == UINT_MAX) + continue; + off += start + search[i].len; + + len = min_t(unsigned int, sizeof(pbuf) - 1, stop - off); + if (skb_copy_bits(*pskb, off, pbuf, len)) + break; + pbuf[len] = '\0'; + + port = htons(simple_strtoul(pbuf, &tmp, 10)); + len = tmp - pbuf; + if (port == 0 || len > 5) + break; + + exp = nf_conntrack_expect_alloc(ct); + if (exp == NULL) { + ret = NF_DROP; + goto out; + } + tuple = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; + nf_conntrack_expect_init(exp, family, + &tuple->src.u3, &tuple->dst.u3, + IPPROTO_TCP, NULL, &port); + + nf_nat_amanda = rcu_dereference(nf_nat_amanda_hook); + if (nf_nat_amanda && ct->status & IPS_NAT_MASK) + ret = nf_nat_amanda(pskb, ctinfo, off - dataoff, + len, exp); + else if (nf_conntrack_expect_related(exp) != 0) + ret = NF_DROP; + nf_conntrack_expect_put(exp); + } + +out: + return ret; +} + +static struct nf_conntrack_helper amanda_helper[2] __read_mostly = { + { + .name = "amanda", + .max_expected = 3, + .timeout = 180, + .me = THIS_MODULE, + .help = amanda_help, + .tuple.src.l3num = AF_INET, + .tuple.src.u.udp.port = __constant_htons(10080), + .tuple.dst.protonum = IPPROTO_UDP, + .mask.src.l3num = 0xFFFF, + .mask.src.u.udp.port = __constant_htons(0xFFFF), + .mask.dst.protonum = 0xFF, + }, + { + .name = "amanda", + .max_expected = 3, + .timeout = 180, + .me = THIS_MODULE, + .help = amanda_help, + .tuple.src.l3num = AF_INET6, + .tuple.src.u.udp.port = __constant_htons(10080), + .tuple.dst.protonum = IPPROTO_UDP, + .mask.src.l3num = 0xFFFF, + .mask.src.u.udp.port = __constant_htons(0xFFFF), + .mask.dst.protonum = 0xFF, + }, +}; + +static void __exit nf_conntrack_amanda_fini(void) +{ + int i; + + nf_conntrack_helper_unregister(&amanda_helper[0]); + nf_conntrack_helper_unregister(&amanda_helper[1]); + for (i = 0; i < ARRAY_SIZE(search); i++) + textsearch_destroy(search[i].ts); +} + +static int __init nf_conntrack_amanda_init(void) +{ + int ret, i; + + ret = -ENOMEM; + for (i = 0; i < ARRAY_SIZE(search); i++) { + search[i].ts = textsearch_prepare(ts_algo, search[i].string, + search[i].len, + GFP_KERNEL, TS_AUTOLOAD); + if (search[i].ts == NULL) + goto err1; + } + ret = nf_conntrack_helper_register(&amanda_helper[0]); + if (ret < 0) + goto err1; + ret = nf_conntrack_helper_register(&amanda_helper[1]); + if (ret < 0) + goto err2; + return 0; + +err2: + nf_conntrack_helper_unregister(&amanda_helper[0]); +err1: + for (; i >= 0; i--) { + if (search[i].ts) + textsearch_destroy(search[i].ts); + } + return ret; +} + +module_init(nf_conntrack_amanda_init); +module_exit(nf_conntrack_amanda_fini); -- cgit v1.2.3-70-g09d2 From f587de0e2feb9eb9b94f98d0a7b7437e4d6617b4 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 2 Dec 2006 22:08:46 -0800 Subject: [NETFILTER]: nf_conntrack/nf_nat: add H.323 helper port Add IPv4 and IPv6 capable nf_conntrack port of the H.323 conntrack/NAT helper. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_h323.h | 92 + include/linux/netfilter/nf_conntrack_h323_asn1.h | 98 + include/linux/netfilter/nf_conntrack_h323_types.h | 951 ++++++++++ include/linux/netfilter_ipv4/Kbuild | 2 - include/linux/netfilter_ipv4/ip_conntrack_h323.h | 2 +- .../netfilter_ipv4/ip_conntrack_helper_h323_asn1.h | 98 - .../ip_conntrack_helper_h323_types.h | 939 ---------- include/net/netfilter/nf_conntrack.h | 2 + include/net/netfilter/nf_conntrack_expect.h | 1 + net/ipv4/netfilter/Kconfig | 5 + net/ipv4/netfilter/Makefile | 3 +- net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c | 874 --------- .../netfilter/ip_conntrack_helper_h323_types.c | 1926 ------------------- net/ipv4/netfilter/nf_nat_h323.c | 596 ++++++ net/netfilter/Kconfig | 19 + net/netfilter/Makefile | 3 + net/netfilter/nf_conntrack_h323_asn1.c | 874 +++++++++ net/netfilter/nf_conntrack_h323_main.c | 1856 +++++++++++++++++++ net/netfilter/nf_conntrack_h323_types.c | 1927 ++++++++++++++++++++ 19 files changed, 6427 insertions(+), 3841 deletions(-) create mode 100644 include/linux/netfilter/nf_conntrack_h323.h create mode 100644 include/linux/netfilter/nf_conntrack_h323_asn1.h create mode 100644 include/linux/netfilter/nf_conntrack_h323_types.h delete mode 100644 include/linux/netfilter_ipv4/ip_conntrack_helper_h323_asn1.h delete mode 100644 include/linux/netfilter_ipv4/ip_conntrack_helper_h323_types.h delete mode 100644 net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c delete mode 100644 net/ipv4/netfilter/ip_conntrack_helper_h323_types.c create mode 100644 net/ipv4/netfilter/nf_nat_h323.c create mode 100644 net/netfilter/nf_conntrack_h323_asn1.c create mode 100644 net/netfilter/nf_conntrack_h323_main.c create mode 100644 net/netfilter/nf_conntrack_h323_types.c (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_h323.h b/include/linux/netfilter/nf_conntrack_h323.h new file mode 100644 index 00000000000..08e2f4977c2 --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_h323.h @@ -0,0 +1,92 @@ +#ifndef _NF_CONNTRACK_H323_H +#define _NF_CONNTRACK_H323_H + +#ifdef __KERNEL__ + +#include + +#define RAS_PORT 1719 +#define Q931_PORT 1720 +#define H323_RTP_CHANNEL_MAX 4 /* Audio, video, FAX and other */ + +/* This structure exists only once per master */ +struct nf_ct_h323_master { + + /* Original and NATed Q.931 or H.245 signal ports */ + __be16 sig_port[IP_CT_DIR_MAX]; + + /* Original and NATed RTP ports */ + __be16 rtp_port[H323_RTP_CHANNEL_MAX][IP_CT_DIR_MAX]; + + union { + /* RAS connection timeout */ + u_int32_t timeout; + + /* Next TPKT length (for separate TPKT header and data) */ + u_int16_t tpkt_len[IP_CT_DIR_MAX]; + }; +}; + +struct nf_conn; + +extern int get_h225_addr(struct nf_conn *ct, unsigned char *data, + TransportAddress *taddr, + union nf_conntrack_address *addr, __be16 *port); +extern void nf_conntrack_h245_expect(struct nf_conn *new, + struct nf_conntrack_expect *this); +extern void nf_conntrack_q931_expect(struct nf_conn *new, + struct nf_conntrack_expect *this); +extern int (*set_h245_addr_hook) (struct sk_buff **pskb, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr, + union nf_conntrack_address *addr, + __be16 port); +extern int (*set_h225_addr_hook) (struct sk_buff **pskb, + unsigned char **data, int dataoff, + TransportAddress *taddr, + union nf_conntrack_address *addr, + __be16 port); +extern int (*set_sig_addr_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, + TransportAddress *taddr, int count); +extern int (*set_ras_addr_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, + TransportAddress *taddr, int count); +extern int (*nat_rtp_rtcp_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr, + __be16 port, __be16 rtp_port, + struct nf_conntrack_expect *rtp_exp, + struct nf_conntrack_expect *rtcp_exp); +extern int (*nat_t120_hook) (struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr, __be16 port, + struct nf_conntrack_expect *exp); +extern int (*nat_h245_hook) (struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + TransportAddress *taddr, __be16 port, + struct nf_conntrack_expect *exp); +extern int (*nat_callforwarding_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + TransportAddress *taddr, + __be16 port, + struct nf_conntrack_expect *exp); +extern int (*nat_q931_hook) (struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, TransportAddress *taddr, + int idx, __be16 port, + struct nf_conntrack_expect *exp); + +#endif + +#endif diff --git a/include/linux/netfilter/nf_conntrack_h323_asn1.h b/include/linux/netfilter/nf_conntrack_h323_asn1.h new file mode 100644 index 00000000000..8dab5968fc7 --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_h323_asn1.h @@ -0,0 +1,98 @@ +/**************************************************************************** + * ip_conntrack_h323_asn1.h - BER and PER decoding library for H.323 + * conntrack/NAT module. + * + * Copyright (c) 2006 by Jing Min Zhao + * + * This source code is licensed under General Public License version 2. + * + * + * This library is based on H.225 version 4, H.235 version 2 and H.245 + * version 7. It is extremely optimized to decode only the absolutely + * necessary objects in a signal for Linux kernel NAT module use, so don't + * expect it to be a full ASN.1 library. + * + * Features: + * + * 1. Small. The total size of code plus data is less than 20 KB (IA32). + * 2. Fast. Decoding Netmeeting's Setup signal 1 million times on a PIII 866 + * takes only 3.9 seconds. + * 3. No memory allocation. It uses a static object. No need to initialize or + * cleanup. + * 4. Thread safe. + * 5. Support embedded architectures that has no misaligned memory access + * support. + * + * Limitations: + * + * 1. At most 30 faststart entries. Actually this is limited by ethernet's MTU. + * If a Setup signal contains more than 30 faststart, the packet size will + * very likely exceed the MTU size, then the TPKT will be fragmented. I + * don't know how to handle this in a Netfilter module. Anybody can help? + * Although I think 30 is enough for most of the cases. + * 2. IPv4 addresses only. + * + ****************************************************************************/ + +#ifndef _NF_CONNTRACK_HELPER_H323_ASN1_H_ +#define _NF_CONNTRACK_HELPER_H323_ASN1_H_ + +/***************************************************************************** + * H.323 Types + ****************************************************************************/ +#include "nf_conntrack_h323_types.h" + +typedef struct { + enum { + Q931_NationalEscape = 0x00, + Q931_Alerting = 0x01, + Q931_CallProceeding = 0x02, + Q931_Connect = 0x07, + Q931_ConnectAck = 0x0F, + Q931_Progress = 0x03, + Q931_Setup = 0x05, + Q931_SetupAck = 0x0D, + Q931_Resume = 0x26, + Q931_ResumeAck = 0x2E, + Q931_ResumeReject = 0x22, + Q931_Suspend = 0x25, + Q931_SuspendAck = 0x2D, + Q931_SuspendReject = 0x21, + Q931_UserInformation = 0x20, + Q931_Disconnect = 0x45, + Q931_Release = 0x4D, + Q931_ReleaseComplete = 0x5A, + Q931_Restart = 0x46, + Q931_RestartAck = 0x4E, + Q931_Segment = 0x60, + Q931_CongestionCtrl = 0x79, + Q931_Information = 0x7B, + Q931_Notify = 0x6E, + Q931_Status = 0x7D, + Q931_StatusEnquiry = 0x75, + Q931_Facility = 0x62 + } MessageType; + H323_UserInformation UUIE; +} Q931; + +/***************************************************************************** + * Decode Functions Return Codes + ****************************************************************************/ + +#define H323_ERROR_NONE 0 /* Decoded successfully */ +#define H323_ERROR_STOP 1 /* Decoding stopped, not really an error */ +#define H323_ERROR_BOUND -1 +#define H323_ERROR_RANGE -2 + + +/***************************************************************************** + * Decode Functions + ****************************************************************************/ + +int DecodeRasMessage(unsigned char *buf, size_t sz, RasMessage * ras); +int DecodeQ931(unsigned char *buf, size_t sz, Q931 * q931); +int DecodeMultimediaSystemControlMessage(unsigned char *buf, size_t sz, + MultimediaSystemControlMessage * + mscm); + +#endif diff --git a/include/linux/netfilter/nf_conntrack_h323_types.h b/include/linux/netfilter/nf_conntrack_h323_types.h new file mode 100644 index 00000000000..38d74d5c970 --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_h323_types.h @@ -0,0 +1,951 @@ +/* Generated by Jing Min Zhao's ASN.1 parser, Apr 20 2006 + * + * Copyright (c) 2006 Jing Min Zhao + * + * This source code is licensed under General Public License version 2. + */ + +typedef struct TransportAddress_ipAddress { /* SEQUENCE */ + int options; /* No use */ + unsigned ip; +} TransportAddress_ipAddress; + +typedef struct TransportAddress_ip6Address { /* SEQUENCE */ + int options; /* No use */ + unsigned ip6; +} TransportAddress_ip6Address; + +typedef struct TransportAddress { /* CHOICE */ + enum { + eTransportAddress_ipAddress, + eTransportAddress_ipSourceRoute, + eTransportAddress_ipxAddress, + eTransportAddress_ip6Address, + eTransportAddress_netBios, + eTransportAddress_nsap, + eTransportAddress_nonStandardAddress, + } choice; + union { + TransportAddress_ipAddress ipAddress; + TransportAddress_ip6Address ip6Address; + }; +} TransportAddress; + +typedef struct DataProtocolCapability { /* CHOICE */ + enum { + eDataProtocolCapability_nonStandard, + eDataProtocolCapability_v14buffered, + eDataProtocolCapability_v42lapm, + eDataProtocolCapability_hdlcFrameTunnelling, + eDataProtocolCapability_h310SeparateVCStack, + eDataProtocolCapability_h310SingleVCStack, + eDataProtocolCapability_transparent, + eDataProtocolCapability_segmentationAndReassembly, + eDataProtocolCapability_hdlcFrameTunnelingwSAR, + eDataProtocolCapability_v120, + eDataProtocolCapability_separateLANStack, + eDataProtocolCapability_v76wCompression, + eDataProtocolCapability_tcp, + eDataProtocolCapability_udp, + } choice; +} DataProtocolCapability; + +typedef struct DataApplicationCapability_application { /* CHOICE */ + enum { + eDataApplicationCapability_application_nonStandard, + eDataApplicationCapability_application_t120, + eDataApplicationCapability_application_dsm_cc, + eDataApplicationCapability_application_userData, + eDataApplicationCapability_application_t84, + eDataApplicationCapability_application_t434, + eDataApplicationCapability_application_h224, + eDataApplicationCapability_application_nlpid, + eDataApplicationCapability_application_dsvdControl, + eDataApplicationCapability_application_h222DataPartitioning, + eDataApplicationCapability_application_t30fax, + eDataApplicationCapability_application_t140, + eDataApplicationCapability_application_t38fax, + eDataApplicationCapability_application_genericDataCapability, + } choice; + union { + DataProtocolCapability t120; + }; +} DataApplicationCapability_application; + +typedef struct DataApplicationCapability { /* SEQUENCE */ + int options; /* No use */ + DataApplicationCapability_application application; +} DataApplicationCapability; + +typedef struct DataType { /* CHOICE */ + enum { + eDataType_nonStandard, + eDataType_nullData, + eDataType_videoData, + eDataType_audioData, + eDataType_data, + eDataType_encryptionData, + eDataType_h235Control, + eDataType_h235Media, + eDataType_multiplexedStream, + } choice; + union { + DataApplicationCapability data; + }; +} DataType; + +typedef struct UnicastAddress_iPAddress { /* SEQUENCE */ + int options; /* No use */ + unsigned network; +} UnicastAddress_iPAddress; + +typedef struct UnicastAddress_iP6Address { /* SEQUENCE */ + int options; /* No use */ + unsigned network; +} UnicastAddress_iP6Address; + +typedef struct UnicastAddress { /* CHOICE */ + enum { + eUnicastAddress_iPAddress, + eUnicastAddress_iPXAddress, + eUnicastAddress_iP6Address, + eUnicastAddress_netBios, + eUnicastAddress_iPSourceRouteAddress, + eUnicastAddress_nsap, + eUnicastAddress_nonStandardAddress, + } choice; + union { + UnicastAddress_iPAddress iPAddress; + UnicastAddress_iP6Address iP6Address; + }; +} UnicastAddress; + +typedef struct H245_TransportAddress { /* CHOICE */ + enum { + eH245_TransportAddress_unicastAddress, + eH245_TransportAddress_multicastAddress, + } choice; + union { + UnicastAddress unicastAddress; + }; +} H245_TransportAddress; + +typedef struct H2250LogicalChannelParameters { /* SEQUENCE */ + enum { + eH2250LogicalChannelParameters_nonStandard = (1 << 31), + eH2250LogicalChannelParameters_associatedSessionID = + (1 << 30), + eH2250LogicalChannelParameters_mediaChannel = (1 << 29), + eH2250LogicalChannelParameters_mediaGuaranteedDelivery = + (1 << 28), + eH2250LogicalChannelParameters_mediaControlChannel = + (1 << 27), + eH2250LogicalChannelParameters_mediaControlGuaranteedDelivery + = (1 << 26), + eH2250LogicalChannelParameters_silenceSuppression = (1 << 25), + eH2250LogicalChannelParameters_destination = (1 << 24), + eH2250LogicalChannelParameters_dynamicRTPPayloadType = + (1 << 23), + eH2250LogicalChannelParameters_mediaPacketization = (1 << 22), + eH2250LogicalChannelParameters_transportCapability = + (1 << 21), + eH2250LogicalChannelParameters_redundancyEncoding = (1 << 20), + eH2250LogicalChannelParameters_source = (1 << 19), + } options; + H245_TransportAddress mediaChannel; + H245_TransportAddress mediaControlChannel; +} H2250LogicalChannelParameters; + +typedef struct OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters { /* CHOICE */ + enum { + eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_h222LogicalChannelParameters, + eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_h223LogicalChannelParameters, + eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_v76LogicalChannelParameters, + eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_h2250LogicalChannelParameters, + eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_none, + } choice; + union { + H2250LogicalChannelParameters h2250LogicalChannelParameters; + }; +} OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters; + +typedef struct OpenLogicalChannel_forwardLogicalChannelParameters { /* SEQUENCE */ + enum { + eOpenLogicalChannel_forwardLogicalChannelParameters_portNumber + = (1 << 31), + eOpenLogicalChannel_forwardLogicalChannelParameters_forwardLogicalChannelDependency + = (1 << 30), + eOpenLogicalChannel_forwardLogicalChannelParameters_replacementFor + = (1 << 29), + } options; + DataType dataType; + OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters + multiplexParameters; +} OpenLogicalChannel_forwardLogicalChannelParameters; + +typedef struct OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters { /* CHOICE */ + enum { + eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters_h223LogicalChannelParameters, + eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters_v76LogicalChannelParameters, + eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters_h2250LogicalChannelParameters, + } choice; + union { + H2250LogicalChannelParameters h2250LogicalChannelParameters; + }; +} OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters; + +typedef struct OpenLogicalChannel_reverseLogicalChannelParameters { /* SEQUENCE */ + enum { + eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters + = (1 << 31), + eOpenLogicalChannel_reverseLogicalChannelParameters_reverseLogicalChannelDependency + = (1 << 30), + eOpenLogicalChannel_reverseLogicalChannelParameters_replacementFor + = (1 << 29), + } options; + OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters + multiplexParameters; +} OpenLogicalChannel_reverseLogicalChannelParameters; + +typedef struct NetworkAccessParameters_networkAddress { /* CHOICE */ + enum { + eNetworkAccessParameters_networkAddress_q2931Address, + eNetworkAccessParameters_networkAddress_e164Address, + eNetworkAccessParameters_networkAddress_localAreaAddress, + } choice; + union { + H245_TransportAddress localAreaAddress; + }; +} NetworkAccessParameters_networkAddress; + +typedef struct NetworkAccessParameters { /* SEQUENCE */ + enum { + eNetworkAccessParameters_distribution = (1 << 31), + eNetworkAccessParameters_externalReference = (1 << 30), + eNetworkAccessParameters_t120SetupProcedure = (1 << 29), + } options; + NetworkAccessParameters_networkAddress networkAddress; +} NetworkAccessParameters; + +typedef struct OpenLogicalChannel { /* SEQUENCE */ + enum { + eOpenLogicalChannel_reverseLogicalChannelParameters = + (1 << 31), + eOpenLogicalChannel_separateStack = (1 << 30), + eOpenLogicalChannel_encryptionSync = (1 << 29), + } options; + OpenLogicalChannel_forwardLogicalChannelParameters + forwardLogicalChannelParameters; + OpenLogicalChannel_reverseLogicalChannelParameters + reverseLogicalChannelParameters; + NetworkAccessParameters separateStack; +} OpenLogicalChannel; + +typedef struct Setup_UUIE_fastStart { /* SEQUENCE OF */ + int count; + OpenLogicalChannel item[30]; +} Setup_UUIE_fastStart; + +typedef struct Setup_UUIE { /* SEQUENCE */ + enum { + eSetup_UUIE_h245Address = (1 << 31), + eSetup_UUIE_sourceAddress = (1 << 30), + eSetup_UUIE_destinationAddress = (1 << 29), + eSetup_UUIE_destCallSignalAddress = (1 << 28), + eSetup_UUIE_destExtraCallInfo = (1 << 27), + eSetup_UUIE_destExtraCRV = (1 << 26), + eSetup_UUIE_callServices = (1 << 25), + eSetup_UUIE_sourceCallSignalAddress = (1 << 24), + eSetup_UUIE_remoteExtensionAddress = (1 << 23), + eSetup_UUIE_callIdentifier = (1 << 22), + eSetup_UUIE_h245SecurityCapability = (1 << 21), + eSetup_UUIE_tokens = (1 << 20), + eSetup_UUIE_cryptoTokens = (1 << 19), + eSetup_UUIE_fastStart = (1 << 18), + eSetup_UUIE_mediaWaitForConnect = (1 << 17), + eSetup_UUIE_canOverlapSend = (1 << 16), + eSetup_UUIE_endpointIdentifier = (1 << 15), + eSetup_UUIE_multipleCalls = (1 << 14), + eSetup_UUIE_maintainConnection = (1 << 13), + eSetup_UUIE_connectionParameters = (1 << 12), + eSetup_UUIE_language = (1 << 11), + eSetup_UUIE_presentationIndicator = (1 << 10), + eSetup_UUIE_screeningIndicator = (1 << 9), + eSetup_UUIE_serviceControl = (1 << 8), + eSetup_UUIE_symmetricOperationRequired = (1 << 7), + eSetup_UUIE_capacity = (1 << 6), + eSetup_UUIE_circuitInfo = (1 << 5), + eSetup_UUIE_desiredProtocols = (1 << 4), + eSetup_UUIE_neededFeatures = (1 << 3), + eSetup_UUIE_desiredFeatures = (1 << 2), + eSetup_UUIE_supportedFeatures = (1 << 1), + eSetup_UUIE_parallelH245Control = (1 << 0), + } options; + TransportAddress h245Address; + TransportAddress destCallSignalAddress; + TransportAddress sourceCallSignalAddress; + Setup_UUIE_fastStart fastStart; +} Setup_UUIE; + +typedef struct CallProceeding_UUIE_fastStart { /* SEQUENCE OF */ + int count; + OpenLogicalChannel item[30]; +} CallProceeding_UUIE_fastStart; + +typedef struct CallProceeding_UUIE { /* SEQUENCE */ + enum { + eCallProceeding_UUIE_h245Address = (1 << 31), + eCallProceeding_UUIE_callIdentifier = (1 << 30), + eCallProceeding_UUIE_h245SecurityMode = (1 << 29), + eCallProceeding_UUIE_tokens = (1 << 28), + eCallProceeding_UUIE_cryptoTokens = (1 << 27), + eCallProceeding_UUIE_fastStart = (1 << 26), + eCallProceeding_UUIE_multipleCalls = (1 << 25), + eCallProceeding_UUIE_maintainConnection = (1 << 24), + eCallProceeding_UUIE_fastConnectRefused = (1 << 23), + eCallProceeding_UUIE_featureSet = (1 << 22), + } options; + TransportAddress h245Address; + CallProceeding_UUIE_fastStart fastStart; +} CallProceeding_UUIE; + +typedef struct Connect_UUIE_fastStart { /* SEQUENCE OF */ + int count; + OpenLogicalChannel item[30]; +} Connect_UUIE_fastStart; + +typedef struct Connect_UUIE { /* SEQUENCE */ + enum { + eConnect_UUIE_h245Address = (1 << 31), + eConnect_UUIE_callIdentifier = (1 << 30), + eConnect_UUIE_h245SecurityMode = (1 << 29), + eConnect_UUIE_tokens = (1 << 28), + eConnect_UUIE_cryptoTokens = (1 << 27), + eConnect_UUIE_fastStart = (1 << 26), + eConnect_UUIE_multipleCalls = (1 << 25), + eConnect_UUIE_maintainConnection = (1 << 24), + eConnect_UUIE_language = (1 << 23), + eConnect_UUIE_connectedAddress = (1 << 22), + eConnect_UUIE_presentationIndicator = (1 << 21), + eConnect_UUIE_screeningIndicator = (1 << 20), + eConnect_UUIE_fastConnectRefused = (1 << 19), + eConnect_UUIE_serviceControl = (1 << 18), + eConnect_UUIE_capacity = (1 << 17), + eConnect_UUIE_featureSet = (1 << 16), + } options; + TransportAddress h245Address; + Connect_UUIE_fastStart fastStart; +} Connect_UUIE; + +typedef struct Alerting_UUIE_fastStart { /* SEQUENCE OF */ + int count; + OpenLogicalChannel item[30]; +} Alerting_UUIE_fastStart; + +typedef struct Alerting_UUIE { /* SEQUENCE */ + enum { + eAlerting_UUIE_h245Address = (1 << 31), + eAlerting_UUIE_callIdentifier = (1 << 30), + eAlerting_UUIE_h245SecurityMode = (1 << 29), + eAlerting_UUIE_tokens = (1 << 28), + eAlerting_UUIE_cryptoTokens = (1 << 27), + eAlerting_UUIE_fastStart = (1 << 26), + eAlerting_UUIE_multipleCalls = (1 << 25), + eAlerting_UUIE_maintainConnection = (1 << 24), + eAlerting_UUIE_alertingAddress = (1 << 23), + eAlerting_UUIE_presentationIndicator = (1 << 22), + eAlerting_UUIE_screeningIndicator = (1 << 21), + eAlerting_UUIE_fastConnectRefused = (1 << 20), + eAlerting_UUIE_serviceControl = (1 << 19), + eAlerting_UUIE_capacity = (1 << 18), + eAlerting_UUIE_featureSet = (1 << 17), + } options; + TransportAddress h245Address; + Alerting_UUIE_fastStart fastStart; +} Alerting_UUIE; + +typedef struct Information_UUIE_fastStart { /* SEQUENCE OF */ + int count; + OpenLogicalChannel item[30]; +} Information_UUIE_fastStart; + +typedef struct Information_UUIE { /* SEQUENCE */ + enum { + eInformation_UUIE_callIdentifier = (1 << 31), + eInformation_UUIE_tokens = (1 << 30), + eInformation_UUIE_cryptoTokens = (1 << 29), + eInformation_UUIE_fastStart = (1 << 28), + eInformation_UUIE_fastConnectRefused = (1 << 27), + eInformation_UUIE_circuitInfo = (1 << 26), + } options; + Information_UUIE_fastStart fastStart; +} Information_UUIE; + +typedef struct FacilityReason { /* CHOICE */ + enum { + eFacilityReason_routeCallToGatekeeper, + eFacilityReason_callForwarded, + eFacilityReason_routeCallToMC, + eFacilityReason_undefinedReason, + eFacilityReason_conferenceListChoice, + eFacilityReason_startH245, + eFacilityReason_noH245, + eFacilityReason_newTokens, + eFacilityReason_featureSetUpdate, + eFacilityReason_forwardedElements, + eFacilityReason_transportedInformation, + } choice; +} FacilityReason; + +typedef struct Facility_UUIE_fastStart { /* SEQUENCE OF */ + int count; + OpenLogicalChannel item[30]; +} Facility_UUIE_fastStart; + +typedef struct Facility_UUIE { /* SEQUENCE */ + enum { + eFacility_UUIE_alternativeAddress = (1 << 31), + eFacility_UUIE_alternativeAliasAddress = (1 << 30), + eFacility_UUIE_conferenceID = (1 << 29), + eFacility_UUIE_callIdentifier = (1 << 28), + eFacility_UUIE_destExtraCallInfo = (1 << 27), + eFacility_UUIE_remoteExtensionAddress = (1 << 26), + eFacility_UUIE_tokens = (1 << 25), + eFacility_UUIE_cryptoTokens = (1 << 24), + eFacility_UUIE_conferences = (1 << 23), + eFacility_UUIE_h245Address = (1 << 22), + eFacility_UUIE_fastStart = (1 << 21), + eFacility_UUIE_multipleCalls = (1 << 20), + eFacility_UUIE_maintainConnection = (1 << 19), + eFacility_UUIE_fastConnectRefused = (1 << 18), + eFacility_UUIE_serviceControl = (1 << 17), + eFacility_UUIE_circuitInfo = (1 << 16), + eFacility_UUIE_featureSet = (1 << 15), + eFacility_UUIE_destinationInfo = (1 << 14), + eFacility_UUIE_h245SecurityMode = (1 << 13), + } options; + TransportAddress alternativeAddress; + FacilityReason reason; + TransportAddress h245Address; + Facility_UUIE_fastStart fastStart; +} Facility_UUIE; + +typedef struct Progress_UUIE_fastStart { /* SEQUENCE OF */ + int count; + OpenLogicalChannel item[30]; +} Progress_UUIE_fastStart; + +typedef struct Progress_UUIE { /* SEQUENCE */ + enum { + eProgress_UUIE_h245Address = (1 << 31), + eProgress_UUIE_h245SecurityMode = (1 << 30), + eProgress_UUIE_tokens = (1 << 29), + eProgress_UUIE_cryptoTokens = (1 << 28), + eProgress_UUIE_fastStart = (1 << 27), + eProgress_UUIE_multipleCalls = (1 << 26), + eProgress_UUIE_maintainConnection = (1 << 25), + eProgress_UUIE_fastConnectRefused = (1 << 24), + } options; + TransportAddress h245Address; + Progress_UUIE_fastStart fastStart; +} Progress_UUIE; + +typedef struct H323_UU_PDU_h323_message_body { /* CHOICE */ + enum { + eH323_UU_PDU_h323_message_body_setup, + eH323_UU_PDU_h323_message_body_callProceeding, + eH323_UU_PDU_h323_message_body_connect, + eH323_UU_PDU_h323_message_body_alerting, + eH323_UU_PDU_h323_message_body_information, + eH323_UU_PDU_h323_message_body_releaseComplete, + eH323_UU_PDU_h323_message_body_facility, + eH323_UU_PDU_h323_message_body_progress, + eH323_UU_PDU_h323_message_body_empty, + eH323_UU_PDU_h323_message_body_status, + eH323_UU_PDU_h323_message_body_statusInquiry, + eH323_UU_PDU_h323_message_body_setupAcknowledge, + eH323_UU_PDU_h323_message_body_notify, + } choice; + union { + Setup_UUIE setup; + CallProceeding_UUIE callProceeding; + Connect_UUIE connect; + Alerting_UUIE alerting; + Information_UUIE information; + Facility_UUIE facility; + Progress_UUIE progress; + }; +} H323_UU_PDU_h323_message_body; + +typedef struct RequestMessage { /* CHOICE */ + enum { + eRequestMessage_nonStandard, + eRequestMessage_masterSlaveDetermination, + eRequestMessage_terminalCapabilitySet, + eRequestMessage_openLogicalChannel, + eRequestMessage_closeLogicalChannel, + eRequestMessage_requestChannelClose, + eRequestMessage_multiplexEntrySend, + eRequestMessage_requestMultiplexEntry, + eRequestMessage_requestMode, + eRequestMessage_roundTripDelayRequest, + eRequestMessage_maintenanceLoopRequest, + eRequestMessage_communicationModeRequest, + eRequestMessage_conferenceRequest, + eRequestMessage_multilinkRequest, + eRequestMessage_logicalChannelRateRequest, + } choice; + union { + OpenLogicalChannel openLogicalChannel; + }; +} RequestMessage; + +typedef struct OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters { /* CHOICE */ + enum { + eOpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters_h222LogicalChannelParameters, + eOpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters_h2250LogicalChannelParameters, + } choice; + union { + H2250LogicalChannelParameters h2250LogicalChannelParameters; + }; +} OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters; + +typedef struct OpenLogicalChannelAck_reverseLogicalChannelParameters { /* SEQUENCE */ + enum { + eOpenLogicalChannelAck_reverseLogicalChannelParameters_portNumber + = (1 << 31), + eOpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters + = (1 << 30), + eOpenLogicalChannelAck_reverseLogicalChannelParameters_replacementFor + = (1 << 29), + } options; + OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters + multiplexParameters; +} OpenLogicalChannelAck_reverseLogicalChannelParameters; + +typedef struct H2250LogicalChannelAckParameters { /* SEQUENCE */ + enum { + eH2250LogicalChannelAckParameters_nonStandard = (1 << 31), + eH2250LogicalChannelAckParameters_sessionID = (1 << 30), + eH2250LogicalChannelAckParameters_mediaChannel = (1 << 29), + eH2250LogicalChannelAckParameters_mediaControlChannel = + (1 << 28), + eH2250LogicalChannelAckParameters_dynamicRTPPayloadType = + (1 << 27), + eH2250LogicalChannelAckParameters_flowControlToZero = + (1 << 26), + eH2250LogicalChannelAckParameters_portNumber = (1 << 25), + } options; + H245_TransportAddress mediaChannel; + H245_TransportAddress mediaControlChannel; +} H2250LogicalChannelAckParameters; + +typedef struct OpenLogicalChannelAck_forwardMultiplexAckParameters { /* CHOICE */ + enum { + eOpenLogicalChannelAck_forwardMultiplexAckParameters_h2250LogicalChannelAckParameters, + } choice; + union { + H2250LogicalChannelAckParameters + h2250LogicalChannelAckParameters; + }; +} OpenLogicalChannelAck_forwardMultiplexAckParameters; + +typedef struct OpenLogicalChannelAck { /* SEQUENCE */ + enum { + eOpenLogicalChannelAck_reverseLogicalChannelParameters = + (1 << 31), + eOpenLogicalChannelAck_separateStack = (1 << 30), + eOpenLogicalChannelAck_forwardMultiplexAckParameters = + (1 << 29), + eOpenLogicalChannelAck_encryptionSync = (1 << 28), + } options; + OpenLogicalChannelAck_reverseLogicalChannelParameters + reverseLogicalChannelParameters; + OpenLogicalChannelAck_forwardMultiplexAckParameters + forwardMultiplexAckParameters; +} OpenLogicalChannelAck; + +typedef struct ResponseMessage { /* CHOICE */ + enum { + eResponseMessage_nonStandard, + eResponseMessage_masterSlaveDeterminationAck, + eResponseMessage_masterSlaveDeterminationReject, + eResponseMessage_terminalCapabilitySetAck, + eResponseMessage_terminalCapabilitySetReject, + eResponseMessage_openLogicalChannelAck, + eResponseMessage_openLogicalChannelReject, + eResponseMessage_closeLogicalChannelAck, + eResponseMessage_requestChannelCloseAck, + eResponseMessage_requestChannelCloseReject, + eResponseMessage_multiplexEntrySendAck, + eResponseMessage_multiplexEntrySendReject, + eResponseMessage_requestMultiplexEntryAck, + eResponseMessage_requestMultiplexEntryReject, + eResponseMessage_requestModeAck, + eResponseMessage_requestModeReject, + eResponseMessage_roundTripDelayResponse, + eResponseMessage_maintenanceLoopAck, + eResponseMessage_maintenanceLoopReject, + eResponseMessage_communicationModeResponse, + eResponseMessage_conferenceResponse, + eResponseMessage_multilinkResponse, + eResponseMessage_logicalChannelRateAcknowledge, + eResponseMessage_logicalChannelRateReject, + } choice; + union { + OpenLogicalChannelAck openLogicalChannelAck; + }; +} ResponseMessage; + +typedef struct MultimediaSystemControlMessage { /* CHOICE */ + enum { + eMultimediaSystemControlMessage_request, + eMultimediaSystemControlMessage_response, + eMultimediaSystemControlMessage_command, + eMultimediaSystemControlMessage_indication, + } choice; + union { + RequestMessage request; + ResponseMessage response; + }; +} MultimediaSystemControlMessage; + +typedef struct H323_UU_PDU_h245Control { /* SEQUENCE OF */ + int count; + MultimediaSystemControlMessage item[4]; +} H323_UU_PDU_h245Control; + +typedef struct H323_UU_PDU { /* SEQUENCE */ + enum { + eH323_UU_PDU_nonStandardData = (1 << 31), + eH323_UU_PDU_h4501SupplementaryService = (1 << 30), + eH323_UU_PDU_h245Tunneling = (1 << 29), + eH323_UU_PDU_h245Control = (1 << 28), + eH323_UU_PDU_nonStandardControl = (1 << 27), + eH323_UU_PDU_callLinkage = (1 << 26), + eH323_UU_PDU_tunnelledSignallingMessage = (1 << 25), + eH323_UU_PDU_provisionalRespToH245Tunneling = (1 << 24), + eH323_UU_PDU_stimulusControl = (1 << 23), + eH323_UU_PDU_genericData = (1 << 22), + } options; + H323_UU_PDU_h323_message_body h323_message_body; + H323_UU_PDU_h245Control h245Control; +} H323_UU_PDU; + +typedef struct H323_UserInformation { /* SEQUENCE */ + enum { + eH323_UserInformation_user_data = (1 << 31), + } options; + H323_UU_PDU h323_uu_pdu; +} H323_UserInformation; + +typedef struct GatekeeperRequest { /* SEQUENCE */ + enum { + eGatekeeperRequest_nonStandardData = (1 << 31), + eGatekeeperRequest_gatekeeperIdentifier = (1 << 30), + eGatekeeperRequest_callServices = (1 << 29), + eGatekeeperRequest_endpointAlias = (1 << 28), + eGatekeeperRequest_alternateEndpoints = (1 << 27), + eGatekeeperRequest_tokens = (1 << 26), + eGatekeeperRequest_cryptoTokens = (1 << 25), + eGatekeeperRequest_authenticationCapability = (1 << 24), + eGatekeeperRequest_algorithmOIDs = (1 << 23), + eGatekeeperRequest_integrity = (1 << 22), + eGatekeeperRequest_integrityCheckValue = (1 << 21), + eGatekeeperRequest_supportsAltGK = (1 << 20), + eGatekeeperRequest_featureSet = (1 << 19), + eGatekeeperRequest_genericData = (1 << 18), + } options; + TransportAddress rasAddress; +} GatekeeperRequest; + +typedef struct GatekeeperConfirm { /* SEQUENCE */ + enum { + eGatekeeperConfirm_nonStandardData = (1 << 31), + eGatekeeperConfirm_gatekeeperIdentifier = (1 << 30), + eGatekeeperConfirm_alternateGatekeeper = (1 << 29), + eGatekeeperConfirm_authenticationMode = (1 << 28), + eGatekeeperConfirm_tokens = (1 << 27), + eGatekeeperConfirm_cryptoTokens = (1 << 26), + eGatekeeperConfirm_algorithmOID = (1 << 25), + eGatekeeperConfirm_integrity = (1 << 24), + eGatekeeperConfirm_integrityCheckValue = (1 << 23), + eGatekeeperConfirm_featureSet = (1 << 22), + eGatekeeperConfirm_genericData = (1 << 21), + } options; + TransportAddress rasAddress; +} GatekeeperConfirm; + +typedef struct RegistrationRequest_callSignalAddress { /* SEQUENCE OF */ + int count; + TransportAddress item[10]; +} RegistrationRequest_callSignalAddress; + +typedef struct RegistrationRequest_rasAddress { /* SEQUENCE OF */ + int count; + TransportAddress item[10]; +} RegistrationRequest_rasAddress; + +typedef struct RegistrationRequest { /* SEQUENCE */ + enum { + eRegistrationRequest_nonStandardData = (1 << 31), + eRegistrationRequest_terminalAlias = (1 << 30), + eRegistrationRequest_gatekeeperIdentifier = (1 << 29), + eRegistrationRequest_alternateEndpoints = (1 << 28), + eRegistrationRequest_timeToLive = (1 << 27), + eRegistrationRequest_tokens = (1 << 26), + eRegistrationRequest_cryptoTokens = (1 << 25), + eRegistrationRequest_integrityCheckValue = (1 << 24), + eRegistrationRequest_keepAlive = (1 << 23), + eRegistrationRequest_endpointIdentifier = (1 << 22), + eRegistrationRequest_willSupplyUUIEs = (1 << 21), + eRegistrationRequest_maintainConnection = (1 << 20), + eRegistrationRequest_alternateTransportAddresses = (1 << 19), + eRegistrationRequest_additiveRegistration = (1 << 18), + eRegistrationRequest_terminalAliasPattern = (1 << 17), + eRegistrationRequest_supportsAltGK = (1 << 16), + eRegistrationRequest_usageReportingCapability = (1 << 15), + eRegistrationRequest_multipleCalls = (1 << 14), + eRegistrationRequest_supportedH248Packages = (1 << 13), + eRegistrationRequest_callCreditCapability = (1 << 12), + eRegistrationRequest_capacityReportingCapability = (1 << 11), + eRegistrationRequest_capacity = (1 << 10), + eRegistrationRequest_featureSet = (1 << 9), + eRegistrationRequest_genericData = (1 << 8), + } options; + RegistrationRequest_callSignalAddress callSignalAddress; + RegistrationRequest_rasAddress rasAddress; + unsigned timeToLive; +} RegistrationRequest; + +typedef struct RegistrationConfirm_callSignalAddress { /* SEQUENCE OF */ + int count; + TransportAddress item[10]; +} RegistrationConfirm_callSignalAddress; + +typedef struct RegistrationConfirm { /* SEQUENCE */ + enum { + eRegistrationConfirm_nonStandardData = (1 << 31), + eRegistrationConfirm_terminalAlias = (1 << 30), + eRegistrationConfirm_gatekeeperIdentifier = (1 << 29), + eRegistrationConfirm_alternateGatekeeper = (1 << 28), + eRegistrationConfirm_timeToLive = (1 << 27), + eRegistrationConfirm_tokens = (1 << 26), + eRegistrationConfirm_cryptoTokens = (1 << 25), + eRegistrationConfirm_integrityCheckValue = (1 << 24), + eRegistrationConfirm_willRespondToIRR = (1 << 23), + eRegistrationConfirm_preGrantedARQ = (1 << 22), + eRegistrationConfirm_maintainConnection = (1 << 21), + eRegistrationConfirm_serviceControl = (1 << 20), + eRegistrationConfirm_supportsAdditiveRegistration = (1 << 19), + eRegistrationConfirm_terminalAliasPattern = (1 << 18), + eRegistrationConfirm_supportedPrefixes = (1 << 17), + eRegistrationConfirm_usageSpec = (1 << 16), + eRegistrationConfirm_featureServerAlias = (1 << 15), + eRegistrationConfirm_capacityReportingSpec = (1 << 14), + eRegistrationConfirm_featureSet = (1 << 13), + eRegistrationConfirm_genericData = (1 << 12), + } options; + RegistrationConfirm_callSignalAddress callSignalAddress; + unsigned timeToLive; +} RegistrationConfirm; + +typedef struct UnregistrationRequest_callSignalAddress { /* SEQUENCE OF */ + int count; + TransportAddress item[10]; +} UnregistrationRequest_callSignalAddress; + +typedef struct UnregistrationRequest { /* SEQUENCE */ + enum { + eUnregistrationRequest_endpointAlias = (1 << 31), + eUnregistrationRequest_nonStandardData = (1 << 30), + eUnregistrationRequest_endpointIdentifier = (1 << 29), + eUnregistrationRequest_alternateEndpoints = (1 << 28), + eUnregistrationRequest_gatekeeperIdentifier = (1 << 27), + eUnregistrationRequest_tokens = (1 << 26), + eUnregistrationRequest_cryptoTokens = (1 << 25), + eUnregistrationRequest_integrityCheckValue = (1 << 24), + eUnregistrationRequest_reason = (1 << 23), + eUnregistrationRequest_endpointAliasPattern = (1 << 22), + eUnregistrationRequest_supportedPrefixes = (1 << 21), + eUnregistrationRequest_alternateGatekeeper = (1 << 20), + eUnregistrationRequest_genericData = (1 << 19), + } options; + UnregistrationRequest_callSignalAddress callSignalAddress; +} UnregistrationRequest; + +typedef struct AdmissionRequest { /* SEQUENCE */ + enum { + eAdmissionRequest_callModel = (1 << 31), + eAdmissionRequest_destinationInfo = (1 << 30), + eAdmissionRequest_destCallSignalAddress = (1 << 29), + eAdmissionRequest_destExtraCallInfo = (1 << 28), + eAdmissionRequest_srcCallSignalAddress = (1 << 27), + eAdmissionRequest_nonStandardData = (1 << 26), + eAdmissionRequest_callServices = (1 << 25), + eAdmissionRequest_canMapAlias = (1 << 24), + eAdmissionRequest_callIdentifier = (1 << 23), + eAdmissionRequest_srcAlternatives = (1 << 22), + eAdmissionRequest_destAlternatives = (1 << 21), + eAdmissionRequest_gatekeeperIdentifier = (1 << 20), + eAdmissionRequest_tokens = (1 << 19), + eAdmissionRequest_cryptoTokens = (1 << 18), + eAdmissionRequest_integrityCheckValue = (1 << 17), + eAdmissionRequest_transportQOS = (1 << 16), + eAdmissionRequest_willSupplyUUIEs = (1 << 15), + eAdmissionRequest_callLinkage = (1 << 14), + eAdmissionRequest_gatewayDataRate = (1 << 13), + eAdmissionRequest_capacity = (1 << 12), + eAdmissionRequest_circuitInfo = (1 << 11), + eAdmissionRequest_desiredProtocols = (1 << 10), + eAdmissionRequest_desiredTunnelledProtocol = (1 << 9), + eAdmissionRequest_featureSet = (1 << 8), + eAdmissionRequest_genericData = (1 << 7), + } options; + TransportAddress destCallSignalAddress; + TransportAddress srcCallSignalAddress; +} AdmissionRequest; + +typedef struct AdmissionConfirm { /* SEQUENCE */ + enum { + eAdmissionConfirm_irrFrequency = (1 << 31), + eAdmissionConfirm_nonStandardData = (1 << 30), + eAdmissionConfirm_destinationInfo = (1 << 29), + eAdmissionConfirm_destExtraCallInfo = (1 << 28), + eAdmissionConfirm_destinationType = (1 << 27), + eAdmissionConfirm_remoteExtensionAddress = (1 << 26), + eAdmissionConfirm_alternateEndpoints = (1 << 25), + eAdmissionConfirm_tokens = (1 << 24), + eAdmissionConfirm_cryptoTokens = (1 << 23), + eAdmissionConfirm_integrityCheckValue = (1 << 22), + eAdmissionConfirm_transportQOS = (1 << 21), + eAdmissionConfirm_willRespondToIRR = (1 << 20), + eAdmissionConfirm_uuiesRequested = (1 << 19), + eAdmissionConfirm_language = (1 << 18), + eAdmissionConfirm_alternateTransportAddresses = (1 << 17), + eAdmissionConfirm_useSpecifiedTransport = (1 << 16), + eAdmissionConfirm_circuitInfo = (1 << 15), + eAdmissionConfirm_usageSpec = (1 << 14), + eAdmissionConfirm_supportedProtocols = (1 << 13), + eAdmissionConfirm_serviceControl = (1 << 12), + eAdmissionConfirm_multipleCalls = (1 << 11), + eAdmissionConfirm_featureSet = (1 << 10), + eAdmissionConfirm_genericData = (1 << 9), + } options; + TransportAddress destCallSignalAddress; +} AdmissionConfirm; + +typedef struct LocationRequest { /* SEQUENCE */ + enum { + eLocationRequest_endpointIdentifier = (1 << 31), + eLocationRequest_nonStandardData = (1 << 30), + eLocationRequest_sourceInfo = (1 << 29), + eLocationRequest_canMapAlias = (1 << 28), + eLocationRequest_gatekeeperIdentifier = (1 << 27), + eLocationRequest_tokens = (1 << 26), + eLocationRequest_cryptoTokens = (1 << 25), + eLocationRequest_integrityCheckValue = (1 << 24), + eLocationRequest_desiredProtocols = (1 << 23), + eLocationRequest_desiredTunnelledProtocol = (1 << 22), + eLocationRequest_featureSet = (1 << 21), + eLocationRequest_genericData = (1 << 20), + eLocationRequest_hopCount = (1 << 19), + eLocationRequest_circuitInfo = (1 << 18), + } options; + TransportAddress replyAddress; +} LocationRequest; + +typedef struct LocationConfirm { /* SEQUENCE */ + enum { + eLocationConfirm_nonStandardData = (1 << 31), + eLocationConfirm_destinationInfo = (1 << 30), + eLocationConfirm_destExtraCallInfo = (1 << 29), + eLocationConfirm_destinationType = (1 << 28), + eLocationConfirm_remoteExtensionAddress = (1 << 27), + eLocationConfirm_alternateEndpoints = (1 << 26), + eLocationConfirm_tokens = (1 << 25), + eLocationConfirm_cryptoTokens = (1 << 24), + eLocationConfirm_integrityCheckValue = (1 << 23), + eLocationConfirm_alternateTransportAddresses = (1 << 22), + eLocationConfirm_supportedProtocols = (1 << 21), + eLocationConfirm_multipleCalls = (1 << 20), + eLocationConfirm_featureSet = (1 << 19), + eLocationConfirm_genericData = (1 << 18), + eLocationConfirm_circuitInfo = (1 << 17), + eLocationConfirm_serviceControl = (1 << 16), + } options; + TransportAddress callSignalAddress; + TransportAddress rasAddress; +} LocationConfirm; + +typedef struct InfoRequestResponse_callSignalAddress { /* SEQUENCE OF */ + int count; + TransportAddress item[10]; +} InfoRequestResponse_callSignalAddress; + +typedef struct InfoRequestResponse { /* SEQUENCE */ + enum { + eInfoRequestResponse_nonStandardData = (1 << 31), + eInfoRequestResponse_endpointAlias = (1 << 30), + eInfoRequestResponse_perCallInfo = (1 << 29), + eInfoRequestResponse_tokens = (1 << 28), + eInfoRequestResponse_cryptoTokens = (1 << 27), + eInfoRequestResponse_integrityCheckValue = (1 << 26), + eInfoRequestResponse_needResponse = (1 << 25), + eInfoRequestResponse_capacity = (1 << 24), + eInfoRequestResponse_irrStatus = (1 << 23), + eInfoRequestResponse_unsolicited = (1 << 22), + eInfoRequestResponse_genericData = (1 << 21), + } options; + TransportAddress rasAddress; + InfoRequestResponse_callSignalAddress callSignalAddress; +} InfoRequestResponse; + +typedef struct RasMessage { /* CHOICE */ + enum { + eRasMessage_gatekeeperRequest, + eRasMessage_gatekeeperConfirm, + eRasMessage_gatekeeperReject, + eRasMessage_registrationRequest, + eRasMessage_registrationConfirm, + eRasMessage_registrationReject, + eRasMessage_unregistrationRequest, + eRasMessage_unregistrationConfirm, + eRasMessage_unregistrationReject, + eRasMessage_admissionRequest, + eRasMessage_admissionConfirm, + eRasMessage_admissionReject, + eRasMessage_bandwidthRequest, + eRasMessage_bandwidthConfirm, + eRasMessage_bandwidthReject, + eRasMessage_disengageRequest, + eRasMessage_disengageConfirm, + eRasMessage_disengageReject, + eRasMessage_locationRequest, + eRasMessage_locationConfirm, + eRasMessage_locationReject, + eRasMessage_infoRequest, + eRasMessage_infoRequestResponse, + eRasMessage_nonStandardMessage, + eRasMessage_unknownMessageResponse, + eRasMessage_requestInProgress, + eRasMessage_resourcesAvailableIndicate, + eRasMessage_resourcesAvailableConfirm, + eRasMessage_infoRequestAck, + eRasMessage_infoRequestNak, + eRasMessage_serviceControlIndication, + eRasMessage_serviceControlResponse, + } choice; + union { + GatekeeperRequest gatekeeperRequest; + GatekeeperConfirm gatekeeperConfirm; + RegistrationRequest registrationRequest; + RegistrationConfirm registrationConfirm; + UnregistrationRequest unregistrationRequest; + AdmissionRequest admissionRequest; + AdmissionConfirm admissionConfirm; + LocationRequest locationRequest; + LocationConfirm locationConfirm; + InfoRequestResponse infoRequestResponse; + }; +} RasMessage; diff --git a/include/linux/netfilter_ipv4/Kbuild b/include/linux/netfilter_ipv4/Kbuild index 591c1a809c0..180337801a8 100644 --- a/include/linux/netfilter_ipv4/Kbuild +++ b/include/linux/netfilter_ipv4/Kbuild @@ -1,6 +1,4 @@ header-y += ip_conntrack_helper.h -header-y += ip_conntrack_helper_h323_asn1.h -header-y += ip_conntrack_helper_h323_types.h header-y += ip_conntrack_protocol.h header-y += ip_conntrack_sctp.h header-y += ip_conntrack_tcp.h diff --git a/include/linux/netfilter_ipv4/ip_conntrack_h323.h b/include/linux/netfilter_ipv4/ip_conntrack_h323.h index 943cc6a4871..18f769818f4 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_h323.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_h323.h @@ -3,7 +3,7 @@ #ifdef __KERNEL__ -#include +#include #define RAS_PORT 1719 #define Q931_PORT 1720 diff --git a/include/linux/netfilter_ipv4/ip_conntrack_helper_h323_asn1.h b/include/linux/netfilter_ipv4/ip_conntrack_helper_h323_asn1.h deleted file mode 100644 index c6e9a0b6d30..00000000000 --- a/include/linux/netfilter_ipv4/ip_conntrack_helper_h323_asn1.h +++ /dev/null @@ -1,98 +0,0 @@ -/**************************************************************************** - * ip_conntrack_helper_h323_asn1.h - BER and PER decoding library for H.323 - * conntrack/NAT module. - * - * Copyright (c) 2006 by Jing Min Zhao - * - * This source code is licensed under General Public License version 2. - * - * - * This library is based on H.225 version 4, H.235 version 2 and H.245 - * version 7. It is extremely optimized to decode only the absolutely - * necessary objects in a signal for Linux kernel NAT module use, so don't - * expect it to be a full ASN.1 library. - * - * Features: - * - * 1. Small. The total size of code plus data is less than 20 KB (IA32). - * 2. Fast. Decoding Netmeeting's Setup signal 1 million times on a PIII 866 - * takes only 3.9 seconds. - * 3. No memory allocation. It uses a static object. No need to initialize or - * cleanup. - * 4. Thread safe. - * 5. Support embedded architectures that has no misaligned memory access - * support. - * - * Limitations: - * - * 1. At most 30 faststart entries. Actually this is limited by ethernet's MTU. - * If a Setup signal contains more than 30 faststart, the packet size will - * very likely exceed the MTU size, then the TPKT will be fragmented. I - * don't know how to handle this in a Netfilter module. Anybody can help? - * Although I think 30 is enough for most of the cases. - * 2. IPv4 addresses only. - * - ****************************************************************************/ - -#ifndef _IP_CONNTRACK_HELPER_H323_ASN1_H_ -#define _IP_CONNTRACK_HELPER_H323_ASN1_H_ - -/***************************************************************************** - * H.323 Types - ****************************************************************************/ -#include "ip_conntrack_helper_h323_types.h" - -typedef struct { - enum { - Q931_NationalEscape = 0x00, - Q931_Alerting = 0x01, - Q931_CallProceeding = 0x02, - Q931_Connect = 0x07, - Q931_ConnectAck = 0x0F, - Q931_Progress = 0x03, - Q931_Setup = 0x05, - Q931_SetupAck = 0x0D, - Q931_Resume = 0x26, - Q931_ResumeAck = 0x2E, - Q931_ResumeReject = 0x22, - Q931_Suspend = 0x25, - Q931_SuspendAck = 0x2D, - Q931_SuspendReject = 0x21, - Q931_UserInformation = 0x20, - Q931_Disconnect = 0x45, - Q931_Release = 0x4D, - Q931_ReleaseComplete = 0x5A, - Q931_Restart = 0x46, - Q931_RestartAck = 0x4E, - Q931_Segment = 0x60, - Q931_CongestionCtrl = 0x79, - Q931_Information = 0x7B, - Q931_Notify = 0x6E, - Q931_Status = 0x7D, - Q931_StatusEnquiry = 0x75, - Q931_Facility = 0x62 - } MessageType; - H323_UserInformation UUIE; -} Q931; - -/***************************************************************************** - * Decode Functions Return Codes - ****************************************************************************/ - -#define H323_ERROR_NONE 0 /* Decoded successfully */ -#define H323_ERROR_STOP 1 /* Decoding stopped, not really an error */ -#define H323_ERROR_BOUND -1 -#define H323_ERROR_RANGE -2 - - -/***************************************************************************** - * Decode Functions - ****************************************************************************/ - -int DecodeRasMessage(unsigned char *buf, size_t sz, RasMessage * ras); -int DecodeQ931(unsigned char *buf, size_t sz, Q931 * q931); -int DecodeMultimediaSystemControlMessage(unsigned char *buf, size_t sz, - MultimediaSystemControlMessage * - mscm); - -#endif diff --git a/include/linux/netfilter_ipv4/ip_conntrack_helper_h323_types.h b/include/linux/netfilter_ipv4/ip_conntrack_helper_h323_types.h deleted file mode 100644 index 3d4a773799f..00000000000 --- a/include/linux/netfilter_ipv4/ip_conntrack_helper_h323_types.h +++ /dev/null @@ -1,939 +0,0 @@ -/* Generated by Jing Min Zhao's ASN.1 parser, Apr 20 2006 - * - * Copyright (c) 2006 Jing Min Zhao - * - * This source code is licensed under General Public License version 2. - */ - -typedef struct TransportAddress_ipAddress { /* SEQUENCE */ - int options; /* No use */ - unsigned ip; -} TransportAddress_ipAddress; - -typedef struct TransportAddress { /* CHOICE */ - enum { - eTransportAddress_ipAddress, - eTransportAddress_ipSourceRoute, - eTransportAddress_ipxAddress, - eTransportAddress_ip6Address, - eTransportAddress_netBios, - eTransportAddress_nsap, - eTransportAddress_nonStandardAddress, - } choice; - union { - TransportAddress_ipAddress ipAddress; - }; -} TransportAddress; - -typedef struct DataProtocolCapability { /* CHOICE */ - enum { - eDataProtocolCapability_nonStandard, - eDataProtocolCapability_v14buffered, - eDataProtocolCapability_v42lapm, - eDataProtocolCapability_hdlcFrameTunnelling, - eDataProtocolCapability_h310SeparateVCStack, - eDataProtocolCapability_h310SingleVCStack, - eDataProtocolCapability_transparent, - eDataProtocolCapability_segmentationAndReassembly, - eDataProtocolCapability_hdlcFrameTunnelingwSAR, - eDataProtocolCapability_v120, - eDataProtocolCapability_separateLANStack, - eDataProtocolCapability_v76wCompression, - eDataProtocolCapability_tcp, - eDataProtocolCapability_udp, - } choice; -} DataProtocolCapability; - -typedef struct DataApplicationCapability_application { /* CHOICE */ - enum { - eDataApplicationCapability_application_nonStandard, - eDataApplicationCapability_application_t120, - eDataApplicationCapability_application_dsm_cc, - eDataApplicationCapability_application_userData, - eDataApplicationCapability_application_t84, - eDataApplicationCapability_application_t434, - eDataApplicationCapability_application_h224, - eDataApplicationCapability_application_nlpid, - eDataApplicationCapability_application_dsvdControl, - eDataApplicationCapability_application_h222DataPartitioning, - eDataApplicationCapability_application_t30fax, - eDataApplicationCapability_application_t140, - eDataApplicationCapability_application_t38fax, - eDataApplicationCapability_application_genericDataCapability, - } choice; - union { - DataProtocolCapability t120; - }; -} DataApplicationCapability_application; - -typedef struct DataApplicationCapability { /* SEQUENCE */ - int options; /* No use */ - DataApplicationCapability_application application; -} DataApplicationCapability; - -typedef struct DataType { /* CHOICE */ - enum { - eDataType_nonStandard, - eDataType_nullData, - eDataType_videoData, - eDataType_audioData, - eDataType_data, - eDataType_encryptionData, - eDataType_h235Control, - eDataType_h235Media, - eDataType_multiplexedStream, - } choice; - union { - DataApplicationCapability data; - }; -} DataType; - -typedef struct UnicastAddress_iPAddress { /* SEQUENCE */ - int options; /* No use */ - unsigned network; -} UnicastAddress_iPAddress; - -typedef struct UnicastAddress { /* CHOICE */ - enum { - eUnicastAddress_iPAddress, - eUnicastAddress_iPXAddress, - eUnicastAddress_iP6Address, - eUnicastAddress_netBios, - eUnicastAddress_iPSourceRouteAddress, - eUnicastAddress_nsap, - eUnicastAddress_nonStandardAddress, - } choice; - union { - UnicastAddress_iPAddress iPAddress; - }; -} UnicastAddress; - -typedef struct H245_TransportAddress { /* CHOICE */ - enum { - eH245_TransportAddress_unicastAddress, - eH245_TransportAddress_multicastAddress, - } choice; - union { - UnicastAddress unicastAddress; - }; -} H245_TransportAddress; - -typedef struct H2250LogicalChannelParameters { /* SEQUENCE */ - enum { - eH2250LogicalChannelParameters_nonStandard = (1 << 31), - eH2250LogicalChannelParameters_associatedSessionID = - (1 << 30), - eH2250LogicalChannelParameters_mediaChannel = (1 << 29), - eH2250LogicalChannelParameters_mediaGuaranteedDelivery = - (1 << 28), - eH2250LogicalChannelParameters_mediaControlChannel = - (1 << 27), - eH2250LogicalChannelParameters_mediaControlGuaranteedDelivery - = (1 << 26), - eH2250LogicalChannelParameters_silenceSuppression = (1 << 25), - eH2250LogicalChannelParameters_destination = (1 << 24), - eH2250LogicalChannelParameters_dynamicRTPPayloadType = - (1 << 23), - eH2250LogicalChannelParameters_mediaPacketization = (1 << 22), - eH2250LogicalChannelParameters_transportCapability = - (1 << 21), - eH2250LogicalChannelParameters_redundancyEncoding = (1 << 20), - eH2250LogicalChannelParameters_source = (1 << 19), - } options; - H245_TransportAddress mediaChannel; - H245_TransportAddress mediaControlChannel; -} H2250LogicalChannelParameters; - -typedef struct OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters { /* CHOICE */ - enum { - eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_h222LogicalChannelParameters, - eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_h223LogicalChannelParameters, - eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_v76LogicalChannelParameters, - eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_h2250LogicalChannelParameters, - eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_none, - } choice; - union { - H2250LogicalChannelParameters h2250LogicalChannelParameters; - }; -} OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters; - -typedef struct OpenLogicalChannel_forwardLogicalChannelParameters { /* SEQUENCE */ - enum { - eOpenLogicalChannel_forwardLogicalChannelParameters_portNumber - = (1 << 31), - eOpenLogicalChannel_forwardLogicalChannelParameters_forwardLogicalChannelDependency - = (1 << 30), - eOpenLogicalChannel_forwardLogicalChannelParameters_replacementFor - = (1 << 29), - } options; - DataType dataType; - OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters - multiplexParameters; -} OpenLogicalChannel_forwardLogicalChannelParameters; - -typedef struct OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters { /* CHOICE */ - enum { - eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters_h223LogicalChannelParameters, - eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters_v76LogicalChannelParameters, - eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters_h2250LogicalChannelParameters, - } choice; - union { - H2250LogicalChannelParameters h2250LogicalChannelParameters; - }; -} OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters; - -typedef struct OpenLogicalChannel_reverseLogicalChannelParameters { /* SEQUENCE */ - enum { - eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters - = (1 << 31), - eOpenLogicalChannel_reverseLogicalChannelParameters_reverseLogicalChannelDependency - = (1 << 30), - eOpenLogicalChannel_reverseLogicalChannelParameters_replacementFor - = (1 << 29), - } options; - OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters - multiplexParameters; -} OpenLogicalChannel_reverseLogicalChannelParameters; - -typedef struct NetworkAccessParameters_networkAddress { /* CHOICE */ - enum { - eNetworkAccessParameters_networkAddress_q2931Address, - eNetworkAccessParameters_networkAddress_e164Address, - eNetworkAccessParameters_networkAddress_localAreaAddress, - } choice; - union { - H245_TransportAddress localAreaAddress; - }; -} NetworkAccessParameters_networkAddress; - -typedef struct NetworkAccessParameters { /* SEQUENCE */ - enum { - eNetworkAccessParameters_distribution = (1 << 31), - eNetworkAccessParameters_externalReference = (1 << 30), - eNetworkAccessParameters_t120SetupProcedure = (1 << 29), - } options; - NetworkAccessParameters_networkAddress networkAddress; -} NetworkAccessParameters; - -typedef struct OpenLogicalChannel { /* SEQUENCE */ - enum { - eOpenLogicalChannel_reverseLogicalChannelParameters = - (1 << 31), - eOpenLogicalChannel_separateStack = (1 << 30), - eOpenLogicalChannel_encryptionSync = (1 << 29), - } options; - OpenLogicalChannel_forwardLogicalChannelParameters - forwardLogicalChannelParameters; - OpenLogicalChannel_reverseLogicalChannelParameters - reverseLogicalChannelParameters; - NetworkAccessParameters separateStack; -} OpenLogicalChannel; - -typedef struct Setup_UUIE_fastStart { /* SEQUENCE OF */ - int count; - OpenLogicalChannel item[30]; -} Setup_UUIE_fastStart; - -typedef struct Setup_UUIE { /* SEQUENCE */ - enum { - eSetup_UUIE_h245Address = (1 << 31), - eSetup_UUIE_sourceAddress = (1 << 30), - eSetup_UUIE_destinationAddress = (1 << 29), - eSetup_UUIE_destCallSignalAddress = (1 << 28), - eSetup_UUIE_destExtraCallInfo = (1 << 27), - eSetup_UUIE_destExtraCRV = (1 << 26), - eSetup_UUIE_callServices = (1 << 25), - eSetup_UUIE_sourceCallSignalAddress = (1 << 24), - eSetup_UUIE_remoteExtensionAddress = (1 << 23), - eSetup_UUIE_callIdentifier = (1 << 22), - eSetup_UUIE_h245SecurityCapability = (1 << 21), - eSetup_UUIE_tokens = (1 << 20), - eSetup_UUIE_cryptoTokens = (1 << 19), - eSetup_UUIE_fastStart = (1 << 18), - eSetup_UUIE_mediaWaitForConnect = (1 << 17), - eSetup_UUIE_canOverlapSend = (1 << 16), - eSetup_UUIE_endpointIdentifier = (1 << 15), - eSetup_UUIE_multipleCalls = (1 << 14), - eSetup_UUIE_maintainConnection = (1 << 13), - eSetup_UUIE_connectionParameters = (1 << 12), - eSetup_UUIE_language = (1 << 11), - eSetup_UUIE_presentationIndicator = (1 << 10), - eSetup_UUIE_screeningIndicator = (1 << 9), - eSetup_UUIE_serviceControl = (1 << 8), - eSetup_UUIE_symmetricOperationRequired = (1 << 7), - eSetup_UUIE_capacity = (1 << 6), - eSetup_UUIE_circuitInfo = (1 << 5), - eSetup_UUIE_desiredProtocols = (1 << 4), - eSetup_UUIE_neededFeatures = (1 << 3), - eSetup_UUIE_desiredFeatures = (1 << 2), - eSetup_UUIE_supportedFeatures = (1 << 1), - eSetup_UUIE_parallelH245Control = (1 << 0), - } options; - TransportAddress h245Address; - TransportAddress destCallSignalAddress; - TransportAddress sourceCallSignalAddress; - Setup_UUIE_fastStart fastStart; -} Setup_UUIE; - -typedef struct CallProceeding_UUIE_fastStart { /* SEQUENCE OF */ - int count; - OpenLogicalChannel item[30]; -} CallProceeding_UUIE_fastStart; - -typedef struct CallProceeding_UUIE { /* SEQUENCE */ - enum { - eCallProceeding_UUIE_h245Address = (1 << 31), - eCallProceeding_UUIE_callIdentifier = (1 << 30), - eCallProceeding_UUIE_h245SecurityMode = (1 << 29), - eCallProceeding_UUIE_tokens = (1 << 28), - eCallProceeding_UUIE_cryptoTokens = (1 << 27), - eCallProceeding_UUIE_fastStart = (1 << 26), - eCallProceeding_UUIE_multipleCalls = (1 << 25), - eCallProceeding_UUIE_maintainConnection = (1 << 24), - eCallProceeding_UUIE_fastConnectRefused = (1 << 23), - eCallProceeding_UUIE_featureSet = (1 << 22), - } options; - TransportAddress h245Address; - CallProceeding_UUIE_fastStart fastStart; -} CallProceeding_UUIE; - -typedef struct Connect_UUIE_fastStart { /* SEQUENCE OF */ - int count; - OpenLogicalChannel item[30]; -} Connect_UUIE_fastStart; - -typedef struct Connect_UUIE { /* SEQUENCE */ - enum { - eConnect_UUIE_h245Address = (1 << 31), - eConnect_UUIE_callIdentifier = (1 << 30), - eConnect_UUIE_h245SecurityMode = (1 << 29), - eConnect_UUIE_tokens = (1 << 28), - eConnect_UUIE_cryptoTokens = (1 << 27), - eConnect_UUIE_fastStart = (1 << 26), - eConnect_UUIE_multipleCalls = (1 << 25), - eConnect_UUIE_maintainConnection = (1 << 24), - eConnect_UUIE_language = (1 << 23), - eConnect_UUIE_connectedAddress = (1 << 22), - eConnect_UUIE_presentationIndicator = (1 << 21), - eConnect_UUIE_screeningIndicator = (1 << 20), - eConnect_UUIE_fastConnectRefused = (1 << 19), - eConnect_UUIE_serviceControl = (1 << 18), - eConnect_UUIE_capacity = (1 << 17), - eConnect_UUIE_featureSet = (1 << 16), - } options; - TransportAddress h245Address; - Connect_UUIE_fastStart fastStart; -} Connect_UUIE; - -typedef struct Alerting_UUIE_fastStart { /* SEQUENCE OF */ - int count; - OpenLogicalChannel item[30]; -} Alerting_UUIE_fastStart; - -typedef struct Alerting_UUIE { /* SEQUENCE */ - enum { - eAlerting_UUIE_h245Address = (1 << 31), - eAlerting_UUIE_callIdentifier = (1 << 30), - eAlerting_UUIE_h245SecurityMode = (1 << 29), - eAlerting_UUIE_tokens = (1 << 28), - eAlerting_UUIE_cryptoTokens = (1 << 27), - eAlerting_UUIE_fastStart = (1 << 26), - eAlerting_UUIE_multipleCalls = (1 << 25), - eAlerting_UUIE_maintainConnection = (1 << 24), - eAlerting_UUIE_alertingAddress = (1 << 23), - eAlerting_UUIE_presentationIndicator = (1 << 22), - eAlerting_UUIE_screeningIndicator = (1 << 21), - eAlerting_UUIE_fastConnectRefused = (1 << 20), - eAlerting_UUIE_serviceControl = (1 << 19), - eAlerting_UUIE_capacity = (1 << 18), - eAlerting_UUIE_featureSet = (1 << 17), - } options; - TransportAddress h245Address; - Alerting_UUIE_fastStart fastStart; -} Alerting_UUIE; - -typedef struct Information_UUIE_fastStart { /* SEQUENCE OF */ - int count; - OpenLogicalChannel item[30]; -} Information_UUIE_fastStart; - -typedef struct Information_UUIE { /* SEQUENCE */ - enum { - eInformation_UUIE_callIdentifier = (1 << 31), - eInformation_UUIE_tokens = (1 << 30), - eInformation_UUIE_cryptoTokens = (1 << 29), - eInformation_UUIE_fastStart = (1 << 28), - eInformation_UUIE_fastConnectRefused = (1 << 27), - eInformation_UUIE_circuitInfo = (1 << 26), - } options; - Information_UUIE_fastStart fastStart; -} Information_UUIE; - -typedef struct FacilityReason { /* CHOICE */ - enum { - eFacilityReason_routeCallToGatekeeper, - eFacilityReason_callForwarded, - eFacilityReason_routeCallToMC, - eFacilityReason_undefinedReason, - eFacilityReason_conferenceListChoice, - eFacilityReason_startH245, - eFacilityReason_noH245, - eFacilityReason_newTokens, - eFacilityReason_featureSetUpdate, - eFacilityReason_forwardedElements, - eFacilityReason_transportedInformation, - } choice; -} FacilityReason; - -typedef struct Facility_UUIE_fastStart { /* SEQUENCE OF */ - int count; - OpenLogicalChannel item[30]; -} Facility_UUIE_fastStart; - -typedef struct Facility_UUIE { /* SEQUENCE */ - enum { - eFacility_UUIE_alternativeAddress = (1 << 31), - eFacility_UUIE_alternativeAliasAddress = (1 << 30), - eFacility_UUIE_conferenceID = (1 << 29), - eFacility_UUIE_callIdentifier = (1 << 28), - eFacility_UUIE_destExtraCallInfo = (1 << 27), - eFacility_UUIE_remoteExtensionAddress = (1 << 26), - eFacility_UUIE_tokens = (1 << 25), - eFacility_UUIE_cryptoTokens = (1 << 24), - eFacility_UUIE_conferences = (1 << 23), - eFacility_UUIE_h245Address = (1 << 22), - eFacility_UUIE_fastStart = (1 << 21), - eFacility_UUIE_multipleCalls = (1 << 20), - eFacility_UUIE_maintainConnection = (1 << 19), - eFacility_UUIE_fastConnectRefused = (1 << 18), - eFacility_UUIE_serviceControl = (1 << 17), - eFacility_UUIE_circuitInfo = (1 << 16), - eFacility_UUIE_featureSet = (1 << 15), - eFacility_UUIE_destinationInfo = (1 << 14), - eFacility_UUIE_h245SecurityMode = (1 << 13), - } options; - TransportAddress alternativeAddress; - FacilityReason reason; - TransportAddress h245Address; - Facility_UUIE_fastStart fastStart; -} Facility_UUIE; - -typedef struct Progress_UUIE_fastStart { /* SEQUENCE OF */ - int count; - OpenLogicalChannel item[30]; -} Progress_UUIE_fastStart; - -typedef struct Progress_UUIE { /* SEQUENCE */ - enum { - eProgress_UUIE_h245Address = (1 << 31), - eProgress_UUIE_h245SecurityMode = (1 << 30), - eProgress_UUIE_tokens = (1 << 29), - eProgress_UUIE_cryptoTokens = (1 << 28), - eProgress_UUIE_fastStart = (1 << 27), - eProgress_UUIE_multipleCalls = (1 << 26), - eProgress_UUIE_maintainConnection = (1 << 25), - eProgress_UUIE_fastConnectRefused = (1 << 24), - } options; - TransportAddress h245Address; - Progress_UUIE_fastStart fastStart; -} Progress_UUIE; - -typedef struct H323_UU_PDU_h323_message_body { /* CHOICE */ - enum { - eH323_UU_PDU_h323_message_body_setup, - eH323_UU_PDU_h323_message_body_callProceeding, - eH323_UU_PDU_h323_message_body_connect, - eH323_UU_PDU_h323_message_body_alerting, - eH323_UU_PDU_h323_message_body_information, - eH323_UU_PDU_h323_message_body_releaseComplete, - eH323_UU_PDU_h323_message_body_facility, - eH323_UU_PDU_h323_message_body_progress, - eH323_UU_PDU_h323_message_body_empty, - eH323_UU_PDU_h323_message_body_status, - eH323_UU_PDU_h323_message_body_statusInquiry, - eH323_UU_PDU_h323_message_body_setupAcknowledge, - eH323_UU_PDU_h323_message_body_notify, - } choice; - union { - Setup_UUIE setup; - CallProceeding_UUIE callProceeding; - Connect_UUIE connect; - Alerting_UUIE alerting; - Information_UUIE information; - Facility_UUIE facility; - Progress_UUIE progress; - }; -} H323_UU_PDU_h323_message_body; - -typedef struct RequestMessage { /* CHOICE */ - enum { - eRequestMessage_nonStandard, - eRequestMessage_masterSlaveDetermination, - eRequestMessage_terminalCapabilitySet, - eRequestMessage_openLogicalChannel, - eRequestMessage_closeLogicalChannel, - eRequestMessage_requestChannelClose, - eRequestMessage_multiplexEntrySend, - eRequestMessage_requestMultiplexEntry, - eRequestMessage_requestMode, - eRequestMessage_roundTripDelayRequest, - eRequestMessage_maintenanceLoopRequest, - eRequestMessage_communicationModeRequest, - eRequestMessage_conferenceRequest, - eRequestMessage_multilinkRequest, - eRequestMessage_logicalChannelRateRequest, - } choice; - union { - OpenLogicalChannel openLogicalChannel; - }; -} RequestMessage; - -typedef struct OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters { /* CHOICE */ - enum { - eOpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters_h222LogicalChannelParameters, - eOpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters_h2250LogicalChannelParameters, - } choice; - union { - H2250LogicalChannelParameters h2250LogicalChannelParameters; - }; -} OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters; - -typedef struct OpenLogicalChannelAck_reverseLogicalChannelParameters { /* SEQUENCE */ - enum { - eOpenLogicalChannelAck_reverseLogicalChannelParameters_portNumber - = (1 << 31), - eOpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters - = (1 << 30), - eOpenLogicalChannelAck_reverseLogicalChannelParameters_replacementFor - = (1 << 29), - } options; - OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters - multiplexParameters; -} OpenLogicalChannelAck_reverseLogicalChannelParameters; - -typedef struct H2250LogicalChannelAckParameters { /* SEQUENCE */ - enum { - eH2250LogicalChannelAckParameters_nonStandard = (1 << 31), - eH2250LogicalChannelAckParameters_sessionID = (1 << 30), - eH2250LogicalChannelAckParameters_mediaChannel = (1 << 29), - eH2250LogicalChannelAckParameters_mediaControlChannel = - (1 << 28), - eH2250LogicalChannelAckParameters_dynamicRTPPayloadType = - (1 << 27), - eH2250LogicalChannelAckParameters_flowControlToZero = - (1 << 26), - eH2250LogicalChannelAckParameters_portNumber = (1 << 25), - } options; - H245_TransportAddress mediaChannel; - H245_TransportAddress mediaControlChannel; -} H2250LogicalChannelAckParameters; - -typedef struct OpenLogicalChannelAck_forwardMultiplexAckParameters { /* CHOICE */ - enum { - eOpenLogicalChannelAck_forwardMultiplexAckParameters_h2250LogicalChannelAckParameters, - } choice; - union { - H2250LogicalChannelAckParameters - h2250LogicalChannelAckParameters; - }; -} OpenLogicalChannelAck_forwardMultiplexAckParameters; - -typedef struct OpenLogicalChannelAck { /* SEQUENCE */ - enum { - eOpenLogicalChannelAck_reverseLogicalChannelParameters = - (1 << 31), - eOpenLogicalChannelAck_separateStack = (1 << 30), - eOpenLogicalChannelAck_forwardMultiplexAckParameters = - (1 << 29), - eOpenLogicalChannelAck_encryptionSync = (1 << 28), - } options; - OpenLogicalChannelAck_reverseLogicalChannelParameters - reverseLogicalChannelParameters; - OpenLogicalChannelAck_forwardMultiplexAckParameters - forwardMultiplexAckParameters; -} OpenLogicalChannelAck; - -typedef struct ResponseMessage { /* CHOICE */ - enum { - eResponseMessage_nonStandard, - eResponseMessage_masterSlaveDeterminationAck, - eResponseMessage_masterSlaveDeterminationReject, - eResponseMessage_terminalCapabilitySetAck, - eResponseMessage_terminalCapabilitySetReject, - eResponseMessage_openLogicalChannelAck, - eResponseMessage_openLogicalChannelReject, - eResponseMessage_closeLogicalChannelAck, - eResponseMessage_requestChannelCloseAck, - eResponseMessage_requestChannelCloseReject, - eResponseMessage_multiplexEntrySendAck, - eResponseMessage_multiplexEntrySendReject, - eResponseMessage_requestMultiplexEntryAck, - eResponseMessage_requestMultiplexEntryReject, - eResponseMessage_requestModeAck, - eResponseMessage_requestModeReject, - eResponseMessage_roundTripDelayResponse, - eResponseMessage_maintenanceLoopAck, - eResponseMessage_maintenanceLoopReject, - eResponseMessage_communicationModeResponse, - eResponseMessage_conferenceResponse, - eResponseMessage_multilinkResponse, - eResponseMessage_logicalChannelRateAcknowledge, - eResponseMessage_logicalChannelRateReject, - } choice; - union { - OpenLogicalChannelAck openLogicalChannelAck; - }; -} ResponseMessage; - -typedef struct MultimediaSystemControlMessage { /* CHOICE */ - enum { - eMultimediaSystemControlMessage_request, - eMultimediaSystemControlMessage_response, - eMultimediaSystemControlMessage_command, - eMultimediaSystemControlMessage_indication, - } choice; - union { - RequestMessage request; - ResponseMessage response; - }; -} MultimediaSystemControlMessage; - -typedef struct H323_UU_PDU_h245Control { /* SEQUENCE OF */ - int count; - MultimediaSystemControlMessage item[4]; -} H323_UU_PDU_h245Control; - -typedef struct H323_UU_PDU { /* SEQUENCE */ - enum { - eH323_UU_PDU_nonStandardData = (1 << 31), - eH323_UU_PDU_h4501SupplementaryService = (1 << 30), - eH323_UU_PDU_h245Tunneling = (1 << 29), - eH323_UU_PDU_h245Control = (1 << 28), - eH323_UU_PDU_nonStandardControl = (1 << 27), - eH323_UU_PDU_callLinkage = (1 << 26), - eH323_UU_PDU_tunnelledSignallingMessage = (1 << 25), - eH323_UU_PDU_provisionalRespToH245Tunneling = (1 << 24), - eH323_UU_PDU_stimulusControl = (1 << 23), - eH323_UU_PDU_genericData = (1 << 22), - } options; - H323_UU_PDU_h323_message_body h323_message_body; - H323_UU_PDU_h245Control h245Control; -} H323_UU_PDU; - -typedef struct H323_UserInformation { /* SEQUENCE */ - enum { - eH323_UserInformation_user_data = (1 << 31), - } options; - H323_UU_PDU h323_uu_pdu; -} H323_UserInformation; - -typedef struct GatekeeperRequest { /* SEQUENCE */ - enum { - eGatekeeperRequest_nonStandardData = (1 << 31), - eGatekeeperRequest_gatekeeperIdentifier = (1 << 30), - eGatekeeperRequest_callServices = (1 << 29), - eGatekeeperRequest_endpointAlias = (1 << 28), - eGatekeeperRequest_alternateEndpoints = (1 << 27), - eGatekeeperRequest_tokens = (1 << 26), - eGatekeeperRequest_cryptoTokens = (1 << 25), - eGatekeeperRequest_authenticationCapability = (1 << 24), - eGatekeeperRequest_algorithmOIDs = (1 << 23), - eGatekeeperRequest_integrity = (1 << 22), - eGatekeeperRequest_integrityCheckValue = (1 << 21), - eGatekeeperRequest_supportsAltGK = (1 << 20), - eGatekeeperRequest_featureSet = (1 << 19), - eGatekeeperRequest_genericData = (1 << 18), - } options; - TransportAddress rasAddress; -} GatekeeperRequest; - -typedef struct GatekeeperConfirm { /* SEQUENCE */ - enum { - eGatekeeperConfirm_nonStandardData = (1 << 31), - eGatekeeperConfirm_gatekeeperIdentifier = (1 << 30), - eGatekeeperConfirm_alternateGatekeeper = (1 << 29), - eGatekeeperConfirm_authenticationMode = (1 << 28), - eGatekeeperConfirm_tokens = (1 << 27), - eGatekeeperConfirm_cryptoTokens = (1 << 26), - eGatekeeperConfirm_algorithmOID = (1 << 25), - eGatekeeperConfirm_integrity = (1 << 24), - eGatekeeperConfirm_integrityCheckValue = (1 << 23), - eGatekeeperConfirm_featureSet = (1 << 22), - eGatekeeperConfirm_genericData = (1 << 21), - } options; - TransportAddress rasAddress; -} GatekeeperConfirm; - -typedef struct RegistrationRequest_callSignalAddress { /* SEQUENCE OF */ - int count; - TransportAddress item[10]; -} RegistrationRequest_callSignalAddress; - -typedef struct RegistrationRequest_rasAddress { /* SEQUENCE OF */ - int count; - TransportAddress item[10]; -} RegistrationRequest_rasAddress; - -typedef struct RegistrationRequest { /* SEQUENCE */ - enum { - eRegistrationRequest_nonStandardData = (1 << 31), - eRegistrationRequest_terminalAlias = (1 << 30), - eRegistrationRequest_gatekeeperIdentifier = (1 << 29), - eRegistrationRequest_alternateEndpoints = (1 << 28), - eRegistrationRequest_timeToLive = (1 << 27), - eRegistrationRequest_tokens = (1 << 26), - eRegistrationRequest_cryptoTokens = (1 << 25), - eRegistrationRequest_integrityCheckValue = (1 << 24), - eRegistrationRequest_keepAlive = (1 << 23), - eRegistrationRequest_endpointIdentifier = (1 << 22), - eRegistrationRequest_willSupplyUUIEs = (1 << 21), - eRegistrationRequest_maintainConnection = (1 << 20), - eRegistrationRequest_alternateTransportAddresses = (1 << 19), - eRegistrationRequest_additiveRegistration = (1 << 18), - eRegistrationRequest_terminalAliasPattern = (1 << 17), - eRegistrationRequest_supportsAltGK = (1 << 16), - eRegistrationRequest_usageReportingCapability = (1 << 15), - eRegistrationRequest_multipleCalls = (1 << 14), - eRegistrationRequest_supportedH248Packages = (1 << 13), - eRegistrationRequest_callCreditCapability = (1 << 12), - eRegistrationRequest_capacityReportingCapability = (1 << 11), - eRegistrationRequest_capacity = (1 << 10), - eRegistrationRequest_featureSet = (1 << 9), - eRegistrationRequest_genericData = (1 << 8), - } options; - RegistrationRequest_callSignalAddress callSignalAddress; - RegistrationRequest_rasAddress rasAddress; - unsigned timeToLive; -} RegistrationRequest; - -typedef struct RegistrationConfirm_callSignalAddress { /* SEQUENCE OF */ - int count; - TransportAddress item[10]; -} RegistrationConfirm_callSignalAddress; - -typedef struct RegistrationConfirm { /* SEQUENCE */ - enum { - eRegistrationConfirm_nonStandardData = (1 << 31), - eRegistrationConfirm_terminalAlias = (1 << 30), - eRegistrationConfirm_gatekeeperIdentifier = (1 << 29), - eRegistrationConfirm_alternateGatekeeper = (1 << 28), - eRegistrationConfirm_timeToLive = (1 << 27), - eRegistrationConfirm_tokens = (1 << 26), - eRegistrationConfirm_cryptoTokens = (1 << 25), - eRegistrationConfirm_integrityCheckValue = (1 << 24), - eRegistrationConfirm_willRespondToIRR = (1 << 23), - eRegistrationConfirm_preGrantedARQ = (1 << 22), - eRegistrationConfirm_maintainConnection = (1 << 21), - eRegistrationConfirm_serviceControl = (1 << 20), - eRegistrationConfirm_supportsAdditiveRegistration = (1 << 19), - eRegistrationConfirm_terminalAliasPattern = (1 << 18), - eRegistrationConfirm_supportedPrefixes = (1 << 17), - eRegistrationConfirm_usageSpec = (1 << 16), - eRegistrationConfirm_featureServerAlias = (1 << 15), - eRegistrationConfirm_capacityReportingSpec = (1 << 14), - eRegistrationConfirm_featureSet = (1 << 13), - eRegistrationConfirm_genericData = (1 << 12), - } options; - RegistrationConfirm_callSignalAddress callSignalAddress; - unsigned timeToLive; -} RegistrationConfirm; - -typedef struct UnregistrationRequest_callSignalAddress { /* SEQUENCE OF */ - int count; - TransportAddress item[10]; -} UnregistrationRequest_callSignalAddress; - -typedef struct UnregistrationRequest { /* SEQUENCE */ - enum { - eUnregistrationRequest_endpointAlias = (1 << 31), - eUnregistrationRequest_nonStandardData = (1 << 30), - eUnregistrationRequest_endpointIdentifier = (1 << 29), - eUnregistrationRequest_alternateEndpoints = (1 << 28), - eUnregistrationRequest_gatekeeperIdentifier = (1 << 27), - eUnregistrationRequest_tokens = (1 << 26), - eUnregistrationRequest_cryptoTokens = (1 << 25), - eUnregistrationRequest_integrityCheckValue = (1 << 24), - eUnregistrationRequest_reason = (1 << 23), - eUnregistrationRequest_endpointAliasPattern = (1 << 22), - eUnregistrationRequest_supportedPrefixes = (1 << 21), - eUnregistrationRequest_alternateGatekeeper = (1 << 20), - eUnregistrationRequest_genericData = (1 << 19), - } options; - UnregistrationRequest_callSignalAddress callSignalAddress; -} UnregistrationRequest; - -typedef struct AdmissionRequest { /* SEQUENCE */ - enum { - eAdmissionRequest_callModel = (1 << 31), - eAdmissionRequest_destinationInfo = (1 << 30), - eAdmissionRequest_destCallSignalAddress = (1 << 29), - eAdmissionRequest_destExtraCallInfo = (1 << 28), - eAdmissionRequest_srcCallSignalAddress = (1 << 27), - eAdmissionRequest_nonStandardData = (1 << 26), - eAdmissionRequest_callServices = (1 << 25), - eAdmissionRequest_canMapAlias = (1 << 24), - eAdmissionRequest_callIdentifier = (1 << 23), - eAdmissionRequest_srcAlternatives = (1 << 22), - eAdmissionRequest_destAlternatives = (1 << 21), - eAdmissionRequest_gatekeeperIdentifier = (1 << 20), - eAdmissionRequest_tokens = (1 << 19), - eAdmissionRequest_cryptoTokens = (1 << 18), - eAdmissionRequest_integrityCheckValue = (1 << 17), - eAdmissionRequest_transportQOS = (1 << 16), - eAdmissionRequest_willSupplyUUIEs = (1 << 15), - eAdmissionRequest_callLinkage = (1 << 14), - eAdmissionRequest_gatewayDataRate = (1 << 13), - eAdmissionRequest_capacity = (1 << 12), - eAdmissionRequest_circuitInfo = (1 << 11), - eAdmissionRequest_desiredProtocols = (1 << 10), - eAdmissionRequest_desiredTunnelledProtocol = (1 << 9), - eAdmissionRequest_featureSet = (1 << 8), - eAdmissionRequest_genericData = (1 << 7), - } options; - TransportAddress destCallSignalAddress; - TransportAddress srcCallSignalAddress; -} AdmissionRequest; - -typedef struct AdmissionConfirm { /* SEQUENCE */ - enum { - eAdmissionConfirm_irrFrequency = (1 << 31), - eAdmissionConfirm_nonStandardData = (1 << 30), - eAdmissionConfirm_destinationInfo = (1 << 29), - eAdmissionConfirm_destExtraCallInfo = (1 << 28), - eAdmissionConfirm_destinationType = (1 << 27), - eAdmissionConfirm_remoteExtensionAddress = (1 << 26), - eAdmissionConfirm_alternateEndpoints = (1 << 25), - eAdmissionConfirm_tokens = (1 << 24), - eAdmissionConfirm_cryptoTokens = (1 << 23), - eAdmissionConfirm_integrityCheckValue = (1 << 22), - eAdmissionConfirm_transportQOS = (1 << 21), - eAdmissionConfirm_willRespondToIRR = (1 << 20), - eAdmissionConfirm_uuiesRequested = (1 << 19), - eAdmissionConfirm_language = (1 << 18), - eAdmissionConfirm_alternateTransportAddresses = (1 << 17), - eAdmissionConfirm_useSpecifiedTransport = (1 << 16), - eAdmissionConfirm_circuitInfo = (1 << 15), - eAdmissionConfirm_usageSpec = (1 << 14), - eAdmissionConfirm_supportedProtocols = (1 << 13), - eAdmissionConfirm_serviceControl = (1 << 12), - eAdmissionConfirm_multipleCalls = (1 << 11), - eAdmissionConfirm_featureSet = (1 << 10), - eAdmissionConfirm_genericData = (1 << 9), - } options; - TransportAddress destCallSignalAddress; -} AdmissionConfirm; - -typedef struct LocationRequest { /* SEQUENCE */ - enum { - eLocationRequest_endpointIdentifier = (1 << 31), - eLocationRequest_nonStandardData = (1 << 30), - eLocationRequest_sourceInfo = (1 << 29), - eLocationRequest_canMapAlias = (1 << 28), - eLocationRequest_gatekeeperIdentifier = (1 << 27), - eLocationRequest_tokens = (1 << 26), - eLocationRequest_cryptoTokens = (1 << 25), - eLocationRequest_integrityCheckValue = (1 << 24), - eLocationRequest_desiredProtocols = (1 << 23), - eLocationRequest_desiredTunnelledProtocol = (1 << 22), - eLocationRequest_featureSet = (1 << 21), - eLocationRequest_genericData = (1 << 20), - eLocationRequest_hopCount = (1 << 19), - eLocationRequest_circuitInfo = (1 << 18), - } options; - TransportAddress replyAddress; -} LocationRequest; - -typedef struct LocationConfirm { /* SEQUENCE */ - enum { - eLocationConfirm_nonStandardData = (1 << 31), - eLocationConfirm_destinationInfo = (1 << 30), - eLocationConfirm_destExtraCallInfo = (1 << 29), - eLocationConfirm_destinationType = (1 << 28), - eLocationConfirm_remoteExtensionAddress = (1 << 27), - eLocationConfirm_alternateEndpoints = (1 << 26), - eLocationConfirm_tokens = (1 << 25), - eLocationConfirm_cryptoTokens = (1 << 24), - eLocationConfirm_integrityCheckValue = (1 << 23), - eLocationConfirm_alternateTransportAddresses = (1 << 22), - eLocationConfirm_supportedProtocols = (1 << 21), - eLocationConfirm_multipleCalls = (1 << 20), - eLocationConfirm_featureSet = (1 << 19), - eLocationConfirm_genericData = (1 << 18), - eLocationConfirm_circuitInfo = (1 << 17), - eLocationConfirm_serviceControl = (1 << 16), - } options; - TransportAddress callSignalAddress; - TransportAddress rasAddress; -} LocationConfirm; - -typedef struct InfoRequestResponse_callSignalAddress { /* SEQUENCE OF */ - int count; - TransportAddress item[10]; -} InfoRequestResponse_callSignalAddress; - -typedef struct InfoRequestResponse { /* SEQUENCE */ - enum { - eInfoRequestResponse_nonStandardData = (1 << 31), - eInfoRequestResponse_endpointAlias = (1 << 30), - eInfoRequestResponse_perCallInfo = (1 << 29), - eInfoRequestResponse_tokens = (1 << 28), - eInfoRequestResponse_cryptoTokens = (1 << 27), - eInfoRequestResponse_integrityCheckValue = (1 << 26), - eInfoRequestResponse_needResponse = (1 << 25), - eInfoRequestResponse_capacity = (1 << 24), - eInfoRequestResponse_irrStatus = (1 << 23), - eInfoRequestResponse_unsolicited = (1 << 22), - eInfoRequestResponse_genericData = (1 << 21), - } options; - TransportAddress rasAddress; - InfoRequestResponse_callSignalAddress callSignalAddress; -} InfoRequestResponse; - -typedef struct RasMessage { /* CHOICE */ - enum { - eRasMessage_gatekeeperRequest, - eRasMessage_gatekeeperConfirm, - eRasMessage_gatekeeperReject, - eRasMessage_registrationRequest, - eRasMessage_registrationConfirm, - eRasMessage_registrationReject, - eRasMessage_unregistrationRequest, - eRasMessage_unregistrationConfirm, - eRasMessage_unregistrationReject, - eRasMessage_admissionRequest, - eRasMessage_admissionConfirm, - eRasMessage_admissionReject, - eRasMessage_bandwidthRequest, - eRasMessage_bandwidthConfirm, - eRasMessage_bandwidthReject, - eRasMessage_disengageRequest, - eRasMessage_disengageConfirm, - eRasMessage_disengageReject, - eRasMessage_locationRequest, - eRasMessage_locationConfirm, - eRasMessage_locationReject, - eRasMessage_infoRequest, - eRasMessage_infoRequestResponse, - eRasMessage_nonStandardMessage, - eRasMessage_unknownMessageResponse, - eRasMessage_requestInProgress, - eRasMessage_resourcesAvailableIndicate, - eRasMessage_resourcesAvailableConfirm, - eRasMessage_infoRequestAck, - eRasMessage_infoRequestNak, - eRasMessage_serviceControlIndication, - eRasMessage_serviceControlResponse, - } choice; - union { - GatekeeperRequest gatekeeperRequest; - GatekeeperConfirm gatekeeperConfirm; - RegistrationRequest registrationRequest; - RegistrationConfirm registrationConfirm; - UnregistrationRequest unregistrationRequest; - AdmissionRequest admissionRequest; - AdmissionConfirm admissionConfirm; - LocationRequest locationRequest; - LocationConfirm locationConfirm; - InfoRequestResponse infoRequestResponse; - }; -} RasMessage; diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 83694cfdfa8..1646076933b 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -41,11 +41,13 @@ union nf_conntrack_expect_proto { /* Add protocol helper include file here */ #include +#include /* per conntrack: application helper private data */ union nf_conntrack_help { /* insert conntrack helper private data (master) here */ struct nf_ct_ftp_master ct_ftp_info; + struct nf_ct_h323_master ct_h323_info; }; #include diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h index 54a3d038bea..cef3136e22a 100644 --- a/include/net/netfilter/nf_conntrack_expect.h +++ b/include/net/netfilter/nf_conntrack_expect.h @@ -41,6 +41,7 @@ struct nf_conntrack_expect unsigned int flags; #ifdef CONFIG_NF_NAT_NEEDED + __be32 saved_ip; /* This is the original per-proto part, used to map the * expected connection the way the recipient expects. */ union nf_conntrack_man_proto saved_proto; diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 6993ec53dc0..e14156d1122 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -529,6 +529,11 @@ config IP_NF_NAT_H323 default IP_NF_NAT if IP_NF_H323=y default m if IP_NF_H323=m +config NF_NAT_H323 + tristate + depends on IP_NF_IPTABLES && NF_CONNTRACK && NF_NAT + default NF_NAT && NF_CONNTRACK_H323 + config IP_NF_NAT_SIP tristate depends on IP_NF_IPTABLES!=n && IP_NF_CONNTRACK!=n && IP_NF_NAT!=n diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index 8893249bbe9..bdaba4700e3 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -15,7 +15,7 @@ endif ip_conntrack_pptp-objs := ip_conntrack_helper_pptp.o ip_conntrack_proto_gre.o ip_nat_pptp-objs := ip_nat_helper_pptp.o ip_nat_proto_gre.o -ip_conntrack_h323-objs := ip_conntrack_helper_h323.o ip_conntrack_helper_h323_asn1.o +ip_conntrack_h323-objs := ip_conntrack_helper_h323.o ../../netfilter/nf_conntrack_h323_asn1.o ip_nat_h323-objs := ip_nat_helper_h323.o # connection tracking @@ -52,6 +52,7 @@ obj-$(CONFIG_IP_NF_NAT_SIP) += ip_nat_sip.o # NAT helpers (nf_conntrack) obj-$(CONFIG_NF_NAT_AMANDA) += nf_nat_amanda.o obj-$(CONFIG_NF_NAT_FTP) += nf_nat_ftp.o +obj-$(CONFIG_NF_NAT_H323) += nf_nat_h323.o # generic IP tables obj-$(CONFIG_IP_NF_IPTABLES) += ip_tables.o diff --git a/net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c b/net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c deleted file mode 100644 index 26dfecadb33..00000000000 --- a/net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c +++ /dev/null @@ -1,874 +0,0 @@ -/**************************************************************************** - * ip_conntrack_helper_h323_asn1.c - BER and PER decoding library for H.323 - * conntrack/NAT module. - * - * Copyright (c) 2006 by Jing Min Zhao - * - * This source code is licensed under General Public License version 2. - * - * See ip_conntrack_helper_h323_asn1.h for details. - * - ****************************************************************************/ - -#ifdef __KERNEL__ -#include -#else -#include -#endif -#include - -/* Trace Flag */ -#ifndef H323_TRACE -#define H323_TRACE 0 -#endif - -#if H323_TRACE -#define TAB_SIZE 4 -#define IFTHEN(cond, act) if(cond){act;} -#ifdef __KERNEL__ -#define PRINT printk -#else -#define PRINT printf -#endif -#define FNAME(name) name, -#else -#define IFTHEN(cond, act) -#define PRINT(fmt, args...) -#define FNAME(name) -#endif - -/* ASN.1 Types */ -#define NUL 0 -#define BOOL 1 -#define OID 2 -#define INT 3 -#define ENUM 4 -#define BITSTR 5 -#define NUMSTR 6 -#define NUMDGT 6 -#define TBCDSTR 6 -#define OCTSTR 7 -#define PRTSTR 7 -#define IA5STR 7 -#define GENSTR 7 -#define BMPSTR 8 -#define SEQ 9 -#define SET 9 -#define SEQOF 10 -#define SETOF 10 -#define CHOICE 11 - -/* Constraint Types */ -#define FIXD 0 -/* #define BITS 1-8 */ -#define BYTE 9 -#define WORD 10 -#define CONS 11 -#define SEMI 12 -#define UNCO 13 - -/* ASN.1 Type Attributes */ -#define SKIP 0 -#define STOP 1 -#define DECODE 2 -#define EXT 4 -#define OPEN 8 -#define OPT 16 - - -/* ASN.1 Field Structure */ -typedef struct field_t { -#if H323_TRACE - char *name; -#endif - unsigned char type; - unsigned char sz; - unsigned char lb; - unsigned char ub; - unsigned short attr; - unsigned short offset; - struct field_t *fields; -} field_t; - -/* Bit Stream */ -typedef struct { - unsigned char *buf; - unsigned char *beg; - unsigned char *end; - unsigned char *cur; - unsigned bit; -} bitstr_t; - -/* Tool Functions */ -#define INC_BIT(bs) if((++bs->bit)>7){bs->cur++;bs->bit=0;} -#define INC_BITS(bs,b) if((bs->bit+=b)>7){bs->cur+=bs->bit>>3;bs->bit&=7;} -#define BYTE_ALIGN(bs) if(bs->bit){bs->cur++;bs->bit=0;} -#define CHECK_BOUND(bs,n) if(bs->cur+(n)>bs->end)return(H323_ERROR_BOUND) -static unsigned get_len(bitstr_t * bs); -static unsigned get_bit(bitstr_t * bs); -static unsigned get_bits(bitstr_t * bs, unsigned b); -static unsigned get_bitmap(bitstr_t * bs, unsigned b); -static unsigned get_uint(bitstr_t * bs, int b); - -/* Decoder Functions */ -static int decode_nul(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_bool(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_oid(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_int(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_enum(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_bitstr(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_numstr(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_octstr(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_bmpstr(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_seq(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_seqof(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_choice(bitstr_t * bs, field_t * f, char *base, int level); - -/* Decoder Functions Vector */ -typedef int (*decoder_t) (bitstr_t *, field_t *, char *, int); -static decoder_t Decoders[] = { - decode_nul, - decode_bool, - decode_oid, - decode_int, - decode_enum, - decode_bitstr, - decode_numstr, - decode_octstr, - decode_bmpstr, - decode_seq, - decode_seqof, - decode_choice, -}; - -/**************************************************************************** - * H.323 Types - ****************************************************************************/ -#include "ip_conntrack_helper_h323_types.c" - -/**************************************************************************** - * Functions - ****************************************************************************/ -/* Assume bs is aligned && v < 16384 */ -unsigned get_len(bitstr_t * bs) -{ - unsigned v; - - v = *bs->cur++; - - if (v & 0x80) { - v &= 0x3f; - v <<= 8; - v += *bs->cur++; - } - - return v; -} - -/****************************************************************************/ -unsigned get_bit(bitstr_t * bs) -{ - unsigned b = (*bs->cur) & (0x80 >> bs->bit); - - INC_BIT(bs); - - return b; -} - -/****************************************************************************/ -/* Assume b <= 8 */ -unsigned get_bits(bitstr_t * bs, unsigned b) -{ - unsigned v, l; - - v = (*bs->cur) & (0xffU >> bs->bit); - l = b + bs->bit; - - if (l < 8) { - v >>= 8 - l; - bs->bit = l; - } else if (l == 8) { - bs->cur++; - bs->bit = 0; - } else { /* l > 8 */ - - v <<= 8; - v += *(++bs->cur); - v >>= 16 - l; - bs->bit = l - 8; - } - - return v; -} - -/****************************************************************************/ -/* Assume b <= 32 */ -unsigned get_bitmap(bitstr_t * bs, unsigned b) -{ - unsigned v, l, shift, bytes; - - if (!b) - return 0; - - l = bs->bit + b; - - if (l < 8) { - v = (unsigned) (*bs->cur) << (bs->bit + 24); - bs->bit = l; - } else if (l == 8) { - v = (unsigned) (*bs->cur++) << (bs->bit + 24); - bs->bit = 0; - } else { - for (bytes = l >> 3, shift = 24, v = 0; bytes; - bytes--, shift -= 8) - v |= (unsigned) (*bs->cur++) << shift; - - if (l < 32) { - v |= (unsigned) (*bs->cur) << shift; - v <<= bs->bit; - } else if (l > 32) { - v <<= bs->bit; - v |= (*bs->cur) >> (8 - bs->bit); - } - - bs->bit = l & 0x7; - } - - v &= 0xffffffff << (32 - b); - - return v; -} - -/**************************************************************************** - * Assume bs is aligned and sizeof(unsigned int) == 4 - ****************************************************************************/ -unsigned get_uint(bitstr_t * bs, int b) -{ - unsigned v = 0; - - switch (b) { - case 4: - v |= *bs->cur++; - v <<= 8; - case 3: - v |= *bs->cur++; - v <<= 8; - case 2: - v |= *bs->cur++; - v <<= 8; - case 1: - v |= *bs->cur++; - break; - } - return v; -} - -/****************************************************************************/ -int decode_nul(bitstr_t * bs, field_t * f, char *base, int level) -{ - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_bool(bitstr_t * bs, field_t * f, char *base, int level) -{ - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - INC_BIT(bs); - - CHECK_BOUND(bs, 0); - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_oid(bitstr_t * bs, field_t * f, char *base, int level) -{ - int len; - - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - BYTE_ALIGN(bs); - CHECK_BOUND(bs, 1); - len = *bs->cur++; - bs->cur += len; - - CHECK_BOUND(bs, 0); - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_int(bitstr_t * bs, field_t * f, char *base, int level) -{ - unsigned len; - - PRINT("%*.s%s", level * TAB_SIZE, " ", f->name); - - switch (f->sz) { - case BYTE: /* Range == 256 */ - BYTE_ALIGN(bs); - bs->cur++; - break; - case WORD: /* 257 <= Range <= 64K */ - BYTE_ALIGN(bs); - bs->cur += 2; - break; - case CONS: /* 64K < Range < 4G */ - len = get_bits(bs, 2) + 1; - BYTE_ALIGN(bs); - if (base && (f->attr & DECODE)) { /* timeToLive */ - unsigned v = get_uint(bs, len) + f->lb; - PRINT(" = %u", v); - *((unsigned *) (base + f->offset)) = v; - } - bs->cur += len; - break; - case UNCO: - BYTE_ALIGN(bs); - CHECK_BOUND(bs, 2); - len = get_len(bs); - bs->cur += len; - break; - default: /* 2 <= Range <= 255 */ - INC_BITS(bs, f->sz); - break; - } - - PRINT("\n"); - - CHECK_BOUND(bs, 0); - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_enum(bitstr_t * bs, field_t * f, char *base, int level) -{ - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - if ((f->attr & EXT) && get_bit(bs)) { - INC_BITS(bs, 7); - } else { - INC_BITS(bs, f->sz); - } - - CHECK_BOUND(bs, 0); - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_bitstr(bitstr_t * bs, field_t * f, char *base, int level) -{ - unsigned len; - - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - BYTE_ALIGN(bs); - switch (f->sz) { - case FIXD: /* fixed length > 16 */ - len = f->lb; - break; - case WORD: /* 2-byte length */ - CHECK_BOUND(bs, 2); - len = (*bs->cur++) << 8; - len += (*bs->cur++) + f->lb; - break; - case SEMI: - CHECK_BOUND(bs, 2); - len = get_len(bs); - break; - default: - len = 0; - break; - } - - bs->cur += len >> 3; - bs->bit = len & 7; - - CHECK_BOUND(bs, 0); - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_numstr(bitstr_t * bs, field_t * f, char *base, int level) -{ - unsigned len; - - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - /* 2 <= Range <= 255 */ - len = get_bits(bs, f->sz) + f->lb; - - BYTE_ALIGN(bs); - INC_BITS(bs, (len << 2)); - - CHECK_BOUND(bs, 0); - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_octstr(bitstr_t * bs, field_t * f, char *base, int level) -{ - unsigned len; - - PRINT("%*.s%s", level * TAB_SIZE, " ", f->name); - - switch (f->sz) { - case FIXD: /* Range == 1 */ - if (f->lb > 2) { - BYTE_ALIGN(bs); - if (base && (f->attr & DECODE)) { - /* The IP Address */ - IFTHEN(f->lb == 4, - PRINT(" = %d.%d.%d.%d:%d", - bs->cur[0], bs->cur[1], - bs->cur[2], bs->cur[3], - bs->cur[4] * 256 + bs->cur[5])); - *((unsigned *) (base + f->offset)) = - bs->cur - bs->buf; - } - } - len = f->lb; - break; - case BYTE: /* Range == 256 */ - BYTE_ALIGN(bs); - CHECK_BOUND(bs, 1); - len = (*bs->cur++) + f->lb; - break; - case SEMI: - BYTE_ALIGN(bs); - CHECK_BOUND(bs, 2); - len = get_len(bs) + f->lb; - break; - default: /* 2 <= Range <= 255 */ - len = get_bits(bs, f->sz) + f->lb; - BYTE_ALIGN(bs); - break; - } - - bs->cur += len; - - PRINT("\n"); - - CHECK_BOUND(bs, 0); - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_bmpstr(bitstr_t * bs, field_t * f, char *base, int level) -{ - unsigned len; - - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - switch (f->sz) { - case BYTE: /* Range == 256 */ - BYTE_ALIGN(bs); - CHECK_BOUND(bs, 1); - len = (*bs->cur++) + f->lb; - break; - default: /* 2 <= Range <= 255 */ - len = get_bits(bs, f->sz) + f->lb; - BYTE_ALIGN(bs); - break; - } - - bs->cur += len << 1; - - CHECK_BOUND(bs, 0); - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_seq(bitstr_t * bs, field_t * f, char *base, int level) -{ - unsigned ext, bmp, i, opt, len = 0, bmp2, bmp2_len; - int err; - field_t *son; - unsigned char *beg = NULL; - - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - /* Decode? */ - base = (base && (f->attr & DECODE)) ? base + f->offset : NULL; - - /* Extensible? */ - ext = (f->attr & EXT) ? get_bit(bs) : 0; - - /* Get fields bitmap */ - bmp = get_bitmap(bs, f->sz); - if (base) - *(unsigned *) base = bmp; - - /* Decode the root components */ - for (i = opt = 0, son = f->fields; i < f->lb; i++, son++) { - if (son->attr & STOP) { - PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", - son->name); - return H323_ERROR_STOP; - } - - if (son->attr & OPT) { /* Optional component */ - if (!((0x80000000U >> (opt++)) & bmp)) /* Not exist */ - continue; - } - - /* Decode */ - if (son->attr & OPEN) { /* Open field */ - CHECK_BOUND(bs, 2); - len = get_len(bs); - CHECK_BOUND(bs, len); - if (!base) { - PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, - " ", son->name); - bs->cur += len; - continue; - } - beg = bs->cur; - - /* Decode */ - if ((err = (Decoders[son->type]) (bs, son, base, - level + 1)) < - H323_ERROR_NONE) - return err; - - bs->cur = beg + len; - bs->bit = 0; - } else if ((err = (Decoders[son->type]) (bs, son, base, - level + 1)) < - H323_ERROR_NONE) - return err; - } - - /* No extension? */ - if (!ext) - return H323_ERROR_NONE; - - /* Get the extension bitmap */ - bmp2_len = get_bits(bs, 7) + 1; - CHECK_BOUND(bs, (bmp2_len + 7) >> 3); - bmp2 = get_bitmap(bs, bmp2_len); - bmp |= bmp2 >> f->sz; - if (base) - *(unsigned *) base = bmp; - BYTE_ALIGN(bs); - - /* Decode the extension components */ - for (opt = 0; opt < bmp2_len; opt++, i++, son++) { - if (i < f->ub && son->attr & STOP) { - PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", - son->name); - return H323_ERROR_STOP; - } - - if (!((0x80000000 >> opt) & bmp2)) /* Not present */ - continue; - - /* Check Range */ - if (i >= f->ub) { /* Newer Version? */ - CHECK_BOUND(bs, 2); - len = get_len(bs); - CHECK_BOUND(bs, len); - bs->cur += len; - continue; - } - - CHECK_BOUND(bs, 2); - len = get_len(bs); - CHECK_BOUND(bs, len); - if (!base || !(son->attr & DECODE)) { - PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", - son->name); - bs->cur += len; - continue; - } - beg = bs->cur; - - if ((err = (Decoders[son->type]) (bs, son, base, - level + 1)) < - H323_ERROR_NONE) - return err; - - bs->cur = beg + len; - bs->bit = 0; - } - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_seqof(bitstr_t * bs, field_t * f, char *base, int level) -{ - unsigned count, effective_count = 0, i, len = 0; - int err; - field_t *son; - unsigned char *beg = NULL; - - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - /* Decode? */ - base = (base && (f->attr & DECODE)) ? base + f->offset : NULL; - - /* Decode item count */ - switch (f->sz) { - case BYTE: - BYTE_ALIGN(bs); - CHECK_BOUND(bs, 1); - count = *bs->cur++; - break; - case WORD: - BYTE_ALIGN(bs); - CHECK_BOUND(bs, 2); - count = *bs->cur++; - count <<= 8; - count = *bs->cur++; - break; - case SEMI: - BYTE_ALIGN(bs); - CHECK_BOUND(bs, 2); - count = get_len(bs); - break; - default: - count = get_bits(bs, f->sz); - break; - } - count += f->lb; - - /* Write Count */ - if (base) { - effective_count = count > f->ub ? f->ub : count; - *(unsigned *) base = effective_count; - base += sizeof(unsigned); - } - - /* Decode nested field */ - son = f->fields; - if (base) - base -= son->offset; - for (i = 0; i < count; i++) { - if (son->attr & OPEN) { - BYTE_ALIGN(bs); - len = get_len(bs); - CHECK_BOUND(bs, len); - if (!base || !(son->attr & DECODE)) { - PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, - " ", son->name); - bs->cur += len; - continue; - } - beg = bs->cur; - - if ((err = (Decoders[son->type]) (bs, son, - i < - effective_count ? - base : NULL, - level + 1)) < - H323_ERROR_NONE) - return err; - - bs->cur = beg + len; - bs->bit = 0; - } else - if ((err = (Decoders[son->type]) (bs, son, - i < - effective_count ? - base : NULL, - level + 1)) < - H323_ERROR_NONE) - return err; - - if (base) - base += son->offset; - } - - return H323_ERROR_NONE; -} - - -/****************************************************************************/ -int decode_choice(bitstr_t * bs, field_t * f, char *base, int level) -{ - unsigned type, ext, len = 0; - int err; - field_t *son; - unsigned char *beg = NULL; - - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - /* Decode? */ - base = (base && (f->attr & DECODE)) ? base + f->offset : NULL; - - /* Decode the choice index number */ - if ((f->attr & EXT) && get_bit(bs)) { - ext = 1; - type = get_bits(bs, 7) + f->lb; - } else { - ext = 0; - type = get_bits(bs, f->sz); - } - - /* Write Type */ - if (base) - *(unsigned *) base = type; - - /* Check Range */ - if (type >= f->ub) { /* Newer version? */ - BYTE_ALIGN(bs); - len = get_len(bs); - CHECK_BOUND(bs, len); - bs->cur += len; - return H323_ERROR_NONE; - } - - /* Transfer to son level */ - son = &f->fields[type]; - if (son->attr & STOP) { - PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", son->name); - return H323_ERROR_STOP; - } - - if (ext || (son->attr & OPEN)) { - BYTE_ALIGN(bs); - len = get_len(bs); - CHECK_BOUND(bs, len); - if (!base || !(son->attr & DECODE)) { - PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", - son->name); - bs->cur += len; - return H323_ERROR_NONE; - } - beg = bs->cur; - - if ((err = (Decoders[son->type]) (bs, son, base, level + 1)) < - H323_ERROR_NONE) - return err; - - bs->cur = beg + len; - bs->bit = 0; - } else if ((err = (Decoders[son->type]) (bs, son, base, level + 1)) < - H323_ERROR_NONE) - return err; - - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int DecodeRasMessage(unsigned char *buf, size_t sz, RasMessage * ras) -{ - static field_t ras_message = { - FNAME("RasMessage") CHOICE, 5, 24, 32, DECODE | EXT, - 0, _RasMessage - }; - bitstr_t bs; - - bs.buf = bs.beg = bs.cur = buf; - bs.end = buf + sz; - bs.bit = 0; - - return decode_choice(&bs, &ras_message, (char *) ras, 0); -} - -/****************************************************************************/ -static int DecodeH323_UserInformation(unsigned char *buf, unsigned char *beg, - size_t sz, H323_UserInformation * uuie) -{ - static field_t h323_userinformation = { - FNAME("H323-UserInformation") SEQ, 1, 2, 2, DECODE | EXT, - 0, _H323_UserInformation - }; - bitstr_t bs; - - bs.buf = buf; - bs.beg = bs.cur = beg; - bs.end = beg + sz; - bs.bit = 0; - - return decode_seq(&bs, &h323_userinformation, (char *) uuie, 0); -} - -/****************************************************************************/ -int DecodeMultimediaSystemControlMessage(unsigned char *buf, size_t sz, - MultimediaSystemControlMessage * - mscm) -{ - static field_t multimediasystemcontrolmessage = { - FNAME("MultimediaSystemControlMessage") CHOICE, 2, 4, 4, - DECODE | EXT, 0, _MultimediaSystemControlMessage - }; - bitstr_t bs; - - bs.buf = bs.beg = bs.cur = buf; - bs.end = buf + sz; - bs.bit = 0; - - return decode_choice(&bs, &multimediasystemcontrolmessage, - (char *) mscm, 0); -} - -/****************************************************************************/ -int DecodeQ931(unsigned char *buf, size_t sz, Q931 * q931) -{ - unsigned char *p = buf; - int len; - - if (!p || sz < 1) - return H323_ERROR_BOUND; - - /* Protocol Discriminator */ - if (*p != 0x08) { - PRINT("Unknown Protocol Discriminator\n"); - return H323_ERROR_RANGE; - } - p++; - sz--; - - /* CallReferenceValue */ - if (sz < 1) - return H323_ERROR_BOUND; - len = *p++; - sz--; - if (sz < len) - return H323_ERROR_BOUND; - p += len; - sz -= len; - - /* Message Type */ - if (sz < 1) - return H323_ERROR_BOUND; - q931->MessageType = *p++; - PRINT("MessageType = %02X\n", q931->MessageType); - if (*p & 0x80) { - p++; - sz--; - } - - /* Decode Information Elements */ - while (sz > 0) { - if (*p == 0x7e) { /* UserUserIE */ - if (sz < 3) - break; - p++; - len = *p++ << 8; - len |= *p++; - sz -= 3; - if (sz < len) - break; - p++; - len--; - return DecodeH323_UserInformation(buf, p, len, - &q931->UUIE); - } - p++; - sz--; - if (sz < 1) - break; - len = *p++; - if (sz < len) - break; - p += len; - sz -= len; - } - - PRINT("Q.931 UUIE not found\n"); - - return H323_ERROR_BOUND; -} diff --git a/net/ipv4/netfilter/ip_conntrack_helper_h323_types.c b/net/ipv4/netfilter/ip_conntrack_helper_h323_types.c deleted file mode 100644 index 4b359618bed..00000000000 --- a/net/ipv4/netfilter/ip_conntrack_helper_h323_types.c +++ /dev/null @@ -1,1926 +0,0 @@ -/* Generated by Jing Min Zhao's ASN.1 parser, Apr 20 2006 - * - * Copyright (c) 2006 Jing Min Zhao - * - * This source code is licensed under General Public License version 2. - */ - -static field_t _TransportAddress_ipAddress[] = { /* SEQUENCE */ - {FNAME("ip") OCTSTR, FIXD, 4, 0, DECODE, - offsetof(TransportAddress_ipAddress, ip), NULL}, - {FNAME("port") INT, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _TransportAddress_ipSourceRoute_route[] = { /* SEQUENCE OF */ - {FNAME("item") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, -}; - -static field_t _TransportAddress_ipSourceRoute_routing[] = { /* CHOICE */ - {FNAME("strict") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("loose") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _TransportAddress_ipSourceRoute[] = { /* SEQUENCE */ - {FNAME("ip") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, - {FNAME("port") INT, WORD, 0, 0, SKIP, 0, NULL}, - {FNAME("route") SEQOF, SEMI, 0, 0, SKIP, 0, - _TransportAddress_ipSourceRoute_route}, - {FNAME("routing") CHOICE, 1, 2, 2, SKIP | EXT, 0, - _TransportAddress_ipSourceRoute_routing}, -}; - -static field_t _TransportAddress_ipxAddress[] = { /* SEQUENCE */ - {FNAME("node") OCTSTR, FIXD, 6, 0, SKIP, 0, NULL}, - {FNAME("netnum") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, - {FNAME("port") OCTSTR, FIXD, 2, 0, SKIP, 0, NULL}, -}; - -static field_t _TransportAddress_ip6Address[] = { /* SEQUENCE */ - {FNAME("ip") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, - {FNAME("port") INT, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H221NonStandard[] = { /* SEQUENCE */ - {FNAME("t35CountryCode") INT, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("t35Extension") INT, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("manufacturerCode") INT, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _NonStandardIdentifier[] = { /* CHOICE */ - {FNAME("object") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("h221NonStandard") SEQ, 0, 3, 3, SKIP | EXT, 0, - _H221NonStandard}, -}; - -static field_t _NonStandardParameter[] = { /* SEQUENCE */ - {FNAME("nonStandardIdentifier") CHOICE, 1, 2, 2, SKIP | EXT, 0, - _NonStandardIdentifier}, - {FNAME("data") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _TransportAddress[] = { /* CHOICE */ - {FNAME("ipAddress") SEQ, 0, 2, 2, DECODE, - offsetof(TransportAddress, ipAddress), _TransportAddress_ipAddress}, - {FNAME("ipSourceRoute") SEQ, 0, 4, 4, SKIP | EXT, 0, - _TransportAddress_ipSourceRoute}, - {FNAME("ipxAddress") SEQ, 0, 3, 3, SKIP, 0, - _TransportAddress_ipxAddress}, - {FNAME("ip6Address") SEQ, 0, 2, 2, SKIP | EXT, 0, - _TransportAddress_ip6Address}, - {FNAME("netBios") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, - {FNAME("nsap") OCTSTR, 5, 1, 0, SKIP, 0, NULL}, - {FNAME("nonStandardAddress") SEQ, 0, 2, 2, SKIP, 0, - _NonStandardParameter}, -}; - -static field_t _AliasAddress[] = { /* CHOICE */ - {FNAME("dialedDigits") NUMDGT, 7, 1, 0, SKIP, 0, NULL}, - {FNAME("h323-ID") BMPSTR, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("url-ID") IA5STR, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("transportID") CHOICE, 3, 7, 7, SKIP | EXT, 0, NULL}, - {FNAME("email-ID") IA5STR, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("partyNumber") CHOICE, 3, 5, 5, SKIP | EXT, 0, NULL}, - {FNAME("mobileUIM") CHOICE, 1, 2, 2, SKIP | EXT, 0, NULL}, -}; - -static field_t _Setup_UUIE_sourceAddress[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _VendorIdentifier[] = { /* SEQUENCE */ - {FNAME("vendor") SEQ, 0, 3, 3, SKIP | EXT, 0, _H221NonStandard}, - {FNAME("productId") OCTSTR, BYTE, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("versionId") OCTSTR, BYTE, 1, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _GatekeeperInfo[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, -}; - -static field_t _H310Caps[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H320Caps[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H321Caps[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H322Caps[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H323Caps[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H324Caps[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _VoiceCaps[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _T120OnlyCaps[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _SupportedProtocols[] = { /* CHOICE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP, 0, - _NonStandardParameter}, - {FNAME("h310") SEQ, 1, 1, 3, SKIP | EXT, 0, _H310Caps}, - {FNAME("h320") SEQ, 1, 1, 3, SKIP | EXT, 0, _H320Caps}, - {FNAME("h321") SEQ, 1, 1, 3, SKIP | EXT, 0, _H321Caps}, - {FNAME("h322") SEQ, 1, 1, 3, SKIP | EXT, 0, _H322Caps}, - {FNAME("h323") SEQ, 1, 1, 3, SKIP | EXT, 0, _H323Caps}, - {FNAME("h324") SEQ, 1, 1, 3, SKIP | EXT, 0, _H324Caps}, - {FNAME("voice") SEQ, 1, 1, 3, SKIP | EXT, 0, _VoiceCaps}, - {FNAME("t120-only") SEQ, 1, 1, 3, SKIP | EXT, 0, _T120OnlyCaps}, - {FNAME("nonStandardProtocol") SEQ, 2, 3, 3, SKIP | EXT, 0, NULL}, - {FNAME("t38FaxAnnexbOnly") SEQ, 2, 5, 5, SKIP | EXT, 0, NULL}, -}; - -static field_t _GatewayInfo_protocol[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 4, 9, 11, SKIP | EXT, 0, _SupportedProtocols}, -}; - -static field_t _GatewayInfo[] = { /* SEQUENCE */ - {FNAME("protocol") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _GatewayInfo_protocol}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, -}; - -static field_t _McuInfo[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("protocol") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _TerminalInfo[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, -}; - -static field_t _EndpointType[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("vendor") SEQ, 2, 3, 3, SKIP | EXT | OPT, 0, - _VendorIdentifier}, - {FNAME("gatekeeper") SEQ, 1, 1, 1, SKIP | EXT | OPT, 0, - _GatekeeperInfo}, - {FNAME("gateway") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, _GatewayInfo}, - {FNAME("mcu") SEQ, 1, 1, 2, SKIP | EXT | OPT, 0, _McuInfo}, - {FNAME("terminal") SEQ, 1, 1, 1, SKIP | EXT | OPT, 0, _TerminalInfo}, - {FNAME("mc") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("undefinedNode") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("set") BITSTR, FIXD, 32, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedTunnelledProtocols") SEQOF, SEMI, 0, 0, SKIP | OPT, - 0, NULL}, -}; - -static field_t _Setup_UUIE_destinationAddress[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _Setup_UUIE_destExtraCallInfo[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _Setup_UUIE_destExtraCRV[] = { /* SEQUENCE OF */ - {FNAME("item") INT, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _Setup_UUIE_conferenceGoal[] = { /* CHOICE */ - {FNAME("create") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("join") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("invite") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("capability-negotiation") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("callIndependentSupplementaryService") NUL, FIXD, 0, 0, SKIP, - 0, NULL}, -}; - -static field_t _Q954Details[] = { /* SEQUENCE */ - {FNAME("conferenceCalling") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("threePartyService") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _QseriesOptions[] = { /* SEQUENCE */ - {FNAME("q932Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("q951Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("q952Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("q953Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("q955Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("q956Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("q957Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("q954Info") SEQ, 0, 2, 2, SKIP | EXT, 0, _Q954Details}, -}; - -static field_t _CallType[] = { /* CHOICE */ - {FNAME("pointToPoint") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("oneToN") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("nToOne") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("nToN") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H245_NonStandardIdentifier_h221NonStandard[] = { /* SEQUENCE */ - {FNAME("t35CountryCode") INT, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("t35Extension") INT, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("manufacturerCode") INT, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H245_NonStandardIdentifier[] = { /* CHOICE */ - {FNAME("object") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("h221NonStandard") SEQ, 0, 3, 3, SKIP, 0, - _H245_NonStandardIdentifier_h221NonStandard}, -}; - -static field_t _H245_NonStandardParameter[] = { /* SEQUENCE */ - {FNAME("nonStandardIdentifier") CHOICE, 1, 2, 2, SKIP, 0, - _H245_NonStandardIdentifier}, - {FNAME("data") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H261VideoCapability[] = { /* SEQUENCE */ - {FNAME("qcifMPI") INT, 2, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("cifMPI") INT, 2, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("temporalSpatialTradeOffCapability") BOOL, FIXD, 0, 0, SKIP, 0, - NULL}, - {FNAME("maxBitRate") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("stillImageTransmission") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("videoBadMBsCap") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H262VideoCapability[] = { /* SEQUENCE */ - {FNAME("profileAndLevel-SPatML") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("profileAndLevel-MPatLL") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("profileAndLevel-MPatML") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("profileAndLevel-MPatH-14") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("profileAndLevel-MPatHL") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("profileAndLevel-SNRatLL") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("profileAndLevel-SNRatML") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("profileAndLevel-SpatialatH-14") BOOL, FIXD, 0, 0, SKIP, 0, - NULL}, - {FNAME("profileAndLevel-HPatML") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("profileAndLevel-HPatH-14") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("profileAndLevel-HPatHL") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("videoBitRate") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("vbvBufferSize") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("samplesPerLine") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("linesPerFrame") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("framesPerSecond") INT, 4, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("luminanceSampleRate") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("videoBadMBsCap") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H263VideoCapability[] = { /* SEQUENCE */ - {FNAME("sqcifMPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("qcifMPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("cifMPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("cif4MPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("cif16MPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("maxBitRate") INT, CONS, 1, 0, SKIP, 0, NULL}, - {FNAME("unrestrictedVector") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("arithmeticCoding") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("advancedPrediction") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("pbFrames") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("temporalSpatialTradeOffCapability") BOOL, FIXD, 0, 0, SKIP, 0, - NULL}, - {FNAME("hrd-B") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("bppMaxKb") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("slowSqcifMPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("slowQcifMPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("slowCifMPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("slowCif4MPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("slowCif16MPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("errorCompensation") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("enhancementLayerInfo") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("h263Options") SEQ, 5, 29, 31, SKIP | EXT | OPT, 0, NULL}, -}; - -static field_t _IS11172VideoCapability[] = { /* SEQUENCE */ - {FNAME("constrainedBitstream") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("videoBitRate") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("vbvBufferSize") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("samplesPerLine") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("linesPerFrame") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("pictureRate") INT, 4, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("luminanceSampleRate") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("videoBadMBsCap") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _VideoCapability[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, - _H245_NonStandardParameter}, - {FNAME("h261VideoCapability") SEQ, 2, 5, 6, SKIP | EXT, 0, - _H261VideoCapability}, - {FNAME("h262VideoCapability") SEQ, 6, 17, 18, SKIP | EXT, 0, - _H262VideoCapability}, - {FNAME("h263VideoCapability") SEQ, 7, 13, 21, SKIP | EXT, 0, - _H263VideoCapability}, - {FNAME("is11172VideoCapability") SEQ, 6, 7, 8, SKIP | EXT, 0, - _IS11172VideoCapability}, - {FNAME("genericVideoCapability") SEQ, 5, 6, 6, SKIP | EXT, 0, NULL}, -}; - -static field_t _AudioCapability_g7231[] = { /* SEQUENCE */ - {FNAME("maxAl-sduAudioFrames") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("silenceSuppression") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _IS11172AudioCapability[] = { /* SEQUENCE */ - {FNAME("audioLayer1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioLayer2") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioLayer3") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioSampling32k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioSampling44k1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioSampling48k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("singleChannel") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("twoChannels") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("bitRate") INT, WORD, 1, 0, SKIP, 0, NULL}, -}; - -static field_t _IS13818AudioCapability[] = { /* SEQUENCE */ - {FNAME("audioLayer1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioLayer2") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioLayer3") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioSampling16k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioSampling22k05") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioSampling24k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioSampling32k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioSampling44k1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioSampling48k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("singleChannel") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("twoChannels") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("threeChannels2-1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("threeChannels3-0") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("fourChannels2-0-2-0") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("fourChannels2-2") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("fourChannels3-1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("fiveChannels3-0-2-0") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("fiveChannels3-2") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("lowFrequencyEnhancement") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("multilingual") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("bitRate") INT, WORD, 1, 0, SKIP, 0, NULL}, -}; - -static field_t _AudioCapability[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, - _H245_NonStandardParameter}, - {FNAME("g711Alaw64k") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g711Alaw56k") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g711Ulaw64k") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g711Ulaw56k") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g722-64k") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g722-56k") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g722-48k") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g7231") SEQ, 0, 2, 2, SKIP, 0, _AudioCapability_g7231}, - {FNAME("g728") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g729") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g729AnnexA") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("is11172AudioCapability") SEQ, 0, 9, 9, SKIP | EXT, 0, - _IS11172AudioCapability}, - {FNAME("is13818AudioCapability") SEQ, 0, 21, 21, SKIP | EXT, 0, - _IS13818AudioCapability}, - {FNAME("g729wAnnexB") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g729AnnexAwAnnexB") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g7231AnnexCCapability") SEQ, 1, 3, 3, SKIP | EXT, 0, NULL}, - {FNAME("gsmFullRate") SEQ, 0, 3, 3, SKIP | EXT, 0, NULL}, - {FNAME("gsmHalfRate") SEQ, 0, 3, 3, SKIP | EXT, 0, NULL}, - {FNAME("gsmEnhancedFullRate") SEQ, 0, 3, 3, SKIP | EXT, 0, NULL}, - {FNAME("genericAudioCapability") SEQ, 5, 6, 6, SKIP | EXT, 0, NULL}, - {FNAME("g729Extensions") SEQ, 1, 8, 8, SKIP | EXT, 0, NULL}, -}; - -static field_t _DataProtocolCapability[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, - _H245_NonStandardParameter}, - {FNAME("v14buffered") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("v42lapm") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("hdlcFrameTunnelling") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("h310SeparateVCStack") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("h310SingleVCStack") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("transparent") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("segmentationAndReassembly") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("hdlcFrameTunnelingwSAR") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("v120") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("separateLANStack") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("v76wCompression") CHOICE, 2, 3, 3, SKIP | EXT, 0, NULL}, - {FNAME("tcp") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("udp") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _T84Profile_t84Restricted[] = { /* SEQUENCE */ - {FNAME("qcif") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("cif") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("ccir601Seq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("ccir601Prog") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("hdtvSeq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("hdtvProg") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("g3FacsMH200x100") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("g3FacsMH200x200") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("g4FacsMMR200x100") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("g4FacsMMR200x200") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("jbig200x200Seq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("jbig200x200Prog") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("jbig300x300Seq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("jbig300x300Prog") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("digPhotoLow") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("digPhotoMedSeq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("digPhotoMedProg") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("digPhotoHighSeq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("digPhotoHighProg") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _T84Profile[] = { /* CHOICE */ - {FNAME("t84Unrestricted") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("t84Restricted") SEQ, 0, 19, 19, SKIP | EXT, 0, - _T84Profile_t84Restricted}, -}; - -static field_t _DataApplicationCapability_application_t84[] = { /* SEQUENCE */ - {FNAME("t84Protocol") CHOICE, 3, 7, 14, SKIP | EXT, 0, - _DataProtocolCapability}, - {FNAME("t84Profile") CHOICE, 1, 2, 2, SKIP, 0, _T84Profile}, -}; - -static field_t _DataApplicationCapability_application_nlpid[] = { /* SEQUENCE */ - {FNAME("nlpidProtocol") CHOICE, 3, 7, 14, SKIP | EXT, 0, - _DataProtocolCapability}, - {FNAME("nlpidData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _DataApplicationCapability_application[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, - _H245_NonStandardParameter}, - {FNAME("t120") CHOICE, 3, 7, 14, DECODE | EXT, - offsetof(DataApplicationCapability_application, t120), - _DataProtocolCapability}, - {FNAME("dsm-cc") CHOICE, 3, 7, 14, SKIP | EXT, 0, - _DataProtocolCapability}, - {FNAME("userData") CHOICE, 3, 7, 14, SKIP | EXT, 0, - _DataProtocolCapability}, - {FNAME("t84") SEQ, 0, 2, 2, SKIP, 0, - _DataApplicationCapability_application_t84}, - {FNAME("t434") CHOICE, 3, 7, 14, SKIP | EXT, 0, - _DataProtocolCapability}, - {FNAME("h224") CHOICE, 3, 7, 14, SKIP | EXT, 0, - _DataProtocolCapability}, - {FNAME("nlpid") SEQ, 0, 2, 2, SKIP, 0, - _DataApplicationCapability_application_nlpid}, - {FNAME("dsvdControl") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("h222DataPartitioning") CHOICE, 3, 7, 14, SKIP | EXT, 0, - _DataProtocolCapability}, - {FNAME("t30fax") CHOICE, 3, 7, 14, SKIP | EXT, 0, NULL}, - {FNAME("t140") CHOICE, 3, 7, 14, SKIP | EXT, 0, NULL}, - {FNAME("t38fax") SEQ, 0, 2, 2, SKIP, 0, NULL}, - {FNAME("genericDataCapability") SEQ, 5, 6, 6, SKIP | EXT, 0, NULL}, -}; - -static field_t _DataApplicationCapability[] = { /* SEQUENCE */ - {FNAME("application") CHOICE, 4, 10, 14, DECODE | EXT, - offsetof(DataApplicationCapability, application), - _DataApplicationCapability_application}, - {FNAME("maxBitRate") INT, CONS, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _EncryptionMode[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, - _H245_NonStandardParameter}, - {FNAME("h233Encryption") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _DataType[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, - _H245_NonStandardParameter}, - {FNAME("nullData") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("videoData") CHOICE, 3, 5, 6, SKIP | EXT, 0, _VideoCapability}, - {FNAME("audioData") CHOICE, 4, 14, 22, SKIP | EXT, 0, - _AudioCapability}, - {FNAME("data") SEQ, 0, 2, 2, DECODE | EXT, offsetof(DataType, data), - _DataApplicationCapability}, - {FNAME("encryptionData") CHOICE, 1, 2, 2, SKIP | EXT, 0, - _EncryptionMode}, - {FNAME("h235Control") SEQ, 0, 2, 2, SKIP, 0, NULL}, - {FNAME("h235Media") SEQ, 0, 2, 2, SKIP | EXT, 0, NULL}, - {FNAME("multiplexedStream") SEQ, 0, 2, 2, SKIP | EXT, 0, NULL}, -}; - -static field_t _H222LogicalChannelParameters[] = { /* SEQUENCE */ - {FNAME("resourceID") INT, WORD, 0, 0, SKIP, 0, NULL}, - {FNAME("subChannelID") INT, WORD, 0, 0, SKIP, 0, NULL}, - {FNAME("pcr-pid") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("programDescriptors") OCTSTR, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("streamDescriptors") OCTSTR, SEMI, 0, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _H223LogicalChannelParameters_adaptationLayerType_al3[] = { /* SEQUENCE */ - {FNAME("controlFieldOctets") INT, 2, 0, 0, SKIP, 0, NULL}, - {FNAME("sendBufferSize") INT, CONS, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H223LogicalChannelParameters_adaptationLayerType[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, - _H245_NonStandardParameter}, - {FNAME("al1Framed") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("al1NotFramed") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("al2WithoutSequenceNumbers") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("al2WithSequenceNumbers") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("al3") SEQ, 0, 2, 2, SKIP, 0, - _H223LogicalChannelParameters_adaptationLayerType_al3}, - {FNAME("al1M") SEQ, 0, 7, 8, SKIP | EXT, 0, NULL}, - {FNAME("al2M") SEQ, 0, 2, 2, SKIP | EXT, 0, NULL}, - {FNAME("al3M") SEQ, 0, 5, 6, SKIP | EXT, 0, NULL}, -}; - -static field_t _H223LogicalChannelParameters[] = { /* SEQUENCE */ - {FNAME("adaptationLayerType") CHOICE, 3, 6, 9, SKIP | EXT, 0, - _H223LogicalChannelParameters_adaptationLayerType}, - {FNAME("segmentableFlag") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CRCLength[] = { /* CHOICE */ - {FNAME("crc8bit") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("crc16bit") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("crc32bit") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _V76HDLCParameters[] = { /* SEQUENCE */ - {FNAME("crcLength") CHOICE, 2, 3, 3, SKIP | EXT, 0, _CRCLength}, - {FNAME("n401") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("loopbackTestProcedure") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _V76LogicalChannelParameters_suspendResume[] = { /* CHOICE */ - {FNAME("noSuspendResume") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("suspendResumewAddress") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("suspendResumewoAddress") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _V76LogicalChannelParameters_mode_eRM_recovery[] = { /* CHOICE */ - {FNAME("rej") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("sREJ") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("mSREJ") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _V76LogicalChannelParameters_mode_eRM[] = { /* SEQUENCE */ - {FNAME("windowSize") INT, 7, 1, 0, SKIP, 0, NULL}, - {FNAME("recovery") CHOICE, 2, 3, 3, SKIP | EXT, 0, - _V76LogicalChannelParameters_mode_eRM_recovery}, -}; - -static field_t _V76LogicalChannelParameters_mode[] = { /* CHOICE */ - {FNAME("eRM") SEQ, 0, 2, 2, SKIP | EXT, 0, - _V76LogicalChannelParameters_mode_eRM}, - {FNAME("uNERM") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _V75Parameters[] = { /* SEQUENCE */ - {FNAME("audioHeaderPresent") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _V76LogicalChannelParameters[] = { /* SEQUENCE */ - {FNAME("hdlcParameters") SEQ, 0, 3, 3, SKIP | EXT, 0, - _V76HDLCParameters}, - {FNAME("suspendResume") CHOICE, 2, 3, 3, SKIP | EXT, 0, - _V76LogicalChannelParameters_suspendResume}, - {FNAME("uIH") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("mode") CHOICE, 1, 2, 2, SKIP | EXT, 0, - _V76LogicalChannelParameters_mode}, - {FNAME("v75Parameters") SEQ, 0, 1, 1, SKIP | EXT, 0, _V75Parameters}, -}; - -static field_t _H2250LogicalChannelParameters_nonStandard[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 0, 2, 2, SKIP, 0, _H245_NonStandardParameter}, -}; - -static field_t _UnicastAddress_iPAddress[] = { /* SEQUENCE */ - {FNAME("network") OCTSTR, FIXD, 4, 0, DECODE, - offsetof(UnicastAddress_iPAddress, network), NULL}, - {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _UnicastAddress_iPXAddress[] = { /* SEQUENCE */ - {FNAME("node") OCTSTR, FIXD, 6, 0, SKIP, 0, NULL}, - {FNAME("netnum") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, - {FNAME("tsapIdentifier") OCTSTR, FIXD, 2, 0, SKIP, 0, NULL}, -}; - -static field_t _UnicastAddress_iP6Address[] = { /* SEQUENCE */ - {FNAME("network") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, - {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _UnicastAddress_iPSourceRouteAddress_routing[] = { /* CHOICE */ - {FNAME("strict") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("loose") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _UnicastAddress_iPSourceRouteAddress_route[] = { /* SEQUENCE OF */ - {FNAME("item") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, -}; - -static field_t _UnicastAddress_iPSourceRouteAddress[] = { /* SEQUENCE */ - {FNAME("routing") CHOICE, 1, 2, 2, SKIP, 0, - _UnicastAddress_iPSourceRouteAddress_routing}, - {FNAME("network") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, - {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, - {FNAME("route") SEQOF, SEMI, 0, 0, SKIP, 0, - _UnicastAddress_iPSourceRouteAddress_route}, -}; - -static field_t _UnicastAddress[] = { /* CHOICE */ - {FNAME("iPAddress") SEQ, 0, 2, 2, DECODE | EXT, - offsetof(UnicastAddress, iPAddress), _UnicastAddress_iPAddress}, - {FNAME("iPXAddress") SEQ, 0, 3, 3, SKIP | EXT, 0, - _UnicastAddress_iPXAddress}, - {FNAME("iP6Address") SEQ, 0, 2, 2, SKIP | EXT, 0, - _UnicastAddress_iP6Address}, - {FNAME("netBios") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, - {FNAME("iPSourceRouteAddress") SEQ, 0, 4, 4, SKIP | EXT, 0, - _UnicastAddress_iPSourceRouteAddress}, - {FNAME("nsap") OCTSTR, 5, 1, 0, SKIP, 0, NULL}, - {FNAME("nonStandardAddress") SEQ, 0, 2, 2, SKIP, 0, NULL}, -}; - -static field_t _MulticastAddress_iPAddress[] = { /* SEQUENCE */ - {FNAME("network") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, - {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _MulticastAddress_iP6Address[] = { /* SEQUENCE */ - {FNAME("network") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, - {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _MulticastAddress[] = { /* CHOICE */ - {FNAME("iPAddress") SEQ, 0, 2, 2, SKIP | EXT, 0, - _MulticastAddress_iPAddress}, - {FNAME("iP6Address") SEQ, 0, 2, 2, SKIP | EXT, 0, - _MulticastAddress_iP6Address}, - {FNAME("nsap") OCTSTR, 5, 1, 0, SKIP, 0, NULL}, - {FNAME("nonStandardAddress") SEQ, 0, 2, 2, SKIP, 0, NULL}, -}; - -static field_t _H245_TransportAddress[] = { /* CHOICE */ - {FNAME("unicastAddress") CHOICE, 3, 5, 7, DECODE | EXT, - offsetof(H245_TransportAddress, unicastAddress), _UnicastAddress}, - {FNAME("multicastAddress") CHOICE, 1, 2, 4, SKIP | EXT, 0, - _MulticastAddress}, -}; - -static field_t _H2250LogicalChannelParameters[] = { /* SEQUENCE */ - {FNAME("nonStandard") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _H2250LogicalChannelParameters_nonStandard}, - {FNAME("sessionID") INT, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("associatedSessionID") INT, 8, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("mediaChannel") CHOICE, 1, 2, 2, DECODE | EXT | OPT, - offsetof(H2250LogicalChannelParameters, mediaChannel), - _H245_TransportAddress}, - {FNAME("mediaGuaranteedDelivery") BOOL, FIXD, 0, 0, SKIP | OPT, 0, - NULL}, - {FNAME("mediaControlChannel") CHOICE, 1, 2, 2, DECODE | EXT | OPT, - offsetof(H2250LogicalChannelParameters, mediaControlChannel), - _H245_TransportAddress}, - {FNAME("mediaControlGuaranteedDelivery") BOOL, FIXD, 0, 0, STOP | OPT, - 0, NULL}, - {FNAME("silenceSuppression") BOOL, FIXD, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("destination") SEQ, 0, 2, 2, STOP | EXT | OPT, 0, NULL}, - {FNAME("dynamicRTPPayloadType") INT, 5, 96, 0, STOP | OPT, 0, NULL}, - {FNAME("mediaPacketization") CHOICE, 0, 1, 2, STOP | EXT | OPT, 0, - NULL}, - {FNAME("transportCapability") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, - NULL}, - {FNAME("redundancyEncoding") SEQ, 1, 2, 2, STOP | EXT | OPT, 0, NULL}, - {FNAME("source") SEQ, 0, 2, 2, SKIP | EXT | OPT, 0, NULL}, -}; - -static field_t _OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters[] = { /* CHOICE */ - {FNAME("h222LogicalChannelParameters") SEQ, 3, 5, 5, SKIP | EXT, 0, - _H222LogicalChannelParameters}, - {FNAME("h223LogicalChannelParameters") SEQ, 0, 2, 2, SKIP | EXT, 0, - _H223LogicalChannelParameters}, - {FNAME("v76LogicalChannelParameters") SEQ, 0, 5, 5, SKIP | EXT, 0, - _V76LogicalChannelParameters}, - {FNAME("h2250LogicalChannelParameters") SEQ, 10, 11, 14, DECODE | EXT, - offsetof - (OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters, - h2250LogicalChannelParameters), _H2250LogicalChannelParameters}, - {FNAME("none") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _OpenLogicalChannel_forwardLogicalChannelParameters[] = { /* SEQUENCE */ - {FNAME("portNumber") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("dataType") CHOICE, 3, 6, 9, DECODE | EXT, - offsetof(OpenLogicalChannel_forwardLogicalChannelParameters, - dataType), _DataType}, - {FNAME("multiplexParameters") CHOICE, 2, 3, 5, DECODE | EXT, - offsetof(OpenLogicalChannel_forwardLogicalChannelParameters, - multiplexParameters), - _OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters}, - {FNAME("forwardLogicalChannelDependency") INT, WORD, 1, 0, SKIP | OPT, - 0, NULL}, - {FNAME("replacementFor") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters[] = { /* CHOICE */ - {FNAME("h223LogicalChannelParameters") SEQ, 0, 2, 2, SKIP | EXT, 0, - _H223LogicalChannelParameters}, - {FNAME("v76LogicalChannelParameters") SEQ, 0, 5, 5, SKIP | EXT, 0, - _V76LogicalChannelParameters}, - {FNAME("h2250LogicalChannelParameters") SEQ, 10, 11, 14, DECODE | EXT, - offsetof - (OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters, - h2250LogicalChannelParameters), _H2250LogicalChannelParameters}, -}; - -static field_t _OpenLogicalChannel_reverseLogicalChannelParameters[] = { /* SEQUENCE */ - {FNAME("dataType") CHOICE, 3, 6, 9, SKIP | EXT, 0, _DataType}, - {FNAME("multiplexParameters") CHOICE, 1, 2, 3, DECODE | EXT | OPT, - offsetof(OpenLogicalChannel_reverseLogicalChannelParameters, - multiplexParameters), - _OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters}, - {FNAME("reverseLogicalChannelDependency") INT, WORD, 1, 0, SKIP | OPT, - 0, NULL}, - {FNAME("replacementFor") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _NetworkAccessParameters_distribution[] = { /* CHOICE */ - {FNAME("unicast") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("multicast") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _Q2931Address_address[] = { /* CHOICE */ - {FNAME("internationalNumber") NUMSTR, 4, 1, 0, SKIP, 0, NULL}, - {FNAME("nsapAddress") OCTSTR, 5, 1, 0, SKIP, 0, NULL}, -}; - -static field_t _Q2931Address[] = { /* SEQUENCE */ - {FNAME("address") CHOICE, 1, 2, 2, SKIP | EXT, 0, - _Q2931Address_address}, - {FNAME("subaddress") OCTSTR, 5, 1, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _NetworkAccessParameters_networkAddress[] = { /* CHOICE */ - {FNAME("q2931Address") SEQ, 1, 2, 2, SKIP | EXT, 0, _Q2931Address}, - {FNAME("e164Address") NUMDGT, 7, 1, 0, SKIP, 0, NULL}, - {FNAME("localAreaAddress") CHOICE, 1, 2, 2, DECODE | EXT, - offsetof(NetworkAccessParameters_networkAddress, localAreaAddress), - _H245_TransportAddress}, -}; - -static field_t _NetworkAccessParameters[] = { /* SEQUENCE */ - {FNAME("distribution") CHOICE, 1, 2, 2, SKIP | EXT | OPT, 0, - _NetworkAccessParameters_distribution}, - {FNAME("networkAddress") CHOICE, 2, 3, 3, DECODE | EXT, - offsetof(NetworkAccessParameters, networkAddress), - _NetworkAccessParameters_networkAddress}, - {FNAME("associateConference") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("externalReference") OCTSTR, 8, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("t120SetupProcedure") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, - NULL}, -}; - -static field_t _OpenLogicalChannel[] = { /* SEQUENCE */ - {FNAME("forwardLogicalChannelNumber") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("forwardLogicalChannelParameters") SEQ, 1, 3, 5, DECODE | EXT, - offsetof(OpenLogicalChannel, forwardLogicalChannelParameters), - _OpenLogicalChannel_forwardLogicalChannelParameters}, - {FNAME("reverseLogicalChannelParameters") SEQ, 1, 2, 4, - DECODE | EXT | OPT, offsetof(OpenLogicalChannel, - reverseLogicalChannelParameters), - _OpenLogicalChannel_reverseLogicalChannelParameters}, - {FNAME("separateStack") SEQ, 2, 4, 5, DECODE | EXT | OPT, - offsetof(OpenLogicalChannel, separateStack), - _NetworkAccessParameters}, - {FNAME("encryptionSync") SEQ, 2, 4, 4, STOP | EXT | OPT, 0, NULL}, -}; - -static field_t _Setup_UUIE_fastStart[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, - sizeof(OpenLogicalChannel), _OpenLogicalChannel} - , -}; - -static field_t _Setup_UUIE[] = { /* SEQUENCE */ - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(Setup_UUIE, h245Address), _TransportAddress}, - {FNAME("sourceAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _Setup_UUIE_sourceAddress}, - {FNAME("sourceInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, _EndpointType}, - {FNAME("destinationAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _Setup_UUIE_destinationAddress}, - {FNAME("destCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(Setup_UUIE, destCallSignalAddress), _TransportAddress}, - {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _Setup_UUIE_destExtraCallInfo}, - {FNAME("destExtraCRV") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _Setup_UUIE_destExtraCRV}, - {FNAME("activeMC") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("conferenceID") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, - {FNAME("conferenceGoal") CHOICE, 2, 3, 5, SKIP | EXT, 0, - _Setup_UUIE_conferenceGoal}, - {FNAME("callServices") SEQ, 0, 8, 8, SKIP | EXT | OPT, 0, - _QseriesOptions}, - {FNAME("callType") CHOICE, 2, 4, 4, SKIP | EXT, 0, _CallType}, - {FNAME("sourceCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(Setup_UUIE, sourceCallSignalAddress), _TransportAddress}, - {FNAME("remoteExtensionAddress") CHOICE, 1, 2, 7, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, - {FNAME("h245SecurityCapability") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, - offsetof(Setup_UUIE, fastStart), _Setup_UUIE_fastStart}, - {FNAME("mediaWaitForConnect") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("canOverlapSend") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, - {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("connectionParameters") SEQ, 0, 3, 3, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("language") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("presentationIndicator") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("screeningIndicator") ENUM, 2, 0, 0, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("symmetricOperationRequired") NUL, FIXD, 0, 0, SKIP | OPT, 0, - NULL}, - {FNAME("capacity") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, NULL}, - {FNAME("circuitInfo") SEQ, 3, 3, 3, SKIP | EXT | OPT, 0, NULL}, - {FNAME("desiredProtocols") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("neededFeatures") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("desiredFeatures") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedFeatures") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("parallelH245Control") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("additionalSourceAddresses") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - NULL}, -}; - -static field_t _CallProceeding_UUIE_fastStart[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, - sizeof(OpenLogicalChannel), _OpenLogicalChannel} - , -}; - -static field_t _CallProceeding_UUIE[] = { /* SEQUENCE */ - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, - _EndpointType}, - {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(CallProceeding_UUIE, h245Address), _TransportAddress}, - {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, - {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, - offsetof(CallProceeding_UUIE, fastStart), - _CallProceeding_UUIE_fastStart}, - {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, -}; - -static field_t _Connect_UUIE_fastStart[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, - sizeof(OpenLogicalChannel), _OpenLogicalChannel} - , -}; - -static field_t _Connect_UUIE[] = { /* SEQUENCE */ - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(Connect_UUIE, h245Address), _TransportAddress}, - {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, - _EndpointType}, - {FNAME("conferenceID") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, - {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, - {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, - offsetof(Connect_UUIE, fastStart), _Connect_UUIE_fastStart}, - {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("language") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("connectedAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("presentationIndicator") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("screeningIndicator") ENUM, 2, 0, 0, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("capacity") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, -}; - -static field_t _Alerting_UUIE_fastStart[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, - sizeof(OpenLogicalChannel), _OpenLogicalChannel} - , -}; - -static field_t _Alerting_UUIE[] = { /* SEQUENCE */ - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, - _EndpointType}, - {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(Alerting_UUIE, h245Address), _TransportAddress}, - {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, - {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, - offsetof(Alerting_UUIE, fastStart), _Alerting_UUIE_fastStart}, - {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("alertingAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("presentationIndicator") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("screeningIndicator") ENUM, 2, 0, 0, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("capacity") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, -}; - -static field_t _Information_UUIE_fastStart[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, - sizeof(OpenLogicalChannel), _OpenLogicalChannel} - , -}; - -static field_t _Information_UUIE[] = { /* SEQUENCE */ - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, - offsetof(Information_UUIE, fastStart), _Information_UUIE_fastStart}, - {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("circuitInfo") SEQ, 3, 3, 3, SKIP | EXT | OPT, 0, NULL}, -}; - -static field_t _ReleaseCompleteReason[] = { /* CHOICE */ - {FNAME("noBandwidth") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("gatekeeperResources") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("unreachableDestination") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("destinationRejection") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("invalidRevision") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("noPermission") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("unreachableGatekeeper") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("gatewayResources") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("badFormatAddress") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("adaptiveBusy") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("inConf") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("undefinedReason") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("facilityCallDeflection") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("securityDenied") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("calledPartyNotRegistered") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("callerNotRegistered") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("newConnectionNeeded") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("nonStandardReason") SEQ, 0, 2, 2, SKIP, 0, NULL}, - {FNAME("replaceWithConferenceInvite") OCTSTR, FIXD, 16, 0, SKIP, 0, - NULL}, - {FNAME("genericDataReason") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("neededFeatureNotSupported") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("tunnelledSignallingRejected") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _ReleaseComplete_UUIE[] = { /* SEQUENCE */ - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("reason") CHOICE, 4, 12, 22, SKIP | EXT | OPT, 0, - _ReleaseCompleteReason}, - {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("busyAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("presentationIndicator") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("screeningIndicator") ENUM, 2, 0, 0, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("capacity") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, NULL}, - {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, -}; - -static field_t _Facility_UUIE_alternativeAliasAddress[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _FacilityReason[] = { /* CHOICE */ - {FNAME("routeCallToGatekeeper") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("callForwarded") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("routeCallToMC") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("undefinedReason") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("conferenceListChoice") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("startH245") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("noH245") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("newTokens") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("featureSetUpdate") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("forwardedElements") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("transportedInformation") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _Facility_UUIE_fastStart[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, - sizeof(OpenLogicalChannel), _OpenLogicalChannel} - , -}; - -static field_t _Facility_UUIE[] = { /* SEQUENCE */ - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("alternativeAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(Facility_UUIE, alternativeAddress), _TransportAddress}, - {FNAME("alternativeAliasAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _Facility_UUIE_alternativeAliasAddress}, - {FNAME("conferenceID") OCTSTR, FIXD, 16, 0, SKIP | OPT, 0, NULL}, - {FNAME("reason") CHOICE, 2, 4, 11, DECODE | EXT, - offsetof(Facility_UUIE, reason), _FacilityReason}, - {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, - {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("remoteExtensionAddress") CHOICE, 1, 2, 7, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("conferences") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(Facility_UUIE, h245Address), _TransportAddress}, - {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, - offsetof(Facility_UUIE, fastStart), _Facility_UUIE_fastStart}, - {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("circuitInfo") SEQ, 3, 3, 3, SKIP | EXT | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, - {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT | OPT, 0, NULL}, - {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, - NULL}, -}; - -static field_t _CallIdentifier[] = { /* SEQUENCE */ - {FNAME("guid") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, -}; - -static field_t _SecurityServiceMode[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, _NonStandardParameter}, - {FNAME("none") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("default") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _SecurityCapabilities[] = { /* SEQUENCE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("encryption") CHOICE, 2, 3, 3, SKIP | EXT, 0, - _SecurityServiceMode}, - {FNAME("authenticaton") CHOICE, 2, 3, 3, SKIP | EXT, 0, - _SecurityServiceMode}, - {FNAME("integrity") CHOICE, 2, 3, 3, SKIP | EXT, 0, - _SecurityServiceMode}, -}; - -static field_t _H245Security[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, _NonStandardParameter}, - {FNAME("noSecurity") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("tls") SEQ, 1, 4, 4, SKIP | EXT, 0, _SecurityCapabilities}, - {FNAME("ipsec") SEQ, 1, 4, 4, SKIP | EXT, 0, _SecurityCapabilities}, -}; - -static field_t _DHset[] = { /* SEQUENCE */ - {FNAME("halfkey") BITSTR, WORD, 0, 0, SKIP, 0, NULL}, - {FNAME("modSize") BITSTR, WORD, 0, 0, SKIP, 0, NULL}, - {FNAME("generator") BITSTR, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _TypedCertificate[] = { /* SEQUENCE */ - {FNAME("type") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("certificate") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H235_NonStandardParameter[] = { /* SEQUENCE */ - {FNAME("nonStandardIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("data") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _ClearToken[] = { /* SEQUENCE */ - {FNAME("tokenOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("timeStamp") INT, CONS, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("password") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("dhkey") SEQ, 0, 3, 3, SKIP | EXT | OPT, 0, _DHset}, - {FNAME("challenge") OCTSTR, 7, 8, 0, SKIP | OPT, 0, NULL}, - {FNAME("random") INT, UNCO, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("certificate") SEQ, 0, 2, 2, SKIP | EXT | OPT, 0, - _TypedCertificate}, - {FNAME("generalID") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP | OPT, 0, - _H235_NonStandardParameter}, - {FNAME("eckasdhkey") CHOICE, 1, 2, 2, SKIP | EXT | OPT, 0, NULL}, - {FNAME("sendersID") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _Progress_UUIE_tokens[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 8, 9, 11, SKIP | EXT, 0, _ClearToken}, -}; - -static field_t _Params[] = { /* SEQUENCE */ - {FNAME("ranInt") INT, UNCO, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("iv8") OCTSTR, FIXD, 8, 0, SKIP | OPT, 0, NULL}, - {FNAME("iv16") OCTSTR, FIXD, 16, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _CryptoH323Token_cryptoEPPwdHash_token[] = { /* SEQUENCE */ - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("hash") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoH323Token_cryptoEPPwdHash[] = { /* SEQUENCE */ - {FNAME("alias") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, - {FNAME("timeStamp") INT, CONS, 1, 0, SKIP, 0, NULL}, - {FNAME("token") SEQ, 0, 3, 3, SKIP, 0, - _CryptoH323Token_cryptoEPPwdHash_token}, -}; - -static field_t _CryptoH323Token_cryptoGKPwdHash_token[] = { /* SEQUENCE */ - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("hash") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoH323Token_cryptoGKPwdHash[] = { /* SEQUENCE */ - {FNAME("gatekeeperId") BMPSTR, 7, 1, 0, SKIP, 0, NULL}, - {FNAME("timeStamp") INT, CONS, 1, 0, SKIP, 0, NULL}, - {FNAME("token") SEQ, 0, 3, 3, SKIP, 0, - _CryptoH323Token_cryptoGKPwdHash_token}, -}; - -static field_t _CryptoH323Token_cryptoEPPwdEncr[] = { /* SEQUENCE */ - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("encryptedData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoH323Token_cryptoGKPwdEncr[] = { /* SEQUENCE */ - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("encryptedData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoH323Token_cryptoEPCert[] = { /* SEQUENCE */ - {FNAME("toBeSigned") SEQ, 8, 9, 11, SKIP | OPEN | EXT, 0, NULL}, - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("signature") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoH323Token_cryptoGKCert[] = { /* SEQUENCE */ - {FNAME("toBeSigned") SEQ, 8, 9, 11, SKIP | OPEN | EXT, 0, NULL}, - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("signature") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoH323Token_cryptoFastStart[] = { /* SEQUENCE */ - {FNAME("toBeSigned") SEQ, 8, 9, 11, SKIP | OPEN | EXT, 0, NULL}, - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("signature") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoToken_cryptoEncryptedToken_token[] = { /* SEQUENCE */ - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("encryptedData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoToken_cryptoEncryptedToken[] = { /* SEQUENCE */ - {FNAME("tokenOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("token") SEQ, 0, 3, 3, SKIP, 0, - _CryptoToken_cryptoEncryptedToken_token}, -}; - -static field_t _CryptoToken_cryptoSignedToken_token[] = { /* SEQUENCE */ - {FNAME("toBeSigned") SEQ, 8, 9, 11, SKIP | OPEN | EXT, 0, NULL}, - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("signature") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoToken_cryptoSignedToken[] = { /* SEQUENCE */ - {FNAME("tokenOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("token") SEQ, 0, 4, 4, SKIP, 0, - _CryptoToken_cryptoSignedToken_token}, -}; - -static field_t _CryptoToken_cryptoHashedToken_token[] = { /* SEQUENCE */ - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("hash") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoToken_cryptoHashedToken[] = { /* SEQUENCE */ - {FNAME("tokenOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("hashedVals") SEQ, 8, 9, 11, SKIP | EXT, 0, _ClearToken}, - {FNAME("token") SEQ, 0, 3, 3, SKIP, 0, - _CryptoToken_cryptoHashedToken_token}, -}; - -static field_t _CryptoToken_cryptoPwdEncr[] = { /* SEQUENCE */ - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("encryptedData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoToken[] = { /* CHOICE */ - {FNAME("cryptoEncryptedToken") SEQ, 0, 2, 2, SKIP, 0, - _CryptoToken_cryptoEncryptedToken}, - {FNAME("cryptoSignedToken") SEQ, 0, 2, 2, SKIP, 0, - _CryptoToken_cryptoSignedToken}, - {FNAME("cryptoHashedToken") SEQ, 0, 3, 3, SKIP, 0, - _CryptoToken_cryptoHashedToken}, - {FNAME("cryptoPwdEncr") SEQ, 0, 3, 3, SKIP, 0, - _CryptoToken_cryptoPwdEncr}, -}; - -static field_t _CryptoH323Token[] = { /* CHOICE */ - {FNAME("cryptoEPPwdHash") SEQ, 0, 3, 3, SKIP, 0, - _CryptoH323Token_cryptoEPPwdHash}, - {FNAME("cryptoGKPwdHash") SEQ, 0, 3, 3, SKIP, 0, - _CryptoH323Token_cryptoGKPwdHash}, - {FNAME("cryptoEPPwdEncr") SEQ, 0, 3, 3, SKIP, 0, - _CryptoH323Token_cryptoEPPwdEncr}, - {FNAME("cryptoGKPwdEncr") SEQ, 0, 3, 3, SKIP, 0, - _CryptoH323Token_cryptoGKPwdEncr}, - {FNAME("cryptoEPCert") SEQ, 0, 4, 4, SKIP, 0, - _CryptoH323Token_cryptoEPCert}, - {FNAME("cryptoGKCert") SEQ, 0, 4, 4, SKIP, 0, - _CryptoH323Token_cryptoGKCert}, - {FNAME("cryptoFastStart") SEQ, 0, 4, 4, SKIP, 0, - _CryptoH323Token_cryptoFastStart}, - {FNAME("nestedcryptoToken") CHOICE, 2, 4, 4, SKIP | EXT, 0, - _CryptoToken}, -}; - -static field_t _Progress_UUIE_cryptoTokens[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 3, 8, 8, SKIP | EXT, 0, _CryptoH323Token}, -}; - -static field_t _Progress_UUIE_fastStart[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, - sizeof(OpenLogicalChannel), _OpenLogicalChannel} - , -}; - -static field_t _Progress_UUIE[] = { /* SEQUENCE */ - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, - _EndpointType}, - {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(Progress_UUIE, h245Address), _TransportAddress}, - {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, - _CallIdentifier}, - {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, - _H245Security}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _Progress_UUIE_tokens}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _Progress_UUIE_cryptoTokens}, - {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, - offsetof(Progress_UUIE, fastStart), _Progress_UUIE_fastStart}, - {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _H323_UU_PDU_h323_message_body[] = { /* CHOICE */ - {FNAME("setup") SEQ, 7, 13, 39, DECODE | EXT, - offsetof(H323_UU_PDU_h323_message_body, setup), _Setup_UUIE}, - {FNAME("callProceeding") SEQ, 1, 3, 12, DECODE | EXT, - offsetof(H323_UU_PDU_h323_message_body, callProceeding), - _CallProceeding_UUIE}, - {FNAME("connect") SEQ, 1, 4, 19, DECODE | EXT, - offsetof(H323_UU_PDU_h323_message_body, connect), _Connect_UUIE}, - {FNAME("alerting") SEQ, 1, 3, 17, DECODE | EXT, - offsetof(H323_UU_PDU_h323_message_body, alerting), _Alerting_UUIE}, - {FNAME("information") SEQ, 0, 1, 7, DECODE | EXT, - offsetof(H323_UU_PDU_h323_message_body, information), - _Information_UUIE}, - {FNAME("releaseComplete") SEQ, 1, 2, 11, SKIP | EXT, 0, - _ReleaseComplete_UUIE}, - {FNAME("facility") SEQ, 3, 5, 21, DECODE | EXT, - offsetof(H323_UU_PDU_h323_message_body, facility), _Facility_UUIE}, - {FNAME("progress") SEQ, 5, 8, 11, DECODE | EXT, - offsetof(H323_UU_PDU_h323_message_body, progress), _Progress_UUIE}, - {FNAME("empty") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("status") SEQ, 2, 4, 4, SKIP | EXT, 0, NULL}, - {FNAME("statusInquiry") SEQ, 2, 4, 4, SKIP | EXT, 0, NULL}, - {FNAME("setupAcknowledge") SEQ, 2, 4, 4, SKIP | EXT, 0, NULL}, - {FNAME("notify") SEQ, 2, 4, 4, SKIP | EXT, 0, NULL}, -}; - -static field_t _RequestMessage[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("masterSlaveDetermination") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, - {FNAME("terminalCapabilitySet") SEQ, 3, 5, 5, STOP | EXT, 0, NULL}, - {FNAME("openLogicalChannel") SEQ, 1, 3, 5, DECODE | EXT, - offsetof(RequestMessage, openLogicalChannel), _OpenLogicalChannel}, - {FNAME("closeLogicalChannel") SEQ, 0, 2, 3, STOP | EXT, 0, NULL}, - {FNAME("requestChannelClose") SEQ, 0, 1, 3, STOP | EXT, 0, NULL}, - {FNAME("multiplexEntrySend") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, - {FNAME("requestMultiplexEntry") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("requestMode") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, - {FNAME("roundTripDelayRequest") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("maintenanceLoopRequest") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("communicationModeRequest") SEQ, 0, 0, 0, STOP | EXT, 0, NULL}, - {FNAME("conferenceRequest") CHOICE, 3, 8, 16, STOP | EXT, 0, NULL}, - {FNAME("multilinkRequest") CHOICE, 3, 5, 5, STOP | EXT, 0, NULL}, - {FNAME("logicalChannelRateRequest") SEQ, 0, 3, 3, STOP | EXT, 0, - NULL}, -}; - -static field_t _OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters[] = { /* CHOICE */ - {FNAME("h222LogicalChannelParameters") SEQ, 3, 5, 5, SKIP | EXT, 0, - _H222LogicalChannelParameters}, - {FNAME("h2250LogicalChannelParameters") SEQ, 10, 11, 14, DECODE | EXT, - offsetof - (OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters, - h2250LogicalChannelParameters), _H2250LogicalChannelParameters}, -}; - -static field_t _OpenLogicalChannelAck_reverseLogicalChannelParameters[] = { /* SEQUENCE */ - {FNAME("reverseLogicalChannelNumber") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("portNumber") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("multiplexParameters") CHOICE, 0, 1, 2, DECODE | EXT | OPT, - offsetof(OpenLogicalChannelAck_reverseLogicalChannelParameters, - multiplexParameters), - _OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters}, - {FNAME("replacementFor") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _H2250LogicalChannelAckParameters_nonStandard[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 0, 2, 2, SKIP, 0, _H245_NonStandardParameter}, -}; - -static field_t _H2250LogicalChannelAckParameters[] = { /* SEQUENCE */ - {FNAME("nonStandard") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _H2250LogicalChannelAckParameters_nonStandard}, - {FNAME("sessionID") INT, 8, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("mediaChannel") CHOICE, 1, 2, 2, DECODE | EXT | OPT, - offsetof(H2250LogicalChannelAckParameters, mediaChannel), - _H245_TransportAddress}, - {FNAME("mediaControlChannel") CHOICE, 1, 2, 2, DECODE | EXT | OPT, - offsetof(H2250LogicalChannelAckParameters, mediaControlChannel), - _H245_TransportAddress}, - {FNAME("dynamicRTPPayloadType") INT, 5, 96, 0, SKIP | OPT, 0, NULL}, - {FNAME("flowControlToZero") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("portNumber") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _OpenLogicalChannelAck_forwardMultiplexAckParameters[] = { /* CHOICE */ - {FNAME("h2250LogicalChannelAckParameters") SEQ, 5, 5, 7, DECODE | EXT, - offsetof(OpenLogicalChannelAck_forwardMultiplexAckParameters, - h2250LogicalChannelAckParameters), - _H2250LogicalChannelAckParameters}, -}; - -static field_t _OpenLogicalChannelAck[] = { /* SEQUENCE */ - {FNAME("forwardLogicalChannelNumber") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("reverseLogicalChannelParameters") SEQ, 2, 3, 4, - DECODE | EXT | OPT, offsetof(OpenLogicalChannelAck, - reverseLogicalChannelParameters), - _OpenLogicalChannelAck_reverseLogicalChannelParameters}, - {FNAME("separateStack") SEQ, 2, 4, 5, SKIP | EXT | OPT, 0, NULL}, - {FNAME("forwardMultiplexAckParameters") CHOICE, 0, 1, 1, - DECODE | EXT | OPT, offsetof(OpenLogicalChannelAck, - forwardMultiplexAckParameters), - _OpenLogicalChannelAck_forwardMultiplexAckParameters}, - {FNAME("encryptionSync") SEQ, 2, 4, 4, STOP | EXT | OPT, 0, NULL}, -}; - -static field_t _ResponseMessage[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("masterSlaveDeterminationAck") SEQ, 0, 1, 1, STOP | EXT, 0, - NULL}, - {FNAME("masterSlaveDeterminationReject") SEQ, 0, 1, 1, STOP | EXT, 0, - NULL}, - {FNAME("terminalCapabilitySetAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("terminalCapabilitySetReject") SEQ, 0, 2, 2, STOP | EXT, 0, - NULL}, - {FNAME("openLogicalChannelAck") SEQ, 1, 2, 5, DECODE | EXT, - offsetof(ResponseMessage, openLogicalChannelAck), - _OpenLogicalChannelAck}, - {FNAME("openLogicalChannelReject") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, - {FNAME("closeLogicalChannelAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("requestChannelCloseAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("requestChannelCloseReject") SEQ, 0, 2, 2, STOP | EXT, 0, - NULL}, - {FNAME("multiplexEntrySendAck") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, - {FNAME("multiplexEntrySendReject") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, - {FNAME("requestMultiplexEntryAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("requestMultiplexEntryReject") SEQ, 0, 2, 2, STOP | EXT, 0, - NULL}, - {FNAME("requestModeAck") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, - {FNAME("requestModeReject") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, - {FNAME("roundTripDelayResponse") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("maintenanceLoopAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("maintenanceLoopReject") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, - {FNAME("communicationModeResponse") CHOICE, 0, 1, 1, STOP | EXT, 0, - NULL}, - {FNAME("conferenceResponse") CHOICE, 3, 8, 16, STOP | EXT, 0, NULL}, - {FNAME("multilinkResponse") CHOICE, 3, 5, 5, STOP | EXT, 0, NULL}, - {FNAME("logicalChannelRateAcknowledge") SEQ, 0, 3, 3, STOP | EXT, 0, - NULL}, - {FNAME("logicalChannelRateReject") SEQ, 1, 4, 4, STOP | EXT, 0, NULL}, -}; - -static field_t _MultimediaSystemControlMessage[] = { /* CHOICE */ - {FNAME("request") CHOICE, 4, 11, 15, DECODE | EXT, - offsetof(MultimediaSystemControlMessage, request), _RequestMessage}, - {FNAME("response") CHOICE, 5, 19, 24, DECODE | EXT, - offsetof(MultimediaSystemControlMessage, response), - _ResponseMessage}, - {FNAME("command") CHOICE, 3, 7, 12, STOP | EXT, 0, NULL}, - {FNAME("indication") CHOICE, 4, 14, 23, STOP | EXT, 0, NULL}, -}; - -static field_t _H323_UU_PDU_h245Control[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 2, 4, 4, DECODE | OPEN | EXT, - sizeof(MultimediaSystemControlMessage), - _MultimediaSystemControlMessage} - , -}; - -static field_t _H323_UU_PDU[] = { /* SEQUENCE */ - {FNAME("h323-message-body") CHOICE, 3, 7, 13, DECODE | EXT, - offsetof(H323_UU_PDU, h323_message_body), - _H323_UU_PDU_h323_message_body}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("h4501SupplementaryService") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - NULL}, - {FNAME("h245Tunneling") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("h245Control") SEQOF, SEMI, 0, 4, DECODE | OPT, - offsetof(H323_UU_PDU, h245Control), _H323_UU_PDU_h245Control}, - {FNAME("nonStandardControl") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("callLinkage") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, - {FNAME("tunnelledSignallingMessage") SEQ, 2, 4, 4, STOP | EXT | OPT, - 0, NULL}, - {FNAME("provisionalRespToH245Tunneling") NUL, FIXD, 0, 0, STOP | OPT, - 0, NULL}, - {FNAME("stimulusControl") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _H323_UserInformation[] = { /* SEQUENCE */ - {FNAME("h323-uu-pdu") SEQ, 1, 2, 11, DECODE | EXT, - offsetof(H323_UserInformation, h323_uu_pdu), _H323_UU_PDU}, - {FNAME("user-data") SEQ, 0, 2, 2, STOP | EXT | OPT, 0, NULL}, -}; - -static field_t _GatekeeperRequest[] = { /* SEQUENCE */ - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("rasAddress") CHOICE, 3, 7, 7, DECODE | EXT, - offsetof(GatekeeperRequest, rasAddress), _TransportAddress}, - {FNAME("endpointType") SEQ, 6, 8, 10, STOP | EXT, 0, NULL}, - {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, - {FNAME("callServices") SEQ, 0, 8, 8, STOP | EXT | OPT, 0, NULL}, - {FNAME("endpointAlias") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("authenticationCapability") SEQOF, SEMI, 0, 0, STOP | OPT, 0, - NULL}, - {FNAME("algorithmOIDs") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrity") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("supportsAltGK") NUL, FIXD, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _GatekeeperConfirm[] = { /* SEQUENCE */ - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("rasAddress") CHOICE, 3, 7, 7, DECODE | EXT, - offsetof(GatekeeperConfirm, rasAddress), _TransportAddress}, - {FNAME("alternateGatekeeper") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("authenticationMode") CHOICE, 3, 7, 8, STOP | EXT | OPT, 0, - NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("algorithmOID") OID, BYTE, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrity") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _RegistrationRequest_callSignalAddress[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, - sizeof(TransportAddress), _TransportAddress} - , -}; - -static field_t _RegistrationRequest_rasAddress[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, - sizeof(TransportAddress), _TransportAddress} - , -}; - -static field_t _RegistrationRequest_terminalAlias[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _RegistrationRequest[] = { /* SEQUENCE */ - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("discoveryComplete") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("callSignalAddress") SEQOF, SEMI, 0, 10, DECODE, - offsetof(RegistrationRequest, callSignalAddress), - _RegistrationRequest_callSignalAddress}, - {FNAME("rasAddress") SEQOF, SEMI, 0, 10, DECODE, - offsetof(RegistrationRequest, rasAddress), - _RegistrationRequest_rasAddress}, - {FNAME("terminalType") SEQ, 6, 8, 10, SKIP | EXT, 0, _EndpointType}, - {FNAME("terminalAlias") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _RegistrationRequest_terminalAlias}, - {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("endpointVendor") SEQ, 2, 3, 3, SKIP | EXT, 0, - _VendorIdentifier}, - {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("timeToLive") INT, CONS, 1, 0, DECODE | OPT, - offsetof(RegistrationRequest, timeToLive), NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("keepAlive") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, - {FNAME("willSupplyUUIEs") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("maintainConnection") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("alternateTransportAddresses") SEQ, 1, 1, 1, STOP | EXT | OPT, - 0, NULL}, - {FNAME("additiveRegistration") NUL, FIXD, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("terminalAliasPattern") SEQOF, SEMI, 0, 0, STOP | OPT, 0, - NULL}, - {FNAME("supportsAltGK") NUL, FIXD, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("usageReportingCapability") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, - NULL}, - {FNAME("multipleCalls") BOOL, FIXD, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("supportedH248Packages") SEQOF, SEMI, 0, 0, STOP | OPT, 0, - NULL}, - {FNAME("callCreditCapability") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, - NULL}, - {FNAME("capacityReportingCapability") SEQ, 0, 1, 1, STOP | EXT | OPT, - 0, NULL}, - {FNAME("capacity") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _RegistrationConfirm_callSignalAddress[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, - sizeof(TransportAddress), _TransportAddress} - , -}; - -static field_t _RegistrationConfirm_terminalAlias[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _RegistrationConfirm[] = { /* SEQUENCE */ - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("callSignalAddress") SEQOF, SEMI, 0, 10, DECODE, - offsetof(RegistrationConfirm, callSignalAddress), - _RegistrationConfirm_callSignalAddress}, - {FNAME("terminalAlias") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _RegistrationConfirm_terminalAlias}, - {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, SKIP, 0, NULL}, - {FNAME("alternateGatekeeper") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("timeToLive") INT, CONS, 1, 0, DECODE | OPT, - offsetof(RegistrationConfirm, timeToLive), NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("willRespondToIRR") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("preGrantedARQ") SEQ, 0, 4, 8, STOP | EXT | OPT, 0, NULL}, - {FNAME("maintainConnection") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("serviceControl") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("supportsAdditiveRegistration") NUL, FIXD, 0, 0, STOP | OPT, 0, - NULL}, - {FNAME("terminalAliasPattern") SEQOF, SEMI, 0, 0, STOP | OPT, 0, - NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("usageSpec") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("featureServerAlias") CHOICE, 1, 2, 7, STOP | EXT | OPT, 0, - NULL}, - {FNAME("capacityReportingSpec") SEQ, 0, 1, 1, STOP | EXT | OPT, 0, - NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _UnregistrationRequest_callSignalAddress[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, - sizeof(TransportAddress), _TransportAddress} - , -}; - -static field_t _UnregistrationRequest[] = { /* SEQUENCE */ - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("callSignalAddress") SEQOF, SEMI, 0, 10, DECODE, - offsetof(UnregistrationRequest, callSignalAddress), - _UnregistrationRequest_callSignalAddress}, - {FNAME("endpointAlias") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, - {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("reason") CHOICE, 2, 4, 5, STOP | EXT | OPT, 0, NULL}, - {FNAME("endpointAliasPattern") SEQOF, SEMI, 0, 0, STOP | OPT, 0, - NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("alternateGatekeeper") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _CallModel[] = { /* CHOICE */ - {FNAME("direct") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("gatekeeperRouted") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _AdmissionRequest_destinationInfo[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _AdmissionRequest_destExtraCallInfo[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _AdmissionRequest_srcInfo[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _AdmissionRequest[] = { /* SEQUENCE */ - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("callType") CHOICE, 2, 4, 4, SKIP | EXT, 0, _CallType}, - {FNAME("callModel") CHOICE, 1, 2, 2, SKIP | EXT | OPT, 0, _CallModel}, - {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, SKIP, 0, NULL}, - {FNAME("destinationInfo") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _AdmissionRequest_destinationInfo}, - {FNAME("destCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(AdmissionRequest, destCallSignalAddress), - _TransportAddress}, - {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _AdmissionRequest_destExtraCallInfo}, - {FNAME("srcInfo") SEQOF, SEMI, 0, 0, SKIP, 0, - _AdmissionRequest_srcInfo}, - {FNAME("srcCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(AdmissionRequest, srcCallSignalAddress), _TransportAddress}, - {FNAME("bandWidth") INT, CONS, 0, 0, STOP, 0, NULL}, - {FNAME("callReferenceValue") INT, WORD, 0, 0, STOP, 0, NULL}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("callServices") SEQ, 0, 8, 8, STOP | EXT | OPT, 0, NULL}, - {FNAME("conferenceID") OCTSTR, FIXD, 16, 0, STOP, 0, NULL}, - {FNAME("activeMC") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("answerCall") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("canMapAlias") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("callIdentifier") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("srcAlternatives") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("destAlternatives") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("transportQOS") CHOICE, 2, 3, 3, STOP | EXT | OPT, 0, NULL}, - {FNAME("willSupplyUUIEs") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("callLinkage") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, - {FNAME("gatewayDataRate") SEQ, 2, 3, 3, STOP | EXT | OPT, 0, NULL}, - {FNAME("capacity") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, - {FNAME("circuitInfo") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, - {FNAME("desiredProtocols") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("desiredTunnelledProtocol") SEQ, 1, 2, 2, STOP | EXT | OPT, 0, - NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _AdmissionConfirm[] = { /* SEQUENCE */ - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("bandWidth") INT, CONS, 0, 0, SKIP, 0, NULL}, - {FNAME("callModel") CHOICE, 1, 2, 2, SKIP | EXT, 0, _CallModel}, - {FNAME("destCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT, - offsetof(AdmissionConfirm, destCallSignalAddress), - _TransportAddress}, - {FNAME("irrFrequency") INT, WORD, 1, 0, STOP | OPT, 0, NULL}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("destinationInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("destinationType") SEQ, 6, 8, 10, STOP | EXT | OPT, 0, NULL}, - {FNAME("remoteExtensionAddress") SEQOF, SEMI, 0, 0, STOP | OPT, 0, - NULL}, - {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("transportQOS") CHOICE, 2, 3, 3, STOP | EXT | OPT, 0, NULL}, - {FNAME("willRespondToIRR") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("uuiesRequested") SEQ, 0, 9, 13, STOP | EXT, 0, NULL}, - {FNAME("language") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("alternateTransportAddresses") SEQ, 1, 1, 1, STOP | EXT | OPT, - 0, NULL}, - {FNAME("useSpecifiedTransport") CHOICE, 1, 2, 2, STOP | EXT | OPT, 0, - NULL}, - {FNAME("circuitInfo") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, - {FNAME("usageSpec") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("supportedProtocols") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("serviceControl") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("multipleCalls") BOOL, FIXD, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _LocationRequest_destinationInfo[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _LocationRequest[] = { /* SEQUENCE */ - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("destinationInfo") SEQOF, SEMI, 0, 0, SKIP, 0, - _LocationRequest_destinationInfo}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("replyAddress") CHOICE, 3, 7, 7, DECODE | EXT, - offsetof(LocationRequest, replyAddress), _TransportAddress}, - {FNAME("sourceInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("canMapAlias") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("desiredProtocols") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("desiredTunnelledProtocol") SEQ, 1, 2, 2, STOP | EXT | OPT, 0, - NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("hopCount") INT, 8, 1, 0, STOP | OPT, 0, NULL}, - {FNAME("circuitInfo") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, -}; - -static field_t _LocationConfirm[] = { /* SEQUENCE */ - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("callSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT, - offsetof(LocationConfirm, callSignalAddress), _TransportAddress}, - {FNAME("rasAddress") CHOICE, 3, 7, 7, DECODE | EXT, - offsetof(LocationConfirm, rasAddress), _TransportAddress}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("destinationInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("destinationType") SEQ, 6, 8, 10, STOP | EXT | OPT, 0, NULL}, - {FNAME("remoteExtensionAddress") SEQOF, SEMI, 0, 0, STOP | OPT, 0, - NULL}, - {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("alternateTransportAddresses") SEQ, 1, 1, 1, STOP | EXT | OPT, - 0, NULL}, - {FNAME("supportedProtocols") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("multipleCalls") BOOL, FIXD, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("circuitInfo") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, - {FNAME("serviceControl") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _InfoRequestResponse_callSignalAddress[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, - sizeof(TransportAddress), _TransportAddress} - , -}; - -static field_t _InfoRequestResponse[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("endpointType") SEQ, 6, 8, 10, SKIP | EXT, 0, _EndpointType}, - {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, SKIP, 0, NULL}, - {FNAME("rasAddress") CHOICE, 3, 7, 7, DECODE | EXT, - offsetof(InfoRequestResponse, rasAddress), _TransportAddress}, - {FNAME("callSignalAddress") SEQOF, SEMI, 0, 10, DECODE, - offsetof(InfoRequestResponse, callSignalAddress), - _InfoRequestResponse_callSignalAddress}, - {FNAME("endpointAlias") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("perCallInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("needResponse") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("capacity") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, - {FNAME("irrStatus") CHOICE, 2, 4, 4, STOP | EXT | OPT, 0, NULL}, - {FNAME("unsolicited") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _RasMessage[] = { /* CHOICE */ - {FNAME("gatekeeperRequest") SEQ, 4, 8, 18, DECODE | EXT, - offsetof(RasMessage, gatekeeperRequest), _GatekeeperRequest}, - {FNAME("gatekeeperConfirm") SEQ, 2, 5, 14, DECODE | EXT, - offsetof(RasMessage, gatekeeperConfirm), _GatekeeperConfirm}, - {FNAME("gatekeeperReject") SEQ, 2, 5, 11, STOP | EXT, 0, NULL}, - {FNAME("registrationRequest") SEQ, 3, 10, 31, DECODE | EXT, - offsetof(RasMessage, registrationRequest), _RegistrationRequest}, - {FNAME("registrationConfirm") SEQ, 3, 7, 24, DECODE | EXT, - offsetof(RasMessage, registrationConfirm), _RegistrationConfirm}, - {FNAME("registrationReject") SEQ, 2, 5, 11, STOP | EXT, 0, NULL}, - {FNAME("unregistrationRequest") SEQ, 3, 5, 15, DECODE | EXT, - offsetof(RasMessage, unregistrationRequest), _UnregistrationRequest}, - {FNAME("unregistrationConfirm") SEQ, 1, 2, 6, STOP | EXT, 0, NULL}, - {FNAME("unregistrationReject") SEQ, 1, 3, 8, STOP | EXT, 0, NULL}, - {FNAME("admissionRequest") SEQ, 7, 16, 34, DECODE | EXT, - offsetof(RasMessage, admissionRequest), _AdmissionRequest}, - {FNAME("admissionConfirm") SEQ, 2, 6, 27, DECODE | EXT, - offsetof(RasMessage, admissionConfirm), _AdmissionConfirm}, - {FNAME("admissionReject") SEQ, 1, 3, 11, STOP | EXT, 0, NULL}, - {FNAME("bandwidthRequest") SEQ, 2, 7, 18, STOP | EXT, 0, NULL}, - {FNAME("bandwidthConfirm") SEQ, 1, 3, 8, STOP | EXT, 0, NULL}, - {FNAME("bandwidthReject") SEQ, 1, 4, 9, STOP | EXT, 0, NULL}, - {FNAME("disengageRequest") SEQ, 1, 6, 19, STOP | EXT, 0, NULL}, - {FNAME("disengageConfirm") SEQ, 1, 2, 9, STOP | EXT, 0, NULL}, - {FNAME("disengageReject") SEQ, 1, 3, 8, STOP | EXT, 0, NULL}, - {FNAME("locationRequest") SEQ, 2, 5, 17, DECODE | EXT, - offsetof(RasMessage, locationRequest), _LocationRequest}, - {FNAME("locationConfirm") SEQ, 1, 4, 19, DECODE | EXT, - offsetof(RasMessage, locationConfirm), _LocationConfirm}, - {FNAME("locationReject") SEQ, 1, 3, 10, STOP | EXT, 0, NULL}, - {FNAME("infoRequest") SEQ, 2, 4, 15, STOP | EXT, 0, NULL}, - {FNAME("infoRequestResponse") SEQ, 3, 8, 16, DECODE | EXT, - offsetof(RasMessage, infoRequestResponse), _InfoRequestResponse}, - {FNAME("nonStandardMessage") SEQ, 0, 2, 7, STOP | EXT, 0, NULL}, - {FNAME("unknownMessageResponse") SEQ, 0, 1, 5, STOP | EXT, 0, NULL}, - {FNAME("requestInProgress") SEQ, 4, 6, 6, STOP | EXT, 0, NULL}, - {FNAME("resourcesAvailableIndicate") SEQ, 4, 9, 11, STOP | EXT, 0, - NULL}, - {FNAME("resourcesAvailableConfirm") SEQ, 4, 6, 7, STOP | EXT, 0, - NULL}, - {FNAME("infoRequestAck") SEQ, 4, 5, 5, STOP | EXT, 0, NULL}, - {FNAME("infoRequestNak") SEQ, 5, 7, 7, STOP | EXT, 0, NULL}, - {FNAME("serviceControlIndication") SEQ, 8, 10, 10, STOP | EXT, 0, - NULL}, - {FNAME("serviceControlResponse") SEQ, 7, 8, 8, STOP | EXT, 0, NULL}, -}; diff --git a/net/ipv4/netfilter/nf_nat_h323.c b/net/ipv4/netfilter/nf_nat_h323.c new file mode 100644 index 00000000000..fb9ab0114c2 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_h323.c @@ -0,0 +1,596 @@ +/* + * H.323 extension for NAT alteration. + * + * Copyright (c) 2006 Jing Min Zhao + * + * This source code is licensed under General Public License version 2. + * + * Based on the 'brute force' H.323 NAT module by + * Jozsef Kadlecsik + */ + +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +/****************************************************************************/ +static int set_addr(struct sk_buff **pskb, + unsigned char **data, int dataoff, + unsigned int addroff, __be32 ip, __be16 port) +{ + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = ip_conntrack_get(*pskb, &ctinfo); + struct { + __be32 ip; + __be16 port; + } __attribute__ ((__packed__)) buf; + struct tcphdr _tcph, *th; + + buf.ip = ip; + buf.port = port; + addroff += dataoff; + + if ((*pskb)->nh.iph->protocol == IPPROTO_TCP) { + if (!nf_nat_mangle_tcp_packet(pskb, ct, ctinfo, + addroff, sizeof(buf), + (char *) &buf, sizeof(buf))) { + if (net_ratelimit()) + printk("nf_nat_h323: nf_nat_mangle_tcp_packet" + " error\n"); + return -1; + } + + /* Relocate data pointer */ + th = skb_header_pointer(*pskb, (*pskb)->nh.iph->ihl * 4, + sizeof(_tcph), &_tcph); + if (th == NULL) + return -1; + *data = (*pskb)->data + (*pskb)->nh.iph->ihl * 4 + + th->doff * 4 + dataoff; + } else { + if (!nf_nat_mangle_udp_packet(pskb, ct, ctinfo, + addroff, sizeof(buf), + (char *) &buf, sizeof(buf))) { + if (net_ratelimit()) + printk("nf_nat_h323: nf_nat_mangle_udp_packet" + " error\n"); + return -1; + } + /* nf_nat_mangle_udp_packet uses skb_make_writable() to copy + * or pull everything in a linear buffer, so we can safely + * use the skb pointers now */ + *data = (*pskb)->data + (*pskb)->nh.iph->ihl * 4 + + sizeof(struct udphdr); + } + + return 0; +} + +/****************************************************************************/ +static int set_h225_addr(struct sk_buff **pskb, + unsigned char **data, int dataoff, + TransportAddress *taddr, + union nf_conntrack_address *addr, __be16 port) +{ + return set_addr(pskb, data, dataoff, taddr->ipAddress.ip, + addr->ip, port); +} + +/****************************************************************************/ +static int set_h245_addr(struct sk_buff **pskb, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr, + union nf_conntrack_address *addr, __be16 port) +{ + return set_addr(pskb, data, dataoff, + taddr->unicastAddress.iPAddress.network, + addr->ip, port); +} + +/****************************************************************************/ +static int set_sig_addr(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, + TransportAddress *taddr, int count) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int dir = CTINFO2DIR(ctinfo); + int i; + __be16 port; + union nf_conntrack_address addr; + + for (i = 0; i < count; i++) { + if (get_h225_addr(ct, *data, &taddr[i], &addr, &port)) { + if (addr.ip == ct->tuplehash[dir].tuple.src.u3.ip && + port == info->sig_port[dir]) { + /* GW->GK */ + + /* Fix for Gnomemeeting */ + if (i > 0 && + get_h225_addr(ct, *data, &taddr[0], + &addr, &port) && + (ntohl(addr.ip) & 0xff000000) == 0x7f000000) + i = 0; + + DEBUGP + ("nf_nat_ras: set signal address " + "%u.%u.%u.%u:%hu->%u.%u.%u.%u:%hu\n", + NIPQUAD(ip), port, + NIPQUAD(ct->tuplehash[!dir].tuple.dst. + ip), info->sig_port[!dir]); + return set_h225_addr(pskb, data, 0, &taddr[i], + &ct->tuplehash[!dir]. + tuple.dst.u3, + info->sig_port[!dir]); + } else if (addr.ip == ct->tuplehash[dir].tuple.dst.u3.ip && + port == info->sig_port[dir]) { + /* GK->GW */ + DEBUGP + ("nf_nat_ras: set signal address " + "%u.%u.%u.%u:%hu->%u.%u.%u.%u:%hu\n", + NIPQUAD(ip), port, + NIPQUAD(ct->tuplehash[!dir].tuple.src. + ip), info->sig_port[!dir]); + return set_h225_addr(pskb, data, 0, &taddr[i], + &ct->tuplehash[!dir]. + tuple.src.u3, + info->sig_port[!dir]); + } + } + } + + return 0; +} + +/****************************************************************************/ +static int set_ras_addr(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, + TransportAddress *taddr, int count) +{ + int dir = CTINFO2DIR(ctinfo); + int i; + __be16 port; + union nf_conntrack_address addr; + + for (i = 0; i < count; i++) { + if (get_h225_addr(ct, *data, &taddr[i], &addr, &port) && + addr.ip == ct->tuplehash[dir].tuple.src.u3.ip && + port == ct->tuplehash[dir].tuple.src.u.udp.port) { + DEBUGP("nf_nat_ras: set rasAddress " + "%u.%u.%u.%u:%hu->%u.%u.%u.%u:%hu\n", + NIPQUAD(ip), ntohs(port), + NIPQUAD(ct->tuplehash[!dir].tuple.dst.u3.ip), + ntohs(ct->tuplehash[!dir].tuple.dst.u.udp. + port)); + return set_h225_addr(pskb, data, 0, &taddr[i], + &ct->tuplehash[!dir].tuple.dst.u3, + ct->tuplehash[!dir].tuple. + dst.u.udp.port); + } + } + + return 0; +} + +/****************************************************************************/ +static int nat_rtp_rtcp(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr, + __be16 port, __be16 rtp_port, + struct nf_conntrack_expect *rtp_exp, + struct nf_conntrack_expect *rtcp_exp) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int dir = CTINFO2DIR(ctinfo); + int i; + u_int16_t nated_port; + + /* Set expectations for NAT */ + rtp_exp->saved_proto.udp.port = rtp_exp->tuple.dst.u.udp.port; + rtp_exp->expectfn = nf_nat_follow_master; + rtp_exp->dir = !dir; + rtcp_exp->saved_proto.udp.port = rtcp_exp->tuple.dst.u.udp.port; + rtcp_exp->expectfn = nf_nat_follow_master; + rtcp_exp->dir = !dir; + + /* Lookup existing expects */ + for (i = 0; i < H323_RTP_CHANNEL_MAX; i++) { + if (info->rtp_port[i][dir] == rtp_port) { + /* Expected */ + + /* Use allocated ports first. This will refresh + * the expects */ + rtp_exp->tuple.dst.u.udp.port = info->rtp_port[i][dir]; + rtcp_exp->tuple.dst.u.udp.port = + htons(ntohs(info->rtp_port[i][dir]) + 1); + break; + } else if (info->rtp_port[i][dir] == 0) { + /* Not expected */ + break; + } + } + + /* Run out of expectations */ + if (i >= H323_RTP_CHANNEL_MAX) { + if (net_ratelimit()) + printk("nf_nat_h323: out of expectations\n"); + return 0; + } + + /* Try to get a pair of ports. */ + for (nated_port = ntohs(rtp_exp->tuple.dst.u.udp.port); + nated_port != 0; nated_port += 2) { + rtp_exp->tuple.dst.u.udp.port = htons(nated_port); + if (nf_conntrack_expect_related(rtp_exp) == 0) { + rtcp_exp->tuple.dst.u.udp.port = + htons(nated_port + 1); + if (nf_conntrack_expect_related(rtcp_exp) == 0) + break; + nf_conntrack_unexpect_related(rtp_exp); + } + } + + if (nated_port == 0) { /* No port available */ + if (net_ratelimit()) + printk("nf_nat_h323: out of RTP ports\n"); + return 0; + } + + /* Modify signal */ + if (set_h245_addr(pskb, data, dataoff, taddr, + &ct->tuplehash[!dir].tuple.dst.u3, + htons((port & htons(1)) ? nated_port + 1 : + nated_port)) == 0) { + /* Save ports */ + info->rtp_port[i][dir] = rtp_port; + info->rtp_port[i][!dir] = htons(nated_port); + } else { + nf_conntrack_unexpect_related(rtp_exp); + nf_conntrack_unexpect_related(rtcp_exp); + return -1; + } + + /* Success */ + DEBUGP("nf_nat_h323: expect RTP %u.%u.%u.%u:%hu->%u.%u.%u.%u:%hu\n", + NIPQUAD(rtp_exp->tuple.src.ip), + ntohs(rtp_exp->tuple.src.u.udp.port), + NIPQUAD(rtp_exp->tuple.dst.ip), + ntohs(rtp_exp->tuple.dst.u.udp.port)); + DEBUGP("nf_nat_h323: expect RTCP %u.%u.%u.%u:%hu->%u.%u.%u.%u:%hu\n", + NIPQUAD(rtcp_exp->tuple.src.ip), + ntohs(rtcp_exp->tuple.src.u.udp.port), + NIPQUAD(rtcp_exp->tuple.dst.ip), + ntohs(rtcp_exp->tuple.dst.u.udp.port)); + + return 0; +} + +/****************************************************************************/ +static int nat_t120(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr, __be16 port, + struct nf_conntrack_expect *exp) +{ + int dir = CTINFO2DIR(ctinfo); + u_int16_t nated_port = ntohs(port); + + /* Set expectations for NAT */ + exp->saved_proto.tcp.port = exp->tuple.dst.u.tcp.port; + exp->expectfn = nf_nat_follow_master; + exp->dir = !dir; + + /* Try to get same port: if not, try to change it. */ + for (; nated_port != 0; nated_port++) { + exp->tuple.dst.u.tcp.port = htons(nated_port); + if (nf_conntrack_expect_related(exp) == 0) + break; + } + + if (nated_port == 0) { /* No port available */ + if (net_ratelimit()) + printk("nf_nat_h323: out of TCP ports\n"); + return 0; + } + + /* Modify signal */ + if (set_h245_addr(pskb, data, dataoff, taddr, + &ct->tuplehash[!dir].tuple.dst.u3, + htons(nated_port)) < 0) { + nf_conntrack_unexpect_related(exp); + return -1; + } + + DEBUGP("nf_nat_h323: expect T.120 %u.%u.%u.%u:%hu->%u.%u.%u.%u:%hu\n", + NIPQUAD(exp->tuple.src.ip), ntohs(exp->tuple.src.u.tcp.port), + NIPQUAD(exp->tuple.dst.ip), ntohs(exp->tuple.dst.u.tcp.port)); + + return 0; +} + +/****************************************************************************/ +static int nat_h245(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + TransportAddress *taddr, __be16 port, + struct nf_conntrack_expect *exp) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int dir = CTINFO2DIR(ctinfo); + u_int16_t nated_port = ntohs(port); + + /* Set expectations for NAT */ + exp->saved_proto.tcp.port = exp->tuple.dst.u.tcp.port; + exp->expectfn = nf_nat_follow_master; + exp->dir = !dir; + + /* Check existing expects */ + if (info->sig_port[dir] == port) + nated_port = ntohs(info->sig_port[!dir]); + + /* Try to get same port: if not, try to change it. */ + for (; nated_port != 0; nated_port++) { + exp->tuple.dst.u.tcp.port = htons(nated_port); + if (nf_conntrack_expect_related(exp) == 0) + break; + } + + if (nated_port == 0) { /* No port available */ + if (net_ratelimit()) + printk("nf_nat_q931: out of TCP ports\n"); + return 0; + } + + /* Modify signal */ + if (set_h225_addr(pskb, data, dataoff, taddr, + &ct->tuplehash[!dir].tuple.dst.u3, + htons(nated_port)) == 0) { + /* Save ports */ + info->sig_port[dir] = port; + info->sig_port[!dir] = htons(nated_port); + } else { + nf_conntrack_unexpect_related(exp); + return -1; + } + + DEBUGP("nf_nat_q931: expect H.245 %u.%u.%u.%u:%hu->%u.%u.%u.%u:%hu\n", + NIPQUAD(exp->tuple.src.ip), ntohs(exp->tuple.src.u.tcp.port), + NIPQUAD(exp->tuple.dst.ip), ntohs(exp->tuple.dst.u.tcp.port)); + + return 0; +} + +/**************************************************************************** + * This conntrack expect function replaces nf_conntrack_q931_expect() + * which was set by nf_conntrack_h323.c. + ****************************************************************************/ +static void ip_nat_q931_expect(struct nf_conn *new, + struct nf_conntrack_expect *this) +{ + struct ip_nat_range range; + + if (this->tuple.src.u3.ip != 0) { /* Only accept calls from GK */ + nf_nat_follow_master(new, this); + return; + } + + /* This must be a fresh one. */ + BUG_ON(new->status & IPS_NAT_DONE_MASK); + + /* Change src to where master sends to */ + range.flags = IP_NAT_RANGE_MAP_IPS; + range.min_ip = range.max_ip = new->tuplehash[!this->dir].tuple.src.u3.ip; + + /* hook doesn't matter, but it has to do source manip */ + nf_nat_setup_info(new, &range, NF_IP_POST_ROUTING); + + /* For DST manip, map port here to where it's expected. */ + range.flags = (IP_NAT_RANGE_MAP_IPS | IP_NAT_RANGE_PROTO_SPECIFIED); + range.min = range.max = this->saved_proto; + range.min_ip = range.max_ip = + new->master->tuplehash[!this->dir].tuple.src.u3.ip; + + /* hook doesn't matter, but it has to do destination manip */ + nf_nat_setup_info(new, &range, NF_IP_PRE_ROUTING); +} + +/****************************************************************************/ +static int nat_q931(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, TransportAddress *taddr, int idx, + __be16 port, struct nf_conntrack_expect *exp) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int dir = CTINFO2DIR(ctinfo); + u_int16_t nated_port = ntohs(port); + union nf_conntrack_address addr; + + /* Set expectations for NAT */ + exp->saved_proto.tcp.port = exp->tuple.dst.u.tcp.port; + exp->expectfn = ip_nat_q931_expect; + exp->dir = !dir; + + /* Check existing expects */ + if (info->sig_port[dir] == port) + nated_port = ntohs(info->sig_port[!dir]); + + /* Try to get same port: if not, try to change it. */ + for (; nated_port != 0; nated_port++) { + exp->tuple.dst.u.tcp.port = htons(nated_port); + if (nf_conntrack_expect_related(exp) == 0) + break; + } + + if (nated_port == 0) { /* No port available */ + if (net_ratelimit()) + printk("nf_nat_ras: out of TCP ports\n"); + return 0; + } + + /* Modify signal */ + if (set_h225_addr(pskb, data, 0, &taddr[idx], + &ct->tuplehash[!dir].tuple.dst.u3, + htons(nated_port)) == 0) { + /* Save ports */ + info->sig_port[dir] = port; + info->sig_port[!dir] = htons(nated_port); + + /* Fix for Gnomemeeting */ + if (idx > 0 && + get_h225_addr(ct, *data, &taddr[0], &addr, &port) && + (ntohl(addr.ip) & 0xff000000) == 0x7f000000) { + set_h225_addr_hook(pskb, data, 0, &taddr[0], + &ct->tuplehash[!dir].tuple.dst.u3, + info->sig_port[!dir]); + } + } else { + nf_conntrack_unexpect_related(exp); + return -1; + } + + /* Success */ + DEBUGP("nf_nat_ras: expect Q.931 %u.%u.%u.%u:%hu->%u.%u.%u.%u:%hu\n", + NIPQUAD(exp->tuple.src.ip), ntohs(exp->tuple.src.u.tcp.port), + NIPQUAD(exp->tuple.dst.ip), ntohs(exp->tuple.dst.u.tcp.port)); + + return 0; +} + +/****************************************************************************/ +static void ip_nat_callforwarding_expect(struct nf_conn *new, + struct nf_conntrack_expect *this) +{ + struct nf_nat_range range; + + /* This must be a fresh one. */ + BUG_ON(new->status & IPS_NAT_DONE_MASK); + + /* Change src to where master sends to */ + range.flags = IP_NAT_RANGE_MAP_IPS; + range.min_ip = range.max_ip = new->tuplehash[!this->dir].tuple.src.u3.ip; + + /* hook doesn't matter, but it has to do source manip */ + nf_nat_setup_info(new, &range, NF_IP_POST_ROUTING); + + /* For DST manip, map port here to where it's expected. */ + range.flags = (IP_NAT_RANGE_MAP_IPS | IP_NAT_RANGE_PROTO_SPECIFIED); + range.min = range.max = this->saved_proto; + range.min_ip = range.max_ip = this->saved_ip; + + /* hook doesn't matter, but it has to do destination manip */ + nf_nat_setup_info(new, &range, NF_IP_PRE_ROUTING); +} + +/****************************************************************************/ +static int nat_callforwarding(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + TransportAddress *taddr, __be16 port, + struct nf_conntrack_expect *exp) +{ + int dir = CTINFO2DIR(ctinfo); + u_int16_t nated_port; + + /* Set expectations for NAT */ + exp->saved_ip = exp->tuple.dst.u3.ip; + exp->tuple.dst.u3.ip = ct->tuplehash[!dir].tuple.dst.u3.ip; + exp->saved_proto.tcp.port = exp->tuple.dst.u.tcp.port; + exp->expectfn = ip_nat_callforwarding_expect; + exp->dir = !dir; + + /* Try to get same port: if not, try to change it. */ + for (nated_port = ntohs(port); nated_port != 0; nated_port++) { + exp->tuple.dst.u.tcp.port = htons(nated_port); + if (nf_conntrack_expect_related(exp) == 0) + break; + } + + if (nated_port == 0) { /* No port available */ + if (net_ratelimit()) + printk("nf_nat_q931: out of TCP ports\n"); + return 0; + } + + /* Modify signal */ + if (!set_h225_addr(pskb, data, dataoff, taddr, + &ct->tuplehash[!dir].tuple.dst.u3, + htons(nated_port)) == 0) { + nf_conntrack_unexpect_related(exp); + return -1; + } + + /* Success */ + DEBUGP("nf_nat_q931: expect Call Forwarding " + "%u.%u.%u.%u:%hu->%u.%u.%u.%u:%hu\n", + NIPQUAD(exp->tuple.src.ip), ntohs(exp->tuple.src.u.tcp.port), + NIPQUAD(exp->tuple.dst.ip), ntohs(exp->tuple.dst.u.tcp.port)); + + return 0; +} + +/****************************************************************************/ +static int __init init(void) +{ + BUG_ON(rcu_dereference(set_h245_addr_hook) != NULL); + BUG_ON(rcu_dereference(set_h225_addr_hook) != NULL); + BUG_ON(rcu_dereference(set_sig_addr_hook) != NULL); + BUG_ON(rcu_dereference(set_ras_addr_hook) != NULL); + BUG_ON(rcu_dereference(nat_rtp_rtcp_hook) != NULL); + BUG_ON(rcu_dereference(nat_t120_hook) != NULL); + BUG_ON(rcu_dereference(nat_h245_hook) != NULL); + BUG_ON(rcu_dereference(nat_callforwarding_hook) != NULL); + BUG_ON(rcu_dereference(nat_q931_hook) != NULL); + + rcu_assign_pointer(set_h245_addr_hook, set_h245_addr); + rcu_assign_pointer(set_h225_addr_hook, set_h225_addr); + rcu_assign_pointer(set_sig_addr_hook, set_sig_addr); + rcu_assign_pointer(set_ras_addr_hook, set_ras_addr); + rcu_assign_pointer(nat_rtp_rtcp_hook, nat_rtp_rtcp); + rcu_assign_pointer(nat_t120_hook, nat_t120); + rcu_assign_pointer(nat_h245_hook, nat_h245); + rcu_assign_pointer(nat_callforwarding_hook, nat_callforwarding); + rcu_assign_pointer(nat_q931_hook, nat_q931); + + DEBUGP("nf_nat_h323: init success\n"); + return 0; +} + +/****************************************************************************/ +static void __exit fini(void) +{ + rcu_assign_pointer(set_h245_addr_hook, NULL); + rcu_assign_pointer(set_h225_addr_hook, NULL); + rcu_assign_pointer(set_sig_addr_hook, NULL); + rcu_assign_pointer(set_ras_addr_hook, NULL); + rcu_assign_pointer(nat_rtp_rtcp_hook, NULL); + rcu_assign_pointer(nat_t120_hook, NULL); + rcu_assign_pointer(nat_h245_hook, NULL); + rcu_assign_pointer(nat_callforwarding_hook, NULL); + rcu_assign_pointer(nat_q931_hook, NULL); + synchronize_rcu(); +} + +/****************************************************************************/ +module_init(init); +module_exit(fini); + +MODULE_AUTHOR("Jing Min Zhao "); +MODULE_DESCRIPTION("H.323 NAT helper"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ip_nat_h323"); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index f85fd43b344..d8f3451c95b 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -160,6 +160,25 @@ config NF_CONNTRACK_FTP To compile it as a module, choose M here. If unsure, say N. +config NF_CONNTRACK_H323 + tristate "H.323 protocol support (EXPERIMENTAL)" + depends on EXPERIMENTAL && NF_CONNTRACK + help + H.323 is a VoIP signalling protocol from ITU-T. As one of the most + important VoIP protocols, it is widely used by voice hardware and + software including voice gateways, IP phones, Netmeeting, OpenPhone, + Gnomemeeting, etc. + + With this module you can support H.323 on a connection tracking/NAT + firewall. + + This module supports RAS, Fast Start, H.245 Tunnelling, Call + Forwarding, RTP/RTCP and T.120 based audio, video, fax, chat, + whiteboard, file transfer, etc. For more information, please + visit http://nath323.sourceforge.net/. + + To compile it as a module, choose M here. If unsure, say N. + config NF_CT_NETLINK tristate 'Connection tracking netlink interface (EXPERIMENTAL)' depends on EXPERIMENTAL && NF_CONNTRACK && NETFILTER_NETLINK diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index a5ee9381742..69b554576b6 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -20,8 +20,11 @@ obj-$(CONFIG_NF_CT_PROTO_SCTP) += nf_conntrack_proto_sctp.o obj-$(CONFIG_NF_CT_NETLINK) += nf_conntrack_netlink.o # connection tracking helpers +nf_conntrack_h323-objs := nf_conntrack_h323_main.o nf_conntrack_h323_asn1.o + obj-$(CONFIG_NF_CONNTRACK_AMANDA) += nf_conntrack_amanda.o obj-$(CONFIG_NF_CONNTRACK_FTP) += nf_conntrack_ftp.o +obj-$(CONFIG_NF_CONNTRACK_H323) += nf_conntrack_h323.o # generic X tables obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o diff --git a/net/netfilter/nf_conntrack_h323_asn1.c b/net/netfilter/nf_conntrack_h323_asn1.c new file mode 100644 index 00000000000..f6fad713d48 --- /dev/null +++ b/net/netfilter/nf_conntrack_h323_asn1.c @@ -0,0 +1,874 @@ +/**************************************************************************** + * ip_conntrack_helper_h323_asn1.c - BER and PER decoding library for H.323 + * conntrack/NAT module. + * + * Copyright (c) 2006 by Jing Min Zhao + * + * This source code is licensed under General Public License version 2. + * + * See ip_conntrack_helper_h323_asn1.h for details. + * + ****************************************************************************/ + +#ifdef __KERNEL__ +#include +#else +#include +#endif +#include + +/* Trace Flag */ +#ifndef H323_TRACE +#define H323_TRACE 0 +#endif + +#if H323_TRACE +#define TAB_SIZE 4 +#define IFTHEN(cond, act) if(cond){act;} +#ifdef __KERNEL__ +#define PRINT printk +#else +#define PRINT printf +#endif +#define FNAME(name) name, +#else +#define IFTHEN(cond, act) +#define PRINT(fmt, args...) +#define FNAME(name) +#endif + +/* ASN.1 Types */ +#define NUL 0 +#define BOOL 1 +#define OID 2 +#define INT 3 +#define ENUM 4 +#define BITSTR 5 +#define NUMSTR 6 +#define NUMDGT 6 +#define TBCDSTR 6 +#define OCTSTR 7 +#define PRTSTR 7 +#define IA5STR 7 +#define GENSTR 7 +#define BMPSTR 8 +#define SEQ 9 +#define SET 9 +#define SEQOF 10 +#define SETOF 10 +#define CHOICE 11 + +/* Constraint Types */ +#define FIXD 0 +/* #define BITS 1-8 */ +#define BYTE 9 +#define WORD 10 +#define CONS 11 +#define SEMI 12 +#define UNCO 13 + +/* ASN.1 Type Attributes */ +#define SKIP 0 +#define STOP 1 +#define DECODE 2 +#define EXT 4 +#define OPEN 8 +#define OPT 16 + + +/* ASN.1 Field Structure */ +typedef struct field_t { +#if H323_TRACE + char *name; +#endif + unsigned char type; + unsigned char sz; + unsigned char lb; + unsigned char ub; + unsigned short attr; + unsigned short offset; + struct field_t *fields; +} field_t; + +/* Bit Stream */ +typedef struct { + unsigned char *buf; + unsigned char *beg; + unsigned char *end; + unsigned char *cur; + unsigned bit; +} bitstr_t; + +/* Tool Functions */ +#define INC_BIT(bs) if((++bs->bit)>7){bs->cur++;bs->bit=0;} +#define INC_BITS(bs,b) if((bs->bit+=b)>7){bs->cur+=bs->bit>>3;bs->bit&=7;} +#define BYTE_ALIGN(bs) if(bs->bit){bs->cur++;bs->bit=0;} +#define CHECK_BOUND(bs,n) if(bs->cur+(n)>bs->end)return(H323_ERROR_BOUND) +static unsigned get_len(bitstr_t * bs); +static unsigned get_bit(bitstr_t * bs); +static unsigned get_bits(bitstr_t * bs, unsigned b); +static unsigned get_bitmap(bitstr_t * bs, unsigned b); +static unsigned get_uint(bitstr_t * bs, int b); + +/* Decoder Functions */ +static int decode_nul(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_bool(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_oid(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_int(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_enum(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_bitstr(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_numstr(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_octstr(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_bmpstr(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_seq(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_seqof(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_choice(bitstr_t * bs, field_t * f, char *base, int level); + +/* Decoder Functions Vector */ +typedef int (*decoder_t) (bitstr_t *, field_t *, char *, int); +static decoder_t Decoders[] = { + decode_nul, + decode_bool, + decode_oid, + decode_int, + decode_enum, + decode_bitstr, + decode_numstr, + decode_octstr, + decode_bmpstr, + decode_seq, + decode_seqof, + decode_choice, +}; + +/**************************************************************************** + * H.323 Types + ****************************************************************************/ +#include "nf_conntrack_h323_types.c" + +/**************************************************************************** + * Functions + ****************************************************************************/ +/* Assume bs is aligned && v < 16384 */ +unsigned get_len(bitstr_t * bs) +{ + unsigned v; + + v = *bs->cur++; + + if (v & 0x80) { + v &= 0x3f; + v <<= 8; + v += *bs->cur++; + } + + return v; +} + +/****************************************************************************/ +unsigned get_bit(bitstr_t * bs) +{ + unsigned b = (*bs->cur) & (0x80 >> bs->bit); + + INC_BIT(bs); + + return b; +} + +/****************************************************************************/ +/* Assume b <= 8 */ +unsigned get_bits(bitstr_t * bs, unsigned b) +{ + unsigned v, l; + + v = (*bs->cur) & (0xffU >> bs->bit); + l = b + bs->bit; + + if (l < 8) { + v >>= 8 - l; + bs->bit = l; + } else if (l == 8) { + bs->cur++; + bs->bit = 0; + } else { /* l > 8 */ + + v <<= 8; + v += *(++bs->cur); + v >>= 16 - l; + bs->bit = l - 8; + } + + return v; +} + +/****************************************************************************/ +/* Assume b <= 32 */ +unsigned get_bitmap(bitstr_t * bs, unsigned b) +{ + unsigned v, l, shift, bytes; + + if (!b) + return 0; + + l = bs->bit + b; + + if (l < 8) { + v = (unsigned) (*bs->cur) << (bs->bit + 24); + bs->bit = l; + } else if (l == 8) { + v = (unsigned) (*bs->cur++) << (bs->bit + 24); + bs->bit = 0; + } else { + for (bytes = l >> 3, shift = 24, v = 0; bytes; + bytes--, shift -= 8) + v |= (unsigned) (*bs->cur++) << shift; + + if (l < 32) { + v |= (unsigned) (*bs->cur) << shift; + v <<= bs->bit; + } else if (l > 32) { + v <<= bs->bit; + v |= (*bs->cur) >> (8 - bs->bit); + } + + bs->bit = l & 0x7; + } + + v &= 0xffffffff << (32 - b); + + return v; +} + +/**************************************************************************** + * Assume bs is aligned and sizeof(unsigned int) == 4 + ****************************************************************************/ +unsigned get_uint(bitstr_t * bs, int b) +{ + unsigned v = 0; + + switch (b) { + case 4: + v |= *bs->cur++; + v <<= 8; + case 3: + v |= *bs->cur++; + v <<= 8; + case 2: + v |= *bs->cur++; + v <<= 8; + case 1: + v |= *bs->cur++; + break; + } + return v; +} + +/****************************************************************************/ +int decode_nul(bitstr_t * bs, field_t * f, char *base, int level) +{ + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_bool(bitstr_t * bs, field_t * f, char *base, int level) +{ + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + INC_BIT(bs); + + CHECK_BOUND(bs, 0); + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_oid(bitstr_t * bs, field_t * f, char *base, int level) +{ + int len; + + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + BYTE_ALIGN(bs); + CHECK_BOUND(bs, 1); + len = *bs->cur++; + bs->cur += len; + + CHECK_BOUND(bs, 0); + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_int(bitstr_t * bs, field_t * f, char *base, int level) +{ + unsigned len; + + PRINT("%*.s%s", level * TAB_SIZE, " ", f->name); + + switch (f->sz) { + case BYTE: /* Range == 256 */ + BYTE_ALIGN(bs); + bs->cur++; + break; + case WORD: /* 257 <= Range <= 64K */ + BYTE_ALIGN(bs); + bs->cur += 2; + break; + case CONS: /* 64K < Range < 4G */ + len = get_bits(bs, 2) + 1; + BYTE_ALIGN(bs); + if (base && (f->attr & DECODE)) { /* timeToLive */ + unsigned v = get_uint(bs, len) + f->lb; + PRINT(" = %u", v); + *((unsigned *) (base + f->offset)) = v; + } + bs->cur += len; + break; + case UNCO: + BYTE_ALIGN(bs); + CHECK_BOUND(bs, 2); + len = get_len(bs); + bs->cur += len; + break; + default: /* 2 <= Range <= 255 */ + INC_BITS(bs, f->sz); + break; + } + + PRINT("\n"); + + CHECK_BOUND(bs, 0); + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_enum(bitstr_t * bs, field_t * f, char *base, int level) +{ + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + if ((f->attr & EXT) && get_bit(bs)) { + INC_BITS(bs, 7); + } else { + INC_BITS(bs, f->sz); + } + + CHECK_BOUND(bs, 0); + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_bitstr(bitstr_t * bs, field_t * f, char *base, int level) +{ + unsigned len; + + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + BYTE_ALIGN(bs); + switch (f->sz) { + case FIXD: /* fixed length > 16 */ + len = f->lb; + break; + case WORD: /* 2-byte length */ + CHECK_BOUND(bs, 2); + len = (*bs->cur++) << 8; + len += (*bs->cur++) + f->lb; + break; + case SEMI: + CHECK_BOUND(bs, 2); + len = get_len(bs); + break; + default: + len = 0; + break; + } + + bs->cur += len >> 3; + bs->bit = len & 7; + + CHECK_BOUND(bs, 0); + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_numstr(bitstr_t * bs, field_t * f, char *base, int level) +{ + unsigned len; + + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + /* 2 <= Range <= 255 */ + len = get_bits(bs, f->sz) + f->lb; + + BYTE_ALIGN(bs); + INC_BITS(bs, (len << 2)); + + CHECK_BOUND(bs, 0); + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_octstr(bitstr_t * bs, field_t * f, char *base, int level) +{ + unsigned len; + + PRINT("%*.s%s", level * TAB_SIZE, " ", f->name); + + switch (f->sz) { + case FIXD: /* Range == 1 */ + if (f->lb > 2) { + BYTE_ALIGN(bs); + if (base && (f->attr & DECODE)) { + /* The IP Address */ + IFTHEN(f->lb == 4, + PRINT(" = %d.%d.%d.%d:%d", + bs->cur[0], bs->cur[1], + bs->cur[2], bs->cur[3], + bs->cur[4] * 256 + bs->cur[5])); + *((unsigned *) (base + f->offset)) = + bs->cur - bs->buf; + } + } + len = f->lb; + break; + case BYTE: /* Range == 256 */ + BYTE_ALIGN(bs); + CHECK_BOUND(bs, 1); + len = (*bs->cur++) + f->lb; + break; + case SEMI: + BYTE_ALIGN(bs); + CHECK_BOUND(bs, 2); + len = get_len(bs) + f->lb; + break; + default: /* 2 <= Range <= 255 */ + len = get_bits(bs, f->sz) + f->lb; + BYTE_ALIGN(bs); + break; + } + + bs->cur += len; + + PRINT("\n"); + + CHECK_BOUND(bs, 0); + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_bmpstr(bitstr_t * bs, field_t * f, char *base, int level) +{ + unsigned len; + + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + switch (f->sz) { + case BYTE: /* Range == 256 */ + BYTE_ALIGN(bs); + CHECK_BOUND(bs, 1); + len = (*bs->cur++) + f->lb; + break; + default: /* 2 <= Range <= 255 */ + len = get_bits(bs, f->sz) + f->lb; + BYTE_ALIGN(bs); + break; + } + + bs->cur += len << 1; + + CHECK_BOUND(bs, 0); + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_seq(bitstr_t * bs, field_t * f, char *base, int level) +{ + unsigned ext, bmp, i, opt, len = 0, bmp2, bmp2_len; + int err; + field_t *son; + unsigned char *beg = NULL; + + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + /* Decode? */ + base = (base && (f->attr & DECODE)) ? base + f->offset : NULL; + + /* Extensible? */ + ext = (f->attr & EXT) ? get_bit(bs) : 0; + + /* Get fields bitmap */ + bmp = get_bitmap(bs, f->sz); + if (base) + *(unsigned *) base = bmp; + + /* Decode the root components */ + for (i = opt = 0, son = f->fields; i < f->lb; i++, son++) { + if (son->attr & STOP) { + PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", + son->name); + return H323_ERROR_STOP; + } + + if (son->attr & OPT) { /* Optional component */ + if (!((0x80000000U >> (opt++)) & bmp)) /* Not exist */ + continue; + } + + /* Decode */ + if (son->attr & OPEN) { /* Open field */ + CHECK_BOUND(bs, 2); + len = get_len(bs); + CHECK_BOUND(bs, len); + if (!base) { + PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, + " ", son->name); + bs->cur += len; + continue; + } + beg = bs->cur; + + /* Decode */ + if ((err = (Decoders[son->type]) (bs, son, base, + level + 1)) < + H323_ERROR_NONE) + return err; + + bs->cur = beg + len; + bs->bit = 0; + } else if ((err = (Decoders[son->type]) (bs, son, base, + level + 1)) < + H323_ERROR_NONE) + return err; + } + + /* No extension? */ + if (!ext) + return H323_ERROR_NONE; + + /* Get the extension bitmap */ + bmp2_len = get_bits(bs, 7) + 1; + CHECK_BOUND(bs, (bmp2_len + 7) >> 3); + bmp2 = get_bitmap(bs, bmp2_len); + bmp |= bmp2 >> f->sz; + if (base) + *(unsigned *) base = bmp; + BYTE_ALIGN(bs); + + /* Decode the extension components */ + for (opt = 0; opt < bmp2_len; opt++, i++, son++) { + if (i < f->ub && son->attr & STOP) { + PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", + son->name); + return H323_ERROR_STOP; + } + + if (!((0x80000000 >> opt) & bmp2)) /* Not present */ + continue; + + /* Check Range */ + if (i >= f->ub) { /* Newer Version? */ + CHECK_BOUND(bs, 2); + len = get_len(bs); + CHECK_BOUND(bs, len); + bs->cur += len; + continue; + } + + CHECK_BOUND(bs, 2); + len = get_len(bs); + CHECK_BOUND(bs, len); + if (!base || !(son->attr & DECODE)) { + PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", + son->name); + bs->cur += len; + continue; + } + beg = bs->cur; + + if ((err = (Decoders[son->type]) (bs, son, base, + level + 1)) < + H323_ERROR_NONE) + return err; + + bs->cur = beg + len; + bs->bit = 0; + } + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_seqof(bitstr_t * bs, field_t * f, char *base, int level) +{ + unsigned count, effective_count = 0, i, len = 0; + int err; + field_t *son; + unsigned char *beg = NULL; + + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + /* Decode? */ + base = (base && (f->attr & DECODE)) ? base + f->offset : NULL; + + /* Decode item count */ + switch (f->sz) { + case BYTE: + BYTE_ALIGN(bs); + CHECK_BOUND(bs, 1); + count = *bs->cur++; + break; + case WORD: + BYTE_ALIGN(bs); + CHECK_BOUND(bs, 2); + count = *bs->cur++; + count <<= 8; + count = *bs->cur++; + break; + case SEMI: + BYTE_ALIGN(bs); + CHECK_BOUND(bs, 2); + count = get_len(bs); + break; + default: + count = get_bits(bs, f->sz); + break; + } + count += f->lb; + + /* Write Count */ + if (base) { + effective_count = count > f->ub ? f->ub : count; + *(unsigned *) base = effective_count; + base += sizeof(unsigned); + } + + /* Decode nested field */ + son = f->fields; + if (base) + base -= son->offset; + for (i = 0; i < count; i++) { + if (son->attr & OPEN) { + BYTE_ALIGN(bs); + len = get_len(bs); + CHECK_BOUND(bs, len); + if (!base || !(son->attr & DECODE)) { + PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, + " ", son->name); + bs->cur += len; + continue; + } + beg = bs->cur; + + if ((err = (Decoders[son->type]) (bs, son, + i < + effective_count ? + base : NULL, + level + 1)) < + H323_ERROR_NONE) + return err; + + bs->cur = beg + len; + bs->bit = 0; + } else + if ((err = (Decoders[son->type]) (bs, son, + i < + effective_count ? + base : NULL, + level + 1)) < + H323_ERROR_NONE) + return err; + + if (base) + base += son->offset; + } + + return H323_ERROR_NONE; +} + + +/****************************************************************************/ +int decode_choice(bitstr_t * bs, field_t * f, char *base, int level) +{ + unsigned type, ext, len = 0; + int err; + field_t *son; + unsigned char *beg = NULL; + + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + /* Decode? */ + base = (base && (f->attr & DECODE)) ? base + f->offset : NULL; + + /* Decode the choice index number */ + if ((f->attr & EXT) && get_bit(bs)) { + ext = 1; + type = get_bits(bs, 7) + f->lb; + } else { + ext = 0; + type = get_bits(bs, f->sz); + } + + /* Write Type */ + if (base) + *(unsigned *) base = type; + + /* Check Range */ + if (type >= f->ub) { /* Newer version? */ + BYTE_ALIGN(bs); + len = get_len(bs); + CHECK_BOUND(bs, len); + bs->cur += len; + return H323_ERROR_NONE; + } + + /* Transfer to son level */ + son = &f->fields[type]; + if (son->attr & STOP) { + PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", son->name); + return H323_ERROR_STOP; + } + + if (ext || (son->attr & OPEN)) { + BYTE_ALIGN(bs); + len = get_len(bs); + CHECK_BOUND(bs, len); + if (!base || !(son->attr & DECODE)) { + PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", + son->name); + bs->cur += len; + return H323_ERROR_NONE; + } + beg = bs->cur; + + if ((err = (Decoders[son->type]) (bs, son, base, level + 1)) < + H323_ERROR_NONE) + return err; + + bs->cur = beg + len; + bs->bit = 0; + } else if ((err = (Decoders[son->type]) (bs, son, base, level + 1)) < + H323_ERROR_NONE) + return err; + + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int DecodeRasMessage(unsigned char *buf, size_t sz, RasMessage * ras) +{ + static field_t ras_message = { + FNAME("RasMessage") CHOICE, 5, 24, 32, DECODE | EXT, + 0, _RasMessage + }; + bitstr_t bs; + + bs.buf = bs.beg = bs.cur = buf; + bs.end = buf + sz; + bs.bit = 0; + + return decode_choice(&bs, &ras_message, (char *) ras, 0); +} + +/****************************************************************************/ +static int DecodeH323_UserInformation(unsigned char *buf, unsigned char *beg, + size_t sz, H323_UserInformation * uuie) +{ + static field_t h323_userinformation = { + FNAME("H323-UserInformation") SEQ, 1, 2, 2, DECODE | EXT, + 0, _H323_UserInformation + }; + bitstr_t bs; + + bs.buf = buf; + bs.beg = bs.cur = beg; + bs.end = beg + sz; + bs.bit = 0; + + return decode_seq(&bs, &h323_userinformation, (char *) uuie, 0); +} + +/****************************************************************************/ +int DecodeMultimediaSystemControlMessage(unsigned char *buf, size_t sz, + MultimediaSystemControlMessage * + mscm) +{ + static field_t multimediasystemcontrolmessage = { + FNAME("MultimediaSystemControlMessage") CHOICE, 2, 4, 4, + DECODE | EXT, 0, _MultimediaSystemControlMessage + }; + bitstr_t bs; + + bs.buf = bs.beg = bs.cur = buf; + bs.end = buf + sz; + bs.bit = 0; + + return decode_choice(&bs, &multimediasystemcontrolmessage, + (char *) mscm, 0); +} + +/****************************************************************************/ +int DecodeQ931(unsigned char *buf, size_t sz, Q931 * q931) +{ + unsigned char *p = buf; + int len; + + if (!p || sz < 1) + return H323_ERROR_BOUND; + + /* Protocol Discriminator */ + if (*p != 0x08) { + PRINT("Unknown Protocol Discriminator\n"); + return H323_ERROR_RANGE; + } + p++; + sz--; + + /* CallReferenceValue */ + if (sz < 1) + return H323_ERROR_BOUND; + len = *p++; + sz--; + if (sz < len) + return H323_ERROR_BOUND; + p += len; + sz -= len; + + /* Message Type */ + if (sz < 1) + return H323_ERROR_BOUND; + q931->MessageType = *p++; + PRINT("MessageType = %02X\n", q931->MessageType); + if (*p & 0x80) { + p++; + sz--; + } + + /* Decode Information Elements */ + while (sz > 0) { + if (*p == 0x7e) { /* UserUserIE */ + if (sz < 3) + break; + p++; + len = *p++ << 8; + len |= *p++; + sz -= 3; + if (sz < len) + break; + p++; + len--; + return DecodeH323_UserInformation(buf, p, len, + &q931->UUIE); + } + p++; + sz--; + if (sz < 1) + break; + len = *p++; + if (sz < len) + break; + p += len; + sz -= len; + } + + PRINT("Q.931 UUIE not found\n"); + + return H323_ERROR_BOUND; +} diff --git a/net/netfilter/nf_conntrack_h323_main.c b/net/netfilter/nf_conntrack_h323_main.c new file mode 100644 index 00000000000..6d8568959f8 --- /dev/null +++ b/net/netfilter/nf_conntrack_h323_main.c @@ -0,0 +1,1856 @@ +/* + * H.323 connection tracking helper + * + * Copyright (c) 2006 Jing Min Zhao + * + * This source code is licensed under General Public License version 2. + * + * Based on the 'brute force' H.323 connection tracking module by + * Jozsef Kadlecsik + * + * For more information, please see http://nath323.sourceforge.net/ + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +/* Parameters */ +static unsigned int default_rrq_ttl __read_mostly = 300; +module_param(default_rrq_ttl, uint, 0600); +MODULE_PARM_DESC(default_rrq_ttl, "use this TTL if it's missing in RRQ"); + +static int gkrouted_only __read_mostly = 1; +module_param(gkrouted_only, int, 0600); +MODULE_PARM_DESC(gkrouted_only, "only accept calls from gatekeeper"); + +static int callforward_filter __read_mostly = 1; +module_param(callforward_filter, bool, 0600); +MODULE_PARM_DESC(callforward_filter, "only create call forwarding expectations " + "if both endpoints are on different sides " + "(determined by routing information)"); + +/* Hooks for NAT */ +int (*set_h245_addr_hook) (struct sk_buff **pskb, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr, + union nf_conntrack_address *addr, __be16 port) + __read_mostly; +int (*set_h225_addr_hook) (struct sk_buff **pskb, + unsigned char **data, int dataoff, + TransportAddress *taddr, + union nf_conntrack_address *addr, __be16 port) + __read_mostly; +int (*set_sig_addr_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, + TransportAddress *taddr, int count) __read_mostly; +int (*set_ras_addr_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, + TransportAddress *taddr, int count) __read_mostly; +int (*nat_rtp_rtcp_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr, + __be16 port, __be16 rtp_port, + struct nf_conntrack_expect *rtp_exp, + struct nf_conntrack_expect *rtcp_exp) __read_mostly; +int (*nat_t120_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr, __be16 port, + struct nf_conntrack_expect *exp) __read_mostly; +int (*nat_h245_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + TransportAddress *taddr, __be16 port, + struct nf_conntrack_expect *exp) __read_mostly; +int (*nat_callforwarding_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + TransportAddress *taddr, __be16 port, + struct nf_conntrack_expect *exp) __read_mostly; +int (*nat_q931_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, TransportAddress *taddr, int idx, + __be16 port, struct nf_conntrack_expect *exp) + __read_mostly; + +static DEFINE_SPINLOCK(nf_h323_lock); +static char *h323_buffer; + +static struct nf_conntrack_helper nf_conntrack_helper_h245; +static struct nf_conntrack_helper nf_conntrack_helper_q931[]; +static struct nf_conntrack_helper nf_conntrack_helper_ras[]; + +/****************************************************************************/ +static int get_tpkt_data(struct sk_buff **pskb, unsigned int protoff, + struct nf_conn *ct, enum ip_conntrack_info ctinfo, + unsigned char **data, int *datalen, int *dataoff) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int dir = CTINFO2DIR(ctinfo); + struct tcphdr _tcph, *th; + int tcpdatalen; + int tcpdataoff; + unsigned char *tpkt; + int tpktlen; + int tpktoff; + + /* Get TCP header */ + th = skb_header_pointer(*pskb, protoff, sizeof(_tcph), &_tcph); + if (th == NULL) + return 0; + + /* Get TCP data offset */ + tcpdataoff = protoff + th->doff * 4; + + /* Get TCP data length */ + tcpdatalen = (*pskb)->len - tcpdataoff; + if (tcpdatalen <= 0) /* No TCP data */ + goto clear_out; + + if (*data == NULL) { /* first TPKT */ + /* Get first TPKT pointer */ + tpkt = skb_header_pointer(*pskb, tcpdataoff, tcpdatalen, + h323_buffer); + BUG_ON(tpkt == NULL); + + /* Validate TPKT identifier */ + if (tcpdatalen < 4 || tpkt[0] != 0x03 || tpkt[1] != 0) { + /* Netmeeting sends TPKT header and data separately */ + if (info->tpkt_len[dir] > 0) { + DEBUGP("nf_ct_h323: previous packet " + "indicated separate TPKT data of %hu " + "bytes\n", info->tpkt_len[dir]); + if (info->tpkt_len[dir] <= tcpdatalen) { + /* Yes, there was a TPKT header + * received */ + *data = tpkt; + *datalen = info->tpkt_len[dir]; + *dataoff = 0; + goto out; + } + + /* Fragmented TPKT */ + if (net_ratelimit()) + printk("nf_ct_h323: " + "fragmented TPKT\n"); + goto clear_out; + } + + /* It is not even a TPKT */ + return 0; + } + tpktoff = 0; + } else { /* Next TPKT */ + tpktoff = *dataoff + *datalen; + tcpdatalen -= tpktoff; + if (tcpdatalen <= 4) /* No more TPKT */ + goto clear_out; + tpkt = *data + *datalen; + + /* Validate TPKT identifier */ + if (tpkt[0] != 0x03 || tpkt[1] != 0) + goto clear_out; + } + + /* Validate TPKT length */ + tpktlen = tpkt[2] * 256 + tpkt[3]; + if (tpktlen < 4) + goto clear_out; + if (tpktlen > tcpdatalen) { + if (tcpdatalen == 4) { /* Separate TPKT header */ + /* Netmeeting sends TPKT header and data separately */ + DEBUGP("nf_ct_h323: separate TPKT header indicates " + "there will be TPKT data of %hu bytes\n", + tpktlen - 4); + info->tpkt_len[dir] = tpktlen - 4; + return 0; + } + + if (net_ratelimit()) + printk("nf_ct_h323: incomplete TPKT (fragmented?)\n"); + goto clear_out; + } + + /* This is the encapsulated data */ + *data = tpkt + 4; + *datalen = tpktlen - 4; + *dataoff = tpktoff + 4; + + out: + /* Clear TPKT length */ + info->tpkt_len[dir] = 0; + return 1; + + clear_out: + info->tpkt_len[dir] = 0; + return 0; +} + +/****************************************************************************/ +static int get_h245_addr(struct nf_conn *ct, unsigned char *data, + H245_TransportAddress *taddr, + union nf_conntrack_address *addr, __be16 *port) +{ + unsigned char *p; + int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; + int len; + + if (taddr->choice != eH245_TransportAddress_unicastAddress) + return 0; + + switch (taddr->unicastAddress.choice) { + case eUnicastAddress_iPAddress: + if (family != AF_INET) + return 0; + p = data + taddr->unicastAddress.iPAddress.network; + len = 4; + break; + case eUnicastAddress_iP6Address: + if (family != AF_INET6) + return 0; + p = data + taddr->unicastAddress.iP6Address.network; + len = 16; + break; + default: + return 0; + } + + memcpy(addr, p, len); + memset((void *)addr + len, 0, sizeof(*addr) - len); + memcpy(port, p + len, sizeof(__be16)); + + return 1; +} + +/****************************************************************************/ +static int expect_rtp_rtcp(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr) +{ + int dir = CTINFO2DIR(ctinfo); + int ret = 0; + __be16 port; + __be16 rtp_port, rtcp_port; + union nf_conntrack_address addr; + struct nf_conntrack_expect *rtp_exp; + struct nf_conntrack_expect *rtcp_exp; + typeof(nat_rtp_rtcp_hook) nat_rtp_rtcp; + + /* Read RTP or RTCP address */ + if (!get_h245_addr(ct, *data, taddr, &addr, &port) || + memcmp(&addr, &ct->tuplehash[dir].tuple.src.u3, sizeof(addr)) || + port == 0) + return 0; + + /* RTP port is even */ + port &= htons(~1); + rtp_port = port; + rtcp_port = htons(ntohs(port) + 1); + + /* Create expect for RTP */ + if ((rtp_exp = nf_conntrack_expect_alloc(ct)) == NULL) + return -1; + nf_conntrack_expect_init(rtp_exp, ct->tuplehash[!dir].tuple.src.l3num, + &ct->tuplehash[!dir].tuple.src.u3, + &ct->tuplehash[!dir].tuple.dst.u3, + IPPROTO_UDP, NULL, &rtp_port); + + /* Create expect for RTCP */ + if ((rtcp_exp = nf_conntrack_expect_alloc(ct)) == NULL) { + nf_conntrack_expect_put(rtp_exp); + return -1; + } + nf_conntrack_expect_init(rtcp_exp, ct->tuplehash[!dir].tuple.src.l3num, + &ct->tuplehash[!dir].tuple.src.u3, + &ct->tuplehash[!dir].tuple.dst.u3, + IPPROTO_UDP, NULL, &rtcp_port); + + if (memcmp(&ct->tuplehash[dir].tuple.src.u3, + &ct->tuplehash[!dir].tuple.dst.u3, + sizeof(ct->tuplehash[dir].tuple.src.u3)) && + (nat_rtp_rtcp = rcu_dereference(nat_rtp_rtcp_hook)) && + ct->status & IPS_NAT_MASK) { + /* NAT needed */ + ret = nat_rtp_rtcp(pskb, ct, ctinfo, data, dataoff, + taddr, port, rtp_port, rtp_exp, rtcp_exp); + } else { /* Conntrack only */ + if (nf_conntrack_expect_related(rtp_exp) == 0) { + if (nf_conntrack_expect_related(rtcp_exp) == 0) { + DEBUGP("nf_ct_h323: expect RTP "); + NF_CT_DUMP_TUPLE(&rtp_exp->tuple); + DEBUGP("nf_ct_h323: expect RTCP "); + NF_CT_DUMP_TUPLE(&rtcp_exp->tuple); + } else { + nf_conntrack_unexpect_related(rtp_exp); + ret = -1; + } + } else + ret = -1; + } + + nf_conntrack_expect_put(rtp_exp); + nf_conntrack_expect_put(rtcp_exp); + + return ret; +} + +/****************************************************************************/ +static int expect_t120(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr) +{ + int dir = CTINFO2DIR(ctinfo); + int ret = 0; + __be16 port; + union nf_conntrack_address addr; + struct nf_conntrack_expect *exp; + typeof(nat_t120_hook) nat_t120; + + /* Read T.120 address */ + if (!get_h245_addr(ct, *data, taddr, &addr, &port) || + memcmp(&addr, &ct->tuplehash[dir].tuple.src.u3, sizeof(addr)) || + port == 0) + return 0; + + /* Create expect for T.120 connections */ + if ((exp = nf_conntrack_expect_alloc(ct)) == NULL) + return -1; + nf_conntrack_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + &ct->tuplehash[!dir].tuple.src.u3, + &ct->tuplehash[!dir].tuple.dst.u3, + IPPROTO_TCP, NULL, &port); + exp->flags = NF_CT_EXPECT_PERMANENT; /* Accept multiple channels */ + + if (memcmp(&ct->tuplehash[dir].tuple.src.u3, + &ct->tuplehash[!dir].tuple.dst.u3, + sizeof(ct->tuplehash[dir].tuple.src.u3)) && + (nat_t120 = rcu_dereference(nat_t120_hook)) && + ct->status & IPS_NAT_MASK) { + /* NAT needed */ + ret = nat_t120(pskb, ct, ctinfo, data, dataoff, taddr, + port, exp); + } else { /* Conntrack only */ + if (nf_conntrack_expect_related(exp) == 0) { + DEBUGP("nf_ct_h323: expect T.120 "); + NF_CT_DUMP_TUPLE(&exp->tuple); + } else + ret = -1; + } + + nf_conntrack_expect_put(exp); + + return ret; +} + +/****************************************************************************/ +static int process_h245_channel(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + H2250LogicalChannelParameters *channel) +{ + int ret; + + if (channel->options & eH2250LogicalChannelParameters_mediaChannel) { + /* RTP */ + ret = expect_rtp_rtcp(pskb, ct, ctinfo, data, dataoff, + &channel->mediaChannel); + if (ret < 0) + return -1; + } + + if (channel-> + options & eH2250LogicalChannelParameters_mediaControlChannel) { + /* RTCP */ + ret = expect_rtp_rtcp(pskb, ct, ctinfo, data, dataoff, + &channel->mediaControlChannel); + if (ret < 0) + return -1; + } + + return 0; +} + +/****************************************************************************/ +static int process_olc(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + OpenLogicalChannel *olc) +{ + int ret; + + DEBUGP("nf_ct_h323: OpenLogicalChannel\n"); + + if (olc->forwardLogicalChannelParameters.multiplexParameters.choice == + eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_h2250LogicalChannelParameters) + { + ret = process_h245_channel(pskb, ct, ctinfo, data, dataoff, + &olc-> + forwardLogicalChannelParameters. + multiplexParameters. + h2250LogicalChannelParameters); + if (ret < 0) + return -1; + } + + if ((olc->options & + eOpenLogicalChannel_reverseLogicalChannelParameters) && + (olc->reverseLogicalChannelParameters.options & + eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters) + && (olc->reverseLogicalChannelParameters.multiplexParameters. + choice == + eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters_h2250LogicalChannelParameters)) + { + ret = + process_h245_channel(pskb, ct, ctinfo, data, dataoff, + &olc-> + reverseLogicalChannelParameters. + multiplexParameters. + h2250LogicalChannelParameters); + if (ret < 0) + return -1; + } + + if ((olc->options & eOpenLogicalChannel_separateStack) && + olc->forwardLogicalChannelParameters.dataType.choice == + eDataType_data && + olc->forwardLogicalChannelParameters.dataType.data.application. + choice == eDataApplicationCapability_application_t120 && + olc->forwardLogicalChannelParameters.dataType.data.application. + t120.choice == eDataProtocolCapability_separateLANStack && + olc->separateStack.networkAddress.choice == + eNetworkAccessParameters_networkAddress_localAreaAddress) { + ret = expect_t120(pskb, ct, ctinfo, data, dataoff, + &olc->separateStack.networkAddress. + localAreaAddress); + if (ret < 0) + return -1; + } + + return 0; +} + +/****************************************************************************/ +static int process_olca(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + OpenLogicalChannelAck *olca) +{ + H2250LogicalChannelAckParameters *ack; + int ret; + + DEBUGP("nf_ct_h323: OpenLogicalChannelAck\n"); + + if ((olca->options & + eOpenLogicalChannelAck_reverseLogicalChannelParameters) && + (olca->reverseLogicalChannelParameters.options & + eOpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters) + && (olca->reverseLogicalChannelParameters.multiplexParameters. + choice == + eOpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters_h2250LogicalChannelParameters)) + { + ret = process_h245_channel(pskb, ct, ctinfo, data, dataoff, + &olca-> + reverseLogicalChannelParameters. + multiplexParameters. + h2250LogicalChannelParameters); + if (ret < 0) + return -1; + } + + if ((olca->options & + eOpenLogicalChannelAck_forwardMultiplexAckParameters) && + (olca->forwardMultiplexAckParameters.choice == + eOpenLogicalChannelAck_forwardMultiplexAckParameters_h2250LogicalChannelAckParameters)) + { + ack = &olca->forwardMultiplexAckParameters. + h2250LogicalChannelAckParameters; + if (ack->options & + eH2250LogicalChannelAckParameters_mediaChannel) { + /* RTP */ + ret = expect_rtp_rtcp(pskb, ct, ctinfo, data, dataoff, + &ack->mediaChannel); + if (ret < 0) + return -1; + } + + if (ack->options & + eH2250LogicalChannelAckParameters_mediaControlChannel) { + /* RTCP */ + ret = expect_rtp_rtcp(pskb, ct, ctinfo, data, dataoff, + &ack->mediaControlChannel); + if (ret < 0) + return -1; + } + } + + return 0; +} + +/****************************************************************************/ +static int process_h245(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + MultimediaSystemControlMessage *mscm) +{ + switch (mscm->choice) { + case eMultimediaSystemControlMessage_request: + if (mscm->request.choice == + eRequestMessage_openLogicalChannel) { + return process_olc(pskb, ct, ctinfo, data, dataoff, + &mscm->request.openLogicalChannel); + } + DEBUGP("nf_ct_h323: H.245 Request %d\n", + mscm->request.choice); + break; + case eMultimediaSystemControlMessage_response: + if (mscm->response.choice == + eResponseMessage_openLogicalChannelAck) { + return process_olca(pskb, ct, ctinfo, data, dataoff, + &mscm->response. + openLogicalChannelAck); + } + DEBUGP("nf_ct_h323: H.245 Response %d\n", + mscm->response.choice); + break; + default: + DEBUGP("nf_ct_h323: H.245 signal %d\n", mscm->choice); + break; + } + + return 0; +} + +/****************************************************************************/ +static int h245_help(struct sk_buff **pskb, unsigned int protoff, + struct nf_conn *ct, enum ip_conntrack_info ctinfo) +{ + static MultimediaSystemControlMessage mscm; + unsigned char *data = NULL; + int datalen; + int dataoff; + int ret; + + /* Until there's been traffic both ways, don't look in packets. */ + if (ctinfo != IP_CT_ESTABLISHED && + ctinfo != IP_CT_ESTABLISHED + IP_CT_IS_REPLY) { + return NF_ACCEPT; + } + DEBUGP("nf_ct_h245: skblen = %u\n", (*pskb)->len); + + spin_lock_bh(&nf_h323_lock); + + /* Process each TPKT */ + while (get_tpkt_data(pskb, protoff, ct, ctinfo, + &data, &datalen, &dataoff)) { + DEBUGP("nf_ct_h245: TPKT len=%d ", datalen); + NF_CT_DUMP_TUPLE(&ct->tuplehash[CTINFO2DIR(ctinfo)].tuple); + + /* Decode H.245 signal */ + ret = DecodeMultimediaSystemControlMessage(data, datalen, + &mscm); + if (ret < 0) { + if (net_ratelimit()) + printk("nf_ct_h245: decoding error: %s\n", + ret == H323_ERROR_BOUND ? + "out of bound" : "out of range"); + /* We don't drop when decoding error */ + break; + } + + /* Process H.245 signal */ + if (process_h245(pskb, ct, ctinfo, &data, dataoff, &mscm) < 0) + goto drop; + } + + spin_unlock_bh(&nf_h323_lock); + return NF_ACCEPT; + + drop: + spin_unlock_bh(&nf_h323_lock); + if (net_ratelimit()) + printk("nf_ct_h245: packet dropped\n"); + return NF_DROP; +} + +/****************************************************************************/ +static struct nf_conntrack_helper nf_conntrack_helper_h245 __read_mostly = { + .name = "H.245", + .me = THIS_MODULE, + .max_expected = H323_RTP_CHANNEL_MAX * 4 + 2 /* T.120 */, + .timeout = 240, + .tuple.dst.protonum = IPPROTO_UDP, + .mask.src.u.udp.port = __constant_htons(0xFFFF), + .mask.dst.protonum = 0xFF, + .help = h245_help +}; + +/****************************************************************************/ +int get_h225_addr(struct nf_conn *ct, unsigned char *data, + TransportAddress *taddr, + union nf_conntrack_address *addr, __be16 *port) +{ + unsigned char *p; + int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; + int len; + + switch (taddr->choice) { + case eTransportAddress_ipAddress: + if (family != AF_INET) + return 0; + p = data + taddr->ipAddress.ip; + len = 4; + break; + case eTransportAddress_ip6Address: + if (family != AF_INET6) + return 0; + p = data + taddr->ip6Address.ip6; + len = 16; + break; + default: + return 0; + } + + memcpy(addr, p, len); + memset((void *)addr + len, 0, sizeof(*addr) - len); + memcpy(port, p + len, sizeof(__be16)); + + return 1; +} + +/****************************************************************************/ +static int expect_h245(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + TransportAddress *taddr) +{ + int dir = CTINFO2DIR(ctinfo); + int ret = 0; + __be16 port; + union nf_conntrack_address addr; + struct nf_conntrack_expect *exp; + typeof(nat_h245_hook) nat_h245; + + /* Read h245Address */ + if (!get_h225_addr(ct, *data, taddr, &addr, &port) || + memcmp(&addr, &ct->tuplehash[dir].tuple.src.u3, sizeof(addr)) || + port == 0) + return 0; + + /* Create expect for h245 connection */ + if ((exp = nf_conntrack_expect_alloc(ct)) == NULL) + return -1; + nf_conntrack_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + &ct->tuplehash[!dir].tuple.src.u3, + &ct->tuplehash[!dir].tuple.dst.u3, + IPPROTO_TCP, NULL, &port); + exp->helper = &nf_conntrack_helper_h245; + + if (memcmp(&ct->tuplehash[dir].tuple.src.u3, + &ct->tuplehash[!dir].tuple.dst.u3, + sizeof(ct->tuplehash[dir].tuple.src.u3)) && + (nat_h245 = rcu_dereference(nat_h245_hook)) && + ct->status & IPS_NAT_MASK) { + /* NAT needed */ + ret = nat_h245(pskb, ct, ctinfo, data, dataoff, taddr, + port, exp); + } else { /* Conntrack only */ + if (nf_conntrack_expect_related(exp) == 0) { + DEBUGP("nf_ct_q931: expect H.245 "); + NF_CT_DUMP_TUPLE(&exp->tuple); + } else + ret = -1; + } + + nf_conntrack_expect_put(exp); + + return ret; +} + +/* If the calling party is on the same side of the forward-to party, + * we don't need to track the second call */ +static int callforward_do_filter(union nf_conntrack_address *src, + union nf_conntrack_address *dst, + int family) +{ + struct flowi fl1, fl2; + int ret = 0; + + memset(&fl1, 0, sizeof(fl1)); + memset(&fl2, 0, sizeof(fl2)); + + switch (family) { + case AF_INET: { + struct rtable *rt1, *rt2; + + fl1.fl4_dst = src->ip; + fl2.fl4_dst = dst->ip; + if (ip_route_output_key(&rt1, &fl1) == 0) { + if (ip_route_output_key(&rt2, &fl2) == 0) { + if (rt1->rt_gateway == rt2->rt_gateway && + rt1->u.dst.dev == rt2->u.dst.dev) + ret = 1; + dst_release(&rt2->u.dst); + } + dst_release(&rt1->u.dst); + } + break; + } +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + case AF_INET6: { + struct rt6_info *rt1, *rt2; + + memcpy(&fl1.fl6_dst, src, sizeof(fl1.fl6_dst)); + memcpy(&fl2.fl6_dst, dst, sizeof(fl2.fl6_dst)); + rt1 = (struct rt6_info *)ip6_route_output(NULL, &fl1); + if (rt1) { + rt2 = (struct rt6_info *)ip6_route_output(NULL, &fl2); + if (rt2) { + if (!memcmp(&rt1->rt6i_gateway, &rt2->rt6i_gateway, + sizeof(rt1->rt6i_gateway)) && + rt1->u.dst.dev == rt2->u.dst.dev) + ret = 1; + dst_release(&rt2->u.dst); + } + dst_release(&rt1->u.dst); + } + break; + } +#endif + } + return ret; + +} + +/****************************************************************************/ +static int expect_callforwarding(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + TransportAddress *taddr) +{ + int dir = CTINFO2DIR(ctinfo); + int ret = 0; + __be16 port; + union nf_conntrack_address addr; + struct nf_conntrack_expect *exp; + typeof(nat_callforwarding_hook) nat_callforwarding; + + /* Read alternativeAddress */ + if (!get_h225_addr(ct, *data, taddr, &addr, &port) || port == 0) + return 0; + + /* If the calling party is on the same side of the forward-to party, + * we don't need to track the second call */ + if (callforward_filter && + callforward_do_filter(&addr, &ct->tuplehash[!dir].tuple.src.u3, + ct->tuplehash[!dir].tuple.src.l3num)) { + DEBUGP("nf_ct_q931: Call Forwarding not tracked\n"); + return 0; + } + + /* Create expect for the second call leg */ + if ((exp = nf_conntrack_expect_alloc(ct)) == NULL) + return -1; + nf_conntrack_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + &ct->tuplehash[!dir].tuple.src.u3, &addr, + IPPROTO_TCP, NULL, &port); + exp->helper = nf_conntrack_helper_q931; + + if (memcmp(&ct->tuplehash[dir].tuple.src.u3, + &ct->tuplehash[!dir].tuple.dst.u3, + sizeof(ct->tuplehash[dir].tuple.src.u3)) && + (nat_callforwarding = rcu_dereference(nat_callforwarding_hook)) && + ct->status & IPS_NAT_MASK) { + /* Need NAT */ + ret = nat_callforwarding(pskb, ct, ctinfo, data, dataoff, + taddr, port, exp); + } else { /* Conntrack only */ + if (nf_conntrack_expect_related(exp) == 0) { + DEBUGP("nf_ct_q931: expect Call Forwarding "); + NF_CT_DUMP_TUPLE(&exp->tuple); + } else + ret = -1; + } + + nf_conntrack_expect_put(exp); + + return ret; +} + +/****************************************************************************/ +static int process_setup(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + Setup_UUIE *setup) +{ + int dir = CTINFO2DIR(ctinfo); + int ret; + int i; + __be16 port; + union nf_conntrack_address addr; + typeof(set_h225_addr_hook) set_h225_addr; + + DEBUGP("nf_ct_q931: Setup\n"); + + if (setup->options & eSetup_UUIE_h245Address) { + ret = expect_h245(pskb, ct, ctinfo, data, dataoff, + &setup->h245Address); + if (ret < 0) + return -1; + } + + set_h225_addr = rcu_dereference(set_h225_addr_hook); + if ((setup->options & eSetup_UUIE_destCallSignalAddress) && + (set_h225_addr) && ct->status && IPS_NAT_MASK && + get_h225_addr(ct, *data, &setup->destCallSignalAddress, + &addr, &port) && + memcmp(&addr, &ct->tuplehash[!dir].tuple.src.u3, sizeof(addr))) { + DEBUGP("nf_ct_q931: set destCallSignalAddress " + NIP6_FMT ":%hu->" NIP6_FMT ":%hu\n", + NIP6(*(struct in6_addr *)&addr), ntohs(port), + NIP6(*(struct in6_addr *)&ct->tuplehash[!dir].tuple.src.u3), + ntohs(ct->tuplehash[!dir].tuple.src.u.tcp.port)); + ret = set_h225_addr(pskb, data, dataoff, + &setup->destCallSignalAddress, + &ct->tuplehash[!dir].tuple.src.u3, + ct->tuplehash[!dir].tuple.src.u.tcp.port); + if (ret < 0) + return -1; + } + + if ((setup->options & eSetup_UUIE_sourceCallSignalAddress) && + (set_h225_addr) && ct->status & IPS_NAT_MASK && + get_h225_addr(ct, *data, &setup->sourceCallSignalAddress, + &addr, &port) && + memcmp(&addr, &ct->tuplehash[!dir].tuple.dst.u3, sizeof(addr))) { + DEBUGP("nf_ct_q931: set sourceCallSignalAddress " + NIP6_FMT ":%hu->" NIP6_FMT ":%hu\n", + NIP6(*(struct in6_addr *)&addr), ntohs(port), + NIP6(*(struct in6_addr *)&ct->tuplehash[!dir].tuple.dst.u3), + ntohs(ct->tuplehash[!dir].tuple.dst.u.tcp.port)); + ret = set_h225_addr(pskb, data, dataoff, + &setup->sourceCallSignalAddress, + &ct->tuplehash[!dir].tuple.dst.u3, + ct->tuplehash[!dir].tuple.dst.u.tcp.port); + if (ret < 0) + return -1; + } + + if (setup->options & eSetup_UUIE_fastStart) { + for (i = 0; i < setup->fastStart.count; i++) { + ret = process_olc(pskb, ct, ctinfo, data, dataoff, + &setup->fastStart.item[i]); + if (ret < 0) + return -1; + } + } + + return 0; +} + +/****************************************************************************/ +static int process_callproceeding(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + CallProceeding_UUIE *callproc) +{ + int ret; + int i; + + DEBUGP("nf_ct_q931: CallProceeding\n"); + + if (callproc->options & eCallProceeding_UUIE_h245Address) { + ret = expect_h245(pskb, ct, ctinfo, data, dataoff, + &callproc->h245Address); + if (ret < 0) + return -1; + } + + if (callproc->options & eCallProceeding_UUIE_fastStart) { + for (i = 0; i < callproc->fastStart.count; i++) { + ret = process_olc(pskb, ct, ctinfo, data, dataoff, + &callproc->fastStart.item[i]); + if (ret < 0) + return -1; + } + } + + return 0; +} + +/****************************************************************************/ +static int process_connect(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + Connect_UUIE *connect) +{ + int ret; + int i; + + DEBUGP("nf_ct_q931: Connect\n"); + + if (connect->options & eConnect_UUIE_h245Address) { + ret = expect_h245(pskb, ct, ctinfo, data, dataoff, + &connect->h245Address); + if (ret < 0) + return -1; + } + + if (connect->options & eConnect_UUIE_fastStart) { + for (i = 0; i < connect->fastStart.count; i++) { + ret = process_olc(pskb, ct, ctinfo, data, dataoff, + &connect->fastStart.item[i]); + if (ret < 0) + return -1; + } + } + + return 0; +} + +/****************************************************************************/ +static int process_alerting(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + Alerting_UUIE *alert) +{ + int ret; + int i; + + DEBUGP("nf_ct_q931: Alerting\n"); + + if (alert->options & eAlerting_UUIE_h245Address) { + ret = expect_h245(pskb, ct, ctinfo, data, dataoff, + &alert->h245Address); + if (ret < 0) + return -1; + } + + if (alert->options & eAlerting_UUIE_fastStart) { + for (i = 0; i < alert->fastStart.count; i++) { + ret = process_olc(pskb, ct, ctinfo, data, dataoff, + &alert->fastStart.item[i]); + if (ret < 0) + return -1; + } + } + + return 0; +} + +/****************************************************************************/ +static int process_information(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + Information_UUIE *info) +{ + int ret; + int i; + + DEBUGP("nf_ct_q931: Information\n"); + + if (info->options & eInformation_UUIE_fastStart) { + for (i = 0; i < info->fastStart.count; i++) { + ret = process_olc(pskb, ct, ctinfo, data, dataoff, + &info->fastStart.item[i]); + if (ret < 0) + return -1; + } + } + + return 0; +} + +/****************************************************************************/ +static int process_facility(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + Facility_UUIE *facility) +{ + int ret; + int i; + + DEBUGP("nf_ct_q931: Facility\n"); + + if (facility->reason.choice == eFacilityReason_callForwarded) { + if (facility->options & eFacility_UUIE_alternativeAddress) + return expect_callforwarding(pskb, ct, ctinfo, data, + dataoff, + &facility-> + alternativeAddress); + return 0; + } + + if (facility->options & eFacility_UUIE_h245Address) { + ret = expect_h245(pskb, ct, ctinfo, data, dataoff, + &facility->h245Address); + if (ret < 0) + return -1; + } + + if (facility->options & eFacility_UUIE_fastStart) { + for (i = 0; i < facility->fastStart.count; i++) { + ret = process_olc(pskb, ct, ctinfo, data, dataoff, + &facility->fastStart.item[i]); + if (ret < 0) + return -1; + } + } + + return 0; +} + +/****************************************************************************/ +static int process_progress(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + Progress_UUIE *progress) +{ + int ret; + int i; + + DEBUGP("nf_ct_q931: Progress\n"); + + if (progress->options & eProgress_UUIE_h245Address) { + ret = expect_h245(pskb, ct, ctinfo, data, dataoff, + &progress->h245Address); + if (ret < 0) + return -1; + } + + if (progress->options & eProgress_UUIE_fastStart) { + for (i = 0; i < progress->fastStart.count; i++) { + ret = process_olc(pskb, ct, ctinfo, data, dataoff, + &progress->fastStart.item[i]); + if (ret < 0) + return -1; + } + } + + return 0; +} + +/****************************************************************************/ +static int process_q931(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, Q931 *q931) +{ + H323_UU_PDU *pdu = &q931->UUIE.h323_uu_pdu; + int i; + int ret = 0; + + switch (pdu->h323_message_body.choice) { + case eH323_UU_PDU_h323_message_body_setup: + ret = process_setup(pskb, ct, ctinfo, data, dataoff, + &pdu->h323_message_body.setup); + break; + case eH323_UU_PDU_h323_message_body_callProceeding: + ret = process_callproceeding(pskb, ct, ctinfo, data, dataoff, + &pdu->h323_message_body. + callProceeding); + break; + case eH323_UU_PDU_h323_message_body_connect: + ret = process_connect(pskb, ct, ctinfo, data, dataoff, + &pdu->h323_message_body.connect); + break; + case eH323_UU_PDU_h323_message_body_alerting: + ret = process_alerting(pskb, ct, ctinfo, data, dataoff, + &pdu->h323_message_body.alerting); + break; + case eH323_UU_PDU_h323_message_body_information: + ret = process_information(pskb, ct, ctinfo, data, dataoff, + &pdu->h323_message_body. + information); + break; + case eH323_UU_PDU_h323_message_body_facility: + ret = process_facility(pskb, ct, ctinfo, data, dataoff, + &pdu->h323_message_body.facility); + break; + case eH323_UU_PDU_h323_message_body_progress: + ret = process_progress(pskb, ct, ctinfo, data, dataoff, + &pdu->h323_message_body.progress); + break; + default: + DEBUGP("nf_ct_q931: Q.931 signal %d\n", + pdu->h323_message_body.choice); + break; + } + + if (ret < 0) + return -1; + + if (pdu->options & eH323_UU_PDU_h245Control) { + for (i = 0; i < pdu->h245Control.count; i++) { + ret = process_h245(pskb, ct, ctinfo, data, dataoff, + &pdu->h245Control.item[i]); + if (ret < 0) + return -1; + } + } + + return 0; +} + +/****************************************************************************/ +static int q931_help(struct sk_buff **pskb, unsigned int protoff, + struct nf_conn *ct, enum ip_conntrack_info ctinfo) +{ + static Q931 q931; + unsigned char *data = NULL; + int datalen; + int dataoff; + int ret; + + /* Until there's been traffic both ways, don't look in packets. */ + if (ctinfo != IP_CT_ESTABLISHED && + ctinfo != IP_CT_ESTABLISHED + IP_CT_IS_REPLY) { + return NF_ACCEPT; + } + DEBUGP("nf_ct_q931: skblen = %u\n", (*pskb)->len); + + spin_lock_bh(&nf_h323_lock); + + /* Process each TPKT */ + while (get_tpkt_data(pskb, protoff, ct, ctinfo, + &data, &datalen, &dataoff)) { + DEBUGP("nf_ct_q931: TPKT len=%d ", datalen); + NF_CT_DUMP_TUPLE(&ct->tuplehash[CTINFO2DIR(ctinfo)].tuple); + + /* Decode Q.931 signal */ + ret = DecodeQ931(data, datalen, &q931); + if (ret < 0) { + if (net_ratelimit()) + printk("nf_ct_q931: decoding error: %s\n", + ret == H323_ERROR_BOUND ? + "out of bound" : "out of range"); + /* We don't drop when decoding error */ + break; + } + + /* Process Q.931 signal */ + if (process_q931(pskb, ct, ctinfo, &data, dataoff, &q931) < 0) + goto drop; + } + + spin_unlock_bh(&nf_h323_lock); + return NF_ACCEPT; + + drop: + spin_unlock_bh(&nf_h323_lock); + if (net_ratelimit()) + printk("nf_ct_q931: packet dropped\n"); + return NF_DROP; +} + +/****************************************************************************/ +static struct nf_conntrack_helper nf_conntrack_helper_q931[] __read_mostly = { + { + .name = "Q.931", + .me = THIS_MODULE, + /* T.120 and H.245 */ + .max_expected = H323_RTP_CHANNEL_MAX * 4 + 4, + .timeout = 240, + .tuple.src.l3num = AF_INET, + .tuple.src.u.tcp.port = __constant_htons(Q931_PORT), + .tuple.dst.protonum = IPPROTO_TCP, + .mask.src.l3num = 0xFFFF, + .mask.src.u.tcp.port = __constant_htons(0xFFFF), + .mask.dst.protonum = 0xFF, + .help = q931_help + }, + { + .name = "Q.931", + .me = THIS_MODULE, + /* T.120 and H.245 */ + .max_expected = H323_RTP_CHANNEL_MAX * 4 + 4, + .timeout = 240, + .tuple.src.l3num = AF_INET6, + .tuple.src.u.tcp.port = __constant_htons(Q931_PORT), + .tuple.dst.protonum = IPPROTO_TCP, + .mask.src.l3num = 0xFFFF, + .mask.src.u.tcp.port = __constant_htons(0xFFFF), + .mask.dst.protonum = 0xFF, + .help = q931_help + }, +}; + +/****************************************************************************/ +static unsigned char *get_udp_data(struct sk_buff **pskb, unsigned int protoff, + int *datalen) +{ + struct udphdr _uh, *uh; + int dataoff; + + uh = skb_header_pointer(*pskb, protoff, sizeof(_uh), &_uh); + if (uh == NULL) + return NULL; + dataoff = protoff + sizeof(_uh); + if (dataoff >= (*pskb)->len) + return NULL; + *datalen = (*pskb)->len - dataoff; + return skb_header_pointer(*pskb, dataoff, *datalen, h323_buffer); +} + +/****************************************************************************/ +static struct nf_conntrack_expect *find_expect(struct nf_conn *ct, + union nf_conntrack_address *addr, + __be16 port) +{ + struct nf_conntrack_expect *exp; + struct nf_conntrack_tuple tuple; + + memset(&tuple.src.u3, 0, sizeof(tuple.src.u3)); + tuple.src.u.tcp.port = 0; + memcpy(&tuple.dst.u3, addr, sizeof(tuple.dst.u3)); + tuple.dst.u.tcp.port = port; + tuple.dst.protonum = IPPROTO_TCP; + + exp = __nf_conntrack_expect_find(&tuple); + if (exp && exp->master == ct) + return exp; + return NULL; +} + +/****************************************************************************/ +static int set_expect_timeout(struct nf_conntrack_expect *exp, + unsigned timeout) +{ + if (!exp || !del_timer(&exp->timeout)) + return 0; + + exp->timeout.expires = jiffies + timeout * HZ; + add_timer(&exp->timeout); + + return 1; +} + +/****************************************************************************/ +static int expect_q931(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, + TransportAddress *taddr, int count) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int dir = CTINFO2DIR(ctinfo); + int ret = 0; + int i; + __be16 port; + union nf_conntrack_address addr; + struct nf_conntrack_expect *exp; + typeof(nat_q931_hook) nat_q931; + + /* Look for the first related address */ + for (i = 0; i < count; i++) { + if (get_h225_addr(ct, *data, &taddr[i], &addr, &port) && + memcmp(&addr, &ct->tuplehash[dir].tuple.src.u3, + sizeof(addr)) == 0 && port != 0) + break; + } + + if (i >= count) /* Not found */ + return 0; + + /* Create expect for Q.931 */ + if ((exp = nf_conntrack_expect_alloc(ct)) == NULL) + return -1; + nf_conntrack_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + gkrouted_only ? /* only accept calls from GK? */ + &ct->tuplehash[!dir].tuple.src.u3 : + NULL, + &ct->tuplehash[!dir].tuple.dst.u3, + IPPROTO_TCP, NULL, &port); + exp->helper = nf_conntrack_helper_q931; + exp->flags = NF_CT_EXPECT_PERMANENT; /* Accept multiple calls */ + + nat_q931 = rcu_dereference(nat_q931_hook); + if (nat_q931 && ct->status & IPS_NAT_MASK) { /* Need NAT */ + ret = nat_q931(pskb, ct, ctinfo, data, taddr, i, port, exp); + } else { /* Conntrack only */ + if (nf_conntrack_expect_related(exp) == 0) { + DEBUGP("nf_ct_ras: expect Q.931 "); + NF_CT_DUMP_TUPLE(&exp->tuple); + + /* Save port for looking up expect in processing RCF */ + info->sig_port[dir] = port; + } else + ret = -1; + } + + nf_conntrack_expect_put(exp); + + return ret; +} + +/****************************************************************************/ +static int process_grq(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, GatekeeperRequest *grq) +{ + typeof(set_ras_addr_hook) set_ras_addr; + + DEBUGP("nf_ct_ras: GRQ\n"); + + set_ras_addr = rcu_dereference(set_ras_addr_hook); + if (set_ras_addr && ct->status & IPS_NAT_MASK) /* NATed */ + return set_ras_addr(pskb, ct, ctinfo, data, + &grq->rasAddress, 1); + return 0; +} + +/****************************************************************************/ +static int process_gcf(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, GatekeeperConfirm *gcf) +{ + int dir = CTINFO2DIR(ctinfo); + int ret = 0; + __be16 port; + union nf_conntrack_address addr; + struct nf_conntrack_expect *exp; + + DEBUGP("nf_ct_ras: GCF\n"); + + if (!get_h225_addr(ct, *data, &gcf->rasAddress, &addr, &port)) + return 0; + + /* Registration port is the same as discovery port */ + if (!memcmp(&addr, &ct->tuplehash[dir].tuple.src.u3, sizeof(addr)) && + port == ct->tuplehash[dir].tuple.src.u.udp.port) + return 0; + + /* Avoid RAS expectation loops. A GCF is never expected. */ + if (test_bit(IPS_EXPECTED_BIT, &ct->status)) + return 0; + + /* Need new expect */ + if ((exp = nf_conntrack_expect_alloc(ct)) == NULL) + return -1; + nf_conntrack_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + &ct->tuplehash[!dir].tuple.src.u3, &addr, + IPPROTO_UDP, NULL, &port); + exp->helper = nf_conntrack_helper_ras; + + if (nf_conntrack_expect_related(exp) == 0) { + DEBUGP("nf_ct_ras: expect RAS "); + NF_CT_DUMP_TUPLE(&exp->tuple); + } else + ret = -1; + + nf_conntrack_expect_put(exp); + + return ret; +} + +/****************************************************************************/ +static int process_rrq(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, RegistrationRequest *rrq) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int ret; + typeof(set_ras_addr_hook) set_ras_addr; + + DEBUGP("nf_ct_ras: RRQ\n"); + + ret = expect_q931(pskb, ct, ctinfo, data, + rrq->callSignalAddress.item, + rrq->callSignalAddress.count); + if (ret < 0) + return -1; + + set_ras_addr = rcu_dereference(set_ras_addr_hook); + if (set_ras_addr && ct->status & IPS_NAT_MASK) { + ret = set_ras_addr(pskb, ct, ctinfo, data, + rrq->rasAddress.item, + rrq->rasAddress.count); + if (ret < 0) + return -1; + } + + if (rrq->options & eRegistrationRequest_timeToLive) { + DEBUGP("nf_ct_ras: RRQ TTL = %u seconds\n", rrq->timeToLive); + info->timeout = rrq->timeToLive; + } else + info->timeout = default_rrq_ttl; + + return 0; +} + +/****************************************************************************/ +static int process_rcf(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, RegistrationConfirm *rcf) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int dir = CTINFO2DIR(ctinfo); + int ret; + struct nf_conntrack_expect *exp; + typeof(set_sig_addr_hook) set_sig_addr; + + DEBUGP("nf_ct_ras: RCF\n"); + + set_sig_addr = rcu_dereference(set_sig_addr_hook); + if (set_sig_addr && ct->status & IPS_NAT_MASK) { + ret = set_sig_addr(pskb, ct, ctinfo, data, + rcf->callSignalAddress.item, + rcf->callSignalAddress.count); + if (ret < 0) + return -1; + } + + if (rcf->options & eRegistrationConfirm_timeToLive) { + DEBUGP("nf_ct_ras: RCF TTL = %u seconds\n", rcf->timeToLive); + info->timeout = rcf->timeToLive; + } + + if (info->timeout > 0) { + DEBUGP + ("nf_ct_ras: set RAS connection timeout to %u seconds\n", + info->timeout); + nf_ct_refresh(ct, *pskb, info->timeout * HZ); + + /* Set expect timeout */ + read_lock_bh(&nf_conntrack_lock); + exp = find_expect(ct, &ct->tuplehash[dir].tuple.dst.u3, + info->sig_port[!dir]); + if (exp) { + DEBUGP("nf_ct_ras: set Q.931 expect " + "timeout to %u seconds for", + info->timeout); + NF_CT_DUMP_TUPLE(&exp->tuple); + set_expect_timeout(exp, info->timeout); + } + read_unlock_bh(&nf_conntrack_lock); + } + + return 0; +} + +/****************************************************************************/ +static int process_urq(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, UnregistrationRequest *urq) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int dir = CTINFO2DIR(ctinfo); + int ret; + typeof(set_sig_addr_hook) set_sig_addr; + + DEBUGP("nf_ct_ras: URQ\n"); + + set_sig_addr = rcu_dereference(set_sig_addr_hook); + if (set_sig_addr && ct->status & IPS_NAT_MASK) { + ret = set_sig_addr(pskb, ct, ctinfo, data, + urq->callSignalAddress.item, + urq->callSignalAddress.count); + if (ret < 0) + return -1; + } + + /* Clear old expect */ + nf_ct_remove_expectations(ct); + info->sig_port[dir] = 0; + info->sig_port[!dir] = 0; + + /* Give it 30 seconds for UCF or URJ */ + nf_ct_refresh(ct, *pskb, 30 * HZ); + + return 0; +} + +/****************************************************************************/ +static int process_arq(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, AdmissionRequest *arq) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int dir = CTINFO2DIR(ctinfo); + __be16 port; + union nf_conntrack_address addr; + typeof(set_h225_addr_hook) set_h225_addr; + + DEBUGP("nf_ct_ras: ARQ\n"); + + set_h225_addr = rcu_dereference(set_h225_addr_hook); + if ((arq->options & eAdmissionRequest_destCallSignalAddress) && + get_h225_addr(ct, *data, &arq->destCallSignalAddress, + &addr, &port) && + !memcmp(&addr, &ct->tuplehash[dir].tuple.src.u3, sizeof(addr)) && + port == info->sig_port[dir] && + set_h225_addr && ct->status & IPS_NAT_MASK) { + /* Answering ARQ */ + return set_h225_addr(pskb, data, 0, + &arq->destCallSignalAddress, + &ct->tuplehash[!dir].tuple.dst.u3, + info->sig_port[!dir]); + } + + if ((arq->options & eAdmissionRequest_srcCallSignalAddress) && + get_h225_addr(ct, *data, &arq->srcCallSignalAddress, + &addr, &port) && + !memcmp(&addr, &ct->tuplehash[dir].tuple.src.u3, sizeof(addr)) && + set_h225_addr && ct->status & IPS_NAT_MASK) { + /* Calling ARQ */ + return set_h225_addr(pskb, data, 0, + &arq->srcCallSignalAddress, + &ct->tuplehash[!dir].tuple.dst.u3, + port); + } + + return 0; +} + +/****************************************************************************/ +static int process_acf(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, AdmissionConfirm *acf) +{ + int dir = CTINFO2DIR(ctinfo); + int ret = 0; + __be16 port; + union nf_conntrack_address addr; + struct nf_conntrack_expect *exp; + typeof(set_sig_addr_hook) set_sig_addr; + + DEBUGP("nf_ct_ras: ACF\n"); + + if (!get_h225_addr(ct, *data, &acf->destCallSignalAddress, + &addr, &port)) + return 0; + + if (!memcmp(&addr, &ct->tuplehash[dir].tuple.dst.u3, sizeof(addr))) { + /* Answering ACF */ + set_sig_addr = rcu_dereference(set_sig_addr_hook); + if (set_sig_addr && ct->status & IPS_NAT_MASK) + return set_sig_addr(pskb, ct, ctinfo, data, + &acf->destCallSignalAddress, 1); + return 0; + } + + /* Need new expect */ + if ((exp = nf_conntrack_expect_alloc(ct)) == NULL) + return -1; + nf_conntrack_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + &ct->tuplehash[!dir].tuple.src.u3, &addr, + IPPROTO_TCP, NULL, &port); + exp->flags = NF_CT_EXPECT_PERMANENT; + exp->helper = nf_conntrack_helper_q931; + + if (nf_conntrack_expect_related(exp) == 0) { + DEBUGP("nf_ct_ras: expect Q.931 "); + NF_CT_DUMP_TUPLE(&exp->tuple); + } else + ret = -1; + + nf_conntrack_expect_put(exp); + + return ret; +} + +/****************************************************************************/ +static int process_lrq(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, LocationRequest *lrq) +{ + typeof(set_ras_addr_hook) set_ras_addr; + + DEBUGP("nf_ct_ras: LRQ\n"); + + set_ras_addr = rcu_dereference(set_ras_addr_hook); + if (set_ras_addr && ct->status & IPS_NAT_MASK) + return set_ras_addr(pskb, ct, ctinfo, data, + &lrq->replyAddress, 1); + return 0; +} + +/****************************************************************************/ +static int process_lcf(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, LocationConfirm *lcf) +{ + int dir = CTINFO2DIR(ctinfo); + int ret = 0; + __be16 port; + union nf_conntrack_address addr; + struct nf_conntrack_expect *exp; + + DEBUGP("nf_ct_ras: LCF\n"); + + if (!get_h225_addr(ct, *data, &lcf->callSignalAddress, + &addr, &port)) + return 0; + + /* Need new expect for call signal */ + if ((exp = nf_conntrack_expect_alloc(ct)) == NULL) + return -1; + nf_conntrack_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + &ct->tuplehash[!dir].tuple.src.u3, &addr, + IPPROTO_TCP, NULL, &port); + exp->flags = NF_CT_EXPECT_PERMANENT; + exp->helper = nf_conntrack_helper_q931; + + if (nf_conntrack_expect_related(exp) == 0) { + DEBUGP("nf_ct_ras: expect Q.931 "); + NF_CT_DUMP_TUPLE(&exp->tuple); + } else + ret = -1; + + nf_conntrack_expect_put(exp); + + /* Ignore rasAddress */ + + return ret; +} + +/****************************************************************************/ +static int process_irr(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, InfoRequestResponse *irr) +{ + int ret; + typeof(set_ras_addr_hook) set_ras_addr; + typeof(set_sig_addr_hook) set_sig_addr; + + DEBUGP("nf_ct_ras: IRR\n"); + + set_ras_addr = rcu_dereference(set_ras_addr_hook); + if (set_ras_addr && ct->status & IPS_NAT_MASK) { + ret = set_ras_addr(pskb, ct, ctinfo, data, + &irr->rasAddress, 1); + if (ret < 0) + return -1; + } + + set_sig_addr = rcu_dereference(set_sig_addr_hook); + if (set_sig_addr && ct->status & IPS_NAT_MASK) { + ret = set_sig_addr(pskb, ct, ctinfo, data, + irr->callSignalAddress.item, + irr->callSignalAddress.count); + if (ret < 0) + return -1; + } + + return 0; +} + +/****************************************************************************/ +static int process_ras(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, RasMessage *ras) +{ + switch (ras->choice) { + case eRasMessage_gatekeeperRequest: + return process_grq(pskb, ct, ctinfo, data, + &ras->gatekeeperRequest); + case eRasMessage_gatekeeperConfirm: + return process_gcf(pskb, ct, ctinfo, data, + &ras->gatekeeperConfirm); + case eRasMessage_registrationRequest: + return process_rrq(pskb, ct, ctinfo, data, + &ras->registrationRequest); + case eRasMessage_registrationConfirm: + return process_rcf(pskb, ct, ctinfo, data, + &ras->registrationConfirm); + case eRasMessage_unregistrationRequest: + return process_urq(pskb, ct, ctinfo, data, + &ras->unregistrationRequest); + case eRasMessage_admissionRequest: + return process_arq(pskb, ct, ctinfo, data, + &ras->admissionRequest); + case eRasMessage_admissionConfirm: + return process_acf(pskb, ct, ctinfo, data, + &ras->admissionConfirm); + case eRasMessage_locationRequest: + return process_lrq(pskb, ct, ctinfo, data, + &ras->locationRequest); + case eRasMessage_locationConfirm: + return process_lcf(pskb, ct, ctinfo, data, + &ras->locationConfirm); + case eRasMessage_infoRequestResponse: + return process_irr(pskb, ct, ctinfo, data, + &ras->infoRequestResponse); + default: + DEBUGP("nf_ct_ras: RAS message %d\n", ras->choice); + break; + } + + return 0; +} + +/****************************************************************************/ +static int ras_help(struct sk_buff **pskb, unsigned int protoff, + struct nf_conn *ct, enum ip_conntrack_info ctinfo) +{ + static RasMessage ras; + unsigned char *data; + int datalen = 0; + int ret; + + DEBUGP("nf_ct_ras: skblen = %u\n", (*pskb)->len); + + spin_lock_bh(&nf_h323_lock); + + /* Get UDP data */ + data = get_udp_data(pskb, protoff, &datalen); + if (data == NULL) + goto accept; + DEBUGP("nf_ct_ras: RAS message len=%d ", datalen); + NF_CT_DUMP_TUPLE(&ct->tuplehash[CTINFO2DIR(ctinfo)].tuple); + + /* Decode RAS message */ + ret = DecodeRasMessage(data, datalen, &ras); + if (ret < 0) { + if (net_ratelimit()) + printk("nf_ct_ras: decoding error: %s\n", + ret == H323_ERROR_BOUND ? + "out of bound" : "out of range"); + goto accept; + } + + /* Process RAS message */ + if (process_ras(pskb, ct, ctinfo, &data, &ras) < 0) + goto drop; + + accept: + spin_unlock_bh(&nf_h323_lock); + return NF_ACCEPT; + + drop: + spin_unlock_bh(&nf_h323_lock); + if (net_ratelimit()) + printk("nf_ct_ras: packet dropped\n"); + return NF_DROP; +} + +/****************************************************************************/ +static struct nf_conntrack_helper nf_conntrack_helper_ras[] __read_mostly = { + { + .name = "RAS", + .me = THIS_MODULE, + .max_expected = 32, + .timeout = 240, + .tuple.src.l3num = AF_INET, + .tuple.src.u.udp.port = __constant_htons(RAS_PORT), + .tuple.dst.protonum = IPPROTO_UDP, + .mask.src.l3num = 0xFFFF, + .mask.src.u.udp.port = __constant_htons(0xFFFF), + .mask.dst.protonum = 0xFF, + .help = ras_help, + }, + { + .name = "RAS", + .me = THIS_MODULE, + .max_expected = 32, + .timeout = 240, + .tuple.src.l3num = AF_INET6, + .tuple.src.u.udp.port = __constant_htons(RAS_PORT), + .tuple.dst.protonum = IPPROTO_UDP, + .mask.src.l3num = 0xFFFF, + .mask.src.u.udp.port = __constant_htons(0xFFFF), + .mask.dst.protonum = 0xFF, + .help = ras_help, + }, +}; + +/****************************************************************************/ +static void __exit nf_conntrack_h323_fini(void) +{ + nf_conntrack_helper_unregister(&nf_conntrack_helper_ras[1]); + nf_conntrack_helper_unregister(&nf_conntrack_helper_ras[0]); + nf_conntrack_helper_unregister(&nf_conntrack_helper_q931[1]); + nf_conntrack_helper_unregister(&nf_conntrack_helper_q931[0]); + kfree(h323_buffer); + DEBUGP("nf_ct_h323: fini\n"); +} + +/****************************************************************************/ +static int __init nf_conntrack_h323_init(void) +{ + int ret; + + h323_buffer = kmalloc(65536, GFP_KERNEL); + if (!h323_buffer) + return -ENOMEM; + ret = nf_conntrack_helper_register(&nf_conntrack_helper_q931[0]); + if (ret < 0) + goto err1; + ret = nf_conntrack_helper_register(&nf_conntrack_helper_q931[1]); + if (ret < 0) + goto err2; + ret = nf_conntrack_helper_register(&nf_conntrack_helper_ras[0]); + if (ret < 0) + goto err3; + ret = nf_conntrack_helper_register(&nf_conntrack_helper_ras[1]); + if (ret < 0) + goto err4; + DEBUGP("nf_ct_h323: init success\n"); + return 0; + +err4: + nf_conntrack_helper_unregister(&nf_conntrack_helper_ras[0]); +err3: + nf_conntrack_helper_unregister(&nf_conntrack_helper_q931[1]); +err2: + nf_conntrack_helper_unregister(&nf_conntrack_helper_q931[0]); +err1: + return ret; +} + +/****************************************************************************/ +module_init(nf_conntrack_h323_init); +module_exit(nf_conntrack_h323_fini); + +EXPORT_SYMBOL_GPL(get_h225_addr); +EXPORT_SYMBOL_GPL(set_h245_addr_hook); +EXPORT_SYMBOL_GPL(set_h225_addr_hook); +EXPORT_SYMBOL_GPL(set_sig_addr_hook); +EXPORT_SYMBOL_GPL(set_ras_addr_hook); +EXPORT_SYMBOL_GPL(nat_rtp_rtcp_hook); +EXPORT_SYMBOL_GPL(nat_t120_hook); +EXPORT_SYMBOL_GPL(nat_h245_hook); +EXPORT_SYMBOL_GPL(nat_callforwarding_hook); +EXPORT_SYMBOL_GPL(nat_q931_hook); + +MODULE_AUTHOR("Jing Min Zhao "); +MODULE_DESCRIPTION("H.323 connection tracking helper"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ip_conntrack_h323"); diff --git a/net/netfilter/nf_conntrack_h323_types.c b/net/netfilter/nf_conntrack_h323_types.c new file mode 100644 index 00000000000..4c6f8b3b120 --- /dev/null +++ b/net/netfilter/nf_conntrack_h323_types.c @@ -0,0 +1,1927 @@ +/* Generated by Jing Min Zhao's ASN.1 parser, Apr 20 2006 + * + * Copyright (c) 2006 Jing Min Zhao + * + * This source code is licensed under General Public License version 2. + */ + +static field_t _TransportAddress_ipAddress[] = { /* SEQUENCE */ + {FNAME("ip") OCTSTR, FIXD, 4, 0, DECODE, + offsetof(TransportAddress_ipAddress, ip), NULL}, + {FNAME("port") INT, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _TransportAddress_ipSourceRoute_route[] = { /* SEQUENCE OF */ + {FNAME("item") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, +}; + +static field_t _TransportAddress_ipSourceRoute_routing[] = { /* CHOICE */ + {FNAME("strict") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("loose") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _TransportAddress_ipSourceRoute[] = { /* SEQUENCE */ + {FNAME("ip") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, + {FNAME("port") INT, WORD, 0, 0, SKIP, 0, NULL}, + {FNAME("route") SEQOF, SEMI, 0, 0, SKIP, 0, + _TransportAddress_ipSourceRoute_route}, + {FNAME("routing") CHOICE, 1, 2, 2, SKIP | EXT, 0, + _TransportAddress_ipSourceRoute_routing}, +}; + +static field_t _TransportAddress_ipxAddress[] = { /* SEQUENCE */ + {FNAME("node") OCTSTR, FIXD, 6, 0, SKIP, 0, NULL}, + {FNAME("netnum") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, + {FNAME("port") OCTSTR, FIXD, 2, 0, SKIP, 0, NULL}, +}; + +static field_t _TransportAddress_ip6Address[] = { /* SEQUENCE */ + {FNAME("ip") OCTSTR, FIXD, 16, 0, DECODE, + offsetof(TransportAddress_ip6Address, ip6), NULL}, + {FNAME("port") INT, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H221NonStandard[] = { /* SEQUENCE */ + {FNAME("t35CountryCode") INT, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("t35Extension") INT, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("manufacturerCode") INT, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _NonStandardIdentifier[] = { /* CHOICE */ + {FNAME("object") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("h221NonStandard") SEQ, 0, 3, 3, SKIP | EXT, 0, + _H221NonStandard}, +}; + +static field_t _NonStandardParameter[] = { /* SEQUENCE */ + {FNAME("nonStandardIdentifier") CHOICE, 1, 2, 2, SKIP | EXT, 0, + _NonStandardIdentifier}, + {FNAME("data") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _TransportAddress[] = { /* CHOICE */ + {FNAME("ipAddress") SEQ, 0, 2, 2, DECODE, + offsetof(TransportAddress, ipAddress), _TransportAddress_ipAddress}, + {FNAME("ipSourceRoute") SEQ, 0, 4, 4, SKIP | EXT, 0, + _TransportAddress_ipSourceRoute}, + {FNAME("ipxAddress") SEQ, 0, 3, 3, SKIP, 0, + _TransportAddress_ipxAddress}, + {FNAME("ip6Address") SEQ, 0, 2, 2, DECODE | EXT, + offsetof(TransportAddress, ip6Address), _TransportAddress_ip6Address}, + {FNAME("netBios") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, + {FNAME("nsap") OCTSTR, 5, 1, 0, SKIP, 0, NULL}, + {FNAME("nonStandardAddress") SEQ, 0, 2, 2, SKIP, 0, + _NonStandardParameter}, +}; + +static field_t _AliasAddress[] = { /* CHOICE */ + {FNAME("dialedDigits") NUMDGT, 7, 1, 0, SKIP, 0, NULL}, + {FNAME("h323-ID") BMPSTR, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("url-ID") IA5STR, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("transportID") CHOICE, 3, 7, 7, SKIP | EXT, 0, NULL}, + {FNAME("email-ID") IA5STR, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("partyNumber") CHOICE, 3, 5, 5, SKIP | EXT, 0, NULL}, + {FNAME("mobileUIM") CHOICE, 1, 2, 2, SKIP | EXT, 0, NULL}, +}; + +static field_t _Setup_UUIE_sourceAddress[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _VendorIdentifier[] = { /* SEQUENCE */ + {FNAME("vendor") SEQ, 0, 3, 3, SKIP | EXT, 0, _H221NonStandard}, + {FNAME("productId") OCTSTR, BYTE, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("versionId") OCTSTR, BYTE, 1, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _GatekeeperInfo[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, +}; + +static field_t _H310Caps[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H320Caps[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H321Caps[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H322Caps[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H323Caps[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H324Caps[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _VoiceCaps[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _T120OnlyCaps[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _SupportedProtocols[] = { /* CHOICE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP, 0, + _NonStandardParameter}, + {FNAME("h310") SEQ, 1, 1, 3, SKIP | EXT, 0, _H310Caps}, + {FNAME("h320") SEQ, 1, 1, 3, SKIP | EXT, 0, _H320Caps}, + {FNAME("h321") SEQ, 1, 1, 3, SKIP | EXT, 0, _H321Caps}, + {FNAME("h322") SEQ, 1, 1, 3, SKIP | EXT, 0, _H322Caps}, + {FNAME("h323") SEQ, 1, 1, 3, SKIP | EXT, 0, _H323Caps}, + {FNAME("h324") SEQ, 1, 1, 3, SKIP | EXT, 0, _H324Caps}, + {FNAME("voice") SEQ, 1, 1, 3, SKIP | EXT, 0, _VoiceCaps}, + {FNAME("t120-only") SEQ, 1, 1, 3, SKIP | EXT, 0, _T120OnlyCaps}, + {FNAME("nonStandardProtocol") SEQ, 2, 3, 3, SKIP | EXT, 0, NULL}, + {FNAME("t38FaxAnnexbOnly") SEQ, 2, 5, 5, SKIP | EXT, 0, NULL}, +}; + +static field_t _GatewayInfo_protocol[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 4, 9, 11, SKIP | EXT, 0, _SupportedProtocols}, +}; + +static field_t _GatewayInfo[] = { /* SEQUENCE */ + {FNAME("protocol") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _GatewayInfo_protocol}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, +}; + +static field_t _McuInfo[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("protocol") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _TerminalInfo[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, +}; + +static field_t _EndpointType[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("vendor") SEQ, 2, 3, 3, SKIP | EXT | OPT, 0, + _VendorIdentifier}, + {FNAME("gatekeeper") SEQ, 1, 1, 1, SKIP | EXT | OPT, 0, + _GatekeeperInfo}, + {FNAME("gateway") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, _GatewayInfo}, + {FNAME("mcu") SEQ, 1, 1, 2, SKIP | EXT | OPT, 0, _McuInfo}, + {FNAME("terminal") SEQ, 1, 1, 1, SKIP | EXT | OPT, 0, _TerminalInfo}, + {FNAME("mc") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("undefinedNode") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("set") BITSTR, FIXD, 32, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedTunnelledProtocols") SEQOF, SEMI, 0, 0, SKIP | OPT, + 0, NULL}, +}; + +static field_t _Setup_UUIE_destinationAddress[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _Setup_UUIE_destExtraCallInfo[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _Setup_UUIE_destExtraCRV[] = { /* SEQUENCE OF */ + {FNAME("item") INT, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _Setup_UUIE_conferenceGoal[] = { /* CHOICE */ + {FNAME("create") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("join") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("invite") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("capability-negotiation") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("callIndependentSupplementaryService") NUL, FIXD, 0, 0, SKIP, + 0, NULL}, +}; + +static field_t _Q954Details[] = { /* SEQUENCE */ + {FNAME("conferenceCalling") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("threePartyService") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _QseriesOptions[] = { /* SEQUENCE */ + {FNAME("q932Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("q951Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("q952Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("q953Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("q955Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("q956Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("q957Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("q954Info") SEQ, 0, 2, 2, SKIP | EXT, 0, _Q954Details}, +}; + +static field_t _CallType[] = { /* CHOICE */ + {FNAME("pointToPoint") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("oneToN") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("nToOne") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("nToN") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H245_NonStandardIdentifier_h221NonStandard[] = { /* SEQUENCE */ + {FNAME("t35CountryCode") INT, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("t35Extension") INT, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("manufacturerCode") INT, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H245_NonStandardIdentifier[] = { /* CHOICE */ + {FNAME("object") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("h221NonStandard") SEQ, 0, 3, 3, SKIP, 0, + _H245_NonStandardIdentifier_h221NonStandard}, +}; + +static field_t _H245_NonStandardParameter[] = { /* SEQUENCE */ + {FNAME("nonStandardIdentifier") CHOICE, 1, 2, 2, SKIP, 0, + _H245_NonStandardIdentifier}, + {FNAME("data") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H261VideoCapability[] = { /* SEQUENCE */ + {FNAME("qcifMPI") INT, 2, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("cifMPI") INT, 2, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("temporalSpatialTradeOffCapability") BOOL, FIXD, 0, 0, SKIP, 0, + NULL}, + {FNAME("maxBitRate") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("stillImageTransmission") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("videoBadMBsCap") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H262VideoCapability[] = { /* SEQUENCE */ + {FNAME("profileAndLevel-SPatML") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("profileAndLevel-MPatLL") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("profileAndLevel-MPatML") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("profileAndLevel-MPatH-14") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("profileAndLevel-MPatHL") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("profileAndLevel-SNRatLL") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("profileAndLevel-SNRatML") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("profileAndLevel-SpatialatH-14") BOOL, FIXD, 0, 0, SKIP, 0, + NULL}, + {FNAME("profileAndLevel-HPatML") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("profileAndLevel-HPatH-14") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("profileAndLevel-HPatHL") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("videoBitRate") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("vbvBufferSize") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("samplesPerLine") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("linesPerFrame") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("framesPerSecond") INT, 4, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("luminanceSampleRate") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("videoBadMBsCap") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H263VideoCapability[] = { /* SEQUENCE */ + {FNAME("sqcifMPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("qcifMPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("cifMPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("cif4MPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("cif16MPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("maxBitRate") INT, CONS, 1, 0, SKIP, 0, NULL}, + {FNAME("unrestrictedVector") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("arithmeticCoding") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("advancedPrediction") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("pbFrames") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("temporalSpatialTradeOffCapability") BOOL, FIXD, 0, 0, SKIP, 0, + NULL}, + {FNAME("hrd-B") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("bppMaxKb") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("slowSqcifMPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("slowQcifMPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("slowCifMPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("slowCif4MPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("slowCif16MPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("errorCompensation") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("enhancementLayerInfo") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("h263Options") SEQ, 5, 29, 31, SKIP | EXT | OPT, 0, NULL}, +}; + +static field_t _IS11172VideoCapability[] = { /* SEQUENCE */ + {FNAME("constrainedBitstream") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("videoBitRate") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("vbvBufferSize") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("samplesPerLine") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("linesPerFrame") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("pictureRate") INT, 4, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("luminanceSampleRate") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("videoBadMBsCap") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _VideoCapability[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, + _H245_NonStandardParameter}, + {FNAME("h261VideoCapability") SEQ, 2, 5, 6, SKIP | EXT, 0, + _H261VideoCapability}, + {FNAME("h262VideoCapability") SEQ, 6, 17, 18, SKIP | EXT, 0, + _H262VideoCapability}, + {FNAME("h263VideoCapability") SEQ, 7, 13, 21, SKIP | EXT, 0, + _H263VideoCapability}, + {FNAME("is11172VideoCapability") SEQ, 6, 7, 8, SKIP | EXT, 0, + _IS11172VideoCapability}, + {FNAME("genericVideoCapability") SEQ, 5, 6, 6, SKIP | EXT, 0, NULL}, +}; + +static field_t _AudioCapability_g7231[] = { /* SEQUENCE */ + {FNAME("maxAl-sduAudioFrames") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("silenceSuppression") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _IS11172AudioCapability[] = { /* SEQUENCE */ + {FNAME("audioLayer1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioLayer2") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioLayer3") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioSampling32k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioSampling44k1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioSampling48k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("singleChannel") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("twoChannels") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("bitRate") INT, WORD, 1, 0, SKIP, 0, NULL}, +}; + +static field_t _IS13818AudioCapability[] = { /* SEQUENCE */ + {FNAME("audioLayer1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioLayer2") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioLayer3") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioSampling16k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioSampling22k05") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioSampling24k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioSampling32k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioSampling44k1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioSampling48k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("singleChannel") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("twoChannels") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("threeChannels2-1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("threeChannels3-0") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("fourChannels2-0-2-0") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("fourChannels2-2") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("fourChannels3-1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("fiveChannels3-0-2-0") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("fiveChannels3-2") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("lowFrequencyEnhancement") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("multilingual") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("bitRate") INT, WORD, 1, 0, SKIP, 0, NULL}, +}; + +static field_t _AudioCapability[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, + _H245_NonStandardParameter}, + {FNAME("g711Alaw64k") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g711Alaw56k") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g711Ulaw64k") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g711Ulaw56k") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g722-64k") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g722-56k") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g722-48k") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g7231") SEQ, 0, 2, 2, SKIP, 0, _AudioCapability_g7231}, + {FNAME("g728") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g729") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g729AnnexA") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("is11172AudioCapability") SEQ, 0, 9, 9, SKIP | EXT, 0, + _IS11172AudioCapability}, + {FNAME("is13818AudioCapability") SEQ, 0, 21, 21, SKIP | EXT, 0, + _IS13818AudioCapability}, + {FNAME("g729wAnnexB") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g729AnnexAwAnnexB") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g7231AnnexCCapability") SEQ, 1, 3, 3, SKIP | EXT, 0, NULL}, + {FNAME("gsmFullRate") SEQ, 0, 3, 3, SKIP | EXT, 0, NULL}, + {FNAME("gsmHalfRate") SEQ, 0, 3, 3, SKIP | EXT, 0, NULL}, + {FNAME("gsmEnhancedFullRate") SEQ, 0, 3, 3, SKIP | EXT, 0, NULL}, + {FNAME("genericAudioCapability") SEQ, 5, 6, 6, SKIP | EXT, 0, NULL}, + {FNAME("g729Extensions") SEQ, 1, 8, 8, SKIP | EXT, 0, NULL}, +}; + +static field_t _DataProtocolCapability[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, + _H245_NonStandardParameter}, + {FNAME("v14buffered") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("v42lapm") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("hdlcFrameTunnelling") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("h310SeparateVCStack") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("h310SingleVCStack") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("transparent") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("segmentationAndReassembly") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("hdlcFrameTunnelingwSAR") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("v120") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("separateLANStack") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("v76wCompression") CHOICE, 2, 3, 3, SKIP | EXT, 0, NULL}, + {FNAME("tcp") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("udp") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _T84Profile_t84Restricted[] = { /* SEQUENCE */ + {FNAME("qcif") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("cif") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("ccir601Seq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("ccir601Prog") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("hdtvSeq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("hdtvProg") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("g3FacsMH200x100") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("g3FacsMH200x200") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("g4FacsMMR200x100") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("g4FacsMMR200x200") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("jbig200x200Seq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("jbig200x200Prog") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("jbig300x300Seq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("jbig300x300Prog") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("digPhotoLow") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("digPhotoMedSeq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("digPhotoMedProg") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("digPhotoHighSeq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("digPhotoHighProg") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _T84Profile[] = { /* CHOICE */ + {FNAME("t84Unrestricted") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("t84Restricted") SEQ, 0, 19, 19, SKIP | EXT, 0, + _T84Profile_t84Restricted}, +}; + +static field_t _DataApplicationCapability_application_t84[] = { /* SEQUENCE */ + {FNAME("t84Protocol") CHOICE, 3, 7, 14, SKIP | EXT, 0, + _DataProtocolCapability}, + {FNAME("t84Profile") CHOICE, 1, 2, 2, SKIP, 0, _T84Profile}, +}; + +static field_t _DataApplicationCapability_application_nlpid[] = { /* SEQUENCE */ + {FNAME("nlpidProtocol") CHOICE, 3, 7, 14, SKIP | EXT, 0, + _DataProtocolCapability}, + {FNAME("nlpidData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _DataApplicationCapability_application[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, + _H245_NonStandardParameter}, + {FNAME("t120") CHOICE, 3, 7, 14, DECODE | EXT, + offsetof(DataApplicationCapability_application, t120), + _DataProtocolCapability}, + {FNAME("dsm-cc") CHOICE, 3, 7, 14, SKIP | EXT, 0, + _DataProtocolCapability}, + {FNAME("userData") CHOICE, 3, 7, 14, SKIP | EXT, 0, + _DataProtocolCapability}, + {FNAME("t84") SEQ, 0, 2, 2, SKIP, 0, + _DataApplicationCapability_application_t84}, + {FNAME("t434") CHOICE, 3, 7, 14, SKIP | EXT, 0, + _DataProtocolCapability}, + {FNAME("h224") CHOICE, 3, 7, 14, SKIP | EXT, 0, + _DataProtocolCapability}, + {FNAME("nlpid") SEQ, 0, 2, 2, SKIP, 0, + _DataApplicationCapability_application_nlpid}, + {FNAME("dsvdControl") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("h222DataPartitioning") CHOICE, 3, 7, 14, SKIP | EXT, 0, + _DataProtocolCapability}, + {FNAME("t30fax") CHOICE, 3, 7, 14, SKIP | EXT, 0, NULL}, + {FNAME("t140") CHOICE, 3, 7, 14, SKIP | EXT, 0, NULL}, + {FNAME("t38fax") SEQ, 0, 2, 2, SKIP, 0, NULL}, + {FNAME("genericDataCapability") SEQ, 5, 6, 6, SKIP | EXT, 0, NULL}, +}; + +static field_t _DataApplicationCapability[] = { /* SEQUENCE */ + {FNAME("application") CHOICE, 4, 10, 14, DECODE | EXT, + offsetof(DataApplicationCapability, application), + _DataApplicationCapability_application}, + {FNAME("maxBitRate") INT, CONS, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _EncryptionMode[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, + _H245_NonStandardParameter}, + {FNAME("h233Encryption") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _DataType[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, + _H245_NonStandardParameter}, + {FNAME("nullData") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("videoData") CHOICE, 3, 5, 6, SKIP | EXT, 0, _VideoCapability}, + {FNAME("audioData") CHOICE, 4, 14, 22, SKIP | EXT, 0, + _AudioCapability}, + {FNAME("data") SEQ, 0, 2, 2, DECODE | EXT, offsetof(DataType, data), + _DataApplicationCapability}, + {FNAME("encryptionData") CHOICE, 1, 2, 2, SKIP | EXT, 0, + _EncryptionMode}, + {FNAME("h235Control") SEQ, 0, 2, 2, SKIP, 0, NULL}, + {FNAME("h235Media") SEQ, 0, 2, 2, SKIP | EXT, 0, NULL}, + {FNAME("multiplexedStream") SEQ, 0, 2, 2, SKIP | EXT, 0, NULL}, +}; + +static field_t _H222LogicalChannelParameters[] = { /* SEQUENCE */ + {FNAME("resourceID") INT, WORD, 0, 0, SKIP, 0, NULL}, + {FNAME("subChannelID") INT, WORD, 0, 0, SKIP, 0, NULL}, + {FNAME("pcr-pid") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("programDescriptors") OCTSTR, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("streamDescriptors") OCTSTR, SEMI, 0, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _H223LogicalChannelParameters_adaptationLayerType_al3[] = { /* SEQUENCE */ + {FNAME("controlFieldOctets") INT, 2, 0, 0, SKIP, 0, NULL}, + {FNAME("sendBufferSize") INT, CONS, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H223LogicalChannelParameters_adaptationLayerType[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, + _H245_NonStandardParameter}, + {FNAME("al1Framed") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("al1NotFramed") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("al2WithoutSequenceNumbers") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("al2WithSequenceNumbers") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("al3") SEQ, 0, 2, 2, SKIP, 0, + _H223LogicalChannelParameters_adaptationLayerType_al3}, + {FNAME("al1M") SEQ, 0, 7, 8, SKIP | EXT, 0, NULL}, + {FNAME("al2M") SEQ, 0, 2, 2, SKIP | EXT, 0, NULL}, + {FNAME("al3M") SEQ, 0, 5, 6, SKIP | EXT, 0, NULL}, +}; + +static field_t _H223LogicalChannelParameters[] = { /* SEQUENCE */ + {FNAME("adaptationLayerType") CHOICE, 3, 6, 9, SKIP | EXT, 0, + _H223LogicalChannelParameters_adaptationLayerType}, + {FNAME("segmentableFlag") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CRCLength[] = { /* CHOICE */ + {FNAME("crc8bit") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("crc16bit") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("crc32bit") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _V76HDLCParameters[] = { /* SEQUENCE */ + {FNAME("crcLength") CHOICE, 2, 3, 3, SKIP | EXT, 0, _CRCLength}, + {FNAME("n401") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("loopbackTestProcedure") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _V76LogicalChannelParameters_suspendResume[] = { /* CHOICE */ + {FNAME("noSuspendResume") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("suspendResumewAddress") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("suspendResumewoAddress") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _V76LogicalChannelParameters_mode_eRM_recovery[] = { /* CHOICE */ + {FNAME("rej") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("sREJ") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("mSREJ") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _V76LogicalChannelParameters_mode_eRM[] = { /* SEQUENCE */ + {FNAME("windowSize") INT, 7, 1, 0, SKIP, 0, NULL}, + {FNAME("recovery") CHOICE, 2, 3, 3, SKIP | EXT, 0, + _V76LogicalChannelParameters_mode_eRM_recovery}, +}; + +static field_t _V76LogicalChannelParameters_mode[] = { /* CHOICE */ + {FNAME("eRM") SEQ, 0, 2, 2, SKIP | EXT, 0, + _V76LogicalChannelParameters_mode_eRM}, + {FNAME("uNERM") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _V75Parameters[] = { /* SEQUENCE */ + {FNAME("audioHeaderPresent") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _V76LogicalChannelParameters[] = { /* SEQUENCE */ + {FNAME("hdlcParameters") SEQ, 0, 3, 3, SKIP | EXT, 0, + _V76HDLCParameters}, + {FNAME("suspendResume") CHOICE, 2, 3, 3, SKIP | EXT, 0, + _V76LogicalChannelParameters_suspendResume}, + {FNAME("uIH") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("mode") CHOICE, 1, 2, 2, SKIP | EXT, 0, + _V76LogicalChannelParameters_mode}, + {FNAME("v75Parameters") SEQ, 0, 1, 1, SKIP | EXT, 0, _V75Parameters}, +}; + +static field_t _H2250LogicalChannelParameters_nonStandard[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 0, 2, 2, SKIP, 0, _H245_NonStandardParameter}, +}; + +static field_t _UnicastAddress_iPAddress[] = { /* SEQUENCE */ + {FNAME("network") OCTSTR, FIXD, 4, 0, DECODE, + offsetof(UnicastAddress_iPAddress, network), NULL}, + {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _UnicastAddress_iPXAddress[] = { /* SEQUENCE */ + {FNAME("node") OCTSTR, FIXD, 6, 0, SKIP, 0, NULL}, + {FNAME("netnum") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, + {FNAME("tsapIdentifier") OCTSTR, FIXD, 2, 0, SKIP, 0, NULL}, +}; + +static field_t _UnicastAddress_iP6Address[] = { /* SEQUENCE */ + {FNAME("network") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, + {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _UnicastAddress_iPSourceRouteAddress_routing[] = { /* CHOICE */ + {FNAME("strict") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("loose") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _UnicastAddress_iPSourceRouteAddress_route[] = { /* SEQUENCE OF */ + {FNAME("item") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, +}; + +static field_t _UnicastAddress_iPSourceRouteAddress[] = { /* SEQUENCE */ + {FNAME("routing") CHOICE, 1, 2, 2, SKIP, 0, + _UnicastAddress_iPSourceRouteAddress_routing}, + {FNAME("network") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, + {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, + {FNAME("route") SEQOF, SEMI, 0, 0, SKIP, 0, + _UnicastAddress_iPSourceRouteAddress_route}, +}; + +static field_t _UnicastAddress[] = { /* CHOICE */ + {FNAME("iPAddress") SEQ, 0, 2, 2, DECODE | EXT, + offsetof(UnicastAddress, iPAddress), _UnicastAddress_iPAddress}, + {FNAME("iPXAddress") SEQ, 0, 3, 3, SKIP | EXT, 0, + _UnicastAddress_iPXAddress}, + {FNAME("iP6Address") SEQ, 0, 2, 2, SKIP | EXT, 0, + _UnicastAddress_iP6Address}, + {FNAME("netBios") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, + {FNAME("iPSourceRouteAddress") SEQ, 0, 4, 4, SKIP | EXT, 0, + _UnicastAddress_iPSourceRouteAddress}, + {FNAME("nsap") OCTSTR, 5, 1, 0, SKIP, 0, NULL}, + {FNAME("nonStandardAddress") SEQ, 0, 2, 2, SKIP, 0, NULL}, +}; + +static field_t _MulticastAddress_iPAddress[] = { /* SEQUENCE */ + {FNAME("network") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, + {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _MulticastAddress_iP6Address[] = { /* SEQUENCE */ + {FNAME("network") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, + {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _MulticastAddress[] = { /* CHOICE */ + {FNAME("iPAddress") SEQ, 0, 2, 2, SKIP | EXT, 0, + _MulticastAddress_iPAddress}, + {FNAME("iP6Address") SEQ, 0, 2, 2, SKIP | EXT, 0, + _MulticastAddress_iP6Address}, + {FNAME("nsap") OCTSTR, 5, 1, 0, SKIP, 0, NULL}, + {FNAME("nonStandardAddress") SEQ, 0, 2, 2, SKIP, 0, NULL}, +}; + +static field_t _H245_TransportAddress[] = { /* CHOICE */ + {FNAME("unicastAddress") CHOICE, 3, 5, 7, DECODE | EXT, + offsetof(H245_TransportAddress, unicastAddress), _UnicastAddress}, + {FNAME("multicastAddress") CHOICE, 1, 2, 4, SKIP | EXT, 0, + _MulticastAddress}, +}; + +static field_t _H2250LogicalChannelParameters[] = { /* SEQUENCE */ + {FNAME("nonStandard") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _H2250LogicalChannelParameters_nonStandard}, + {FNAME("sessionID") INT, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("associatedSessionID") INT, 8, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("mediaChannel") CHOICE, 1, 2, 2, DECODE | EXT | OPT, + offsetof(H2250LogicalChannelParameters, mediaChannel), + _H245_TransportAddress}, + {FNAME("mediaGuaranteedDelivery") BOOL, FIXD, 0, 0, SKIP | OPT, 0, + NULL}, + {FNAME("mediaControlChannel") CHOICE, 1, 2, 2, DECODE | EXT | OPT, + offsetof(H2250LogicalChannelParameters, mediaControlChannel), + _H245_TransportAddress}, + {FNAME("mediaControlGuaranteedDelivery") BOOL, FIXD, 0, 0, STOP | OPT, + 0, NULL}, + {FNAME("silenceSuppression") BOOL, FIXD, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("destination") SEQ, 0, 2, 2, STOP | EXT | OPT, 0, NULL}, + {FNAME("dynamicRTPPayloadType") INT, 5, 96, 0, STOP | OPT, 0, NULL}, + {FNAME("mediaPacketization") CHOICE, 0, 1, 2, STOP | EXT | OPT, 0, + NULL}, + {FNAME("transportCapability") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, + NULL}, + {FNAME("redundancyEncoding") SEQ, 1, 2, 2, STOP | EXT | OPT, 0, NULL}, + {FNAME("source") SEQ, 0, 2, 2, SKIP | EXT | OPT, 0, NULL}, +}; + +static field_t _OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters[] = { /* CHOICE */ + {FNAME("h222LogicalChannelParameters") SEQ, 3, 5, 5, SKIP | EXT, 0, + _H222LogicalChannelParameters}, + {FNAME("h223LogicalChannelParameters") SEQ, 0, 2, 2, SKIP | EXT, 0, + _H223LogicalChannelParameters}, + {FNAME("v76LogicalChannelParameters") SEQ, 0, 5, 5, SKIP | EXT, 0, + _V76LogicalChannelParameters}, + {FNAME("h2250LogicalChannelParameters") SEQ, 10, 11, 14, DECODE | EXT, + offsetof + (OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters, + h2250LogicalChannelParameters), _H2250LogicalChannelParameters}, + {FNAME("none") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _OpenLogicalChannel_forwardLogicalChannelParameters[] = { /* SEQUENCE */ + {FNAME("portNumber") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("dataType") CHOICE, 3, 6, 9, DECODE | EXT, + offsetof(OpenLogicalChannel_forwardLogicalChannelParameters, + dataType), _DataType}, + {FNAME("multiplexParameters") CHOICE, 2, 3, 5, DECODE | EXT, + offsetof(OpenLogicalChannel_forwardLogicalChannelParameters, + multiplexParameters), + _OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters}, + {FNAME("forwardLogicalChannelDependency") INT, WORD, 1, 0, SKIP | OPT, + 0, NULL}, + {FNAME("replacementFor") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters[] = { /* CHOICE */ + {FNAME("h223LogicalChannelParameters") SEQ, 0, 2, 2, SKIP | EXT, 0, + _H223LogicalChannelParameters}, + {FNAME("v76LogicalChannelParameters") SEQ, 0, 5, 5, SKIP | EXT, 0, + _V76LogicalChannelParameters}, + {FNAME("h2250LogicalChannelParameters") SEQ, 10, 11, 14, DECODE | EXT, + offsetof + (OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters, + h2250LogicalChannelParameters), _H2250LogicalChannelParameters}, +}; + +static field_t _OpenLogicalChannel_reverseLogicalChannelParameters[] = { /* SEQUENCE */ + {FNAME("dataType") CHOICE, 3, 6, 9, SKIP | EXT, 0, _DataType}, + {FNAME("multiplexParameters") CHOICE, 1, 2, 3, DECODE | EXT | OPT, + offsetof(OpenLogicalChannel_reverseLogicalChannelParameters, + multiplexParameters), + _OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters}, + {FNAME("reverseLogicalChannelDependency") INT, WORD, 1, 0, SKIP | OPT, + 0, NULL}, + {FNAME("replacementFor") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _NetworkAccessParameters_distribution[] = { /* CHOICE */ + {FNAME("unicast") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("multicast") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _Q2931Address_address[] = { /* CHOICE */ + {FNAME("internationalNumber") NUMSTR, 4, 1, 0, SKIP, 0, NULL}, + {FNAME("nsapAddress") OCTSTR, 5, 1, 0, SKIP, 0, NULL}, +}; + +static field_t _Q2931Address[] = { /* SEQUENCE */ + {FNAME("address") CHOICE, 1, 2, 2, SKIP | EXT, 0, + _Q2931Address_address}, + {FNAME("subaddress") OCTSTR, 5, 1, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _NetworkAccessParameters_networkAddress[] = { /* CHOICE */ + {FNAME("q2931Address") SEQ, 1, 2, 2, SKIP | EXT, 0, _Q2931Address}, + {FNAME("e164Address") NUMDGT, 7, 1, 0, SKIP, 0, NULL}, + {FNAME("localAreaAddress") CHOICE, 1, 2, 2, DECODE | EXT, + offsetof(NetworkAccessParameters_networkAddress, localAreaAddress), + _H245_TransportAddress}, +}; + +static field_t _NetworkAccessParameters[] = { /* SEQUENCE */ + {FNAME("distribution") CHOICE, 1, 2, 2, SKIP | EXT | OPT, 0, + _NetworkAccessParameters_distribution}, + {FNAME("networkAddress") CHOICE, 2, 3, 3, DECODE | EXT, + offsetof(NetworkAccessParameters, networkAddress), + _NetworkAccessParameters_networkAddress}, + {FNAME("associateConference") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("externalReference") OCTSTR, 8, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("t120SetupProcedure") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, + NULL}, +}; + +static field_t _OpenLogicalChannel[] = { /* SEQUENCE */ + {FNAME("forwardLogicalChannelNumber") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("forwardLogicalChannelParameters") SEQ, 1, 3, 5, DECODE | EXT, + offsetof(OpenLogicalChannel, forwardLogicalChannelParameters), + _OpenLogicalChannel_forwardLogicalChannelParameters}, + {FNAME("reverseLogicalChannelParameters") SEQ, 1, 2, 4, + DECODE | EXT | OPT, offsetof(OpenLogicalChannel, + reverseLogicalChannelParameters), + _OpenLogicalChannel_reverseLogicalChannelParameters}, + {FNAME("separateStack") SEQ, 2, 4, 5, DECODE | EXT | OPT, + offsetof(OpenLogicalChannel, separateStack), + _NetworkAccessParameters}, + {FNAME("encryptionSync") SEQ, 2, 4, 4, STOP | EXT | OPT, 0, NULL}, +}; + +static field_t _Setup_UUIE_fastStart[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, + sizeof(OpenLogicalChannel), _OpenLogicalChannel} + , +}; + +static field_t _Setup_UUIE[] = { /* SEQUENCE */ + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(Setup_UUIE, h245Address), _TransportAddress}, + {FNAME("sourceAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _Setup_UUIE_sourceAddress}, + {FNAME("sourceInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, _EndpointType}, + {FNAME("destinationAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _Setup_UUIE_destinationAddress}, + {FNAME("destCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(Setup_UUIE, destCallSignalAddress), _TransportAddress}, + {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _Setup_UUIE_destExtraCallInfo}, + {FNAME("destExtraCRV") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _Setup_UUIE_destExtraCRV}, + {FNAME("activeMC") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("conferenceID") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, + {FNAME("conferenceGoal") CHOICE, 2, 3, 5, SKIP | EXT, 0, + _Setup_UUIE_conferenceGoal}, + {FNAME("callServices") SEQ, 0, 8, 8, SKIP | EXT | OPT, 0, + _QseriesOptions}, + {FNAME("callType") CHOICE, 2, 4, 4, SKIP | EXT, 0, _CallType}, + {FNAME("sourceCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(Setup_UUIE, sourceCallSignalAddress), _TransportAddress}, + {FNAME("remoteExtensionAddress") CHOICE, 1, 2, 7, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, + {FNAME("h245SecurityCapability") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, + offsetof(Setup_UUIE, fastStart), _Setup_UUIE_fastStart}, + {FNAME("mediaWaitForConnect") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("canOverlapSend") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, + {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("connectionParameters") SEQ, 0, 3, 3, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("language") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("presentationIndicator") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("screeningIndicator") ENUM, 2, 0, 0, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("symmetricOperationRequired") NUL, FIXD, 0, 0, SKIP | OPT, 0, + NULL}, + {FNAME("capacity") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, NULL}, + {FNAME("circuitInfo") SEQ, 3, 3, 3, SKIP | EXT | OPT, 0, NULL}, + {FNAME("desiredProtocols") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("neededFeatures") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("desiredFeatures") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedFeatures") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("parallelH245Control") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("additionalSourceAddresses") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + NULL}, +}; + +static field_t _CallProceeding_UUIE_fastStart[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, + sizeof(OpenLogicalChannel), _OpenLogicalChannel} + , +}; + +static field_t _CallProceeding_UUIE[] = { /* SEQUENCE */ + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, + _EndpointType}, + {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(CallProceeding_UUIE, h245Address), _TransportAddress}, + {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, + {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, + offsetof(CallProceeding_UUIE, fastStart), + _CallProceeding_UUIE_fastStart}, + {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, +}; + +static field_t _Connect_UUIE_fastStart[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, + sizeof(OpenLogicalChannel), _OpenLogicalChannel} + , +}; + +static field_t _Connect_UUIE[] = { /* SEQUENCE */ + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(Connect_UUIE, h245Address), _TransportAddress}, + {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, + _EndpointType}, + {FNAME("conferenceID") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, + {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, + {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, + offsetof(Connect_UUIE, fastStart), _Connect_UUIE_fastStart}, + {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("language") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("connectedAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("presentationIndicator") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("screeningIndicator") ENUM, 2, 0, 0, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("capacity") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, +}; + +static field_t _Alerting_UUIE_fastStart[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, + sizeof(OpenLogicalChannel), _OpenLogicalChannel} + , +}; + +static field_t _Alerting_UUIE[] = { /* SEQUENCE */ + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, + _EndpointType}, + {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(Alerting_UUIE, h245Address), _TransportAddress}, + {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, + {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, + offsetof(Alerting_UUIE, fastStart), _Alerting_UUIE_fastStart}, + {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("alertingAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("presentationIndicator") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("screeningIndicator") ENUM, 2, 0, 0, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("capacity") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, +}; + +static field_t _Information_UUIE_fastStart[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, + sizeof(OpenLogicalChannel), _OpenLogicalChannel} + , +}; + +static field_t _Information_UUIE[] = { /* SEQUENCE */ + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, + offsetof(Information_UUIE, fastStart), _Information_UUIE_fastStart}, + {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("circuitInfo") SEQ, 3, 3, 3, SKIP | EXT | OPT, 0, NULL}, +}; + +static field_t _ReleaseCompleteReason[] = { /* CHOICE */ + {FNAME("noBandwidth") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("gatekeeperResources") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("unreachableDestination") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("destinationRejection") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("invalidRevision") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("noPermission") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("unreachableGatekeeper") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("gatewayResources") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("badFormatAddress") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("adaptiveBusy") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("inConf") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("undefinedReason") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("facilityCallDeflection") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("securityDenied") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("calledPartyNotRegistered") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("callerNotRegistered") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("newConnectionNeeded") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("nonStandardReason") SEQ, 0, 2, 2, SKIP, 0, NULL}, + {FNAME("replaceWithConferenceInvite") OCTSTR, FIXD, 16, 0, SKIP, 0, + NULL}, + {FNAME("genericDataReason") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("neededFeatureNotSupported") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("tunnelledSignallingRejected") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _ReleaseComplete_UUIE[] = { /* SEQUENCE */ + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("reason") CHOICE, 4, 12, 22, SKIP | EXT | OPT, 0, + _ReleaseCompleteReason}, + {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("busyAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("presentationIndicator") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("screeningIndicator") ENUM, 2, 0, 0, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("capacity") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, NULL}, + {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, +}; + +static field_t _Facility_UUIE_alternativeAliasAddress[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _FacilityReason[] = { /* CHOICE */ + {FNAME("routeCallToGatekeeper") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("callForwarded") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("routeCallToMC") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("undefinedReason") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("conferenceListChoice") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("startH245") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("noH245") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("newTokens") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("featureSetUpdate") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("forwardedElements") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("transportedInformation") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _Facility_UUIE_fastStart[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, + sizeof(OpenLogicalChannel), _OpenLogicalChannel} + , +}; + +static field_t _Facility_UUIE[] = { /* SEQUENCE */ + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("alternativeAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(Facility_UUIE, alternativeAddress), _TransportAddress}, + {FNAME("alternativeAliasAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _Facility_UUIE_alternativeAliasAddress}, + {FNAME("conferenceID") OCTSTR, FIXD, 16, 0, SKIP | OPT, 0, NULL}, + {FNAME("reason") CHOICE, 2, 4, 11, DECODE | EXT, + offsetof(Facility_UUIE, reason), _FacilityReason}, + {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, + {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("remoteExtensionAddress") CHOICE, 1, 2, 7, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("conferences") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(Facility_UUIE, h245Address), _TransportAddress}, + {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, + offsetof(Facility_UUIE, fastStart), _Facility_UUIE_fastStart}, + {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("circuitInfo") SEQ, 3, 3, 3, SKIP | EXT | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, + {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT | OPT, 0, NULL}, + {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, + NULL}, +}; + +static field_t _CallIdentifier[] = { /* SEQUENCE */ + {FNAME("guid") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, +}; + +static field_t _SecurityServiceMode[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, _NonStandardParameter}, + {FNAME("none") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("default") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _SecurityCapabilities[] = { /* SEQUENCE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("encryption") CHOICE, 2, 3, 3, SKIP | EXT, 0, + _SecurityServiceMode}, + {FNAME("authenticaton") CHOICE, 2, 3, 3, SKIP | EXT, 0, + _SecurityServiceMode}, + {FNAME("integrity") CHOICE, 2, 3, 3, SKIP | EXT, 0, + _SecurityServiceMode}, +}; + +static field_t _H245Security[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, _NonStandardParameter}, + {FNAME("noSecurity") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("tls") SEQ, 1, 4, 4, SKIP | EXT, 0, _SecurityCapabilities}, + {FNAME("ipsec") SEQ, 1, 4, 4, SKIP | EXT, 0, _SecurityCapabilities}, +}; + +static field_t _DHset[] = { /* SEQUENCE */ + {FNAME("halfkey") BITSTR, WORD, 0, 0, SKIP, 0, NULL}, + {FNAME("modSize") BITSTR, WORD, 0, 0, SKIP, 0, NULL}, + {FNAME("generator") BITSTR, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _TypedCertificate[] = { /* SEQUENCE */ + {FNAME("type") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("certificate") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H235_NonStandardParameter[] = { /* SEQUENCE */ + {FNAME("nonStandardIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("data") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _ClearToken[] = { /* SEQUENCE */ + {FNAME("tokenOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("timeStamp") INT, CONS, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("password") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("dhkey") SEQ, 0, 3, 3, SKIP | EXT | OPT, 0, _DHset}, + {FNAME("challenge") OCTSTR, 7, 8, 0, SKIP | OPT, 0, NULL}, + {FNAME("random") INT, UNCO, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("certificate") SEQ, 0, 2, 2, SKIP | EXT | OPT, 0, + _TypedCertificate}, + {FNAME("generalID") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP | OPT, 0, + _H235_NonStandardParameter}, + {FNAME("eckasdhkey") CHOICE, 1, 2, 2, SKIP | EXT | OPT, 0, NULL}, + {FNAME("sendersID") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _Progress_UUIE_tokens[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 8, 9, 11, SKIP | EXT, 0, _ClearToken}, +}; + +static field_t _Params[] = { /* SEQUENCE */ + {FNAME("ranInt") INT, UNCO, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("iv8") OCTSTR, FIXD, 8, 0, SKIP | OPT, 0, NULL}, + {FNAME("iv16") OCTSTR, FIXD, 16, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _CryptoH323Token_cryptoEPPwdHash_token[] = { /* SEQUENCE */ + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("hash") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoH323Token_cryptoEPPwdHash[] = { /* SEQUENCE */ + {FNAME("alias") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, + {FNAME("timeStamp") INT, CONS, 1, 0, SKIP, 0, NULL}, + {FNAME("token") SEQ, 0, 3, 3, SKIP, 0, + _CryptoH323Token_cryptoEPPwdHash_token}, +}; + +static field_t _CryptoH323Token_cryptoGKPwdHash_token[] = { /* SEQUENCE */ + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("hash") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoH323Token_cryptoGKPwdHash[] = { /* SEQUENCE */ + {FNAME("gatekeeperId") BMPSTR, 7, 1, 0, SKIP, 0, NULL}, + {FNAME("timeStamp") INT, CONS, 1, 0, SKIP, 0, NULL}, + {FNAME("token") SEQ, 0, 3, 3, SKIP, 0, + _CryptoH323Token_cryptoGKPwdHash_token}, +}; + +static field_t _CryptoH323Token_cryptoEPPwdEncr[] = { /* SEQUENCE */ + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("encryptedData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoH323Token_cryptoGKPwdEncr[] = { /* SEQUENCE */ + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("encryptedData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoH323Token_cryptoEPCert[] = { /* SEQUENCE */ + {FNAME("toBeSigned") SEQ, 8, 9, 11, SKIP | OPEN | EXT, 0, NULL}, + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("signature") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoH323Token_cryptoGKCert[] = { /* SEQUENCE */ + {FNAME("toBeSigned") SEQ, 8, 9, 11, SKIP | OPEN | EXT, 0, NULL}, + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("signature") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoH323Token_cryptoFastStart[] = { /* SEQUENCE */ + {FNAME("toBeSigned") SEQ, 8, 9, 11, SKIP | OPEN | EXT, 0, NULL}, + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("signature") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoToken_cryptoEncryptedToken_token[] = { /* SEQUENCE */ + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("encryptedData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoToken_cryptoEncryptedToken[] = { /* SEQUENCE */ + {FNAME("tokenOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("token") SEQ, 0, 3, 3, SKIP, 0, + _CryptoToken_cryptoEncryptedToken_token}, +}; + +static field_t _CryptoToken_cryptoSignedToken_token[] = { /* SEQUENCE */ + {FNAME("toBeSigned") SEQ, 8, 9, 11, SKIP | OPEN | EXT, 0, NULL}, + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("signature") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoToken_cryptoSignedToken[] = { /* SEQUENCE */ + {FNAME("tokenOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("token") SEQ, 0, 4, 4, SKIP, 0, + _CryptoToken_cryptoSignedToken_token}, +}; + +static field_t _CryptoToken_cryptoHashedToken_token[] = { /* SEQUENCE */ + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("hash") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoToken_cryptoHashedToken[] = { /* SEQUENCE */ + {FNAME("tokenOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("hashedVals") SEQ, 8, 9, 11, SKIP | EXT, 0, _ClearToken}, + {FNAME("token") SEQ, 0, 3, 3, SKIP, 0, + _CryptoToken_cryptoHashedToken_token}, +}; + +static field_t _CryptoToken_cryptoPwdEncr[] = { /* SEQUENCE */ + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("encryptedData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoToken[] = { /* CHOICE */ + {FNAME("cryptoEncryptedToken") SEQ, 0, 2, 2, SKIP, 0, + _CryptoToken_cryptoEncryptedToken}, + {FNAME("cryptoSignedToken") SEQ, 0, 2, 2, SKIP, 0, + _CryptoToken_cryptoSignedToken}, + {FNAME("cryptoHashedToken") SEQ, 0, 3, 3, SKIP, 0, + _CryptoToken_cryptoHashedToken}, + {FNAME("cryptoPwdEncr") SEQ, 0, 3, 3, SKIP, 0, + _CryptoToken_cryptoPwdEncr}, +}; + +static field_t _CryptoH323Token[] = { /* CHOICE */ + {FNAME("cryptoEPPwdHash") SEQ, 0, 3, 3, SKIP, 0, + _CryptoH323Token_cryptoEPPwdHash}, + {FNAME("cryptoGKPwdHash") SEQ, 0, 3, 3, SKIP, 0, + _CryptoH323Token_cryptoGKPwdHash}, + {FNAME("cryptoEPPwdEncr") SEQ, 0, 3, 3, SKIP, 0, + _CryptoH323Token_cryptoEPPwdEncr}, + {FNAME("cryptoGKPwdEncr") SEQ, 0, 3, 3, SKIP, 0, + _CryptoH323Token_cryptoGKPwdEncr}, + {FNAME("cryptoEPCert") SEQ, 0, 4, 4, SKIP, 0, + _CryptoH323Token_cryptoEPCert}, + {FNAME("cryptoGKCert") SEQ, 0, 4, 4, SKIP, 0, + _CryptoH323Token_cryptoGKCert}, + {FNAME("cryptoFastStart") SEQ, 0, 4, 4, SKIP, 0, + _CryptoH323Token_cryptoFastStart}, + {FNAME("nestedcryptoToken") CHOICE, 2, 4, 4, SKIP | EXT, 0, + _CryptoToken}, +}; + +static field_t _Progress_UUIE_cryptoTokens[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 3, 8, 8, SKIP | EXT, 0, _CryptoH323Token}, +}; + +static field_t _Progress_UUIE_fastStart[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, + sizeof(OpenLogicalChannel), _OpenLogicalChannel} + , +}; + +static field_t _Progress_UUIE[] = { /* SEQUENCE */ + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, + _EndpointType}, + {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(Progress_UUIE, h245Address), _TransportAddress}, + {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, + _CallIdentifier}, + {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, + _H245Security}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _Progress_UUIE_tokens}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _Progress_UUIE_cryptoTokens}, + {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, + offsetof(Progress_UUIE, fastStart), _Progress_UUIE_fastStart}, + {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _H323_UU_PDU_h323_message_body[] = { /* CHOICE */ + {FNAME("setup") SEQ, 7, 13, 39, DECODE | EXT, + offsetof(H323_UU_PDU_h323_message_body, setup), _Setup_UUIE}, + {FNAME("callProceeding") SEQ, 1, 3, 12, DECODE | EXT, + offsetof(H323_UU_PDU_h323_message_body, callProceeding), + _CallProceeding_UUIE}, + {FNAME("connect") SEQ, 1, 4, 19, DECODE | EXT, + offsetof(H323_UU_PDU_h323_message_body, connect), _Connect_UUIE}, + {FNAME("alerting") SEQ, 1, 3, 17, DECODE | EXT, + offsetof(H323_UU_PDU_h323_message_body, alerting), _Alerting_UUIE}, + {FNAME("information") SEQ, 0, 1, 7, DECODE | EXT, + offsetof(H323_UU_PDU_h323_message_body, information), + _Information_UUIE}, + {FNAME("releaseComplete") SEQ, 1, 2, 11, SKIP | EXT, 0, + _ReleaseComplete_UUIE}, + {FNAME("facility") SEQ, 3, 5, 21, DECODE | EXT, + offsetof(H323_UU_PDU_h323_message_body, facility), _Facility_UUIE}, + {FNAME("progress") SEQ, 5, 8, 11, DECODE | EXT, + offsetof(H323_UU_PDU_h323_message_body, progress), _Progress_UUIE}, + {FNAME("empty") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("status") SEQ, 2, 4, 4, SKIP | EXT, 0, NULL}, + {FNAME("statusInquiry") SEQ, 2, 4, 4, SKIP | EXT, 0, NULL}, + {FNAME("setupAcknowledge") SEQ, 2, 4, 4, SKIP | EXT, 0, NULL}, + {FNAME("notify") SEQ, 2, 4, 4, SKIP | EXT, 0, NULL}, +}; + +static field_t _RequestMessage[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("masterSlaveDetermination") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, + {FNAME("terminalCapabilitySet") SEQ, 3, 5, 5, STOP | EXT, 0, NULL}, + {FNAME("openLogicalChannel") SEQ, 1, 3, 5, DECODE | EXT, + offsetof(RequestMessage, openLogicalChannel), _OpenLogicalChannel}, + {FNAME("closeLogicalChannel") SEQ, 0, 2, 3, STOP | EXT, 0, NULL}, + {FNAME("requestChannelClose") SEQ, 0, 1, 3, STOP | EXT, 0, NULL}, + {FNAME("multiplexEntrySend") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, + {FNAME("requestMultiplexEntry") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("requestMode") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, + {FNAME("roundTripDelayRequest") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("maintenanceLoopRequest") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("communicationModeRequest") SEQ, 0, 0, 0, STOP | EXT, 0, NULL}, + {FNAME("conferenceRequest") CHOICE, 3, 8, 16, STOP | EXT, 0, NULL}, + {FNAME("multilinkRequest") CHOICE, 3, 5, 5, STOP | EXT, 0, NULL}, + {FNAME("logicalChannelRateRequest") SEQ, 0, 3, 3, STOP | EXT, 0, + NULL}, +}; + +static field_t _OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters[] = { /* CHOICE */ + {FNAME("h222LogicalChannelParameters") SEQ, 3, 5, 5, SKIP | EXT, 0, + _H222LogicalChannelParameters}, + {FNAME("h2250LogicalChannelParameters") SEQ, 10, 11, 14, DECODE | EXT, + offsetof + (OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters, + h2250LogicalChannelParameters), _H2250LogicalChannelParameters}, +}; + +static field_t _OpenLogicalChannelAck_reverseLogicalChannelParameters[] = { /* SEQUENCE */ + {FNAME("reverseLogicalChannelNumber") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("portNumber") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("multiplexParameters") CHOICE, 0, 1, 2, DECODE | EXT | OPT, + offsetof(OpenLogicalChannelAck_reverseLogicalChannelParameters, + multiplexParameters), + _OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters}, + {FNAME("replacementFor") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _H2250LogicalChannelAckParameters_nonStandard[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 0, 2, 2, SKIP, 0, _H245_NonStandardParameter}, +}; + +static field_t _H2250LogicalChannelAckParameters[] = { /* SEQUENCE */ + {FNAME("nonStandard") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _H2250LogicalChannelAckParameters_nonStandard}, + {FNAME("sessionID") INT, 8, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("mediaChannel") CHOICE, 1, 2, 2, DECODE | EXT | OPT, + offsetof(H2250LogicalChannelAckParameters, mediaChannel), + _H245_TransportAddress}, + {FNAME("mediaControlChannel") CHOICE, 1, 2, 2, DECODE | EXT | OPT, + offsetof(H2250LogicalChannelAckParameters, mediaControlChannel), + _H245_TransportAddress}, + {FNAME("dynamicRTPPayloadType") INT, 5, 96, 0, SKIP | OPT, 0, NULL}, + {FNAME("flowControlToZero") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("portNumber") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _OpenLogicalChannelAck_forwardMultiplexAckParameters[] = { /* CHOICE */ + {FNAME("h2250LogicalChannelAckParameters") SEQ, 5, 5, 7, DECODE | EXT, + offsetof(OpenLogicalChannelAck_forwardMultiplexAckParameters, + h2250LogicalChannelAckParameters), + _H2250LogicalChannelAckParameters}, +}; + +static field_t _OpenLogicalChannelAck[] = { /* SEQUENCE */ + {FNAME("forwardLogicalChannelNumber") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("reverseLogicalChannelParameters") SEQ, 2, 3, 4, + DECODE | EXT | OPT, offsetof(OpenLogicalChannelAck, + reverseLogicalChannelParameters), + _OpenLogicalChannelAck_reverseLogicalChannelParameters}, + {FNAME("separateStack") SEQ, 2, 4, 5, SKIP | EXT | OPT, 0, NULL}, + {FNAME("forwardMultiplexAckParameters") CHOICE, 0, 1, 1, + DECODE | EXT | OPT, offsetof(OpenLogicalChannelAck, + forwardMultiplexAckParameters), + _OpenLogicalChannelAck_forwardMultiplexAckParameters}, + {FNAME("encryptionSync") SEQ, 2, 4, 4, STOP | EXT | OPT, 0, NULL}, +}; + +static field_t _ResponseMessage[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("masterSlaveDeterminationAck") SEQ, 0, 1, 1, STOP | EXT, 0, + NULL}, + {FNAME("masterSlaveDeterminationReject") SEQ, 0, 1, 1, STOP | EXT, 0, + NULL}, + {FNAME("terminalCapabilitySetAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("terminalCapabilitySetReject") SEQ, 0, 2, 2, STOP | EXT, 0, + NULL}, + {FNAME("openLogicalChannelAck") SEQ, 1, 2, 5, DECODE | EXT, + offsetof(ResponseMessage, openLogicalChannelAck), + _OpenLogicalChannelAck}, + {FNAME("openLogicalChannelReject") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, + {FNAME("closeLogicalChannelAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("requestChannelCloseAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("requestChannelCloseReject") SEQ, 0, 2, 2, STOP | EXT, 0, + NULL}, + {FNAME("multiplexEntrySendAck") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, + {FNAME("multiplexEntrySendReject") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, + {FNAME("requestMultiplexEntryAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("requestMultiplexEntryReject") SEQ, 0, 2, 2, STOP | EXT, 0, + NULL}, + {FNAME("requestModeAck") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, + {FNAME("requestModeReject") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, + {FNAME("roundTripDelayResponse") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("maintenanceLoopAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("maintenanceLoopReject") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, + {FNAME("communicationModeResponse") CHOICE, 0, 1, 1, STOP | EXT, 0, + NULL}, + {FNAME("conferenceResponse") CHOICE, 3, 8, 16, STOP | EXT, 0, NULL}, + {FNAME("multilinkResponse") CHOICE, 3, 5, 5, STOP | EXT, 0, NULL}, + {FNAME("logicalChannelRateAcknowledge") SEQ, 0, 3, 3, STOP | EXT, 0, + NULL}, + {FNAME("logicalChannelRateReject") SEQ, 1, 4, 4, STOP | EXT, 0, NULL}, +}; + +static field_t _MultimediaSystemControlMessage[] = { /* CHOICE */ + {FNAME("request") CHOICE, 4, 11, 15, DECODE | EXT, + offsetof(MultimediaSystemControlMessage, request), _RequestMessage}, + {FNAME("response") CHOICE, 5, 19, 24, DECODE | EXT, + offsetof(MultimediaSystemControlMessage, response), + _ResponseMessage}, + {FNAME("command") CHOICE, 3, 7, 12, STOP | EXT, 0, NULL}, + {FNAME("indication") CHOICE, 4, 14, 23, STOP | EXT, 0, NULL}, +}; + +static field_t _H323_UU_PDU_h245Control[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 2, 4, 4, DECODE | OPEN | EXT, + sizeof(MultimediaSystemControlMessage), + _MultimediaSystemControlMessage} + , +}; + +static field_t _H323_UU_PDU[] = { /* SEQUENCE */ + {FNAME("h323-message-body") CHOICE, 3, 7, 13, DECODE | EXT, + offsetof(H323_UU_PDU, h323_message_body), + _H323_UU_PDU_h323_message_body}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("h4501SupplementaryService") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + NULL}, + {FNAME("h245Tunneling") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("h245Control") SEQOF, SEMI, 0, 4, DECODE | OPT, + offsetof(H323_UU_PDU, h245Control), _H323_UU_PDU_h245Control}, + {FNAME("nonStandardControl") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("callLinkage") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, + {FNAME("tunnelledSignallingMessage") SEQ, 2, 4, 4, STOP | EXT | OPT, + 0, NULL}, + {FNAME("provisionalRespToH245Tunneling") NUL, FIXD, 0, 0, STOP | OPT, + 0, NULL}, + {FNAME("stimulusControl") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _H323_UserInformation[] = { /* SEQUENCE */ + {FNAME("h323-uu-pdu") SEQ, 1, 2, 11, DECODE | EXT, + offsetof(H323_UserInformation, h323_uu_pdu), _H323_UU_PDU}, + {FNAME("user-data") SEQ, 0, 2, 2, STOP | EXT | OPT, 0, NULL}, +}; + +static field_t _GatekeeperRequest[] = { /* SEQUENCE */ + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("rasAddress") CHOICE, 3, 7, 7, DECODE | EXT, + offsetof(GatekeeperRequest, rasAddress), _TransportAddress}, + {FNAME("endpointType") SEQ, 6, 8, 10, STOP | EXT, 0, NULL}, + {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, + {FNAME("callServices") SEQ, 0, 8, 8, STOP | EXT | OPT, 0, NULL}, + {FNAME("endpointAlias") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("authenticationCapability") SEQOF, SEMI, 0, 0, STOP | OPT, 0, + NULL}, + {FNAME("algorithmOIDs") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrity") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("supportsAltGK") NUL, FIXD, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _GatekeeperConfirm[] = { /* SEQUENCE */ + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("rasAddress") CHOICE, 3, 7, 7, DECODE | EXT, + offsetof(GatekeeperConfirm, rasAddress), _TransportAddress}, + {FNAME("alternateGatekeeper") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("authenticationMode") CHOICE, 3, 7, 8, STOP | EXT | OPT, 0, + NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("algorithmOID") OID, BYTE, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrity") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _RegistrationRequest_callSignalAddress[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, + sizeof(TransportAddress), _TransportAddress} + , +}; + +static field_t _RegistrationRequest_rasAddress[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, + sizeof(TransportAddress), _TransportAddress} + , +}; + +static field_t _RegistrationRequest_terminalAlias[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _RegistrationRequest[] = { /* SEQUENCE */ + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("discoveryComplete") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("callSignalAddress") SEQOF, SEMI, 0, 10, DECODE, + offsetof(RegistrationRequest, callSignalAddress), + _RegistrationRequest_callSignalAddress}, + {FNAME("rasAddress") SEQOF, SEMI, 0, 10, DECODE, + offsetof(RegistrationRequest, rasAddress), + _RegistrationRequest_rasAddress}, + {FNAME("terminalType") SEQ, 6, 8, 10, SKIP | EXT, 0, _EndpointType}, + {FNAME("terminalAlias") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _RegistrationRequest_terminalAlias}, + {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("endpointVendor") SEQ, 2, 3, 3, SKIP | EXT, 0, + _VendorIdentifier}, + {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("timeToLive") INT, CONS, 1, 0, DECODE | OPT, + offsetof(RegistrationRequest, timeToLive), NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("keepAlive") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, + {FNAME("willSupplyUUIEs") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("maintainConnection") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("alternateTransportAddresses") SEQ, 1, 1, 1, STOP | EXT | OPT, + 0, NULL}, + {FNAME("additiveRegistration") NUL, FIXD, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("terminalAliasPattern") SEQOF, SEMI, 0, 0, STOP | OPT, 0, + NULL}, + {FNAME("supportsAltGK") NUL, FIXD, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("usageReportingCapability") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, + NULL}, + {FNAME("multipleCalls") BOOL, FIXD, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("supportedH248Packages") SEQOF, SEMI, 0, 0, STOP | OPT, 0, + NULL}, + {FNAME("callCreditCapability") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, + NULL}, + {FNAME("capacityReportingCapability") SEQ, 0, 1, 1, STOP | EXT | OPT, + 0, NULL}, + {FNAME("capacity") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _RegistrationConfirm_callSignalAddress[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, + sizeof(TransportAddress), _TransportAddress} + , +}; + +static field_t _RegistrationConfirm_terminalAlias[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _RegistrationConfirm[] = { /* SEQUENCE */ + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("callSignalAddress") SEQOF, SEMI, 0, 10, DECODE, + offsetof(RegistrationConfirm, callSignalAddress), + _RegistrationConfirm_callSignalAddress}, + {FNAME("terminalAlias") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _RegistrationConfirm_terminalAlias}, + {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, SKIP, 0, NULL}, + {FNAME("alternateGatekeeper") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("timeToLive") INT, CONS, 1, 0, DECODE | OPT, + offsetof(RegistrationConfirm, timeToLive), NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("willRespondToIRR") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("preGrantedARQ") SEQ, 0, 4, 8, STOP | EXT | OPT, 0, NULL}, + {FNAME("maintainConnection") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("serviceControl") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("supportsAdditiveRegistration") NUL, FIXD, 0, 0, STOP | OPT, 0, + NULL}, + {FNAME("terminalAliasPattern") SEQOF, SEMI, 0, 0, STOP | OPT, 0, + NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("usageSpec") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("featureServerAlias") CHOICE, 1, 2, 7, STOP | EXT | OPT, 0, + NULL}, + {FNAME("capacityReportingSpec") SEQ, 0, 1, 1, STOP | EXT | OPT, 0, + NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _UnregistrationRequest_callSignalAddress[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, + sizeof(TransportAddress), _TransportAddress} + , +}; + +static field_t _UnregistrationRequest[] = { /* SEQUENCE */ + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("callSignalAddress") SEQOF, SEMI, 0, 10, DECODE, + offsetof(UnregistrationRequest, callSignalAddress), + _UnregistrationRequest_callSignalAddress}, + {FNAME("endpointAlias") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, + {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("reason") CHOICE, 2, 4, 5, STOP | EXT | OPT, 0, NULL}, + {FNAME("endpointAliasPattern") SEQOF, SEMI, 0, 0, STOP | OPT, 0, + NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("alternateGatekeeper") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _CallModel[] = { /* CHOICE */ + {FNAME("direct") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("gatekeeperRouted") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _AdmissionRequest_destinationInfo[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _AdmissionRequest_destExtraCallInfo[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _AdmissionRequest_srcInfo[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _AdmissionRequest[] = { /* SEQUENCE */ + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("callType") CHOICE, 2, 4, 4, SKIP | EXT, 0, _CallType}, + {FNAME("callModel") CHOICE, 1, 2, 2, SKIP | EXT | OPT, 0, _CallModel}, + {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, SKIP, 0, NULL}, + {FNAME("destinationInfo") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _AdmissionRequest_destinationInfo}, + {FNAME("destCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(AdmissionRequest, destCallSignalAddress), + _TransportAddress}, + {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _AdmissionRequest_destExtraCallInfo}, + {FNAME("srcInfo") SEQOF, SEMI, 0, 0, SKIP, 0, + _AdmissionRequest_srcInfo}, + {FNAME("srcCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(AdmissionRequest, srcCallSignalAddress), _TransportAddress}, + {FNAME("bandWidth") INT, CONS, 0, 0, STOP, 0, NULL}, + {FNAME("callReferenceValue") INT, WORD, 0, 0, STOP, 0, NULL}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("callServices") SEQ, 0, 8, 8, STOP | EXT | OPT, 0, NULL}, + {FNAME("conferenceID") OCTSTR, FIXD, 16, 0, STOP, 0, NULL}, + {FNAME("activeMC") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("answerCall") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("canMapAlias") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("callIdentifier") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("srcAlternatives") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("destAlternatives") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("transportQOS") CHOICE, 2, 3, 3, STOP | EXT | OPT, 0, NULL}, + {FNAME("willSupplyUUIEs") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("callLinkage") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, + {FNAME("gatewayDataRate") SEQ, 2, 3, 3, STOP | EXT | OPT, 0, NULL}, + {FNAME("capacity") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, + {FNAME("circuitInfo") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, + {FNAME("desiredProtocols") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("desiredTunnelledProtocol") SEQ, 1, 2, 2, STOP | EXT | OPT, 0, + NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _AdmissionConfirm[] = { /* SEQUENCE */ + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("bandWidth") INT, CONS, 0, 0, SKIP, 0, NULL}, + {FNAME("callModel") CHOICE, 1, 2, 2, SKIP | EXT, 0, _CallModel}, + {FNAME("destCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT, + offsetof(AdmissionConfirm, destCallSignalAddress), + _TransportAddress}, + {FNAME("irrFrequency") INT, WORD, 1, 0, STOP | OPT, 0, NULL}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("destinationInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("destinationType") SEQ, 6, 8, 10, STOP | EXT | OPT, 0, NULL}, + {FNAME("remoteExtensionAddress") SEQOF, SEMI, 0, 0, STOP | OPT, 0, + NULL}, + {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("transportQOS") CHOICE, 2, 3, 3, STOP | EXT | OPT, 0, NULL}, + {FNAME("willRespondToIRR") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("uuiesRequested") SEQ, 0, 9, 13, STOP | EXT, 0, NULL}, + {FNAME("language") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("alternateTransportAddresses") SEQ, 1, 1, 1, STOP | EXT | OPT, + 0, NULL}, + {FNAME("useSpecifiedTransport") CHOICE, 1, 2, 2, STOP | EXT | OPT, 0, + NULL}, + {FNAME("circuitInfo") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, + {FNAME("usageSpec") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("supportedProtocols") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("serviceControl") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("multipleCalls") BOOL, FIXD, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _LocationRequest_destinationInfo[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _LocationRequest[] = { /* SEQUENCE */ + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("destinationInfo") SEQOF, SEMI, 0, 0, SKIP, 0, + _LocationRequest_destinationInfo}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("replyAddress") CHOICE, 3, 7, 7, DECODE | EXT, + offsetof(LocationRequest, replyAddress), _TransportAddress}, + {FNAME("sourceInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("canMapAlias") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("desiredProtocols") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("desiredTunnelledProtocol") SEQ, 1, 2, 2, STOP | EXT | OPT, 0, + NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("hopCount") INT, 8, 1, 0, STOP | OPT, 0, NULL}, + {FNAME("circuitInfo") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, +}; + +static field_t _LocationConfirm[] = { /* SEQUENCE */ + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("callSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT, + offsetof(LocationConfirm, callSignalAddress), _TransportAddress}, + {FNAME("rasAddress") CHOICE, 3, 7, 7, DECODE | EXT, + offsetof(LocationConfirm, rasAddress), _TransportAddress}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("destinationInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("destinationType") SEQ, 6, 8, 10, STOP | EXT | OPT, 0, NULL}, + {FNAME("remoteExtensionAddress") SEQOF, SEMI, 0, 0, STOP | OPT, 0, + NULL}, + {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("alternateTransportAddresses") SEQ, 1, 1, 1, STOP | EXT | OPT, + 0, NULL}, + {FNAME("supportedProtocols") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("multipleCalls") BOOL, FIXD, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("circuitInfo") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, + {FNAME("serviceControl") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _InfoRequestResponse_callSignalAddress[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, + sizeof(TransportAddress), _TransportAddress} + , +}; + +static field_t _InfoRequestResponse[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("endpointType") SEQ, 6, 8, 10, SKIP | EXT, 0, _EndpointType}, + {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, SKIP, 0, NULL}, + {FNAME("rasAddress") CHOICE, 3, 7, 7, DECODE | EXT, + offsetof(InfoRequestResponse, rasAddress), _TransportAddress}, + {FNAME("callSignalAddress") SEQOF, SEMI, 0, 10, DECODE, + offsetof(InfoRequestResponse, callSignalAddress), + _InfoRequestResponse_callSignalAddress}, + {FNAME("endpointAlias") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("perCallInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("needResponse") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("capacity") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, + {FNAME("irrStatus") CHOICE, 2, 4, 4, STOP | EXT | OPT, 0, NULL}, + {FNAME("unsolicited") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _RasMessage[] = { /* CHOICE */ + {FNAME("gatekeeperRequest") SEQ, 4, 8, 18, DECODE | EXT, + offsetof(RasMessage, gatekeeperRequest), _GatekeeperRequest}, + {FNAME("gatekeeperConfirm") SEQ, 2, 5, 14, DECODE | EXT, + offsetof(RasMessage, gatekeeperConfirm), _GatekeeperConfirm}, + {FNAME("gatekeeperReject") SEQ, 2, 5, 11, STOP | EXT, 0, NULL}, + {FNAME("registrationRequest") SEQ, 3, 10, 31, DECODE | EXT, + offsetof(RasMessage, registrationRequest), _RegistrationRequest}, + {FNAME("registrationConfirm") SEQ, 3, 7, 24, DECODE | EXT, + offsetof(RasMessage, registrationConfirm), _RegistrationConfirm}, + {FNAME("registrationReject") SEQ, 2, 5, 11, STOP | EXT, 0, NULL}, + {FNAME("unregistrationRequest") SEQ, 3, 5, 15, DECODE | EXT, + offsetof(RasMessage, unregistrationRequest), _UnregistrationRequest}, + {FNAME("unregistrationConfirm") SEQ, 1, 2, 6, STOP | EXT, 0, NULL}, + {FNAME("unregistrationReject") SEQ, 1, 3, 8, STOP | EXT, 0, NULL}, + {FNAME("admissionRequest") SEQ, 7, 16, 34, DECODE | EXT, + offsetof(RasMessage, admissionRequest), _AdmissionRequest}, + {FNAME("admissionConfirm") SEQ, 2, 6, 27, DECODE | EXT, + offsetof(RasMessage, admissionConfirm), _AdmissionConfirm}, + {FNAME("admissionReject") SEQ, 1, 3, 11, STOP | EXT, 0, NULL}, + {FNAME("bandwidthRequest") SEQ, 2, 7, 18, STOP | EXT, 0, NULL}, + {FNAME("bandwidthConfirm") SEQ, 1, 3, 8, STOP | EXT, 0, NULL}, + {FNAME("bandwidthReject") SEQ, 1, 4, 9, STOP | EXT, 0, NULL}, + {FNAME("disengageRequest") SEQ, 1, 6, 19, STOP | EXT, 0, NULL}, + {FNAME("disengageConfirm") SEQ, 1, 2, 9, STOP | EXT, 0, NULL}, + {FNAME("disengageReject") SEQ, 1, 3, 8, STOP | EXT, 0, NULL}, + {FNAME("locationRequest") SEQ, 2, 5, 17, DECODE | EXT, + offsetof(RasMessage, locationRequest), _LocationRequest}, + {FNAME("locationConfirm") SEQ, 1, 4, 19, DECODE | EXT, + offsetof(RasMessage, locationConfirm), _LocationConfirm}, + {FNAME("locationReject") SEQ, 1, 3, 10, STOP | EXT, 0, NULL}, + {FNAME("infoRequest") SEQ, 2, 4, 15, STOP | EXT, 0, NULL}, + {FNAME("infoRequestResponse") SEQ, 3, 8, 16, DECODE | EXT, + offsetof(RasMessage, infoRequestResponse), _InfoRequestResponse}, + {FNAME("nonStandardMessage") SEQ, 0, 2, 7, STOP | EXT, 0, NULL}, + {FNAME("unknownMessageResponse") SEQ, 0, 1, 5, STOP | EXT, 0, NULL}, + {FNAME("requestInProgress") SEQ, 4, 6, 6, STOP | EXT, 0, NULL}, + {FNAME("resourcesAvailableIndicate") SEQ, 4, 9, 11, STOP | EXT, 0, + NULL}, + {FNAME("resourcesAvailableConfirm") SEQ, 4, 6, 7, STOP | EXT, 0, + NULL}, + {FNAME("infoRequestAck") SEQ, 4, 5, 5, STOP | EXT, 0, NULL}, + {FNAME("infoRequestNak") SEQ, 5, 7, 7, STOP | EXT, 0, NULL}, + {FNAME("serviceControlIndication") SEQ, 8, 10, 10, STOP | EXT, 0, + NULL}, + {FNAME("serviceControlResponse") SEQ, 7, 8, 8, STOP | EXT, 0, NULL}, +}; -- cgit v1.2.3-70-g09d2 From 869f37d8e48f3911eb70f38a994feaa8f8380008 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 2 Dec 2006 22:09:06 -0800 Subject: [NETFILTER]: nf_conntrack/nf_nat: add IRC helper port Add nf_conntrack port of the IRC conntrack/NAT helper. Since DCC doesn't support IPv6 yet, the helper is still IPv4 only. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_irc.h | 15 ++ net/ipv4/netfilter/Kconfig | 5 + net/ipv4/netfilter/Makefile | 1 + net/ipv4/netfilter/nf_nat_irc.c | 101 +++++++++++ net/netfilter/Kconfig | 15 ++ net/netfilter/Makefile | 1 + net/netfilter/nf_conntrack_irc.c | 281 +++++++++++++++++++++++++++++ 7 files changed, 419 insertions(+) create mode 100644 include/linux/netfilter/nf_conntrack_irc.h create mode 100644 net/ipv4/netfilter/nf_nat_irc.c create mode 100644 net/netfilter/nf_conntrack_irc.c (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_irc.h b/include/linux/netfilter/nf_conntrack_irc.h new file mode 100644 index 00000000000..2ab6b825591 --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_irc.h @@ -0,0 +1,15 @@ +#ifndef _NF_CONNTRACK_IRC_H +#define _NF_CONNTRACK_IRC_H + +#ifdef __KERNEL__ + +#define IRC_PORT 6667 + +extern unsigned int (*nf_nat_irc_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conntrack_expect *exp); + +#endif /* __KERNEL__ */ +#endif /* _NF_CONNTRACK_IRC_H */ diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index e14156d1122..4555f721dfc 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -500,6 +500,11 @@ config IP_NF_NAT_IRC default IP_NF_NAT if IP_NF_IRC=y default m if IP_NF_IRC=m +config NF_NAT_IRC + tristate + depends on IP_NF_IPTABLES && NF_CONNTRACK && NF_NAT + default NF_NAT && NF_CONNTRACK_IRC + config IP_NF_NAT_TFTP tristate depends on IP_NF_IPTABLES!=n && IP_NF_CONNTRACK!=n && IP_NF_NAT!=n diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index bdaba4700e3..56733c37032 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -53,6 +53,7 @@ obj-$(CONFIG_IP_NF_NAT_SIP) += ip_nat_sip.o obj-$(CONFIG_NF_NAT_AMANDA) += nf_nat_amanda.o obj-$(CONFIG_NF_NAT_FTP) += nf_nat_ftp.o obj-$(CONFIG_NF_NAT_H323) += nf_nat_h323.o +obj-$(CONFIG_NF_NAT_IRC) += nf_nat_irc.o # generic IP tables obj-$(CONFIG_IP_NF_IPTABLES) += ip_tables.o diff --git a/net/ipv4/netfilter/nf_nat_irc.c b/net/ipv4/netfilter/nf_nat_irc.c new file mode 100644 index 00000000000..9b8c0daea74 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_irc.c @@ -0,0 +1,101 @@ +/* IRC extension for TCP NAT alteration. + * + * (C) 2000-2001 by Harald Welte + * (C) 2004 Rusty Russell IBM Corporation + * based on a copy of RR's ip_nat_ftp.c + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +MODULE_AUTHOR("Harald Welte "); +MODULE_DESCRIPTION("IRC (DCC) NAT helper"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ip_nat_irc"); + +static unsigned int help(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conntrack_expect *exp) +{ + char buffer[sizeof("4294967296 65635")]; + u_int32_t ip; + u_int16_t port; + unsigned int ret; + + DEBUGP("IRC_NAT: info (seq %u + %u) in %u\n", + expect->seq, exp_irc_info->len, ntohl(tcph->seq)); + + /* Reply comes from server. */ + exp->saved_proto.tcp.port = exp->tuple.dst.u.tcp.port; + exp->dir = IP_CT_DIR_REPLY; + exp->expectfn = nf_nat_follow_master; + + /* Try to get same port: if not, try to change it. */ + for (port = ntohs(exp->saved_proto.tcp.port); port != 0; port++) { + exp->tuple.dst.u.tcp.port = htons(port); + if (nf_conntrack_expect_related(exp) == 0) + break; + } + + if (port == 0) + return NF_DROP; + + ip = ntohl(exp->master->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip); + sprintf(buffer, "%u %u", ip, port); + DEBUGP("nf_nat_irc: inserting '%s' == %u.%u.%u.%u, port %u\n", + buffer, NIPQUAD(ip), port); + + ret = nf_nat_mangle_tcp_packet(pskb, exp->master, ctinfo, + matchoff, matchlen, buffer, + strlen(buffer)); + if (ret != NF_ACCEPT) + nf_conntrack_unexpect_related(exp); + return ret; +} + +static void __exit nf_nat_irc_fini(void) +{ + rcu_assign_pointer(nf_nat_irc_hook, NULL); + synchronize_rcu(); +} + +static int __init nf_nat_irc_init(void) +{ + BUG_ON(rcu_dereference(nf_nat_irc_hook)); + rcu_assign_pointer(nf_nat_irc_hook, help); + return 0; +} + +/* Prior to 2.6.11, we had a ports param. No longer, but don't break users. */ +static int warn_set(const char *val, struct kernel_param *kp) +{ + printk(KERN_INFO KBUILD_MODNAME + ": kernel >= 2.6.10 only uses 'ports' for conntrack modules\n"); + return 0; +} +module_param_call(ports, warn_set, NULL, NULL, 0); + +module_init(nf_nat_irc_init); +module_exit(nf_nat_irc_fini); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index d8f3451c95b..bd50897d8fb 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -179,6 +179,21 @@ config NF_CONNTRACK_H323 To compile it as a module, choose M here. If unsure, say N. +config NF_CONNTRACK_IRC + tristate "IRC protocol support (EXPERIMENTAL)" + depends on EXPERIMENTAL && NF_CONNTRACK + help + There is a commonly-used extension to IRC called + Direct Client-to-Client Protocol (DCC). This enables users to send + files to each other, and also chat to each other without the need + of a server. DCC Sending is used anywhere you send files over IRC, + and DCC Chat is most commonly used by Eggdrop bots. If you are + using NAT, this extension will enable you to send files and initiate + chats. Note that you do NOT need this extension to get files or + have others initiate chats, or everything else in IRC. + + To compile it as a module, choose M here. If unsure, say N. + config NF_CT_NETLINK tristate 'Connection tracking netlink interface (EXPERIMENTAL)' depends on EXPERIMENTAL && NF_CONNTRACK && NETFILTER_NETLINK diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 69b554576b6..343fd489640 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -25,6 +25,7 @@ nf_conntrack_h323-objs := nf_conntrack_h323_main.o nf_conntrack_h323_asn1.o obj-$(CONFIG_NF_CONNTRACK_AMANDA) += nf_conntrack_amanda.o obj-$(CONFIG_NF_CONNTRACK_FTP) += nf_conntrack_ftp.o obj-$(CONFIG_NF_CONNTRACK_H323) += nf_conntrack_h323.o +obj-$(CONFIG_NF_CONNTRACK_IRC) += nf_conntrack_irc.o # generic X tables obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o diff --git a/net/netfilter/nf_conntrack_irc.c b/net/netfilter/nf_conntrack_irc.c new file mode 100644 index 00000000000..ed01db63439 --- /dev/null +++ b/net/netfilter/nf_conntrack_irc.c @@ -0,0 +1,281 @@ +/* IRC extension for IP connection tracking, Version 1.21 + * (C) 2000-2002 by Harald Welte + * based on RR's ip_conntrack_ftp.c + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include + +#define MAX_PORTS 8 +static unsigned short ports[MAX_PORTS]; +static int ports_c; +static unsigned int max_dcc_channels = 8; +static unsigned int dcc_timeout __read_mostly = 300; +/* This is slow, but it's simple. --RR */ +static char *irc_buffer; +static DEFINE_SPINLOCK(irc_buffer_lock); + +unsigned int (*nf_nat_irc_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conntrack_expect *exp) __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_irc_hook); + +MODULE_AUTHOR("Harald Welte "); +MODULE_DESCRIPTION("IRC (DCC) connection tracking helper"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ip_conntrack_irc"); + +module_param_array(ports, ushort, &ports_c, 0400); +MODULE_PARM_DESC(ports, "port numbers of IRC servers"); +module_param(max_dcc_channels, uint, 0400); +MODULE_PARM_DESC(max_dcc_channels, "max number of expected DCC channels per " + "IRC session"); +module_param(dcc_timeout, uint, 0400); +MODULE_PARM_DESC(dcc_timeout, "timeout on for unestablished DCC channels"); + +static const char *dccprotos[] = { + "SEND ", "CHAT ", "MOVE ", "TSEND ", "SCHAT " +}; + +#define MINMATCHLEN 5 + +#if 0 +#define DEBUGP(format, args...) printk(KERN_DEBUG "%s:%s:" format, \ + __FILE__, __FUNCTION__ , ## args) +#else +#define DEBUGP(format, args...) +#endif + +/* tries to get the ip_addr and port out of a dcc command + * return value: -1 on failure, 0 on success + * data pointer to first byte of DCC command data + * data_end pointer to last byte of dcc command data + * ip returns parsed ip of dcc command + * port returns parsed port of dcc command + * ad_beg_p returns pointer to first byte of addr data + * ad_end_p returns pointer to last byte of addr data + */ +static int parse_dcc(char *data, char *data_end, u_int32_t *ip, + u_int16_t *port, char **ad_beg_p, char **ad_end_p) +{ + /* at least 12: "AAAAAAAA P\1\n" */ + while (*data++ != ' ') + if (data > data_end - 12) + return -1; + + *ad_beg_p = data; + *ip = simple_strtoul(data, &data, 10); + + /* skip blanks between ip and port */ + while (*data == ' ') { + if (data >= data_end) + return -1; + data++; + } + + *port = simple_strtoul(data, &data, 10); + *ad_end_p = data; + + return 0; +} + +static int help(struct sk_buff **pskb, unsigned int protoff, + struct nf_conn *ct, enum ip_conntrack_info ctinfo) +{ + unsigned int dataoff; + struct tcphdr _tcph, *th; + char *data, *data_limit, *ib_ptr; + int dir = CTINFO2DIR(ctinfo); + struct nf_conntrack_expect *exp; + struct nf_conntrack_tuple *tuple; + u_int32_t dcc_ip; + u_int16_t dcc_port; + __be16 port; + int i, ret = NF_ACCEPT; + char *addr_beg_p, *addr_end_p; + typeof(nf_nat_irc_hook) nf_nat_irc; + + /* If packet is coming from IRC server */ + if (dir == IP_CT_DIR_REPLY) + return NF_ACCEPT; + + /* Until there's been traffic both ways, don't look in packets. */ + if (ctinfo != IP_CT_ESTABLISHED && + ctinfo != IP_CT_ESTABLISHED + IP_CT_IS_REPLY) + return NF_ACCEPT; + + /* Not a full tcp header? */ + th = skb_header_pointer(*pskb, protoff, sizeof(_tcph), &_tcph); + if (th == NULL) + return NF_ACCEPT; + + /* No data? */ + dataoff = protoff + th->doff*4; + if (dataoff >= (*pskb)->len) + return NF_ACCEPT; + + spin_lock_bh(&irc_buffer_lock); + ib_ptr = skb_header_pointer(*pskb, dataoff, (*pskb)->len - dataoff, + irc_buffer); + BUG_ON(ib_ptr == NULL); + + data = ib_ptr; + data_limit = ib_ptr + (*pskb)->len - dataoff; + + /* strlen("\1DCC SENT t AAAAAAAA P\1\n")=24 + * 5+MINMATCHLEN+strlen("t AAAAAAAA P\1\n")=14 */ + while (data < data_limit - (19 + MINMATCHLEN)) { + if (memcmp(data, "\1DCC ", 5)) { + data++; + continue; + } + data += 5; + /* we have at least (19+MINMATCHLEN)-5 bytes valid data left */ + + DEBUGP("DCC found in master %u.%u.%u.%u:%u %u.%u.%u.%u:%u...\n", + NIPQUAD(iph->saddr), ntohs(th->source), + NIPQUAD(iph->daddr), ntohs(th->dest)); + + for (i = 0; i < ARRAY_SIZE(dccprotos); i++) { + if (memcmp(data, dccprotos[i], strlen(dccprotos[i]))) { + /* no match */ + continue; + } + data += strlen(dccprotos[i]); + DEBUGP("DCC %s detected\n", dccprotos[i]); + + /* we have at least + * (19+MINMATCHLEN)-5-dccprotos[i].matchlen bytes valid + * data left (== 14/13 bytes) */ + if (parse_dcc((char *)data, data_limit, &dcc_ip, + &dcc_port, &addr_beg_p, &addr_end_p)) { + DEBUGP("unable to parse dcc command\n"); + continue; + } + DEBUGP("DCC bound ip/port: %u.%u.%u.%u:%u\n", + HIPQUAD(dcc_ip), dcc_port); + + /* dcc_ip can be the internal OR external (NAT'ed) IP */ + tuple = &ct->tuplehash[dir].tuple; + if (tuple->src.u3.ip != htonl(dcc_ip) && + tuple->dst.u3.ip != htonl(dcc_ip)) { + if (net_ratelimit()) + printk(KERN_WARNING + "Forged DCC command from " + "%u.%u.%u.%u: %u.%u.%u.%u:%u\n", + NIPQUAD(tuple->src.u3.ip), + HIPQUAD(dcc_ip), dcc_port); + continue; + } + + exp = nf_conntrack_expect_alloc(ct); + if (exp == NULL) { + ret = NF_DROP; + goto out; + } + tuple = &ct->tuplehash[!dir].tuple; + port = htons(dcc_port); + nf_conntrack_expect_init(exp, tuple->src.l3num, + NULL, &tuple->dst.u3, + IPPROTO_TCP, NULL, &port); + + nf_nat_irc = rcu_dereference(nf_nat_irc_hook); + if (nf_nat_irc && ct->status & IPS_NAT_MASK) + ret = nf_nat_irc(pskb, ctinfo, + addr_beg_p - ib_ptr, + addr_end_p - addr_beg_p, + exp); + else if (nf_conntrack_expect_related(exp) != 0) + ret = NF_DROP; + nf_conntrack_expect_put(exp); + goto out; + } + } + out: + spin_unlock_bh(&irc_buffer_lock); + return ret; +} + +static struct nf_conntrack_helper irc[MAX_PORTS] __read_mostly; +static char irc_names[MAX_PORTS][sizeof("irc-65535")] __read_mostly; + +static void nf_conntrack_irc_fini(void); + +static int __init nf_conntrack_irc_init(void) +{ + int i, ret; + char *tmpname; + + if (max_dcc_channels < 1) { + printk("nf_ct_irc: max_dcc_channels must not be zero\n"); + return -EINVAL; + } + + irc_buffer = kmalloc(65536, GFP_KERNEL); + if (!irc_buffer) + return -ENOMEM; + + /* If no port given, default to standard irc port */ + if (ports_c == 0) + ports[ports_c++] = IRC_PORT; + + for (i = 0; i < ports_c; i++) { + irc[i].tuple.src.l3num = AF_INET; + irc[i].tuple.src.u.tcp.port = htons(ports[i]); + irc[i].tuple.dst.protonum = IPPROTO_TCP; + irc[i].mask.src.l3num = 0xFFFF; + irc[i].mask.src.u.tcp.port = htons(0xFFFF); + irc[i].mask.dst.protonum = 0xFF; + irc[i].max_expected = max_dcc_channels; + irc[i].timeout = dcc_timeout; + irc[i].me = THIS_MODULE; + irc[i].help = help; + + tmpname = &irc_names[i][0]; + if (ports[i] == IRC_PORT) + sprintf(tmpname, "irc"); + else + sprintf(tmpname, "irc-%u", i); + irc[i].name = tmpname; + + ret = nf_conntrack_helper_register(&irc[i]); + if (ret) { + printk("nf_ct_irc: failed to register helper " + "for pf: %u port: %u\n", + irc[i].tuple.src.l3num, ports[i]); + nf_conntrack_irc_fini(); + return ret; + } + } + return 0; +} + +/* This function is intentionally _NOT_ defined as __exit, because + * it is needed by the init function */ +static void nf_conntrack_irc_fini(void) +{ + int i; + + for (i = 0; i < ports_c; i++) + nf_conntrack_helper_unregister(&irc[i]); + kfree(irc_buffer); +} + +module_init(nf_conntrack_irc_init); +module_exit(nf_conntrack_irc_fini); -- cgit v1.2.3-70-g09d2 From f09943fefe6b702e40893d35b4f10fd1064037fe Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 2 Dec 2006 22:09:41 -0800 Subject: [NETFILTER]: nf_conntrack/nf_nat: add PPTP helper port Add nf_conntrack port of the PPtP conntrack/NAT helper. Since there seems to be no IPv6-capable PPtP implementation the helper only support IPv4. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_pptp.h | 321 ++++++++++++ include/linux/netfilter/nf_conntrack_proto_gre.h | 112 +++++ include/net/netfilter/ipv4/nf_conntrack_ipv4.h | 2 + include/net/netfilter/nf_conntrack.h | 4 + include/net/netfilter/nf_conntrack_helper.h | 2 + include/net/netfilter/nf_conntrack_tuple.h | 6 + net/ipv4/netfilter/Kconfig | 10 + net/ipv4/netfilter/Makefile | 4 + net/ipv4/netfilter/nf_nat_pptp.c | 315 ++++++++++++ net/ipv4/netfilter/nf_nat_proto_gre.c | 179 +++++++ net/netfilter/Kconfig | 23 + net/netfilter/Makefile | 2 + net/netfilter/nf_conntrack_core.c | 4 + net/netfilter/nf_conntrack_pptp.c | 607 +++++++++++++++++++++++ net/netfilter/nf_conntrack_proto_gre.c | 305 ++++++++++++ 15 files changed, 1896 insertions(+) create mode 100644 include/linux/netfilter/nf_conntrack_pptp.h create mode 100644 include/linux/netfilter/nf_conntrack_proto_gre.h create mode 100644 net/ipv4/netfilter/nf_nat_pptp.c create mode 100644 net/ipv4/netfilter/nf_nat_proto_gre.c create mode 100644 net/netfilter/nf_conntrack_pptp.c create mode 100644 net/netfilter/nf_conntrack_proto_gre.c (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_pptp.h b/include/linux/netfilter/nf_conntrack_pptp.h new file mode 100644 index 00000000000..fb049ec11ff --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_pptp.h @@ -0,0 +1,321 @@ +/* PPTP constants and structs */ +#ifndef _NF_CONNTRACK_PPTP_H +#define _NF_CONNTRACK_PPTP_H + +/* state of the control session */ +enum pptp_ctrlsess_state { + PPTP_SESSION_NONE, /* no session present */ + PPTP_SESSION_ERROR, /* some session error */ + PPTP_SESSION_STOPREQ, /* stop_sess request seen */ + PPTP_SESSION_REQUESTED, /* start_sess request seen */ + PPTP_SESSION_CONFIRMED, /* session established */ +}; + +/* state of the call inside the control session */ +enum pptp_ctrlcall_state { + PPTP_CALL_NONE, + PPTP_CALL_ERROR, + PPTP_CALL_OUT_REQ, + PPTP_CALL_OUT_CONF, + PPTP_CALL_IN_REQ, + PPTP_CALL_IN_REP, + PPTP_CALL_IN_CONF, + PPTP_CALL_CLEAR_REQ, +}; + +/* conntrack private data */ +struct nf_ct_pptp_master { + enum pptp_ctrlsess_state sstate; /* session state */ + enum pptp_ctrlcall_state cstate; /* call state */ + __be16 pac_call_id; /* call id of PAC */ + __be16 pns_call_id; /* call id of PNS */ + + /* in pre-2.6.11 this used to be per-expect. Now it is per-conntrack + * and therefore imposes a fixed limit on the number of maps */ + struct nf_ct_gre_keymap *keymap[IP_CT_DIR_MAX]; +}; + +struct nf_nat_pptp { + __be16 pns_call_id; /* NAT'ed PNS call id */ + __be16 pac_call_id; /* NAT'ed PAC call id */ +}; + +#ifdef __KERNEL__ + +#define PPTP_CONTROL_PORT 1723 + +#define PPTP_PACKET_CONTROL 1 +#define PPTP_PACKET_MGMT 2 + +#define PPTP_MAGIC_COOKIE 0x1a2b3c4d + +struct pptp_pkt_hdr { + __u16 packetLength; + __be16 packetType; + __be32 magicCookie; +}; + +/* PptpControlMessageType values */ +#define PPTP_START_SESSION_REQUEST 1 +#define PPTP_START_SESSION_REPLY 2 +#define PPTP_STOP_SESSION_REQUEST 3 +#define PPTP_STOP_SESSION_REPLY 4 +#define PPTP_ECHO_REQUEST 5 +#define PPTP_ECHO_REPLY 6 +#define PPTP_OUT_CALL_REQUEST 7 +#define PPTP_OUT_CALL_REPLY 8 +#define PPTP_IN_CALL_REQUEST 9 +#define PPTP_IN_CALL_REPLY 10 +#define PPTP_IN_CALL_CONNECT 11 +#define PPTP_CALL_CLEAR_REQUEST 12 +#define PPTP_CALL_DISCONNECT_NOTIFY 13 +#define PPTP_WAN_ERROR_NOTIFY 14 +#define PPTP_SET_LINK_INFO 15 + +#define PPTP_MSG_MAX 15 + +/* PptpGeneralError values */ +#define PPTP_ERROR_CODE_NONE 0 +#define PPTP_NOT_CONNECTED 1 +#define PPTP_BAD_FORMAT 2 +#define PPTP_BAD_VALUE 3 +#define PPTP_NO_RESOURCE 4 +#define PPTP_BAD_CALLID 5 +#define PPTP_REMOVE_DEVICE_ERROR 6 + +struct PptpControlHeader { + __be16 messageType; + __u16 reserved; +}; + +/* FramingCapability Bitmap Values */ +#define PPTP_FRAME_CAP_ASYNC 0x1 +#define PPTP_FRAME_CAP_SYNC 0x2 + +/* BearerCapability Bitmap Values */ +#define PPTP_BEARER_CAP_ANALOG 0x1 +#define PPTP_BEARER_CAP_DIGITAL 0x2 + +struct PptpStartSessionRequest { + __be16 protocolVersion; + __u16 reserved1; + __be32 framingCapability; + __be32 bearerCapability; + __be16 maxChannels; + __be16 firmwareRevision; + __u8 hostName[64]; + __u8 vendorString[64]; +}; + +/* PptpStartSessionResultCode Values */ +#define PPTP_START_OK 1 +#define PPTP_START_GENERAL_ERROR 2 +#define PPTP_START_ALREADY_CONNECTED 3 +#define PPTP_START_NOT_AUTHORIZED 4 +#define PPTP_START_UNKNOWN_PROTOCOL 5 + +struct PptpStartSessionReply { + __be16 protocolVersion; + __u8 resultCode; + __u8 generalErrorCode; + __be32 framingCapability; + __be32 bearerCapability; + __be16 maxChannels; + __be16 firmwareRevision; + __u8 hostName[64]; + __u8 vendorString[64]; +}; + +/* PptpStopReasons */ +#define PPTP_STOP_NONE 1 +#define PPTP_STOP_PROTOCOL 2 +#define PPTP_STOP_LOCAL_SHUTDOWN 3 + +struct PptpStopSessionRequest { + __u8 reason; + __u8 reserved1; + __u16 reserved2; +}; + +/* PptpStopSessionResultCode */ +#define PPTP_STOP_OK 1 +#define PPTP_STOP_GENERAL_ERROR 2 + +struct PptpStopSessionReply { + __u8 resultCode; + __u8 generalErrorCode; + __u16 reserved1; +}; + +struct PptpEchoRequest { + __be32 identNumber; +}; + +/* PptpEchoReplyResultCode */ +#define PPTP_ECHO_OK 1 +#define PPTP_ECHO_GENERAL_ERROR 2 + +struct PptpEchoReply { + __be32 identNumber; + __u8 resultCode; + __u8 generalErrorCode; + __u16 reserved; +}; + +/* PptpFramingType */ +#define PPTP_ASYNC_FRAMING 1 +#define PPTP_SYNC_FRAMING 2 +#define PPTP_DONT_CARE_FRAMING 3 + +/* PptpCallBearerType */ +#define PPTP_ANALOG_TYPE 1 +#define PPTP_DIGITAL_TYPE 2 +#define PPTP_DONT_CARE_BEARER_TYPE 3 + +struct PptpOutCallRequest { + __be16 callID; + __be16 callSerialNumber; + __be32 minBPS; + __be32 maxBPS; + __be32 bearerType; + __be32 framingType; + __be16 packetWindow; + __be16 packetProcDelay; + __be16 phoneNumberLength; + __u16 reserved1; + __u8 phoneNumber[64]; + __u8 subAddress[64]; +}; + +/* PptpCallResultCode */ +#define PPTP_OUTCALL_CONNECT 1 +#define PPTP_OUTCALL_GENERAL_ERROR 2 +#define PPTP_OUTCALL_NO_CARRIER 3 +#define PPTP_OUTCALL_BUSY 4 +#define PPTP_OUTCALL_NO_DIAL_TONE 5 +#define PPTP_OUTCALL_TIMEOUT 6 +#define PPTP_OUTCALL_DONT_ACCEPT 7 + +struct PptpOutCallReply { + __be16 callID; + __be16 peersCallID; + __u8 resultCode; + __u8 generalErrorCode; + __be16 causeCode; + __be32 connectSpeed; + __be16 packetWindow; + __be16 packetProcDelay; + __be32 physChannelID; +}; + +struct PptpInCallRequest { + __be16 callID; + __be16 callSerialNumber; + __be32 callBearerType; + __be32 physChannelID; + __be16 dialedNumberLength; + __be16 dialingNumberLength; + __u8 dialedNumber[64]; + __u8 dialingNumber[64]; + __u8 subAddress[64]; +}; + +/* PptpInCallResultCode */ +#define PPTP_INCALL_ACCEPT 1 +#define PPTP_INCALL_GENERAL_ERROR 2 +#define PPTP_INCALL_DONT_ACCEPT 3 + +struct PptpInCallReply { + __be16 callID; + __be16 peersCallID; + __u8 resultCode; + __u8 generalErrorCode; + __be16 packetWindow; + __be16 packetProcDelay; + __u16 reserved; +}; + +struct PptpInCallConnected { + __be16 peersCallID; + __u16 reserved; + __be32 connectSpeed; + __be16 packetWindow; + __be16 packetProcDelay; + __be32 callFramingType; +}; + +struct PptpClearCallRequest { + __be16 callID; + __u16 reserved; +}; + +struct PptpCallDisconnectNotify { + __be16 callID; + __u8 resultCode; + __u8 generalErrorCode; + __be16 causeCode; + __u16 reserved; + __u8 callStatistics[128]; +}; + +struct PptpWanErrorNotify { + __be16 peersCallID; + __u16 reserved; + __be32 crcErrors; + __be32 framingErrors; + __be32 hardwareOverRuns; + __be32 bufferOverRuns; + __be32 timeoutErrors; + __be32 alignmentErrors; +}; + +struct PptpSetLinkInfo { + __be16 peersCallID; + __u16 reserved; + __be32 sendAccm; + __be32 recvAccm; +}; + +union pptp_ctrl_union { + struct PptpStartSessionRequest sreq; + struct PptpStartSessionReply srep; + struct PptpStopSessionRequest streq; + struct PptpStopSessionReply strep; + struct PptpOutCallRequest ocreq; + struct PptpOutCallReply ocack; + struct PptpInCallRequest icreq; + struct PptpInCallReply icack; + struct PptpInCallConnected iccon; + struct PptpClearCallRequest clrreq; + struct PptpCallDisconnectNotify disc; + struct PptpWanErrorNotify wanerr; + struct PptpSetLinkInfo setlink; +}; + +/* crap needed for nf_conntrack_compat.h */ +struct nf_conn; +struct nf_conntrack_expect; +enum ip_conntrack_info; + +extern int +(*nf_nat_pptp_hook_outbound)(struct sk_buff **pskb, + struct nf_conn *ct, enum ip_conntrack_info ctinfo, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq); + +extern int +(*nf_nat_pptp_hook_inbound)(struct sk_buff **pskb, + struct nf_conn *ct, enum ip_conntrack_info ctinfo, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq); + +extern void +(*nf_nat_pptp_hook_exp_gre)(struct nf_conntrack_expect *exp_orig, + struct nf_conntrack_expect *exp_reply); + +extern void +(*nf_nat_pptp_hook_expectfn)(struct nf_conn *ct, + struct nf_conntrack_expect *exp); + +#endif /* __KERNEL__ */ +#endif /* _NF_CONNTRACK_PPTP_H */ diff --git a/include/linux/netfilter/nf_conntrack_proto_gre.h b/include/linux/netfilter/nf_conntrack_proto_gre.h new file mode 100644 index 00000000000..4e6bbce04ff --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_proto_gre.h @@ -0,0 +1,112 @@ +#ifndef _CONNTRACK_PROTO_GRE_H +#define _CONNTRACK_PROTO_GRE_H +#include + +/* GRE PROTOCOL HEADER */ + +/* GRE Version field */ +#define GRE_VERSION_1701 0x0 +#define GRE_VERSION_PPTP 0x1 + +/* GRE Protocol field */ +#define GRE_PROTOCOL_PPTP 0x880B + +/* GRE Flags */ +#define GRE_FLAG_C 0x80 +#define GRE_FLAG_R 0x40 +#define GRE_FLAG_K 0x20 +#define GRE_FLAG_S 0x10 +#define GRE_FLAG_A 0x80 + +#define GRE_IS_C(f) ((f)&GRE_FLAG_C) +#define GRE_IS_R(f) ((f)&GRE_FLAG_R) +#define GRE_IS_K(f) ((f)&GRE_FLAG_K) +#define GRE_IS_S(f) ((f)&GRE_FLAG_S) +#define GRE_IS_A(f) ((f)&GRE_FLAG_A) + +/* GRE is a mess: Four different standards */ +struct gre_hdr { +#if defined(__LITTLE_ENDIAN_BITFIELD) + __u16 rec:3, + srr:1, + seq:1, + key:1, + routing:1, + csum:1, + version:3, + reserved:4, + ack:1; +#elif defined(__BIG_ENDIAN_BITFIELD) + __u16 csum:1, + routing:1, + key:1, + seq:1, + srr:1, + rec:3, + ack:1, + reserved:4, + version:3; +#else +#error "Adjust your defines" +#endif + __be16 protocol; +}; + +/* modified GRE header for PPTP */ +struct gre_hdr_pptp { + __u8 flags; /* bitfield */ + __u8 version; /* should be GRE_VERSION_PPTP */ + __be16 protocol; /* should be GRE_PROTOCOL_PPTP */ + __be16 payload_len; /* size of ppp payload, not inc. gre header */ + __be16 call_id; /* peer's call_id for this session */ + __be32 seq; /* sequence number. Present if S==1 */ + __be32 ack; /* seq number of highest packet recieved by */ + /* sender in this session */ +}; + +struct nf_ct_gre { + unsigned int stream_timeout; + unsigned int timeout; +}; + +#ifdef __KERNEL__ +#include + +struct nf_conn; + +/* structure for original <-> reply keymap */ +struct nf_ct_gre_keymap { + struct list_head list; + struct nf_conntrack_tuple tuple; +}; + +/* add new tuple->key_reply pair to keymap */ +int nf_ct_gre_keymap_add(struct nf_conn *ct, enum ip_conntrack_dir dir, + struct nf_conntrack_tuple *t); + +/* delete keymap entries */ +void nf_ct_gre_keymap_destroy(struct nf_conn *ct); + +/* get pointer to gre key, if present */ +static inline __be32 *gre_key(struct gre_hdr *greh) +{ + if (!greh->key) + return NULL; + if (greh->csum || greh->routing) + return (__be32 *)(greh+sizeof(*greh)+4); + return (__be32 *)(greh+sizeof(*greh)); +} + +/* get pointer ot gre csum, if present */ +static inline __sum16 *gre_csum(struct gre_hdr *greh) +{ + if (!greh->csum) + return NULL; + return (__sum16 *)(greh+sizeof(*greh)); +} + +extern void nf_ct_gre_keymap_flush(void); +extern void nf_nat_need_gre(void); + +#endif /* __KERNEL__ */ +#endif /* _CONNTRACK_PROTO_GRE_H */ diff --git a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h index a1c57ee0a4f..1401ccc051c 100644 --- a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h +++ b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h @@ -11,10 +11,12 @@ #ifdef CONFIG_NF_NAT_NEEDED #include +#include /* per conntrack: nat application helper private data */ union nf_conntrack_nat_help { /* insert nat helper private data here */ + struct nf_nat_pptp nat_pptp_info; }; struct nf_conn_nat { diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 1646076933b..032b36a0e37 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -21,6 +21,7 @@ #include #include +#include #include #include @@ -33,6 +34,7 @@ union nf_conntrack_proto { struct ip_ct_tcp tcp; struct ip_ct_icmp icmp; struct nf_ct_icmpv6 icmpv6; + struct nf_ct_gre gre; }; union nf_conntrack_expect_proto { @@ -41,12 +43,14 @@ union nf_conntrack_expect_proto { /* Add protocol helper include file here */ #include +#include #include /* per conntrack: application helper private data */ union nf_conntrack_help { /* insert conntrack helper private data (master) here */ struct nf_ct_ftp_master ct_ftp_info; + struct nf_ct_pptp_master ct_pptp_info; struct nf_ct_h323_master ct_h323_info; }; diff --git a/include/net/netfilter/nf_conntrack_helper.h b/include/net/netfilter/nf_conntrack_helper.h index fbba9e8b95f..8c72ac9f0ab 100644 --- a/include/net/netfilter/nf_conntrack_helper.h +++ b/include/net/netfilter/nf_conntrack_helper.h @@ -34,6 +34,8 @@ struct nf_conntrack_helper struct nf_conn *ct, enum ip_conntrack_info conntrackinfo); + void (*destroy)(struct nf_conn *ct); + int (*to_nfattr)(struct sk_buff *skb, const struct nf_conn *ct); }; diff --git a/include/net/netfilter/nf_conntrack_tuple.h b/include/net/netfilter/nf_conntrack_tuple.h index c96a9c57673..5d72b16e876 100644 --- a/include/net/netfilter/nf_conntrack_tuple.h +++ b/include/net/netfilter/nf_conntrack_tuple.h @@ -49,6 +49,9 @@ union nf_conntrack_man_proto struct { __be16 port; } sctp; + struct { + __be16 key; /* GRE key is 32bit, PPtP only uses 16bit */ + } gre; }; /* The manipulable part of the tuple. */ @@ -84,6 +87,9 @@ struct nf_conntrack_tuple struct { __be16 port; } sctp; + struct { + __be16 key; + } gre; } u; /* The protocol. */ diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 4555f721dfc..c3327ac024d 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -484,6 +484,10 @@ config IP_NF_NAT_SNMP_BASIC # '&&' (6) # # (6) Returns the result of min(/expr/, /expr/). +config NF_NAT_PROTO_GRE + tristate + depends on NF_NAT && NF_CT_PROTO_GRE + config IP_NF_NAT_FTP tristate depends on IP_NF_IPTABLES && IP_NF_CONNTRACK && IP_NF_NAT @@ -528,6 +532,12 @@ config IP_NF_NAT_PPTP default IP_NF_NAT if IP_NF_PPTP=y default m if IP_NF_PPTP=m +config NF_NAT_PPTP + tristate + depends on IP_NF_IPTABLES && NF_CONNTRACK && NF_NAT + default NF_NAT && NF_CONNTRACK_PPTP + select NF_NAT_PROTO_GRE + config IP_NF_NAT_H323 tristate depends on IP_NF_IPTABLES!=n && IP_NF_CONNTRACK!=n && IP_NF_NAT!=n diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index 56733c37032..ef33ff2cdda 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -54,6 +54,10 @@ obj-$(CONFIG_NF_NAT_AMANDA) += nf_nat_amanda.o obj-$(CONFIG_NF_NAT_FTP) += nf_nat_ftp.o obj-$(CONFIG_NF_NAT_H323) += nf_nat_h323.o obj-$(CONFIG_NF_NAT_IRC) += nf_nat_irc.o +obj-$(CONFIG_NF_NAT_PPTP) += nf_nat_pptp.o + +# NAT protocols (nf_nat) +obj-$(CONFIG_NF_NAT_PROTO_GRE) += nf_nat_proto_gre.o # generic IP tables obj-$(CONFIG_IP_NF_IPTABLES) += ip_tables.o diff --git a/net/ipv4/netfilter/nf_nat_pptp.c b/net/ipv4/netfilter/nf_nat_pptp.c new file mode 100644 index 00000000000..0ae45b79a4e --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_pptp.c @@ -0,0 +1,315 @@ +/* + * nf_nat_pptp.c + * + * NAT support for PPTP (Point to Point Tunneling Protocol). + * PPTP is a a protocol for creating virtual private networks. + * It is a specification defined by Microsoft and some vendors + * working with Microsoft. PPTP is built on top of a modified + * version of the Internet Generic Routing Encapsulation Protocol. + * GRE is defined in RFC 1701 and RFC 1702. Documentation of + * PPTP can be found in RFC 2637 + * + * (C) 2000-2005 by Harald Welte + * + * Development of this code funded by Astaro AG (http://www.astaro.com/) + * + * TODO: - NAT to a unique tuple, not to TCP source port + * (needs netfilter tuple reservation) + */ + +#include +#include + +#include +#include +#include +#include +#include +#include +#include + +#define NF_NAT_PPTP_VERSION "3.0" + +#define REQ_CID(req, off) (*(__be16 *)((char *)(req) + (off))) + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Harald Welte "); +MODULE_DESCRIPTION("Netfilter NAT helper module for PPTP"); +MODULE_ALIAS("ip_nat_pptp"); + +#if 0 +extern const char *pptp_msg_name[]; +#define DEBUGP(format, args...) printk(KERN_DEBUG "%s:%s: " format, __FILE__, \ + __FUNCTION__, ## args) +#else +#define DEBUGP(format, args...) +#endif + +static void pptp_nat_expected(struct nf_conn *ct, + struct nf_conntrack_expect *exp) +{ + struct nf_conn *master = ct->master; + struct nf_conntrack_expect *other_exp; + struct nf_conntrack_tuple t; + struct nf_ct_pptp_master *ct_pptp_info; + struct nf_nat_pptp *nat_pptp_info; + struct ip_nat_range range; + + ct_pptp_info = &nfct_help(master)->help.ct_pptp_info; + nat_pptp_info = &nfct_nat(master)->help.nat_pptp_info; + + /* And here goes the grand finale of corrosion... */ + if (exp->dir == IP_CT_DIR_ORIGINAL) { + DEBUGP("we are PNS->PAC\n"); + /* therefore, build tuple for PAC->PNS */ + t.src.l3num = AF_INET; + t.src.u3.ip = master->tuplehash[!exp->dir].tuple.src.u3.ip; + t.src.u.gre.key = ct_pptp_info->pac_call_id; + t.dst.u3.ip = master->tuplehash[!exp->dir].tuple.dst.u3.ip; + t.dst.u.gre.key = ct_pptp_info->pns_call_id; + t.dst.protonum = IPPROTO_GRE; + } else { + DEBUGP("we are PAC->PNS\n"); + /* build tuple for PNS->PAC */ + t.src.l3num = AF_INET; + t.src.u3.ip = master->tuplehash[exp->dir].tuple.src.u3.ip; + t.src.u.gre.key = nat_pptp_info->pns_call_id; + t.dst.u3.ip = master->tuplehash[exp->dir].tuple.dst.u3.ip; + t.dst.u.gre.key = nat_pptp_info->pac_call_id; + t.dst.protonum = IPPROTO_GRE; + } + + DEBUGP("trying to unexpect other dir: "); + NF_CT_DUMP_TUPLE(&t); + other_exp = nf_conntrack_expect_find_get(&t); + if (other_exp) { + nf_conntrack_unexpect_related(other_exp); + nf_conntrack_expect_put(other_exp); + DEBUGP("success\n"); + } else { + DEBUGP("not found!\n"); + } + + /* This must be a fresh one. */ + BUG_ON(ct->status & IPS_NAT_DONE_MASK); + + /* Change src to where master sends to */ + range.flags = IP_NAT_RANGE_MAP_IPS; + range.min_ip = range.max_ip + = ct->master->tuplehash[!exp->dir].tuple.dst.u3.ip; + if (exp->dir == IP_CT_DIR_ORIGINAL) { + range.flags |= IP_NAT_RANGE_PROTO_SPECIFIED; + range.min = range.max = exp->saved_proto; + } + /* hook doesn't matter, but it has to do source manip */ + nf_nat_setup_info(ct, &range, NF_IP_POST_ROUTING); + + /* For DST manip, map port here to where it's expected. */ + range.flags = IP_NAT_RANGE_MAP_IPS; + range.min_ip = range.max_ip + = ct->master->tuplehash[!exp->dir].tuple.src.u3.ip; + if (exp->dir == IP_CT_DIR_REPLY) { + range.flags |= IP_NAT_RANGE_PROTO_SPECIFIED; + range.min = range.max = exp->saved_proto; + } + /* hook doesn't matter, but it has to do destination manip */ + nf_nat_setup_info(ct, &range, NF_IP_PRE_ROUTING); +} + +/* outbound packets == from PNS to PAC */ +static int +pptp_outbound_pkt(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq) + +{ + struct nf_ct_pptp_master *ct_pptp_info; + struct nf_nat_pptp *nat_pptp_info; + u_int16_t msg; + __be16 new_callid; + unsigned int cid_off; + + ct_pptp_info = &nfct_help(ct)->help.ct_pptp_info; + nat_pptp_info = &nfct_nat(ct)->help.nat_pptp_info; + + new_callid = ct_pptp_info->pns_call_id; + + switch (msg = ntohs(ctlh->messageType)) { + case PPTP_OUT_CALL_REQUEST: + cid_off = offsetof(union pptp_ctrl_union, ocreq.callID); + /* FIXME: ideally we would want to reserve a call ID + * here. current netfilter NAT core is not able to do + * this :( For now we use TCP source port. This breaks + * multiple calls within one control session */ + + /* save original call ID in nat_info */ + nat_pptp_info->pns_call_id = ct_pptp_info->pns_call_id; + + /* don't use tcph->source since we are at a DSTmanip + * hook (e.g. PREROUTING) and pkt is not mangled yet */ + new_callid = ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u.tcp.port; + + /* save new call ID in ct info */ + ct_pptp_info->pns_call_id = new_callid; + break; + case PPTP_IN_CALL_REPLY: + cid_off = offsetof(union pptp_ctrl_union, icack.callID); + break; + case PPTP_CALL_CLEAR_REQUEST: + cid_off = offsetof(union pptp_ctrl_union, clrreq.callID); + break; + default: + DEBUGP("unknown outbound packet 0x%04x:%s\n", msg, + (msg <= PPTP_MSG_MAX)? + pptp_msg_name[msg]:pptp_msg_name[0]); + /* fall through */ + case PPTP_SET_LINK_INFO: + /* only need to NAT in case PAC is behind NAT box */ + case PPTP_START_SESSION_REQUEST: + case PPTP_START_SESSION_REPLY: + case PPTP_STOP_SESSION_REQUEST: + case PPTP_STOP_SESSION_REPLY: + case PPTP_ECHO_REQUEST: + case PPTP_ECHO_REPLY: + /* no need to alter packet */ + return NF_ACCEPT; + } + + /* only OUT_CALL_REQUEST, IN_CALL_REPLY, CALL_CLEAR_REQUEST pass + * down to here */ + DEBUGP("altering call id from 0x%04x to 0x%04x\n", + ntohs(REQ_CID(pptpReq, cid_off)), ntohs(new_callid)); + + /* mangle packet */ + if (nf_nat_mangle_tcp_packet(pskb, ct, ctinfo, + cid_off + sizeof(struct pptp_pkt_hdr) + + sizeof(struct PptpControlHeader), + sizeof(new_callid), (char *)&new_callid, + sizeof(new_callid)) == 0) + return NF_DROP; + return NF_ACCEPT; +} + +static void +pptp_exp_gre(struct nf_conntrack_expect *expect_orig, + struct nf_conntrack_expect *expect_reply) +{ + struct nf_conn *ct = expect_orig->master; + struct nf_ct_pptp_master *ct_pptp_info; + struct nf_nat_pptp *nat_pptp_info; + + ct_pptp_info = &nfct_help(ct)->help.ct_pptp_info; + nat_pptp_info = &nfct_nat(ct)->help.nat_pptp_info; + + /* save original PAC call ID in nat_info */ + nat_pptp_info->pac_call_id = ct_pptp_info->pac_call_id; + + /* alter expectation for PNS->PAC direction */ + expect_orig->saved_proto.gre.key = ct_pptp_info->pns_call_id; + expect_orig->tuple.src.u.gre.key = nat_pptp_info->pns_call_id; + expect_orig->tuple.dst.u.gre.key = ct_pptp_info->pac_call_id; + expect_orig->dir = IP_CT_DIR_ORIGINAL; + + /* alter expectation for PAC->PNS direction */ + expect_reply->saved_proto.gre.key = nat_pptp_info->pns_call_id; + expect_reply->tuple.src.u.gre.key = nat_pptp_info->pac_call_id; + expect_reply->tuple.dst.u.gre.key = ct_pptp_info->pns_call_id; + expect_reply->dir = IP_CT_DIR_REPLY; +} + +/* inbound packets == from PAC to PNS */ +static int +pptp_inbound_pkt(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq) +{ + struct nf_nat_pptp *nat_pptp_info; + u_int16_t msg; + __be16 new_pcid; + unsigned int pcid_off; + + nat_pptp_info = &nfct_nat(ct)->help.nat_pptp_info; + new_pcid = nat_pptp_info->pns_call_id; + + switch (msg = ntohs(ctlh->messageType)) { + case PPTP_OUT_CALL_REPLY: + pcid_off = offsetof(union pptp_ctrl_union, ocack.peersCallID); + break; + case PPTP_IN_CALL_CONNECT: + pcid_off = offsetof(union pptp_ctrl_union, iccon.peersCallID); + break; + case PPTP_IN_CALL_REQUEST: + /* only need to nat in case PAC is behind NAT box */ + return NF_ACCEPT; + case PPTP_WAN_ERROR_NOTIFY: + pcid_off = offsetof(union pptp_ctrl_union, wanerr.peersCallID); + break; + case PPTP_CALL_DISCONNECT_NOTIFY: + pcid_off = offsetof(union pptp_ctrl_union, disc.callID); + break; + case PPTP_SET_LINK_INFO: + pcid_off = offsetof(union pptp_ctrl_union, setlink.peersCallID); + break; + default: + DEBUGP("unknown inbound packet %s\n", (msg <= PPTP_MSG_MAX)? + pptp_msg_name[msg]:pptp_msg_name[0]); + /* fall through */ + case PPTP_START_SESSION_REQUEST: + case PPTP_START_SESSION_REPLY: + case PPTP_STOP_SESSION_REQUEST: + case PPTP_STOP_SESSION_REPLY: + case PPTP_ECHO_REQUEST: + case PPTP_ECHO_REPLY: + /* no need to alter packet */ + return NF_ACCEPT; + } + + /* only OUT_CALL_REPLY, IN_CALL_CONNECT, IN_CALL_REQUEST, + * WAN_ERROR_NOTIFY, CALL_DISCONNECT_NOTIFY pass down here */ + + /* mangle packet */ + DEBUGP("altering peer call id from 0x%04x to 0x%04x\n", + ntohs(REQ_CID(pptpReq, pcid_off)), ntohs(new_pcid)); + + if (nf_nat_mangle_tcp_packet(pskb, ct, ctinfo, + pcid_off + sizeof(struct pptp_pkt_hdr) + + sizeof(struct PptpControlHeader), + sizeof(new_pcid), (char *)&new_pcid, + sizeof(new_pcid)) == 0) + return NF_DROP; + return NF_ACCEPT; +} + +static int __init nf_nat_helper_pptp_init(void) +{ + nf_nat_need_gre(); + + BUG_ON(rcu_dereference(nf_nat_pptp_hook_outbound)); + rcu_assign_pointer(nf_nat_pptp_hook_outbound, pptp_outbound_pkt); + + BUG_ON(rcu_dereference(nf_nat_pptp_hook_inbound)); + rcu_assign_pointer(nf_nat_pptp_hook_inbound, pptp_inbound_pkt); + + BUG_ON(rcu_dereference(nf_nat_pptp_hook_exp_gre)); + rcu_assign_pointer(nf_nat_pptp_hook_exp_gre, pptp_exp_gre); + + BUG_ON(rcu_dereference(nf_nat_pptp_hook_expectfn)); + rcu_assign_pointer(nf_nat_pptp_hook_expectfn, pptp_nat_expected); + return 0; +} + +static void __exit nf_nat_helper_pptp_fini(void) +{ + rcu_assign_pointer(nf_nat_pptp_hook_expectfn, NULL); + rcu_assign_pointer(nf_nat_pptp_hook_exp_gre, NULL); + rcu_assign_pointer(nf_nat_pptp_hook_inbound, NULL); + rcu_assign_pointer(nf_nat_pptp_hook_outbound, NULL); + synchronize_rcu(); +} + +module_init(nf_nat_helper_pptp_init); +module_exit(nf_nat_helper_pptp_fini); diff --git a/net/ipv4/netfilter/nf_nat_proto_gre.c b/net/ipv4/netfilter/nf_nat_proto_gre.c new file mode 100644 index 00000000000..d3de579e09d --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_proto_gre.c @@ -0,0 +1,179 @@ +/* + * nf_nat_proto_gre.c + * + * NAT protocol helper module for GRE. + * + * GRE is a generic encapsulation protocol, which is generally not very + * suited for NAT, as it has no protocol-specific part as port numbers. + * + * It has an optional key field, which may help us distinguishing two + * connections between the same two hosts. + * + * GRE is defined in RFC 1701 and RFC 1702, as well as RFC 2784 + * + * PPTP is built on top of a modified version of GRE, and has a mandatory + * field called "CallID", which serves us for the same purpose as the key + * field in plain GRE. + * + * Documentation about PPTP can be found in RFC 2637 + * + * (C) 2000-2005 by Harald Welte + * + * Development of this code funded by Astaro AG (http://www.astaro.com/) + * + */ + +#include +#include +#include + +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Harald Welte "); +MODULE_DESCRIPTION("Netfilter NAT protocol helper module for GRE"); + +#if 0 +#define DEBUGP(format, args...) printk(KERN_DEBUG "%s:%s: " format, __FILE__, \ + __FUNCTION__, ## args) +#else +#define DEBUGP(x, args...) +#endif + +/* is key in given range between min and max */ +static int +gre_in_range(const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype, + const union nf_conntrack_man_proto *min, + const union nf_conntrack_man_proto *max) +{ + __be16 key; + + if (maniptype == IP_NAT_MANIP_SRC) + key = tuple->src.u.gre.key; + else + key = tuple->dst.u.gre.key; + + return ntohs(key) >= ntohs(min->gre.key) && + ntohs(key) <= ntohs(max->gre.key); +} + +/* generate unique tuple ... */ +static int +gre_unique_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + enum nf_nat_manip_type maniptype, + const struct nf_conn *conntrack) +{ + static u_int16_t key; + __be16 *keyptr; + unsigned int min, i, range_size; + + if (maniptype == IP_NAT_MANIP_SRC) + keyptr = &tuple->src.u.gre.key; + else + keyptr = &tuple->dst.u.gre.key; + + if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED)) { + DEBUGP("%p: NATing GRE PPTP\n", conntrack); + min = 1; + range_size = 0xffff; + } else { + min = ntohs(range->min.gre.key); + range_size = ntohs(range->max.gre.key) - min + 1; + } + + DEBUGP("min = %u, range_size = %u\n", min, range_size); + + for (i = 0; i < range_size; i++, key++) { + *keyptr = htons(min + key % range_size); + if (!nf_nat_used_tuple(tuple, conntrack)) + return 1; + } + + DEBUGP("%p: no NAT mapping\n", conntrack); + return 0; +} + +/* manipulate a GRE packet according to maniptype */ +static int +gre_manip_pkt(struct sk_buff **pskb, unsigned int iphdroff, + const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype) +{ + struct gre_hdr *greh; + struct gre_hdr_pptp *pgreh; + struct iphdr *iph = (struct iphdr *)((*pskb)->data + iphdroff); + unsigned int hdroff = iphdroff + iph->ihl * 4; + + /* pgreh includes two optional 32bit fields which are not required + * to be there. That's where the magic '8' comes from */ + if (!skb_make_writable(pskb, hdroff + sizeof(*pgreh) - 8)) + return 0; + + greh = (void *)(*pskb)->data + hdroff; + pgreh = (struct gre_hdr_pptp *)greh; + + /* we only have destination manip of a packet, since 'source key' + * is not present in the packet itself */ + if (maniptype != IP_NAT_MANIP_DST) + return 1; + switch (greh->version) { + case 0: + if (!greh->key) { + DEBUGP("can't nat GRE w/o key\n"); + break; + } + if (greh->csum) { + /* FIXME: Never tested this code... */ + nf_proto_csum_replace4(gre_csum(greh), *pskb, + *(gre_key(greh)), + tuple->dst.u.gre.key, 0); + } + *(gre_key(greh)) = tuple->dst.u.gre.key; + break; + case GRE_VERSION_PPTP: + DEBUGP("call_id -> 0x%04x\n", ntohs(tuple->dst.u.gre.key)); + pgreh->call_id = tuple->dst.u.gre.key; + break; + default: + DEBUGP("can't nat unknown GRE version\n"); + return 0; + } + return 1; +} + +static struct nf_nat_protocol gre __read_mostly = { + .name = "GRE", + .protonum = IPPROTO_GRE, + .manip_pkt = gre_manip_pkt, + .in_range = gre_in_range, + .unique_tuple = gre_unique_tuple, +#if defined(CONFIG_IP_NF_CONNTRACK_NETLINK) || \ + defined(CONFIG_IP_NF_CONNTRACK_NETLINK_MODULE) + .range_to_nfattr = nf_nat_port_range_to_nfattr, + .nfattr_to_range = nf_nat_port_nfattr_to_range, +#endif +}; + +int __init nf_nat_proto_gre_init(void) +{ + return nf_nat_protocol_register(&gre); +} + +void __exit nf_nat_proto_gre_fini(void) +{ + nf_nat_protocol_unregister(&gre); +} + +module_init(nf_nat_proto_gre_init); +module_exit(nf_nat_proto_gre_fini); + +void nf_nat_need_gre(void) +{ + return; +} +EXPORT_SYMBOL_GPL(nf_nat_need_gre); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index d1a365d83c5..6b2eb26ae03 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -120,6 +120,10 @@ config NF_CONNTRACK_EVENTS If unsure, say `N'. +config NF_CT_PROTO_GRE + tristate + depends on EXPERIMENTAL && NF_CONNTRACK + config NF_CT_PROTO_SCTP tristate 'SCTP protocol on new connection tracking support (EXPERIMENTAL)' depends on EXPERIMENTAL && NF_CONNTRACK @@ -213,6 +217,25 @@ config NF_CONNTRACK_NETBIOS_NS To compile it as a module, choose M here. If unsure, say N. +config NF_CONNTRACK_PPTP + tristate "PPtP protocol support (EXPERIMENTAL)" + depends on EXPERIMENTAL && NF_CONNTRACK + select NF_CT_PROTO_GRE + help + This module adds support for PPTP (Point to Point Tunnelling + Protocol, RFC2637) connection tracking and NAT. + + If you are running PPTP sessions over a stateful firewall or NAT + box, you may want to enable this feature. + + Please note that not all PPTP modes of operation are supported yet. + Specifically these limitations exist: + - Blindy assumes that control connections are always established + in PNS->PAC direction. This is a violation of RFC2637. + - Only supports a single call within each session + + To compile it as a module, choose M here. If unsure, say N. + config NF_CT_NETLINK tristate 'Connection tracking netlink interface (EXPERIMENTAL)' depends on EXPERIMENTAL && NF_CONNTRACK && NETFILTER_NETLINK diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 67144b2af64..897bed4cbd7 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -14,6 +14,7 @@ obj-$(CONFIG_NETFILTER_NETLINK_LOG) += nfnetlink_log.o obj-$(CONFIG_NF_CONNTRACK) += nf_conntrack.o # SCTP protocol connection tracking +obj-$(CONFIG_NF_CT_PROTO_GRE) += nf_conntrack_proto_gre.o obj-$(CONFIG_NF_CT_PROTO_SCTP) += nf_conntrack_proto_sctp.o # netlink interface for nf_conntrack @@ -27,6 +28,7 @@ obj-$(CONFIG_NF_CONNTRACK_FTP) += nf_conntrack_ftp.o obj-$(CONFIG_NF_CONNTRACK_H323) += nf_conntrack_h323.o obj-$(CONFIG_NF_CONNTRACK_IRC) += nf_conntrack_irc.o obj-$(CONFIG_NF_CONNTRACK_NETBIOS_NS) += nf_conntrack_netbios_ns.o +obj-$(CONFIG_NF_CONNTRACK_PPTP) += nf_conntrack_pptp.o # generic X tables obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index aa8beabfeeb..ed756c928bc 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -300,6 +300,7 @@ static void destroy_conntrack(struct nf_conntrack *nfct) { struct nf_conn *ct = (struct nf_conn *)nfct; + struct nf_conn_help *help = nfct_help(ct); struct nf_conntrack_l3proto *l3proto; struct nf_conntrack_l4proto *l4proto; @@ -310,6 +311,9 @@ destroy_conntrack(struct nf_conntrack *nfct) nf_conntrack_event(IPCT_DESTROY, ct); set_bit(IPS_DYING_BIT, &ct->status); + if (help && help->helper && help->helper->destroy) + help->helper->destroy(ct); + /* To make sure we don't get any weird locking issues here: * destroy_conntrack() MUST NOT be called with a write lock * to nf_conntrack_lock!!! -HW */ diff --git a/net/netfilter/nf_conntrack_pptp.c b/net/netfilter/nf_conntrack_pptp.c new file mode 100644 index 00000000000..f0ff00e0d05 --- /dev/null +++ b/net/netfilter/nf_conntrack_pptp.c @@ -0,0 +1,607 @@ +/* + * Connection tracking support for PPTP (Point to Point Tunneling Protocol). + * PPTP is a a protocol for creating virtual private networks. + * It is a specification defined by Microsoft and some vendors + * working with Microsoft. PPTP is built on top of a modified + * version of the Internet Generic Routing Encapsulation Protocol. + * GRE is defined in RFC 1701 and RFC 1702. Documentation of + * PPTP can be found in RFC 2637 + * + * (C) 2000-2005 by Harald Welte + * + * Development of this code funded by Astaro AG (http://www.astaro.com/) + * + * Limitations: + * - We blindly assume that control connections are always + * established in PNS->PAC direction. This is a violation + * of RFFC2673 + * - We can only support one single call within each session + * TODO: + * - testing of incoming PPTP calls + */ + +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +#define NF_CT_PPTP_VERSION "3.1" + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Harald Welte "); +MODULE_DESCRIPTION("Netfilter connection tracking helper module for PPTP"); +MODULE_ALIAS("ip_conntrack_pptp"); + +static DEFINE_SPINLOCK(nf_pptp_lock); + +int +(*nf_nat_pptp_hook_outbound)(struct sk_buff **pskb, + struct nf_conn *ct, enum ip_conntrack_info ctinfo, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq) __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_pptp_hook_outbound); + +int +(*nf_nat_pptp_hook_inbound)(struct sk_buff **pskb, + struct nf_conn *ct, enum ip_conntrack_info ctinfo, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq) __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_pptp_hook_inbound); + +void +(*nf_nat_pptp_hook_exp_gre)(struct nf_conntrack_expect *expect_orig, + struct nf_conntrack_expect *expect_reply) + __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_pptp_hook_exp_gre); + +void +(*nf_nat_pptp_hook_expectfn)(struct nf_conn *ct, + struct nf_conntrack_expect *exp) __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_pptp_hook_expectfn); + +#if 0 +/* PptpControlMessageType names */ +const char *pptp_msg_name[] = { + "UNKNOWN_MESSAGE", + "START_SESSION_REQUEST", + "START_SESSION_REPLY", + "STOP_SESSION_REQUEST", + "STOP_SESSION_REPLY", + "ECHO_REQUEST", + "ECHO_REPLY", + "OUT_CALL_REQUEST", + "OUT_CALL_REPLY", + "IN_CALL_REQUEST", + "IN_CALL_REPLY", + "IN_CALL_CONNECT", + "CALL_CLEAR_REQUEST", + "CALL_DISCONNECT_NOTIFY", + "WAN_ERROR_NOTIFY", + "SET_LINK_INFO" +}; +EXPORT_SYMBOL(pptp_msg_name); +#define DEBUGP(format, args...) printk(KERN_DEBUG "%s:%s: " format, __FILE__, __FUNCTION__, ## args) +#else +#define DEBUGP(format, args...) +#endif + +#define SECS *HZ +#define MINS * 60 SECS +#define HOURS * 60 MINS + +#define PPTP_GRE_TIMEOUT (10 MINS) +#define PPTP_GRE_STREAM_TIMEOUT (5 HOURS) + +static void pptp_expectfn(struct nf_conn *ct, + struct nf_conntrack_expect *exp) +{ + typeof(nf_nat_pptp_hook_expectfn) nf_nat_pptp_expectfn; + DEBUGP("increasing timeouts\n"); + + /* increase timeout of GRE data channel conntrack entry */ + ct->proto.gre.timeout = PPTP_GRE_TIMEOUT; + ct->proto.gre.stream_timeout = PPTP_GRE_STREAM_TIMEOUT; + + /* Can you see how rusty this code is, compared with the pre-2.6.11 + * one? That's what happened to my shiny newnat of 2002 ;( -HW */ + + rcu_read_lock(); + nf_nat_pptp_expectfn = rcu_dereference(nf_nat_pptp_hook_expectfn); + if (nf_nat_pptp_expectfn && ct->status & IPS_NAT_MASK) + nf_nat_pptp_expectfn(ct, exp); + else { + struct nf_conntrack_tuple inv_t; + struct nf_conntrack_expect *exp_other; + + /* obviously this tuple inversion only works until you do NAT */ + nf_ct_invert_tuplepr(&inv_t, &exp->tuple); + DEBUGP("trying to unexpect other dir: "); + NF_CT_DUMP_TUPLE(&inv_t); + + exp_other = nf_conntrack_expect_find_get(&inv_t); + if (exp_other) { + /* delete other expectation. */ + DEBUGP("found\n"); + nf_conntrack_unexpect_related(exp_other); + nf_conntrack_expect_put(exp_other); + } else { + DEBUGP("not found\n"); + } + } + rcu_read_unlock(); +} + +static int destroy_sibling_or_exp(const struct nf_conntrack_tuple *t) +{ + struct nf_conntrack_tuple_hash *h; + struct nf_conntrack_expect *exp; + struct nf_conn *sibling; + + DEBUGP("trying to timeout ct or exp for tuple "); + NF_CT_DUMP_TUPLE(t); + + h = nf_conntrack_find_get(t, NULL); + if (h) { + sibling = nf_ct_tuplehash_to_ctrack(h); + DEBUGP("setting timeout of conntrack %p to 0\n", sibling); + sibling->proto.gre.timeout = 0; + sibling->proto.gre.stream_timeout = 0; + if (del_timer(&sibling->timeout)) + sibling->timeout.function((unsigned long)sibling); + nf_ct_put(sibling); + return 1; + } else { + exp = nf_conntrack_expect_find_get(t); + if (exp) { + DEBUGP("unexpect_related of expect %p\n", exp); + nf_conntrack_unexpect_related(exp); + nf_conntrack_expect_put(exp); + return 1; + } + } + return 0; +} + +/* timeout GRE data connections */ +static void pptp_destroy_siblings(struct nf_conn *ct) +{ + struct nf_conn_help *help = nfct_help(ct); + struct nf_conntrack_tuple t; + + nf_ct_gre_keymap_destroy(ct); + + /* try original (pns->pac) tuple */ + memcpy(&t, &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple, sizeof(t)); + t.dst.protonum = IPPROTO_GRE; + t.src.u.gre.key = help->help.ct_pptp_info.pns_call_id; + t.dst.u.gre.key = help->help.ct_pptp_info.pac_call_id; + if (!destroy_sibling_or_exp(&t)) + DEBUGP("failed to timeout original pns->pac ct/exp\n"); + + /* try reply (pac->pns) tuple */ + memcpy(&t, &ct->tuplehash[IP_CT_DIR_REPLY].tuple, sizeof(t)); + t.dst.protonum = IPPROTO_GRE; + t.src.u.gre.key = help->help.ct_pptp_info.pac_call_id; + t.dst.u.gre.key = help->help.ct_pptp_info.pns_call_id; + if (!destroy_sibling_or_exp(&t)) + DEBUGP("failed to timeout reply pac->pns ct/exp\n"); +} + +/* expect GRE connections (PNS->PAC and PAC->PNS direction) */ +static int exp_gre(struct nf_conn *ct, __be16 callid, __be16 peer_callid) +{ + struct nf_conntrack_expect *exp_orig, *exp_reply; + enum ip_conntrack_dir dir; + int ret = 1; + typeof(nf_nat_pptp_hook_exp_gre) nf_nat_pptp_exp_gre; + + exp_orig = nf_conntrack_expect_alloc(ct); + if (exp_orig == NULL) + goto out; + + exp_reply = nf_conntrack_expect_alloc(ct); + if (exp_reply == NULL) + goto out_put_orig; + + /* original direction, PNS->PAC */ + dir = IP_CT_DIR_ORIGINAL; + nf_conntrack_expect_init(exp_orig, ct->tuplehash[dir].tuple.src.l3num, + &ct->tuplehash[dir].tuple.src.u3, + &ct->tuplehash[dir].tuple.dst.u3, + IPPROTO_GRE, &peer_callid, &callid); + exp_orig->expectfn = pptp_expectfn; + + /* reply direction, PAC->PNS */ + dir = IP_CT_DIR_REPLY; + nf_conntrack_expect_init(exp_reply, ct->tuplehash[dir].tuple.src.l3num, + &ct->tuplehash[dir].tuple.src.u3, + &ct->tuplehash[dir].tuple.dst.u3, + IPPROTO_GRE, &callid, &peer_callid); + exp_reply->expectfn = pptp_expectfn; + + nf_nat_pptp_exp_gre = rcu_dereference(nf_nat_pptp_hook_exp_gre); + if (nf_nat_pptp_exp_gre && ct->status & IPS_NAT_MASK) + nf_nat_pptp_exp_gre(exp_orig, exp_reply); + if (nf_conntrack_expect_related(exp_orig) != 0) + goto out_put_both; + if (nf_conntrack_expect_related(exp_reply) != 0) + goto out_unexpect_orig; + + /* Add GRE keymap entries */ + if (nf_ct_gre_keymap_add(ct, IP_CT_DIR_ORIGINAL, &exp_orig->tuple) != 0) + goto out_unexpect_both; + if (nf_ct_gre_keymap_add(ct, IP_CT_DIR_REPLY, &exp_reply->tuple) != 0) { + nf_ct_gre_keymap_destroy(ct); + goto out_unexpect_both; + } + ret = 0; + +out_put_both: + nf_conntrack_expect_put(exp_reply); +out_put_orig: + nf_conntrack_expect_put(exp_orig); +out: + return ret; + +out_unexpect_both: + nf_conntrack_unexpect_related(exp_reply); +out_unexpect_orig: + nf_conntrack_unexpect_related(exp_orig); + goto out_put_both; +} + +static inline int +pptp_inbound_pkt(struct sk_buff **pskb, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq, + unsigned int reqlen, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo) +{ + struct nf_ct_pptp_master *info = &nfct_help(ct)->help.ct_pptp_info; + u_int16_t msg; + __be16 cid = 0, pcid = 0; + typeof(nf_nat_pptp_hook_inbound) nf_nat_pptp_inbound; + + msg = ntohs(ctlh->messageType); + DEBUGP("inbound control message %s\n", pptp_msg_name[msg]); + + switch (msg) { + case PPTP_START_SESSION_REPLY: + /* server confirms new control session */ + if (info->sstate < PPTP_SESSION_REQUESTED) + goto invalid; + if (pptpReq->srep.resultCode == PPTP_START_OK) + info->sstate = PPTP_SESSION_CONFIRMED; + else + info->sstate = PPTP_SESSION_ERROR; + break; + + case PPTP_STOP_SESSION_REPLY: + /* server confirms end of control session */ + if (info->sstate > PPTP_SESSION_STOPREQ) + goto invalid; + if (pptpReq->strep.resultCode == PPTP_STOP_OK) + info->sstate = PPTP_SESSION_NONE; + else + info->sstate = PPTP_SESSION_ERROR; + break; + + case PPTP_OUT_CALL_REPLY: + /* server accepted call, we now expect GRE frames */ + if (info->sstate != PPTP_SESSION_CONFIRMED) + goto invalid; + if (info->cstate != PPTP_CALL_OUT_REQ && + info->cstate != PPTP_CALL_OUT_CONF) + goto invalid; + + cid = pptpReq->ocack.callID; + pcid = pptpReq->ocack.peersCallID; + if (info->pns_call_id != pcid) + goto invalid; + DEBUGP("%s, CID=%X, PCID=%X\n", pptp_msg_name[msg], + ntohs(cid), ntohs(pcid)); + + if (pptpReq->ocack.resultCode == PPTP_OUTCALL_CONNECT) { + info->cstate = PPTP_CALL_OUT_CONF; + info->pac_call_id = cid; + exp_gre(ct, cid, pcid); + } else + info->cstate = PPTP_CALL_NONE; + break; + + case PPTP_IN_CALL_REQUEST: + /* server tells us about incoming call request */ + if (info->sstate != PPTP_SESSION_CONFIRMED) + goto invalid; + + cid = pptpReq->icreq.callID; + DEBUGP("%s, CID=%X\n", pptp_msg_name[msg], ntohs(cid)); + info->cstate = PPTP_CALL_IN_REQ; + info->pac_call_id = cid; + break; + + case PPTP_IN_CALL_CONNECT: + /* server tells us about incoming call established */ + if (info->sstate != PPTP_SESSION_CONFIRMED) + goto invalid; + if (info->cstate != PPTP_CALL_IN_REP && + info->cstate != PPTP_CALL_IN_CONF) + goto invalid; + + pcid = pptpReq->iccon.peersCallID; + cid = info->pac_call_id; + + if (info->pns_call_id != pcid) + goto invalid; + + DEBUGP("%s, PCID=%X\n", pptp_msg_name[msg], ntohs(pcid)); + info->cstate = PPTP_CALL_IN_CONF; + + /* we expect a GRE connection from PAC to PNS */ + exp_gre(ct, cid, pcid); + break; + + case PPTP_CALL_DISCONNECT_NOTIFY: + /* server confirms disconnect */ + cid = pptpReq->disc.callID; + DEBUGP("%s, CID=%X\n", pptp_msg_name[msg], ntohs(cid)); + info->cstate = PPTP_CALL_NONE; + + /* untrack this call id, unexpect GRE packets */ + pptp_destroy_siblings(ct); + break; + + case PPTP_WAN_ERROR_NOTIFY: + case PPTP_ECHO_REQUEST: + case PPTP_ECHO_REPLY: + /* I don't have to explain these ;) */ + break; + + default: + goto invalid; + } + + nf_nat_pptp_inbound = rcu_dereference(nf_nat_pptp_hook_inbound); + if (nf_nat_pptp_inbound && ct->status & IPS_NAT_MASK) + return nf_nat_pptp_inbound(pskb, ct, ctinfo, ctlh, pptpReq); + return NF_ACCEPT; + +invalid: + DEBUGP("invalid %s: type=%d cid=%u pcid=%u " + "cstate=%d sstate=%d pns_cid=%u pac_cid=%u\n", + msg <= PPTP_MSG_MAX ? pptp_msg_name[msg] : pptp_msg_name[0], + msg, ntohs(cid), ntohs(pcid), info->cstate, info->sstate, + ntohs(info->pns_call_id), ntohs(info->pac_call_id)); + return NF_ACCEPT; +} + +static inline int +pptp_outbound_pkt(struct sk_buff **pskb, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq, + unsigned int reqlen, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo) +{ + struct nf_ct_pptp_master *info = &nfct_help(ct)->help.ct_pptp_info; + u_int16_t msg; + __be16 cid = 0, pcid = 0; + typeof(nf_nat_pptp_hook_outbound) nf_nat_pptp_outbound; + + msg = ntohs(ctlh->messageType); + DEBUGP("outbound control message %s\n", pptp_msg_name[msg]); + + switch (msg) { + case PPTP_START_SESSION_REQUEST: + /* client requests for new control session */ + if (info->sstate != PPTP_SESSION_NONE) + goto invalid; + info->sstate = PPTP_SESSION_REQUESTED; + break; + + case PPTP_STOP_SESSION_REQUEST: + /* client requests end of control session */ + info->sstate = PPTP_SESSION_STOPREQ; + break; + + case PPTP_OUT_CALL_REQUEST: + /* client initiating connection to server */ + if (info->sstate != PPTP_SESSION_CONFIRMED) + goto invalid; + info->cstate = PPTP_CALL_OUT_REQ; + /* track PNS call id */ + cid = pptpReq->ocreq.callID; + DEBUGP("%s, CID=%X\n", pptp_msg_name[msg], ntohs(cid)); + info->pns_call_id = cid; + break; + + case PPTP_IN_CALL_REPLY: + /* client answers incoming call */ + if (info->cstate != PPTP_CALL_IN_REQ && + info->cstate != PPTP_CALL_IN_REP) + goto invalid; + + cid = pptpReq->icack.callID; + pcid = pptpReq->icack.peersCallID; + if (info->pac_call_id != pcid) + goto invalid; + DEBUGP("%s, CID=%X PCID=%X\n", pptp_msg_name[msg], + ntohs(cid), ntohs(pcid)); + + if (pptpReq->icack.resultCode == PPTP_INCALL_ACCEPT) { + /* part two of the three-way handshake */ + info->cstate = PPTP_CALL_IN_REP; + info->pns_call_id = cid; + } else + info->cstate = PPTP_CALL_NONE; + break; + + case PPTP_CALL_CLEAR_REQUEST: + /* client requests hangup of call */ + if (info->sstate != PPTP_SESSION_CONFIRMED) + goto invalid; + /* FUTURE: iterate over all calls and check if + * call ID is valid. We don't do this without newnat, + * because we only know about last call */ + info->cstate = PPTP_CALL_CLEAR_REQ; + break; + + case PPTP_SET_LINK_INFO: + case PPTP_ECHO_REQUEST: + case PPTP_ECHO_REPLY: + /* I don't have to explain these ;) */ + break; + + default: + goto invalid; + } + + nf_nat_pptp_outbound = rcu_dereference(nf_nat_pptp_hook_outbound); + if (nf_nat_pptp_outbound && ct->status & IPS_NAT_MASK) + return nf_nat_pptp_outbound(pskb, ct, ctinfo, ctlh, pptpReq); + return NF_ACCEPT; + +invalid: + DEBUGP("invalid %s: type=%d cid=%u pcid=%u " + "cstate=%d sstate=%d pns_cid=%u pac_cid=%u\n", + msg <= PPTP_MSG_MAX ? pptp_msg_name[msg] : pptp_msg_name[0], + msg, ntohs(cid), ntohs(pcid), info->cstate, info->sstate, + ntohs(info->pns_call_id), ntohs(info->pac_call_id)); + return NF_ACCEPT; +} + +static const unsigned int pptp_msg_size[] = { + [PPTP_START_SESSION_REQUEST] = sizeof(struct PptpStartSessionRequest), + [PPTP_START_SESSION_REPLY] = sizeof(struct PptpStartSessionReply), + [PPTP_STOP_SESSION_REQUEST] = sizeof(struct PptpStopSessionRequest), + [PPTP_STOP_SESSION_REPLY] = sizeof(struct PptpStopSessionReply), + [PPTP_OUT_CALL_REQUEST] = sizeof(struct PptpOutCallRequest), + [PPTP_OUT_CALL_REPLY] = sizeof(struct PptpOutCallReply), + [PPTP_IN_CALL_REQUEST] = sizeof(struct PptpInCallRequest), + [PPTP_IN_CALL_REPLY] = sizeof(struct PptpInCallReply), + [PPTP_IN_CALL_CONNECT] = sizeof(struct PptpInCallConnected), + [PPTP_CALL_CLEAR_REQUEST] = sizeof(struct PptpClearCallRequest), + [PPTP_CALL_DISCONNECT_NOTIFY] = sizeof(struct PptpCallDisconnectNotify), + [PPTP_WAN_ERROR_NOTIFY] = sizeof(struct PptpWanErrorNotify), + [PPTP_SET_LINK_INFO] = sizeof(struct PptpSetLinkInfo), +}; + +/* track caller id inside control connection, call expect_related */ +static int +conntrack_pptp_help(struct sk_buff **pskb, unsigned int protoff, + struct nf_conn *ct, enum ip_conntrack_info ctinfo) + +{ + int dir = CTINFO2DIR(ctinfo); + struct nf_ct_pptp_master *info = &nfct_help(ct)->help.ct_pptp_info; + struct tcphdr _tcph, *tcph; + struct pptp_pkt_hdr _pptph, *pptph; + struct PptpControlHeader _ctlh, *ctlh; + union pptp_ctrl_union _pptpReq, *pptpReq; + unsigned int tcplen = (*pskb)->len - protoff; + unsigned int datalen, reqlen, nexthdr_off; + int oldsstate, oldcstate; + int ret; + u_int16_t msg; + + /* don't do any tracking before tcp handshake complete */ + if (ctinfo != IP_CT_ESTABLISHED && + ctinfo != IP_CT_ESTABLISHED + IP_CT_IS_REPLY) + return NF_ACCEPT; + + nexthdr_off = protoff; + tcph = skb_header_pointer(*pskb, nexthdr_off, sizeof(_tcph), &_tcph); + BUG_ON(!tcph); + nexthdr_off += tcph->doff * 4; + datalen = tcplen - tcph->doff * 4; + + pptph = skb_header_pointer(*pskb, nexthdr_off, sizeof(_pptph), &_pptph); + if (!pptph) { + DEBUGP("no full PPTP header, can't track\n"); + return NF_ACCEPT; + } + nexthdr_off += sizeof(_pptph); + datalen -= sizeof(_pptph); + + /* if it's not a control message we can't do anything with it */ + if (ntohs(pptph->packetType) != PPTP_PACKET_CONTROL || + ntohl(pptph->magicCookie) != PPTP_MAGIC_COOKIE) { + DEBUGP("not a control packet\n"); + return NF_ACCEPT; + } + + ctlh = skb_header_pointer(*pskb, nexthdr_off, sizeof(_ctlh), &_ctlh); + if (!ctlh) + return NF_ACCEPT; + nexthdr_off += sizeof(_ctlh); + datalen -= sizeof(_ctlh); + + reqlen = datalen; + msg = ntohs(ctlh->messageType); + if (msg > 0 && msg <= PPTP_MSG_MAX && reqlen < pptp_msg_size[msg]) + return NF_ACCEPT; + if (reqlen > sizeof(*pptpReq)) + reqlen = sizeof(*pptpReq); + + pptpReq = skb_header_pointer(*pskb, nexthdr_off, reqlen, &_pptpReq); + if (!pptpReq) + return NF_ACCEPT; + + oldsstate = info->sstate; + oldcstate = info->cstate; + + spin_lock_bh(&nf_pptp_lock); + + /* FIXME: We just blindly assume that the control connection is always + * established from PNS->PAC. However, RFC makes no guarantee */ + if (dir == IP_CT_DIR_ORIGINAL) + /* client -> server (PNS -> PAC) */ + ret = pptp_outbound_pkt(pskb, ctlh, pptpReq, reqlen, ct, + ctinfo); + else + /* server -> client (PAC -> PNS) */ + ret = pptp_inbound_pkt(pskb, ctlh, pptpReq, reqlen, ct, + ctinfo); + DEBUGP("sstate: %d->%d, cstate: %d->%d\n", + oldsstate, info->sstate, oldcstate, info->cstate); + spin_unlock_bh(&nf_pptp_lock); + + return ret; +} + +/* control protocol helper */ +static struct nf_conntrack_helper pptp __read_mostly = { + .name = "pptp", + .me = THIS_MODULE, + .max_expected = 2, + .timeout = 5 * 60, + .tuple.src.l3num = AF_INET, + .tuple.src.u.tcp.port = __constant_htons(PPTP_CONTROL_PORT), + .tuple.dst.protonum = IPPROTO_TCP, + .mask.src.l3num = 0xffff, + .mask.src.u.tcp.port = __constant_htons(0xffff), + .mask.dst.protonum = 0xff, + .help = conntrack_pptp_help, + .destroy = pptp_destroy_siblings, +}; + +static int __init nf_conntrack_pptp_init(void) +{ + return nf_conntrack_helper_register(&pptp); +} + +static void __exit nf_conntrack_pptp_fini(void) +{ + nf_conntrack_helper_unregister(&pptp); + nf_ct_gre_keymap_flush(); +} + +module_init(nf_conntrack_pptp_init); +module_exit(nf_conntrack_pptp_fini); diff --git a/net/netfilter/nf_conntrack_proto_gre.c b/net/netfilter/nf_conntrack_proto_gre.c new file mode 100644 index 00000000000..ac193ce7024 --- /dev/null +++ b/net/netfilter/nf_conntrack_proto_gre.c @@ -0,0 +1,305 @@ +/* + * ip_conntrack_proto_gre.c - Version 3.0 + * + * Connection tracking protocol helper module for GRE. + * + * GRE is a generic encapsulation protocol, which is generally not very + * suited for NAT, as it has no protocol-specific part as port numbers. + * + * It has an optional key field, which may help us distinguishing two + * connections between the same two hosts. + * + * GRE is defined in RFC 1701 and RFC 1702, as well as RFC 2784 + * + * PPTP is built on top of a modified version of GRE, and has a mandatory + * field called "CallID", which serves us for the same purpose as the key + * field in plain GRE. + * + * Documentation about PPTP can be found in RFC 2637 + * + * (C) 2000-2005 by Harald Welte + * + * Development of this code funded by Astaro AG (http://www.astaro.com/) + * + */ + +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +#define GRE_TIMEOUT (30 * HZ) +#define GRE_STREAM_TIMEOUT (180 * HZ) + +#if 0 +#define DEBUGP(format, args...) printk(KERN_DEBUG "%s:%s: " format, __FILE__, __FUNCTION__, ## args) +#else +#define DEBUGP(x, args...) +#endif + +static DEFINE_RWLOCK(nf_ct_gre_lock); +static LIST_HEAD(gre_keymap_list); + +void nf_ct_gre_keymap_flush(void) +{ + struct list_head *pos, *n; + + write_lock_bh(&nf_ct_gre_lock); + list_for_each_safe(pos, n, &gre_keymap_list) { + list_del(pos); + kfree(pos); + } + write_unlock_bh(&nf_ct_gre_lock); +} +EXPORT_SYMBOL(nf_ct_gre_keymap_flush); + +static inline int gre_key_cmpfn(const struct nf_ct_gre_keymap *km, + const struct nf_conntrack_tuple *t) +{ + return km->tuple.src.l3num == t->src.l3num && + !memcmp(&km->tuple.src.u3, &t->src.u3, sizeof(t->src.u3)) && + !memcmp(&km->tuple.dst.u3, &t->dst.u3, sizeof(t->dst.u3)) && + km->tuple.dst.protonum == t->dst.protonum && + km->tuple.dst.u.all == t->dst.u.all; +} + +/* look up the source key for a given tuple */ +static __be16 gre_keymap_lookup(struct nf_conntrack_tuple *t) +{ + struct nf_ct_gre_keymap *km; + __be16 key = 0; + + read_lock_bh(&nf_ct_gre_lock); + list_for_each_entry(km, &gre_keymap_list, list) { + if (gre_key_cmpfn(km, t)) { + key = km->tuple.src.u.gre.key; + break; + } + } + read_unlock_bh(&nf_ct_gre_lock); + + DEBUGP("lookup src key 0x%x for ", key); + NF_CT_DUMP_TUPLE(t); + + return key; +} + +/* add a single keymap entry, associate with specified master ct */ +int nf_ct_gre_keymap_add(struct nf_conn *ct, enum ip_conntrack_dir dir, + struct nf_conntrack_tuple *t) +{ + struct nf_conn_help *help = nfct_help(ct); + struct nf_ct_gre_keymap **kmp, *km; + + BUG_ON(strcmp(help->helper->name, "pptp")); + kmp = &help->help.ct_pptp_info.keymap[dir]; + if (*kmp) { + /* check whether it's a retransmission */ + list_for_each_entry(km, &gre_keymap_list, list) { + if (gre_key_cmpfn(km, t) && km == *kmp) + return 0; + } + DEBUGP("trying to override keymap_%s for ct %p\n", + dir == IP_CT_DIR_REPLY ? "reply" : "orig", ct); + return -EEXIST; + } + + km = kmalloc(sizeof(*km), GFP_ATOMIC); + if (!km) + return -ENOMEM; + memcpy(&km->tuple, t, sizeof(*t)); + *kmp = km; + + DEBUGP("adding new entry %p: ", km); + NF_CT_DUMP_TUPLE(&km->tuple); + + write_lock_bh(&nf_ct_gre_lock); + list_add_tail(&km->list, &gre_keymap_list); + write_unlock_bh(&nf_ct_gre_lock); + + return 0; +} +EXPORT_SYMBOL_GPL(nf_ct_gre_keymap_add); + +/* destroy the keymap entries associated with specified master ct */ +void nf_ct_gre_keymap_destroy(struct nf_conn *ct) +{ + struct nf_conn_help *help = nfct_help(ct); + enum ip_conntrack_dir dir; + + DEBUGP("entering for ct %p\n", ct); + BUG_ON(strcmp(help->helper->name, "pptp")); + + write_lock_bh(&nf_ct_gre_lock); + for (dir = IP_CT_DIR_ORIGINAL; dir < IP_CT_DIR_MAX; dir++) { + if (help->help.ct_pptp_info.keymap[dir]) { + DEBUGP("removing %p from list\n", + help->help.ct_pptp_info.keymap[dir]); + list_del(&help->help.ct_pptp_info.keymap[dir]->list); + kfree(help->help.ct_pptp_info.keymap[dir]); + help->help.ct_pptp_info.keymap[dir] = NULL; + } + } + write_unlock_bh(&nf_ct_gre_lock); +} +EXPORT_SYMBOL_GPL(nf_ct_gre_keymap_destroy); + +/* PUBLIC CONNTRACK PROTO HELPER FUNCTIONS */ + +/* invert gre part of tuple */ +static int gre_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + tuple->dst.u.gre.key = orig->src.u.gre.key; + tuple->src.u.gre.key = orig->dst.u.gre.key; + return 1; +} + +/* gre hdr info to tuple */ +static int gre_pkt_to_tuple(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple) +{ + struct gre_hdr_pptp _pgrehdr, *pgrehdr; + __be16 srckey; + struct gre_hdr _grehdr, *grehdr; + + /* first only delinearize old RFC1701 GRE header */ + grehdr = skb_header_pointer(skb, dataoff, sizeof(_grehdr), &_grehdr); + if (!grehdr || grehdr->version != GRE_VERSION_PPTP) { + /* try to behave like "nf_conntrack_proto_generic" */ + tuple->src.u.all = 0; + tuple->dst.u.all = 0; + return 1; + } + + /* PPTP header is variable length, only need up to the call_id field */ + pgrehdr = skb_header_pointer(skb, dataoff, 8, &_pgrehdr); + if (!pgrehdr) + return 1; + + if (ntohs(grehdr->protocol) != GRE_PROTOCOL_PPTP) { + DEBUGP("GRE_VERSION_PPTP but unknown proto\n"); + return 0; + } + + tuple->dst.u.gre.key = pgrehdr->call_id; + srckey = gre_keymap_lookup(tuple); + tuple->src.u.gre.key = srckey; + + return 1; +} + +/* print gre part of tuple */ +static int gre_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return seq_printf(s, "srckey=0x%x dstkey=0x%x ", + ntohs(tuple->src.u.gre.key), + ntohs(tuple->dst.u.gre.key)); +} + +/* print private data for conntrack */ +static int gre_print_conntrack(struct seq_file *s, + const struct nf_conn *ct) +{ + return seq_printf(s, "timeout=%u, stream_timeout=%u ", + (ct->proto.gre.timeout / HZ), + (ct->proto.gre.stream_timeout / HZ)); +} + +/* Returns verdict for packet, and may modify conntrack */ +static int gre_packet(struct nf_conn *ct, + const struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info ctinfo, + int pf, + unsigned int hooknum) +{ + /* If we've seen traffic both ways, this is a GRE connection. + * Extend timeout. */ + if (ct->status & IPS_SEEN_REPLY) { + nf_ct_refresh_acct(ct, ctinfo, skb, + ct->proto.gre.stream_timeout); + /* Also, more likely to be important, and not a probe. */ + set_bit(IPS_ASSURED_BIT, &ct->status); + nf_conntrack_event_cache(IPCT_STATUS, skb); + } else + nf_ct_refresh_acct(ct, ctinfo, skb, + ct->proto.gre.timeout); + + return NF_ACCEPT; +} + +/* Called when a new connection for this protocol found. */ +static int gre_new(struct nf_conn *ct, const struct sk_buff *skb, + unsigned int dataoff) +{ + DEBUGP(": "); + NF_CT_DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + + /* initialize to sane value. Ideally a conntrack helper + * (e.g. in case of pptp) is increasing them */ + ct->proto.gre.stream_timeout = GRE_STREAM_TIMEOUT; + ct->proto.gre.timeout = GRE_TIMEOUT; + + return 1; +} + +/* Called when a conntrack entry has already been removed from the hashes + * and is about to be deleted from memory */ +static void gre_destroy(struct nf_conn *ct) +{ + struct nf_conn *master = ct->master; + DEBUGP(" entering\n"); + + if (!master) + DEBUGP("no master !?!\n"); + else + nf_ct_gre_keymap_destroy(master); +} + +/* protocol helper struct */ +static struct nf_conntrack_l4proto nf_conntrack_l4proto_gre4 = { + .l3proto = AF_INET, + .l4proto = IPPROTO_GRE, + .name = "gre", + .pkt_to_tuple = gre_pkt_to_tuple, + .invert_tuple = gre_invert_tuple, + .print_tuple = gre_print_tuple, + .print_conntrack = gre_print_conntrack, + .packet = gre_packet, + .new = gre_new, + .destroy = gre_destroy, + .me = THIS_MODULE, +#if defined(CONFIG_NF_CONNTRACK_NETLINK) || \ + defined(CONFIG_NF_CONNTRACK_NETLINK_MODULE) + .tuple_to_nfattr = nf_ct_port_tuple_to_nfattr, + .nfattr_to_tuple = nf_ct_port_nfattr_to_tuple, +#endif +}; + +static int __init nf_ct_proto_gre_init(void) +{ + return nf_conntrack_l4proto_register(&nf_conntrack_l4proto_gre4); +} + +static void nf_ct_proto_gre_fini(void) +{ + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_gre4); + nf_ct_gre_keymap_flush(); +} + +module_init(nf_ct_proto_gre_init); +module_exit(nf_ct_proto_gre_fini); + +MODULE_LICENSE("GPL"); -- cgit v1.2.3-70-g09d2 From 9fafcd7b203229c3f3893a475741afc27e276306 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 2 Dec 2006 22:09:57 -0800 Subject: [NETFILTER]: nf_conntrack/nf_nat: add SIP helper port Add IPv4 and IPv6 capable nf_conntrack port of the SIP conntrack/NAT helper. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_sip.h | 41 +++ net/ipv4/netfilter/Kconfig | 5 + net/ipv4/netfilter/Makefile | 1 + net/ipv4/netfilter/nf_nat_sip.c | 283 +++++++++++++++ net/netfilter/Kconfig | 12 + net/netfilter/Makefile | 1 + net/netfilter/nf_conntrack_sip.c | 530 +++++++++++++++++++++++++++++ 7 files changed, 873 insertions(+) create mode 100644 include/linux/netfilter/nf_conntrack_sip.h create mode 100644 net/ipv4/netfilter/nf_nat_sip.c create mode 100644 net/netfilter/nf_conntrack_sip.c (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h new file mode 100644 index 00000000000..bb7f2041db7 --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -0,0 +1,41 @@ +#ifndef __NF_CONNTRACK_SIP_H__ +#define __NF_CONNTRACK_SIP_H__ +#ifdef __KERNEL__ + +#define SIP_PORT 5060 +#define SIP_TIMEOUT 3600 + +enum sip_header_pos { + POS_REG_REQ_URI, + POS_REQ_URI, + POS_FROM, + POS_TO, + POS_VIA, + POS_CONTACT, + POS_CONTENT, + POS_MEDIA, + POS_OWNER_IP4, + POS_CONNECTION_IP4, + POS_OWNER_IP6, + POS_CONNECTION_IP6, + POS_SDP_HEADER, +}; + +extern unsigned int (*nf_nat_sip_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conn *ct, + const char **dptr); +extern unsigned int (*nf_nat_sdp_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conntrack_expect *exp, + const char *dptr); + +extern int ct_sip_get_info(struct nf_conn *ct, const char *dptr, size_t dlen, + unsigned int *matchoff, unsigned int *matchlen, + enum sip_header_pos pos); +extern int ct_sip_lnlen(const char *line, const char *limit); +extern const char *ct_sip_search(const char *needle, const char *haystack, + size_t needle_len, size_t haystack_len, + int case_sensitive); +#endif /* __KERNEL__ */ +#endif /* __NF_CONNTRACK_SIP_H__ */ diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index c3327ac024d..83e83f553cc 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -555,6 +555,11 @@ config IP_NF_NAT_SIP default IP_NF_NAT if IP_NF_SIP=y default m if IP_NF_SIP=m +config NF_NAT_SIP + tristate + depends on IP_NF_IPTABLES && NF_CONNTRACK && NF_NAT + default NF_NAT && NF_CONNTRACK_SIP + # mangle + specific targets config IP_NF_MANGLE tristate "Packet mangling" diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index ef33ff2cdda..167b65e89ac 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -55,6 +55,7 @@ obj-$(CONFIG_NF_NAT_FTP) += nf_nat_ftp.o obj-$(CONFIG_NF_NAT_H323) += nf_nat_h323.o obj-$(CONFIG_NF_NAT_IRC) += nf_nat_irc.o obj-$(CONFIG_NF_NAT_PPTP) += nf_nat_pptp.o +obj-$(CONFIG_NF_NAT_SIP) += nf_nat_sip.o # NAT protocols (nf_nat) obj-$(CONFIG_NF_NAT_PROTO_GRE) += nf_nat_proto_gre.o diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c new file mode 100644 index 00000000000..3d524b95731 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_sip.c @@ -0,0 +1,283 @@ +/* SIP extension for UDP NAT alteration. + * + * (C) 2005 by Christian Hentschel + * based on RR's ip_nat_ftp.c and other modules. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Christian Hentschel "); +MODULE_DESCRIPTION("SIP NAT helper"); +MODULE_ALIAS("ip_nat_sip"); + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +struct addr_map { + struct { + char src[sizeof("nnn.nnn.nnn.nnn:nnnnn")]; + char dst[sizeof("nnn.nnn.nnn.nnn:nnnnn")]; + unsigned int srclen, srciplen; + unsigned int dstlen, dstiplen; + } addr[IP_CT_DIR_MAX]; +}; + +static void addr_map_init(struct nf_conn *ct, struct addr_map *map) +{ + struct nf_conntrack_tuple *t; + enum ip_conntrack_dir dir; + unsigned int n; + + for (dir = 0; dir < IP_CT_DIR_MAX; dir++) { + t = &ct->tuplehash[dir].tuple; + + n = sprintf(map->addr[dir].src, "%u.%u.%u.%u", + NIPQUAD(t->src.u3.ip)); + map->addr[dir].srciplen = n; + n += sprintf(map->addr[dir].src + n, ":%u", + ntohs(t->src.u.udp.port)); + map->addr[dir].srclen = n; + + n = sprintf(map->addr[dir].dst, "%u.%u.%u.%u", + NIPQUAD(t->dst.u3.ip)); + map->addr[dir].dstiplen = n; + n += sprintf(map->addr[dir].dst + n, ":%u", + ntohs(t->dst.u.udp.port)); + map->addr[dir].dstlen = n; + } +} + +static int map_sip_addr(struct sk_buff **pskb, enum ip_conntrack_info ctinfo, + struct nf_conn *ct, const char **dptr, size_t dlen, + enum sip_header_pos pos, struct addr_map *map) +{ + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + unsigned int matchlen, matchoff, addrlen; + char *addr; + + if (ct_sip_get_info(ct, *dptr, dlen, &matchoff, &matchlen, pos) <= 0) + return 1; + + if ((matchlen == map->addr[dir].srciplen || + matchlen == map->addr[dir].srclen) && + memcmp(*dptr + matchoff, map->addr[dir].src, matchlen) == 0) { + addr = map->addr[!dir].dst; + addrlen = map->addr[!dir].dstlen; + } else if ((matchlen == map->addr[dir].dstiplen || + matchlen == map->addr[dir].dstlen) && + memcmp(*dptr + matchoff, map->addr[dir].dst, matchlen) == 0) { + addr = map->addr[!dir].src; + addrlen = map->addr[!dir].srclen; + } else + return 1; + + if (!nf_nat_mangle_udp_packet(pskb, ct, ctinfo, + matchoff, matchlen, addr, addrlen)) + return 0; + *dptr = (*pskb)->data + (*pskb)->nh.iph->ihl*4 + sizeof(struct udphdr); + return 1; + +} + +static unsigned int ip_nat_sip(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conn *ct, + const char **dptr) +{ + enum sip_header_pos pos; + struct addr_map map; + int dataoff, datalen; + + dataoff = (*pskb)->nh.iph->ihl*4 + sizeof(struct udphdr); + datalen = (*pskb)->len - dataoff; + if (datalen < sizeof("SIP/2.0") - 1) + return NF_DROP; + + addr_map_init(ct, &map); + + /* Basic rules: requests and responses. */ + if (strncmp(*dptr, "SIP/2.0", sizeof("SIP/2.0") - 1) != 0) { + /* 10.2: Constructing the REGISTER Request: + * + * The "userinfo" and "@" components of the SIP URI MUST NOT + * be present. + */ + if (datalen >= sizeof("REGISTER") - 1 && + strncmp(*dptr, "REGISTER", sizeof("REGISTER") - 1) == 0) + pos = POS_REG_REQ_URI; + else + pos = POS_REQ_URI; + + if (!map_sip_addr(pskb, ctinfo, ct, dptr, datalen, pos, &map)) + return NF_DROP; + } + + if (!map_sip_addr(pskb, ctinfo, ct, dptr, datalen, POS_FROM, &map) || + !map_sip_addr(pskb, ctinfo, ct, dptr, datalen, POS_TO, &map) || + !map_sip_addr(pskb, ctinfo, ct, dptr, datalen, POS_VIA, &map) || + !map_sip_addr(pskb, ctinfo, ct, dptr, datalen, POS_CONTACT, &map)) + return NF_DROP; + return NF_ACCEPT; +} + +static unsigned int mangle_sip_packet(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conn *ct, + const char **dptr, size_t dlen, + char *buffer, int bufflen, + enum sip_header_pos pos) +{ + unsigned int matchlen, matchoff; + + if (ct_sip_get_info(ct, *dptr, dlen, &matchoff, &matchlen, pos) <= 0) + return 0; + + if (!nf_nat_mangle_udp_packet(pskb, ct, ctinfo, + matchoff, matchlen, buffer, bufflen)) + return 0; + + /* We need to reload this. Thanks Patrick. */ + *dptr = (*pskb)->data + (*pskb)->nh.iph->ihl*4 + sizeof(struct udphdr); + return 1; +} + +static int mangle_content_len(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conn *ct, + const char *dptr) +{ + unsigned int dataoff, matchoff, matchlen; + char buffer[sizeof("65536")]; + int bufflen; + + dataoff = (*pskb)->nh.iph->ihl*4 + sizeof(struct udphdr); + + /* Get actual SDP lenght */ + if (ct_sip_get_info(ct, dptr, (*pskb)->len - dataoff, &matchoff, + &matchlen, POS_SDP_HEADER) > 0) { + + /* since ct_sip_get_info() give us a pointer passing 'v=' + we need to add 2 bytes in this count. */ + int c_len = (*pskb)->len - dataoff - matchoff + 2; + + /* Now, update SDP length */ + if (ct_sip_get_info(ct, dptr, (*pskb)->len - dataoff, &matchoff, + &matchlen, POS_CONTENT) > 0) { + + bufflen = sprintf(buffer, "%u", c_len); + return nf_nat_mangle_udp_packet(pskb, ct, ctinfo, + matchoff, matchlen, + buffer, bufflen); + } + } + return 0; +} + +static unsigned int mangle_sdp(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conn *ct, + __be32 newip, u_int16_t port, + const char *dptr) +{ + char buffer[sizeof("nnn.nnn.nnn.nnn")]; + unsigned int dataoff, bufflen; + + dataoff = (*pskb)->nh.iph->ihl*4 + sizeof(struct udphdr); + + /* Mangle owner and contact info. */ + bufflen = sprintf(buffer, "%u.%u.%u.%u", NIPQUAD(newip)); + if (!mangle_sip_packet(pskb, ctinfo, ct, &dptr, (*pskb)->len - dataoff, + buffer, bufflen, POS_OWNER_IP4)) + return 0; + + if (!mangle_sip_packet(pskb, ctinfo, ct, &dptr, (*pskb)->len - dataoff, + buffer, bufflen, POS_CONNECTION_IP4)) + return 0; + + /* Mangle media port. */ + bufflen = sprintf(buffer, "%u", port); + if (!mangle_sip_packet(pskb, ctinfo, ct, &dptr, (*pskb)->len - dataoff, + buffer, bufflen, POS_MEDIA)) + return 0; + + return mangle_content_len(pskb, ctinfo, ct, dptr); +} + +/* So, this packet has hit the connection tracking matching code. + Mangle it, and change the expectation to match the new version. */ +static unsigned int ip_nat_sdp(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conntrack_expect *exp, + const char *dptr) +{ + struct nf_conn *ct = exp->master; + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + __be32 newip; + u_int16_t port; + + DEBUGP("ip_nat_sdp():\n"); + + /* Connection will come from reply */ + newip = ct->tuplehash[!dir].tuple.dst.u3.ip; + + exp->tuple.dst.u3.ip = newip; + exp->saved_proto.udp.port = exp->tuple.dst.u.udp.port; + exp->dir = !dir; + + /* When you see the packet, we need to NAT it the same as the + this one. */ + exp->expectfn = nf_nat_follow_master; + + /* Try to get same port: if not, try to change it. */ + for (port = ntohs(exp->saved_proto.udp.port); port != 0; port++) { + exp->tuple.dst.u.udp.port = htons(port); + if (nf_conntrack_expect_related(exp) == 0) + break; + } + + if (port == 0) + return NF_DROP; + + if (!mangle_sdp(pskb, ctinfo, ct, newip, port, dptr)) { + nf_conntrack_unexpect_related(exp); + return NF_DROP; + } + return NF_ACCEPT; +} + +static void __exit nf_nat_sip_fini(void) +{ + rcu_assign_pointer(nf_nat_sip_hook, NULL); + rcu_assign_pointer(nf_nat_sdp_hook, NULL); + synchronize_rcu(); +} + +static int __init nf_nat_sip_init(void) +{ + BUG_ON(rcu_dereference(nf_nat_sip_hook)); + BUG_ON(rcu_dereference(nf_nat_sdp_hook)); + rcu_assign_pointer(nf_nat_sip_hook, ip_nat_sip); + rcu_assign_pointer(nf_nat_sdp_hook, ip_nat_sdp); + return 0; +} + +module_init(nf_nat_sip_init); +module_exit(nf_nat_sip_fini); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 6b2eb26ae03..e73891194b3 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -236,6 +236,18 @@ config NF_CONNTRACK_PPTP To compile it as a module, choose M here. If unsure, say N. +config NF_CONNTRACK_SIP + tristate "SIP protocol support (EXPERIMENTAL)" + depends on EXPERIMENTAL && NF_CONNTRACK + help + SIP is an application-layer control protocol that can establish, + modify, and terminate multimedia sessions (conferences) such as + Internet telephony calls. With the ip_conntrack_sip and + the nf_nat_sip modules you can support the protocol on a connection + tracking/NATing firewall. + + To compile it as a module, choose M here. If unsure, say N. + config NF_CT_NETLINK tristate 'Connection tracking netlink interface (EXPERIMENTAL)' depends on EXPERIMENTAL && NF_CONNTRACK && NETFILTER_NETLINK diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 897bed4cbd7..86da21700aa 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -29,6 +29,7 @@ obj-$(CONFIG_NF_CONNTRACK_H323) += nf_conntrack_h323.o obj-$(CONFIG_NF_CONNTRACK_IRC) += nf_conntrack_irc.o obj-$(CONFIG_NF_CONNTRACK_NETBIOS_NS) += nf_conntrack_netbios_ns.o obj-$(CONFIG_NF_CONNTRACK_PPTP) += nf_conntrack_pptp.o +obj-$(CONFIG_NF_CONNTRACK_SIP) += nf_conntrack_sip.o # generic X tables obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c new file mode 100644 index 00000000000..e0c06795a94 --- /dev/null +++ b/net/netfilter/nf_conntrack_sip.c @@ -0,0 +1,530 @@ +/* SIP extension for IP connection tracking. + * + * (C) 2005 by Christian Hentschel + * based on RR's ip_conntrack_ftp.c and other modules. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Christian Hentschel "); +MODULE_DESCRIPTION("SIP connection tracking helper"); +MODULE_ALIAS("ip_conntrack_sip"); + +#define MAX_PORTS 8 +static unsigned short ports[MAX_PORTS]; +static int ports_c; +module_param_array(ports, ushort, &ports_c, 0400); +MODULE_PARM_DESC(ports, "port numbers of SIP servers"); + +static unsigned int sip_timeout __read_mostly = SIP_TIMEOUT; +module_param(sip_timeout, uint, 0600); +MODULE_PARM_DESC(sip_timeout, "timeout for the master SIP session"); + +unsigned int (*nf_nat_sip_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conn *ct, + const char **dptr) __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_sip_hook); + +unsigned int (*nf_nat_sdp_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conntrack_expect *exp, + const char *dptr) __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_sdp_hook); + +static int digits_len(struct nf_conn *, const char *, const char *, int *); +static int epaddr_len(struct nf_conn *, const char *, const char *, int *); +static int skp_digits_len(struct nf_conn *, const char *, const char *, int *); +static int skp_epaddr_len(struct nf_conn *, const char *, const char *, int *); + +struct sip_header_nfo { + const char *lname; + const char *sname; + const char *ln_str; + size_t lnlen; + size_t snlen; + size_t ln_strlen; + int case_sensitive; + int (*match_len)(struct nf_conn *, const char *, + const char *, int *); +}; + +static const struct sip_header_nfo ct_sip_hdrs[] = { + [POS_REG_REQ_URI] = { /* SIP REGISTER request URI */ + .lname = "sip:", + .lnlen = sizeof("sip:") - 1, + .ln_str = ":", + .ln_strlen = sizeof(":") - 1, + .match_len = epaddr_len, + }, + [POS_REQ_URI] = { /* SIP request URI */ + .lname = "sip:", + .lnlen = sizeof("sip:") - 1, + .ln_str = "@", + .ln_strlen = sizeof("@") - 1, + .match_len = epaddr_len, + }, + [POS_FROM] = { /* SIP From header */ + .lname = "From:", + .lnlen = sizeof("From:") - 1, + .sname = "\r\nf:", + .snlen = sizeof("\r\nf:") - 1, + .ln_str = "sip:", + .ln_strlen = sizeof("sip:") - 1, + .match_len = skp_epaddr_len, + }, + [POS_TO] = { /* SIP To header */ + .lname = "To:", + .lnlen = sizeof("To:") - 1, + .sname = "\r\nt:", + .snlen = sizeof("\r\nt:") - 1, + .ln_str = "sip:", + .ln_strlen = sizeof("sip:") - 1, + .match_len = skp_epaddr_len + }, + [POS_VIA] = { /* SIP Via header */ + .lname = "Via:", + .lnlen = sizeof("Via:") - 1, + .sname = "\r\nv:", + .snlen = sizeof("\r\nv:") - 1, /* rfc3261 "\r\n" */ + .ln_str = "UDP ", + .ln_strlen = sizeof("UDP ") - 1, + .match_len = epaddr_len, + }, + [POS_CONTACT] = { /* SIP Contact header */ + .lname = "Contact:", + .lnlen = sizeof("Contact:") - 1, + .sname = "\r\nm:", + .snlen = sizeof("\r\nm:") - 1, + .ln_str = "sip:", + .ln_strlen = sizeof("sip:") - 1, + .match_len = skp_epaddr_len + }, + [POS_CONTENT] = { /* SIP Content length header */ + .lname = "Content-Length:", + .lnlen = sizeof("Content-Length:") - 1, + .sname = "\r\nl:", + .snlen = sizeof("\r\nl:") - 1, + .ln_str = ":", + .ln_strlen = sizeof(":") - 1, + .match_len = skp_digits_len + }, + [POS_MEDIA] = { /* SDP media info */ + .case_sensitive = 1, + .lname = "\nm=", + .lnlen = sizeof("\nm=") - 1, + .sname = "\rm=", + .snlen = sizeof("\rm=") - 1, + .ln_str = "audio ", + .ln_strlen = sizeof("audio ") - 1, + .match_len = digits_len + }, + [POS_OWNER_IP4] = { /* SDP owner address*/ + .case_sensitive = 1, + .lname = "\no=", + .lnlen = sizeof("\no=") - 1, + .sname = "\ro=", + .snlen = sizeof("\ro=") - 1, + .ln_str = "IN IP4 ", + .ln_strlen = sizeof("IN IP4 ") - 1, + .match_len = epaddr_len + }, + [POS_CONNECTION_IP4] = {/* SDP connection info */ + .case_sensitive = 1, + .lname = "\nc=", + .lnlen = sizeof("\nc=") - 1, + .sname = "\rc=", + .snlen = sizeof("\rc=") - 1, + .ln_str = "IN IP4 ", + .ln_strlen = sizeof("IN IP4 ") - 1, + .match_len = epaddr_len + }, + [POS_OWNER_IP6] = { /* SDP owner address*/ + .case_sensitive = 1, + .lname = "\no=", + .lnlen = sizeof("\no=") - 1, + .sname = "\ro=", + .snlen = sizeof("\ro=") - 1, + .ln_str = "IN IP6 ", + .ln_strlen = sizeof("IN IP6 ") - 1, + .match_len = epaddr_len + }, + [POS_CONNECTION_IP6] = {/* SDP connection info */ + .case_sensitive = 1, + .lname = "\nc=", + .lnlen = sizeof("\nc=") - 1, + .sname = "\rc=", + .snlen = sizeof("\rc=") - 1, + .ln_str = "IN IP6 ", + .ln_strlen = sizeof("IN IP6 ") - 1, + .match_len = epaddr_len + }, + [POS_SDP_HEADER] = { /* SDP version header */ + .case_sensitive = 1, + .lname = "\nv=", + .lnlen = sizeof("\nv=") - 1, + .sname = "\rv=", + .snlen = sizeof("\rv=") - 1, + .ln_str = "=", + .ln_strlen = sizeof("=") - 1, + .match_len = digits_len + } +}; + +/* get line lenght until first CR or LF seen. */ +int ct_sip_lnlen(const char *line, const char *limit) +{ + const char *k = line; + + while ((line <= limit) && (*line == '\r' || *line == '\n')) + line++; + + while (line <= limit) { + if (*line == '\r' || *line == '\n') + break; + line++; + } + return line - k; +} +EXPORT_SYMBOL_GPL(ct_sip_lnlen); + +/* Linear string search, case sensitive. */ +const char *ct_sip_search(const char *needle, const char *haystack, + size_t needle_len, size_t haystack_len, + int case_sensitive) +{ + const char *limit = haystack + (haystack_len - needle_len); + + while (haystack <= limit) { + if (case_sensitive) { + if (strncmp(haystack, needle, needle_len) == 0) + return haystack; + } else { + if (strnicmp(haystack, needle, needle_len) == 0) + return haystack; + } + haystack++; + } + return NULL; +} +EXPORT_SYMBOL_GPL(ct_sip_search); + +static int digits_len(struct nf_conn *ct, const char *dptr, + const char *limit, int *shift) +{ + int len = 0; + while (dptr <= limit && isdigit(*dptr)) { + dptr++; + len++; + } + return len; +} + +/* get digits lenght, skiping blank spaces. */ +static int skp_digits_len(struct nf_conn *ct, const char *dptr, + const char *limit, int *shift) +{ + for (; dptr <= limit && *dptr == ' '; dptr++) + (*shift)++; + + return digits_len(ct, dptr, limit, shift); +} + +static int parse_addr(struct nf_conn *ct, const char *cp, const char **endp, + union nf_conntrack_address *addr, const char *limit) +{ + const char *end; + int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; + int ret = 0; + + switch (family) { + case AF_INET: + ret = in4_pton(cp, limit - cp, (u8 *)&addr->ip, -1, &end); + break; + case AF_INET6: + ret = in6_pton(cp, limit - cp, (u8 *)&addr->ip6, -1, &end); + break; + default: + BUG(); + } + + if (ret == 0 || end == cp) + return 0; + if (endp) + *endp = end; + return 1; +} + +/* skip ip address. returns its length. */ +static int epaddr_len(struct nf_conn *ct, const char *dptr, + const char *limit, int *shift) +{ + union nf_conntrack_address addr; + const char *aux = dptr; + + if (!parse_addr(ct, dptr, &dptr, &addr, limit)) { + DEBUGP("ip: %s parse failed.!\n", dptr); + return 0; + } + + /* Port number */ + if (*dptr == ':') { + dptr++; + dptr += digits_len(ct, dptr, limit, shift); + } + return dptr - aux; +} + +/* get address length, skiping user info. */ +static int skp_epaddr_len(struct nf_conn *ct, const char *dptr, + const char *limit, int *shift) +{ + int s = *shift; + + for (; dptr <= limit && *dptr != '@'; dptr++) + (*shift)++; + + if (*dptr == '@') { + dptr++; + (*shift)++; + } else + *shift = s; + + return epaddr_len(ct, dptr, limit, shift); +} + +/* Returns 0 if not found, -1 error parsing. */ +int ct_sip_get_info(struct nf_conn *ct, + const char *dptr, size_t dlen, + unsigned int *matchoff, + unsigned int *matchlen, + enum sip_header_pos pos) +{ + const struct sip_header_nfo *hnfo = &ct_sip_hdrs[pos]; + const char *limit, *aux, *k = dptr; + int shift = 0; + + limit = dptr + (dlen - hnfo->lnlen); + + while (dptr <= limit) { + if ((strncmp(dptr, hnfo->lname, hnfo->lnlen) != 0) && + (strncmp(dptr, hnfo->sname, hnfo->snlen) != 0)) { + dptr++; + continue; + } + aux = ct_sip_search(hnfo->ln_str, dptr, hnfo->ln_strlen, + ct_sip_lnlen(dptr, limit), + hnfo->case_sensitive); + if (!aux) { + DEBUGP("'%s' not found in '%s'.\n", hnfo->ln_str, + hnfo->lname); + return -1; + } + aux += hnfo->ln_strlen; + + *matchlen = hnfo->match_len(ct, aux, limit, &shift); + if (!*matchlen) + return -1; + + *matchoff = (aux - k) + shift; + + DEBUGP("%s match succeeded! - len: %u\n", hnfo->lname, + *matchlen); + return 1; + } + DEBUGP("%s header not found.\n", hnfo->lname); + return 0; +} +EXPORT_SYMBOL_GPL(ct_sip_get_info); + +static int set_expected_rtp(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + union nf_conntrack_address *addr, + __be16 port, + const char *dptr) +{ + struct nf_conntrack_expect *exp; + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + int family = ct->tuplehash[!dir].tuple.src.l3num; + int ret; + typeof(nf_nat_sdp_hook) nf_nat_sdp; + + exp = nf_conntrack_expect_alloc(ct); + if (exp == NULL) + return NF_DROP; + nf_conntrack_expect_init(exp, family, + &ct->tuplehash[!dir].tuple.src.u3, addr, + IPPROTO_UDP, NULL, &port); + + nf_nat_sdp = rcu_dereference(nf_nat_sdp_hook); + if (nf_nat_sdp && ct->status & IPS_NAT_MASK) + ret = nf_nat_sdp(pskb, ctinfo, exp, dptr); + else { + if (nf_conntrack_expect_related(exp) != 0) + ret = NF_DROP; + else + ret = NF_ACCEPT; + } + nf_conntrack_expect_put(exp); + + return ret; +} + +static int sip_help(struct sk_buff **pskb, + unsigned int protoff, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo) +{ + int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; + union nf_conntrack_address addr; + unsigned int dataoff, datalen; + const char *dptr; + int ret = NF_ACCEPT; + int matchoff, matchlen; + u_int16_t port; + enum sip_header_pos pos; + typeof(nf_nat_sip_hook) nf_nat_sip; + + /* No Data ? */ + dataoff = protoff + sizeof(struct udphdr); + if (dataoff >= (*pskb)->len) + return NF_ACCEPT; + + nf_ct_refresh(ct, *pskb, sip_timeout * HZ); + + if (!skb_is_nonlinear(*pskb)) + dptr = (*pskb)->data + dataoff; + else { + DEBUGP("Copy of skbuff not supported yet.\n"); + goto out; + } + + nf_nat_sip = rcu_dereference(nf_nat_sip_hook); + if (nf_nat_sip && ct->status & IPS_NAT_MASK) { + if (!nf_nat_sip(pskb, ctinfo, ct, &dptr)) { + ret = NF_DROP; + goto out; + } + } + + datalen = (*pskb)->len - dataoff; + if (datalen < sizeof("SIP/2.0 200") - 1) + goto out; + + /* RTP info only in some SDP pkts */ + if (memcmp(dptr, "INVITE", sizeof("INVITE") - 1) != 0 && + memcmp(dptr, "SIP/2.0 200", sizeof("SIP/2.0 200") - 1) != 0) { + goto out; + } + /* Get address and port from SDP packet. */ + pos = family == AF_INET ? POS_CONNECTION_IP4 : POS_CONNECTION_IP6; + if (ct_sip_get_info(ct, dptr, datalen, &matchoff, &matchlen, pos) > 0) { + + /* We'll drop only if there are parse problems. */ + if (!parse_addr(ct, dptr + matchoff, NULL, &addr, + dptr + datalen)) { + ret = NF_DROP; + goto out; + } + if (ct_sip_get_info(ct, dptr, datalen, &matchoff, &matchlen, + POS_MEDIA) > 0) { + + port = simple_strtoul(dptr + matchoff, NULL, 10); + if (port < 1024) { + ret = NF_DROP; + goto out; + } + ret = set_expected_rtp(pskb, ct, ctinfo, &addr, + htons(port), dptr); + } + } +out: + return ret; +} + +static struct nf_conntrack_helper sip[MAX_PORTS][2] __read_mostly; +static char sip_names[MAX_PORTS][2][sizeof("sip-65535")] __read_mostly; + +static void nf_conntrack_sip_fini(void) +{ + int i, j; + + for (i = 0; i < ports_c; i++) { + for (j = 0; j < 2; j++) { + if (sip[i][j].me == NULL) + continue; + nf_conntrack_helper_unregister(&sip[i][j]); + } + } +} + +static int __init nf_conntrack_sip_init(void) +{ + int i, j, ret; + char *tmpname; + + if (ports_c == 0) + ports[ports_c++] = SIP_PORT; + + for (i = 0; i < ports_c; i++) { + memset(&sip[i], 0, sizeof(sip[i])); + + sip[i][0].tuple.src.l3num = AF_INET; + sip[i][1].tuple.src.l3num = AF_INET6; + for (j = 0; j < 2; j++) { + sip[i][j].tuple.dst.protonum = IPPROTO_UDP; + sip[i][j].tuple.src.u.udp.port = htons(ports[i]); + sip[i][j].mask.src.l3num = 0xFFFF; + sip[i][j].mask.src.u.udp.port = htons(0xFFFF); + sip[i][j].mask.dst.protonum = 0xFF; + sip[i][j].max_expected = 2; + sip[i][j].timeout = 3 * 60; /* 3 minutes */ + sip[i][j].me = THIS_MODULE; + sip[i][j].help = sip_help; + + tmpname = &sip_names[i][j][0]; + if (ports[i] == SIP_PORT) + sprintf(tmpname, "sip"); + else + sprintf(tmpname, "sip-%u", i); + sip[i][j].name = tmpname; + + DEBUGP("port #%u: %u\n", i, ports[i]); + + ret = nf_conntrack_helper_register(&sip[i][j]); + if (ret) { + printk("nf_ct_sip: failed to register helper " + "for pf: %u port: %u\n", + sip[i][j].tuple.src.l3num, ports[i]); + nf_conntrack_sip_fini(); + return ret; + } + } + } + return 0; +} + +module_init(nf_conntrack_sip_init); +module_exit(nf_conntrack_sip_fini); -- cgit v1.2.3-70-g09d2 From a536df35b3a58caa9015bf7887a374b20f658368 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 2 Dec 2006 22:10:18 -0800 Subject: [NETFILTER]: nf_conntrack/nf_nat: add TFTP helper port Add IPv4 and IPv6 capable nf_conntrack port of the TFTP conntrack/NAT helper. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_tftp.h | 20 ++++ net/ipv4/netfilter/Kconfig | 5 + net/ipv4/netfilter/Makefile | 1 + net/ipv4/netfilter/nf_nat_tftp.c | 52 +++++++++ net/netfilter/Kconfig | 11 ++ net/netfilter/Makefile | 1 + net/netfilter/nf_conntrack_tftp.c | 159 ++++++++++++++++++++++++++++ 7 files changed, 249 insertions(+) create mode 100644 include/linux/netfilter/nf_conntrack_tftp.h create mode 100644 net/ipv4/netfilter/nf_nat_tftp.c create mode 100644 net/netfilter/nf_conntrack_tftp.c (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_tftp.h b/include/linux/netfilter/nf_conntrack_tftp.h new file mode 100644 index 00000000000..0d79b7ae051 --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_tftp.h @@ -0,0 +1,20 @@ +#ifndef _NF_CONNTRACK_TFTP_H +#define _NF_CONNTRACK_TFTP_H + +#define TFTP_PORT 69 + +struct tftphdr { + __be16 opcode; +}; + +#define TFTP_OPCODE_READ 1 +#define TFTP_OPCODE_WRITE 2 +#define TFTP_OPCODE_DATA 3 +#define TFTP_OPCODE_ACK 4 +#define TFTP_OPCODE_ERROR 5 + +extern unsigned int (*nf_nat_tftp_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conntrack_expect *exp); + +#endif /* _NF_CONNTRACK_TFTP_H */ diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 83e83f553cc..1c0018b6bea 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -515,6 +515,11 @@ config IP_NF_NAT_TFTP default IP_NF_NAT if IP_NF_TFTP=y default m if IP_NF_TFTP=m +config NF_NAT_TFTP + tristate + depends on IP_NF_IPTABLES && NF_CONNTRACK && NF_NAT + default NF_NAT && NF_CONNTRACK_TFTP + config IP_NF_NAT_AMANDA tristate depends on IP_NF_IPTABLES!=n && IP_NF_CONNTRACK!=n && IP_NF_NAT!=n diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index 167b65e89ac..3ca0af0ed5f 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -56,6 +56,7 @@ obj-$(CONFIG_NF_NAT_H323) += nf_nat_h323.o obj-$(CONFIG_NF_NAT_IRC) += nf_nat_irc.o obj-$(CONFIG_NF_NAT_PPTP) += nf_nat_pptp.o obj-$(CONFIG_NF_NAT_SIP) += nf_nat_sip.o +obj-$(CONFIG_NF_NAT_TFTP) += nf_nat_tftp.o # NAT protocols (nf_nat) obj-$(CONFIG_NF_NAT_PROTO_GRE) += nf_nat_proto_gre.o diff --git a/net/ipv4/netfilter/nf_nat_tftp.c b/net/ipv4/netfilter/nf_nat_tftp.c new file mode 100644 index 00000000000..2566b79de22 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_tftp.c @@ -0,0 +1,52 @@ +/* (C) 2001-2002 Magnus Boden + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include + +#include +#include +#include +#include +#include + +MODULE_AUTHOR("Magnus Boden "); +MODULE_DESCRIPTION("TFTP NAT helper"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ip_nat_tftp"); + +static unsigned int help(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conntrack_expect *exp) +{ + struct nf_conn *ct = exp->master; + + exp->saved_proto.udp.port + = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u.udp.port; + exp->dir = IP_CT_DIR_REPLY; + exp->expectfn = nf_nat_follow_master; + if (nf_conntrack_expect_related(exp) != 0) + return NF_DROP; + return NF_ACCEPT; +} + +static void __exit nf_nat_tftp_fini(void) +{ + rcu_assign_pointer(nf_nat_tftp_hook, NULL); + synchronize_rcu(); +} + +static int __init nf_nat_tftp_init(void) +{ + BUG_ON(rcu_dereference(nf_nat_tftp_hook)); + rcu_assign_pointer(nf_nat_tftp_hook, help); + return 0; +} + +module_init(nf_nat_tftp_init); +module_exit(nf_nat_tftp_fini); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index e73891194b3..e0deac17df1 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -248,6 +248,17 @@ config NF_CONNTRACK_SIP To compile it as a module, choose M here. If unsure, say N. +config NF_CONNTRACK_TFTP + tristate "TFTP protocol support (EXPERIMENTAL)" + depends on EXPERIMENTAL && NF_CONNTRACK + help + TFTP connection tracking helper, this is required depending + on how restrictive your ruleset is. + If you are using a tftp client behind -j SNAT or -j MASQUERADING + you will need this. + + To compile it as a module, choose M here. If unsure, say N. + config NF_CT_NETLINK tristate 'Connection tracking netlink interface (EXPERIMENTAL)' depends on EXPERIMENTAL && NF_CONNTRACK && NETFILTER_NETLINK diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 86da21700aa..5dc5574f7e9 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -30,6 +30,7 @@ obj-$(CONFIG_NF_CONNTRACK_IRC) += nf_conntrack_irc.o obj-$(CONFIG_NF_CONNTRACK_NETBIOS_NS) += nf_conntrack_netbios_ns.o obj-$(CONFIG_NF_CONNTRACK_PPTP) += nf_conntrack_pptp.o obj-$(CONFIG_NF_CONNTRACK_SIP) += nf_conntrack_sip.o +obj-$(CONFIG_NF_CONNTRACK_TFTP) += nf_conntrack_tftp.o # generic X tables obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o diff --git a/net/netfilter/nf_conntrack_tftp.c b/net/netfilter/nf_conntrack_tftp.c new file mode 100644 index 00000000000..d46a1263fcb --- /dev/null +++ b/net/netfilter/nf_conntrack_tftp.c @@ -0,0 +1,159 @@ +/* (C) 2001-2002 Magnus Boden + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +MODULE_AUTHOR("Magnus Boden "); +MODULE_DESCRIPTION("TFTP connection tracking helper"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ip_conntrack_tftp"); + +#define MAX_PORTS 8 +static unsigned short ports[MAX_PORTS]; +static int ports_c; +module_param_array(ports, ushort, &ports_c, 0400); +MODULE_PARM_DESC(ports, "Port numbers of TFTP servers"); + +#if 0 +#define DEBUGP(format, args...) printk("%s:%s:" format, \ + __FILE__, __FUNCTION__ , ## args) +#else +#define DEBUGP(format, args...) +#endif + +unsigned int (*nf_nat_tftp_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conntrack_expect *exp) __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_tftp_hook); + +static int tftp_help(struct sk_buff **pskb, + unsigned int protoff, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo) +{ + struct tftphdr _tftph, *tfh; + struct nf_conntrack_expect *exp; + struct nf_conntrack_tuple *tuple; + unsigned int ret = NF_ACCEPT; + int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; + typeof(nf_nat_tftp_hook) nf_nat_tftp; + + tfh = skb_header_pointer(*pskb, protoff + sizeof(struct udphdr), + sizeof(_tftph), &_tftph); + if (tfh == NULL) + return NF_ACCEPT; + + switch (ntohs(tfh->opcode)) { + case TFTP_OPCODE_READ: + case TFTP_OPCODE_WRITE: + /* RRQ and WRQ works the same way */ + DEBUGP(""); + NF_CT_DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + NF_CT_DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_REPLY].tuple); + + exp = nf_conntrack_expect_alloc(ct); + if (exp == NULL) + return NF_DROP; + tuple = &ct->tuplehash[IP_CT_DIR_REPLY].tuple; + nf_conntrack_expect_init(exp, family, + &tuple->src.u3, &tuple->dst.u3, + IPPROTO_UDP, + NULL, &tuple->dst.u.udp.port); + + DEBUGP("expect: "); + NF_CT_DUMP_TUPLE(&exp->tuple); + NF_CT_DUMP_TUPLE(&exp->mask); + + nf_nat_tftp = rcu_dereference(nf_nat_tftp_hook); + if (nf_nat_tftp && ct->status & IPS_NAT_MASK) + ret = nf_nat_tftp(pskb, ctinfo, exp); + else if (nf_conntrack_expect_related(exp) != 0) + ret = NF_DROP; + nf_conntrack_expect_put(exp); + break; + case TFTP_OPCODE_DATA: + case TFTP_OPCODE_ACK: + DEBUGP("Data/ACK opcode\n"); + break; + case TFTP_OPCODE_ERROR: + DEBUGP("Error opcode\n"); + break; + default: + DEBUGP("Unknown opcode\n"); + } + return ret; +} + +static struct nf_conntrack_helper tftp[MAX_PORTS][2] __read_mostly; +static char tftp_names[MAX_PORTS][2][sizeof("tftp-65535")] __read_mostly; + +static void nf_conntrack_tftp_fini(void) +{ + int i, j; + + for (i = 0; i < ports_c; i++) { + for (j = 0; j < 2; j++) + nf_conntrack_helper_unregister(&tftp[i][j]); + } +} + +static int __init nf_conntrack_tftp_init(void) +{ + int i, j, ret; + char *tmpname; + + if (ports_c == 0) + ports[ports_c++] = TFTP_PORT; + + for (i = 0; i < ports_c; i++) { + memset(&tftp[i], 0, sizeof(tftp[i])); + + tftp[i][0].tuple.src.l3num = AF_INET; + tftp[i][1].tuple.src.l3num = AF_INET6; + for (j = 0; j < 2; j++) { + tftp[i][j].tuple.dst.protonum = IPPROTO_UDP; + tftp[i][j].tuple.src.u.udp.port = htons(ports[i]); + tftp[i][j].mask.src.l3num = 0xFFFF; + tftp[i][j].mask.dst.protonum = 0xFF; + tftp[i][j].mask.src.u.udp.port = htons(0xFFFF); + tftp[i][j].max_expected = 1; + tftp[i][j].timeout = 5 * 60; /* 5 minutes */ + tftp[i][j].me = THIS_MODULE; + tftp[i][j].help = tftp_help; + + tmpname = &tftp_names[i][j][0]; + if (ports[i] == TFTP_PORT) + sprintf(tmpname, "tftp"); + else + sprintf(tmpname, "tftp-%u", i); + tftp[i][j].name = tmpname; + + ret = nf_conntrack_helper_register(&tftp[i][j]); + if (ret) { + printk("nf_ct_tftp: failed to register helper " + "for pf: %u port: %u\n", + tftp[i][j].tuple.src.l3num, ports[i]); + nf_conntrack_tftp_fini(); + return ret; + } + } + } + return 0; +} + +module_init(nf_conntrack_tftp_init); +module_exit(nf_conntrack_tftp_fini); -- cgit v1.2.3-70-g09d2 From 2b5f6dcce5bf94b9b119e9ed8d537098ec61c3d2 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Sat, 2 Dec 2006 22:22:25 -0800 Subject: [XFRM]: Fix aevent structuring to be more complete. aevents can not uniquely identify an SA. We break the ABI with this patch, but consensus is that since it is not yet utilized by any (known) application then it is fine (better do it now than later). Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller --- Documentation/networking/xfrm_sync.txt | 5 ++++- include/linux/xfrm.h | 2 ++ net/xfrm/xfrm_user.c | 4 +++- 3 files changed, 9 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/Documentation/networking/xfrm_sync.txt b/Documentation/networking/xfrm_sync.txt index 8be626f7c0b..d7aac9dedeb 100644 --- a/Documentation/networking/xfrm_sync.txt +++ b/Documentation/networking/xfrm_sync.txt @@ -47,10 +47,13 @@ aevent_id structure looks like: struct xfrm_aevent_id { struct xfrm_usersa_id sa_id; + xfrm_address_t saddr; __u32 flags; + __u32 reqid; }; -xfrm_usersa_id in this message layout identifies the SA. +The unique SA is identified by the combination of xfrm_usersa_id, +reqid and saddr. flags are used to indicate different things. The possible flags are: diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h index 7907c42bd4e..088ba8113f7 100644 --- a/include/linux/xfrm.h +++ b/include/linux/xfrm.h @@ -289,7 +289,9 @@ struct xfrm_usersa_id { struct xfrm_aevent_id { struct xfrm_usersa_id sa_id; + xfrm_address_t saddr; __u32 flags; + __u32 reqid; }; struct xfrm_userspi_info { diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 055b72fb37b..6f97665983d 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -1281,10 +1281,12 @@ static int build_aevent(struct sk_buff *skb, struct xfrm_state *x, struct km_eve id = NLMSG_DATA(nlh); nlh->nlmsg_flags = 0; - id->sa_id.daddr = x->id.daddr; + memcpy(&id->sa_id.daddr, &x->id.daddr,sizeof(x->id.daddr)); id->sa_id.spi = x->id.spi; id->sa_id.family = x->props.family; id->sa_id.proto = x->id.proto; + memcpy(&id->saddr, &x->props.saddr,sizeof(x->props.saddr)); + id->reqid = x->props.reqid; id->flags = c->data.aevent; RTA_PUT(skb, XFRMA_REPLAY_VAL, sizeof(x->replay), &x->replay); -- cgit v1.2.3-70-g09d2 From d916faace3efc0bf19fe9a615a1ab8fa1a24cd93 Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Sun, 3 Dec 2006 22:22:41 -0500 Subject: Remove long-unmaintained ftape driver subsystem. It's bitrotten, long unmaintained, long hidden under BROKEN_ON_SMP, etc. As scheduled in feature-removal-schedule.txt, and ack'd several times on lkml. Signed-off-by: Jeff Garzik --- Documentation/00-INDEX | 2 - Documentation/feature-removal-schedule.txt | 8 - Documentation/ftape.txt | 307 ----- Documentation/kernel-parameters.txt | 3 - MAINTAINERS | 5 - drivers/char/Kconfig | 33 - drivers/char/Makefile | 1 - drivers/char/ftape/Kconfig | 330 ------ drivers/char/ftape/Makefile | 28 - drivers/char/ftape/README.PCI | 81 -- drivers/char/ftape/RELEASE-NOTES | 966 --------------- drivers/char/ftape/compressor/Makefile | 31 - drivers/char/ftape/compressor/lzrw3.c | 743 ------------ drivers/char/ftape/compressor/lzrw3.h | 253 ---- drivers/char/ftape/compressor/zftape-compress.c | 1203 ------------------- drivers/char/ftape/compressor/zftape-compress.h | 83 -- drivers/char/ftape/lowlevel/Makefile | 43 - drivers/char/ftape/lowlevel/fc-10.c | 175 --- drivers/char/ftape/lowlevel/fc-10.h | 39 - drivers/char/ftape/lowlevel/fdc-io.c | 1349 --------------------- drivers/char/ftape/lowlevel/fdc-io.h | 252 ---- drivers/char/ftape/lowlevel/fdc-isr.c | 1170 ------------------- drivers/char/ftape/lowlevel/fdc-isr.h | 55 - drivers/char/ftape/lowlevel/ftape-bsm.c | 491 -------- drivers/char/ftape/lowlevel/ftape-bsm.h | 66 -- drivers/char/ftape/lowlevel/ftape-buffer.c | 130 --- drivers/char/ftape/lowlevel/ftape-buffer.h | 32 - drivers/char/ftape/lowlevel/ftape-calibr.c | 275 ----- drivers/char/ftape/lowlevel/ftape-calibr.h | 37 - drivers/char/ftape/lowlevel/ftape-ctl.c | 896 -------------- drivers/char/ftape/lowlevel/ftape-ctl.h | 162 --- drivers/char/ftape/lowlevel/ftape-ecc.c | 853 -------------- drivers/char/ftape/lowlevel/ftape-ecc.h | 84 -- drivers/char/ftape/lowlevel/ftape-format.c | 344 ------ drivers/char/ftape/lowlevel/ftape-format.h | 37 - drivers/char/ftape/lowlevel/ftape-init.c | 160 --- drivers/char/ftape/lowlevel/ftape-init.h | 43 - drivers/char/ftape/lowlevel/ftape-io.c | 992 ---------------- drivers/char/ftape/lowlevel/ftape-io.h | 90 -- drivers/char/ftape/lowlevel/ftape-proc.c | 214 ---- drivers/char/ftape/lowlevel/ftape-proc.h | 35 - drivers/char/ftape/lowlevel/ftape-read.c | 621 ---------- drivers/char/ftape/lowlevel/ftape-read.h | 51 - drivers/char/ftape/lowlevel/ftape-rw.c | 1092 ----------------- drivers/char/ftape/lowlevel/ftape-rw.h | 111 -- drivers/char/ftape/lowlevel/ftape-setup.c | 104 -- drivers/char/ftape/lowlevel/ftape-tracing.c | 118 -- drivers/char/ftape/lowlevel/ftape-tracing.h | 179 --- drivers/char/ftape/lowlevel/ftape-write.c | 336 ------ drivers/char/ftape/lowlevel/ftape-write.h | 53 - drivers/char/ftape/lowlevel/ftape_syms.c | 87 -- drivers/char/ftape/zftape/Makefile | 36 - drivers/char/ftape/zftape/zftape-buffers.c | 149 --- drivers/char/ftape/zftape/zftape-buffers.h | 55 - drivers/char/ftape/zftape/zftape-ctl.c | 1417 ----------------------- drivers/char/ftape/zftape/zftape-ctl.h | 58 - drivers/char/ftape/zftape/zftape-eof.c | 199 ---- drivers/char/ftape/zftape/zftape-eof.h | 52 - drivers/char/ftape/zftape/zftape-init.c | 377 ------ drivers/char/ftape/zftape/zftape-init.h | 77 -- drivers/char/ftape/zftape/zftape-read.c | 377 ------ drivers/char/ftape/zftape/zftape-read.h | 53 - drivers/char/ftape/zftape/zftape-rw.c | 375 ------ drivers/char/ftape/zftape/zftape-rw.h | 101 -- drivers/char/ftape/zftape/zftape-vtbl.c | 757 ------------ drivers/char/ftape/zftape/zftape-vtbl.h | 227 ---- drivers/char/ftape/zftape/zftape-write.c | 483 -------- drivers/char/ftape/zftape/zftape-write.h | 38 - drivers/char/ftape/zftape/zftape_syms.c | 43 - include/linux/Kbuild | 4 - include/linux/ftape-header-segment.h | 122 -- include/linux/ftape-vendors.h | 137 --- include/linux/ftape.h | 201 ---- include/linux/zftape.h | 87 -- 74 files changed, 20278 deletions(-) delete mode 100644 Documentation/ftape.txt delete mode 100644 drivers/char/ftape/Kconfig delete mode 100644 drivers/char/ftape/Makefile delete mode 100644 drivers/char/ftape/README.PCI delete mode 100644 drivers/char/ftape/RELEASE-NOTES delete mode 100644 drivers/char/ftape/compressor/Makefile delete mode 100644 drivers/char/ftape/compressor/lzrw3.c delete mode 100644 drivers/char/ftape/compressor/lzrw3.h delete mode 100644 drivers/char/ftape/compressor/zftape-compress.c delete mode 100644 drivers/char/ftape/compressor/zftape-compress.h delete mode 100644 drivers/char/ftape/lowlevel/Makefile delete mode 100644 drivers/char/ftape/lowlevel/fc-10.c delete mode 100644 drivers/char/ftape/lowlevel/fc-10.h delete mode 100644 drivers/char/ftape/lowlevel/fdc-io.c delete mode 100644 drivers/char/ftape/lowlevel/fdc-io.h delete mode 100644 drivers/char/ftape/lowlevel/fdc-isr.c delete mode 100644 drivers/char/ftape/lowlevel/fdc-isr.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-bsm.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-bsm.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-buffer.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-buffer.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-calibr.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-calibr.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-ctl.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-ctl.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-ecc.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-ecc.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-format.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-format.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-init.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-init.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-io.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-io.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-proc.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-proc.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-read.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-read.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-rw.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-rw.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-setup.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-tracing.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-tracing.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-write.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-write.h delete mode 100644 drivers/char/ftape/lowlevel/ftape_syms.c delete mode 100644 drivers/char/ftape/zftape/Makefile delete mode 100644 drivers/char/ftape/zftape/zftape-buffers.c delete mode 100644 drivers/char/ftape/zftape/zftape-buffers.h delete mode 100644 drivers/char/ftape/zftape/zftape-ctl.c delete mode 100644 drivers/char/ftape/zftape/zftape-ctl.h delete mode 100644 drivers/char/ftape/zftape/zftape-eof.c delete mode 100644 drivers/char/ftape/zftape/zftape-eof.h delete mode 100644 drivers/char/ftape/zftape/zftape-init.c delete mode 100644 drivers/char/ftape/zftape/zftape-init.h delete mode 100644 drivers/char/ftape/zftape/zftape-read.c delete mode 100644 drivers/char/ftape/zftape/zftape-read.h delete mode 100644 drivers/char/ftape/zftape/zftape-rw.c delete mode 100644 drivers/char/ftape/zftape/zftape-rw.h delete mode 100644 drivers/char/ftape/zftape/zftape-vtbl.c delete mode 100644 drivers/char/ftape/zftape/zftape-vtbl.h delete mode 100644 drivers/char/ftape/zftape/zftape-write.c delete mode 100644 drivers/char/ftape/zftape/zftape-write.h delete mode 100644 drivers/char/ftape/zftape/zftape_syms.c delete mode 100644 include/linux/ftape-header-segment.h delete mode 100644 include/linux/ftape-vendors.h delete mode 100644 include/linux/ftape.h delete mode 100644 include/linux/zftape.h (limited to 'include') diff --git a/Documentation/00-INDEX b/Documentation/00-INDEX index 02457ec9c94..f08ca953573 100644 --- a/Documentation/00-INDEX +++ b/Documentation/00-INDEX @@ -104,8 +104,6 @@ firmware_class/ - request_firmware() hotplug interface info. floppy.txt - notes and driver options for the floppy disk driver. -ftape.txt - - notes about the floppy tape device driver. hayes-esp.txt - info on using the Hayes ESP serial driver. highuid.txt diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index f81819364b7..226ecf2ffd5 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -234,14 +234,6 @@ Who: Jean Delvare --------------------------- -What: ftape -When: 2.6.20 -Why: Orphaned for ages. SMP bugs long unfixed. Few users left - in the world. -Who: Jeff Garzik - ---------------------------- - What: IPv4 only connection tracking/NAT/helpers When: 2.6.22 Why: The new layer 3 independant connection tracking replaces the old diff --git a/Documentation/ftape.txt b/Documentation/ftape.txt deleted file mode 100644 index 7d8bb338403..00000000000 --- a/Documentation/ftape.txt +++ /dev/null @@ -1,307 +0,0 @@ -Intro -===== - -This file describes some issues involved when using the "ftape" -floppy tape device driver that comes with the Linux kernel. - -ftape has a home page at - -http://ftape.dot-heine.de/ - -which contains further information about ftape. Please cross check -this WWW address against the address given (if any) in the MAINTAINERS -file located in the top level directory of the Linux kernel source -tree. - -NOTE: This is an unmaintained set of drivers, and it is not guaranteed to work. -If you are interested in taking over maintenance, contact Claus-Justus Heine -, the former maintainer. - -Contents -======== - -A minus 1: Ftape documentation - -A. Changes - 1. Goal - 2. I/O Block Size - 3. Write Access when not at EOD (End Of Data) or BOT (Begin Of Tape) - 4. Formatting - 5. Interchanging cartridges with other operating systems - -B. Debugging Output - 1. Introduction - 2. Tuning the debugging output - -C. Boot and load time configuration - 1. Setting boot time parameters - 2. Module load time parameters - 3. Ftape boot- and load time options - 4. Example kernel parameter setting - 5. Example module parameter setting - -D. Support and contacts - -******************************************************************************* - -A minus 1. Ftape documentation -============================== - -Unluckily, the ftape-HOWTO is out of date. This really needs to be -changed. Up to date documentation as well as recent development -versions of ftape and useful links to related topics can be found at -the ftape home page at - -http://ftape.dot-heine.de/ - -******************************************************************************* - -A. Changes -========== - -1. Goal - ~~~~ - The goal of all that incompatibilities was to give ftape an interface - that resembles the interface provided by SCSI tape drives as close - as possible. Thus any Unix backup program that is known to work - with SCSI tape drives should also work. - - The concept of a fixed block size for read/write transfers is - rather unrelated to this SCSI tape compatibility at the file system - interface level. It developed out of a feature of zftape, a - block wise user transparent on-the-fly compression. That compression - support will not be dropped in future releases for compatibility - reasons with previous releases of zftape. - -2. I/O Block Size - ~~~~~~~~~~~~~~ - The block size defaults to 10k which is the default block size of - GNU tar. - - The block size can be tuned either during kernel configuration or - at runtime with the MTIOCTOP ioctl using the MTSETBLK operation - (i.e. do "mt -f /dev/qft0" setblk #BLKSZ). A block size of 0 - switches to variable block size mode i.e. "mt setblk 0" switches - off the block size restriction. However, this disables zftape's - built in on-the-fly compression which doesn't work with variable - block size mode. - - The BLKSZ parameter must be given as a byte count and must be a - multiple of 32k or 0, i.e. use "mt setblk 32768" to switch to a - block size of 32k. - - The typical symptom of a block size mismatch is an "invalid - argument" error message. - -3. Write Access when not at EOD (End Of Data) or BOT (Begin Of Tape) - ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - zftape (the file system interface of ftape-3.x) denies write access - to the tape cartridge when it isn't positioned either at BOT or - EOD. - -4. Formatting - ~~~~~~~~~~ - ftape DOES support formatting of floppy tape cartridges. You need the - `ftformat' program that is shipped with the modules version of ftape. - Please get the latest version of ftape from - - ftp://sunsite.unc.edu/pub/Linux/kernel/tapes - - or from the ftape home page at - - http://ftape.dot-heine.de/ - - `ftformat' is contained in the `./contrib/' subdirectory of that - separate ftape package. - -5. Interchanging cartridges with other operating systems - ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - - The internal emulation of Unix tape device file marks has changed - completely. ftape now uses the volume table segment as specified - by the QIC-40/80/3010/3020/113 standards to emulate file marks. As - a consequence there is limited support to interchange cartridges - with other operating systems. - - To be more precise: ftape will detect volumes written by other OS's - programs and other OS's programs will detect volumes written by - ftape. - - However, it isn't possible to extract the data dumped to the tape - by some MSDOS program with ftape. This exceeds the scope of a - kernel device driver. If you need such functionality, then go ahead - and write a user space utility that is able to do that. ftape already - provides all kernel level support necessary to do that. - -******************************************************************************* - -B. Debugging Output - ================ - -1. Introduction - ~~~~~~~~~~~~ - The ftape driver can be very noisy in that is can print lots of - debugging messages to the kernel log files and the system console. - While this is useful for debugging it might be annoying during - normal use and enlarges the size of the driver by several kilobytes. - - To reduce the size of the driver you can trim the maximal amount of - debugging information available during kernel configuration. Please - refer to the kernel configuration script and its on-line help - functionality. - - The amount of debugging output maps to the "tracing" boot time - option and the "ft_tracing" modules option as follows: - - 0 bugs - 1 + errors (with call-stack dump) - 2 + warnings - 3 + information - 4 + more information - 5 + program flow - 6 + fdc/dma info - 7 + data flow - 8 + everything else - -2. Tuning the debugging output - ~~~~~~~~~~~~~~~~~~~~~~~~~~~ - To reduce the amount of debugging output printed to the system - console you can - - i) trim the debugging output at run-time with - - mt -f /dev/nqft0 setdensity #DBGLVL - - where "#DBGLVL" is a number between 0 and 9 - - ii) trim the debugging output at module load time with - - modprobe ftape ft_tracing=#DBGLVL - - Of course, this applies only if you have configured ftape to be - compiled as a module. - - iii) trim the debugging output during system boot time. Add the - following to the kernel command line: - - ftape=#DBGLVL,tracing - - Please refer also to the next section if you don't know how to - set boot time parameters. - -******************************************************************************* - -C. Boot and load time configuration - ================================ - -1. Setting boot time parameters - ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - Assuming that you use lilo, the LI)nux LO)ader, boot time kernel - parameters can be set by adding a line - - append some_kernel_boot_time_parameter - - to `/etc/lilo.conf' or at real boot time by typing in the options - at the prompt provided by LILO. I can't give you advice on how to - specify those parameters with other loaders as I don't use them. - - For ftape, each "some_kernel_boot_time_parameter" looks like - "ftape=value,option". As an example, the debugging output can be - increased with - - ftape=4,tracing - - NOTE: the value precedes the option name. - -2. Module load time parameters - ~~~~~~~~~~~~~~~~~~~~~~~~~~~ - Module parameters can be specified either directly when invoking - the program 'modprobe' at the shell prompt: - - modprobe ftape ft_tracing=4 - - or by editing the file `/etc/modprobe.conf' in which case they take - effect each time when the module is loaded with `modprobe' (please - refer to the respective manual pages). Thus, you should add a line - - options ftape ft_tracing=4 - - to `/etc/modprobe.conf` if you intend to increase the debugging - output of the driver. - - -3. Ftape boot- and load time options - ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - - i. Controlling the amount of debugging output - DBGLVL has to be replaced by a number between 0 and 8. - - module | kernel command line - -----------------------|---------------------- - ft_tracing=DBGLVL | ftape=DBGLVL,tracing - - ii. Hardware setup - BASE is the base address of your floppy disk controller, - IRQ and DMA give its interrupt and DMA channel, respectively. - BOOL is an integer, "0" means "no"; any other value means - "yes". You don't need to specify anything if connecting your tape - drive to the standard floppy disk controller. All of these - values have reasonable defaults. The defaults can be modified - during kernel configuration, i.e. while running "make config", - "make menuconfig" or "make xconfig" in the top level directory - of the Linux kernel source tree. Please refer also to the on - line documentation provided during that kernel configuration - process. - - ft_probe_fc10 is set to a non-zero value if you wish for ftape to - probe for a Colorado FC-10 or FC-20 controller. - - ft_mach2 is set to a non-zero value if you wish for ftape to probe - for a Mountain MACH-2 controller. - - module | kernel command line - -----------------------|---------------------- - ft_fdc_base=BASE | ftape=BASE,ioport - ft_fdc_irq=IRQ | ftape=IRQ,irq - ft_fdc_dma=DMA | ftape=DMA,dma - ft_probe_fc10=BOOL | ftape=BOOL,fc10 - ft_mach2=BOOL | ftape=BOOL,mach2 - ft_fdc_threshold=THR | ftape=THR,threshold - ft_fdc_rate_limit=RATE | ftape=RATE,datarate - -4. Example kernel parameter setting - ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - To configure ftape to probe for a Colorado FC-10/FC-20 controller - and to increase the amount of debugging output a little bit, add - the following line to `/etc/lilo.conf': - - append ftape=1,fc10 ftape=4,tracing - -5. Example module parameter setting - ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - To do the same, but with ftape compiled as a loadable kernel - module, add the following line to `/etc/modprobe.conf': - - options ftape ft_probe_fc10=1 ft_tracing=4 - -******************************************************************************* - -D. Support and contacts - ==================== - - Ftape is distributed under the GNU General Public License. There is - absolutely no warranty for this software. However, you can reach - the current maintainer of the ftape package under the email address - given in the MAINTAINERS file which is located in the top level - directory of the Linux kernel source tree. There you'll find also - the relevant mailing list to use as a discussion forum and the web - page to query for the most recent documentation, related work and - development versions of ftape. - - Changelog: - ========== - -~1996: Original Document - -10-24-2004: General cleanup and updating, noting additional module options. - James Nelson diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 67473849f20..15e4fed127f 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -557,9 +557,6 @@ and is between 256 and 4096 characters. It is defined in the file floppy= [HW] See Documentation/floppy.txt. - ftape= [HW] Floppy Tape subsystem debugging options. - See Documentation/ftape.txt. - gamecon.map[2|3]= [HW,JOY] Multisystem joystick and NES/SNES/PSX pad support via parallel port (up to 5 devices per port) diff --git a/MAINTAINERS b/MAINTAINERS index 45df5d4e2ab..8385a69138a 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1166,11 +1166,6 @@ P: David Howells M: dhowells@redhat.com S: Maintained -FTAPE/QIC-117 -L: linux-tape@vger.kernel.org -W: http://sourceforge.net/projects/ftape -S: Orphan - FUSE: FILESYSTEM IN USERSPACE P: Miklos Szeredi M: miklos@szeredi.hu diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig index ad8b537ad47..24f922f1278 100644 --- a/drivers/char/Kconfig +++ b/drivers/char/Kconfig @@ -855,39 +855,6 @@ config TANBAC_TB0219 depends TANBAC_TB022X select GPIO_VR41XX -menu "Ftape, the floppy tape device driver" - -config FTAPE - tristate "Ftape (QIC-80/Travan) support" - depends on BROKEN_ON_SMP && (ALPHA || X86) - ---help--- - If you have a tape drive that is connected to your floppy - controller, say Y here. - - Some tape drives (like the Seagate "Tape Store 3200" or the Iomega - "Ditto 3200" or the Exabyte "Eagle TR-3") come with a "high speed" - controller of their own. These drives (and their companion - controllers) are also supported if you say Y here. - - If you have a special controller (such as the CMS FC-10, FC-20, - Mountain Mach-II, or any controller that is based on the Intel 82078 - FDC like the high speed controllers by Seagate and Exabyte and - Iomega's "Ditto Dash") you must configure it by selecting the - appropriate entries from the "Floppy tape controllers" sub-menu - below and possibly modify the default values for the IRQ and DMA - channel and the IO base in ftape's configuration menu. - - If you want to use your floppy tape drive on a PCI-bus based system, - please read the file . - - The ftape kernel driver is also available as a runtime loadable - module. To compile this driver as a module, choose M here: the - module will be called ftape. - -source "drivers/char/ftape/Kconfig" - -endmenu - source "drivers/char/agp/Kconfig" source "drivers/char/drm/Kconfig" diff --git a/drivers/char/Makefile b/drivers/char/Makefile index 777cad04509..b1fcdab9094 100644 --- a/drivers/char/Makefile +++ b/drivers/char/Makefile @@ -78,7 +78,6 @@ obj-$(CONFIG_TOSHIBA) += toshiba.o obj-$(CONFIG_I8K) += i8k.o obj-$(CONFIG_DS1620) += ds1620.o obj-$(CONFIG_HW_RANDOM) += hw_random/ -obj-$(CONFIG_FTAPE) += ftape/ obj-$(CONFIG_COBALT_LCD) += lcd.o obj-$(CONFIG_PPDEV) += ppdev.o obj-$(CONFIG_NWBUTTON) += nwbutton.o diff --git a/drivers/char/ftape/Kconfig b/drivers/char/ftape/Kconfig deleted file mode 100644 index 0d65189a7ae..00000000000 --- a/drivers/char/ftape/Kconfig +++ /dev/null @@ -1,330 +0,0 @@ -# -# Ftape configuration -# -config ZFTAPE - tristate "Zftape, the VFS interface" - depends on FTAPE - ---help--- - Normally, you want to say Y or M. DON'T say N here or you - WON'T BE ABLE TO USE YOUR FLOPPY TAPE DRIVE. - - The ftape module itself no longer contains the routines necessary - to interface with the kernel VFS layer (i.e. to actually write data - to and read data from the tape drive). Instead the file system - interface (i.e. the hardware independent part of the driver) has - been moved to a separate module. - - To compile this driver as a module, choose M here: the - module will be called zftape. - - Regardless of whether you say Y or M here, an additional runtime - loadable module called `zft-compressor' which contains code to - support user transparent on-the-fly compression based on Ross - William's lzrw3 algorithm will be produced. If you have enabled the - kernel module loader (i.e. have said Y to "Kernel module loader - support", above) then `zft-compressor' will be loaded - automatically by zftape when needed. - - Despite its name, zftape does NOT use compression by default. - -config ZFT_DFLT_BLK_SZ - int "Default block size" - depends on ZFTAPE - default "10240" - ---help--- - If unsure leave this at its default value, i.e. 10240. Note that - you specify only the default block size here. The block size can be - changed at run time using the MTSETBLK tape operation with the - MTIOCTOP ioctl (i.e. with "mt -f /dev/qft0 setblk #BLKSZ" from the - shell command line). - - The probably most striking difference between zftape and previous - versions of ftape is the fact that all data must be written or read - in multiples of a fixed block size. The block size defaults to - 10240 which is what GNU tar uses. The values for the block size - should be either 1 or multiples of 1024 up to a maximum value of - 63488 (i.e. 62 K). If you specify `1' then zftape's builtin - compression will be disabled. - - Reasonable values are `10240' (GNU tar's default block size), - `5120' (afio's default block size), `32768' (default block size some - backup programs assume for SCSI tape drives) or `1' (no restriction - on block size, but disables builtin compression). - -comment "The compressor will be built as a module only!" - depends on FTAPE && ZFTAPE - -config ZFT_COMPRESSOR - tristate - depends on FTAPE!=n && ZFTAPE!=n - default m - -config FT_NR_BUFFERS - int "Number of ftape buffers (EXPERIMENTAL)" - depends on FTAPE && EXPERIMENTAL - default "3" - help - Please leave this at `3' unless you REALLY know what you are doing. - It is not necessary to change this value. Values below 3 make the - proper use of ftape impossible, values greater than 3 are a waste of - memory. You can change the amount of DMA memory used by ftape at - runtime with "mt -f /dev/qft0 setdrvbuffer #NUMBUFFERS". Each buffer - wastes 32 KB of memory. Please note that this memory cannot be - swapped out. - -config FT_PROC_FS - bool "Enable procfs status report (+2kb)" - depends on FTAPE && PROC_FS - ---help--- - Optional. Saying Y will result in creation of a directory - `/proc/ftape' under the /proc file system. The files can be viewed - with your favorite pager (i.e. use "more /proc/ftape/history" or - "less /proc/ftape/history" or simply "cat /proc/ftape/history"). The - file will contain some status information about the inserted - cartridge, the kernel driver, your tape drive, the floppy disk - controller and the error history for the most recent use of the - kernel driver. Saying Y will enlarge the size of the ftape driver - by approximately 2 KB. - - WARNING: When compiling ftape as a module (i.e. saying M to "Floppy - tape drive") it is dangerous to use ftape's /proc file system - interface. Accessing `/proc/ftape' while the module is unloaded will - result in a kernel Oops. This cannot be fixed from inside ftape. - -choice - prompt "Debugging output" - depends on FTAPE - default FT_NORMAL_DEBUG - -config FT_NORMAL_DEBUG - bool "Normal" - ---help--- - This option controls the amount of debugging output the ftape driver - is ABLE to produce; it does not increase or diminish the debugging - level itself. If unsure, leave this at its default setting, - i.e. choose "Normal". - - Ftape can print lots of debugging messages to the system console - resp. kernel log files. Reducing the amount of possible debugging - output reduces the size of the kernel module by some KB, so it might - be a good idea to use "None" for emergency boot floppies. - - If you want to save memory then the following strategy is - recommended: leave this option at its default setting "Normal" until - you know that the driver works as expected, afterwards reconfigure - the kernel, this time specifying "Reduced" or "None" and recompile - and install the kernel as usual. Note that choosing "Excessive" - debugging output does not increase the amount of debugging output - printed to the console but only makes it possible to produce - "Excessive" debugging output. - - Please read for a short description - how to control the amount of debugging output. - -config FT_FULL_DEBUG - bool "Excessive" - help - Extremely verbose output for driver debugging purposes. - -config FT_NO_TRACE - bool "Reduced" - help - Reduced tape driver debugging output. - -config FT_NO_TRACE_AT_ALL - bool "None" - help - Suppress all debugging output from the tape drive. - -endchoice - -comment "Hardware configuration" - depends on FTAPE - -choice - prompt "Floppy tape controllers" - depends on FTAPE - default FT_STD_FDC - -config FT_STD_FDC - bool "Standard" - ---help--- - Only change this setting if you have a special controller. If you - didn't plug any add-on card into your computer system but just - plugged the floppy tape cable into the already existing floppy drive - controller then you don't want to change the default setting, - i.e. choose "Standard". - - Choose "MACH-2" if you have a Mountain Mach-2 controller. - Choose "FC-10/FC-20" if you have a Colorado FC-10 or FC-20 - controller. - Choose "Alt/82078" if you have another controller that is located at - an IO base address different from the standard floppy drive - controller's base address of `0x3f0', or uses an IRQ (interrupt) - channel different from `6', or a DMA channel different from - `2'. This is necessary for any controller card that is based on - Intel's 82078 FDC such as Seagate's, Exabyte's and Iomega's "high - speed" controllers. - - If you choose something other than "Standard" then please make - sure that the settings for the IO base address and the IRQ and DMA - channel in the configuration menus below are correct. Use the manual - of your tape drive to determine the correct settings! - - If you are already successfully using your tape drive with another - operating system then you definitely should use the same settings - for the IO base, the IRQ and DMA channel that have proven to work - with that other OS. - - Note that this menu lets you specify only the default setting for - the hardware setup. The hardware configuration can be changed at - boot time (when ftape is compiled into the kernel, i.e. if you - have said Y to "Floppy tape drive") or module load time (i.e. if you - have said M to "Floppy tape drive"). - - Please read also the file which - contains a short description of the parameters that can be set at - boot or load time. If you want to use your floppy tape drive on a - PCI-bus based system, please read the file - . - -config FT_MACH2 - bool "MACH-2" - -config FT_PROBE_FC10 - bool "FC-10/FC-20" - -config FT_ALT_FDC - bool "Alt/82078" - -endchoice - -comment "Consult the manuals of your tape drive for the correct settings!" - depends on FTAPE && !FT_STD_FDC - -config FT_FDC_BASE - hex "IO base of the floppy disk controller" - depends on FTAPE && !FT_STD_FDC - default "0" - ---help--- - You don't need to specify a value if the following default - settings for the base IO address are correct: - <<< MACH-2 : 0x1E0 >>> - <<< FC-10/FC-20: 0x180 >>> - <<< Secondary : 0x370 >>> - Secondary refers to a secondary FDC controller like the "high speed" - controllers delivered by Seagate or Exabyte or Iomega's Ditto Dash. - Please make sure that the setting for the IO base address - specified here is correct. USE THE MANUAL OF YOUR TAPE DRIVE OR - CONTROLLER CARD TO DETERMINE THE CORRECT SETTING. If you are already - successfully using the tape drive with another operating system then - you definitely should use the same settings for the IO base that has - proven to work with that other OS. - - Note that this menu lets you specify only the default setting for - the IO base. The hardware configuration can be changed at boot time - (when ftape is compiled into the kernel, i.e. if you specified Y to - "Floppy tape drive") or module load time (i.e. if you have said M to - "Floppy tape drive"). - - Please read also the file which - contains a short description of the parameters that can be set at - boot or load time. - -config FT_FDC_IRQ - int "IRQ channel of the floppy disk controller" - depends on FTAPE && !FT_STD_FDC - default "0" - ---help--- - You don't need to specify a value if the following default - settings for the interrupt channel are correct: - <<< MACH-2 : 6 >>> - <<< FC-10/FC-20: 9 >>> - <<< Secondary : 6 >>> - Secondary refers to secondary a FDC controller like the "high speed" - controllers delivered by Seagate or Exabyte or Iomega's Ditto Dash. - Please make sure that the setting for the IO base address - specified here is correct. USE THE MANUAL OF YOUR TAPE DRIVE OR - CONTROLLER CARD TO DETERMINE THE CORRECT SETTING. If you are already - successfully using the tape drive with another operating system then - you definitely should use the same settings for the IO base that has - proven to work with that other OS. - - Note that this menu lets you specify only the default setting for - the IRQ channel. The hardware configuration can be changed at boot - time (when ftape is compiled into the kernel, i.e. if you said Y to - "Floppy tape drive") or module load time (i.e. if you said M to - "Floppy tape drive"). - - Please read also the file which - contains a short description of the parameters that can be set at - boot or load time. - -config FT_FDC_DMA - int "DMA channel of the floppy disk controller" - depends on FTAPE && !FT_STD_FDC - default "0" - ---help--- - You don't need to specify a value if the following default - settings for the DMA channel are correct: - <<< MACH-2 : 2 >>> - <<< FC-10/FC-20: 3 >>> - <<< Secondary : 2 >>> - Secondary refers to a secondary FDC controller like the "high speed" - controllers delivered by Seagate or Exabyte or Iomega's Ditto Dash. - Please make sure that the setting for the IO base address - specified here is correct. USE THE MANUAL OF YOUR TAPE DRIVE OR - CONTROLLER CARD TO DETERMINE THE CORRECT SETTING. If you are already - successfully using the tape drive with another operating system then - you definitely should use the same settings for the IO base that has - proven to work with that other OS. - - Note that this menu lets you specify only the default setting for - the DMA channel. The hardware configuration can be changed at boot - time (when ftape is compiled into the kernel, i.e. if you said Y to - "Floppy tape drive") or module load time (i.e. if you said M to - "Floppy tape drive"). - - Please read also the file which - contains a short description of the parameters that can be set at - boot or load time. - -config FT_FDC_THR - int "Default FIFO threshold (EXPERIMENTAL)" - depends on FTAPE && EXPERIMENTAL - default "8" - help - Set the FIFO threshold of the FDC. If this is higher the DMA - controller may serve the FDC after a higher latency time. If this is - lower, fewer DMA transfers occur leading to less bus contention. - You may try to tune this if ftape annoys you with "reduced data - rate because of excessive overrun errors" messages. However, this - doesn't seem to have too much effect. - - If unsure, don't touch the initial value, i.e. leave it at "8". - -config FT_FDC_MAX_RATE - int "Maximal data rate to use (EXPERIMENTAL)" - depends on FTAPE && EXPERIMENTAL - default "2000" - ---help--- - With some motherboard/FDC combinations ftape will not be able to - run your FDC/tape drive combination at the highest available - speed. If this is the case you'll encounter "reduced data rate - because of excessive overrun errors" messages and lots of retries - before ftape finally decides to reduce the data rate. - - In this case it might be desirable to tell ftape beforehand that - it need not try to run the tape drive at the highest available - speed. If unsure, leave this disabled, i.e. leave it at 2000 - bits/sec. - -config FT_ALPHA_CLOCK - int "CPU clock frequency of your DEC Alpha" if ALPHA - depends on FTAPE - default "0" - help - On some DEC Alpha machines the CPU clock frequency cannot be - determined automatically, so you need to specify it here ONLY if - running a DEC Alpha, otherwise this setting has no effect. - diff --git a/drivers/char/ftape/Makefile b/drivers/char/ftape/Makefile deleted file mode 100644 index 0e67d2f8b7e..00000000000 --- a/drivers/char/ftape/Makefile +++ /dev/null @@ -1,28 +0,0 @@ -# -# Copyright (C) 1997 Claus Heine. -# -# This program is free software; you can redistribute it and/or modify -# it under the terms of the GNU General Public License as published by -# the Free Software Foundation; either version 2, or (at your option) -# any later version. -# -# This program is distributed in the hope that it will be useful, -# but WITHOUT ANY WARRANTY; without even the implied warranty of -# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -# GNU General Public License for more details. -# -# You should have received a copy of the GNU General Public License -# along with this program; see the file COPYING. If not, write to -# the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. -# -# $Source: /homes/cvs/ftape-stacked/ftape/Makefile,v $ -# $Revision: 1.4 $ -# $Date: 1997/10/05 19:17:56 $ -# -# Makefile for the QIC-40/80/3010/3020 floppy-tape driver for -# Linux. -# - -obj-$(CONFIG_FTAPE) += lowlevel/ -obj-$(CONFIG_ZFTAPE) += zftape/ -obj-$(CONFIG_ZFT_COMPRESSOR) += compressor/ diff --git a/drivers/char/ftape/README.PCI b/drivers/char/ftape/README.PCI deleted file mode 100644 index 18de159d36e..00000000000 --- a/drivers/char/ftape/README.PCI +++ /dev/null @@ -1,81 +0,0 @@ -Some notes for ftape users with PCI motherboards: -================================================= - -The problem: ------------- - -There have been some problem reports from people using PCI-bus based -systems getting overrun errors. -I wasn't able to reproduce these until I ran ftape on a Intel Plato -(Premiere PCI II) motherboard with bios version 1.00.08AX1. -It turned out that if GAT (Guaranteed Access Timing) is enabled (?) -ftape gets a lot of overrun errors. -The problem disappears when disabling GAT in the bios. -Note that Intel removed this setting (permanently disabled) from the -1.00.10AX1 bios ! - -It looks like that if GAT is enabled there are often large periods -(greater than 120 us !??) on the ISA bus that the DMA controller cannot -service the floppy disk controller. -I cannot imagine this being acceptable in a decent PCI implementation. -Maybe this is a `feature' of the chipset. I can only speculate why -Intel choose to remove the option from the latest Bios... - -The lesson of this all is that there may be other motherboard -implementations having the same of similar problems. -If you experience a lot of overrun errors during a backup to tape, -see if there is some setting in the Bios that may influence the -bus timing. - -I judge this a hardware problem and not a limitation of ftape ;-) -My DOS backup software seems to be suffering from the same problems -and even refuses to run at 1 Mbps ! -Ftape will reduce the data-rate from 1 Mbps to 500 Kbps if the number -of overrun errors on a track exceeds a threshold. - - -Possible solutions: -------------------- - -Some of the problems were solved by upgrading the (flash) bios. -Other suggest that it has to do with the FDC being on the PCI -bus, but that is not the case with the Intel Premiere II boards. -[If upgrading the bios doesn't solve the problem you could try -a floppy disk controller on the isa-bus]. - -Here is a list of systems and recommended BIOS settings: - - - Intel Premiere PCI (Revenge): - -Bios version 1.00.09.AF2 is reported to work. - - - - Intel Premiere PCI II (Plato): - -Bios version 1.00.10.AX1 and version 11 beta are ok. -If using version 1.00.08.AX1, GAT must be disabled ! - - - - ASUS PCI/I-SP3G: - -Preferred settings: ISA-GAT-mode : disabled - DMA-linebuffer-mode : standard - ISA-masterbuffer-mode : standard - - - DELL Dimension XPS P90 - -Bios version A2 is reported to be broken, while bios version A5 works. -You can get a flash bios upgrade from http://www.dell.com - - -To see if you're having the GAT problem, try making a backup -under DOS. If it's very slow and often repositions you're -probably having this problem. - - --//-- - LocalWords: ftape PCI bios GAT ISA DMA chipset Mbps Kbps FDC isa AF ok ASUS - LocalWords: SP linebuffer masterbuffer XPS http www com diff --git a/drivers/char/ftape/RELEASE-NOTES b/drivers/char/ftape/RELEASE-NOTES deleted file mode 100644 index 03799dbc05a..00000000000 --- a/drivers/char/ftape/RELEASE-NOTES +++ /dev/null @@ -1,966 +0,0 @@ -Hey, Emacs, we're -*-Text-*- mode! - -===== Release notes for ftape-3.04d 25/11/97 ===== -- The correct pre-processor statement for "else if" is "#elif" not - "elsif". -- Need to call zft_reset_position() when overwriting cartridges - previously written with ftape-2.x, sftape, or ancient - (pre-ftape-3.x) versions of zftape. - -===== Release notes for ftape-3.04c 16/11/97 ===== -- fdc_probe() was calling DUMPREGS with a result length of "1" which - was just fine. Undo previous change. - -===== Release notes for ftape-3.04b 14/11/97 ===== - -- patches/2.x.x/floppy.c.diff was somewhat broken, releasing i/o - regions it never had allocated. -- fdc_probe() was calling DUMPREGS with a result length of "1" instead - of "10" -- Writing deleted data marks if the first segents on track zero are - should work now. -- ftformat should now be able to handle those cases where the tape - drive sets the read only status bit (QIC-40/80 cartridges with - QIC-3010/3020 tape drives) because the header segment is damaged. -- the MTIOCFTCMD ioctl may now be issued by the superuser ONLY. - -===== Release notes for ftape-3.04a 12/11/97 ===== -- Fix an "infinite loop can't be killed by signal" bug in - ftape_get_drive_status(). Only relevant when trying to access - buggy/misconfigured hardware -- Try to compensate a bug in the HP Colorado T3000's firmware: it - doesn't set the write protect bit for QIC80/QIC40 cartridges. - -===== Release notes for ftape-3.04 06/11/97 ===== -- If positioning with fast seeking fails fall back to a slow seek - before giving up. -- (nearly) no retries on "no data errors" when verifying after - formatting. Improved tuning of the bad sector map after formatting. -- the directory layout has changed again to allow for easier kernel - integration -- Module parameter "ftape_tracing" now is called "ft_tracing" because - the "ftape_tracing" variable has the version checksum attached to it. -- `/proc/ftape' interface for 2.0.* kernels. `/proc/ftape' no longer - is a directory but a file that contains all the information formerly - provided in separate files under the `/proc/ftape/' directory. -- Most of the configuration options have been prefixed by "CONFIG_FT_" - in preparation of the kernel inclusion. The Makefiles under - "./ftape/" should be directly usable by the kernel. -- The MODVERSIONS stuff is now auto-detected. -- Broke backslashed multi line options in MCONFIG into separate lines - using GNU-make's "+=" feature. -- The html and dvi version of the manual is now installed under - '/usr/doc/ftape` with 'make install` -- New SMP define in MCONFIG. ftape works with SMP if this is defined. -- attempt to cope with "excessive overrun errors" by gradually - increasing FDC FIFO threshold. But this doesn't seem to have too - much an effect. -- New load time configuration parameter "ft_fdc_rate_limit". If you - encounter too many overrun errors with a 2Mb controller then you - might want to set this to 1000. -- overrun errors on the last sector in a segment sometimes result in - a zero DMA residue. Dunno why, but compensate for it. -- there were still fdc_read() timeout errors. I think I have fixed it - now, please FIXME. -- Sometimes ftape_write() failed to re-start the tape drive when a - segment without a good sector was reached ("wait for empty segment - failed"). This is fixed. Especially important for > QIC-3010. -- sftape (aka ftape-2.x) has vanished. I didn't work on it for - ages. It is probably still possible to use the old code with - ftape-3.04, if one really needs it (BUT RECOMPILE IT) -- zftape no longer alters the contents of already existing volume - table entries, which makes it possible to fill in missing fields, - like time stamps using some user space program. -- ./contrib/vtblc/ contains such a program. -- new perl script ./contrib/scripts/listtape that list the contents of a - floppy tape cartridge parsing the output of "mt volinfo" + "mt fsf" -- the MTWEOF implementation has changed a little bit (after I had a - look at amanda). Calling MTWEOF while the tape is still held open - after writing something to the tape now will terminate the current - volume, and start a new one at the current position. -- the volume table maintained by zftape now is a doubly linked list - that grows dynamically as needed. - - formatting floppy tape cartridges - --------------------------------- - * there is a new user space formatting program that does most of the - dirty work in user space (auto-detect, computing the sector - coordinates, adjusting time stamps and statistics). It has a - simple command line interface. - * ftape-format.o has vanished, it has been folded into the low level - ftape.o module, and the ioctl interface into zftape.o. Most of the - complicated stuff has been moved to user space, so there was no - need for a separate module anymore. - * there is a new ioctl MTIOCFTCMD that sends a bare QIC-117 command - to the tape drive. - * there is a new mmap() feature to map the dma buffers into user - space to be used by the user level formatting program. - * Formatting of yet unformatted or totally degaussed cartridges - should be possible now. FIXME. - -===== Release notes for ftape-3.03b, ==== - -ftape-3.03b was released as a beta release only. Its main new feature -was support of the DITTO-2GB drive. This was made possible by reverse -engineering done by after Iomega failed to support -ftape. Although they had promised to do so (this makes me feel a bit -sad and uncomfortable about Iomega). - -===== Release notes for ftape-3.03a, 22/05/97 ==== - -- Finally fixed auto-un-loading of modules for kernels > 2.1.18 -- Add an "uninstall" target to the Makefile -- removed the kdtime hack -- texi2www didn't properly set the back-reference from a footnote back - to the regular text. - - zftape specific - --------------- - * hide the old compression map volume. Taper doesn't accept the - presence of non-Taper volumes and Taper-written volume on the same - tape. - * EOD (End Of Data) handling was still broken: the expected behavior - is to return a zero byte count at the first attempt to read past - EOD, return a zero byte count at the second attempt to read past - EOD and THEN return -EIO. - - ftape-format specific - --------------------- - * Detection of QIC-40 cartridges in select_tape_format() was broken - and made it impossible to format QIC-3010/3020 cartridges. - * There are strange "TR-1 Extra" cartridges out there which weren't - detected properly because the don't strictly conform to the - QIC-80, Rev. N, spec. - -===== Release notes for ftape-3.03, 30/04/97 ===== - -- Removed kernel integration code from the package. I plan to provide - a package that can be integrated into the stock kernel separately - (hopefully soon). - As a result, a simple `make' command now will build everything. -- ALL compile time configuration options have been moved to the file - `MCONFIG'. -- Quite a few `low level' changes to allow formatting of cartridges. -- formatting is implemented as a separate module `ftape-format.o'. The - modified `mt' program contains sample code that shows how to use it. -- The VFS interface has been moved from the `ftape.o' module to the - high level modules `zftape.o' resp. `sftape.o'. `ftape.o' contains - the hardware support only. -- A bit of /proc support for kernels > 2.1.28 -- Moved documentation to Doc subdir. INSTALL now contains some real - installation notes. -- `install' target in Makefile. - -zftape specific: ----------------- - -- zftape works for large cartridges now ( > 2^31 bytes) -- MTIOCVOLINFO and MTIOCGETSIZE now return the size in KILOBYTES, - NO LONGER in bytes. - -- permissions for write access to a cartridge have changed: - * zftape now also takes the file access mode into account - * zftape no longer allows writing in the middle of the recorded - media. The tape has to be positioned at BOT or EOD for write - access. - -- MTBSF has changed. It used to position at the beginning of the - previous file when called with count 1. This was different from the - expected behavior for other Un*x tape drivers (i.e. SCSI). MTBSF - with count 1 should merely position at the beginning of the current - volume. Fixed. As a result, `tar --verify' now produces the desired - result: it verifies the last written volume, not the pre-last - written volume. - -- The compression map has vanished --> no need for `mt erase' any - more. Fast seeking in a compressed volume is still be possible, but - takes slightly longer. As a side effect, you may experience an - additional volume showing up in front of all others for old - cartridges. This is the tape volume that holds the compression map. - -- The compression support for zftape has been moved to a separate - module `zft-compressor'. DON'T forget to load it before trying to - read back compressed volumes. The stock `zftape.o' module probes for - the module `zft-compressor' using the kerneld message channel; you - have to install `zft-compressor.o' in a place where modprobe can - find it if you want to use this. - -- New experimental feature that tries to get the broken down GMT time - from user space via a kernel daemon message channel. You need to - compile and start the `kdtime' daemon contained in the contrib - directory to use it. Needed (?) for time stamps in the header - segments and the volume table. - -- variable block size mode via MTSETBLK 0 - -- keep modules locked in memory after the block size has been changed - -sftape specific: ----------------- - -- end of tape handling should be fixed, i.e. multi volume archives - written with `afio' can be read back now. - - -===== Release notes for ftape-3.02a, 09/01/97 ===== - -No big news: -- call zft_init() resp. sft_init() when compiling the entire stuff - into the kernel image. -- fix bug in ftape-setup.c when NO_TRACE_AT_ALL was defined. -- fix bug in sftape-eof.c/zftape-eof.c for old kernels (1.2.*) -- add support for new module interface for recent kernels - -===== Release notes for ftape-3.02, 16/12/96 ===== -- Fixed the `FDC unlock command failed' bug in fdc-io.c. When the FIFO - was already locked when ftape was loaded, ftape failed to unlock it. -- Fixed compilation of `contrib/gnumt'. It now finds `mtio.h' even if - ftape is NOT included into the kernel source tree. -- fc-10.c: include for inb() and outb(). -- ftape/sftape/zftape: all global variable now have either a `ftape_', - a `ft_', `sft_', `zft_' or `qic_' prefix to prevent name clashes - with other parts of the kernel when including ftape into the kernel - source tree. -- Kerneld support has changed. `ftape' now searches for a module - `ftape-frontend' when none of the frontend (`sftape' or `zftape') is - loaded. Please refer to the `Installation/Loading ftape' section of - the TeXinfo manual. -- Add load resp. boot-time configuration of ftape. There are now - variables ft_fdc_base, ft_fdc_dma and ft_fdc_irq corresponding to - the former FDC_BASE etc. compile time definitions. One can also use - the kernel command line parameters to configure the driver if it is - compiled into the kernel. Also, the FC-10/FC-20 support is load-time - configurable now as well as the MACH-II hack (ft_probe_fc10, - resp. ft_mach2). Please refer to the section `Installation/Configure - ftape' of the TeXinfo manual. -- I removed the MODVERSIONS option from `Makefile.module'. Let me alone - with ftape and MODVERSIONS unless you include the ftape sources into - the kernel source tree. -- new vendors in `vendors.h': - * HP Colorado T3000 - * ComByte DoublePlay (including a bug fix for their broken - formatting software, thanks to whraven@njackn.com) - * Iomega DITTO 2GIG. NOTE: this drive cannot work with ftape because - the logical data layout of the cartridges used by this drive does - NOT conform to the QIC standards, it is a special Iomega specific - format. I've sent mail to Iomega but didn't receive an answer - yet. If you want this drive to be supported by ftape, ask Iomega - to give me information about it. -- zftape: - * re-introduced the MTIOC_ZFTAPE_GETBLKSZ ioctl for compatibility - with zftape 1.06a and earlier. Please don't use it when writing - new software, use the MTIOCVOLINFO ioctl instead. - * Major overhaul of the code that updates the header segments. Never - change the tape label unless erasing the tape. Thus we almost - never need to write the header segments, unless we would modify - the bad sector map which isn't done yet. Updating of volume table - and compression map more secure now although it takes a bit - longer. - * Fixed bug when aborting a write operation with a signal: zftape - now finishes the current volume (i.e. writes an eof marker) at the - current position. It didn't before which led to somehow *strange* - behavior in this cases. - * Keep module locked in memory when using it with the non-rewinding - devices and the tape is not logical at BOT. Needed for kerneld - support. -- sftape: - * Keep module locked in memory when using it with the non-rewinding - devices and the tape is not logical at BOT. Needed for kerneld - support. - -===== Release notes for ftape-3.01, 14/11/96 ===== - -- Fixed silly bugs in ftape-3.00: - * MAKEDEV.ftape: major device number must be 27, not 23 - * sftape/sftape-read.c: sftape_read_header_segments() called - itself recursively instead of calling ftape_read_header_segment() - * zftape/qic-vtbl.h: conversion of ftape's file marks to zftape's - internal volume table was broken. - * patches/2.x.x/linux-2.0.21.dif: my RCS (resp. CVS) system replaced - the `$Revison:' etc. macros in the `ftape.h' concerning part of the - patch :-( Fixed. - * info/ftape.info: Fixed misspellings (`cp' <-> `cp -r' etc.) - * when ftape/sftape or ftape/zftape was compiled into the kernel the - variable ftape_status was declared twice. Fixed. - * removed reference to undeclared variable kernel_version when not - compiling as module - * fixed a bug introduced by the use of bit-fields for some flags - (i.e. write_protected, no_cartridge, formatted) - * flag `header_read' is now reset correctly to zero when tape is - removed. -- fixed a bug in sftape/sftape-eof.c that was already in the original - ftape code. MTFSF/BSF was not handled correctly when positioned - right before the file mark (think of tar) -- Changed TRACE macros (following a suggestion of Marcin Dalecki) to use - the predefined __FUNCTION__ macro of GCC. Spares about 4k of code. -- added new vendor id for Iomega DITTO 2GIG -- fixed a bug already present in zftape-1.06 when aborting a write - with a signal: we now finish the current volume at that - position. Header segments remain NOT up to date until an explicit call - to MTREW or MTOFFL is done. - -===== Release notes for ftape-3.00, 14/10/96 ===== - -- Merged ftape with zftape. There are three modules now: - ftape for the hardware support, sftape for the implementation of the - original ftape eof mark stuff and zftape that implements zftape's way - of handling things (compression, volume table, tape blocks of - constant length) -- Documentation in TeXinfo format in the `info' subdirectory. -- New ioctls for zftape. See zftape/zftape.h -- Dummy formatting ioctl for ftape. See ftape.h -- Kernel patch files for the 2.*.* series to include ftape-3.00 in the - kernel source tree. These includes a kernel compatible Config.in - script and fairly large online information for the kernel configure - script. -- Support for compiling with Linux-1.2.13. -- Modified GNU mt from their cpio package that can handle the new - ioctls. -- ftape/sftape/zftape is kerneld save now! - -Notes on sftape: -- sftape implements the eof handling code of the original ftape. If - you like to stick with the original ftape stuff, you have to use - this module, not zftape. -- sftape is kerneld save, unlike the original ftape. -- we keep the entire header segment now in memory, so no need to read - it before updating the header segments. Additional memory - consumption: 256 bytes. - -Notes for zftape: -- zftape has support for tapes with format code 6 now, which use a - slightly different volume table format compared with other floppy - tapes. -- new ioctls for zftape. Have a look at zftape/zftape.h -- The internal volume table representation has changed for zftape. Old - cartridges are converted automatically. -- zftape no longer uses compression map segments, which have vanished - from the QIC specs, but creates volume table entry that reserves - enough space for the compression map. -- zftape is kerneld save now. -- we keep the entire header segment now in memory, so no need to read - it before updating the header segments. Additional memory - consumption: 256 bytes. - -Notes for contrib/gnumt: -- modified mt from the GNU cpio package that supports all the new - ioctls of zftape. -Notes for contrib/swapout: -- This contains the swapout.c program that was written by Kai - Harrekilde-Pederson. I simply added a Makefile. - -===== Release notes for ftape-2.10, 14/10/96 ===== - -The ftape maintainer has changed. -Kai Harrekilde-Petersen -has resigned from maintaining ftape, and I, -Claus-Justus Heine , -have taken over. - -- Added support for tapes with `format code 6', i.e. QIC-3020 tapes - with more than 2^16 segments. -- merged changes made by Bas Laarhoven with ftape-2.09. Refer - to his release notes below. I've included them into this - file unchanged for your reference. -- disabled call stack back trace for now. This new feature - introduced by the interim release 2.0.x still seems to - be buggy. -- Tried to minimize differences between the ftape version - to be included into the kernel source tree and the standalone - module version. -- Reintroduced support for Linux-1.2.13. Please refer to the - Install-guide. - -===== Release notes for ftape-2.09, 16/06/96 ===== - -There aren't any really big news in this release, mostly just that I -(the maintainer) have changed my email address (due to a new job). My -new address is - -- The CLK_48MHZ and FDC_82078SL options has gone (all 2Mbps cards seem - to use a 48MHz oscillator anyway and I haven't heard of an 'SL - chip out there). -- The S82078B has been `downgraded' to i82077AA compability. -- TESTING option revived. Right now, it'll enable the (seriously broken) - 2Mbps code. If you enable it, you'll experience a tape drive that's - *really* out to lunch! -- Some (bold) changes in the init code. Please notify me if they - break things for you. - -===== Release notes for ftape-2.08, 14/03/96 ===== - -If you correct a problem with ftape, please send your patch to -khp@dolphinics.no too. - -- Updated to reflect that NR_MEM_LISTS is gone in 1.3.74 -- Teac 700 added to list of known drives. -- The registered device name is now "ft" rather than "ftape". - -===== Release notes for ftape-2.07a, 14/03/96 ===== - -Bugfixes by Marcin Dalecki : -- In the last release it just compiled against 1.3.70; - now the params to request_irq() and free_irq are() are fixed, so it also - works in 1.3.73 :-) -- Support for modules is now correct for newer kernels. - -===== Release notes for ftape-2.07, 04/03/96 ===== - - -- ftape updated to compile against 1.3.70. -- Iomega 700 and Wangtek 3200 recognised. - - -===== Release notes for ftape-2.06b, 13/02/96 ===== - -Another simple bugfix version. - -- Jumbo 700 recognised. -- Typo in vendors.h fixed. - - -===== Release notes for ftape-2.06a, 10/02/96 ===== - -This release is a simple bugfix version. - -- Linux/SMP: ftape *should* work. -- FC-10/20: Only accepts IRQs 3-7, or 9. If IRQ 9, properly tell the card - to use IRQ 2. Thanks to Greg Crider (gcrider@iclnet.org) for finding and - locating this bug and testing the patch. -- Insight drive recognised correctly again. -- Motor-on wakeup version of the Iomega 250 drive added - - -===== Release notes for ftape-2.06, 28/01/96 ===== - -Special thanks go to Neal Friedman and Steven Sorbom for their -help in producing and testing this release. - -I have continued to clean up the code, with an eye towards inclusion -of ftape in Linus' official kernel (In fact, as I type this, I am -running on a kernel with ftape support statically linked). I have -test-compiled ftape against my 1.2.13 tree without problems. -Hopefully, everything should be OK for the v1.2.x people. - -WARNING! Alan Cox has mailed me that ftape does *NOT* work with -Linux/SMP. If you try to run ftape under Linux/SMP, it will cause a -kernel deadlock (which is worse than a panic). - -- QIC-3020/TR-3: 1Mbps support works. Neal is capable of reading and - writing data to a tape. ftape will automatically detect the type of - tape (e.g. TR-3 vs QIC-80) and move the fdc in and out of - "perpendicular mode" as necessary. -- 2Mbps support is disabled by default, since it is not fully - debugged. If you are adventurous, remove -DFDC_82078SL in the - Makefile and see what happens :-) -- fdc detection: silly bugs removed (Only 2Mbps fdcs were affected) - and added detection of the National Semiconductors PC8744 fdc chip - (used in the PC873xx "super-IO" chips). -- Removed warning about incompatible types when compiling with Linux - 1.2.x. -- README.PCI updated with info about the DELL Dimension XPS P90. -- Connor TST3200R added to detected drives. -- `swapout' utility added to distribution. It will dirty 5Meg of - memory, trying to swap out other programs. Just say `make swapout' - to build it. ftape will do this automatically Real Soon Now (ie: - when I have found out which kernel memory alloc function to call). - - -===== Release notes for ftape-2.05, 08/01/96 ===== - -- For v1.2.x Kernels, you must apply the patch linux-1.2/ksyms.patch to - the kernel and rebuild it (it adds the __get_dma_pages symbol to - ksyms.c). -- Included new asm-i386/io.h file from v1.3.x kernel series, to enable - gcc v.2.7.[12] to compile v1.2.x kernels (linux-1.2/io.h). -- Module versions: If you wish to compile ftape as a versioned module, - you must first compile your kernel with CONFIG_MODVERSIONS=y. - Otherwise, you will get complaints that does not - exist (if that happens, a `touch modversions.h' will help you out). -- CLK_48MHZ: new define in the Makefile (default: non-zero). If you have - a tape controller card that uses the i82078(-1) chip, but cannot get - it to work with ftape, try set it to 0 (and please report this). -- QIC-3010/3020: Complete support is still missing, but will hopefully - come soon. Steven Sorbom has kindly provided me with hints about - this. Writing of QIC-3020 tapes definitely does NOT work (do not try - it! - the drive will not be in "perpendicular mode" and this will ruin - the formatting info on the tape). -- ftape_num_buffers is out of fashion: use NR_BUFFERS instead (and - recompile if you want to change it :-). - - -===== Release notes for ftape-2.04, 01/01/96 ===== - -This version by Kai Harrekilde-Petersen - -- ALERT! Support for Kernels earlier then v1.1.85 is about to go away. - I intend to clean up some of the code (getting rid of an annoyingly - large numbers of #ifdef mostly), which means that support for - pre-1.1.85 kernels must go as well. -- NR_FTAPE_BUFFERS is gone; You can instead select the number of dma - buffers by saying `insmod ftape.o ftape_num_buffer=' instead. -- Configure script gone. ftape will now automagically determine your - kernel version by /usr/include/linux/version.h instead. -- CONFIG_MODVERSIONS now work. All combinations of versioned / - unversioned kernel and ftape module works (at least with my 1.3.52 - kernel). -- If you have problems with inserting ftape into an old (1.2.x) - kernel (e.g. insmod says "1.2.8 does not match 1.2.8), recompile - your modules utilities with your new compiler. -- Reveal TB1400 drive added to vendors.h -- Support for the i82078-1 (2Mbps) chip is coming along. The - biggest problem is that I don't have such a card, which makes - testing / debugging somewhat problematic. The second biggest - problem is that I do not have the QIC-3010/3020 standards either. - Status right now is that the chip is detected, and it should be - possible to put it into 2Mbps mode. However, I do not know what - "extras" are needed to complete the support. Although putting the - i82078 into 1Mbps mode ought to work out of the box, it doesn't - (right now, ftape complains about id am errors). - - -===== Release notes for ftape-2.04beta5, 29/12/95 ===== - -Bas offline linux-tape ----------------------- -For reasons only known to the majordomo mail list processor, Bas was -kicked off the linux-tape list sometime during the summer. Being -overworked at his for-pay job, he didn't notice it much. Instead I -(Kai, khp@dolphinics.no) has worked on ftape to produce the 2.04(beta) -version. - -zftape ------- -Note that there exists a much improved version of ftape, written by -Claus-Justus Heine which is named -zftape, which conforms to the QIC-80 specs on how to mark backups, and -is capable of doing automatic compression. However, zftape makes -substantial changes to ftape, and I (Kai) have therefore declined to -integrate zftape into ftape. Hopefully, this will happen soon. - -CONFIG_QIC117 removed from the kernel -------------------------------------- -The biggest change of all is that ftape now will allocate its dma -buffers when it is inserted. The means that the CONFIG_QIC117 option -has disappeared from the Linux kernel as of v1.3.34. If you have an -earlier kernel, simply answer 'no' to the question will do the trick -(if you get complains about __get_free_pages() missing, contact the -linux-tape mailing list). - -Note that ftape-2.04beta will work equally well on kernels with and -without `ftape support'. The only catch is, that you will waste -around 96-128Kb of precious DMA'able memory on a box that has ftape -support compiled in. - -Now for the real changes: - -- FC-20 can now use DMA channels 1, 2, and 3. Thanks to Daniel - Cohen, catman@wpi.edu. -- ftape no longer requires a (gigantic) 96Kb buffer to be statically - allocated by the kernel. -- Added new Iomega drive (8882) to vendors.h -- -fno-strength-reduce added to Makefile, since GCC is broken. -- i82078-1 (2Mbps) FDC support started. - - -===== Release notes for ftape-2.03b, 27/05/95 ===== - -- Prevented verify_area to return error if called with zero length. -- Fixed a bug in flush_buffers that caused too much padding to be - written when a final segment had bad sectors. -- Increased maximum fast-seek overshoot value from 5 to 10 segments. -- Breaking loop after 5 retries when positioning fails. -- Fixed wrong calculation of tape length for QIC-3010 and QIC-3020 - tapes (densities were swapped). -- Fixed wrong calculation of overshoot on seek_forward: Wrong sign - of error. -- Suppress (false) error message due to new tape loaded. -- Added two new CMS drives (11c3 and 11c5) to vendors.h. - - -===== Release notes for ftape-2.03a, 09/05/95 ===== - -- Fixed display of old error (even if already cleared) in ftape_open. -- Improved tape length detection, ioctls would fail for 425 ft tapes. - Until the tape length is calculated with data from the header - segment, we'll use worst-case values. -- Clear eof_mark after rewinding ioctls. -- Fixed wrong version message (2.03 had 2.02g id). -- Fixed bug that caused the fdc to be reset very frequently. - This shouldn't affect normal operation but the timing of the - report routines has changed again and that may cause problems. - We'll just have to find out.... -- Implemented correct write precompensation setting for QIC-3010/3020. -- Cleaned up fdc_interrupt_wait routine. Hope it still works :-) -- Finally removed (already disabled) special eof mark handling for - gnu tar. -- Changed order of get_dma_residue and disable_dma in fdc-isr.c - because the current order would fail on at least one system. - We're back to the original order again, hope (and expect) this - doesn't break any other system. - - -===== Release notes for ftape-2.03, 07/05/95 ===== - -(Changes refer to the first ftape-2.02 release) - -Support for wide and extended length tapes ------------------------------------------- -The Conner TSM 420 and 850 drives are reported to be working. -I haven't received any reports about other brands; the TSM 420 -and 850 seem to be the most widely used wide drives. -Extended length tapes (425 ft) with normal QIC-80 drives -are operating too (At least I've had no reports stating otherwise). -_Not_ yet completely supported (although they may work) are -QIC-3020 drives and 2 Mbps floppy disk controllers won't work at -the highest speed. -If someone is kind enough to send me one of these, I'll include -support for it too ;-) - -Easier configuration --------------------- -Problems due to wrong settings in the Makefile are prevented -by using a configuration script that sets the necessary (kernel -version dependent) compile time options. -This kernel version is now determined from the sources found -at /usr/src/linux, or if not found, the old way using -/proc/version. -Versioned modules will be used automatically when supported -by- and configured in- the kernel. -Note that the current modules code (1.1.87) is still broken -and _needs_ the fix included in the insmod directory. -Please don't send me any more Oops reports caused by insmod :-( - -Reduced module size -------------------- -The standard module size is much reduced and some compile time -options can even reduce it further. (I don't recommend this -for normal use but it can be handy for rescue diskettes) - -Option: Approx. module size: - - 150 Kb -NO_TRACE 125 Kb -NO_TRACE_AT_ALL 67 Kb - - -Much improved driver interruption ---------------------------------- -Most possible loops have been broken and signal detection -has been improved. -In most cases the driver can be aborted by ^C (SIGINT) and -SIGKILL (kill -9) will generate be a sure kill. -(Note that aborting a tape operation may damage the last -data written to tape) - -Improved error recovery ------------------------ -Ftape now returns an error (ENODATA) to the application if -a segment proves to be unrecoverable and then skips the -bad segment. -This causes most applications to continue to work (tar -and afio) loosing only a small amount (up to 29 Kb) of data. -Retried read operations will now be done slightly off-track -to improve the chance of success. Serious head off-track -errors will be detected. - -FC-10 and FC-20 controllers ---------------------------- -Ftape now supports both the old CMS FC-10 and the newer FC-20 -controllers. -Because the operation of these cards is still undocumented, -thus far they will only work with the default settings (See -Makefile). Any feed-back on how to use them with other settings -will be welcome ! -Compilation will fail if one changes the settings to illegal -values. - -Kernels and compilers ---------------------- -Ftape is currently being developed using the 2.5.8 compiler. -The older 2.4.5 probably works too (Set option in Makefile!). -I have no experience with any later compilers nor Elf support. -Any information on this is welcome. -The latest kernel I have tested ftape with is 1.2.6. - -Compression ------------ -An impressive collection of changes for ftape including -on-the-fly compression is still lying on my desk. -If 2.03 proves to be reliable I might start integrating these -but as usual, I'm short in time :-( - -Formatting ----------- -There is still no way to format tapes under Linux. As far as -I know all attempts to write such a program have died now. -Since formatted tapes are rather common now, I think all we -need is a utility that writes a worst case pattern and verifies -that with the drive put in verify mode, reducing margins. -Any takers ? - -Furthermore ------------ -Cleaned up messages. -Prepared to support multiple tape drives on one fdc. -Thanks to all the people who sent bug reports and helped me -improve the driver. Without trying to be complete I'll mention -Gary Anderson (without his accurate reports and unreliable -hardware there wouldn't be a 2.03), Stefan Kneifel (FC-20), -Robert Broughton (FC-20, you were almost there ;-), Bjorn -Ekwall (for the versioned modules and buggy insmod ;-), Peter -Fox, Christopher Oliver, Ralph Whittaker and not the least -Linus Torvalds (for Linux and keeping me busy because of -changes to the kernel ;-) -Thanks to anyone I forgot, for the bug reports, the ftape -bashing and the mental support... - - -That's it for now. Have Fun, - -Bas. - - -===== Release notes for ftape-2.02g, 06/05/95 ===== - -- Added extra test to break read-id loop with signal. -- Changed rewind code to handle negative overshoot for drives - that take very long to start or stop. -- Let use of get/set i/o-regions depend on kernel version. -- Changed code to use a more general test for conditional - compilations depending on kernel version. -- Improved micro-step functionality to go off-track only - while reading (id & data). -- Added failure on tape-not-referenced bit in ftape_command. -- Added FOREVER option to read-wait routine. -- Changed read-id to use shorter timeout causing smaller - rewinds on timeout. -- Made kernel-interface functions static. - - -===== Release notes for ftape-2.02f, 03/05/95 ===== - -- Added support for dual tape drives on my system, extended Configure - script to detect host 'dodo'. -- Log media defect in history if ecc failed and no data was returned. -- Fixed Configure script that was failing for kernel versions with - double digit version or revision numbers. - - -===== Release notes for ftape-2.02e, 01/05/95 ===== - -- Fixed reposition loop at logical eot (failing read_id). -- Fixed 34 segment offset when rewinding. -- Added fast seek capability for more than 255 segments. -- Fixed wrong busy result from ftape_command causing reverse - seek to fail. -- Added breakout from infinite rewind loop (if something fails). - - -===== Release notes for ftape-2.02d, 30/04/95 ===== - -- Improved abortion on signals: Interrupt will make a graceful - exit, Kill will be less nice and should be used if everything - else fails. -- Included check for tape-head off track. -- Implemented exit from tape-start loop. -- Added kernel io-port registration. -- Implemented skip of failing segment (ENODATA) on ecc failure. - This allows afio and tar to continue when the tape is damaged. -- Made distinction between drive names with different codes. - - -===== Release notes for ftape-2.02c, 22/04/95 ===== - -- Fixed too tight command queueing after tape stop/pause command - issued from within interrupt service routine (Showed as timeout - on Acknowledge errors during retries on some systems) -- Tried to fix timeouts when using 425 ft tape because the extended - length doesn't seem to be detected by the hardware. - We now use the format code from the header segment so adjust the - timing after reading the header segment. -- Fixed some messages stating 'unexpected something...' being not - unexpected anymore. -- Started preparations for merge of dynamic buffer allocation and - compression code. -- Changed some debug messages to include relevant segment information - at level 4. -- Included early bail-out when drive offline, preventing a lot of - false messages. -- Moved ftape_parameter_xxx() offsets into function instead of in calls. -- Removed 'weird, drive busy but no data' error when caused by - an error during a read-id. -- Improved 'timeout on acknowledge' diagnostics. -- Moved MODULE option into Configure. -- Reduced code size when no tracing at all was set (Claus Heine). -- No longer log error code 0 (no error) as an error. - - -===== Release notes for ftape-2.02b, 09/04/95 ===== - -- Relaxed timing for status operation and displaying - abnormal results. Hopefully this shows what's going - wrong with the Conner TSM850R drives. -- Created script for configuration, using version number - of kernel source if available, otherwise /proc/version. -- Fixed conditionals in kernel-interface.c. -- Removed unavoidable TRACE output. - - -===== Release notes for ftape-2.02a, 01/04/95 ===== - -- Implemented `new-style' (versioned) modules support for new - kernels. -- Reduced size of module by moving static data to bss. -- Now using version number of kernel source instead of running - kernel for kernel versions >= 1.1.82 -- Added feedback on drive speeds to vendor information. -- Included fixed insmod sources to distribution (Let's hope - the modules distribution get fixed soon :-/). - -Note that I haven't yet implemented any of the code extension I -received. I hope to find some time to do this soon. - - -===== Release notes for ftape-2.02, 15/01/95 ===== - - -- Fixed failing repositioning when overshoot was incremented. -- Fixed rate selection: Because of a deficiency in the QIC-117 - specification one cannot distinguish between a not implemented - and a failing command. Therefor we now try to find out if the - drive does support this command before usage. -- Fixed error retry using wrong offset in fdc-isr. -- Improved retry code to retry only once on a single no-data - error in a segment. -- Validate sector number extracted from eof mark because an - invalid file mark (due to ???) could cause kernel panic. -- Split ftape-io.c into ftape-io.c and ftape-ctl.c files. -- Corrected too high media error count after writing to - a bad tape. -- Added #include again because old kernel versions - need it. -- Fixed fdc not being disabled when open failed because no tape - drive was found. -- Fixed problem with soft error in sector 32 (shift operator with - shiftcount 32 is not defined). - - -===== Release notes for ftape-2.01, 08/01/95 ===== - - -- Removed TESTING setting from distributed Makefile. -- Fixed `mt asf' failure: Rewind was deferred to close which - overruled the fsf ioctl. -- Prevented non-interruptible commands being interrupted. -- Added missing timeout.pause setting. -- Maximum tape speed read from drive type information table. - If the information is not in the table (0) the drive will - determine the speed itself and put a message in the logfile. - This information should then be added to the table in the - vendors.h file (and reported to me). -- Added call to ftape_init_drive after soft reset for those - (antique) drives that don't do an implicit seek_load_point - after a reset or power up. -- Don't try to set data rate if reset failed. -- Prevent update of seek variables when starting from the - beginning or the end of the tape. -- Fixed wrong adjustment of overshoot in seek_forward(). -- Added sync to Makefile (again). -- Added code to diagnose timer problems (calibr.c). -- Replaced time differences by timediff calls. -- Removed reference to do_floppy from object for recent kernels. -- Fixed wrong display of 'failing dma controller' message. -- Removed various no longer used #include statements. -- Added max. tape speed value to vendor-struct. -- Changed ftape-command to check pre-conditions and wait - if needed. -- Further updated qic117.h to rev G. -- Combined command name table and restrictions table to one. - Extended this table with some new fields. -- Increased timeout on Ack timer value and included code to - report out of spec behaviour. -- Increased rewind timeout margin to calculated + 20%. -- Improved data rate selection so it won't fail on some - older (pre standard) drives. -- Changed initialisation code so drive will be rewound if the - driver is reloaded and the tape is not at bot. -- Moved some of the flush operations from close to the ioctls. -- Added exit code value to failing verify area message. -- Loop until tape halted in smart-stop. -- Fast seek handled specially if located at bot or eot. -- Being more conservative on overshoot value. - - -===== Release notes for ftape-2.00, 31/12/94 ===== - - The Install-guide is completely rewritten and now also includes -some information on how to use the driver. If you're either new -to ftape or new to Unix tape devices make sure to read it ! - - If you own a pci system and experience problems with the -ftape driver make sure to read the README.PCI file. It contains -some hints on how to fix your hardware. - - For anybody who hasn't noticed: The version number of the -driver has been incremented (The latest released version has -been version 1.14d). - This has been done for two major reasons: - - o A new (better) error recovery scheme is implemented. - o Support for new drive types has been added. - - All these improvements/changes will probably include a couple -of new (and old?) bugs. If you encounter any problems that you think -I'm not yet aware of, feel free to send a report to . - I recommend keeping a version of ftape-1.14d available, just -in case ;-) - - This version should work with all kernel versions from 1.0.9 up -to 1.1.72 (and probably earlier and later versions too). - - -Major new features: - -- Better handling of tapes with defects: When a sector repeatedly - (SOFT_RETRIES in ftape.h) cannot be written to or read from it is - marked as an hard error and gets skipped. - The error correction code can handle up to three of these hard - errors provided there are no other errors in that segment (32 Kb). - -- Allows writing to tapes with defects (although the risk of loosing - data increases !) - Look for the media-defects entry printed with the statistics when - the tape is closed. A non-zero value here shows a bad tape. - [the actual count is wrong (too high), this is a known bug]. - -- Use of backup header segment if first one is failing. - -- Support for extended length tapes with QIC-80: both 425 and 1100 ft. - 0.25 inch tapes are now recognized and handled. - -- Support for new QIC-80 drives with 8 mm `wide' tapes (e.g. Conner - TSM 420). - -- Support for new QIC-3010 and QIC-3020 drives (experimental) with - both 0.25 inch and 8 mm tapes. - -Some minor features were added, a couple of small bugs were fixed and -probably some new ones introduced ;-). - -[lseek() didn't make it into this version] - -Have fun, - -Bas. ----- - LocalWords: ftape MCONFIG mt VFS zftape resp sftape proc subdir MTIOCVOLINFO - LocalWords: MTIOCGETSIZE BOT EOD MTBSF zft kerneld modprobe kdtime contrib TR - LocalWords: MTSETBLK afio uninstall texi www EIO QIC init sft eof aka dma GB - LocalWords: SIGKILL MTIOCFTCMD mmap Iomega FDC fdc io gnumt mtio fc asm inb - LocalWords: outb ft qic frontend TeXinfo irq mach MODVERSIONS CONFIG html dvi - LocalWords: usr doc SMP Mb Dunno FIXME vtblc perl listtape volinfo fsf MTWEOF - LocalWords: amanda degaussed ComByte DoublePlay whraven njackn com MTIOC vtbl - LocalWords: GETBLKSZ MAKEDEV zftape's linux dif CVS Revison cp MTREW MTOFFL - LocalWords: MTFSF BSF Marcin Dalecki GCC Config cpio swapout Kai Harrekilde - LocalWords: Pederson khp dolphinics Justus claus momo rwth aachen Laarhoven diff --git a/drivers/char/ftape/compressor/Makefile b/drivers/char/ftape/compressor/Makefile deleted file mode 100644 index 1fbd6c4019d..00000000000 --- a/drivers/char/ftape/compressor/Makefile +++ /dev/null @@ -1,31 +0,0 @@ -# -# Copyright (C) 1997 Claus-Justus Heine. -# -# This program is free software; you can redistribute it and/or modify -# it under the terms of the GNU General Public License as published by -# the Free Software Foundation; either version 2, or (at your option) -# any later version. -# -# This program is distributed in the hope that it will be useful, -# but WITHOUT ANY WARRANTY; without even the implied warranty of -# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -# GNU General Public License for more details. -# -# You should have received a copy of the GNU General Public License -# along with this program; see the file COPYING. If not, write to -# the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. -# -# $Source: /homes/cvs/ftape-stacked/ftape/compressor/Makefile,v $ -# $Revision: 1.1 $ -# $Date: 1997/10/05 19:12:28 $ -# -# Makefile for the optional compressor for th zftape VFS -# interface to the QIC-40/80/3010/3020 floppy-tape driver for -# Linux. -# - -obj-$(CONFIG_ZFT_COMPRESSOR) += zft-compressor.o - -zft-compressor-objs := zftape-compress.o lzrw3.o - -CFLAGS_lzrw3.o := -O6 -funroll-all-loops diff --git a/drivers/char/ftape/compressor/lzrw3.c b/drivers/char/ftape/compressor/lzrw3.c deleted file mode 100644 index a032a0ee2a9..00000000000 --- a/drivers/char/ftape/compressor/lzrw3.c +++ /dev/null @@ -1,743 +0,0 @@ -/* - * $Source: /homes/cvs/ftape-stacked/ftape/compressor/lzrw3.c,v $ - * $Revision: 1.1 $ - * $Date: 1997/10/05 19:12:29 $ - * - * Implementation of Ross Williams lzrw3 algorithm. Adaption for zftape. - * - */ - -#include "../compressor/lzrw3.h" /* Defines single exported function "compress". */ - -/******************************************************************************/ -/* */ -/* LZRW3.C */ -/* */ -/******************************************************************************/ -/* */ -/* Author : Ross Williams. */ -/* Date : 30-Jun-1991. */ -/* Release : 1. */ -/* */ -/******************************************************************************/ -/* */ -/* This file contains an implementation of the LZRW3 data compression */ -/* algorithm in C. */ -/* */ -/* The algorithm is a general purpose compression algorithm that runs fast */ -/* and gives reasonable compression. The algorithm is a member of the Lempel */ -/* Ziv family of algorithms and bases its compression on the presence in the */ -/* data of repeated substrings. */ -/* */ -/* This algorithm is unpatented and the code is public domain. As the */ -/* algorithm is based on the LZ77 class of algorithms, it is unlikely to be */ -/* the subject of a patent challenge. */ -/* */ -/* Unlike the LZRW1 and LZRW1-A algorithms, the LZRW3 algorithm is */ -/* deterministic and is guaranteed to yield the same compressed */ -/* representation for a given file each time it is run. */ -/* */ -/* The LZRW3 algorithm was originally designed and implemented */ -/* by Ross Williams on 31-Dec-1990. */ -/* */ -/* Here are the results of applying this code, compiled under THINK C 4.0 */ -/* and running on a Mac-SE (8MHz 68000), to the standard calgary corpus. */ -/* */ -/* +----------------------------------------------------------------+ */ -/* | DATA COMPRESSION TEST | */ -/* | ===================== | */ -/* | Time of run : Sun 30-Jun-1991 09:31PM | */ -/* | Timing accuracy : One part in 100 | */ -/* | Context length : 262144 bytes (= 256.0000K) | */ -/* | Test suite : Calgary Corpus Suite | */ -/* | Files in suite : 14 | */ -/* | Algorithm : LZRW3 | */ -/* | Note: All averages are calculated from the un-rounded values. | */ -/* +----------------------------------------------------------------+ */ -/* | File Name Length CxB ComLen %Remn Bits Com K/s Dec K/s | */ -/* | ---------- ------ --- ------ ----- ---- ------- ------- | */ -/* | rpus:Bib.D 111261 1 55033 49.5 3.96 19.46 32.27 | */ -/* | us:Book1.D 768771 3 467962 60.9 4.87 17.03 31.07 | */ -/* | us:Book2.D 610856 3 317102 51.9 4.15 19.39 34.15 | */ -/* | rpus:Geo.D 102400 1 82424 80.5 6.44 11.65 18.18 | */ -/* | pus:News.D 377109 2 205670 54.5 4.36 17.14 27.47 | */ -/* | pus:Obj1.D 21504 1 13027 60.6 4.85 13.40 18.95 | */ -/* | pus:Obj2.D 246814 1 116286 47.1 3.77 19.31 30.10 | */ -/* | s:Paper1.D 53161 1 27522 51.8 4.14 18.60 31.15 | */ -/* | s:Paper2.D 82199 1 45160 54.9 4.40 18.45 32.84 | */ -/* | rpus:Pic.D 513216 2 122388 23.8 1.91 35.29 51.05 | */ -/* | us:Progc.D 39611 1 19669 49.7 3.97 18.87 30.64 | */ -/* | us:Progl.D 71646 1 28247 39.4 3.15 24.34 40.66 | */ -/* | us:Progp.D 49379 1 19377 39.2 3.14 23.91 39.23 | */ -/* | us:Trans.D 93695 1 33481 35.7 2.86 25.48 40.37 | */ -/* +----------------------------------------------------------------+ */ -/* | Average 224401 1 110953 50.0 4.00 20.17 32.72 | */ -/* +----------------------------------------------------------------+ */ -/* */ -/******************************************************************************/ - -/******************************************************************************/ - -/* The following structure is returned by the "compress" function below when */ -/* the user asks the function to return identifying information. */ -/* The most important field in the record is the working memory field which */ -/* tells the calling program how much working memory should be passed to */ -/* "compress" when it is called to perform a compression or decompression. */ -/* LZRW3 uses the same amount of memory during compression and decompression. */ -/* For more information on this structure see "compress.h". */ - -#define U(X) ((ULONG) X) -#define SIZE_P_BYTE (U(sizeof(UBYTE *))) -#define SIZE_WORD (U(sizeof(UWORD ))) -#define ALIGNMENT_FUDGE (U(16)) -#define MEM_REQ ( U(4096)*(SIZE_P_BYTE) + ALIGNMENT_FUDGE ) - -static struct compress_identity identity = -{ - U(0x032DDEA8), /* Algorithm identification number. */ - MEM_REQ, /* Working memory (bytes) required. */ - "LZRW3", /* Name of algorithm. */ - "1.0", /* Version number of algorithm. */ - "31-Dec-1990", /* Date of algorithm. */ - "Public Domain", /* Copyright notice. */ - "Ross N. Williams", /* Author of algorithm. */ - "Renaissance Software", /* Affiliation of author. */ - "Public Domain" /* Vendor of algorithm. */ -}; - -LOCAL void compress_compress (UBYTE *,UBYTE *,ULONG,UBYTE *, LONG *); -LOCAL void compress_decompress(UBYTE *,UBYTE *,LONG, UBYTE *, ULONG *); - -/******************************************************************************/ - -/* This function is the only function exported by this module. */ -/* Depending on its first parameter, the function can be requested to */ -/* compress a block of memory, decompress a block of memory, or to identify */ -/* itself. For more information, see the specification file "compress.h". */ - -EXPORT void lzrw3_compress( - UWORD action, /* Action to be performed. */ - UBYTE *wrk_mem, /* Address of working memory we can use.*/ - UBYTE *src_adr, /* Address of input data. */ - LONG src_len, /* Length of input data. */ - UBYTE *dst_adr, /* Address to put output data. */ - void *p_dst_len /* Address of longword for length of output data.*/ -) -{ - switch (action) - { - case COMPRESS_ACTION_IDENTITY: - *((struct compress_identity **)p_dst_len)= &identity; - break; - case COMPRESS_ACTION_COMPRESS: - compress_compress(wrk_mem,src_adr,src_len,dst_adr,(LONG *)p_dst_len); - break; - case COMPRESS_ACTION_DECOMPRESS: - compress_decompress(wrk_mem,src_adr,src_len,dst_adr,(LONG *)p_dst_len); - break; - } -} - -/******************************************************************************/ -/* */ -/* BRIEF DESCRIPTION OF THE LZRW3 ALGORITHM */ -/* ======================================== */ -/* The LZRW3 algorithm is identical to the LZRW1-A algorithm except that */ -/* instead of transmitting history offsets, it transmits hash table indexes. */ -/* In order to decode the indexes, the decompressor must maintain an */ -/* identical hash table. Copy items are straightforward:when the decompressor */ -/* receives a copy item, it simply looks up the hash table to translate the */ -/* index into a pointer into the data already decompressed. To update the */ -/* hash table, it replaces the same table entry with a pointer to the start */ -/* of the newly decoded phrase. The tricky part is with literal items, for at */ -/* the time that the decompressor receives a literal item the decompressor */ -/* does not have the three bytes in the Ziv (that the compressor has) to */ -/* perform the three-byte hash. To solve this problem, in LZRW3, both the */ -/* compressor and decompressor are wired up so that they "buffer" these */ -/* literals and update their hash tables only when three bytes are available. */ -/* This makes the maximum buffering 2 bytes. */ -/* */ -/* Replacement of offsets by hash table indexes yields a few percent extra */ -/* compression at the cost of some speed. LZRW3 is slower than LZRW1, LZRW1-A */ -/* and LZRW2, but yields better compression. */ -/* */ -/* Extra compression could be obtained by using a hash table of depth two. */ -/* However, increasing the depth above one incurs a significant decrease in */ -/* compression speed which was not considered worthwhile. Another reason for */ -/* keeping the depth down to one was to allow easy comparison with the */ -/* LZRW1-A and LZRW2 algorithms so as to demonstrate the exact effect of the */ -/* use of direct hash indexes. */ -/* */ -/* +---+ */ -/* |___|4095 */ -/* |___| */ -/* +---------------------*_|<---+ /----+---\ */ -/* | |___| +---|Hash | */ -/* | |___| |Function| */ -/* | |___| \--------/ */ -/* | |___|0 ^ */ -/* | +---+ | */ -/* | Hash +-----+ */ -/* | Table | */ -/* | --- */ -/* v ^^^ */ -/* +-------------------------------------|----------------+ */ -/* |||||||||||||||||||||||||||||||||||||||||||||||||||||||| */ -/* +-------------------------------------|----------------+ */ -/* | |1......18| | */ -/* |<------- Lempel=History ------------>|<--Ziv-->| | */ -/* | (=bytes already processed) |<-Still to go-->| */ -/* |<-------------------- INPUT BLOCK ------------------->| */ -/* */ -/* The diagram above for LZRW3 looks almost identical to the diagram for */ -/* LZRW1. The difference is that in LZRW3, the compressor transmits hash */ -/* table indices instead of Lempel offsets. For this to work, the */ -/* decompressor must maintain a hash table as well as the compressor and both */ -/* compressor and decompressor must "buffer" literals, as the decompressor */ -/* cannot hash phrases commencing with a literal until another two bytes have */ -/* arrived. */ -/* */ -/* LZRW3 Algorithm Execution Summary */ -/* --------------------------------- */ -/* 1. Hash the first three bytes of the Ziv to yield a hash table index h. */ -/* 2. Look up the hash table yielding history pointer p. */ -/* 3. Match where p points with the Ziv. If there is a match of three or */ -/* more bytes, code those bytes (in the Ziv) as a copy item, otherwise */ -/* code the next byte in the Ziv as a literal item. */ -/* 4. Update the hash table as possible subject to the constraint that only */ -/* phrases commencing three bytes back from the Ziv can be hashed and */ -/* entered into the hash table. (This enables the decompressor to keep */ -/* pace). See the description and code for more details. */ -/* */ -/******************************************************************************/ -/* */ -/* DEFINITION OF COMPRESSED FILE FORMAT */ -/* ==================================== */ -/* * A compressed file consists of a COPY FLAG followed by a REMAINDER. */ -/* * The copy flag CF uses up four bytes with the first byte being the */ -/* least significant. */ -/* * If CF=1, then the compressed file represents the remainder of the file */ -/* exactly. Otherwise CF=0 and the remainder of the file consists of zero */ -/* or more GROUPS, each of which represents one or more bytes. */ -/* * Each group consists of two bytes of CONTROL information followed by */ -/* sixteen ITEMs except for the last group which can contain from one */ -/* to sixteen items. */ -/* * An item can be either a LITERAL item or a COPY item. */ -/* * Each item corresponds to a bit in the control bytes. */ -/* * The first control byte corresponds to the first 8 items in the group */ -/* with bit 0 corresponding to the first item in the group and bit 7 to */ -/* the eighth item in the group. */ -/* * The second control byte corresponds to the second 8 items in the group */ -/* with bit 0 corresponding to the ninth item in the group and bit 7 to */ -/* the sixteenth item in the group. */ -/* * A zero bit in a control word means that the corresponding item is a */ -/* literal item. A one bit corresponds to a copy item. */ -/* * A literal item consists of a single byte which represents itself. */ -/* * A copy item consists of two bytes that represent from 3 to 18 bytes. */ -/* * The first byte in a copy item will be denoted C1. */ -/* * The second byte in a copy item will be denoted C2. */ -/* * Bits will be selected using square brackets. */ -/* For example: C1[0..3] is the low nibble of the first control byte. */ -/* of copy item C1. */ -/* * The LENGTH of a copy item is defined to be C1[0..3]+3 which is a number */ -/* in the range [3,18]. */ -/* * The INDEX of a copy item is defined to be C1[4..7]*256+C2[0..8] which */ -/* is a number in the range [0,4095]. */ -/* * A copy item represents the sequence of bytes */ -/* text[POS-OFFSET..POS-OFFSET+LENGTH-1] where */ -/* text is the entire text of the uncompressed string. */ -/* POS is the index in the text of the character following the */ -/* string represented by all the items preceeding the item */ -/* being defined. */ -/* OFFSET is obtained from INDEX by looking up the hash table. */ -/* */ -/******************************************************************************/ - -/* The following #define defines the length of the copy flag that appears at */ -/* the start of the compressed file. The value of four bytes was chosen */ -/* because the fast_copy routine on my Macintosh runs faster if the source */ -/* and destination blocks are relatively longword aligned. */ -/* The actual flag data appears in the first byte. The rest are zeroed so as */ -/* to normalize the compressed representation (i.e. not non-deterministic). */ -#define FLAG_BYTES 4 - -/* The following #defines define the meaning of the values of the copy */ -/* flag at the start of the compressed file. */ -#define FLAG_COMPRESS 0 /* Signals that output was result of compression. */ -#define FLAG_COPY 1 /* Signals that output was simply copied over. */ - -/* The 68000 microprocessor (on which this algorithm was originally developed */ -/* is fussy about non-aligned arrays of words. To avoid these problems the */ -/* following macro can be used to "waste" from 0 to 3 bytes so as to align */ -/* the argument pointer. */ -#define ULONG_ALIGN_UP(X) ((((ULONG)X)+sizeof(ULONG)-1)&~(sizeof(ULONG)-1)) - - -/* The following constant defines the maximum length of an uncompressed item. */ -/* This definition must not be changed; its value is hardwired into the code. */ -/* The longest number of bytes that can be spanned by a single item is 18 */ -/* for the longest copy item. */ -#define MAX_RAW_ITEM (18) - -/* The following constant defines the maximum length of an uncompressed group.*/ -/* This definition must not be changed; its value is hardwired into the code. */ -/* A group contains at most 16 items which explains this definition. */ -#define MAX_RAW_GROUP (16*MAX_RAW_ITEM) - -/* The following constant defines the maximum length of a compressed group. */ -/* This definition must not be changed; its value is hardwired into the code. */ -/* A compressed group consists of two control bytes followed by up to 16 */ -/* compressed items each of which can have a maximum length of two bytes. */ -#define MAX_CMP_GROUP (2+16*2) - -/* The following constant defines the number of entries in the hash table. */ -/* This definition must not be changed; its value is hardwired into the code. */ -#define HASH_TABLE_LENGTH (4096) - -/* LZRW3, unlike LZRW1(-A), must initialize its hash table so as to enable */ -/* the compressor and decompressor to stay in step maintaining identical hash */ -/* tables. In an early version of the algorithm, the tables were simply */ -/* initialized to zero and a check for zero was included just before the */ -/* matching code. However, this test costs time. A better solution is to */ -/* initialize all the entries in the hash table to point to a constant */ -/* string. The decompressor does the same. This solution requires no extra */ -/* test. The contents of the string do not matter so long as the string is */ -/* the same for the compressor and decompressor and contains at least */ -/* MAX_RAW_ITEM bytes. I chose consecutive decimal digits because they do not */ -/* have white space problems (e.g. there is no chance that the compiler will */ -/* replace more than one space by a TAB) and because they make the length of */ -/* the string obvious by inspection. */ -#define START_STRING_18 ((UBYTE *) "123456789012345678") - -/* In this algorithm, hash values have to be calculated at more than one */ -/* point. The following macro neatens the code up for this. */ -#define HASH(PTR) \ - (((40543*(((*(PTR))<<8)^((*((PTR)+1))<<4)^(*((PTR)+2))))>>4) & 0xFFF) - -/******************************************************************************/ - -/* Input : Hand over the required amount of working memory in p_wrk_mem. */ -/* Input : Specify input block using p_src_first and src_len. */ -/* Input : Point p_dst_first to the start of the output zone (OZ). */ -/* Input : Point p_dst_len to a ULONG to receive the output length. */ -/* Input : Input block and output zone must not overlap. */ -/* Output : Length of output block written to *p_dst_len. */ -/* Output : Output block in Mem[p_dst_first..p_dst_first+*p_dst_len-1]. May */ -/* Output : write in OZ=Mem[p_dst_first..p_dst_first+src_len+MAX_CMP_GROUP-1].*/ -/* Output : Upon completion guaranteed *p_dst_len<=src_len+FLAG_BYTES. */ -LOCAL void compress_compress(UBYTE *p_wrk_mem, - UBYTE *p_src_first, ULONG src_len, - UBYTE *p_dst_first, LONG *p_dst_len) -{ - /* p_src and p_dst step through the source and destination blocks. */ - register UBYTE *p_src = p_src_first; - register UBYTE *p_dst = p_dst_first; - - /* The following variables are never modified and are used in the */ - /* calculations that determine when the main loop terminates. */ - UBYTE *p_src_post = p_src_first+src_len; - UBYTE *p_dst_post = p_dst_first+src_len; - UBYTE *p_src_max1 = p_src_first+src_len-MAX_RAW_ITEM; - UBYTE *p_src_max16 = p_src_first+src_len-MAX_RAW_ITEM*16; - - /* The variables 'p_control' and 'control' are used to buffer control bits. */ - /* Before each group is processed, the next two bytes of the output block */ - /* are set aside for the control word for the group about to be processed. */ - /* 'p_control' is set to point to the first byte of that word. Meanwhile, */ - /* 'control' buffers the control bits being generated during the processing */ - /* of the group. Instead of having a counter to keep track of how many items */ - /* have been processed (=the number of bits in the control word), at the */ - /* start of each group, the top word of 'control' is filled with 1 bits. */ - /* As 'control' is shifted for each item, the 1 bits in the top word are */ - /* absorbed or destroyed. When they all run out (i.e. when the top word is */ - /* all zero bits, we know that we are at the end of a group. */ -# define TOPWORD 0xFFFF0000 - UBYTE *p_control; - register ULONG control=TOPWORD; - - /* THe variable 'hash' always points to the first element of the hash table. */ - UBYTE **hash= (UBYTE **) ULONG_ALIGN_UP(p_wrk_mem); - - /* The following two variables represent the literal buffer. p_h1 points to */ - /* the hash table entry corresponding to the youngest literal. p_h2 points */ - /* to the hash table entry corresponding to the second youngest literal. */ - /* Note: p_h1=0=>p_h2=0 because zero values denote absence of a pending */ - /* literal. The variables are initialized to zero meaning an empty "buffer". */ - UBYTE **p_h1=NULL; - UBYTE **p_h2=NULL; - - /* To start, we write the flag bytes. Being optimistic, we set the flag to */ - /* FLAG_COMPRESS. The remaining flag bytes are zeroed so as to keep the */ - /* algorithm deterministic. */ - *p_dst++=FLAG_COMPRESS; - {UWORD i; for (i=2;i<=FLAG_BYTES;i++) *p_dst++=0;} - - /* Reserve the first word of output as the control word for the first group. */ - /* Note: This is undone at the end if the input block is empty. */ - p_control=p_dst; p_dst+=2; - - /* Initialize all elements of the hash table to point to a constant string. */ - /* Use of an unrolled loop speeds this up considerably. */ - {UWORD i; UBYTE **p_h=hash; -# define ZH *p_h++=START_STRING_18 - for (i=0;i<256;i++) /* 256=HASH_TABLE_LENGTH/16. */ - {ZH;ZH;ZH;ZH; - ZH;ZH;ZH;ZH; - ZH;ZH;ZH;ZH; - ZH;ZH;ZH;ZH;} - } - - /* The main loop processes either 1 or 16 items per iteration. As its */ - /* termination logic is complicated, I have opted for an infinite loop */ - /* structure containing 'break' and 'goto' statements. */ - while (TRUE) - {/* Begin main processing loop. */ - - /* Note: All the variables here except unroll should be defined within */ - /* the inner loop. Unfortunately the loop hasn't got a block. */ - register UBYTE *p; /* Scans through targ phrase during matching. */ - register UBYTE *p_ziv= NULL ; /* Points to first byte of current Ziv. */ - register UWORD unroll; /* Loop counter for unrolled inner loop. */ - register UWORD index; /* Index of current hash table entry. */ - register UBYTE **p_h0 = NULL ; /* Pointer to current hash table entry. */ - - /* Test for overrun and jump to overrun code if necessary. */ - if (p_dst>p_dst_post) - goto overrun; - - /* The following cascade of if statements efficiently catches and deals */ - /* with varying degrees of closeness to the end of the input block. */ - /* When we get very close to the end, we stop updating the table and */ - /* code the remaining bytes as literals. This makes the code simpler. */ - unroll=16; - if (p_src>p_src_max16) - { - unroll=1; - if (p_src>p_src_max1) - { - if (p_src==p_src_post) - break; - else - goto literal; - } - } - - /* This inner unrolled loop processes 'unroll' (whose value is either 1 */ - /* or 16) items. I have chosen to implement this loop with labels and */ - /* gotos to heighten the ease with which the loop may be implemented with */ - /* a single decrement and branch instruction in assembly language and */ - /* also because the labels act as highly readable place markers. */ - /* (Also because we jump into the loop for endgame literals (see above)). */ - - begin_unrolled_loop: - - /* To process the next phrase, we hash the next three bytes and use */ - /* the resultant hash table index to look up the hash table. A pointer */ - /* to the entry is stored in p_h0 so as to avoid an array lookup. The */ - /* hash table entry *p_h0 is looked up yielding a pointer p to a */ - /* potential match of the Ziv in the history. */ - index=HASH(p_src); - p_h0=&hash[index]; - p=*p_h0; - - /* Having looked up the candidate position, we are in a position to */ - /* attempt a match. The match loop has been unrolled using the PS */ - /* macro so that failure within the first three bytes automatically */ - /* results in the literal branch being taken. The coding is simple. */ - /* p_ziv saves p_src so we can let p_src wander. */ -# define PS *p++!=*p_src++ - p_ziv=p_src; - if (PS || PS || PS) - { - /* Literal. */ - - /* Code the literal byte as itself and a zero control bit. */ - p_src=p_ziv; literal: *p_dst++=*p_src++; control&=0xFFFEFFFF; - - /* We have just coded a literal. If we had two pending ones, that */ - /* makes three and we can update the hash table. */ - if (p_h2!=0) - {*p_h2=p_ziv-2;} - - /* In any case, rotate the hash table pointers for next time. */ - p_h2=p_h1; p_h1=p_h0; - - } - else - { - /* Copy */ - - /* Match up to 15 remaining bytes using an unrolled loop and code. */ -#if 0 - PS || PS || PS || PS || PS || PS || PS || PS || - PS || PS || PS || PS || PS || PS || PS || p_src++; -#else - if ( - !( PS || PS || PS || PS || PS || PS || PS || PS || - PS || PS || PS || PS || PS || PS || PS ) - ) p_src++; -#endif - *p_dst++=((index&0xF00)>>4)|(--p_src-p_ziv-3); - *p_dst++=index&0xFF; - - /* As we have just coded three bytes, we are now in a position to */ - /* update the hash table with the literal bytes that were pending */ - /* upon the arrival of extra context bytes. */ - if (p_h1!=0) - { - if (p_h2) - {*p_h2=p_ziv-2; p_h2=NULL;} - *p_h1=p_ziv-1; p_h1=NULL; - } - - /* In any case, we can update the hash table based on the current */ - /* position as we just coded at least three bytes in a copy items. */ - *p_h0=p_ziv; - - } - control>>=1; - - /* This loop is all set up for a decrement and jump instruction! */ -#ifndef linux -` end_unrolled_loop: if (--unroll) goto begin_unrolled_loop; -#else - /* end_unrolled_loop: */ if (--unroll) goto begin_unrolled_loop; -#endif - - /* At this point it will nearly always be the end of a group in which */ - /* case, we have to do some control-word processing. However, near the */ - /* end of the input block, the inner unrolled loop is only executed once. */ - /* This necessitates the 'if' test. */ - if ((control&TOPWORD)==0) - { - /* Write the control word to the place we saved for it in the output. */ - *p_control++= control &0xFF; - *p_control = (control>>8) &0xFF; - - /* Reserve the next word in the output block for the control word */ - /* for the group about to be processed. */ - p_control=p_dst; p_dst+=2; - - /* Reset the control bits buffer. */ - control=TOPWORD; - } - - } /* End main processing loop. */ - - /* After the main processing loop has executed, all the input bytes have */ - /* been processed. However, the control word has still to be written to the */ - /* word reserved for it in the output at the start of the most recent group. */ - /* Before writing, the control word has to be shifted so that all the bits */ - /* are in the right place. The "empty" bit positions are filled with 1s */ - /* which partially fill the top word. */ - while(control&TOPWORD) control>>=1; - *p_control++= control &0xFF; - *p_control++=(control>>8) &0xFF; - - /* If the last group contained no items, delete the control word too. */ - if (p_control==p_dst) p_dst-=2; - - /* Write the length of the output block to the dst_len parameter and return. */ - *p_dst_len=p_dst-p_dst_first; - return; - - /* Jump here as soon as an overrun is detected. An overrun is defined to */ - /* have occurred if p_dst>p_dst_first+src_len. That is, the moment the */ - /* length of the output written so far exceeds the length of the input block.*/ - /* The algorithm checks for overruns at least at the end of each group */ - /* which means that the maximum overrun is MAX_CMP_GROUP bytes. */ - /* Once an overrun occurs, the only thing to do is to set the copy flag and */ - /* copy the input over. */ - overrun: -#if 0 - *p_dst_first=FLAG_COPY; - fast_copy(p_src_first,p_dst_first+FLAG_BYTES,src_len); - *p_dst_len=src_len+FLAG_BYTES; -#else - fast_copy(p_src_first,p_dst_first,src_len); - *p_dst_len= -src_len; /* return a negative number to indicate uncompressed data */ -#endif -} - -/******************************************************************************/ - -/* Input : Hand over the required amount of working memory in p_wrk_mem. */ -/* Input : Specify input block using p_src_first and src_len. */ -/* Input : Point p_dst_first to the start of the output zone. */ -/* Input : Point p_dst_len to a ULONG to receive the output length. */ -/* Input : Input block and output zone must not overlap. User knows */ -/* Input : upperbound on output block length from earlier compression. */ -/* Input : In any case, maximum expansion possible is nine times. */ -/* Output : Length of output block written to *p_dst_len. */ -/* Output : Output block in Mem[p_dst_first..p_dst_first+*p_dst_len-1]. */ -/* Output : Writes only in Mem[p_dst_first..p_dst_first+*p_dst_len-1]. */ -LOCAL void compress_decompress( UBYTE *p_wrk_mem, - UBYTE *p_src_first, LONG src_len, - UBYTE *p_dst_first, ULONG *p_dst_len) -{ - /* Byte pointers p_src and p_dst scan through the input and output blocks. */ - register UBYTE *p_src = p_src_first+FLAG_BYTES; - register UBYTE *p_dst = p_dst_first; - /* we need to avoid a SEGV when trying to uncompress corrupt data */ - register UBYTE *p_dst_post = p_dst_first + *p_dst_len; - - /* The following two variables are never modified and are used to control */ - /* the main loop. */ - UBYTE *p_src_post = p_src_first+src_len; - UBYTE *p_src_max16 = p_src_first+src_len-(MAX_CMP_GROUP-2); - - /* The hash table is the only resident of the working memory. The hash table */ - /* contains HASH_TABLE_LENGTH=4096 pointers to positions in the history. To */ - /* keep Macintoshes happy, it is longword aligned. */ - UBYTE **hash = (UBYTE **) ULONG_ALIGN_UP(p_wrk_mem); - - /* The variable 'control' is used to buffer the control bits which appear in */ - /* groups of 16 bits (control words) at the start of each compressed group. */ - /* When each group is read, bit 16 of the register is set to one. Whenever */ - /* a new bit is needed, the register is shifted right. When the value of the */ - /* register becomes 1, we know that we have reached the end of a group. */ - /* Initializing the register to 1 thus instructs the code to follow that it */ - /* should read a new control word immediately. */ - register ULONG control=1; - - /* The value of 'literals' is always in the range 0..3. It is the number of */ - /* consecutive literal items just seen. We have to record this number so as */ - /* to know when to update the hash table. When literals gets to 3, there */ - /* have been three consecutive literals and we can update at the position of */ - /* the oldest of the three. */ - register UWORD literals=0; - - /* Check the leading copy flag to see if the compressor chose to use a copy */ - /* operation instead of a compression operation. If a copy operation was */ - /* used, then all we need to do is copy the data over, set the output length */ - /* and return. */ -#if 0 - if (*p_src_first==FLAG_COPY) - { - fast_copy(p_src_first+FLAG_BYTES,p_dst_first,src_len-FLAG_BYTES); - *p_dst_len=src_len-FLAG_BYTES; - return; - } -#else - if ( src_len < 0 ) - { - fast_copy(p_src_first,p_dst_first,-src_len ); - *p_dst_len = (ULONG)-src_len; - return; - } -#endif - - /* Initialize all elements of the hash table to point to a constant string. */ - /* Use of an unrolled loop speeds this up considerably. */ - {UWORD i; UBYTE **p_h=hash; -# define ZJ *p_h++=START_STRING_18 - for (i=0;i<256;i++) /* 256=HASH_TABLE_LENGTH/16. */ - {ZJ;ZJ;ZJ;ZJ; - ZJ;ZJ;ZJ;ZJ; - ZJ;ZJ;ZJ;ZJ; - ZJ;ZJ;ZJ;ZJ;} - } - - /* The outer loop processes either 1 or 16 items per iteration depending on */ - /* how close p_src is to the end of the input block. */ - while (p_src!=p_src_post) - {/* Start of outer loop */ - - register UWORD unroll; /* Counts unrolled loop executions. */ - - /* When 'control' has the value 1, it means that the 16 buffered control */ - /* bits that were read in at the start of the current group have all been */ - /* shifted out and that all that is left is the 1 bit that was injected */ - /* into bit 16 at the start of the current group. When we reach the end */ - /* of a group, we have to load a new control word and inject a new 1 bit. */ - if (control==1) - { - control=0x10000|*p_src++; - control|=(*p_src++)<<8; - } - - /* If it is possible that we are within 16 groups from the end of the */ - /* input, execute the unrolled loop only once, else process a whole group */ - /* of 16 items by looping 16 times. */ - unroll= p_src<=p_src_max16 ? 16 : 1; - - /* This inner loop processes one phrase (item) per iteration. */ - while (unroll--) - { /* Begin unrolled inner loop. */ - - /* Process a literal or copy item depending on the next control bit. */ - if (control&1) - { - /* Copy item. */ - - register UBYTE *p; /* Points to place from which to copy. */ - register UWORD lenmt; /* Length of copy item minus three. */ - register UBYTE **p_hte; /* Pointer to current hash table entry.*/ - register UBYTE *p_ziv=p_dst; /* Pointer to start of current Ziv. */ - - /* Read and dismantle the copy word. Work out from where to copy. */ - lenmt=*p_src++; - p_hte=&hash[((lenmt&0xF0)<<4)|*p_src++]; - p=*p_hte; - lenmt&=0xF; - - /* Now perform the copy using a half unrolled loop. */ - *p_dst++=*p++; - *p_dst++=*p++; - *p_dst++=*p++; - while (lenmt--) - *p_dst++=*p++; - - /* Because we have just received 3 or more bytes in a copy item */ - /* (whose bytes we have just installed in the output), we are now */ - /* in a position to flush all the pending literal hashings that had */ - /* been postponed for lack of bytes. */ - if (literals>0) - { - register UBYTE *r=p_ziv-literals; - hash[HASH(r)]=r; - if (literals==2) - {r++; hash[HASH(r)]=r;} - literals=0; - } - - /* In any case, we can immediately update the hash table with the */ - /* current position. We don't need to do a HASH(...) to work out */ - /* where to put the pointer, as the compressor just told us!!! */ - *p_hte=p_ziv; - - } - else - { - /* Literal item. */ - - /* Copy over the literal byte. */ - *p_dst++=*p_src++; - - /* If we now have three literals waiting to be hashed into the hash */ - /* table, we can do one of them now (because there are three). */ - if (++literals == 3) - {register UBYTE *p=p_dst-3; hash[HASH(p)]=p; literals=2;} - } - - /* Shift the control buffer so the next control bit is in bit 0. */ - control>>=1; -#if 1 - if (p_dst > p_dst_post) - { - /* Shit: we tried to decompress corrupt data */ - *p_dst_len = 0; - return; - } -#endif - } /* End unrolled inner loop. */ - - } /* End of outer loop */ - - /* Write the length of the decompressed data before returning. */ - *p_dst_len=p_dst-p_dst_first; -} - -/******************************************************************************/ -/* End of LZRW3.C */ -/******************************************************************************/ diff --git a/drivers/char/ftape/compressor/lzrw3.h b/drivers/char/ftape/compressor/lzrw3.h deleted file mode 100644 index 533feba4752..00000000000 --- a/drivers/char/ftape/compressor/lzrw3.h +++ /dev/null @@ -1,253 +0,0 @@ -#ifndef _LZRW3_H -#define _LZRW3_H -/* - * $Source: /homes/cvs/ftape-stacked/ftape/compressor/lzrw3.h,v $ - * $Revision: 1.1 $ - * $Date: 1997/10/05 19:12:30 $ - * - * include files for lzrw3. Only slighty modified from the original - * version. Assembles the three include files compress.h, port.h and - * fastcopy.h from the original lzrw3 package. - * - */ - -#include -#include - -/******************************************************************************/ -/* */ -/* COMPRESS.H */ -/* */ -/******************************************************************************/ -/* */ -/* Author : Ross Williams. */ -/* Date : December 1989. */ -/* */ -/* This header file defines the interface to a set of functions called */ -/* 'compress', each member of which implements a particular data compression */ -/* algorithm. */ -/* */ -/* Normally in C programming, for each .H file, there is a corresponding .C */ -/* file that implements the functions promised in the .H file. */ -/* Here, there are many .C files corresponding to this header file. */ -/* Each comforming implementation file contains a single function */ -/* called 'compress' that implements a single data compression */ -/* algorithm that conforms with the interface specified in this header file. */ -/* Only one algorithm can be linked in at a time in this organization. */ -/* */ -/******************************************************************************/ -/* */ -/* DEFINITION OF FUNCTION COMPRESS */ -/* =============================== */ -/* */ -/* Summary of Function Compress */ -/* ---------------------------- */ -/* The action that 'compress' takes depends on its first argument called */ -/* 'action'. The function provides three actions: */ -/* */ -/* - Return information about the algorithm. */ -/* - Compress a block of memory. */ -/* - Decompress a block of memory. */ -/* */ -/* Parameters */ -/* ---------- */ -/* See the formal C definition later for a description of the parameters. */ -/* */ -/* Constants */ -/* --------- */ -/* COMPRESS_OVERRUN: The constant COMPRESS_OVERRUN defines by how many bytes */ -/* an algorithm is allowed to expand a block during a compression operation. */ -/* */ -/* Although compression algorithms usually compress data, there will always */ -/* be data that a given compressor will expand (this can be proven). */ -/* Fortunately, the degree of expansion can be limited to a single bit, by */ -/* copying over the input data if the data gets bigger during compression. */ -/* To allow for this possibility, the first bit of a compressed */ -/* representation can be used as a flag indicating whether the */ -/* input data was copied over, or truly compressed. In practice, the first */ -/* byte would be used to store this bit so as to maintain byte alignment. */ -/* */ -/* Unfortunately, in general, the only way to tell if an algorithm will */ -/* expand a particular block of data is to run the algorithm on the data. */ -/* If the algorithm does not continuously monitor how many output bytes it */ -/* has written, it might write an output block far larger than the input */ -/* block before realizing that it has done so. */ -/* On the other hand, continuous checks on output length are inefficient. */ -/* */ -/* To cater for all these problems, this interface definition: */ -/* > Allows a compression algorithm to return an output block that is up to */ -/* COMPRESS_OVERRUN bytes longer than the input block. */ -/* > Allows a compression algorithm to write up to COMPRESS_OVERRUN bytes */ -/* more than the length of the input block to the memory of the output */ -/* block regardless of the length of the output block eventually returned. */ -/* This allows an algorithm to overrun the length of the input block in the */ -/* output block by up to COMPRESS_OVERRUN bytes between expansion checks. */ -/* */ -/* The problem does not arise for decompression. */ -/* */ -/* Identity Action */ -/* --------------- */ -/* > action must be COMPRESS_ACTION_IDENTITY. */ -/* > p_dst_len must point to a longword to receive a longword address. */ -/* > The value of the other parameters does not matter. */ -/* > After execution, the longword that p_dst_len points to will be a pointer */ -/* to a structure of type compress_identity. */ -/* Thus, for example, after the call, (*p_dst_len)->memory will return the */ -/* number of bytes of working memory that the algorithm requires to run. */ -/* > The values of the identity structure returned are fixed constant */ -/* attributes of the algorithm and must not vary from call to call. */ -/* */ -/* Common Requirements for Compression and Decompression Actions */ -/* ------------------------------------------------------------- */ -/* > wrk_mem must point to an unused block of memory of a length specified in */ -/* the algorithm's identity block. The identity block can be obtained by */ -/* making a separate call to compress, specifying the identity action. */ -/* > The INPUT BLOCK is defined to be Memory[src_addr,src_addr+src_len-1]. */ -/* > dst_len will be used to denote *p_dst_len. */ -/* > dst_len is not read by compress, only written. */ -/* > The value of dst_len is defined only upon termination. */ -/* > The OUTPUT BLOCK is defined to be Memory[dst_addr,dst_addr+dst_len-1]. */ -/* */ -/* Compression Action */ -/* ------------------ */ -/* > action must be COMPRESS_ACTION_COMPRESS. */ -/* > src_len must be in the range [0,COMPRESS_MAX_ORG]. */ -/* > The OUTPUT ZONE is defined to be */ -/* Memory[dst_addr,dst_addr+src_len-1+COMPRESS_OVERRUN]. */ -/* > The function can modify any part of the output zone regardless of the */ -/* final length of the output block. */ -/* > The input block and the output zone must not overlap. */ -/* > dst_len will be in the range [0,src_len+COMPRESS_OVERRUN]. */ -/* > dst_len will be in the range [0,COMPRESS_MAX_COM] (from prev fact). */ -/* > The output block will consist of a representation of the input block. */ -/* */ -/* Decompression Action */ -/* -------------------- */ -/* > action must be COMPRESS_ACTION_DECOMPRESS. */ -/* > The input block must be the result of an earlier compression operation. */ -/* > If the previous fact is true, the following facts must also be true: */ -/* > src_len will be in the range [0,COMPRESS_MAX_COM]. */ -/* > dst_len will be in the range [0,COMPRESS_MAX_ORG]. */ -/* > The input and output blocks must not overlap. */ -/* > Only the output block is modified. */ -/* > Upon termination, the output block will consist of the bytes contained */ -/* in the input block passed to the earlier compression operation. */ -/* */ -/******************************************************************************/ - -/******************************************************************************/ -/* */ -/* PORT.H */ -/* */ -/******************************************************************************/ -/* */ -/* This module contains macro definitions and types that are likely to */ -/* change between computers. */ -/* */ -/******************************************************************************/ - -#ifndef DONE_PORT /* Only do this if not previously done. */ - - #ifdef THINK_C - #define UBYTE unsigned char /* Unsigned byte */ - #define UWORD unsigned int /* Unsigned word (2 bytes) */ - #define ULONG unsigned long /* Unsigned word (4 bytes) */ - #define BOOL unsigned char /* Boolean */ - #define FOPEN_BINARY_READ "rb" /* Mode string for binary reading. */ - #define FOPEN_BINARY_WRITE "wb" /* Mode string for binary writing. */ - #define FOPEN_TEXT_APPEND "a" /* Mode string for text appending. */ - #define REAL double /* USed for floating point stuff. */ - #endif - #if defined(LINUX) || defined(linux) - #define UBYTE __u8 /* Unsigned byte */ - #define UWORD __u16 /* Unsigned word (2 bytes) */ - #define ULONG __u32 /* Unsigned word (4 bytes) */ - #define LONG __s32 /* Signed word (4 bytes) */ - #define BOOL is not used here /* Boolean */ - #define FOPEN_BINARY_READ not used /* Mode string for binary reading. */ - #define FOPEN_BINARY_WRITE not used /* Mode string for binary writing. */ - #define FOPEN_TEXT_APPEND not used /* Mode string for text appending. */ - #define REAL not used /* USed for floating point stuff. */ - #ifndef TRUE - #define TRUE 1 - #endif - #endif - - #define DONE_PORT /* Don't do all this again. */ - #define MALLOC_FAIL NULL /* Failure status from malloc() */ - #define LOCAL static /* For non-exported routines. */ - #define EXPORT /* Signals exported function. */ - #define then /* Useful for aligning ifs. */ - -#endif - -/******************************************************************************/ -/* End of PORT.H */ -/******************************************************************************/ - -#define COMPRESS_ACTION_IDENTITY 0 -#define COMPRESS_ACTION_COMPRESS 1 -#define COMPRESS_ACTION_DECOMPRESS 2 - -#define COMPRESS_OVERRUN 1024 -#define COMPRESS_MAX_COM 0x70000000 -#define COMPRESS_MAX_ORG (COMPRESS_MAX_COM-COMPRESS_OVERRUN) - -#define COMPRESS_MAX_STRLEN 255 - -/* The following structure provides information about the algorithm. */ -/* > The top bit of id must be zero. The remaining bits must be chosen by */ -/* the author of the algorithm by tossing a coin 31 times. */ -/* > The amount of memory requested by the algorithm is specified in bytes */ -/* and must be in the range [0,0x70000000]. */ -/* > All strings s must be such that strlen(s)<=COMPRESS_MAX_STRLEN. */ -struct compress_identity - { - ULONG id; /* Identifying number of algorithm. */ - ULONG memory; /* Number of bytes of working memory required. */ - - char *name; /* Name of algorithm. */ - char *version; /* Version number. */ - char *date; /* Date of release of this version. */ - char *copyright; /* Copyright message. */ - - char *author; /* Author of algorithm. */ - char *affiliation; /* Affiliation of author. */ - char *vendor; /* Where the algorithm can be obtained. */ - }; - -void lzrw3_compress( /* Single function interface to compression algorithm. */ -UWORD action, /* Action to be performed. */ -UBYTE *wrk_mem, /* Working memory temporarily given to routine to use. */ -UBYTE *src_adr, /* Address of input data. */ -LONG src_len, /* Length of input data. */ -UBYTE *dst_adr, /* Address of output data. */ -void *p_dst_len /* Pointer to a longword where routine will write: */ - /* If action=..IDENTITY => Adr of id structure. */ - /* If action=..COMPRESS => Length of output data. */ - /* If action=..DECOMPRESS => Length of output data. */ -); - -/******************************************************************************/ -/* End of COMPRESS.H */ -/******************************************************************************/ - - -/******************************************************************************/ -/* fast_copy.h */ -/******************************************************************************/ - -/* This function copies a block of memory very quickly. */ -/* The exact speed depends on the relative alignment of the blocks of memory. */ -/* PRE : 0<=src_len<=(2^32)-1 . */ -/* PRE : Source and destination blocks must not overlap. */ -/* POST : MEM[dst_adr,dst_adr+src_len-1]=MEM[src_adr,src_adr+src_len-1]. */ -/* POST : MEM[dst_adr,dst_adr+src_len-1] is the only memory changed. */ - -#define fast_copy(src,dst,len) memcpy(dst,src,len) - -/******************************************************************************/ -/* End of fast_copy.h */ -/******************************************************************************/ - -#endif diff --git a/drivers/char/ftape/compressor/zftape-compress.c b/drivers/char/ftape/compressor/zftape-compress.c deleted file mode 100644 index 65ffc0be3df..00000000000 --- a/drivers/char/ftape/compressor/zftape-compress.c +++ /dev/null @@ -1,1203 +0,0 @@ -/* - * Copyright (C) 1994-1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or - modify it under the terms of the GNU General Public License as - published by the Free Software Foundation; either version 2, or (at - your option) any later version. - - This program is distributed in the hope that it will be useful, but - WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, - USA. - - * - * This file implements a "generic" interface between the * - * zftape-driver and a compression-algorithm. The * - * compression-algorithm currently used is a LZ77. I use the * - * implementation lzrw3 by Ross N. Williams (Renaissance * - * Software). The compression program itself is in the file - * lzrw3.c * and lzrw3.h. To adopt another compression algorithm - * the functions * zft_compress() and zft_uncompress() must be - * changed * appropriately. See below. - */ - -#include -#include -#include - -#include - -#include - -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-eof.h" -#include "../zftape/zftape-ctl.h" -#include "../zftape/zftape-write.h" -#include "../zftape/zftape-read.h" -#include "../zftape/zftape-rw.h" -#include "../compressor/zftape-compress.h" -#include "../zftape/zftape-vtbl.h" -#include "../compressor/lzrw3.h" - -/* - * global variables - */ - -/* I handle the allocation of this buffer as a special case, because - * it's size varies depending on the tape length inserted. - */ - -/* local variables - */ -static void *zftc_wrk_mem = NULL; -static __u8 *zftc_buf = NULL; -static void *zftc_scratch_buf = NULL; - -/* compression statistics - */ -static unsigned int zftc_wr_uncompressed = 0; -static unsigned int zftc_wr_compressed = 0; -static unsigned int zftc_rd_uncompressed = 0; -static unsigned int zftc_rd_compressed = 0; - -/* forward */ -static int zftc_write(int *write_cnt, - __u8 *dst_buf, const int seg_sz, - const __u8 __user *src_buf, const int req_len, - const zft_position *pos, const zft_volinfo *volume); -static int zftc_read(int *read_cnt, - __u8 __user *dst_buf, const int to_do, - const __u8 *src_buf, const int seg_sz, - const zft_position *pos, const zft_volinfo *volume); -static int zftc_seek(unsigned int new_block_pos, - zft_position *pos, const zft_volinfo *volume, - __u8 *buffer); -static void zftc_lock (void); -static void zftc_reset (void); -static void zftc_cleanup(void); -static void zftc_stats (void); - -/* compressed segment. This conforms to QIC-80-MC, Revision K. - * - * Rev. K applies to tapes with `fixed length format' which is - * indicated by format code 2,3 and 5. See below for format code 4 and 6 - * - * 2 bytes: offset of compression segment structure - * 29k > offset >= 29k-18: data from previous segment ens in this - * segment and no compressed block starts - * in this segment - * offset == 0: data from previous segment occupies entire - * segment and continues in next segment - * n bytes: remainder from previous segment - * - * Rev. K: - * 4 bytes: 4 bytes: files set byte offset - * Post Rev. K and QIC-3020/3020: - * 8 bytes: 8 bytes: files set byte offset - * 2 bytes: byte count N (amount of data following) - * bit 15 is set if data is compressed, bit 15 is not - * set if data is uncompressed - * N bytes: data (as much as specified in the byte count) - * 2 bytes: byte count N_1 of next cluster - * N_1 bytes: data of next cluset - * 2 bytes: byte count N_2 of next cluster - * N_2 bytes: ... - * - * Note that the `N' byte count accounts only for the bytes that in the - * current segment if the cluster spans to the next segment. - */ - -typedef struct -{ - int cmpr_pos; /* actual position in compression buffer */ - int cmpr_sz; /* what is left in the compression buffer - * when copying the compressed data to the - * deblock buffer - */ - unsigned int first_block; /* location of header information in - * this segment - */ - unsigned int count; /* amount of data of current block - * contained in current segment - */ - unsigned int offset; /* offset in current segment */ - unsigned int spans:1; /* might continue in next segment */ - unsigned int uncmpr; /* 0x8000 if this block contains - * uncompressed data - */ - __s64 foffs; /* file set byte offset, same as in - * compression map segment - */ -} cmpr_info; - -static cmpr_info cseg; /* static data. Must be kept uptodate and shared by - * read, write and seek functions - */ - -#define DUMP_CMPR_INFO(level, msg, info) \ - TRACE(level, msg "\n" \ - KERN_INFO "cmpr_pos : %d\n" \ - KERN_INFO "cmpr_sz : %d\n" \ - KERN_INFO "first_block: %d\n" \ - KERN_INFO "count : %d\n" \ - KERN_INFO "offset : %d\n" \ - KERN_INFO "spans : %d\n" \ - KERN_INFO "uncmpr : 0x%04x\n" \ - KERN_INFO "foffs : " LL_X, \ - (info)->cmpr_pos, (info)->cmpr_sz, (info)->first_block, \ - (info)->count, (info)->offset, (info)->spans == 1, \ - (info)->uncmpr, LL((info)->foffs)) - -/* dispatch compression segment info, return error code - * - * afterwards, cseg->offset points to start of data of the NEXT - * compressed block, and cseg->count contains the amount of data - * left in the actual compressed block. cseg->spans is set to 1 if - * the block is continued in the following segment. Otherwise it is - * set to 0. - */ -static int get_cseg (cmpr_info *cinfo, const __u8 *buff, - const unsigned int seg_sz, - const zft_volinfo *volume) -{ - TRACE_FUN(ft_t_flow); - - cinfo->first_block = GET2(buff, 0); - if (cinfo->first_block == 0) { /* data spans to next segment */ - cinfo->count = seg_sz - sizeof(__u16); - cinfo->offset = seg_sz; - cinfo->spans = 1; - } else { /* cluster definetely ends in this segment */ - if (cinfo->first_block > seg_sz) { - /* data corrupted */ - TRACE_ABORT(-EIO, ft_t_err, "corrupted data:\n" - KERN_INFO "segment size: %d\n" - KERN_INFO "first block : %d", - seg_sz, cinfo->first_block); - } - cinfo->count = cinfo->first_block - sizeof(__u16); - cinfo->offset = cinfo->first_block; - cinfo->spans = 0; - } - /* now get the offset the first block should have in the - * uncompressed data stream. - * - * For this magic `18' refer to CRF-3 standard or QIC-80MC, - * Rev. K. - */ - if ((seg_sz - cinfo->offset) > 18) { - if (volume->qic113) { /* > revision K */ - TRACE(ft_t_data_flow, "New QIC-113 compliance"); - cinfo->foffs = GET8(buff, cinfo->offset); - cinfo->offset += sizeof(__s64); - } else { - TRACE(/* ft_t_data_flow */ ft_t_noise, "pre QIC-113 version"); - cinfo->foffs = (__s64)GET4(buff, cinfo->offset); - cinfo->offset += sizeof(__u32); - } - } - if (cinfo->foffs > volume->size) { - TRACE_ABORT(-EIO, ft_t_err, "Inconsistency:\n" - KERN_INFO "offset in current volume: %d\n" - KERN_INFO "size of current volume : %d", - (int)(cinfo->foffs>>10), (int)(volume->size>>10)); - } - if (cinfo->cmpr_pos + cinfo->count > volume->blk_sz) { - TRACE_ABORT(-EIO, ft_t_err, "Inconsistency:\n" - KERN_INFO "block size : %d\n" - KERN_INFO "data record: %d", - volume->blk_sz, cinfo->cmpr_pos + cinfo->count); - } - DUMP_CMPR_INFO(ft_t_noise /* ft_t_any */, "", cinfo); - TRACE_EXIT 0; -} - -/* This one is called, when a new cluster starts in same segment. - * - * Note: if this is the first cluster in the current segment, we must - * not check whether there are more than 18 bytes available because - * this have already been done in get_cseg() and there may be less - * than 18 bytes available due to header information. - * - */ -static void get_next_cluster(cmpr_info *cluster, const __u8 *buff, - const int seg_sz, const int finish) -{ - TRACE_FUN(ft_t_flow); - - if (seg_sz - cluster->offset > 18 || cluster->foffs != 0) { - cluster->count = GET2(buff, cluster->offset); - cluster->uncmpr = cluster->count & 0x8000; - cluster->count -= cluster->uncmpr; - cluster->offset += sizeof(__u16); - cluster->foffs = 0; - if ((cluster->offset + cluster->count) < seg_sz) { - cluster->spans = 0; - } else if (cluster->offset + cluster->count == seg_sz) { - cluster->spans = !finish; - } else { - /* either an error or a volume written by an - * old version. If this is a data error, then we'll - * catch it later. - */ - TRACE(ft_t_data_flow, "Either error or old volume"); - cluster->spans = 1; - cluster->count = seg_sz - cluster->offset; - } - } else { - cluster->count = 0; - cluster->spans = 0; - cluster->foffs = 0; - } - DUMP_CMPR_INFO(ft_t_noise /* ft_t_any */ , "", cluster); - TRACE_EXIT; -} - -static void zftc_lock(void) -{ -} - -/* this function is needed for zftape_reset_position in zftape-io.c - */ -static void zftc_reset(void) -{ - TRACE_FUN(ft_t_flow); - - memset((void *)&cseg, '\0', sizeof(cseg)); - zftc_stats(); - TRACE_EXIT; -} - -static int cmpr_mem_initialized = 0; -static unsigned int alloc_blksz = 0; - -static int zft_allocate_cmpr_mem(unsigned int blksz) -{ - TRACE_FUN(ft_t_flow); - - if (cmpr_mem_initialized && blksz == alloc_blksz) { - TRACE_EXIT 0; - } - TRACE_CATCH(zft_vmalloc_once(&zftc_wrk_mem, CMPR_WRK_MEM_SIZE), - zftc_cleanup()); - TRACE_CATCH(zft_vmalloc_always(&zftc_buf, blksz + CMPR_OVERRUN), - zftc_cleanup()); - alloc_blksz = blksz; - TRACE_CATCH(zft_vmalloc_always(&zftc_scratch_buf, blksz+CMPR_OVERRUN), - zftc_cleanup()); - cmpr_mem_initialized = 1; - TRACE_EXIT 0; -} - -static void zftc_cleanup(void) -{ - TRACE_FUN(ft_t_flow); - - zft_vfree(&zftc_wrk_mem, CMPR_WRK_MEM_SIZE); - zft_vfree(&zftc_buf, alloc_blksz + CMPR_OVERRUN); - zft_vfree(&zftc_scratch_buf, alloc_blksz + CMPR_OVERRUN); - cmpr_mem_initialized = alloc_blksz = 0; - TRACE_EXIT; -} - -/***************************************************************************** - * * - * The following two functions "ftape_compress()" and * - * "ftape_uncompress()" are the interface to the actual compression * - * algorithm (i.e. they are calling the "compress()" function from * - * the lzrw3 package for now). These routines could quite easily be * - * changed to adopt another compression algorithm instead of lzrw3, * - * which currently is used. * - * * - *****************************************************************************/ - -/* called by zft_compress_write() to perform the compression. Must - * return the size of the compressed data. - * - * NOTE: The size of the compressed data should not exceed the size of - * the uncompressed data. Most compression algorithms have means - * to store data unchanged if the "compressed" data amount would - * exceed the original one. Mostly this is done by storing some - * flag-bytes in front of the compressed data to indicate if it - * is compressed or not. Thus the worst compression result - * length is the original length plus those flag-bytes. - * - * We don't want that, as the QIC-80 standard provides a means - * of marking uncompressed blocks by simply setting bit 15 of - * the compressed block's length. Thus a compessed block can - * have at most a length of 2^15-1 bytes. The QIC-80 standard - * restricts the block-length even further, allowing only 29k - - * 6 bytes. - * - * Currently, the maximum blocksize used by zftape is 28k. - * - * In short: don't exceed the length of the input-package, set - * bit 15 of the compressed size to 1 if you have copied data - * instead of compressing it. - */ -static int zft_compress(__u8 *in_buffer, unsigned int in_sz, __u8 *out_buffer) -{ - __s32 compressed_sz; - TRACE_FUN(ft_t_flow); - - - lzrw3_compress(COMPRESS_ACTION_COMPRESS, zftc_wrk_mem, - in_buffer, in_sz, out_buffer, &compressed_sz); - if (TRACE_LEVEL >= ft_t_info) { - /* the compiler will optimize this away when - * compiled with NO_TRACE_AT_ALL option - */ - TRACE(ft_t_data_flow, "\n" - KERN_INFO "before compression: %d bytes\n" - KERN_INFO "after compresison : %d bytes", - in_sz, - (int)(compressed_sz < 0 - ? -compressed_sz : compressed_sz)); - /* for statistical purposes - */ - zftc_wr_compressed += (compressed_sz < 0 - ? -compressed_sz : compressed_sz); - zftc_wr_uncompressed += in_sz; - } - TRACE_EXIT (int)compressed_sz; -} - -/* called by zft_compress_read() to decompress the data. Must - * return the size of the decompressed data for sanity checks - * (compared with zft_blk_sz) - * - * NOTE: Read the note for zft_compress() above! If bit 15 of the - * parameter in_sz is set, then the data in in_buffer isn't - * compressed, which must be handled by the un-compression - * algorithm. (I changed lzrw3 to handle this.) - * - * The parameter max_out_sz is needed to prevent buffer overruns when - * uncompressing corrupt data. - */ -static unsigned int zft_uncompress(__u8 *in_buffer, - int in_sz, - __u8 *out_buffer, - unsigned int max_out_sz) -{ - TRACE_FUN(ft_t_flow); - - lzrw3_compress(COMPRESS_ACTION_DECOMPRESS, zftc_wrk_mem, - in_buffer, (__s32)in_sz, - out_buffer, (__u32 *)&max_out_sz); - - if (TRACE_LEVEL >= ft_t_info) { - TRACE(ft_t_data_flow, "\n" - KERN_INFO "before decompression: %d bytes\n" - KERN_INFO "after decompression : %d bytes", - in_sz < 0 ? -in_sz : in_sz,(int)max_out_sz); - /* for statistical purposes - */ - zftc_rd_compressed += in_sz < 0 ? -in_sz : in_sz; - zftc_rd_uncompressed += max_out_sz; - } - TRACE_EXIT (unsigned int)max_out_sz; -} - -/* print some statistics about the efficiency of the compression to - * the kernel log - */ -static void zftc_stats(void) -{ - TRACE_FUN(ft_t_flow); - - if (TRACE_LEVEL < ft_t_info) { - TRACE_EXIT; - } - if (zftc_wr_uncompressed != 0) { - if (zftc_wr_compressed > (1<<14)) { - TRACE(ft_t_info, "compression statistics (writing):\n" - KERN_INFO " compr./uncmpr. : %3d %%", - (((zftc_wr_compressed>>10) * 100) - / (zftc_wr_uncompressed>>10))); - } else { - TRACE(ft_t_info, "compression statistics (writing):\n" - KERN_INFO " compr./uncmpr. : %3d %%", - ((zftc_wr_compressed * 100) - / zftc_wr_uncompressed)); - } - } - if (zftc_rd_uncompressed != 0) { - if (zftc_rd_compressed > (1<<14)) { - TRACE(ft_t_info, "compression statistics (reading):\n" - KERN_INFO " compr./uncmpr. : %3d %%", - (((zftc_rd_compressed>>10) * 100) - / (zftc_rd_uncompressed>>10))); - } else { - TRACE(ft_t_info, "compression statistics (reading):\n" - KERN_INFO " compr./uncmpr. : %3d %%", - ((zftc_rd_compressed * 100) - / zftc_rd_uncompressed)); - } - } - /* only print it once: */ - zftc_wr_uncompressed = - zftc_wr_compressed = - zftc_rd_uncompressed = - zftc_rd_compressed = 0; - TRACE_EXIT; -} - -/* start new compressed block - */ -static int start_new_cseg(cmpr_info *cluster, - char *dst_buf, - const zft_position *pos, - const unsigned int blk_sz, - const char *src_buf, - const int this_segs_sz, - const int qic113) -{ - int size_left; - int cp_cnt; - int buf_pos; - TRACE_FUN(ft_t_flow); - - size_left = this_segs_sz - sizeof(__u16) - cluster->cmpr_sz; - TRACE(ft_t_data_flow,"\n" - KERN_INFO "segment size : %d\n" - KERN_INFO "compressed_sz: %d\n" - KERN_INFO "size_left : %d", - this_segs_sz, cluster->cmpr_sz, size_left); - if (size_left > 18) { /* start a new cluseter */ - cp_cnt = cluster->cmpr_sz; - cluster->cmpr_sz = 0; - buf_pos = cp_cnt + sizeof(__u16); - PUT2(dst_buf, 0, buf_pos); - - if (qic113) { - __s64 foffs = pos->volume_pos; - if (cp_cnt) foffs += (__s64)blk_sz; - - TRACE(ft_t_data_flow, "new style QIC-113 header"); - PUT8(dst_buf, buf_pos, foffs); - buf_pos += sizeof(__s64); - } else { - __u32 foffs = (__u32)pos->volume_pos; - if (cp_cnt) foffs += (__u32)blk_sz; - - TRACE(ft_t_data_flow, "old style QIC-80MC header"); - PUT4(dst_buf, buf_pos, foffs); - buf_pos += sizeof(__u32); - } - } else if (size_left >= 0) { - cp_cnt = cluster->cmpr_sz; - cluster->cmpr_sz = 0; - buf_pos = cp_cnt + sizeof(__u16); - PUT2(dst_buf, 0, buf_pos); - /* zero unused part of segment. */ - memset(dst_buf + buf_pos, '\0', size_left); - buf_pos = this_segs_sz; - } else { /* need entire segment and more space */ - PUT2(dst_buf, 0, 0); - cp_cnt = this_segs_sz - sizeof(__u16); - cluster->cmpr_sz -= cp_cnt; - buf_pos = this_segs_sz; - } - memcpy(dst_buf + sizeof(__u16), src_buf + cluster->cmpr_pos, cp_cnt); - cluster->cmpr_pos += cp_cnt; - TRACE_EXIT buf_pos; -} - -/* return-value: the number of bytes removed from the user-buffer - * `src_buf' or error code - * - * int *write_cnt : how much actually has been moved to the - * dst_buf. Need not be initialized when - * function returns with an error code - * (negativ return value) - * __u8 *dst_buf : kernel space buffer where the has to be - * copied to. The contents of this buffers - * goes to a specific segment. - * const int seg_sz : the size of the segment dst_buf will be - * copied to. - * const zft_position *pos : struct containing the coordinates in - * the current volume (byte position, - * segment id of current segment etc) - * const zft_volinfo *volume: information about the current volume, - * size etc. - * const __u8 *src_buf : user space buffer that contains the - * data the user wants to be written to - * tape. - * const int req_len : the amount of data the user wants to be - * written to tape. - */ -static int zftc_write(int *write_cnt, - __u8 *dst_buf, const int seg_sz, - const __u8 __user *src_buf, const int req_len, - const zft_position *pos, const zft_volinfo *volume) -{ - int req_len_left = req_len; - int result; - int len_left; - int buf_pos_write = pos->seg_byte_pos; - TRACE_FUN(ft_t_flow); - - /* Note: we do not unlock the module because - * there are some values cached in that `cseg' variable. We - * don't don't want to use this information when being - * unloaded by kerneld even when the tape is full or when we - * cannot allocate enough memory. - */ - if (pos->tape_pos > (volume->size-volume->blk_sz-ZFT_CMPR_OVERHEAD)) { - TRACE_EXIT -ENOSPC; - } - if (zft_allocate_cmpr_mem(volume->blk_sz) < 0) { - /* should we unlock the module? But it shouldn't - * be locked anyway ... - */ - TRACE_EXIT -ENOMEM; - } - if (buf_pos_write == 0) { /* fill a new segment */ - *write_cnt = buf_pos_write = start_new_cseg(&cseg, - dst_buf, - pos, - volume->blk_sz, - zftc_buf, - seg_sz, - volume->qic113); - if (cseg.cmpr_sz == 0 && cseg.cmpr_pos != 0) { - req_len_left -= result = volume->blk_sz; - cseg.cmpr_pos = 0; - } else { - result = 0; - } - } else { - *write_cnt = result = 0; - } - - len_left = seg_sz - buf_pos_write; - while ((req_len_left > 0) && (len_left > 18)) { - /* now we have some size left for a new compressed - * block. We know, that the compression buffer is - * empty (else there wouldn't be any space left). - */ - if (copy_from_user(zftc_scratch_buf, src_buf + result, - volume->blk_sz) != 0) { - TRACE_EXIT -EFAULT; - } - req_len_left -= volume->blk_sz; - cseg.cmpr_sz = zft_compress(zftc_scratch_buf, volume->blk_sz, - zftc_buf); - if (cseg.cmpr_sz < 0) { - cseg.uncmpr = 0x8000; - cseg.cmpr_sz = -cseg.cmpr_sz; - } else { - cseg.uncmpr = 0; - } - /* increment "result" iff we copied the entire - * compressed block to the zft_deblock_buf - */ - len_left -= sizeof(__u16); - if (len_left >= cseg.cmpr_sz) { - len_left -= cseg.count = cseg.cmpr_sz; - cseg.cmpr_pos = cseg.cmpr_sz = 0; - result += volume->blk_sz; - } else { - cseg.cmpr_sz -= - cseg.cmpr_pos = - cseg.count = len_left; - len_left = 0; - } - PUT2(dst_buf, buf_pos_write, cseg.uncmpr | cseg.count); - buf_pos_write += sizeof(__u16); - memcpy(dst_buf + buf_pos_write, zftc_buf, cseg.count); - buf_pos_write += cseg.count; - *write_cnt += cseg.count + sizeof(__u16); - FT_SIGNAL_EXIT(_DONT_BLOCK); - } - /* erase the remainder of the segment if less than 18 bytes - * left (18 bytes is due to the QIC-80 standard) - */ - if (len_left <= 18) { - memset(dst_buf + buf_pos_write, '\0', len_left); - (*write_cnt) += len_left; - } - TRACE(ft_t_data_flow, "returning %d", result); - TRACE_EXIT result; -} - -/* out: - * - * int *read_cnt: the number of bytes we removed from the zft_deblock_buf - * (result) - * int *to_do : the remaining size of the read-request. - * - * in: - * - * char *buff : buff is the address of the upper part of the user - * buffer, that hasn't been filled with data yet. - - * int buf_pos_read : copy of from _ftape_read() - * int buf_len_read : copy of buf_len_rd from _ftape_read() - * char *zft_deblock_buf: zft_deblock_buf - * unsigned short blk_sz: the block size valid for this volume, may differ - * from zft_blk_sz. - * int finish: if != 0 means that this is the last segment belonging - * to this volume - * returns the amount of data actually copied to the user-buffer - * - * to_do MUST NOT SHRINK except to indicate an EOF. In this case *to_do has to - * be set to 0 - */ -static int zftc_read (int *read_cnt, - __u8 __user *dst_buf, const int to_do, - const __u8 *src_buf, const int seg_sz, - const zft_position *pos, const zft_volinfo *volume) -{ - int uncompressed_sz; - int result = 0; - int remaining = to_do; - TRACE_FUN(ft_t_flow); - - TRACE_CATCH(zft_allocate_cmpr_mem(volume->blk_sz),); - if (pos->seg_byte_pos == 0) { - /* new segment just read - */ - TRACE_CATCH(get_cseg(&cseg, src_buf, seg_sz, volume), - *read_cnt = 0); - memcpy(zftc_buf + cseg.cmpr_pos, src_buf + sizeof(__u16), - cseg.count); - cseg.cmpr_pos += cseg.count; - *read_cnt = cseg.offset; - DUMP_CMPR_INFO(ft_t_noise /* ft_t_any */, "", &cseg); - } else { - *read_cnt = 0; - } - /* loop and uncompress until user buffer full or - * deblock-buffer empty - */ - TRACE(ft_t_data_flow, "compressed_sz: %d, compos : %d, *read_cnt: %d", - cseg.cmpr_sz, cseg.cmpr_pos, *read_cnt); - while ((cseg.spans == 0) && (remaining > 0)) { - if (cseg.cmpr_pos != 0) { /* cmpr buf is not empty */ - uncompressed_sz = - zft_uncompress(zftc_buf, - cseg.uncmpr == 0x8000 ? - -cseg.cmpr_pos : cseg.cmpr_pos, - zftc_scratch_buf, - volume->blk_sz); - if (uncompressed_sz != volume->blk_sz) { - *read_cnt = 0; - TRACE_ABORT(-EIO, ft_t_warn, - "Uncompressed blk (%d) != blk size (%d)", - uncompressed_sz, volume->blk_sz); - } - if (copy_to_user(dst_buf + result, - zftc_scratch_buf, - uncompressed_sz) != 0 ) { - TRACE_EXIT -EFAULT; - } - remaining -= uncompressed_sz; - result += uncompressed_sz; - cseg.cmpr_pos = 0; - } - if (remaining > 0) { - get_next_cluster(&cseg, src_buf, seg_sz, - volume->end_seg == pos->seg_pos); - if (cseg.count != 0) { - memcpy(zftc_buf, src_buf + cseg.offset, - cseg.count); - cseg.cmpr_pos = cseg.count; - cseg.offset += cseg.count; - *read_cnt += cseg.count + sizeof(__u16); - } else { - remaining = 0; - } - } - TRACE(ft_t_data_flow, "\n" - KERN_INFO "compressed_sz: %d\n" - KERN_INFO "compos : %d\n" - KERN_INFO "*read_cnt : %d", - cseg.cmpr_sz, cseg.cmpr_pos, *read_cnt); - } - if (seg_sz - cseg.offset <= 18) { - *read_cnt += seg_sz - cseg.offset; - TRACE(ft_t_data_flow, "expanding read cnt to: %d", *read_cnt); - } - TRACE(ft_t_data_flow, "\n" - KERN_INFO "segment size : %d\n" - KERN_INFO "read count : %d\n" - KERN_INFO "buf_pos_read : %d\n" - KERN_INFO "remaining : %d", - seg_sz, *read_cnt, pos->seg_byte_pos, - seg_sz - *read_cnt - pos->seg_byte_pos); - TRACE(ft_t_data_flow, "returning: %d", result); - TRACE_EXIT result; -} - -/* seeks to the new data-position. Reads sometimes a segment. - * - * start_seg and end_seg give the boundaries of the current volume - * blk_sz is the blk_sz of the current volume as stored in the - * volume label - * - * We don't allow blocksizes less than 1024 bytes, therefore we don't need - * a 64 bit argument for new_block_pos. - */ - -static int seek_in_segment(const unsigned int to_do, cmpr_info *c_info, - const char *src_buf, const int seg_sz, - const int seg_pos, const zft_volinfo *volume); -static int slow_seek_forward_until_error(const unsigned int distance, - cmpr_info *c_info, zft_position *pos, - const zft_volinfo *volume, __u8 *buf); -static int search_valid_segment(unsigned int segment, - const unsigned int end_seg, - const unsigned int max_foffs, - zft_position *pos, cmpr_info *c_info, - const zft_volinfo *volume, __u8 *buf); -static int slow_seek_forward(unsigned int dest, cmpr_info *c_info, - zft_position *pos, const zft_volinfo *volume, - __u8 *buf); -static int compute_seg_pos(unsigned int dest, zft_position *pos, - const zft_volinfo *volume); - -#define ZFT_SLOW_SEEK_THRESHOLD 10 /* segments */ -#define ZFT_FAST_SEEK_MAX_TRIALS 10 /* times */ -#define ZFT_FAST_SEEK_BACKUP 10 /* segments */ - -static int zftc_seek(unsigned int new_block_pos, - zft_position *pos, const zft_volinfo *volume, __u8 *buf) -{ - unsigned int dest; - int limit; - int distance; - int result = 0; - int seg_dist; - int new_seg; - int old_seg = 0; - int fast_seek_trials = 0; - TRACE_FUN(ft_t_flow); - - if (new_block_pos == 0) { - pos->seg_pos = volume->start_seg; - pos->seg_byte_pos = 0; - pos->volume_pos = 0; - zftc_reset(); - TRACE_EXIT 0; - } - dest = new_block_pos * (volume->blk_sz >> 10); - distance = dest - (pos->volume_pos >> 10); - while (distance != 0) { - seg_dist = compute_seg_pos(dest, pos, volume); - TRACE(ft_t_noise, "\n" - KERN_INFO "seg_dist: %d\n" - KERN_INFO "distance: %d\n" - KERN_INFO "dest : %d\n" - KERN_INFO "vpos : %d\n" - KERN_INFO "seg_pos : %d\n" - KERN_INFO "trials : %d", - seg_dist, distance, dest, - (unsigned int)(pos->volume_pos>>10), pos->seg_pos, - fast_seek_trials); - if (distance > 0) { - if (seg_dist < 0) { - TRACE(ft_t_bug, "BUG: distance %d > 0, " - "segment difference %d < 0", - distance, seg_dist); - result = -EIO; - break; - } - new_seg = pos->seg_pos + seg_dist; - if (new_seg > volume->end_seg) { - new_seg = volume->end_seg; - } - if (old_seg == new_seg || /* loop */ - seg_dist <= ZFT_SLOW_SEEK_THRESHOLD || - fast_seek_trials >= ZFT_FAST_SEEK_MAX_TRIALS) { - TRACE(ft_t_noise, "starting slow seek:\n" - KERN_INFO "fast seek failed too often: %s\n" - KERN_INFO "near target position : %s\n" - KERN_INFO "looping between two segs : %s", - (fast_seek_trials >= - ZFT_FAST_SEEK_MAX_TRIALS) - ? "yes" : "no", - (seg_dist <= ZFT_SLOW_SEEK_THRESHOLD) - ? "yes" : "no", - (old_seg == new_seg) - ? "yes" : "no"); - result = slow_seek_forward(dest, &cseg, - pos, volume, buf); - break; - } - old_seg = new_seg; - limit = volume->end_seg; - fast_seek_trials ++; - for (;;) { - result = search_valid_segment(new_seg, limit, - volume->size, - pos, &cseg, - volume, buf); - if (result == 0 || result == -EINTR) { - break; - } - if (new_seg == volume->start_seg) { - result = -EIO; /* set errror - * condition - */ - break; - } - limit = new_seg; - new_seg -= ZFT_FAST_SEEK_BACKUP; - if (new_seg < volume->start_seg) { - new_seg = volume->start_seg; - } - } - if (result < 0) { - TRACE(ft_t_warn, - "Couldn't find a readable segment"); - break; - } - } else /* if (distance < 0) */ { - if (seg_dist > 0) { - TRACE(ft_t_bug, "BUG: distance %d < 0, " - "segment difference %d >0", - distance, seg_dist); - result = -EIO; - break; - } - new_seg = pos->seg_pos + seg_dist; - if (fast_seek_trials > 0 && seg_dist == 0) { - /* this avoids sticking to the same - * segment all the time. On the other hand: - * if we got here for the first time, and the - * deblock_buffer still contains a valid - * segment, then there is no need to skip to - * the previous segment if the desired position - * is inside this segment. - */ - new_seg --; - } - if (new_seg < volume->start_seg) { - new_seg = volume->start_seg; - } - limit = pos->seg_pos; - fast_seek_trials ++; - for (;;) { - result = search_valid_segment(new_seg, limit, - pos->volume_pos, - pos, &cseg, - volume, buf); - if (result == 0 || result == -EINTR) { - break; - } - if (new_seg == volume->start_seg) { - result = -EIO; /* set errror - * condition - */ - break; - } - limit = new_seg; - new_seg -= ZFT_FAST_SEEK_BACKUP; - if (new_seg < volume->start_seg) { - new_seg = volume->start_seg; - } - } - if (result < 0) { - TRACE(ft_t_warn, - "Couldn't find a readable segment"); - break; - } - } - distance = dest - (pos->volume_pos >> 10); - } - TRACE_EXIT result; -} - - -/* advance inside the given segment at most to_do bytes. - * of kilobytes moved - */ - -static int seek_in_segment(const unsigned int to_do, - cmpr_info *c_info, - const char *src_buf, - const int seg_sz, - const int seg_pos, - const zft_volinfo *volume) -{ - int result = 0; - int blk_sz = volume->blk_sz >> 10; - int remaining = to_do; - TRACE_FUN(ft_t_flow); - - if (c_info->offset == 0) { - /* new segment just read - */ - TRACE_CATCH(get_cseg(c_info, src_buf, seg_sz, volume),); - c_info->cmpr_pos += c_info->count; - DUMP_CMPR_INFO(ft_t_noise, "", c_info); - } - /* loop and uncompress until user buffer full or - * deblock-buffer empty - */ - TRACE(ft_t_noise, "compressed_sz: %d, compos : %d", - c_info->cmpr_sz, c_info->cmpr_pos); - while (c_info->spans == 0 && remaining > 0) { - if (c_info->cmpr_pos != 0) { /* cmpr buf is not empty */ - result += blk_sz; - remaining -= blk_sz; - c_info->cmpr_pos = 0; - } - if (remaining > 0) { - get_next_cluster(c_info, src_buf, seg_sz, - volume->end_seg == seg_pos); - if (c_info->count != 0) { - c_info->cmpr_pos = c_info->count; - c_info->offset += c_info->count; - } else { - break; - } - } - /* Allow escape from this loop on signal! - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - DUMP_CMPR_INFO(ft_t_noise, "", c_info); - TRACE(ft_t_noise, "to_do: %d", remaining); - } - if (seg_sz - c_info->offset <= 18) { - c_info->offset = seg_sz; - } - TRACE(ft_t_noise, "\n" - KERN_INFO "segment size : %d\n" - KERN_INFO "buf_pos_read : %d\n" - KERN_INFO "remaining : %d", - seg_sz, c_info->offset, - seg_sz - c_info->offset); - TRACE_EXIT result; -} - -static int slow_seek_forward_until_error(const unsigned int distance, - cmpr_info *c_info, - zft_position *pos, - const zft_volinfo *volume, - __u8 *buf) -{ - unsigned int remaining = distance; - int seg_sz; - int seg_pos; - int result; - TRACE_FUN(ft_t_flow); - - seg_pos = pos->seg_pos; - do { - TRACE_CATCH(seg_sz = zft_fetch_segment(seg_pos, buf, - FT_RD_AHEAD),); - /* now we have the contents of the actual segment in - * the deblock buffer - */ - TRACE_CATCH(result = seek_in_segment(remaining, c_info, buf, - seg_sz, seg_pos,volume),); - remaining -= result; - pos->volume_pos += result<<10; - pos->seg_pos = seg_pos; - pos->seg_byte_pos = c_info->offset; - seg_pos ++; - if (seg_pos <= volume->end_seg && c_info->offset == seg_sz) { - pos->seg_pos ++; - pos->seg_byte_pos = 0; - c_info->offset = 0; - } - /* Allow escape from this loop on signal! - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - TRACE(ft_t_noise, "\n" - KERN_INFO "remaining: %d\n" - KERN_INFO "seg_pos: %d\n" - KERN_INFO "end_seg: %d\n" - KERN_INFO "result: %d", - remaining, seg_pos, volume->end_seg, result); - } while (remaining > 0 && seg_pos <= volume->end_seg); - TRACE_EXIT 0; -} - -/* return segment id of next segment containing valid data, -EIO otherwise - */ -static int search_valid_segment(unsigned int segment, - const unsigned int end_seg, - const unsigned int max_foffs, - zft_position *pos, - cmpr_info *c_info, - const zft_volinfo *volume, - __u8 *buf) -{ - cmpr_info tmp_info; - int seg_sz; - TRACE_FUN(ft_t_flow); - - memset(&tmp_info, 0, sizeof(cmpr_info)); - while (segment <= end_seg) { - FT_SIGNAL_EXIT(_DONT_BLOCK); - TRACE(ft_t_noise, - "Searching readable segment between %d and %d", - segment, end_seg); - seg_sz = zft_fetch_segment(segment, buf, FT_RD_AHEAD); - if ((seg_sz > 0) && - (get_cseg (&tmp_info, buf, seg_sz, volume) >= 0) && - (tmp_info.foffs != 0 || segment == volume->start_seg)) { - if ((tmp_info.foffs>>10) > max_foffs) { - TRACE_ABORT(-EIO, ft_t_noise, "\n" - KERN_INFO "cseg.foff: %d\n" - KERN_INFO "dest : %d", - (int)(tmp_info.foffs >> 10), - max_foffs); - } - DUMP_CMPR_INFO(ft_t_noise, "", &tmp_info); - *c_info = tmp_info; - pos->seg_pos = segment; - pos->volume_pos = c_info->foffs; - pos->seg_byte_pos = c_info->offset; - TRACE(ft_t_noise, "found segment at %d", segment); - TRACE_EXIT 0; - } - segment++; - } - TRACE_EXIT -EIO; -} - -static int slow_seek_forward(unsigned int dest, - cmpr_info *c_info, - zft_position *pos, - const zft_volinfo *volume, - __u8 *buf) -{ - unsigned int distance; - int result = 0; - TRACE_FUN(ft_t_flow); - - distance = dest - (pos->volume_pos >> 10); - while ((distance > 0) && - (result = slow_seek_forward_until_error(distance, - c_info, - pos, - volume, - buf)) < 0) { - if (result == -EINTR) { - break; - } - TRACE(ft_t_noise, "seg_pos: %d", pos->seg_pos); - /* the failing segment is either pos->seg_pos or - * pos->seg_pos + 1. There is no need to further try - * that segment, because ftape_read_segment() already - * has tried very much to read it. So we start with - * following segment, which is pos->seg_pos + 1 - */ - if(search_valid_segment(pos->seg_pos+1, volume->end_seg, dest, - pos, c_info, - volume, buf) < 0) { - TRACE(ft_t_noise, "search_valid_segment() failed"); - result = -EIO; - break; - } - distance = dest - (pos->volume_pos >> 10); - result = 0; - TRACE(ft_t_noise, "segment: %d", pos->seg_pos); - /* found valid segment, retry the seek */ - } - TRACE_EXIT result; -} - -static int compute_seg_pos(const unsigned int dest, - zft_position *pos, - const zft_volinfo *volume) -{ - int segment; - int distance = dest - (pos->volume_pos >> 10); - unsigned int raw_size; - unsigned int virt_size; - unsigned int factor; - TRACE_FUN(ft_t_flow); - - if (distance >= 0) { - raw_size = volume->end_seg - pos->seg_pos + 1; - virt_size = ((unsigned int)(volume->size>>10) - - (unsigned int)(pos->volume_pos>>10) - + FT_SECTORS_PER_SEGMENT - FT_ECC_SECTORS - 1); - virt_size /= FT_SECTORS_PER_SEGMENT - FT_ECC_SECTORS; - if (virt_size == 0 || raw_size == 0) { - TRACE_EXIT 0; - } - if (raw_size >= (1<<25)) { - factor = raw_size/(virt_size>>7); - } else { - factor = (raw_size<<7)/virt_size; - } - segment = distance/(FT_SECTORS_PER_SEGMENT-FT_ECC_SECTORS); - segment = (segment * factor)>>7; - } else { - raw_size = pos->seg_pos - volume->start_seg + 1; - virt_size = ((unsigned int)(pos->volume_pos>>10) - + FT_SECTORS_PER_SEGMENT - FT_ECC_SECTORS - 1); - virt_size /= FT_SECTORS_PER_SEGMENT - FT_ECC_SECTORS; - if (virt_size == 0 || raw_size == 0) { - TRACE_EXIT 0; - } - if (raw_size >= (1<<25)) { - factor = raw_size/(virt_size>>7); - } else { - factor = (raw_size<<7)/virt_size; - } - segment = distance/(FT_SECTORS_PER_SEGMENT-FT_ECC_SECTORS); - } - TRACE(ft_t_noise, "factor: %d/%d", factor, 1<<7); - TRACE_EXIT segment; -} - -static struct zft_cmpr_ops cmpr_ops = { - zftc_write, - zftc_read, - zftc_seek, - zftc_lock, - zftc_reset, - zftc_cleanup -}; - -int zft_compressor_init(void) -{ - TRACE_FUN(ft_t_flow); - -#ifdef MODULE - printk(KERN_INFO "zftape compressor v1.00a 970514 for " FTAPE_VERSION "\n"); - if (TRACE_LEVEL >= ft_t_info) { - printk( -KERN_INFO "(c) 1997 Claus-Justus Heine (claus@momo.math.rwth-aachen.de)\n" -KERN_INFO "Compressor for zftape (lzrw3 algorithm)\n"); - } -#else /* !MODULE */ - /* print a short no-nonsense boot message */ - printk(KERN_INFO "zftape compressor v1.00a 970514\n"); - printk(KERN_INFO "For use with " FTAPE_VERSION "\n"); -#endif /* MODULE */ - TRACE(ft_t_info, "zft_compressor_init @ 0x%p", zft_compressor_init); - TRACE(ft_t_info, "installing compressor for zftape ..."); - TRACE_CATCH(zft_cmpr_register(&cmpr_ops),); - TRACE_EXIT 0; -} - -#ifdef MODULE - -MODULE_AUTHOR( - "(c) 1996, 1997 Claus-Justus Heine (claus@momo.math.rwth-aachen.de"); -MODULE_DESCRIPTION( -"Compression routines for zftape. Uses the lzrw3 algorithm by Ross Williams"); -MODULE_LICENSE("GPL"); - -/* Called by modules package when installing the driver - */ -int init_module(void) -{ - return zft_compressor_init(); -} - -#endif /* MODULE */ diff --git a/drivers/char/ftape/compressor/zftape-compress.h b/drivers/char/ftape/compressor/zftape-compress.h deleted file mode 100644 index f200741e33b..00000000000 --- a/drivers/char/ftape/compressor/zftape-compress.h +++ /dev/null @@ -1,83 +0,0 @@ -#ifndef _ZFTAPE_COMPRESS_H -#define _ZFTAPE_COMPRESS_H -/* - * Copyright (c) 1994-1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or - modify it under the terms of the GNU General Public License as - published by the Free Software Foundation; either version 2, or (at - your option) any later version. - - This program is distributed in the hope that it will be useful, but - WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, - USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/compressor/zftape-compress.h,v $ - * $Revision: 1.1 $ - * $Date: 1997/10/05 19:12:32 $ - * - * This file contains macros and definitions for zftape's - * builtin compression code. - * - */ - -#include "../zftape/zftape-buffers.h" -#include "../zftape/zftape-vtbl.h" -#include "../compressor/lzrw3.h" - -/* CMPR_WRK_MEM_SIZE gives the size of the compression wrk_mem */ -/* I got these out of lzrw3.c */ -#define U(X) ((__u32) X) -#define SIZE_P_BYTE (U(sizeof(__u8 *))) -#define ALIGNMENT_FUDGE (U(16)) - -#define CMPR_WRK_MEM_SIZE (U(4096)*(SIZE_P_BYTE) + ALIGNMENT_FUDGE) - -/* the maximum number of bytes the size of the "compressed" data can - * exceed the uncompressed data. As it is quite useless to compress - * data twice it is sometimes the case that it is more efficient to - * copy a block of data but to feed it to the "compression" - * algorithm. In this case there are some flag bytes or the like - * proceding the "compressed" data. THAT MUST NOT BE THE CASE for the - * algorithm we use for this driver. Instead, the high bit 15 of - * compressed_size: - * - * compressed_size = ftape_compress() - * - * must be set in such a case. - * - * Nevertheless, it might also be as for lzrw3 that there is an - * "intermediate" overrun that exceeds the amount of the compressed - * data that is actually produced. During the algorithm we need in the - * worst case MAX_CMP_GROUP bytes more than the input-size. - */ -#define MAX_CMP_GROUP (2+16*2) /* from lzrw3.c */ - -#define CMPR_OVERRUN MAX_CMP_GROUP /* during compression */ - -/****************************************************/ - -#define CMPR_BUFFER_SIZE (MAX_BLOCK_SIZE + CMPR_OVERRUN) - -/* the compression map stores the byte offset compressed blocks within - * the current volume for catridges with format code 2,3 and 5 - * (and old versions of zftape) and the offset measured in kilobytes for - * format code 4 and 6. This gives us a possible max. size of a - * compressed volume of 1024*4GIG which should be enough. - */ -typedef __u32 CmprMap; - -/* globals - */ - -/* exported functions - */ - -#endif /* _ZFTAPE_COMPRESS_H */ diff --git a/drivers/char/ftape/lowlevel/Makefile b/drivers/char/ftape/lowlevel/Makefile deleted file mode 100644 index febab07ba42..00000000000 --- a/drivers/char/ftape/lowlevel/Makefile +++ /dev/null @@ -1,43 +0,0 @@ -# -# Copyright (C) 1996, 1997 Clau-Justus Heine. -# -# This program is free software; you can redistribute it and/or modify -# it under the terms of the GNU General Public License as published by -# the Free Software Foundation; either version 2, or (at your option) -# any later version. -# -# This program is distributed in the hope that it will be useful, -# but WITHOUT ANY WARRANTY; without even the implied warranty of -# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -# GNU General Public License for more details. -# -# You should have received a copy of the GNU General Public License -# along with this program; see the file COPYING. If not, write to -# the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. -# -# $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/Makefile,v $ -# $Revision: 1.4 $ -# $Date: 1997/10/07 09:26:02 $ -# -# Makefile for the lowlevel part QIC-40/80/3010/3020 floppy-tape -# driver for Linux. -# - -obj-$(CONFIG_FTAPE) += ftape.o - -ftape-objs := ftape-init.o fdc-io.o fdc-isr.o \ - ftape-bsm.o ftape-ctl.o ftape-read.o ftape-rw.o \ - ftape-write.o ftape-io.o ftape-calibr.o ftape-ecc.o fc-10.o \ - ftape-buffer.o ftape-format.o ftape_syms.o - -ifeq ($(CONFIG_FTAPE),y) -ftape-objs += ftape-setup.o -endif - -ifndef CONFIG_FT_NO_TRACE_AT_ALL -ftape-objs += ftape-tracing.o -endif - -ifeq ($(CONFIG_FT_PROC_FS),y) -ftape-objs += ftape-proc.o -endif diff --git a/drivers/char/ftape/lowlevel/fc-10.c b/drivers/char/ftape/lowlevel/fc-10.c deleted file mode 100644 index 9bc1cddade7..00000000000 --- a/drivers/char/ftape/lowlevel/fc-10.c +++ /dev/null @@ -1,175 +0,0 @@ -/* - * - - Copyright (C) 1993,1994 Jon Tombs. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - The entire guts of this program was written by dosemu, modified to - record reads and writes to the ports in the 0x180-0x188 address space, - while running the CMS program TAPE.EXE V2.0.5 supplied with the drive. - - Modified to use an array of addresses and generally cleaned up (made - much shorter) 4 June 94, dosemu isn't that good at writing short code it - would seem :-). Made independent of 0x180, but I doubt it will work - at any other address. - - Modified for distribution with ftape source. 21 June 94, SJL. - - Modifications on 20 October 95, by Daniel Cohen (catman@wpi.edu): - Modified to support different DMA, IRQ, and IO Ports. Borland's - Turbo Debugger in virtual 8086 mode (TD386.EXE with hardware breakpoints - provided by the TDH386.SYS Device Driver) was used on the CMS program - TAPE V4.0.5. I set breakpoints on I/O to ports 0x180-0x187. Note that - CMS's program will not successfully configure the tape drive if you set - breakpoints on IO Reads, but you can set them on IO Writes without problems. - Known problems: - - You can not use DMA Channels 5 or 7. - - Modification on 29 January 96, by Daniel Cohen (catman@wpi.edu): - Modified to only accept IRQs 3 - 7, or 9. Since we can only send a 3 bit - number representing the IRQ to the card, special handling is required when - IRQ 9 is selected. IRQ 2 and 9 are the same, and we should request IRQ 9 - from the kernel while telling the card to use IRQ 2. Thanks to Greg - Crider (gcrider@iclnet.org) for finding and locating this bug, as well as - testing the patch. - - Modification on 11 December 96, by Claus Heine (claus@momo.math.rwth-aachen.de): - Modified a little to use variahle ft_fdc_base, ft_fdc_irq, ft_fdc_dma - instead of preprocessor symbols. Thus we can compile this into the module - or kernel and let the user specify the options as command line arguments. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/fc-10.c,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:04 $ - * - * This file contains code for the CMS FC-10/FC-20 card. - */ - -#include -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/fdc-io.h" -#include "../lowlevel/fc-10.h" - -static __u16 inbs_magic[] = { - 0x3, 0x3, 0x0, 0x4, 0x7, 0x2, 0x5, 0x3, 0x1, 0x4, - 0x3, 0x5, 0x2, 0x0, 0x3, 0x7, 0x4, 0x2, - 0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7 -}; - -static __u16 fc10_ports[] = { - 0x180, 0x210, 0x2A0, 0x300, 0x330, 0x340, 0x370 -}; - -int fc10_enable(void) -{ - int i; - __u8 cardConfig = 0x00; - __u8 x; - TRACE_FUN(ft_t_flow); - -/* This code will only work if the FC-10 (or FC-20) is set to - * use DMA channels 1, 2, or 3. DMA channels 5 and 7 seem to be - * initialized by the same command as channels 1 and 3, respectively. - */ - if (ft_fdc_dma > 3) { - TRACE_ABORT(0, ft_t_err, -"Error: The FC-10/20 must be set to use DMA channels 1, 2, or 3!"); - } -/* Only allow the FC-10/20 to use IRQ 3-7, or 9. Note that CMS's program - * only accepts IRQ's 2-7, but in linux, IRQ 2 is the same as IRQ 9. - */ - if (ft_fdc_irq < 3 || ft_fdc_irq == 8 || ft_fdc_irq > 9) { - TRACE_ABORT(0, ft_t_err, -"Error: The FC-10/20 must be set to use IRQ levels 3 - 7, or 9!\n" -KERN_INFO "Note: IRQ 9 is the same as IRQ 2"); - } - /* Clear state machine ??? - */ - for (i = 0; i < NR_ITEMS(inbs_magic); i++) { - inb(ft_fdc_base + inbs_magic[i]); - } - outb(0x0, ft_fdc_base); - - x = inb(ft_fdc_base); - if (x == 0x13 || x == 0x93) { - for (i = 1; i < 8; i++) { - if (inb(ft_fdc_base + i) != x) { - TRACE_EXIT 0; - } - } - } else { - TRACE_EXIT 0; - } - - outb(0x8, ft_fdc_base); - - for (i = 0; i < 8; i++) { - if (inb(ft_fdc_base + i) != 0x0) { - TRACE_EXIT 0; - } - } - outb(0x10, ft_fdc_base); - - for (i = 0; i < 8; i++) { - if (inb(ft_fdc_base + i) != 0xff) { - TRACE_EXIT 0; - } - } - - /* Okay, we found a FC-10 card ! ??? - */ - outb(0x0, fdc.ccr); - - /* Clear state machine again ??? - */ - for (i = 0; i < NR_ITEMS(inbs_magic); i++) { - inb(ft_fdc_base + inbs_magic[i]); - } - /* Send io port */ - for (i = 0; i < NR_ITEMS(fc10_ports); i++) - if (ft_fdc_base == fc10_ports[i]) - cardConfig = i + 1; - if (cardConfig == 0) { - TRACE_EXIT 0; /* Invalid I/O Port */ - } - /* and IRQ - If using IRQ 9, tell the FC card it is actually IRQ 2 */ - if (ft_fdc_irq != 9) - cardConfig |= ft_fdc_irq << 3; - else - cardConfig |= 2 << 3; - - /* and finally DMA Channel */ - cardConfig |= ft_fdc_dma << 6; - outb(cardConfig, ft_fdc_base); /* DMA [2 bits]/IRQ [3 bits]/BASE [3 bits] */ - - /* Enable FC-10 ??? - */ - outb(0, fdc.ccr); - outb(0, fdc.dor2); - outb(FDC_DMA_MODE /* 8 */, fdc.dor); - outb(FDC_DMA_MODE /* 8 */, fdc.dor); - outb(1, fdc.dor2); - - /************************************* - * - * cH: why the hell should this be necessary? This is done - * by fdc_reset()!!! - * - *************************************/ - /* Initialize fdc, select drive B: - */ - outb(FDC_DMA_MODE, fdc.dor); /* assert reset, dma & irq enabled */ - /* 0x08 */ - outb(FDC_DMA_MODE|FDC_RESET_NOT, fdc.dor); /* release reset */ - /* 0x08 | 0x04 = 0x0c */ - outb(FDC_DMA_MODE|FDC_RESET_NOT|FDC_MOTOR_1|FTAPE_SEL_B, fdc.dor); - /* 0x08 | 0x04 | 0x20 | 0x01 = 0x2d */ - /* select drive 1 */ /* why not drive 0 ???? */ - TRACE_EXIT (x == 0x93) ? 2 : 1; -} diff --git a/drivers/char/ftape/lowlevel/fc-10.h b/drivers/char/ftape/lowlevel/fc-10.h deleted file mode 100644 index da7b88bca88..00000000000 --- a/drivers/char/ftape/lowlevel/fc-10.h +++ /dev/null @@ -1,39 +0,0 @@ -#ifndef _FC_10_H -#define _FC_10_H - -/* - * Copyright (C) 1994-1996 Bas Laarhoven. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/fc-10.h,v $ - * $Revision: 1.1 $ - * $Date: 1997/09/19 09:05:22 $ - * - * This file contains definitions for the FC-10 code - * of the QIC-40/80 floppy-tape driver for Linux. - */ - -/* - * fc-10.c defined global vars. - */ - -/* - * fc-10.c defined global functions. - */ -extern int fc10_enable(void); - -#endif diff --git a/drivers/char/ftape/lowlevel/fdc-io.c b/drivers/char/ftape/lowlevel/fdc-io.c deleted file mode 100644 index bbcf918f056..00000000000 --- a/drivers/char/ftape/lowlevel/fdc-io.c +++ /dev/null @@ -1,1349 +0,0 @@ -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/fdc-io.c,v $ - * $Revision: 1.7.4.2 $ - * $Date: 1997/11/16 14:48:17 $ - * - * This file contains the low-level floppy disk interface code - * for the QIC-40/80/3010/3020 floppy-tape driver "ftape" for - * Linux. - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/fdc-io.h" -#include "../lowlevel/fdc-isr.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-calibr.h" -#include "../lowlevel/fc-10.h" - -/* Global vars. - */ -static int ftape_motor; -volatile int ftape_current_cylinder = -1; -volatile fdc_mode_enum fdc_mode = fdc_idle; -fdc_config_info fdc; -DECLARE_WAIT_QUEUE_HEAD(ftape_wait_intr); - -unsigned int ft_fdc_base = CONFIG_FT_FDC_BASE; -unsigned int ft_fdc_irq = CONFIG_FT_FDC_IRQ; -unsigned int ft_fdc_dma = CONFIG_FT_FDC_DMA; -unsigned int ft_fdc_threshold = CONFIG_FT_FDC_THR; /* bytes */ -unsigned int ft_fdc_rate_limit = CONFIG_FT_FDC_MAX_RATE; /* bits/sec */ -int ft_probe_fc10 = CONFIG_FT_PROBE_FC10; -int ft_mach2 = CONFIG_FT_MACH2; - -/* Local vars. - */ -static spinlock_t fdc_io_lock; -static unsigned int fdc_calibr_count; -static unsigned int fdc_calibr_time; -static int fdc_status; -volatile __u8 fdc_head; /* FDC head from sector id */ -volatile __u8 fdc_cyl; /* FDC track from sector id */ -volatile __u8 fdc_sect; /* FDC sector from sector id */ -static int fdc_data_rate = 500; /* data rate (Kbps) */ -static int fdc_rate_code; /* data rate code (0 == 500 Kbps) */ -static int fdc_seek_rate = 2; /* step rate (msec) */ -static void (*do_ftape) (void); -static int fdc_fifo_state; /* original fifo setting - fifo enabled */ -static int fdc_fifo_thr; /* original fifo setting - threshold */ -static int fdc_lock_state; /* original lock setting - locked */ -static int fdc_fifo_locked; /* has fifo && lock set ? */ -static __u8 fdc_precomp; /* default precomp. value (nsec) */ -static __u8 fdc_prec_code; /* fdc precomp. select code */ - -static char ftape_id[] = "ftape"; /* used by request irq and free irq */ - -static int fdc_set_seek_rate(int seek_rate); - -void fdc_catch_stray_interrupts(int count) -{ - unsigned long flags; - - spin_lock_irqsave(&fdc_io_lock, flags); - if (count == 0) { - ft_expected_stray_interrupts = 0; - } else { - ft_expected_stray_interrupts += count; - } - spin_unlock_irqrestore(&fdc_io_lock, flags); -} - -/* Wait during a timeout period for a given FDC status. - * If usecs == 0 then just test status, else wait at least for usecs. - * Returns -ETIME on timeout. Function must be calibrated first ! - */ -static int fdc_wait(unsigned int usecs, __u8 mask, __u8 state) -{ - int count_1 = (fdc_calibr_count * usecs + - fdc_calibr_count - 1) / fdc_calibr_time; - - do { - fdc_status = inb_p(fdc.msr); - if ((fdc_status & mask) == state) { - return 0; - } - } while (count_1-- >= 0); - return -ETIME; -} - -int fdc_ready_wait(unsigned int usecs) -{ - return fdc_wait(usecs, FDC_DATA_READY | FDC_BUSY, FDC_DATA_READY); -} - -/* Why can't we just use udelay()? - */ -static void fdc_usec_wait(unsigned int usecs) -{ - fdc_wait(usecs, 0, 1); /* will always timeout ! */ -} - -static int fdc_ready_out_wait(unsigned int usecs) -{ - fdc_usec_wait(FT_RQM_DELAY); /* wait for valid RQM status */ - return fdc_wait(usecs, FDC_DATA_OUT_READY, FDC_DATA_OUT_READY); -} - -void fdc_wait_calibrate(void) -{ - ftape_calibrate("fdc_wait", - fdc_usec_wait, &fdc_calibr_count, &fdc_calibr_time); -} - -/* Wait for a (short) while for the FDC to become ready - * and transfer the next command byte. - * Return -ETIME on timeout on getting ready (depends on hardware!). - */ -static int fdc_write(const __u8 data) -{ - fdc_usec_wait(FT_RQM_DELAY); /* wait for valid RQM status */ - if (fdc_wait(150, FDC_DATA_READY_MASK, FDC_DATA_IN_READY) < 0) { - return -ETIME; - } else { - outb(data, fdc.fifo); - return 0; - } -} - -/* Wait for a (short) while for the FDC to become ready - * and transfer the next result byte. - * Return -ETIME if timeout on getting ready (depends on hardware!). - */ -static int fdc_read(__u8 * data) -{ - fdc_usec_wait(FT_RQM_DELAY); /* wait for valid RQM status */ - if (fdc_wait(150, FDC_DATA_READY_MASK, FDC_DATA_OUT_READY) < 0) { - return -ETIME; - } else { - *data = inb(fdc.fifo); - return 0; - } -} - -/* Output a cmd_len long command string to the FDC. - * The FDC should be ready to receive a new command or - * an error (EBUSY or ETIME) will occur. - */ -int fdc_command(const __u8 * cmd_data, int cmd_len) -{ - int result = 0; - unsigned long flags; - int count = cmd_len; - int retry = 0; -#ifdef TESTING - static unsigned int last_time; - unsigned int time; -#endif - TRACE_FUN(ft_t_any); - - fdc_usec_wait(FT_RQM_DELAY); /* wait for valid RQM status */ - spin_lock_irqsave(&fdc_io_lock, flags); - if (!in_interrupt()) - /* Yes, I know, too much comments inside this function - * ... - * - * Yet another bug in the original driver. All that - * havoc is caused by the fact that the isr() sends - * itself a command to the floppy tape driver (pause, - * micro step pause). Now, the problem is that - * commands are transmitted via the fdc_seek - * command. But: the fdc performs seeks in the - * background i.e. it doesn't signal busy while - * sending the step pulses to the drive. Therefore the - * non-interrupt level driver has no chance to tell - * whether the isr() just has issued a seek. Therefore - * we HAVE TO have a look at the ft_hide_interrupt - * flag: it signals the non-interrupt level part of - * the driver that it has to wait for the fdc until it - * has completet seeking. - * - * THIS WAS PRESUMABLY THE REASON FOR ALL THAT - * "fdc_read timeout" errors, I HOPE :-) - */ - if (ft_hide_interrupt) { - restore_flags(flags); - TRACE(ft_t_info, - "Waiting for the isr() completing fdc_seek()"); - if (fdc_interrupt_wait(2 * FT_SECOND) < 0) { - TRACE(ft_t_warn, - "Warning: timeout waiting for isr() seek to complete"); - } - if (ft_hide_interrupt || !ft_seek_completed) { - /* There cannot be another - * interrupt. The isr() only stops - * the tape and the next interrupt - * won't come until we have send our - * command to the drive. - */ - TRACE_ABORT(-EIO, ft_t_bug, - "BUG? isr() is still seeking?\n" - KERN_INFO "hide: %d\n" - KERN_INFO "seek: %d", - ft_hide_interrupt, - ft_seek_completed); - - } - fdc_usec_wait(FT_RQM_DELAY); /* wait for valid RQM status */ - spin_lock_irqsave(&fdc_io_lock, flags); - } - fdc_status = inb(fdc.msr); - if ((fdc_status & FDC_DATA_READY_MASK) != FDC_DATA_IN_READY) { - spin_unlock_irqrestore(&fdc_io_lock, flags); - TRACE_ABORT(-EBUSY, ft_t_err, "fdc not ready"); - } - fdc_mode = *cmd_data; /* used by isr */ -#ifdef TESTING - if (fdc_mode == FDC_SEEK) { - time = ftape_timediff(last_time, ftape_timestamp()); - if (time < 6000) { - TRACE(ft_t_bug,"Warning: short timeout between seek commands: %d", - time); - } - } -#endif - if (!in_interrupt()) { - /* shouldn't be cleared if called from isr - */ - ft_interrupt_seen = 0; - } - while (count) { - result = fdc_write(*cmd_data); - if (result < 0) { - TRACE(ft_t_fdc_dma, - "fdc_mode = %02x, status = %02x at index %d", - (int) fdc_mode, (int) fdc_status, - cmd_len - count); - if (++retry <= 3) { - TRACE(ft_t_warn, "fdc_write timeout, retry"); - } else { - TRACE(ft_t_err, "fdc_write timeout, fatal"); - /* recover ??? */ - break; - } - } else { - --count; - ++cmd_data; - } - } -#ifdef TESTING - if (fdc_mode == FDC_SEEK) { - last_time = ftape_timestamp(); - } -#endif - spin_unlock_irqrestore(&fdc_io_lock, flags); - TRACE_EXIT result; -} - -/* Input a res_len long result string from the FDC. - * The FDC should be ready to send the result or an error - * (EBUSY or ETIME) will occur. - */ -int fdc_result(__u8 * res_data, int res_len) -{ - int result = 0; - unsigned long flags; - int count = res_len; - int retry = 0; - TRACE_FUN(ft_t_any); - - spin_lock_irqsave(&fdc_io_lock, flags); - fdc_status = inb(fdc.msr); - if ((fdc_status & FDC_DATA_READY_MASK) != FDC_DATA_OUT_READY) { - TRACE(ft_t_err, "fdc not ready"); - result = -EBUSY; - } else while (count) { - if (!(fdc_status & FDC_BUSY)) { - spin_unlock_irqrestore(&fdc_io_lock, flags); - TRACE_ABORT(-EIO, ft_t_err, "premature end of result phase"); - } - result = fdc_read(res_data); - if (result < 0) { - TRACE(ft_t_fdc_dma, - "fdc_mode = %02x, status = %02x at index %d", - (int) fdc_mode, - (int) fdc_status, - res_len - count); - if (++retry <= 3) { - TRACE(ft_t_warn, "fdc_read timeout, retry"); - } else { - TRACE(ft_t_err, "fdc_read timeout, fatal"); - /* recover ??? */ - break; - ++retry; - } - } else { - --count; - ++res_data; - } - } - spin_unlock_irqrestore(&fdc_io_lock, flags); - fdc_usec_wait(FT_RQM_DELAY); /* allow FDC to negate BSY */ - TRACE_EXIT result; -} - -/* Handle command and result phases for - * commands without data phase. - */ -static int fdc_issue_command(const __u8 * out_data, int out_count, - __u8 * in_data, int in_count) -{ - TRACE_FUN(ft_t_any); - - if (out_count > 0) { - TRACE_CATCH(fdc_command(out_data, out_count),); - } - /* will take 24 - 30 usec for fdc_sense_drive_status and - * fdc_sense_interrupt_status commands. - * 35 fails sometimes (5/9/93 SJL) - * On a loaded system it incidentally takes longer than - * this for the fdc to get ready ! ?????? WHY ?????? - * So until we know what's going on use a very long timeout. - */ - TRACE_CATCH(fdc_ready_out_wait(500 /* usec */),); - if (in_count > 0) { - TRACE_CATCH(fdc_result(in_data, in_count), - TRACE(ft_t_err, "result phase aborted")); - } - TRACE_EXIT 0; -} - -/* Wait for FDC interrupt with timeout (in milliseconds). - * Signals are blocked so the wait will not be aborted. - * Note: interrupts must be enabled ! (23/05/93 SJL) - */ -int fdc_interrupt_wait(unsigned int time) -{ - DECLARE_WAITQUEUE(wait,current); - sigset_t old_sigmask; - static int resetting; - long timeout; - - TRACE_FUN(ft_t_fdc_dma); - - if (waitqueue_active(&ftape_wait_intr)) { - TRACE_ABORT(-EIO, ft_t_err, "error: nested call"); - } - /* timeout time will be up to USPT microseconds too long ! */ - timeout = (1000 * time + FT_USPT - 1) / FT_USPT; - - spin_lock_irq(¤t->sighand->siglock); - old_sigmask = current->blocked; - sigfillset(¤t->blocked); - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - set_current_state(TASK_INTERRUPTIBLE); - add_wait_queue(&ftape_wait_intr, &wait); - while (!ft_interrupt_seen && timeout) - timeout = schedule_timeout_interruptible(timeout); - - spin_lock_irq(¤t->sighand->siglock); - current->blocked = old_sigmask; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - remove_wait_queue(&ftape_wait_intr, &wait); - /* the following IS necessary. True: as well - * wake_up_interruptible() as the schedule() set TASK_RUNNING - * when they wakeup a task, BUT: it may very well be that - * ft_interrupt_seen is already set to 1 when we enter here - * in which case schedule() gets never called, and - * TASK_RUNNING never set. This has the funny effect that we - * execute all the code until we leave kernel space, but then - * the task is stopped (a task CANNOT be preempted while in - * kernel mode. Sending a pair of SIGSTOP/SIGCONT to the - * tasks wakes it up again. Funny! :-) - */ - current->state = TASK_RUNNING; - if (ft_interrupt_seen) { /* woken up by interrupt */ - ft_interrupt_seen = 0; - TRACE_EXIT 0; - } - /* Original comment: - * In first instance, next statement seems unnecessary since - * it will be cleared in fdc_command. However, a small part of - * the software seems to rely on this being cleared here - * (ftape_close might fail) so stick to it until things get fixed ! - */ - /* My deeply sought of knowledge: - * Behold NO! It is obvious. fdc_reset() doesn't call fdc_command() - * but nevertheless uses fdc_interrupt_wait(). OF COURSE this needs to - * be reset here. - */ - ft_interrupt_seen = 0; /* clear for next call */ - if (!resetting) { - resetting = 1; /* break infinite recursion if reset fails */ - TRACE(ft_t_any, "cleanup reset"); - fdc_reset(); - resetting = 0; - } - TRACE_EXIT (signal_pending(current)) ? -EINTR : -ETIME; -} - -/* Start/stop drive motor. Enable DMA mode. - */ -void fdc_motor(int motor) -{ - int unit = ft_drive_sel; - int data = unit | FDC_RESET_NOT | FDC_DMA_MODE; - TRACE_FUN(ft_t_any); - - ftape_motor = motor; - if (ftape_motor) { - data |= FDC_MOTOR_0 << unit; - TRACE(ft_t_noise, "turning motor %d on", unit); - } else { - TRACE(ft_t_noise, "turning motor %d off", unit); - } - if (ft_mach2) { - outb_p(data, fdc.dor2); - } else { - outb_p(data, fdc.dor); - } - ftape_sleep(10 * FT_MILLISECOND); - TRACE_EXIT; -} - -static void fdc_update_dsr(void) -{ - TRACE_FUN(ft_t_any); - - TRACE(ft_t_flow, "rate = %d Kbps, precomp = %d ns", - fdc_data_rate, fdc_precomp); - if (fdc.type >= i82077) { - outb_p((fdc_rate_code & 0x03) | fdc_prec_code, fdc.dsr); - } else { - outb_p(fdc_rate_code & 0x03, fdc.ccr); - } - TRACE_EXIT; -} - -void fdc_set_write_precomp(int precomp) -{ - TRACE_FUN(ft_t_any); - - TRACE(ft_t_noise, "New precomp: %d nsec", precomp); - fdc_precomp = precomp; - /* write precompensation can be set in multiples of 41.67 nsec. - * round the parameter to the nearest multiple and convert it - * into a fdc setting. Note that 0 means default to the fdc, - * 7 is used instead of that. - */ - fdc_prec_code = ((fdc_precomp + 21) / 42) << 2; - if (fdc_prec_code == 0 || fdc_prec_code > (6 << 2)) { - fdc_prec_code = 7 << 2; - } - fdc_update_dsr(); - TRACE_EXIT; -} - -/* Reprogram the 82078 registers to use Data Rate Table 1 on all drives. - */ -static void fdc_set_drive_specs(void) -{ - __u8 cmd[] = { FDC_DRIVE_SPEC, 0x00, 0x00, 0x00, 0x00, 0xc0}; - int result; - TRACE_FUN(ft_t_any); - - TRACE(ft_t_flow, "Setting of drive specs called"); - if (fdc.type >= i82078_1) { - cmd[1] = (0 << 5) | (2 << 2); - cmd[2] = (1 << 5) | (2 << 2); - cmd[3] = (2 << 5) | (2 << 2); - cmd[4] = (3 << 5) | (2 << 2); - result = fdc_command(cmd, NR_ITEMS(cmd)); - if (result < 0) { - TRACE(ft_t_err, "Setting of drive specs failed"); - } - } - TRACE_EXIT; -} - -/* Select clock for fdc, must correspond with tape drive setting ! - * This also influences the fdc timing so we must adjust some values. - */ -int fdc_set_data_rate(int rate) -{ - int bad_rate = 0; - TRACE_FUN(ft_t_any); - - /* Select clock for fdc, must correspond with tape drive setting ! - * This also influences the fdc timing so we must adjust some values. - */ - TRACE(ft_t_fdc_dma, "new rate = %d", rate); - switch (rate) { - case 250: - fdc_rate_code = fdc_data_rate_250; - break; - case 500: - fdc_rate_code = fdc_data_rate_500; - break; - case 1000: - if (fdc.type < i82077) { - bad_rate = 1; - } else { - fdc_rate_code = fdc_data_rate_1000; - } - break; - case 2000: - if (fdc.type < i82078_1) { - bad_rate = 1; - } else { - fdc_rate_code = fdc_data_rate_2000; - } - break; - default: - bad_rate = 1; - } - if (bad_rate) { - TRACE_ABORT(-EIO, - ft_t_fdc_dma, "%d is not a valid data rate", rate); - } - fdc_data_rate = rate; - fdc_update_dsr(); - fdc_set_seek_rate(fdc_seek_rate); /* clock changed! */ - ftape_udelay(1000); - TRACE_EXIT 0; -} - -/* keep the unit select if keep_select is != 0, - */ -static void fdc_dor_reset(int keep_select) -{ - __u8 fdc_ctl = ft_drive_sel; - - if (keep_select != 0) { - fdc_ctl |= FDC_DMA_MODE; - if (ftape_motor) { - fdc_ctl |= FDC_MOTOR_0 << ft_drive_sel; - } - } - ftape_udelay(10); /* ??? but seems to be necessary */ - if (ft_mach2) { - outb_p(fdc_ctl & 0x0f, fdc.dor); - outb_p(fdc_ctl, fdc.dor2); - } else { - outb_p(fdc_ctl, fdc.dor); - } - fdc_usec_wait(10); /* delay >= 14 fdc clocks */ - if (keep_select == 0) { - fdc_ctl = 0; - } - fdc_ctl |= FDC_RESET_NOT; - if (ft_mach2) { - outb_p(fdc_ctl & 0x0f, fdc.dor); - outb_p(fdc_ctl, fdc.dor2); - } else { - outb_p(fdc_ctl, fdc.dor); - } -} - -/* Reset the floppy disk controller. Leave the ftape_unit selected. - */ -void fdc_reset(void) -{ - int st0; - int i; - int dummy; - unsigned long flags; - TRACE_FUN(ft_t_any); - - spin_lock_irqsave(&fdc_io_lock, flags); - - fdc_dor_reset(1); /* keep unit selected */ - - fdc_mode = fdc_idle; - - /* maybe the spin_lock_irq* pair is not necessary, BUT: - * the following line MUST be here. Otherwise fdc_interrupt_wait() - * won't wait. Note that fdc_reset() is called from - * ftape_dumb_stop() when the fdc is busy transferring data. In this - * case fdc_isr() MOST PROBABLY sets ft_interrupt_seen, and tries - * to get the result bytes from the fdc etc. CLASH. - */ - ft_interrupt_seen = 0; - - /* Program data rate - */ - fdc_update_dsr(); /* restore data rate and precomp */ - - spin_unlock_irqrestore(&fdc_io_lock, flags); - - /* - * Wait for first polling cycle to complete - */ - if (fdc_interrupt_wait(1 * FT_SECOND) < 0) { - TRACE(ft_t_err, "no drive polling interrupt!"); - } else { /* clear all disk-changed statuses */ - for (i = 0; i < 4; ++i) { - if(fdc_sense_interrupt_status(&st0, &dummy) != 0) { - TRACE(ft_t_err, "sense failed for %d", i); - } - if (i == ft_drive_sel) { - ftape_current_cylinder = dummy; - } - } - TRACE(ft_t_noise, "drive polling completed"); - } - /* - * SPECIFY COMMAND - */ - fdc_set_seek_rate(fdc_seek_rate); - /* - * DRIVE SPECIFICATION COMMAND (if fdc type known) - */ - if (fdc.type >= i82078_1) { - fdc_set_drive_specs(); - } - TRACE_EXIT; -} - -#if !defined(CLK_48MHZ) -# define CLK_48MHZ 1 -#endif - -/* When we're done, put the fdc into reset mode so that the regular - * floppy disk driver will figure out that something is wrong and - * initialize the controller the way it wants. - */ -void fdc_disable(void) -{ - __u8 cmd1[] = {FDC_CONFIGURE, 0x00, 0x00, 0x00}; - __u8 cmd2[] = {FDC_LOCK}; - __u8 cmd3[] = {FDC_UNLOCK}; - __u8 stat[1]; - TRACE_FUN(ft_t_flow); - - if (!fdc_fifo_locked) { - fdc_reset(); - TRACE_EXIT; - } - if (fdc_issue_command(cmd3, 1, stat, 1) < 0 || stat[0] != 0x00) { - fdc_dor_reset(0); - TRACE_ABORT(/**/, ft_t_bug, - "couldn't unlock fifo, configuration remains changed"); - } - fdc_fifo_locked = 0; - if (CLK_48MHZ && fdc.type >= i82078) { - cmd1[0] |= FDC_CLK48_BIT; - } - cmd1[2] = ((fdc_fifo_state) ? 0 : 0x20) + (fdc_fifo_thr - 1); - if (fdc_command(cmd1, NR_ITEMS(cmd1)) < 0) { - fdc_dor_reset(0); - TRACE_ABORT(/**/, ft_t_bug, - "couldn't reconfigure fifo to old state"); - } - if (fdc_lock_state && - fdc_issue_command(cmd2, 1, stat, 1) < 0) { - fdc_dor_reset(0); - TRACE_ABORT(/**/, ft_t_bug, "couldn't lock old state again"); - } - TRACE(ft_t_noise, "fifo restored: %sabled, thr. %d, %slocked", - fdc_fifo_state ? "en" : "dis", - fdc_fifo_thr, (fdc_lock_state) ? "" : "not "); - fdc_dor_reset(0); - TRACE_EXIT; -} - -/* Specify FDC seek-rate (milliseconds) - */ -static int fdc_set_seek_rate(int seek_rate) -{ - /* set step rate, dma mode, and minimal head load and unload times - */ - __u8 in[3] = { FDC_SPECIFY, 1, (1 << 1)}; - - fdc_seek_rate = seek_rate; - in[1] |= (16 - (fdc_data_rate * fdc_seek_rate) / 500) << 4; - - return fdc_command(in, 3); -} - -/* Sense drive status: get unit's drive status (ST3) - */ -int fdc_sense_drive_status(int *st3) -{ - __u8 out[2]; - __u8 in[1]; - TRACE_FUN(ft_t_any); - - out[0] = FDC_SENSED; - out[1] = ft_drive_sel; - TRACE_CATCH(fdc_issue_command(out, 2, in, 1),); - *st3 = in[0]; - TRACE_EXIT 0; -} - -/* Sense Interrupt Status command: - * should be issued at the end of each seek. - * get ST0 and current cylinder. - */ -int fdc_sense_interrupt_status(int *st0, int *current_cylinder) -{ - __u8 out[1]; - __u8 in[2]; - TRACE_FUN(ft_t_any); - - out[0] = FDC_SENSEI; - TRACE_CATCH(fdc_issue_command(out, 1, in, 2),); - *st0 = in[0]; - *current_cylinder = in[1]; - TRACE_EXIT 0; -} - -/* step to track - */ -int fdc_seek(int track) -{ - __u8 out[3]; - int st0, pcn; -#ifdef TESTING - unsigned int time; -#endif - TRACE_FUN(ft_t_any); - - out[0] = FDC_SEEK; - out[1] = ft_drive_sel; - out[2] = track; -#ifdef TESTING - time = ftape_timestamp(); -#endif - /* We really need this command to work ! - */ - ft_seek_completed = 0; - TRACE_CATCH(fdc_command(out, 3), - fdc_reset(); - TRACE(ft_t_noise, "destination was: %d, resetting FDC...", - track)); - /* Handle interrupts until ft_seek_completed or timeout. - */ - for (;;) { - TRACE_CATCH(fdc_interrupt_wait(2 * FT_SECOND),); - if (ft_seek_completed) { - TRACE_CATCH(fdc_sense_interrupt_status(&st0, &pcn),); - if ((st0 & ST0_SEEK_END) == 0) { - TRACE_ABORT(-EIO, ft_t_err, - "no seek-end after seek completion !??"); - } - break; - } - } -#ifdef TESTING - time = ftape_timediff(time, ftape_timestamp()) / abs(track - ftape_current_cylinder); - if ((time < 900 || time > 3100) && abs(track - ftape_current_cylinder) > 5) { - TRACE(ft_t_warn, "Wrong FDC STEP interval: %d usecs (%d)", - time, track - ftape_current_cylinder); - } -#endif - /* Verify whether we issued the right tape command. - */ - /* Verify that we seek to the proper track. */ - if (pcn != track) { - TRACE_ABORT(-EIO, ft_t_err, "bad seek.."); - } - ftape_current_cylinder = track; - TRACE_EXIT 0; -} - -static int perpend_mode; /* set if fdc is in perpendicular mode */ - -static int perpend_off(void) -{ - __u8 perpend[] = {FDC_PERPEND, 0x00}; - TRACE_FUN(ft_t_any); - - if (perpend_mode) { - /* Turn off perpendicular mode */ - perpend[1] = 0x80; - TRACE_CATCH(fdc_command(perpend, 2), - TRACE(ft_t_err,"Perpendicular mode exit failed!")); - perpend_mode = 0; - } - TRACE_EXIT 0; -} - -static int handle_perpend(int segment_id) -{ - __u8 perpend[] = {FDC_PERPEND, 0x00}; - TRACE_FUN(ft_t_any); - - /* When writing QIC-3020 tapes, turn on perpendicular mode - * if tape is moving in forward direction (even tracks). - */ - if (ft_qic_std == QIC_TAPE_QIC3020 && - ((segment_id / ft_segments_per_track) & 1) == 0) { -/* FIXME: some i82077 seem to support perpendicular mode as - * well. - */ -#if 0 - if (fdc.type < i82077AA) {} -#else - if (fdc.type < i82077 && ft_data_rate < 1000) { -#endif - /* fdc does not support perpendicular mode: complain - */ - TRACE_ABORT(-EIO, ft_t_err, - "Your FDC does not support QIC-3020."); - } - perpend[1] = 0x03 /* 0x83 + (0x4 << ft_drive_sel) */ ; - TRACE_CATCH(fdc_command(perpend, 2), - TRACE(ft_t_err,"Perpendicular mode entry failed!")); - TRACE(ft_t_flow, "Perpendicular mode set"); - perpend_mode = 1; - TRACE_EXIT 0; - } - TRACE_EXIT perpend_off(); -} - -static inline void fdc_setup_dma(char mode, - volatile void *addr, unsigned int count) -{ - /* Program the DMA controller. - */ - disable_dma(fdc.dma); - clear_dma_ff(fdc.dma); - set_dma_mode(fdc.dma, mode); - set_dma_addr(fdc.dma, virt_to_bus((void*)addr)); - set_dma_count(fdc.dma, count); - enable_dma(fdc.dma); -} - -/* Setup fdc and dma for formatting the next segment - */ -int fdc_setup_formatting(buffer_struct * buff) -{ - unsigned long flags; - __u8 out[6] = { - FDC_FORMAT, 0x00, 3, 4 * FT_SECTORS_PER_SEGMENT, 0x00, 0x6b - }; - TRACE_FUN(ft_t_any); - - TRACE_CATCH(handle_perpend(buff->segment_id),); - /* Program the DMA controller. - */ - TRACE(ft_t_fdc_dma, - "phys. addr. = %lx", virt_to_bus((void*) buff->ptr)); - spin_lock_irqsave(&fdc_io_lock, flags); - fdc_setup_dma(DMA_MODE_WRITE, buff->ptr, FT_SECTORS_PER_SEGMENT * 4); - /* Issue FDC command to start reading/writing. - */ - out[1] = ft_drive_sel; - out[4] = buff->gap3; - TRACE_CATCH(fdc_setup_error = fdc_command(out, sizeof(out)), - restore_flags(flags); fdc_mode = fdc_idle); - spin_unlock_irqrestore(&fdc_io_lock, flags); - TRACE_EXIT 0; -} - - -/* Setup Floppy Disk Controller and DMA to read or write the next cluster - * of good sectors from or to the current segment. - */ -int fdc_setup_read_write(buffer_struct * buff, __u8 operation) -{ - unsigned long flags; - __u8 out[9]; - int dma_mode; - TRACE_FUN(ft_t_any); - - switch(operation) { - case FDC_VERIFY: - if (fdc.type < i82077) { - operation = FDC_READ; - } - case FDC_READ: - case FDC_READ_DELETED: - dma_mode = DMA_MODE_READ; - TRACE(ft_t_fdc_dma, "xfer %d sectors to 0x%p", - buff->sector_count, buff->ptr); - TRACE_CATCH(perpend_off(),); - break; - case FDC_WRITE_DELETED: - TRACE(ft_t_noise, "deleting segment %d", buff->segment_id); - case FDC_WRITE: - dma_mode = DMA_MODE_WRITE; - /* When writing QIC-3020 tapes, turn on perpendicular mode - * if tape is moving in forward direction (even tracks). - */ - TRACE_CATCH(handle_perpend(buff->segment_id),); - TRACE(ft_t_fdc_dma, "xfer %d sectors from 0x%p", - buff->sector_count, buff->ptr); - break; - default: - TRACE_ABORT(-EIO, - ft_t_bug, "bug: invalid operation parameter"); - } - TRACE(ft_t_fdc_dma, "phys. addr. = %lx",virt_to_bus((void*)buff->ptr)); - spin_lock_irqsave(&fdc_io_lock, flags); - if (operation != FDC_VERIFY) { - fdc_setup_dma(dma_mode, buff->ptr, - FT_SECTOR_SIZE * buff->sector_count); - } - /* Issue FDC command to start reading/writing. - */ - out[0] = operation; - out[1] = ft_drive_sel; - out[2] = buff->cyl; - out[3] = buff->head; - out[4] = buff->sect + buff->sector_offset; - out[5] = 3; /* Sector size of 1K. */ - out[6] = out[4] + buff->sector_count - 1; /* last sector */ - out[7] = 109; /* Gap length. */ - out[8] = 0xff; /* No limit to transfer size. */ - TRACE(ft_t_fdc_dma, "C: 0x%02x, H: 0x%02x, R: 0x%02x, cnt: 0x%02x", - out[2], out[3], out[4], out[6] - out[4] + 1); - spin_unlock_irqrestore(&fdc_io_lock, flags); - TRACE_CATCH(fdc_setup_error = fdc_command(out, 9),fdc_mode = fdc_idle); - TRACE_EXIT 0; -} - -int fdc_fifo_threshold(__u8 threshold, - int *fifo_state, int *lock_state, int *fifo_thr) -{ - const __u8 cmd0[] = {FDC_DUMPREGS}; - __u8 cmd1[] = {FDC_CONFIGURE, 0, (0x0f & (threshold - 1)), 0}; - const __u8 cmd2[] = {FDC_LOCK}; - const __u8 cmd3[] = {FDC_UNLOCK}; - __u8 reg[10]; - __u8 stat; - int i; - int result; - TRACE_FUN(ft_t_any); - - if (CLK_48MHZ && fdc.type >= i82078) { - cmd1[0] |= FDC_CLK48_BIT; - } - /* Dump fdc internal registers for examination - */ - TRACE_CATCH(fdc_command(cmd0, NR_ITEMS(cmd0)), - TRACE(ft_t_warn, "dumpreg cmd failed, fifo unchanged")); - /* Now read fdc internal registers from fifo - */ - for (i = 0; i < (int)NR_ITEMS(reg); ++i) { - fdc_read(®[i]); - TRACE(ft_t_fdc_dma, "Register %d = 0x%02x", i, reg[i]); - } - if (fifo_state && lock_state && fifo_thr) { - *fifo_state = (reg[8] & 0x20) == 0; - *lock_state = reg[7] & 0x80; - *fifo_thr = 1 + (reg[8] & 0x0f); - } - TRACE(ft_t_noise, - "original fifo state: %sabled, threshold %d, %slocked", - ((reg[8] & 0x20) == 0) ? "en" : "dis", - 1 + (reg[8] & 0x0f), (reg[7] & 0x80) ? "" : "not "); - /* If fdc is already locked, unlock it first ! */ - if (reg[7] & 0x80) { - fdc_ready_wait(100); - TRACE_CATCH(fdc_issue_command(cmd3, NR_ITEMS(cmd3), &stat, 1), - TRACE(ft_t_bug, "FDC unlock command failed, " - "configuration unchanged")); - } - fdc_fifo_locked = 0; - /* Enable fifo and set threshold at xx bytes to allow a - * reasonably large latency and reduce number of dma bursts. - */ - fdc_ready_wait(100); - if ((result = fdc_command(cmd1, NR_ITEMS(cmd1))) < 0) { - TRACE(ft_t_bug, "configure cmd failed, fifo unchanged"); - } - /* Now lock configuration so reset will not change it - */ - if(fdc_issue_command(cmd2, NR_ITEMS(cmd2), &stat, 1) < 0 || - stat != 0x10) { - TRACE_ABORT(-EIO, ft_t_bug, - "FDC lock command failed, stat = 0x%02x", stat); - } - fdc_fifo_locked = 1; - TRACE_EXIT result; -} - -static int fdc_fifo_enable(void) -{ - TRACE_FUN(ft_t_any); - - if (fdc_fifo_locked) { - TRACE_ABORT(0, ft_t_warn, "Fifo not enabled because locked"); - } - TRACE_CATCH(fdc_fifo_threshold(ft_fdc_threshold /* bytes */, - &fdc_fifo_state, - &fdc_lock_state, - &fdc_fifo_thr),); - TRACE_CATCH(fdc_fifo_threshold(ft_fdc_threshold /* bytes */, - NULL, NULL, NULL),); - TRACE_EXIT 0; -} - -/* Determine fd controller type - */ -static __u8 fdc_save_state[2]; - -static int fdc_probe(void) -{ - __u8 cmd[1]; - __u8 stat[16]; /* must be able to hold dumpregs & save results */ - int i; - TRACE_FUN(ft_t_any); - - /* Try to find out what kind of fd controller we have to deal with - * Scheme borrowed from floppy driver: - * first try if FDC_DUMPREGS command works - * (this indicates that we have a 82072 or better) - * then try the FDC_VERSION command (82072 doesn't support this) - * then try the FDC_UNLOCK command (some older 82077's don't support this) - * then try the FDC_PARTID command (82078's support this) - */ - cmd[0] = FDC_DUMPREGS; - if (fdc_issue_command(cmd, 1, stat, 1) != 0) { - TRACE_ABORT(no_fdc, ft_t_bug, "No FDC found"); - } - if (stat[0] == 0x80) { - /* invalid command: must be pre 82072 */ - TRACE_ABORT(i8272, - ft_t_warn, "Type 8272A/765A compatible FDC found"); - } - fdc_result(&stat[1], 9); - fdc_save_state[0] = stat[7]; - fdc_save_state[1] = stat[8]; - cmd[0] = FDC_VERSION; - if (fdc_issue_command(cmd, 1, stat, 1) < 0 || stat[0] == 0x80) { - TRACE_ABORT(i8272, ft_t_warn, "Type 82072 FDC found"); - } - if (*stat != 0x90) { - TRACE_ABORT(i8272, ft_t_warn, "Unknown FDC found"); - } - cmd[0] = FDC_UNLOCK; - if(fdc_issue_command(cmd, 1, stat, 1) < 0 || stat[0] != 0x00) { - TRACE_ABORT(i8272, ft_t_warn, - "Type pre-1991 82077 FDC found, " - "treating it like a 82072"); - } - if (fdc_save_state[0] & 0x80) { /* was locked */ - cmd[0] = FDC_LOCK; /* restore lock */ - (void)fdc_issue_command(cmd, 1, stat, 1); - TRACE(ft_t_warn, "FDC is already locked"); - } - /* Test for a i82078 FDC */ - cmd[0] = FDC_PARTID; - if (fdc_issue_command(cmd, 1, stat, 1) < 0 || stat[0] == 0x80) { - /* invalid command: not a i82078xx type FDC */ - for (i = 0; i < 4; ++i) { - outb_p(i, fdc.tdr); - if ((inb_p(fdc.tdr) & 0x03) != i) { - TRACE_ABORT(i82077, - ft_t_warn, "Type 82077 FDC found"); - } - } - TRACE_ABORT(i82077AA, ft_t_warn, "Type 82077AA FDC found"); - } - /* FDC_PARTID cmd succeeded */ - switch (stat[0] >> 5) { - case 0x0: - /* i82078SL or i82078-1. The SL part cannot run at - * 2Mbps (the SL and -1 dies are identical; they are - * speed graded after production, according to Intel). - * Some SL's can be detected by doing a SAVE cmd and - * look at bit 7 of the first byte (the SEL3V# bit). - * If it is 0, the part runs off 3Volts, and hence it - * is a SL. - */ - cmd[0] = FDC_SAVE; - if(fdc_issue_command(cmd, 1, stat, 16) < 0) { - TRACE(ft_t_err, "FDC_SAVE failed. Dunno why"); - /* guess we better claim the fdc to be a i82078 */ - TRACE_ABORT(i82078, - ft_t_warn, - "Type i82078 FDC (i suppose) found"); - } - if ((stat[0] & FDC_SEL3V_BIT)) { - /* fdc running off 5Volts; Pray that it's a i82078-1 - */ - TRACE_ABORT(i82078_1, ft_t_warn, - "Type i82078-1 or 5Volt i82078SL FDC found"); - } - TRACE_ABORT(i82078, ft_t_warn, - "Type 3Volt i82078SL FDC (1Mbps) found"); - case 0x1: - case 0x2: /* S82078B */ - /* The '78B isn't '78 compatible. Detect it as a '77AA */ - TRACE_ABORT(i82077AA, ft_t_warn, "Type i82077AA FDC found"); - case 0x3: /* NSC PC8744 core; used in several super-IO chips */ - TRACE_ABORT(i82077AA, - ft_t_warn, "Type 82077AA compatible FDC found"); - default: - TRACE(ft_t_warn, "A previously undetected FDC found"); - TRACE_ABORT(i82077AA, ft_t_warn, - "Treating it as a 82077AA. Please report partid= %d", - stat[0]); - } /* switch(stat[ 0] >> 5) */ - TRACE_EXIT no_fdc; -} - -static int fdc_request_regions(void) -{ - TRACE_FUN(ft_t_flow); - - if (ft_mach2 || ft_probe_fc10) { - if (!request_region(fdc.sra, 8, "fdc (ft)")) { -#ifndef BROKEN_FLOPPY_DRIVER - TRACE_EXIT -EBUSY; -#else - TRACE(ft_t_warn, -"address 0x%03x occupied (by floppy driver?), using it anyway", fdc.sra); -#endif - } - } else { - if (!request_region(fdc.sra, 6, "fdc (ft)")) { -#ifndef BROKEN_FLOPPY_DRIVER - TRACE_EXIT -EBUSY; -#else - TRACE(ft_t_warn, -"address 0x%03x occupied (by floppy driver?), using it anyway", fdc.sra); -#endif - } - if (!request_region(fdc.sra + 7, 1, "fdc (ft)")) { -#ifndef BROKEN_FLOPPY_DRIVER - release_region(fdc.sra, 6); - TRACE_EXIT -EBUSY; -#else - TRACE(ft_t_warn, -"address 0x%03x occupied (by floppy driver?), using it anyway", fdc.sra + 7); -#endif - } - } - TRACE_EXIT 0; -} - -void fdc_release_regions(void) -{ - TRACE_FUN(ft_t_flow); - - if (fdc.sra != 0) { - if (fdc.dor2 != 0) { - release_region(fdc.sra, 8); - } else { - release_region(fdc.sra, 6); - release_region(fdc.dir, 1); - } - } - TRACE_EXIT; -} - -static int fdc_config_regs(unsigned int fdc_base, - unsigned int fdc_irq, - unsigned int fdc_dma) -{ - TRACE_FUN(ft_t_flow); - - fdc.irq = fdc_irq; - fdc.dma = fdc_dma; - fdc.sra = fdc_base; - fdc.srb = fdc_base + 1; - fdc.dor = fdc_base + 2; - fdc.tdr = fdc_base + 3; - fdc.msr = fdc.dsr = fdc_base + 4; - fdc.fifo = fdc_base + 5; - fdc.dir = fdc.ccr = fdc_base + 7; - fdc.dor2 = (ft_mach2 || ft_probe_fc10) ? fdc_base + 6 : 0; - TRACE_CATCH(fdc_request_regions(), fdc.sra = 0); - TRACE_EXIT 0; -} - -static int fdc_config(void) -{ - static int already_done; - TRACE_FUN(ft_t_any); - - if (already_done) { - TRACE_CATCH(fdc_request_regions(),); - *(fdc.hook) = fdc_isr; /* hook our handler in */ - TRACE_EXIT 0; - } - if (ft_probe_fc10) { - int fc_type; - - TRACE_CATCH(fdc_config_regs(ft_fdc_base, - ft_fdc_irq, ft_fdc_dma),); - fc_type = fc10_enable(); - if (fc_type != 0) { - TRACE(ft_t_warn, "FC-%c0 controller found", '0' + fc_type); - fdc.type = fc10; - fdc.hook = &do_ftape; - *(fdc.hook) = fdc_isr; /* hook our handler in */ - already_done = 1; - TRACE_EXIT 0; - } else { - TRACE(ft_t_warn, "FC-10/20 controller not found"); - fdc_release_regions(); - fdc.type = no_fdc; - ft_probe_fc10 = 0; - ft_fdc_base = 0x3f0; - ft_fdc_irq = 6; - ft_fdc_dma = 2; - } - } - TRACE(ft_t_warn, "fdc base: 0x%x, irq: %d, dma: %d", - ft_fdc_base, ft_fdc_irq, ft_fdc_dma); - TRACE_CATCH(fdc_config_regs(ft_fdc_base, ft_fdc_irq, ft_fdc_dma),); - fdc.hook = &do_ftape; - *(fdc.hook) = fdc_isr; /* hook our handler in */ - already_done = 1; - TRACE_EXIT 0; -} - -static irqreturn_t ftape_interrupt(int irq, void *dev_id) -{ - void (*handler) (void) = *fdc.hook; - int handled = 0; - TRACE_FUN(ft_t_any); - - *fdc.hook = NULL; - if (handler) { - handled = 1; - handler(); - } else { - TRACE(ft_t_bug, "Unexpected ftape interrupt"); - } - TRACE_EXIT IRQ_RETVAL(handled); -} - -static int fdc_grab_irq_and_dma(void) -{ - TRACE_FUN(ft_t_any); - - if (fdc.hook == &do_ftape) { - /* Get fast interrupt handler. - */ - if (request_irq(fdc.irq, ftape_interrupt, - IRQF_DISABLED, "ft", ftape_id)) { - TRACE_ABORT(-EIO, ft_t_bug, - "Unable to grab IRQ%d for ftape driver", - fdc.irq); - } - if (request_dma(fdc.dma, ftape_id)) { - free_irq(fdc.irq, ftape_id); - TRACE_ABORT(-EIO, ft_t_bug, - "Unable to grab DMA%d for ftape driver", - fdc.dma); - } - } - if (ft_fdc_base != 0x3f0 && (ft_fdc_dma == 2 || ft_fdc_irq == 6)) { - /* Using same dma channel or irq as standard fdc, need - * to disable the dma-gate on the std fdc. This - * couldn't be done in the floppy driver as some - * laptops are using the dma-gate to enter a low power - * or even suspended state :-( - */ - outb_p(FDC_RESET_NOT, 0x3f2); - TRACE(ft_t_noise, "DMA-gate on standard fdc disabled"); - } - TRACE_EXIT 0; -} - -int fdc_release_irq_and_dma(void) -{ - TRACE_FUN(ft_t_any); - - if (fdc.hook == &do_ftape) { - disable_dma(fdc.dma); /* just in case... */ - free_dma(fdc.dma); - free_irq(fdc.irq, ftape_id); - } - if (ft_fdc_base != 0x3f0 && (ft_fdc_dma == 2 || ft_fdc_irq == 6)) { - /* Using same dma channel as standard fdc, need to - * disable the dma-gate on the std fdc. This couldn't - * be done in the floppy driver as some laptops are - * using the dma-gate to enter a low power or even - * suspended state :-( - */ - outb_p(FDC_RESET_NOT | FDC_DMA_MODE, 0x3f2); - TRACE(ft_t_noise, "DMA-gate on standard fdc enabled again"); - } - TRACE_EXIT 0; -} - -int fdc_init(void) -{ - TRACE_FUN(ft_t_any); - - /* find a FDC to use */ - TRACE_CATCH(fdc_config(),); - TRACE_CATCH(fdc_grab_irq_and_dma(), fdc_release_regions()); - ftape_motor = 0; - fdc_catch_stray_interrupts(0); /* clear number of awainted - * stray interrupte - */ - fdc_catch_stray_interrupts(1); /* one always comes (?) */ - TRACE(ft_t_flow, "resetting fdc"); - fdc_set_seek_rate(2); /* use nominal QIC step rate */ - fdc_reset(); /* init fdc & clear track counters */ - if (fdc.type == no_fdc) { /* no FC-10 or FC-20 found */ - fdc.type = fdc_probe(); - fdc_reset(); /* update with new knowledge */ - } - if (fdc.type == no_fdc) { - fdc_release_irq_and_dma(); - fdc_release_regions(); - TRACE_EXIT -ENXIO; - } - if (fdc.type >= i82077) { - if (fdc_fifo_enable() < 0) { - TRACE(ft_t_warn, "couldn't enable fdc fifo !"); - } else { - TRACE(ft_t_flow, "fdc fifo enabled and locked"); - } - } - TRACE_EXIT 0; -} diff --git a/drivers/char/ftape/lowlevel/fdc-io.h b/drivers/char/ftape/lowlevel/fdc-io.h deleted file mode 100644 index 7ec3c72178b..00000000000 --- a/drivers/char/ftape/lowlevel/fdc-io.h +++ /dev/null @@ -1,252 +0,0 @@ -#ifndef _FDC_IO_H -#define _FDC_IO_H - -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/fdc-io.h,v $ - * $Revision: 1.3 $ - * $Date: 1997/10/05 19:18:06 $ - * - * This file contains the declarations for the low level - * functions that communicate with the floppy disk controller, - * for the QIC-40/80/3010/3020 floppy-tape driver "ftape" for - * Linux. - */ - -#include - -#include "../lowlevel/ftape-bsm.h" - -#define FDC_SK_BIT (0x20) -#define FDC_MT_BIT (0x80) - -#define FDC_READ (FD_READ & ~(FDC_SK_BIT | FDC_MT_BIT)) -#define FDC_WRITE (FD_WRITE & ~FDC_MT_BIT) -#define FDC_READ_DELETED (0x4c) -#define FDC_WRITE_DELETED (0x49) -#define FDC_VERIFY (0x56) -#define FDC_READID (0x4a) -#define FDC_SENSED (0x04) -#define FDC_SENSEI (FD_SENSEI) -#define FDC_FORMAT (FD_FORMAT) -#define FDC_RECAL (FD_RECALIBRATE) -#define FDC_SEEK (FD_SEEK) -#define FDC_SPECIFY (FD_SPECIFY) -#define FDC_RECALIBR (FD_RECALIBRATE) -#define FDC_VERSION (FD_VERSION) -#define FDC_PERPEND (FD_PERPENDICULAR) -#define FDC_DUMPREGS (FD_DUMPREGS) -#define FDC_LOCK (FD_LOCK) -#define FDC_UNLOCK (FD_UNLOCK) -#define FDC_CONFIGURE (FD_CONFIGURE) -#define FDC_DRIVE_SPEC (0x8e) /* i82078 has this (any others?) */ -#define FDC_PARTID (0x18) /* i82078 has this */ -#define FDC_SAVE (0x2e) /* i82078 has this (any others?) */ -#define FDC_RESTORE (0x4e) /* i82078 has this (any others?) */ - -#define FDC_STATUS_MASK (STATUS_BUSY | STATUS_DMA | STATUS_DIR | STATUS_READY) -#define FDC_DATA_READY (STATUS_READY) -#define FDC_DATA_OUTPUT (STATUS_DIR) -#define FDC_DATA_READY_MASK (STATUS_READY | STATUS_DIR) -#define FDC_DATA_OUT_READY (STATUS_READY | STATUS_DIR) -#define FDC_DATA_IN_READY (STATUS_READY) -#define FDC_BUSY (STATUS_BUSY) -#define FDC_CLK48_BIT (0x80) -#define FDC_SEL3V_BIT (0x40) - -#define ST0_INT_MASK (ST0_INTR) -#define FDC_INT_NORMAL (ST0_INTR & 0x00) -#define FDC_INT_ABNORMAL (ST0_INTR & 0x40) -#define FDC_INT_INVALID (ST0_INTR & 0x80) -#define FDC_INT_READYCH (ST0_INTR & 0xC0) -#define ST0_SEEK_END (ST0_SE) -#define ST3_TRACK_0 (ST3_TZ) - -#define FDC_RESET_NOT (0x04) -#define FDC_DMA_MODE (0x08) -#define FDC_MOTOR_0 (0x10) -#define FDC_MOTOR_1 (0x20) - -typedef struct { - void (**hook) (void); /* our wedge into the isr */ - enum { - no_fdc, i8272, i82077, i82077AA, fc10, - i82078, i82078_1 - } type; /* FDC type */ - unsigned int irq; /* FDC irq nr */ - unsigned int dma; /* FDC dma channel nr */ - __u16 sra; /* Status register A (PS/2 only) */ - __u16 srb; /* Status register B (PS/2 only) */ - __u16 dor; /* Digital output register */ - __u16 tdr; /* Tape Drive Register (82077SL-1 & - 82078 only) */ - __u16 msr; /* Main Status Register */ - __u16 dsr; /* Datarate Select Register (8207x only) */ - __u16 fifo; /* Data register / Fifo on 8207x */ - __u16 dir; /* Digital Input Register */ - __u16 ccr; /* Configuration Control Register */ - __u16 dor2; /* Alternate dor on MACH-2 controller, - also used with FC-10, meaning unknown */ -} fdc_config_info; - -typedef enum { - fdc_data_rate_250 = 2, - fdc_data_rate_300 = 1, /* any fdc in default configuration */ - fdc_data_rate_500 = 0, - fdc_data_rate_1000 = 3, - fdc_data_rate_2000 = 1, /* i82078-1: when using Data Rate Table #2 */ -} fdc_data_rate_type; - -typedef enum { - fdc_idle = 0, - fdc_reading_data = FDC_READ, - fdc_seeking = FDC_SEEK, - fdc_writing_data = FDC_WRITE, - fdc_deleting = FDC_WRITE_DELETED, - fdc_reading_id = FDC_READID, - fdc_recalibrating = FDC_RECAL, - fdc_formatting = FDC_FORMAT, - fdc_verifying = FDC_VERIFY -} fdc_mode_enum; - -typedef enum { - waiting = 0, - reading, - writing, - formatting, - verifying, - deleting, - done, - error, - mmapped, -} buffer_state_enum; - -typedef struct { - __u8 *address; - volatile buffer_state_enum status; - volatile __u8 *ptr; - volatile unsigned int bytes; - volatile unsigned int segment_id; - - /* bitmap for remainder of segment not yet handled. - * one bit set for each bad sector that must be skipped. - */ - volatile SectorMap bad_sector_map; - - /* bitmap with bad data blocks in data buffer. - * the errors in this map may be retried. - */ - volatile SectorMap soft_error_map; - - /* bitmap with bad data blocks in data buffer - * the errors in this map may not be retried. - */ - volatile SectorMap hard_error_map; - - /* retry counter for soft errors. - */ - volatile int retry; - - /* sectors to skip on retry ??? - */ - volatile unsigned int skip; - - /* nr of data blocks in data buffer - */ - volatile unsigned int data_offset; - - /* offset in segment for first sector to be handled. - */ - volatile unsigned int sector_offset; - - /* size of cluster of good sectors to be handled. - */ - volatile unsigned int sector_count; - - /* size of remaining part of segment to be handled. - */ - volatile unsigned int remaining; - - /* points to next segment (contiguous) to be handled, - * or is zero if no read-ahead is allowed. - */ - volatile unsigned int next_segment; - - /* flag being set if deleted data was read. - */ - volatile int deleted; - - /* floppy coordinates of first sector in segment */ - volatile __u8 head; - volatile __u8 cyl; - volatile __u8 sect; - - /* gap to use when formatting */ - __u8 gap3; - /* flag set when buffer is mmaped */ - int mmapped; -} buffer_struct; - -/* - * fdc-io.c defined public variables - */ -extern volatile fdc_mode_enum fdc_mode; -extern int fdc_setup_error; /* outdated ??? */ -extern wait_queue_head_t ftape_wait_intr; -extern volatile int ftape_current_cylinder; /* track nr FDC thinks we're on */ -extern volatile __u8 fdc_head; /* FDC head */ -extern volatile __u8 fdc_cyl; /* FDC track */ -extern volatile __u8 fdc_sect; /* FDC sector */ -extern fdc_config_info fdc; /* FDC hardware configuration */ - -extern unsigned int ft_fdc_base; -extern unsigned int ft_fdc_irq; -extern unsigned int ft_fdc_dma; -extern unsigned int ft_fdc_threshold; -extern unsigned int ft_fdc_rate_limit; -extern int ft_probe_fc10; -extern int ft_mach2; -/* - * fdc-io.c defined public functions - */ -extern void fdc_catch_stray_interrupts(int count); -extern int fdc_ready_wait(unsigned int timeout); -extern int fdc_command(const __u8 * cmd_data, int cmd_len); -extern int fdc_result(__u8 * res_data, int res_len); -extern int fdc_interrupt_wait(unsigned int time); -extern int fdc_seek(int track); -extern int fdc_sense_drive_status(int *st3); -extern void fdc_motor(int motor); -extern void fdc_reset(void); -extern void fdc_disable(void); -extern int fdc_fifo_threshold(__u8 threshold, - int *fifo_state, int *lock_state, int *fifo_thr); -extern void fdc_wait_calibrate(void); -extern int fdc_sense_interrupt_status(int *st0, int *current_cylinder); -extern void fdc_save_drive_specs(void); -extern void fdc_restore_drive_specs(void); -extern int fdc_set_data_rate(int rate); -extern void fdc_set_write_precomp(int precomp); -extern int fdc_release_irq_and_dma(void); -extern void fdc_release_regions(void); -extern int fdc_init(void); -extern int fdc_setup_read_write(buffer_struct * buff, __u8 operation); -extern int fdc_setup_formatting(buffer_struct * buff); -#endif diff --git a/drivers/char/ftape/lowlevel/fdc-isr.c b/drivers/char/ftape/lowlevel/fdc-isr.c deleted file mode 100644 index ad2bc733ae1..00000000000 --- a/drivers/char/ftape/lowlevel/fdc-isr.c +++ /dev/null @@ -1,1170 +0,0 @@ -/* - * Copyright (C) 1994-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/fdc-isr.c,v $ - * $Revision: 1.9 $ - * $Date: 1997/10/17 23:01:53 $ - * - * This file contains the interrupt service routine and - * associated code for the QIC-40/80/3010/3020 floppy-tape driver - * "ftape" for Linux. - */ - -#include -#include - -#define volatile /* */ - -#include -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/fdc-isr.h" -#include "../lowlevel/fdc-io.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-calibr.h" -#include "../lowlevel/ftape-bsm.h" - -/* Global vars. - */ -volatile int ft_expected_stray_interrupts; -volatile int ft_interrupt_seen; -volatile int ft_seek_completed; -volatile int ft_hide_interrupt; -/* Local vars. - */ -typedef enum { - no_error = 0, id_am_error = 0x01, id_crc_error = 0x02, - data_am_error = 0x04, data_crc_error = 0x08, - no_data_error = 0x10, overrun_error = 0x20, -} error_cause; -static int stop_read_ahead; - - -static void print_error_cause(int cause) -{ - TRACE_FUN(ft_t_any); - - switch (cause) { - case no_data_error: - TRACE(ft_t_noise, "no data error"); - break; - case id_am_error: - TRACE(ft_t_noise, "id am error"); - break; - case id_crc_error: - TRACE(ft_t_noise, "id crc error"); - break; - case data_am_error: - TRACE(ft_t_noise, "data am error"); - break; - case data_crc_error: - TRACE(ft_t_noise, "data crc error"); - break; - case overrun_error: - TRACE(ft_t_noise, "overrun error"); - break; - default:; - } - TRACE_EXIT; -} - -static char *fdc_mode_txt(fdc_mode_enum mode) -{ - switch (mode) { - case fdc_idle: - return "fdc_idle"; - case fdc_reading_data: - return "fdc_reading_data"; - case fdc_seeking: - return "fdc_seeking"; - case fdc_writing_data: - return "fdc_writing_data"; - case fdc_reading_id: - return "fdc_reading_id"; - case fdc_recalibrating: - return "fdc_recalibrating"; - case fdc_formatting: - return "fdc_formatting"; - case fdc_verifying: - return "fdc_verifying"; - default: - return "unknown"; - } -} - -static inline error_cause decode_irq_cause(fdc_mode_enum mode, __u8 st[]) -{ - error_cause cause = no_error; - TRACE_FUN(ft_t_any); - - /* Valid st[], decode cause of interrupt. - */ - switch (st[0] & ST0_INT_MASK) { - case FDC_INT_NORMAL: - TRACE(ft_t_fdc_dma,"normal completion: %s",fdc_mode_txt(mode)); - break; - case FDC_INT_ABNORMAL: - TRACE(ft_t_flow, "abnormal completion %s", fdc_mode_txt(mode)); - TRACE(ft_t_fdc_dma, "ST0: 0x%02x, ST1: 0x%02x, ST2: 0x%02x", - st[0], st[1], st[2]); - TRACE(ft_t_fdc_dma, - "C: 0x%02x, H: 0x%02x, R: 0x%02x, N: 0x%02x", - st[3], st[4], st[5], st[6]); - if (st[1] & 0x01) { - if (st[2] & 0x01) { - cause = data_am_error; - } else { - cause = id_am_error; - } - } else if (st[1] & 0x20) { - if (st[2] & 0x20) { - cause = data_crc_error; - } else { - cause = id_crc_error; - } - } else if (st[1] & 0x04) { - cause = no_data_error; - } else if (st[1] & 0x10) { - cause = overrun_error; - } - print_error_cause(cause); - break; - case FDC_INT_INVALID: - TRACE(ft_t_flow, "invalid completion %s", fdc_mode_txt(mode)); - break; - case FDC_INT_READYCH: - if (st[0] & ST0_SEEK_END) { - TRACE(ft_t_flow, "drive poll completed"); - } else { - TRACE(ft_t_flow, "ready change %s",fdc_mode_txt(mode)); - } - break; - default: - break; - } - TRACE_EXIT cause; -} - -static void update_history(error_cause cause) -{ - switch (cause) { - case id_am_error: - ft_history.id_am_errors++; - break; - case id_crc_error: - ft_history.id_crc_errors++; - break; - case data_am_error: - ft_history.data_am_errors++; - break; - case data_crc_error: - ft_history.data_crc_errors++; - break; - case overrun_error: - ft_history.overrun_errors++; - break; - case no_data_error: - ft_history.no_data_errors++; - break; - default:; - } -} - -static void skip_bad_sector(buffer_struct * buff) -{ - TRACE_FUN(ft_t_any); - - /* Mark sector as soft error and skip it - */ - if (buff->remaining > 0) { - ++buff->sector_offset; - ++buff->data_offset; - --buff->remaining; - buff->ptr += FT_SECTOR_SIZE; - buff->bad_sector_map >>= 1; - } else { - /* Hey, what is this????????????? C code: if we shift - * more than 31 bits, we get no shift. That's bad!!!!!! - */ - ++buff->sector_offset; /* hack for error maps */ - TRACE(ft_t_warn, "skipping last sector in segment"); - } - TRACE_EXIT; -} - -static void update_error_maps(buffer_struct * buff, unsigned int error_offset) -{ - int hard = 0; - TRACE_FUN(ft_t_any); - - if (buff->retry < FT_SOFT_RETRIES) { - buff->soft_error_map |= (1 << error_offset); - } else { - buff->hard_error_map |= (1 << error_offset); - buff->soft_error_map &= ~buff->hard_error_map; - buff->retry = -1; /* will be set to 0 in setup_segment */ - hard = 1; - } - TRACE(ft_t_noise, "sector %d : %s error\n" - KERN_INFO "hard map: 0x%08lx\n" - KERN_INFO "soft map: 0x%08lx", - FT_SECTOR(error_offset), hard ? "hard" : "soft", - (long) buff->hard_error_map, (long) buff->soft_error_map); - TRACE_EXIT; -} - -static void print_progress(buffer_struct *buff, error_cause cause) -{ - TRACE_FUN(ft_t_any); - - switch (cause) { - case no_error: - TRACE(ft_t_flow,"%d Sector(s) transferred", buff->sector_count); - break; - case no_data_error: - TRACE(ft_t_flow, "Sector %d not found", - FT_SECTOR(buff->sector_offset)); - break; - case overrun_error: - /* got an overrun error on the first byte, must be a - * hardware problem - */ - TRACE(ft_t_bug, - "Unexpected error: failing DMA or FDC controller ?"); - break; - case data_crc_error: - TRACE(ft_t_flow, "Error in sector %d", - FT_SECTOR(buff->sector_offset - 1)); - break; - case id_crc_error: - case id_am_error: - case data_am_error: - TRACE(ft_t_flow, "Error in sector %d", - FT_SECTOR(buff->sector_offset)); - break; - default: - TRACE(ft_t_flow, "Unexpected error at sector %d", - FT_SECTOR(buff->sector_offset)); - break; - } - TRACE_EXIT; -} - -/* - * Error cause: Amount xferred: Action: - * - * id_am_error 0 mark bad and skip - * id_crc_error 0 mark bad and skip - * data_am_error 0 mark bad and skip - * data_crc_error % 1024 mark bad and skip - * no_data_error 0 retry on write - * mark bad and skip on read - * overrun_error [ 0..all-1 ] mark bad and skip - * no_error all continue - */ - -/* the arg `sector' is returned by the fdc and tells us at which sector we - * are positioned at (relative to starting sector of segment) - */ -static void determine_verify_progress(buffer_struct *buff, - error_cause cause, - __u8 sector) -{ - TRACE_FUN(ft_t_any); - - if (cause == no_error && sector == 1) { - buff->sector_offset = FT_SECTORS_PER_SEGMENT; - buff->remaining = 0; - if (TRACE_LEVEL >= ft_t_flow) { - print_progress(buff, cause); - } - } else { - buff->sector_offset = sector - buff->sect; - buff->remaining = FT_SECTORS_PER_SEGMENT - buff->sector_offset; - TRACE(ft_t_noise, "%ssector offset: 0x%04x", - (cause == no_error) ? "unexpected " : "", - buff->sector_offset); - switch (cause) { - case overrun_error: - break; -#if 0 - case no_data_error: - buff->retry = FT_SOFT_RETRIES; - if (buff->hard_error_map && - buff->sector_offset > 1 && - (buff->hard_error_map & - (1 << (buff->sector_offset-2)))) { - buff->retry --; - } - break; -#endif - default: - buff->retry = FT_SOFT_RETRIES; - break; - } - if (TRACE_LEVEL >= ft_t_flow) { - print_progress(buff, cause); - } - /* Sector_offset points to the problem area Now adjust - * sector_offset so it always points one past he failing - * sector. I.e. skip the bad sector. - */ - ++buff->sector_offset; - --buff->remaining; - update_error_maps(buff, buff->sector_offset - 1); - } - TRACE_EXIT; -} - -static void determine_progress(buffer_struct *buff, - error_cause cause, - __u8 sector) -{ - unsigned int dma_residue; - TRACE_FUN(ft_t_any); - - /* Using less preferred order of disable_dma and - * get_dma_residue because this seems to fail on at least one - * system if reversed! - */ - dma_residue = get_dma_residue(fdc.dma); - disable_dma(fdc.dma); - if (cause != no_error || dma_residue != 0) { - TRACE(ft_t_noise, "%sDMA residue: 0x%04x", - (cause == no_error) ? "unexpected " : "", - dma_residue); - /* adjust to actual value: */ - if (dma_residue == 0) { - /* this happens sometimes with overrun errors. - * I don't know whether we could ignore the - * overrun error. Play save. - */ - buff->sector_count --; - } else { - buff->sector_count -= ((dma_residue + - (FT_SECTOR_SIZE - 1)) / - FT_SECTOR_SIZE); - } - } - /* Update var's influenced by the DMA operation. - */ - if (buff->sector_count > 0) { - buff->sector_offset += buff->sector_count; - buff->data_offset += buff->sector_count; - buff->ptr += (buff->sector_count * - FT_SECTOR_SIZE); - buff->remaining -= buff->sector_count; - buff->bad_sector_map >>= buff->sector_count; - } - if (TRACE_LEVEL >= ft_t_flow) { - print_progress(buff, cause); - } - if (cause != no_error) { - if (buff->remaining == 0) { - TRACE(ft_t_warn, "foo?\n" - KERN_INFO "count : %d\n" - KERN_INFO "offset: %d\n" - KERN_INFO "soft : %08x\n" - KERN_INFO "hard : %08x", - buff->sector_count, - buff->sector_offset, - buff->soft_error_map, - buff->hard_error_map); - } - /* Sector_offset points to the problem area, except if we got - * a data_crc_error. In that case it points one past the - * failing sector. - * - * Now adjust sector_offset so it always points one past he - * failing sector. I.e. skip the bad sector. - */ - if (cause != data_crc_error) { - skip_bad_sector(buff); - } - update_error_maps(buff, buff->sector_offset - 1); - } - TRACE_EXIT; -} - -static int calc_steps(int cmd) -{ - if (ftape_current_cylinder > cmd) { - return ftape_current_cylinder - cmd; - } else { - return ftape_current_cylinder + cmd; - } -} - -static void pause_tape(int retry, int mode) -{ - int result; - __u8 out[3] = {FDC_SEEK, ft_drive_sel, 0}; - TRACE_FUN(ft_t_any); - - /* We'll use a raw seek command to get the tape to rewind and - * stop for a retry. - */ - ++ft_history.rewinds; - if (qic117_cmds[ftape_current_command].non_intr) { - TRACE(ft_t_warn, "motion command may be issued too soon"); - } - if (retry && (mode == fdc_reading_data || - mode == fdc_reading_id || - mode == fdc_verifying)) { - ftape_current_command = QIC_MICRO_STEP_PAUSE; - ftape_might_be_off_track = 1; - } else { - ftape_current_command = QIC_PAUSE; - } - out[2] = calc_steps(ftape_current_command); - result = fdc_command(out, 3); /* issue QIC_117 command */ - ftape_current_cylinder = out[ 2]; - if (result < 0) { - TRACE(ft_t_noise, "qic-pause failed, status = %d", result); - } else { - ft_location.known = 0; - ft_runner_status = idle; - ft_hide_interrupt = 1; - ftape_tape_running = 0; - } - TRACE_EXIT; -} - -static void continue_xfer(buffer_struct *buff, - fdc_mode_enum mode, - unsigned int skip) -{ - int write = 0; - TRACE_FUN(ft_t_any); - - if (mode == fdc_writing_data || mode == fdc_deleting) { - write = 1; - } - /* This part can be removed if it never happens - */ - if (skip > 0 && - (ft_runner_status != running || - (write && (buff->status != writing)) || - (!write && (buff->status != reading && - buff->status != verifying)))) { - TRACE(ft_t_err, "unexpected runner/buffer state %d/%d", - ft_runner_status, buff->status); - buff->status = error; - /* finish this buffer: */ - (void)ftape_next_buffer(ft_queue_head); - ft_runner_status = aborting; - fdc_mode = fdc_idle; - } else if (buff->remaining > 0 && ftape_calc_next_cluster(buff) > 0) { - /* still sectors left in current segment, continue - * with this segment - */ - if (fdc_setup_read_write(buff, mode) < 0) { - /* failed, abort operation - */ - buff->bytes = buff->ptr - buff->address; - buff->status = error; - /* finish this buffer: */ - (void)ftape_next_buffer(ft_queue_head); - ft_runner_status = aborting; - fdc_mode = fdc_idle; - } - } else { - /* current segment completed - */ - unsigned int last_segment = buff->segment_id; - int eot = ((last_segment + 1) % ft_segments_per_track) == 0; - unsigned int next = buff->next_segment; /* 0 means stop ! */ - - buff->bytes = buff->ptr - buff->address; - buff->status = done; - buff = ftape_next_buffer(ft_queue_head); - if (eot) { - /* finished last segment on current track, - * can't continue - */ - ft_runner_status = logical_eot; - fdc_mode = fdc_idle; - TRACE_EXIT; - } - if (next <= 0) { - /* don't continue with next segment - */ - TRACE(ft_t_noise, "no %s allowed, stopping tape", - (write) ? "write next" : "read ahead"); - pause_tape(0, mode); - ft_runner_status = idle; /* not quite true until - * next irq - */ - TRACE_EXIT; - } - /* continue with next segment - */ - if (buff->status != waiting) { - TRACE(ft_t_noise, "all input buffers %s, pausing tape", - (write) ? "empty" : "full"); - pause_tape(0, mode); - ft_runner_status = idle; /* not quite true until - * next irq - */ - TRACE_EXIT; - } - if (write && next != buff->segment_id) { - TRACE(ft_t_noise, - "segments out of order, aborting write"); - ft_runner_status = do_abort; - fdc_mode = fdc_idle; - TRACE_EXIT; - } - ftape_setup_new_segment(buff, next, 0); - if (stop_read_ahead) { - buff->next_segment = 0; - stop_read_ahead = 0; - } - if (ftape_calc_next_cluster(buff) == 0 || - fdc_setup_read_write(buff, mode) != 0) { - TRACE(ft_t_err, "couldn't start %s-ahead", - write ? "write" : "read"); - ft_runner_status = do_abort; - fdc_mode = fdc_idle; - } else { - /* keep on going */ - switch (ft_driver_state) { - case reading: buff->status = reading; break; - case verifying: buff->status = verifying; break; - case writing: buff->status = writing; break; - case deleting: buff->status = deleting; break; - default: - TRACE(ft_t_err, - "BUG: ft_driver_state %d should be one out of " - "{reading, writing, verifying, deleting}", - ft_driver_state); - buff->status = write ? writing : reading; - break; - } - } - } - TRACE_EXIT; -} - -static void retry_sector(buffer_struct *buff, - int mode, - unsigned int skip) -{ - TRACE_FUN(ft_t_any); - - TRACE(ft_t_noise, "%s error, will retry", - (mode == fdc_writing_data || mode == fdc_deleting) ? "write" : "read"); - pause_tape(1, mode); - ft_runner_status = aborting; - buff->status = error; - buff->skip = skip; - TRACE_EXIT; -} - -static unsigned int find_resume_point(buffer_struct *buff) -{ - int i = 0; - SectorMap mask; - SectorMap map; - TRACE_FUN(ft_t_any); - - /* This function is to be called after all variables have been - * updated to point past the failing sector. - * If there are any soft errors before the failing sector, - * find the first soft error and return the sector offset. - * Otherwise find the last hard error. - * Note: there should always be at least one hard or soft error ! - */ - if (buff->sector_offset < 1 || buff->sector_offset > 32) { - TRACE(ft_t_bug, "BUG: sector_offset = %d", - buff->sector_offset); - TRACE_EXIT 0; - } - if (buff->sector_offset >= 32) { /* C-limitation on shift ! */ - mask = 0xffffffff; - } else { - mask = (1 << buff->sector_offset) - 1; - } - map = buff->soft_error_map & mask; - if (map) { - while ((map & (1 << i)) == 0) { - ++i; - } - TRACE(ft_t_noise, "at sector %d", FT_SECTOR(i)); - } else { - map = buff->hard_error_map & mask; - i = buff->sector_offset - 1; - if (map) { - while ((map & (1 << i)) == 0) { - --i; - } - TRACE(ft_t_noise, "after sector %d", FT_SECTOR(i)); - ++i; /* first sector after last hard error */ - } else { - TRACE(ft_t_bug, "BUG: no soft or hard errors"); - } - } - TRACE_EXIT i; -} - -/* check possible dma residue when formatting, update position record in - * buffer struct. This is, of course, modelled after determine_progress(), but - * we don't need to set up for retries because the format process cannot be - * interrupted (except at the end of the tape track). - */ -static int determine_fmt_progress(buffer_struct *buff, error_cause cause) -{ - unsigned int dma_residue; - TRACE_FUN(ft_t_any); - - /* Using less preferred order of disable_dma and - * get_dma_residue because this seems to fail on at least one - * system if reversed! - */ - dma_residue = get_dma_residue(fdc.dma); - disable_dma(fdc.dma); - if (cause != no_error || dma_residue != 0) { - TRACE(ft_t_info, "DMA residue = 0x%04x", dma_residue); - fdc_mode = fdc_idle; - switch(cause) { - case no_error: - ft_runner_status = aborting; - buff->status = idle; - break; - case overrun_error: - /* got an overrun error on the first byte, must be a - * hardware problem - */ - TRACE(ft_t_bug, - "Unexpected error: failing DMA controller ?"); - ft_runner_status = do_abort; - buff->status = error; - break; - default: - TRACE(ft_t_noise, "Unexpected error at segment %d", - buff->segment_id); - ft_runner_status = do_abort; - buff->status = error; - break; - } - TRACE_EXIT -EIO; /* can only retry entire track in format mode - */ - } - /* Update var's influenced by the DMA operation. - */ - buff->ptr += FT_SECTORS_PER_SEGMENT * 4; - buff->bytes -= FT_SECTORS_PER_SEGMENT * 4; - buff->remaining -= FT_SECTORS_PER_SEGMENT; - buff->segment_id ++; /* done with segment */ - TRACE_EXIT 0; -} - -/* - * Continue formatting, switch buffers if there is no data left in - * current buffer. This is, of course, modelled after - * continue_xfer(), but we don't need to set up for retries because - * the format process cannot be interrupted (except at the end of the - * tape track). - */ -static void continue_formatting(buffer_struct *buff) -{ - TRACE_FUN(ft_t_any); - - if (buff->remaining <= 0) { /* no space left in dma buffer */ - unsigned int next = buff->next_segment; - - if (next == 0) { /* end of tape track */ - buff->status = done; - ft_runner_status = logical_eot; - fdc_mode = fdc_idle; - TRACE(ft_t_noise, "Done formatting track %d", - ft_location.track); - TRACE_EXIT; - } - /* - * switch to next buffer! - */ - buff->status = done; - buff = ftape_next_buffer(ft_queue_head); - - if (buff->status != waiting || next != buff->segment_id) { - goto format_setup_error; - } - } - if (fdc_setup_formatting(buff) < 0) { - goto format_setup_error; - } - buff->status = formatting; - TRACE(ft_t_fdc_dma, "Formatting segment %d on track %d", - buff->segment_id, ft_location.track); - TRACE_EXIT; - format_setup_error: - ft_runner_status = do_abort; - fdc_mode = fdc_idle; - buff->status = error; - TRACE(ft_t_err, "Error setting up for segment %d on track %d", - buff->segment_id, ft_location.track); - TRACE_EXIT; - -} - -/* this handles writing, read id, reading and formatting - */ -static void handle_fdc_busy(buffer_struct *buff) -{ - static int no_data_error_count; - int retry = 0; - error_cause cause; - __u8 in[7]; - int skip; - fdc_mode_enum fmode = fdc_mode; - TRACE_FUN(ft_t_any); - - if (fdc_result(in, 7) < 0) { /* better get it fast ! */ - TRACE(ft_t_err, - "Probably fatal error during FDC Result Phase\n" - KERN_INFO - "drive may hang until (power on) reset :-("); - /* what to do next ???? - */ - TRACE_EXIT; - } - cause = decode_irq_cause(fdc_mode, in); -#ifdef TESTING - { int i; - for (i = 0; i < (int)ft_nr_buffers; ++i) - TRACE(ft_t_any, "buffer[%d] status: %d, segment_id: %d", - i, ft_buffer[i]->status, ft_buffer[i]->segment_id); - } -#endif - if (fmode == fdc_reading_data && ft_driver_state == verifying) { - fmode = fdc_verifying; - } - switch (fmode) { - case fdc_verifying: - if (ft_runner_status == aborting || - ft_runner_status == do_abort) { - TRACE(ft_t_noise,"aborting %s",fdc_mode_txt(fdc_mode)); - break; - } - if (buff->retry > 0) { - TRACE(ft_t_flow, "this is retry nr %d", buff->retry); - } - switch (cause) { - case no_error: - no_data_error_count = 0; - determine_verify_progress(buff, cause, in[5]); - if (in[2] & 0x40) { - /* This should not happen when verifying - */ - TRACE(ft_t_warn, - "deleted data in segment %d/%d", - buff->segment_id, - FT_SECTOR(buff->sector_offset - 1)); - buff->remaining = 0; /* abort transfer */ - buff->hard_error_map = EMPTY_SEGMENT; - skip = 1; - } else { - skip = 0; - } - continue_xfer(buff, fdc_mode, skip); - break; - case no_data_error: - no_data_error_count ++; - case overrun_error: - retry ++; - case id_am_error: - case id_crc_error: - case data_am_error: - case data_crc_error: - determine_verify_progress(buff, cause, in[5]); - if (cause == no_data_error) { - if (no_data_error_count >= 2) { - TRACE(ft_t_warn, - "retrying because of successive " - "no data errors"); - no_data_error_count = 0; - } else { - retry --; - } - } else { - no_data_error_count = 0; - } - if (retry) { - skip = find_resume_point(buff); - } else { - skip = buff->sector_offset; - } - if (retry && skip < 32) { - retry_sector(buff, fdc_mode, skip); - } else { - continue_xfer(buff, fdc_mode, skip); - } - update_history(cause); - break; - default: - /* Don't know why this could happen - * but find out. - */ - determine_verify_progress(buff, cause, in[5]); - retry_sector(buff, fdc_mode, 0); - TRACE(ft_t_err, "Error: unexpected error"); - break; - } - break; - case fdc_reading_data: -#ifdef TESTING - /* I'm sorry, but: NOBODY ever used this trace - * messages for ages. I guess that Bas was the last person - * that ever really used this (thank you, between the lines) - */ - if (cause == no_error) { - TRACE(ft_t_flow,"reading segment %d",buff->segment_id); - } else { - TRACE(ft_t_noise, "error reading segment %d", - buff->segment_id); - TRACE(ft_t_noise, "\n" - KERN_INFO - "IRQ:C: 0x%02x, H: 0x%02x, R: 0x%02x, N: 0x%02x\n" - KERN_INFO - "BUF:C: 0x%02x, H: 0x%02x, R: 0x%02x", - in[3], in[4], in[5], in[6], - buff->cyl, buff->head, buff->sect); - } -#endif - if (ft_runner_status == aborting || - ft_runner_status == do_abort) { - TRACE(ft_t_noise,"aborting %s",fdc_mode_txt(fdc_mode)); - break; - } - if (buff->bad_sector_map == FAKE_SEGMENT) { - /* This condition occurs when reading a `fake' - * sector that's not accessible. Doesn't - * really matter as we would have ignored it - * anyway ! - * - * Chance is that we're past the next segment - * now, so the next operation may fail and - * result in a retry. - */ - buff->remaining = 0; /* skip failing sector */ - /* buff->ptr = buff->address; */ - /* fake success: */ - continue_xfer(buff, fdc_mode, 1); - /* trace calls are expensive: place them AFTER - * the real stuff has been done. - * - */ - TRACE(ft_t_noise, "skipping empty segment %d (read), size? %d", - buff->segment_id, buff->ptr - buff->address); - TRACE_EXIT; - } - if (buff->retry > 0) { - TRACE(ft_t_flow, "this is retry nr %d", buff->retry); - } - switch (cause) { - case no_error: - determine_progress(buff, cause, in[5]); - if (in[2] & 0x40) { - /* Handle deleted data in header segments. - * Skip segment and force read-ahead. - */ - TRACE(ft_t_warn, - "deleted data in segment %d/%d", - buff->segment_id, - FT_SECTOR(buff->sector_offset - 1)); - buff->deleted = 1; - buff->remaining = 0;/*abort transfer */ - buff->soft_error_map |= - (-1L << buff->sector_offset); - if (buff->segment_id == 0) { - /* stop on next segment */ - stop_read_ahead = 1; - } - /* force read-ahead: */ - buff->next_segment = - buff->segment_id + 1; - skip = (FT_SECTORS_PER_SEGMENT - - buff->sector_offset); - } else { - skip = 0; - } - continue_xfer(buff, fdc_mode, skip); - break; - case no_data_error: - /* Tape started too far ahead of or behind the - * right sector. This may also happen in the - * middle of a segment ! - * - * Handle no-data as soft error. If next - * sector fails too, a retry (with needed - * reposition) will follow. - */ - retry ++; - case id_am_error: - case id_crc_error: - case data_am_error: - case data_crc_error: - case overrun_error: - retry += (buff->soft_error_map != 0 || - buff->hard_error_map != 0); - determine_progress(buff, cause, in[5]); -#if 1 || defined(TESTING) - if (cause == overrun_error) retry ++; -#endif - if (retry) { - skip = find_resume_point(buff); - } else { - skip = buff->sector_offset; - } - /* Try to resume with next sector on single - * errors (let ecc correct it), but retry on - * no_data (we'll be past the target when we - * get here so we cannot retry) or on - * multiple errors (reduce chance on ecc - * failure). - */ - /* cH: 23/02/97: if the last sector in the - * segment was a hard error, then there is - * no sense in a retry. This occasion seldom - * occurs but ... @:³²¸`@%&§$ - */ - if (retry && skip < 32) { - retry_sector(buff, fdc_mode, skip); - } else { - continue_xfer(buff, fdc_mode, skip); - } - update_history(cause); - break; - default: - /* Don't know why this could happen - * but find out. - */ - determine_progress(buff, cause, in[5]); - retry_sector(buff, fdc_mode, 0); - TRACE(ft_t_err, "Error: unexpected error"); - break; - } - break; - case fdc_reading_id: - if (cause == no_error) { - fdc_cyl = in[3]; - fdc_head = in[4]; - fdc_sect = in[5]; - TRACE(ft_t_fdc_dma, - "id read: C: 0x%02x, H: 0x%02x, R: 0x%02x", - fdc_cyl, fdc_head, fdc_sect); - } else { /* no valid information, use invalid sector */ - fdc_cyl = fdc_head = fdc_sect = 0; - TRACE(ft_t_flow, "Didn't find valid sector Id"); - } - fdc_mode = fdc_idle; - break; - case fdc_deleting: - case fdc_writing_data: -#ifdef TESTING - if (cause == no_error) { - TRACE(ft_t_flow, "writing segment %d", buff->segment_id); - } else { - TRACE(ft_t_noise, "error writing segment %d", - buff->segment_id); - } -#endif - if (ft_runner_status == aborting || - ft_runner_status == do_abort) { - TRACE(ft_t_flow, "aborting %s",fdc_mode_txt(fdc_mode)); - break; - } - if (buff->retry > 0) { - TRACE(ft_t_flow, "this is retry nr %d", buff->retry); - } - if (buff->bad_sector_map == FAKE_SEGMENT) { - /* This condition occurs when trying to write to a - * `fake' sector that's not accessible. Doesn't really - * matter as it isn't used anyway ! Might be located - * at wrong segment, then we'll fail on the next - * segment. - */ - TRACE(ft_t_noise, "skipping empty segment (write)"); - buff->remaining = 0; /* skip failing sector */ - /* fake success: */ - continue_xfer(buff, fdc_mode, 1); - break; - } - switch (cause) { - case no_error: - determine_progress(buff, cause, in[5]); - continue_xfer(buff, fdc_mode, 0); - break; - case no_data_error: - case id_am_error: - case id_crc_error: - case data_am_error: - case overrun_error: - update_history(cause); - determine_progress(buff, cause, in[5]); - skip = find_resume_point(buff); - retry_sector(buff, fdc_mode, skip); - break; - default: - if (in[1] & 0x02) { - TRACE(ft_t_err, "media not writable"); - } else { - TRACE(ft_t_bug, "unforeseen write error"); - } - fdc_mode = fdc_idle; - break; - } - break; /* fdc_deleting || fdc_writing_data */ - case fdc_formatting: - /* The interrupt comes after formatting a segment. We then - * have to set up QUICKLY for the next segment. But - * afterwards, there is plenty of time. - */ - switch (cause) { - case no_error: - /* would like to keep most of the formatting stuff - * outside the isr code, but timing is too critical - */ - if (determine_fmt_progress(buff, cause) >= 0) { - continue_formatting(buff); - } - break; - case no_data_error: - case id_am_error: - case id_crc_error: - case data_am_error: - case overrun_error: - default: - determine_fmt_progress(buff, cause); - update_history(cause); - if (in[1] & 0x02) { - TRACE(ft_t_err, "media not writable"); - } else { - TRACE(ft_t_bug, "unforeseen write error"); - } - break; - } /* cause */ - break; - default: - TRACE(ft_t_warn, "Warning: unexpected irq during: %s", - fdc_mode_txt(fdc_mode)); - fdc_mode = fdc_idle; - break; - } - TRACE_EXIT; -} - -/* FDC interrupt service routine. - */ -void fdc_isr(void) -{ - static int isr_active; -#ifdef TESTING - unsigned int t0 = ftape_timestamp(); -#endif - TRACE_FUN(ft_t_any); - - if (isr_active++) { - --isr_active; - TRACE(ft_t_bug, "BUG: nested interrupt, not good !"); - *fdc.hook = fdc_isr; /* hook our handler into the fdc - * code again - */ - TRACE_EXIT; - } - sti(); - if (inb_p(fdc.msr) & FDC_BUSY) { /* Entering Result Phase */ - ft_hide_interrupt = 0; - handle_fdc_busy(ftape_get_buffer(ft_queue_head)); - if (ft_runner_status == do_abort) { - /* cease operation, remember tape position - */ - TRACE(ft_t_flow, "runner aborting"); - ft_runner_status = aborting; - ++ft_expected_stray_interrupts; - } - } else { /* !FDC_BUSY */ - /* clear interrupt, cause should be gotten by issuing - * a Sense Interrupt Status command. - */ - if (fdc_mode == fdc_recalibrating || fdc_mode == fdc_seeking) { - if (ft_hide_interrupt) { - int st0; - int pcn; - - if (fdc_sense_interrupt_status(&st0, &pcn) < 0) - TRACE(ft_t_err, - "sense interrupt status failed"); - ftape_current_cylinder = pcn; - TRACE(ft_t_flow, "handled hidden interrupt"); - } - ft_seek_completed = 1; - fdc_mode = fdc_idle; - } else if (!waitqueue_active(&ftape_wait_intr)) { - if (ft_expected_stray_interrupts == 0) { - TRACE(ft_t_warn, "unexpected stray interrupt"); - } else { - TRACE(ft_t_flow, "expected stray interrupt"); - --ft_expected_stray_interrupts; - } - } else { - if (fdc_mode == fdc_reading_data || - fdc_mode == fdc_verifying || - fdc_mode == fdc_writing_data || - fdc_mode == fdc_deleting || - fdc_mode == fdc_formatting || - fdc_mode == fdc_reading_id) { - if (inb_p(fdc.msr) & FDC_BUSY) { - TRACE(ft_t_bug, - "***** FDC failure, busy too late"); - } else { - TRACE(ft_t_bug, - "***** FDC failure, no busy"); - } - } else { - TRACE(ft_t_fdc_dma, "awaited stray interrupt"); - } - } - ft_hide_interrupt = 0; - } - /* Handle sleep code. - */ - if (!ft_hide_interrupt) { - ft_interrupt_seen ++; - if (waitqueue_active(&ftape_wait_intr)) { - wake_up_interruptible(&ftape_wait_intr); - } - } else { - TRACE(ft_t_flow, "hiding interrupt while %s", - waitqueue_active(&ftape_wait_intr) ? "waiting":"active"); - } -#ifdef TESTING - t0 = ftape_timediff(t0, ftape_timestamp()); - if (t0 >= 1000) { - /* only tell us about long calls */ - TRACE(ft_t_noise, "isr() duration: %5d usec", t0); - } -#endif - *fdc.hook = fdc_isr; /* hook our handler into the fdc code again */ - --isr_active; - TRACE_EXIT; -} diff --git a/drivers/char/ftape/lowlevel/fdc-isr.h b/drivers/char/ftape/lowlevel/fdc-isr.h deleted file mode 100644 index 065aa978942..00000000000 --- a/drivers/char/ftape/lowlevel/fdc-isr.h +++ /dev/null @@ -1,55 +0,0 @@ -#ifndef _FDC_ISR_H -#define _FDC_ISR_H - -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/fdc-isr.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:07 $ - * - * This file declares the global variables necessary to - * synchronize the interrupt service routine (isr) with the - * remainder of the QIC-40/80/3010/3020 floppy-tape driver - * "ftape" for Linux. - */ - -/* - * fdc-isr.c defined public variables - */ -extern volatile int ft_expected_stray_interrupts; /* masks stray interrupts */ -extern volatile int ft_seek_completed; /* flag set by isr */ -extern volatile int ft_interrupt_seen; /* flag set by isr */ -extern volatile int ft_hide_interrupt; /* flag set by isr */ - -/* - * fdc-io.c defined public functions - */ -extern void fdc_isr(void); - -/* - * A kernel hook that steals one interrupt from the floppy - * driver (Should be fixed when the new fdc driver gets ready) - * See the linux kernel source files: - * drivers/block/floppy.c & drivers/block/blk.h - * for the details. - */ -extern void (*do_floppy) (void); - -#endif diff --git a/drivers/char/ftape/lowlevel/ftape-bsm.c b/drivers/char/ftape/lowlevel/ftape-bsm.c deleted file mode 100644 index d1a301cc344..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-bsm.c +++ /dev/null @@ -1,491 +0,0 @@ -/* - * Copyright (C) 1994-1996 Bas Laarhoven, - * (C) 1996-1997 Claus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-bsm.c,v $ - * $Revision: 1.3 $ - * $Date: 1997/10/05 19:15:15 $ - * - * This file contains the bad-sector map handling code for - * the QIC-117 floppy tape driver for Linux. - * QIC-40, QIC-80, QIC-3010 and QIC-3020 maps are implemented. - */ - -#include - -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-bsm.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-rw.h" - -/* Global vars. - */ - -/* Local vars. - */ -static __u8 *bad_sector_map; -static SectorCount *bsm_hash_ptr; - -typedef enum { - forward, backward -} mode_type; - -#if 0 -static void ftape_put_bad_sector_entry(int segment_id, SectorMap new_map); -#endif - -#if 0 -/* fix_tape converts a normal QIC-80 tape into a 'wide' tape. - * For testing purposes only ! - */ -void fix_tape(__u8 * buffer, ft_format_type new_code) -{ - static __u8 list[BAD_SECTOR_MAP_SIZE]; - SectorMap *src_ptr = (SectorMap *) list; - __u8 *dst_ptr = bad_sector_map; - SectorMap map; - unsigned int sector = 1; - int i; - - if (format_code != fmt_var && format_code != fmt_big) { - memcpy(list, bad_sector_map, sizeof(list)); - memset(bad_sector_map, 0, sizeof(bad_sector_map)); - while ((__u8 *) src_ptr - list < sizeof(list)) { - map = *src_ptr++; - if (map == EMPTY_SEGMENT) { - *(SectorMap *) dst_ptr = 0x800000 + sector; - dst_ptr += 3; - sector += SECTORS_PER_SEGMENT; - } else { - for (i = 0; i < SECTORS_PER_SEGMENT; ++i) { - if (map & 1) { - *(SewctorMap *) dst_ptr = sector; - dst_ptr += 3; - } - map >>= 1; - ++sector; - } - } - } - } - bad_sector_map_changed = 1; - *(buffer + 4) = new_code; /* put new format code */ - if (format_code != fmt_var && new_code == fmt_big) { - PUT4(buffer, FT_6_HSEG_1, (__u32)GET2(buffer, 6)); - PUT4(buffer, FT_6_HSEG_2, (__u32)GET2(buffer, 8)); - PUT4(buffer, FT_6_FRST_SEG, (__u32)GET2(buffer, 10)); - PUT4(buffer, FT_6_LAST_SEG, (__u32)GET2(buffer, 12)); - memset(buffer+6, '\0', 8); - } - format_code = new_code; -} - -#endif - -/* given buffer that contains a header segment, find the end of - * of the bsm list - */ -__u8 * ftape_find_end_of_bsm_list(__u8 * address) -{ - __u8 *ptr = address + FT_HEADER_END; /* start of bsm list */ - __u8 *limit = address + FT_SEGMENT_SIZE; - while (ptr + 2 < limit) { - if (ptr[0] || ptr[1] || ptr[2]) { - ptr += 3; - } else { - return ptr; - } - } - return NULL; -} - -static inline void put_sector(SectorCount *ptr, unsigned int sector) -{ - ptr->bytes[0] = sector & 0xff; - sector >>= 8; - ptr->bytes[1] = sector & 0xff; - sector >>= 8; - ptr->bytes[2] = sector & 0xff; -} - -static inline unsigned int get_sector(SectorCount *ptr) -{ -#if 1 - unsigned int sector; - - sector = ptr->bytes[0]; - sector += ptr->bytes[1] << 8; - sector += ptr->bytes[2] << 16; - - return sector; -#else - /* GET4 gets the next four bytes in Intel little endian order - * and converts them to host byte order and handles unaligned - * access. - */ - return (GET4(ptr, 0) & 0x00ffffff); /* back to host byte order */ -#endif -} - -static void bsm_debug_fake(void) -{ - /* for testing of bad sector handling at end of tape - */ -#if 0 - ftape_put_bad_sector_entry(segments_per_track * tracks_per_tape - 3, - 0x000003e0; - ftape_put_bad_sector_entry(segments_per_track * tracks_per_tape - 2, - 0xff3fffff; - ftape_put_bad_sector_entry(segments_per_track * tracks_per_tape - 1, - 0xffffe000; -#endif - /* Enable to test bad sector handling - */ -#if 0 - ftape_put_bad_sector_entry(30, 0xfffffffe) - ftape_put_bad_sector_entry(32, 0x7fffffff); - ftape_put_bad_sector_entry(34, 0xfffeffff); - ftape_put_bad_sector_entry(36, 0x55555555); - ftape_put_bad_sector_entry(38, 0xffffffff); - ftape_put_bad_sector_entry(50, 0xffff0000); - ftape_put_bad_sector_entry(51, 0xffffffff); - ftape_put_bad_sector_entry(52, 0xffffffff); - ftape_put_bad_sector_entry(53, 0x0000ffff); -#endif - /* Enable when testing multiple volume tar dumps. - */ -#if 0 - { - int i; - - for (i = ft_first_data_segment; - i <= ft_last_data_segment - 7; ++i) { - ftape_put_bad_sector_entry(i, EMPTY_SEGMENT); - } - } -#endif - /* Enable when testing bit positions in *_error_map - */ -#if 0 - { - int i; - - for (i = first_data_segment; i <= last_data_segment; ++i) { - ftape_put_bad_sector_entry(i, - ftape_get_bad_sector_entry(i) - | 0x00ff00ff); - } - } -#endif -} - -static void print_bad_sector_map(void) -{ - unsigned int good_sectors; - unsigned int total_bad = 0; - int i; - TRACE_FUN(ft_t_flow); - - if (ft_format_code == fmt_big || - ft_format_code == fmt_var || - ft_format_code == fmt_1100ft) { - SectorCount *ptr = (SectorCount *)bad_sector_map; - unsigned int sector; - __u16 *ptr16; - - while((sector = get_sector(ptr++)) != 0) { - if ((ft_format_code == fmt_big || - ft_format_code == fmt_var) && - sector & 0x800000) { - total_bad += FT_SECTORS_PER_SEGMENT - 3; - TRACE(ft_t_noise, "bad segment at sector: %6d", - sector & 0x7fffff); - } else { - ++total_bad; - TRACE(ft_t_noise, "bad sector: %6d", sector); - } - } - /* Display old ftape's end-of-file marks - */ - ptr16 = (__u16*)ptr; - while ((sector = get_unaligned(ptr16++)) != 0) { - TRACE(ft_t_noise, "Old ftape eof mark: %4d/%2d", - sector, get_unaligned(ptr16++)); - } - } else { /* fixed size format */ - for (i = ft_first_data_segment; - i < (int)(ft_segments_per_track * ft_tracks_per_tape); ++i) { - SectorMap map = ((SectorMap *) bad_sector_map)[i]; - - if (map) { - TRACE(ft_t_noise, - "bsm for segment %4d: 0x%08x", i, (unsigned int)map); - total_bad += ((map == EMPTY_SEGMENT) - ? FT_SECTORS_PER_SEGMENT - 3 - : count_ones(map)); - } - } - } - good_sectors = - ((ft_segments_per_track * ft_tracks_per_tape - ft_first_data_segment) - * (FT_SECTORS_PER_SEGMENT - 3)) - total_bad; - TRACE(ft_t_info, "%d Kb usable on this tape", good_sectors); - if (total_bad == 0) { - TRACE(ft_t_info, - "WARNING: this tape has no bad blocks registered !"); - } else { - TRACE(ft_t_info, "%d bad sectors", total_bad); - } - TRACE_EXIT; -} - - -void ftape_extract_bad_sector_map(__u8 * buffer) -{ - TRACE_FUN(ft_t_any); - - /* Fill the bad sector map with the contents of buffer. - */ - if (ft_format_code == fmt_var || ft_format_code == fmt_big) { - /* QIC-3010/3020 and wide QIC-80 tapes no longer have a failed - * sector log but use this area to extend the bad sector map. - */ - bad_sector_map = &buffer[FT_HEADER_END]; - } else { - /* non-wide QIC-80 tapes have a failed sector log area that - * mustn't be included in the bad sector map. - */ - bad_sector_map = &buffer[FT_FSL + FT_FSL_SIZE]; - } - if (ft_format_code == fmt_1100ft || - ft_format_code == fmt_var || - ft_format_code == fmt_big) { - bsm_hash_ptr = (SectorCount *)bad_sector_map; - } else { - bsm_hash_ptr = NULL; - } - bsm_debug_fake(); - if (TRACE_LEVEL >= ft_t_info) { - print_bad_sector_map(); - } - TRACE_EXIT; -} - -static inline SectorMap cvt2map(unsigned int sector) -{ - return 1 << (((sector & 0x7fffff) - 1) % FT_SECTORS_PER_SEGMENT); -} - -static inline int cvt2segment(unsigned int sector) -{ - return ((sector & 0x7fffff) - 1) / FT_SECTORS_PER_SEGMENT; -} - -static int forward_seek_entry(int segment_id, - SectorCount **ptr, - SectorMap *map) -{ - unsigned int sector; - int segment; - - do { - sector = get_sector((*ptr)++); - segment = cvt2segment(sector); - } while (sector != 0 && segment < segment_id); - (*ptr) --; /* point to first sector >= segment_id */ - /* Get all sectors in segment_id - */ - if (sector == 0 || segment != segment_id) { - *map = 0; - return 0; - } else if ((sector & 0x800000) && - (ft_format_code == fmt_var || ft_format_code == fmt_big)) { - *map = EMPTY_SEGMENT; - return FT_SECTORS_PER_SEGMENT; - } else { - int count = 1; - SectorCount *tmp_ptr = (*ptr) + 1; - - *map = cvt2map(sector); - while ((sector = get_sector(tmp_ptr++)) != 0 && - (segment = cvt2segment(sector)) == segment_id) { - *map |= cvt2map(sector); - ++count; - } - return count; - } -} - -static int backwards_seek_entry(int segment_id, - SectorCount **ptr, - SectorMap *map) -{ - unsigned int sector; - int segment; /* max unsigned int */ - - if (*ptr <= (SectorCount *)bad_sector_map) { - *map = 0; - return 0; - } - do { - sector = get_sector(--(*ptr)); - segment = cvt2segment(sector); - } while (*ptr > (SectorCount *)bad_sector_map && segment > segment_id); - if (segment > segment_id) { /* at start of list, no entry found */ - *map = 0; - return 0; - } else if (segment < segment_id) { - /* before smaller entry, adjust for overshoot */ - (*ptr) ++; - *map = 0; - return 0; - } else if ((sector & 0x800000) && - (ft_format_code == fmt_big || ft_format_code == fmt_var)) { - *map = EMPTY_SEGMENT; - return FT_SECTORS_PER_SEGMENT; - } else { /* get all sectors in segment_id */ - int count = 1; - - *map = cvt2map(sector); - while(*ptr > (SectorCount *)bad_sector_map) { - sector = get_sector(--(*ptr)); - segment = cvt2segment(sector); - if (segment != segment_id) { - break; - } - *map |= cvt2map(sector); - ++count; - } - if (segment < segment_id) { - (*ptr) ++; - } - return count; - } -} - -#if 0 -static void ftape_put_bad_sector_entry(int segment_id, SectorMap new_map) -{ - SectorCount *ptr = (SectorCount *)bad_sector_map; - int count; - int new_count; - SectorMap map; - TRACE_FUN(ft_t_any); - - if (ft_format_code == fmt_1100ft || - ft_format_code == fmt_var || - ft_format_code == fmt_big) { - count = forward_seek_entry(segment_id, &ptr, &map); - new_count = count_ones(new_map); - /* If format code == 4 put empty segment instead of 32 - * bad sectors. - */ - if (ft_format_code == fmt_var || ft_format_code == fmt_big) { - if (new_count == FT_SECTORS_PER_SEGMENT) { - new_count = 1; - } - if (count == FT_SECTORS_PER_SEGMENT) { - count = 1; - } - } - if (count != new_count) { - /* insert (or delete if < 0) new_count - count - * entries. Move trailing part of list - * including terminating 0. - */ - SectorCount *hi_ptr = ptr; - - do { - } while (get_sector(hi_ptr++) != 0); - /* Note: ptr is of type byte *, and each bad sector - * consumes 3 bytes. - */ - memmove(ptr + new_count, ptr + count, - (size_t)(hi_ptr - (ptr + count))*sizeof(SectorCount)); - } - TRACE(ft_t_noise, "putting map 0x%08x at %p, segment %d", - (unsigned int)new_map, ptr, segment_id); - if (new_count == 1 && new_map == EMPTY_SEGMENT) { - put_sector(ptr++, (0x800001 + - segment_id * - FT_SECTORS_PER_SEGMENT)); - } else { - int i = 0; - - while (new_map) { - if (new_map & 1) { - put_sector(ptr++, - 1 + segment_id * - FT_SECTORS_PER_SEGMENT + i); - } - ++i; - new_map >>= 1; - } - } - } else { - ((SectorMap *) bad_sector_map)[segment_id] = new_map; - } - TRACE_EXIT; -} -#endif /* 0 */ - -SectorMap ftape_get_bad_sector_entry(int segment_id) -{ - if (ft_used_header_segment == -1) { - /* When reading header segment we'll need a blank map. - */ - return 0; - } else if (bsm_hash_ptr != NULL) { - /* Invariants: - * map - mask value returned on last call. - * bsm_hash_ptr - points to first sector greater or equal to - * first sector in last_referenced segment. - * last_referenced - segment id used in the last call, - * sector and map belong to this id. - * This code is designed for sequential access and retries. - * For true random access it may have to be redesigned. - */ - static int last_reference = -1; - static SectorMap map; - - if (segment_id > last_reference) { - /* Skip all sectors before segment_id - */ - forward_seek_entry(segment_id, &bsm_hash_ptr, &map); - } else if (segment_id < last_reference) { - /* Skip backwards until begin of buffer or - * first sector in segment_id - */ - backwards_seek_entry(segment_id, &bsm_hash_ptr, &map); - } /* segment_id == last_reference : keep map */ - last_reference = segment_id; - return map; - } else { - return ((SectorMap *) bad_sector_map)[segment_id]; - } -} - -/* This is simply here to prevent us from overwriting other kernel - * data. Writes will result in NULL Pointer dereference. - */ -void ftape_init_bsm(void) -{ - bad_sector_map = NULL; - bsm_hash_ptr = NULL; -} diff --git a/drivers/char/ftape/lowlevel/ftape-bsm.h b/drivers/char/ftape/lowlevel/ftape-bsm.h deleted file mode 100644 index ed45465af4d..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-bsm.h +++ /dev/null @@ -1,66 +0,0 @@ -#ifndef _FTAPE_BSM_H -#define _FTAPE_BSM_H - -/* - * Copyright (C) 1994-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-bsm.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:07 $ - * - * This file contains definitions for the bad sector map handling - * routines for the QIC-117 floppy-tape driver for Linux. - */ - -#include -#include - -#define EMPTY_SEGMENT (0xffffffff) -#define FAKE_SEGMENT (0xfffffffe) - -/* maximum (format code 4) bad sector map size (bytes). - */ -#define BAD_SECTOR_MAP_SIZE (29 * SECTOR_SIZE - 256) - -/* format code 4 bad sector entry, ftape uses this - * internally for all format codes - */ -typedef __u32 SectorMap; -/* variable and 1100 ft bad sector map entry. These three bytes represent - * a single sector address measured from BOT. - */ -typedef struct NewSectorMap { - __u8 bytes[3]; -} SectorCount; - - -/* - * ftape-bsm.c defined global vars. - */ - -/* - * ftape-bsm.c defined global functions. - */ -extern void update_bad_sector_map(__u8 * buffer); -extern void ftape_extract_bad_sector_map(__u8 * buffer); -extern SectorMap ftape_get_bad_sector_entry(int segment_id); -extern __u8 *ftape_find_end_of_bsm_list(__u8 * address); -extern void ftape_init_bsm(void); - -#endif diff --git a/drivers/char/ftape/lowlevel/ftape-buffer.c b/drivers/char/ftape/lowlevel/ftape-buffer.c deleted file mode 100644 index c706ff16277..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-buffer.c +++ /dev/null @@ -1,130 +0,0 @@ -/* - * Copyright (C) 1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-buffer.c,v $ - * $Revision: 1.3 $ - * $Date: 1997/10/16 23:33:11 $ - * - * This file contains the allocator/dealloctor for ftape's dynamic dma - * buffer. - */ - -#include -#include -#include -#include - -#include -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/ftape-read.h" -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-buffer.h" - -/* DMA'able memory allocation stuff. - */ - -static inline void *dmaalloc(size_t size) -{ - unsigned long addr; - - if (size == 0) { - return NULL; - } - addr = __get_dma_pages(GFP_KERNEL, get_order(size)); - if (addr) { - struct page *page; - - for (page = virt_to_page(addr); page < virt_to_page(addr+size); page++) - SetPageReserved(page); - } - return (void *)addr; -} - -static inline void dmafree(void *addr, size_t size) -{ - if (size > 0) { - struct page *page; - - for (page = virt_to_page((unsigned long)addr); - page < virt_to_page((unsigned long)addr+size); page++) - ClearPageReserved(page); - free_pages((unsigned long) addr, get_order(size)); - } -} - -static int add_one_buffer(void) -{ - TRACE_FUN(ft_t_flow); - - if (ft_nr_buffers >= FT_MAX_NR_BUFFERS) { - TRACE_EXIT -ENOMEM; - } - ft_buffer[ft_nr_buffers] = kmalloc(sizeof(buffer_struct), GFP_KERNEL); - if (ft_buffer[ft_nr_buffers] == NULL) { - TRACE_EXIT -ENOMEM; - } - memset(ft_buffer[ft_nr_buffers], 0, sizeof(buffer_struct)); - ft_buffer[ft_nr_buffers]->address = dmaalloc(FT_BUFF_SIZE); - if (ft_buffer[ft_nr_buffers]->address == NULL) { - kfree(ft_buffer[ft_nr_buffers]); - ft_buffer[ft_nr_buffers] = NULL; - TRACE_EXIT -ENOMEM; - } - ft_nr_buffers ++; - TRACE(ft_t_info, "buffer nr #%d @ %p, dma area @ %p", - ft_nr_buffers, - ft_buffer[ft_nr_buffers-1], - ft_buffer[ft_nr_buffers-1]->address); - TRACE_EXIT 0; -} - -static void del_one_buffer(void) -{ - TRACE_FUN(ft_t_flow); - if (ft_nr_buffers > 0) { - TRACE(ft_t_info, "releasing buffer nr #%d @ %p, dma area @ %p", - ft_nr_buffers, - ft_buffer[ft_nr_buffers-1], - ft_buffer[ft_nr_buffers-1]->address); - ft_nr_buffers --; - dmafree(ft_buffer[ft_nr_buffers]->address, FT_BUFF_SIZE); - kfree(ft_buffer[ft_nr_buffers]); - ft_buffer[ft_nr_buffers] = NULL; - } - TRACE_EXIT; -} - -int ftape_set_nr_buffers(int cnt) -{ - int delta = cnt - ft_nr_buffers; - TRACE_FUN(ft_t_flow); - - if (delta > 0) { - while (delta--) { - if (add_one_buffer() < 0) { - TRACE_EXIT -ENOMEM; - } - } - } else if (delta < 0) { - while (delta++) { - del_one_buffer(); - } - } - ftape_zap_read_buffers(); - TRACE_EXIT 0; -} diff --git a/drivers/char/ftape/lowlevel/ftape-buffer.h b/drivers/char/ftape/lowlevel/ftape-buffer.h deleted file mode 100644 index eec99cee8f8..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-buffer.h +++ /dev/null @@ -1,32 +0,0 @@ -#ifndef _FTAPE_BUFFER_H -#define _FTAPE_BUFFER_H - -/* - * Copyright (C) 1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-buffer.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:08 $ - * - * This file contains the allocator/dealloctor for ftape's dynamic dma - * buffer. - */ - -extern int ftape_set_nr_buffers(int cnt); - -#endif diff --git a/drivers/char/ftape/lowlevel/ftape-calibr.c b/drivers/char/ftape/lowlevel/ftape-calibr.c deleted file mode 100644 index 8e50bfd35a5..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-calibr.c +++ /dev/null @@ -1,275 +0,0 @@ -/* - * Copyright (C) 1993-1996 Bas Laarhoven. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-calibr.c,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:08 $ - * - * GP calibration routine for processor speed dependent - * functions. - */ - -#include -#include -#include -#include -#if defined(__alpha__) -# include -#elif defined(__x86_64__) -# include -# include -#elif defined(__i386__) -# include -#endif -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-calibr.h" -#include "../lowlevel/fdc-io.h" - -#undef DEBUG - -#if !defined(__alpha__) && !defined(__i386__) && !defined(__x86_64__) -# error Ftape is not implemented for this architecture! -#endif - -#if defined(__alpha__) || defined(__x86_64__) -static unsigned long ps_per_cycle = 0; -#endif - -static spinlock_t calibr_lock; - -/* - * Note: On Intel PCs, the clock ticks at 100 Hz (HZ==100) which is - * too slow for certain timeouts (and that clock doesn't even tick - * when interrupts are disabled). For that reason, the 8254 timer is - * used directly to implement fine-grained timeouts. However, on - * Alpha PCs, the 8254 is *not* used to implement the clock tick - * (which is 1024 Hz, normally) and the 8254 timer runs at some - * "random" frequency (it seems to run at 18Hz, but it's not safe to - * rely on this value). Instead, we use the Alpha's "rpcc" - * instruction to read cycle counts. As this is a 32 bit counter, - * it will overflow only once per 30 seconds (on a 200MHz machine), - * which is plenty. - */ - -unsigned int ftape_timestamp(void) -{ -#if defined(__alpha__) - unsigned long r; - - asm volatile ("rpcc %0" : "=r" (r)); - return r; -#elif defined(__x86_64__) - unsigned long r; - rdtscl(r); - return r; -#elif defined(__i386__) - -/* - * Note that there is some time between counter underflowing and jiffies - * increasing, so the code below won't always give correct output. - * -Vojtech - */ - - unsigned long flags; - __u16 lo; - __u16 hi; - - spin_lock_irqsave(&calibr_lock, flags); - outb_p(0x00, 0x43); /* latch the count ASAP */ - lo = inb_p(0x40); /* read the latched count */ - lo |= inb(0x40) << 8; - hi = jiffies; - spin_unlock_irqrestore(&calibr_lock, flags); - return ((hi + 1) * (unsigned int) LATCH) - lo; /* downcounter ! */ -#endif -} - -static unsigned int short_ftape_timestamp(void) -{ -#if defined(__alpha__) || defined(__x86_64__) - return ftape_timestamp(); -#elif defined(__i386__) - unsigned int count; - unsigned long flags; - - spin_lock_irqsave(&calibr_lock, flags); - outb_p(0x00, 0x43); /* latch the count ASAP */ - count = inb_p(0x40); /* read the latched count */ - count |= inb(0x40) << 8; - spin_unlock_irqrestore(&calibr_lock, flags); - return (LATCH - count); /* normal: downcounter */ -#endif -} - -static unsigned int diff(unsigned int t0, unsigned int t1) -{ -#if defined(__alpha__) || defined(__x86_64__) - return (t1 - t0); -#elif defined(__i386__) - /* - * This is tricky: to work for both short and full ftape_timestamps - * we'll have to discriminate between these. - * If it _looks_ like short stamps with wrapping around we'll - * asume it are. This will generate a small error if it really - * was a (very large) delta from full ftape_timestamps. - */ - return (t1 <= t0 && t0 <= LATCH) ? t1 + LATCH - t0 : t1 - t0; -#endif -} - -static unsigned int usecs(unsigned int count) -{ -#if defined(__alpha__) || defined(__x86_64__) - return (ps_per_cycle * count) / 1000000UL; -#elif defined(__i386__) - return (10000 * count) / ((CLOCK_TICK_RATE + 50) / 100); -#endif -} - -unsigned int ftape_timediff(unsigned int t0, unsigned int t1) -{ - /* - * Calculate difference in usec for ftape_timestamp results t0 & t1. - * Note that on the i386 platform with short time-stamps, the - * maximum allowed timespan is 1/HZ or we'll lose ticks! - */ - return usecs(diff(t0, t1)); -} - -/* To get an indication of the I/O performance, - * measure the duration of the inb() function. - */ -static void time_inb(void) -{ - int i; - int t0, t1; - unsigned long flags; - int status; - TRACE_FUN(ft_t_any); - - spin_lock_irqsave(&calibr_lock, flags); - t0 = short_ftape_timestamp(); - for (i = 0; i < 1000; ++i) { - status = inb(fdc.msr); - } - t1 = short_ftape_timestamp(); - spin_unlock_irqrestore(&calibr_lock, flags); - TRACE(ft_t_info, "inb() duration: %d nsec", ftape_timediff(t0, t1)); - TRACE_EXIT; -} - -static void init_clock(void) -{ - TRACE_FUN(ft_t_any); - -#if defined(__x86_64__) - ps_per_cycle = 1000000000UL / cpu_khz; -#elif defined(__alpha__) - extern struct hwrpb_struct *hwrpb; - ps_per_cycle = (1000*1000*1000*1000UL) / hwrpb->cycle_freq; -#endif - TRACE_EXIT; -} - -/* - * Input: function taking int count as parameter. - * pointers to calculated calibration variables. - */ -void ftape_calibrate(char *name, - void (*fun) (unsigned int), - unsigned int *calibr_count, - unsigned int *calibr_time) -{ - static int first_time = 1; - int i; - unsigned int tc = 0; - unsigned int count; - unsigned int time; -#if defined(__i386__) - unsigned int old_tc = 0; - unsigned int old_count = 1; - unsigned int old_time = 1; -#endif - TRACE_FUN(ft_t_flow); - - if (first_time) { /* get idea of I/O performance */ - init_clock(); - time_inb(); - first_time = 0; - } - /* value of timeout must be set so that on very slow systems - * it will give a time less than one jiffy, and on - * very fast systems it'll give reasonable precision. - */ - - count = 40; - for (i = 0; i < 15; ++i) { - unsigned int t0; - unsigned int t1; - unsigned int once; - unsigned int multiple; - unsigned long flags; - - *calibr_count = - *calibr_time = count; /* set TC to 1 */ - spin_lock_irqsave(&calibr_lock, flags); - fun(0); /* dummy, get code into cache */ - t0 = short_ftape_timestamp(); - fun(0); /* overhead + one test */ - t1 = short_ftape_timestamp(); - once = diff(t0, t1); - t0 = short_ftape_timestamp(); - fun(count); /* overhead + count tests */ - t1 = short_ftape_timestamp(); - multiple = diff(t0, t1); - spin_unlock_irqrestore(&calibr_lock, flags); - time = ftape_timediff(0, multiple - once); - tc = (1000 * time) / (count - 1); - TRACE(ft_t_any, "once:%3d us,%6d times:%6d us, TC:%5d ns", - usecs(once), count - 1, usecs(multiple), tc); -#if defined(__alpha__) || defined(__x86_64__) - /* - * Increase the calibration count exponentially until the - * calibration time exceeds 100 ms. - */ - if (time >= 100*1000) { - break; - } -#elif defined(__i386__) - /* - * increase the count until the resulting time nears 2/HZ, - * then the tc will drop sharply because we lose LATCH counts. - */ - if (tc <= old_tc / 2) { - time = old_time; - count = old_count; - break; - } - old_tc = tc; - old_count = count; - old_time = time; -#endif - count *= 2; - } - *calibr_count = count - 1; - *calibr_time = time; - TRACE(ft_t_info, "TC for `%s()' = %d nsec (at %d counts)", - name, (1000 * *calibr_time) / *calibr_count, *calibr_count); - TRACE_EXIT; -} diff --git a/drivers/char/ftape/lowlevel/ftape-calibr.h b/drivers/char/ftape/lowlevel/ftape-calibr.h deleted file mode 100644 index 0c7e75246c7..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-calibr.h +++ /dev/null @@ -1,37 +0,0 @@ -#ifndef _FTAPE_CALIBR_H -#define _FTAPE_CALIBR_H - -/* - * Copyright (C) 1993-1996 Bas Laarhoven. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-calibr.h,v $ - * $Revision: 1.1 $ - * $Date: 1997/09/19 09:05:26 $ - * - * This file contains a gp calibration routine for - * hardware dependent timeout functions. - */ - -extern void ftape_calibrate(char *name, - void (*fun) (unsigned int), - unsigned int *calibr_count, - unsigned int *calibr_time); -extern unsigned int ftape_timestamp(void); -extern unsigned int ftape_timediff(unsigned int t0, unsigned int t1); - -#endif /* _FTAPE_CALIBR_H */ diff --git a/drivers/char/ftape/lowlevel/ftape-ctl.c b/drivers/char/ftape/lowlevel/ftape-ctl.c deleted file mode 100644 index 5d7c1ce92d5..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-ctl.c +++ /dev/null @@ -1,896 +0,0 @@ -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-ctl.c,v $ - * $Revision: 1.4 $ - * $Date: 1997/11/11 14:37:44 $ - * - * This file contains the non-read/write ftape functions for the - * QIC-40/80/3010/3020 floppy-tape driver "ftape" for Linux. - */ - -#include -#include -#include - -#include -#include -#include -#include - -/* ease porting between pre-2.4.x and later kernels */ -#define vma_get_pgoff(v) ((v)->vm_pgoff) - -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-write.h" -#include "../lowlevel/ftape-read.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/ftape-bsm.h" - -/* Global vars. - */ -ftape_info ftape_status = { -/* vendor information */ - { 0, }, /* drive type */ -/* data rates */ - 500, /* used data rate */ - 500, /* drive max rate */ - 500, /* fdc max rate */ -/* drive selection, either FTAPE_SEL_A/B/C/D */ - -1, /* drive selection */ -/* flags set after decode the drive and tape status */ - 0, /* formatted */ - 1, /* no tape */ - 1, /* write protected */ - 1, /* new tape */ -/* values of last queried drive/tape status and error */ - {{0,}}, /* last error code */ - {{0,}}, /* drive status, configuration, tape status */ -/* cartridge geometry */ - 20, /* tracks_per_tape */ - 102, /* segments_per_track */ -/* location of header segments, etc. */ - -1, /* used_header_segment */ - -1, /* header_segment_1 */ - -1, /* header_segment_2 */ - -1, /* first_data_segment */ - -1, /* last_data_segment */ -/* the format code as stored in the header segment */ - fmt_normal, /* format code */ -/* the default for the qic std: unknown */ - -1, -/* is tape running? */ - idle, /* runner_state */ -/* is tape reading/writing/verifying/formatting/deleting */ - idle, /* driver state */ -/* flags fatal hardware error */ - 1, /* failure */ -/* history record */ - { 0, } /* history record */ -}; - -int ftape_segments_per_head = 1020; -int ftape_segments_per_cylinder = 4; -int ftape_init_drive_needed = 1; /* need to be global for ftape_reset_drive() - * in ftape-io.c - */ - -/* Local vars. - */ -static const vendor_struct vendors[] = QIC117_VENDORS; -static const wakeup_method methods[] = WAKEUP_METHODS; - -const ftape_info *ftape_get_status(void) -{ -#if defined(STATUS_PARANOYA) - static ftape_info get_status; - - get_status = ftape_status; - return &get_status; -#else - return &ftape_status; /* maybe return only a copy of it to assure - * read only access - */ -#endif -} - -static int ftape_not_operational(int status) -{ - /* return true if status indicates tape can not be used. - */ - return ((status ^ QIC_STATUS_CARTRIDGE_PRESENT) & - (QIC_STATUS_ERROR | - QIC_STATUS_CARTRIDGE_PRESENT | - QIC_STATUS_NEW_CARTRIDGE)); -} - -int ftape_seek_to_eot(void) -{ - int status; - TRACE_FUN(ft_t_any); - - TRACE_CATCH(ftape_ready_wait(ftape_timeout.pause, &status),); - while ((status & QIC_STATUS_AT_EOT) == 0) { - if (ftape_not_operational(status)) { - TRACE_EXIT -EIO; - } - TRACE_CATCH(ftape_command_wait(QIC_PHYSICAL_FORWARD, - ftape_timeout.rewind,&status),); - } - TRACE_EXIT 0; -} - -int ftape_seek_to_bot(void) -{ - int status; - TRACE_FUN(ft_t_any); - - TRACE_CATCH(ftape_ready_wait(ftape_timeout.pause, &status),); - while ((status & QIC_STATUS_AT_BOT) == 0) { - if (ftape_not_operational(status)) { - TRACE_EXIT -EIO; - } - TRACE_CATCH(ftape_command_wait(QIC_PHYSICAL_REVERSE, - ftape_timeout.rewind,&status),); - } - TRACE_EXIT 0; -} - -static int ftape_new_cartridge(void) -{ - ft_location.track = -1; /* force seek on first access */ - ftape_zap_read_buffers(); - ftape_zap_write_buffers(); - return 0; -} - -int ftape_abort_operation(void) -{ - int result = 0; - int status; - TRACE_FUN(ft_t_flow); - - if (ft_runner_status == running) { - TRACE(ft_t_noise, "aborting runner, waiting"); - - ft_runner_status = do_abort; - /* set timeout so that the tape will run to logical EOT - * if we missed the last sector and there are no queue pulses. - */ - result = ftape_dumb_stop(); - } - if (ft_runner_status != idle) { - if (ft_runner_status == do_abort) { - TRACE(ft_t_noise, "forcing runner abort"); - } - TRACE(ft_t_noise, "stopping tape"); - result = ftape_stop_tape(&status); - ft_location.known = 0; - ft_runner_status = idle; - } - ftape_reset_buffer(); - ftape_zap_read_buffers(); - ftape_set_state(idle); - TRACE_EXIT result; -} - -static int lookup_vendor_id(unsigned int vendor_id) -{ - int i = 0; - - while (vendors[i].vendor_id != vendor_id) { - if (++i >= NR_ITEMS(vendors)) { - return -1; - } - } - return i; -} - -static void ftape_detach_drive(void) -{ - TRACE_FUN(ft_t_any); - - TRACE(ft_t_flow, "disabling tape drive and fdc"); - ftape_put_drive_to_sleep(ft_drive_type.wake_up); - fdc_catch_stray_interrupts(1); /* one always comes */ - fdc_disable(); - fdc_release_irq_and_dma(); - fdc_release_regions(); - TRACE_EXIT; -} - -static void clear_history(void) -{ - ft_history.used = 0; - ft_history.id_am_errors = - ft_history.id_crc_errors = - ft_history.data_am_errors = - ft_history.data_crc_errors = - ft_history.overrun_errors = - ft_history.no_data_errors = - ft_history.retries = - ft_history.crc_errors = - ft_history.crc_failures = - ft_history.ecc_failures = - ft_history.corrected = - ft_history.defects = - ft_history.rewinds = 0; -} - -static int ftape_activate_drive(vendor_struct * drive_type) -{ - int result = 0; - TRACE_FUN(ft_t_flow); - - /* If we already know the drive type, wake it up. - * Else try to find out what kind of drive is attached. - */ - if (drive_type->wake_up != unknown_wake_up) { - TRACE(ft_t_flow, "enabling tape drive and fdc"); - result = ftape_wakeup_drive(drive_type->wake_up); - if (result < 0) { - TRACE(ft_t_err, "known wakeup method failed"); - } - } else { - wake_up_types method; - const ft_trace_t old_tracing = TRACE_LEVEL; - if (TRACE_LEVEL < ft_t_flow) { - SET_TRACE_LEVEL(ft_t_bug); - } - - /* Try to awaken the drive using all known methods. - * Lower tracing for a while. - */ - for (method=no_wake_up; method < NR_ITEMS(methods); ++method) { - drive_type->wake_up = method; -#ifdef CONFIG_FT_TWO_DRIVES - /* Test setup for dual drive configuration. - * /dev/rft2 uses mountain wakeup - * /dev/rft3 uses colorado wakeup - * Other systems will use the normal scheme. - */ - if ((ft_drive_sel < 2) || - (ft_drive_sel == 2 && method == FT_WAKE_UP_1) || - (ft_drive_sel == 3 && method == FT_WAKE_UP_2)) { - result=ftape_wakeup_drive(drive_type->wake_up); - } else { - result = -EIO; - } -#else - result = ftape_wakeup_drive(drive_type->wake_up); -#endif - if (result >= 0) { - TRACE(ft_t_warn, "drive wakeup method: %s", - methods[drive_type->wake_up].name); - break; - } - } - SET_TRACE_LEVEL(old_tracing); - - if (method >= NR_ITEMS(methods)) { - /* no response at all, cannot open this drive */ - drive_type->wake_up = unknown_wake_up; - TRACE(ft_t_err, "no tape drive found !"); - result = -ENODEV; - } - } - TRACE_EXIT result; -} - -static int ftape_get_drive_status(void) -{ - int result; - int status; - TRACE_FUN(ft_t_flow); - - ft_no_tape = ft_write_protected = 0; - /* Tape drive is activated now. - * First clear error status if present. - */ - do { - result = ftape_ready_wait(ftape_timeout.reset, &status); - if (result < 0) { - if (result == -ETIME) { - TRACE(ft_t_err, "ftape_ready_wait timeout"); - } else if (result == -EINTR) { - TRACE(ft_t_err, "ftape_ready_wait aborted"); - } else { - TRACE(ft_t_err, "ftape_ready_wait failed"); - } - TRACE_EXIT -EIO; - } - /* Clear error condition (drive is ready !) - */ - if (status & QIC_STATUS_ERROR) { - unsigned int error; - qic117_cmd_t command; - - TRACE(ft_t_err, "error status set"); - result = ftape_report_error(&error, &command, 1); - if (result < 0) { - TRACE(ft_t_err, - "report_error_code failed: %d", result); - /* hope it's working next time */ - ftape_reset_drive(); - TRACE_EXIT -EIO; - } else if (error != 0) { - TRACE(ft_t_noise, "error code : %d", error); - TRACE(ft_t_noise, "error command: %d", command); - } - } - if (status & QIC_STATUS_NEW_CARTRIDGE) { - unsigned int error; - qic117_cmd_t command; - const ft_trace_t old_tracing = TRACE_LEVEL; - SET_TRACE_LEVEL(ft_t_bug); - - /* Undocumented feature: Must clear (not present!) - * error here or we'll fail later. - */ - ftape_report_error(&error, &command, 1); - - SET_TRACE_LEVEL(old_tracing); - TRACE(ft_t_info, "status: new cartridge"); - ft_new_tape = 1; - } else { - ft_new_tape = 0; - } - FT_SIGNAL_EXIT(_DONT_BLOCK); - } while (status & QIC_STATUS_ERROR); - - ft_no_tape = !(status & QIC_STATUS_CARTRIDGE_PRESENT); - ft_write_protected = (status & QIC_STATUS_WRITE_PROTECT) != 0; - if (ft_no_tape) { - TRACE(ft_t_warn, "no cartridge present"); - } else { - if (ft_write_protected) { - TRACE(ft_t_noise, "Write protected cartridge"); - } - } - TRACE_EXIT 0; -} - -static void ftape_log_vendor_id(void) -{ - int vendor_index; - TRACE_FUN(ft_t_flow); - - ftape_report_vendor_id(&ft_drive_type.vendor_id); - vendor_index = lookup_vendor_id(ft_drive_type.vendor_id); - if (ft_drive_type.vendor_id == UNKNOWN_VENDOR && - ft_drive_type.wake_up == wake_up_colorado) { - vendor_index = 0; - /* hack to get rid of all this mail */ - ft_drive_type.vendor_id = 0; - } - if (vendor_index < 0) { - /* Unknown vendor id, first time opening device. The - * drive_type remains set to type found at wakeup - * time, this will probably keep the driver operating - * for this new vendor. - */ - TRACE(ft_t_warn, "\n" - KERN_INFO "============ unknown vendor id ===========\n" - KERN_INFO "A new, yet unsupported tape drive is found\n" - KERN_INFO "Please report the following values:\n" - KERN_INFO " Vendor id : 0x%04x\n" - KERN_INFO " Wakeup method : %s\n" - KERN_INFO "And a description of your tape drive\n" - KERN_INFO "to "THE_FTAPE_MAINTAINER"\n" - KERN_INFO "==========================================", - ft_drive_type.vendor_id, - methods[ft_drive_type.wake_up].name); - ft_drive_type.speed = 0; /* unknown */ - } else { - ft_drive_type.name = vendors[vendor_index].name; - ft_drive_type.speed = vendors[vendor_index].speed; - TRACE(ft_t_info, "tape drive type: %s", ft_drive_type.name); - /* scan all methods for this vendor_id in table */ - while(ft_drive_type.wake_up != vendors[vendor_index].wake_up) { - if (vendor_index < NR_ITEMS(vendors) - 1 && - vendors[vendor_index + 1].vendor_id - == - ft_drive_type.vendor_id) { - ++vendor_index; - } else { - break; - } - } - if (ft_drive_type.wake_up != vendors[vendor_index].wake_up) { - TRACE(ft_t_warn, "\n" - KERN_INFO "==========================================\n" - KERN_INFO "wakeup type mismatch:\n" - KERN_INFO "found: %s, expected: %s\n" - KERN_INFO "please report this to "THE_FTAPE_MAINTAINER"\n" - KERN_INFO "==========================================", - methods[ft_drive_type.wake_up].name, - methods[vendors[vendor_index].wake_up].name); - } - } - TRACE_EXIT; -} - -void ftape_calc_timeouts(unsigned int qic_std, - unsigned int data_rate, - unsigned int tape_len) -{ - int speed; /* deci-ips ! */ - int ff_speed; - int length; - TRACE_FUN(ft_t_any); - - /* tape transport speed - * data rate: QIC-40 QIC-80 QIC-3010 QIC-3020 - * - * 250 Kbps 25 ips n/a n/a n/a - * 500 Kbps 50 ips 34 ips 22.6 ips n/a - * 1 Mbps n/a 68 ips 45.2 ips 22.6 ips - * 2 Mbps n/a n/a n/a 45.2 ips - * - * fast tape transport speed is at least 68 ips. - */ - switch (qic_std) { - case QIC_TAPE_QIC40: - speed = (data_rate == 250) ? 250 : 500; - break; - case QIC_TAPE_QIC80: - speed = (data_rate == 500) ? 340 : 680; - break; - case QIC_TAPE_QIC3010: - speed = (data_rate == 500) ? 226 : 452; - break; - case QIC_TAPE_QIC3020: - speed = (data_rate == 1000) ? 226 : 452; - break; - default: - TRACE(ft_t_bug, "Unknown qic_std (bug) ?"); - speed = 500; - break; - } - if (ft_drive_type.speed == 0) { - unsigned long t0; - static int dt = 0; /* keep gcc from complaining */ - static int first_time = 1; - - /* Measure the time it takes to wind to EOT and back to BOT. - * If the tape length is known, calculate the rewind speed. - * Else keep the time value for calculation of the rewind - * speed later on, when the length _is_ known. - * Ask for a report only when length and speed are both known. - */ - if (first_time) { - ftape_seek_to_bot(); - t0 = jiffies; - ftape_seek_to_eot(); - ftape_seek_to_bot(); - dt = (int) (((jiffies - t0) * FT_USPT) / 1000); - if (dt < 1) { - dt = 1; /* prevent div by zero on failures */ - } - first_time = 0; - TRACE(ft_t_info, - "trying to determine seek timeout, got %d msec", - dt); - } - if (tape_len != 0) { - ft_drive_type.speed = - (2 * 12 * tape_len * 1000) / dt; - TRACE(ft_t_warn, "\n" - KERN_INFO "==========================================\n" - KERN_INFO "drive type: %s\n" - KERN_INFO "delta time = %d ms, length = %d ft\n" - KERN_INFO "has a maximum tape speed of %d ips\n" - KERN_INFO "please report this to "THE_FTAPE_MAINTAINER"\n" - KERN_INFO "==========================================", - ft_drive_type.name, dt, tape_len, - ft_drive_type.speed); - } - } - /* Handle unknown length tapes as very long ones. We'll - * determine the actual length from a header segment later. - * This is normal for all modern (Wide,TR1/2/3) formats. - */ - if (tape_len <= 0) { - TRACE(ft_t_noise, - "Unknown tape length, using maximal timeouts"); - length = QIC_TOP_TAPE_LEN; /* use worst case values */ - } else { - length = tape_len; /* use actual values */ - } - if (ft_drive_type.speed == 0) { - ff_speed = speed; - } else { - ff_speed = ft_drive_type.speed; - } - /* time to go from bot to eot at normal speed (data rate): - * time = (1+delta) * length (ft) * 12 (inch/ft) / speed (ips) - * delta = 10 % for seek speed, 20 % for rewind speed. - */ - ftape_timeout.seek = (length * 132 * FT_SECOND) / speed; - ftape_timeout.rewind = (length * 144 * FT_SECOND) / (10 * ff_speed); - ftape_timeout.reset = 20 * FT_SECOND + ftape_timeout.rewind; - TRACE(ft_t_noise, "timeouts for speed = %d, length = %d\n" - KERN_INFO "seek timeout : %d sec\n" - KERN_INFO "rewind timeout: %d sec\n" - KERN_INFO "reset timeout : %d sec", - speed, length, - (ftape_timeout.seek + 500) / 1000, - (ftape_timeout.rewind + 500) / 1000, - (ftape_timeout.reset + 500) / 1000); - TRACE_EXIT; -} - -/* This function calibrates the datarate (i.e. determines the maximal - * usable data rate) and sets the global variable ft_qic_std to qic_std - * - */ -int ftape_calibrate_data_rate(unsigned int qic_std) -{ - int rate = ft_fdc_rate_limit; - int result; - TRACE_FUN(ft_t_flow); - - ft_qic_std = qic_std; - - if (ft_qic_std == -1) { - TRACE_ABORT(-EIO, ft_t_err, - "Unable to determine data rate if QIC standard is unknown"); - } - - /* Select highest rate supported by both fdc and drive. - * Start with highest rate supported by the fdc. - */ - while (fdc_set_data_rate(rate) < 0 && rate > 250) { - rate /= 2; - } - TRACE(ft_t_info, - "Highest FDC supported data rate: %d Kbps", rate); - ft_fdc_max_rate = rate; - do { - result = ftape_set_data_rate(rate, ft_qic_std); - } while (result == -EINVAL && (rate /= 2) > 250); - if (result < 0) { - TRACE_ABORT(-EIO, ft_t_err, "set datarate failed"); - } - ft_data_rate = rate; - TRACE_EXIT 0; -} - -static int ftape_init_drive(void) -{ - int status; - qic_model model; - unsigned int qic_std; - unsigned int data_rate; - TRACE_FUN(ft_t_flow); - - ftape_init_drive_needed = 0; /* don't retry if this fails ? */ - TRACE_CATCH(ftape_report_raw_drive_status(&status),); - if (status & QIC_STATUS_CARTRIDGE_PRESENT) { - if (!(status & QIC_STATUS_AT_BOT)) { - /* Antique drives will get here after a soft reset, - * modern ones only if the driver is loaded when the - * tape wasn't rewound properly. - */ - /* Tape should be at bot if new cartridge ! */ - ftape_seek_to_bot(); - } - if (!(status & QIC_STATUS_REFERENCED)) { - TRACE(ft_t_flow, "starting seek_load_point"); - TRACE_CATCH(ftape_command_wait(QIC_SEEK_LOAD_POINT, - ftape_timeout.reset, - &status),); - } - } - ft_formatted = (status & QIC_STATUS_REFERENCED) != 0; - if (!ft_formatted) { - TRACE(ft_t_warn, "Warning: tape is not formatted !"); - } - - /* report configuration aborts when ftape_tape_len == -1 - * unknown qic_std is okay if not formatted. - */ - TRACE_CATCH(ftape_report_configuration(&model, - &data_rate, - &qic_std, - &ftape_tape_len),); - - /* Maybe add the following to the /proc entry - */ - TRACE(ft_t_info, "%s drive @ %d Kbps", - (model == prehistoric) ? "prehistoric" : - ((model == pre_qic117c) ? "pre QIC-117C" : - ((model == post_qic117b) ? "post QIC-117B" : - "post QIC-117D")), data_rate); - - if (ft_formatted) { - /* initialize ft_used_data_rate to maximum value - * and set ft_qic_std - */ - TRACE_CATCH(ftape_calibrate_data_rate(qic_std),); - if (ftape_tape_len == 0) { - TRACE(ft_t_info, "unknown length QIC-%s tape", - (ft_qic_std == QIC_TAPE_QIC40) ? "40" : - ((ft_qic_std == QIC_TAPE_QIC80) ? "80" : - ((ft_qic_std == QIC_TAPE_QIC3010) - ? "3010" : "3020"))); - } else { - TRACE(ft_t_info, "%d ft. QIC-%s tape", ftape_tape_len, - (ft_qic_std == QIC_TAPE_QIC40) ? "40" : - ((ft_qic_std == QIC_TAPE_QIC80) ? "80" : - ((ft_qic_std == QIC_TAPE_QIC3010) - ? "3010" : "3020"))); - } - ftape_calc_timeouts(ft_qic_std, ft_data_rate, ftape_tape_len); - /* soft write-protect QIC-40/QIC-80 cartridges used with a - * Colorado T3000 drive. Buggy hardware! - */ - if ((ft_drive_type.vendor_id == 0x011c6) && - ((ft_qic_std == QIC_TAPE_QIC40 || - ft_qic_std == QIC_TAPE_QIC80) && - !ft_write_protected)) { - TRACE(ft_t_warn, "\n" - KERN_INFO "The famous Colorado T3000 bug:\n" - KERN_INFO "%s drives can't write QIC40 and QIC80\n" - KERN_INFO "cartridges but don't set the write-protect flag!", - ft_drive_type.name); - ft_write_protected = 1; - } - } else { - /* Doesn't make too much sense to set the data rate - * because we don't know what to use for the write - * precompensation. - * Need to do this again when formatting the cartridge. - */ - ft_data_rate = data_rate; - ftape_calc_timeouts(QIC_TAPE_QIC40, - data_rate, - ftape_tape_len); - } - ftape_new_cartridge(); - TRACE_EXIT 0; -} - -static void ftape_munmap(void) -{ - int i; - TRACE_FUN(ft_t_flow); - - for (i = 0; i < ft_nr_buffers; i++) { - ft_buffer[i]->mmapped = 0; - } - TRACE_EXIT; -} - -/* Map the dma buffers into the virtual address range given by vma. - * We only check the caller doesn't map non-existent buffers. We - * don't check for multiple mappings. - */ -int ftape_mmap(struct vm_area_struct *vma) -{ - int num_buffers; - int i; - TRACE_FUN(ft_t_flow); - - if (ft_failure) { - TRACE_EXIT -ENODEV; - } - if (!(vma->vm_flags & (VM_READ|VM_WRITE))) { - TRACE_ABORT(-EINVAL, ft_t_err, "Undefined mmap() access"); - } - if (vma_get_pgoff(vma) != 0) { - TRACE_ABORT(-EINVAL, ft_t_err, "page offset must be 0"); - } - if ((vma->vm_end - vma->vm_start) % FT_BUFF_SIZE != 0) { - TRACE_ABORT(-EINVAL, ft_t_err, - "size = %ld, should be a multiple of %d", - vma->vm_end - vma->vm_start, - FT_BUFF_SIZE); - } - num_buffers = (vma->vm_end - vma->vm_start) / FT_BUFF_SIZE; - if (num_buffers > ft_nr_buffers) { - TRACE_ABORT(-EINVAL, - ft_t_err, "size = %ld, should be less than %d", - vma->vm_end - vma->vm_start, - ft_nr_buffers * FT_BUFF_SIZE); - } - if (ft_driver_state != idle) { - /* this also clears the buffer states - */ - ftape_abort_operation(); - } else { - ftape_reset_buffer(); - } - for (i = 0; i < num_buffers; i++) { - unsigned long pfn; - - pfn = virt_to_phys(ft_buffer[i]->address) >> PAGE_SHIFT; - TRACE_CATCH(remap_pfn_range(vma, vma->vm_start + - i * FT_BUFF_SIZE, - pfn, - FT_BUFF_SIZE, - vma->vm_page_prot), - _res = -EAGAIN); - TRACE(ft_t_noise, "remapped dma buffer @ %p to location @ %p", - ft_buffer[i]->address, - (void *)(vma->vm_start + i * FT_BUFF_SIZE)); - } - for (i = 0; i < num_buffers; i++) { - memset(ft_buffer[i]->address, 0xAA, FT_BUFF_SIZE); - ft_buffer[i]->mmapped++; - } - TRACE_EXIT 0; -} - -static void ftape_init_driver(void); /* forward declaration */ - -/* OPEN routine called by kernel-interface code - */ -int ftape_enable(int drive_selection) -{ - TRACE_FUN(ft_t_any); - - if (ft_drive_sel == -1 || ft_drive_sel != drive_selection) { - /* Other selection than last time - */ - ftape_init_driver(); - } - ft_drive_sel = FTAPE_SEL(drive_selection); - ft_failure = 0; - TRACE_CATCH(fdc_init(),); /* init & detect fdc */ - TRACE_CATCH(ftape_activate_drive(&ft_drive_type), - fdc_disable(); - fdc_release_irq_and_dma(); - fdc_release_regions()); - TRACE_CATCH(ftape_get_drive_status(), ftape_detach_drive()); - if (ft_drive_type.vendor_id == UNKNOWN_VENDOR) { - ftape_log_vendor_id(); - } - if (ft_new_tape) { - ftape_init_drive_needed = 1; - } - if (!ft_no_tape && ftape_init_drive_needed) { - TRACE_CATCH(ftape_init_drive(), ftape_detach_drive()); - } - ftape_munmap(); /* clear the mmap flag */ - clear_history(); - TRACE_EXIT 0; -} - -/* release routine called by the high level interface modules - * zftape or sftape. - */ -void ftape_disable(void) -{ - int i; - TRACE_FUN(ft_t_any); - - for (i = 0; i < ft_nr_buffers; i++) { - if (ft_buffer[i]->mmapped) { - TRACE(ft_t_noise, "first byte of buffer %d: 0x%02x", - i, *ft_buffer[i]->address); - } - } - if (sigtestsetmask(¤t->pending.signal, _DONT_BLOCK) && - !(sigtestsetmask(¤t->pending.signal, _NEVER_BLOCK)) && - ftape_tape_running) { - TRACE(ft_t_warn, - "Interrupted by fatal signal and tape still running"); - ftape_dumb_stop(); - ftape_abort_operation(); /* it's annoying */ - } else { - ftape_set_state(idle); - } - ftape_detach_drive(); - if (ft_history.used) { - TRACE(ft_t_info, "== Non-fatal errors this run: =="); - TRACE(ft_t_info, "fdc isr statistics:\n" - KERN_INFO " id_am_errors : %3d\n" - KERN_INFO " id_crc_errors : %3d\n" - KERN_INFO " data_am_errors : %3d\n" - KERN_INFO " data_crc_errors : %3d\n" - KERN_INFO " overrun_errors : %3d\n" - KERN_INFO " no_data_errors : %3d\n" - KERN_INFO " retries : %3d", - ft_history.id_am_errors, ft_history.id_crc_errors, - ft_history.data_am_errors, ft_history.data_crc_errors, - ft_history.overrun_errors, ft_history.no_data_errors, - ft_history.retries); - if (ft_history.used & 1) { - TRACE(ft_t_info, "ecc statistics:\n" - KERN_INFO " crc_errors : %3d\n" - KERN_INFO " crc_failures : %3d\n" - KERN_INFO " ecc_failures : %3d\n" - KERN_INFO " sectors corrected: %3d", - ft_history.crc_errors, ft_history.crc_failures, - ft_history.ecc_failures, ft_history.corrected); - } - if (ft_history.defects > 0) { - TRACE(ft_t_warn, "Warning: %d media defects!", - ft_history.defects); - } - if (ft_history.rewinds > 0) { - TRACE(ft_t_info, "tape motion statistics:\n" - KERN_INFO "repositions : %3d", - ft_history.rewinds); - } - } - ft_failure = 1; - TRACE_EXIT; -} - -static void ftape_init_driver(void) -{ - TRACE_FUN(ft_t_flow); - - ft_drive_type.vendor_id = UNKNOWN_VENDOR; - ft_drive_type.speed = 0; - ft_drive_type.wake_up = unknown_wake_up; - ft_drive_type.name = "Unknown"; - - ftape_timeout.seek = 650 * FT_SECOND; - ftape_timeout.reset = 670 * FT_SECOND; - ftape_timeout.rewind = 650 * FT_SECOND; - ftape_timeout.head_seek = 15 * FT_SECOND; - ftape_timeout.stop = 5 * FT_SECOND; - ftape_timeout.pause = 16 * FT_SECOND; - - ft_qic_std = -1; - ftape_tape_len = 0; /* unknown */ - ftape_current_command = 0; - ftape_current_cylinder = -1; - - ft_segments_per_track = 102; - ftape_segments_per_head = 1020; - ftape_segments_per_cylinder = 4; - ft_tracks_per_tape = 20; - - ft_failure = 1; - - ft_formatted = 0; - ft_no_tape = 1; - ft_write_protected = 1; - ft_new_tape = 1; - - ft_driver_state = idle; - - ft_data_rate = - ft_fdc_max_rate = 500; - ft_drive_max_rate = 0; /* triggers set_rate_test() */ - - ftape_init_drive_needed = 1; - - ft_header_segment_1 = -1; - ft_header_segment_2 = -1; - ft_used_header_segment = -1; - ft_first_data_segment = -1; - ft_last_data_segment = -1; - - ft_location.track = -1; - ft_location.known = 0; - - ftape_tape_running = 0; - ftape_might_be_off_track = 1; - - ftape_new_cartridge(); /* init some tape related variables */ - ftape_init_bsm(); - TRACE_EXIT; -} diff --git a/drivers/char/ftape/lowlevel/ftape-ctl.h b/drivers/char/ftape/lowlevel/ftape-ctl.h deleted file mode 100644 index 5f5e30bc361..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-ctl.h +++ /dev/null @@ -1,162 +0,0 @@ -#ifndef _FTAPE_CTL_H -#define _FTAPE_CTL_H - -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-ctl.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:09 $ - * - * This file contains the non-standard IOCTL related definitions - * for the QIC-40/80/3010/3020 floppy-tape driver "ftape" for - * Linux. - */ - -#include -#include -#include - -#include "../lowlevel/ftape-rw.h" -#include - -typedef struct { - int used; /* any reading or writing done */ - /* isr statistics */ - unsigned int id_am_errors; /* id address mark not found */ - unsigned int id_crc_errors; /* crc error in id address mark */ - unsigned int data_am_errors; /* data address mark not found */ - unsigned int data_crc_errors; /* crc error in data field */ - unsigned int overrun_errors; /* fdc access timing problem */ - unsigned int no_data_errors; /* sector not found */ - unsigned int retries; /* number of tape retries */ - /* ecc statistics */ - unsigned int crc_errors; /* crc error in data */ - unsigned int crc_failures; /* bad data without crc error */ - unsigned int ecc_failures; /* failed to correct */ - unsigned int corrected; /* total sectors corrected */ - /* general statistics */ - unsigned int rewinds; /* number of tape rewinds */ - unsigned int defects; /* bad sectors due to media defects */ -} history_record; - -/* this structure contains * ALL * information that we want - * our child modules to know about, but don't want them to - * modify. - */ -typedef struct { - /* vendor information */ - vendor_struct fti_drive_type; - /* data rates */ - unsigned int fti_used_data_rate; - unsigned int fti_drive_max_rate; - unsigned int fti_fdc_max_rate; - /* drive selection, either FTAPE_SEL_A/B/C/D */ - int fti_drive_sel; - /* flags set after decode the drive and tape status */ - unsigned int fti_formatted :1; - unsigned int fti_no_tape :1; - unsigned int fti_write_protected:1; - unsigned int fti_new_tape :1; - /* values of last queried drive/tape status and error */ - ft_drive_error fti_last_error; - ft_drive_status fti_last_status; - /* cartridge geometry */ - unsigned int fti_tracks_per_tape; - unsigned int fti_segments_per_track; - /* location of header segments, etc. */ - int fti_used_header_segment; - int fti_header_segment_1; - int fti_header_segment_2; - int fti_first_data_segment; - int fti_last_data_segment; - /* the format code as stored in the header segment */ - ft_format_type fti_format_code; - /* the following is the sole reason for the ftape_set_status() call */ - unsigned int fti_qic_std; - /* is tape running? */ - volatile enum runner_status_enum fti_runner_status; - /* is tape reading/writing/verifying/formatting/deleting */ - buffer_state_enum fti_state; - /* flags fatal hardware error */ - unsigned int fti_failure:1; - /* history record */ - history_record fti_history; -} ftape_info; - -/* vendor information */ -#define ft_drive_type ftape_status.fti_drive_type -/* data rates */ -#define ft_data_rate ftape_status.fti_used_data_rate -#define ft_drive_max_rate ftape_status.fti_drive_max_rate -#define ft_fdc_max_rate ftape_status.fti_fdc_max_rate -/* drive selection, either FTAPE_SEL_A/B/C/D */ -#define ft_drive_sel ftape_status.fti_drive_sel -/* flags set after decode the drive and tape status */ -#define ft_formatted ftape_status.fti_formatted -#define ft_no_tape ftape_status.fti_no_tape -#define ft_write_protected ftape_status.fti_write_protected -#define ft_new_tape ftape_status.fti_new_tape -/* values of last queried drive/tape status and error */ -#define ft_last_error ftape_status.fti_last_error -#define ft_last_status ftape_status.fti_last_status -/* cartridge geometry */ -#define ft_tracks_per_tape ftape_status.fti_tracks_per_tape -#define ft_segments_per_track ftape_status.fti_segments_per_track -/* the format code as stored in the header segment */ -#define ft_format_code ftape_status.fti_format_code -/* the qic status as returned by report drive configuration */ -#define ft_qic_std ftape_status.fti_qic_std -#define ft_used_header_segment ftape_status.fti_used_header_segment -#define ft_header_segment_1 ftape_status.fti_header_segment_1 -#define ft_header_segment_2 ftape_status.fti_header_segment_2 -#define ft_first_data_segment ftape_status.fti_first_data_segment -#define ft_last_data_segment ftape_status.fti_last_data_segment -/* is tape running? */ -#define ft_runner_status ftape_status.fti_runner_status -/* is tape reading/writing/verifying/formatting/deleting */ -#define ft_driver_state ftape_status.fti_state -/* flags fatal hardware error */ -#define ft_failure ftape_status.fti_failure -/* history record */ -#define ft_history ftape_status.fti_history - -/* - * ftape-ctl.c defined global vars. - */ -extern ftape_info ftape_status; -extern int ftape_segments_per_head; -extern int ftape_segments_per_cylinder; -extern int ftape_init_drive_needed; - -/* - * ftape-ctl.c defined global functions. - */ -extern int ftape_mmap(struct vm_area_struct *vma); -extern int ftape_enable(int drive_selection); -extern void ftape_disable(void); -extern int ftape_seek_to_bot(void); -extern int ftape_seek_to_eot(void); -extern int ftape_abort_operation(void); -extern void ftape_calc_timeouts(unsigned int qic_std, - unsigned int data_rate, - unsigned int tape_len); -extern int ftape_calibrate_data_rate(unsigned int qic_std); -extern const ftape_info *ftape_get_status(void); -#endif diff --git a/drivers/char/ftape/lowlevel/ftape-ecc.c b/drivers/char/ftape/lowlevel/ftape-ecc.c deleted file mode 100644 index e5632f674bc..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-ecc.c +++ /dev/null @@ -1,853 +0,0 @@ -/* - * - * Copyright (c) 1993 Ning and David Mosberger. - - This is based on code originally written by Bas Laarhoven (bas@vimec.nl) - and David L. Brown, Jr., and incorporates improvements suggested by - Kai Harrekilde-Petersen. - - This program is free software; you can redistribute it and/or - modify it under the terms of the GNU General Public License as - published by the Free Software Foundation; either version 2, or (at - your option) any later version. - - This program is distributed in the hope that it will be useful, but - WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, - USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-ecc.c,v $ - * $Revision: 1.3 $ - * $Date: 1997/10/05 19:18:10 $ - * - * This file contains the Reed-Solomon error correction code - * for the QIC-40/80 floppy-tape driver for Linux. - */ - -#include - -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-ecc.h" - -/* Machines that are big-endian should define macro BIG_ENDIAN. - * Unfortunately, there doesn't appear to be a standard include file - * that works for all OSs. - */ - -#if defined(__sparc__) || defined(__hppa) -#define BIG_ENDIAN -#endif /* __sparc__ || __hppa */ - -#if defined(__mips__) -#error Find a smart way to determine the Endianness of the MIPS CPU -#endif - -/* Notice: to minimize the potential for confusion, we use r to - * denote the independent variable of the polynomials in the - * Galois Field GF(2^8). We reserve x for polynomials that - * that have coefficients in GF(2^8). - * - * The Galois Field in which coefficient arithmetic is performed are - * the polynomials over Z_2 (i.e., 0 and 1) modulo the irreducible - * polynomial f(r), where f(r)=r^8 + r^7 + r^2 + r + 1. A polynomial - * is represented as a byte with the MSB as the coefficient of r^7 and - * the LSB as the coefficient of r^0. For example, the binary - * representation of f(x) is 0x187 (of course, this doesn't fit into 8 - * bits). In this field, the polynomial r is a primitive element. - * That is, r^i with i in 0,...,255 enumerates all elements in the - * field. - * - * The generator polynomial for the QIC-80 ECC is - * - * g(x) = x^3 + r^105*x^2 + r^105*x + 1 - * - * which can be factored into: - * - * g(x) = (x-r^-1)(x-r^0)(x-r^1) - * - * the byte representation of the coefficients are: - * - * r^105 = 0xc0 - * r^-1 = 0xc3 - * r^0 = 0x01 - * r^1 = 0x02 - * - * Notice that r^-1 = r^254 as exponent arithmetic is performed - * modulo 2^8-1 = 255. - * - * For more information on Galois Fields and Reed-Solomon codes, refer - * to any good book. I found _An Introduction to Error Correcting - * Codes with Applications_ by S. A. Vanstone and P. C. van Oorschot - * to be a good introduction into the former. _CODING THEORY: The - * Essentials_ I found very useful for its concise description of - * Reed-Solomon encoding/decoding. - * - */ - -typedef __u8 Matrix[3][3]; - -/* - * gfpow[] is defined such that gfpow[i] returns r^i if - * i is in the range [0..255]. - */ -static const __u8 gfpow[] = -{ - 0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80, - 0x87, 0x89, 0x95, 0xad, 0xdd, 0x3d, 0x7a, 0xf4, - 0x6f, 0xde, 0x3b, 0x76, 0xec, 0x5f, 0xbe, 0xfb, - 0x71, 0xe2, 0x43, 0x86, 0x8b, 0x91, 0xa5, 0xcd, - 0x1d, 0x3a, 0x74, 0xe8, 0x57, 0xae, 0xdb, 0x31, - 0x62, 0xc4, 0x0f, 0x1e, 0x3c, 0x78, 0xf0, 0x67, - 0xce, 0x1b, 0x36, 0x6c, 0xd8, 0x37, 0x6e, 0xdc, - 0x3f, 0x7e, 0xfc, 0x7f, 0xfe, 0x7b, 0xf6, 0x6b, - 0xd6, 0x2b, 0x56, 0xac, 0xdf, 0x39, 0x72, 0xe4, - 0x4f, 0x9e, 0xbb, 0xf1, 0x65, 0xca, 0x13, 0x26, - 0x4c, 0x98, 0xb7, 0xe9, 0x55, 0xaa, 0xd3, 0x21, - 0x42, 0x84, 0x8f, 0x99, 0xb5, 0xed, 0x5d, 0xba, - 0xf3, 0x61, 0xc2, 0x03, 0x06, 0x0c, 0x18, 0x30, - 0x60, 0xc0, 0x07, 0x0e, 0x1c, 0x38, 0x70, 0xe0, - 0x47, 0x8e, 0x9b, 0xb1, 0xe5, 0x4d, 0x9a, 0xb3, - 0xe1, 0x45, 0x8a, 0x93, 0xa1, 0xc5, 0x0d, 0x1a, - 0x34, 0x68, 0xd0, 0x27, 0x4e, 0x9c, 0xbf, 0xf9, - 0x75, 0xea, 0x53, 0xa6, 0xcb, 0x11, 0x22, 0x44, - 0x88, 0x97, 0xa9, 0xd5, 0x2d, 0x5a, 0xb4, 0xef, - 0x59, 0xb2, 0xe3, 0x41, 0x82, 0x83, 0x81, 0x85, - 0x8d, 0x9d, 0xbd, 0xfd, 0x7d, 0xfa, 0x73, 0xe6, - 0x4b, 0x96, 0xab, 0xd1, 0x25, 0x4a, 0x94, 0xaf, - 0xd9, 0x35, 0x6a, 0xd4, 0x2f, 0x5e, 0xbc, 0xff, - 0x79, 0xf2, 0x63, 0xc6, 0x0b, 0x16, 0x2c, 0x58, - 0xb0, 0xe7, 0x49, 0x92, 0xa3, 0xc1, 0x05, 0x0a, - 0x14, 0x28, 0x50, 0xa0, 0xc7, 0x09, 0x12, 0x24, - 0x48, 0x90, 0xa7, 0xc9, 0x15, 0x2a, 0x54, 0xa8, - 0xd7, 0x29, 0x52, 0xa4, 0xcf, 0x19, 0x32, 0x64, - 0xc8, 0x17, 0x2e, 0x5c, 0xb8, 0xf7, 0x69, 0xd2, - 0x23, 0x46, 0x8c, 0x9f, 0xb9, 0xf5, 0x6d, 0xda, - 0x33, 0x66, 0xcc, 0x1f, 0x3e, 0x7c, 0xf8, 0x77, - 0xee, 0x5b, 0xb6, 0xeb, 0x51, 0xa2, 0xc3, 0x01 -}; - -/* - * This is a log table. That is, gflog[r^i] returns i (modulo f(r)). - * gflog[0] is undefined and the first element is therefore not valid. - */ -static const __u8 gflog[256] = -{ - 0xff, 0x00, 0x01, 0x63, 0x02, 0xc6, 0x64, 0x6a, - 0x03, 0xcd, 0xc7, 0xbc, 0x65, 0x7e, 0x6b, 0x2a, - 0x04, 0x8d, 0xce, 0x4e, 0xc8, 0xd4, 0xbd, 0xe1, - 0x66, 0xdd, 0x7f, 0x31, 0x6c, 0x20, 0x2b, 0xf3, - 0x05, 0x57, 0x8e, 0xe8, 0xcf, 0xac, 0x4f, 0x83, - 0xc9, 0xd9, 0xd5, 0x41, 0xbe, 0x94, 0xe2, 0xb4, - 0x67, 0x27, 0xde, 0xf0, 0x80, 0xb1, 0x32, 0x35, - 0x6d, 0x45, 0x21, 0x12, 0x2c, 0x0d, 0xf4, 0x38, - 0x06, 0x9b, 0x58, 0x1a, 0x8f, 0x79, 0xe9, 0x70, - 0xd0, 0xc2, 0xad, 0xa8, 0x50, 0x75, 0x84, 0x48, - 0xca, 0xfc, 0xda, 0x8a, 0xd6, 0x54, 0x42, 0x24, - 0xbf, 0x98, 0x95, 0xf9, 0xe3, 0x5e, 0xb5, 0x15, - 0x68, 0x61, 0x28, 0xba, 0xdf, 0x4c, 0xf1, 0x2f, - 0x81, 0xe6, 0xb2, 0x3f, 0x33, 0xee, 0x36, 0x10, - 0x6e, 0x18, 0x46, 0xa6, 0x22, 0x88, 0x13, 0xf7, - 0x2d, 0xb8, 0x0e, 0x3d, 0xf5, 0xa4, 0x39, 0x3b, - 0x07, 0x9e, 0x9c, 0x9d, 0x59, 0x9f, 0x1b, 0x08, - 0x90, 0x09, 0x7a, 0x1c, 0xea, 0xa0, 0x71, 0x5a, - 0xd1, 0x1d, 0xc3, 0x7b, 0xae, 0x0a, 0xa9, 0x91, - 0x51, 0x5b, 0x76, 0x72, 0x85, 0xa1, 0x49, 0xeb, - 0xcb, 0x7c, 0xfd, 0xc4, 0xdb, 0x1e, 0x8b, 0xd2, - 0xd7, 0x92, 0x55, 0xaa, 0x43, 0x0b, 0x25, 0xaf, - 0xc0, 0x73, 0x99, 0x77, 0x96, 0x5c, 0xfa, 0x52, - 0xe4, 0xec, 0x5f, 0x4a, 0xb6, 0xa2, 0x16, 0x86, - 0x69, 0xc5, 0x62, 0xfe, 0x29, 0x7d, 0xbb, 0xcc, - 0xe0, 0xd3, 0x4d, 0x8c, 0xf2, 0x1f, 0x30, 0xdc, - 0x82, 0xab, 0xe7, 0x56, 0xb3, 0x93, 0x40, 0xd8, - 0x34, 0xb0, 0xef, 0x26, 0x37, 0x0c, 0x11, 0x44, - 0x6f, 0x78, 0x19, 0x9a, 0x47, 0x74, 0xa7, 0xc1, - 0x23, 0x53, 0x89, 0xfb, 0x14, 0x5d, 0xf8, 0x97, - 0x2e, 0x4b, 0xb9, 0x60, 0x0f, 0xed, 0x3e, 0xe5, - 0xf6, 0x87, 0xa5, 0x17, 0x3a, 0xa3, 0x3c, 0xb7 -}; - -/* This is a multiplication table for the factor 0xc0 (i.e., r^105 (mod f(r)). - * gfmul_c0[f] returns r^105 * f(r) (modulo f(r)). - */ -static const __u8 gfmul_c0[256] = -{ - 0x00, 0xc0, 0x07, 0xc7, 0x0e, 0xce, 0x09, 0xc9, - 0x1c, 0xdc, 0x1b, 0xdb, 0x12, 0xd2, 0x15, 0xd5, - 0x38, 0xf8, 0x3f, 0xff, 0x36, 0xf6, 0x31, 0xf1, - 0x24, 0xe4, 0x23, 0xe3, 0x2a, 0xea, 0x2d, 0xed, - 0x70, 0xb0, 0x77, 0xb7, 0x7e, 0xbe, 0x79, 0xb9, - 0x6c, 0xac, 0x6b, 0xab, 0x62, 0xa2, 0x65, 0xa5, - 0x48, 0x88, 0x4f, 0x8f, 0x46, 0x86, 0x41, 0x81, - 0x54, 0x94, 0x53, 0x93, 0x5a, 0x9a, 0x5d, 0x9d, - 0xe0, 0x20, 0xe7, 0x27, 0xee, 0x2e, 0xe9, 0x29, - 0xfc, 0x3c, 0xfb, 0x3b, 0xf2, 0x32, 0xf5, 0x35, - 0xd8, 0x18, 0xdf, 0x1f, 0xd6, 0x16, 0xd1, 0x11, - 0xc4, 0x04, 0xc3, 0x03, 0xca, 0x0a, 0xcd, 0x0d, - 0x90, 0x50, 0x97, 0x57, 0x9e, 0x5e, 0x99, 0x59, - 0x8c, 0x4c, 0x8b, 0x4b, 0x82, 0x42, 0x85, 0x45, - 0xa8, 0x68, 0xaf, 0x6f, 0xa6, 0x66, 0xa1, 0x61, - 0xb4, 0x74, 0xb3, 0x73, 0xba, 0x7a, 0xbd, 0x7d, - 0x47, 0x87, 0x40, 0x80, 0x49, 0x89, 0x4e, 0x8e, - 0x5b, 0x9b, 0x5c, 0x9c, 0x55, 0x95, 0x52, 0x92, - 0x7f, 0xbf, 0x78, 0xb8, 0x71, 0xb1, 0x76, 0xb6, - 0x63, 0xa3, 0x64, 0xa4, 0x6d, 0xad, 0x6a, 0xaa, - 0x37, 0xf7, 0x30, 0xf0, 0x39, 0xf9, 0x3e, 0xfe, - 0x2b, 0xeb, 0x2c, 0xec, 0x25, 0xe5, 0x22, 0xe2, - 0x0f, 0xcf, 0x08, 0xc8, 0x01, 0xc1, 0x06, 0xc6, - 0x13, 0xd3, 0x14, 0xd4, 0x1d, 0xdd, 0x1a, 0xda, - 0xa7, 0x67, 0xa0, 0x60, 0xa9, 0x69, 0xae, 0x6e, - 0xbb, 0x7b, 0xbc, 0x7c, 0xb5, 0x75, 0xb2, 0x72, - 0x9f, 0x5f, 0x98, 0x58, 0x91, 0x51, 0x96, 0x56, - 0x83, 0x43, 0x84, 0x44, 0x8d, 0x4d, 0x8a, 0x4a, - 0xd7, 0x17, 0xd0, 0x10, 0xd9, 0x19, 0xde, 0x1e, - 0xcb, 0x0b, 0xcc, 0x0c, 0xc5, 0x05, 0xc2, 0x02, - 0xef, 0x2f, 0xe8, 0x28, 0xe1, 0x21, 0xe6, 0x26, - 0xf3, 0x33, 0xf4, 0x34, 0xfd, 0x3d, 0xfa, 0x3a -}; - - -/* Returns V modulo 255 provided V is in the range -255,-254,...,509. - */ -static inline __u8 mod255(int v) -{ - if (v > 0) { - if (v < 255) { - return v; - } else { - return v - 255; - } - } else { - return v + 255; - } -} - - -/* Add two numbers in the field. Addition in this field is equivalent - * to a bit-wise exclusive OR operation---subtraction is therefore - * identical to addition. - */ -static inline __u8 gfadd(__u8 a, __u8 b) -{ - return a ^ b; -} - - -/* Add two vectors of numbers in the field. Each byte in A and B gets - * added individually. - */ -static inline unsigned long gfadd_long(unsigned long a, unsigned long b) -{ - return a ^ b; -} - - -/* Multiply two numbers in the field: - */ -static inline __u8 gfmul(__u8 a, __u8 b) -{ - if (a && b) { - return gfpow[mod255(gflog[a] + gflog[b])]; - } else { - return 0; - } -} - - -/* Just like gfmul, except we have already looked up the log of the - * second number. - */ -static inline __u8 gfmul_exp(__u8 a, int b) -{ - if (a) { - return gfpow[mod255(gflog[a] + b)]; - } else { - return 0; - } -} - - -/* Just like gfmul_exp, except that A is a vector of numbers. That - * is, each byte in A gets multiplied by gfpow[mod255(B)]. - */ -static inline unsigned long gfmul_exp_long(unsigned long a, int b) -{ - __u8 t; - - if (sizeof(long) == 4) { - return ( - ((t = (__u32)a >> 24 & 0xff) ? - (((__u32) gfpow[mod255(gflog[t] + b)]) << 24) : 0) | - ((t = (__u32)a >> 16 & 0xff) ? - (((__u32) gfpow[mod255(gflog[t] + b)]) << 16) : 0) | - ((t = (__u32)a >> 8 & 0xff) ? - (((__u32) gfpow[mod255(gflog[t] + b)]) << 8) : 0) | - ((t = (__u32)a >> 0 & 0xff) ? - (((__u32) gfpow[mod255(gflog[t] + b)]) << 0) : 0)); - } else if (sizeof(long) == 8) { - return ( - ((t = (__u64)a >> 56 & 0xff) ? - (((__u64) gfpow[mod255(gflog[t] + b)]) << 56) : 0) | - ((t = (__u64)a >> 48 & 0xff) ? - (((__u64) gfpow[mod255(gflog[t] + b)]) << 48) : 0) | - ((t = (__u64)a >> 40 & 0xff) ? - (((__u64) gfpow[mod255(gflog[t] + b)]) << 40) : 0) | - ((t = (__u64)a >> 32 & 0xff) ? - (((__u64) gfpow[mod255(gflog[t] + b)]) << 32) : 0) | - ((t = (__u64)a >> 24 & 0xff) ? - (((__u64) gfpow[mod255(gflog[t] + b)]) << 24) : 0) | - ((t = (__u64)a >> 16 & 0xff) ? - (((__u64) gfpow[mod255(gflog[t] + b)]) << 16) : 0) | - ((t = (__u64)a >> 8 & 0xff) ? - (((__u64) gfpow[mod255(gflog[t] + b)]) << 8) : 0) | - ((t = (__u64)a >> 0 & 0xff) ? - (((__u64) gfpow[mod255(gflog[t] + b)]) << 0) : 0)); - } else { - TRACE_FUN(ft_t_any); - TRACE_ABORT(-1, ft_t_err, "Error: size of long is %d bytes", - (int)sizeof(long)); - } -} - - -/* Divide two numbers in the field. Returns a/b (modulo f(x)). - */ -static inline __u8 gfdiv(__u8 a, __u8 b) -{ - if (!b) { - TRACE_FUN(ft_t_any); - TRACE_ABORT(0xff, ft_t_bug, "Error: division by zero"); - } else if (a == 0) { - return 0; - } else { - return gfpow[mod255(gflog[a] - gflog[b])]; - } -} - - -/* The following functions return the inverse of the matrix of the - * linear system that needs to be solved to determine the error - * magnitudes. The first deals with matrices of rank 3, while the - * second deals with matrices of rank 2. The error indices are passed - * in arguments L0,..,L2 (0=first sector, 31=last sector). The error - * indices must be sorted in ascending order, i.e., L00 CRC failures)"); - } - log_det = 255 - gflog[det]; - - /* Now, calculate all of the coefficients: - */ - Ainv[0][0]= gfmul_exp(gfadd(gfpow[l1], gfpow[l2]), log_det); - Ainv[0][1]= gfmul_exp(gfadd(t21, t12), log_det); - Ainv[0][2]= gfmul_exp(gfadd(gfpow[255 - l1], gfpow[255 - l2]),log_det); - - Ainv[1][0]= gfmul_exp(gfadd(gfpow[l0], gfpow[l2]), log_det); - Ainv[1][1]= gfmul_exp(gfadd(t20, t02), log_det); - Ainv[1][2]= gfmul_exp(gfadd(gfpow[255 - l0], gfpow[255 - l2]),log_det); - - Ainv[2][0]= gfmul_exp(gfadd(gfpow[l0], gfpow[l1]), log_det); - Ainv[2][1]= gfmul_exp(gfadd(t10, t01), log_det); - Ainv[2][2]= gfmul_exp(gfadd(gfpow[255 - l0], gfpow[255 - l1]),log_det); - - return 1; -} - - -static inline int gfinv2(__u8 l0, __u8 l1, Matrix Ainv) -{ - __u8 det; - __u8 t1, t2; - int log_det; - - t1 = gfpow[255 - l0]; - t2 = gfpow[255 - l1]; - det = gfadd(t1, t2); - if (!det) { - TRACE_FUN(ft_t_any); - TRACE_ABORT(0, ft_t_err, - "Inversion failed (2 CRC errors, >0 CRC failures)"); - } - log_det = 255 - gflog[det]; - - /* Now, calculate all of the coefficients: - */ - Ainv[0][0] = Ainv[1][0] = gfpow[log_det]; - - Ainv[0][1] = gfmul_exp(t2, log_det); - Ainv[1][1] = gfmul_exp(t1, log_det); - - return 1; -} - - -/* Multiply matrix A by vector S and return result in vector B. M is - * assumed to be of order NxN, S and B of order Nx1. - */ -static inline void gfmat_mul(int n, Matrix A, - __u8 *s, __u8 *b) -{ - int i, j; - __u8 dot_prod; - - for (i = 0; i < n; ++i) { - dot_prod = 0; - for (j = 0; j < n; ++j) { - dot_prod = gfadd(dot_prod, gfmul(A[i][j], s[j])); - } - b[i] = dot_prod; - } -} - - - -/* The Reed Solomon ECC codes are computed over the N-th byte of each - * block, where N=SECTOR_SIZE. There are up to 29 blocks of data, and - * 3 blocks of ECC. The blocks are stored contiguously in memory. A - * segment, consequently, is assumed to have at least 4 blocks: one or - * more data blocks plus three ECC blocks. - * - * Notice: In QIC-80 speak, a CRC error is a sector with an incorrect - * CRC. A CRC failure is a sector with incorrect data, but - * a valid CRC. In the error control literature, the former - * is usually called "erasure", the latter "error." - */ -/* Compute the parity bytes for C columns of data, where C is the - * number of bytes that fit into a long integer. We use a linear - * feed-back register to do this. The parity bytes P[0], P[STRIDE], - * P[2*STRIDE] are computed such that: - * - * x^k * p(x) + m(x) = 0 (modulo g(x)) - * - * where k = NBLOCKS, - * p(x) = P[0] + P[STRIDE]*x + P[2*STRIDE]*x^2, and - * m(x) = sum_{i=0}^k m_i*x^i. - * m_i = DATA[i*SECTOR_SIZE] - */ -static inline void set_parity(unsigned long *data, - int nblocks, - unsigned long *p, - int stride) -{ - unsigned long p0, p1, p2, t1, t2, *end; - - end = data + nblocks * (FT_SECTOR_SIZE / sizeof(long)); - p0 = p1 = p2 = 0; - while (data < end) { - /* The new parity bytes p0_i, p1_i, p2_i are computed - * from the old values p0_{i-1}, p1_{i-1}, p2_{i-1} - * recursively as: - * - * p0_i = p1_{i-1} + r^105 * (m_{i-1} - p0_{i-1}) - * p1_i = p2_{i-1} + r^105 * (m_{i-1} - p0_{i-1}) - * p2_i = (m_{i-1} - p0_{i-1}) - * - * With the initial condition: p0_0 = p1_0 = p2_0 = 0. - */ - t1 = gfadd_long(*data, p0); - /* - * Multiply each byte in t1 by 0xc0: - */ - if (sizeof(long) == 4) { - t2= (((__u32) gfmul_c0[(__u32)t1 >> 24 & 0xff]) << 24 | - ((__u32) gfmul_c0[(__u32)t1 >> 16 & 0xff]) << 16 | - ((__u32) gfmul_c0[(__u32)t1 >> 8 & 0xff]) << 8 | - ((__u32) gfmul_c0[(__u32)t1 >> 0 & 0xff]) << 0); - } else if (sizeof(long) == 8) { - t2= (((__u64) gfmul_c0[(__u64)t1 >> 56 & 0xff]) << 56 | - ((__u64) gfmul_c0[(__u64)t1 >> 48 & 0xff]) << 48 | - ((__u64) gfmul_c0[(__u64)t1 >> 40 & 0xff]) << 40 | - ((__u64) gfmul_c0[(__u64)t1 >> 32 & 0xff]) << 32 | - ((__u64) gfmul_c0[(__u64)t1 >> 24 & 0xff]) << 24 | - ((__u64) gfmul_c0[(__u64)t1 >> 16 & 0xff]) << 16 | - ((__u64) gfmul_c0[(__u64)t1 >> 8 & 0xff]) << 8 | - ((__u64) gfmul_c0[(__u64)t1 >> 0 & 0xff]) << 0); - } else { - TRACE_FUN(ft_t_any); - TRACE(ft_t_err, "Error: long is of size %d", - (int) sizeof(long)); - TRACE_EXIT; - } - p0 = gfadd_long(t2, p1); - p1 = gfadd_long(t2, p2); - p2 = t1; - data += FT_SECTOR_SIZE / sizeof(long); - } - *p = p0; - p += stride; - *p = p1; - p += stride; - *p = p2; - return; -} - - -/* Compute the 3 syndrome values. DATA should point to the first byte - * of the column for which the syndromes are desired. The syndromes - * are computed over the first NBLOCKS of rows. The three bytes will - * be placed in S[0], S[1], and S[2]. - * - * S[i] is the value of the "message" polynomial m(x) evaluated at the - * i-th root of the generator polynomial g(x). - * - * As g(x)=(x-r^-1)(x-1)(x-r^1) we evaluate the message polynomial at - * x=r^-1 to get S[0], at x=r^0=1 to get S[1], and at x=r to get S[2]. - * This could be done directly and efficiently via the Horner scheme. - * However, it would require multiplication tables for the factors - * r^-1 (0xc3) and r (0x02). The following scheme does not require - * any multiplication tables beyond what's needed for set_parity() - * anyway and is slightly faster if there are no errors and slightly - * slower if there are errors. The latter is hopefully the infrequent - * case. - * - * To understand the alternative algorithm, notice that set_parity(m, - * k, p) computes parity bytes such that: - * - * x^k * p(x) = m(x) (modulo g(x)). - * - * That is, to evaluate m(r^m), where r^m is a root of g(x), we can - * simply evaluate (r^m)^k*p(r^m). Also, notice that p is 0 if and - * only if s is zero. That is, if all parity bytes are 0, we know - * there is no error in the data and consequently there is no need to - * compute s(x) at all! In all other cases, we compute s(x) from p(x) - * by evaluating (r^m)^k*p(r^m) for m=-1, m=0, and m=1. The p(x) - * polynomial is evaluated via the Horner scheme. - */ -static int compute_syndromes(unsigned long *data, int nblocks, unsigned long *s) -{ - unsigned long p[3]; - - set_parity(data, nblocks, p, 1); - if (p[0] | p[1] | p[2]) { - /* Some of the checked columns do not have a zero - * syndrome. For simplicity, we compute the syndromes - * for all columns that we have computed the - * remainders for. - */ - s[0] = gfmul_exp_long( - gfadd_long(p[0], - gfmul_exp_long( - gfadd_long(p[1], - gfmul_exp_long(p[2], -1)), - -1)), - -nblocks); - s[1] = gfadd_long(gfadd_long(p[2], p[1]), p[0]); - s[2] = gfmul_exp_long( - gfadd_long(p[0], - gfmul_exp_long( - gfadd_long(p[1], - gfmul_exp_long(p[2], 1)), - 1)), - nblocks); - return 0; - } else { - return 1; - } -} - - -/* Correct the block in the column pointed to by DATA. There are NBAD - * CRC errors and their indices are in BAD_LOC[0], up to - * BAD_LOC[NBAD-1]. If NBAD>1, Ainv holds the inverse of the matrix - * of the linear system that needs to be solved to determine the error - * magnitudes. S[0], S[1], and S[2] are the syndrome values. If row - * j gets corrected, then bit j will be set in CORRECTION_MAP. - */ -static inline int correct_block(__u8 *data, int nblocks, - int nbad, int *bad_loc, Matrix Ainv, - __u8 *s, - SectorMap * correction_map) -{ - int ncorrected = 0; - int i; - __u8 t1, t2; - __u8 c0, c1, c2; /* check bytes */ - __u8 error_mag[3], log_error_mag; - __u8 *dp, l, e; - TRACE_FUN(ft_t_any); - - switch (nbad) { - case 0: - /* might have a CRC failure: */ - if (s[0] == 0) { - /* more than one error */ - TRACE_ABORT(-1, ft_t_err, - "ECC failed (0 CRC errors, >1 CRC failures)"); - } - t1 = gfdiv(s[1], s[0]); - if ((bad_loc[nbad++] = gflog[t1]) >= nblocks) { - TRACE(ft_t_err, - "ECC failed (0 CRC errors, >1 CRC failures)"); - TRACE_ABORT(-1, ft_t_err, - "attempt to correct data at %d", bad_loc[0]); - } - error_mag[0] = s[1]; - break; - case 1: - t1 = gfadd(gfmul_exp(s[1], bad_loc[0]), s[2]); - t2 = gfadd(gfmul_exp(s[0], bad_loc[0]), s[1]); - if (t1 == 0 && t2 == 0) { - /* one erasure, no error: */ - Ainv[0][0] = gfpow[bad_loc[0]]; - } else if (t1 == 0 || t2 == 0) { - /* one erasure and more than one error: */ - TRACE_ABORT(-1, ft_t_err, - "ECC failed (1 erasure, >1 error)"); - } else { - /* one erasure, one error: */ - if ((bad_loc[nbad++] = gflog[gfdiv(t1, t2)]) - >= nblocks) { - TRACE(ft_t_err, "ECC failed " - "(1 CRC errors, >1 CRC failures)"); - TRACE_ABORT(-1, ft_t_err, - "attempt to correct data at %d", - bad_loc[1]); - } - if (!gfinv2(bad_loc[0], bad_loc[1], Ainv)) { - /* inversion failed---must have more - * than one error - */ - TRACE_EXIT -1; - } - } - /* FALL THROUGH TO ERROR MAGNITUDE COMPUTATION: - */ - case 2: - case 3: - /* compute error magnitudes: */ - gfmat_mul(nbad, Ainv, s, error_mag); - break; - - default: - TRACE_ABORT(-1, ft_t_err, - "Internal Error: number of CRC errors > 3"); - } - - /* Perform correction by adding ERROR_MAG[i] to the byte at - * offset BAD_LOC[i]. Also add the value of the computed - * error polynomial to the syndrome values. If the correction - * was successful, the resulting check bytes should be zero - * (i.e., the corrected data is a valid code word). - */ - c0 = s[0]; - c1 = s[1]; - c2 = s[2]; - for (i = 0; i < nbad; ++i) { - e = error_mag[i]; - if (e) { - /* correct the byte at offset L by magnitude E: */ - l = bad_loc[i]; - dp = &data[l * FT_SECTOR_SIZE]; - *dp = gfadd(*dp, e); - *correction_map |= 1 << l; - ++ncorrected; - - log_error_mag = gflog[e]; - c0 = gfadd(c0, gfpow[mod255(log_error_mag - l)]); - c1 = gfadd(c1, e); - c2 = gfadd(c2, gfpow[mod255(log_error_mag + l)]); - } - } - if (c0 || c1 || c2) { - TRACE_ABORT(-1, ft_t_err, - "ECC self-check failed, too many errors"); - } - TRACE_EXIT ncorrected; -} - - -#if defined(ECC_SANITY_CHECK) || defined(ECC_PARANOID) - -/* Perform a sanity check on the computed parity bytes: - */ -static int sanity_check(unsigned long *data, int nblocks) -{ - TRACE_FUN(ft_t_any); - unsigned long s[3]; - - if (!compute_syndromes(data, nblocks, s)) { - TRACE_ABORT(0, ft_bug, - "Internal Error: syndrome self-check failed"); - } - TRACE_EXIT 1; -} - -#endif /* defined(ECC_SANITY_CHECK) || defined(ECC_PARANOID) */ - -/* Compute the parity for an entire segment of data. - */ -int ftape_ecc_set_segment_parity(struct memory_segment *mseg) -{ - int i; - __u8 *parity_bytes; - - parity_bytes = &mseg->data[(mseg->blocks - 3) * FT_SECTOR_SIZE]; - for (i = 0; i < FT_SECTOR_SIZE; i += sizeof(long)) { - set_parity((unsigned long *) &mseg->data[i], mseg->blocks - 3, - (unsigned long *) &parity_bytes[i], - FT_SECTOR_SIZE / sizeof(long)); -#ifdef ECC_PARANOID - if (!sanity_check((unsigned long *) &mseg->data[i], - mseg->blocks)) { - return -1; - } -#endif /* ECC_PARANOID */ - } - return 0; -} - - -/* Checks and corrects (if possible) the segment MSEG. Returns one of - * ECC_OK, ECC_CORRECTED, and ECC_FAILED. - */ -int ftape_ecc_correct_data(struct memory_segment *mseg) -{ - int col, i, result; - int ncorrected = 0; - int nerasures = 0; /* # of erasures (CRC errors) */ - int erasure_loc[3]; /* erasure locations */ - unsigned long ss[3]; - __u8 s[3]; - Matrix Ainv; - TRACE_FUN(ft_t_flow); - - mseg->corrected = 0; - - /* find first column that has non-zero syndromes: */ - for (col = 0; col < FT_SECTOR_SIZE; col += sizeof(long)) { - if (!compute_syndromes((unsigned long *) &mseg->data[col], - mseg->blocks, ss)) { - /* something is wrong---have to fix things */ - break; - } - } - if (col >= FT_SECTOR_SIZE) { - /* all syndromes are ok, therefore nothing to correct */ - TRACE_EXIT ECC_OK; - } - /* count the number of CRC errors if there were any: */ - if (mseg->read_bad) { - for (i = 0; i < mseg->blocks; i++) { - if (BAD_CHECK(mseg->read_bad, i)) { - if (nerasures >= 3) { - /* this is too much for ECC */ - TRACE_ABORT(ECC_FAILED, ft_t_err, - "ECC failed (>3 CRC errors)"); - } /* if */ - erasure_loc[nerasures++] = i; - } - } - } - /* - * If there are at least 2 CRC errors, determine inverse of matrix - * of linear system to be solved: - */ - switch (nerasures) { - case 2: - if (!gfinv2(erasure_loc[0], erasure_loc[1], Ainv)) { - TRACE_EXIT ECC_FAILED; - } - break; - case 3: - if (!gfinv3(erasure_loc[0], erasure_loc[1], - erasure_loc[2], Ainv)) { - TRACE_EXIT ECC_FAILED; - } - break; - default: - /* this is not an error condition... */ - break; - } - - do { - for (i = 0; i < sizeof(long); ++i) { - s[0] = ss[0]; - s[1] = ss[1]; - s[2] = ss[2]; - if (s[0] | s[1] | s[2]) { -#ifdef BIG_ENDIAN - result = correct_block( - &mseg->data[col + sizeof(long) - 1 - i], - mseg->blocks, - nerasures, - erasure_loc, - Ainv, - s, - &mseg->corrected); -#else - result = correct_block(&mseg->data[col + i], - mseg->blocks, - nerasures, - erasure_loc, - Ainv, - s, - &mseg->corrected); -#endif - if (result < 0) { - TRACE_EXIT ECC_FAILED; - } - ncorrected += result; - } - ss[0] >>= 8; - ss[1] >>= 8; - ss[2] >>= 8; - } - -#ifdef ECC_SANITY_CHECK - if (!sanity_check((unsigned long *) &mseg->data[col], - mseg->blocks)) { - TRACE_EXIT ECC_FAILED; - } -#endif /* ECC_SANITY_CHECK */ - - /* find next column with non-zero syndromes: */ - while ((col += sizeof(long)) < FT_SECTOR_SIZE) { - if (!compute_syndromes((unsigned long *) - &mseg->data[col], mseg->blocks, ss)) { - /* something is wrong---have to fix things */ - break; - } - } - } while (col < FT_SECTOR_SIZE); - if (ncorrected && nerasures == 0) { - TRACE(ft_t_warn, "block contained error not caught by CRC"); - } - TRACE((ncorrected > 0) ? ft_t_noise : ft_t_any, "number of corrections: %d", ncorrected); - TRACE_EXIT ncorrected ? ECC_CORRECTED : ECC_OK; -} diff --git a/drivers/char/ftape/lowlevel/ftape-ecc.h b/drivers/char/ftape/lowlevel/ftape-ecc.h deleted file mode 100644 index 4829146fe9a..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-ecc.h +++ /dev/null @@ -1,84 +0,0 @@ -#ifndef _FTAPE_ECC_H_ -#define _FTAPE_ECC_H_ - -/* - * Copyright (C) 1993 Ning and David Mosberger. - * Original: - * Copyright (C) 1993 Bas Laarhoven. - * Copyright (C) 1992 David L. Brown, Jr. - - This program is free software; you can redistribute it and/or - modify it under the terms of the GNU General Public License as - published by the Free Software Foundation; either version 2, or (at - your option) any later version. - - This program is distributed in the hope that it will be useful, but - WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, - USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-ecc.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:11 $ - * - * This file contains the definitions for the - * Reed-Solomon error correction code - * for the QIC-40/80 tape streamer device driver. - */ - -#include "../lowlevel/ftape-bsm.h" - -#define BAD_CLEAR(entry) ((entry)=0) -#define BAD_SET(entry,sector) ((entry)|=(1<<(sector))) -#define BAD_CHECK(entry,sector) ((entry)&(1<<(sector))) - -/* - * Return values for ecc_correct_data: - */ -enum { - ECC_OK, /* Data was correct. */ - ECC_CORRECTED, /* Correctable error in data. */ - ECC_FAILED, /* Could not correct data. */ -}; - -/* - * Representation of an in memory segment. MARKED_BAD lists the - * sectors that were marked bad during formatting. If the N-th sector - * in a segment is marked bad, bit 1< -#include - -#include -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/ftape-ecc.h" -#include "../lowlevel/ftape-bsm.h" -#include "../lowlevel/ftape-format.h" - -#if defined(TESTING) -#define FT_FMT_SEGS_PER_BUF 50 -#else -#define FT_FMT_SEGS_PER_BUF (FT_BUFF_SIZE/(4*FT_SECTORS_PER_SEGMENT)) -#endif - -static spinlock_t ftape_format_lock; - -/* - * first segment of the new buffer - */ -static int switch_segment; - -/* - * at most 256 segments fit into one 32 kb buffer. Even TR-1 cartridges have - * more than this many segments per track, so better be careful. - * - * buffer_struct *buff: buffer to store the formatting coordinates in - * int start: starting segment for this buffer. - * int spt: segments per track - * - * Note: segment ids are relative to the start of the track here. - */ -static void setup_format_buffer(buffer_struct *buff, int start, int spt, - __u8 gap3) -{ - int to_do = spt - start; - TRACE_FUN(ft_t_flow); - - if (to_do > FT_FMT_SEGS_PER_BUF) { - to_do = FT_FMT_SEGS_PER_BUF; - } - buff->ptr = buff->address; - buff->remaining = to_do * FT_SECTORS_PER_SEGMENT; /* # sectors */ - buff->bytes = buff->remaining * 4; /* need 4 bytes per sector */ - buff->gap3 = gap3; - buff->segment_id = start; - buff->next_segment = start + to_do; - if (buff->next_segment >= spt) { - buff->next_segment = 0; /* 0 means: stop runner */ - } - buff->status = waiting; /* tells the isr that it can use - * this buffer - */ - TRACE_EXIT; -} - - -/* - * start formatting a new track. - */ -int ftape_format_track(const unsigned int track, const __u8 gap3) -{ - unsigned long flags; - buffer_struct *tail, *head; - int status; - TRACE_FUN(ft_t_flow); - - TRACE_CATCH(ftape_ready_wait(ftape_timeout.pause, &status),); - if (track & 1) { - if (!(status & QIC_STATUS_AT_EOT)) { - TRACE_CATCH(ftape_seek_to_eot(),); - } - } else { - if (!(status & QIC_STATUS_AT_BOT)) { - TRACE_CATCH(ftape_seek_to_bot(),); - } - } - ftape_abort_operation(); /* this sets ft_head = ft_tail = 0 */ - ftape_set_state(formatting); - - TRACE(ft_t_noise, - "Formatting track %d, logical: from segment %d to %d", - track, track * ft_segments_per_track, - (track + 1) * ft_segments_per_track - 1); - - /* - * initialize the buffer switching protocol for this track - */ - head = ftape_get_buffer(ft_queue_head); /* tape isn't running yet */ - tail = ftape_get_buffer(ft_queue_tail); /* tape isn't running yet */ - switch_segment = 0; - do { - FT_SIGNAL_EXIT(_DONT_BLOCK); - setup_format_buffer(tail, switch_segment, - ft_segments_per_track, gap3); - switch_segment = tail->next_segment; - } while ((switch_segment != 0) && - ((tail = ftape_next_buffer(ft_queue_tail)) != head)); - /* go */ - head->status = formatting; - TRACE_CATCH(ftape_seek_head_to_track(track),); - TRACE_CATCH(ftape_command(QIC_LOGICAL_FORWARD),); - spin_lock_irqsave(&ftape_format_lock, flags); - TRACE_CATCH(fdc_setup_formatting(head), restore_flags(flags)); - spin_unlock_irqrestore(&ftape_format_lock, flags); - TRACE_EXIT 0; -} - -/* return segment id of segment currently being formatted and do the - * buffer switching stuff. - */ -int ftape_format_status(unsigned int *segment_id) -{ - buffer_struct *tail = ftape_get_buffer(ft_queue_tail); - int result; - TRACE_FUN(ft_t_flow); - - while (switch_segment != 0 && - ftape_get_buffer(ft_queue_head) != tail) { - FT_SIGNAL_EXIT(_DONT_BLOCK); - /* need more buffers, first wait for empty buffer - */ - TRACE_CATCH(ftape_wait_segment(formatting),); - /* don't worry for gap3. If we ever hit this piece of code, - * then all buffer already have the correct gap3 set! - */ - setup_format_buffer(tail, switch_segment, - ft_segments_per_track, tail->gap3); - switch_segment = tail->next_segment; - if (switch_segment != 0) { - tail = ftape_next_buffer(ft_queue_tail); - } - } - /* should runner stop ? - */ - if (ft_runner_status == aborting || ft_runner_status == do_abort) { - buffer_struct *head = ftape_get_buffer(ft_queue_head); - TRACE(ft_t_warn, "Error formatting segment %d", - ftape_get_buffer(ft_queue_head)->segment_id); - (void)ftape_abort_operation(); - TRACE_EXIT (head->status != error) ? -EAGAIN : -EIO; - } - /* - * don't care if the timer expires, this is just kind of a - * "select" operation that lets the calling process sleep - * until something has happened - */ - if (fdc_interrupt_wait(5 * FT_SECOND) < 0) { - TRACE(ft_t_noise, "End of track %d at segment %d", - ft_location.track, - ftape_get_buffer(ft_queue_head)->segment_id); - result = 1; /* end of track, unlock module */ - } else { - result = 0; - } - /* - * the calling process should use the seg id to determine - * which parts of the dma buffers can be safely overwritten - * with new data. - */ - *segment_id = ftape_get_buffer(ft_queue_head)->segment_id; - /* - * Internally we start counting segment ids from the start of - * each track when formatting, but externally we keep them - * relative to the start of the tape: - */ - *segment_id += ft_location.track * ft_segments_per_track; - TRACE_EXIT result; -} - -/* - * The segment id is relative to the start of the tape - */ -int ftape_verify_segment(const unsigned int segment_id, SectorMap *bsm) -{ - int result; - int verify_done = 0; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "Verifying segment %d", segment_id); - - if (ft_driver_state != verifying) { - TRACE(ft_t_noise, "calling ftape_abort_operation"); - if (ftape_abort_operation() < 0) { - TRACE(ft_t_err, "ftape_abort_operation failed"); - TRACE_EXIT -EIO; - } - } - *bsm = 0x00000000; - ftape_set_state(verifying); - for (;;) { - buffer_struct *tail; - /* - * Allow escape from this loop on signal - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - /* - * Search all full buffers for the first matching the - * wanted segment. Clear other buffers on the fly. - */ - tail = ftape_get_buffer(ft_queue_tail); - while (!verify_done && tail->status == done) { - /* - * Allow escape from this loop on signal ! - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - if (tail->segment_id == segment_id) { - /* If out buffer is already full, - * return its contents. - */ - TRACE(ft_t_flow, "found segment in cache: %d", - segment_id); - if ((tail->soft_error_map | - tail->hard_error_map) != 0) { - TRACE(ft_t_info,"bsm[%d] = 0x%08lx", - segment_id, - (unsigned long) - (tail->soft_error_map | - tail->hard_error_map)); - *bsm = (tail->soft_error_map | - tail->hard_error_map); - } - verify_done = 1; - } else { - TRACE(ft_t_flow,"zapping segment in cache: %d", - tail->segment_id); - } - tail->status = waiting; - tail = ftape_next_buffer(ft_queue_tail); - } - if (!verify_done && tail->status == verifying) { - if (tail->segment_id == segment_id) { - switch(ftape_wait_segment(verifying)) { - case 0: - break; - case -EINTR: - TRACE_ABORT(-EINTR, ft_t_warn, - "interrupted by " - "non-blockable signal"); - break; - default: - ftape_abort_operation(); - ftape_set_state(verifying); - /* be picky */ - TRACE_ABORT(-EIO, ft_t_warn, - "wait_segment failed"); - } - } else { - /* We're reading the wrong segment, - * stop runner. - */ - TRACE(ft_t_noise, "verifying wrong segment"); - ftape_abort_operation(); - ftape_set_state(verifying); - } - } - /* should runner stop ? - */ - if (ft_runner_status == aborting) { - buffer_struct *head = ftape_get_buffer(ft_queue_head); - if (head->status == error || - head->status == verifying) { - /* no data or overrun error */ - head->status = waiting; - } - TRACE_CATCH(ftape_dumb_stop(),); - } else { - /* If just passed last segment on tape: wait - * for BOT or EOT mark. Sets ft_runner_status to - * idle if at lEOT and successful - */ - TRACE_CATCH(ftape_handle_logical_eot(),); - } - if (verify_done) { - TRACE_EXIT 0; - } - /* Now at least one buffer is idle! - * Restart runner & tape if needed. - */ - /* We could optimize the following a little bit. We know that - * the bad sector map is empty. - */ - tail = ftape_get_buffer(ft_queue_tail); - if (tail->status == waiting) { - buffer_struct *head = ftape_get_buffer(ft_queue_head); - - ftape_setup_new_segment(head, segment_id, -1); - ftape_calc_next_cluster(head); - if (ft_runner_status == idle) { - result = ftape_start_tape(segment_id, - head->sector_offset); - switch(result) { - case 0: - break; - case -ETIME: - case -EINTR: - TRACE_ABORT(result, ft_t_err, "Error: " - "segment %d unreachable", - segment_id); - break; - default: - *bsm = EMPTY_SEGMENT; - TRACE_EXIT 0; - break; - } - } - head->status = verifying; - fdc_setup_read_write(head, FDC_VERIFY); - } - } - /* not reached */ - TRACE_EXIT -EIO; -} diff --git a/drivers/char/ftape/lowlevel/ftape-format.h b/drivers/char/ftape/lowlevel/ftape-format.h deleted file mode 100644 index f1516156664..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-format.h +++ /dev/null @@ -1,37 +0,0 @@ -#ifndef _FTAPE_FORMAT_H -#define _FTAPE_FORMAT_H - -/* - * Copyright (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-format.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:13 $ - * - * This file contains the low level definitions for the - * formatting support for the QIC-40/80/3010/3020 floppy-tape - * driver "ftape" for Linux. - */ - -#ifdef __KERNEL__ -extern int ftape_format_track(const unsigned int track, const __u8 gap3); -extern int ftape_format_status(unsigned int *segment_id); -extern int ftape_verify_segment(const unsigned int segment_id, SectorMap *bsm); -#endif /* __KERNEL__ */ - -#endif diff --git a/drivers/char/ftape/lowlevel/ftape-init.c b/drivers/char/ftape/lowlevel/ftape-init.c deleted file mode 100644 index 4998132a81d..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-init.c +++ /dev/null @@ -1,160 +0,0 @@ -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * This file contains the code that interfaces the kernel - * for the QIC-40/80/3010/3020 floppy-tape driver for Linux. - */ - -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#ifdef CONFIG_ZFTAPE -#include -#endif - -#include "../lowlevel/ftape-init.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-read.h" -#include "../lowlevel/ftape-write.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/fdc-io.h" -#include "../lowlevel/ftape-buffer.h" -#include "../lowlevel/ftape-proc.h" -#include "../lowlevel/ftape-tracing.h" - - -#if defined(MODULE) && !defined(CONFIG_FT_NO_TRACE_AT_ALL) -static int ft_tracing = -1; -#endif - - -/* Called by modules package when installing the driver - * or by kernel during the initialization phase - */ -static int __init ftape_init(void) -{ - TRACE_FUN(ft_t_flow); - -#ifdef MODULE -#ifndef CONFIG_FT_NO_TRACE_AT_ALL - if (ft_tracing != -1) { - ftape_tracing = ft_tracing; - } -#endif - printk(KERN_INFO FTAPE_VERSION "\n"); - if (TRACE_LEVEL >= ft_t_info) { - printk( -KERN_INFO "(c) 1993-1996 Bas Laarhoven (bas@vimec.nl)\n" -KERN_INFO "(c) 1995-1996 Kai Harrekilde-Petersen (khp@dolphinics.no)\n" -KERN_INFO "(c) 1996-1997 Claus-Justus Heine (claus@momo.math.rwth-aachen.de)\n" -KERN_INFO "QIC-117 driver for QIC-40/80/3010/3020 floppy tape drives\n"); - } -#else /* !MODULE */ - /* print a short no-nonsense boot message */ - printk(KERN_INFO FTAPE_VERSION "\n"); -#endif /* MODULE */ - TRACE(ft_t_info, "installing QIC-117 floppy tape hardware drive ... "); - TRACE(ft_t_info, "ftape_init @ 0x%p", ftape_init); - /* Allocate the DMA buffers. They are deallocated at cleanup() time. - */ -#ifdef TESTING -#ifdef MODULE - while (ftape_set_nr_buffers(CONFIG_FT_NR_BUFFERS) < 0) { - ftape_sleep(FT_SECOND/20); - if (signal_pending(current)) { - (void)ftape_set_nr_buffers(0); - TRACE(ft_t_bug, - "Killed by signal while allocating buffers."); - TRACE_ABORT(-EINTR, - ft_t_bug, "Free up memory and retry"); - } - } -#else - TRACE_CATCH(ftape_set_nr_buffers(CONFIG_FT_NR_BUFFERS), - (void)ftape_set_nr_buffers(0)); -#endif -#else - TRACE_CATCH(ftape_set_nr_buffers(CONFIG_FT_NR_BUFFERS), - (void)ftape_set_nr_buffers(0)); -#endif - ft_drive_sel = -1; - ft_failure = 1; /* inhibit any operation but open */ - ftape_udelay_calibrate(); /* must be before fdc_wait_calibrate ! */ - fdc_wait_calibrate(); -#if defined(CONFIG_PROC_FS) && defined(CONFIG_FT_PROC_FS) - (void)ftape_proc_init(); -#endif -#ifdef CONFIG_ZFTAPE - (void)zft_init(); -#endif - TRACE_EXIT 0; -} - -module_param(ft_fdc_base, uint, 0); -MODULE_PARM_DESC(ft_fdc_base, "Base address of FDC controller."); -module_param(ft_fdc_irq, uint, 0); -MODULE_PARM_DESC(ft_fdc_irq, "IRQ (interrupt channel) to use."); -module_param(ft_fdc_dma, uint, 0); -MODULE_PARM_DESC(ft_fdc_dma, "DMA channel to use."); -module_param(ft_fdc_threshold, uint, 0); -MODULE_PARM_DESC(ft_fdc_threshold, "Threshold of the FDC Fifo."); -module_param(ft_fdc_rate_limit, uint, 0); -MODULE_PARM_DESC(ft_fdc_rate_limit, "Maximal data rate for FDC."); -module_param(ft_probe_fc10, bool, 0); -MODULE_PARM_DESC(ft_probe_fc10, - "If non-zero, probe for a Colorado FC-10/FC-20 controller."); -module_param(ft_mach2, bool, 0); -MODULE_PARM_DESC(ft_mach2, - "If non-zero, probe for a Mountain MACH-2 controller."); -#if defined(MODULE) && !defined(CONFIG_FT_NO_TRACE_AT_ALL) -module_param(ft_tracing, int, 0644); -MODULE_PARM_DESC(ft_tracing, - "Amount of debugging output, 0 <= tracing <= 8, default 3."); -#endif - -MODULE_AUTHOR( - "(c) 1993-1996 Bas Laarhoven (bas@vimec.nl), " - "(c) 1995-1996 Kai Harrekilde-Petersen (khp@dolphinics.no), " - "(c) 1996, 1997 Claus-Justus Heine (claus@momo.math.rwth-aachen.de)"); -MODULE_DESCRIPTION( - "QIC-117 driver for QIC-40/80/3010/3020 floppy tape drives."); -MODULE_LICENSE("GPL"); - -static void __exit ftape_exit(void) -{ - TRACE_FUN(ft_t_flow); - -#if defined(CONFIG_PROC_FS) && defined(CONFIG_FT_PROC_FS) - ftape_proc_destroy(); -#endif - (void)ftape_set_nr_buffers(0); - printk(KERN_INFO "ftape: unloaded.\n"); - TRACE_EXIT; -} - -module_init(ftape_init); -module_exit(ftape_exit); diff --git a/drivers/char/ftape/lowlevel/ftape-init.h b/drivers/char/ftape/lowlevel/ftape-init.h deleted file mode 100644 index 99a7b8ab086..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-init.h +++ /dev/null @@ -1,43 +0,0 @@ -#ifndef _FTAPE_INIT_H -#define _FTAPE_INIT_H - -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-init.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:16 $ - * - * This file contains the definitions for the interface to - * the Linux kernel for floppy tape driver ftape. - * - */ - -#include -#include - -#define _NEVER_BLOCK (sigmask(SIGKILL) | sigmask(SIGSTOP)) -#define _DONT_BLOCK (_NEVER_BLOCK | sigmask(SIGINT)) -#define _DO_BLOCK (sigmask(SIGPIPE)) - -#ifndef QIC117_TAPE_MAJOR -#define QIC117_TAPE_MAJOR 27 -#endif - -#endif diff --git a/drivers/char/ftape/lowlevel/ftape-io.c b/drivers/char/ftape/lowlevel/ftape-io.c deleted file mode 100644 index 259015aeff5..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-io.c +++ /dev/null @@ -1,992 +0,0 @@ -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996 Kai Harrekilde-Petersen, - * (C) 1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-io.c,v $ - * $Revision: 1.4 $ - * $Date: 1997/11/11 14:02:36 $ - * - * This file contains the general control functions for the - * QIC-40/80/3010/3020 floppy-tape driver "ftape" for Linux. - */ - -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/fdc-io.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/ftape-write.h" -#include "../lowlevel/ftape-read.h" -#include "../lowlevel/ftape-init.h" -#include "../lowlevel/ftape-calibr.h" - -/* Global vars. - */ -/* NOTE: sectors start numbering at 1, all others at 0 ! */ -ft_timeout_table ftape_timeout; -unsigned int ftape_tape_len; -volatile qic117_cmd_t ftape_current_command; -const struct qic117_command_table qic117_cmds[] = QIC117_COMMANDS; -int ftape_might_be_off_track; - -/* Local vars. - */ -static int diagnostic_mode; -static unsigned int ftape_udelay_count; -static unsigned int ftape_udelay_time; - -void ftape_udelay(unsigned int usecs) -{ - volatile int count = (ftape_udelay_count * usecs + - ftape_udelay_count - 1) / ftape_udelay_time; - volatile int i; - - while (count-- > 0) { - for (i = 0; i < 20; ++i); - } -} - -void ftape_udelay_calibrate(void) -{ - ftape_calibrate("ftape_udelay", - ftape_udelay, &ftape_udelay_count, &ftape_udelay_time); -} - -/* Delay (msec) routine. - */ -void ftape_sleep(unsigned int time) -{ - TRACE_FUN(ft_t_any); - - time *= 1000; /* msecs -> usecs */ - if (time < FT_USPT) { - /* Time too small for scheduler, do a busy wait ! */ - ftape_udelay(time); - } else { - long timeout; - unsigned long flags; - unsigned int ticks = (time + FT_USPT - 1) / FT_USPT; - - TRACE(ft_t_any, "%d msec, %d ticks", time/1000, ticks); - timeout = ticks; - save_flags(flags); - sti(); - msleep_interruptible(jiffies_to_msecs(timeout)); - /* Mmm. Isn't current->blocked == 0xffffffff ? - */ - if (signal_pending(current)) { - TRACE(ft_t_err, "awoken by non-blocked signal :-("); - } - restore_flags(flags); - } - TRACE_EXIT; -} - -/* send a command or parameter to the drive - * Generates # of step pulses. - */ -static inline int ft_send_to_drive(int arg) -{ - /* Always wait for a command_timeout period to separate - * individuals commands and/or parameters. - */ - ftape_sleep(3 * FT_MILLISECOND); - /* Keep cylinder nr within range, step towards home if possible. - */ - if (ftape_current_cylinder >= arg) { - return fdc_seek(ftape_current_cylinder - arg); - } else { - return fdc_seek(ftape_current_cylinder + arg); - } -} - -/* forward */ int ftape_report_raw_drive_status(int *status); - -static int ft_check_cmd_restrictions(qic117_cmd_t command) -{ - int status = -1; - TRACE_FUN(ft_t_any); - - TRACE(ft_t_flow, "%s", qic117_cmds[command].name); - /* A new motion command during an uninterruptible (motion) - * command requires a ready status before the new command can - * be issued. Otherwise a new motion command needs to be - * checked against required status. - */ - if (qic117_cmds[command].cmd_type == motion && - qic117_cmds[ftape_current_command].non_intr) { - ftape_report_raw_drive_status(&status); - if ((status & QIC_STATUS_READY) == 0) { - TRACE(ft_t_noise, - "motion cmd (%d) during non-intr cmd (%d)", - command, ftape_current_command); - TRACE(ft_t_noise, "waiting until drive gets ready"); - ftape_ready_wait(ftape_timeout.seek, - &status); - } - } - if (qic117_cmds[command].mask != 0) { - __u8 difference; - /* Some commands do require a certain status: - */ - if (status == -1) { /* not yet set */ - ftape_report_raw_drive_status(&status); - } - difference = ((status ^ qic117_cmds[command].state) & - qic117_cmds[command].mask); - /* Wait until the drive gets - * ready. This may last forever if - * the drive never gets ready... - */ - while ((difference & QIC_STATUS_READY) != 0) { - TRACE(ft_t_noise, "command %d issued while not ready", - command); - TRACE(ft_t_noise, "waiting until drive gets ready"); - if (ftape_ready_wait(ftape_timeout.seek, - &status) == -EINTR) { - /* Bail out on signal ! - */ - TRACE_ABORT(-EINTR, ft_t_warn, - "interrupted by non-blockable signal"); - } - difference = ((status ^ qic117_cmds[command].state) & - qic117_cmds[command].mask); - } - while ((difference & QIC_STATUS_ERROR) != 0) { - int err; - qic117_cmd_t cmd; - - TRACE(ft_t_noise, - "command %d issued while error pending", - command); - TRACE(ft_t_noise, "clearing error status"); - ftape_report_error(&err, &cmd, 1); - ftape_report_raw_drive_status(&status); - difference = ((status ^ qic117_cmds[command].state) & - qic117_cmds[command].mask); - if ((difference & QIC_STATUS_ERROR) != 0) { - /* Bail out on fatal signal ! - */ - FT_SIGNAL_EXIT(_NEVER_BLOCK); - } - } - if (difference) { - /* Any remaining difference can't be solved - * here. - */ - if (difference & (QIC_STATUS_CARTRIDGE_PRESENT | - QIC_STATUS_NEW_CARTRIDGE | - QIC_STATUS_REFERENCED)) { - TRACE(ft_t_warn, - "Fatal: tape removed or reinserted !"); - ft_failure = 1; - } else { - TRACE(ft_t_err, "wrong state: 0x%02x should be: 0x%02x", - status & qic117_cmds[command].mask, - qic117_cmds[command].state); - } - TRACE_EXIT -EIO; - } - if (~status & QIC_STATUS_READY & qic117_cmds[command].mask) { - TRACE_ABORT(-EBUSY, ft_t_err, "Bad: still busy!"); - } - } - TRACE_EXIT 0; -} - -/* Issue a tape command: - */ -int ftape_command(qic117_cmd_t command) -{ - int result = 0; - static int level; - TRACE_FUN(ft_t_any); - - if ((unsigned int)command > NR_ITEMS(qic117_cmds)) { - /* This is a bug we'll want to know about too. - */ - TRACE_ABORT(-EIO, ft_t_bug, "bug - bad command: %d", command); - } - if (++level > 5) { /* This is a bug we'll want to know about. */ - --level; - TRACE_ABORT(-EIO, ft_t_bug, "bug - recursion for command: %d", - command); - } - /* disable logging and restriction check for some commands, - * check all other commands that have a prescribed starting - * status. - */ - if (diagnostic_mode) { - TRACE(ft_t_flow, "diagnostic command %d", command); - } else if (command == QIC_REPORT_DRIVE_STATUS || - command == QIC_REPORT_NEXT_BIT) { - TRACE(ft_t_any, "%s", qic117_cmds[command].name); - } else { - TRACE_CATCH(ft_check_cmd_restrictions(command), --level); - } - /* Now all conditions are met or result was < 0. - */ - result = ft_send_to_drive((unsigned int)command); - if (qic117_cmds[command].cmd_type == motion && - command != QIC_LOGICAL_FORWARD && command != QIC_STOP_TAPE) { - ft_location.known = 0; - } - ftape_current_command = command; - --level; - TRACE_EXIT result; -} - -/* Send a tape command parameter: - * Generates command # of step pulses. - * Skips tape-status call ! - */ -int ftape_parameter(unsigned int parameter) -{ - TRACE_FUN(ft_t_any); - - TRACE(ft_t_flow, "called with parameter = %d", parameter); - TRACE_EXIT ft_send_to_drive(parameter + 2); -} - -/* Wait for the drive to get ready. - * timeout time in milli-seconds - * Returned status is valid if result != -EIO - * - * Should we allow to be killed by SIGINT? (^C) - * Would be nice at least for large timeouts. - */ -int ftape_ready_wait(unsigned int timeout, int *status) -{ - unsigned long t0; - unsigned int poll_delay; - int signal_retries; - TRACE_FUN(ft_t_any); - - /* the following ** REALLY ** reduces the system load when - * e.g. one simply rewinds or retensions. The tape is slow - * anyway. It is really not necessary to detect error - * conditions with 1/10 seconds granularity - * - * On my AMD 133MHZ 486: 100 ms: 23% system load - * 1 sec: 5% - * 5 sec: 0.6%, yeah - */ - if (timeout <= FT_SECOND) { - poll_delay = 100 * FT_MILLISECOND; - signal_retries = 20; /* two seconds */ - } else if (timeout < 20 * FT_SECOND) { - TRACE(ft_t_flow, "setting poll delay to 1 second"); - poll_delay = FT_SECOND; - signal_retries = 2; /* two seconds */ - } else { - TRACE(ft_t_flow, "setting poll delay to 5 seconds"); - poll_delay = 5 * FT_SECOND; - signal_retries = 1; /* five seconds */ - } - for (;;) { - t0 = jiffies; - TRACE_CATCH(ftape_report_raw_drive_status(status),); - if (*status & QIC_STATUS_READY) { - TRACE_EXIT 0; - } - if (!signal_retries--) { - FT_SIGNAL_EXIT(_NEVER_BLOCK); - } - if ((int)timeout >= 0) { - /* this will fail when jiffies wraps around about - * once every year :-) - */ - timeout -= ((jiffies - t0) * FT_SECOND) / HZ; - if (timeout <= 0) { - TRACE_ABORT(-ETIME, ft_t_err, "timeout"); - } - ftape_sleep(poll_delay); - timeout -= poll_delay; - } else { - ftape_sleep(poll_delay); - } - } - TRACE_EXIT -ETIME; -} - -/* Issue command and wait up to timeout milli seconds for drive ready - */ -int ftape_command_wait(qic117_cmd_t command, unsigned int timeout, int *status) -{ - int result; - - /* Drive should be ready, issue command - */ - result = ftape_command(command); - if (result >= 0) { - result = ftape_ready_wait(timeout, status); - } - return result; -} - -static int ftape_parameter_wait(unsigned int parm, unsigned int timeout, int *status) -{ - int result; - - /* Drive should be ready, issue command - */ - result = ftape_parameter(parm); - if (result >= 0) { - result = ftape_ready_wait(timeout, status); - } - return result; -} - -/*-------------------------------------------------------------------------- - * Report operations - */ - -/* Query the drive about its status. The command is sent and - result_length bits of status are returned (2 extra bits are read - for start and stop). */ - -int ftape_report_operation(int *status, - qic117_cmd_t command, - int result_length) -{ - int i, st3; - unsigned int t0; - unsigned int dt; - TRACE_FUN(ft_t_any); - - TRACE_CATCH(ftape_command(command),); - t0 = ftape_timestamp(); - i = 0; - do { - ++i; - ftape_sleep(3 * FT_MILLISECOND); /* see remark below */ - TRACE_CATCH(fdc_sense_drive_status(&st3),); - dt = ftape_timediff(t0, ftape_timestamp()); - /* Ack should be asserted within Ttimout + Tack = 6 msec. - * Looks like some drives fail to do this so extend this - * period to 300 msec. - */ - } while (!(st3 & ST3_TRACK_0) && dt < 300000); - if (!(st3 & ST3_TRACK_0)) { - TRACE(ft_t_err, - "No acknowledge after %u msec. (%i iter)", dt / 1000, i); - TRACE_ABORT(-EIO, ft_t_err, "timeout on Acknowledge"); - } - /* dt may be larger than expected because of other tasks - * scheduled while we were sleeping. - */ - if (i > 1 && dt > 6000) { - TRACE(ft_t_err, "Acknowledge after %u msec. (%i iter)", - dt / 1000, i); - } - *status = 0; - for (i = 0; i < result_length + 1; i++) { - TRACE_CATCH(ftape_command(QIC_REPORT_NEXT_BIT),); - TRACE_CATCH(fdc_sense_drive_status(&st3),); - if (i < result_length) { - *status |= ((st3 & ST3_TRACK_0) ? 1 : 0) << i; - } else if ((st3 & ST3_TRACK_0) == 0) { - TRACE_ABORT(-EIO, ft_t_err, "missing status stop bit"); - } - } - /* this command will put track zero and index back into normal state */ - (void)ftape_command(QIC_REPORT_NEXT_BIT); - TRACE_EXIT 0; -} - -/* Report the current drive status. */ - -int ftape_report_raw_drive_status(int *status) -{ - int result; - int count = 0; - TRACE_FUN(ft_t_any); - - do { - result = ftape_report_operation(status, - QIC_REPORT_DRIVE_STATUS, 8); - } while (result < 0 && ++count <= 3); - if (result < 0) { - TRACE_ABORT(-EIO, ft_t_err, - "report_operation failed after %d trials", count); - } - if ((*status & 0xff) == 0xff) { - TRACE_ABORT(-EIO, ft_t_err, - "impossible drive status 0xff"); - } - if (*status & QIC_STATUS_READY) { - ftape_current_command = QIC_NO_COMMAND; /* completed */ - } - ft_last_status.status.drive_status = (__u8)(*status & 0xff); - TRACE_EXIT 0; -} - -int ftape_report_drive_status(int *status) -{ - TRACE_FUN(ft_t_any); - - TRACE_CATCH(ftape_report_raw_drive_status(status),); - if (*status & QIC_STATUS_NEW_CARTRIDGE || - !(*status & QIC_STATUS_CARTRIDGE_PRESENT)) { - ft_failure = 1; /* will inhibit further operations */ - TRACE_EXIT -EIO; - } - if (*status & QIC_STATUS_READY && *status & QIC_STATUS_ERROR) { - /* Let caller handle all errors */ - TRACE_ABORT(1, ft_t_warn, "warning: error status set!"); - } - TRACE_EXIT 0; -} - -int ftape_report_error(unsigned int *error, - qic117_cmd_t *command, int report) -{ - static const ftape_error ftape_errors[] = QIC117_ERRORS; - int code; - TRACE_FUN(ft_t_any); - - TRACE_CATCH(ftape_report_operation(&code, QIC_REPORT_ERROR_CODE, 16),); - *error = (unsigned int)(code & 0xff); - *command = (qic117_cmd_t)((code>>8)&0xff); - /* remember hardware status, maybe useful for status ioctls - */ - ft_last_error.error.command = (__u8)*command; - ft_last_error.error.error = (__u8)*error; - if (!report) { - TRACE_EXIT 0; - } - if (*error == 0) { - TRACE_ABORT(0, ft_t_info, "No error"); - } - TRACE(ft_t_info, "errorcode: %d", *error); - if (*error < NR_ITEMS(ftape_errors)) { - TRACE(ft_t_noise, "%sFatal ERROR:", - (ftape_errors[*error].fatal ? "" : "Non-")); - TRACE(ft_t_noise, "%s ...", ftape_errors[*error].message); - } else { - TRACE(ft_t_noise, "Unknown ERROR !"); - } - if ((unsigned int)*command < NR_ITEMS(qic117_cmds) && - qic117_cmds[*command].name != NULL) { - TRACE(ft_t_noise, "... caused by command \'%s\'", - qic117_cmds[*command].name); - } else { - TRACE(ft_t_noise, "... caused by unknown command %d", - *command); - } - TRACE_EXIT 0; -} - -int ftape_report_configuration(qic_model *model, - unsigned int *rate, - int *qic_std, - int *tape_len) -{ - int result; - int config; - int status; - static const unsigned int qic_rates[ 4] = { 250, 2000, 500, 1000 }; - TRACE_FUN(ft_t_any); - - result = ftape_report_operation(&config, - QIC_REPORT_DRIVE_CONFIGURATION, 8); - if (result < 0) { - ft_last_status.status.drive_config = (__u8)0x00; - *model = prehistoric; - *rate = 500; - *qic_std = QIC_TAPE_QIC40; - *tape_len = 205; - TRACE_EXIT 0; - } else { - ft_last_status.status.drive_config = (__u8)(config & 0xff); - } - *rate = qic_rates[(config & QIC_CONFIG_RATE_MASK) >> QIC_CONFIG_RATE_SHIFT]; - result = ftape_report_operation(&status, QIC_REPORT_TAPE_STATUS, 8); - if (result < 0) { - ft_last_status.status.tape_status = (__u8)0x00; - /* pre- QIC117 rev C spec. drive, QIC_CONFIG_80 bit is valid. - */ - *qic_std = (config & QIC_CONFIG_80) ? - QIC_TAPE_QIC80 : QIC_TAPE_QIC40; - /* ?? how's about 425ft tapes? */ - *tape_len = (config & QIC_CONFIG_LONG) ? 307 : 0; - *model = pre_qic117c; - result = 0; - } else { - ft_last_status.status.tape_status = (__u8)(status & 0xff); - *model = post_qic117b; - TRACE(ft_t_any, "report tape status result = %02x", status); - /* post- QIC117 rev C spec. drive, QIC_CONFIG_80 bit is - * invalid. - */ - switch (status & QIC_TAPE_STD_MASK) { - case QIC_TAPE_QIC40: - case QIC_TAPE_QIC80: - case QIC_TAPE_QIC3020: - case QIC_TAPE_QIC3010: - *qic_std = status & QIC_TAPE_STD_MASK; - break; - default: - *qic_std = -1; - break; - } - switch (status & QIC_TAPE_LEN_MASK) { - case QIC_TAPE_205FT: - /* 205 or 425+ ft 550 Oe tape */ - *tape_len = 0; - break; - case QIC_TAPE_307FT: - /* 307.5 ft 550 Oe Extended Length (XL) tape */ - *tape_len = 307; - break; - case QIC_TAPE_VARIABLE: - /* Variable length 550 Oe tape */ - *tape_len = 0; - break; - case QIC_TAPE_1100FT: - /* 1100 ft 550 Oe tape */ - *tape_len = 1100; - break; - case QIC_TAPE_FLEX: - /* Variable length 900 Oe tape */ - *tape_len = 0; - break; - default: - *tape_len = -1; - break; - } - if (*qic_std == -1 || *tape_len == -1) { - TRACE(ft_t_any, - "post qic-117b spec drive with unknown tape"); - } - result = *tape_len == -1 ? -EIO : 0; - if (status & QIC_TAPE_WIDE) { - switch (*qic_std) { - case QIC_TAPE_QIC80: - TRACE(ft_t_info, "TR-1 tape detected"); - break; - case QIC_TAPE_QIC3010: - TRACE(ft_t_info, "TR-2 tape detected"); - break; - case QIC_TAPE_QIC3020: - TRACE(ft_t_info, "TR-3 tape detected"); - break; - default: - TRACE(ft_t_warn, - "Unknown Travan tape type detected"); - break; - } - } - } - TRACE_EXIT (result < 0) ? -EIO : 0; -} - -static int ftape_report_rom_version(int *version) -{ - - if (ftape_report_operation(version, QIC_REPORT_ROM_VERSION, 8) < 0) { - return -EIO; - } else { - return 0; - } -} - -void ftape_report_vendor_id(unsigned int *id) -{ - int result; - TRACE_FUN(ft_t_any); - - /* We'll try to get a vendor id from the drive. First - * according to the QIC-117 spec, a 16-bit id is requested. - * If that fails we'll try an 8-bit version, otherwise we'll - * try an undocumented query. - */ - result = ftape_report_operation((int *) id, QIC_REPORT_VENDOR_ID, 16); - if (result < 0) { - result = ftape_report_operation((int *) id, - QIC_REPORT_VENDOR_ID, 8); - if (result < 0) { - /* The following is an undocumented call found - * in the CMS code. - */ - result = ftape_report_operation((int *) id, 24, 8); - if (result < 0) { - *id = UNKNOWN_VENDOR; - } else { - TRACE(ft_t_noise, "got old 8 bit id: %04x", - *id); - *id |= 0x20000; - } - } else { - TRACE(ft_t_noise, "got 8 bit id: %04x", *id); - *id |= 0x10000; - } - } else { - TRACE(ft_t_noise, "got 16 bit id: %04x", *id); - } - if (*id == 0x0047) { - int version; - int sign; - - if (ftape_report_rom_version(&version) < 0) { - TRACE(ft_t_bug, "report rom version failed"); - TRACE_EXIT; - } - TRACE(ft_t_noise, "CMS rom version: %d", version); - ftape_command(QIC_ENTER_DIAGNOSTIC_1); - ftape_command(QIC_ENTER_DIAGNOSTIC_1); - diagnostic_mode = 1; - if (ftape_report_operation(&sign, 9, 8) < 0) { - unsigned int error; - qic117_cmd_t command; - - ftape_report_error(&error, &command, 1); - ftape_command(QIC_ENTER_PRIMARY_MODE); - diagnostic_mode = 0; - TRACE_EXIT; /* failure ! */ - } else { - TRACE(ft_t_noise, "CMS signature: %02x", sign); - } - if (sign == 0xa5) { - result = ftape_report_operation(&sign, 37, 8); - if (result < 0) { - if (version >= 63) { - *id = 0x8880; - TRACE(ft_t_noise, - "This is an Iomega drive !"); - } else { - *id = 0x0047; - TRACE(ft_t_noise, - "This is a real CMS drive !"); - } - } else { - *id = 0x0047; - TRACE(ft_t_noise, "CMS status: %d", sign); - } - } else { - *id = UNKNOWN_VENDOR; - } - ftape_command(QIC_ENTER_PRIMARY_MODE); - diagnostic_mode = 0; - } - TRACE_EXIT; -} - -static int qic_rate_code(unsigned int rate) -{ - switch (rate) { - case 250: - return QIC_CONFIG_RATE_250; - case 500: - return QIC_CONFIG_RATE_500; - case 1000: - return QIC_CONFIG_RATE_1000; - case 2000: - return QIC_CONFIG_RATE_2000; - default: - return QIC_CONFIG_RATE_500; - } -} - -static int ftape_set_rate_test(unsigned int *max_rate) -{ - unsigned int error; - qic117_cmd_t command; - int status; - int supported = 0; - TRACE_FUN(ft_t_any); - - /* Check if the drive does support the select rate command - * by testing all different settings. If any one is accepted - * we assume the command is supported, else not. - */ - for (*max_rate = 2000; *max_rate >= 250; *max_rate /= 2) { - if (ftape_command(QIC_SELECT_RATE) < 0) { - continue; - } - if (ftape_parameter_wait(qic_rate_code(*max_rate), - 1 * FT_SECOND, &status) < 0) { - continue; - } - if (status & QIC_STATUS_ERROR) { - ftape_report_error(&error, &command, 0); - continue; - } - supported = 1; /* did accept a request */ - break; - } - TRACE(ft_t_noise, "Select Rate command is%s supported", - supported ? "" : " not"); - TRACE_EXIT supported; -} - -int ftape_set_data_rate(unsigned int new_rate /* Kbps */, unsigned int qic_std) -{ - int status; - int result = 0; - unsigned int data_rate = new_rate; - static int supported; - int rate_changed = 0; - qic_model dummy_model; - unsigned int dummy_qic_std, dummy_tape_len; - TRACE_FUN(ft_t_any); - - if (ft_drive_max_rate == 0) { /* first time */ - supported = ftape_set_rate_test(&ft_drive_max_rate); - } - if (supported) { - ftape_command(QIC_SELECT_RATE); - result = ftape_parameter_wait(qic_rate_code(new_rate), - 1 * FT_SECOND, &status); - if (result >= 0 && !(status & QIC_STATUS_ERROR)) { - rate_changed = 1; - } - } - TRACE_CATCH(result = ftape_report_configuration(&dummy_model, - &data_rate, - &dummy_qic_std, - &dummy_tape_len),); - if (data_rate != new_rate) { - if (!supported) { - TRACE(ft_t_warn, "Rate change not supported!"); - } else if (rate_changed) { - TRACE(ft_t_warn, "Requested: %d, got %d", - new_rate, data_rate); - } else { - TRACE(ft_t_warn, "Rate change failed!"); - } - result = -EINVAL; - } - /* - * Set data rate and write precompensation as specified: - * - * | QIC-40/80 | QIC-3010/3020 - * rate | precomp | precomp - * ----------+-------------+-------------- - * 250 Kbps. | 250 ns. | 0 ns. - * 500 Kbps. | 125 ns. | 0 ns. - * 1 Mbps. | 42 ns. | 0 ns. - * 2 Mbps | N/A | 0 ns. - */ - if ((qic_std == QIC_TAPE_QIC40 && data_rate > 500) || - (qic_std == QIC_TAPE_QIC80 && data_rate > 1000)) { - TRACE_ABORT(-EINVAL, - ft_t_warn, "Datarate too high for QIC-mode"); - } - TRACE_CATCH(fdc_set_data_rate(data_rate),_res = -EINVAL); - ft_data_rate = data_rate; - if (qic_std == QIC_TAPE_QIC40 || qic_std == QIC_TAPE_QIC80) { - switch (data_rate) { - case 250: - fdc_set_write_precomp(250); - break; - default: - case 500: - fdc_set_write_precomp(125); - break; - case 1000: - fdc_set_write_precomp(42); - break; - } - } else { - fdc_set_write_precomp(0); - } - TRACE_EXIT result; -} - -/* The next two functions are used to cope with excessive overrun errors - */ -int ftape_increase_threshold(void) -{ - TRACE_FUN(ft_t_flow); - - if (fdc.type < i82077 || ft_fdc_threshold >= 12) { - TRACE_ABORT(-EIO, ft_t_err, "cannot increase fifo threshold"); - } - if (fdc_fifo_threshold(++ft_fdc_threshold, NULL, NULL, NULL) < 0) { - TRACE(ft_t_err, "cannot increase fifo threshold"); - ft_fdc_threshold --; - fdc_reset(); - } - TRACE(ft_t_info, "New FIFO threshold: %d", ft_fdc_threshold); - TRACE_EXIT 0; -} - -int ftape_half_data_rate(void) -{ - if (ft_data_rate < 500) { - return -1; - } - if (ftape_set_data_rate(ft_data_rate / 2, ft_qic_std) < 0) { - return -EIO; - } - ftape_calc_timeouts(ft_qic_std, ft_data_rate, ftape_tape_len); - return 0; -} - -/* Seek the head to the specified track. - */ -int ftape_seek_head_to_track(unsigned int track) -{ - int status; - TRACE_FUN(ft_t_any); - - ft_location.track = -1; /* remains set in case of error */ - if (track >= ft_tracks_per_tape) { - TRACE_ABORT(-EINVAL, ft_t_bug, "track out of bounds"); - } - TRACE(ft_t_flow, "seeking track %d", track); - TRACE_CATCH(ftape_command(QIC_SEEK_HEAD_TO_TRACK),); - TRACE_CATCH(ftape_parameter_wait(track, ftape_timeout.head_seek, - &status),); - ft_location.track = track; - ftape_might_be_off_track = 0; - TRACE_EXIT 0; -} - -int ftape_wakeup_drive(wake_up_types method) -{ - int status; - int motor_on = 0; - TRACE_FUN(ft_t_any); - - switch (method) { - case wake_up_colorado: - TRACE_CATCH(ftape_command(QIC_PHANTOM_SELECT),); - TRACE_CATCH(ftape_parameter(0 /* ft_drive_sel ?? */),); - break; - case wake_up_mountain: - TRACE_CATCH(ftape_command(QIC_SOFT_SELECT),); - ftape_sleep(FT_MILLISECOND); /* NEEDED */ - TRACE_CATCH(ftape_parameter(18),); - break; - case wake_up_insight: - ftape_sleep(100 * FT_MILLISECOND); - motor_on = 1; - fdc_motor(motor_on); /* enable is done by motor-on */ - case no_wake_up: - break; - default: - TRACE_EXIT -ENODEV; /* unknown wakeup method */ - break; - } - /* If wakeup succeeded we shouldn't get an error here.. - */ - TRACE_CATCH(ftape_report_raw_drive_status(&status), - if (motor_on) { - fdc_motor(0); - }); - TRACE_EXIT 0; -} - -int ftape_put_drive_to_sleep(wake_up_types method) -{ - TRACE_FUN(ft_t_any); - - switch (method) { - case wake_up_colorado: - TRACE_CATCH(ftape_command(QIC_PHANTOM_DESELECT),); - break; - case wake_up_mountain: - TRACE_CATCH(ftape_command(QIC_SOFT_DESELECT),); - break; - case wake_up_insight: - fdc_motor(0); /* enable is done by motor-on */ - case no_wake_up: /* no wakeup / no sleep ! */ - break; - default: - TRACE_EXIT -ENODEV; /* unknown wakeup method */ - } - TRACE_EXIT 0; -} - -int ftape_reset_drive(void) -{ - int result = 0; - int status; - unsigned int err_code; - qic117_cmd_t err_command; - int i; - TRACE_FUN(ft_t_any); - - /* We want to re-establish contact with our drive. Fire a - * number of reset commands (single step pulses) and pray for - * success. - */ - for (i = 0; i < 2; ++i) { - TRACE(ft_t_flow, "Resetting fdc"); - fdc_reset(); - ftape_sleep(10 * FT_MILLISECOND); - TRACE(ft_t_flow, "Reset command to drive"); - result = ftape_command(QIC_RESET); - if (result == 0) { - ftape_sleep(1 * FT_SECOND); /* drive not - * accessible - * during 1 second - */ - TRACE(ft_t_flow, "Re-selecting drive"); - - /* Strange, the QIC-117 specs don't mention - * this but the drive gets deselected after a - * soft reset ! So we need to enable it - * again. - */ - if (ftape_wakeup_drive(ft_drive_type.wake_up) < 0) { - TRACE(ft_t_err, "Wakeup failed !"); - } - TRACE(ft_t_flow, "Waiting until drive gets ready"); - result= ftape_ready_wait(ftape_timeout.reset, &status); - if (result == 0 && (status & QIC_STATUS_ERROR)) { - result = ftape_report_error(&err_code, - &err_command, 1); - if (result == 0 && err_code == 27) { - /* Okay, drive saw reset - * command and responded as it - * should - */ - break; - } else { - result = -EIO; - } - } else { - result = -EIO; - } - } - FT_SIGNAL_EXIT(_DONT_BLOCK); - } - if (result != 0) { - TRACE(ft_t_err, "General failure to reset tape drive"); - } else { - /* Restore correct settings: keep original rate - */ - ftape_set_data_rate(ft_data_rate, ft_qic_std); - } - ftape_init_drive_needed = 1; - TRACE_EXIT result; -} diff --git a/drivers/char/ftape/lowlevel/ftape-io.h b/drivers/char/ftape/lowlevel/ftape-io.h deleted file mode 100644 index 26a7baad871..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-io.h +++ /dev/null @@ -1,90 +0,0 @@ -#ifndef _FTAPE_IO_H -#define _FTAPE_IO_H - -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-io.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:18 $ - * - * This file contains definitions for the glue part of the - * QIC-40/80/3010/3020 floppy-tape driver "ftape" for Linux. - */ - -#include -#include - -typedef struct { - unsigned int seek; - unsigned int reset; - unsigned int rewind; - unsigned int head_seek; - unsigned int stop; - unsigned int pause; -} ft_timeout_table; - -typedef enum { - prehistoric, pre_qic117c, post_qic117b, post_qic117d -} qic_model; - -/* - * ftape-io.c defined global vars. - */ -extern ft_timeout_table ftape_timeout; -extern unsigned int ftape_tape_len; -extern volatile qic117_cmd_t ftape_current_command; -extern const struct qic117_command_table qic117_cmds[]; -extern int ftape_might_be_off_track; - -/* - * ftape-io.c defined global functions. - */ -extern void ftape_udelay(unsigned int usecs); -extern void ftape_udelay_calibrate(void); -extern void ftape_sleep(unsigned int time); -extern void ftape_report_vendor_id(unsigned int *id); -extern int ftape_command(qic117_cmd_t command); -extern int ftape_command_wait(qic117_cmd_t command, - unsigned int timeout, - int *status); -extern int ftape_parameter(unsigned int parameter); -extern int ftape_report_operation(int *status, - qic117_cmd_t command, - int result_length); -extern int ftape_report_configuration(qic_model *model, - unsigned int *rate, - int *qic_std, - int *tape_len); -extern int ftape_report_drive_status(int *status); -extern int ftape_report_raw_drive_status(int *status); -extern int ftape_report_status(int *status); -extern int ftape_ready_wait(unsigned int timeout, int *status); -extern int ftape_seek_head_to_track(unsigned int track); -extern int ftape_set_data_rate(unsigned int new_rate, unsigned int qic_std); -extern int ftape_report_error(unsigned int *error, - qic117_cmd_t *command, - int report); -extern int ftape_reset_drive(void); -extern int ftape_put_drive_to_sleep(wake_up_types method); -extern int ftape_wakeup_drive(wake_up_types method); -extern int ftape_increase_threshold(void); -extern int ftape_half_data_rate(void); - -#endif diff --git a/drivers/char/ftape/lowlevel/ftape-proc.c b/drivers/char/ftape/lowlevel/ftape-proc.c deleted file mode 100644 index e805b15e0a1..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-proc.c +++ /dev/null @@ -1,214 +0,0 @@ -/* - * Copyright (C) 1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-proc.c,v $ - * $Revision: 1.11 $ - * $Date: 1997/10/24 14:47:37 $ - * - * This file contains the procfs interface for the - * QIC-40/80/3010/3020 floppy-tape driver "ftape" for Linux. - - * Old code removed, switched to dynamic proc entry. - */ - - -#if defined(CONFIG_PROC_FS) && defined(CONFIG_FT_PROC_FS) - -#include - -#include -#include -#include - -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-proc.h" -#include "../lowlevel/ftape-tracing.h" - -static size_t get_driver_info(char *buf) -{ - const char *debug_level[] = { "bugs" , - "errors", - "warnings", - "informational", - "noisy", - "program flow", - "fdc and dma", - "data flow", - "anything" }; - - return sprintf(buf, - "version : %s\n" - "used data rate: %d kbit/sec\n" - "dma memory : %d kb\n" - "debug messages: %s\n", - FTAPE_VERSION, - ft_data_rate, - FT_BUFF_SIZE * ft_nr_buffers >> 10, - debug_level[TRACE_LEVEL]); -} - -static size_t get_tapedrive_info(char *buf) -{ - return sprintf(buf, - "vendor id : 0x%04x\n" - "drive name: %s\n" - "wind speed: %d ips\n" - "wakeup : %s\n" - "max. rate : %d kbit/sec\n", - ft_drive_type.vendor_id, - ft_drive_type.name, - ft_drive_type.speed, - ((ft_drive_type.wake_up == no_wake_up) - ? "No wakeup needed" : - ((ft_drive_type.wake_up == wake_up_colorado) - ? "Colorado" : - ((ft_drive_type.wake_up == wake_up_mountain) - ? "Mountain" : - ((ft_drive_type.wake_up == wake_up_insight) - ? "Motor on" : - "Unknown")))), - ft_drive_max_rate); -} - -static size_t get_cartridge_info(char *buf) -{ - if (ftape_init_drive_needed) { - return sprintf(buf, "uninitialized\n"); - } - if (ft_no_tape) { - return sprintf(buf, "no cartridge inserted\n"); - } - return sprintf(buf, - "segments : %5d\n" - "tracks : %5d\n" - "length : %5dft\n" - "formatted : %3s\n" - "writable : %3s\n" - "QIC spec. : QIC-%s\n" - "fmt-code : %1d\n", - ft_segments_per_track, - ft_tracks_per_tape, - ftape_tape_len, - (ft_formatted == 1) ? "yes" : "no", - (ft_write_protected == 1) ? "no" : "yes", - ((ft_qic_std == QIC_TAPE_QIC40) ? "40" : - ((ft_qic_std == QIC_TAPE_QIC80) ? "80" : - ((ft_qic_std == QIC_TAPE_QIC3010) ? "3010" : - ((ft_qic_std == QIC_TAPE_QIC3020) ? "3020" : - "???")))), - ft_format_code); -} - -static size_t get_controller_info(char *buf) -{ - const char *fdc_name[] = { "no fdc", - "i8272", - "i82077", - "i82077AA", - "Colorado FC-10 or FC-20", - "i82078", - "i82078_1" }; - - return sprintf(buf, - "FDC type : %s\n" - "FDC base : 0x%03x\n" - "FDC irq : %d\n" - "FDC dma : %d\n" - "FDC thr. : %d\n" - "max. rate : %d kbit/sec\n", - ft_mach2 ? "Mountain MACH-2" : fdc_name[fdc.type], - fdc.sra, fdc.irq, fdc.dma, - ft_fdc_threshold, ft_fdc_max_rate); -} - -static size_t get_history_info(char *buf) -{ - size_t len; - - len = sprintf(buf, - "\nFDC isr statistics\n" - " id_am_errors : %3d\n" - " id_crc_errors : %3d\n" - " data_am_errors : %3d\n" - " data_crc_errors : %3d\n" - " overrun_errors : %3d\n" - " no_data_errors : %3d\n" - " retries : %3d\n", - ft_history.id_am_errors, ft_history.id_crc_errors, - ft_history.data_am_errors, ft_history.data_crc_errors, - ft_history.overrun_errors, ft_history.no_data_errors, - ft_history.retries); - len += sprintf(buf + len, - "\nECC statistics\n" - " crc_errors : %3d\n" - " crc_failures : %3d\n" - " ecc_failures : %3d\n" - " sectors corrected: %3d\n", - ft_history.crc_errors, ft_history.crc_failures, - ft_history.ecc_failures, ft_history.corrected); - len += sprintf(buf + len, - "\ntape quality statistics\n" - " media defects : %3d\n", - ft_history.defects); - len += sprintf(buf + len, - "\ntape motion statistics\n" - " repositions : %3d\n", - ft_history.rewinds); - return len; -} - -static int ftape_read_proc(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - char *ptr = page; - size_t len; - - ptr += sprintf(ptr, "Kernel Driver\n\n"); - ptr += get_driver_info(ptr); - ptr += sprintf(ptr, "\nTape Drive\n\n"); - ptr += get_tapedrive_info(ptr); - ptr += sprintf(ptr, "\nFDC Controller\n\n"); - ptr += get_controller_info(ptr); - ptr += sprintf(ptr, "\nTape Cartridge\n\n"); - ptr += get_cartridge_info(ptr); - ptr += sprintf(ptr, "\nHistory Record\n\n"); - ptr += get_history_info(ptr); - - len = strlen(page); - *start = NULL; - if (off+count >= len) { - *eof = 1; - } else { - *eof = 0; - } - return len; -} - -int __init ftape_proc_init(void) -{ - return create_proc_read_entry("ftape", 0, &proc_root, - ftape_read_proc, NULL) != NULL; -} - -void ftape_proc_destroy(void) -{ - remove_proc_entry("ftape", &proc_root); -} - -#endif /* defined(CONFIG_PROC_FS) && defined(CONFIG_FT_PROC_FS) */ diff --git a/drivers/char/ftape/lowlevel/ftape-proc.h b/drivers/char/ftape/lowlevel/ftape-proc.h deleted file mode 100644 index 264dfcc1d22..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-proc.h +++ /dev/null @@ -1,35 +0,0 @@ -#ifndef _FTAPE_PROC_H -#define _FTAPE_PROC_H - -/* - * Copyright (C) 1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-proc.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:20 $ - * - * This file contains definitions for the procfs interface of the - * QIC-40/80/3010/3020 floppy-tape driver "ftape" for Linux. - */ - -#include - -extern int ftape_proc_init(void); -extern void ftape_proc_destroy(void); - -#endif diff --git a/drivers/char/ftape/lowlevel/ftape-read.c b/drivers/char/ftape/lowlevel/ftape-read.c deleted file mode 100644 index d967d8cd86d..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-read.c +++ /dev/null @@ -1,621 +0,0 @@ -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-read.c,v $ - * $Revision: 1.6 $ - * $Date: 1997/10/21 14:39:22 $ - * - * This file contains the reading code - * for the QIC-117 floppy-tape driver for Linux. - * - */ - -#include -#include -#include - -#include -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-read.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/ftape-write.h" -#include "../lowlevel/ftape-ecc.h" -#include "../lowlevel/ftape-bsm.h" - -/* Global vars. - */ - -/* Local vars. - */ - -void ftape_zap_read_buffers(void) -{ - int i; - - for (i = 0; i < ft_nr_buffers; ++i) { -/* changed to "fit" with dynamic allocation of tape_buffer. --khp */ - ft_buffer[i]->status = waiting; - ft_buffer[i]->bytes = 0; - ft_buffer[i]->skip = 0; - ft_buffer[i]->retry = 0; - } -/* ftape_reset_buffer(); */ -} - -static SectorMap convert_sector_map(buffer_struct * buff) -{ - int i = 0; - SectorMap bad_map = ftape_get_bad_sector_entry(buff->segment_id); - SectorMap src_map = buff->soft_error_map | buff->hard_error_map; - SectorMap dst_map = 0; - TRACE_FUN(ft_t_any); - - if (bad_map || src_map) { - TRACE(ft_t_flow, "bad_map = 0x%08lx", (long) bad_map); - TRACE(ft_t_flow, "src_map = 0x%08lx", (long) src_map); - } - while (bad_map) { - while ((bad_map & 1) == 0) { - if (src_map & 1) { - dst_map |= (1 << i); - } - src_map >>= 1; - bad_map >>= 1; - ++i; - } - /* (bad_map & 1) == 1 */ - src_map >>= 1; - bad_map >>= 1; - } - if (src_map) { - dst_map |= (src_map << i); - } - if (dst_map) { - TRACE(ft_t_flow, "dst_map = 0x%08lx", (long) dst_map); - } - TRACE_EXIT dst_map; -} - -static int correct_and_copy_fraction(buffer_struct *buff, __u8 * destination, - int start, int size) -{ - struct memory_segment mseg; - int result; - SectorMap read_bad; - TRACE_FUN(ft_t_any); - - mseg.read_bad = convert_sector_map(buff); - mseg.marked_bad = 0; /* not used... */ - mseg.blocks = buff->bytes / FT_SECTOR_SIZE; - mseg.data = buff->address; - /* If there are no data sectors we can skip this segment. - */ - if (mseg.blocks <= 3) { - TRACE_ABORT(0, ft_t_noise, "empty segment"); - } - read_bad = mseg.read_bad; - ft_history.crc_errors += count_ones(read_bad); - result = ftape_ecc_correct_data(&mseg); - if (read_bad != 0 || mseg.corrected != 0) { - TRACE(ft_t_noise, "crc error map: 0x%08lx", (unsigned long)read_bad); - TRACE(ft_t_noise, "corrected map: 0x%08lx", (unsigned long)mseg.corrected); - ft_history.corrected += count_ones(mseg.corrected); - } - if (result == ECC_CORRECTED || result == ECC_OK) { - if (result == ECC_CORRECTED) { - TRACE(ft_t_info, "ecc corrected segment: %d", buff->segment_id); - } - if(start < 0) { - start= 0; - } - if((start+size) > ((mseg.blocks - 3) * FT_SECTOR_SIZE)) { - size = (mseg.blocks - 3) * FT_SECTOR_SIZE - start; - } - if (size < 0) { - size= 0; - } - if(size > 0) { - memcpy(destination + start, mseg.data + start, size); - } - if ((read_bad ^ mseg.corrected) & mseg.corrected) { - /* sectors corrected without crc errors set */ - ft_history.crc_failures++; - } - TRACE_EXIT size; /* (mseg.blocks - 3) * FT_SECTOR_SIZE; */ - } else { - ft_history.ecc_failures++; - TRACE_ABORT(-EAGAIN, - ft_t_err, "ecc failure on segment %d", - buff->segment_id); - } - TRACE_EXIT 0; -} - -/* Read given segment into buffer at address. - */ -int ftape_read_segment_fraction(const int segment_id, - void *address, - const ft_read_mode_t read_mode, - const int start, - const int size) -{ - int result = 0; - int retry = 0; - int bytes_read = 0; - int read_done = 0; - TRACE_FUN(ft_t_flow); - - ft_history.used |= 1; - TRACE(ft_t_data_flow, "segment_id = %d", segment_id); - if (ft_driver_state != reading) { - TRACE(ft_t_noise, "calling ftape_abort_operation"); - TRACE_CATCH(ftape_abort_operation(),); - ftape_set_state(reading); - } - for(;;) { - buffer_struct *tail; - /* Allow escape from this loop on signal ! - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - /* Search all full buffers for the first matching the - * wanted segment. Clear other buffers on the fly. - */ - tail = ftape_get_buffer(ft_queue_tail); - while (!read_done && tail->status == done) { - /* Allow escape from this loop on signal ! - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - if (tail->segment_id == segment_id) { - /* If out buffer is already full, - * return its contents. - */ - TRACE(ft_t_flow, "found segment in cache: %d", - segment_id); - if (tail->deleted) { - /* Return a value that - * read_header_segment - * understands. As this - * should only occur when - * searching for the header - * segments it shouldn't be - * misinterpreted elsewhere. - */ - TRACE_EXIT 0; - } - result = correct_and_copy_fraction( - tail, - address, - start, - size); - TRACE(ft_t_flow, "segment contains (bytes): %d", - result); - if (result < 0) { - if (result != -EAGAIN) { - TRACE_EXIT result; - } - /* keep read_done == 0, will - * trigger - * ftape_abort_operation - * because reading wrong - * segment. - */ - TRACE(ft_t_err, "ecc failed, retry"); - ++retry; - } else { - read_done = 1; - bytes_read = result; - } - } else { - TRACE(ft_t_flow,"zapping segment in cache: %d", - tail->segment_id); - } - tail->status = waiting; - tail = ftape_next_buffer(ft_queue_tail); - } - if (!read_done && tail->status == reading) { - if (tail->segment_id == segment_id) { - switch(ftape_wait_segment(reading)) { - case 0: - break; - case -EINTR: - TRACE_ABORT(-EINTR, ft_t_warn, - "interrupted by " - "non-blockable signal"); - break; - default: - TRACE(ft_t_noise, - "wait_segment failed"); - ftape_abort_operation(); - ftape_set_state(reading); - break; - } - } else { - /* We're reading the wrong segment, - * stop runner. - */ - TRACE(ft_t_noise, "reading wrong segment"); - ftape_abort_operation(); - ftape_set_state(reading); - } - } - /* should runner stop ? - */ - if (ft_runner_status == aborting) { - buffer_struct *head = ftape_get_buffer(ft_queue_head); - switch(head->status) { - case error: - ft_history.defects += - count_ones(head->hard_error_map); - case reading: - head->status = waiting; - break; - default: - break; - } - TRACE_CATCH(ftape_dumb_stop(),); - } else { - /* If just passed last segment on tape: wait - * for BOT or EOT mark. Sets ft_runner_status to - * idle if at lEOT and successful - */ - TRACE_CATCH(ftape_handle_logical_eot(),); - } - /* If we got a segment: quit, or else retry up to limit. - * - * If segment to read is empty, do not start runner for it, - * but wait for next read call. - */ - if (read_done || - ftape_get_bad_sector_entry(segment_id) == EMPTY_SEGMENT ) { - /* bytes_read = 0; should still be zero */ - TRACE_EXIT bytes_read; - - } - if (retry > FT_RETRIES_ON_ECC_ERROR) { - ft_history.defects++; - TRACE_ABORT(-ENODATA, ft_t_err, - "too many retries on ecc failure"); - } - /* Now at least one buffer is empty ! - * Restart runner & tape if needed. - */ - TRACE(ft_t_any, "head: %d, tail: %d, ft_runner_status: %d", - ftape_buffer_id(ft_queue_head), - ftape_buffer_id(ft_queue_tail), - ft_runner_status); - TRACE(ft_t_any, "buffer[].status, [head]: %d, [tail]: %d", - ftape_get_buffer(ft_queue_head)->status, - ftape_get_buffer(ft_queue_tail)->status); - tail = ftape_get_buffer(ft_queue_tail); - if (tail->status == waiting) { - buffer_struct *head = ftape_get_buffer(ft_queue_head); - - ftape_setup_new_segment(head, segment_id, -1); - if (read_mode == FT_RD_SINGLE) { - /* disable read-ahead */ - head->next_segment = 0; - } - ftape_calc_next_cluster(head); - if (ft_runner_status == idle) { - result = ftape_start_tape(segment_id, - head->sector_offset); - if (result < 0) { - TRACE_ABORT(result, ft_t_err, "Error: " - "segment %d unreachable", - segment_id); - } - } - head->status = reading; - fdc_setup_read_write(head, FDC_READ); - } - } - /* not reached */ - TRACE_EXIT -EIO; -} - -int ftape_read_header_segment(__u8 *address) -{ - int result; - int header_segment; - int first_failed = 0; - int status; - TRACE_FUN(ft_t_flow); - - ft_used_header_segment = -1; - TRACE_CATCH(ftape_report_drive_status(&status),); - TRACE(ft_t_flow, "reading..."); - /* We're looking for the first header segment. - * A header segment cannot contain bad sectors, therefor at the - * tape start, segments with bad sectors are (according to QIC-40/80) - * written with deleted data marks and must be skipped. - */ - memset(address, '\0', (FT_SECTORS_PER_SEGMENT - 3) * FT_SECTOR_SIZE); - result = 0; -#define HEADER_SEGMENT_BOUNDARY 68 /* why not 42? */ - for (header_segment = 0; - header_segment < HEADER_SEGMENT_BOUNDARY && result == 0; - ++header_segment) { - /* Set no read-ahead, the isr will force read-ahead whenever - * it encounters deleted data ! - */ - result = ftape_read_segment(header_segment, - address, - FT_RD_SINGLE); - if (result < 0 && !first_failed) { - TRACE(ft_t_err, "header segment damaged, trying backup"); - first_failed = 1; - result = 0; /* force read of next (backup) segment */ - } - } - if (result < 0 || header_segment >= HEADER_SEGMENT_BOUNDARY) { - TRACE_ABORT(-EIO, ft_t_err, - "no readable header segment found"); - } - TRACE_CATCH(ftape_abort_operation(),); - ft_used_header_segment = header_segment; - result = ftape_decode_header_segment(address); - TRACE_EXIT result; -} - -int ftape_decode_header_segment(__u8 *address) -{ - unsigned int max_floppy_side; - unsigned int max_floppy_track; - unsigned int max_floppy_sector; - unsigned int new_tape_len; - TRACE_FUN(ft_t_flow); - - if (GET4(address, FT_SIGNATURE) == FT_D2G_MAGIC) { - /* Ditto 2GB header segment. They encrypt the bad sector map. - * We decrypt it and store them in normal format. - * I hope this is correct. - */ - int i; - TRACE(ft_t_warn, - "Found Ditto 2GB tape, " - "trying to decrypt bad sector map"); - for (i=256; i < 29 * FT_SECTOR_SIZE; i++) { - address[i] = ~(address[i] - (i&0xff)); - } - PUT4(address, 0,FT_HSEG_MAGIC); - } else if (GET4(address, FT_SIGNATURE) != FT_HSEG_MAGIC) { - TRACE_ABORT(-EIO, ft_t_err, - "wrong signature in header segment"); - } - ft_format_code = (ft_format_type) address[FT_FMT_CODE]; - if (ft_format_code != fmt_big) { - ft_header_segment_1 = GET2(address, FT_HSEG_1); - ft_header_segment_2 = GET2(address, FT_HSEG_2); - ft_first_data_segment = GET2(address, FT_FRST_SEG); - ft_last_data_segment = GET2(address, FT_LAST_SEG); - } else { - ft_header_segment_1 = GET4(address, FT_6_HSEG_1); - ft_header_segment_2 = GET4(address, FT_6_HSEG_2); - ft_first_data_segment = GET4(address, FT_6_FRST_SEG); - ft_last_data_segment = GET4(address, FT_6_LAST_SEG); - } - TRACE(ft_t_noise, "first data segment: %d", ft_first_data_segment); - TRACE(ft_t_noise, "last data segment: %d", ft_last_data_segment); - TRACE(ft_t_noise, "header segments are %d and %d", - ft_header_segment_1, ft_header_segment_2); - - /* Verify tape parameters... - * QIC-40/80 spec: tape_parameters: - * - * segments-per-track segments_per_track - * tracks-per-cartridge tracks_per_tape - * max-floppy-side (segments_per_track * - * tracks_per_tape - 1) / - * ftape_segments_per_head - * max-floppy-track ftape_segments_per_head / - * ftape_segments_per_cylinder - 1 - * max-floppy-sector ftape_segments_per_cylinder * - * FT_SECTORS_PER_SEGMENT - */ - ft_segments_per_track = GET2(address, FT_SPT); - ft_tracks_per_tape = address[FT_TPC]; - max_floppy_side = address[FT_FHM]; - max_floppy_track = address[FT_FTM]; - max_floppy_sector = address[FT_FSM]; - TRACE(ft_t_noise, "(fmt/spt/tpc/fhm/ftm/fsm) = %d/%d/%d/%d/%d/%d", - ft_format_code, ft_segments_per_track, ft_tracks_per_tape, - max_floppy_side, max_floppy_track, max_floppy_sector); - new_tape_len = ftape_tape_len; - switch (ft_format_code) { - case fmt_425ft: - new_tape_len = 425; - break; - case fmt_normal: - if (ftape_tape_len == 0) { /* otherwise 307 ft */ - new_tape_len = 205; - } - break; - case fmt_1100ft: - new_tape_len = 1100; - break; - case fmt_var:{ - int segments_per_1000_inch = 1; /* non-zero default for switch */ - switch (ft_qic_std) { - case QIC_TAPE_QIC40: - segments_per_1000_inch = 332; - break; - case QIC_TAPE_QIC80: - segments_per_1000_inch = 488; - break; - case QIC_TAPE_QIC3010: - segments_per_1000_inch = 730; - break; - case QIC_TAPE_QIC3020: - segments_per_1000_inch = 1430; - break; - } - new_tape_len = (1000 * ft_segments_per_track + - (segments_per_1000_inch - 1)) / segments_per_1000_inch; - break; - } - case fmt_big:{ - int segments_per_1000_inch = 1; /* non-zero default for switch */ - switch (ft_qic_std) { - case QIC_TAPE_QIC40: - segments_per_1000_inch = 332; - break; - case QIC_TAPE_QIC80: - segments_per_1000_inch = 488; - break; - case QIC_TAPE_QIC3010: - segments_per_1000_inch = 730; - break; - case QIC_TAPE_QIC3020: - segments_per_1000_inch = 1430; - break; - default: - TRACE_ABORT(-EIO, ft_t_bug, - "%x QIC-standard with fmt-code %d, please report", - ft_qic_std, ft_format_code); - } - new_tape_len = ((1000 * ft_segments_per_track + - (segments_per_1000_inch - 1)) / - segments_per_1000_inch); - break; - } - default: - TRACE_ABORT(-EIO, ft_t_err, - "unknown tape format, please report !"); - } - if (new_tape_len != ftape_tape_len) { - ftape_tape_len = new_tape_len; - TRACE(ft_t_info, "calculated tape length is %d ft", - ftape_tape_len); - ftape_calc_timeouts(ft_qic_std, ft_data_rate, ftape_tape_len); - } - if (ft_segments_per_track == 0 && ft_tracks_per_tape == 0 && - max_floppy_side == 0 && max_floppy_track == 0 && - max_floppy_sector == 0) { - /* QIC-40 Rev E and earlier has no values in the header. - */ - ft_segments_per_track = 68; - ft_tracks_per_tape = 20; - max_floppy_side = 1; - max_floppy_track = 169; - max_floppy_sector = 128; - } - /* This test will compensate for the wrong parameter on tapes - * formatted by Conner software. - */ - if (ft_segments_per_track == 150 && - ft_tracks_per_tape == 28 && - max_floppy_side == 7 && - max_floppy_track == 149 && - max_floppy_sector == 128) { -TRACE(ft_t_info, "the famous CONNER bug: max_floppy_side off by one !"); - max_floppy_side = 6; - } - /* These tests will compensate for the wrong parameter on tapes - * formatted by ComByte Windows software. - * - * First, for 205 foot tapes - */ - if (ft_segments_per_track == 100 && - ft_tracks_per_tape == 28 && - max_floppy_side == 9 && - max_floppy_track == 149 && - max_floppy_sector == 128) { -TRACE(ft_t_info, "the ComByte bug: max_floppy_side incorrect!"); - max_floppy_side = 4; - } - /* Next, for 307 foot tapes. */ - if (ft_segments_per_track == 150 && - ft_tracks_per_tape == 28 && - max_floppy_side == 9 && - max_floppy_track == 149 && - max_floppy_sector == 128) { -TRACE(ft_t_info, "the ComByte bug: max_floppy_side incorrect!"); - max_floppy_side = 6; - } - /* This test will compensate for the wrong parameter on tapes - * formatted by Colorado Windows software. - */ - if (ft_segments_per_track == 150 && - ft_tracks_per_tape == 28 && - max_floppy_side == 6 && - max_floppy_track == 150 && - max_floppy_sector == 128) { -TRACE(ft_t_info, "the famous Colorado bug: max_floppy_track off by one !"); - max_floppy_track = 149; - } - ftape_segments_per_head = ((max_floppy_sector/FT_SECTORS_PER_SEGMENT) * - (max_floppy_track + 1)); - /* This test will compensate for some bug reported by Dima - * Brodsky. Seems to be a Colorado bug, either. (freebee - * Imation tape shipped together with Colorado T3000 - */ - if ((ft_format_code == fmt_var || ft_format_code == fmt_big) && - ft_tracks_per_tape == 50 && - max_floppy_side == 54 && - max_floppy_track == 255 && - max_floppy_sector == 128) { -TRACE(ft_t_info, "the famous ??? bug: max_floppy_track off by one !"); - max_floppy_track = 254; - } - /* - * Verify drive_configuration with tape parameters - */ - if (ftape_segments_per_head == 0 || ftape_segments_per_cylinder == 0 || - ((ft_segments_per_track * ft_tracks_per_tape - 1) / ftape_segments_per_head - != max_floppy_side) || - (ftape_segments_per_head / ftape_segments_per_cylinder - 1 != max_floppy_track) || - (ftape_segments_per_cylinder * FT_SECTORS_PER_SEGMENT != max_floppy_sector) -#ifdef TESTING - || ((ft_format_code == fmt_var || ft_format_code == fmt_big) && - (max_floppy_track != 254 || max_floppy_sector != 128)) -#endif - ) { - char segperheadz = ftape_segments_per_head ? ' ' : '?'; - char segpercylz = ftape_segments_per_cylinder ? ' ' : '?'; - TRACE(ft_t_err,"Tape parameters inconsistency, please report"); - TRACE(ft_t_err, "reported = %d/%d/%d/%d/%d/%d", - ft_format_code, - ft_segments_per_track, - ft_tracks_per_tape, - max_floppy_side, - max_floppy_track, - max_floppy_sector); - TRACE(ft_t_err, "required = %d/%d/%d/%d%c/%d%c/%d", - ft_format_code, - ft_segments_per_track, - ft_tracks_per_tape, - ftape_segments_per_head ? - ((ft_segments_per_track * ft_tracks_per_tape -1) / - ftape_segments_per_head ) : - (ft_segments_per_track * ft_tracks_per_tape -1), - segperheadz, - ftape_segments_per_cylinder ? - (ftape_segments_per_head / - ftape_segments_per_cylinder - 1 ) : - ftape_segments_per_head - 1, - segpercylz, - (ftape_segments_per_cylinder * FT_SECTORS_PER_SEGMENT)); - TRACE_EXIT -EIO; - } - ftape_extract_bad_sector_map(address); - TRACE_EXIT 0; -} diff --git a/drivers/char/ftape/lowlevel/ftape-read.h b/drivers/char/ftape/lowlevel/ftape-read.h deleted file mode 100644 index 069f99f2a98..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-read.h +++ /dev/null @@ -1,51 +0,0 @@ -#ifndef _FTAPE_READ_H -#define _FTAPE_READ_H - -/* - * Copyright (C) 1994-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-read.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:22 $ - * - * This file contains the definitions for the read functions - * for the QIC-117 floppy-tape driver for Linux. - * - */ - -/* ftape-read.c defined global functions. - */ -typedef enum { - FT_RD_SINGLE = 0, - FT_RD_AHEAD = 1, -} ft_read_mode_t; - -extern int ftape_read_header_segment(__u8 *address); -extern int ftape_decode_header_segment(__u8 *address); -extern int ftape_read_segment_fraction(const int segment, - void *address, - const ft_read_mode_t read_mode, - const int start, - const int size); -#define ftape_read_segment(segment, address, read_mode) \ - ftape_read_segment_fraction(segment, address, read_mode, \ - 0, FT_SEGMENT_SIZE) -extern void ftape_zap_read_buffers(void); - -#endif /* _FTAPE_READ_H */ diff --git a/drivers/char/ftape/lowlevel/ftape-rw.c b/drivers/char/ftape/lowlevel/ftape-rw.c deleted file mode 100644 index c0d6dc2cbfd..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-rw.c +++ /dev/null @@ -1,1092 +0,0 @@ -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-rw.c,v $ - * $Revision: 1.7 $ - * $Date: 1997/10/28 14:26:49 $ - * - * This file contains some common code for the segment read and - * segment write routines for the QIC-117 floppy-tape driver for - * Linux. - */ - -#include -#include - -#include -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/fdc-io.h" -#include "../lowlevel/ftape-init.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-read.h" -#include "../lowlevel/ftape-ecc.h" -#include "../lowlevel/ftape-bsm.h" - -/* Global vars. - */ -int ft_nr_buffers; -buffer_struct *ft_buffer[FT_MAX_NR_BUFFERS]; -static volatile int ft_head; -static volatile int ft_tail; /* not volatile but need same type as head */ -int fdc_setup_error; -location_record ft_location = {-1, 0}; -volatile int ftape_tape_running; - -/* Local vars. - */ -static int overrun_count_offset; -static int inhibit_correction; - -/* maxmimal allowed overshoot when fast seeking - */ -#define OVERSHOOT_LIMIT 10 - -/* Increment cyclic buffer nr. - */ -buffer_struct *ftape_next_buffer(ft_buffer_queue_t pos) -{ - switch (pos) { - case ft_queue_head: - if (++ft_head >= ft_nr_buffers) { - ft_head = 0; - } - return ft_buffer[ft_head]; - case ft_queue_tail: - if (++ft_tail >= ft_nr_buffers) { - ft_tail = 0; - } - return ft_buffer[ft_tail]; - default: - return NULL; - } -} -int ftape_buffer_id(ft_buffer_queue_t pos) -{ - switch(pos) { - case ft_queue_head: return ft_head; - case ft_queue_tail: return ft_tail; - default: return -1; - } -} -buffer_struct *ftape_get_buffer(ft_buffer_queue_t pos) -{ - switch(pos) { - case ft_queue_head: return ft_buffer[ft_head]; - case ft_queue_tail: return ft_buffer[ft_tail]; - default: return NULL; - } -} -void ftape_reset_buffer(void) -{ - ft_head = ft_tail = 0; -} - -buffer_state_enum ftape_set_state(buffer_state_enum new_state) -{ - buffer_state_enum old_state = ft_driver_state; - - ft_driver_state = new_state; - return old_state; -} -/* Calculate Floppy Disk Controller and DMA parameters for a segment. - * head: selects buffer struct in array. - * offset: number of physical sectors to skip (including bad ones). - * count: number of physical sectors to handle (including bad ones). - */ -static int setup_segment(buffer_struct * buff, - int segment_id, - unsigned int sector_offset, - unsigned int sector_count, - int retry) -{ - SectorMap offset_mask; - SectorMap mask; - TRACE_FUN(ft_t_any); - - buff->segment_id = segment_id; - buff->sector_offset = sector_offset; - buff->remaining = sector_count; - buff->head = segment_id / ftape_segments_per_head; - buff->cyl = (segment_id % ftape_segments_per_head) / ftape_segments_per_cylinder; - buff->sect = (segment_id % ftape_segments_per_cylinder) * FT_SECTORS_PER_SEGMENT + 1; - buff->deleted = 0; - offset_mask = (1 << buff->sector_offset) - 1; - mask = ftape_get_bad_sector_entry(segment_id) & offset_mask; - while (mask) { - if (mask & 1) { - offset_mask >>= 1; /* don't count bad sector */ - } - mask >>= 1; - } - buff->data_offset = count_ones(offset_mask); /* good sectors to skip */ - buff->ptr = buff->address + buff->data_offset * FT_SECTOR_SIZE; - TRACE(ft_t_flow, "data offset = %d sectors", buff->data_offset); - if (retry) { - buff->soft_error_map &= offset_mask; /* keep skipped part */ - } else { - buff->hard_error_map = buff->soft_error_map = 0; - } - buff->bad_sector_map = ftape_get_bad_sector_entry(buff->segment_id); - if (buff->bad_sector_map != 0) { - TRACE(ft_t_noise, "segment: %d, bad sector map: %08lx", - buff->segment_id, (long)buff->bad_sector_map); - } else { - TRACE(ft_t_flow, "segment: %d", buff->segment_id); - } - if (buff->sector_offset > 0) { - buff->bad_sector_map >>= buff->sector_offset; - } - if (buff->sector_offset != 0 || buff->remaining != FT_SECTORS_PER_SEGMENT) { - TRACE(ft_t_flow, "sector offset = %d, count = %d", - buff->sector_offset, buff->remaining); - } - /* Segments with 3 or less sectors are not written with valid - * data because there is no space left for the ecc. The - * data written is whatever happens to be in the buffer. - * Reading such a segment will return a zero byte-count. - * To allow us to read/write segments with all bad sectors - * we fake one readable sector in the segment. This - * prevents having to handle these segments in a very - * special way. It is not important if the reading of this - * bad sector fails or not (the data is ignored). It is - * only read to keep the driver running. - * - * The QIC-40/80 spec. has no information on how to handle - * this case, so this is my interpretation. - */ - if (buff->bad_sector_map == EMPTY_SEGMENT) { - TRACE(ft_t_flow, "empty segment %d, fake first sector good", - buff->segment_id); - if (buff->ptr != buff->address) { - TRACE(ft_t_bug, "This is a bug: %p/%p", - buff->ptr, buff->address); - } - buff->bad_sector_map = FAKE_SEGMENT; - } - fdc_setup_error = 0; - buff->next_segment = segment_id + 1; - TRACE_EXIT 0; -} - -/* Calculate Floppy Disk Controller and DMA parameters for a new segment. - */ -int ftape_setup_new_segment(buffer_struct * buff, int segment_id, int skip) -{ - int result = 0; - static int old_segment_id = -1; - static buffer_state_enum old_ft_driver_state = idle; - int retry = 0; - unsigned offset = 0; - int count = FT_SECTORS_PER_SEGMENT; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_flow, "%s segment %d (old = %d)", - (ft_driver_state == reading || ft_driver_state == verifying) - ? "reading" : "writing", - segment_id, old_segment_id); - if (ft_driver_state != old_ft_driver_state) { /* when verifying */ - old_segment_id = -1; - old_ft_driver_state = ft_driver_state; - } - if (segment_id == old_segment_id) { - ++buff->retry; - ++ft_history.retries; - TRACE(ft_t_flow, "setting up for retry nr %d", buff->retry); - retry = 1; - if (skip && buff->skip > 0) { /* allow skip on retry */ - offset = buff->skip; - count -= offset; - TRACE(ft_t_flow, "skipping %d sectors", offset); - } - } else { - buff->retry = 0; - buff->skip = 0; - old_segment_id = segment_id; - } - result = setup_segment(buff, segment_id, offset, count, retry); - TRACE_EXIT result; -} - -/* Determine size of next cluster of good sectors. - */ -int ftape_calc_next_cluster(buffer_struct * buff) -{ - /* Skip bad sectors. - */ - while (buff->remaining > 0 && (buff->bad_sector_map & 1) != 0) { - buff->bad_sector_map >>= 1; - ++buff->sector_offset; - --buff->remaining; - } - /* Find next cluster of good sectors - */ - if (buff->bad_sector_map == 0) { /* speed up */ - buff->sector_count = buff->remaining; - } else { - SectorMap map = buff->bad_sector_map; - - buff->sector_count = 0; - while (buff->sector_count < buff->remaining && (map & 1) == 0) { - ++buff->sector_count; - map >>= 1; - } - } - return buff->sector_count; -} - -/* if just passed the last segment on a track, wait for BOT - * or EOT mark. - */ -int ftape_handle_logical_eot(void) -{ - TRACE_FUN(ft_t_flow); - - if (ft_runner_status == logical_eot) { - int status; - - TRACE(ft_t_noise, "tape at logical EOT"); - TRACE_CATCH(ftape_ready_wait(ftape_timeout.seek, &status),); - if ((status & (QIC_STATUS_AT_BOT | QIC_STATUS_AT_EOT)) == 0) { - TRACE_ABORT(-EIO, ft_t_err, "eot/bot not reached"); - } - ft_runner_status = end_of_tape; - } - if (ft_runner_status == end_of_tape) { - TRACE(ft_t_noise, "runner stopped because of logical EOT"); - ft_runner_status = idle; - } - TRACE_EXIT 0; -} - -static int check_bot_eot(int status) -{ - TRACE_FUN(ft_t_flow); - - if (status & (QIC_STATUS_AT_BOT | QIC_STATUS_AT_EOT)) { - ft_location.bot = ((ft_location.track & 1) == 0 ? - (status & QIC_STATUS_AT_BOT) != 0: - (status & QIC_STATUS_AT_EOT) != 0); - ft_location.eot = !ft_location.bot; - ft_location.segment = (ft_location.track + - (ft_location.bot ? 0 : 1)) * ft_segments_per_track - 1; - ft_location.sector = -1; - ft_location.known = 1; - TRACE(ft_t_flow, "tape at logical %s", - ft_location.bot ? "bot" : "eot"); - TRACE(ft_t_flow, "segment = %d", ft_location.segment); - } else { - ft_location.known = 0; - } - TRACE_EXIT ft_location.known; -} - -/* Read Id of first sector passing tape head. - */ -static int ftape_read_id(void) -{ - int status; - __u8 out[2]; - TRACE_FUN(ft_t_any); - - /* Assume tape is running on entry, be able to handle - * situation where it stopped or is stopping. - */ - ft_location.known = 0; /* default is location not known */ - out[0] = FDC_READID; - out[1] = ft_drive_sel; - TRACE_CATCH(fdc_command(out, 2),); - switch (fdc_interrupt_wait(20 * FT_SECOND)) { - case 0: - if (fdc_sect == 0) { - if (ftape_report_drive_status(&status) >= 0 && - (status & QIC_STATUS_READY)) { - ftape_tape_running = 0; - TRACE(ft_t_flow, "tape has stopped"); - check_bot_eot(status); - } - } else { - ft_location.known = 1; - ft_location.segment = (ftape_segments_per_head - * fdc_head - + ftape_segments_per_cylinder - * fdc_cyl - + (fdc_sect - 1) - / FT_SECTORS_PER_SEGMENT); - ft_location.sector = ((fdc_sect - 1) - % FT_SECTORS_PER_SEGMENT); - ft_location.eot = ft_location.bot = 0; - } - break; - case -ETIME: - /* Didn't find id on tape, must be near end: Wait - * until stopped. - */ - if (ftape_ready_wait(FT_FOREVER, &status) >= 0) { - ftape_tape_running = 0; - TRACE(ft_t_flow, "tape has stopped"); - check_bot_eot(status); - } - break; - default: - /* Interrupted or otherwise failing - * fdc_interrupt_wait() - */ - TRACE(ft_t_err, "fdc_interrupt_wait failed"); - break; - } - if (!ft_location.known) { - TRACE_ABORT(-EIO, ft_t_flow, "no id found"); - } - if (ft_location.sector == 0) { - TRACE(ft_t_flow, "passing segment %d/%d", - ft_location.segment, ft_location.sector); - } else { - TRACE(ft_t_fdc_dma, "passing segment %d/%d", - ft_location.segment, ft_location.sector); - } - TRACE_EXIT 0; -} - -static int logical_forward(void) -{ - ftape_tape_running = 1; - return ftape_command(QIC_LOGICAL_FORWARD); -} - -int ftape_stop_tape(int *pstatus) -{ - int retry = 0; - int result; - TRACE_FUN(ft_t_flow); - - do { - result = ftape_command_wait(QIC_STOP_TAPE, - ftape_timeout.stop, pstatus); - if (result == 0) { - if ((*pstatus & QIC_STATUS_READY) == 0) { - result = -EIO; - } else { - ftape_tape_running = 0; - } - } - } while (result < 0 && ++retry <= 3); - if (result < 0) { - TRACE(ft_t_err, "failed ! (fatal)"); - } - TRACE_EXIT result; -} - -int ftape_dumb_stop(void) -{ - int result; - int status; - TRACE_FUN(ft_t_flow); - - /* Abort current fdc operation if it's busy (probably read - * or write operation pending) with a reset. - */ - if (fdc_ready_wait(100 /* usec */) < 0) { - TRACE(ft_t_noise, "aborting fdc operation"); - fdc_reset(); - } - /* Reading id's after the last segment on a track may fail - * but eventually the drive will become ready (logical eot). - */ - result = ftape_report_drive_status(&status); - ft_location.known = 0; - do { - if (result == 0 && status & QIC_STATUS_READY) { - /* Tape is not running any more. - */ - TRACE(ft_t_noise, "tape already halted"); - check_bot_eot(status); - ftape_tape_running = 0; - } else if (ftape_tape_running) { - /* Tape is (was) still moving. - */ -#ifdef TESTING - ftape_read_id(); -#endif - result = ftape_stop_tape(&status); - } else { - /* Tape not yet ready but stopped. - */ - result = ftape_ready_wait(ftape_timeout.pause,&status); - } - } while (ftape_tape_running - && !(sigtestsetmask(¤t->pending.signal, _NEVER_BLOCK))); -#ifndef TESTING - ft_location.known = 0; -#endif - if (ft_runner_status == aborting || ft_runner_status == do_abort) { - ft_runner_status = idle; - } - TRACE_EXIT result; -} - -/* Wait until runner has finished tail buffer. - * - */ -int ftape_wait_segment(buffer_state_enum state) -{ - int status; - int result = 0; - TRACE_FUN(ft_t_flow); - - while (ft_buffer[ft_tail]->status == state) { - TRACE(ft_t_flow, "state: %d", ft_buffer[ft_tail]->status); - /* First buffer still being worked on, wait up to timeout. - * - * Note: we check two times for being killed. 50 - * seconds are quite long. Note that - * fdc_interrupt_wait() is not killable by any - * means. ftape_read_segment() wants us to return - * -EINTR in case of a signal. - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - result = fdc_interrupt_wait(50 * FT_SECOND); - FT_SIGNAL_EXIT(_DONT_BLOCK); - if (result < 0) { - TRACE_ABORT(result, - ft_t_err, "fdc_interrupt_wait failed"); - } - if (fdc_setup_error) { - /* recover... FIXME */ - TRACE_ABORT(-EIO, ft_t_err, "setup error"); - } - } - if (ft_buffer[ft_tail]->status != error) { - TRACE_EXIT 0; - } - TRACE_CATCH(ftape_report_drive_status(&status),); - TRACE(ft_t_noise, "ftape_report_drive_status: 0x%02x", status); - if ((status & QIC_STATUS_READY) && - (status & QIC_STATUS_ERROR)) { - unsigned int error; - qic117_cmd_t command; - - /* Report and clear error state. - * In case the drive can't operate at the selected - * rate, select the next lower data rate. - */ - ftape_report_error(&error, &command, 1); - if (error == 31 && command == QIC_LOGICAL_FORWARD) { - /* drive does not accept this data rate */ - if (ft_data_rate > 250) { - TRACE(ft_t_info, - "Probable data rate conflict"); - TRACE(ft_t_info, - "Lowering data rate to %d Kbps", - ft_data_rate / 2); - ftape_half_data_rate(); - if (ft_buffer[ft_tail]->retry > 0) { - /* give it a chance */ - --ft_buffer[ft_tail]->retry; - } - } else { - /* no rate is accepted... */ - TRACE(ft_t_err, "We're dead :("); - } - } else { - TRACE(ft_t_err, "Unknown error"); - } - TRACE_EXIT -EIO; /* g.p. error */ - } - TRACE_EXIT 0; -} - -/* forward */ static int seek_forward(int segment_id, int fast); - -static int fast_seek(int count, int reverse) -{ - int result = 0; - int status; - TRACE_FUN(ft_t_flow); - - if (count > 0) { - /* If positioned at begin or end of tape, fast seeking needs - * special treatment. - * Starting from logical bot needs a (slow) seek to the first - * segment before the high speed seek. Most drives do this - * automatically but some older don't, so we treat them - * all the same. - * Starting from logical eot is even more difficult because - * we cannot (slow) reverse seek to the last segment. - * TO BE IMPLEMENTED. - */ - inhibit_correction = 0; - if (ft_location.known && - ((ft_location.bot && !reverse) || - (ft_location.eot && reverse))) { - if (!reverse) { - /* (slow) skip to first segment on a track - */ - seek_forward(ft_location.track * ft_segments_per_track, 0); - --count; - } else { - /* When seeking backwards from - * end-of-tape the number of erased - * gaps found seems to be higher than - * expected. Therefor the drive must - * skip some more segments than - * calculated, but we don't know how - * many. Thus we will prevent the - * re-calculation of offset and - * overshoot when seeking backwards. - */ - inhibit_correction = 1; - count += 3; /* best guess */ - } - } - } else { - TRACE(ft_t_flow, "warning: zero or negative count: %d", count); - } - if (count > 0) { - int i; - int nibbles = count > 255 ? 3 : 2; - - if (count > 4095) { - TRACE(ft_t_noise, "skipping clipped at 4095 segment"); - count = 4095; - } - /* Issue this tape command first. */ - if (!reverse) { - TRACE(ft_t_noise, "skipping %d segment(s)", count); - result = ftape_command(nibbles == 3 ? - QIC_SKIP_EXTENDED_FORWARD : QIC_SKIP_FORWARD); - } else { - TRACE(ft_t_noise, "backing up %d segment(s)", count); - result = ftape_command(nibbles == 3 ? - QIC_SKIP_EXTENDED_REVERSE : QIC_SKIP_REVERSE); - } - if (result < 0) { - TRACE(ft_t_noise, "Skip command failed"); - } else { - --count; /* 0 means one gap etc. */ - for (i = 0; i < nibbles; ++i) { - if (result >= 0) { - result = ftape_parameter(count & 15); - count /= 16; - } - } - result = ftape_ready_wait(ftape_timeout.rewind, &status); - if (result >= 0) { - ftape_tape_running = 0; - } - } - } - TRACE_EXIT result; -} - -static int validate(int id) -{ - /* Check to see if position found is off-track as reported - * once. Because all tracks in one direction lie next to - * each other, if off-track the error will be approximately - * 2 * ft_segments_per_track. - */ - if (ft_location.track == -1) { - return 1; /* unforseen situation, don't generate error */ - } else { - /* Use margin of ft_segments_per_track on both sides - * because ftape needs some margin and the error we're - * looking for is much larger ! - */ - int lo = (ft_location.track - 1) * ft_segments_per_track; - int hi = (ft_location.track + 2) * ft_segments_per_track; - - return (id >= lo && id < hi); - } -} - -static int seek_forward(int segment_id, int fast) -{ - int failures = 0; - int count; - static int margin = 1; /* fixed: stop this before target */ - static int overshoot = 1; - static int min_count = 8; - int expected = -1; - int target = segment_id - margin; - int fast_seeking; - int prev_segment = ft_location.segment; - TRACE_FUN(ft_t_flow); - - if (!ft_location.known) { - TRACE_ABORT(-EIO, ft_t_err, - "fatal: cannot seek from unknown location"); - } - if (!validate(segment_id)) { - ftape_sleep(1 * FT_SECOND); - ft_failure = 1; - TRACE_ABORT(-EIO, ft_t_err, - "fatal: head off track (bad hardware?)"); - } - TRACE(ft_t_noise, "from %d/%d to %d/0 - %d", - ft_location.segment, ft_location.sector,segment_id,margin); - count = target - ft_location.segment - overshoot; - fast_seeking = (fast && - count > (min_count + (ft_location.bot ? 1 : 0))); - if (fast_seeking) { - TRACE(ft_t_noise, "fast skipping %d segments", count); - expected = segment_id - margin; - fast_seek(count, 0); - } - if (!ftape_tape_running) { - logical_forward(); - } - while (ft_location.segment < segment_id) { - /* This requires at least one sector in a (bad) segment to - * have a valid and readable sector id ! - * It looks like this is not guaranteed, so we must try - * to find a way to skip an EMPTY_SEGMENT. !!! FIXME !!! - */ - if (ftape_read_id() < 0 || !ft_location.known || - sigtestsetmask(¤t->pending.signal, _DONT_BLOCK)) { - ft_location.known = 0; - if (!ftape_tape_running || - ++failures > FT_SECTORS_PER_SEGMENT) { - TRACE_ABORT(-EIO, ft_t_err, - "read_id failed completely"); - } - FT_SIGNAL_EXIT(_DONT_BLOCK); - TRACE(ft_t_flow, "read_id failed, retry (%d)", - failures); - continue; - } - if (fast_seeking) { - TRACE(ft_t_noise, "ended at %d/%d (%d,%d)", - ft_location.segment, ft_location.sector, - overshoot, inhibit_correction); - if (!inhibit_correction && - (ft_location.segment < expected || - ft_location.segment > expected + margin)) { - int error = ft_location.segment - expected; - TRACE(ft_t_noise, - "adjusting overshoot from %d to %d", - overshoot, overshoot + error); - overshoot += error; - /* All overshoots have the same - * direction, so it should never - * become negative, but who knows. - */ - if (overshoot < -5 || - overshoot > OVERSHOOT_LIMIT) { - if (overshoot < 0) { - /* keep sane value */ - overshoot = -5; - } else { - /* keep sane value */ - overshoot = OVERSHOOT_LIMIT; - } - TRACE(ft_t_noise, - "clipped overshoot to %d", - overshoot); - } - } - fast_seeking = 0; - } - if (ft_location.known) { - if (ft_location.segment > prev_segment + 1) { - TRACE(ft_t_noise, - "missed segment %d while skipping", - prev_segment + 1); - } - prev_segment = ft_location.segment; - } - } - if (ft_location.segment > segment_id) { - TRACE_ABORT(-EIO, - ft_t_noise, "failed: skip ended at segment %d/%d", - ft_location.segment, ft_location.sector); - } - TRACE_EXIT 0; -} - -static int skip_reverse(int segment_id, int *pstatus) -{ - int failures = 0; - static int overshoot = 1; - static int min_rewind = 2; /* 1 + overshoot */ - static const int margin = 1; /* stop this before target */ - int expected = 0; - int count = 1; - int short_seek; - int target = segment_id - margin; - TRACE_FUN(ft_t_flow); - - if (ft_location.known && !validate(segment_id)) { - ftape_sleep(1 * FT_SECOND); - ft_failure = 1; - TRACE_ABORT(-EIO, ft_t_err, - "fatal: head off track (bad hardware?)"); - } - do { - if (!ft_location.known) { - TRACE(ft_t_warn, "warning: location not known"); - } - TRACE(ft_t_noise, "from %d/%d to %d/0 - %d", - ft_location.segment, ft_location.sector, - segment_id, margin); - /* min_rewind == 1 + overshoot_when_doing_minimum_rewind - * overshoot == overshoot_when_doing_larger_rewind - * Initially min_rewind == 1 + overshoot, optimization - * of both values will be done separately. - * overshoot and min_rewind can be negative as both are - * sums of three components: - * any_overshoot == rewind_overshoot - - * stop_overshoot - - * start_overshoot - */ - if (ft_location.segment - target - (min_rewind - 1) < 1) { - short_seek = 1; - } else { - count = ft_location.segment - target - overshoot; - short_seek = (count < 1); - } - if (short_seek) { - count = 1; /* do shortest rewind */ - expected = ft_location.segment - min_rewind; - if (expected/ft_segments_per_track != ft_location.track) { - expected = (ft_location.track * - ft_segments_per_track); - } - } else { - expected = target; - } - fast_seek(count, 1); - logical_forward(); - if (ftape_read_id() < 0 || !ft_location.known || - (sigtestsetmask(¤t->pending.signal, _DONT_BLOCK))) { - if ((!ftape_tape_running && !ft_location.known) || - ++failures > FT_SECTORS_PER_SEGMENT) { - TRACE_ABORT(-EIO, ft_t_err, - "read_id failed completely"); - } - FT_SIGNAL_EXIT(_DONT_BLOCK); - TRACE_CATCH(ftape_report_drive_status(pstatus),); - TRACE(ft_t_noise, "ftape_read_id failed, retry (%d)", - failures); - continue; - } - TRACE(ft_t_noise, "ended at %d/%d (%d,%d,%d)", - ft_location.segment, ft_location.sector, - min_rewind, overshoot, inhibit_correction); - if (!inhibit_correction && - (ft_location.segment < expected || - ft_location.segment > expected + margin)) { - int error = expected - ft_location.segment; - if (short_seek) { - TRACE(ft_t_noise, - "adjusting min_rewind from %d to %d", - min_rewind, min_rewind + error); - min_rewind += error; - if (min_rewind < -5) { - /* is this right ? FIXME ! */ - /* keep sane value */ - min_rewind = -5; - TRACE(ft_t_noise, - "clipped min_rewind to %d", - min_rewind); - } - } else { - TRACE(ft_t_noise, - "adjusting overshoot from %d to %d", - overshoot, overshoot + error); - overshoot += error; - if (overshoot < -5 || - overshoot > OVERSHOOT_LIMIT) { - if (overshoot < 0) { - /* keep sane value */ - overshoot = -5; - } else { - /* keep sane value */ - overshoot = OVERSHOOT_LIMIT; - } - TRACE(ft_t_noise, - "clipped overshoot to %d", - overshoot); - } - } - } - } while (ft_location.segment > segment_id); - if (ft_location.known) { - TRACE(ft_t_noise, "current location: %d/%d", - ft_location.segment, ft_location.sector); - } - TRACE_EXIT 0; -} - -static int determine_position(void) -{ - int retry = 0; - int status; - int result; - TRACE_FUN(ft_t_flow); - - if (!ftape_tape_running) { - /* This should only happen if tape is stopped by isr. - */ - TRACE(ft_t_flow, "waiting for tape stop"); - if (ftape_ready_wait(ftape_timeout.pause, &status) < 0) { - TRACE(ft_t_flow, "drive still running (fatal)"); - ftape_tape_running = 1; /* ? */ - } - } else { - ftape_report_drive_status(&status); - } - if (status & QIC_STATUS_READY) { - /* Drive must be ready to check error state ! - */ - TRACE(ft_t_flow, "drive is ready"); - if (status & QIC_STATUS_ERROR) { - unsigned int error; - qic117_cmd_t command; - - /* Report and clear error state, try to continue. - */ - TRACE(ft_t_flow, "error status set"); - ftape_report_error(&error, &command, 1); - ftape_ready_wait(ftape_timeout.reset, &status); - ftape_tape_running = 0; /* ? */ - } - if (check_bot_eot(status)) { - if (ft_location.bot) { - if ((status & QIC_STATUS_READY) == 0) { - /* tape moving away from - * bot/eot, let's see if we - * can catch up with the first - * segment on this track. - */ - } else { - TRACE(ft_t_flow, - "start tape from logical bot"); - logical_forward(); /* start moving */ - } - } else { - if ((status & QIC_STATUS_READY) == 0) { - TRACE(ft_t_noise, "waiting for logical end of track"); - result = ftape_ready_wait(ftape_timeout.reset, &status); - /* error handling needed ? */ - } else { - TRACE(ft_t_noise, - "tape at logical end of track"); - } - } - } else { - TRACE(ft_t_flow, "start tape"); - logical_forward(); /* start moving */ - ft_location.known = 0; /* not cleared by logical forward ! */ - } - } - /* tape should be moving now, start reading id's - */ - while (!ft_location.known && - retry++ < FT_SECTORS_PER_SEGMENT && - (result = ftape_read_id()) < 0) { - - TRACE(ft_t_flow, "location unknown"); - - /* exit on signal - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - - /* read-id somehow failed, tape may - * have reached end or some other - * error happened. - */ - TRACE(ft_t_flow, "read-id failed"); - TRACE_CATCH(ftape_report_drive_status(&status),); - TRACE(ft_t_err, "ftape_report_drive_status: 0x%02x", status); - if (status & QIC_STATUS_READY) { - ftape_tape_running = 0; - TRACE(ft_t_noise, "tape stopped for unknown reason! " - "status = 0x%02x", status); - if (status & QIC_STATUS_ERROR || - !check_bot_eot(status)) { - /* oops, tape stopped but not at end! - */ - TRACE_EXIT -EIO; - } - } - } - TRACE(ft_t_flow, - "tape is positioned at segment %d", ft_location.segment); - TRACE_EXIT ft_location.known ? 0 : -EIO; -} - -/* Get the tape running and position it just before the - * requested segment. - * Seek tape-track and reposition as needed. - */ -int ftape_start_tape(int segment_id, int sector_offset) -{ - int track = segment_id / ft_segments_per_track; - int result = -EIO; - int status; - static int last_segment = -1; - static int bad_bus_timing = 0; - /* number of segments passing the head between starting the tape - * and being able to access the first sector. - */ - static int start_offset = 1; - int retry; - TRACE_FUN(ft_t_flow); - - /* If sector_offset > 0, seek into wanted segment instead of - * into previous. - * This allows error recovery if a part of the segment is bad - * (erased) causing the tape drive to generate an index pulse - * thus causing a no-data error before the requested sector - * is reached. - */ - ftape_tape_running = 0; - TRACE(ft_t_noise, "target segment: %d/%d%s", segment_id, sector_offset, - ft_buffer[ft_head]->retry > 0 ? " retry" : ""); - if (ft_buffer[ft_head]->retry > 0) { /* this is a retry */ - int dist = segment_id - last_segment; - - if ((int)ft_history.overrun_errors < overrun_count_offset) { - overrun_count_offset = ft_history.overrun_errors; - } else if (dist < 0 || dist > 50) { - overrun_count_offset = ft_history.overrun_errors; - } else if ((ft_history.overrun_errors - - overrun_count_offset) >= 8) { - if (ftape_increase_threshold() >= 0) { - --ft_buffer[ft_head]->retry; - overrun_count_offset = - ft_history.overrun_errors; - TRACE(ft_t_warn, "increased threshold because " - "of excessive overrun errors"); - } else if (!bad_bus_timing && ft_data_rate >= 1000) { - ftape_half_data_rate(); - --ft_buffer[ft_head]->retry; - bad_bus_timing = 1; - overrun_count_offset = - ft_history.overrun_errors; - TRACE(ft_t_warn, "reduced datarate because " - "of excessive overrun errors"); - } - } - } - last_segment = segment_id; - if (ft_location.track != track || - (ftape_might_be_off_track && ft_buffer[ft_head]->retry== 0)) { - /* current track unknown or not equal to destination - */ - ftape_ready_wait(ftape_timeout.seek, &status); - ftape_seek_head_to_track(track); - /* overrun_count_offset = ft_history.overrun_errors; */ - } - result = -EIO; - retry = 0; - while (result < 0 && - retry++ <= 5 && - !ft_failure && - !(sigtestsetmask(¤t->pending.signal, _DONT_BLOCK))) { - - if (retry && start_offset < 5) { - start_offset ++; - } - /* Check if we are able to catch the requested - * segment in time. - */ - if ((ft_location.known || (determine_position() == 0)) && - ft_location.segment >= - (segment_id - - ((ftape_tape_running || ft_location.bot) - ? 0 : start_offset))) { - /* Too far ahead (in or past target segment). - */ - if (ftape_tape_running) { - if ((result = ftape_stop_tape(&status)) < 0) { - TRACE(ft_t_err, - "stop tape failed with code %d", - result); - break; - } - TRACE(ft_t_noise, "tape stopped"); - ftape_tape_running = 0; - } - TRACE(ft_t_noise, "repositioning"); - ++ft_history.rewinds; - if (segment_id % ft_segments_per_track < start_offset){ - TRACE(ft_t_noise, "end of track condition\n" - KERN_INFO "segment_id : %d\n" - KERN_INFO "ft_segments_per_track: %d\n" - KERN_INFO "start_offset : %d", - segment_id, ft_segments_per_track, - start_offset); - - /* If seeking to first segments on - * track better do a complete rewind - * to logical begin of track to get a - * more steady tape motion. - */ - result = ftape_command_wait( - (ft_location.track & 1) - ? QIC_PHYSICAL_FORWARD - : QIC_PHYSICAL_REVERSE, - ftape_timeout.rewind, &status); - check_bot_eot(status); /* update location */ - } else { - result= skip_reverse(segment_id - start_offset, - &status); - } - } - if (!ft_location.known) { - TRACE(ft_t_bug, "panic: location not known"); - result = -EIO; - continue; /* while() will check for failure */ - } - TRACE(ft_t_noise, "current segment: %d/%d", - ft_location.segment, ft_location.sector); - /* We're on the right track somewhere before the - * wanted segment. Start tape movement if needed and - * skip to just before or inside the requested - * segment. Keep tape running. - */ - result = 0; - if (ft_location.segment < - (segment_id - ((ftape_tape_running || ft_location.bot) - ? 0 : start_offset))) { - if (sector_offset > 0) { - result = seek_forward(segment_id, - retry <= 3); - } else { - result = seek_forward(segment_id - 1, - retry <= 3); - } - } - if (result == 0 && - ft_location.segment != - (segment_id - (sector_offset > 0 ? 0 : 1))) { - result = -EIO; - } - } - if (result < 0) { - TRACE(ft_t_err, "failed to reposition"); - } else { - ft_runner_status = running; - } - TRACE_EXIT result; -} diff --git a/drivers/char/ftape/lowlevel/ftape-rw.h b/drivers/char/ftape/lowlevel/ftape-rw.h deleted file mode 100644 index 32f4feeb887..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-rw.h +++ /dev/null @@ -1,111 +0,0 @@ -#ifndef _FTAPE_RW_H -#define _FTAPE_RW_H - -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-rw.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:25 $ - * - * This file contains the definitions for the read and write - * functions for the QIC-117 floppy-tape driver for Linux. - * - * Claus-Justus Heine (1996/09/20): Add definition of format code 6 - * Claus-Justus Heine (1996/10/04): Changed GET/PUT macros to cast to (__u8 *) - * - */ - -#include "../lowlevel/fdc-io.h" -#include "../lowlevel/ftape-init.h" -#include "../lowlevel/ftape-bsm.h" - -#include - -#define GET2(address, offset) get_unaligned((__u16*)((__u8 *)address + offset)) -#define GET4(address, offset) get_unaligned((__u32*)((__u8 *)address + offset)) -#define GET8(address, offset) get_unaligned((__u64*)((__u8 *)address + offset)) -#define PUT2(address, offset , value) put_unaligned((value), (__u16*)((__u8 *)address + offset)) -#define PUT4(address, offset , value) put_unaligned((value), (__u32*)((__u8 *)address + offset)) -#define PUT8(address, offset , value) put_unaligned((value), (__u64*)((__u8 *)address + offset)) - -enum runner_status_enum { - idle = 0, - running, - do_abort, - aborting, - logical_eot, - end_of_tape, -}; - -typedef enum ft_buffer_queue { - ft_queue_head = 0, - ft_queue_tail = 1 -} ft_buffer_queue_t; - - -typedef struct { - int track; /* tape head position */ - volatile int segment; /* current segment */ - volatile int sector; /* sector offset within current segment */ - volatile unsigned int bot; /* logical begin of track */ - volatile unsigned int eot; /* logical end of track */ - volatile unsigned int known; /* validates bot, segment, sector */ -} location_record; - -/* Count nr of 1's in pattern. - */ -static inline int count_ones(unsigned long mask) -{ - int bits; - - for (bits = 0; mask != 0; mask >>= 1) { - if (mask & 1) { - ++bits; - } - } - return bits; -} - -#define FT_MAX_NR_BUFFERS 16 /* arbitrary value */ -/* ftape-rw.c defined global vars. - */ -extern buffer_struct *ft_buffer[FT_MAX_NR_BUFFERS]; -extern int ft_nr_buffers; -extern location_record ft_location; -extern volatile int ftape_tape_running; - -/* ftape-rw.c defined global functions. - */ -extern int ftape_setup_new_segment(buffer_struct * buff, - int segment_id, - int offset); -extern int ftape_calc_next_cluster(buffer_struct * buff); -extern buffer_struct *ftape_next_buffer (ft_buffer_queue_t pos); -extern buffer_struct *ftape_get_buffer (ft_buffer_queue_t pos); -extern int ftape_buffer_id (ft_buffer_queue_t pos); -extern void ftape_reset_buffer(void); -extern void ftape_tape_parameters(__u8 drive_configuration); -extern int ftape_wait_segment(buffer_state_enum state); -extern int ftape_dumb_stop(void); -extern int ftape_start_tape(int segment_id, int offset); -extern int ftape_stop_tape(int *pstatus); -extern int ftape_handle_logical_eot(void); -extern buffer_state_enum ftape_set_state(buffer_state_enum new_state); -#endif /* _FTAPE_RW_H */ diff --git a/drivers/char/ftape/lowlevel/ftape-setup.c b/drivers/char/ftape/lowlevel/ftape-setup.c deleted file mode 100644 index 678340acd0b..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-setup.c +++ /dev/null @@ -1,104 +0,0 @@ -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-setup.c,v $ - * $Revision: 1.7 $ - * $Date: 1997/10/10 09:57:06 $ - * - * This file contains the code for processing the kernel command - * line options for the QIC-40/80/3010/3020 floppy-tape driver - * "ftape" for Linux. - */ - -#include -#include -#include - -#include -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/fdc-io.h" - -static struct param_table { - const char *name; - int *var; - int def_param; - int min; - int max; -} config_params[] __initdata = { -#ifndef CONFIG_FT_NO_TRACE_AT_ALL - { "tracing", &ftape_tracing, 3, ft_t_bug, ft_t_any}, -#endif - { "ioport", &ft_fdc_base, CONFIG_FT_FDC_BASE, 0x0, 0xfff}, - { "irq", &ft_fdc_irq, CONFIG_FT_FDC_IRQ, 2, 15}, - { "dma", &ft_fdc_dma, CONFIG_FT_FDC_DMA, 0, 3}, - { "threshold", &ft_fdc_threshold, CONFIG_FT_FDC_THR, 1, 16}, - { "datarate", &ft_fdc_rate_limit, CONFIG_FT_FDC_MAX_RATE, 500, 2000}, - { "fc10", &ft_probe_fc10, CONFIG_FT_PROBE_FC10, 0, 1}, - { "mach2", &ft_mach2, CONFIG_FT_MACH2, 0, 1} -}; - -static int __init ftape_setup(char *str) -{ - int i; - int param; - int ints[2]; - - TRACE_FUN(ft_t_flow); - - str = get_options(str, ARRAY_SIZE(ints), ints); - if (str) { - for (i=0; i < NR_ITEMS(config_params); i++) { - if (strcmp(str,config_params[i].name) == 0){ - if (ints[0]) { - param = ints[1]; - } else { - param = config_params[i].def_param; - } - if (param < config_params[i].min || - param > config_params[i].max) { - TRACE(ft_t_err, - "parameter %s out of range %d ... %d", - config_params[i].name, - config_params[i].min, - config_params[i].max); - goto out; - } - if(config_params[i].var) { - TRACE(ft_t_info, "%s=%d", str, param); - *config_params[i].var = param; - } - goto out; - } - } - } - if (str) { - TRACE(ft_t_err, "unknown ftape option [%s]", str); - - TRACE(ft_t_err, "allowed options are:"); - for (i=0; i < NR_ITEMS(config_params); i++) { - TRACE(ft_t_err, " %s",config_params[i].name); - } - } else { - TRACE(ft_t_err, "botched ftape option"); - } - out: - TRACE_EXIT 1; -} - -__setup("ftape=", ftape_setup); diff --git a/drivers/char/ftape/lowlevel/ftape-tracing.c b/drivers/char/ftape/lowlevel/ftape-tracing.c deleted file mode 100644 index 7fdc6567440..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-tracing.c +++ /dev/null @@ -1,118 +0,0 @@ -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-tracing.c,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:27 $ - * - * This file contains the reading code - * for the QIC-117 floppy-tape driver for Linux. - */ - -#include -#include "../lowlevel/ftape-tracing.h" - -/* Global vars. - */ -/* tracing - * set it to: to log : - * 0 bugs - * 1 + errors - * 2 + warnings - * 3 + information - * 4 + more information - * 5 + program flow - * 6 + fdc/dma info - * 7 + data flow - * 8 + everything else - */ -ft_trace_t ftape_tracing = ft_t_info; /* Default level: information and up */ -int ftape_function_nest_level; - -/* Local vars. - */ -static __u8 trace_id; -static char spacing[] = "* "; - -void ftape_trace_call(const char *file, const char *name) -{ - char *indent; - - /* Since printk seems not to work with "%*s" format - * we'll use this work-around. - */ - if (ftape_function_nest_level < 0) { - printk(KERN_INFO "function nest level (%d) < 0\n", - ftape_function_nest_level); - ftape_function_nest_level = 0; - } - if (ftape_function_nest_level < sizeof(spacing)) { - indent = (spacing + - sizeof(spacing) - 1 - - ftape_function_nest_level); - } else { - indent = spacing; - } - printk(KERN_INFO "[%03d]%s+%s (%s)\n", - (int) trace_id++, indent, file, name); -} - -void ftape_trace_exit(const char *file, const char *name) -{ - char *indent; - - /* Since printk seems not to work with "%*s" format - * we'll use this work-around. - */ - if (ftape_function_nest_level < 0) { - printk(KERN_INFO "function nest level (%d) < 0\n", ftape_function_nest_level); - ftape_function_nest_level = 0; - } - if (ftape_function_nest_level < sizeof(spacing)) { - indent = (spacing + - sizeof(spacing) - 1 - - ftape_function_nest_level); - } else { - indent = spacing; - } - printk(KERN_INFO "[%03d]%s-%s (%s)\n", - (int) trace_id++, indent, file, name); -} - -void ftape_trace_log(const char *file, const char *function) -{ - char *indent; - - /* Since printk seems not to work with "%*s" format - * we'll use this work-around. - */ - if (ftape_function_nest_level < 0) { - printk(KERN_INFO "function nest level (%d) < 0\n", ftape_function_nest_level); - ftape_function_nest_level = 0; - } - if (ftape_function_nest_level < sizeof(spacing)) { - indent = (spacing + - sizeof(spacing) - 1 - - ftape_function_nest_level); - } else { - indent = spacing; - } - printk(KERN_INFO "[%03d]%s%s (%s) - ", - (int) trace_id++, indent, file, function); -} diff --git a/drivers/char/ftape/lowlevel/ftape-tracing.h b/drivers/char/ftape/lowlevel/ftape-tracing.h deleted file mode 100644 index 2950810c708..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-tracing.h +++ /dev/null @@ -1,179 +0,0 @@ -#ifndef _FTAPE_TRACING_H -#define _FTAPE_TRACING_H - -/* - * Copyright (C) 1994-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-tracing.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:28 $ - * - * This file contains definitions that eases the debugging of the - * QIC-40/80/3010/3020 floppy-tape driver "ftape" for Linux. - */ - -#include - -/* - * Be very careful with TRACE_EXIT and TRACE_ABORT. - * - * if (something) TRACE_EXIT error; - * - * will NOT work. Use - * - * if (something) { - * TRACE_EXIT error; - * } - * - * instead. Maybe a bit dangerous, but save lots of lines of code. - */ - -#define LL_X "%d/%d KB" -#define LL(x) (unsigned int)((__u64)(x)>>10), (unsigned int)((x)&1023) - -typedef enum { - ft_t_nil = -1, - ft_t_bug, - ft_t_err, - ft_t_warn, - ft_t_info, - ft_t_noise, - ft_t_flow, - ft_t_fdc_dma, - ft_t_data_flow, - ft_t_any -} ft_trace_t; - -#ifdef CONFIG_FT_NO_TRACE_AT_ALL -/* the compiler will optimize away most TRACE() macros - */ -#define FT_TRACE_TOP_LEVEL ft_t_bug -#define TRACE_FUN(level) do {} while(0) -#define TRACE_EXIT return -#define TRACE(l, m, i...) \ -{ \ - if ((ft_trace_t)(l) == FT_TRACE_TOP_LEVEL) { \ - printk(KERN_INFO"ftape%s(%s):\n" \ - KERN_INFO m".\n" ,__FILE__, __FUNCTION__ , ##i); \ - } \ -} -#define SET_TRACE_LEVEL(l) if ((l) == (l)) do {} while(0) -#define TRACE_LEVEL FT_TRACE_TOP_LEVEL - -#else - -#ifdef CONFIG_FT_NO_TRACE -/* the compiler will optimize away many TRACE() macros - * the ftape_simple_trace_call() function simply increments - * the function nest level. - */ -#define FT_TRACE_TOP_LEVEL ft_t_warn -#define TRACE_FUN(level) ftape_function_nest_level++ -#define TRACE_EXIT ftape_function_nest_level--; return - -#else -#ifdef CONFIG_FT_FULL_DEBUG -#define FT_TRACE_TOP_LEVEL ft_t_any -#else -#define FT_TRACE_TOP_LEVEL ft_t_flow -#endif -#define TRACE_FUN(level) \ - const ft_trace_t _tracing = level; \ - if (ftape_tracing >= (ft_trace_t)(level) && \ - (ft_trace_t)(level) <= FT_TRACE_TOP_LEVEL) \ - ftape_trace_call(__FILE__, __FUNCTION__); \ - ftape_function_nest_level ++; - -#define TRACE_EXIT \ - --ftape_function_nest_level; \ - if (ftape_tracing >= (ft_trace_t)(_tracing) && \ - (ft_trace_t)(_tracing) <= FT_TRACE_TOP_LEVEL) \ - ftape_trace_exit(__FILE__, __FUNCTION__); \ - return - -#endif - -#define TRACE(l, m, i...) \ -{ \ - if (ftape_tracing >= (ft_trace_t)(l) && \ - (ft_trace_t)(l) <= FT_TRACE_TOP_LEVEL) { \ - ftape_trace_log(__FILE__, __FUNCTION__); \ - printk(m".\n" ,##i); \ - } \ -} - -#define SET_TRACE_LEVEL(l) \ -{ \ - if ((ft_trace_t)(l) <= FT_TRACE_TOP_LEVEL) { \ - ftape_tracing = (ft_trace_t)(l); \ - } else { \ - ftape_tracing = FT_TRACE_TOP_LEVEL; \ - } \ -} -#define TRACE_LEVEL \ -((ftape_tracing <= FT_TRACE_TOP_LEVEL) ? ftape_tracing : FT_TRACE_TOP_LEVEL) - - -/* Global variables declared in tracing.c - */ -extern ft_trace_t ftape_tracing; /* sets default level */ -extern int ftape_function_nest_level; - -/* Global functions declared in tracing.c - */ -extern void ftape_trace_call(const char *file, const char *name); -extern void ftape_trace_exit(const char *file, const char *name); -extern void ftape_trace_log (const char *file, const char *name); - -#endif /* !defined(CONFIG_FT_NO_TRACE_AT_ALL) */ - -/* - * Abort with a message. - */ -#define TRACE_ABORT(res, i...) \ -{ \ - TRACE(i); \ - TRACE_EXIT res; \ -} - -/* The following transforms the common "if(result < 0) ... " into a - * one-liner. - */ -#define _TRACE_CATCH(level, fun, action) \ -{ \ - int _res = (fun); \ - if (_res < 0) { \ - do { action /* */ ; } while(0); \ - TRACE_ABORT(_res, level, "%s failed: %d", #fun, _res); \ - } \ -} - -#define TRACE_CATCH(fun, fail) _TRACE_CATCH(ft_t_err, fun, fail) - -/* Abort the current function when signalled. This doesn't belong here, - * but rather into ftape-rw.h (maybe) - */ -#define FT_SIGNAL_EXIT(sig_mask) \ - if (sigtestsetmask(¤t->pending.signal, sig_mask)) { \ - TRACE_ABORT(-EINTR, \ - ft_t_warn, \ - "interrupted by non-blockable signal"); \ - } - -#endif /* _FTAPE_TRACING_H */ diff --git a/drivers/char/ftape/lowlevel/ftape-write.c b/drivers/char/ftape/lowlevel/ftape-write.c deleted file mode 100644 index 45601ec801e..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-write.c +++ /dev/null @@ -1,336 +0,0 @@ -/* - * Copyright (C) 1993-1995 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-write.c,v $ - * $Revision: 1.3.4.1 $ - * $Date: 1997/11/14 18:07:04 $ - * - * This file contains the writing code - * for the QIC-117 floppy-tape driver for Linux. - */ - -#include -#include -#include - -#include -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-write.h" -#include "../lowlevel/ftape-read.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/ftape-ecc.h" -#include "../lowlevel/ftape-bsm.h" -#include "../lowlevel/fdc-isr.h" - -/* Global vars. - */ - -/* Local vars. - */ -static int last_write_failed; - -void ftape_zap_write_buffers(void) -{ - int i; - - for (i = 0; i < ft_nr_buffers; ++i) { - ft_buffer[i]->status = done; - } - ftape_reset_buffer(); -} - -static int copy_and_gen_ecc(void *destination, - const void *source, - const SectorMap bad_sector_map) -{ - int result; - struct memory_segment mseg; - int bads = count_ones(bad_sector_map); - TRACE_FUN(ft_t_any); - - if (bads > 0) { - TRACE(ft_t_noise, "bad sectors in map: %d", bads); - } - if (bads + 3 >= FT_SECTORS_PER_SEGMENT) { - TRACE(ft_t_noise, "empty segment"); - mseg.blocks = 0; /* skip entire segment */ - result = 0; /* nothing written */ - } else { - mseg.blocks = FT_SECTORS_PER_SEGMENT - bads; - mseg.data = destination; - memcpy(mseg.data, source, (mseg.blocks - 3) * FT_SECTOR_SIZE); - result = ftape_ecc_set_segment_parity(&mseg); - if (result < 0) { - TRACE(ft_t_err, "ecc_set_segment_parity failed"); - } else { - result = (mseg.blocks - 3) * FT_SECTOR_SIZE; - } - } - TRACE_EXIT result; -} - - -int ftape_start_writing(const ft_write_mode_t mode) -{ - buffer_struct *head = ftape_get_buffer(ft_queue_head); - int segment_id = head->segment_id; - int result; - buffer_state_enum wanted_state = (mode == FT_WR_DELETE - ? deleting - : writing); - TRACE_FUN(ft_t_flow); - - if ((ft_driver_state != wanted_state) || head->status != waiting) { - TRACE_EXIT 0; - } - ftape_setup_new_segment(head, segment_id, 1); - if (mode == FT_WR_SINGLE) { - /* stop tape instead of pause */ - head->next_segment = 0; - } - ftape_calc_next_cluster(head); /* prepare */ - head->status = ft_driver_state; /* either writing or deleting */ - if (ft_runner_status == idle) { - TRACE(ft_t_noise, - "starting runner for segment %d", segment_id); - TRACE_CATCH(ftape_start_tape(segment_id,head->sector_offset),); - } else { - TRACE(ft_t_noise, "runner not idle, not starting tape"); - } - /* go */ - result = fdc_setup_read_write(head, (mode == FT_WR_DELETE - ? FDC_WRITE_DELETED : FDC_WRITE)); - ftape_set_state(wanted_state); /* should not be necessary */ - TRACE_EXIT result; -} - -/* Wait until all data is actually written to tape. - * - * There is a problem: when the tape runs into logical EOT, then this - * failes. We need to restart the runner in this case. - */ -int ftape_loop_until_writes_done(void) -{ - buffer_struct *head; - TRACE_FUN(ft_t_flow); - - while ((ft_driver_state == writing || ft_driver_state == deleting) && - ftape_get_buffer(ft_queue_head)->status != done) { - /* set the runner status to idle if at lEOT */ - TRACE_CATCH(ftape_handle_logical_eot(), last_write_failed = 1); - /* restart the tape if necessary */ - if (ft_runner_status == idle) { - TRACE(ft_t_noise, "runner is idle, restarting"); - if (ft_driver_state == deleting) { - TRACE_CATCH(ftape_start_writing(FT_WR_DELETE), - last_write_failed = 1); - } else { - TRACE_CATCH(ftape_start_writing(FT_WR_MULTI), - last_write_failed = 1); - } - } - TRACE(ft_t_noise, "tail: %d, head: %d", - ftape_buffer_id(ft_queue_tail), - ftape_buffer_id(ft_queue_head)); - TRACE_CATCH(fdc_interrupt_wait(5 * FT_SECOND), - last_write_failed = 1); - head = ftape_get_buffer(ft_queue_head); - if (head->status == error) { - /* Allow escape from loop when signaled ! - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - if (head->hard_error_map != 0) { - /* Implement hard write error recovery here - */ - } - /* retry this one */ - head->status = waiting; - if (ft_runner_status == aborting) { - ftape_dumb_stop(); - } - if (ft_runner_status != idle) { - TRACE_ABORT(-EIO, ft_t_err, - "unexpected state: " - "ft_runner_status != idle"); - } - ftape_start_writing(ft_driver_state == deleting - ? FT_WR_MULTI : FT_WR_DELETE); - } - TRACE(ft_t_noise, "looping until writes done"); - } - ftape_set_state(idle); - TRACE_EXIT 0; -} - -/* Write given segment from buffer at address to tape. - */ -static int write_segment(const int segment_id, - const void *address, - const ft_write_mode_t write_mode) -{ - int bytes_written = 0; - buffer_struct *tail; - buffer_state_enum wanted_state = (write_mode == FT_WR_DELETE - ? deleting : writing); - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "segment_id = %d", segment_id); - if (ft_driver_state != wanted_state) { - if (ft_driver_state == deleting || - wanted_state == deleting) { - TRACE_CATCH(ftape_loop_until_writes_done(),); - } - TRACE(ft_t_noise, "calling ftape_abort_operation"); - TRACE_CATCH(ftape_abort_operation(),); - ftape_zap_write_buffers(); - ftape_set_state(wanted_state); - } - /* if all buffers full we'll have to wait... - */ - ftape_wait_segment(wanted_state); - tail = ftape_get_buffer(ft_queue_tail); - switch(tail->status) { - case done: - ft_history.defects += count_ones(tail->hard_error_map); - break; - case waiting: - /* this could happen with multiple EMPTY_SEGMENTs, but - * shouldn't happen any more as we re-start the runner even - * with an empty segment. - */ - bytes_written = -EAGAIN; - break; - case error: - /* setup for a retry - */ - tail->status = waiting; - bytes_written = -EAGAIN; /* force retry */ - if (tail->hard_error_map != 0) { - TRACE(ft_t_warn, - "warning: %d hard error(s) in written segment", - count_ones(tail->hard_error_map)); - TRACE(ft_t_noise, "hard_error_map = 0x%08lx", - (long)tail->hard_error_map); - /* Implement hard write error recovery here - */ - } - break; - default: - TRACE_ABORT(-EIO, ft_t_err, - "wait for empty segment failed, tail status: %d", - tail->status); - } - /* should runner stop ? - */ - if (ft_runner_status == aborting) { - buffer_struct *head = ftape_get_buffer(ft_queue_head); - if (head->status == wanted_state) { - head->status = done; /* ???? */ - } - /* don't call abort_operation(), we don't want to zap - * the dma buffers - */ - TRACE_CATCH(ftape_dumb_stop(),); - } else { - /* If just passed last segment on tape: wait for BOT - * or EOT mark. Sets ft_runner_status to idle if at lEOT - * and successful - */ - TRACE_CATCH(ftape_handle_logical_eot(),); - } - if (tail->status == done) { - /* now at least one buffer is empty, fill it with our - * data. skip bad sectors and generate ecc. - * copy_and_gen_ecc return nr of bytes written, range - * 0..29 Kb inclusive! - * - * Empty segments are handled inside coyp_and_gen_ecc() - */ - if (write_mode != FT_WR_DELETE) { - TRACE_CATCH(bytes_written = copy_and_gen_ecc( - tail->address, address, - ftape_get_bad_sector_entry(segment_id)),); - } - tail->segment_id = segment_id; - tail->status = waiting; - tail = ftape_next_buffer(ft_queue_tail); - } - /* Start tape only if all buffers full or flush mode. - * This will give higher probability of streaming. - */ - if (ft_runner_status != running && - ((tail->status == waiting && - ftape_get_buffer(ft_queue_head) == tail) || - write_mode != FT_WR_ASYNC)) { - TRACE_CATCH(ftape_start_writing(write_mode),); - } - TRACE_EXIT bytes_written; -} - -/* Write as much as fits from buffer to the given segment on tape - * and handle retries. - * Return the number of bytes written (>= 0), or: - * -EIO write failed - * -EINTR interrupted by signal - * -ENOSPC device full - */ -int ftape_write_segment(const int segment_id, - const void *buffer, - const ft_write_mode_t flush) -{ - int retry = 0; - int result; - TRACE_FUN(ft_t_flow); - - ft_history.used |= 2; - if (segment_id >= ft_tracks_per_tape*ft_segments_per_track) { - /* tape full */ - TRACE_ABORT(-ENOSPC, ft_t_err, - "invalid segment id: %d (max %d)", - segment_id, - ft_tracks_per_tape * ft_segments_per_track -1); - } - for (;;) { - if ((result = write_segment(segment_id, buffer, flush)) >= 0) { - if (result == 0) { /* empty segment */ - TRACE(ft_t_noise, - "empty segment, nothing written"); - } - TRACE_EXIT result; - } - if (result == -EAGAIN) { - if (++retry > 100) { /* give up */ - TRACE_ABORT(-EIO, ft_t_err, - "write failed, >100 retries in segment"); - } - TRACE(ft_t_warn, "write error, retry %d (%d)", - retry, - ftape_get_buffer(ft_queue_tail)->segment_id); - } else { - TRACE_ABORT(result, ft_t_err, - "write_segment failed, error: %d", result); - } - /* Allow escape from loop when signaled ! - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - } -} diff --git a/drivers/char/ftape/lowlevel/ftape-write.h b/drivers/char/ftape/lowlevel/ftape-write.h deleted file mode 100644 index 0e7f898b7af..00000000000 --- a/drivers/char/ftape/lowlevel/ftape-write.h +++ /dev/null @@ -1,53 +0,0 @@ -#ifndef _FTAPE_WRITE_H -#define _FTAPE_WRITE_H - -/* - * Copyright (C) 1994-1995 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-write.h,v $ - $Author: claus $ - * - $Revision: 1.2 $ - $Date: 1997/10/05 19:18:30 $ - $State: Exp $ - * - * This file contains the definitions for the write functions - * for the QIC-117 floppy-tape driver for Linux. - * - */ - - -/* ftape-write.c defined global functions. - */ -typedef enum { - FT_WR_ASYNC = 0, /* start tape only when all buffers are full */ - FT_WR_MULTI = 1, /* start tape, but don't necessarily stop */ - FT_WR_SINGLE = 2, /* write a single segment and stop afterwards */ - FT_WR_DELETE = 3 /* write deleted data marks */ -} ft_write_mode_t; - -extern int ftape_start_writing(const ft_write_mode_t mode); -extern int ftape_write_segment(const int segment, - const void *address, - const ft_write_mode_t flushing); -extern void ftape_zap_write_buffers(void); -extern int ftape_loop_until_writes_done(void); - -#endif /* _FTAPE_WRITE_H */ - diff --git a/drivers/char/ftape/lowlevel/ftape_syms.c b/drivers/char/ftape/lowlevel/ftape_syms.c deleted file mode 100644 index 8e0dc4a07ca..00000000000 --- a/drivers/char/ftape/lowlevel/ftape_syms.c +++ /dev/null @@ -1,87 +0,0 @@ -/* - * Copyright (C) 1996-1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape_syms.c,v $ - * $Revision: 1.4 $ - * $Date: 1997/10/17 00:03:51 $ - * - * This file contains the symbols that the ftape low level - * part of the QIC-40/80/3010/3020 floppy-tape driver "ftape" - * exports to its high level clients - */ - -#include - -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-init.h" -#include "../lowlevel/fdc-io.h" -#include "../lowlevel/ftape-read.h" -#include "../lowlevel/ftape-write.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/ftape-bsm.h" -#include "../lowlevel/ftape-buffer.h" -#include "../lowlevel/ftape-format.h" - -/* bad sector handling from ftape-bsm.c */ -EXPORT_SYMBOL(ftape_get_bad_sector_entry); -EXPORT_SYMBOL(ftape_find_end_of_bsm_list); -/* from ftape-rw.c */ -EXPORT_SYMBOL(ftape_set_state); -/* from ftape-ctl.c */ -EXPORT_SYMBOL(ftape_seek_to_bot); -EXPORT_SYMBOL(ftape_seek_to_eot); -EXPORT_SYMBOL(ftape_abort_operation); -EXPORT_SYMBOL(ftape_get_status); -EXPORT_SYMBOL(ftape_enable); -EXPORT_SYMBOL(ftape_disable); -EXPORT_SYMBOL(ftape_mmap); -EXPORT_SYMBOL(ftape_calibrate_data_rate); -/* from ftape-io.c */ -EXPORT_SYMBOL(ftape_reset_drive); -EXPORT_SYMBOL(ftape_command); -EXPORT_SYMBOL(ftape_parameter); -EXPORT_SYMBOL(ftape_ready_wait); -EXPORT_SYMBOL(ftape_report_operation); -EXPORT_SYMBOL(ftape_report_error); -/* from ftape-read.c */ -EXPORT_SYMBOL(ftape_read_segment_fraction); -EXPORT_SYMBOL(ftape_zap_read_buffers); -EXPORT_SYMBOL(ftape_read_header_segment); -EXPORT_SYMBOL(ftape_decode_header_segment); -/* from ftape-write.c */ -EXPORT_SYMBOL(ftape_write_segment); -EXPORT_SYMBOL(ftape_start_writing); -EXPORT_SYMBOL(ftape_loop_until_writes_done); -/* from ftape-buffer.h */ -EXPORT_SYMBOL(ftape_set_nr_buffers); -/* from ftape-format.h */ -EXPORT_SYMBOL(ftape_format_track); -EXPORT_SYMBOL(ftape_format_status); -EXPORT_SYMBOL(ftape_verify_segment); -/* from tracing.c */ -#ifndef CONFIG_FT_NO_TRACE_AT_ALL -EXPORT_SYMBOL(ftape_tracing); -EXPORT_SYMBOL(ftape_function_nest_level); -EXPORT_SYMBOL(ftape_trace_call); -EXPORT_SYMBOL(ftape_trace_exit); -EXPORT_SYMBOL(ftape_trace_log); -#endif - diff --git a/drivers/char/ftape/zftape/Makefile b/drivers/char/ftape/zftape/Makefile deleted file mode 100644 index 6d91c1f77c0..00000000000 --- a/drivers/char/ftape/zftape/Makefile +++ /dev/null @@ -1,36 +0,0 @@ -# -# Copyright (C) 1996, 1997 Claus-Justus Heine. -# -# This program is free software; you can redistribute it and/or modify -# it under the terms of the GNU General Public License as published by -# the Free Software Foundation; either version 2, or (at your option) -# any later version. -# -# This program is distributed in the hope that it will be useful, -# but WITHOUT ANY WARRANTY; without even the implied warranty of -# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -# GNU General Public License for more details. -# -# You should have received a copy of the GNU General Public License -# along with this program; see the file COPYING. If not, write to -# the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. -# -# $Source: /homes/cvs/ftape-stacked/ftape/zftape/Makefile,v $ -# $Revision: 1.4 $ -# $Date: 1997/10/05 19:18:58 $ -# -# Makefile for the QIC-40/80/3010/3020 zftape interface VFS to -# ftape -# - - -# ZFT_OBSOLETE - enable the MTIOC_ZFTAPE_GETBLKSZ ioctl. You should -# leave this enabled for compatibility with taper. - -obj-$(CONFIG_ZFTAPE) += zftape.o - -zftape-objs := zftape-rw.o zftape-ctl.o zftape-read.o \ - zftape-write.o zftape-vtbl.o zftape-eof.o \ - zftape-init.o zftape-buffers.o zftape_syms.o - -EXTRA_CFLAGS := -DZFT_OBSOLETE diff --git a/drivers/char/ftape/zftape/zftape-buffers.c b/drivers/char/ftape/zftape/zftape-buffers.c deleted file mode 100644 index 7ebce2ec789..00000000000 --- a/drivers/char/ftape/zftape/zftape-buffers.c +++ /dev/null @@ -1,149 +0,0 @@ -/* - * Copyright (C) 1995-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-buffers.c,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:59 $ - * - * This file contains the dynamic buffer allocation routines - * of zftape - */ - -#include -#include -#include -#include - -#include - -#include - -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-eof.h" -#include "../zftape/zftape-ctl.h" -#include "../zftape/zftape-write.h" -#include "../zftape/zftape-read.h" -#include "../zftape/zftape-rw.h" -#include "../zftape/zftape-vtbl.h" - -/* global variables - */ - -/* local varibales - */ -static unsigned int used_memory; -static unsigned int peak_memory; - -void zft_memory_stats(void) -{ - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "Memory usage (vmalloc allocations):\n" - KERN_INFO "total allocated: %d\n" - KERN_INFO "peak allocation: %d", - used_memory, peak_memory); - peak_memory = used_memory; - TRACE_EXIT; -} - -int zft_vcalloc_once(void *new, size_t size) -{ - TRACE_FUN(ft_t_flow); - if (zft_vmalloc_once(new, size) < 0) { - TRACE_EXIT -ENOMEM; - } - memset(*(void **)new, '\0', size); - TRACE_EXIT 0; -} -int zft_vmalloc_once(void *new, size_t size) -{ - TRACE_FUN(ft_t_flow); - - if (*(void **)new != NULL || size == 0) { - TRACE_EXIT 0; - } - if ((*(void **)new = vmalloc(size)) == NULL) { - TRACE_EXIT -ENOMEM; - } - used_memory += size; - if (peak_memory < used_memory) { - peak_memory = used_memory; - } - TRACE_ABORT(0, ft_t_noise, - "allocated buffer @ %p, %zd bytes", *(void **)new, size); -} -int zft_vmalloc_always(void *new, size_t size) -{ - TRACE_FUN(ft_t_flow); - - zft_vfree(new, size); - TRACE_EXIT zft_vmalloc_once(new, size); -} -void zft_vfree(void *old, size_t size) -{ - TRACE_FUN(ft_t_flow); - - if (*(void **)old) { - vfree(*(void **)old); - used_memory -= size; - TRACE(ft_t_noise, "released buffer @ %p, %zd bytes", - *(void **)old, size); - *(void **)old = NULL; - } - TRACE_EXIT; -} - -void *zft_kmalloc(size_t size) -{ - void *new; - - while ((new = kmalloc(size, GFP_KERNEL)) == NULL) { - msleep_interruptible(100); - } - memset(new, 0, size); - used_memory += size; - if (peak_memory < used_memory) { - peak_memory = used_memory; - } - return new; -} - -void zft_kfree(void *old, size_t size) -{ - kfree(old); - used_memory -= size; -} - -/* there are some more buffers that are allocated on demand. - * cleanup_module() calles this function to be sure to have released - * them - */ -void zft_uninit_mem(void) -{ - TRACE_FUN(ft_t_flow); - - zft_vfree(&zft_hseg_buf, FT_SEGMENT_SIZE); - zft_vfree(&zft_deblock_buf, FT_SEGMENT_SIZE); zft_deblock_segment = -1; - zft_free_vtbl(); - if (zft_cmpr_lock(0 /* don't load */) == 0) { - (*zft_cmpr_ops->cleanup)(); - (*zft_cmpr_ops->reset)(); /* unlock it again */ - } - zft_memory_stats(); - TRACE_EXIT; -} diff --git a/drivers/char/ftape/zftape/zftape-buffers.h b/drivers/char/ftape/zftape/zftape-buffers.h deleted file mode 100644 index 798e3128c68..00000000000 --- a/drivers/char/ftape/zftape/zftape-buffers.h +++ /dev/null @@ -1,55 +0,0 @@ -#ifndef _FTAPE_DYNMEM_H -#define _FTAPE_DYNMEM_H - -/* - * Copyright (C) 1995-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-buffers.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:59 $ - * - * memory allocation routines. - * - */ - -/* we do not allocate all of the really large buffer memory before - * someone tries to open the drive. ftape_open() may fail with - * -ENOMEM, but that's better having 200k of vmalloced memory which - * cannot be swapped out. - */ - -extern void zft_memory_stats(void); -extern int zft_vmalloc_once(void *new, size_t size); -extern int zft_vcalloc_once(void *new, size_t size); -extern int zft_vmalloc_always(void *new, size_t size); -extern void zft_vfree(void *old, size_t size); -extern void *zft_kmalloc(size_t size); -extern void zft_kfree(void *old, size_t size); - -/* called by cleanup_module() - */ -extern void zft_uninit_mem(void); - -#endif - - - - - - - diff --git a/drivers/char/ftape/zftape/zftape-ctl.c b/drivers/char/ftape/zftape/zftape-ctl.c deleted file mode 100644 index 22ba0f5d00c..00000000000 --- a/drivers/char/ftape/zftape/zftape-ctl.c +++ /dev/null @@ -1,1417 +0,0 @@ -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-ctl.c,v $ - * $Revision: 1.2.6.2 $ - * $Date: 1997/11/14 18:07:33 $ - * - * This file contains the non-read/write zftape functions - * for the QIC-40/80/3010/3020 floppy-tape driver for Linux. - */ - -#include -#include -#include -#include - -#include - -#include - -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-eof.h" -#include "../zftape/zftape-ctl.h" -#include "../zftape/zftape-write.h" -#include "../zftape/zftape-read.h" -#include "../zftape/zftape-rw.h" -#include "../zftape/zftape-vtbl.h" - -/* Global vars. - */ -int zft_write_protected; /* this is when cartridge rdonly or O_RDONLY */ -int zft_header_read; -int zft_offline; -unsigned int zft_unit; -int zft_resid; -int zft_mt_compression; - -/* Local vars. - */ -static int going_offline; - -typedef int (mt_fun)(int *argptr); -typedef int (*mt_funp)(int *argptr); -typedef struct -{ - mt_funp function; - unsigned offline : 1; /* op permitted if offline or no_tape */ - unsigned write_protected : 1; /* op permitted if write-protected */ - unsigned not_formatted : 1; /* op permitted if tape not formatted */ - unsigned raw_mode : 1; /* op permitted if zft_mode == 0 */ - unsigned need_idle_state : 1; /* need to call def_idle_state */ - char *name; -} fun_entry; - -static mt_fun mt_dummy, mt_reset, mt_fsr, mt_bsr, mt_rew, mt_offl, mt_nop, - mt_weof, mt_erase, mt_ras2, mt_setblk, mt_setdensity, - mt_seek, mt_tell, mt_reten, mt_eom, mt_fsf, mt_bsf, - mt_fsfm, mt_bsfm, mt_setdrvbuffer, mt_compression; - -static fun_entry mt_funs[]= -{ - {mt_reset , 1, 1, 1, 1, 0, "MT_RESET" }, /* 0 */ - {mt_fsf , 0, 1, 0, 0, 1, "MT_FSF" }, - {mt_bsf , 0, 1, 0, 0, 1, "MT_BSF" }, - {mt_fsr , 0, 1, 0, 1, 1, "MT_FSR" }, - {mt_bsr , 0, 1, 0, 1, 1, "MT_BSR" }, - {mt_weof , 0, 0, 0, 0, 0, "MT_WEOF" }, /* 5 */ - {mt_rew , 0, 1, 1, 1, 0, "MT_REW" }, - {mt_offl , 0, 1, 1, 1, 0, "MT_OFFL" }, - {mt_nop , 1, 1, 1, 1, 0, "MT_NOP" }, - {mt_reten , 0, 1, 1, 1, 0, "MT_RETEN" }, - {mt_bsfm , 0, 1, 0, 0, 1, "MT_BSFM" }, /* 10 */ - {mt_fsfm , 0, 1, 0, 0, 1, "MT_FSFM" }, - {mt_eom , 0, 1, 0, 0, 1, "MT_EOM" }, - {mt_erase , 0, 0, 0, 1, 0, "MT_ERASE" }, - {mt_dummy , 1, 1, 1, 1, 0, "MT_RAS1" }, - {mt_ras2 , 0, 0, 0, 1, 0, "MT_RAS2" }, - {mt_dummy , 1, 1, 1, 1, 0, "MT_RAS3" }, - {mt_dummy , 1, 1, 1, 1, 0, "UNKNOWN" }, - {mt_dummy , 1, 1, 1, 1, 0, "UNKNOWN" }, - {mt_dummy , 1, 1, 1, 1, 0, "UNKNOWN" }, - {mt_setblk , 1, 1, 1, 1, 1, "MT_SETBLK"}, /* 20 */ - {mt_setdensity , 1, 1, 1, 1, 0, "MT_SETDENSITY"}, - {mt_seek , 0, 1, 0, 1, 1, "MT_SEEK" }, - {mt_dummy , 0, 1, 0, 1, 1, "MT_TELL" }, /* wr-only ?! */ - {mt_setdrvbuffer, 1, 1, 1, 1, 0, "MT_SETDRVBUFFER" }, - {mt_dummy , 1, 1, 1, 1, 0, "MT_FSS" }, /* 25 */ - {mt_dummy , 1, 1, 1, 1, 0, "MT_BSS" }, - {mt_dummy , 1, 1, 1, 1, 0, "MT_WSM" }, - {mt_dummy , 1, 1, 1, 1, 0, "MT_LOCK" }, - {mt_dummy , 1, 1, 1, 1, 0, "MT_UNLOCK"}, - {mt_dummy , 1, 1, 1, 1, 0, "MT_LOAD" }, /* 30 */ - {mt_dummy , 1, 1, 1, 1, 0, "MT_UNLOAD"}, - {mt_compression , 1, 1, 1, 0, 1, "MT_COMPRESSION"}, - {mt_dummy , 1, 1, 1, 1, 0, "MT_SETPART"}, - {mt_dummy , 1, 1, 1, 1, 0, "MT_MKPART"} -}; - -#define NR_MT_CMDS NR_ITEMS(mt_funs) - -void zft_reset_position(zft_position *pos) -{ - TRACE_FUN(ft_t_flow); - - pos->seg_byte_pos = - pos->volume_pos = 0; - if (zft_header_read) { - /* need to keep track of the volume table and - * compression map. We therefor simply - * position at the beginning of the first - * volume. This covers old ftape archives as - * well has various flavours of the - * compression map segments. The worst case is - * that the compression map shows up as a - * additional volume in front of all others. - */ - pos->seg_pos = zft_find_volume(0)->start_seg; - pos->tape_pos = zft_calc_tape_pos(pos->seg_pos); - } else { - pos->tape_pos = 0; - pos->seg_pos = -1; - } - zft_just_before_eof = 0; - zft_deblock_segment = -1; - zft_io_state = zft_idle; - zft_zap_read_buffers(); - zft_prevent_flush(); - /* unlock the compresison module if it is loaded. - * The zero arg means not to try to load the module. - */ - if (zft_cmpr_lock(0) == 0) { - (*zft_cmpr_ops->reset)(); /* unlock */ - } - TRACE_EXIT; -} - -static void zft_init_driver(void) -{ - TRACE_FUN(ft_t_flow); - - zft_resid = - zft_header_read = - zft_old_ftape = - zft_offline = - zft_write_protected = - going_offline = - zft_mt_compression = - zft_header_changed = - zft_volume_table_changed = - zft_written_segments = 0; - zft_blk_sz = CONFIG_ZFT_DFLT_BLK_SZ; - zft_reset_position(&zft_pos); /* does most of the stuff */ - ftape_zap_read_buffers(); - ftape_set_state(idle); - TRACE_EXIT; -} - -int zft_def_idle_state(void) -{ - int result = 0; - TRACE_FUN(ft_t_flow); - - if (!zft_header_read) { - result = zft_read_header_segments(); - } else if ((result = zft_flush_buffers()) >= 0 && zft_qic_mode) { - /* don't move past eof - */ - (void)zft_close_volume(&zft_pos); - } - if (ftape_abort_operation() < 0) { - TRACE(ft_t_warn, "ftape_abort_operation() failed"); - result = -EIO; - } - /* clear remaining read buffers */ - zft_zap_read_buffers(); - zft_io_state = zft_idle; - TRACE_EXIT result; -} - -/***************************************************************************** - * * - * functions for the MTIOCTOP commands * - * * - *****************************************************************************/ - -static int mt_dummy(int *dummy) -{ - TRACE_FUN(ft_t_flow); - - TRACE_EXIT -ENOSYS; -} - -static int mt_reset(int *dummy) -{ - TRACE_FUN(ft_t_flow); - - (void)ftape_seek_to_bot(); - TRACE_CATCH(ftape_reset_drive(), - zft_init_driver(); zft_uninit_mem(); zft_offline = 1); - /* fake a re-open of the device. This will set all flage and - * allocate buffers as appropriate. The new tape condition will - * force the open routine to do anything we need. - */ - TRACE_CATCH(_zft_open(-1 /* fake reopen */, 0 /* dummy */),); - TRACE_EXIT 0; -} - -static int mt_fsf(int *arg) -{ - int result; - TRACE_FUN(ft_t_flow); - - result = zft_skip_volumes(*arg, &zft_pos); - zft_just_before_eof = 0; - TRACE_EXIT result; -} - -static int mt_bsf(int *arg) -{ - int result = 0; - TRACE_FUN(ft_t_flow); - - if (*arg != 0) { - result = zft_skip_volumes(-*arg + 1, &zft_pos); - } - TRACE_EXIT result; -} - -static int seek_block(__s64 data_offset, - __s64 block_increment, - zft_position *pos) -{ - int result = 0; - __s64 new_block_pos; - __s64 vol_block_count; - const zft_volinfo *volume; - int exceed; - TRACE_FUN(ft_t_flow); - - volume = zft_find_volume(pos->seg_pos); - if (volume->start_seg == 0 || volume->end_seg == 0) { - TRACE_EXIT -EIO; - } - new_block_pos = (zft_div_blksz(data_offset, volume->blk_sz) - + block_increment); - vol_block_count = zft_div_blksz(volume->size, volume->blk_sz); - if (new_block_pos < 0) { - TRACE(ft_t_noise, - "new_block_pos " LL_X " < 0", LL(new_block_pos)); - zft_resid = (int)new_block_pos; - new_block_pos = 0; - exceed = 1; - } else if (new_block_pos > vol_block_count) { - TRACE(ft_t_noise, - "new_block_pos " LL_X " exceeds size of volume " LL_X, - LL(new_block_pos), LL(vol_block_count)); - zft_resid = (int)(vol_block_count - new_block_pos); - new_block_pos = vol_block_count; - exceed = 1; - } else { - exceed = 0; - } - if (zft_use_compression && volume->use_compression) { - TRACE_CATCH(zft_cmpr_lock(1 /* try to load */),); - result = (*zft_cmpr_ops->seek)(new_block_pos, pos, volume, - zft_deblock_buf); - pos->tape_pos = zft_calc_tape_pos(pos->seg_pos); - pos->tape_pos += pos->seg_byte_pos; - } else { - pos->volume_pos = zft_mul_blksz(new_block_pos, volume->blk_sz); - pos->tape_pos = zft_calc_tape_pos(volume->start_seg); - pos->tape_pos += pos->volume_pos; - pos->seg_pos = zft_calc_seg_byte_coord(&pos->seg_byte_pos, - pos->tape_pos); - } - zft_just_before_eof = volume->size == pos->volume_pos; - if (zft_just_before_eof) { - /* why this? because zft_file_no checks agains start - * and end segment of a volume. We do not want to - * advance to the next volume with this function. - */ - TRACE(ft_t_noise, "set zft_just_before_eof"); - zft_position_before_eof(pos, volume); - } - TRACE(ft_t_noise, "\n" - KERN_INFO "new_seg_pos : %d\n" - KERN_INFO "new_tape_pos: " LL_X "\n" - KERN_INFO "vol_size : " LL_X "\n" - KERN_INFO "seg_byte_pos: %d\n" - KERN_INFO "blk_sz : %d", - pos->seg_pos, LL(pos->tape_pos), - LL(volume->size), pos->seg_byte_pos, - volume->blk_sz); - if (!exceed) { - zft_resid = new_block_pos - zft_div_blksz(pos->volume_pos, - volume->blk_sz); - } - if (zft_resid < 0) { - zft_resid = -zft_resid; - } - TRACE_EXIT ((exceed || zft_resid != 0) && result >= 0) ? -EINVAL : result; -} - -static int mt_fsr(int *arg) -{ - int result; - TRACE_FUN(ft_t_flow); - - result = seek_block(zft_pos.volume_pos, *arg, &zft_pos); - TRACE_EXIT result; -} - -static int mt_bsr(int *arg) -{ - int result; - TRACE_FUN(ft_t_flow); - - result = seek_block(zft_pos.volume_pos, -*arg, &zft_pos); - TRACE_EXIT result; -} - -static int mt_weof(int *arg) -{ - int result; - TRACE_FUN(ft_t_flow); - - TRACE_CATCH(zft_flush_buffers(),); - result = zft_weof(*arg, &zft_pos); - TRACE_EXIT result; -} - -static int mt_rew(int *dummy) -{ - int result; - TRACE_FUN(ft_t_flow); - - if(zft_header_read) { - (void)zft_def_idle_state(); - } - result = ftape_seek_to_bot(); - zft_reset_position(&zft_pos); - TRACE_EXIT result; -} - -static int mt_offl(int *dummy) -{ - int result; - TRACE_FUN(ft_t_flow); - - going_offline= 1; - result = mt_rew(NULL); - TRACE_EXIT result; -} - -static int mt_nop(int *dummy) -{ - TRACE_FUN(ft_t_flow); - /* should we set tape status? - */ - if (!zft_offline) { /* offline includes no_tape */ - (void)zft_def_idle_state(); - } - TRACE_EXIT 0; -} - -static int mt_reten(int *dummy) -{ - int result; - TRACE_FUN(ft_t_flow); - - if(zft_header_read) { - (void)zft_def_idle_state(); - } - result = ftape_seek_to_eot(); - if (result >= 0) { - result = ftape_seek_to_bot(); - } - TRACE_EXIT(result); -} - -static int fsfbsfm(int arg, zft_position *pos) -{ - const zft_volinfo *vtbl; - __s64 block_pos; - TRACE_FUN(ft_t_flow); - - /* What to do? This should seek to the next file-mark and - * position BEFORE. That is, a next write would just extend - * the current file. Well. Let's just seek to the end of the - * current file, if count == 1. If count > 1, then do a - * "mt_fsf(count - 1)", and then seek to the end of that file. - * If count == 0, do nothing - */ - if (arg == 0) { - TRACE_EXIT 0; - } - zft_just_before_eof = 0; - TRACE_CATCH(zft_skip_volumes(arg < 0 ? arg : arg-1, pos), - if (arg > 0) { - zft_resid ++; - }); - vtbl = zft_find_volume(pos->seg_pos); - block_pos = zft_div_blksz(vtbl->size, vtbl->blk_sz); - (void)seek_block(0, block_pos, pos); - if (pos->volume_pos != vtbl->size) { - zft_just_before_eof = 0; - zft_resid = 1; - /* we didn't managed to go there */ - TRACE_ABORT(-EIO, ft_t_err, - "wanted file position " LL_X ", arrived at " LL_X, - LL(vtbl->size), LL(pos->volume_pos)); - } - zft_just_before_eof = 1; - TRACE_EXIT 0; -} - -static int mt_bsfm(int *arg) -{ - int result; - TRACE_FUN(ft_t_flow); - - result = fsfbsfm(-*arg, &zft_pos); - TRACE_EXIT result; -} - -static int mt_fsfm(int *arg) -{ - int result; - TRACE_FUN(ft_t_flow); - - result = fsfbsfm(*arg, &zft_pos); - TRACE_EXIT result; -} - -static int mt_eom(int *dummy) -{ - TRACE_FUN(ft_t_flow); - - zft_skip_to_eom(&zft_pos); - TRACE_EXIT 0; -} - -static int mt_erase(int *dummy) -{ - int result; - TRACE_FUN(ft_t_flow); - - result = zft_erase(); - TRACE_EXIT result; -} - -static int mt_ras2(int *dummy) -{ - int result; - TRACE_FUN(ft_t_flow); - - result = -ENOSYS; - TRACE_EXIT result; -} - -/* Sets the new blocksize in BYTES - * - */ -static int mt_setblk(int *new_size) -{ - TRACE_FUN(ft_t_flow); - - if((unsigned int)(*new_size) > ZFT_MAX_BLK_SZ) { - TRACE_ABORT(-EINVAL, ft_t_info, - "desired blk_sz (%d) should be <= %d bytes", - *new_size, ZFT_MAX_BLK_SZ); - } - if ((*new_size & (FT_SECTOR_SIZE-1)) != 0) { - TRACE_ABORT(-EINVAL, ft_t_info, - "desired blk_sz (%d) must be a multiple of %d bytes", - *new_size, FT_SECTOR_SIZE); - } - if (*new_size == 0) { - if (zft_use_compression) { - TRACE_ABORT(-EINVAL, ft_t_info, - "Variable block size not yet " - "supported with compression"); - } - *new_size = 1; - } - zft_blk_sz = *new_size; - TRACE_EXIT 0; -} - -static int mt_setdensity(int *arg) -{ - TRACE_FUN(ft_t_flow); - - SET_TRACE_LEVEL(*arg); - TRACE(TRACE_LEVEL, "tracing set to %d", TRACE_LEVEL); - if ((int)TRACE_LEVEL != *arg) { - TRACE_EXIT -EINVAL; - } - TRACE_EXIT 0; -} - -static int mt_seek(int *new_block_pos) -{ - int result= 0; - TRACE_FUN(ft_t_any); - - result = seek_block(0, (__s64)*new_block_pos, &zft_pos); - TRACE_EXIT result; -} - -/* OK, this is totally different from SCSI, but the worst thing that can - * happen is that there is not enough defragmentated memory that can be - * allocated. Also, there is a hardwired limit of 16 dma buffers in the - * stock ftape module. This shouldn't bring the system down. - * - * NOTE: the argument specifies the total number of dma buffers to use. - * The driver needs at least 3 buffers to function at all. - * - */ -static int mt_setdrvbuffer(int *cnt) -{ - TRACE_FUN(ft_t_flow); - - if (*cnt < 3) { - TRACE_EXIT -EINVAL; - } - TRACE_CATCH(ftape_set_nr_buffers(*cnt),); - TRACE_EXIT 0; -} -/* return the block position from start of volume - */ -static int mt_tell(int *arg) -{ - TRACE_FUN(ft_t_flow); - - *arg = zft_div_blksz(zft_pos.volume_pos, - zft_find_volume(zft_pos.seg_pos)->blk_sz); - TRACE_EXIT 0; -} - -static int mt_compression(int *arg) -{ - TRACE_FUN(ft_t_flow); - - /* Ok. We could also check whether compression is available at - * all by trying to load the compression module. We could - * also check for a block size of 1 byte which is illegal - * with compression. Instead of doing it here we rely on - * zftape_write() to do the proper checks. - */ - if ((unsigned int)*arg > 1) { - TRACE_EXIT -EINVAL; - } - if (*arg != 0 && zft_blk_sz == 1) { /* variable block size */ - TRACE_ABORT(-EINVAL, ft_t_info, - "Compression not yet supported " - "with variable block size"); - } - zft_mt_compression = *arg; - if ((zft_unit & ZFT_ZIP_MODE) == 0) { - zft_use_compression = zft_mt_compression; - } - TRACE_EXIT 0; -} - -/* check whether write access is allowed. Write access is denied when - * + zft_write_protected == 1 -- this accounts for either hard write - * protection of the cartridge or for - * O_RDONLY access mode of the tape device - * + zft_offline == 1 -- this meany that there is either no tape - * or that the MTOFFLINE ioctl has been - * previously issued (`soft eject') - * + ft_formatted == 0 -- this means that the cartridge is not - * formatted - * Then we distinuguish two cases. When zft_qic_mode is TRUE, then we try - * to emulate a `traditional' (aka SCSI like) UN*X tape device. Therefore we - * deny writes when - * + zft_qic_mode ==1 && - * (!zft_tape_at_lbot() && -- tape no at logical BOT - * !(zft_tape_at_eom() || -- tape not at logical EOM (or EOD) - * (zft_tape_at_eom() && - * zft_old_ftape()))) -- we can't add new volume to tapes - * written by old ftape because ftape - * don't use the volume table - * - * when the drive is in true raw mode (aka /dev/rawft0) then we don't - * care about LBOT and EOM conditions. This device is intended for a - * user level program that wants to truly implement the QIC-80 compliance - * at the logical data layout level of the cartridge, i.e. implement all - * that volume table and volume directory stuff etc.< - */ -int zft_check_write_access(zft_position *pos) -{ - TRACE_FUN(ft_t_flow); - - if (zft_offline) { /* offline includes no_tape */ - TRACE_ABORT(-ENXIO, - ft_t_info, "tape is offline or no cartridge"); - } - if (!ft_formatted) { - TRACE_ABORT(-EACCES, ft_t_info, "tape is not formatted"); - } - if (zft_write_protected) { - TRACE_ABORT(-EACCES, ft_t_info, "cartridge write protected"); - } - if (zft_qic_mode) { - /* check BOT condition */ - if (!zft_tape_at_lbot(pos)) { - /* protect cartridges written by old ftape if - * not at BOT because they use the vtbl - * segment for storing data - */ - if (zft_old_ftape) { - TRACE_ABORT(-EACCES, ft_t_warn, - "Cannot write to cartridges written by old ftape when not at BOT"); - } - /* not at BOT, but allow writes at EOD, of course - */ - if (!zft_tape_at_eod(pos)) { - TRACE_ABORT(-EACCES, ft_t_info, - "tape not at BOT and not at EOD"); - } - } - /* fine. Now the tape is either at BOT or at EOD. */ - } - /* or in raw mode in which case we don't care about BOT and EOD */ - TRACE_EXIT 0; -} - -/* OPEN routine called by kernel-interface code - * - * NOTE: this is also called by mt_reset() with dev_minor == -1 - * to fake a reopen after a reset. - */ -int _zft_open(unsigned int dev_minor, unsigned int access_mode) -{ - static unsigned int tape_unit; - static unsigned int file_access_mode; - int result; - TRACE_FUN(ft_t_flow); - - if ((int)dev_minor == -1) { - /* fake reopen */ - zft_unit = tape_unit; - access_mode = file_access_mode; - zft_init_driver(); /* reset all static data to defaults */ - } else { - tape_unit = dev_minor; - file_access_mode = access_mode; - if ((result = ftape_enable(FTAPE_SEL(dev_minor))) < 0) { - TRACE_ABORT(-ENXIO, ft_t_err, - "ftape_enable failed: %d", result); - } - if (ft_new_tape || ft_no_tape || !ft_formatted || - (FTAPE_SEL(zft_unit) != FTAPE_SEL(dev_minor)) || - (zft_unit & ZFT_RAW_MODE) != (dev_minor & ZFT_RAW_MODE)) { - /* reset all static data to defaults, - */ - zft_init_driver(); - } - zft_unit = dev_minor; - } - zft_set_flags(zft_unit); /* decode the minor bits */ - if (zft_blk_sz == 1 && zft_use_compression) { - ftape_disable(); /* resets ft_no_tape */ - TRACE_ABORT(-ENODEV, ft_t_warn, "Variable block size not yet " - "supported with compression"); - } - /* no need for most of the buffers when no tape or not - * formatted. for the read/write operations, it is the - * regardless whether there is no tape, a not-formatted tape - * or the whether the driver is soft offline. - * Nevertheless we allow some ioctls with non-formatted tapes, - * like rewind and reset. - */ - if (ft_no_tape || !ft_formatted) { - zft_uninit_mem(); - } - if (ft_no_tape) { - zft_offline = 1; /* so we need not test two variables */ - } - if ((access_mode == O_WRONLY || access_mode == O_RDWR) && - (ft_write_protected || ft_no_tape)) { - ftape_disable(); /* resets ft_no_tape */ - TRACE_ABORT(ft_no_tape ? -ENXIO : -EROFS, - ft_t_warn, "wrong access mode %s cartridge", - ft_no_tape ? "without a" : "with write protected"); - } - zft_write_protected = (access_mode == O_RDONLY || - ft_write_protected != 0); - if (zft_write_protected) { - TRACE(ft_t_noise, - "read only access mode: %d, " - "drive write protected: %d", - access_mode == O_RDONLY, - ft_write_protected != 0); - } - if (!zft_offline) { - TRACE_CATCH(zft_vmalloc_once(&zft_deblock_buf,FT_SEGMENT_SIZE), - ftape_disable()); - } - /* zft_seg_pos should be greater than the vtbl segpos but not - * if in compatibility mode and only after we read in the - * header segments - * - * might also be a problem if the user makes a backup with a - * *qft* device and rewinds it with a raw device. - */ - if (zft_qic_mode && - !zft_old_ftape && - zft_pos.seg_pos >= 0 && - zft_header_read && - zft_pos.seg_pos <= ft_first_data_segment) { - TRACE(ft_t_noise, "you probably mixed up the zftape devices!"); - zft_reset_position(&zft_pos); - } - TRACE_EXIT 0; -} - -/* RELEASE routine called by kernel-interface code - */ -int _zft_close(void) -{ - int result = 0; - TRACE_FUN(ft_t_flow); - - if (zft_offline) { - /* call the hardware release routine. Puts the drive offline */ - ftape_disable(); - TRACE_EXIT 0; - } - if (!(ft_write_protected || zft_old_ftape)) { - result = zft_flush_buffers(); - TRACE(ft_t_noise, "writing file mark at current position"); - if (zft_qic_mode && zft_close_volume(&zft_pos) == 0) { - zft_move_past_eof(&zft_pos); - } - if ((zft_tape_at_lbot(&zft_pos) || - !(zft_unit & FTAPE_NO_REWIND))) { - if (result >= 0) { - result = zft_update_header_segments(); - } else { - TRACE(ft_t_err, - "Error: unable to update header segments"); - } - } - } - ftape_abort_operation(); - if (!(zft_unit & FTAPE_NO_REWIND)) { - TRACE(ft_t_noise, "rewinding tape"); - if (ftape_seek_to_bot() < 0 && result >= 0) { - result = -EIO; /* keep old value */ - } - zft_reset_position(&zft_pos); - } - zft_zap_read_buffers(); - /* now free up memory as much as possible. We don't destroy - * the deblock buffer if it containes a valid segment. - */ - if (zft_deblock_segment == -1) { - zft_vfree(&zft_deblock_buf, FT_SEGMENT_SIZE); - } - /* high level driver status, forces creation of a new volume - * when calling ftape_write again and not zft_just_before_eof - */ - zft_io_state = zft_idle; - if (going_offline) { - zft_init_driver(); - zft_uninit_mem(); - going_offline = 0; - zft_offline = 1; - } else if (zft_cmpr_lock(0 /* don't load */) == 0) { - (*zft_cmpr_ops->reset)(); /* unlock it again */ - } - zft_memory_stats(); - /* call the hardware release routine. Puts the drive offline */ - ftape_disable(); - TRACE_EXIT result; -} - -/* - * the wrapper function around the wrapper MTIOCTOP ioctl - */ -static int mtioctop(struct mtop *mtop, int arg_size) -{ - int result = 0; - fun_entry *mt_fun_entry; - TRACE_FUN(ft_t_flow); - - if (arg_size != sizeof(struct mtop) || mtop->mt_op >= NR_MT_CMDS) { - TRACE_EXIT -EINVAL; - } - TRACE(ft_t_noise, "calling MTIOCTOP command: %s", - mt_funs[mtop->mt_op].name); - mt_fun_entry= &mt_funs[mtop->mt_op]; - zft_resid = mtop->mt_count; - if (!mt_fun_entry->offline && zft_offline) { - if (ft_no_tape) { - TRACE_ABORT(-ENXIO, ft_t_info, "no tape present"); - } else { - TRACE_ABORT(-ENXIO, ft_t_info, "drive is offline"); - } - } - if (!mt_fun_entry->not_formatted && !ft_formatted) { - TRACE_ABORT(-EACCES, ft_t_info, "tape is not formatted"); - } - if (!mt_fun_entry->write_protected) { - TRACE_CATCH(zft_check_write_access(&zft_pos),); - } - if (mt_fun_entry->need_idle_state && !(zft_offline || !ft_formatted)) { - TRACE_CATCH(zft_def_idle_state(),); - } - if (!zft_qic_mode && !mt_fun_entry->raw_mode) { - TRACE_ABORT(-EACCES, ft_t_info, -"Drive needs to be in QIC-80 compatibility mode for this command"); - } - result = (mt_fun_entry->function)(&mtop->mt_count); - if (zft_tape_at_lbot(&zft_pos)) { - TRACE_CATCH(zft_update_header_segments(),); - } - if (result >= 0) { - zft_resid = 0; - } - TRACE_EXIT result; -} - -/* - * standard MTIOCGET ioctl - */ -static int mtiocget(struct mtget *mtget, int arg_size) -{ - const zft_volinfo *volume; - __s64 max_tape_pos; - TRACE_FUN(ft_t_flow); - - if (arg_size != sizeof(struct mtget)) { - TRACE_ABORT(-EINVAL, ft_t_info, "bad argument size: %d", - arg_size); - } - mtget->mt_type = ft_drive_type.vendor_id + 0x800000; - mtget->mt_dsreg = ft_last_status.space; - mtget->mt_erreg = ft_last_error.space; /* error register */ - mtget->mt_resid = zft_resid; /* residuum of writes, reads and - * MTIOCTOP commands - */ - if (!zft_offline) { /* neither no_tape nor soft offline */ - mtget->mt_gstat = GMT_ONLINE(~0UL); - /* should rather return the status of the cartridge - * than the access mode of the file, therefor use - * ft_write_protected, not zft_write_protected - */ - if (ft_write_protected) { - mtget->mt_gstat |= GMT_WR_PROT(~0UL); - } - if(zft_header_read) { /* this catches non-formatted */ - volume = zft_find_volume(zft_pos.seg_pos); - mtget->mt_fileno = volume->count; - max_tape_pos = zft_capacity - zft_blk_sz; - if (zft_use_compression) { - max_tape_pos -= ZFT_CMPR_OVERHEAD; - } - if (zft_tape_at_eod(&zft_pos)) { - mtget->mt_gstat |= GMT_EOD(~0UL); - } - if (zft_pos.tape_pos > max_tape_pos) { - mtget->mt_gstat |= GMT_EOT(~0UL); - } - mtget->mt_blkno = zft_div_blksz(zft_pos.volume_pos, - volume->blk_sz); - if (zft_just_before_eof) { - mtget->mt_gstat |= GMT_EOF(~0UL); - } - if (zft_tape_at_lbot(&zft_pos)) { - mtget->mt_gstat |= GMT_BOT(~0UL); - } - } else { - mtget->mt_fileno = mtget->mt_blkno = -1; - if (mtget->mt_dsreg & QIC_STATUS_AT_BOT) { - mtget->mt_gstat |= GMT_BOT(~0UL); - } - } - } else { - if (ft_no_tape) { - mtget->mt_gstat = GMT_DR_OPEN(~0UL); - } else { - mtget->mt_gstat = 0UL; - } - mtget->mt_fileno = mtget->mt_blkno = -1; - } - TRACE_EXIT 0; -} - -#ifdef MTIOCRDFTSEG -/* - * Read a floppy tape segment. This is useful for manipulating the - * volume table, and read the old header segment before re-formatting - * the cartridge. - */ -static int mtiocrdftseg(struct mtftseg * mtftseg, int arg_size) -{ - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "Mag tape ioctl command: MTIOCRDFTSEG"); - if (zft_qic_mode) { - TRACE_ABORT(-EACCES, ft_t_info, - "driver needs to be in raw mode for this ioctl"); - } - if (arg_size != sizeof(struct mtftseg)) { - TRACE_ABORT(-EINVAL, ft_t_info, "bad argument size: %d", - arg_size); - } - if (zft_offline) { - TRACE_EXIT -ENXIO; - } - if (mtftseg->mt_mode != FT_RD_SINGLE && - mtftseg->mt_mode != FT_RD_AHEAD) { - TRACE_ABORT(-EINVAL, ft_t_info, "invalid read mode"); - } - if (!ft_formatted) { - TRACE_EXIT -EACCES; /* -ENXIO ? */ - - } - if (!zft_header_read) { - TRACE_CATCH(zft_def_idle_state(),); - } - if (mtftseg->mt_segno > ft_last_data_segment) { - TRACE_ABORT(-EINVAL, ft_t_info, "segment number is too large"); - } - mtftseg->mt_result = ftape_read_segment(mtftseg->mt_segno, - zft_deblock_buf, - mtftseg->mt_mode); - if (mtftseg->mt_result < 0) { - /* a negativ result is not an ioctl error. if - * the user wants to read damaged tapes, - * it's up to her/him - */ - TRACE_EXIT 0; - } - if (copy_to_user(mtftseg->mt_data, - zft_deblock_buf, - mtftseg->mt_result) != 0) { - TRACE_EXIT -EFAULT; - } - TRACE_EXIT 0; -} -#endif - -#ifdef MTIOCWRFTSEG -/* - * write a floppy tape segment. This version features writing of - * deleted address marks, and gracefully ignores the (software) - * ft_formatted flag to support writing of header segments after - * formatting. - */ -static int mtiocwrftseg(struct mtftseg * mtftseg, int arg_size) -{ - int result; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "Mag tape ioctl command: MTIOCWRFTSEG"); - if (zft_write_protected || zft_qic_mode) { - TRACE_EXIT -EACCES; - } - if (arg_size != sizeof(struct mtftseg)) { - TRACE_ABORT(-EINVAL, ft_t_info, "bad argument size: %d", - arg_size); - } - if (zft_offline) { - TRACE_EXIT -ENXIO; - } - if (mtftseg->mt_mode != FT_WR_ASYNC && - mtftseg->mt_mode != FT_WR_MULTI && - mtftseg->mt_mode != FT_WR_SINGLE && - mtftseg->mt_mode != FT_WR_DELETE) { - TRACE_ABORT(-EINVAL, ft_t_info, "invalid write mode"); - } - /* - * We don't check for ft_formatted, because this gives - * only the software status of the driver. - * - * We assume that the user knows what it is - * doing. And rely on the low level stuff to fail - * when the tape isn't formatted. We only make sure - * that The header segment buffer is allocated, - * because it holds the bad sector map. - */ - if (zft_hseg_buf == NULL) { - TRACE_EXIT -ENXIO; - } - if (mtftseg->mt_mode != FT_WR_DELETE) { - if (copy_from_user(zft_deblock_buf, - mtftseg->mt_data, - FT_SEGMENT_SIZE) != 0) { - TRACE_EXIT -EFAULT; - } - } - mtftseg->mt_result = ftape_write_segment(mtftseg->mt_segno, - zft_deblock_buf, - mtftseg->mt_mode); - if (mtftseg->mt_result >= 0 && mtftseg->mt_mode == FT_WR_SINGLE) { - /* - * a negativ result is not an ioctl error. if - * the user wants to write damaged tapes, - * it's up to her/him - */ - if ((result = ftape_loop_until_writes_done()) < 0) { - mtftseg->mt_result = result; - } - } - TRACE_EXIT 0; -} -#endif - -#ifdef MTIOCVOLINFO -/* - * get information about volume positioned at. - */ -static int mtiocvolinfo(struct mtvolinfo *volinfo, int arg_size) -{ - const zft_volinfo *volume; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "Mag tape ioctl command: MTIOCVOLINFO"); - if (arg_size != sizeof(struct mtvolinfo)) { - TRACE_ABORT(-EINVAL, - ft_t_info, "bad argument size: %d", arg_size); - } - if (zft_offline) { - TRACE_EXIT -ENXIO; - } - if (!ft_formatted) { - TRACE_EXIT -EACCES; - } - TRACE_CATCH(zft_def_idle_state(),); - volume = zft_find_volume(zft_pos.seg_pos); - volinfo->mt_volno = volume->count; - volinfo->mt_blksz = volume->blk_sz == 1 ? 0 : volume->blk_sz; - volinfo->mt_size = volume->size >> 10; - volinfo->mt_rawsize = ((zft_calc_tape_pos(volume->end_seg + 1) >> 10) - - (zft_calc_tape_pos(volume->start_seg) >> 10)); - volinfo->mt_cmpr = volume->use_compression; - TRACE_EXIT 0; -} -#endif - -#ifdef ZFT_OBSOLETE -static int mtioc_zftape_getblksz(struct mtblksz *blksz, int arg_size) -{ - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "\n" - KERN_INFO "Mag tape ioctl command: MTIOC_ZTAPE_GETBLKSZ\n" - KERN_INFO "This ioctl is here merely for compatibility.\n" - KERN_INFO "Please use MTIOCVOLINFO instead"); - if (arg_size != sizeof(struct mtblksz)) { - TRACE_ABORT(-EINVAL, - ft_t_info, "bad argument size: %d", arg_size); - } - if (zft_offline) { - TRACE_EXIT -ENXIO; - } - if (!ft_formatted) { - TRACE_EXIT -EACCES; - } - TRACE_CATCH(zft_def_idle_state(),); - blksz->mt_blksz = zft_find_volume(zft_pos.seg_pos)->blk_sz; - TRACE_EXIT 0; -} -#endif - -#ifdef MTIOCGETSIZE -/* - * get the capacity of the tape cartridge. - */ -static int mtiocgetsize(struct mttapesize *size, int arg_size) -{ - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "Mag tape ioctl command: MTIOC_ZFTAPE_GETSIZE"); - if (arg_size != sizeof(struct mttapesize)) { - TRACE_ABORT(-EINVAL, - ft_t_info, "bad argument size: %d", arg_size); - } - if (zft_offline) { - TRACE_EXIT -ENXIO; - } - if (!ft_formatted) { - TRACE_EXIT -EACCES; - } - TRACE_CATCH(zft_def_idle_state(),); - size->mt_capacity = (unsigned int)(zft_capacity>>10); - size->mt_used = (unsigned int)(zft_get_eom_pos()>>10); - TRACE_EXIT 0; -} -#endif - -static int mtiocpos(struct mtpos *mtpos, int arg_size) -{ - int result; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "Mag tape ioctl command: MTIOCPOS"); - if (arg_size != sizeof(struct mtpos)) { - TRACE_ABORT(-EINVAL, - ft_t_info, "bad argument size: %d", arg_size); - } - result = mt_tell((int *)&mtpos->mt_blkno); - TRACE_EXIT result; -} - -#ifdef MTIOCFTFORMAT -/* - * formatting of floppy tape cartridges. This is intended to be used - * together with the MTIOCFTCMD ioctl and the new mmap feature - */ - -/* - * This function uses ftape_decode_header_segment() to inform the low - * level ftape module about the new parameters. - * - * It erases the hseg_buf. The calling process must specify all - * parameters to assure proper operation. - * - * return values: -EINVAL - wrong argument size - * -EINVAL - if ftape_decode_header_segment() failed. - */ -static int set_format_parms(struct ftfmtparms *p, __u8 *hseg_buf) -{ - ft_trace_t old_level = TRACE_LEVEL; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "MTIOCFTFORMAT operation FTFMT_SETPARMS"); - memset(hseg_buf, 0, FT_SEGMENT_SIZE); - PUT4(hseg_buf, FT_SIGNATURE, FT_HSEG_MAGIC); - - /* fill in user specified parameters - */ - hseg_buf[FT_FMT_CODE] = (__u8)p->ft_fmtcode; - PUT2(hseg_buf, FT_SPT, p->ft_spt); - hseg_buf[FT_TPC] = (__u8)p->ft_tpc; - hseg_buf[FT_FHM] = (__u8)p->ft_fhm; - hseg_buf[FT_FTM] = (__u8)p->ft_ftm; - - /* fill in sane defaults to make ftape happy. - */ - hseg_buf[FT_FSM] = (__u8)128; /* 128 is hard wired all over ftape */ - if (p->ft_fmtcode == fmt_big) { - PUT4(hseg_buf, FT_6_HSEG_1, 0); - PUT4(hseg_buf, FT_6_HSEG_2, 1); - PUT4(hseg_buf, FT_6_FRST_SEG, 2); - PUT4(hseg_buf, FT_6_LAST_SEG, p->ft_spt * p->ft_tpc - 1); - } else { - PUT2(hseg_buf, FT_HSEG_1, 0); - PUT2(hseg_buf, FT_HSEG_2, 1); - PUT2(hseg_buf, FT_FRST_SEG, 2); - PUT2(hseg_buf, FT_LAST_SEG, p->ft_spt * p->ft_tpc - 1); - } - - /* Synchronize with the low level module. This is particularly - * needed for unformatted cartridges as the QIC std was previously - * unknown BUT is needed to set data rate and to calculate timeouts. - */ - TRACE_CATCH(ftape_calibrate_data_rate(p->ft_qicstd&QIC_TAPE_STD_MASK), - _res = -EINVAL); - - /* The following will also recalcualte the timeouts for the tape - * length and QIC std we want to format to. - * abort with -EINVAL rather than -EIO - */ - SET_TRACE_LEVEL(ft_t_warn); - TRACE_CATCH(ftape_decode_header_segment(hseg_buf), - SET_TRACE_LEVEL(old_level); _res = -EINVAL); - SET_TRACE_LEVEL(old_level); - TRACE_EXIT 0; -} - -/* - * Return the internal SOFTWARE status of the kernel driver. This does - * NOT query the tape drive about its status. - */ -static int get_format_parms(struct ftfmtparms *p, __u8 *hseg_buffer) -{ - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "MTIOCFTFORMAT operation FTFMT_GETPARMS"); - p->ft_qicstd = ft_qic_std; - p->ft_fmtcode = ft_format_code; - p->ft_fhm = hseg_buffer[FT_FHM]; - p->ft_ftm = hseg_buffer[FT_FTM]; - p->ft_spt = ft_segments_per_track; - p->ft_tpc = ft_tracks_per_tape; - TRACE_EXIT 0; -} - -static int mtiocftformat(struct mtftformat *mtftformat, int arg_size) -{ - int result; - union fmt_arg *arg = &mtftformat->fmt_arg; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "Mag tape ioctl command: MTIOCFTFORMAT"); - if (zft_offline) { - if (ft_no_tape) { - TRACE_ABORT(-ENXIO, ft_t_info, "no tape present"); - } else { - TRACE_ABORT(-ENXIO, ft_t_info, "drive is offline"); - } - } - if (zft_qic_mode) { - TRACE_ABORT(-EACCES, ft_t_info, - "driver needs to be in raw mode for this ioctl"); - } - if (zft_hseg_buf == NULL) { - TRACE_CATCH(zft_vcalloc_once(&zft_hseg_buf, FT_SEGMENT_SIZE),); - } - zft_header_read = 0; - switch(mtftformat->fmt_op) { - case FTFMT_SET_PARMS: - TRACE_CATCH(set_format_parms(&arg->fmt_parms, zft_hseg_buf),); - TRACE_EXIT 0; - case FTFMT_GET_PARMS: - TRACE_CATCH(get_format_parms(&arg->fmt_parms, zft_hseg_buf),); - TRACE_EXIT 0; - case FTFMT_FORMAT_TRACK: - if ((ft_formatted && zft_check_write_access(&zft_pos) < 0) || - (!ft_formatted && zft_write_protected)) { - TRACE_ABORT(-EACCES, ft_t_info, "Write access denied"); - } - TRACE_CATCH(ftape_format_track(arg->fmt_track.ft_track, - arg->fmt_track.ft_gap3),); - TRACE_EXIT 0; - case FTFMT_STATUS: - TRACE_CATCH(ftape_format_status(&arg->fmt_status.ft_segment),); - TRACE_EXIT 0; - case FTFMT_VERIFY: - TRACE_CATCH(ftape_verify_segment(arg->fmt_verify.ft_segment, - (SectorMap *)&arg->fmt_verify.ft_bsm),); - TRACE_EXIT 0; - default: - TRACE_ABORT(-EINVAL, ft_t_err, "Invalid format operation"); - } - TRACE_EXIT result; -} -#endif - -#ifdef MTIOCFTCMD -/* - * send a QIC-117 command to the drive, with optional timeouts, - * parameter and result bits. This is intended to be used together - * with the formatting ioctl. - */ -static int mtiocftcmd(struct mtftcmd *ftcmd, int arg_size) -{ - int i; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "Mag tape ioctl command: MTIOCFTCMD"); - if (!capable(CAP_SYS_ADMIN)) { - TRACE_ABORT(-EPERM, ft_t_info, - "need CAP_SYS_ADMIN capability to send raw qic-117 commands"); - } - if (zft_qic_mode) { - TRACE_ABORT(-EACCES, ft_t_info, - "driver needs to be in raw mode for this ioctl"); - } - if (arg_size != sizeof(struct mtftcmd)) { - TRACE_ABORT(-EINVAL, - ft_t_info, "bad argument size: %d", arg_size); - } - if (ftcmd->ft_wait_before) { - TRACE_CATCH(ftape_ready_wait(ftcmd->ft_wait_before, - &ftcmd->ft_status),); - } - if (ftcmd->ft_status & QIC_STATUS_ERROR) - goto ftmtcmd_error; - if (ftcmd->ft_result_bits != 0) { - TRACE_CATCH(ftape_report_operation(&ftcmd->ft_result, - ftcmd->ft_cmd, - ftcmd->ft_result_bits),); - } else { - TRACE_CATCH(ftape_command(ftcmd->ft_cmd),); - if (ftcmd->ft_status & QIC_STATUS_ERROR) - goto ftmtcmd_error; - for (i = 0; i < ftcmd->ft_parm_cnt; i++) { - TRACE_CATCH(ftape_parameter(ftcmd->ft_parms[i]&0x0f),); - if (ftcmd->ft_status & QIC_STATUS_ERROR) - goto ftmtcmd_error; - } - } - if (ftcmd->ft_wait_after != 0) { - TRACE_CATCH(ftape_ready_wait(ftcmd->ft_wait_after, - &ftcmd->ft_status),); - } -ftmtcmd_error: - if (ftcmd->ft_status & QIC_STATUS_ERROR) { - TRACE(ft_t_noise, "error status set"); - TRACE_CATCH(ftape_report_error(&ftcmd->ft_error, - &ftcmd->ft_cmd, 1),); - } - TRACE_EXIT 0; /* this is not an i/o error */ -} -#endif - -/* IOCTL routine called by kernel-interface code - */ -int _zft_ioctl(unsigned int command, void __user * arg) -{ - int result; - union { struct mtop mtop; - struct mtget mtget; - struct mtpos mtpos; -#ifdef MTIOCRDFTSEG - struct mtftseg mtftseg; -#endif -#ifdef MTIOCVOLINFO - struct mtvolinfo mtvolinfo; -#endif -#ifdef MTIOCGETSIZE - struct mttapesize mttapesize; -#endif -#ifdef MTIOCFTFORMAT - struct mtftformat mtftformat; -#endif -#ifdef ZFT_OBSOLETE - struct mtblksz mtblksz; -#endif -#ifdef MTIOCFTCMD - struct mtftcmd mtftcmd; -#endif - } krnl_arg; - int arg_size = _IOC_SIZE(command); - int dir = _IOC_DIR(command); - TRACE_FUN(ft_t_flow); - - /* This check will only catch arguments that are too large ! - */ - if (dir & (_IOC_READ | _IOC_WRITE) && arg_size > sizeof(krnl_arg)) { - TRACE_ABORT(-EINVAL, - ft_t_info, "bad argument size: %d", arg_size); - } - if (dir & _IOC_WRITE) { - if (copy_from_user(&krnl_arg, arg, arg_size) != 0) { - TRACE_EXIT -EFAULT; - } - } - TRACE(ft_t_flow, "called with ioctl command: 0x%08x", command); - switch (command) { - case MTIOCTOP: - result = mtioctop(&krnl_arg.mtop, arg_size); - break; - case MTIOCGET: - result = mtiocget(&krnl_arg.mtget, arg_size); - break; - case MTIOCPOS: - result = mtiocpos(&krnl_arg.mtpos, arg_size); - break; -#ifdef MTIOCVOLINFO - case MTIOCVOLINFO: - result = mtiocvolinfo(&krnl_arg.mtvolinfo, arg_size); - break; -#endif -#ifdef ZFT_OBSOLETE - case MTIOC_ZFTAPE_GETBLKSZ: - result = mtioc_zftape_getblksz(&krnl_arg.mtblksz, arg_size); - break; -#endif -#ifdef MTIOCRDFTSEG - case MTIOCRDFTSEG: /* read a segment via ioctl */ - result = mtiocrdftseg(&krnl_arg.mtftseg, arg_size); - break; -#endif -#ifdef MTIOCWRFTSEG - case MTIOCWRFTSEG: /* write a segment via ioctl */ - result = mtiocwrftseg(&krnl_arg.mtftseg, arg_size); - break; -#endif -#ifdef MTIOCGETSIZE - case MTIOCGETSIZE: - result = mtiocgetsize(&krnl_arg.mttapesize, arg_size); - break; -#endif -#ifdef MTIOCFTFORMAT - case MTIOCFTFORMAT: - result = mtiocftformat(&krnl_arg.mtftformat, arg_size); - break; -#endif -#ifdef MTIOCFTCMD - case MTIOCFTCMD: - result = mtiocftcmd(&krnl_arg.mtftcmd, arg_size); - break; -#endif - default: - result = -EINVAL; - break; - } - if ((result >= 0) && (dir & _IOC_READ)) { - if (copy_to_user(arg, &krnl_arg, arg_size) != 0) { - TRACE_EXIT -EFAULT; - } - } - TRACE_EXIT result; -} diff --git a/drivers/char/ftape/zftape/zftape-ctl.h b/drivers/char/ftape/zftape/zftape-ctl.h deleted file mode 100644 index 8e6f2d7ac74..00000000000 --- a/drivers/char/ftape/zftape/zftape-ctl.h +++ /dev/null @@ -1,58 +0,0 @@ -#ifndef _ZFTAPE_CTL_H -#define _ZFTAPE_CTL_H - -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-ctl.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:02 $ - * - * This file contains the non-standard IOCTL related definitions - * for the QIC-40/80 floppy-tape driver for Linux. - */ - -#include -#include - -#include "../zftape/zftape-rw.h" - -#ifdef CONFIG_ZFTAPE_MODULE -#define ftape_status (*zft_status) -#endif - -extern int zft_offline; -extern int zft_mt_compression; -extern int zft_write_protected; -extern int zft_header_read; -extern unsigned int zft_unit; -extern int zft_resid; - -extern void zft_reset_position(zft_position *pos); -extern int zft_check_write_access(zft_position *pos); -extern int zft_def_idle_state(void); - -/* hooks for the VFS interface - */ -extern int _zft_open(unsigned int dev_minor, unsigned int access_mode); -extern int _zft_close(void); -extern int _zft_ioctl(unsigned int command, void __user *arg); -#endif - - - diff --git a/drivers/char/ftape/zftape/zftape-eof.c b/drivers/char/ftape/zftape/zftape-eof.c deleted file mode 100644 index dcadcaee9ac..00000000000 --- a/drivers/char/ftape/zftape/zftape-eof.c +++ /dev/null @@ -1,199 +0,0 @@ -/* - * I use these routines just to decide when I have to fake a - * volume-table to preserve compatibility to original ftape. - */ -/* - * Copyright (C) 1994-1995 Bas Laarhoven. - * - * Modified for zftape 1996, 1997 Claus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-eof.c,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:02 $ - * - * This file contains the eof mark handling code - * for the QIC-40/80 floppy-tape driver for Linux. - */ - -#include -#include - -#include - -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-rw.h" -#include "../zftape/zftape-eof.h" - -/* Global vars. - */ - -/* a copy of the failed sector log from the header segment. - */ -eof_mark_union *zft_eof_map; - -/* number of eof marks (entries in bad sector log) on tape. - */ -int zft_nr_eof_marks = -1; - - -/* Local vars. - */ - -static char linux_tape_label[] = "Linux raw format V"; -enum { - min_fmt_version = 1, max_fmt_version = 2 -}; -static unsigned ftape_fmt_version = 0; - - -/* Ftape (mis)uses the bad sector log to record end-of-file marks. - * Initially (when the tape is erased) all entries in the bad sector - * log are added to the tape's bad sector map. The bad sector log then - * is cleared. - * - * The bad sector log normally contains entries of the form: - * even 16-bit word: segment number of bad sector - * odd 16-bit word: encoded date - * There can be a total of 448 entries (1792 bytes). - * - * My guess is that no program is using this bad sector log (the * - * format seems useless as there is no indication of the bad sector - * itself, only the segment) However, if any program does use the bad - * sector log, the format used by ftape will let the program think - * there are some bad sectors and no harm is done. - * - * The eof mark entries that ftape stores in the bad sector log: even - * 16-bit word: segment number of eof mark odd 16-bit word: sector - * number of eof mark [1..32] - * - * The zft_eof_map as maintained is a sorted list of eof mark entries. - * - * - * The tape name field in the header segments is used to store a linux - * tape identification string and a version number. This way the tape - * can be recognized as a Linux raw format tape when using tools under - * other OS's. - * - * 'Wide' QIC tapes (format code 4) don't have a failed sector list - * anymore. That space is used for the (longer) bad sector map that - * now is a variable length list too. We now store our end-of-file - * marker list after the bad-sector-map on tape. The list is delimited - * by a (__u32) 0 entry. - */ - -int zft_ftape_validate_label(char *label) -{ - static char tmp_label[45]; - int result = 0; - TRACE_FUN(ft_t_any); - - memcpy(tmp_label, label, FT_LABEL_SZ); - tmp_label[FT_LABEL_SZ] = '\0'; - TRACE(ft_t_noise, "tape label = `%s'", tmp_label); - ftape_fmt_version = 0; - if (memcmp(label, linux_tape_label, strlen(linux_tape_label)) == 0) { - int pos = strlen(linux_tape_label); - while (label[pos] >= '0' && label[pos] <= '9') { - ftape_fmt_version *= 10; - ftape_fmt_version = label[ pos++] - '0'; - } - result = (ftape_fmt_version >= min_fmt_version && - ftape_fmt_version <= max_fmt_version); - } - TRACE(ft_t_noise, "format version = %d", ftape_fmt_version); - TRACE_EXIT result; -} - -static __u8 * find_end_of_eof_list(__u8 * ptr, __u8 * limit) -{ - while (ptr + 3 < limit) { - - if (get_unaligned((__u32*)ptr)) { - ptr += sizeof(__u32); - } else { - return ptr; - } - } - return NULL; -} - -void zft_ftape_extract_file_marks(__u8* address) -{ - int i; - TRACE_FUN(ft_t_any); - - zft_eof_map = NULL; - if (ft_format_code == fmt_var || ft_format_code == fmt_big) { - __u8* end; - __u8* start = ftape_find_end_of_bsm_list(address); - - zft_nr_eof_marks = 0; - if (start) { - start += 3; /* skip end of list mark */ - end = find_end_of_eof_list(start, - address + FT_SEGMENT_SIZE); - if (end && end - start <= FT_FSL_SIZE) { - zft_nr_eof_marks = ((end - start) / - sizeof(eof_mark_union)); - zft_eof_map = (eof_mark_union *)start; - } else { - TRACE(ft_t_err, - "EOF Mark List is too long or damaged!"); - } - } else { - TRACE(ft_t_err, - "Bad Sector List is too long or damaged !"); - } - } else { - zft_eof_map = (eof_mark_union *)&address[FT_FSL]; - zft_nr_eof_marks = GET2(address, FT_FSL_CNT); - } - TRACE(ft_t_noise, "number of file marks: %d", zft_nr_eof_marks); - if (ftape_fmt_version == 1) { - TRACE(ft_t_info, "swapping version 1 fields"); - /* version 1 format uses swapped sector and segment - * fields, correct that ! - */ - for (i = 0; i < zft_nr_eof_marks; ++i) { - __u16 tmp = GET2(&zft_eof_map[i].mark.segment,0); - PUT2(&zft_eof_map[i].mark.segment, 0, - GET2(&zft_eof_map[i].mark.date,0)); - PUT2(&zft_eof_map[i].mark.date, 0, tmp); - } - } - for (i = 0; i < zft_nr_eof_marks; ++i) { - TRACE(ft_t_noise, "eof mark: %5d/%2d", - GET2(&zft_eof_map[i].mark.segment, 0), - GET2(&zft_eof_map[i].mark.date,0)); - } - TRACE_EXIT; -} - -void zft_clear_ftape_file_marks(void) -{ - TRACE_FUN(ft_t_flow); - /* Clear failed sector log: remove all tape marks. We - * don't use old ftape-style EOF-marks. - */ - TRACE(ft_t_info, "Clearing old ftape's eof map"); - memset(zft_eof_map, 0, zft_nr_eof_marks * sizeof(__u32)); - zft_nr_eof_marks = 0; - PUT2(zft_hseg_buf, FT_FSL_CNT, 0); /* nr of eof-marks */ - zft_header_changed = 1; - zft_update_label(zft_hseg_buf); - TRACE_EXIT; -} diff --git a/drivers/char/ftape/zftape/zftape-eof.h b/drivers/char/ftape/zftape/zftape-eof.h deleted file mode 100644 index 26568c26c51..00000000000 --- a/drivers/char/ftape/zftape/zftape-eof.h +++ /dev/null @@ -1,52 +0,0 @@ -#ifndef _ZFTAPE_EOF_H -#define _ZFTAPE_EOF_H - -/* - * Copyright (C) 1994-1995 Bas Laarhoven. - * adaptaed for zftape 1996, 1997 by Claus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-eof.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:03 $ - * - * Definitions and declarations for the end of file markers - * for the QIC-40/80 floppy-tape driver for Linux. - */ - -#include -#include "../zftape/zftape-buffers.h" -/* failed sector log size (only used if format code != 4). - */ - -typedef union { - ft_fsl_entry mark; - __u32 entry; -} eof_mark_union; - -/* ftape-eof.c defined global vars. - */ -extern int zft_nr_eof_marks; -extern eof_mark_union *zft_eof_map; - -/* ftape-eof.c defined global functions. - */ -extern void zft_ftape_extract_file_marks(__u8* address); -extern int zft_ftape_validate_label(char* label); -extern void zft_clear_ftape_file_marks(void); - -#endif diff --git a/drivers/char/ftape/zftape/zftape-init.c b/drivers/char/ftape/zftape/zftape-init.c deleted file mode 100644 index 164a1aa77a2..00000000000 --- a/drivers/char/ftape/zftape/zftape-init.c +++ /dev/null @@ -1,377 +0,0 @@ -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * This file contains the code that registers the zftape frontend - * to the ftape floppy tape driver for Linux - */ - -#include -#include -#include -#include -#include -#include -#include -#ifdef CONFIG_KMOD -#include -#endif -#include -#include - -#include -#include -#include - -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-read.h" -#include "../zftape/zftape-write.h" -#include "../zftape/zftape-ctl.h" -#include "../zftape/zftape-buffers.h" - -MODULE_AUTHOR("(c) 1996, 1997 Claus-Justus Heine " - "(claus@momo.math.rwth-aachen.de)"); -MODULE_DESCRIPTION(ZFTAPE_VERSION " - " - "VFS interface for the Linux floppy tape driver. " - "Support for QIC-113 compatible volume table " - "and builtin compression (lzrw3 algorithm)"); -MODULE_SUPPORTED_DEVICE("char-major-27"); -MODULE_LICENSE("GPL"); - -/* Global vars. - */ -struct zft_cmpr_ops *zft_cmpr_ops = NULL; -const ftape_info *zft_status; - -/* Local vars. - */ -static unsigned long busy_flag; - -static sigset_t orig_sigmask; - -/* the interface to the kernel vfs layer - */ - -/* Note about llseek(): - * - * st.c and tpqic.c update fp->f_pos but don't implment llseek() and - * initialize the llseek component of the file_ops struct with NULL. - * This means that the user will get the default seek, but the tape - * device will not respect the new position, but happily read from the - * old position. Think a zftape specific llseek() function would be - * better, returning -ESPIPE. TODO. - */ - -static int zft_open (struct inode *ino, struct file *filep); -static int zft_close(struct inode *ino, struct file *filep); -static int zft_ioctl(struct inode *ino, struct file *filep, - unsigned int command, unsigned long arg); -static int zft_mmap(struct file *filep, struct vm_area_struct *vma); -static ssize_t zft_read (struct file *fp, char __user *buff, - size_t req_len, loff_t *ppos); -static ssize_t zft_write(struct file *fp, const char __user *buff, - size_t req_len, loff_t *ppos); - -static const struct file_operations zft_cdev = -{ - .owner = THIS_MODULE, - .read = zft_read, - .write = zft_write, - .ioctl = zft_ioctl, - .mmap = zft_mmap, - .open = zft_open, - .release = zft_close, -}; - -static struct class *zft_class; - -/* Open floppy tape device - */ -static int zft_open(struct inode *ino, struct file *filep) -{ - int result; - TRACE_FUN(ft_t_flow); - - nonseekable_open(ino, filep); - TRACE(ft_t_flow, "called for minor %d", iminor(ino)); - if ( test_and_set_bit(0,&busy_flag) ) { - TRACE_ABORT(-EBUSY, ft_t_warn, "failed: already busy"); - } - if ((iminor(ino) & ~(ZFT_MINOR_OP_MASK | FTAPE_NO_REWIND)) - > - FTAPE_SEL_D) { - clear_bit(0,&busy_flag); - TRACE_ABORT(-ENXIO, ft_t_err, "failed: invalid unit nr"); - } - orig_sigmask = current->blocked; - sigfillset(¤t->blocked); - result = _zft_open(iminor(ino), filep->f_flags & O_ACCMODE); - if (result < 0) { - current->blocked = orig_sigmask; /* restore mask */ - clear_bit(0,&busy_flag); - TRACE_ABORT(result, ft_t_err, "_ftape_open failed"); - } else { - /* Mask signals that will disturb proper operation of the - * program that is calling. - */ - current->blocked = orig_sigmask; - sigaddsetmask (¤t->blocked, _DO_BLOCK); - TRACE_EXIT 0; - } -} - -/* Close floppy tape device - */ -static int zft_close(struct inode *ino, struct file *filep) -{ - int result; - TRACE_FUN(ft_t_flow); - - if ( !test_bit(0,&busy_flag) || iminor(ino) != zft_unit) { - TRACE(ft_t_err, "failed: not busy or wrong unit"); - TRACE_EXIT 0; - } - sigfillset(¤t->blocked); - result = _zft_close(); - if (result < 0) { - TRACE(ft_t_err, "_zft_close failed"); - } - current->blocked = orig_sigmask; /* restore before open state */ - clear_bit(0,&busy_flag); - TRACE_EXIT 0; -} - -/* Ioctl for floppy tape device - */ -static int zft_ioctl(struct inode *ino, struct file *filep, - unsigned int command, unsigned long arg) -{ - int result = -EIO; - sigset_t old_sigmask; - TRACE_FUN(ft_t_flow); - - if ( !test_bit(0,&busy_flag) || iminor(ino) != zft_unit || ft_failure) { - TRACE_ABORT(-EIO, ft_t_err, - "failed: not busy, failure or wrong unit"); - } - old_sigmask = current->blocked; /* save mask */ - sigfillset(¤t->blocked); - /* This will work as long as sizeof(void *) == sizeof(long) */ - result = _zft_ioctl(command, (void __user *) arg); - current->blocked = old_sigmask; /* restore mask */ - TRACE_EXIT result; -} - -/* Ioctl for floppy tape device - */ -static int zft_mmap(struct file *filep, struct vm_area_struct *vma) -{ - int result = -EIO; - sigset_t old_sigmask; - TRACE_FUN(ft_t_flow); - - if ( !test_bit(0,&busy_flag) || - iminor(filep->f_dentry->d_inode) != zft_unit || - ft_failure) - { - TRACE_ABORT(-EIO, ft_t_err, - "failed: not busy, failure or wrong unit"); - } - old_sigmask = current->blocked; /* save mask */ - sigfillset(¤t->blocked); - if ((result = ftape_mmap(vma)) >= 0) { -#ifndef MSYNC_BUG_WAS_FIXED - static struct vm_operations_struct dummy = { NULL, }; - vma->vm_ops = &dummy; -#endif - } - current->blocked = old_sigmask; /* restore mask */ - TRACE_EXIT result; -} - -/* Read from floppy tape device - */ -static ssize_t zft_read(struct file *fp, char __user *buff, - size_t req_len, loff_t *ppos) -{ - int result = -EIO; - sigset_t old_sigmask; - struct inode *ino = fp->f_dentry->d_inode; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_data_flow, "called with count: %ld", (unsigned long)req_len); - if (!test_bit(0,&busy_flag) || iminor(ino) != zft_unit || ft_failure) { - TRACE_ABORT(-EIO, ft_t_err, - "failed: not busy, failure or wrong unit"); - } - old_sigmask = current->blocked; /* save mask */ - sigfillset(¤t->blocked); - result = _zft_read(buff, req_len); - current->blocked = old_sigmask; /* restore mask */ - TRACE(ft_t_data_flow, "return with count: %d", result); - TRACE_EXIT result; -} - -/* Write to tape device - */ -static ssize_t zft_write(struct file *fp, const char __user *buff, - size_t req_len, loff_t *ppos) -{ - int result = -EIO; - sigset_t old_sigmask; - struct inode *ino = fp->f_dentry->d_inode; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_flow, "called with count: %ld", (unsigned long)req_len); - if (!test_bit(0,&busy_flag) || iminor(ino) != zft_unit || ft_failure) { - TRACE_ABORT(-EIO, ft_t_err, - "failed: not busy, failure or wrong unit"); - } - old_sigmask = current->blocked; /* save mask */ - sigfillset(¤t->blocked); - result = _zft_write(buff, req_len); - current->blocked = old_sigmask; /* restore mask */ - TRACE(ft_t_data_flow, "return with count: %d", result); - TRACE_EXIT result; -} - -/* END OF VFS INTERFACE - * - *****************************************************************************/ - -/* driver/module initialization - */ - -/* the compression module has to call this function to hook into the zftape - * code - */ -int zft_cmpr_register(struct zft_cmpr_ops *new_ops) -{ - TRACE_FUN(ft_t_flow); - - if (zft_cmpr_ops != NULL) { - TRACE_EXIT -EBUSY; - } else { - zft_cmpr_ops = new_ops; - TRACE_EXIT 0; - } -} - -/* lock the zft-compressor() module. - */ -int zft_cmpr_lock(int try_to_load) -{ - if (zft_cmpr_ops == NULL) { -#ifdef CONFIG_KMOD - if (try_to_load) { - request_module("zft-compressor"); - if (zft_cmpr_ops == NULL) { - return -ENOSYS; - } - } else { - return -ENOSYS; - } -#else - return -ENOSYS; -#endif - } - (*zft_cmpr_ops->lock)(); - return 0; -} - -#ifdef CONFIG_ZFT_COMPRESSOR -extern int zft_compressor_init(void); -#endif - -/* Called by modules package when installing the driver or by kernel - * during the initialization phase - */ -int __init zft_init(void) -{ - int i; - TRACE_FUN(ft_t_flow); - -#ifdef MODULE - printk(KERN_INFO ZFTAPE_VERSION "\n"); - if (TRACE_LEVEL >= ft_t_info) { - printk( -KERN_INFO -"(c) 1996, 1997 Claus-Justus Heine (claus@momo.math.rwth-aachen.de)\n" -KERN_INFO -"vfs interface for ftape floppy tape driver.\n" -KERN_INFO -"Support for QIC-113 compatible volume table, dynamic memory allocation\n" -KERN_INFO -"and builtin compression (lzrw3 algorithm).\n"); - } -#else /* !MODULE */ - /* print a short no-nonsense boot message */ - printk(KERN_INFO ZFTAPE_VERSION "\n"); -#endif /* MODULE */ - TRACE(ft_t_info, "zft_init @ 0x%p", zft_init); - TRACE(ft_t_info, - "installing zftape VFS interface for ftape driver ..."); - TRACE_CATCH(register_chrdev(QIC117_TAPE_MAJOR, "zft", &zft_cdev),); - - zft_class = class_create(THIS_MODULE, "zft"); - for (i = 0; i < 4; i++) { - class_device_create(zft_class, NULL, MKDEV(QIC117_TAPE_MAJOR, i), NULL, "qft%i", i); - class_device_create(zft_class, NULL, MKDEV(QIC117_TAPE_MAJOR, i + 4), NULL, "nqft%i", i); - class_device_create(zft_class, NULL, MKDEV(QIC117_TAPE_MAJOR, i + 16), NULL, "zqft%i", i); - class_device_create(zft_class, NULL, MKDEV(QIC117_TAPE_MAJOR, i + 20), NULL, "nzqft%i", i); - class_device_create(zft_class, NULL, MKDEV(QIC117_TAPE_MAJOR, i + 32), NULL, "rawqft%i", i); - class_device_create(zft_class, NULL, MKDEV(QIC117_TAPE_MAJOR, i + 36), NULL, "nrawrawqft%i", i); - } - -#ifdef CONFIG_ZFT_COMPRESSOR - (void)zft_compressor_init(); -#endif - zft_status = ftape_get_status(); /* fetch global data of ftape - * hardware driver - */ - TRACE_EXIT 0; -} - - -/* Called by modules package when removing the driver - */ -static void zft_exit(void) -{ - int i; - TRACE_FUN(ft_t_flow); - - if (unregister_chrdev(QIC117_TAPE_MAJOR, "zft") != 0) { - TRACE(ft_t_warn, "failed"); - } else { - TRACE(ft_t_info, "successful"); - } - for (i = 0; i < 4; i++) { - class_device_destroy(zft_class, MKDEV(QIC117_TAPE_MAJOR, i)); - class_device_destroy(zft_class, MKDEV(QIC117_TAPE_MAJOR, i + 4)); - class_device_destroy(zft_class, MKDEV(QIC117_TAPE_MAJOR, i + 16)); - class_device_destroy(zft_class, MKDEV(QIC117_TAPE_MAJOR, i + 20)); - class_device_destroy(zft_class, MKDEV(QIC117_TAPE_MAJOR, i + 32)); - class_device_destroy(zft_class, MKDEV(QIC117_TAPE_MAJOR, i + 36)); - } - class_destroy(zft_class); - zft_uninit_mem(); /* release remaining memory, if any */ - printk(KERN_INFO "zftape successfully unloaded.\n"); - TRACE_EXIT; -} - -module_init(zft_init); -module_exit(zft_exit); diff --git a/drivers/char/ftape/zftape/zftape-init.h b/drivers/char/ftape/zftape/zftape-init.h deleted file mode 100644 index 937e5d48c20..00000000000 --- a/drivers/char/ftape/zftape/zftape-init.h +++ /dev/null @@ -1,77 +0,0 @@ -#ifndef _ZFTAPE_INIT_H -#define _ZFTAPE_INIT_H - -/* - * Copyright (C) 1996, 1997 Claus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-init.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:05 $ - * - * This file contains definitions and macro for the vfs - * interface defined by zftape - * - */ - -#include - -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-read.h" -#include "../lowlevel/ftape-write.h" -#include "../lowlevel/ftape-bsm.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-buffer.h" -#include "../lowlevel/ftape-format.h" - -#include "../zftape/zftape-rw.h" - -#ifdef MODULE -#define ftape_status (*zft_status) -#endif - -extern const ftape_info *zft_status; /* needed for zftape-vtbl.h */ - -#include "../zftape/zftape-vtbl.h" - -struct zft_cmpr_ops { - int (*write)(int *write_cnt, - __u8 *dst_buf, const int seg_sz, - const __u8 __user *src_buf, const int req_len, - const zft_position *pos, const zft_volinfo *volume); - int (*read)(int *read_cnt, - __u8 __user *dst_buf, const int req_len, - const __u8 *src_buf, const int seg_sz, - const zft_position *pos, const zft_volinfo *volume); - int (*seek)(unsigned int new_block_pos, - zft_position *pos, const zft_volinfo *volume, - __u8 *buffer); - void (*lock) (void); - void (*reset) (void); - void (*cleanup)(void); -}; - -extern struct zft_cmpr_ops *zft_cmpr_ops; -/* zftape-init.c defined global functions. - */ -extern int zft_cmpr_register(struct zft_cmpr_ops *new_ops); -extern int zft_cmpr_lock(int try_to_load); - -#endif - - diff --git a/drivers/char/ftape/zftape/zftape-read.c b/drivers/char/ftape/zftape/zftape-read.c deleted file mode 100644 index 214bf03dce6..00000000000 --- a/drivers/char/ftape/zftape/zftape-read.c +++ /dev/null @@ -1,377 +0,0 @@ -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-read.c,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:06 $ - * - * This file contains the high level reading code - * for the QIC-117 floppy-tape driver for Linux. - */ - -#include -#include - -#include - -#include - -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-eof.h" -#include "../zftape/zftape-ctl.h" -#include "../zftape/zftape-write.h" -#include "../zftape/zftape-read.h" -#include "../zftape/zftape-rw.h" -#include "../zftape/zftape-vtbl.h" - -/* Global vars. - */ -int zft_just_before_eof; - -/* Local vars. - */ -static int buf_len_rd; - -void zft_zap_read_buffers(void) -{ - buf_len_rd = 0; -} - -int zft_read_header_segments(void) -{ - TRACE_FUN(ft_t_flow); - - zft_header_read = 0; - TRACE_CATCH(zft_vmalloc_once(&zft_hseg_buf, FT_SEGMENT_SIZE),); - TRACE_CATCH(ftape_read_header_segment(zft_hseg_buf),); - TRACE(ft_t_info, "Segments written since first format: %d", - (int)GET4(zft_hseg_buf, FT_SEG_CNT)); - zft_qic113 = (ft_format_code != fmt_normal && - ft_format_code != fmt_1100ft && - ft_format_code != fmt_425ft); - TRACE(ft_t_info, "ft_first_data_segment: %d, ft_last_data_segment: %d", - ft_first_data_segment, ft_last_data_segment); - zft_capacity = zft_get_capacity(); - zft_old_ftape = zft_ftape_validate_label(&zft_hseg_buf[FT_LABEL]); - if (zft_old_ftape) { - TRACE(ft_t_info, -"Found old ftaped tape, emulating eof marks, entering read-only mode"); - zft_ftape_extract_file_marks(zft_hseg_buf); - TRACE_CATCH(zft_fake_volume_headers(zft_eof_map, - zft_nr_eof_marks),); - } else { - /* the specs say that the volume table must be - * initialized with zeroes during formatting, so it - * MUST be readable, i.e. contain vaid ECC - * information. - */ - TRACE_CATCH(ftape_read_segment(ft_first_data_segment, - zft_deblock_buf, - FT_RD_SINGLE),); - TRACE_CATCH(zft_extract_volume_headers(zft_deblock_buf),); - } - zft_header_read = 1; - zft_set_flags(zft_unit); - zft_reset_position(&zft_pos); - TRACE_EXIT 0; -} - -int zft_fetch_segment_fraction(const unsigned int segment, void *buffer, - const ft_read_mode_t read_mode, - const unsigned int start, - const unsigned int size) -{ - int seg_sz; - TRACE_FUN(ft_t_flow); - - if (segment == zft_deblock_segment) { - TRACE(ft_t_data_flow, - "re-using segment %d already in deblock buffer", - segment); - seg_sz = zft_get_seg_sz(segment); - if (start > seg_sz) { - TRACE_ABORT(-EINVAL, ft_t_bug, - "trying to read beyond end of segment:\n" - KERN_INFO "seg_sz : %d\n" - KERN_INFO "start : %d\n" - KERN_INFO "segment: %d", - seg_sz, start, segment); - } - if ((start + size) > seg_sz) { - TRACE_EXIT seg_sz - start; - } - TRACE_EXIT size; - } - seg_sz = ftape_read_segment_fraction(segment, buffer, read_mode, - start, size); - TRACE(ft_t_data_flow, "segment %d, result %d", segment, seg_sz); - if ((int)seg_sz >= 0 && start == 0 && size == FT_SEGMENT_SIZE) { - /* this implicitly assumes that we are always called with - * buffer == zft_deblock_buf - */ - zft_deblock_segment = segment; - } else { - zft_deblock_segment = -1; - } - TRACE_EXIT seg_sz; -} - -/* - * out: - * - * int *read_cnt: the number of bytes we removed from the - * zft_deblock_buf (result) - * - * int *to_do : the remaining size of the read-request. Is changed. - * - * in: - * - * char *buff : buff is the address of the upper part of the user - * buffer, that hasn't been filled with data yet. - * int buf_pos_read: copy of buf_pos_rd - * int buf_len_read: copy of buf_len_rd - * char *zft_deblock_buf: ftape_zft_deblock_buf - * - * returns the amount of data actually copied to the user-buffer - * - * to_do MUST NOT SHRINK except to indicate an EOT. In this case to_do - * has to be set to 0. We cannot return -ENOSPC, because we return the - * amount of data actually * copied to the user-buffer - */ -static int zft_simple_read (int *read_cnt, - __u8 __user *dst_buf, - const int to_do, - const __u8 *src_buf, - const int seg_sz, - const zft_position *pos, - const zft_volinfo *volume) -{ - TRACE_FUN(ft_t_flow); - - if (seg_sz - pos->seg_byte_pos < to_do) { - *read_cnt = seg_sz - pos->seg_byte_pos; - } else { - *read_cnt = to_do; - } - if (copy_to_user(dst_buf, - src_buf + pos->seg_byte_pos, *read_cnt) != 0) { - TRACE_EXIT -EFAULT; - } - TRACE(ft_t_noise, "nr bytes just read: %d", *read_cnt); - TRACE_EXIT *read_cnt; -} - -/* req_len: gets clipped due to EOT of EOF. - * req_clipped: is a flag indicating whether req_len was clipped or not - * volume: contains information on current volume (blk_sz etc.) - */ -static int check_read_access(int *req_len, - const zft_volinfo **volume, - int *req_clipped, - const zft_position *pos) -{ - static __s64 remaining; - static int eod; - TRACE_FUN(ft_t_flow); - - if (zft_io_state != zft_reading) { - if (zft_offline) { /* offline includes no_tape */ - TRACE_ABORT(-ENXIO, ft_t_warn, - "tape is offline or no cartridge"); - } - if (!ft_formatted) { - TRACE_ABORT(-EACCES, - ft_t_warn, "tape is not formatted"); - } - /* now enter defined state, read header segment if not - * already done and flush write buffers - */ - TRACE_CATCH(zft_def_idle_state(),); - zft_io_state = zft_reading; - if (zft_tape_at_eod(pos)) { - eod = 1; - TRACE_EXIT 1; - } - eod = 0; - *volume = zft_find_volume(pos->seg_pos); - /* get the space left until EOF */ - remaining = zft_check_for_eof(*volume, pos); - buf_len_rd = 0; - TRACE(ft_t_noise, "remaining: " LL_X ", vol_no: %d", - LL(remaining), (*volume)->count); - } else if (zft_tape_at_eod(pos)) { - if (++eod > 2) { - TRACE_EXIT -EIO; /* st.c also returns -EIO */ - } else { - TRACE_EXIT 1; - } - } - if ((*req_len % (*volume)->blk_sz) != 0) { - /* this message is informational only. The user gets the - * proper return value - */ - TRACE_ABORT(-EINVAL, ft_t_info, - "req_len %d not a multiple of block size %d", - *req_len, (*volume)->blk_sz); - } - /* As GNU tar doesn't accept partial read counts when the - * multiple volume flag is set, we make sure to return the - * requested amount of data. Except, of course, at the end of - * the tape or file mark. - */ - remaining -= *req_len; - if (remaining <= 0) { - TRACE(ft_t_noise, - "clipped request from %d to %d.", - *req_len, (int)(*req_len + remaining)); - *req_len += remaining; - *req_clipped = 1; - } else { - *req_clipped = 0; - } - TRACE_EXIT 0; -} - -/* this_segs_size: the current segment's size. - * buff: the USER-SPACE buffer provided by the calling function. - * req_len: how much data should be read at most. - * volume: contains information on current volume (blk_sz etc.) - */ -static int empty_deblock_buf(__u8 __user *usr_buf, const int req_len, - const __u8 *src_buf, const int seg_sz, - zft_position *pos, - const zft_volinfo *volume) -{ - int cnt; - int result = 0; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_data_flow, "this_segs_size: %d", seg_sz); - if (zft_use_compression && volume->use_compression) { - TRACE_CATCH(zft_cmpr_lock(1 /* try to load */),); - TRACE_CATCH(result= (*zft_cmpr_ops->read)(&cnt, - usr_buf, req_len, - src_buf, seg_sz, - pos, volume),); - } else { - TRACE_CATCH(result= zft_simple_read (&cnt, - usr_buf, req_len, - src_buf, seg_sz, - pos, volume),); - } - pos->volume_pos += result; - pos->tape_pos += cnt; - pos->seg_byte_pos += cnt; - buf_len_rd -= cnt; /* remaining bytes in buffer */ - TRACE(ft_t_data_flow, "buf_len_rd: %d, cnt: %d", buf_len_rd, cnt); - if(pos->seg_byte_pos >= seg_sz) { - pos->seg_pos++; - pos->seg_byte_pos = 0; - } - TRACE(ft_t_data_flow, "bytes moved out of deblock-buffer: %d", cnt); - TRACE_EXIT result; -} - - -/* note: we store the segment id of the segment that is inside the - * deblock buffer. This spares a lot of ftape_read_segment()s when we - * use small block-sizes. The block-size may be 1kb (SECTOR_SIZE). In - * this case a MTFSR 28 maybe still inside the same segment. - */ -int _zft_read(char __user *buff, int req_len) -{ - int req_clipped; - int result = 0; - int bytes_read = 0; - static unsigned int seg_sz = 0; - static const zft_volinfo *volume = NULL; - TRACE_FUN(ft_t_flow); - - zft_resid = req_len; - result = check_read_access(&req_len, &volume, - &req_clipped, &zft_pos); - switch(result) { - case 0: - break; /* nothing special */ - case 1: - TRACE(ft_t_noise, "EOD reached"); - TRACE_EXIT 0; /* EOD */ - default: - TRACE_ABORT(result, ft_t_noise, - "check_read_access() failed with result %d", - result); - TRACE_EXIT result; - } - while (req_len > 0) { - /* Allow escape from this loop on signal ! - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - /* buf_len_rd == 0 means that we need to read a new - * segment. - */ - if (buf_len_rd == 0) { - while((result = zft_fetch_segment(zft_pos.seg_pos, - zft_deblock_buf, - FT_RD_AHEAD)) == 0) { - zft_pos.seg_pos ++; - zft_pos.seg_byte_pos = 0; - } - if (result < 0) { - zft_resid -= bytes_read; - TRACE_ABORT(result, ft_t_noise, - "zft_fetch_segment(): %d", - result); - } - seg_sz = result; - buf_len_rd = seg_sz - zft_pos.seg_byte_pos; - } - TRACE_CATCH(result = empty_deblock_buf(buff, - req_len, - zft_deblock_buf, - seg_sz, - &zft_pos, - volume), - zft_resid -= bytes_read); - TRACE(ft_t_data_flow, "bytes just read: %d", result); - bytes_read += result; /* what we got so far */ - buff += result; /* index in user-buffer */ - req_len -= result; /* what's left from req_len */ - } /* while (req_len > 0) */ - if (req_clipped) { - TRACE(ft_t_data_flow, - "maybe partial count because of eof mark"); - if (zft_just_before_eof && bytes_read == 0) { - /* req_len was > 0, but user didn't get - * anything the user has read in the eof-mark - */ - zft_move_past_eof(&zft_pos); - ftape_abort_operation(); - } else { - /* don't skip to the next file before the user - * tried to read a second time past EOF Just - * mark that we are at EOF and maybe decrement - * zft_seg_pos to stay in the same volume; - */ - zft_just_before_eof = 1; - zft_position_before_eof(&zft_pos, volume); - TRACE(ft_t_noise, "just before eof"); - } - } - zft_resid -= result; /* for MTSTATUS */ - TRACE_EXIT bytes_read; -} diff --git a/drivers/char/ftape/zftape/zftape-read.h b/drivers/char/ftape/zftape/zftape-read.h deleted file mode 100644 index 42941de0c23..00000000000 --- a/drivers/char/ftape/zftape/zftape-read.h +++ /dev/null @@ -1,53 +0,0 @@ -#ifndef _ZFTAPE_READ_H -#define _ZFTAPE_READ_H - -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-read.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:07 $ - * - * This file contains the definitions for the read functions - * for the zftape driver for Linux. - * - */ - -#include "../lowlevel/ftape-read.h" - -/* ftape-read.c defined global vars. - */ -extern int zft_just_before_eof; - -/* ftape-read.c defined global functions. - */ -extern void zft_zap_read_buffers(void); -extern int zft_read_header_segments(void); -extern int zft_fetch_segment_fraction(const unsigned int segment, - void *buffer, - const ft_read_mode_t read_mode, - const unsigned int start, - const unsigned int size); -#define zft_fetch_segment(segment, address, read_mode) \ - zft_fetch_segment_fraction(segment, address, read_mode, \ - 0, FT_SEGMENT_SIZE) -/* hook for the VFS interface - */ -extern int _zft_read(char __user *buff, int req_len); - -#endif /* _ZFTAPE_READ_H */ diff --git a/drivers/char/ftape/zftape/zftape-rw.c b/drivers/char/ftape/zftape/zftape-rw.c deleted file mode 100644 index dab63468688..00000000000 --- a/drivers/char/ftape/zftape/zftape-rw.c +++ /dev/null @@ -1,375 +0,0 @@ -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-rw.c,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:08 $ - * - * This file contains some common code for the r/w code for - * zftape. - */ - -#include -#include - -#include -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-eof.h" -#include "../zftape/zftape-ctl.h" -#include "../zftape/zftape-write.h" -#include "../zftape/zftape-read.h" -#include "../zftape/zftape-rw.h" -#include "../zftape/zftape-vtbl.h" - -/* Global vars. - */ - -__u8 *zft_deblock_buf; -__u8 *zft_hseg_buf; -int zft_deblock_segment = -1; -zft_status_enum zft_io_state = zft_idle; -int zft_header_changed; -int zft_qic113; /* conform to old specs. and old zftape */ -int zft_use_compression; -zft_position zft_pos = { - -1, /* seg_pos */ - 0, /* seg_byte_pos */ - 0, /* tape_pos */ - 0 /* volume_pos */ -}; -unsigned int zft_blk_sz = CONFIG_ZFT_DFLT_BLK_SZ; -__s64 zft_capacity; - -unsigned int zft_written_segments; -int zft_label_changed; - -/* Local vars. - */ - -unsigned int zft_get_seg_sz(unsigned int segment) -{ - int size; - TRACE_FUN(ft_t_any); - - size = FT_SEGMENT_SIZE - - count_ones(ftape_get_bad_sector_entry(segment))*FT_SECTOR_SIZE; - if (size > 0) { - TRACE_EXIT (unsigned)size; - } else { - TRACE_EXIT 0; - } -} - -/* ftape_set_flags(). Claus-Justus Heine, 1994/1995 - */ -void zft_set_flags(unsigned minor_unit) -{ - TRACE_FUN(ft_t_flow); - - zft_use_compression = zft_qic_mode = 0; - switch (minor_unit & ZFT_MINOR_OP_MASK) { - case (ZFT_Q80_MODE | ZFT_ZIP_MODE): - case ZFT_ZIP_MODE: - zft_use_compression = 1; - case 0: - case ZFT_Q80_MODE: - zft_qic_mode = 1; - if (zft_mt_compression) { /* override the default */ - zft_use_compression = 1; - } - break; - case ZFT_RAW_MODE: - TRACE(ft_t_noise, "switching to raw mode"); - break; - default: - TRACE(ft_t_warn, "Warning:\n" - KERN_INFO "Wrong combination of minor device bits.\n" - KERN_INFO "Switching to raw read-only mode."); - zft_write_protected = 1; - break; - } - TRACE_EXIT; -} - -/* computes the segment and byte offset inside the segment - * corresponding to tape_pos. - * - * tape_pos gives the offset in bytes from the beginning of the - * ft_first_data_segment *seg_byte_pos is the offset in the current - * segment in bytes - * - * Of, if this routine was called often one should cache the last data - * pos it was called with, but actually this is only needed in - * ftape_seek_block(), that is, almost never. - */ -int zft_calc_seg_byte_coord(int *seg_byte_pos, __s64 tape_pos) -{ - int segment; - int seg_sz; - TRACE_FUN(ft_t_flow); - - if (tape_pos == 0) { - *seg_byte_pos = 0; - segment = ft_first_data_segment; - } else { - seg_sz = 0; - - for (segment = ft_first_data_segment; - ((tape_pos > 0) && (segment <= ft_last_data_segment)); - segment++) { - seg_sz = zft_get_seg_sz(segment); - tape_pos -= seg_sz; - } - if(tape_pos >= 0) { - /* the case tape_pos > != 0 means that the - * argument tape_pos lies beyond the EOT. - */ - *seg_byte_pos= 0; - } else { /* tape_pos < 0 */ - segment--; - *seg_byte_pos= tape_pos + seg_sz; - } - } - TRACE_EXIT(segment); -} - -/* ftape_calc_tape_pos(). - * - * computes the offset in bytes from the beginning of the - * ft_first_data_segment inverse to ftape_calc_seg_byte_coord - * - * We should do some caching. But how: - * - * Each time the header segments are read in, this routine is called - * with ft_tracks_per_tape*segments_per_track argumnet. So this should be - * the time to reset the cache. - * - * Also, it might be in the future that the bad sector map gets - * changed. -> reset the cache - */ -static int seg_pos; -static __s64 tape_pos; - -__s64 zft_get_capacity(void) -{ - seg_pos = ft_first_data_segment; - tape_pos = 0; - - while (seg_pos <= ft_last_data_segment) { - tape_pos += zft_get_seg_sz(seg_pos ++); - } - return tape_pos; -} - -__s64 zft_calc_tape_pos(int segment) -{ - int d1, d2, d3; - TRACE_FUN(ft_t_any); - - if (segment > ft_last_data_segment) { - TRACE_EXIT zft_capacity; - } - if (segment < ft_first_data_segment) { - TRACE_EXIT 0; - } - d2 = segment - seg_pos; - if (-d2 > 10) { - d1 = segment - ft_first_data_segment; - if (-d2 > d1) { - tape_pos = 0; - seg_pos = ft_first_data_segment; - d2 = d1; - } - } - if (d2 > 10) { - d3 = ft_last_data_segment - segment; - if (d2 > d3) { - tape_pos = zft_capacity; - seg_pos = ft_last_data_segment + 1; - d2 = -d3; - } - } - if (d2 > 0) { - while (seg_pos < segment) { - tape_pos += zft_get_seg_sz(seg_pos++); - } - } else { - while (seg_pos > segment) { - tape_pos -= zft_get_seg_sz(--seg_pos); - } - } - TRACE(ft_t_noise, "new cached pos: %d", seg_pos); - - TRACE_EXIT tape_pos; -} - -/* copy Z-label string to buffer, keeps track of the correct offset in - * `buffer' - */ -void zft_update_label(__u8 *buffer) -{ - TRACE_FUN(ft_t_flow); - - if (strncmp(&buffer[FT_LABEL], ZFTAPE_LABEL, - sizeof(ZFTAPE_LABEL)-1) != 0) { - TRACE(ft_t_info, "updating label from \"%s\" to \"%s\"", - &buffer[FT_LABEL], ZFTAPE_LABEL); - strcpy(&buffer[FT_LABEL], ZFTAPE_LABEL); - memset(&buffer[FT_LABEL] + sizeof(ZFTAPE_LABEL) - 1, ' ', - FT_LABEL_SZ - sizeof(ZFTAPE_LABEL + 1)); - PUT4(buffer, FT_LABEL_DATE, 0); - zft_label_changed = zft_header_changed = 1; /* changed */ - } - TRACE_EXIT; -} - -int zft_verify_write_segments(unsigned int segment, - __u8 *data, size_t size, - __u8 *buffer) -{ - int result; - __u8 *write_buf; - __u8 *src_buf; - int single; - int seg_pos; - int seg_sz; - int remaining; - ft_write_mode_t write_mode; - TRACE_FUN(ft_t_flow); - - seg_pos = segment; - seg_sz = zft_get_seg_sz(seg_pos); - src_buf = data; - single = size <= seg_sz; - remaining = size; - do { - TRACE(ft_t_noise, "\n" - KERN_INFO "remaining: %d\n" - KERN_INFO "seg_sz : %d\n" - KERN_INFO "segment : %d", - remaining, seg_sz, seg_pos); - if (remaining == seg_sz) { - write_buf = src_buf; - write_mode = single ? FT_WR_SINGLE : FT_WR_MULTI; - remaining = 0; - } else if (remaining > seg_sz) { - write_buf = src_buf; - write_mode = FT_WR_ASYNC; /* don't start tape */ - remaining -= seg_sz; - } else { /* remaining < seg_sz */ - write_buf = buffer; - memcpy(write_buf, src_buf, remaining); - memset(&write_buf[remaining],'\0',seg_sz-remaining); - write_mode = single ? FT_WR_SINGLE : FT_WR_MULTI; - remaining = 0; - } - if ((result = ftape_write_segment(seg_pos, - write_buf, - write_mode)) != seg_sz) { - TRACE(ft_t_err, "Error: " - "Couldn't write segment %d", seg_pos); - TRACE_EXIT result < 0 ? result : -EIO; /* bail out */ - } - zft_written_segments ++; - seg_sz = zft_get_seg_sz(++seg_pos); - src_buf += result; - } while (remaining > 0); - if (ftape_get_status()->fti_state == writing) { - TRACE_CATCH(ftape_loop_until_writes_done(),); - TRACE_CATCH(ftape_abort_operation(),); - zft_prevent_flush(); - } - seg_pos = segment; - src_buf = data; - remaining = size; - do { - TRACE_CATCH(result = ftape_read_segment(seg_pos, buffer, - single ? FT_RD_SINGLE - : FT_RD_AHEAD),); - if (memcmp(src_buf, buffer, - remaining > result ? result : remaining) != 0) { - TRACE_ABORT(-EIO, ft_t_err, - "Failed to verify written segment %d", - seg_pos); - } - remaining -= result; - TRACE(ft_t_noise, "verify successful:\n" - KERN_INFO "segment : %d\n" - KERN_INFO "segsize : %d\n" - KERN_INFO "remaining: %d", - seg_pos, result, remaining); - src_buf += seg_sz; - seg_pos++; - } while (remaining > 0); - TRACE_EXIT size; -} - - -/* zft_erase(). implemented compression-handling - * - * calculate the first data-segment when using/not using compression. - * - * update header-segment and compression-map-segment. - */ -int zft_erase(void) -{ - int result = 0; - TRACE_FUN(ft_t_flow); - - if (!zft_header_read) { - TRACE_CATCH(zft_vmalloc_once((void **)&zft_hseg_buf, - FT_SEGMENT_SIZE),); - /* no need to read the vtbl and compression map */ - TRACE_CATCH(ftape_read_header_segment(zft_hseg_buf),); - if ((zft_old_ftape = - zft_ftape_validate_label(&zft_hseg_buf[FT_LABEL]))) { - zft_ftape_extract_file_marks(zft_hseg_buf); - } - TRACE(ft_t_noise, - "ft_first_data_segment: %d, ft_last_data_segment: %d", - ft_first_data_segment, ft_last_data_segment); - zft_qic113 = (ft_format_code != fmt_normal && - ft_format_code != fmt_1100ft && - ft_format_code != fmt_425ft); - } - if (zft_old_ftape) { - zft_clear_ftape_file_marks(); - zft_old_ftape = 0; /* no longer old ftape */ - } - PUT2(zft_hseg_buf, FT_CMAP_START, 0); - zft_volume_table_changed = 1; - zft_capacity = zft_get_capacity(); - zft_init_vtbl(); - /* the rest must be done in ftape_update_header_segments - */ - zft_header_read = 1; - zft_header_changed = 1; /* force update of timestamp */ - result = zft_update_header_segments(); - - ftape_abort_operation(); - - zft_reset_position(&zft_pos); - zft_set_flags (zft_unit); - TRACE_EXIT result; -} - -unsigned int zft_get_time(void) -{ - unsigned int date = FT_TIME_STAMP(2097, 11, 30, 23, 59, 59); /* fun */ - return date; -} diff --git a/drivers/char/ftape/zftape/zftape-rw.h b/drivers/char/ftape/zftape/zftape-rw.h deleted file mode 100644 index 1ceec22b60b..00000000000 --- a/drivers/char/ftape/zftape/zftape-rw.h +++ /dev/null @@ -1,101 +0,0 @@ -#ifndef _ZFTAPE_RW_H -#define _ZFTAPE_RW_H - -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-rw.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:09 $ - * - * This file contains the definitions for the read and write - * functions for the QIC-117 floppy-tape driver for Linux. - * - */ - -#include "../zftape/zftape-buffers.h" - -#define SEGMENTS_PER_TAPE (ft_segments_per_track * ft_tracks_per_tape) - -/* QIC-113 Rev. G says that `a maximum of 63488 raw bytes may be - * compressed into a single frame'. - * Maybe we should stick to 32kb to make it more `beautiful' - */ -#define ZFT_MAX_BLK_SZ (62*1024) /* bytes */ -#if !defined(CONFIG_ZFT_DFLT_BLK_SZ) -# define CONFIG_ZFT_DFLT_BLK_SZ (10*1024) /* bytes, default of gnu tar */ -#elif CONFIG_ZFT_DFLT_BLK_SZ == 0 -# undef CONFIG_ZFT_DFLT_BLK_SZ -# define CONFIG_ZFT_DFLT_BLK_SZ 1 -#elif (CONFIG_ZFT_DFLT_BLK_SZ % 1024) != 0 -# error CONFIG_ZFT_DFLT_BLK_SZ must be 1 or a multiple of 1024 -#endif -/* The *optional* compression routines need some overhead per tape - * block for their purposes. Instead of asking the actual compression - * implementation how much it needs, we restrict this overhead to be - * maximal of ZFT_CMPT_OVERHEAD size. We need this for EOT - * conditions. The tape is assumed to be logical at EOT when the - * distance from the physical EOT is less than - * one tape block + ZFT_CMPR_OVERHEAD - */ -#define ZFT_CMPR_OVERHEAD 16 /* bytes */ - -typedef enum -{ - zft_idle = 0, - zft_reading, - zft_writing, -} zft_status_enum; - -typedef struct /* all values measured in bytes */ -{ - int seg_pos; /* segment currently positioned at */ - int seg_byte_pos; /* offset in current segment */ - __s64 tape_pos; /* real offset from BOT */ - __s64 volume_pos; /* pos. in uncompressed data stream in - * current volume - */ -} zft_position; - -extern zft_position zft_pos; -extern __u8 *zft_deblock_buf; -extern __u8 *zft_hseg_buf; -extern int zft_deblock_segment; -extern zft_status_enum zft_io_state; -extern int zft_header_changed; -extern int zft_qic113; /* conform to old specs. and old zftape */ -extern int zft_use_compression; -extern unsigned int zft_blk_sz; -extern __s64 zft_capacity; -extern unsigned int zft_written_segments; -extern int zft_label_changed; - -/* zftape-rw.c exported functions - */ -extern unsigned int zft_get_seg_sz(unsigned int segment); -extern void zft_set_flags(unsigned int minor_unit); -extern int zft_calc_seg_byte_coord(int *seg_byte_pos, __s64 tape_pos); -extern __s64 zft_calc_tape_pos(int segment); -extern __s64 zft_get_capacity(void); -extern void zft_update_label(__u8 *buffer); -extern int zft_erase(void); -extern int zft_verify_write_segments(unsigned int segment, - __u8 *data, size_t size, __u8 *buffer); -extern unsigned int zft_get_time(void); -#endif /* _ZFTAPE_RW_H */ - diff --git a/drivers/char/ftape/zftape/zftape-vtbl.c b/drivers/char/ftape/zftape/zftape-vtbl.c deleted file mode 100644 index ad7f8be6340..00000000000 --- a/drivers/char/ftape/zftape/zftape-vtbl.c +++ /dev/null @@ -1,757 +0,0 @@ -/* - * Copyright (c) 1995-1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or - modify it under the terms of the GNU General Public License as - published by the Free Software Foundation; either version 2, or (at - your option) any later version. - - This program is distributed in the hope that it will be useful, but - WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, - USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-vtbl.c,v $ - * $Revision: 1.7.6.1 $ - * $Date: 1997/11/24 13:48:31 $ - * - * This file defines a volume table as defined in various QIC - * standards. - * - * This is a minimal implementation, just allowing ordinary DOS - * :( prgrams to identify the cartridge as used. - */ - -#include -#include -#include - -#include -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-eof.h" -#include "../zftape/zftape-ctl.h" -#include "../zftape/zftape-write.h" -#include "../zftape/zftape-read.h" -#include "../zftape/zftape-rw.h" -#include "../zftape/zftape-vtbl.h" - -#define ZFT_CMAP_HACK /* leave this defined to hide the compression map */ - -/* - * global variables - */ -int zft_qic_mode = 1; /* use the vtbl */ -int zft_old_ftape; /* prevents old ftaped tapes to be overwritten */ -int zft_volume_table_changed; /* for write_header_segments() */ - -/* - * private variables (only exported for inline functions) - */ -LIST_HEAD(zft_vtbl); - -/* We could also allocate these dynamically when extracting the volume table - * sizeof(zft_volinfo) is about 32 or something close to that - */ -static zft_volinfo tape_vtbl; -static zft_volinfo eot_vtbl; -static zft_volinfo *cur_vtbl; - -static inline void zft_new_vtbl_entry(void) -{ - struct list_head *tmp = &zft_last_vtbl->node; - zft_volinfo *new = zft_kmalloc(sizeof(zft_volinfo)); - - list_add(&new->node, tmp); - new->count = zft_eom_vtbl->count ++; -} - -void zft_free_vtbl(void) -{ - for (;;) { - struct list_head *tmp = zft_vtbl.prev; - zft_volinfo *vtbl; - - if (tmp == &zft_vtbl) - break; - list_del(tmp); - vtbl = list_entry(tmp, zft_volinfo, node); - zft_kfree(vtbl, sizeof(zft_volinfo)); - } - INIT_LIST_HEAD(&zft_vtbl); - cur_vtbl = NULL; -} - -/* initialize vtbl, called by ftape_new_cartridge() - */ -void zft_init_vtbl(void) -{ - zft_volinfo *new; - - zft_free_vtbl(); - - /* Create the two dummy vtbl entries - */ - new = zft_kmalloc(sizeof(zft_volinfo)); - list_add(&new->node, &zft_vtbl); - new = zft_kmalloc(sizeof(zft_volinfo)); - list_add(&new->node, &zft_vtbl); - zft_head_vtbl->end_seg = ft_first_data_segment; - zft_head_vtbl->blk_sz = zft_blk_sz; - zft_head_vtbl->count = -1; - zft_eom_vtbl->start_seg = ft_first_data_segment + 1; - zft_eom_vtbl->end_seg = ft_last_data_segment + 1; - zft_eom_vtbl->blk_sz = zft_blk_sz; - zft_eom_vtbl->count = 0; - - /* Reset the pointer for zft_find_volume() - */ - cur_vtbl = zft_eom_vtbl; - - /* initialize the dummy vtbl entries for zft_qic_mode == 0 - */ - eot_vtbl.start_seg = ft_last_data_segment + 1; - eot_vtbl.end_seg = ft_last_data_segment + 1; - eot_vtbl.blk_sz = zft_blk_sz; - eot_vtbl.count = -1; - tape_vtbl.start_seg = ft_first_data_segment; - tape_vtbl.end_seg = ft_last_data_segment; - tape_vtbl.blk_sz = zft_blk_sz; - tape_vtbl.size = zft_capacity; - tape_vtbl.count = 0; -} - -/* check for a valid VTBL signature. - */ -static int vtbl_signature_valid(__u8 signature[4]) -{ - const char *vtbl_ids[] = VTBL_IDS; /* valid signatures */ - int j; - - for (j = 0; - (j < NR_ITEMS(vtbl_ids)) && (memcmp(signature, vtbl_ids[j], 4) != 0); - j++); - return j < NR_ITEMS(vtbl_ids); -} - -/* We used to store the block-size of the volume in the volume-label, - * using the keyword "blocksize". The blocksize written to the - * volume-label is in bytes. - * - * We use this now only for compatibility with old zftape version. We - * store the blocksize directly as binary number in the vendor - * extension part of the volume entry. - */ -static int check_volume_label(const char *label, int *blk_sz) -{ - int valid_format; - char *blocksize; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "called with \"%s\" / \"%s\"", label, ZFT_VOL_NAME); - if (strncmp(label, ZFT_VOL_NAME, strlen(ZFT_VOL_NAME)) != 0) { - *blk_sz = 1; /* smallest block size that we allow */ - valid_format = 0; - } else { - TRACE(ft_t_noise, "got old style zftape vtbl entry"); - /* get the default blocksize */ - /* use the kernel strstr() */ - blocksize= strstr(label, " blocksize "); - if (blocksize) { - blocksize += strlen(" blocksize "); - for(*blk_sz= 0; - *blocksize >= '0' && *blocksize <= '9'; - blocksize++) { - *blk_sz *= 10; - *blk_sz += *blocksize - '0'; - } - if (*blk_sz > ZFT_MAX_BLK_SZ) { - *blk_sz= 1; - valid_format= 0; - } else { - valid_format = 1; - } - } else { - *blk_sz= 1; - valid_format= 0; - } - } - TRACE_EXIT valid_format; -} - -/* check for a zftape volume - */ -static int check_volume(__u8 *entry, zft_volinfo *volume) -{ - TRACE_FUN(ft_t_flow); - - if(strncmp(&entry[VTBL_EXT+EXT_ZFTAPE_SIG], ZFTAPE_SIG, - strlen(ZFTAPE_SIG)) == 0) { - TRACE(ft_t_noise, "got new style zftape vtbl entry"); - volume->blk_sz = GET2(entry, VTBL_EXT+EXT_ZFTAPE_BLKSZ); - volume->qic113 = entry[VTBL_EXT+EXT_ZFTAPE_QIC113]; - TRACE_EXIT 1; - } else { - TRACE_EXIT check_volume_label(&entry[VTBL_DESC], &volume->blk_sz); - } -} - - -/* create zftape specific vtbl entry, the volume bounds are inserted - * in the calling function, zft_create_volume_headers() - */ -static void create_zft_volume(__u8 *entry, zft_volinfo *vtbl) -{ - TRACE_FUN(ft_t_flow); - - memset(entry, 0, VTBL_SIZE); - memcpy(&entry[VTBL_SIG], VTBL_ID, 4); - sprintf(&entry[VTBL_DESC], ZFT_VOL_NAME" %03d", vtbl->count); - entry[VTBL_FLAGS] = (VTBL_FL_NOT_VERIFIED | VTBL_FL_SEG_SPANNING); - entry[VTBL_M_NO] = 1; /* multi_cartridge_count */ - strcpy(&entry[VTBL_EXT+EXT_ZFTAPE_SIG], ZFTAPE_SIG); - PUT2(entry, VTBL_EXT+EXT_ZFTAPE_BLKSZ, vtbl->blk_sz); - if (zft_qic113) { - PUT8(entry, VTBL_DATA_SIZE, vtbl->size); - entry[VTBL_CMPR] = VTBL_CMPR_UNREG; - if (vtbl->use_compression) { /* use compression: */ - entry[VTBL_CMPR] |= VTBL_CMPR_USED; - } - entry[VTBL_EXT+EXT_ZFTAPE_QIC113] = 1; - } else { - PUT4(entry, VTBL_DATA_SIZE, vtbl->size); - entry[VTBL_K_CMPR] = VTBL_CMPR_UNREG; - if (vtbl->use_compression) { /* use compression: */ - entry[VTBL_K_CMPR] |= VTBL_CMPR_USED; - } - } - if (ft_format_code == fmt_big) { - /* SCSI like vtbl, store the number of used - * segments as 4 byte value - */ - PUT4(entry, VTBL_SCSI_SEGS, vtbl->end_seg-vtbl->start_seg + 1); - } else { - /* normal, QIC-80MC like vtbl - */ - PUT2(entry, VTBL_START, vtbl->start_seg); - PUT2(entry, VTBL_END, vtbl->end_seg); - } - TRACE_EXIT; -} - -/* this one creates the volume headers for each volume. It is assumed - * that buffer already contains the old volume-table, so that vtbl - * entries without the zft_volume flag set can savely be ignored. - */ -static void zft_create_volume_headers(__u8 *buffer) -{ - __u8 *entry; - struct list_head *tmp; - zft_volinfo *vtbl; - TRACE_FUN(ft_t_flow); - -#ifdef ZFT_CMAP_HACK - if((strncmp(&buffer[VTBL_EXT+EXT_ZFTAPE_SIG], ZFTAPE_SIG, - strlen(ZFTAPE_SIG)) == 0) && - buffer[VTBL_EXT+EXT_ZFTAPE_CMAP] != 0) { - TRACE(ft_t_noise, "deleting cmap volume"); - memmove(buffer, buffer + VTBL_SIZE, - FT_SEGMENT_SIZE - VTBL_SIZE); - } -#endif - entry = buffer; - for (tmp = zft_head_vtbl->node.next; - tmp != &zft_eom_vtbl->node; - tmp = tmp->next) { - vtbl = list_entry(tmp, zft_volinfo, node); - /* we now fill in the values only for newly created volumes. - */ - if (vtbl->new_volume) { - create_zft_volume(entry, vtbl); - vtbl->new_volume = 0; /* clear the flag */ - } - - DUMP_VOLINFO(ft_t_noise, &entry[VTBL_DESC], vtbl); - entry += VTBL_SIZE; - } - memset(entry, 0, FT_SEGMENT_SIZE - zft_eom_vtbl->count * VTBL_SIZE); - TRACE_EXIT; -} - -/* write volume table to tape. Calls zft_create_volume_headers() - */ -int zft_update_volume_table(unsigned int segment) -{ - int result = 0; - __u8 *verify_buf = NULL; - TRACE_FUN(ft_t_flow); - - TRACE_CATCH(result = ftape_read_segment(ft_first_data_segment, - zft_deblock_buf, - FT_RD_SINGLE),); - zft_create_volume_headers(zft_deblock_buf); - TRACE(ft_t_noise, "writing volume table segment %d", segment); - if (zft_vmalloc_once(&verify_buf, FT_SEGMENT_SIZE) == 0) { - TRACE_CATCH(zft_verify_write_segments(segment, - zft_deblock_buf, result, - verify_buf), - zft_vfree(&verify_buf, FT_SEGMENT_SIZE)); - zft_vfree(&verify_buf, FT_SEGMENT_SIZE); - } else { - TRACE_CATCH(ftape_write_segment(segment, zft_deblock_buf, - FT_WR_SINGLE),); - } - TRACE_EXIT 0; -} - -/* non zftape volumes are handled in raw mode. Thus we need to - * calculate the raw amount of data contained in those segments. - */ -static void extract_alien_volume(__u8 *entry, zft_volinfo *vtbl) -{ - TRACE_FUN(ft_t_flow); - - vtbl->size = (zft_calc_tape_pos(zft_last_vtbl->end_seg+1) - - zft_calc_tape_pos(zft_last_vtbl->start_seg)); - vtbl->use_compression = 0; - vtbl->qic113 = zft_qic113; - if (vtbl->qic113) { - TRACE(ft_t_noise, - "Fake alien volume's size from " LL_X " to " LL_X, - LL(GET8(entry, VTBL_DATA_SIZE)), LL(vtbl->size)); - } else { - TRACE(ft_t_noise, - "Fake alien volume's size from %d to " LL_X, - (int)GET4(entry, VTBL_DATA_SIZE), LL(vtbl->size)); - } - TRACE_EXIT; -} - - -/* extract an zftape specific volume - */ -static void extract_zft_volume(__u8 *entry, zft_volinfo *vtbl) -{ - TRACE_FUN(ft_t_flow); - - if (vtbl->qic113) { - vtbl->size = GET8(entry, VTBL_DATA_SIZE); - vtbl->use_compression = - (entry[VTBL_CMPR] & VTBL_CMPR_USED) != 0; - } else { - vtbl->size = GET4(entry, VTBL_DATA_SIZE); - if (entry[VTBL_K_CMPR] & VTBL_CMPR_UNREG) { - vtbl->use_compression = - (entry[VTBL_K_CMPR] & VTBL_CMPR_USED) != 0; - } else if (entry[VTBL_CMPR] & VTBL_CMPR_UNREG) { - vtbl->use_compression = - (entry[VTBL_CMPR] & VTBL_CMPR_USED) != 0; - } else { - TRACE(ft_t_warn, "Geeh! There is something wrong:\n" - KERN_INFO "QIC compression (Rev = K): %x\n" - KERN_INFO "QIC compression (Rev > K): %x", - entry[VTBL_K_CMPR], entry[VTBL_CMPR]); - } - } - TRACE_EXIT; -} - -/* extract the volume table from buffer. "buffer" must already contain - * the vtbl-segment - */ -int zft_extract_volume_headers(__u8 *buffer) -{ - __u8 *entry; - TRACE_FUN(ft_t_flow); - - zft_init_vtbl(); - entry = buffer; -#ifdef ZFT_CMAP_HACK - if ((strncmp(&entry[VTBL_EXT+EXT_ZFTAPE_SIG], ZFTAPE_SIG, - strlen(ZFTAPE_SIG)) == 0) && - entry[VTBL_EXT+EXT_ZFTAPE_CMAP] != 0) { - TRACE(ft_t_noise, "ignoring cmap volume"); - entry += VTBL_SIZE; - } -#endif - /* the end of the vtbl is indicated by an invalid signature - */ - while (vtbl_signature_valid(&entry[VTBL_SIG]) && - (entry - buffer) < FT_SEGMENT_SIZE) { - zft_new_vtbl_entry(); - if (ft_format_code == fmt_big) { - /* SCSI like vtbl, stores only the number of - * segments used - */ - unsigned int num_segments= GET4(entry, VTBL_SCSI_SEGS); - zft_last_vtbl->start_seg = zft_eom_vtbl->start_seg; - zft_last_vtbl->end_seg = - zft_last_vtbl->start_seg + num_segments - 1; - } else { - /* `normal', QIC-80 like vtbl - */ - zft_last_vtbl->start_seg = GET2(entry, VTBL_START); - zft_last_vtbl->end_seg = GET2(entry, VTBL_END); - } - zft_eom_vtbl->start_seg = zft_last_vtbl->end_seg + 1; - /* check if we created this volume and get the - * blk_sz - */ - zft_last_vtbl->zft_volume = check_volume(entry, zft_last_vtbl); - if (zft_last_vtbl->zft_volume == 0) { - extract_alien_volume(entry, zft_last_vtbl); - } else { - extract_zft_volume(entry, zft_last_vtbl); - } - DUMP_VOLINFO(ft_t_noise, &entry[VTBL_DESC], zft_last_vtbl); - entry +=VTBL_SIZE; - } -#if 0 -/* - * undefine to test end of tape handling - */ - zft_new_vtbl_entry(); - zft_last_vtbl->start_seg = zft_eom_vtbl->start_seg; - zft_last_vtbl->end_seg = ft_last_data_segment - 10; - zft_last_vtbl->blk_sz = zft_blk_sz; - zft_last_vtbl->zft_volume = 1; - zft_last_vtbl->qic113 = zft_qic113; - zft_last_vtbl->size = (zft_calc_tape_pos(zft_last_vtbl->end_seg+1) - - zft_calc_tape_pos(zft_last_vtbl->start_seg)); -#endif - TRACE_EXIT 0; -} - -/* this functions translates the failed_sector_log, misused as - * EOF-marker list, into a virtual volume table. The table mustn't be - * written to tape, because this would occupy the first data segment, - * which should be the volume table, but is actually the first segment - * that is filled with data (when using standard ftape). We assume, - * that we get a non-empty failed_sector_log. - */ -int zft_fake_volume_headers (eof_mark_union *eof_map, int num_failed_sectors) -{ - unsigned int segment, sector; - int have_eom = 0; - int vol_no; - TRACE_FUN(ft_t_flow); - - if ((num_failed_sectors >= 2) && - (GET2(&eof_map[num_failed_sectors - 1].mark.segment, 0) - == - GET2(&eof_map[num_failed_sectors - 2].mark.segment, 0) + 1) && - (GET2(&eof_map[num_failed_sectors - 1].mark.date, 0) == 1)) { - /* this should be eom. We keep the remainder of the - * tape as another volume. - */ - have_eom = 1; - } - zft_init_vtbl(); - zft_eom_vtbl->start_seg = ft_first_data_segment; - for(vol_no = 0; vol_no < num_failed_sectors - have_eom; vol_no ++) { - zft_new_vtbl_entry(); - - segment = GET2(&eof_map[vol_no].mark.segment, 0); - sector = GET2(&eof_map[vol_no].mark.date, 0); - - zft_last_vtbl->start_seg = zft_eom_vtbl->start_seg; - zft_last_vtbl->end_seg = segment; - zft_eom_vtbl->start_seg = segment + 1; - zft_last_vtbl->blk_sz = 1; - zft_last_vtbl->size = - (zft_calc_tape_pos(zft_last_vtbl->end_seg) - - zft_calc_tape_pos(zft_last_vtbl->start_seg) - + (sector-1) * FT_SECTOR_SIZE); - TRACE(ft_t_noise, - "failed sector log: segment: %d, sector: %d", - segment, sector); - DUMP_VOLINFO(ft_t_noise, "Faked volume", zft_last_vtbl); - } - if (!have_eom) { - zft_new_vtbl_entry(); - zft_last_vtbl->start_seg = zft_eom_vtbl->start_seg; - zft_last_vtbl->end_seg = ft_last_data_segment; - zft_eom_vtbl->start_seg = ft_last_data_segment + 1; - zft_last_vtbl->size = zft_capacity; - zft_last_vtbl->size -= zft_calc_tape_pos(zft_last_vtbl->start_seg); - zft_last_vtbl->blk_sz = 1; - DUMP_VOLINFO(ft_t_noise, "Faked volume",zft_last_vtbl); - } - TRACE_EXIT 0; -} - -/* update the internal volume table - * - * if before start of last volume: erase all following volumes if - * inside a volume: set end of volume to infinity - * - * this function is intended to be called every time _ftape_write() is - * called - * - * return: 0 if no new volume was created, 1 if a new volume was - * created - * - * NOTE: we don't need to check for zft_mode as ftape_write() does - * that already. This function gets never called without accessing - * zftape via the *qft* devices - */ - -int zft_open_volume(zft_position *pos, int blk_sz, int use_compression) -{ - TRACE_FUN(ft_t_flow); - - if (!zft_qic_mode) { - TRACE_EXIT 0; - } - if (zft_tape_at_lbot(pos)) { - zft_init_vtbl(); - if(zft_old_ftape) { - /* clear old ftape's eof marks */ - zft_clear_ftape_file_marks(); - zft_old_ftape = 0; /* no longer old ftape */ - } - zft_reset_position(pos); - } - if (pos->seg_pos != zft_last_vtbl->end_seg + 1) { - TRACE_ABORT(-EIO, ft_t_bug, - "BUG: seg_pos: %d, zft_last_vtbl->end_seg: %d", - pos->seg_pos, zft_last_vtbl->end_seg); - } - TRACE(ft_t_noise, "create new volume"); - if (zft_eom_vtbl->count >= ZFT_MAX_VOLUMES) { - TRACE_ABORT(-ENOSPC, ft_t_err, - "Error: maxmimal number of volumes exhausted " - "(maxmimum is %d)", ZFT_MAX_VOLUMES); - } - zft_new_vtbl_entry(); - pos->volume_pos = pos->seg_byte_pos = 0; - zft_last_vtbl->start_seg = pos->seg_pos; - zft_last_vtbl->end_seg = ft_last_data_segment; /* infinity */ - zft_last_vtbl->blk_sz = blk_sz; - zft_last_vtbl->size = zft_capacity; - zft_last_vtbl->zft_volume = 1; - zft_last_vtbl->use_compression = use_compression; - zft_last_vtbl->qic113 = zft_qic113; - zft_last_vtbl->new_volume = 1; - zft_last_vtbl->open = 1; - zft_volume_table_changed = 1; - zft_eom_vtbl->start_seg = ft_last_data_segment + 1; - TRACE_EXIT 0; -} - -/* perform mtfsf, mtbsf, not allowed without zft_qic_mode - */ -int zft_skip_volumes(int count, zft_position *pos) -{ - const zft_volinfo *vtbl; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "count: %d", count); - - vtbl= zft_find_volume(pos->seg_pos); - while (count > 0 && vtbl != zft_eom_vtbl) { - vtbl = list_entry(vtbl->node.next, zft_volinfo, node); - count --; - } - while (count < 0 && vtbl != zft_first_vtbl) { - vtbl = list_entry(vtbl->node.prev, zft_volinfo, node); - count ++; - } - pos->seg_pos = vtbl->start_seg; - pos->seg_byte_pos = 0; - pos->volume_pos = 0; - pos->tape_pos = zft_calc_tape_pos(pos->seg_pos); - zft_just_before_eof = vtbl->size == 0; - if (zft_cmpr_ops) { - (*zft_cmpr_ops->reset)(); - } - zft_deblock_segment = -1; /* no need to keep cache */ - TRACE(ft_t_noise, "repositioning to:\n" - KERN_INFO "zft_seg_pos : %d\n" - KERN_INFO "zft_seg_byte_pos : %d\n" - KERN_INFO "zft_tape_pos : " LL_X "\n" - KERN_INFO "zft_volume_pos : " LL_X "\n" - KERN_INFO "file number : %d", - pos->seg_pos, pos->seg_byte_pos, - LL(pos->tape_pos), LL(pos->volume_pos), vtbl->count); - zft_resid = count < 0 ? -count : count; - TRACE_EXIT zft_resid ? -EINVAL : 0; -} - -/* the following simply returns the raw data position of the EOM - * marker, MTIOCSIZE ioctl - */ -__s64 zft_get_eom_pos(void) -{ - if (zft_qic_mode) { - return zft_calc_tape_pos(zft_eom_vtbl->start_seg); - } else { - /* there is only one volume in raw mode */ - return zft_capacity; - } -} - -/* skip to eom, used for MTEOM - */ -void zft_skip_to_eom(zft_position *pos) -{ - TRACE_FUN(ft_t_flow); - pos->seg_pos = zft_eom_vtbl->start_seg; - pos->seg_byte_pos = - pos->volume_pos = - zft_just_before_eof = 0; - pos->tape_pos = zft_calc_tape_pos(pos->seg_pos); - TRACE(ft_t_noise, "ftape positioned to segment %d, data pos " LL_X, - pos->seg_pos, LL(pos->tape_pos)); - TRACE_EXIT; -} - -/* write an EOF-marker by setting zft_last_vtbl->end_seg to seg_pos. - * NOTE: this function assumes that zft_last_vtbl points to a valid - * vtbl entry - * - * NOTE: this routine always positions before the EOF marker - */ -int zft_close_volume(zft_position *pos) -{ - TRACE_FUN(ft_t_any); - - if (zft_vtbl_empty || !zft_last_vtbl->open) { /* should not happen */ - TRACE(ft_t_noise, "There are no volumes to finish"); - TRACE_EXIT -EIO; - } - if (pos->seg_byte_pos == 0 && - pos->seg_pos != zft_last_vtbl->start_seg) { - pos->seg_pos --; - pos->seg_byte_pos = zft_get_seg_sz(pos->seg_pos); - } - zft_last_vtbl->end_seg = pos->seg_pos; - zft_last_vtbl->size = pos->volume_pos; - zft_volume_table_changed = 1; - zft_just_before_eof = 1; - zft_eom_vtbl->start_seg = zft_last_vtbl->end_seg + 1; - zft_last_vtbl->open = 0; /* closed */ - TRACE_EXIT 0; -} - -/* write count file-marks at current position. - * - * The tape is positioned after the eof-marker, that is at byte 0 of - * the segment following the eof-marker - * - * this function is only allowed in zft_qic_mode - * - * Only allowed when tape is at BOT or EOD. - */ -int zft_weof(unsigned int count, zft_position *pos) -{ - - TRACE_FUN(ft_t_flow); - - if (!count) { /* write zero EOF marks should be a real no-op */ - TRACE_EXIT 0; - } - zft_volume_table_changed = 1; - if (zft_tape_at_lbot(pos)) { - zft_init_vtbl(); - if(zft_old_ftape) { - /* clear old ftape's eof marks */ - zft_clear_ftape_file_marks(); - zft_old_ftape = 0; /* no longer old ftape */ - } - } - if (zft_last_vtbl->open) { - zft_close_volume(pos); - zft_move_past_eof(pos); - count --; - } - /* now it's easy, just append eof-marks, that is empty - * volumes, to the end of the already recorded media. - */ - while (count > 0 && - pos->seg_pos <= ft_last_data_segment && - zft_eom_vtbl->count < ZFT_MAX_VOLUMES) { - TRACE(ft_t_noise, - "Writing zero sized file at segment %d", pos->seg_pos); - zft_new_vtbl_entry(); - zft_last_vtbl->start_seg = pos->seg_pos; - zft_last_vtbl->end_seg = pos->seg_pos; - zft_last_vtbl->size = 0; - zft_last_vtbl->blk_sz = zft_blk_sz; - zft_last_vtbl->zft_volume = 1; - zft_last_vtbl->use_compression = 0; - pos->tape_pos += zft_get_seg_sz(pos->seg_pos); - zft_eom_vtbl->start_seg = ++ pos->seg_pos; - count --; - } - if (count > 0) { - /* there are two possibilities: end of tape, or the - * maximum number of files is exhausted. - */ - zft_resid = count; - TRACE(ft_t_noise,"Number of marks NOT written: %d", zft_resid); - if (zft_eom_vtbl->count == ZFT_MAX_VOLUMES) { - TRACE_ABORT(-EINVAL, ft_t_warn, - "maximum allowed number of files " - "exhausted: %d", ZFT_MAX_VOLUMES); - } else { - TRACE_ABORT(-ENOSPC, - ft_t_noise, "reached end of tape"); - } - } - TRACE_EXIT 0; -} - -const zft_volinfo *zft_find_volume(unsigned int seg_pos) -{ - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_any, "called with seg_pos %d",seg_pos); - if (!zft_qic_mode) { - if (seg_pos > ft_last_data_segment) { - TRACE_EXIT &eot_vtbl; - } - tape_vtbl.blk_sz = zft_blk_sz; - TRACE_EXIT &tape_vtbl; - } - if (seg_pos < zft_first_vtbl->start_seg) { - TRACE_EXIT (cur_vtbl = zft_first_vtbl); - } - while (seg_pos > cur_vtbl->end_seg) { - cur_vtbl = list_entry(cur_vtbl->node.next, zft_volinfo, node); - TRACE(ft_t_noise, "%d - %d", cur_vtbl->start_seg, cur_vtbl->end_seg); - } - while (seg_pos < cur_vtbl->start_seg) { - cur_vtbl = list_entry(cur_vtbl->node.prev, zft_volinfo, node); - TRACE(ft_t_noise, "%d - %d", cur_vtbl->start_seg, cur_vtbl->end_seg); - } - if (seg_pos > cur_vtbl->end_seg || seg_pos < cur_vtbl->start_seg) { - TRACE(ft_t_bug, "This cannot happen"); - } - DUMP_VOLINFO(ft_t_noise, "", cur_vtbl); - TRACE_EXIT cur_vtbl; -} - -/* this function really assumes that we are just before eof - */ -void zft_move_past_eof(zft_position *pos) -{ - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "old seg. pos: %d", pos->seg_pos); - pos->tape_pos += zft_get_seg_sz(pos->seg_pos++) - pos->seg_byte_pos; - pos->seg_byte_pos = 0; - pos->volume_pos = 0; - if (zft_cmpr_ops) { - (*zft_cmpr_ops->reset)(); - } - zft_just_before_eof = 0; - zft_deblock_segment = -1; /* no need to cache it anymore */ - TRACE(ft_t_noise, "new seg. pos: %d", pos->seg_pos); - TRACE_EXIT; -} diff --git a/drivers/char/ftape/zftape/zftape-vtbl.h b/drivers/char/ftape/zftape/zftape-vtbl.h deleted file mode 100644 index f31d196d175..00000000000 --- a/drivers/char/ftape/zftape/zftape-vtbl.h +++ /dev/null @@ -1,227 +0,0 @@ -#ifndef _ZFTAPE_VTBL_H -#define _ZFTAPE_VTBL_H - -/* - * Copyright (c) 1995-1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or - modify it under the terms of the GNU General Public License as - published by the Free Software Foundation; either version 2, or (at - your option) any later version. - - This program is distributed in the hope that it will be useful, but - WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, - USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-vtbl.h,v $ - * $Revision: 1.3 $ - * $Date: 1997/10/28 14:30:09 $ - * - * This file defines a volume table as defined in the QIC-80 - * development standards. - */ - -#include - -#include "../lowlevel/ftape-tracing.h" - -#include "../zftape/zftape-eof.h" -#include "../zftape/zftape-ctl.h" -#include "../zftape/zftape-rw.h" - -#define VTBL_SIZE 128 /* bytes */ - -/* The following are offsets in the vtbl. */ -#define VTBL_SIG 0 -#define VTBL_START 4 -#define VTBL_END 6 -#define VTBL_DESC 8 -#define VTBL_DATE 52 -#define VTBL_FLAGS 56 -#define VTBL_FL_VENDOR_SPECIFIC (1<<0) -#define VTBL_FL_MUTLI_CARTRIDGE (1<<1) -#define VTBL_FL_NOT_VERIFIED (1<<2) -#define VTBL_FL_REDIR_INHIBIT (1<<3) -#define VTBL_FL_SEG_SPANNING (1<<4) -#define VTBL_FL_DIRECTORY_LAST (1<<5) -#define VTBL_FL_RESERVED_6 (1<<6) -#define VTBL_FL_RESERVED_7 (1<<7) -#define VTBL_M_NO 57 -#define VTBL_EXT 58 -#define EXT_ZFTAPE_SIG 0 -#define EXT_ZFTAPE_BLKSZ 10 -#define EXT_ZFTAPE_CMAP 12 -#define EXT_ZFTAPE_QIC113 13 -#define VTBL_PWD 84 -#define VTBL_DIR_SIZE 92 -#define VTBL_DATA_SIZE 96 -#define VTBL_OS_VERSION 104 -#define VTBL_SRC_DRIVE 106 -#define VTBL_DEV 122 -#define VTBL_RESERVED_1 123 -#define VTBL_CMPR 124 -#define VTBL_CMPR_UNREG 0x3f -#define VTBL_CMPR_USED 0x80 -#define VTBL_FMT 125 -#define VTBL_RESERVED_2 126 -#define VTBL_RESERVED_3 127 -/* compatibility with pre revision K */ -#define VTBL_K_CMPR 120 - -/* the next is used by QIC-3020 tapes with format code 6 (>2^16 - * segments) It is specified in QIC-113, Rev. G, Section 5 (SCSI - * volume table). The difference is simply, that we only store the - * number of segments used, not the starting segment. - */ -#define VTBL_SCSI_SEGS 4 /* is a 4 byte value */ - -/* one vtbl is 128 bytes, that results in a maximum number of - * 29*1024/128 = 232 volumes. - */ -#define ZFT_MAX_VOLUMES (FT_SEGMENT_SIZE/VTBL_SIZE) -#define VTBL_ID "VTBL" -#define VTBL_IDS { VTBL_ID, "XTBL", "UTID", "EXVT" } /* other valid ids */ -#define ZFT_VOL_NAME "zftape volume" /* volume label used by me */ -#define ZFTAPE_SIG "LINUX ZFT" - -/* global variables - */ -typedef struct zft_internal_vtbl -{ - struct list_head node; - int count; - unsigned int start_seg; /* 32 bits are enough for now */ - unsigned int end_seg; /* 32 bits are enough for now */ - __s64 size; /* uncompressed size */ - unsigned int blk_sz; /* block size for this volume */ - unsigned int zft_volume :1; /* zftape created this volume */ - unsigned int use_compression:1; /* compressed volume */ - unsigned int qic113 :1; /* layout of compressed block - * info and vtbl conforms to - * QIC-113, Rev. G - */ - unsigned int new_volume :1; /* it was created by us, this - * run. this allows the - * fields that aren't really - * used by zftape to be filled - * in by some user level - * program. - */ - unsigned int open :1; /* just in progress of being - * written - */ -} zft_volinfo; - -extern struct list_head zft_vtbl; -#define zft_head_vtbl list_entry(zft_vtbl.next, zft_volinfo, node) -#define zft_eom_vtbl list_entry(zft_vtbl.prev, zft_volinfo, node) -#define zft_last_vtbl list_entry(zft_eom_vtbl->node.prev, zft_volinfo, node) -#define zft_first_vtbl list_entry(zft_head_vtbl->node.next, zft_volinfo, node) -#define zft_vtbl_empty (zft_eom_vtbl->node.prev == &zft_head_vtbl->node) - -#define DUMP_VOLINFO(level, desc, info) \ -{ \ - char tmp[21]; \ - strlcpy(tmp, desc, sizeof(tmp)); \ - TRACE(level, "Volume %d:\n" \ - KERN_INFO "description : %s\n" \ - KERN_INFO "first segment: %d\n" \ - KERN_INFO "last segment: %d\n" \ - KERN_INFO "size : " LL_X "\n" \ - KERN_INFO "block size : %d\n" \ - KERN_INFO "compression : %d\n" \ - KERN_INFO "zftape volume: %d\n" \ - KERN_INFO "QIC-113 conf.: %d", \ - (info)->count, tmp, (info)->start_seg, (info)->end_seg, \ - LL((info)->size), (info)->blk_sz, \ - (info)->use_compression != 0, (info)->zft_volume != 0, \ - (info)->qic113 != 0); \ -} - -extern int zft_qic_mode; -extern int zft_old_ftape; -extern int zft_volume_table_changed; - -/* exported functions */ -extern void zft_init_vtbl (void); -extern void zft_free_vtbl (void); -extern int zft_extract_volume_headers(__u8 *buffer); -extern int zft_update_volume_table (unsigned int segment); -extern int zft_open_volume (zft_position *pos, - int blk_sz, int use_compression); -extern int zft_close_volume (zft_position *pos); -extern const zft_volinfo *zft_find_volume(unsigned int seg_pos); -extern int zft_skip_volumes (int count, zft_position *pos); -extern __s64 zft_get_eom_pos (void); -extern void zft_skip_to_eom (zft_position *pos); -extern int zft_fake_volume_headers (eof_mark_union *eof_map, - int num_failed_sectors); -extern int zft_weof (unsigned int count, zft_position *pos); -extern void zft_move_past_eof (zft_position *pos); - -static inline int zft_tape_at_eod (const zft_position *pos); -static inline int zft_tape_at_lbot (const zft_position *pos); -static inline void zft_position_before_eof (zft_position *pos, - const zft_volinfo *volume); -static inline __s64 zft_check_for_eof(const zft_volinfo *vtbl, - const zft_position *pos); - -/* this function decrements the zft_seg_pos counter if we are right - * at the beginning of a segment. This is to handle fsfm/bsfm -- we - * need to position before the eof mark. NOTE: zft_tape_pos is not - * changed - */ -static inline void zft_position_before_eof(zft_position *pos, - const zft_volinfo *volume) -{ - TRACE_FUN(ft_t_flow); - - if (pos->seg_pos == volume->end_seg + 1 && pos->seg_byte_pos == 0) { - pos->seg_pos --; - pos->seg_byte_pos = zft_get_seg_sz(pos->seg_pos); - } - TRACE_EXIT; -} - -/* Mmmh. Is the position at the end of the last volume, that is right - * before the last EOF mark also logical an EOD condition? - */ -static inline int zft_tape_at_eod(const zft_position *pos) -{ - TRACE_FUN(ft_t_any); - - if (zft_qic_mode) { - TRACE_EXIT (pos->seg_pos >= zft_eom_vtbl->start_seg || - zft_last_vtbl->open); - } else { - TRACE_EXIT pos->seg_pos > ft_last_data_segment; - } -} - -static inline int zft_tape_at_lbot(const zft_position *pos) -{ - if (zft_qic_mode) { - return (pos->seg_pos <= zft_first_vtbl->start_seg && - pos->volume_pos == 0); - } else { - return (pos->seg_pos <= ft_first_data_segment && - pos->volume_pos == 0); - } -} - -/* This one checks for EOF. return remaing space (may be negative) - */ -static inline __s64 zft_check_for_eof(const zft_volinfo *vtbl, - const zft_position *pos) -{ - return (__s64)(vtbl->size - pos->volume_pos); -} - -#endif /* _ZFTAPE_VTBL_H */ diff --git a/drivers/char/ftape/zftape/zftape-write.c b/drivers/char/ftape/zftape/zftape-write.c deleted file mode 100644 index 94327b8c97b..00000000000 --- a/drivers/char/ftape/zftape/zftape-write.c +++ /dev/null @@ -1,483 +0,0 @@ -/* - * Copyright (C) 1996, 1997 Claus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-write.c,v $ - * $Revision: 1.3 $ - * $Date: 1997/11/06 00:50:29 $ - * - * This file contains the writing code - * for the QIC-117 floppy-tape driver for Linux. - */ - -#include -#include - -#include - -#include - -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-eof.h" -#include "../zftape/zftape-ctl.h" -#include "../zftape/zftape-write.h" -#include "../zftape/zftape-read.h" -#include "../zftape/zftape-rw.h" -#include "../zftape/zftape-vtbl.h" - -/* Global vars. - */ - -/* Local vars. - */ -static int last_write_failed; -static int need_flush; - -void zft_prevent_flush(void) -{ - need_flush = 0; -} - -static int zft_write_header_segments(__u8* buffer) -{ - int header_1_ok = 0; - int header_2_ok = 0; - unsigned int time_stamp; - TRACE_FUN(ft_t_noise); - - TRACE_CATCH(ftape_abort_operation(),); - ftape_seek_to_bot(); /* prevents extra rewind */ - if (GET4(buffer, 0) != FT_HSEG_MAGIC) { - TRACE_ABORT(-EIO, ft_t_err, - "wrong header signature found, aborting"); - } - /* Be optimistic: */ - PUT4(buffer, FT_SEG_CNT, - zft_written_segments + GET4(buffer, FT_SEG_CNT) + 2); - if ((time_stamp = zft_get_time()) != 0) { - PUT4(buffer, FT_WR_DATE, time_stamp); - if (zft_label_changed) { - PUT4(buffer, FT_LABEL_DATE, time_stamp); - } - } - TRACE(ft_t_noise, - "writing first header segment %d", ft_header_segment_1); - header_1_ok = zft_verify_write_segments(ft_header_segment_1, - buffer, FT_SEGMENT_SIZE, - zft_deblock_buf) >= 0; - TRACE(ft_t_noise, - "writing second header segment %d", ft_header_segment_2); - header_2_ok = zft_verify_write_segments(ft_header_segment_2, - buffer, FT_SEGMENT_SIZE, - zft_deblock_buf) >= 0; - if (!header_1_ok) { - TRACE(ft_t_warn, "Warning: " - "update of first header segment failed"); - } - if (!header_2_ok) { - TRACE(ft_t_warn, "Warning: " - "update of second header segment failed"); - } - if (!header_1_ok && !header_2_ok) { - TRACE_ABORT(-EIO, ft_t_err, "Error: " - "update of both header segments failed."); - } - TRACE_EXIT 0; -} - -int zft_update_header_segments(void) -{ - TRACE_FUN(ft_t_noise); - - /* must NOT use zft_write_protected, as it also includes the - * file access mode. But we also want to update when soft - * write protection is enabled (O_RDONLY) - */ - if (ft_write_protected || zft_old_ftape) { - TRACE_ABORT(0, ft_t_noise, "Tape set read-only: no update"); - } - if (!zft_header_read) { - TRACE_ABORT(0, ft_t_noise, "Nothing to update"); - } - if (!zft_header_changed) { - zft_header_changed = zft_written_segments > 0; - } - if (!zft_header_changed && !zft_volume_table_changed) { - TRACE_ABORT(0, ft_t_noise, "Nothing to update"); - } - TRACE(ft_t_noise, "Updating header segments"); - if (ftape_get_status()->fti_state == writing) { - TRACE_CATCH(ftape_loop_until_writes_done(),); - } - TRACE_CATCH(ftape_abort_operation(),); - - zft_deblock_segment = -1; /* invalidate the cache */ - if (zft_header_changed) { - TRACE_CATCH(zft_write_header_segments(zft_hseg_buf),); - } - if (zft_volume_table_changed) { - TRACE_CATCH(zft_update_volume_table(ft_first_data_segment),); - } - zft_header_changed = - zft_volume_table_changed = - zft_label_changed = - zft_written_segments = 0; - TRACE_CATCH(ftape_abort_operation(),); - ftape_seek_to_bot(); - TRACE_EXIT 0; -} - -static int read_merge_buffer(int seg_pos, __u8 *buffer, int offset, int seg_sz) -{ - int result = 0; - const ft_trace_t old_tracing = TRACE_LEVEL; - TRACE_FUN(ft_t_flow); - - if (zft_qic_mode) { - /* writing in the middle of a volume is NOT allowed - * - */ - TRACE(ft_t_noise, "No need to read a segment"); - memset(buffer + offset, 0, seg_sz - offset); - TRACE_EXIT 0; - } - TRACE(ft_t_any, "waiting"); - ftape_start_writing(FT_WR_MULTI); - TRACE_CATCH(ftape_loop_until_writes_done(),); - - TRACE(ft_t_noise, "trying to read segment %d from offset %d", - seg_pos, offset); - SET_TRACE_LEVEL(ft_t_bug); - result = zft_fetch_segment_fraction(seg_pos, buffer, - FT_RD_SINGLE, - offset, seg_sz - offset); - SET_TRACE_LEVEL(old_tracing); - if (result != (seg_sz - offset)) { - TRACE(ft_t_noise, "Ignore error: read_segment() result: %d", - result); - memset(buffer + offset, 0, seg_sz - offset); - } - TRACE_EXIT 0; -} - -/* flush the write buffer to tape and write an eof-marker at the - * current position if not in raw mode. This function always - * positions the tape before the eof-marker. _ftape_close() should - * then advance to the next segment. - * - * the parameter "finish_volume" describes whether to position before - * or after the possibly created file-mark. We always position after - * the file-mark when called from ftape_close() and a flush was needed - * (that is ftape_write() was the last tape operation before calling - * ftape_flush) But we always position before the file-mark when this - * function get's called from outside ftape_close() - */ -int zft_flush_buffers(void) -{ - int result; - int data_remaining; - int this_segs_size; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_data_flow, - "entered, ftape_state = %d", ftape_get_status()->fti_state); - if (ftape_get_status()->fti_state != writing && !need_flush) { - TRACE_ABORT(0, ft_t_noise, "no need for flush"); - } - zft_io_state = zft_idle; /* triggers some initializations for the - * read and write routines - */ - if (last_write_failed) { - ftape_abort_operation(); - TRACE_EXIT -EIO; - } - TRACE(ft_t_noise, "flushing write buffers"); - this_segs_size = zft_get_seg_sz(zft_pos.seg_pos); - if (this_segs_size == zft_pos.seg_byte_pos) { - zft_pos.seg_pos ++; - data_remaining = zft_pos.seg_byte_pos = 0; - } else { - data_remaining = zft_pos.seg_byte_pos; - } - /* If there is any data not written to tape yet, append zero's - * up to the end of the sector (if using compression) or merge - * it with the data existing on the tape Then write the - * segment(s) to tape. - */ - TRACE(ft_t_noise, "Position:\n" - KERN_INFO "seg_pos : %d\n" - KERN_INFO "byte pos : %d\n" - KERN_INFO "remaining: %d", - zft_pos.seg_pos, zft_pos.seg_byte_pos, data_remaining); - if (data_remaining > 0) { - do { - this_segs_size = zft_get_seg_sz(zft_pos.seg_pos); - if (this_segs_size > data_remaining) { - TRACE_CATCH(read_merge_buffer(zft_pos.seg_pos, - zft_deblock_buf, - data_remaining, - this_segs_size), - last_write_failed = 1); - } - result = ftape_write_segment(zft_pos.seg_pos, - zft_deblock_buf, - FT_WR_MULTI); - if (result != this_segs_size) { - TRACE(ft_t_err, "flush buffers failed"); - zft_pos.tape_pos -= zft_pos.seg_byte_pos; - zft_pos.seg_byte_pos = 0; - - last_write_failed = 1; - TRACE_EXIT result; - } - zft_written_segments ++; - TRACE(ft_t_data_flow, - "flush, moved out buffer: %d", result); - /* need next segment for more data (empty segments?) - */ - if (result < data_remaining) { - if (result > 0) { - /* move remainder to buffer beginning - */ - memmove(zft_deblock_buf, - zft_deblock_buf + result, - FT_SEGMENT_SIZE - result); - } - } - data_remaining -= result; - zft_pos.seg_pos ++; - } while (data_remaining > 0); - TRACE(ft_t_any, "result: %d", result); - zft_deblock_segment = --zft_pos.seg_pos; - if (data_remaining == 0) { /* first byte next segment */ - zft_pos.seg_byte_pos = this_segs_size; - } else { /* after data previous segment, data_remaining < 0 */ - zft_pos.seg_byte_pos = data_remaining + result; - } - } else { - TRACE(ft_t_noise, "zft_deblock_buf empty"); - zft_pos.seg_pos --; - zft_pos.seg_byte_pos = zft_get_seg_sz (zft_pos.seg_pos); - ftape_start_writing(FT_WR_MULTI); - } - TRACE(ft_t_any, "waiting"); - if ((result = ftape_loop_until_writes_done()) < 0) { - /* that's really bad. What to to with zft_tape_pos? - */ - TRACE(ft_t_err, "flush buffers failed"); - } - TRACE(ft_t_any, "zft_seg_pos: %d, zft_seg_byte_pos: %d", - zft_pos.seg_pos, zft_pos.seg_byte_pos); - last_write_failed = - need_flush = 0; - TRACE_EXIT result; -} - -/* return-value: the number of bytes removed from the user-buffer - * - * out: - * int *write_cnt: how much actually has been moved to the - * zft_deblock_buf - * int req_len : MUST NOT BE CHANGED, except at EOT, in - * which case it may be adjusted - * in : - * char *buff : the user buffer - * int buf_pos_write : copy of buf_len_wr int - * this_segs_size : the size in bytes of the actual segment - * char - * *zft_deblock_buf : zft_deblock_buf - */ -static int zft_simple_write(int *cnt, - __u8 *dst_buf, const int seg_sz, - const __u8 __user *src_buf, const int req_len, - const zft_position *pos,const zft_volinfo *volume) -{ - int space_left; - TRACE_FUN(ft_t_flow); - - /* volume->size holds the tape capacity while volume is open */ - if (pos->tape_pos + volume->blk_sz > volume->size) { - TRACE_EXIT -ENOSPC; - } - /* remaining space in this segment, NOT zft_deblock_buf - */ - space_left = seg_sz - pos->seg_byte_pos; - *cnt = req_len < space_left ? req_len : space_left; - if (copy_from_user(dst_buf + pos->seg_byte_pos, src_buf, *cnt) != 0) { - TRACE_EXIT -EFAULT; - } - TRACE_EXIT *cnt; -} - -static int check_write_access(int req_len, - const zft_volinfo **volume, - zft_position *pos, - const unsigned int blk_sz) -{ - int result; - TRACE_FUN(ft_t_flow); - - if ((req_len % zft_blk_sz) != 0) { - TRACE_ABORT(-EINVAL, ft_t_info, - "write-count %d must be multiple of block-size %d", - req_len, blk_sz); - } - if (zft_io_state == zft_writing) { - /* all other error conditions have been checked earlier - */ - TRACE_EXIT 0; - } - zft_io_state = zft_idle; - TRACE_CATCH(zft_check_write_access(pos),); - /* If we haven't read the header segment yet, do it now. - * This will verify the configuration, get the bad sector - * table and read the volume table segment - */ - if (!zft_header_read) { - TRACE_CATCH(zft_read_header_segments(),); - } - /* fine. Now the tape is either at BOT or at EOD, - * Write start of volume now - */ - TRACE_CATCH(zft_open_volume(pos, blk_sz, zft_use_compression),); - *volume = zft_find_volume(pos->seg_pos); - DUMP_VOLINFO(ft_t_noise, "", *volume); - zft_just_before_eof = 0; - /* now merge with old data if necessary */ - if (!zft_qic_mode && pos->seg_byte_pos != 0){ - result = zft_fetch_segment(pos->seg_pos, - zft_deblock_buf, - FT_RD_SINGLE); - if (result < 0) { - if (result == -EINTR || result == -ENOSPC) { - TRACE_EXIT result; - } - TRACE(ft_t_noise, - "ftape_read_segment() result: %d. " - "This might be normal when using " - "a newly\nformatted tape", result); - memset(zft_deblock_buf, '\0', pos->seg_byte_pos); - } - } - zft_io_state = zft_writing; - TRACE_EXIT 0; -} - -static int fill_deblock_buf(__u8 *dst_buf, const int seg_sz, - zft_position *pos, const zft_volinfo *volume, - const char __user *usr_buf, const int req_len) -{ - int cnt = 0; - int result = 0; - TRACE_FUN(ft_t_flow); - - if (seg_sz == 0) { - TRACE_ABORT(0, ft_t_data_flow, "empty segment"); - } - TRACE(ft_t_data_flow, "\n" - KERN_INFO "remaining req_len: %d\n" - KERN_INFO " buf_pos: %d", - req_len, pos->seg_byte_pos); - /* zft_deblock_buf will not contain a valid segment any longer */ - zft_deblock_segment = -1; - if (zft_use_compression) { - TRACE_CATCH(zft_cmpr_lock(1 /* try to load */),); - TRACE_CATCH(result= (*zft_cmpr_ops->write)(&cnt, - dst_buf, seg_sz, - usr_buf, req_len, - pos, volume),); - } else { - TRACE_CATCH(result= zft_simple_write(&cnt, - dst_buf, seg_sz, - usr_buf, req_len, - pos, volume),); - } - pos->volume_pos += result; - pos->seg_byte_pos += cnt; - pos->tape_pos += cnt; - TRACE(ft_t_data_flow, "\n" - KERN_INFO "removed from user-buffer : %d bytes.\n" - KERN_INFO "copied to zft_deblock_buf: %d bytes.\n" - KERN_INFO "zft_tape_pos : " LL_X " bytes.", - result, cnt, LL(pos->tape_pos)); - TRACE_EXIT result; -} - - -/* called by the kernel-interface routine "zft_write()" - */ -int _zft_write(const char __user *buff, int req_len) -{ - int result = 0; - int written = 0; - int write_cnt; - int seg_sz; - static const zft_volinfo *volume = NULL; - TRACE_FUN(ft_t_flow); - - zft_resid = req_len; - last_write_failed = 1; /* reset to 0 when successful */ - /* check if write is allowed - */ - TRACE_CATCH(check_write_access(req_len, &volume,&zft_pos,zft_blk_sz),); - while (req_len > 0) { - /* Allow us to escape from this loop with a signal ! - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - seg_sz = zft_get_seg_sz(zft_pos.seg_pos); - if ((write_cnt = fill_deblock_buf(zft_deblock_buf, - seg_sz, - &zft_pos, - volume, - buff, - req_len)) < 0) { - zft_resid -= written; - if (write_cnt == -ENOSPC) { - /* leave the remainder to flush_buffers() - */ - TRACE(ft_t_info, "No space left on device"); - last_write_failed = 0; - if (!need_flush) { - need_flush = written > 0; - } - TRACE_EXIT written > 0 ? written : -ENOSPC; - } else { - TRACE_EXIT result; - } - } - if (zft_pos.seg_byte_pos == seg_sz) { - TRACE_CATCH(ftape_write_segment(zft_pos.seg_pos, - zft_deblock_buf, - FT_WR_ASYNC), - zft_resid -= written); - zft_written_segments ++; - zft_pos.seg_byte_pos = 0; - zft_deblock_segment = zft_pos.seg_pos; - ++zft_pos.seg_pos; - } - written += write_cnt; - buff += write_cnt; - req_len -= write_cnt; - } /* while (req_len > 0) */ - TRACE(ft_t_data_flow, "remaining in blocking buffer: %d", - zft_pos.seg_byte_pos); - TRACE(ft_t_data_flow, "just written bytes: %d", written); - last_write_failed = 0; - zft_resid -= written; - need_flush = need_flush || written > 0; - TRACE_EXIT written; /* bytes written */ -} diff --git a/drivers/char/ftape/zftape/zftape-write.h b/drivers/char/ftape/zftape/zftape-write.h deleted file mode 100644 index ea887015b49..00000000000 --- a/drivers/char/ftape/zftape/zftape-write.h +++ /dev/null @@ -1,38 +0,0 @@ -#ifndef _ZFTAPE_WRITE_H -#define _ZFTAPE_WRITE_H - -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-write.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:13 $ - * - * This file contains the definitions for the write functions - * for the zftape driver for Linux. - * - */ - -extern int zft_flush_buffers(void); -extern int zft_update_header_segments(void); -extern void zft_prevent_flush(void); - -/* hook for the VFS interface - */ -extern int _zft_write(const char __user *buff, int req_len); -#endif /* _ZFTAPE_WRITE_H */ diff --git a/drivers/char/ftape/zftape/zftape_syms.c b/drivers/char/ftape/zftape/zftape_syms.c deleted file mode 100644 index 2db1401682d..00000000000 --- a/drivers/char/ftape/zftape/zftape_syms.c +++ /dev/null @@ -1,43 +0,0 @@ -/* - * Copyright (C) 1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape_syms.c,v $ - * $Revision: 1.3 $ - * $Date: 1997/10/05 19:19:14 $ - * - * This file contains the symbols that the zftape frontend to - * the ftape floppy tape driver exports - */ - -#include - -#include - -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-read.h" -#include "../zftape/zftape-buffers.h" -#include "../zftape/zftape-ctl.h" - -/* zftape-init.c */ -EXPORT_SYMBOL(zft_cmpr_register); -/* zftape-read.c */ -EXPORT_SYMBOL(zft_fetch_segment_fraction); -/* zftape-buffers.c */ -EXPORT_SYMBOL(zft_vmalloc_once); -EXPORT_SYMBOL(zft_vmalloc_always); -EXPORT_SYMBOL(zft_vfree); diff --git a/include/linux/Kbuild b/include/linux/Kbuild index d7e04689304..ff433126361 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -60,8 +60,6 @@ header-y += fadvise.h header-y += fd.h header-y += fdreg.h header-y += fib_rules.h -header-y += ftape-header-segment.h -header-y += ftape-vendors.h header-y += fuse.h header-y += futex.h header-y += genetlink.h @@ -206,7 +204,6 @@ unifdef-y += fcntl.h unifdef-y += filter.h unifdef-y += flat.h unifdef-y += fs.h -unifdef-y += ftape.h unifdef-y += gameport.h unifdef-y += generic_serial.h unifdef-y += genhd.h @@ -341,6 +338,5 @@ unifdef-y += wait.h unifdef-y += wanrouter.h unifdef-y += watchdog.h unifdef-y += xfrm.h -unifdef-y += zftape.h objhdr-y += version.h diff --git a/include/linux/ftape-header-segment.h b/include/linux/ftape-header-segment.h deleted file mode 100644 index 4732218f070..00000000000 --- a/include/linux/ftape-header-segment.h +++ /dev/null @@ -1,122 +0,0 @@ -#ifndef _FTAPE_HEADER_SEGMENT_H -#define _FTAPE_HEADER_SEGMENT_H - -/* - * Copyright (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/include/linux/ftape-header-segment.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:28 $ - * - * This file defines some offsets into the header segment of a - * floppy tape cartridge. For use with the QIC-40/80/3010/3020 - * floppy-tape driver "ftape" for Linux. - */ - -#define FT_SIGNATURE 0 /* must be 0xaa55aa55 */ -#define FT_FMT_CODE 4 -#define FT_REV_LEVEL 5 /* only for QIC-80 since. Rev. L (== 0x0c) */ -#define FT_HSEG_1 6 /* first header segment, except for format code 6 */ -#define FT_HSEG_2 8 /* second header segment, except for format code 6 */ -#define FT_FRST_SEG 10 /* first data segment, except for format code 6 */ -#define FT_LAST_SEG 12 /* last data segment, except for format code 6 */ -#define FT_FMT_DATE 14 /* date and time of most recent format, see below */ -#define FT_WR_DATE 18 /* date and time of most recent write or format */ -#define FT_SPT 24 /* segments per track */ -#define FT_TPC 26 /* tracks per cartridge */ -#define FT_FHM 27 /* floppy drive head (maximum of it) */ -#define FT_FTM 28 /* floppy track max. */ -#define FT_FSM 29 /* floppy sector max. (128) */ -#define FT_LABEL 30 /* floppy tape label */ -#define FT_LABEL_DATE 74 /* date and time the tape label was written */ -#define FT_LABEL_SZ (FT_LABEL_DATE - FT_LABEL) -#define FT_CMAP_START 78 /* starting segment of compression map */ -#define FT_FMT_ERROR 128 /* must be set to 0xff if remainder gets lost during - * tape format - */ -#define FT_SEG_CNT 130 /* number of seg. written, formatted or verified - * through lifetime of tape (why not read?) - */ -#define FT_INIT_DATE 138 /* date and time of initial tape format */ -#define FT_FMT_CNT 142 /* number of times tape has been formatted */ -#define FT_FSL_CNT 144 /* number of segments in failed sector log */ -#define FT_MK_CODE 146 /* id string of tape manufacturer */ -#define FT_LOT_CODE 190 /* tape manufacturer lot code */ -#define FT_6_HSEG_1 234 /* first header segment for format code 6 */ -#define FT_6_HSEG_2 238 /* second header segment for format code 6 */ -#define FT_6_FRST_SEG 242 /* first data segment for format code 6 */ -#define FT_6_LAST_SEG 246 /* last data segment for format code 6 */ - -#define FT_FSL 256 -#define FT_HEADER_END 256 /* space beyond this point: - * format codes 2, 3 and 5: - * - failed sector log until byte 2047 - * - bad sector map in the reamining part of segment - * format codes 4 and 6: - * - bad sector map starts hear - */ - - -/* value to be stored at the FT_SIGNATURE offset - */ -#define FT_HSEG_MAGIC 0xaa55aa55 -#define FT_D2G_MAGIC 0x82288228 /* Ditto 2GB */ - -/* data and time encoding: */ -#define FT_YEAR_SHIFT 25 -#define FT_YEAR_MASK 0xfe000000 -#define FT_YEAR_0 1970 -#define FT_YEAR_MAX 127 -#define FT_YEAR(year) ((((year)-FT_YEAR_0)< -#include -#endif -#include -#include - -#define FT_SECTOR(x) (x+1) /* sector offset into real sector */ -#define FT_SECTOR_SIZE 1024 -#define FT_SECTORS_PER_SEGMENT 32 -#define FT_ECC_SECTORS 3 -#define FT_SEGMENT_SIZE ((FT_SECTORS_PER_SEGMENT - FT_ECC_SECTORS) * FT_SECTOR_SIZE) -#define FT_BUFF_SIZE (FT_SECTORS_PER_SEGMENT * FT_SECTOR_SIZE) - -/* - * bits of the minor device number that define drive selection - * methods. Could be used one day to access multiple tape - * drives on the same controller. - */ -#define FTAPE_SEL_A 0 -#define FTAPE_SEL_B 1 -#define FTAPE_SEL_C 2 -#define FTAPE_SEL_D 3 -#define FTAPE_SEL_MASK 3 -#define FTAPE_SEL(unit) ((unit) & FTAPE_SEL_MASK) -#define FTAPE_NO_REWIND 4 /* mask for minor nr */ - -/* the following two may be reported when MTIOCGET is requested ... */ -typedef union { - struct { - __u8 error; - __u8 command; - } error; - long space; -} ft_drive_error; -typedef union { - struct { - __u8 drive_status; - __u8 drive_config; - __u8 tape_status; - } status; - long space; -} ft_drive_status; - -#ifdef __KERNEL__ - -#define FT_RQM_DELAY 12 -#define FT_MILLISECOND 1 -#define FT_SECOND 1000 -#define FT_FOREVER -1 -#ifndef HZ -#error "HZ undefined." -#endif -#define FT_USPT (1000000/HZ) /* microseconds per tick */ - -/* This defines the number of retries that the driver will allow - * before giving up (and letting a higher level handle the error). - */ -#ifdef TESTING -#define FT_SOFT_RETRIES 1 /* number of low level retries */ -#define FT_RETRIES_ON_ECC_ERROR 3 /* ecc error when correcting segment */ -#else -#define FT_SOFT_RETRIES 6 /* number of low level retries (triple) */ -#define FT_RETRIES_ON_ECC_ERROR 3 /* ecc error when correcting segment */ -#endif - -#ifndef THE_FTAPE_MAINTAINER -#define THE_FTAPE_MAINTAINER "the ftape maintainer" -#endif - -/* Initialize missing configuration parameters. - */ -#ifndef CONFIG_FT_NR_BUFFERS -# define CONFIG_FT_NR_BUFFERS 3 -#endif -#ifndef CONFIG_FT_FDC_THR -# define CONFIG_FT_FDC_THR 8 -#endif -#ifndef CONFIG_FT_FDC_MAX_RATE -# define CONFIG_FT_FDC_MAX_RATE 2000 -#endif -#ifndef CONFIG_FT_FDC_BASE -# define CONFIG_FT_FDC_BASE 0 -#endif -#ifndef CONFIG_FT_FDC_IRQ -# define CONFIG_FT_FDC_IRQ 0 -#endif -#ifndef CONFIG_FT_FDC_DMA -# define CONFIG_FT_FDC_DMA 0 -#endif - -/* Turn some booleans into numbers. - */ -#ifdef CONFIG_FT_PROBE_FC10 -# undef CONFIG_FT_PROBE_FC10 -# define CONFIG_FT_PROBE_FC10 1 -#else -# define CONFIG_FT_PROBE_FC10 0 -#endif -#ifdef CONFIG_FT_MACH2 -# undef CONFIG_FT_MACH2 -# define CONFIG_FT_MACH2 1 -#else -# define CONFIG_FT_MACH2 0 -#endif - -/* Insert default settings - */ -#if CONFIG_FT_PROBE_FC10 == 1 -# if CONFIG_FT_FDC_BASE == 0 -# undef CONFIG_FT_FDC_BASE -# define CONFIG_FT_FDC_BASE 0x180 -# endif -# if CONFIG_FT_FDC_IRQ == 0 -# undef CONFIG_FT_FDC_IRQ -# define CONFIG_FT_FDC_IRQ 9 -# endif -# if CONFIG_FT_FDC_DMA == 0 -# undef CONFIG_FT_FDC_DMA -# define CONFIG_FT_FDC_DMA 3 -# endif -#elif CONFIG_FT_MACH2 == 1 /* CONFIG_FT_PROBE_FC10 == 1 */ -# if CONFIG_FT_FDC_BASE == 0 -# undef CONFIG_FT_FDC_BASE -# define CONFIG_FT_FDC_BASE 0x1E0 -# endif -# if CONFIG_FT_FDC_IRQ == 0 -# undef CONFIG_FT_FDC_IRQ -# define CONFIG_FT_FDC_IRQ 6 -# endif -# if CONFIG_FT_FDC_DMA == 0 -# undef CONFIG_FT_FDC_DMA -# define CONFIG_FT_FDC_DMA 2 -# endif -#elif defined(CONFIG_FT_ALT_FDC) /* CONFIG_FT_MACH2 */ -# if CONFIG_FT_FDC_BASE == 0 -# undef CONFIG_FT_FDC_BASE -# define CONFIG_FT_FDC_BASE 0x370 -# endif -# if CONFIG_FT_FDC_IRQ == 0 -# undef CONFIG_FT_FDC_IRQ -# define CONFIG_FT_FDC_IRQ 6 -# endif -# if CONFIG_FT_FDC_DMA == 0 -# undef CONFIG_FT_FDC_DMA -# define CONFIG_FT_FDC_DMA 2 -# endif -#else /* CONFIG_FT_ALT_FDC */ -# if CONFIG_FT_FDC_BASE == 0 -# undef CONFIG_FT_FDC_BASE -# define CONFIG_FT_FDC_BASE 0x3f0 -# endif -# if CONFIG_FT_FDC_IRQ == 0 -# undef CONFIG_FT_FDC_IRQ -# define CONFIG_FT_FDC_IRQ 6 -# endif -# if CONFIG_FT_FDC_DMA == 0 -# undef CONFIG_FT_FDC_DMA -# define CONFIG_FT_FDC_DMA 2 -# endif -#endif /* standard FDC */ - -/* some useful macro's - */ -#define NR_ITEMS(x) (int)(sizeof(x)/ sizeof(*x)) - -#endif /* __KERNEL__ */ - -#endif diff --git a/include/linux/zftape.h b/include/linux/zftape.h deleted file mode 100644 index b057c65366c..00000000000 --- a/include/linux/zftape.h +++ /dev/null @@ -1,87 +0,0 @@ -#ifndef _ZFTAPE_H -#define _ZFTAPE_H - -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/include/linux/zftape.h,v $ - * $Revision: 1.12 $ - * $Date: 1997/10/21 11:02:37 $ - * - * Special ioctl and other global info for the zftape VFS - * interface for the QIC-40/80/3010/3020 floppy-tape driver for - * Linux. - */ - -#define ZFTAPE_VERSION "zftape for " FTAPE_VERSION - -#include - -#define ZFTAPE_LABEL "Ftape - The Linux Floppy Tape Project!" - -/* Bits of the minor device number that control the operation mode */ -#define ZFT_Q80_MODE (1 << 3) -#define ZFT_ZIP_MODE (1 << 4) -#define ZFT_RAW_MODE (1 << 5) -#define ZFT_MINOR_OP_MASK (ZFT_Q80_MODE | \ - ZFT_ZIP_MODE | \ - ZFT_RAW_MODE) -#define ZFT_MINOR_MASK (FTAPE_SEL_MASK | \ - ZFT_MINOR_OP_MASK | \ - FTAPE_NO_REWIND) - -#ifdef ZFT_OBSOLETE -struct mtblksz { - unsigned int mt_blksz; -}; -#define MTIOC_ZFTAPE_GETBLKSZ _IOR('m', 104, struct mtblksz) -#endif - -#ifdef __KERNEL__ - -extern int zft_init(void); - -static inline __s64 zft_div_blksz(__s64 value, __u32 blk_sz) -{ - if (blk_sz == 1) { - return value; - } else { - return (__s64)(((__u32)(value >> 10) + (blk_sz >> 10) - 1) - / (blk_sz >> 10)); - } -} - -static inline __s64 zft_mul_blksz(__s64 value, __u32 blk_sz) -{ - if (blk_sz == 1) { - return value; - } else { - /* if blk_sz != 1, then it is a multiple of 1024. In - * this case, `value' will also fit into 32 bits. - * - * Actually, this limits the capacity to 42 - * bits. This is (2^32)*1024, roughly a thousand - * times 2GB, or 3 Terabytes. Hopefully this is enough - */ - return(__s64)(((__u32)(value)*(blk_sz>>10))<<10); - } -} - -#endif - -#endif -- cgit v1.2.3-70-g09d2