From 092cab7e2cd868cb0b30209a0337689c3ffd6133 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 16 Dec 2008 01:19:41 -0800 Subject: netfilter: ctnetlink: fix missing CTA_NAT_SEQ_UNSPEC This patch fixes an inconsistency in nfnetlink_conntrack.h that I introduced myself. The problem is that CTA_NAT_SEQ_UNSPEC is missing from enum ctattr_natseq. This inconsistency may lead to problems in the message parsing in userspace (if the message contains the CTA_NAT_SEQ_* attributes, of course). This patch breaks backward compatibility, however, the only known client of this code is libnetfilter_conntrack which indeed crashes because it assumes the existence of CTA_NAT_SEQ_UNSPEC to do the parsing. The CTA_NAT_SEQ_* attributes were introduced in 2.6.25. Signed-off-by: Pablo Neira Ayuso Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nfnetlink_conntrack.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/netfilter/nfnetlink_conntrack.h b/include/linux/netfilter/nfnetlink_conntrack.h index c19595c8930..29fe9ea1d34 100644 --- a/include/linux/netfilter/nfnetlink_conntrack.h +++ b/include/linux/netfilter/nfnetlink_conntrack.h @@ -141,6 +141,7 @@ enum ctattr_protonat { #define CTA_PROTONAT_MAX (__CTA_PROTONAT_MAX - 1) enum ctattr_natseq { + CTA_NAT_SEQ_UNSPEC, CTA_NAT_SEQ_CORRECTION_POS, CTA_NAT_SEQ_OFFSET_BEFORE, CTA_NAT_SEQ_OFFSET_AFTER, -- cgit v1.2.3-70-g09d2 From bf53de907dfdaac178c92d774aae7370d7b97d20 Mon Sep 17 00:00:00 2001 From: Markus Metzger Date: Fri, 19 Dec 2008 15:10:24 +0100 Subject: x86, bts: add fork and exit handling Impact: introduce new ptrace facility Add arch_ptrace_untrace() function that is called when the tracer detaches (either voluntarily or when the tracing task dies); ptrace_disable() is only called on a voluntary detach. Add ptrace_fork() and arch_ptrace_fork(). They are called when a traced task is forked. Clear DS and BTS related fields on fork. Release DS resources and reclaim memory in ptrace_untrace(). This releases resources already when the tracing task dies. We used to do that when the traced task dies. Signed-off-by: Markus Metzger Signed-off-by: Ingo Molnar --- arch/x86/include/asm/ds.h | 9 ++++++++ arch/x86/include/asm/ptrace.h | 7 ++++++ arch/x86/kernel/ds.c | 11 ++++++++++ arch/x86/kernel/process_32.c | 20 ++++++++--------- arch/x86/kernel/process_64.c | 20 ++++++++--------- arch/x86/kernel/ptrace.c | 50 ++++++++++++++++++++++++++++++++++--------- include/linux/ptrace.h | 22 +++++++++++++++++++ kernel/fork.c | 2 ++ kernel/ptrace.c | 12 +++++++++++ 9 files changed, 121 insertions(+), 32 deletions(-) (limited to 'include/linux') diff --git a/arch/x86/include/asm/ds.h b/arch/x86/include/asm/ds.h index ee0ea3a96c1..a8f672ba100 100644 --- a/arch/x86/include/asm/ds.h +++ b/arch/x86/include/asm/ds.h @@ -252,12 +252,21 @@ extern void __cpuinit ds_init_intel(struct cpuinfo_x86 *); */ extern void ds_switch_to(struct task_struct *prev, struct task_struct *next); +/* + * Task clone/init and cleanup work + */ +extern void ds_copy_thread(struct task_struct *tsk, struct task_struct *father); +extern void ds_exit_thread(struct task_struct *tsk); + #else /* CONFIG_X86_DS */ struct cpuinfo_x86; static inline void __cpuinit ds_init_intel(struct cpuinfo_x86 *ignored) {} static inline void ds_switch_to(struct task_struct *prev, struct task_struct *next) {} +static inline void ds_copy_thread(struct task_struct *tsk, + struct task_struct *father) {} +static inline void ds_exit_thread(struct task_struct *tsk) {} #endif /* CONFIG_X86_DS */ #endif /* _ASM_X86_DS_H */ diff --git a/arch/x86/include/asm/ptrace.h b/arch/x86/include/asm/ptrace.h index fbf74421591..6d34d954c22 100644 --- a/arch/x86/include/asm/ptrace.h +++ b/arch/x86/include/asm/ptrace.h @@ -235,6 +235,13 @@ extern int do_get_thread_area(struct task_struct *p, int idx, extern int do_set_thread_area(struct task_struct *p, int idx, struct user_desc __user *info, int can_allocate); +extern void x86_ptrace_untrace(struct task_struct *); +extern void x86_ptrace_fork(struct task_struct *child, + unsigned long clone_flags); + +#define arch_ptrace_untrace(tsk) x86_ptrace_untrace(tsk) +#define arch_ptrace_fork(child, flags) x86_ptrace_fork(child, flags) + #endif /* __KERNEL__ */ #endif /* !__ASSEMBLY__ */ diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c index 98d271e60e0..da91701a234 100644 --- a/arch/x86/kernel/ds.c +++ b/arch/x86/kernel/ds.c @@ -1017,3 +1017,14 @@ void ds_switch_to(struct task_struct *prev, struct task_struct *next) update_debugctlmsr(next->thread.debugctlmsr); } + +void ds_copy_thread(struct task_struct *tsk, struct task_struct *father) +{ + clear_tsk_thread_flag(tsk, TIF_DS_AREA_MSR); + tsk->thread.ds_ctx = NULL; +} + +void ds_exit_thread(struct task_struct *tsk) +{ + WARN_ON(tsk->thread.ds_ctx); +} diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index 605eff9a8ac..3ba155d2488 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -60,6 +60,7 @@ #include #include #include +#include asmlinkage void ret_from_fork(void) __asm__("ret_from_fork"); @@ -251,17 +252,8 @@ void exit_thread(void) tss->x86_tss.io_bitmap_base = INVALID_IO_BITMAP_OFFSET; put_cpu(); } -#ifdef CONFIG_X86_DS - /* Free any BTS tracers that have not been properly released. */ - if (unlikely(current->bts)) { - ds_release_bts(current->bts); - current->bts = NULL; - - kfree(current->bts_buffer); - current->bts_buffer = NULL; - current->bts_size = 0; - } -#endif /* CONFIG_X86_DS */ + + ds_exit_thread(current); } void flush_thread(void) @@ -343,6 +335,12 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long sp, kfree(p->thread.io_bitmap_ptr); p->thread.io_bitmap_max = 0; } + + ds_copy_thread(p, current); + + clear_tsk_thread_flag(p, TIF_DEBUGCTLMSR); + p->thread.debugctlmsr = 0; + return err; } diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 1cfd2a4bf85..416fb9282f4 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -53,6 +53,7 @@ #include #include #include +#include asmlinkage extern void ret_from_fork(void); @@ -236,17 +237,8 @@ void exit_thread(void) t->io_bitmap_max = 0; put_cpu(); } -#ifdef CONFIG_X86_DS - /* Free any BTS tracers that have not been properly released. */ - if (unlikely(current->bts)) { - ds_release_bts(current->bts); - current->bts = NULL; - - kfree(current->bts_buffer); - current->bts_buffer = NULL; - current->bts_size = 0; - } -#endif /* CONFIG_X86_DS */ + + ds_exit_thread(current); } void flush_thread(void) @@ -376,6 +368,12 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long sp, if (err) goto out; } + + ds_copy_thread(p, me); + + clear_tsk_thread_flag(p, TIF_DEBUGCTLMSR); + p->thread.debugctlmsr = 0; + err = 0; out: if (err && p->thread.io_bitmap_ptr) { diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index 45e9855da2d..6ad2bb60765 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -769,8 +769,47 @@ static int ptrace_bts_size(struct task_struct *child) return (trace->ds.top - trace->ds.begin) / trace->ds.size; } + +static void ptrace_bts_fork(struct task_struct *tsk) +{ + tsk->bts = NULL; + tsk->bts_buffer = NULL; + tsk->bts_size = 0; + tsk->thread.bts_ovfl_signal = 0; +} + +static void ptrace_bts_untrace(struct task_struct *child) +{ + if (unlikely(child->bts)) { + ds_release_bts(child->bts); + child->bts = NULL; + + kfree(child->bts_buffer); + child->bts_buffer = NULL; + child->bts_size = 0; + } +} + +static void ptrace_bts_detach(struct task_struct *child) +{ + ptrace_bts_untrace(child); +} +#else +static inline void ptrace_bts_fork(struct task_struct *tsk) {} +static inline void ptrace_bts_detach(struct task_struct *child) {} +static inline void ptrace_bts_untrace(struct task_struct *child) {} #endif /* CONFIG_X86_PTRACE_BTS */ +void x86_ptrace_fork(struct task_struct *child, unsigned long clone_flags) +{ + ptrace_bts_fork(child); +} + +void x86_ptrace_untrace(struct task_struct *child) +{ + ptrace_bts_untrace(child); +} + /* * Called by kernel/ptrace.c when detaching.. * @@ -782,16 +821,7 @@ void ptrace_disable(struct task_struct *child) #ifdef TIF_SYSCALL_EMU clear_tsk_thread_flag(child, TIF_SYSCALL_EMU); #endif -#ifdef CONFIG_X86_PTRACE_BTS - if (child->bts) { - ds_release_bts(child->bts); - child->bts = NULL; - - kfree(child->bts_buffer); - child->bts_buffer = NULL; - child->bts_size = 0; - } -#endif /* CONFIG_X86_PTRACE_BTS */ + ptrace_bts_detach(child); } #if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h index 22641d5d45d..98b93ca4db0 100644 --- a/include/linux/ptrace.h +++ b/include/linux/ptrace.h @@ -94,6 +94,7 @@ extern void ptrace_notify(int exit_code); extern void __ptrace_link(struct task_struct *child, struct task_struct *new_parent); extern void __ptrace_unlink(struct task_struct *child); +extern void ptrace_fork(struct task_struct *task, unsigned long clone_flags); #define PTRACE_MODE_READ 1 #define PTRACE_MODE_ATTACH 2 /* Returns 0 on success, -errno on denial. */ @@ -313,6 +314,27 @@ static inline void user_enable_block_step(struct task_struct *task) #define arch_ptrace_stop(code, info) do { } while (0) #endif +#ifndef arch_ptrace_untrace +/* + * Do machine-specific work before untracing child. + * + * This is called for a normal detach as well as from ptrace_exit() + * when the tracing task dies. + * + * Called with write_lock(&tasklist_lock) held. + */ +#define arch_ptrace_untrace(task) do { } while (0) +#endif + +#ifndef arch_ptrace_fork +/* + * Do machine-specific work to initialize a new task. + * + * This is called from copy_process(). + */ +#define arch_ptrace_fork(child, clone_flags) do { } while (0) +#endif + extern int task_current_syscall(struct task_struct *target, long *callno, unsigned long args[6], unsigned int maxargs, unsigned long *sp, unsigned long *pc); diff --git a/kernel/fork.c b/kernel/fork.c index 7b93da72d4a..65ce60adc8e 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1096,6 +1096,8 @@ static struct task_struct *copy_process(unsigned long clone_flags, #ifdef CONFIG_DEBUG_MUTEXES p->blocked_on = NULL; /* not blocked yet */ #endif + if (unlikely(ptrace_reparented(current))) + ptrace_fork(p, clone_flags); /* Perform scheduler related setup. Assign this task to a CPU. */ sched_fork(p, clone_flags); diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 4c8bcd7dd8e..100a71cfdab 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -25,6 +25,17 @@ #include #include + +/* + * Initialize a new task whose father had been ptraced. + * + * Called from copy_process(). + */ +void ptrace_fork(struct task_struct *child, unsigned long clone_flags) +{ + arch_ptrace_fork(child, clone_flags); +} + /* * ptrace a task: make the debugger its new parent and * move it to the ptrace list. @@ -72,6 +83,7 @@ void __ptrace_unlink(struct task_struct *child) child->parent = child->real_parent; list_del_init(&child->ptrace_entry); + arch_ptrace_untrace(child); if (task_is_traced(child)) ptrace_untrace(child); } -- cgit v1.2.3-70-g09d2 From c5dee6177f4bd2095aab7d9be9f6ebdddd6deee9 Mon Sep 17 00:00:00 2001 From: Markus Metzger Date: Fri, 19 Dec 2008 15:17:02 +0100 Subject: x86, bts: memory accounting Impact: move the BTS buffer accounting to the mlock bucket Add alloc_locked_buffer() and free_locked_buffer() functions to mm/mlock.c to kalloc a buffer and account the locked memory to current. Account the memory for the BTS buffer to the tracer. Signed-off-by: Markus Metzger Signed-off-by: Ingo Molnar --- arch/x86/kernel/ptrace.c | 45 ++++++++++++++++++++++++++++++++++----------- include/linux/mm.h | 2 ++ mm/mlock.c | 45 +++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 81 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index 6ad2bb60765..0a5df5f82fb 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -650,6 +650,24 @@ static int ptrace_bts_drain(struct task_struct *child, return drained; } +static int ptrace_bts_allocate_buffer(struct task_struct *child, size_t size) +{ + child->bts_buffer = alloc_locked_buffer(size); + if (!child->bts_buffer) + return -ENOMEM; + + child->bts_size = size; + + return 0; +} + +static void ptrace_bts_free_buffer(struct task_struct *child) +{ + free_locked_buffer(child->bts_buffer, child->bts_size); + child->bts_buffer = NULL; + child->bts_size = 0; +} + static int ptrace_bts_config(struct task_struct *child, long cfg_size, const struct ptrace_bts_config __user *ucfg) @@ -679,14 +697,13 @@ static int ptrace_bts_config(struct task_struct *child, if ((cfg.flags & PTRACE_BTS_O_ALLOC) && (cfg.size != child->bts_size)) { - kfree(child->bts_buffer); + int error; - child->bts_size = cfg.size; - child->bts_buffer = kzalloc(cfg.size, GFP_KERNEL); - if (!child->bts_buffer) { - child->bts_size = 0; - return -ENOMEM; - } + ptrace_bts_free_buffer(child); + + error = ptrace_bts_allocate_buffer(child, cfg.size); + if (error < 0) + return error; } if (cfg.flags & PTRACE_BTS_O_TRACE) @@ -701,10 +718,8 @@ static int ptrace_bts_config(struct task_struct *child, if (IS_ERR(child->bts)) { int error = PTR_ERR(child->bts); - kfree(child->bts_buffer); + ptrace_bts_free_buffer(child); child->bts = NULL; - child->bts_buffer = NULL; - child->bts_size = 0; return error; } @@ -784,6 +799,9 @@ static void ptrace_bts_untrace(struct task_struct *child) ds_release_bts(child->bts); child->bts = NULL; + /* We cannot update total_vm and locked_vm since + child's mm is already gone. But we can reclaim the + memory. */ kfree(child->bts_buffer); child->bts_buffer = NULL; child->bts_size = 0; @@ -792,7 +810,12 @@ static void ptrace_bts_untrace(struct task_struct *child) static void ptrace_bts_detach(struct task_struct *child) { - ptrace_bts_untrace(child); + if (unlikely(child->bts)) { + ds_release_bts(child->bts); + child->bts = NULL; + + ptrace_bts_free_buffer(child); + } } #else static inline void ptrace_bts_fork(struct task_struct *tsk) {} diff --git a/include/linux/mm.h b/include/linux/mm.h index ffee2f74341..9979d3fab6e 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1286,5 +1286,7 @@ int vmemmap_populate_basepages(struct page *start_page, int vmemmap_populate(struct page *start_page, unsigned long pages, int node); void vmemmap_populate_print_last(void); +extern void *alloc_locked_buffer(size_t size); +extern void free_locked_buffer(void *buffer, size_t size); #endif /* __KERNEL__ */ #endif /* _LINUX_MM_H */ diff --git a/mm/mlock.c b/mm/mlock.c index 1ada366570c..3035a56e761 100644 --- a/mm/mlock.c +++ b/mm/mlock.c @@ -667,3 +667,48 @@ void user_shm_unlock(size_t size, struct user_struct *user) spin_unlock(&shmlock_user_lock); free_uid(user); } + +void *alloc_locked_buffer(size_t size) +{ + unsigned long rlim, vm, pgsz; + void *buffer = NULL; + + pgsz = PAGE_ALIGN(size) >> PAGE_SHIFT; + + down_write(¤t->mm->mmap_sem); + + rlim = current->signal->rlim[RLIMIT_AS].rlim_cur >> PAGE_SHIFT; + vm = current->mm->total_vm + pgsz; + if (rlim < vm) + goto out; + + rlim = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur >> PAGE_SHIFT; + vm = current->mm->locked_vm + pgsz; + if (rlim < vm) + goto out; + + buffer = kzalloc(size, GFP_KERNEL); + if (!buffer) + goto out; + + current->mm->total_vm += pgsz; + current->mm->locked_vm += pgsz; + + out: + up_write(¤t->mm->mmap_sem); + return buffer; +} + +void free_locked_buffer(void *buffer, size_t size) +{ + unsigned long pgsz = PAGE_ALIGN(size) >> PAGE_SHIFT; + + down_write(¤t->mm->mmap_sem); + + current->mm->total_vm -= pgsz; + current->mm->locked_vm -= pgsz; + + up_write(¤t->mm->mmap_sem); + + kfree(buffer); +} -- cgit v1.2.3-70-g09d2 From 3ddeb912f41801fd1968c7880d031702a396e4d0 Mon Sep 17 00:00:00 2001 From: Lai Jiangshan Date: Sat, 20 Dec 2008 17:15:14 +0800 Subject: ftrace: enable format arguments checking Impact: broaden gcc printf format checks for ftrace_printk() format arguments checking for ftrace_printk() is __printf(1, 2), not __printf(1, 0). Signed-off-by: Lai Jiangshan Signed-off-by: Ingo Molnar --- include/linux/ftrace.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h index 04b52e6ebc6..677432b9cb7 100644 --- a/include/linux/ftrace.h +++ b/include/linux/ftrace.h @@ -303,7 +303,7 @@ extern void ftrace_dump(void); static inline void ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3) { } static inline int -ftrace_printk(const char *fmt, ...) __attribute__ ((format (printf, 1, 0))); +ftrace_printk(const char *fmt, ...) __attribute__ ((format (printf, 1, 2))); static inline void tracing_start(void) { } static inline void tracing_stop(void) { } -- cgit v1.2.3-70-g09d2