From 7974891db234467eaf1fec613ec0129cb4ac2332 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 28 Jun 2010 14:15:54 +0200 Subject: x86: Always use irq stacks IRQ stacks provide much better safety against unexpected stack use from interrupts, at the minimal downside of slightly higher memory usage. Enable irq stacks also for the default 8k stack on 32-bit kernels to minimize the problem of stack overflows through interrupt activity. This is what the 64-bit kernel and various other architectures already do. Signed-off-by: Christoph Hellwig LKML-Reference: <20100628121554.GA6605@lst.de> Signed-off-by: Thomas Gleixner --- arch/x86/Kconfig.debug | 3 +-- arch/x86/include/asm/irq.h | 12 +++++------- arch/x86/kernel/irq_32.c | 6 ------ 3 files changed, 6 insertions(+), 15 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug index 75085080b63..badda8e20e7 100644 --- a/arch/x86/Kconfig.debug +++ b/arch/x86/Kconfig.debug @@ -128,8 +128,7 @@ config 4KSTACKS If you say Y here the kernel will use a 4Kb stacksize for the kernel stack attached to each process/thread. This facilitates running more threads on a system and also reduces the pressure - on the VM subsystem for higher order allocations. This option - will also use IRQ stacks to compensate for the reduced stackspace. + on the VM subsystem for higher order allocations. config DOUBLEFAULT default y diff --git a/arch/x86/include/asm/irq.h b/arch/x86/include/asm/irq.h index 5458380b6ef..0bf5b008365 100644 --- a/arch/x86/include/asm/irq.h +++ b/arch/x86/include/asm/irq.h @@ -19,18 +19,16 @@ static inline int irq_canonicalize(int irq) # define ARCH_HAS_NMI_WATCHDOG #endif -#ifdef CONFIG_4KSTACKS - extern void irq_ctx_init(int cpu); - extern void irq_ctx_exit(int cpu); -# define __ARCH_HAS_DO_SOFTIRQ +#ifdef CONFIG_X86_32 +extern void irq_ctx_init(int cpu); +extern void irq_ctx_exit(int cpu); #else # define irq_ctx_init(cpu) do { } while (0) # define irq_ctx_exit(cpu) do { } while (0) -# ifdef CONFIG_X86_64 -# define __ARCH_HAS_DO_SOFTIRQ -# endif #endif +#define __ARCH_HAS_DO_SOFTIRQ + #ifdef CONFIG_HOTPLUG_CPU #include extern void fixup_irqs(void); diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c index 10709f29d16..67f5f9f5299 100644 --- a/arch/x86/kernel/irq_32.c +++ b/arch/x86/kernel/irq_32.c @@ -49,7 +49,6 @@ static inline int check_stack_overflow(void) { return 0; } static inline void print_stack_overflow(void) { } #endif -#ifdef CONFIG_4KSTACKS /* * per-CPU IRQ handling contexts (thread information and stack) */ @@ -187,11 +186,6 @@ asmlinkage void do_softirq(void) local_irq_restore(flags); } -#else -static inline int -execute_on_irq_stack(int overflow, struct irq_desc *desc, int irq) { return 0; } -#endif - bool handle_irq(unsigned irq, struct pt_regs *regs) { struct irq_desc *desc; -- cgit v1.2.3-70-g09d2 From dcfa726280116dd31adad37da940f542663567d0 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 28 Jun 2010 14:16:14 +0200 Subject: x86: Remove CONFIG_4KSTACKS These days 4 kilobytes of stack just aren't enough for reliably operation, and people using lots of threads have long switched to 64-bit kernels, so remove the CONFIG_4KSTACKS option. Signed-off-by: Christoph Hellwig LKML-Reference: <20100628121614.GB6605@lst.de> Signed-off-by: Thomas Gleixner --- arch/x86/Kconfig.debug | 9 --------- arch/x86/include/asm/module.h | 7 +------ arch/x86/include/asm/page_32_types.h | 4 ---- 3 files changed, 1 insertion(+), 19 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug index badda8e20e7..7f1530838bc 100644 --- a/arch/x86/Kconfig.debug +++ b/arch/x86/Kconfig.debug @@ -121,15 +121,6 @@ config DEBUG_NX_TEST and the software setup of this feature. If in doubt, say "N" -config 4KSTACKS - bool "Use 4Kb for kernel stacks instead of 8Kb" - depends on X86_32 - ---help--- - If you say Y here the kernel will use a 4Kb stacksize for the - kernel stack attached to each process/thread. This facilitates - running more threads on a system and also reduces the pressure - on the VM subsystem for higher order allocations. - config DOUBLEFAULT default y bool "Enable doublefault exception handler" if EMBEDDED diff --git a/arch/x86/include/asm/module.h b/arch/x86/include/asm/module.h index 3e2ce58a31a..67763c5d8b4 100644 --- a/arch/x86/include/asm/module.h +++ b/arch/x86/include/asm/module.h @@ -60,12 +60,7 @@ #endif #ifdef CONFIG_X86_32 -# ifdef CONFIG_4KSTACKS -# define MODULE_STACKSIZE "4KSTACKS " -# else -# define MODULE_STACKSIZE "" -# endif -# define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY MODULE_STACKSIZE +# define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY #endif #endif /* _ASM_X86_MODULE_H */ diff --git a/arch/x86/include/asm/page_32_types.h b/arch/x86/include/asm/page_32_types.h index 6f1b7331313..ade619ff9e2 100644 --- a/arch/x86/include/asm/page_32_types.h +++ b/arch/x86/include/asm/page_32_types.h @@ -15,11 +15,7 @@ */ #define __PAGE_OFFSET _AC(CONFIG_PAGE_OFFSET, UL) -#ifdef CONFIG_4KSTACKS -#define THREAD_ORDER 0 -#else #define THREAD_ORDER 1 -#endif #define THREAD_SIZE (PAGE_SIZE << THREAD_ORDER) #define STACKFAULT_STACK 0 -- cgit v1.2.3-70-g09d2 From 25897374297906eeebef8864299406bdcb5859c3 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 27 Jul 2010 14:13:13 +0200 Subject: x86-32: Align IRQ stacks properly As suggested by Steven Rostedt we need to align the IRQ stacks to the stack size, not just the page size to make them work for stack traces and other things that depend on finding the stack slot itself with 8k stacks. Signed-off-by: Christoph Hellwig LKML-Reference: <20100727121313.GA19976@lst.de> Signed-off-by: H. Peter Anvin --- arch/x86/kernel/irq_32.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c index 67f5f9f5299..3b5609f54c4 100644 --- a/arch/x86/kernel/irq_32.c +++ b/arch/x86/kernel/irq_32.c @@ -55,7 +55,7 @@ static inline void print_stack_overflow(void) { } union irq_ctx { struct thread_info tinfo; u32 stack[THREAD_SIZE/sizeof(u32)]; -} __attribute__((aligned(PAGE_SIZE))); +} __attribute__((aligned(THREAD_SIZE))); static DEFINE_PER_CPU(union irq_ctx *, hardirq_ctx); static DEFINE_PER_CPU(union irq_ctx *, softirq_ctx); -- cgit v1.2.3-70-g09d2 From f6e9456c9272bb570df6e217cdbe007e270b1c4e Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Wed, 21 Jul 2010 19:03:58 +0200 Subject: x86, cleanup: Remove obsolete boot_cpu_id variable boot_cpu_id is there for historical reasons and was renamed to boot_cpu_physical_apicid in patch: c70dcb7 x86: change boot_cpu_id to boot_cpu_physical_apicid However, there are some remaining occurrences of boot_cpu_id that are never touched in the kernel and thus its value is always 0. This patch removes boot_cpu_id completely. Signed-off-by: Robert Richter LKML-Reference: <1279731838-1522-8-git-send-email-robert.richter@amd.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/apb_timer.h | 1 - arch/x86/include/asm/cpu.h | 1 - arch/x86/kernel/apb_timer.c | 2 +- arch/x86/kernel/apic/io_apic.c | 12 ++++++------ arch/x86/kernel/cpu/amd.c | 2 +- arch/x86/kernel/cpu/common.c | 2 +- arch/x86/kernel/cpu/intel.c | 2 +- arch/x86/kernel/reboot.c | 2 +- arch/x86/kernel/setup.c | 1 - arch/x86/kernel/setup_percpu.c | 2 +- arch/x86/mm/k8topology_64.c | 6 +++--- 11 files changed, 15 insertions(+), 18 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/apb_timer.h b/arch/x86/include/asm/apb_timer.h index a69b1ac9eaf..2fefa501d3b 100644 --- a/arch/x86/include/asm/apb_timer.h +++ b/arch/x86/include/asm/apb_timer.h @@ -54,7 +54,6 @@ extern struct clock_event_device *global_clock_event; extern unsigned long apbt_quick_calibrate(void); extern int arch_setup_apbt_irqs(int irq, int trigger, int mask, int cpu); extern void apbt_setup_secondary_clock(void); -extern unsigned int boot_cpu_id; extern struct sfi_timer_table_entry *sfi_get_mtmr(int hint); extern void sfi_free_mtmr(struct sfi_timer_table_entry *mtmr); diff --git a/arch/x86/include/asm/cpu.h b/arch/x86/include/asm/cpu.h index b185091bf19..4fab24de26b 100644 --- a/arch/x86/include/asm/cpu.h +++ b/arch/x86/include/asm/cpu.h @@ -32,6 +32,5 @@ extern void arch_unregister_cpu(int); DECLARE_PER_CPU(int, cpu_state); -extern unsigned int boot_cpu_id; #endif /* _ASM_X86_CPU_H */ diff --git a/arch/x86/kernel/apb_timer.c b/arch/x86/kernel/apb_timer.c index 8dd77800ff5..08f75fb4f50 100644 --- a/arch/x86/kernel/apb_timer.c +++ b/arch/x86/kernel/apb_timer.c @@ -343,7 +343,7 @@ void apbt_setup_secondary_clock(void) /* Don't register boot CPU clockevent */ cpu = smp_processor_id(); - if (cpu == boot_cpu_id) + if (!cpu) return; /* * We need to calculate the scaled math multiplication factor for diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 4dc0084ec1b..8884928d7bc 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -162,7 +162,7 @@ int __init arch_early_irq_init(void) cfg = irq_cfgx; count = ARRAY_SIZE(irq_cfgx); - node= cpu_to_node(boot_cpu_id); + node = cpu_to_node(0); for (i = 0; i < count; i++) { desc = irq_to_desc(i); @@ -1483,7 +1483,7 @@ static void __init setup_IO_APIC_irqs(void) int notcon = 0; struct irq_desc *desc; struct irq_cfg *cfg; - int node = cpu_to_node(boot_cpu_id); + int node = cpu_to_node(0); apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n"); @@ -1548,7 +1548,7 @@ static void __init setup_IO_APIC_irqs(void) void setup_IO_APIC_irq_extra(u32 gsi) { int apic_id = 0, pin, idx, irq; - int node = cpu_to_node(boot_cpu_id); + int node = cpu_to_node(0); struct irq_desc *desc; struct irq_cfg *cfg; @@ -2925,7 +2925,7 @@ static inline void __init check_timer(void) { struct irq_desc *desc = irq_to_desc(0); struct irq_cfg *cfg = desc->chip_data; - int node = cpu_to_node(boot_cpu_id); + int node = cpu_to_node(0); int apic1, pin1, apic2, pin2; unsigned long flags; int no_pin1 = 0; @@ -3279,7 +3279,7 @@ unsigned int create_irq_nr(unsigned int irq_want, int node) int create_irq(void) { - int node = cpu_to_node(boot_cpu_id); + int node = cpu_to_node(0); unsigned int irq_want; int irq; @@ -3901,7 +3901,7 @@ static int __io_apic_set_pci_routing(struct device *dev, int irq, if (dev) node = dev_to_node(dev); else - node = cpu_to_node(boot_cpu_id); + node = cpu_to_node(0); desc = irq_to_desc_alloc_node(irq, node); if (!desc) { diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 60a57b13082..3a7c852f021 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -148,7 +148,7 @@ static void __cpuinit amd_k7_smp_check(struct cpuinfo_x86 *c) { #ifdef CONFIG_SMP /* calling is from identify_secondary_cpu() ? */ - if (c->cpu_index == boot_cpu_id) + if (!c->cpu_index) return; /* diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 490dac63c2d..787b3c7c662 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -665,7 +665,7 @@ static void __init early_identify_cpu(struct cpuinfo_x86 *c) this_cpu->c_early_init(c); #ifdef CONFIG_SMP - c->cpu_index = boot_cpu_id; + c->cpu_index = 0; #endif filter_cpuid_features(c, false); } diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 85f69cdeae1..3a683ea5267 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -169,7 +169,7 @@ static void __cpuinit intel_smp_check(struct cpuinfo_x86 *c) { #ifdef CONFIG_SMP /* calling is from identify_secondary_cpu() ? */ - if (c->cpu_index == boot_cpu_id) + if (!c->cpu_index) return; /* diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index e3af342fe83..7a4cf14223b 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -84,7 +84,7 @@ static int __init reboot_setup(char *str) } /* we will leave sorting out the final value when we are ready to reboot, since we might not - have set up boot_cpu_id or smp_num_cpu */ + have detected BSP APIC ID or smp_num_cpu */ break; #endif /* CONFIG_SMP */ diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index b008e788320..dede5c4bae4 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -125,7 +125,6 @@ unsigned long max_pfn_mapped; RESERVE_BRK(dmi_alloc, 65536); #endif -unsigned int boot_cpu_id __read_mostly; static __initdata unsigned long _brk_start = (unsigned long)__brk_base; unsigned long _brk_end = (unsigned long)__brk_base; diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c index a60df9ae645..2335c15c93a 100644 --- a/arch/x86/kernel/setup_percpu.c +++ b/arch/x86/kernel/setup_percpu.c @@ -253,7 +253,7 @@ void __init setup_per_cpu_areas(void) * Up to this point, the boot CPU has been using .init.data * area. Reload any changed state for the boot CPU. */ - if (cpu == boot_cpu_id) + if (!cpu) switch_to_new_gdt(cpu); } diff --git a/arch/x86/mm/k8topology_64.c b/arch/x86/mm/k8topology_64.c index 970ed579d4e..240f86462a8 100644 --- a/arch/x86/mm/k8topology_64.c +++ b/arch/x86/mm/k8topology_64.c @@ -54,8 +54,8 @@ static __init int find_northbridge(void) static __init void early_get_boot_cpu_id(void) { /* - * need to get boot_cpu_id so can use that to create apicid_to_node - * in k8_scan_nodes() + * need to get the APIC ID of the BSP so can use that to + * create apicid_to_node in k8_scan_nodes() */ #ifdef CONFIG_X86_MPPARSE /* @@ -212,7 +212,7 @@ int __init k8_scan_nodes(void) bits = boot_cpu_data.x86_coreid_bits; cores = (1< 0) { pr_info("BSP APIC ID: %02x\n", boot_cpu_physical_apicid); -- cgit v1.2.3-70-g09d2 From 1f999ab5a1360afc388868cc0ef9afe8edeef3be Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Wed, 21 Jul 2010 19:03:57 +0200 Subject: x86, xsave: Disable xsave in i387 emulation mode xsave is broken for (!HAVE_HWFP). This is the case if config MATH_EMULATION is enabled, 'no387' kernel parameter is set and xsave exists. xsave will not work because x86/math-emu and xsave share the same memory. As this case can be treated as corner case we simply disable xsave then. Signed-off-by: Robert Richter LKML-Reference: <1279731838-1522-7-git-send-email-robert.richter@amd.com> Signed-off-by: H. Peter Anvin --- arch/x86/kernel/i387.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index 1f11f5ce668..2605c50b11d 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -67,6 +67,12 @@ static void __cpuinit init_thread_xstate(void) */ if (!HAVE_HWFP) { + /* + * Disable xsave as we do not support it if i387 + * emulation is enabled. + */ + setup_clear_cpu_cap(X86_FEATURE_XSAVE); + setup_clear_cpu_cap(X86_FEATURE_XSAVEOPT); xstate_size = sizeof(struct i387_soft_struct); return; } -- cgit v1.2.3-70-g09d2 From c1a65932fd7216fdc9a0db8bbffe1d47842f862c Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Tue, 29 Jun 2010 18:08:13 +0200 Subject: perf: Drop unappropriate tests on arch callchains Drop the TASK_RUNNING test on user tasks for callchains as this check doesn't seem to make any sense. Also remove the tests for !current that is not supposed to happen and current->pid as this should be handled at the generic level, with exclude_idle attribute. Signed-off-by: Frederic Weisbecker Tested-by: Will Deacon Cc: Ingo Molnar Cc: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: Paul Mackerras Cc: Stephane Eranian Cc: David Miller Cc: Paul Mundt Cc: Borislav Petkov --- arch/arm/kernel/perf_event.c | 6 ------ arch/sh/kernel/perf_callchain.c | 3 --- arch/x86/kernel/cpu/perf_event.c | 3 --- 3 files changed, 12 deletions(-) (limited to 'arch/x86') diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index 417c392ddf1..fdcb0be47df 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -3107,12 +3107,6 @@ perf_do_callchain(struct pt_regs *regs, is_user = user_mode(regs); - if (!current || !current->pid) - return; - - if (is_user && current->state != TASK_RUNNING) - return; - if (!is_user) perf_callchain_kernel(regs, entry); diff --git a/arch/sh/kernel/perf_callchain.c b/arch/sh/kernel/perf_callchain.c index a9dd3abde28..1d6dbce7a3b 100644 --- a/arch/sh/kernel/perf_callchain.c +++ b/arch/sh/kernel/perf_callchain.c @@ -68,9 +68,6 @@ perf_do_callchain(struct pt_regs *regs, struct perf_callchain_entry *entry) is_user = user_mode(regs); - if (is_user && current->state != TASK_RUNNING) - return; - /* * Only the kernel side is implemented for now. */ diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index f2da20fda02..4a4d191f949 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -1703,9 +1703,6 @@ perf_do_callchain(struct pt_regs *regs, struct perf_callchain_entry *entry) is_user = user_mode(regs); - if (is_user && current->state != TASK_RUNNING) - return; - if (!is_user) perf_callchain_kernel(regs, entry); -- cgit v1.2.3-70-g09d2 From 70791ce9ba68a5921c9905ef05d23f62a90bc10c Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Tue, 29 Jun 2010 19:34:05 +0200 Subject: perf: Generalize callchain_store() callchain_store() is the same on every archs, inline it in perf_event.h and rename it to perf_callchain_store() to avoid any collision. This removes repetitive code. Signed-off-by: Frederic Weisbecker Acked-by: Paul Mackerras Tested-by: Will Deacon Cc: Ingo Molnar Cc: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: Stephane Eranian Cc: David Miller Cc: Paul Mundt Cc: Borislav Petkov --- arch/arm/kernel/perf_event.c | 15 ++++---------- arch/powerpc/kernel/perf_callchain.c | 40 +++++++++++++----------------------- arch/sh/kernel/perf_callchain.c | 11 +++------- arch/sparc/kernel/perf_event.c | 26 +++++++++-------------- arch/x86/kernel/cpu/perf_event.c | 20 +++++++----------- include/linux/perf_event.h | 7 +++++++ 6 files changed, 45 insertions(+), 74 deletions(-) (limited to 'arch/x86') diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index fdcb0be47df..a07c3b1955f 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -3001,13 +3001,6 @@ arch_initcall(init_hw_perf_events); /* * Callchain handling code. */ -static inline void -callchain_store(struct perf_callchain_entry *entry, - u64 ip) -{ - if (entry->nr < PERF_MAX_STACK_DEPTH) - entry->ip[entry->nr++] = ip; -} /* * The registers we're interested in are at the end of the variable @@ -3039,7 +3032,7 @@ user_backtrace(struct frame_tail *tail, if (__copy_from_user_inatomic(&buftail, tail, sizeof(buftail))) return NULL; - callchain_store(entry, buftail.lr); + perf_callchain_store(entry, buftail.lr); /* * Frame pointers should strictly progress back up the stack @@ -3057,7 +3050,7 @@ perf_callchain_user(struct pt_regs *regs, { struct frame_tail *tail; - callchain_store(entry, PERF_CONTEXT_USER); + perf_callchain_store(entry, PERF_CONTEXT_USER); if (!user_mode(regs)) regs = task_pt_regs(current); @@ -3078,7 +3071,7 @@ callchain_trace(struct stackframe *fr, void *data) { struct perf_callchain_entry *entry = data; - callchain_store(entry, fr->pc); + perf_callchain_store(entry, fr->pc); return 0; } @@ -3088,7 +3081,7 @@ perf_callchain_kernel(struct pt_regs *regs, { struct stackframe fr; - callchain_store(entry, PERF_CONTEXT_KERNEL); + perf_callchain_store(entry, PERF_CONTEXT_KERNEL); fr.fp = regs->ARM_fp; fr.sp = regs->ARM_sp; fr.lr = regs->ARM_lr; diff --git a/arch/powerpc/kernel/perf_callchain.c b/arch/powerpc/kernel/perf_callchain.c index 95ad9dad298..a286c2e5a3e 100644 --- a/arch/powerpc/kernel/perf_callchain.c +++ b/arch/powerpc/kernel/perf_callchain.c @@ -23,18 +23,6 @@ #include "ppc32.h" #endif -/* - * Store another value in a callchain_entry. - */ -static inline void callchain_store(struct perf_callchain_entry *entry, u64 ip) -{ - unsigned int nr = entry->nr; - - if (nr < PERF_MAX_STACK_DEPTH) { - entry->ip[nr] = ip; - entry->nr = nr + 1; - } -} /* * Is sp valid as the address of the next kernel stack frame after prev_sp? @@ -69,8 +57,8 @@ static void perf_callchain_kernel(struct pt_regs *regs, lr = regs->link; sp = regs->gpr[1]; - callchain_store(entry, PERF_CONTEXT_KERNEL); - callchain_store(entry, regs->nip); + perf_callchain_store(entry, PERF_CONTEXT_KERNEL); + perf_callchain_store(entry, regs->nip); if (!validate_sp(sp, current, STACK_FRAME_OVERHEAD)) return; @@ -89,7 +77,7 @@ static void perf_callchain_kernel(struct pt_regs *regs, next_ip = regs->nip; lr = regs->link; level = 0; - callchain_store(entry, PERF_CONTEXT_KERNEL); + perf_callchain_store(entry, PERF_CONTEXT_KERNEL); } else { if (level == 0) @@ -111,7 +99,7 @@ static void perf_callchain_kernel(struct pt_regs *regs, ++level; } - callchain_store(entry, next_ip); + perf_callchain_store(entry, next_ip); if (!valid_next_sp(next_sp, sp)) return; sp = next_sp; @@ -246,8 +234,8 @@ static void perf_callchain_user_64(struct pt_regs *regs, next_ip = regs->nip; lr = regs->link; sp = regs->gpr[1]; - callchain_store(entry, PERF_CONTEXT_USER); - callchain_store(entry, next_ip); + perf_callchain_store(entry, PERF_CONTEXT_USER); + perf_callchain_store(entry, next_ip); for (;;) { fp = (unsigned long __user *) sp; @@ -276,14 +264,14 @@ static void perf_callchain_user_64(struct pt_regs *regs, read_user_stack_64(&uregs[PT_R1], &sp)) return; level = 0; - callchain_store(entry, PERF_CONTEXT_USER); - callchain_store(entry, next_ip); + perf_callchain_store(entry, PERF_CONTEXT_USER); + perf_callchain_store(entry, next_ip); continue; } if (level == 0) next_ip = lr; - callchain_store(entry, next_ip); + perf_callchain_store(entry, next_ip); ++level; sp = next_sp; } @@ -447,8 +435,8 @@ static void perf_callchain_user_32(struct pt_regs *regs, next_ip = regs->nip; lr = regs->link; sp = regs->gpr[1]; - callchain_store(entry, PERF_CONTEXT_USER); - callchain_store(entry, next_ip); + perf_callchain_store(entry, PERF_CONTEXT_USER); + perf_callchain_store(entry, next_ip); while (entry->nr < PERF_MAX_STACK_DEPTH) { fp = (unsigned int __user *) (unsigned long) sp; @@ -470,14 +458,14 @@ static void perf_callchain_user_32(struct pt_regs *regs, read_user_stack_32(&uregs[PT_R1], &sp)) return; level = 0; - callchain_store(entry, PERF_CONTEXT_USER); - callchain_store(entry, next_ip); + perf_callchain_store(entry, PERF_CONTEXT_USER); + perf_callchain_store(entry, next_ip); continue; } if (level == 0) next_ip = lr; - callchain_store(entry, next_ip); + perf_callchain_store(entry, next_ip); ++level; sp = next_sp; } diff --git a/arch/sh/kernel/perf_callchain.c b/arch/sh/kernel/perf_callchain.c index 1d6dbce7a3b..00143f3dd19 100644 --- a/arch/sh/kernel/perf_callchain.c +++ b/arch/sh/kernel/perf_callchain.c @@ -14,11 +14,6 @@ #include #include -static inline void callchain_store(struct perf_callchain_entry *entry, u64 ip) -{ - if (entry->nr < PERF_MAX_STACK_DEPTH) - entry->ip[entry->nr++] = ip; -} static void callchain_warning(void *data, char *msg) { @@ -39,7 +34,7 @@ static void callchain_address(void *data, unsigned long addr, int reliable) struct perf_callchain_entry *entry = data; if (reliable) - callchain_store(entry, addr); + perf_callchain_store(entry, addr); } static const struct stacktrace_ops callchain_ops = { @@ -52,8 +47,8 @@ static const struct stacktrace_ops callchain_ops = { static void perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry) { - callchain_store(entry, PERF_CONTEXT_KERNEL); - callchain_store(entry, regs->pc); + perf_callchain_store(entry, PERF_CONTEXT_KERNEL); + perf_callchain_store(entry, regs->pc); unwind_stack(NULL, regs, NULL, &callchain_ops, entry); } diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c index 357ced3c33f..2a95a907986 100644 --- a/arch/sparc/kernel/perf_event.c +++ b/arch/sparc/kernel/perf_event.c @@ -1283,12 +1283,6 @@ void __init init_hw_perf_events(void) register_die_notifier(&perf_event_nmi_notifier); } -static inline void callchain_store(struct perf_callchain_entry *entry, u64 ip) -{ - if (entry->nr < PERF_MAX_STACK_DEPTH) - entry->ip[entry->nr++] = ip; -} - static void perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry) { @@ -1297,8 +1291,8 @@ static void perf_callchain_kernel(struct pt_regs *regs, int graph = 0; #endif - callchain_store(entry, PERF_CONTEXT_KERNEL); - callchain_store(entry, regs->tpc); + perf_callchain_store(entry, PERF_CONTEXT_KERNEL); + perf_callchain_store(entry, regs->tpc); ksp = regs->u_regs[UREG_I6]; fp = ksp + STACK_BIAS; @@ -1322,13 +1316,13 @@ static void perf_callchain_kernel(struct pt_regs *regs, pc = sf->callers_pc; fp = (unsigned long)sf->fp + STACK_BIAS; } - callchain_store(entry, pc); + perf_callchain_store(entry, pc); #ifdef CONFIG_FUNCTION_GRAPH_TRACER if ((pc + 8UL) == (unsigned long) &return_to_handler) { int index = current->curr_ret_stack; if (current->ret_stack && index >= graph) { pc = current->ret_stack[index - graph].ret; - callchain_store(entry, pc); + perf_callchain_store(entry, pc); graph++; } } @@ -1341,8 +1335,8 @@ static void perf_callchain_user_64(struct pt_regs *regs, { unsigned long ufp; - callchain_store(entry, PERF_CONTEXT_USER); - callchain_store(entry, regs->tpc); + perf_callchain_store(entry, PERF_CONTEXT_USER); + perf_callchain_store(entry, regs->tpc); ufp = regs->u_regs[UREG_I6] + STACK_BIAS; do { @@ -1355,7 +1349,7 @@ static void perf_callchain_user_64(struct pt_regs *regs, pc = sf.callers_pc; ufp = (unsigned long)sf.fp + STACK_BIAS; - callchain_store(entry, pc); + perf_callchain_store(entry, pc); } while (entry->nr < PERF_MAX_STACK_DEPTH); } @@ -1364,8 +1358,8 @@ static void perf_callchain_user_32(struct pt_regs *regs, { unsigned long ufp; - callchain_store(entry, PERF_CONTEXT_USER); - callchain_store(entry, regs->tpc); + perf_callchain_store(entry, PERF_CONTEXT_USER); + perf_callchain_store(entry, regs->tpc); ufp = regs->u_regs[UREG_I6] & 0xffffffffUL; do { @@ -1378,7 +1372,7 @@ static void perf_callchain_user_32(struct pt_regs *regs, pc = sf.callers_pc; ufp = (unsigned long)sf.fp; - callchain_store(entry, pc); + perf_callchain_store(entry, pc); } while (entry->nr < PERF_MAX_STACK_DEPTH); } diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 4a4d191f949..8af28caeafc 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -1571,12 +1571,6 @@ const struct pmu *hw_perf_event_init(struct perf_event *event) * callchain support */ -static inline -void callchain_store(struct perf_callchain_entry *entry, u64 ip) -{ - if (entry->nr < PERF_MAX_STACK_DEPTH) - entry->ip[entry->nr++] = ip; -} static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_irq_entry); static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_nmi_entry); @@ -1602,7 +1596,7 @@ static void backtrace_address(void *data, unsigned long addr, int reliable) { struct perf_callchain_entry *entry = data; - callchain_store(entry, addr); + perf_callchain_store(entry, addr); } static const struct stacktrace_ops backtrace_ops = { @@ -1616,8 +1610,8 @@ static const struct stacktrace_ops backtrace_ops = { static void perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry) { - callchain_store(entry, PERF_CONTEXT_KERNEL); - callchain_store(entry, regs->ip); + perf_callchain_store(entry, PERF_CONTEXT_KERNEL); + perf_callchain_store(entry, regs->ip); dump_trace(NULL, regs, NULL, regs->bp, &backtrace_ops, entry); } @@ -1646,7 +1640,7 @@ perf_callchain_user32(struct pt_regs *regs, struct perf_callchain_entry *entry) if (fp < compat_ptr(regs->sp)) break; - callchain_store(entry, frame.return_address); + perf_callchain_store(entry, frame.return_address); fp = compat_ptr(frame.next_frame); } return 1; @@ -1670,8 +1664,8 @@ perf_callchain_user(struct pt_regs *regs, struct perf_callchain_entry *entry) fp = (void __user *)regs->bp; - callchain_store(entry, PERF_CONTEXT_USER); - callchain_store(entry, regs->ip); + perf_callchain_store(entry, PERF_CONTEXT_USER); + perf_callchain_store(entry, regs->ip); if (perf_callchain_user32(regs, entry)) return; @@ -1688,7 +1682,7 @@ perf_callchain_user(struct pt_regs *regs, struct perf_callchain_entry *entry) if ((unsigned long)fp < regs->sp) break; - callchain_store(entry, frame.return_address); + perf_callchain_store(entry, frame.return_address); fp = frame.next_frame; } } diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 937495c2507..358880404b4 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -978,6 +978,13 @@ extern void perf_event_fork(struct task_struct *tsk); extern struct perf_callchain_entry *perf_callchain(struct pt_regs *regs); +static inline void +perf_callchain_store(struct perf_callchain_entry *entry, u64 ip) +{ + if (entry->nr < PERF_MAX_STACK_DEPTH) + entry->ip[entry->nr++] = ip; +} + extern int sysctl_perf_event_paranoid; extern int sysctl_perf_event_mlock; extern int sysctl_perf_event_sample_rate; -- cgit v1.2.3-70-g09d2 From 56962b4449af34070bb1994621ef4f0265eed4d8 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Wed, 30 Jun 2010 23:03:51 +0200 Subject: perf: Generalize some arch callchain code - Most archs use one callchain buffer per cpu, except x86 that needs to deal with NMIs. Provide a default perf_callchain_buffer() implementation that x86 overrides. - Centralize all the kernel/user regs handling and invoke new arch handlers from there: perf_callchain_user() / perf_callchain_kernel() That avoid all the user_mode(), current->mm checks and so... - Invert some parameters in perf_callchain_*() helpers: entry to the left, regs to the right, following the traditional (dst, src). Signed-off-by: Frederic Weisbecker Acked-by: Paul Mackerras Tested-by: Will Deacon Cc: Ingo Molnar Cc: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: Stephane Eranian Cc: David Miller Cc: Paul Mundt Cc: Borislav Petkov --- arch/arm/kernel/perf_event.c | 43 +++---------------------------- arch/powerpc/kernel/perf_callchain.c | 49 +++++++++++------------------------- arch/sh/kernel/perf_callchain.c | 37 ++------------------------- arch/sparc/kernel/perf_event.c | 46 +++++++++++---------------------- arch/x86/kernel/cpu/perf_event.c | 45 ++++++--------------------------- include/linux/perf_event.h | 10 +++++++- kernel/perf_event.c | 40 +++++++++++++++++++++++++++-- 7 files changed, 90 insertions(+), 180 deletions(-) (limited to 'arch/x86') diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index a07c3b1955f..0e3bbdb1592 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -3044,17 +3044,13 @@ user_backtrace(struct frame_tail *tail, return buftail.fp - 1; } -static void -perf_callchain_user(struct pt_regs *regs, - struct perf_callchain_entry *entry) +void +perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs) { struct frame_tail *tail; perf_callchain_store(entry, PERF_CONTEXT_USER); - if (!user_mode(regs)) - regs = task_pt_regs(current); - tail = (struct frame_tail *)regs->ARM_fp - 1; while (tail && !((unsigned long)tail & 0x3)) @@ -3075,9 +3071,8 @@ callchain_trace(struct stackframe *fr, return 0; } -static void -perf_callchain_kernel(struct pt_regs *regs, - struct perf_callchain_entry *entry) +void +perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs) { struct stackframe fr; @@ -3088,33 +3083,3 @@ perf_callchain_kernel(struct pt_regs *regs, fr.pc = regs->ARM_pc; walk_stackframe(&fr, callchain_trace, entry); } - -static void -perf_do_callchain(struct pt_regs *regs, - struct perf_callchain_entry *entry) -{ - int is_user; - - if (!regs) - return; - - is_user = user_mode(regs); - - if (!is_user) - perf_callchain_kernel(regs, entry); - - if (current->mm) - perf_callchain_user(regs, entry); -} - -static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_irq_entry); - -struct perf_callchain_entry * -perf_callchain(struct pt_regs *regs) -{ - struct perf_callchain_entry *entry = &__get_cpu_var(pmc_irq_entry); - - entry->nr = 0; - perf_do_callchain(regs, entry); - return entry; -} diff --git a/arch/powerpc/kernel/perf_callchain.c b/arch/powerpc/kernel/perf_callchain.c index a286c2e5a3e..f7a85ede840 100644 --- a/arch/powerpc/kernel/perf_callchain.c +++ b/arch/powerpc/kernel/perf_callchain.c @@ -46,8 +46,8 @@ static int valid_next_sp(unsigned long sp, unsigned long prev_sp) return 0; } -static void perf_callchain_kernel(struct pt_regs *regs, - struct perf_callchain_entry *entry) +void +perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs) { unsigned long sp, next_sp; unsigned long next_ip; @@ -221,8 +221,8 @@ static int sane_signal_64_frame(unsigned long sp) puc == (unsigned long) &sf->uc; } -static void perf_callchain_user_64(struct pt_regs *regs, - struct perf_callchain_entry *entry) +static void perf_callchain_user_64(struct perf_callchain_entry *entry, + struct pt_regs *regs) { unsigned long sp, next_sp; unsigned long next_ip; @@ -303,8 +303,8 @@ static int read_user_stack_32(unsigned int __user *ptr, unsigned int *ret) return __get_user_inatomic(*ret, ptr); } -static inline void perf_callchain_user_64(struct pt_regs *regs, - struct perf_callchain_entry *entry) +static inline void perf_callchain_user_64(struct perf_callchain_entry *entry, + struct pt_regs *regs) { } @@ -423,8 +423,8 @@ static unsigned int __user *signal_frame_32_regs(unsigned int sp, return mctx->mc_gregs; } -static void perf_callchain_user_32(struct pt_regs *regs, - struct perf_callchain_entry *entry) +static void perf_callchain_user_32(struct perf_callchain_entry *entry, + struct pt_regs *regs) { unsigned int sp, next_sp; unsigned int next_ip; @@ -471,32 +471,11 @@ static void perf_callchain_user_32(struct pt_regs *regs, } } -/* - * Since we can't get PMU interrupts inside a PMU interrupt handler, - * we don't need separate irq and nmi entries here. - */ -static DEFINE_PER_CPU(struct perf_callchain_entry, cpu_perf_callchain); - -struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) +void +perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs) { - struct perf_callchain_entry *entry = &__get_cpu_var(cpu_perf_callchain); - - entry->nr = 0; - - if (!user_mode(regs)) { - perf_callchain_kernel(regs, entry); - if (current->mm) - regs = task_pt_regs(current); - else - regs = NULL; - } - - if (regs) { - if (current_is_64bit()) - perf_callchain_user_64(regs, entry); - else - perf_callchain_user_32(regs, entry); - } - - return entry; + if (current_is_64bit()) + perf_callchain_user_64(entry, regs); + else + perf_callchain_user_32(entry, regs); } diff --git a/arch/sh/kernel/perf_callchain.c b/arch/sh/kernel/perf_callchain.c index 00143f3dd19..ef076a91292 100644 --- a/arch/sh/kernel/perf_callchain.c +++ b/arch/sh/kernel/perf_callchain.c @@ -44,44 +44,11 @@ static const struct stacktrace_ops callchain_ops = { .address = callchain_address, }; -static void -perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry) +void +perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs) { perf_callchain_store(entry, PERF_CONTEXT_KERNEL); perf_callchain_store(entry, regs->pc); unwind_stack(NULL, regs, NULL, &callchain_ops, entry); } - -static void -perf_do_callchain(struct pt_regs *regs, struct perf_callchain_entry *entry) -{ - int is_user; - - if (!regs) - return; - - is_user = user_mode(regs); - - /* - * Only the kernel side is implemented for now. - */ - if (!is_user) - perf_callchain_kernel(regs, entry); -} - -/* - * No need for separate IRQ and NMI entries. - */ -static DEFINE_PER_CPU(struct perf_callchain_entry, callchain); - -struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) -{ - struct perf_callchain_entry *entry = &__get_cpu_var(callchain); - - entry->nr = 0; - - perf_do_callchain(regs, entry); - - return entry; -} diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c index 2a95a907986..460162d74ab 100644 --- a/arch/sparc/kernel/perf_event.c +++ b/arch/sparc/kernel/perf_event.c @@ -1283,14 +1283,16 @@ void __init init_hw_perf_events(void) register_die_notifier(&perf_event_nmi_notifier); } -static void perf_callchain_kernel(struct pt_regs *regs, - struct perf_callchain_entry *entry) +void perf_callchain_kernel(struct perf_callchain_entry *entry, + struct pt_regs *regs) { unsigned long ksp, fp; #ifdef CONFIG_FUNCTION_GRAPH_TRACER int graph = 0; #endif + stack_trace_flush(); + perf_callchain_store(entry, PERF_CONTEXT_KERNEL); perf_callchain_store(entry, regs->tpc); @@ -1330,8 +1332,8 @@ static void perf_callchain_kernel(struct pt_regs *regs, } while (entry->nr < PERF_MAX_STACK_DEPTH); } -static void perf_callchain_user_64(struct pt_regs *regs, - struct perf_callchain_entry *entry) +static void perf_callchain_user_64(struct perf_callchain_entry *entry, + struct pt_regs *regs) { unsigned long ufp; @@ -1353,8 +1355,8 @@ static void perf_callchain_user_64(struct pt_regs *regs, } while (entry->nr < PERF_MAX_STACK_DEPTH); } -static void perf_callchain_user_32(struct pt_regs *regs, - struct perf_callchain_entry *entry) +static void perf_callchain_user_32(struct perf_callchain_entry *entry, + struct pt_regs *regs) { unsigned long ufp; @@ -1376,30 +1378,12 @@ static void perf_callchain_user_32(struct pt_regs *regs, } while (entry->nr < PERF_MAX_STACK_DEPTH); } -/* Like powerpc we can't get PMU interrupts within the PMU handler, - * so no need for separate NMI and IRQ chains as on x86. - */ -static DEFINE_PER_CPU(struct perf_callchain_entry, callchain); - -struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) +void +perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs) { - struct perf_callchain_entry *entry = &__get_cpu_var(callchain); - - entry->nr = 0; - if (!user_mode(regs)) { - stack_trace_flush(); - perf_callchain_kernel(regs, entry); - if (current->mm) - regs = task_pt_regs(current); - else - regs = NULL; - } - if (regs) { - flushw_user(); - if (test_thread_flag(TIF_32BIT)) - perf_callchain_user_32(regs, entry); - else - perf_callchain_user_64(regs, entry); - } - return entry; + flushw_user(); + if (test_thread_flag(TIF_32BIT)) + perf_callchain_user_32(entry, regs); + else + perf_callchain_user_64(entry, regs); } diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 8af28caeafc..39f8421b86e 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -1571,9 +1571,7 @@ const struct pmu *hw_perf_event_init(struct perf_event *event) * callchain support */ - -static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_irq_entry); -static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_nmi_entry); +static DEFINE_PER_CPU(struct perf_callchain_entry, perf_callchain_entry_nmi); static void @@ -1607,8 +1605,8 @@ static const struct stacktrace_ops backtrace_ops = { .walk_stack = print_context_stack_bp, }; -static void -perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry) +void +perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs) { perf_callchain_store(entry, PERF_CONTEXT_KERNEL); perf_callchain_store(entry, regs->ip); @@ -1653,14 +1651,12 @@ perf_callchain_user32(struct pt_regs *regs, struct perf_callchain_entry *entry) } #endif -static void -perf_callchain_user(struct pt_regs *regs, struct perf_callchain_entry *entry) +void +perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs) { struct stack_frame frame; const void __user *fp; - if (!user_mode(regs)) - regs = task_pt_regs(current); fp = (void __user *)regs->bp; @@ -1687,42 +1683,17 @@ perf_callchain_user(struct pt_regs *regs, struct perf_callchain_entry *entry) } } -static void -perf_do_callchain(struct pt_regs *regs, struct perf_callchain_entry *entry) -{ - int is_user; - - if (!regs) - return; - - is_user = user_mode(regs); - - if (!is_user) - perf_callchain_kernel(regs, entry); - - if (current->mm) - perf_callchain_user(regs, entry); -} - -struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) +struct perf_callchain_entry *perf_callchain_buffer(void) { - struct perf_callchain_entry *entry; - if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) { /* TODO: We don't support guest os callchain now */ return NULL; } if (in_nmi()) - entry = &__get_cpu_var(pmc_nmi_entry); - else - entry = &__get_cpu_var(pmc_irq_entry); - - entry->nr = 0; - - perf_do_callchain(regs, entry); + return &__get_cpu_var(perf_callchain_entry_nmi); - return entry; + return &__get_cpu_var(perf_callchain_entry); } unsigned long perf_instruction_pointer(struct pt_regs *regs) diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 358880404b4..4db61dded38 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -976,7 +976,15 @@ extern int perf_unregister_guest_info_callbacks(struct perf_guest_info_callbacks extern void perf_event_comm(struct task_struct *tsk); extern void perf_event_fork(struct task_struct *tsk); -extern struct perf_callchain_entry *perf_callchain(struct pt_regs *regs); +/* Callchains */ +DECLARE_PER_CPU(struct perf_callchain_entry, perf_callchain_entry); + +extern void perf_callchain_user(struct perf_callchain_entry *entry, + struct pt_regs *regs); +extern void perf_callchain_kernel(struct perf_callchain_entry *entry, + struct pt_regs *regs); +extern struct perf_callchain_entry *perf_callchain_buffer(void); + static inline void perf_callchain_store(struct perf_callchain_entry *entry, u64 ip) diff --git a/kernel/perf_event.c b/kernel/perf_event.c index c772a3d4000..02efde6c879 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -2937,13 +2937,49 @@ void perf_event_do_pending(void) __perf_pending_run(); } +DEFINE_PER_CPU(struct perf_callchain_entry, perf_callchain_entry); + /* * Callchain support -- arch specific */ -__weak struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) +__weak struct perf_callchain_entry *perf_callchain_buffer(void) { - return NULL; + return &__get_cpu_var(perf_callchain_entry); +} + +__weak void perf_callchain_kernel(struct perf_callchain_entry *entry, + struct pt_regs *regs) +{ +} + +__weak void perf_callchain_user(struct perf_callchain_entry *entry, + struct pt_regs *regs) +{ +} + +static struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) +{ + struct perf_callchain_entry *entry; + + entry = perf_callchain_buffer(); + if (!entry) + return NULL; + + entry->nr = 0; + + if (!user_mode(regs)) { + perf_callchain_kernel(entry, regs); + if (current->mm) + regs = task_pt_regs(current); + else + regs = NULL; + } + + if (regs) + perf_callchain_user(entry, regs); + + return entry; } -- cgit v1.2.3-70-g09d2 From f72c1a931e311bb7780fee19e41a89ac42cab50e Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Thu, 1 Jul 2010 02:31:21 +0200 Subject: perf: Factorize callchain context handling Store the kernel and user contexts from the generic layer instead of archs, this gathers some repetitive code. Signed-off-by: Frederic Weisbecker Acked-by: Paul Mackerras Tested-by: Will Deacon Cc: Ingo Molnar Cc: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: Stephane Eranian Cc: David Miller Cc: Paul Mundt Cc: Borislav Petkov --- arch/arm/kernel/perf_event.c | 2 -- arch/powerpc/kernel/perf_callchain.c | 3 --- arch/sh/kernel/perf_callchain.c | 1 - arch/sparc/kernel/perf_event.c | 3 --- arch/x86/kernel/cpu/perf_event.c | 2 -- kernel/perf_event.c | 5 ++++- 6 files changed, 4 insertions(+), 12 deletions(-) (limited to 'arch/x86') diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index 0e3bbdb1592..64ca8c3ab94 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -3049,7 +3049,6 @@ perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs) { struct frame_tail *tail; - perf_callchain_store(entry, PERF_CONTEXT_USER); tail = (struct frame_tail *)regs->ARM_fp - 1; @@ -3076,7 +3075,6 @@ perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs) { struct stackframe fr; - perf_callchain_store(entry, PERF_CONTEXT_KERNEL); fr.fp = regs->ARM_fp; fr.sp = regs->ARM_sp; fr.lr = regs->ARM_lr; diff --git a/arch/powerpc/kernel/perf_callchain.c b/arch/powerpc/kernel/perf_callchain.c index f7a85ede840..d05ae4204bb 100644 --- a/arch/powerpc/kernel/perf_callchain.c +++ b/arch/powerpc/kernel/perf_callchain.c @@ -57,7 +57,6 @@ perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs) lr = regs->link; sp = regs->gpr[1]; - perf_callchain_store(entry, PERF_CONTEXT_KERNEL); perf_callchain_store(entry, regs->nip); if (!validate_sp(sp, current, STACK_FRAME_OVERHEAD)) @@ -234,7 +233,6 @@ static void perf_callchain_user_64(struct perf_callchain_entry *entry, next_ip = regs->nip; lr = regs->link; sp = regs->gpr[1]; - perf_callchain_store(entry, PERF_CONTEXT_USER); perf_callchain_store(entry, next_ip); for (;;) { @@ -435,7 +433,6 @@ static void perf_callchain_user_32(struct perf_callchain_entry *entry, next_ip = regs->nip; lr = regs->link; sp = regs->gpr[1]; - perf_callchain_store(entry, PERF_CONTEXT_USER); perf_callchain_store(entry, next_ip); while (entry->nr < PERF_MAX_STACK_DEPTH) { diff --git a/arch/sh/kernel/perf_callchain.c b/arch/sh/kernel/perf_callchain.c index ef076a91292..d5ca1ef50fa 100644 --- a/arch/sh/kernel/perf_callchain.c +++ b/arch/sh/kernel/perf_callchain.c @@ -47,7 +47,6 @@ static const struct stacktrace_ops callchain_ops = { void perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs) { - perf_callchain_store(entry, PERF_CONTEXT_KERNEL); perf_callchain_store(entry, regs->pc); unwind_stack(NULL, regs, NULL, &callchain_ops, entry); diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c index 460162d74ab..4bc40293857 100644 --- a/arch/sparc/kernel/perf_event.c +++ b/arch/sparc/kernel/perf_event.c @@ -1293,7 +1293,6 @@ void perf_callchain_kernel(struct perf_callchain_entry *entry, stack_trace_flush(); - perf_callchain_store(entry, PERF_CONTEXT_KERNEL); perf_callchain_store(entry, regs->tpc); ksp = regs->u_regs[UREG_I6]; @@ -1337,7 +1336,6 @@ static void perf_callchain_user_64(struct perf_callchain_entry *entry, { unsigned long ufp; - perf_callchain_store(entry, PERF_CONTEXT_USER); perf_callchain_store(entry, regs->tpc); ufp = regs->u_regs[UREG_I6] + STACK_BIAS; @@ -1360,7 +1358,6 @@ static void perf_callchain_user_32(struct perf_callchain_entry *entry, { unsigned long ufp; - perf_callchain_store(entry, PERF_CONTEXT_USER); perf_callchain_store(entry, regs->tpc); ufp = regs->u_regs[UREG_I6] & 0xffffffffUL; diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 39f8421b86e..a3c922288cc 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -1608,7 +1608,6 @@ static const struct stacktrace_ops backtrace_ops = { void perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs) { - perf_callchain_store(entry, PERF_CONTEXT_KERNEL); perf_callchain_store(entry, regs->ip); dump_trace(NULL, regs, NULL, regs->bp, &backtrace_ops, entry); @@ -1660,7 +1659,6 @@ perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs) fp = (void __user *)regs->bp; - perf_callchain_store(entry, PERF_CONTEXT_USER); perf_callchain_store(entry, regs->ip); if (perf_callchain_user32(regs, entry)) diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 02efde6c879..615d024894c 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -2969,6 +2969,7 @@ static struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) entry->nr = 0; if (!user_mode(regs)) { + perf_callchain_store(entry, PERF_CONTEXT_KERNEL); perf_callchain_kernel(entry, regs); if (current->mm) regs = task_pt_regs(current); @@ -2976,8 +2977,10 @@ static struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) regs = NULL; } - if (regs) + if (regs) { + perf_callchain_store(entry, PERF_CONTEXT_USER); perf_callchain_user(entry, regs); + } return entry; } -- cgit v1.2.3-70-g09d2 From 927c7a9e92c4f69097a6e9e086d11fc2f8a5b40b Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Thu, 1 Jul 2010 16:20:36 +0200 Subject: perf: Fix race in callchains Now that software events don't have interrupt disabled anymore in the event path, callchains can nest on any context. So seperating nmi and others contexts in two buffers has become racy. Fix this by providing one buffer per nesting level. Given the size of the callchain entries (2040 bytes * 4), we now need to allocate them dynamically. v2: Fixed put_callchain_entry call after recursion. Fix the type of the recursion, it must be an array. v3: Use a manual pr cpu allocation (temporary solution until NMIs can safely access vmalloc'ed memory). Do a better separation between callchain reference tracking and allocation. Make the "put" path lockless for non-release cases. v4: Protect the callchain buffers with rcu. v5: Do the cpu buffers allocations node affine. Signed-off-by: Frederic Weisbecker Tested-by: Will Deacon Cc: Ingo Molnar Cc: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: Paul Mackerras Cc: Stephane Eranian Cc: Paul Mundt Cc: David Miller Cc: Borislav Petkov --- arch/x86/kernel/cpu/perf_event.c | 22 ++- include/linux/perf_event.h | 1 - kernel/perf_event.c | 298 ++++++++++++++++++++++++++++++--------- 3 files changed, 238 insertions(+), 83 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index a3c922288cc..8e91cf34a9c 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -1608,6 +1608,11 @@ static const struct stacktrace_ops backtrace_ops = { void perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs) { + if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) { + /* TODO: We don't support guest os callchain now */ + return NULL; + } + perf_callchain_store(entry, regs->ip); dump_trace(NULL, regs, NULL, regs->bp, &backtrace_ops, entry); @@ -1656,6 +1661,10 @@ perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs) struct stack_frame frame; const void __user *fp; + if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) { + /* TODO: We don't support guest os callchain now */ + return NULL; + } fp = (void __user *)regs->bp; @@ -1681,19 +1690,6 @@ perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs) } } -struct perf_callchain_entry *perf_callchain_buffer(void) -{ - if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) { - /* TODO: We don't support guest os callchain now */ - return NULL; - } - - if (in_nmi()) - return &__get_cpu_var(perf_callchain_entry_nmi); - - return &__get_cpu_var(perf_callchain_entry); -} - unsigned long perf_instruction_pointer(struct pt_regs *regs) { unsigned long ip; diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 4db61dded38..d7e8ea69086 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -983,7 +983,6 @@ extern void perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs); extern void perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs); -extern struct perf_callchain_entry *perf_callchain_buffer(void); static inline void diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 615d024894c..75ab8a2df6b 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -1763,6 +1763,216 @@ static u64 perf_event_read(struct perf_event *event) return perf_event_count(event); } +/* + * Callchain support + */ + +struct callchain_cpus_entries { + struct rcu_head rcu_head; + struct perf_callchain_entry *cpu_entries[0]; +}; + +static DEFINE_PER_CPU(int, callchain_recursion[4]); +static atomic_t nr_callchain_events; +static DEFINE_MUTEX(callchain_mutex); +struct callchain_cpus_entries *callchain_cpus_entries; + + +__weak void perf_callchain_kernel(struct perf_callchain_entry *entry, + struct pt_regs *regs) +{ +} + +__weak void perf_callchain_user(struct perf_callchain_entry *entry, + struct pt_regs *regs) +{ +} + +static void release_callchain_buffers_rcu(struct rcu_head *head) +{ + struct callchain_cpus_entries *entries; + int cpu; + + entries = container_of(head, struct callchain_cpus_entries, rcu_head); + + for_each_possible_cpu(cpu) + kfree(entries->cpu_entries[cpu]); + + kfree(entries); +} + +static void release_callchain_buffers(void) +{ + struct callchain_cpus_entries *entries; + + entries = callchain_cpus_entries; + rcu_assign_pointer(callchain_cpus_entries, NULL); + call_rcu(&entries->rcu_head, release_callchain_buffers_rcu); +} + +static int alloc_callchain_buffers(void) +{ + int cpu; + int size; + struct callchain_cpus_entries *entries; + + /* + * We can't use the percpu allocation API for data that can be + * accessed from NMI. Use a temporary manual per cpu allocation + * until that gets sorted out. + */ + size = sizeof(*entries) + sizeof(struct perf_callchain_entry *) * + num_possible_cpus(); + + entries = kzalloc(size, GFP_KERNEL); + if (!entries) + return -ENOMEM; + + size = sizeof(struct perf_callchain_entry) * 4; + + for_each_possible_cpu(cpu) { + entries->cpu_entries[cpu] = kmalloc_node(size, GFP_KERNEL, + cpu_to_node(cpu)); + if (!entries->cpu_entries[cpu]) + goto fail; + } + + rcu_assign_pointer(callchain_cpus_entries, entries); + + return 0; + +fail: + for_each_possible_cpu(cpu) + kfree(entries->cpu_entries[cpu]); + kfree(entries); + + return -ENOMEM; +} + +static int get_callchain_buffers(void) +{ + int err = 0; + int count; + + mutex_lock(&callchain_mutex); + + count = atomic_inc_return(&nr_callchain_events); + if (WARN_ON_ONCE(count < 1)) { + err = -EINVAL; + goto exit; + } + + if (count > 1) { + /* If the allocation failed, give up */ + if (!callchain_cpus_entries) + err = -ENOMEM; + goto exit; + } + + err = alloc_callchain_buffers(); + if (err) + release_callchain_buffers(); +exit: + mutex_unlock(&callchain_mutex); + + return err; +} + +static void put_callchain_buffers(void) +{ + if (atomic_dec_and_mutex_lock(&nr_callchain_events, &callchain_mutex)) { + release_callchain_buffers(); + mutex_unlock(&callchain_mutex); + } +} + +static int get_recursion_context(int *recursion) +{ + int rctx; + + if (in_nmi()) + rctx = 3; + else if (in_irq()) + rctx = 2; + else if (in_softirq()) + rctx = 1; + else + rctx = 0; + + if (recursion[rctx]) + return -1; + + recursion[rctx]++; + barrier(); + + return rctx; +} + +static inline void put_recursion_context(int *recursion, int rctx) +{ + barrier(); + recursion[rctx]--; +} + +static struct perf_callchain_entry *get_callchain_entry(int *rctx) +{ + int cpu; + struct callchain_cpus_entries *entries; + + *rctx = get_recursion_context(__get_cpu_var(callchain_recursion)); + if (*rctx == -1) + return NULL; + + entries = rcu_dereference(callchain_cpus_entries); + if (!entries) + return NULL; + + cpu = smp_processor_id(); + + return &entries->cpu_entries[cpu][*rctx]; +} + +static void +put_callchain_entry(int rctx) +{ + put_recursion_context(__get_cpu_var(callchain_recursion), rctx); +} + +static struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) +{ + int rctx; + struct perf_callchain_entry *entry; + + + entry = get_callchain_entry(&rctx); + if (rctx == -1) + return NULL; + + if (!entry) + goto exit_put; + + entry->nr = 0; + + if (!user_mode(regs)) { + perf_callchain_store(entry, PERF_CONTEXT_KERNEL); + perf_callchain_kernel(entry, regs); + if (current->mm) + regs = task_pt_regs(current); + else + regs = NULL; + } + + if (regs) { + perf_callchain_store(entry, PERF_CONTEXT_USER); + perf_callchain_user(entry, regs); + } + +exit_put: + put_callchain_entry(rctx); + + return entry; +} + /* * Initialize the perf_event context in a task_struct: */ @@ -1895,6 +2105,8 @@ static void free_event(struct perf_event *event) atomic_dec(&nr_comm_events); if (event->attr.task) atomic_dec(&nr_task_events); + if (event->attr.sample_type & PERF_SAMPLE_CALLCHAIN) + put_callchain_buffers(); } if (event->buffer) { @@ -2937,55 +3149,6 @@ void perf_event_do_pending(void) __perf_pending_run(); } -DEFINE_PER_CPU(struct perf_callchain_entry, perf_callchain_entry); - -/* - * Callchain support -- arch specific - */ - -__weak struct perf_callchain_entry *perf_callchain_buffer(void) -{ - return &__get_cpu_var(perf_callchain_entry); -} - -__weak void perf_callchain_kernel(struct perf_callchain_entry *entry, - struct pt_regs *regs) -{ -} - -__weak void perf_callchain_user(struct perf_callchain_entry *entry, - struct pt_regs *regs) -{ -} - -static struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) -{ - struct perf_callchain_entry *entry; - - entry = perf_callchain_buffer(); - if (!entry) - return NULL; - - entry->nr = 0; - - if (!user_mode(regs)) { - perf_callchain_store(entry, PERF_CONTEXT_KERNEL); - perf_callchain_kernel(entry, regs); - if (current->mm) - regs = task_pt_regs(current); - else - regs = NULL; - } - - if (regs) { - perf_callchain_store(entry, PERF_CONTEXT_USER); - perf_callchain_user(entry, regs); - } - - return entry; -} - - /* * We assume there is only KVM supporting the callbacks. * Later on, we might change it to a list if there is @@ -3480,14 +3643,20 @@ static void perf_event_output(struct perf_event *event, int nmi, struct perf_output_handle handle; struct perf_event_header header; + /* protect the callchain buffers */ + rcu_read_lock(); + perf_prepare_sample(&header, data, event, regs); if (perf_output_begin(&handle, event, header.size, nmi, 1)) - return; + goto exit; perf_output_sample(&handle, &header, data, event); perf_output_end(&handle); + +exit: + rcu_read_unlock(); } /* @@ -4243,32 +4412,16 @@ end: int perf_swevent_get_recursion_context(void) { struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context); - int rctx; - if (in_nmi()) - rctx = 3; - else if (in_irq()) - rctx = 2; - else if (in_softirq()) - rctx = 1; - else - rctx = 0; - - if (cpuctx->recursion[rctx]) - return -1; - - cpuctx->recursion[rctx]++; - barrier(); - - return rctx; + return get_recursion_context(cpuctx->recursion); } EXPORT_SYMBOL_GPL(perf_swevent_get_recursion_context); void inline perf_swevent_put_recursion_context(int rctx) { struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context); - barrier(); - cpuctx->recursion[rctx]--; + + put_recursion_context(cpuctx->recursion, rctx); } void __perf_sw_event(u32 event_id, u64 nr, int nmi, @@ -4968,6 +5121,13 @@ done: atomic_inc(&nr_comm_events); if (event->attr.task) atomic_inc(&nr_task_events); + if (event->attr.sample_type & PERF_SAMPLE_CALLCHAIN) { + err = get_callchain_buffers(); + if (err) { + free_event(event); + return ERR_PTR(err); + } + } } return event; -- cgit v1.2.3-70-g09d2 From ed8052616680e72f58bca678d4d1678cb12a7e47 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 20 Aug 2010 14:30:41 +0200 Subject: perf: Remove superfluous return values from perf_callchain_*() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fixes these build warnings introduced by the callchain rework: arch/x86/kernel/cpu/perf_event.c: In function ‘perf_callchain_kernel’: arch/x86/kernel/cpu/perf_event.c:1646: warning: ‘return’ with a value, in function returning void arch/x86/kernel/cpu/perf_event.c: In function ‘perf_callchain_user’: arch/x86/kernel/cpu/perf_event.c:1699: warning: ‘return’ with a value, in function returning void arch/x86/kernel/cpu/perf_event.c: At top level: arch/x86/kernel/cpu/perf_event.c:1607: warning: ‘perf_callchain_entry_nmi’ defined but not used Signed-off-by: Peter Zijlstra Acked-by: Frederic Weisbecker LKML-Reference: Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 8e91cf34a9c..acc52afaf7b 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -1610,7 +1610,7 @@ perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs) { if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) { /* TODO: We don't support guest os callchain now */ - return NULL; + return; } perf_callchain_store(entry, regs->ip); @@ -1663,7 +1663,7 @@ perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs) if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) { /* TODO: We don't support guest os callchain now */ - return NULL; + return; } fp = (void __user *)regs->bp; -- cgit v1.2.3-70-g09d2 From 61c77326d1df079f202fa79403c3ccd8c5966a81 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Mon, 16 Aug 2010 09:16:55 +0800 Subject: x86, mm: Avoid unnecessary TLB flush In x86, access and dirty bits are set automatically by CPU when CPU accesses memory. When we go into the code path of below flush_tlb_fix_spurious_fault(), we already set dirty bit for pte and don't need flush tlb. This might mean tlb entry in some CPUs hasn't dirty bit set, but this doesn't matter. When the CPUs do page write, they will automatically check the bit and no software involved. On the other hand, flush tlb in below position is harmful. Test creates CPU number of threads, each thread writes to a same but random address in same vma range and we measure the total time. Under a 4 socket system, original time is 1.96s, while with the patch, the time is 0.8s. Under a 2 socket system, there is 20% time cut too. perf shows a lot of time are taking to send ipi/handle ipi for tlb flush. Signed-off-by: Shaohua Li LKML-Reference: <20100816011655.GA362@sli10-desk.sh.intel.com> Acked-by: Suresh Siddha Cc: Andrea Archangeli Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/pgtable.h | 2 ++ include/asm-generic/pgtable.h | 4 ++++ mm/memory.c | 2 +- 3 files changed, 7 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h index a34c785c5a6..2d0a33bd297 100644 --- a/arch/x86/include/asm/pgtable.h +++ b/arch/x86/include/asm/pgtable.h @@ -603,6 +603,8 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, pte_update(mm, addr, ptep); } +#define flush_tlb_fix_spurious_fault(vma, address) + /* * clone_pgd_range(pgd_t *dst, pgd_t *src, int count); * diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h index e2bd73e8f9c..f4d4120e512 100644 --- a/include/asm-generic/pgtable.h +++ b/include/asm-generic/pgtable.h @@ -129,6 +129,10 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addres #define move_pte(pte, prot, old_addr, new_addr) (pte) #endif +#ifndef flush_tlb_fix_spurious_fault +#define flush_tlb_fix_spurious_fault(vma, address) flush_tlb_page(vma, address) +#endif + #ifndef pgprot_noncached #define pgprot_noncached(prot) (prot) #endif diff --git a/mm/memory.c b/mm/memory.c index 2ed2267439d..a40da698396 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -3147,7 +3147,7 @@ static inline int handle_pte_fault(struct mm_struct *mm, * with threads. */ if (flags & FAULT_FLAG_WRITE) - flush_tlb_page(vma, address); + flush_tlb_fix_spurious_fault(vma, address); } unlock: pte_unmap_unlock(pte, ptl); -- cgit v1.2.3-70-g09d2 From fdf4289679fd41d76553ce224750e9737cd80eea Mon Sep 17 00:00:00 2001 From: "Ma, Ling" Date: Mon, 23 Aug 2010 14:11:12 -0700 Subject: x86, mem: Don't implement forward memmove() as memcpy() memmove() allow source and destination address to be overlap, but there is no such limitation for memcpy(). Therefore, explicitly implement memmove() in both the forwards and backward directions, to give us the ability to optimize memcpy(). Signed-off-by: Ma Ling LKML-Reference: Signed-off-by: H. Peter Anvin --- arch/x86/lib/memcpy_32.c | 38 +++++++++++++++++++++++++++----------- arch/x86/lib/memmove_64.c | 46 +++++++++++++++++++++++++++++++++++++++++----- 2 files changed, 68 insertions(+), 16 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/lib/memcpy_32.c b/arch/x86/lib/memcpy_32.c index 5415a9d06f5..be424dfcf36 100644 --- a/arch/x86/lib/memcpy_32.c +++ b/arch/x86/lib/memcpy_32.c @@ -25,19 +25,35 @@ void *memmove(void *dest, const void *src, size_t n) int d0, d1, d2; if (dest < src) { - memcpy(dest, src, n); + if ((dest + n) < src) + return memcpy(dest, src, n); + else + __asm__ __volatile__( + "rep\n\t" + "movsb\n\t" + : "=&c" (d0), "=&S" (d1), "=&D" (d2) + :"0" (n), + "1" (src), + "2" (dest) + :"memory"); + } else { - __asm__ __volatile__( - "std\n\t" - "rep\n\t" - "movsb\n\t" - "cld" - : "=&c" (d0), "=&S" (d1), "=&D" (d2) - :"0" (n), - "1" (n-1+src), - "2" (n-1+dest) - :"memory"); + + if((src + count) < dest) + return memcpy(dest, src, count); + else + __asm__ __volatile__( + "std\n\t" + "rep\n\t" + "movsb\n\t" + "cld" + : "=&c" (d0), "=&S" (d1), "=&D" (d2) + :"0" (n), + "1" (n-1+src), + "2" (n-1+dest) + :"memory"); } + return dest; } EXPORT_SYMBOL(memmove); diff --git a/arch/x86/lib/memmove_64.c b/arch/x86/lib/memmove_64.c index 0a33909bf12..ecacc4b3d9e 100644 --- a/arch/x86/lib/memmove_64.c +++ b/arch/x86/lib/memmove_64.c @@ -8,13 +8,49 @@ #undef memmove void *memmove(void *dest, const void *src, size_t count) { + unsigned long d0, d1, d2, d3; if (dest < src) { - return memcpy(dest, src, count); + if ((dest + count) < src) + return memcpy(dest, src, count); + else + __asm__ __volatile__( + "movq %0, %3\n\t" + "shr $3, %0\n\t" + "andq $7, %3\n\t" + "rep\n\t" + "movsq\n\t" + "movq %3, %0\n\t" + "rep\n\t" + "movsb" + : "=&c" (d0), "=&S" (d1), "=&D" (d2), "=r" (d3) + :"0" (count), + "1" (src), + "2" (dest) + :"memory"); } else { - char *p = dest + count; - const char *s = src + count; - while (count--) - *--p = *--s; + if((src + count) < dest) + return memcpy(dest, src, count); + else + __asm__ __volatile__( + "movq %0, %3\n\t" + "lea -8(%1, %0), %1\n\t" + "lea -8(%2, %0), %2\n\t" + "shr $3, %0\n\t" + "andq $7, %3\n\t" + "std\n\t" + "rep\n\t" + "movsq\n\t" + "lea 7(%1), %1\n\t" + "lea 7(%2), %2\n\t" + "movq %3, %0\n\t" + "rep\n\t" + "movsb\n\t" + "cld" + : "=&c" (d0), "=&S" (d1), "=&D" (d2), "=r" (d3) + :"0" (count), + "1" (src), + "2" (dest) + :"memory"); } return dest; } -- cgit v1.2.3-70-g09d2 From 59daa706fbec745684702741b9f5373142dd9fdc Mon Sep 17 00:00:00 2001 From: Ma Ling Date: Tue, 29 Jun 2010 03:24:25 +0800 Subject: x86, mem: Optimize memcpy by avoiding memory false dependece All read operations after allocation stage can run speculatively, all write operation will run in program order, and if addresses are different read may run before older write operation, otherwise wait until write commit. However CPU don't check each address bit, so read could fail to recognize different address even they are in different page.For example if rsi is 0xf004, rdi is 0xe008, in following operation there will generate big performance latency. 1. movq (%rsi), %rax 2. movq %rax, (%rdi) 3. movq 8(%rsi), %rax 4. movq %rax, 8(%rdi) If %rsi and rdi were in really the same meory page, there are TRUE read-after-write dependence because instruction 2 write 0x008 and instruction 3 read 0x00c, the two address are overlap partially. Actually there are in different page and no any issues, but without checking each address bit CPU could think they are in the same page, and instruction 3 have to wait for instruction 2 to write data into cache from write buffer, then load data from cache, the cost time read spent is equal to mfence instruction. We may avoid it by tuning operation sequence as follow. 1. movq 8(%rsi), %rax 2. movq %rax, 8(%rdi) 3. movq (%rsi), %rax 4. movq %rax, (%rdi) Instruction 3 read 0x004, instruction 2 write address 0x010, no any dependence. At last on Core2 we gain 1.83x speedup compared with original instruction sequence. In this patch we first handle small size(less 20bytes), then jump to different copy mode. Based on our micro-benchmark small bytes from 1 to 127 bytes, we got up to 2X improvement, and up to 1.5X improvement for 1024 bytes on Corei7. (We use our micro-benchmark, and will do further test according to your requirment) Signed-off-by: Ma Ling LKML-Reference: <1277753065-18610-1-git-send-email-ling.ma@intel.com> Signed-off-by: H. Peter Anvin --- arch/x86/lib/memcpy_32.c | 6 +- arch/x86/lib/memcpy_64.S | 158 ++++++++++++++++++++++++++++++----------------- 2 files changed, 105 insertions(+), 59 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/lib/memcpy_32.c b/arch/x86/lib/memcpy_32.c index be424dfcf36..81130d477ee 100644 --- a/arch/x86/lib/memcpy_32.c +++ b/arch/x86/lib/memcpy_32.c @@ -36,11 +36,9 @@ void *memmove(void *dest, const void *src, size_t n) "1" (src), "2" (dest) :"memory"); - } else { - - if((src + count) < dest) - return memcpy(dest, src, count); + if((src + n) < dest) + return memcpy(dest, src, n); else __asm__ __volatile__( "std\n\t" diff --git a/arch/x86/lib/memcpy_64.S b/arch/x86/lib/memcpy_64.S index bcbcd1e0f7d..75ef61e35e3 100644 --- a/arch/x86/lib/memcpy_64.S +++ b/arch/x86/lib/memcpy_64.S @@ -40,84 +40,132 @@ ENTRY(__memcpy) ENTRY(memcpy) CFI_STARTPROC + movq %rdi, %rax /* - * Put the number of full 64-byte blocks into %ecx. - * Tail portion is handled at the end: + * Use 32bit CMP here to avoid long NOP padding. */ - movq %rdi, %rax - movl %edx, %ecx - shrl $6, %ecx - jz .Lhandle_tail + cmp $0x20, %edx + jb .Lhandle_tail - .p2align 4 -.Lloop_64: /* - * We decrement the loop index here - and the zero-flag is - * checked at the end of the loop (instructions inbetween do - * not change the zero flag): + * We check whether memory false dependece could occur, + * then jump to corresponding copy mode. */ - decl %ecx + cmp %dil, %sil + jl .Lcopy_backward + subl $0x20, %edx +.Lcopy_forward_loop: + subq $0x20, %rdx /* - * Move in blocks of 4x16 bytes: + * Move in blocks of 4x8 bytes: */ - movq 0*8(%rsi), %r11 - movq 1*8(%rsi), %r8 - movq %r11, 0*8(%rdi) - movq %r8, 1*8(%rdi) - - movq 2*8(%rsi), %r9 - movq 3*8(%rsi), %r10 - movq %r9, 2*8(%rdi) - movq %r10, 3*8(%rdi) - - movq 4*8(%rsi), %r11 - movq 5*8(%rsi), %r8 - movq %r11, 4*8(%rdi) - movq %r8, 5*8(%rdi) - - movq 6*8(%rsi), %r9 - movq 7*8(%rsi), %r10 - movq %r9, 6*8(%rdi) - movq %r10, 7*8(%rdi) - - leaq 64(%rsi), %rsi - leaq 64(%rdi), %rdi - - jnz .Lloop_64 + movq 0*8(%rsi), %r8 + movq 1*8(%rsi), %r9 + movq 2*8(%rsi), %r10 + movq 3*8(%rsi), %r11 + leaq 4*8(%rsi), %rsi + + movq %r8, 0*8(%rdi) + movq %r9, 1*8(%rdi) + movq %r10, 2*8(%rdi) + movq %r11, 3*8(%rdi) + leaq 4*8(%rdi), %rdi + jae .Lcopy_forward_loop + addq $0x20, %rdx + jmp .Lhandle_tail + +.Lcopy_backward: + /* + * Calculate copy position to tail. + */ + addq %rdx, %rsi + addq %rdx, %rdi + subq $0x20, %rdx + /* + * At most 3 ALU operations in one cycle, + * so append NOPS in the same 16bytes trunk. + */ + .p2align 4 +.Lcopy_backward_loop: + subq $0x20, %rdx + movq -1*8(%rsi), %r8 + movq -2*8(%rsi), %r9 + movq -3*8(%rsi), %r10 + movq -4*8(%rsi), %r11 + leaq -4*8(%rsi), %rsi + movq %r8, -1*8(%rdi) + movq %r9, -2*8(%rdi) + movq %r10, -3*8(%rdi) + movq %r11, -4*8(%rdi) + leaq -4*8(%rdi), %rdi + jae .Lcopy_backward_loop + /* + * Calculate copy position to head. + */ + addq $0x20, %rdx + subq %rdx, %rsi + subq %rdx, %rdi .Lhandle_tail: - movl %edx, %ecx - andl $63, %ecx - shrl $3, %ecx - jz .Lhandle_7 + cmpq $16, %rdx + jb .Lless_16bytes + /* + * Move data from 16 bytes to 31 bytes. + */ + movq 0*8(%rsi), %r8 + movq 1*8(%rsi), %r9 + movq -2*8(%rsi, %rdx), %r10 + movq -1*8(%rsi, %rdx), %r11 + movq %r8, 0*8(%rdi) + movq %r9, 1*8(%rdi) + movq %r10, -2*8(%rdi, %rdx) + movq %r11, -1*8(%rdi, %rdx) + retq .p2align 4 -.Lloop_8: - decl %ecx - movq (%rsi), %r8 - movq %r8, (%rdi) - leaq 8(%rdi), %rdi - leaq 8(%rsi), %rsi - jnz .Lloop_8 - -.Lhandle_7: - movl %edx, %ecx - andl $7, %ecx - jz .Lend +.Lless_16bytes: + cmpq $8, %rdx + jb .Lless_8bytes + /* + * Move data from 8 bytes to 15 bytes. + */ + movq 0*8(%rsi), %r8 + movq -1*8(%rsi, %rdx), %r9 + movq %r8, 0*8(%rdi) + movq %r9, -1*8(%rdi, %rdx) + retq + .p2align 4 +.Lless_8bytes: + cmpq $4, %rdx + jb .Lless_3bytes + /* + * Move data from 4 bytes to 7 bytes. + */ + movl (%rsi), %ecx + movl -4(%rsi, %rdx), %r8d + movl %ecx, (%rdi) + movl %r8d, -4(%rdi, %rdx) + retq .p2align 4 +.Lless_3bytes: + cmpl $0, %edx + je .Lend + /* + * Move data from 1 bytes to 3 bytes. + */ .Lloop_1: movb (%rsi), %r8b movb %r8b, (%rdi) incq %rdi incq %rsi - decl %ecx + decl %edx jnz .Lloop_1 .Lend: - ret + retq CFI_ENDPROC ENDPROC(memcpy) ENDPROC(__memcpy) -- cgit v1.2.3-70-g09d2 From 9863c90f682fba34cdc26c3437e8c00da6c83fa4 Mon Sep 17 00:00:00 2001 From: Alok Kataria Date: Mon, 23 Aug 2010 14:49:11 -0700 Subject: x86, vmware: Remove deprecated VMI kernel support With the recent innovations in CPU hardware acceleration technologies from Intel and AMD, VMware ran a few experiments to compare these techniques to guest paravirtualization technique on VMware's platform. These hardware assisted virtualization techniques have outperformed the performance benefits provided by VMI in most of the workloads. VMware expects that these hardware features will be ubiquitous in a couple of years, as a result, VMware has started a phased retirement of this feature from the hypervisor. Please note that VMI has always been an optimization and non-VMI kernels still work fine on VMware's platform. Latest versions of VMware's product which support VMI are, Workstation 7.0 and VSphere 4.0 on ESX side, future maintainence releases for these products will continue supporting VMI. For more details about VMI retirement take a look at this, http://blogs.vmware.com/guestosguide/2009/09/vmi-retirement.html This feature removal was scheduled for 2.6.37 back in September 2009. Signed-off-by: Alok N Kataria LKML-Reference: <1282600151.19396.22.camel@ank32.eng.vmware.com> Signed-off-by: H. Peter Anvin --- Documentation/feature-removal-schedule.txt | 28 - Documentation/kernel-parameters.txt | 2 +- arch/x86/Kconfig | 19 - arch/x86/include/asm/vmi.h | 269 --------- arch/x86/include/asm/vmi_time.h | 98 ---- arch/x86/kernel/Makefile | 1 - arch/x86/kernel/setup.c | 12 +- arch/x86/kernel/smpboot.c | 2 - arch/x86/kernel/vmi_32.c | 893 ----------------------------- arch/x86/kernel/vmiclock_32.c | 317 ---------- 10 files changed, 5 insertions(+), 1636 deletions(-) delete mode 100644 arch/x86/include/asm/vmi.h delete mode 100644 arch/x86/include/asm/vmi_time.h delete mode 100644 arch/x86/kernel/vmi_32.c delete mode 100644 arch/x86/kernel/vmiclock_32.c (limited to 'arch/x86') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 842aa9de84a..5e2bc4ab897 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -386,34 +386,6 @@ Who: Tejun Heo ---------------------------- -What: Support for VMware's guest paravirtuliazation technique [VMI] will be - dropped. -When: 2.6.37 or earlier. -Why: With the recent innovations in CPU hardware acceleration technologies - from Intel and AMD, VMware ran a few experiments to compare these - techniques to guest paravirtualization technique on VMware's platform. - These hardware assisted virtualization techniques have outperformed the - performance benefits provided by VMI in most of the workloads. VMware - expects that these hardware features will be ubiquitous in a couple of - years, as a result, VMware has started a phased retirement of this - feature from the hypervisor. We will be removing this feature from the - Kernel too. Right now we are targeting 2.6.37 but can retire earlier if - technical reasons (read opportunity to remove major chunk of pvops) - arise. - - Please note that VMI has always been an optimization and non-VMI kernels - still work fine on VMware's platform. - Latest versions of VMware's product which support VMI are, - Workstation 7.0 and VSphere 4.0 on ESX side, future maintainence - releases for these products will continue supporting VMI. - - For more details about VMI retirement take a look at this, - http://blogs.vmware.com/guestosguide/2009/09/vmi-retirement.html - -Who: Alok N Kataria - ----------------------------- - What: Support for lcd_switch and display_get in asus-laptop driver When: March 2010 Why: These two features use non-standard interfaces. There are the diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 2c85c0692b0..582409801a4 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -455,7 +455,7 @@ and is between 256 and 4096 characters. It is defined in the file [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2, pxa_timer,timer3,32k_counter,timer0_1 [AVR32] avr32 - [X86-32] pit,hpet,tsc,vmi-timer; + [X86-32] pit,hpet,tsc; scx200_hrt on Geode; cyclone on IBM x440 [MIPS] MIPS [PARISC] cr16 diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index cea0cd9a316..f0ee331feea 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -517,25 +517,6 @@ if PARAVIRT_GUEST source "arch/x86/xen/Kconfig" -config VMI - bool "VMI Guest support (DEPRECATED)" - select PARAVIRT - depends on X86_32 - ---help--- - VMI provides a paravirtualized interface to the VMware ESX server - (it could be used by other hypervisors in theory too, but is not - at the moment), by linking the kernel to a GPL-ed ROM module - provided by the hypervisor. - - As of September 2009, VMware has started a phased retirement - of this feature from VMware's products. Please see - feature-removal-schedule.txt for details. If you are - planning to enable this option, please note that you cannot - live migrate a VMI enabled VM to a future VMware product, - which doesn't support VMI. So if you expect your kernel to - seamlessly migrate to newer VMware products, keep this - disabled. - config KVM_CLOCK bool "KVM paravirtualized clock" select PARAVIRT diff --git a/arch/x86/include/asm/vmi.h b/arch/x86/include/asm/vmi.h deleted file mode 100644 index 61e08c0a290..00000000000 --- a/arch/x86/include/asm/vmi.h +++ /dev/null @@ -1,269 +0,0 @@ -/* - * VMI interface definition - * - * Copyright (C) 2005, VMware, 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 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, GOOD TITLE or - * NON INFRINGEMENT. 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., 675 Mass Ave, Cambridge, MA 02139, USA. - * - * Maintained by: Zachary Amsden zach@vmware.com - * - */ -#include - -/* - *--------------------------------------------------------------------- - * - * VMI Option ROM API - * - *--------------------------------------------------------------------- - */ -#define VMI_SIGNATURE 0x696d5663 /* "cVmi" */ - -#define PCI_VENDOR_ID_VMWARE 0x15AD -#define PCI_DEVICE_ID_VMWARE_VMI 0x0801 - -/* - * We use two version numbers for compatibility, with the major - * number signifying interface breakages, and the minor number - * interface extensions. - */ -#define VMI_API_REV_MAJOR 3 -#define VMI_API_REV_MINOR 0 - -#define VMI_CALL_CPUID 0 -#define VMI_CALL_WRMSR 1 -#define VMI_CALL_RDMSR 2 -#define VMI_CALL_SetGDT 3 -#define VMI_CALL_SetLDT 4 -#define VMI_CALL_SetIDT 5 -#define VMI_CALL_SetTR 6 -#define VMI_CALL_GetGDT 7 -#define VMI_CALL_GetLDT 8 -#define VMI_CALL_GetIDT 9 -#define VMI_CALL_GetTR 10 -#define VMI_CALL_WriteGDTEntry 11 -#define VMI_CALL_WriteLDTEntry 12 -#define VMI_CALL_WriteIDTEntry 13 -#define VMI_CALL_UpdateKernelStack 14 -#define VMI_CALL_SetCR0 15 -#define VMI_CALL_SetCR2 16 -#define VMI_CALL_SetCR3 17 -#define VMI_CALL_SetCR4 18 -#define VMI_CALL_GetCR0 19 -#define VMI_CALL_GetCR2 20 -#define VMI_CALL_GetCR3 21 -#define VMI_CALL_GetCR4 22 -#define VMI_CALL_WBINVD 23 -#define VMI_CALL_SetDR 24 -#define VMI_CALL_GetDR 25 -#define VMI_CALL_RDPMC 26 -#define VMI_CALL_RDTSC 27 -#define VMI_CALL_CLTS 28 -#define VMI_CALL_EnableInterrupts 29 -#define VMI_CALL_DisableInterrupts 30 -#define VMI_CALL_GetInterruptMask 31 -#define VMI_CALL_SetInterruptMask 32 -#define VMI_CALL_IRET 33 -#define VMI_CALL_SYSEXIT 34 -#define VMI_CALL_Halt 35 -#define VMI_CALL_Reboot 36 -#define VMI_CALL_Shutdown 37 -#define VMI_CALL_SetPxE 38 -#define VMI_CALL_SetPxELong 39 -#define VMI_CALL_UpdatePxE 40 -#define VMI_CALL_UpdatePxELong 41 -#define VMI_CALL_MachineToPhysical 42 -#define VMI_CALL_PhysicalToMachine 43 -#define VMI_CALL_AllocatePage 44 -#define VMI_CALL_ReleasePage 45 -#define VMI_CALL_InvalPage 46 -#define VMI_CALL_FlushTLB 47 -#define VMI_CALL_SetLinearMapping 48 - -#define VMI_CALL_SetIOPLMask 61 -#define VMI_CALL_SetInitialAPState 62 -#define VMI_CALL_APICWrite 63 -#define VMI_CALL_APICRead 64 -#define VMI_CALL_IODelay 65 -#define VMI_CALL_SetLazyMode 73 - -/* - *--------------------------------------------------------------------- - * - * MMU operation flags - * - *--------------------------------------------------------------------- - */ - -/* Flags used by VMI_{Allocate|Release}Page call */ -#define VMI_PAGE_PAE 0x10 /* Allocate PAE shadow */ -#define VMI_PAGE_CLONE 0x20 /* Clone from another shadow */ -#define VMI_PAGE_ZEROED 0x40 /* Page is pre-zeroed */ - - -/* Flags shared by Allocate|Release Page and PTE updates */ -#define VMI_PAGE_PT 0x01 -#define VMI_PAGE_PD 0x02 -#define VMI_PAGE_PDP 0x04 -#define VMI_PAGE_PML4 0x08 - -#define VMI_PAGE_NORMAL 0x00 /* for debugging */ - -/* Flags used by PTE updates */ -#define VMI_PAGE_CURRENT_AS 0x10 /* implies VMI_PAGE_VA_MASK is valid */ -#define VMI_PAGE_DEFER 0x20 /* may queue update until TLB inval */ -#define VMI_PAGE_VA_MASK 0xfffff000 - -#ifdef CONFIG_X86_PAE -#define VMI_PAGE_L1 (VMI_PAGE_PT | VMI_PAGE_PAE | VMI_PAGE_ZEROED) -#define VMI_PAGE_L2 (VMI_PAGE_PD | VMI_PAGE_PAE | VMI_PAGE_ZEROED) -#else -#define VMI_PAGE_L1 (VMI_PAGE_PT | VMI_PAGE_ZEROED) -#define VMI_PAGE_L2 (VMI_PAGE_PD | VMI_PAGE_ZEROED) -#endif - -/* Flags used by VMI_FlushTLB call */ -#define VMI_FLUSH_TLB 0x01 -#define VMI_FLUSH_GLOBAL 0x02 - -/* - *--------------------------------------------------------------------- - * - * VMI relocation definitions for ROM call get_reloc - * - *--------------------------------------------------------------------- - */ - -/* VMI Relocation types */ -#define VMI_RELOCATION_NONE 0 -#define VMI_RELOCATION_CALL_REL 1 -#define VMI_RELOCATION_JUMP_REL 2 -#define VMI_RELOCATION_NOP 3 - -#ifndef __ASSEMBLY__ -struct vmi_relocation_info { - unsigned char *eip; - unsigned char type; - unsigned char reserved[3]; -}; -#endif - - -/* - *--------------------------------------------------------------------- - * - * Generic ROM structures and definitions - * - *--------------------------------------------------------------------- - */ - -#ifndef __ASSEMBLY__ - -struct vrom_header { - u16 rom_signature; /* option ROM signature */ - u8 rom_length; /* ROM length in 512 byte chunks */ - u8 rom_entry[4]; /* 16-bit code entry point */ - u8 rom_pad0; /* 4-byte align pad */ - u32 vrom_signature; /* VROM identification signature */ - u8 api_version_min;/* Minor version of API */ - u8 api_version_maj;/* Major version of API */ - u8 jump_slots; /* Number of jump slots */ - u8 reserved1; /* Reserved for expansion */ - u32 virtual_top; /* Hypervisor virtual address start */ - u16 reserved2; /* Reserved for expansion */ - u16 license_offs; /* Offset to License string */ - u16 pci_header_offs;/* Offset to PCI OPROM header */ - u16 pnp_header_offs;/* Offset to PnP OPROM header */ - u32 rom_pad3; /* PnP reserverd / VMI reserved */ - u8 reserved[96]; /* Reserved for headers */ - char vmi_init[8]; /* VMI_Init jump point */ - char get_reloc[8]; /* VMI_GetRelocationInfo jump point */ -} __attribute__((packed)); - -struct pnp_header { - char sig[4]; - char rev; - char size; - short next; - short res; - long devID; - unsigned short manufacturer_offset; - unsigned short product_offset; -} __attribute__((packed)); - -struct pci_header { - char sig[4]; - short vendorID; - short deviceID; - short vpdData; - short size; - char rev; - char class; - char subclass; - char interface; - short chunks; - char rom_version_min; - char rom_version_maj; - char codetype; - char lastRom; - short reserved; -} __attribute__((packed)); - -/* Function prototypes for bootstrapping */ -#ifdef CONFIG_VMI -extern void vmi_init(void); -extern void vmi_activate(void); -extern void vmi_bringup(void); -#else -static inline void vmi_init(void) {} -static inline void vmi_activate(void) {} -static inline void vmi_bringup(void) {} -#endif - -/* State needed to start an application processor in an SMP system. */ -struct vmi_ap_state { - u32 cr0; - u32 cr2; - u32 cr3; - u32 cr4; - - u64 efer; - - u32 eip; - u32 eflags; - u32 eax; - u32 ebx; - u32 ecx; - u32 edx; - u32 esp; - u32 ebp; - u32 esi; - u32 edi; - u16 cs; - u16 ss; - u16 ds; - u16 es; - u16 fs; - u16 gs; - u16 ldtr; - - u16 gdtr_limit; - u32 gdtr_base; - u32 idtr_base; - u16 idtr_limit; -}; - -#endif diff --git a/arch/x86/include/asm/vmi_time.h b/arch/x86/include/asm/vmi_time.h deleted file mode 100644 index c6e0bee93e3..00000000000 --- a/arch/x86/include/asm/vmi_time.h +++ /dev/null @@ -1,98 +0,0 @@ -/* - * VMI Time wrappers - * - * Copyright (C) 2006, VMware, 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 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, GOOD TITLE or - * NON INFRINGEMENT. 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., 675 Mass Ave, Cambridge, MA 02139, USA. - * - * Send feedback to dhecht@vmware.com - * - */ - -#ifndef _ASM_X86_VMI_TIME_H -#define _ASM_X86_VMI_TIME_H - -/* - * Raw VMI call indices for timer functions - */ -#define VMI_CALL_GetCycleFrequency 66 -#define VMI_CALL_GetCycleCounter 67 -#define VMI_CALL_SetAlarm 68 -#define VMI_CALL_CancelAlarm 69 -#define VMI_CALL_GetWallclockTime 70 -#define VMI_CALL_WallclockUpdated 71 - -/* Cached VMI timer operations */ -extern struct vmi_timer_ops { - u64 (*get_cycle_frequency)(void); - u64 (*get_cycle_counter)(int); - u64 (*get_wallclock)(void); - int (*wallclock_updated)(void); - void (*set_alarm)(u32 flags, u64 expiry, u64 period); - void (*cancel_alarm)(u32 flags); -} vmi_timer_ops; - -/* Prototypes */ -extern void __init vmi_time_init(void); -extern unsigned long vmi_get_wallclock(void); -extern int vmi_set_wallclock(unsigned long now); -extern unsigned long long vmi_sched_clock(void); -extern unsigned long vmi_tsc_khz(void); - -#ifdef CONFIG_X86_LOCAL_APIC -extern void __devinit vmi_time_bsp_init(void); -extern void __devinit vmi_time_ap_init(void); -#endif - -/* - * When run under a hypervisor, a vcpu is always in one of three states: - * running, halted, or ready. The vcpu is in the 'running' state if it - * is executing. When the vcpu executes the halt interface, the vcpu - * enters the 'halted' state and remains halted until there is some work - * pending for the vcpu (e.g. an alarm expires, host I/O completes on - * behalf of virtual I/O). At this point, the vcpu enters the 'ready' - * state (waiting for the hypervisor to reschedule it). Finally, at any - * time when the vcpu is not in the 'running' state nor the 'halted' - * state, it is in the 'ready' state. - * - * Real time is advances while the vcpu is 'running', 'ready', or - * 'halted'. Stolen time is the time in which the vcpu is in the - * 'ready' state. Available time is the remaining time -- the vcpu is - * either 'running' or 'halted'. - * - * All three views of time are accessible through the VMI cycle - * counters. - */ - -/* The cycle counters. */ -#define VMI_CYCLES_REAL 0 -#define VMI_CYCLES_AVAILABLE 1 -#define VMI_CYCLES_STOLEN 2 - -/* The alarm interface 'flags' bits */ -#define VMI_ALARM_COUNTERS 2 - -#define VMI_ALARM_COUNTER_MASK 0x000000ff - -#define VMI_ALARM_WIRED_IRQ0 0x00000000 -#define VMI_ALARM_WIRED_LVTT 0x00010000 - -#define VMI_ALARM_IS_ONESHOT 0x00000000 -#define VMI_ALARM_IS_PERIODIC 0x00000100 - -#define CONFIG_VMI_ALARM_HZ 100 - -#endif /* _ASM_X86_VMI_TIME_H */ diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 0925676266b..801127cd975 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -91,7 +91,6 @@ obj-$(CONFIG_K8_NB) += k8.o obj-$(CONFIG_DEBUG_RODATA_TEST) += test_rodata.o obj-$(CONFIG_DEBUG_NX_TEST) += test_nx.o -obj-$(CONFIG_VMI) += vmi_32.o vmiclock_32.o obj-$(CONFIG_KVM_GUEST) += kvm.o obj-$(CONFIG_KVM_CLOCK) += kvmclock.o obj-$(CONFIG_PARAVIRT) += paravirt.o paravirt_patch_$(BITS).o diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index c3a4fbb2b99..feb4c21e499 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -83,7 +83,6 @@ #include #include #include -#include #include #include #include @@ -734,10 +733,10 @@ void __init setup_arch(char **cmdline_p) printk(KERN_INFO "Command line: %s\n", boot_command_line); #endif - /* VMI may relocate the fixmap; do this before touching ioremap area */ - vmi_init(); - - /* OFW also may relocate the fixmap */ + /* + * If we have OLPC OFW, we might end up relocating the fixmap due to + * reserve_top(), so do this before touching the ioremap area. + */ olpc_ofw_detect(); early_trap_init(); @@ -838,9 +837,6 @@ void __init setup_arch(char **cmdline_p) x86_report_nx(); - /* Must be before kernel pagetables are setup */ - vmi_activate(); - /* after early param, so could get panic from serial */ reserve_early_setup_data(); diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 8b3bfc4dd70..63a1a5596ac 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -62,7 +62,6 @@ #include #include #include -#include #include #include #include @@ -311,7 +310,6 @@ notrace static void __cpuinit start_secondary(void *unused) __flush_tlb_all(); #endif - vmi_bringup(); cpu_init(); preempt_disable(); smp_callin(); diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c deleted file mode 100644 index ce9fbacb752..00000000000 --- a/arch/x86/kernel/vmi_32.c +++ /dev/null @@ -1,893 +0,0 @@ -/* - * VMI specific paravirt-ops implementation - * - * Copyright (C) 2005, VMware, 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 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, GOOD TITLE or - * NON INFRINGEMENT. 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., 675 Mass Ave, Cambridge, MA 02139, USA. - * - * Send feedback to zach@vmware.com - * - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -/* Convenient for calling VMI functions indirectly in the ROM */ -typedef u32 __attribute__((regparm(1))) (VROMFUNC)(void); -typedef u64 __attribute__((regparm(2))) (VROMLONGFUNC)(int); - -#define call_vrom_func(rom,func) \ - (((VROMFUNC *)(rom->func))()) - -#define call_vrom_long_func(rom,func,arg) \ - (((VROMLONGFUNC *)(rom->func)) (arg)) - -static struct vrom_header *vmi_rom; -static int disable_pge; -static int disable_pse; -static int disable_sep; -static int disable_tsc; -static int disable_mtrr; -static int disable_noidle; -static int disable_vmi_timer; - -/* Cached VMI operations */ -static struct { - void (*cpuid)(void /* non-c */); - void (*_set_ldt)(u32 selector); - void (*set_tr)(u32 selector); - void (*write_idt_entry)(struct desc_struct *, int, u32, u32); - void (*write_gdt_entry)(struct desc_struct *, int, u32, u32); - void (*write_ldt_entry)(struct desc_struct *, int, u32, u32); - void (*set_kernel_stack)(u32 selector, u32 sp0); - void (*allocate_page)(u32, u32, u32, u32, u32); - void (*release_page)(u32, u32); - void (*set_pte)(pte_t, pte_t *, unsigned); - void (*update_pte)(pte_t *, unsigned); - void (*set_linear_mapping)(int, void *, u32, u32); - void (*_flush_tlb)(int); - void (*set_initial_ap_state)(int, int); - void (*halt)(void); - void (*set_lazy_mode)(int mode); -} vmi_ops; - -/* Cached VMI operations */ -struct vmi_timer_ops vmi_timer_ops; - -/* - * VMI patching routines. - */ -#define MNEM_CALL 0xe8 -#define MNEM_JMP 0xe9 -#define MNEM_RET 0xc3 - -#define IRQ_PATCH_INT_MASK 0 -#define IRQ_PATCH_DISABLE 5 - -static inline void patch_offset(void *insnbuf, - unsigned long ip, unsigned long dest) -{ - *(unsigned long *)(insnbuf+1) = dest-ip-5; -} - -static unsigned patch_internal(int call, unsigned len, void *insnbuf, - unsigned long ip) -{ - u64 reloc; - struct vmi_relocation_info *const rel = (struct vmi_relocation_info *)&reloc; - reloc = call_vrom_long_func(vmi_rom, get_reloc, call); - switch(rel->type) { - case VMI_RELOCATION_CALL_REL: - BUG_ON(len < 5); - *(char *)insnbuf = MNEM_CALL; - patch_offset(insnbuf, ip, (unsigned long)rel->eip); - return 5; - - case VMI_RELOCATION_JUMP_REL: - BUG_ON(len < 5); - *(char *)insnbuf = MNEM_JMP; - patch_offset(insnbuf, ip, (unsigned long)rel->eip); - return 5; - - case VMI_RELOCATION_NOP: - /* obliterate the whole thing */ - return 0; - - case VMI_RELOCATION_NONE: - /* leave native code in place */ - break; - - default: - BUG(); - } - return len; -} - -/* - * Apply patch if appropriate, return length of new instruction - * sequence. The callee does nop padding for us. - */ -static unsigned vmi_patch(u8 type, u16 clobbers, void *insns, - unsigned long ip, unsigned len) -{ - switch (type) { - case PARAVIRT_PATCH(pv_irq_ops.irq_disable): - return patch_internal(VMI_CALL_DisableInterrupts, len, - insns, ip); - case PARAVIRT_PATCH(pv_irq_ops.irq_enable): - return patch_internal(VMI_CALL_EnableInterrupts, len, - insns, ip); - case PARAVIRT_PATCH(pv_irq_ops.restore_fl): - return patch_internal(VMI_CALL_SetInterruptMask, len, - insns, ip); - case PARAVIRT_PATCH(pv_irq_ops.save_fl): - return patch_internal(VMI_CALL_GetInterruptMask, len, - insns, ip); - case PARAVIRT_PATCH(pv_cpu_ops.iret): - return patch_internal(VMI_CALL_IRET, len, insns, ip); - case PARAVIRT_PATCH(pv_cpu_ops.irq_enable_sysexit): - return patch_internal(VMI_CALL_SYSEXIT, len, insns, ip); - default: - break; - } - return len; -} - -/* CPUID has non-C semantics, and paravirt-ops API doesn't match hardware ISA */ -static void vmi_cpuid(unsigned int *ax, unsigned int *bx, - unsigned int *cx, unsigned int *dx) -{ - int override = 0; - if (*ax == 1) - override = 1; - asm volatile ("call *%6" - : "=a" (*ax), - "=b" (*bx), - "=c" (*cx), - "=d" (*dx) - : "0" (*ax), "2" (*cx), "r" (vmi_ops.cpuid)); - if (override) { - if (disable_pse) - *dx &= ~X86_FEATURE_PSE; - if (disable_pge) - *dx &= ~X86_FEATURE_PGE; - if (disable_sep) - *dx &= ~X86_FEATURE_SEP; - if (disable_tsc) - *dx &= ~X86_FEATURE_TSC; - if (disable_mtrr) - *dx &= ~X86_FEATURE_MTRR; - } -} - -static inline void vmi_maybe_load_tls(struct desc_struct *gdt, int nr, struct desc_struct *new) -{ - if (gdt[nr].a != new->a || gdt[nr].b != new->b) - write_gdt_entry(gdt, nr, new, 0); -} - -static void vmi_load_tls(struct thread_struct *t, unsigned int cpu) -{ - struct desc_struct *gdt = get_cpu_gdt_table(cpu); - vmi_maybe_load_tls(gdt, GDT_ENTRY_TLS_MIN + 0, &t->tls_array[0]); - vmi_maybe_load_tls(gdt, GDT_ENTRY_TLS_MIN + 1, &t->tls_array[1]); - vmi_maybe_load_tls(gdt, GDT_ENTRY_TLS_MIN + 2, &t->tls_array[2]); -} - -static void vmi_set_ldt(const void *addr, unsigned entries) -{ - unsigned cpu = smp_processor_id(); - struct desc_struct desc; - - pack_descriptor(&desc, (unsigned long)addr, - entries * sizeof(struct desc_struct) - 1, - DESC_LDT, 0); - write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_LDT, &desc, DESC_LDT); - vmi_ops._set_ldt(entries ? GDT_ENTRY_LDT*sizeof(struct desc_struct) : 0); -} - -static void vmi_set_tr(void) -{ - vmi_ops.set_tr(GDT_ENTRY_TSS*sizeof(struct desc_struct)); -} - -static void vmi_write_idt_entry(gate_desc *dt, int entry, const gate_desc *g) -{ - u32 *idt_entry = (u32 *)g; - vmi_ops.write_idt_entry(dt, entry, idt_entry[0], idt_entry[1]); -} - -static void vmi_write_gdt_entry(struct desc_struct *dt, int entry, - const void *desc, int type) -{ - u32 *gdt_entry = (u32 *)desc; - vmi_ops.write_gdt_entry(dt, entry, gdt_entry[0], gdt_entry[1]); -} - -static void vmi_write_ldt_entry(struct desc_struct *dt, int entry, - const void *desc) -{ - u32 *ldt_entry = (u32 *)desc; - vmi_ops.write_ldt_entry(dt, entry, ldt_entry[0], ldt_entry[1]); -} - -static void vmi_load_sp0(struct tss_struct *tss, - struct thread_struct *thread) -{ - tss->x86_tss.sp0 = thread->sp0; - - /* This can only happen when SEP is enabled, no need to test "SEP"arately */ - if (unlikely(tss->x86_tss.ss1 != thread->sysenter_cs)) { - tss->x86_tss.ss1 = thread->sysenter_cs; - wrmsr(MSR_IA32_SYSENTER_CS, thread->sysenter_cs, 0); - } - vmi_ops.set_kernel_stack(__KERNEL_DS, tss->x86_tss.sp0); -} - -static void vmi_flush_tlb_user(void) -{ - vmi_ops._flush_tlb(VMI_FLUSH_TLB); -} - -static void vmi_flush_tlb_kernel(void) -{ - vmi_ops._flush_tlb(VMI_FLUSH_TLB | VMI_FLUSH_GLOBAL); -} - -/* Stub to do nothing at all; used for delays and unimplemented calls */ -static void vmi_nop(void) -{ -} - -static void vmi_allocate_pte(struct mm_struct *mm, unsigned long pfn) -{ - vmi_ops.allocate_page(pfn, VMI_PAGE_L1, 0, 0, 0); -} - -static void vmi_allocate_pmd(struct mm_struct *mm, unsigned long pfn) -{ - /* - * This call comes in very early, before mem_map is setup. - * It is called only for swapper_pg_dir, which already has - * data on it. - */ - vmi_ops.allocate_page(pfn, VMI_PAGE_L2, 0, 0, 0); -} - -static void vmi_allocate_pmd_clone(unsigned long pfn, unsigned long clonepfn, unsigned long start, unsigned long count) -{ - vmi_ops.allocate_page(pfn, VMI_PAGE_L2 | VMI_PAGE_CLONE, clonepfn, start, count); -} - -static void vmi_release_pte(unsigned long pfn) -{ - vmi_ops.release_page(pfn, VMI_PAGE_L1); -} - -static void vmi_release_pmd(unsigned long pfn) -{ - vmi_ops.release_page(pfn, VMI_PAGE_L2); -} - -/* - * We use the pgd_free hook for releasing the pgd page: - */ -static void vmi_pgd_free(struct mm_struct *mm, pgd_t *pgd) -{ - unsigned long pfn = __pa(pgd) >> PAGE_SHIFT; - - vmi_ops.release_page(pfn, VMI_PAGE_L2); -} - -/* - * Helper macros for MMU update flags. We can defer updates until a flush - * or page invalidation only if the update is to the current address space - * (otherwise, there is no flush). We must check against init_mm, since - * this could be a kernel update, which usually passes init_mm, although - * sometimes this check can be skipped if we know the particular function - * is only called on user mode PTEs. We could change the kernel to pass - * current->active_mm here, but in particular, I was unsure if changing - * mm/highmem.c to do this would still be correct on other architectures. - */ -#define is_current_as(mm, mustbeuser) ((mm) == current->active_mm || \ - (!mustbeuser && (mm) == &init_mm)) -#define vmi_flags_addr(mm, addr, level, user) \ - ((level) | (is_current_as(mm, user) ? \ - (VMI_PAGE_CURRENT_AS | ((addr) & VMI_PAGE_VA_MASK)) : 0)) -#define vmi_flags_addr_defer(mm, addr, level, user) \ - ((level) | (is_current_as(mm, user) ? \ - (VMI_PAGE_DEFER | VMI_PAGE_CURRENT_AS | ((addr) & VMI_PAGE_VA_MASK)) : 0)) - -static void vmi_update_pte(struct mm_struct *mm, unsigned long addr, pte_t *ptep) -{ - vmi_ops.update_pte(ptep, vmi_flags_addr(mm, addr, VMI_PAGE_PT, 0)); -} - -static void vmi_update_pte_defer(struct mm_struct *mm, unsigned long addr, pte_t *ptep) -{ - vmi_ops.update_pte(ptep, vmi_flags_addr_defer(mm, addr, VMI_PAGE_PT, 0)); -} - -static void vmi_set_pte(pte_t *ptep, pte_t pte) -{ - /* XXX because of set_pmd_pte, this can be called on PT or PD layers */ - vmi_ops.set_pte(pte, ptep, VMI_PAGE_PT); -} - -static void vmi_set_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte) -{ - vmi_ops.set_pte(pte, ptep, vmi_flags_addr(mm, addr, VMI_PAGE_PT, 0)); -} - -static void vmi_set_pmd(pmd_t *pmdp, pmd_t pmdval) -{ -#ifdef CONFIG_X86_PAE - const pte_t pte = { .pte = pmdval.pmd }; -#else - const pte_t pte = { pmdval.pud.pgd.pgd }; -#endif - vmi_ops.set_pte(pte, (pte_t *)pmdp, VMI_PAGE_PD); -} - -#ifdef CONFIG_X86_PAE - -static void vmi_set_pte_atomic(pte_t *ptep, pte_t pteval) -{ - /* - * XXX This is called from set_pmd_pte, but at both PT - * and PD layers so the VMI_PAGE_PT flag is wrong. But - * it is only called for large page mapping changes, - * the Xen backend, doesn't support large pages, and the - * ESX backend doesn't depend on the flag. - */ - set_64bit((unsigned long long *)ptep,pte_val(pteval)); - vmi_ops.update_pte(ptep, VMI_PAGE_PT); -} - -static void vmi_set_pud(pud_t *pudp, pud_t pudval) -{ - /* Um, eww */ - const pte_t pte = { .pte = pudval.pgd.pgd }; - vmi_ops.set_pte(pte, (pte_t *)pudp, VMI_PAGE_PDP); -} - -static void vmi_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep) -{ - const pte_t pte = { .pte = 0 }; - vmi_ops.set_pte(pte, ptep, vmi_flags_addr(mm, addr, VMI_PAGE_PT, 0)); -} - -static void vmi_pmd_clear(pmd_t *pmd) -{ - const pte_t pte = { .pte = 0 }; - vmi_ops.set_pte(pte, (pte_t *)pmd, VMI_PAGE_PD); -} -#endif - -#ifdef CONFIG_SMP -static void __devinit -vmi_startup_ipi_hook(int phys_apicid, unsigned long start_eip, - unsigned long start_esp) -{ - struct vmi_ap_state ap; - - /* Default everything to zero. This is fine for most GPRs. */ - memset(&ap, 0, sizeof(struct vmi_ap_state)); - - ap.gdtr_limit = GDT_SIZE - 1; - ap.gdtr_base = (unsigned long) get_cpu_gdt_table(phys_apicid); - - ap.idtr_limit = IDT_ENTRIES * 8 - 1; - ap.idtr_base = (unsigned long) idt_table; - - ap.ldtr = 0; - - ap.cs = __KERNEL_CS; - ap.eip = (unsigned long) start_eip; - ap.ss = __KERNEL_DS; - ap.esp = (unsigned long) start_esp; - - ap.ds = __USER_DS; - ap.es = __USER_DS; - ap.fs = __KERNEL_PERCPU; - ap.gs = __KERNEL_STACK_CANARY; - - ap.eflags = 0; - -#ifdef CONFIG_X86_PAE - /* efer should match BSP efer. */ - if (cpu_has_nx) { - unsigned l, h; - rdmsr(MSR_EFER, l, h); - ap.efer = (unsigned long long) h << 32 | l; - } -#endif - - ap.cr3 = __pa(swapper_pg_dir); - /* Protected mode, paging, AM, WP, NE, MP. */ - ap.cr0 = 0x80050023; - ap.cr4 = mmu_cr4_features; - vmi_ops.set_initial_ap_state((u32)&ap, phys_apicid); -} -#endif - -static void vmi_start_context_switch(struct task_struct *prev) -{ - paravirt_start_context_switch(prev); - vmi_ops.set_lazy_mode(2); -} - -static void vmi_end_context_switch(struct task_struct *next) -{ - vmi_ops.set_lazy_mode(0); - paravirt_end_context_switch(next); -} - -static void vmi_enter_lazy_mmu(void) -{ - paravirt_enter_lazy_mmu(); - vmi_ops.set_lazy_mode(1); -} - -static void vmi_leave_lazy_mmu(void) -{ - vmi_ops.set_lazy_mode(0); - paravirt_leave_lazy_mmu(); -} - -static inline int __init check_vmi_rom(struct vrom_header *rom) -{ - struct pci_header *pci; - struct pnp_header *pnp; - const char *manufacturer = "UNKNOWN"; - const char *product = "UNKNOWN"; - const char *license = "unspecified"; - - if (rom->rom_signature != 0xaa55) - return 0; - if (rom->vrom_signature != VMI_SIGNATURE) - return 0; - if (rom->api_version_maj != VMI_API_REV_MAJOR || - rom->api_version_min+1 < VMI_API_REV_MINOR+1) { - printk(KERN_WARNING "VMI: Found mismatched rom version %d.%d\n", - rom->api_version_maj, - rom->api_version_min); - return 0; - } - - /* - * Relying on the VMI_SIGNATURE field is not 100% safe, so check - * the PCI header and device type to make sure this is really a - * VMI device. - */ - if (!rom->pci_header_offs) { - printk(KERN_WARNING "VMI: ROM does not contain PCI header.\n"); - return 0; - } - - pci = (struct pci_header *)((char *)rom+rom->pci_header_offs); - if (pci->vendorID != PCI_VENDOR_ID_VMWARE || - pci->deviceID != PCI_DEVICE_ID_VMWARE_VMI) { - /* Allow it to run... anyways, but warn */ - printk(KERN_WARNING "VMI: ROM from unknown manufacturer\n"); - } - - if (rom->pnp_header_offs) { - pnp = (struct pnp_header *)((char *)rom+rom->pnp_header_offs); - if (pnp->manufacturer_offset) - manufacturer = (const char *)rom+pnp->manufacturer_offset; - if (pnp->product_offset) - product = (const char *)rom+pnp->product_offset; - } - - if (rom->license_offs) - license = (char *)rom+rom->license_offs; - - printk(KERN_INFO "VMI: Found %s %s, API version %d.%d, ROM version %d.%d\n", - manufacturer, product, - rom->api_version_maj, rom->api_version_min, - pci->rom_version_maj, pci->rom_version_min); - - /* Don't allow BSD/MIT here for now because we don't want to end up - with any binary only shim layers */ - if (strcmp(license, "GPL") && strcmp(license, "GPL v2")) { - printk(KERN_WARNING "VMI: Non GPL license `%s' found for ROM. Not used.\n", - license); - return 0; - } - - return 1; -} - -/* - * Probe for the VMI option ROM - */ -static inline int __init probe_vmi_rom(void) -{ - unsigned long base; - - /* VMI ROM is in option ROM area, check signature */ - for (base = 0xC0000; base < 0xE0000; base += 2048) { - struct vrom_header *romstart; - romstart = (struct vrom_header *)isa_bus_to_virt(base); - if (check_vmi_rom(romstart)) { - vmi_rom = romstart; - return 1; - } - } - return 0; -} - -/* - * VMI setup common to all processors - */ -void vmi_bringup(void) -{ - /* We must establish the lowmem mapping for MMU ops to work */ - if (vmi_ops.set_linear_mapping) - vmi_ops.set_linear_mapping(0, (void *)__PAGE_OFFSET, MAXMEM_PFN, 0); -} - -/* - * Return a pointer to a VMI function or NULL if unimplemented - */ -static void *vmi_get_function(int vmicall) -{ - u64 reloc; - const struct vmi_relocation_info *rel = (struct vmi_relocation_info *)&reloc; - reloc = call_vrom_long_func(vmi_rom, get_reloc, vmicall); - BUG_ON(rel->type == VMI_RELOCATION_JUMP_REL); - if (rel->type == VMI_RELOCATION_CALL_REL) - return (void *)rel->eip; - else - return NULL; -} - -/* - * Helper macro for making the VMI paravirt-ops fill code readable. - * For unimplemented operations, fall back to default, unless nop - * is returned by the ROM. - */ -#define para_fill(opname, vmicall) \ -do { \ - reloc = call_vrom_long_func(vmi_rom, get_reloc, \ - VMI_CALL_##vmicall); \ - if (rel->type == VMI_RELOCATION_CALL_REL) \ - opname = (void *)rel->eip; \ - else if (rel->type == VMI_RELOCATION_NOP) \ - opname = (void *)vmi_nop; \ - else if (rel->type != VMI_RELOCATION_NONE) \ - printk(KERN_WARNING "VMI: Unknown relocation " \ - "type %d for " #vmicall"\n",\ - rel->type); \ -} while (0) - -/* - * Helper macro for making the VMI paravirt-ops fill code readable. - * For cached operations which do not match the VMI ROM ABI and must - * go through a tranlation stub. Ignore NOPs, since it is not clear - * a NOP * VMI function corresponds to a NOP paravirt-op when the - * functions are not in 1-1 correspondence. - */ -#define para_wrap(opname, wrapper, cache, vmicall) \ -do { \ - reloc = call_vrom_long_func(vmi_rom, get_reloc, \ - VMI_CALL_##vmicall); \ - BUG_ON(rel->type == VMI_RELOCATION_JUMP_REL); \ - if (rel->type == VMI_RELOCATION_CALL_REL) { \ - opname = wrapper; \ - vmi_ops.cache = (void *)rel->eip; \ - } \ -} while (0) - -/* - * Activate the VMI interface and switch into paravirtualized mode - */ -static inline int __init activate_vmi(void) -{ - short kernel_cs; - u64 reloc; - const struct vmi_relocation_info *rel = (struct vmi_relocation_info *)&reloc; - - /* - * Prevent page tables from being allocated in highmem, even if - * CONFIG_HIGHPTE is enabled. - */ - __userpte_alloc_gfp &= ~__GFP_HIGHMEM; - - if (call_vrom_func(vmi_rom, vmi_init) != 0) { - printk(KERN_ERR "VMI ROM failed to initialize!"); - return 0; - } - savesegment(cs, kernel_cs); - - pv_info.paravirt_enabled = 1; - pv_info.kernel_rpl = kernel_cs & SEGMENT_RPL_MASK; - pv_info.name = "vmi [deprecated]"; - - pv_init_ops.patch = vmi_patch; - - /* - * Many of these operations are ABI compatible with VMI. - * This means we can fill in the paravirt-ops with direct - * pointers into the VMI ROM. If the calling convention for - * these operations changes, this code needs to be updated. - * - * Exceptions - * CPUID paravirt-op uses pointers, not the native ISA - * halt has no VMI equivalent; all VMI halts are "safe" - * no MSR support yet - just trap and emulate. VMI uses the - * same ABI as the native ISA, but Linux wants exceptions - * from bogus MSR read / write handled - * rdpmc is not yet used in Linux - */ - - /* CPUID is special, so very special it gets wrapped like a present */ - para_wrap(pv_cpu_ops.cpuid, vmi_cpuid, cpuid, CPUID); - - para_fill(pv_cpu_ops.clts, CLTS); - para_fill(pv_cpu_ops.get_debugreg, GetDR); - para_fill(pv_cpu_ops.set_debugreg, SetDR); - para_fill(pv_cpu_ops.read_cr0, GetCR0); - para_fill(pv_mmu_ops.read_cr2, GetCR2); - para_fill(pv_mmu_ops.read_cr3, GetCR3); - para_fill(pv_cpu_ops.read_cr4, GetCR4); - para_fill(pv_cpu_ops.write_cr0, SetCR0); - para_fill(pv_mmu_ops.write_cr2, SetCR2); - para_fill(pv_mmu_ops.write_cr3, SetCR3); - para_fill(pv_cpu_ops.write_cr4, SetCR4); - - para_fill(pv_irq_ops.save_fl.func, GetInterruptMask); - para_fill(pv_irq_ops.restore_fl.func, SetInterruptMask); - para_fill(pv_irq_ops.irq_disable.func, DisableInterrupts); - para_fill(pv_irq_ops.irq_enable.func, EnableInterrupts); - - para_fill(pv_cpu_ops.wbinvd, WBINVD); - para_fill(pv_cpu_ops.read_tsc, RDTSC); - - /* The following we emulate with trap and emulate for now */ - /* paravirt_ops.read_msr = vmi_rdmsr */ - /* paravirt_ops.write_msr = vmi_wrmsr */ - /* paravirt_ops.rdpmc = vmi_rdpmc */ - - /* TR interface doesn't pass TR value, wrap */ - para_wrap(pv_cpu_ops.load_tr_desc, vmi_set_tr, set_tr, SetTR); - - /* LDT is special, too */ - para_wrap(pv_cpu_ops.set_ldt, vmi_set_ldt, _set_ldt, SetLDT); - - para_fill(pv_cpu_ops.load_gdt, SetGDT); - para_fill(pv_cpu_ops.load_idt, SetIDT); - para_fill(pv_cpu_ops.store_gdt, GetGDT); - para_fill(pv_cpu_ops.store_idt, GetIDT); - para_fill(pv_cpu_ops.store_tr, GetTR); - pv_cpu_ops.load_tls = vmi_load_tls; - para_wrap(pv_cpu_ops.write_ldt_entry, vmi_write_ldt_entry, - write_ldt_entry, WriteLDTEntry); - para_wrap(pv_cpu_ops.write_gdt_entry, vmi_write_gdt_entry, - write_gdt_entry, WriteGDTEntry); - para_wrap(pv_cpu_ops.write_idt_entry, vmi_write_idt_entry, - write_idt_entry, WriteIDTEntry); - para_wrap(pv_cpu_ops.load_sp0, vmi_load_sp0, set_kernel_stack, UpdateKernelStack); - para_fill(pv_cpu_ops.set_iopl_mask, SetIOPLMask); - para_fill(pv_cpu_ops.io_delay, IODelay); - - para_wrap(pv_cpu_ops.start_context_switch, vmi_start_context_switch, - set_lazy_mode, SetLazyMode); - para_wrap(pv_cpu_ops.end_context_switch, vmi_end_context_switch, - set_lazy_mode, SetLazyMode); - - para_wrap(pv_mmu_ops.lazy_mode.enter, vmi_enter_lazy_mmu, - set_lazy_mode, SetLazyMode); - para_wrap(pv_mmu_ops.lazy_mode.leave, vmi_leave_lazy_mmu, - set_lazy_mode, SetLazyMode); - - /* user and kernel flush are just handled with different flags to FlushTLB */ - para_wrap(pv_mmu_ops.flush_tlb_user, vmi_flush_tlb_user, _flush_tlb, FlushTLB); - para_wrap(pv_mmu_ops.flush_tlb_kernel, vmi_flush_tlb_kernel, _flush_tlb, FlushTLB); - para_fill(pv_mmu_ops.flush_tlb_single, InvalPage); - - /* - * Until a standard flag format can be agreed on, we need to - * implement these as wrappers in Linux. Get the VMI ROM - * function pointers for the two backend calls. - */ -#ifdef CONFIG_X86_PAE - vmi_ops.set_pte = vmi_get_function(VMI_CALL_SetPxELong); - vmi_ops.update_pte = vmi_get_function(VMI_CALL_UpdatePxELong); -#else - vmi_ops.set_pte = vmi_get_function(VMI_CALL_SetPxE); - vmi_ops.update_pte = vmi_get_function(VMI_CALL_UpdatePxE); -#endif - - if (vmi_ops.set_pte) { - pv_mmu_ops.set_pte = vmi_set_pte; - pv_mmu_ops.set_pte_at = vmi_set_pte_at; - pv_mmu_ops.set_pmd = vmi_set_pmd; -#ifdef CONFIG_X86_PAE - pv_mmu_ops.set_pte_atomic = vmi_set_pte_atomic; - pv_mmu_ops.set_pud = vmi_set_pud; - pv_mmu_ops.pte_clear = vmi_pte_clear; - pv_mmu_ops.pmd_clear = vmi_pmd_clear; -#endif - } - - if (vmi_ops.update_pte) { - pv_mmu_ops.pte_update = vmi_update_pte; - pv_mmu_ops.pte_update_defer = vmi_update_pte_defer; - } - - vmi_ops.allocate_page = vmi_get_function(VMI_CALL_AllocatePage); - if (vmi_ops.allocate_page) { - pv_mmu_ops.alloc_pte = vmi_allocate_pte; - pv_mmu_ops.alloc_pmd = vmi_allocate_pmd; - pv_mmu_ops.alloc_pmd_clone = vmi_allocate_pmd_clone; - } - - vmi_ops.release_page = vmi_get_function(VMI_CALL_ReleasePage); - if (vmi_ops.release_page) { - pv_mmu_ops.release_pte = vmi_release_pte; - pv_mmu_ops.release_pmd = vmi_release_pmd; - pv_mmu_ops.pgd_free = vmi_pgd_free; - } - - /* Set linear is needed in all cases */ - vmi_ops.set_linear_mapping = vmi_get_function(VMI_CALL_SetLinearMapping); - - /* - * These MUST always be patched. Don't support indirect jumps - * through these operations, as the VMI interface may use either - * a jump or a call to get to these operations, depending on - * the backend. They are performance critical anyway, so requiring - * a patch is not a big problem. - */ - pv_cpu_ops.irq_enable_sysexit = (void *)0xfeedbab0; - pv_cpu_ops.iret = (void *)0xbadbab0; - -#ifdef CONFIG_SMP - para_wrap(pv_apic_ops.startup_ipi_hook, vmi_startup_ipi_hook, set_initial_ap_state, SetInitialAPState); -#endif - -#ifdef CONFIG_X86_LOCAL_APIC - para_fill(apic->read, APICRead); - para_fill(apic->write, APICWrite); -#endif - - /* - * Check for VMI timer functionality by probing for a cycle frequency method - */ - reloc = call_vrom_long_func(vmi_rom, get_reloc, VMI_CALL_GetCycleFrequency); - if (!disable_vmi_timer && rel->type != VMI_RELOCATION_NONE) { - vmi_timer_ops.get_cycle_frequency = (void *)rel->eip; - vmi_timer_ops.get_cycle_counter = - vmi_get_function(VMI_CALL_GetCycleCounter); - vmi_timer_ops.get_wallclock = - vmi_get_function(VMI_CALL_GetWallclockTime); - vmi_timer_ops.wallclock_updated = - vmi_get_function(VMI_CALL_WallclockUpdated); - vmi_timer_ops.set_alarm = vmi_get_function(VMI_CALL_SetAlarm); - vmi_timer_ops.cancel_alarm = - vmi_get_function(VMI_CALL_CancelAlarm); - x86_init.timers.timer_init = vmi_time_init; -#ifdef CONFIG_X86_LOCAL_APIC - x86_init.timers.setup_percpu_clockev = vmi_time_bsp_init; - x86_cpuinit.setup_percpu_clockev = vmi_time_ap_init; -#endif - pv_time_ops.sched_clock = vmi_sched_clock; - x86_platform.calibrate_tsc = vmi_tsc_khz; - x86_platform.get_wallclock = vmi_get_wallclock; - x86_platform.set_wallclock = vmi_set_wallclock; - - /* We have true wallclock functions; disable CMOS clock sync */ - no_sync_cmos_clock = 1; - } else { - disable_noidle = 1; - disable_vmi_timer = 1; - } - - para_fill(pv_irq_ops.safe_halt, Halt); - - /* - * Alternative instruction rewriting doesn't happen soon enough - * to convert VMI_IRET to a call instead of a jump; so we have - * to do this before IRQs get reenabled. Fortunately, it is - * idempotent. - */ - apply_paravirt(__parainstructions, __parainstructions_end); - - vmi_bringup(); - - return 1; -} - -#undef para_fill - -void __init vmi_init(void) -{ - if (!vmi_rom) - probe_vmi_rom(); - else - check_vmi_rom(vmi_rom); - - /* In case probing for or validating the ROM failed, basil */ - if (!vmi_rom) - return; - - reserve_top_address(-vmi_rom->virtual_top); - -#ifdef CONFIG_X86_IO_APIC - /* This is virtual hardware; timer routing is wired correctly */ - no_timer_check = 1; -#endif -} - -void __init vmi_activate(void) -{ - unsigned long flags; - - if (!vmi_rom) - return; - - local_irq_save(flags); - activate_vmi(); - local_irq_restore(flags & X86_EFLAGS_IF); -} - -static int __init parse_vmi(char *arg) -{ - if (!arg) - return -EINVAL; - - if (!strcmp(arg, "disable_pge")) { - clear_cpu_cap(&boot_cpu_data, X86_FEATURE_PGE); - disable_pge = 1; - } else if (!strcmp(arg, "disable_pse")) { - clear_cpu_cap(&boot_cpu_data, X86_FEATURE_PSE); - disable_pse = 1; - } else if (!strcmp(arg, "disable_sep")) { - clear_cpu_cap(&boot_cpu_data, X86_FEATURE_SEP); - disable_sep = 1; - } else if (!strcmp(arg, "disable_tsc")) { - clear_cpu_cap(&boot_cpu_data, X86_FEATURE_TSC); - disable_tsc = 1; - } else if (!strcmp(arg, "disable_mtrr")) { - clear_cpu_cap(&boot_cpu_data, X86_FEATURE_MTRR); - disable_mtrr = 1; - } else if (!strcmp(arg, "disable_timer")) { - disable_vmi_timer = 1; - disable_noidle = 1; - } else if (!strcmp(arg, "disable_noidle")) - disable_noidle = 1; - return 0; -} - -early_param("vmi", parse_vmi); diff --git a/arch/x86/kernel/vmiclock_32.c b/arch/x86/kernel/vmiclock_32.c deleted file mode 100644 index 5e1ff66ecd7..00000000000 --- a/arch/x86/kernel/vmiclock_32.c +++ /dev/null @@ -1,317 +0,0 @@ -/* - * VMI paravirtual timer support routines. - * - * Copyright (C) 2007, VMware, 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 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, GOOD TITLE or - * NON INFRINGEMENT. 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., 675 Mass Ave, Cambridge, MA 02139, USA. - * - */ - -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include -#include -#include - -#define VMI_ONESHOT (VMI_ALARM_IS_ONESHOT | VMI_CYCLES_REAL | vmi_get_alarm_wiring()) -#define VMI_PERIODIC (VMI_ALARM_IS_PERIODIC | VMI_CYCLES_REAL | vmi_get_alarm_wiring()) - -static DEFINE_PER_CPU(struct clock_event_device, local_events); - -static inline u32 vmi_counter(u32 flags) -{ - /* Given VMI_ONESHOT or VMI_PERIODIC, return the corresponding - * cycle counter. */ - return flags & VMI_ALARM_COUNTER_MASK; -} - -/* paravirt_ops.get_wallclock = vmi_get_wallclock */ -unsigned long vmi_get_wallclock(void) -{ - unsigned long long wallclock; - wallclock = vmi_timer_ops.get_wallclock(); // nsec - (void)do_div(wallclock, 1000000000); // sec - - return wallclock; -} - -/* paravirt_ops.set_wallclock = vmi_set_wallclock */ -int vmi_set_wallclock(unsigned long now) -{ - return 0; -} - -/* paravirt_ops.sched_clock = vmi_sched_clock */ -unsigned long long vmi_sched_clock(void) -{ - return cycles_2_ns(vmi_timer_ops.get_cycle_counter(VMI_CYCLES_AVAILABLE)); -} - -/* x86_platform.calibrate_tsc = vmi_tsc_khz */ -unsigned long vmi_tsc_khz(void) -{ - unsigned long long khz; - khz = vmi_timer_ops.get_cycle_frequency(); - (void)do_div(khz, 1000); - return khz; -} - -static inline unsigned int vmi_get_timer_vector(void) -{ - return IRQ0_VECTOR; -} - -/** vmi clockchip */ -#ifdef CONFIG_X86_LOCAL_APIC -static unsigned int startup_timer_irq(unsigned int irq) -{ - unsigned long val = apic_read(APIC_LVTT); - apic_write(APIC_LVTT, vmi_get_timer_vector()); - - return (val & APIC_SEND_PENDING); -} - -static void mask_timer_irq(unsigned int irq) -{ - unsigned long val = apic_read(APIC_LVTT); - apic_write(APIC_LVTT, val | APIC_LVT_MASKED); -} - -static void unmask_timer_irq(unsigned int irq) -{ - unsigned long val = apic_read(APIC_LVTT); - apic_write(APIC_LVTT, val & ~APIC_LVT_MASKED); -} - -static void ack_timer_irq(unsigned int irq) -{ - ack_APIC_irq(); -} - -static struct irq_chip vmi_chip __read_mostly = { - .name = "VMI-LOCAL", - .startup = startup_timer_irq, - .mask = mask_timer_irq, - .unmask = unmask_timer_irq, - .ack = ack_timer_irq -}; -#endif - -/** vmi clockevent */ -#define VMI_ALARM_WIRED_IRQ0 0x00000000 -#define VMI_ALARM_WIRED_LVTT 0x00010000 -static int vmi_wiring = VMI_ALARM_WIRED_IRQ0; - -static inline int vmi_get_alarm_wiring(void) -{ - return vmi_wiring; -} - -static void vmi_timer_set_mode(enum clock_event_mode mode, - struct clock_event_device *evt) -{ - cycle_t now, cycles_per_hz; - BUG_ON(!irqs_disabled()); - - switch (mode) { - case CLOCK_EVT_MODE_ONESHOT: - case CLOCK_EVT_MODE_RESUME: - break; - case CLOCK_EVT_MODE_PERIODIC: - cycles_per_hz = vmi_timer_ops.get_cycle_frequency(); - (void)do_div(cycles_per_hz, HZ); - now = vmi_timer_ops.get_cycle_counter(vmi_counter(VMI_PERIODIC)); - vmi_timer_ops.set_alarm(VMI_PERIODIC, now, cycles_per_hz); - break; - case CLOCK_EVT_MODE_UNUSED: - case CLOCK_EVT_MODE_SHUTDOWN: - switch (evt->mode) { - case CLOCK_EVT_MODE_ONESHOT: - vmi_timer_ops.cancel_alarm(VMI_ONESHOT); - break; - case CLOCK_EVT_MODE_PERIODIC: - vmi_timer_ops.cancel_alarm(VMI_PERIODIC); - break; - default: - break; - } - break; - default: - break; - } -} - -static int vmi_timer_next_event(unsigned long delta, - struct clock_event_device *evt) -{ - /* Unfortunately, set_next_event interface only passes relative - * expiry, but we want absolute expiry. It'd be better if were - * were passed an absolute expiry, since a bunch of time may - * have been stolen between the time the delta is computed and - * when we set the alarm below. */ - cycle_t now = vmi_timer_ops.get_cycle_counter(vmi_counter(VMI_ONESHOT)); - - BUG_ON(evt->mode != CLOCK_EVT_MODE_ONESHOT); - vmi_timer_ops.set_alarm(VMI_ONESHOT, now + delta, 0); - return 0; -} - -static struct clock_event_device vmi_clockevent = { - .name = "vmi-timer", - .features = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT, - .shift = 22, - .set_mode = vmi_timer_set_mode, - .set_next_event = vmi_timer_next_event, - .rating = 1000, - .irq = 0, -}; - -static irqreturn_t vmi_timer_interrupt(int irq, void *dev_id) -{ - struct clock_event_device *evt = &__get_cpu_var(local_events); - evt->event_handler(evt); - return IRQ_HANDLED; -} - -static struct irqaction vmi_clock_action = { - .name = "vmi-timer", - .handler = vmi_timer_interrupt, - .flags = IRQF_DISABLED | IRQF_NOBALANCING | IRQF_TIMER, -}; - -static void __devinit vmi_time_init_clockevent(void) -{ - cycle_t cycles_per_msec; - struct clock_event_device *evt; - - int cpu = smp_processor_id(); - evt = &__get_cpu_var(local_events); - - /* Use cycles_per_msec since div_sc params are 32-bits. */ - cycles_per_msec = vmi_timer_ops.get_cycle_frequency(); - (void)do_div(cycles_per_msec, 1000); - - memcpy(evt, &vmi_clockevent, sizeof(*evt)); - /* Must pick .shift such that .mult fits in 32-bits. Choosing - * .shift to be 22 allows 2^(32-22) cycles per nano-seconds - * before overflow. */ - evt->mult = div_sc(cycles_per_msec, NSEC_PER_MSEC, evt->shift); - /* Upper bound is clockevent's use of ulong for cycle deltas. */ - evt->max_delta_ns = clockevent_delta2ns(ULONG_MAX, evt); - evt->min_delta_ns = clockevent_delta2ns(1, evt); - evt->cpumask = cpumask_of(cpu); - - printk(KERN_WARNING "vmi: registering clock event %s. mult=%u shift=%u\n", - evt->name, evt->mult, evt->shift); - clockevents_register_device(evt); -} - -void __init vmi_time_init(void) -{ - unsigned int cpu; - /* Disable PIT: BIOSes start PIT CH0 with 18.2hz peridic. */ - outb_pit(0x3a, PIT_MODE); /* binary, mode 5, LSB/MSB, ch 0 */ - - vmi_time_init_clockevent(); - setup_irq(0, &vmi_clock_action); - for_each_possible_cpu(cpu) - per_cpu(vector_irq, cpu)[vmi_get_timer_vector()] = 0; -} - -#ifdef CONFIG_X86_LOCAL_APIC -void __devinit vmi_time_bsp_init(void) -{ - /* - * On APIC systems, we want local timers to fire on each cpu. We do - * this by programming LVTT to deliver timer events to the IRQ handler - * for IRQ-0, since we can't re-use the APIC local timer handler - * without interfering with that code. - */ - clockevents_notify(CLOCK_EVT_NOTIFY_SUSPEND, NULL); - local_irq_disable(); -#ifdef CONFIG_SMP - /* - * XXX handle_percpu_irq only defined for SMP; we need to switch over - * to using it, since this is a local interrupt, which each CPU must - * handle individually without locking out or dropping simultaneous - * local timers on other CPUs. We also don't want to trigger the - * quirk workaround code for interrupts which gets invoked from - * handle_percpu_irq via eoi, so we use our own IRQ chip. - */ - set_irq_chip_and_handler_name(0, &vmi_chip, handle_percpu_irq, "lvtt"); -#else - set_irq_chip_and_handler_name(0, &vmi_chip, handle_edge_irq, "lvtt"); -#endif - vmi_wiring = VMI_ALARM_WIRED_LVTT; - apic_write(APIC_LVTT, vmi_get_timer_vector()); - local_irq_enable(); - clockevents_notify(CLOCK_EVT_NOTIFY_RESUME, NULL); -} - -void __devinit vmi_time_ap_init(void) -{ - vmi_time_init_clockevent(); - apic_write(APIC_LVTT, vmi_get_timer_vector()); -} -#endif - -/** vmi clocksource */ -static struct clocksource clocksource_vmi; - -static cycle_t read_real_cycles(struct clocksource *cs) -{ - cycle_t ret = (cycle_t)vmi_timer_ops.get_cycle_counter(VMI_CYCLES_REAL); - return max(ret, clocksource_vmi.cycle_last); -} - -static struct clocksource clocksource_vmi = { - .name = "vmi-timer", - .rating = 450, - .read = read_real_cycles, - .mask = CLOCKSOURCE_MASK(64), - .mult = 0, /* to be set */ - .shift = 22, - .flags = CLOCK_SOURCE_IS_CONTINUOUS, -}; - -static int __init init_vmi_clocksource(void) -{ - cycle_t cycles_per_msec; - - if (!vmi_timer_ops.get_cycle_frequency) - return 0; - /* Use khz2mult rather than hz2mult since hz arg is only 32-bits. */ - cycles_per_msec = vmi_timer_ops.get_cycle_frequency(); - (void)do_div(cycles_per_msec, 1000); - - /* Note that clocksource.{mult, shift} converts in the opposite direction - * as clockevents. */ - clocksource_vmi.mult = clocksource_khz2mult(cycles_per_msec, - clocksource_vmi.shift); - - printk(KERN_WARNING "vmi: registering clock source khz=%lld\n", cycles_per_msec); - return clocksource_register(&clocksource_vmi); - -} -module_init(init_vmi_clocksource); -- cgit v1.2.3-70-g09d2 From b0f4c062fb6dd4c02b1fe6de73319ed50a09b27d Mon Sep 17 00:00:00 2001 From: Alok Kataria Date: Mon, 23 Aug 2010 17:05:57 -0700 Subject: x86, paravirt: Remove alloc_pmd_clone hook, only used by VMI VMI was the only user of the alloc_pmd_clone hook, given that VMI is now removed we can also remove this hook. Signed-off-by: Alok N Kataria LKML-Reference: <1282608357.19396.36.camel@ank32.eng.vmware.com> Cc: Jeremy Fitzhardinge Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/paravirt.h | 5 ----- arch/x86/include/asm/paravirt_types.h | 1 - arch/x86/kernel/paravirt.c | 1 - arch/x86/mm/pgtable.c | 4 ---- arch/x86/xen/mmu.c | 1 - 5 files changed, 12 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h index 5653f43d90e..edecb4ed221 100644 --- a/arch/x86/include/asm/paravirt.h +++ b/arch/x86/include/asm/paravirt.h @@ -416,11 +416,6 @@ static inline void paravirt_alloc_pmd(struct mm_struct *mm, unsigned long pfn) PVOP_VCALL2(pv_mmu_ops.alloc_pmd, mm, pfn); } -static inline void paravirt_alloc_pmd_clone(unsigned long pfn, unsigned long clonepfn, - unsigned long start, unsigned long count) -{ - PVOP_VCALL4(pv_mmu_ops.alloc_pmd_clone, pfn, clonepfn, start, count); -} static inline void paravirt_release_pmd(unsigned long pfn) { PVOP_VCALL1(pv_mmu_ops.release_pmd, pfn); diff --git a/arch/x86/include/asm/paravirt_types.h b/arch/x86/include/asm/paravirt_types.h index db9ef553234..b82bac97525 100644 --- a/arch/x86/include/asm/paravirt_types.h +++ b/arch/x86/include/asm/paravirt_types.h @@ -255,7 +255,6 @@ struct pv_mmu_ops { */ void (*alloc_pte)(struct mm_struct *mm, unsigned long pfn); void (*alloc_pmd)(struct mm_struct *mm, unsigned long pfn); - void (*alloc_pmd_clone)(unsigned long pfn, unsigned long clonepfn, unsigned long start, unsigned long count); void (*alloc_pud)(struct mm_struct *mm, unsigned long pfn); void (*release_pte)(unsigned long pfn); void (*release_pmd)(unsigned long pfn); diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index 1db183ed7c0..c5b250011fd 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -413,7 +413,6 @@ struct pv_mmu_ops pv_mmu_ops = { .alloc_pte = paravirt_nop, .alloc_pmd = paravirt_nop, - .alloc_pmd_clone = paravirt_nop, .alloc_pud = paravirt_nop, .release_pte = paravirt_nop, .release_pmd = paravirt_nop, diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c index 5c4ee422590..a96023e872a 100644 --- a/arch/x86/mm/pgtable.c +++ b/arch/x86/mm/pgtable.c @@ -98,10 +98,6 @@ static void pgd_ctor(pgd_t *pgd) clone_pgd_range(pgd + KERNEL_PGD_BOUNDARY, swapper_pg_dir + KERNEL_PGD_BOUNDARY, KERNEL_PGD_PTRS); - paravirt_alloc_pmd_clone(__pa(pgd) >> PAGE_SHIFT, - __pa(swapper_pg_dir) >> PAGE_SHIFT, - KERNEL_PGD_BOUNDARY, - KERNEL_PGD_PTRS); } /* list required to sync kernel mapping updates */ diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index 42086ac406a..b2363fcbcd0 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -1969,7 +1969,6 @@ static const struct pv_mmu_ops xen_mmu_ops __initdata = { .alloc_pte = xen_alloc_pte_init, .release_pte = xen_release_pte_init, .alloc_pmd = xen_alloc_pmd_init, - .alloc_pmd_clone = paravirt_nop, .release_pmd = xen_release_pmd_init, #ifdef CONFIG_X86_64 -- cgit v1.2.3-70-g09d2 From d0cd7425fab774a480cce17c2f649984312d0b55 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Tue, 24 Aug 2010 17:32:04 -0700 Subject: x86, bios: By default, reserve the low 64K for all BIOSes The laundry list of BIOSes that need the low 64K reserved is getting very long, so make it the default across all BIOSes. This also allows the code to be simplified and unified with the reservation code for the first 4K. This resolves kernel bugzilla 16661 and who knows what else... Signed-off-by: H. Peter Anvin LKML-Reference: --- arch/x86/Kconfig | 47 ++++++++++++++++----------- arch/x86/kernel/setup.c | 84 +++++-------------------------------------------- 2 files changed, 35 insertions(+), 96 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index cea0cd9a316..683ae8f9bd0 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -1326,25 +1326,34 @@ config X86_BOOTPARAM_MEMORY_CORRUPTION_CHECK Set whether the default state of memory_corruption_check is on or off. -config X86_RESERVE_LOW_64K - bool "Reserve low 64K of RAM on AMI/Phoenix BIOSen" - default y - ---help--- - Reserve the first 64K of physical RAM on BIOSes that are known - to potentially corrupt that memory range. A numbers of BIOSes are - known to utilize this area during suspend/resume, so it must not - be used by the kernel. - - Set this to N if you are absolutely sure that you trust the BIOS - to get all its memory reservations and usages right. - - If you have doubts about the BIOS (e.g. suspend/resume does not - work or there's kernel crashes after certain hardware hotplug - events) and it's not AMI or Phoenix, then you might want to enable - X86_CHECK_BIOS_CORRUPTION=y to allow the kernel to check typical - corruption patterns. - - Say Y if unsure. +config X86_LOW_RESERVE + int "Amount of low memory, in kilobytes, to reserve for the BIOS" + default 64 + range 4 640 + ---help--- + Specify the amount of low memory to reserve for the BIOS. + + The first page contains BIOS data structures that the kernel + must not use, so that page must always be reserved. + + By default we reserve the first 64K of physical RAM, as a + number of BIOSes are known to corrupt that memory range + during events such as suspend/resume or monitor cable + insertion, so it must not be used by the kernel. + + You can set this to 4 if you are absolutely sure that you + trust the BIOS to get all its memory reservations and usages + right. If you know your BIOS have problems beyond the + default 64K area, you can set this to 640 to avoid using the + entire low memory range. + + If you have doubts about the BIOS (e.g. suspend/resume does + not work or there's kernel crashes after certain hardware + hotplug events) then you might want to enable + X86_CHECK_BIOS_CORRUPTION=y to allow the kernel to check + typical corruption patterns. + + Leave this to the default value of 64 if you are unsure. config MATH_EMULATION bool diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index c3a4fbb2b99..eb87f1c83f9 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -618,88 +618,20 @@ static __init void reserve_ibft_region(void) reserve_early_overlap_ok(addr, addr + size, "ibft"); } -#ifdef CONFIG_X86_RESERVE_LOW_64K -static int __init dmi_low_memory_corruption(const struct dmi_system_id *d) -{ - printk(KERN_NOTICE - "%s detected: BIOS may corrupt low RAM, working around it.\n", - d->ident); - - e820_update_range(0, 0x10000, E820_RAM, E820_RESERVED); - sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); - - return 0; -} -#endif - -/* List of systems that have known low memory corruption BIOS problems */ -static struct dmi_system_id __initdata bad_bios_dmi_table[] = { -#ifdef CONFIG_X86_RESERVE_LOW_64K - { - .callback = dmi_low_memory_corruption, - .ident = "AMI BIOS", - .matches = { - DMI_MATCH(DMI_BIOS_VENDOR, "American Megatrends Inc."), - }, - }, - { - .callback = dmi_low_memory_corruption, - .ident = "Phoenix BIOS", - .matches = { - DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix Technologies"), - }, - }, - { - .callback = dmi_low_memory_corruption, - .ident = "Phoenix/MSC BIOS", - .matches = { - DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix/MSC"), - }, - }, - /* - * AMI BIOS with low memory corruption was found on Intel DG45ID and - * DG45FC boards. - * It has a different DMI_BIOS_VENDOR = "Intel Corp.", for now we will - * match only DMI_BOARD_NAME and see if there is more bad products - * with this vendor. - */ - { - .callback = dmi_low_memory_corruption, - .ident = "AMI BIOS", - .matches = { - DMI_MATCH(DMI_BOARD_NAME, "DG45ID"), - }, - }, - { - .callback = dmi_low_memory_corruption, - .ident = "AMI BIOS", - .matches = { - DMI_MATCH(DMI_BOARD_NAME, "DG45FC"), - }, - }, - /* - * The Dell Inspiron Mini 1012 has DMI_BIOS_VENDOR = "Dell Inc.", so - * match on the product name. - */ - { - .callback = dmi_low_memory_corruption, - .ident = "Phoenix BIOS", - .matches = { - DMI_MATCH(DMI_PRODUCT_NAME, "Inspiron 1012"), - }, - }, -#endif - {} -}; - static void __init trim_bios_range(void) { /* * A special case is the first 4Kb of memory; * This is a BIOS owned area, not kernel ram, but generally * not listed as such in the E820 table. + * + * This typically reserves additional memory (64KiB by default) + * since some BIOSes are known to corrupt low memory. See the + * Kconfig help text for X86_LOW_RESERVE. */ - e820_update_range(0, PAGE_SIZE, E820_RAM, E820_RESERVED); + e820_update_range(0, ALIGN(CONFIG_X86_LOW_RESERVE << 10, PAGE_SIZE), + E820_RAM, E820_RESERVED); + /* * special case: Some BIOSen report the PC BIOS * area (640->1Mb) as ram even though it is not. @@ -863,8 +795,6 @@ void __init setup_arch(char **cmdline_p) dmi_scan_machine(); - dmi_check_system(bad_bios_dmi_table); - /* * VMware detection requires dmi to be available, so this * needs to be done after dmi_scan_machine, for the BP. -- cgit v1.2.3-70-g09d2 From 04fba67163a9e6132614b72b33bb2743bd33ffb3 Mon Sep 17 00:00:00 2001 From: Lin Ming Date: Wed, 25 Aug 2010 14:49:09 +0800 Subject: perf: Remove unused variable MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This fixes the following build warning introduced by the callchain rework: arch/x86/kernel/cpu/perf_event.c:1574: warning: ‘perf_callchain_entry_nmi’ defined but not used Signed-off-by: Lin Ming Cc: Peter Zijlstra Cc: Frederic Weisbecker LKML-Reference: <1282718949.16443.75.camel@minggr.sh.intel.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index acc52afaf7b..bcddac61422 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -1571,9 +1571,6 @@ const struct pmu *hw_perf_event_init(struct perf_event *event) * callchain support */ -static DEFINE_PER_CPU(struct perf_callchain_entry, perf_callchain_entry_nmi); - - static void backtrace_warning_symbol(void *data, char *msg, unsigned long symbol) { -- cgit v1.2.3-70-g09d2 From acf01734b1747b1ec4be6f159aff579ea5f7f8e2 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Wed, 25 Aug 2010 18:28:23 +0200 Subject: x86, tsc: Remove CPU frequency calibration on AMD 6b37f5a20c0e5c334c010a587058354215433e92 introduced the CPU frequency calibration code for AMD CPUs whose TSCs didn't increment with the core's P0 frequency. From F10h, revB onward, however, the TSC increment rate is denoted by MSRC001_0015[24] and when this bit is set (which should be done by the BIOS) the TSC increments with the P0 frequency so the calibration is not needed and booting can be a couple of mcecs faster on those machines. Besides, there should be virtually no machines out there which don't have this bit set, therefore this calibration can be safely removed. It is a shaky hack anyway since it assumes implicitly that the core is in P0 when BIOS hands off to the OS, which might not always be the case. Signed-off-by: Borislav Petkov LKML-Reference: <20100825162823.GE26438@aftab> Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/amd.c | 17 ++++++++++++++ arch/x86/kernel/tsc.c | 58 ----------------------------------------------- 2 files changed, 17 insertions(+), 58 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index ba5f62f45f0..fc563fabde6 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -412,6 +412,23 @@ static void __cpuinit early_init_amd(struct cpuinfo_x86 *c) set_cpu_cap(c, X86_FEATURE_EXTD_APICID); } #endif + + /* We need to do the following only once */ + if (c != &boot_cpu_data) + return; + + if (cpu_has(c, X86_FEATURE_CONSTANT_TSC)) { + + if (c->x86 > 0x10 || + (c->x86 == 0x10 && c->x86_model >= 0x2)) { + u64 val; + + rdmsrl(MSR_K7_HWCR, val); + if (!(val & BIT(24))) + printk(KERN_WARNING FW_BUG "TSC doesn't count " + "with P0 frequency!\n"); + } + } } static void __cpuinit init_amd(struct cpuinfo_x86 *c) diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index ce8e5023933..13b6a6cc77f 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -854,60 +854,6 @@ static void __init init_tsc_clocksource(void) clocksource_register_khz(&clocksource_tsc, tsc_khz); } -#ifdef CONFIG_X86_64 -/* - * calibrate_cpu is used on systems with fixed rate TSCs to determine - * processor frequency - */ -#define TICK_COUNT 100000000 -static unsigned long __init calibrate_cpu(void) -{ - int tsc_start, tsc_now; - int i, no_ctr_free; - unsigned long evntsel3 = 0, pmc3 = 0, pmc_now = 0; - unsigned long flags; - - for (i = 0; i < 4; i++) - if (avail_to_resrv_perfctr_nmi_bit(i)) - break; - no_ctr_free = (i == 4); - if (no_ctr_free) { - WARN(1, KERN_WARNING "Warning: AMD perfctrs busy ... " - "cpu_khz value may be incorrect.\n"); - i = 3; - rdmsrl(MSR_K7_EVNTSEL3, evntsel3); - wrmsrl(MSR_K7_EVNTSEL3, 0); - rdmsrl(MSR_K7_PERFCTR3, pmc3); - } else { - reserve_perfctr_nmi(MSR_K7_PERFCTR0 + i); - reserve_evntsel_nmi(MSR_K7_EVNTSEL0 + i); - } - local_irq_save(flags); - /* start measuring cycles, incrementing from 0 */ - wrmsrl(MSR_K7_PERFCTR0 + i, 0); - wrmsrl(MSR_K7_EVNTSEL0 + i, 1 << 22 | 3 << 16 | 0x76); - rdtscl(tsc_start); - do { - rdmsrl(MSR_K7_PERFCTR0 + i, pmc_now); - tsc_now = get_cycles(); - } while ((tsc_now - tsc_start) < TICK_COUNT); - - local_irq_restore(flags); - if (no_ctr_free) { - wrmsrl(MSR_K7_EVNTSEL3, 0); - wrmsrl(MSR_K7_PERFCTR3, pmc3); - wrmsrl(MSR_K7_EVNTSEL3, evntsel3); - } else { - release_perfctr_nmi(MSR_K7_PERFCTR0 + i); - release_evntsel_nmi(MSR_K7_EVNTSEL0 + i); - } - - return pmc_now * tsc_khz / (tsc_now - tsc_start); -} -#else -static inline unsigned long calibrate_cpu(void) { return cpu_khz; } -#endif - void __init tsc_init(void) { u64 lpj; @@ -926,10 +872,6 @@ void __init tsc_init(void) return; } - if (cpu_has(&boot_cpu_data, X86_FEATURE_CONSTANT_TSC) && - (boot_cpu_data.x86_vendor == X86_VENDOR_AMD)) - cpu_khz = calibrate_cpu(); - printk("Detected %lu.%03lu MHz processor.\n", (unsigned long)cpu_khz / 1000, (unsigned long)cpu_khz % 1000); -- cgit v1.2.3-70-g09d2 From 9ea77bdb39b62c9bf9fd3cdd1c25a9420bccd380 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Wed, 25 Aug 2010 16:38:20 -0700 Subject: x86, bios: Make the x86 early memory reservation a kernel option Add a kernel command-line option so the x86 early memory reservation size can be adjusted at runtime instead of only at compile time. Suggested-by: Andrew Morton LKML-Reference: Signed-off-by: H. Peter Anvin --- Documentation/kernel-parameters.txt | 5 +++++ arch/x86/Kconfig | 2 +- arch/x86/kernel/setup.c | 28 ++++++++++++++++++++++++++-- 3 files changed, 32 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 2c85c0692b0..41ce93e71de 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -2150,6 +2150,11 @@ and is between 256 and 4096 characters. It is defined in the file Reserves a hole at the top of the kernel virtual address space. + reservelow= [X86] + Format: nn[K] + Set the amount of memory to reserve for BIOS at + the bottom of the address space. + reset_devices [KNL] Force drivers to reset the underlying device during initialization. diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 683ae8f9bd0..d3590008c5d 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -1326,7 +1326,7 @@ config X86_BOOTPARAM_MEMORY_CORRUPTION_CHECK Set whether the default state of memory_corruption_check is on or off. -config X86_LOW_RESERVE +config X86_RESERVE_LOW int "Amount of low memory, in kilobytes, to reserve for the BIOS" default 64 range 4 640 diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index eb87f1c83f9..af277e369de 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -618,6 +618,8 @@ static __init void reserve_ibft_region(void) reserve_early_overlap_ok(addr, addr + size, "ibft"); } +static unsigned reserve_low = CONFIG_X86_RESERVE_LOW << 10; + static void __init trim_bios_range(void) { /* @@ -627,9 +629,9 @@ static void __init trim_bios_range(void) * * This typically reserves additional memory (64KiB by default) * since some BIOSes are known to corrupt low memory. See the - * Kconfig help text for X86_LOW_RESERVE. + * Kconfig help text for X86_RESERVE_LOW. */ - e820_update_range(0, ALIGN(CONFIG_X86_LOW_RESERVE << 10, PAGE_SIZE), + e820_update_range(0, ALIGN(reserve_low, PAGE_SIZE), E820_RAM, E820_RESERVED); /* @@ -641,6 +643,28 @@ static void __init trim_bios_range(void) sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); } +static int __init parse_reservelow(char *p) +{ + unsigned long long size; + + if (!p) + return -EINVAL; + + size = memparse(p, &p); + + if (size < 4096) + size = 4096; + + if (size > 640*1024) + size = 640*1024; + + reserve_low = size; + + return 0; +} + +early_param("reservelow", parse_reservelow); + /* * Determine if we were loaded by an EFI loader. If so, then we have also been * passed the efi memmap, systab, etc., so we should use these data structures -- cgit v1.2.3-70-g09d2 From 6afb5157b9eba4092e2f0f54d24a3806409bdde5 Mon Sep 17 00:00:00 2001 From: Haicheng Li Date: Wed, 19 May 2010 17:42:14 +0800 Subject: x86, mm: Separate x86_64 vmalloc_sync_all() into separate functions No behavior change. Move some of vmalloc_sync_all() code into a new function sync_global_pgds() that will be useful for memory hotplug. Signed-off-by: Haicheng Li LKML-Reference: <4C6E4ECD.1090607@linux.intel.com> Reviewed-by: Wu Fengguang Reviewed-by: Andi Kleen Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/pgtable_64.h | 2 ++ arch/x86/mm/fault.c | 24 +----------------------- arch/x86/mm/init_64.c | 30 ++++++++++++++++++++++++++++++ 3 files changed, 33 insertions(+), 23 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/pgtable_64.h b/arch/x86/include/asm/pgtable_64.h index 076052cd62b..f96ac9bedf7 100644 --- a/arch/x86/include/asm/pgtable_64.h +++ b/arch/x86/include/asm/pgtable_64.h @@ -102,6 +102,8 @@ static inline void native_pgd_clear(pgd_t *pgd) native_set_pgd(pgd, native_make_pgd(0)); } +extern void sync_global_pgds(unsigned long start, unsigned long end); + /* * Conversion functions: convert a page and protection to a page entry, * and a page entry and page directory to the page they refer to. diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 4c4508e8a20..51f7ee71d6c 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -326,29 +326,7 @@ out: void vmalloc_sync_all(void) { - unsigned long address; - - for (address = VMALLOC_START & PGDIR_MASK; address <= VMALLOC_END; - address += PGDIR_SIZE) { - - const pgd_t *pgd_ref = pgd_offset_k(address); - unsigned long flags; - struct page *page; - - if (pgd_none(*pgd_ref)) - continue; - - spin_lock_irqsave(&pgd_lock, flags); - list_for_each_entry(page, &pgd_list, lru) { - pgd_t *pgd; - pgd = (pgd_t *)page_address(page) + pgd_index(address); - if (pgd_none(*pgd)) - set_pgd(pgd, *pgd_ref); - else - BUG_ON(pgd_page_vaddr(*pgd) != pgd_page_vaddr(*pgd_ref)); - } - spin_unlock_irqrestore(&pgd_lock, flags); - } + sync_global_pgds(VMALLOC_START & PGDIR_MASK, VMALLOC_END); } /* diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 9a6674689a2..61a1b4fdecb 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -97,6 +97,36 @@ static int __init nonx32_setup(char *str) } __setup("noexec32=", nonx32_setup); +/* + * When memory was added/removed make sure all the processes MM have + * suitable PGD entries in the local PGD level page. + */ +void sync_global_pgds(unsigned long start, unsigned long end) +{ + unsigned long address; + + for (address = start; address <= end; address += PGDIR_SIZE) { + const pgd_t *pgd_ref = pgd_offset_k(address); + unsigned long flags; + struct page *page; + + if (pgd_none(*pgd_ref)) + continue; + + spin_lock_irqsave(&pgd_lock, flags); + list_for_each_entry(page, &pgd_list, lru) { + pgd_t *pgd; + pgd = (pgd_t *)page_address(page) + pgd_index(address); + if (pgd_none(*pgd)) + set_pgd(pgd, *pgd_ref); + else + BUG_ON(pgd_page_vaddr(*pgd) + != pgd_page_vaddr(*pgd_ref)); + } + spin_unlock_irqrestore(&pgd_lock, flags); + } +} + /* * NOTE: This function is marked __ref because it calls __init function * (alloc_bootmem_pages). It's safe to do it ONLY when after_bootmem == 0. -- cgit v1.2.3-70-g09d2 From 9b861528a8012e7bc4d1f7bae07395b225331477 Mon Sep 17 00:00:00 2001 From: Haicheng Li Date: Fri, 20 Aug 2010 17:50:16 +0800 Subject: x86-64, mem: Update all PGDs for direct mapping and vmemmap mapping changes When memory hotplug-adding happens for a large enough area that a new PGD entry is needed for the direct mapping, the PGDs of other processes would not get updated. This leads to some CPUs oopsing like below when they have to access the unmapped areas. [ 1139.243192] BUG: soft lockup - CPU#0 stuck for 61s! [bash:6534] [ 1139.243195] Modules linked in: ipv6 autofs4 rfcomm l2cap crc16 bluetooth rfkill binfmt_misc dm_mirror dm_region_hash dm_log dm_multipath dm_mod video output sbs sbshc fan battery ac parport_pc lp parport joydev usbhid processor thermal thermal_sys container button rtc_cmos rtc_core rtc_lib i2c_i801 i2c_core pcspkr uhci_hcd ohci_hcd ehci_hcd usbcore [ 1139.243229] irq event stamp: 8538759 [ 1139.243230] hardirqs last enabled at (8538759): [] restore_args+0x0/0x30 [ 1139.243236] hardirqs last disabled at (8538757): [] __do_softirq+0x106/0x146 [ 1139.243240] softirqs last enabled at (8538758): [] __do_softirq+0x137/0x146 [ 1139.243245] softirqs last disabled at (8538743): [] call_softirq+0x1c/0x34 [ 1139.243249] CPU 0: [ 1139.243250] Modules linked in: ipv6 autofs4 rfcomm l2cap crc16 bluetooth rfkill binfmt_misc dm_mirror dm_region_hash dm_log dm_multipath dm_mod video output sbs sbshc fan battery ac parport_pc lp parport joydev usbhid processor thermal thermal_sys container button rtc_cmos rtc_core rtc_lib i2c_i801 i2c_core pcspkr uhci_hcd ohci_hcd ehci_hcd usbcore [ 1139.243284] Pid: 6534, comm: bash Tainted: G M 2.6.32-haicheng-cpuhp #7 QSSC-S4R [ 1139.243287] RIP: 0010:[] [] alloc_arraycache+0x35/0x69 [ 1139.243292] RSP: 0018:ffff8802799f9d78 EFLAGS: 00010286 [ 1139.243295] RAX: ffff8884ffc00000 RBX: ffff8802799f9d98 RCX: 0000000000000000 [ 1139.243297] RDX: 0000000000190018 RSI: 0000000000000001 RDI: ffff8884ffc00010 [ 1139.243300] RBP: ffffffff8100c34e R08: 0000000000000002 R09: 0000000000000000 [ 1139.243303] R10: ffffffff8246dda0 R11: 000000d08246dda0 R12: ffff8802599bfff0 [ 1139.243305] R13: ffff88027904c040 R14: ffff8802799f8000 R15: 0000000000000001 [ 1139.243308] FS: 00007fe81bfe86e0(0000) GS:ffff88000d800000(0000) knlGS:0000000000000000 [ 1139.243311] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 1139.243313] CR2: ffff8884ffc00000 CR3: 000000026cf2d000 CR4: 00000000000006f0 [ 1139.243316] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 1139.243318] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 [ 1139.243321] Call Trace: [ 1139.243324] [] ? alloc_arraycache+0x29/0x69 [ 1139.243328] [] ? cpuup_callback+0x1b0/0x32a [ 1139.243333] [] ? notifier_call_chain+0x33/0x5b [ 1139.243337] [] ? __raw_notifier_call_chain+0x9/0xb [ 1139.243340] [] ? cpu_up+0xb3/0x152 [ 1139.243344] [] ? store_online+0x4d/0x75 [ 1139.243348] [] ? sysdev_store+0x1b/0x1d [ 1139.243351] [] ? sysfs_write_file+0xe5/0x121 [ 1139.243355] [] ? vfs_write+0xae/0x14a [ 1139.243358] [] ? sys_write+0x47/0x6f [ 1139.243362] [] ? system_call_fastpath+0x16/0x1b This patch makes sure to always replicate new direct mapping PGD entries to the PGDs of all processes, as well as ensures corresponding vmemmap mapping gets synced. V1: initial code by Andi Kleen. V2: fix several issues found in testing. V3: as suggested by Wu Fengguang, reuse common code of vmalloc_sync_all(). [ hpa: changed pgd_change from int to bool ] Originally-by: Andi Kleen Signed-off-by: Haicheng Li LKML-Reference: <4C6E4FD8.6080100@linux.intel.com> Reviewed-by: Wu Fengguang Reviewed-by: Andi Kleen Signed-off-by: H. Peter Anvin --- arch/x86/mm/init_64.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 61a1b4fdecb..64e7bc2ded9 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -564,8 +564,9 @@ kernel_physical_mapping_init(unsigned long start, unsigned long end, unsigned long page_size_mask) { - + bool pgd_changed = false; unsigned long next, last_map_addr = end; + unsigned long addr; start = (unsigned long)__va(start); end = (unsigned long)__va(end); @@ -593,7 +594,12 @@ kernel_physical_mapping_init(unsigned long start, spin_lock(&init_mm.page_table_lock); pgd_populate(&init_mm, pgd, __va(pud_phys)); spin_unlock(&init_mm.page_table_lock); + pgd_changed = true; } + + if (pgd_changed) + sync_global_pgds(addr, end); + __flush_tlb_all(); return last_map_addr; @@ -1033,6 +1039,7 @@ vmemmap_populate(struct page *start_page, unsigned long size, int node) } } + sync_global_pgds((unsigned long)start_page, end); return 0; } -- cgit v1.2.3-70-g09d2 From 0444ad93ea2449963132d68753020a6a24d69895 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Thu, 26 Aug 2010 13:57:56 -0400 Subject: x86, iommu: Add IOMMU_INIT macros, .iommu_table section, and iommu_table_entry structure This patch set adds a mechanism to "modularize" the IOMMUs we have on X86. Currently the count of IOMMUs is up to six and they have a complex relationship that requires careful execution order. 'pci_iommu_alloc' does that today, but most folks are unhappy with how it does it. This patch set addresses this and also paves a mechanism to jettison unused IOMMUs during run-time. For details that sparked this, please refer to: http://lkml.org/lkml/2010/8/2/282 The first solution that comes to mind is to convert wholesale the IOMMU detection routines to be called during initcall time frame. Unfortunately that misses the dependency relationship that some of the IOMMUs have (for example: for AMD-Vi IOMMU to work, GART detection MUST run first, and before all of that SWIOTLB MUST run). The second solution would be to introduce a registration call wherein the IOMMU would provide its detection/init routines and as well on what MUST run before it. That would work, except that the 'pci_iommu_alloc' which would run through this list, is called during mem_init. This means we don't have any memory allocator, and it is so early that we haven't yet started running through the initcall_t list. This solution borrows concepts from the 2nd idea and from how MODULE_INIT works. A macro is provided that each IOMMU uses to define it's detect function and early_init (before the memory allocate is active), and as well what other IOMMU MUST run before us. Since most IOMMUs depend on having SWIOTLB run first ("pci_swiotlb_detect") a convenience macro to depends on that is also provided. This macro is similar in design to MODULE_PARAM macro wherein we setup a .iommu_table section in which we populate it with the values that match a struct iommu_table_entry. During bootup we will sort through the array so that the IOMMUs that MUST run before us are first elements in the array. And then we just iterate through them calling the detection routine and if appropiate, the init routines. Signed-off-by: Konrad Rzeszutek Wilk LKML-Reference: <1282845485-8991-2-git-send-email-konrad.wilk@oracle.com> CC: H. Peter Anvin CC: Fujita Tomonori CC: Thomas Gleixner CC: Ingo Molnar Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/iommu_table.h | 95 ++++++++++++++++++++++++++++++++++++++ arch/x86/kernel/vmlinux.lds.S | 7 +++ 2 files changed, 102 insertions(+) create mode 100644 arch/x86/include/asm/iommu_table.h (limited to 'arch/x86') diff --git a/arch/x86/include/asm/iommu_table.h b/arch/x86/include/asm/iommu_table.h new file mode 100644 index 00000000000..435176f96a5 --- /dev/null +++ b/arch/x86/include/asm/iommu_table.h @@ -0,0 +1,95 @@ + +#ifndef _ASM_X86_IOMMU_TABLE_H +#define _ASM_X86_IOMMU_TABLE_H + +#include + +/* + * History lesson: + * The execution chain of IOMMUs in 2.6.36 looks as so: + * + * [xen-swiotlb] + * | + * +----[swiotlb *]--+ + * / | \ + * / | \ + * [GART] [Calgary] [Intel VT-d] + * / + * / + * [AMD-Vi] + * + * *: if SWIOTLB detected 'iommu=soft'/'swiotlb=force' it would skip + * over the rest of IOMMUs and unconditionally initialize the SWIOTLB. + * Also it would surreptitiously initialize set the swiotlb=1 if there were + * more than 4GB and if the user did not pass in 'iommu=off'. The swiotlb + * flag would be turned off by all IOMMUs except the Calgary one. + * + * The IOMMU_INIT* macros allow a similar tree (or more complex if desired) + * to be built by defining who we depend on. + * + * And all that needs to be done is to use one of the macros in the IOMMU + * and the pci-dma.c will take care of the rest. + */ + +struct iommu_table_entry { + initcall_t detect; + initcall_t depend; + void (*early_init)(void); /* No memory allocate available. */ + void (*late_init)(void); /* Yes, can allocate memory. */ +#define IOMMU_FINISH_IF_DETECTED (1<<0) +#define IOMMU_DETECTED (1<<1) + int flags; +}; +/* + * Macro fills out an entry in the .iommu_table that is equivalent + * to the fields that 'struct iommu_table_entry' has. The entries + * that are put in the .iommu_table section are not put in any order + * hence during boot-time we will have to resort them based on + * dependency. */ + + +#define __IOMMU_INIT(_detect, _depend, _early_init, _late_init, _finish)\ + static const struct iommu_table_entry const \ + __iommu_entry_##_detect __used \ + __attribute__ ((unused, __section__(".iommu_table"), \ + aligned((sizeof(void *))))) \ + = {_detect, _depend, _early_init, _late_init, \ + _finish ? IOMMU_FINISH_IF_DETECTED : 0} +/* + * The simplest IOMMU definition. Provide the detection routine + * and it will be run after the SWIOTLB and the other IOMMUs + * that utilize this macro. If the IOMMU is detected (ie, the + * detect routine returns a positive value), the other IOMMUs + * are also checked. You can use IOMMU_INIT_FINISH if you prefer + * to stop detecting the other IOMMUs after yours has been detected. + */ +#define IOMMU_INIT_POST(_detect) \ + __IOMMU_INIT(_detect, pci_swiotlb_detect, 0, 0, 0) + +#define IOMMU_INIT_POST_FINISH(detect) \ + __IOMMU_INIT(_detect, pci_swiotlb_detect, 0, 0, 1) + +/* + * A more sophisticated version of IOMMU_INIT. This variant requires: + * a). A detection routine function. + * b). The name of the detection routine we depend on to get called + * before us. + * c). The init routine which gets called if the detection routine + * returns a positive value from the pci_iommu_alloc. This means + * no presence of a memory allocator. + * d). Similar to the 'init', except that this gets called from pci_iommu_init + * where we do have a memory allocator. + * + * The _CONT vs the _EXIT differs in that the _CONT variant will + * continue detecting other IOMMUs in the call list after the + * the detection routine returns a positive number. The _EXIT will + * stop the execution chain. Both will still call the 'init' and + * 'late_init' functions if they are set. + */ +#define IOMMU_INIT_FINISH(_detect, _depend, _init, _late_init) \ + __IOMMU_INIT(_detect, _depend, _init, _late_init, 1) + +#define IOMMU_INIT(_detect, _depend, _init, _late_init) \ + __IOMMU_INIT(_detect, _depend, _init, _late_init, 0) + +#endif /* _ASM_X86_IOMMU_TABLE_H */ diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S index d0bb52296fa..b92e040466c 100644 --- a/arch/x86/kernel/vmlinux.lds.S +++ b/arch/x86/kernel/vmlinux.lds.S @@ -260,6 +260,13 @@ SECTIONS *(.altinstr_replacement) } + .iommu_table : AT(ADDR(.iommu_table) - LOAD_OFFSET) { + __iommu_table = .; + *(.iommu_table) + . = ALIGN(8); + __iommu_table_end = .; + } + /* * .exit.text is discard at runtime, not link time, to deal with * references from .altinstructions and .eh_frame -- cgit v1.2.3-70-g09d2 From 480125ba49ba62be93beea37770f266846e077ab Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Thu, 26 Aug 2010 13:57:57 -0400 Subject: x86, iommu: Make all IOMMU's detection routines return a value. We return 1 if the IOMMU has been detected. Zero or an error number if we failed to find it. This is in preperation of using the IOMMU_INIT so that we can detect whether an IOMMU is present. I have not tested this for regression on Calgary, nor on AMD Vi chipsets as I don't have that hardware. CC: Muli Ben-Yehuda CC: "Jon D. Mason" CC: "Darrick J. Wong" CC: Jesse Barnes CC: David Woodhouse CC: Chris Wright CC: Yinghai Lu CC: Joerg Roedel CC: H. Peter Anvin CC: Fujita Tomonori Signed-off-by: Konrad Rzeszutek Wilk LKML-Reference: <1282845485-8991-3-git-send-email-konrad.wilk@oracle.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/amd_iommu.h | 4 ++-- arch/x86/include/asm/calgary.h | 4 ++-- arch/x86/include/asm/gart.h | 5 +++-- arch/x86/kernel/amd_iommu_init.c | 8 +++++--- arch/x86/kernel/aperture_64.c | 11 +++++++---- arch/x86/kernel/pci-calgary_64.c | 15 ++++++++------- drivers/pci/dmar.c | 4 +++- include/linux/dmar.h | 6 +++--- 8 files changed, 33 insertions(+), 24 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/amd_iommu.h b/arch/x86/include/asm/amd_iommu.h index 5af2982133b..2798142cdb4 100644 --- a/arch/x86/include/asm/amd_iommu.h +++ b/arch/x86/include/asm/amd_iommu.h @@ -24,11 +24,11 @@ #ifdef CONFIG_AMD_IOMMU -extern void amd_iommu_detect(void); +extern int amd_iommu_detect(void); #else -static inline void amd_iommu_detect(void) { } +static inline int amd_iommu_detect(void) { return -ENODEV; } #endif diff --git a/arch/x86/include/asm/calgary.h b/arch/x86/include/asm/calgary.h index 0918654305a..0d467b33883 100644 --- a/arch/x86/include/asm/calgary.h +++ b/arch/x86/include/asm/calgary.h @@ -62,9 +62,9 @@ struct cal_chipset_ops { extern int use_calgary; #ifdef CONFIG_CALGARY_IOMMU -extern void detect_calgary(void); +extern int detect_calgary(void); #else -static inline void detect_calgary(void) { return; } +static inline int detect_calgary(void) { return -ENODEV; } #endif #endif /* _ASM_X86_CALGARY_H */ diff --git a/arch/x86/include/asm/gart.h b/arch/x86/include/asm/gart.h index 4ac5b0f33fc..d7d1d4c438a 100644 --- a/arch/x86/include/asm/gart.h +++ b/arch/x86/include/asm/gart.h @@ -37,7 +37,7 @@ extern int gart_iommu_aperture_disabled; extern void early_gart_iommu_check(void); extern int gart_iommu_init(void); extern void __init gart_parse_options(char *); -extern void gart_iommu_hole_init(void); +extern int gart_iommu_hole_init(void); #else #define gart_iommu_aperture 0 @@ -50,8 +50,9 @@ static inline void early_gart_iommu_check(void) static inline void gart_parse_options(char *options) { } -static inline void gart_iommu_hole_init(void) +static inline int gart_iommu_hole_init(void) { + return -ENODEV; } #endif diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index 3cc63e2b8dd..0b9e2dc4fc9 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -1382,13 +1382,13 @@ static int __init early_amd_iommu_detect(struct acpi_table_header *table) return 0; } -void __init amd_iommu_detect(void) +int __init amd_iommu_detect(void) { if (no_iommu || (iommu_detected && !gart_iommu_aperture)) - return; + return -ENODEV; if (amd_iommu_disabled) - return; + return -ENODEV; if (acpi_table_parse("IVRS", early_amd_iommu_detect) == 0) { iommu_detected = 1; @@ -1397,7 +1397,9 @@ void __init amd_iommu_detect(void) /* Make sure ACS will be enabled */ pci_request_acs(); + return 1; } + return -ENODEV; } /**************************************************************************** diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c index a2e0caf26e1..afa0dab3302 100644 --- a/arch/x86/kernel/aperture_64.c +++ b/arch/x86/kernel/aperture_64.c @@ -371,7 +371,7 @@ void __init early_gart_iommu_check(void) static int __initdata printed_gart_size_msg; -void __init gart_iommu_hole_init(void) +int __init gart_iommu_hole_init(void) { u32 agp_aper_base = 0, agp_aper_order = 0; u32 aper_size, aper_alloc = 0, aper_order = 0, last_aper_order = 0; @@ -381,7 +381,7 @@ void __init gart_iommu_hole_init(void) if (gart_iommu_aperture_disabled || !fix_aperture || !early_pci_allowed()) - return; + return -ENODEV; printk(KERN_INFO "Checking aperture...\n"); @@ -463,8 +463,9 @@ out: unsigned long n = (32 * 1024 * 1024) << last_aper_order; insert_aperture_resource((u32)last_aper_base, n); + return 1; } - return; + return 0; } if (!fallback_aper_force) { @@ -500,7 +501,7 @@ out: panic("Not enough memory for aperture"); } } else { - return; + return 0; } /* Fix up the north bridges */ @@ -524,4 +525,6 @@ out: } set_up_gart_resume(aper_order, aper_alloc); + + return 1; } diff --git a/arch/x86/kernel/pci-calgary_64.c b/arch/x86/kernel/pci-calgary_64.c index 078d4ec1a9d..28c6b389fee 100644 --- a/arch/x86/kernel/pci-calgary_64.c +++ b/arch/x86/kernel/pci-calgary_64.c @@ -1364,7 +1364,7 @@ static int __init calgary_iommu_init(void) return 0; } -void __init detect_calgary(void) +int __init detect_calgary(void) { int bus; void *tbl; @@ -1378,13 +1378,13 @@ void __init detect_calgary(void) * another HW IOMMU already, bail out. */ if (no_iommu || iommu_detected) - return; + return -ENODEV; if (!use_calgary) - return; + return -ENODEV; if (!early_pci_allowed()) - return; + return -ENODEV; printk(KERN_DEBUG "Calgary: detecting Calgary via BIOS EBDA area\n"); @@ -1410,13 +1410,13 @@ void __init detect_calgary(void) if (!rio_table_hdr) { printk(KERN_DEBUG "Calgary: Unable to locate Rio Grande table " "in EBDA - bailing!\n"); - return; + return -ENODEV; } ret = build_detail_arrays(); if (ret) { printk(KERN_DEBUG "Calgary: build_detail_arrays ret %d\n", ret); - return; + return -ENOMEM; } specified_table_size = determine_tce_table_size((is_kdump_kernel() ? @@ -1464,7 +1464,7 @@ void __init detect_calgary(void) x86_init.iommu.iommu_init = calgary_iommu_init; } - return; + return calgary_found; cleanup: for (--bus; bus >= 0; --bus) { @@ -1473,6 +1473,7 @@ cleanup: if (info->tce_space) free_tce_table(info->tce_space); } + return -ENOMEM; } static int __init calgary_parse_options(char *p) diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c index 0a19708074c..5fa64ea5416 100644 --- a/drivers/pci/dmar.c +++ b/drivers/pci/dmar.c @@ -687,7 +687,7 @@ failed: return 0; } -void __init detect_intel_iommu(void) +int __init detect_intel_iommu(void) { int ret; @@ -723,6 +723,8 @@ void __init detect_intel_iommu(void) } early_acpi_os_unmap_memory(dmar_tbl, dmar_tbl_size); dmar_tbl = NULL; + + return (ret ? 1 : -ENODEV); } diff --git a/include/linux/dmar.h b/include/linux/dmar.h index d7cecc90ed3..a2060204151 100644 --- a/include/linux/dmar.h +++ b/include/linux/dmar.h @@ -57,15 +57,15 @@ extern int dmar_table_init(void); extern int dmar_dev_scope_init(void); /* Intel IOMMU detection */ -extern void detect_intel_iommu(void); +extern int detect_intel_iommu(void); extern int enable_drhd_fault_handling(void); extern int parse_ioapics_under_ir(void); extern int alloc_iommu(struct dmar_drhd_unit *); #else -static inline void detect_intel_iommu(void) +static inline int detect_intel_iommu(void) { - return; + return -ENODEV; } static inline int dmar_table_init(void) -- cgit v1.2.3-70-g09d2 From 5bef80a4b826b9cee1c6aec7ecc371ec395260cc Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Thu, 26 Aug 2010 13:57:58 -0400 Subject: x86, iommu: Add proper dependency sort routine (and sanity check). We are using a very simple sort routine which sorts the .iommu_table array in the order of dependencies. Specifically each structure of iommu_table_entry has a field 'depend' which contains the function pointer to the IOMMU that MUST be run before us. We sort the array of structures so that the struct iommu_table_entry with no 'depend' field are first, and then the subsequent ones are the ones for which the 'depend' function has been already invoked (in other words, precede us). Using the kernel's version 'sort', which is a mergeheap is feasible, but would require making the comparison operator scan recursivly the array to satisfy the "heapify" process: setting the levels properly. The end result would much more complex than it should be an it is just much simpler to utilize this simple sort routine. Signed-off-by: Konrad Rzeszutek Wilk LKML-Reference: <1282845485-8991-4-git-send-email-konrad.wilk@oracle.com> CC: H. Peter Anvin CC: Fujita Tomonori Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/iommu_table.h | 6 +++ arch/x86/kernel/Makefile | 1 + arch/x86/kernel/pci-iommu_table.c | 89 ++++++++++++++++++++++++++++++++++++++ 3 files changed, 96 insertions(+) create mode 100644 arch/x86/kernel/pci-iommu_table.c (limited to 'arch/x86') diff --git a/arch/x86/include/asm/iommu_table.h b/arch/x86/include/asm/iommu_table.h index 435176f96a5..2124e3ef6f9 100644 --- a/arch/x86/include/asm/iommu_table.h +++ b/arch/x86/include/asm/iommu_table.h @@ -92,4 +92,10 @@ struct iommu_table_entry { #define IOMMU_INIT(_detect, _depend, _init, _late_init) \ __IOMMU_INIT(_detect, _depend, _init, _late_init, 0) +void sort_iommu_table(struct iommu_table_entry *start, + struct iommu_table_entry *finish); + +void check_iommu_entries(struct iommu_table_entry *start, + struct iommu_table_entry *finish); + #endif /* _ASM_X86_IOMMU_TABLE_H */ diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 0925676266b..6817546595e 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -42,6 +42,7 @@ obj-y += bootflag.o e820.o obj-y += pci-dma.o quirks.o i8237.o topology.o kdebugfs.o obj-y += alternative.o i8253.o pci-nommu.o hw_breakpoint.o obj-y += tsc.o io_delay.o rtc.o +obj-y += pci-iommu_table.o obj-$(CONFIG_X86_TRAMPOLINE) += trampoline.o obj-y += process.o diff --git a/arch/x86/kernel/pci-iommu_table.c b/arch/x86/kernel/pci-iommu_table.c new file mode 100644 index 00000000000..55d745ec118 --- /dev/null +++ b/arch/x86/kernel/pci-iommu_table.c @@ -0,0 +1,89 @@ +#include +#include +#include +#include + + +#define DEBUG 1 + +static struct iommu_table_entry * __init +find_dependents_of(struct iommu_table_entry *start, + struct iommu_table_entry *finish, + struct iommu_table_entry *q) +{ + struct iommu_table_entry *p; + + if (!q) + return NULL; + + for (p = start; p < finish; p++) + if (p->detect == q->depend) + return p; + + return NULL; +} + + +void __init sort_iommu_table(struct iommu_table_entry *start, + struct iommu_table_entry *finish) { + + struct iommu_table_entry *p, *q, tmp; + + for (p = start; p < finish; p++) { +again: + q = find_dependents_of(start, finish, p); + /* We are bit sneaky here. We use the memory address to figure + * out if the node we depend on is past our point, if so, swap. + */ + if (q > p) { + tmp = *p; + memmove(p, q, sizeof(*p)); + *q = tmp; + goto again; + } + } + +} + +#ifdef DEBUG +void __init check_iommu_entries(struct iommu_table_entry *start, + struct iommu_table_entry *finish) +{ + struct iommu_table_entry *p, *q, *x; + char sym_p[KSYM_SYMBOL_LEN]; + char sym_q[KSYM_SYMBOL_LEN]; + + /* Simple cyclic dependency checker. */ + for (p = start; p < finish; p++) { + q = find_dependents_of(start, finish, p); + x = find_dependents_of(start, finish, q); + if (p == x) { + sprint_symbol(sym_p, (unsigned long)p->detect); + sprint_symbol(sym_q, (unsigned long)q->detect); + + printk(KERN_ERR "CYCLIC DEPENDENCY FOUND! %s depends" \ + " on %s and vice-versa. BREAKING IT.\n", + sym_p, sym_q); + /* Heavy handed way..*/ + x->depend = 0; + } + } + + for (p = start; p < finish; p++) { + q = find_dependents_of(p, finish, p); + if (q && q > p) { + sprint_symbol(sym_p, (unsigned long)p->detect); + sprint_symbol(sym_q, (unsigned long)q->detect); + + printk(KERN_ERR "EXECUTION ORDER INVALID! %s "\ + "should be called before %s!\n", + sym_p, sym_q); + } + } +} +#else +inline void check_iommu_entries(struct iommu_table_entry *start, + struct iommu_table_entry *finish) +{ +} +#endif -- cgit v1.2.3-70-g09d2 From efa631c26d3bb1162b8f95008801db602217f52b Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Thu, 26 Aug 2010 13:57:59 -0400 Subject: x86, swiotlb: Simplify SWIOTLB pci_swiotlb_detect routine. In 'pci_swiotlb_detect' we used to do two different things: a). If user provided 'iommu=soft' or 'swiotlb=force' we would set swiotlb=1 and return 1 (and forcing pci-dma.c to call pci_swiotlb_init() immediately). b). If 4GB or more would be detected and if user did not specify iommu=off, we would set 'swiotlb=1' and return whatever 'a)' figured out. We simplify this by splitting a) and b) in two different routines. CC: Fujita Tomonori Signed-off-by: Konrad Rzeszutek Wilk LKML-Reference: <1282845485-8991-5-git-send-email-konrad.wilk@oracle.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/iommu_table.h | 4 ++-- arch/x86/include/asm/swiotlb.h | 13 +++++++++++-- arch/x86/kernel/pci-dma.c | 4 +++- arch/x86/kernel/pci-swiotlb.c | 33 +++++++++++++++++++++++++++------ 4 files changed, 43 insertions(+), 11 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/iommu_table.h b/arch/x86/include/asm/iommu_table.h index 2124e3ef6f9..df55a78888e 100644 --- a/arch/x86/include/asm/iommu_table.h +++ b/arch/x86/include/asm/iommu_table.h @@ -64,10 +64,10 @@ struct iommu_table_entry { * to stop detecting the other IOMMUs after yours has been detected. */ #define IOMMU_INIT_POST(_detect) \ - __IOMMU_INIT(_detect, pci_swiotlb_detect, 0, 0, 0) + __IOMMU_INIT(_detect, pci_swiotlb_detect_4gb, 0, 0, 0) #define IOMMU_INIT_POST_FINISH(detect) \ - __IOMMU_INIT(_detect, pci_swiotlb_detect, 0, 0, 1) + __IOMMU_INIT(_detect, pci_swiotlb_detect_4gb, 0, 0, 1) /* * A more sophisticated version of IOMMU_INIT. This variant requires: diff --git a/arch/x86/include/asm/swiotlb.h b/arch/x86/include/asm/swiotlb.h index 8085277e1b8..977f1761a25 100644 --- a/arch/x86/include/asm/swiotlb.h +++ b/arch/x86/include/asm/swiotlb.h @@ -5,17 +5,26 @@ #ifdef CONFIG_SWIOTLB extern int swiotlb; -extern int __init pci_swiotlb_detect(void); +extern int __init pci_swiotlb_detect_override(void); +extern int __init pci_swiotlb_detect_4gb(void); extern void __init pci_swiotlb_init(void); +extern void __init pci_swiotlb_late_init(void); #else #define swiotlb 0 -static inline int pci_swiotlb_detect(void) +static inline int pci_swiotlb_detect_override(void) +{ + return 0; +} +static inline int pci_swiotlb_detect_4gb(void) { return 0; } static inline void pci_swiotlb_init(void) { } +static inline void pci_swiotlb_late_init(void) +{ +} #endif static inline void dma_mark_clean(void *addr, size_t size) {} diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index 9f07cfcbd3a..1b3beb5075e 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -133,9 +133,11 @@ void __init pci_iommu_alloc(void) /* free the range so iommu could get some range less than 4G */ dma32_free_bootmem(); - if (pci_xen_swiotlb_detect() || pci_swiotlb_detect()) + if (pci_xen_swiotlb_detect() || pci_swiotlb_detect_override()) goto out; + pci_swiotlb_detect_4gb(); + gart_iommu_hole_init(); detect_calgary(); diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c index a5bc528d432..c7a72faeb14 100644 --- a/arch/x86/kernel/pci-swiotlb.c +++ b/arch/x86/kernel/pci-swiotlb.c @@ -41,24 +41,33 @@ static struct dma_map_ops swiotlb_dma_ops = { }; /* - * pci_swiotlb_detect - set swiotlb to 1 if necessary + * pci_swiotlb_detect_override - set swiotlb to 1 if necessary * * This returns non-zero if we are forced to use swiotlb (by the boot * option). */ -int __init pci_swiotlb_detect(void) +int __init pci_swiotlb_detect_override(void) { int use_swiotlb = swiotlb | swiotlb_force; + if (swiotlb_force) + swiotlb = 1; + + return use_swiotlb; +} + +/* + * if 4GB or more detected (and iommu=off not set) return 1 + * and set swiotlb to 1. + */ +int __init pci_swiotlb_detect_4gb(void) +{ /* don't initialize swiotlb if iommu=off (no_iommu=1) */ #ifdef CONFIG_X86_64 if (!no_iommu && max_pfn > MAX_DMA32_PFN) swiotlb = 1; #endif - if (swiotlb_force) - swiotlb = 1; - - return use_swiotlb; + return swiotlb; } void __init pci_swiotlb_init(void) @@ -68,3 +77,15 @@ void __init pci_swiotlb_init(void) dma_ops = &swiotlb_dma_ops; } } + +void __init pci_swiotlb_late_init(void) +{ + /* An IOMMU turned us off. */ + if (!swiotlb) + swiotlb_free(); + else { + printk(KERN_INFO "PCI-DMA: " + "Using software bounce buffering for IO (SWIOTLB)\n"); + swiotlb_print_info(); + } +} -- cgit v1.2.3-70-g09d2 From c116c5457c46edb767df6f4e36d4905e3514ad37 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Thu, 26 Aug 2010 13:58:00 -0400 Subject: x86, swiotlb: Make SWIOTLB use IOMMU_INIT_* macros. We utilize the IOMMU_INIT macros to create this dependency: [pci_xen_swiotlb_detect] | [pci_swiotlb_detect_override] | [pci_swiotlb_detect_4gb] And set the SWIOTLB IOMMU_INIT to utilize 'pci_swiotlb_init' for .init and 'pci_swiotlb_late_init' for .late_init. Signed-off-by: Konrad Rzeszutek Wilk LKML-Reference: <1282845485-8991-6-git-send-email-konrad.wilk@oracle.com> CC: Fujita Tomonori Signed-off-by: H. Peter Anvin --- arch/x86/kernel/pci-swiotlb.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c index c7a72faeb14..8f972cbddef 100644 --- a/arch/x86/kernel/pci-swiotlb.c +++ b/arch/x86/kernel/pci-swiotlb.c @@ -10,7 +10,8 @@ #include #include #include - +#include +#include int swiotlb __read_mostly; static void *x86_swiotlb_alloc_coherent(struct device *hwdev, size_t size, @@ -55,6 +56,10 @@ int __init pci_swiotlb_detect_override(void) return use_swiotlb; } +IOMMU_INIT_FINISH(pci_swiotlb_detect_override, + pci_xen_swiotlb_detect, + pci_swiotlb_init, + pci_swiotlb_late_init); /* * if 4GB or more detected (and iommu=off not set) return 1 @@ -69,6 +74,10 @@ int __init pci_swiotlb_detect_4gb(void) #endif return swiotlb; } +IOMMU_INIT(pci_swiotlb_detect_4gb, + pci_swiotlb_detect_override, + pci_swiotlb_init, + pci_swiotlb_late_init); void __init pci_swiotlb_init(void) { -- cgit v1.2.3-70-g09d2 From 5cb3a267939a223eb84692d229569d2ef493d7ca Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Thu, 26 Aug 2010 13:58:01 -0400 Subject: x86, xen-swiotlb: Make Xen-SWIOTLB use IOMMU_INIT_* macros. We utilize the IOMMU_INIT macros to create this dependency: [null] | [pci_xen_swiotlb_detect] | [pci_swiotlb_detect_override] | [pci_swiotlb_detect_4gb] In other words, we set 'pci_xen_swiotlb_detect' to be the first detection to be run during start. CC: Fujita Tomonori Cc: Jeremy Fitzhardinge Signed-off-by: Konrad Rzeszutek Wilk LKML-Reference: <1282845485-8991-7-git-send-email-konrad.wilk@oracle.com> Signed-off-by: H. Peter Anvin --- arch/x86/xen/pci-swiotlb-xen.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/xen/pci-swiotlb-xen.c b/arch/x86/xen/pci-swiotlb-xen.c index a013ec9d0c5..22471001b74 100644 --- a/arch/x86/xen/pci-swiotlb-xen.c +++ b/arch/x86/xen/pci-swiotlb-xen.c @@ -5,6 +5,7 @@ #include #include +#include int xen_swiotlb __read_mostly; @@ -56,3 +57,7 @@ void __init pci_xen_swiotlb_init(void) dma_ops = &xen_swiotlb_dma_ops; } } +IOMMU_INIT_FINISH(pci_xen_swiotlb_detect, + 0, + pci_xen_swiotlb_init, + 0); -- cgit v1.2.3-70-g09d2 From d2aa232f3d0b5a3e22f91b736fe68eddcf0d5ea3 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Thu, 26 Aug 2010 13:58:02 -0400 Subject: x86, calgary: Make Calgary IOMMU use IOMMU_INIT_* macros. We utilize the IOMMU_INIT macros to create this dependency: [pci_xen_swiotlb_detect] | [pci_swiotlb_detect_override] | [pci_swiotlb_detect_4gb] | [detect_calgary] Meaning that 'detect_calgary' is going to be called after 'pci_swiotlb_detect'. Signed-off-by: Konrad Rzeszutek Wilk LKML-Reference: <1282845485-8991-8-git-send-email-konrad.wilk@oracle.com> CC: Muli Ben-Yehuda CC: "Jon D. Mason" CC: "Darrick J. Wong" CC: Fujita Tomonori Signed-off-by: H. Peter Anvin --- arch/x86/kernel/pci-calgary_64.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/pci-calgary_64.c b/arch/x86/kernel/pci-calgary_64.c index 28c6b389fee..f56a117cef6 100644 --- a/arch/x86/kernel/pci-calgary_64.c +++ b/arch/x86/kernel/pci-calgary_64.c @@ -47,6 +47,7 @@ #include #include #include +#include #ifdef CONFIG_CALGARY_IOMMU_ENABLED_BY_DEFAULT int use_calgary __read_mostly = 1; @@ -1595,3 +1596,5 @@ static int __init calgary_fixup_tce_spaces(void) * and before device_initcall. */ rootfs_initcall(calgary_fixup_tce_spaces); + +IOMMU_INIT_POST(detect_calgary); -- cgit v1.2.3-70-g09d2 From 22e6daf41ba28ddc06295e42859b266f737b3e99 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Thu, 26 Aug 2010 13:58:03 -0400 Subject: x86, GART/AMD-VI: Make AMD GART and IOMMU use IOMMU_INIT_* macros. We utilize the IOMMU_INIT macros to create this dependency: [null] | [pci_xen_swiotlb_detect] | [pci_swiotlb_detect_override] | [pci_swiotlb_detect_4gb] | +-------+--------+ / \ [detect_calgary] [gart_iommu_hole_init] | [amd_iommu_detect] Meaning that 'amd_iommu_detect' will be called after 'gart_iommu_hole_init'. Signed-off-by: Konrad Rzeszutek Wilk LKML-Reference: <1282845485-8991-9-git-send-email-konrad.wilk@oracle.com> CC: Fujita Tomonori CC: Joerg Roedel CC: Thomas Gleixner CC: Ingo Molnar Signed-off-by: H. Peter Anvin --- arch/x86/kernel/amd_iommu_init.c | 7 ++++++- arch/x86/kernel/pci-gart_64.c | 2 ++ 2 files changed, 8 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index 0b9e2dc4fc9..26a5e438521 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -31,7 +31,7 @@ #include #include #include - +#include /* * definitions for the ACPI scanning code */ @@ -1430,3 +1430,8 @@ static int __init parse_amd_iommu_options(char *str) __setup("amd_iommu_dump", parse_amd_iommu_dump); __setup("amd_iommu=", parse_amd_iommu_options); + +IOMMU_INIT_FINISH(amd_iommu_detect, + gart_iommu_hole_init, + 0, + 0); diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index 0f7f130caa6..de9734b100a 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -41,6 +41,7 @@ #include #include #include +#include static unsigned long iommu_bus_base; /* GART remapping area (physical) */ static unsigned long iommu_size; /* size of remapping area bytes */ @@ -896,3 +897,4 @@ void __init gart_parse_options(char *p) } } } +IOMMU_INIT_POST(gart_iommu_hole_init); -- cgit v1.2.3-70-g09d2 From ee1f284f38c8dfcbc7b656915a039dde016de7d3 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Thu, 26 Aug 2010 13:58:05 -0400 Subject: x86, iommu: Utilize the IOMMU_INIT macros functionality. We remove all of the sub-platform detection/init routines and instead use on the .iommu_table array of structs to call the .early_init if .detect returned a positive value. Also we can stop detecting other IOMMUs if the IOMMU used the _FINISH type macro. During the 'pci_iommu_init' stage, we call .init for the second-stage initialization if it was defined. Currently only SWIOTLB has this defined and it used to de-allocate the SWIOTLB if the other detected IOMMUs have deemed it unnecessary to use SWIOTLB. Signed-off-by: Konrad Rzeszutek Wilk LKML-Reference: <1282845485-8991-11-git-send-email-konrad.wilk@oracle.com> CC: Fujita Tomonori CC: Thomas Gleixner CC: Ingo Molnar Signed-off-by: Konrad Rzeszutek Wilk Signed-off-by: H. Peter Anvin --- arch/x86/kernel/pci-dma.c | 46 +++++++++++++++++++++------------------------- 1 file changed, 21 insertions(+), 25 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index 1b3beb5075e..9ea999a4dcc 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -11,9 +11,8 @@ #include #include #include -#include #include -#include +#include static int forbid_dac __read_mostly; @@ -45,6 +44,8 @@ int iommu_detected __read_mostly = 0; */ int iommu_pass_through __read_mostly; +extern struct iommu_table_entry __iommu_table[], __iommu_table_end[]; + /* Dummy device used for NULL arguments (normally ISA). */ struct device x86_dma_fallback_dev = { .init_name = "fallback device", @@ -130,28 +131,24 @@ static void __init dma32_free_bootmem(void) void __init pci_iommu_alloc(void) { + struct iommu_table_entry *p; + /* free the range so iommu could get some range less than 4G */ dma32_free_bootmem(); - if (pci_xen_swiotlb_detect() || pci_swiotlb_detect_override()) - goto out; - - pci_swiotlb_detect_4gb(); - - gart_iommu_hole_init(); - - detect_calgary(); - - detect_intel_iommu(); + sort_iommu_table(__iommu_table, __iommu_table_end); + check_iommu_entries(__iommu_table, __iommu_table_end); - /* needs to be called after gart_iommu_hole_init */ - amd_iommu_detect(); -out: - pci_xen_swiotlb_init(); - - pci_swiotlb_init(); + for (p = __iommu_table; p < __iommu_table_end; p++) { + if (p && p->detect && p->detect() > 0) { + p->flags |= IOMMU_DETECTED; + if (p->early_init) + p->early_init(); + if (p->flags & IOMMU_FINISH_IF_DETECTED) + break; + } + } } - void *dma_generic_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_addr, gfp_t flag) { @@ -294,6 +291,7 @@ EXPORT_SYMBOL(dma_supported); static int __init pci_iommu_init(void) { + struct iommu_table_entry *p; dma_debug_init(PREALLOC_DMA_DEBUG_ENTRIES); #ifdef CONFIG_PCI @@ -301,12 +299,10 @@ static int __init pci_iommu_init(void) #endif x86_init.iommu.iommu_init(); - if (swiotlb || xen_swiotlb) { - printk(KERN_INFO "PCI-DMA: " - "Using software bounce buffering for IO (SWIOTLB)\n"); - swiotlb_print_info(); - } else - swiotlb_free(); + for (p = __iommu_table; p < __iommu_table_end; p++) { + if (p && (p->flags & IOMMU_DETECTED) && p->late_init) + p->late_init(); + } return 0; } -- cgit v1.2.3-70-g09d2 From 660a293ea9be709b893d371fbc0328fcca33c33a Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Tue, 27 Jul 2010 16:06:28 +0800 Subject: x86, mm: Make spurious_fault check explicitly check the PRESENT bit pte_present() returns true even present bit isn't set but _PAGE_PROTNONE (global bit) bit is set. While with CONFIG_DEBUG_PAGEALLOC, free pages have global bit set but present bit clear. This patch makes we could catch free pages access with CONFIG_DEBUG_PAGEALLOC enabled. [ hpa: added a comment in the code as a warning to janitors ] Signed-off-by: Shaohua Li LKML-Reference: <1280217988.32400.75.camel@sli10-desk.sh.intel.com> Signed-off-by: H. Peter Anvin --- arch/x86/mm/fault.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 51f7ee71d6c..caec22906d7 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -872,8 +872,14 @@ spurious_fault(unsigned long error_code, unsigned long address) if (pmd_large(*pmd)) return spurious_fault_check(error_code, (pte_t *) pmd); + /* + * Note: don't use pte_present() here, since it returns true + * if the _PAGE_PROTNONE bit is set. However, this aliases the + * _PAGE_GLOBAL bit, which for kernel pages give false positives + * when CONFIG_DEBUG_PAGEALLOC is used. + */ pte = pte_offset_kernel(pmd, address); - if (!pte_present(*pte)) + if (!(pte_flags(*pte) & _PAGE_PRESENT)) return 0; ret = spurious_fault_check(error_code, pte); -- cgit v1.2.3-70-g09d2 From fb74fb6db91abc3c1ceeb9d2c17b44866a12c63e Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 25 Aug 2010 13:39:15 -0700 Subject: x86, memblock: Add memblock_x86_find_in_range_size() size is returned according free range. Will be used to find free ranges for early_memtest and memory corruption check Do not mess it up with lib/memblock.c yet. Signed-off-by: Yinghai Lu Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/memblock.h | 8 ++++ arch/x86/mm/Makefile | 2 + arch/x86/mm/memblock.c | 87 +++++++++++++++++++++++++++++++++++++++++ 3 files changed, 97 insertions(+) create mode 100644 arch/x86/include/asm/memblock.h create mode 100644 arch/x86/mm/memblock.c (limited to 'arch/x86') diff --git a/arch/x86/include/asm/memblock.h b/arch/x86/include/asm/memblock.h new file mode 100644 index 00000000000..c14219a2616 --- /dev/null +++ b/arch/x86/include/asm/memblock.h @@ -0,0 +1,8 @@ +#ifndef _X86_MEMBLOCK_H +#define _X86_MEMBLOCK_H + +#define ARCH_DISCARD_MEMBLOCK + +u64 memblock_x86_find_in_range_size(u64 start, u64 *sizep, u64 align); + +#endif diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile index a4c768397ba..55543397a8a 100644 --- a/arch/x86/mm/Makefile +++ b/arch/x86/mm/Makefile @@ -26,4 +26,6 @@ obj-$(CONFIG_NUMA) += numa.o numa_$(BITS).o obj-$(CONFIG_K8_NUMA) += k8topology_64.o obj-$(CONFIG_ACPI_NUMA) += srat_$(BITS).o +obj-$(CONFIG_HAVE_MEMBLOCK) += memblock.o + obj-$(CONFIG_MEMTEST) += memtest.o diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c new file mode 100644 index 00000000000..26ba46234cb --- /dev/null +++ b/arch/x86/mm/memblock.c @@ -0,0 +1,87 @@ +#include +#include +#include +#include +#include +#include +#include +#include + +/* Check for already reserved areas */ +static inline bool __init bad_addr_size(u64 *addrp, u64 *sizep, u64 align) +{ + struct memblock_region *r; + u64 addr = *addrp, last; + u64 size = *sizep; + bool changed = false; + +again: + last = addr + size; + for_each_memblock(reserved, r) { + if (last > r->base && addr < r->base) { + size = r->base - addr; + changed = true; + goto again; + } + if (last > (r->base + r->size) && addr < (r->base + r->size)) { + addr = round_up(r->base + r->size, align); + size = last - addr; + changed = true; + goto again; + } + if (last <= (r->base + r->size) && addr >= r->base) { + (*sizep)++; + return false; + } + } + if (changed) { + *addrp = addr; + *sizep = size; + } + return changed; +} + +static u64 __init __memblock_x86_find_in_range_size(u64 ei_start, u64 ei_last, u64 start, + u64 *sizep, u64 align) +{ + u64 addr, last; + + addr = round_up(ei_start, align); + if (addr < start) + addr = round_up(start, align); + if (addr >= ei_last) + goto out; + *sizep = ei_last - addr; + while (bad_addr_size(&addr, sizep, align) && addr + *sizep <= ei_last) + ; + last = addr + *sizep; + if (last > ei_last) + goto out; + + return addr; + +out: + return MEMBLOCK_ERROR; +} + +/* + * Find next free range after start, and size is returned in *sizep + */ +u64 __init memblock_x86_find_in_range_size(u64 start, u64 *sizep, u64 align) +{ + struct memblock_region *r; + + for_each_memblock(memory, r) { + u64 ei_start = r->base; + u64 ei_last = ei_start + r->size; + u64 addr; + + addr = __memblock_x86_find_in_range_size(ei_start, ei_last, start, + sizep, align); + + if (addr != MEMBLOCK_ERROR) + return addr; + } + + return MEMBLOCK_ERROR; +} -- cgit v1.2.3-70-g09d2 From f88eff74aa848e58b1ea49768c0bbb874b31357f Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 25 Aug 2010 13:39:15 -0700 Subject: bootmem, x86: Add weak version of reserve_bootmem_generic It will be used memblock_x86_to_bootmem converting It is an wrapper for reserve_bootmem, and x86 64bit is using special one. Also clean up that version for x86_64. We don't need to take care of numa path for that, bootmem can handle it how Signed-off-by: Yinghai Lu Signed-off-by: H. Peter Anvin --- arch/x86/mm/init_32.c | 6 ------ arch/x86/mm/init_64.c | 20 ++------------------ mm/bootmem.c | 6 ++++++ 3 files changed, 8 insertions(+), 24 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index bca79091b9d..90e054589aa 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -1069,9 +1069,3 @@ void mark_rodata_ro(void) #endif } #endif - -int __init reserve_bootmem_generic(unsigned long phys, unsigned long len, - int flags) -{ - return reserve_bootmem(phys, len, flags); -} diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index ee41bba315d..634fa0884a4 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -799,13 +799,10 @@ void mark_rodata_ro(void) #endif +#ifndef CONFIG_NO_BOOTMEM int __init reserve_bootmem_generic(unsigned long phys, unsigned long len, int flags) { -#ifdef CONFIG_NUMA - int nid, next_nid; - int ret; -#endif unsigned long pfn = phys >> PAGE_SHIFT; if (pfn >= max_pfn) { @@ -821,21 +818,7 @@ int __init reserve_bootmem_generic(unsigned long phys, unsigned long len, return -EFAULT; } - /* Should check here against the e820 map to avoid double free */ -#ifdef CONFIG_NUMA - nid = phys_to_nid(phys); - next_nid = phys_to_nid(phys + len - 1); - if (nid == next_nid) - ret = reserve_bootmem_node(NODE_DATA(nid), phys, len, flags); - else - ret = reserve_bootmem(phys, len, flags); - - if (ret != 0) - return ret; - -#else reserve_bootmem(phys, len, flags); -#endif if (phys+len <= MAX_DMA_PFN*PAGE_SIZE) { dma_reserve += len / PAGE_SIZE; @@ -844,6 +827,7 @@ int __init reserve_bootmem_generic(unsigned long phys, unsigned long len, return 0; } +#endif int kern_addr_valid(unsigned long addr) { diff --git a/mm/bootmem.c b/mm/bootmem.c index 142c84a5499..bde170dd2fd 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -526,6 +526,12 @@ int __init reserve_bootmem(unsigned long addr, unsigned long size, } #ifndef CONFIG_NO_BOOTMEM +int __weak __init reserve_bootmem_generic(unsigned long phys, unsigned long len, + int flags) +{ + return reserve_bootmem(phys, len, flags); +} + static unsigned long __init align_idx(struct bootmem_data *bdata, unsigned long idx, unsigned long step) { -- cgit v1.2.3-70-g09d2 From 27de794365786b4cdc3461ed4e23af2a33f40612 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 25 Aug 2010 13:39:15 -0700 Subject: x86, memblock: Add memblock_x86_to_bootmem() memblock_x86_to_bootmem() will reserve memblock.reserved.region in bootmem after bootmem is set up. We can use it to with all arches that support memblock later. Signed-off-by: Yinghai Lu Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/memblock.h | 1 + arch/x86/mm/memblock.c | 29 +++++++++++++++++++++++++++++ 2 files changed, 30 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/memblock.h b/arch/x86/include/asm/memblock.h index c14219a2616..69cf853e931 100644 --- a/arch/x86/include/asm/memblock.h +++ b/arch/x86/include/asm/memblock.h @@ -4,5 +4,6 @@ #define ARCH_DISCARD_MEMBLOCK u64 memblock_x86_find_in_range_size(u64 start, u64 *sizep, u64 align); +void memblock_x86_to_bootmem(u64 start, u64 end); #endif diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index 26ba46234cb..8101084d452 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -85,3 +85,32 @@ u64 __init memblock_x86_find_in_range_size(u64 start, u64 *sizep, u64 align) return MEMBLOCK_ERROR; } + +#ifndef CONFIG_NO_BOOTMEM +void __init memblock_x86_to_bootmem(u64 start, u64 end) +{ + int count; + u64 final_start, final_end; + struct memblock_region *r; + + /* Take out region array itself */ + memblock_free_reserved_regions(); + + count = memblock.reserved.cnt; + pr_info("(%d early reservations) ==> bootmem [%010llx-%010llx]\n", count, start, end - 1); + for_each_memblock(reserved, r) { + pr_info(" [%010llx-%010llx] ", (u64)r->base, (u64)r->base + r->size - 1); + final_start = max(start, r->base); + final_end = min(end, r->base + r->size); + if (final_start >= final_end) { + pr_cont("\n"); + continue; + } + pr_cont(" ==> [%010llx-%010llx]\n", final_start, final_end - 1); + reserve_bootmem_generic(final_start, final_end - final_start, BOOTMEM_DEFAULT); + } + + /* Put region array back ? */ + memblock_reserve_reserved_regions(); +} +#endif -- cgit v1.2.3-70-g09d2 From 9dc5d569c133819c1ce069ebb1d771c62de32580 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 25 Aug 2010 13:39:15 -0700 Subject: x86, memblock: Add memblock_x86_reserve_range/memblock_x86_free_range They are wrappers for core versions, which take start/end/name instead of base/size. This will make x86 conversion eaasier. could add more debug print out -v2: change get_max_mapped() to memblock.default_alloc_limit according to Michael Ellerman and Ben change to memblock_x86_reserve_range and memblock_x86_free_range according to Michael Ellerman -v3: call check_and_double after reserve/free, so could avoid to use find_memblock_area. Suggested by Michael Ellerman Signed-off-by: Yinghai Lu Cc: Benjamin Herrenschmidt Cc: Michael Ellerman Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/memblock.h | 3 +++ arch/x86/mm/memblock.c | 22 ++++++++++++++++++++++ 2 files changed, 25 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/memblock.h b/arch/x86/include/asm/memblock.h index 69cf853e931..e11ddf059fa 100644 --- a/arch/x86/include/asm/memblock.h +++ b/arch/x86/include/asm/memblock.h @@ -6,4 +6,7 @@ u64 memblock_x86_find_in_range_size(u64 start, u64 *sizep, u64 align); void memblock_x86_to_bootmem(u64 start, u64 end); +void memblock_x86_reserve_range(u64 start, u64 end, char *name); +void memblock_x86_free_range(u64 start, u64 end); + #endif diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index 8101084d452..9829eaf1dbd 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -114,3 +114,25 @@ void __init memblock_x86_to_bootmem(u64 start, u64 end) memblock_reserve_reserved_regions(); } #endif + +void __init memblock_x86_reserve_range(u64 start, u64 end, char *name) +{ + if (start == end) + return; + + if (WARN_ONCE(start > end, "memblock_x86_reserve_range: wrong range [%#llx, %#llx]\n", start, end)) + return; + + memblock_reserve(start, end - start); +} + +void __init memblock_x86_free_range(u64 start, u64 end) +{ + if (start == end) + return; + + if (WARN_ONCE(start > end, "memblock_x86_free_range: wrong range [%#llx, %#llx]\n", start, end)) + return; + + memblock_free(start, end - start); +} -- cgit v1.2.3-70-g09d2 From 4d5cf86ce187c0d3a4cdf233ab0cc6526ccbe01f Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 25 Aug 2010 13:39:16 -0700 Subject: x86, memblock: Add get_free_all_memory_range() get_free_all_memory_range is for CONFIG_NO_BOOTMEM=y, and will be called by free_all_memory_core_early(). It will use early_node_map aka active ranges subtract memblock.reserved to get all free range, and those ranges will convert to slab pages. -v4: increase range size Signed-off-by: Yinghai Lu Cc: Jan Beulich Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/memblock.h | 2 + arch/x86/mm/memblock.c | 98 ++++++++++++++++++++++++++++++++++++++++- 2 files changed, 99 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/memblock.h b/arch/x86/include/asm/memblock.h index e11ddf059fa..72639ce65e8 100644 --- a/arch/x86/include/asm/memblock.h +++ b/arch/x86/include/asm/memblock.h @@ -8,5 +8,7 @@ void memblock_x86_to_bootmem(u64 start, u64 end); void memblock_x86_reserve_range(u64 start, u64 end, char *name); void memblock_x86_free_range(u64 start, u64 end); +struct range; +int get_free_all_memory_range(struct range **rangep, int nodeid); #endif diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index 9829eaf1dbd..b4500604ab3 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -86,7 +86,103 @@ u64 __init memblock_x86_find_in_range_size(u64 start, u64 *sizep, u64 align) return MEMBLOCK_ERROR; } -#ifndef CONFIG_NO_BOOTMEM +static __init struct range *find_range_array(int count) +{ + u64 end, size, mem; + struct range *range; + + size = sizeof(struct range) * count; + end = memblock.current_limit; + + mem = memblock_find_in_range(0, end, size, sizeof(struct range)); + if (mem == MEMBLOCK_ERROR) + panic("can not find more space for range array"); + + /* + * This range is tempoaray, so don't reserve it, it will not be + * overlapped because We will not alloccate new buffer before + * We discard this one + */ + range = __va(mem); + memset(range, 0, size); + + return range; +} + +#ifdef CONFIG_NO_BOOTMEM +static void __init memblock_x86_subtract_reserved(struct range *range, int az) +{ + u64 final_start, final_end; + struct memblock_region *r; + + /* Take out region array itself at first*/ + memblock_free_reserved_regions(); + + pr_info("Subtract (%ld early reservations)\n", memblock.reserved.cnt); + + for_each_memblock(reserved, r) { + pr_info(" [%010llx-%010llx]\n", (u64)r->base, (u64)r->base + r->size - 1); + final_start = PFN_DOWN(r->base); + final_end = PFN_UP(r->base + r->size); + if (final_start >= final_end) + continue; + subtract_range(range, az, final_start, final_end); + } + + /* Put region array back ? */ + memblock_reserve_reserved_regions(); +} + +struct count_data { + int nr; +}; + +static int __init count_work_fn(unsigned long start_pfn, + unsigned long end_pfn, void *datax) +{ + struct count_data *data = datax; + + data->nr++; + + return 0; +} + +static int __init count_early_node_map(int nodeid) +{ + struct count_data data; + + data.nr = 0; + work_with_active_regions(nodeid, count_work_fn, &data); + + return data.nr; +} + +int __init get_free_all_memory_range(struct range **rangep, int nodeid) +{ + int count; + struct range *range; + int nr_range; + + count = (memblock.reserved.cnt + count_early_node_map(nodeid)) * 2; + + range = find_range_array(count); + nr_range = 0; + + /* + * Use early_node_map[] and memblock.reserved.region to get range array + * at first + */ + nr_range = add_from_early_node_map(range, count, nr_range, nodeid); +#ifdef CONFIG_X86_32 + subtract_range(range, count, max_low_pfn, -1ULL); +#endif + memblock_x86_subtract_reserved(range, count); + nr_range = clean_sort_range(range, count); + + *rangep = range; + return nr_range; +} +#else void __init memblock_x86_to_bootmem(u64 start, u64 end) { int count; -- cgit v1.2.3-70-g09d2 From 88ba088c18457caaf8d2e5f8d36becc731a3d4f6 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 25 Aug 2010 13:39:16 -0700 Subject: x86, memblock: Add memblock_x86_register_active_regions() and memblock_x86_hole_size() memblock_x86_register_active_regions() will be used to fill early_node_map, the result will be memblock.memory.region AND numa data memblock_x86_hole_size will be used to find hole size on memblock.memory.region with specified range. Signed-off-by: Yinghai Lu Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/memblock.h | 4 +++ arch/x86/mm/memblock.c | 66 +++++++++++++++++++++++++++++++++++++++++ 2 files changed, 70 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/memblock.h b/arch/x86/include/asm/memblock.h index 72639ce65e8..16af28d3607 100644 --- a/arch/x86/include/asm/memblock.h +++ b/arch/x86/include/asm/memblock.h @@ -11,4 +11,8 @@ void memblock_x86_free_range(u64 start, u64 end); struct range; int get_free_all_memory_range(struct range **rangep, int nodeid); +void memblock_x86_register_active_regions(int nid, unsigned long start_pfn, + unsigned long last_pfn); +u64 memblock_x86_hole_size(u64 start, u64 end); + #endif diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index b4500604ab3..53a7a5aebd6 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -232,3 +232,69 @@ void __init memblock_x86_free_range(u64 start, u64 end) memblock_free(start, end - start); } + +/* + * Finds an active region in the address range from start_pfn to last_pfn and + * returns its range in ei_startpfn and ei_endpfn for the memblock entry. + */ +static int __init memblock_x86_find_active_region(const struct memblock_region *ei, + unsigned long start_pfn, + unsigned long last_pfn, + unsigned long *ei_startpfn, + unsigned long *ei_endpfn) +{ + u64 align = PAGE_SIZE; + + *ei_startpfn = round_up(ei->base, align) >> PAGE_SHIFT; + *ei_endpfn = round_down(ei->base + ei->size, align) >> PAGE_SHIFT; + + /* Skip map entries smaller than a page */ + if (*ei_startpfn >= *ei_endpfn) + return 0; + + /* Skip if map is outside the node */ + if (*ei_endpfn <= start_pfn || *ei_startpfn >= last_pfn) + return 0; + + /* Check for overlaps */ + if (*ei_startpfn < start_pfn) + *ei_startpfn = start_pfn; + if (*ei_endpfn > last_pfn) + *ei_endpfn = last_pfn; + + return 1; +} + +/* Walk the memblock.memory map and register active regions within a node */ +void __init memblock_x86_register_active_regions(int nid, unsigned long start_pfn, + unsigned long last_pfn) +{ + unsigned long ei_startpfn; + unsigned long ei_endpfn; + struct memblock_region *r; + + for_each_memblock(memory, r) + if (memblock_x86_find_active_region(r, start_pfn, last_pfn, + &ei_startpfn, &ei_endpfn)) + add_active_range(nid, ei_startpfn, ei_endpfn); +} + +/* + * Find the hole size (in bytes) in the memory range. + * @start: starting address of the memory range to scan + * @end: ending address of the memory range to scan + */ +u64 __init memblock_x86_hole_size(u64 start, u64 end) +{ + unsigned long start_pfn = start >> PAGE_SHIFT; + unsigned long last_pfn = end >> PAGE_SHIFT; + unsigned long ei_startpfn, ei_endpfn, ram = 0; + struct memblock_region *r; + + for_each_memblock(memory, r) + if (memblock_x86_find_active_region(r, start_pfn, last_pfn, + &ei_startpfn, &ei_endpfn)) + ram += ei_endpfn - ei_startpfn; + + return end - start - ((u64)ram << PAGE_SHIFT); +} -- cgit v1.2.3-70-g09d2 From 6bcc8176d07f108da3b1af17fb2c0e82c80e948e Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 25 Aug 2010 13:39:16 -0700 Subject: x86, memblock: Add memblock_x86_find_in_range_node() It can be used to find NODE_DATA for numa. Need to make sure early_node_map[] is filled before it is called, otherwise it will fallback to memblock_find_in_range(), with node range. Signed-off-by: Yinghai Lu Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/memblock.h | 1 + arch/x86/mm/memblock.c | 15 +++++++++++++++ 2 files changed, 16 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/memblock.h b/arch/x86/include/asm/memblock.h index 16af28d3607..3a86b10380f 100644 --- a/arch/x86/include/asm/memblock.h +++ b/arch/x86/include/asm/memblock.h @@ -14,5 +14,6 @@ int get_free_all_memory_range(struct range **rangep, int nodeid); void memblock_x86_register_active_regions(int nid, unsigned long start_pfn, unsigned long last_pfn); u64 memblock_x86_hole_size(u64 start, u64 end); +u64 memblock_x86_find_in_range_node(int nid, u64 start, u64 end, u64 size, u64 align); #endif diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index 53a7a5aebd6..22ff0a39b22 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -233,6 +233,21 @@ void __init memblock_x86_free_range(u64 start, u64 end) memblock_free(start, end - start); } +/* + * Need to call this function after memblock_x86_register_active_regions, + * so early_node_map[] is filled already. + */ +u64 __init memblock_x86_find_in_range_node(int nid, u64 start, u64 end, u64 size, u64 align) +{ + u64 addr; + addr = find_memory_core_early(nid, size, align, start, end); + if (addr != MEMBLOCK_ERROR) + return addr; + + /* Fallback, should already have start end within node range */ + return memblock_find_in_range(start, end, size, align); +} + /* * Finds an active region in the address range from start_pfn to last_pfn and * returns its range in ei_startpfn and ei_endpfn for the memblock entry. -- cgit v1.2.3-70-g09d2 From b52c17ce854125700c4e19d4427d39bf2504ff63 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 25 Aug 2010 13:39:16 -0700 Subject: x86, memblock: Add memblock_x86_free_memory_in_range() It will return free memory size in specified range. We can not use memory_size - reserved_size here, because some reserved area may not be in the scope of memblock.memory.region. Use memblock.memory.region subtracting memblock.reserved.region to get free range array. then count size of all free ranges. -v2: Ben insist on using _in_range Signed-off-by: Yinghai Lu Cc: Benjamin Herrenschmidt Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/memblock.h | 1 + arch/x86/mm/memblock.c | 48 +++++++++++++++++++++++++++++++++++++++++ 2 files changed, 49 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/memblock.h b/arch/x86/include/asm/memblock.h index 3a86b10380f..fc3c230812e 100644 --- a/arch/x86/include/asm/memblock.h +++ b/arch/x86/include/asm/memblock.h @@ -15,5 +15,6 @@ void memblock_x86_register_active_regions(int nid, unsigned long start_pfn, unsigned long last_pfn); u64 memblock_x86_hole_size(u64 start, u64 end); u64 memblock_x86_find_in_range_node(int nid, u64 start, u64 end, u64 size, u64 align); +u64 memblock_x86_free_memory_in_range(u64 addr, u64 limit); #endif diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index 22ff0a39b22..30d60cf29ce 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -211,6 +211,54 @@ void __init memblock_x86_to_bootmem(u64 start, u64 end) } #endif +u64 __init memblock_x86_free_memory_in_range(u64 addr, u64 limit) +{ + int i, count; + struct range *range; + int nr_range; + u64 final_start, final_end; + u64 free_size; + struct memblock_region *r; + + count = (memblock.reserved.cnt + memblock.memory.cnt) * 2; + + range = find_range_array(count); + nr_range = 0; + + addr = PFN_UP(addr); + limit = PFN_DOWN(limit); + + for_each_memblock(memory, r) { + final_start = PFN_UP(r->base); + final_end = PFN_DOWN(r->base + r->size); + if (final_start >= final_end) + continue; + if (final_start >= limit || final_end <= addr) + continue; + + nr_range = add_range(range, count, nr_range, final_start, final_end); + } + subtract_range(range, count, 0, addr); + subtract_range(range, count, limit, -1ULL); + for_each_memblock(reserved, r) { + final_start = PFN_DOWN(r->base); + final_end = PFN_UP(r->base + r->size); + if (final_start >= final_end) + continue; + if (final_start >= limit || final_end <= addr) + continue; + + subtract_range(range, count, final_start, final_end); + } + nr_range = clean_sort_range(range, count); + + free_size = 0; + for (i = 0; i < nr_range; i++) + free_size += range[i].end - range[i].start; + + return free_size << PAGE_SHIFT; +} + void __init memblock_x86_reserve_range(u64 start, u64 end, char *name) { if (start == end) -- cgit v1.2.3-70-g09d2 From e82d42be24bd5d75bf6f81045636e6ca95ab55f2 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 25 Aug 2010 13:39:17 -0700 Subject: x86, memblock: Add memblock_x86_memory_in_range() It will return memory size in specified range according to memblock.memory.region Try to share some code with memblock_x86_free_memory_in_range() by passing get_free to __memblock_x86_memory_in_range(). -v2: Ben want _in_range in the name instead of size Signed-off-by: Yinghai Lu Cc: Benjamin Herrenschmidt Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/memblock.h | 1 + arch/x86/mm/memblock.c | 18 +++++++++++++++++- 2 files changed, 18 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/memblock.h b/arch/x86/include/asm/memblock.h index fc3c230812e..2c304bb6e07 100644 --- a/arch/x86/include/asm/memblock.h +++ b/arch/x86/include/asm/memblock.h @@ -16,5 +16,6 @@ void memblock_x86_register_active_regions(int nid, unsigned long start_pfn, u64 memblock_x86_hole_size(u64 start, u64 end); u64 memblock_x86_find_in_range_node(int nid, u64 start, u64 end, u64 size, u64 align); u64 memblock_x86_free_memory_in_range(u64 addr, u64 limit); +u64 memblock_x86_memory_in_range(u64 addr, u64 limit); #endif diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index 30d60cf29ce..32ddad5dc93 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -211,7 +211,7 @@ void __init memblock_x86_to_bootmem(u64 start, u64 end) } #endif -u64 __init memblock_x86_free_memory_in_range(u64 addr, u64 limit) +static u64 __init __memblock_x86_memory_in_range(u64 addr, u64 limit, bool get_free) { int i, count; struct range *range; @@ -240,6 +240,10 @@ u64 __init memblock_x86_free_memory_in_range(u64 addr, u64 limit) } subtract_range(range, count, 0, addr); subtract_range(range, count, limit, -1ULL); + + /* Subtract memblock.reserved.region in range ? */ + if (!get_free) + goto sort_and_count_them; for_each_memblock(reserved, r) { final_start = PFN_DOWN(r->base); final_end = PFN_UP(r->base + r->size); @@ -250,6 +254,8 @@ u64 __init memblock_x86_free_memory_in_range(u64 addr, u64 limit) subtract_range(range, count, final_start, final_end); } + +sort_and_count_them: nr_range = clean_sort_range(range, count); free_size = 0; @@ -259,6 +265,16 @@ u64 __init memblock_x86_free_memory_in_range(u64 addr, u64 limit) return free_size << PAGE_SHIFT; } +u64 __init memblock_x86_free_memory_in_range(u64 addr, u64 limit) +{ + return __memblock_x86_memory_in_range(addr, limit, true); +} + +u64 __init memblock_x86_memory_in_range(u64 addr, u64 limit) +{ + return __memblock_x86_memory_in_range(addr, limit, false); +} + void __init memblock_x86_reserve_range(u64 start, u64 end, char *name) { if (start == end) -- cgit v1.2.3-70-g09d2 From 301ff3e88ef9ff4bdb92f36a3e6170fce4c9dd34 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 25 Aug 2010 13:39:17 -0700 Subject: x86, memblock: Use memblock_debug to control debug message print out Also let memblock_x86_reserve_range/memblock_x86_free_range could print out name if memblock=debug is specified will also print ther name when reserve_memblock_area/free_memblock_area are called. -v2: according to Ingo, put " if (memblock_debug) " in one place Signed-off-by: Yinghai Lu Signed-off-by: H. Peter Anvin --- arch/x86/mm/memblock.c | 20 ++++++++++++-------- 1 file changed, 12 insertions(+), 8 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index 32ddad5dc93..aaff3932588 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -118,10 +118,10 @@ static void __init memblock_x86_subtract_reserved(struct range *range, int az) /* Take out region array itself at first*/ memblock_free_reserved_regions(); - pr_info("Subtract (%ld early reservations)\n", memblock.reserved.cnt); + memblock_dbg("Subtract (%ld early reservations)\n", memblock.reserved.cnt); for_each_memblock(reserved, r) { - pr_info(" [%010llx-%010llx]\n", (u64)r->base, (u64)r->base + r->size - 1); + memblock_dbg(" [%010llx-%010llx]\n", (u64)r->base, (u64)r->base + r->size - 1); final_start = PFN_DOWN(r->base); final_end = PFN_UP(r->base + r->size); if (final_start >= final_end) @@ -193,16 +193,16 @@ void __init memblock_x86_to_bootmem(u64 start, u64 end) memblock_free_reserved_regions(); count = memblock.reserved.cnt; - pr_info("(%d early reservations) ==> bootmem [%010llx-%010llx]\n", count, start, end - 1); + memblock_dbg("(%d early reservations) ==> bootmem [%#010llx-%#010llx]\n", count, start, end - 1); for_each_memblock(reserved, r) { - pr_info(" [%010llx-%010llx] ", (u64)r->base, (u64)r->base + r->size - 1); + memblock_dbg(" [%#010llx-%#010llx] ", (u64)r->base, (u64)r->base + r->size - 1); final_start = max(start, r->base); final_end = min(end, r->base + r->size); if (final_start >= final_end) { - pr_cont("\n"); + memblock_dbg("\n"); continue; } - pr_cont(" ==> [%010llx-%010llx]\n", final_start, final_end - 1); + memblock_dbg(" ==> [%#010llx-%#010llx]\n", final_start, final_end - 1); reserve_bootmem_generic(final_start, final_end - final_start, BOOTMEM_DEFAULT); } @@ -280,9 +280,11 @@ void __init memblock_x86_reserve_range(u64 start, u64 end, char *name) if (start == end) return; - if (WARN_ONCE(start > end, "memblock_x86_reserve_range: wrong range [%#llx, %#llx]\n", start, end)) + if (WARN_ONCE(start > end, "memblock_x86_reserve_range: wrong range [%#llx, %#llx)\n", start, end)) return; + memblock_dbg(" memblock_x86_reserve_range: [%#010llx-%#010llx] %16s\n", start, end - 1, name); + memblock_reserve(start, end - start); } @@ -291,9 +293,11 @@ void __init memblock_x86_free_range(u64 start, u64 end) if (start == end) return; - if (WARN_ONCE(start > end, "memblock_x86_free_range: wrong range [%#llx, %#llx]\n", start, end)) + if (WARN_ONCE(start > end, "memblock_x86_free_range: wrong range [%#llx, %#llx)\n", start, end)) return; + memblock_dbg(" memblock_x86_free_range: [%#010llx-%#010llx]\n", start, end - 1); + memblock_free(start, end - start); } -- cgit v1.2.3-70-g09d2 From 72d7c3b33c980843e756681fb4867dc1efd62a76 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 25 Aug 2010 13:39:17 -0700 Subject: x86: Use memblock to replace early_res 1. replace find_e820_area with memblock_find_in_range 2. replace reserve_early with memblock_x86_reserve_range 3. replace free_early with memblock_x86_free_range. 4. NO_BOOTMEM will switch to use memblock too. 5. use _e820, _early wrap in the patch, in following patch, will replace them all 6. because memblock_x86_free_range support partial free, we can remove some special care 7. Need to make sure that memblock_find_in_range() is called after memblock_x86_fill() so adjust some calling later in setup.c::setup_arch() -- corruption_check and mptable_update -v2: Move reserve_brk() early Before fill_memblock_area, to avoid overlap between brk and memblock_find_in_range() that could happen We have more then 128 RAM entry in E820 tables, and memblock_x86_fill() could use memblock_find_in_range() to find a new place for memblock.memory.region array. and We don't need to use extend_brk() after fill_memblock_area() So move reserve_brk() early before fill_memblock_area(). -v3: Move find_smp_config early To make sure memblock_find_in_range not find wrong place, if BIOS doesn't put mptable in right place. -v4: Treat RESERVED_KERN as RAM in memblock.memory. and they are already in memblock.reserved already.. use __NOT_KEEP_MEMBLOCK to make sure memblock related code could be freed later. -v5: Generic version __memblock_find_in_range() is going from high to low, and for 32bit active_region for 32bit does include high pages need to replace the limit with memblock.default_alloc_limit, aka get_max_mapped() -v6: Use current_limit instead -v7: check with MEMBLOCK_ERROR instead of -1ULL or -1L -v8: Set memblock_can_resize early to handle EFI with more RAM entries -v9: update after kmemleak changes in mainline Suggested-by: David S. Miller Suggested-by: Benjamin Herrenschmidt Suggested-by: Thomas Gleixner Signed-off-by: Yinghai Lu Signed-off-by: H. Peter Anvin --- arch/x86/Kconfig | 9 +-- arch/x86/include/asm/e820.h | 14 ++-- arch/x86/kernel/check.c | 16 +++-- arch/x86/kernel/e820.c | 159 ++++++++++++++--------------------------- arch/x86/kernel/head.c | 3 +- arch/x86/kernel/head32.c | 6 +- arch/x86/kernel/head64.c | 3 + arch/x86/kernel/mpparse.c | 5 +- arch/x86/kernel/setup.c | 46 ++++++++---- arch/x86/kernel/setup_percpu.c | 6 -- arch/x86/mm/numa_64.c | 9 +-- mm/bootmem.c | 3 + mm/page_alloc.c | 50 ++++--------- mm/sparse-vmemmap.c | 11 --- 14 files changed, 141 insertions(+), 199 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index dcb0593b4a6..542bb2610cb 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -27,6 +27,7 @@ config X86 select HAVE_PERF_EVENTS if (!M386 && !M486) select HAVE_IOREMAP_PROT select HAVE_KPROBES + select HAVE_MEMBLOCK select ARCH_WANT_OPTIONAL_GPIOLIB select ARCH_WANT_FRAME_POINTERS select HAVE_DMA_ATTRS @@ -195,9 +196,6 @@ config ARCH_SUPPORTS_OPTIMIZED_INLINING config ARCH_SUPPORTS_DEBUG_PAGEALLOC def_bool y -config HAVE_EARLY_RES - def_bool y - config HAVE_INTEL_TXT def_bool y depends on EXPERIMENTAL && DMAR && ACPI @@ -590,14 +588,13 @@ config NO_BOOTMEM default y bool "Disable Bootmem code" ---help--- - Use early_res directly instead of bootmem before slab is ready. + Use memblock directly instead of bootmem before slab is ready. - allocator (buddy) [generic] - early allocator (bootmem) [generic] - - very early allocator (reserve_early*()) [x86] + - very early allocator (memblock) [some generic] - very very early allocator (early brk model) [x86] So reduce one layer between early allocator to final allocator - config MEMTEST bool "Memtest" ---help--- diff --git a/arch/x86/include/asm/e820.h b/arch/x86/include/asm/e820.h index ec8a52d14ab..388fed29146 100644 --- a/arch/x86/include/asm/e820.h +++ b/arch/x86/include/asm/e820.h @@ -117,24 +117,26 @@ extern unsigned long end_user_pfn; extern u64 find_e820_area(u64 start, u64 end, u64 size, u64 align); extern u64 find_e820_area_size(u64 start, u64 *sizep, u64 align); extern u64 early_reserve_e820(u64 startt, u64 sizet, u64 align); -#include extern unsigned long e820_end_of_ram_pfn(void); extern unsigned long e820_end_of_low_ram_pfn(void); -extern int e820_find_active_region(const struct e820entry *ei, - unsigned long start_pfn, - unsigned long last_pfn, - unsigned long *ei_startpfn, - unsigned long *ei_endpfn); extern void e820_register_active_regions(int nid, unsigned long start_pfn, unsigned long end_pfn); extern u64 e820_hole_size(u64 start, u64 end); + +extern u64 early_reserve_e820(u64 startt, u64 sizet, u64 align); + +void memblock_x86_fill(void); + extern void finish_e820_parsing(void); extern void e820_reserve_resources(void); extern void e820_reserve_resources_late(void); extern void setup_memory_map(void); extern char *default_machine_specific_memory_setup(void); +void reserve_early(u64 start, u64 end, char *name); +void free_early(u64 start, u64 end); + /* * Returns true iff the specified range [s,e) is completely contained inside * the ISA region. diff --git a/arch/x86/kernel/check.c b/arch/x86/kernel/check.c index fc999e6fc46..13a38917951 100644 --- a/arch/x86/kernel/check.c +++ b/arch/x86/kernel/check.c @@ -2,7 +2,8 @@ #include #include #include -#include +#include + #include /* @@ -18,10 +19,12 @@ static int __read_mostly memory_corruption_check = -1; static unsigned __read_mostly corruption_check_size = 64*1024; static unsigned __read_mostly corruption_check_period = 60; /* seconds */ -static struct e820entry scan_areas[MAX_SCAN_AREAS]; +static struct scan_area { + u64 addr; + u64 size; +} scan_areas[MAX_SCAN_AREAS]; static int num_scan_areas; - static __init int set_corruption_check(char *arg) { char *end; @@ -81,9 +84,9 @@ void __init setup_bios_corruption_check(void) while (addr < corruption_check_size && num_scan_areas < MAX_SCAN_AREAS) { u64 size; - addr = find_e820_area_size(addr, &size, PAGE_SIZE); + addr = memblock_x86_find_in_range_size(addr, &size, PAGE_SIZE); - if (!(addr + 1)) + if (addr == MEMBLOCK_ERROR) break; if (addr >= corruption_check_size) @@ -92,7 +95,7 @@ void __init setup_bios_corruption_check(void) if ((addr + size) > corruption_check_size) size = corruption_check_size - addr; - e820_update_range(addr, size, E820_RAM, E820_RESERVED); + memblock_x86_reserve_range(addr, addr + size, "SCAN RAM"); scan_areas[num_scan_areas].addr = addr; scan_areas[num_scan_areas].size = size; num_scan_areas++; @@ -105,7 +108,6 @@ void __init setup_bios_corruption_check(void) printk(KERN_INFO "Scanning %d areas for low memory corruption\n", num_scan_areas); - update_e820(); } diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index 0d6fc71bedb..a9221d18a5e 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include @@ -742,69 +743,29 @@ core_initcall(e820_mark_nvs_memory); */ u64 __init find_e820_area(u64 start, u64 end, u64 size, u64 align) { - int i; - - for (i = 0; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - u64 addr; - u64 ei_start, ei_last; + u64 mem = memblock_find_in_range(start, end, size, align); - if (ei->type != E820_RAM) - continue; - - ei_last = ei->addr + ei->size; - ei_start = ei->addr; - addr = find_early_area(ei_start, ei_last, start, end, - size, align); - - if (addr != -1ULL) - return addr; - } - return -1ULL; -} + if (mem == MEMBLOCK_ERROR) + return -1ULL; -u64 __init find_fw_memmap_area(u64 start, u64 end, u64 size, u64 align) -{ - return find_e820_area(start, end, size, align); + return mem; } -u64 __init get_max_mapped(void) -{ - u64 end = max_pfn_mapped; - - end <<= PAGE_SHIFT; - - return end; -} /* * Find next free range after *start */ u64 __init find_e820_area_size(u64 start, u64 *sizep, u64 align) { - int i; + u64 mem = memblock_x86_find_in_range_size(start, sizep, align); - for (i = 0; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - u64 addr; - u64 ei_start, ei_last; - - if (ei->type != E820_RAM) - continue; - - ei_last = ei->addr + ei->size; - ei_start = ei->addr; - addr = find_early_area_size(ei_start, ei_last, start, - sizep, align); + if (mem == MEMBLOCK_ERROR) + return -1ULL - if (addr != -1ULL) - return addr; - } - - return -1ULL; + return mem; } /* - * pre allocated 4k and reserved it in e820 + * pre allocated 4k and reserved it in memblock and e820_saved */ u64 __init early_reserve_e820(u64 startt, u64 sizet, u64 align) { @@ -813,8 +774,8 @@ u64 __init early_reserve_e820(u64 startt, u64 sizet, u64 align) u64 start; for (start = startt; ; start += size) { - start = find_e820_area_size(start, &size, align); - if (!(start + 1)) + start = memblock_x86_find_in_range_size(start, &size, align); + if (start == MEMBLOCK_ERROR) return 0; if (size >= sizet) break; @@ -830,10 +791,9 @@ u64 __init early_reserve_e820(u64 startt, u64 sizet, u64 align) addr = round_down(start + size - sizet, align); if (addr < start) return 0; - e820_update_range(addr, sizet, E820_RAM, E820_RESERVED); + memblock_x86_reserve_range(addr, addr + sizet, "new next"); e820_update_range_saved(addr, sizet, E820_RAM, E820_RESERVED); - printk(KERN_INFO "update e820 for early_reserve_e820\n"); - update_e820(); + printk(KERN_INFO "update e820_saved for early_reserve_e820\n"); update_e820_saved(); return addr; @@ -895,52 +855,12 @@ unsigned long __init e820_end_of_low_ram_pfn(void) { return e820_end_pfn(1UL<<(32 - PAGE_SHIFT), E820_RAM); } -/* - * Finds an active region in the address range from start_pfn to last_pfn and - * returns its range in ei_startpfn and ei_endpfn for the e820 entry. - */ -int __init e820_find_active_region(const struct e820entry *ei, - unsigned long start_pfn, - unsigned long last_pfn, - unsigned long *ei_startpfn, - unsigned long *ei_endpfn) -{ - u64 align = PAGE_SIZE; - - *ei_startpfn = round_up(ei->addr, align) >> PAGE_SHIFT; - *ei_endpfn = round_down(ei->addr + ei->size, align) >> PAGE_SHIFT; - - /* Skip map entries smaller than a page */ - if (*ei_startpfn >= *ei_endpfn) - return 0; - - /* Skip if map is outside the node */ - if (ei->type != E820_RAM || *ei_endpfn <= start_pfn || - *ei_startpfn >= last_pfn) - return 0; - - /* Check for overlaps */ - if (*ei_startpfn < start_pfn) - *ei_startpfn = start_pfn; - if (*ei_endpfn > last_pfn) - *ei_endpfn = last_pfn; - - return 1; -} /* Walk the e820 map and register active regions within a node */ void __init e820_register_active_regions(int nid, unsigned long start_pfn, unsigned long last_pfn) { - unsigned long ei_startpfn; - unsigned long ei_endpfn; - int i; - - for (i = 0; i < e820.nr_map; i++) - if (e820_find_active_region(&e820.map[i], - start_pfn, last_pfn, - &ei_startpfn, &ei_endpfn)) - add_active_range(nid, ei_startpfn, ei_endpfn); + memblock_x86_register_active_regions(nid, start_pfn, last_pfn); } /* @@ -950,18 +870,16 @@ void __init e820_register_active_regions(int nid, unsigned long start_pfn, */ u64 __init e820_hole_size(u64 start, u64 end) { - unsigned long start_pfn = start >> PAGE_SHIFT; - unsigned long last_pfn = end >> PAGE_SHIFT; - unsigned long ei_startpfn, ei_endpfn, ram = 0; - int i; + return memblock_x86_hole_size(start, end); +} - for (i = 0; i < e820.nr_map; i++) { - if (e820_find_active_region(&e820.map[i], - start_pfn, last_pfn, - &ei_startpfn, &ei_endpfn)) - ram += ei_endpfn - ei_startpfn; - } - return end - start - ((u64)ram << PAGE_SHIFT); +void reserve_early(u64 start, u64 end, char *name) +{ + memblock_x86_reserve_range(start, end, name); +} +void free_early(u64 start, u64 end) +{ + memblock_x86_free_range(start, end); } static void early_panic(char *msg) @@ -1210,3 +1128,32 @@ void __init setup_memory_map(void) printk(KERN_INFO "BIOS-provided physical RAM map:\n"); e820_print_map(who); } + +void __init memblock_x86_fill(void) +{ + int i; + u64 end; + + /* + * EFI may have more than 128 entries + * We are safe to enable resizing, beause memblock_x86_fill() + * is rather later for x86 + */ + memblock_can_resize = 1; + + for (i = 0; i < e820.nr_map; i++) { + struct e820entry *ei = &e820.map[i]; + + end = ei->addr + ei->size; + if (end != (resource_size_t)end) + continue; + + if (ei->type != E820_RAM && ei->type != E820_RESERVED_KERN) + continue; + + memblock_add(ei->addr, ei->size); + } + + memblock_analyze(); + memblock_dump_all(); +} diff --git a/arch/x86/kernel/head.c b/arch/x86/kernel/head.c index 3e66bd364a9..af0699ba48c 100644 --- a/arch/x86/kernel/head.c +++ b/arch/x86/kernel/head.c @@ -1,5 +1,6 @@ #include #include +#include #include #include @@ -51,5 +52,5 @@ void __init reserve_ebda_region(void) lowmem = 0x9f000; /* reserve all memory between lowmem and the 1MB mark */ - reserve_early_overlap_ok(lowmem, 0x100000, "BIOS reserved"); + memblock_x86_reserve_range(lowmem, 0x100000, "* BIOS reserved"); } diff --git a/arch/x86/kernel/head32.c b/arch/x86/kernel/head32.c index b2e24603739..da60aa8a850 100644 --- a/arch/x86/kernel/head32.c +++ b/arch/x86/kernel/head32.c @@ -8,6 +8,7 @@ #include #include #include +#include #include #include @@ -30,14 +31,15 @@ static void __init i386_default_early_setup(void) void __init i386_start_kernel(void) { + memblock_init(); + #ifdef CONFIG_X86_TRAMPOLINE /* * But first pinch a few for the stack/trampoline stuff * FIXME: Don't need the extra page at 4K, but need to fix * trampoline before removing it. (see the GDT stuff) */ - reserve_early_overlap_ok(PAGE_SIZE, PAGE_SIZE + PAGE_SIZE, - "EX TRAMPOLINE"); + memblock_x86_reserve_range(PAGE_SIZE, PAGE_SIZE + PAGE_SIZE, "EX TRAMPOLINE"); #endif reserve_early(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index 7147143fd61..8ee930fdeeb 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include @@ -98,6 +99,8 @@ void __init x86_64_start_reservations(char *real_mode_data) { copy_bootdata(__va(real_mode_data)); + memblock_init(); + reserve_early(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); #ifdef CONFIG_BLK_DEV_INITRD diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c index d86dbf7e54b..8252545ae6f 100644 --- a/arch/x86/kernel/mpparse.c +++ b/arch/x86/kernel/mpparse.c @@ -11,6 +11,7 @@ #include #include #include +#include #include #include #include @@ -641,7 +642,7 @@ static void __init smp_reserve_memory(struct mpf_intel *mpf) { unsigned long size = get_mpc_size(mpf->physptr); - reserve_early_overlap_ok(mpf->physptr, mpf->physptr+size, "MP-table mpc"); + memblock_x86_reserve_range(mpf->physptr, mpf->physptr+size, "* MP-table mpc"); } static int __init smp_scan_config(unsigned long base, unsigned long length) @@ -670,7 +671,7 @@ static int __init smp_scan_config(unsigned long base, unsigned long length) mpf, (u64)virt_to_phys(mpf)); mem = virt_to_phys(mpf); - reserve_early_overlap_ok(mem, mem + sizeof(*mpf), "MP-table mpf"); + memblock_x86_reserve_range(mem, mem + sizeof(*mpf), "* MP-table mpf"); if (mpf->physptr) smp_reserve_memory(mpf); diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index b4ae4acbd03..bbe0aaf7749 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -31,6 +31,7 @@ #include #include #include +#include #include #include #include @@ -614,7 +615,7 @@ static __init void reserve_ibft_region(void) addr = find_ibft_region(&size); if (size) - reserve_early_overlap_ok(addr, addr + size, "ibft"); + memblock_x86_reserve_range(addr, addr + size, "* ibft"); } #ifdef CONFIG_X86_RESERVE_LOW_64K @@ -708,6 +709,15 @@ static void __init trim_bios_range(void) sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); } +static u64 __init get_max_mapped(void) +{ + u64 end = max_pfn_mapped; + + end <<= PAGE_SHIFT; + + return end; +} + /* * Determine if we were loaded by an EFI loader. If so, then we have also been * passed the efi memmap, systab, etc., so we should use these data structures @@ -891,8 +901,6 @@ void __init setup_arch(char **cmdline_p) */ max_pfn = e820_end_of_ram_pfn(); - /* preallocate 4k for mptable mpc */ - early_reserve_e820_mpc_new(); /* update e820 for memory not covered by WB MTRRs */ mtrr_bp_init(); if (mtrr_trim_uncached_memory(max_pfn)) @@ -917,15 +925,6 @@ void __init setup_arch(char **cmdline_p) max_pfn_mapped = KERNEL_IMAGE_SIZE >> PAGE_SHIFT; #endif -#ifdef CONFIG_X86_CHECK_BIOS_CORRUPTION - setup_bios_corruption_check(); -#endif - - printk(KERN_DEBUG "initial memory mapped : 0 - %08lx\n", - max_pfn_mapped< #include #include +#include #include #include #include @@ -171,8 +172,8 @@ static void * __init early_node_mem(int nodeid, unsigned long start, if (start < (MAX_DMA32_PFN< (MAX_DMA32_PFN< #include #include +#include #include #include @@ -434,6 +435,7 @@ void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, unsigned long size) { #ifdef CONFIG_NO_BOOTMEM + kmemleak_free_part(__va(physaddr), size); free_early(physaddr, physaddr + size); #else unsigned long start, end; @@ -459,6 +461,7 @@ void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, void __init free_bootmem(unsigned long addr, unsigned long size) { #ifdef CONFIG_NO_BOOTMEM + kmemleak_free_part(__va(addr), size); free_early(addr, addr + size); #else unsigned long start, end; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 8c9b34674d8..f2cd7450fa7 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -3667,46 +3667,26 @@ int __init add_from_early_node_map(struct range *range, int az, void * __init __alloc_memory_core_early(int nid, u64 size, u64 align, u64 goal, u64 limit) { - int i; void *ptr; + u64 addr; - if (limit > get_max_mapped()) - limit = get_max_mapped(); + if (limit > memblock.current_limit) + limit = memblock.current_limit; - /* need to go over early_node_map to find out good range for node */ - for_each_active_range_index_in_nid(i, nid) { - u64 addr; - u64 ei_start, ei_last; + addr = find_memory_core_early(nid, size, align, goal, limit); - ei_last = early_node_map[i].end_pfn; - ei_last <<= PAGE_SHIFT; - ei_start = early_node_map[i].start_pfn; - ei_start <<= PAGE_SHIFT; - addr = find_early_area(ei_start, ei_last, - goal, limit, size, align); - - if (addr == -1ULL) - continue; - -#if 0 - printk(KERN_DEBUG "alloc (nid=%d %llx - %llx) (%llx - %llx) %llx %llx => %llx\n", - nid, - ei_start, ei_last, goal, limit, size, - align, addr); -#endif - - ptr = phys_to_virt(addr); - memset(ptr, 0, size); - reserve_early_without_check(addr, addr + size, "BOOTMEM"); - /* - * The min_count is set to 0 so that bootmem allocated blocks - * are never reported as leaks. - */ - kmemleak_alloc(ptr, size, 0, 0); - return ptr; - } + if (addr == MEMBLOCK_ERROR) + return NULL; - return NULL; + ptr = phys_to_virt(addr); + memset(ptr, 0, size); + memblock_x86_reserve_range(addr, addr + size, "BOOTMEM"); + /* + * The min_count is set to 0 so that bootmem allocated blocks + * are never reported as leaks. + */ + kmemleak_alloc(ptr, size, 0, 0); + return ptr; } #endif diff --git a/mm/sparse-vmemmap.c b/mm/sparse-vmemmap.c index aa33fd67fa4..29d6cbffb28 100644 --- a/mm/sparse-vmemmap.c +++ b/mm/sparse-vmemmap.c @@ -220,18 +220,7 @@ void __init sparse_mem_maps_populate_node(struct page **map_map, if (vmemmap_buf_start) { /* need to free left buf */ -#ifdef CONFIG_NO_BOOTMEM - free_early(__pa(vmemmap_buf_start), __pa(vmemmap_buf_end)); - if (vmemmap_buf_start < vmemmap_buf) { - char name[15]; - - snprintf(name, sizeof(name), "MEMMAP %d", nodeid); - reserve_early_without_check(__pa(vmemmap_buf_start), - __pa(vmemmap_buf), name); - } -#else free_bootmem(__pa(vmemmap_buf), vmemmap_buf_end - vmemmap_buf); -#endif vmemmap_buf = NULL; vmemmap_buf_end = NULL; } -- cgit v1.2.3-70-g09d2 From a9ce6bc15100023b411f8117e53a016d61889800 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 25 Aug 2010 13:39:17 -0700 Subject: x86, memblock: Replace e820_/_early string with memblock_ 1.include linux/memblock.h directly. so later could reduce e820.h reference. 2 this patch is done by sed scripts mainly -v2: use MEMBLOCK_ERROR instead of -1ULL or -1UL Signed-off-by: Yinghai Lu Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/efi.h | 2 +- arch/x86/kernel/acpi/sleep.c | 9 +++++---- arch/x86/kernel/apic/numaq_32.c | 3 ++- arch/x86/kernel/efi.c | 5 +++-- arch/x86/kernel/head32.c | 4 ++-- arch/x86/kernel/head64.c | 4 ++-- arch/x86/kernel/setup.c | 29 ++++++++++++++--------------- arch/x86/kernel/trampoline.c | 10 +++++----- arch/x86/mm/init.c | 10 ++++++---- arch/x86/mm/init_32.c | 14 ++++++++------ arch/x86/mm/init_64.c | 11 ++++++----- arch/x86/mm/k8topology_64.c | 4 +++- arch/x86/mm/memtest.c | 7 +++---- arch/x86/mm/numa_32.c | 25 +++++++++++++------------ arch/x86/mm/numa_64.c | 34 +++++++++++++++++----------------- arch/x86/mm/srat_32.c | 3 ++- arch/x86/mm/srat_64.c | 11 ++++++----- arch/x86/xen/mmu.c | 5 +++-- arch/x86/xen/setup.c | 3 ++- mm/bootmem.c | 4 ++-- 20 files changed, 105 insertions(+), 92 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h index 8406ed7f992..8e4a16508d4 100644 --- a/arch/x86/include/asm/efi.h +++ b/arch/x86/include/asm/efi.h @@ -90,7 +90,7 @@ extern void __iomem *efi_ioremap(unsigned long addr, unsigned long size, #endif /* CONFIG_X86_32 */ extern int add_efi_memmap; -extern void efi_reserve_early(void); +extern void efi_memblock_x86_reserve_range(void); extern void efi_call_phys_prelog(void); extern void efi_call_phys_epilog(void); diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c index fcc3c61fdec..d829e75f968 100644 --- a/arch/x86/kernel/acpi/sleep.c +++ b/arch/x86/kernel/acpi/sleep.c @@ -7,6 +7,7 @@ #include #include +#include #include #include #include @@ -125,7 +126,7 @@ void acpi_restore_state_mem(void) */ void __init acpi_reserve_wakeup_memory(void) { - unsigned long mem; + phys_addr_t mem; if ((&wakeup_code_end - &wakeup_code_start) > WAKEUP_SIZE) { printk(KERN_ERR @@ -133,15 +134,15 @@ void __init acpi_reserve_wakeup_memory(void) return; } - mem = find_e820_area(0, 1<<20, WAKEUP_SIZE, PAGE_SIZE); + mem = memblock_find_in_range(0, 1<<20, WAKEUP_SIZE, PAGE_SIZE); - if (mem == -1L) { + if (mem == MEMBLOCK_ERROR) { printk(KERN_ERR "ACPI: Cannot allocate lowmem, S3 disabled.\n"); return; } acpi_realmode = (unsigned long) phys_to_virt(mem); acpi_wakeup_address = mem; - reserve_early(mem, mem + WAKEUP_SIZE, "ACPI WAKEUP"); + memblock_x86_reserve_range(mem, mem + WAKEUP_SIZE, "ACPI WAKEUP"); } diff --git a/arch/x86/kernel/apic/numaq_32.c b/arch/x86/kernel/apic/numaq_32.c index 3e28401f161..960f26ab5c9 100644 --- a/arch/x86/kernel/apic/numaq_32.c +++ b/arch/x86/kernel/apic/numaq_32.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include #include @@ -88,7 +89,7 @@ static inline void numaq_register_node(int node, struct sys_cfg_data *scd) node_end_pfn[node] = MB_TO_PAGES(eq->hi_shrd_mem_start + eq->hi_shrd_mem_size); - e820_register_active_regions(node, node_start_pfn[node], + memblock_x86_register_active_regions(node, node_start_pfn[node], node_end_pfn[node]); memory_present(node, node_start_pfn[node], node_end_pfn[node]); diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c index c2fa9b8b497..0fe27d7c625 100644 --- a/arch/x86/kernel/efi.c +++ b/arch/x86/kernel/efi.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include #include @@ -275,7 +276,7 @@ static void __init do_add_efi_memmap(void) sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); } -void __init efi_reserve_early(void) +void __init efi_memblock_x86_reserve_range(void) { unsigned long pmap; @@ -290,7 +291,7 @@ void __init efi_reserve_early(void) boot_params.efi_info.efi_memdesc_size; memmap.desc_version = boot_params.efi_info.efi_memdesc_version; memmap.desc_size = boot_params.efi_info.efi_memdesc_size; - reserve_early(pmap, pmap + memmap.nr_map * memmap.desc_size, + memblock_x86_reserve_range(pmap, pmap + memmap.nr_map * memmap.desc_size, "EFI memmap"); } diff --git a/arch/x86/kernel/head32.c b/arch/x86/kernel/head32.c index da60aa8a850..74e4cf65043 100644 --- a/arch/x86/kernel/head32.c +++ b/arch/x86/kernel/head32.c @@ -42,7 +42,7 @@ void __init i386_start_kernel(void) memblock_x86_reserve_range(PAGE_SIZE, PAGE_SIZE + PAGE_SIZE, "EX TRAMPOLINE"); #endif - reserve_early(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); + memblock_x86_reserve_range(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); #ifdef CONFIG_BLK_DEV_INITRD /* Reserve INITRD */ @@ -51,7 +51,7 @@ void __init i386_start_kernel(void) u64 ramdisk_image = boot_params.hdr.ramdisk_image; u64 ramdisk_size = boot_params.hdr.ramdisk_size; u64 ramdisk_end = PAGE_ALIGN(ramdisk_image + ramdisk_size); - reserve_early(ramdisk_image, ramdisk_end, "RAMDISK"); + memblock_x86_reserve_range(ramdisk_image, ramdisk_end, "RAMDISK"); } #endif diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index 8ee930fdeeb..97adf9828b9 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -101,7 +101,7 @@ void __init x86_64_start_reservations(char *real_mode_data) memblock_init(); - reserve_early(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); + memblock_x86_reserve_range(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); #ifdef CONFIG_BLK_DEV_INITRD /* Reserve INITRD */ @@ -110,7 +110,7 @@ void __init x86_64_start_reservations(char *real_mode_data) unsigned long ramdisk_image = boot_params.hdr.ramdisk_image; unsigned long ramdisk_size = boot_params.hdr.ramdisk_size; unsigned long ramdisk_end = PAGE_ALIGN(ramdisk_image + ramdisk_size); - reserve_early(ramdisk_image, ramdisk_end, "RAMDISK"); + memblock_x86_reserve_range(ramdisk_image, ramdisk_end, "RAMDISK"); } #endif diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index bbe0aaf7749..a4f01733e87 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -302,7 +302,7 @@ static inline void init_gbpages(void) static void __init reserve_brk(void) { if (_brk_end > _brk_start) - reserve_early(__pa(_brk_start), __pa(_brk_end), "BRK"); + memblock_x86_reserve_range(__pa(_brk_start), __pa(_brk_end), "BRK"); /* Mark brk area as locked down and no longer taking any new allocations */ @@ -324,17 +324,16 @@ static void __init relocate_initrd(void) char *p, *q; /* We need to move the initrd down into lowmem */ - ramdisk_here = find_e820_area(0, end_of_lowmem, area_size, + ramdisk_here = memblock_find_in_range(0, end_of_lowmem, area_size, PAGE_SIZE); - if (ramdisk_here == -1ULL) + if (ramdisk_here == MEMBLOCK_ERROR) panic("Cannot find place for new RAMDISK of size %lld\n", ramdisk_size); /* Note: this includes all the lowmem currently occupied by the initrd, we rely on that fact to keep the data intact. */ - reserve_early(ramdisk_here, ramdisk_here + area_size, - "NEW RAMDISK"); + memblock_x86_reserve_range(ramdisk_here, ramdisk_here + area_size, "NEW RAMDISK"); initrd_start = ramdisk_here + PAGE_OFFSET; initrd_end = initrd_start + ramdisk_size; printk(KERN_INFO "Allocated new RAMDISK: %08llx - %08llx\n", @@ -390,7 +389,7 @@ static void __init reserve_initrd(void) initrd_start = 0; if (ramdisk_size >= (end_of_lowmem>>1)) { - free_early(ramdisk_image, ramdisk_end); + memblock_x86_free_range(ramdisk_image, ramdisk_end); printk(KERN_ERR "initrd too large to handle, " "disabling initrd\n"); return; @@ -413,7 +412,7 @@ static void __init reserve_initrd(void) relocate_initrd(); - free_early(ramdisk_image, ramdisk_end); + memblock_x86_free_range(ramdisk_image, ramdisk_end); } #else static void __init reserve_initrd(void) @@ -469,7 +468,7 @@ static void __init e820_reserve_setup_data(void) e820_print_map("reserve setup_data"); } -static void __init reserve_early_setup_data(void) +static void __init memblock_x86_reserve_range_setup_data(void) { struct setup_data *data; u64 pa_data; @@ -481,7 +480,7 @@ static void __init reserve_early_setup_data(void) while (pa_data) { data = early_memremap(pa_data, sizeof(*data)); sprintf(buf, "setup data %x", data->type); - reserve_early(pa_data, pa_data+sizeof(*data)+data->len, buf); + memblock_x86_reserve_range(pa_data, pa_data+sizeof(*data)+data->len, buf); pa_data = data->next; early_iounmap(data, sizeof(*data)); } @@ -519,23 +518,23 @@ static void __init reserve_crashkernel(void) if (crash_base <= 0) { const unsigned long long alignment = 16<<20; /* 16M */ - crash_base = find_e820_area(alignment, ULONG_MAX, crash_size, + crash_base = memblock_find_in_range(alignment, ULONG_MAX, crash_size, alignment); - if (crash_base == -1ULL) { + if (crash_base == MEMBLOCK_ERROR) { pr_info("crashkernel reservation failed - No suitable area found.\n"); return; } } else { unsigned long long start; - start = find_e820_area(crash_base, ULONG_MAX, crash_size, + start = memblock_find_in_range(crash_base, ULONG_MAX, crash_size, 1<<20); if (start != crash_base) { pr_info("crashkernel reservation failed - memory is in use.\n"); return; } } - reserve_early(crash_base, crash_base + crash_size, "CRASH KERNEL"); + memblock_x86_reserve_range(crash_base, crash_base + crash_size, "CRASH KERNEL"); printk(KERN_INFO "Reserving %ldMB of memory at %ldMB " "for crashkernel (System RAM: %ldMB)\n", @@ -786,7 +785,7 @@ void __init setup_arch(char **cmdline_p) #endif 4)) { efi_enabled = 1; - efi_reserve_early(); + efi_memblock_x86_reserve_range(); } #endif @@ -846,7 +845,7 @@ void __init setup_arch(char **cmdline_p) vmi_activate(); /* after early param, so could get panic from serial */ - reserve_early_setup_data(); + memblock_x86_reserve_range_setup_data(); if (acpi_mps_check()) { #ifdef CONFIG_X86_LOCAL_APIC diff --git a/arch/x86/kernel/trampoline.c b/arch/x86/kernel/trampoline.c index c652ef62742..7c2102c2aad 100644 --- a/arch/x86/kernel/trampoline.c +++ b/arch/x86/kernel/trampoline.c @@ -1,7 +1,7 @@ #include +#include #include -#include #if defined(CONFIG_X86_64) && defined(CONFIG_ACPI_SLEEP) #define __trampinit @@ -16,15 +16,15 @@ unsigned char *__trampinitdata trampoline_base; void __init reserve_trampoline_memory(void) { - unsigned long mem; + phys_addr_t mem; /* Has to be in very low memory so we can execute real-mode AP code. */ - mem = find_e820_area(0, 1<<20, TRAMPOLINE_SIZE, PAGE_SIZE); - if (mem == -1L) + mem = memblock_find_in_range(0, 1<<20, TRAMPOLINE_SIZE, PAGE_SIZE); + if (mem == MEMBLOCK_ERROR) panic("Cannot allocate trampoline\n"); trampoline_base = __va(mem); - reserve_early(mem, mem + TRAMPOLINE_SIZE, "TRAMPOLINE"); + memblock_x86_reserve_range(mem, mem + TRAMPOLINE_SIZE, "TRAMPOLINE"); } /* diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c index b278535b14a..c0e28a13de7 100644 --- a/arch/x86/mm/init.c +++ b/arch/x86/mm/init.c @@ -2,6 +2,7 @@ #include #include #include +#include #include #include @@ -33,6 +34,7 @@ static void __init find_early_table_space(unsigned long end, int use_pse, int use_gbpages) { unsigned long puds, pmds, ptes, tables, start; + phys_addr_t base; puds = (end + PUD_SIZE - 1) >> PUD_SHIFT; tables = roundup(puds * sizeof(pud_t), PAGE_SIZE); @@ -75,12 +77,12 @@ static void __init find_early_table_space(unsigned long end, int use_pse, #else start = 0x8000; #endif - e820_table_start = find_e820_area(start, max_pfn_mapped<>= PAGE_SHIFT; + e820_table_start = base >> PAGE_SHIFT; e820_table_end = e820_table_start; e820_table_top = e820_table_start + (tables >> PAGE_SHIFT); @@ -299,7 +301,7 @@ unsigned long __init_refok init_memory_mapping(unsigned long start, __flush_tlb_all(); if (!after_bootmem && e820_table_end > e820_table_start) - reserve_early(e820_table_start << PAGE_SHIFT, + memblock_x86_reserve_range(e820_table_start << PAGE_SHIFT, e820_table_end << PAGE_SHIFT, "PGTABLE"); if (!after_bootmem) diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index 90e054589aa..63b09bae250 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -25,6 +25,7 @@ #include #include #include +#include #include #include #include @@ -712,14 +713,14 @@ void __init initmem_init(unsigned long start_pfn, unsigned long end_pfn, highstart_pfn = highend_pfn = max_pfn; if (max_pfn > max_low_pfn) highstart_pfn = max_low_pfn; - e820_register_active_regions(0, 0, highend_pfn); + memblock_x86_register_active_regions(0, 0, highend_pfn); sparse_memory_present_with_active_regions(0); printk(KERN_NOTICE "%ldMB HIGHMEM available.\n", pages_to_mb(highend_pfn - highstart_pfn)); num_physpages = highend_pfn; high_memory = (void *) __va(highstart_pfn * PAGE_SIZE - 1) + 1; #else - e820_register_active_regions(0, 0, max_low_pfn); + memblock_x86_register_active_regions(0, 0, max_low_pfn); sparse_memory_present_with_active_regions(0); num_physpages = max_low_pfn; high_memory = (void *) __va(max_low_pfn * PAGE_SIZE - 1) + 1; @@ -776,16 +777,16 @@ void __init setup_bootmem_allocator(void) { #ifndef CONFIG_NO_BOOTMEM int nodeid; - unsigned long bootmap_size, bootmap; + phys_addr_t bootmap_size, bootmap; /* * Initialize the boot-time allocator (with low memory only): */ bootmap_size = bootmem_bootmap_pages(max_low_pfn)< #include #include +#include #include #include #include @@ -577,18 +578,18 @@ void __init initmem_init(unsigned long start_pfn, unsigned long end_pfn, unsigned long bootmap_size, bootmap; bootmap_size = bootmem_bootmap_pages(end_pfn)<> PAGE_SHIFT, 0, end_pfn); - e820_register_active_regions(0, start_pfn, end_pfn); + memblock_x86_register_active_regions(0, start_pfn, end_pfn); free_bootmem_with_active_regions(0, end_pfn); #else - e820_register_active_regions(0, start_pfn, end_pfn); + memblock_x86_register_active_regions(0, start_pfn, end_pfn); #endif } #endif diff --git a/arch/x86/mm/k8topology_64.c b/arch/x86/mm/k8topology_64.c index 970ed579d4e..966de9372e8 100644 --- a/arch/x86/mm/k8topology_64.c +++ b/arch/x86/mm/k8topology_64.c @@ -11,6 +11,8 @@ #include #include #include +#include + #include #include #include @@ -222,7 +224,7 @@ int __init k8_scan_nodes(void) for_each_node_mask(i, node_possible_map) { int j; - e820_register_active_regions(i, + memblock_x86_register_active_regions(i, nodes[i].start >> PAGE_SHIFT, nodes[i].end >> PAGE_SHIFT); for (j = apicid_base; j < cores + apicid_base; j++) diff --git a/arch/x86/mm/memtest.c b/arch/x86/mm/memtest.c index 18d244f7020..92faf3a1c53 100644 --- a/arch/x86/mm/memtest.c +++ b/arch/x86/mm/memtest.c @@ -6,8 +6,7 @@ #include #include #include - -#include +#include static u64 patterns[] __initdata = { 0, @@ -35,7 +34,7 @@ static void __init reserve_bad_mem(u64 pattern, u64 start_bad, u64 end_bad) (unsigned long long) pattern, (unsigned long long) start_bad, (unsigned long long) end_bad); - reserve_early(start_bad, end_bad, "BAD RAM"); + memblock_x86_reserve_range(start_bad, end_bad, "BAD RAM"); } static void __init memtest(u64 pattern, u64 start_phys, u64 size) @@ -74,7 +73,7 @@ static void __init do_one_pass(u64 pattern, u64 start, u64 end) u64 size = 0; while (start < end) { - start = find_e820_area_size(start, &size, 1); + start = memblock_x86_find_in_range_size(start, &size, 1); /* done ? */ if (start >= end) diff --git a/arch/x86/mm/numa_32.c b/arch/x86/mm/numa_32.c index 809baaaf48b..ddf9730b206 100644 --- a/arch/x86/mm/numa_32.c +++ b/arch/x86/mm/numa_32.c @@ -24,6 +24,7 @@ #include #include +#include #include #include #include @@ -120,7 +121,7 @@ int __init get_memcfg_numa_flat(void) node_start_pfn[0] = 0; node_end_pfn[0] = max_pfn; - e820_register_active_regions(0, 0, max_pfn); + memblock_x86_register_active_regions(0, 0, max_pfn); memory_present(0, 0, max_pfn); node_remap_size[0] = node_memmap_size_bytes(0, 0, max_pfn); @@ -161,14 +162,14 @@ static void __init allocate_pgdat(int nid) NODE_DATA(nid) = (pg_data_t *)node_remap_start_vaddr[nid]; else { unsigned long pgdat_phys; - pgdat_phys = find_e820_area(min_low_pfn<>PAGE_SHIFT)); memset(buf, 0, sizeof(buf)); sprintf(buf, "NODE_DATA %d", nid); - reserve_early(pgdat_phys, pgdat_phys + sizeof(pg_data_t), buf); + memblock_x86_reserve_range(pgdat_phys, pgdat_phys + sizeof(pg_data_t), buf); } printk(KERN_DEBUG "allocate_pgdat: node %d NODE_DATA %08lx\n", nid, (unsigned long)NODE_DATA(nid)); @@ -291,15 +292,15 @@ static __init unsigned long calculate_numa_remap_pages(void) PTRS_PER_PTE); node_kva_target <<= PAGE_SHIFT; do { - node_kva_final = find_e820_area(node_kva_target, + node_kva_final = memblock_find_in_range(node_kva_target, ((u64)node_end_pfn[nid])<>PAGE_SHIFT) > (node_start_pfn[nid])); - if (node_kva_final == -1ULL) + if (node_kva_final == MEMBLOCK_ERROR) panic("Can not get kva ram\n"); node_remap_size[nid] = size; @@ -318,9 +319,9 @@ static __init unsigned long calculate_numa_remap_pages(void) * but we could have some hole in high memory, and it will only * check page_is_ram(pfn) && !page_is_reserved_early(pfn) to decide * to use it as free. - * So reserve_early here, hope we don't run out of that array + * So memblock_x86_reserve_range here, hope we don't run out of that array */ - reserve_early(node_kva_final, + memblock_x86_reserve_range(node_kva_final, node_kva_final+(((u64)size)<> PAGE_SHIFT; kva_target_pfn -= PTRS_PER_PTE; - } while (kva_start_pfn == -1UL && kva_target_pfn > min_low_pfn); + } while (kva_start_pfn == MEMBLOCK_ERROR && kva_target_pfn > min_low_pfn); - if (kva_start_pfn == -1UL) + if (kva_start_pfn == MEMBLOCK_ERROR) panic("Can not get kva space\n"); printk(KERN_INFO "kva_start_pfn ~ %lx max_low_pfn ~ %lx\n", @@ -382,7 +383,7 @@ void __init initmem_init(unsigned long start_pfn, unsigned long end_pfn, printk(KERN_INFO "max_pfn = %lx\n", max_pfn); /* avoid clash with initrd */ - reserve_early(kva_start_pfn< physnodes[i].end) { end = physnodes[i].end; @@ -467,7 +467,7 @@ static int __init split_nodes_interleave(u64 addr, u64 max_addr, * this one must extend to the boundary. */ if (end < dma32_end && dma32_end - end - - e820_hole_size(end, dma32_end) < FAKE_NODE_MIN_SIZE) + memblock_x86_hole_size(end, dma32_end) < FAKE_NODE_MIN_SIZE) end = dma32_end; /* @@ -476,7 +476,7 @@ static int __init split_nodes_interleave(u64 addr, u64 max_addr, * physical node. */ if (physnodes[i].end - end - - e820_hole_size(end, physnodes[i].end) < size) + memblock_x86_hole_size(end, physnodes[i].end) < size) end = physnodes[i].end; /* @@ -504,7 +504,7 @@ static u64 __init find_end_of_node(u64 start, u64 max_addr, u64 size) { u64 end = start + size; - while (end - start - e820_hole_size(start, end) < size) { + while (end - start - memblock_x86_hole_size(start, end) < size) { end += FAKE_NODE_MIN_SIZE; if (end > max_addr) { end = max_addr; @@ -533,7 +533,7 @@ static int __init split_nodes_size_interleave(u64 addr, u64 max_addr, u64 size) * creates a uniform distribution of node sizes across the entire * machine (but not necessarily over physical nodes). */ - min_size = (max_addr - addr - e820_hole_size(addr, max_addr)) / + min_size = (max_addr - addr - memblock_x86_hole_size(addr, max_addr)) / MAX_NUMNODES; min_size = max(min_size, FAKE_NODE_MIN_SIZE); if ((min_size & FAKE_NODE_MIN_HASH_MASK) < min_size) @@ -566,7 +566,7 @@ static int __init split_nodes_size_interleave(u64 addr, u64 max_addr, u64 size) * this one must extend to the boundary. */ if (end < dma32_end && dma32_end - end - - e820_hole_size(end, dma32_end) < FAKE_NODE_MIN_SIZE) + memblock_x86_hole_size(end, dma32_end) < FAKE_NODE_MIN_SIZE) end = dma32_end; /* @@ -575,7 +575,7 @@ static int __init split_nodes_size_interleave(u64 addr, u64 max_addr, u64 size) * physical node. */ if (physnodes[i].end - end - - e820_hole_size(end, physnodes[i].end) < size) + memblock_x86_hole_size(end, physnodes[i].end) < size) end = physnodes[i].end; /* @@ -639,7 +639,7 @@ static int __init numa_emulation(unsigned long start_pfn, */ remove_all_active_ranges(); for_each_node_mask(i, node_possible_map) { - e820_register_active_regions(i, nodes[i].start >> PAGE_SHIFT, + memblock_x86_register_active_regions(i, nodes[i].start >> PAGE_SHIFT, nodes[i].end >> PAGE_SHIFT); setup_node_bootmem(i, nodes[i].start, nodes[i].end); } @@ -692,7 +692,7 @@ void __init initmem_init(unsigned long start_pfn, unsigned long last_pfn, node_set(0, node_possible_map); for (i = 0; i < nr_cpu_ids; i++) numa_set_node(i, 0); - e820_register_active_regions(0, start_pfn, last_pfn); + memblock_x86_register_active_regions(0, start_pfn, last_pfn); setup_node_bootmem(0, start_pfn << PAGE_SHIFT, last_pfn << PAGE_SHIFT); } diff --git a/arch/x86/mm/srat_32.c b/arch/x86/mm/srat_32.c index 9324f13492d..a17dffd136c 100644 --- a/arch/x86/mm/srat_32.c +++ b/arch/x86/mm/srat_32.c @@ -25,6 +25,7 @@ */ #include #include +#include #include #include #include @@ -264,7 +265,7 @@ int __init get_memcfg_from_srat(void) if (node_read_chunk(chunk->nid, chunk)) continue; - e820_register_active_regions(chunk->nid, chunk->start_pfn, + memblock_x86_register_active_regions(chunk->nid, chunk->start_pfn, min(chunk->end_pfn, max_pfn)); } /* for out of order entries in SRAT */ diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat_64.c index f9897f7a9ef..7f44eb62a5e 100644 --- a/arch/x86/mm/srat_64.c +++ b/arch/x86/mm/srat_64.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include @@ -98,15 +99,15 @@ void __init acpi_numa_slit_init(struct acpi_table_slit *slit) unsigned long phys; length = slit->header.length; - phys = find_e820_area(0, max_pfn_mapped< x2APIC mapping */ @@ -324,7 +325,7 @@ static int __init nodes_cover_memory(const struct bootnode *nodes) pxmram = 0; } - e820ram = max_pfn - (e820_hole_size(0, max_pfn<>PAGE_SHIFT); + e820ram = max_pfn - (memblock_x86_hole_size(0, max_pfn<>PAGE_SHIFT); /* We seem to lose 3 pages somewhere. Allow 1M of slack. */ if ((long)(e820ram - pxmram) >= (1<<(20 - PAGE_SHIFT))) { printk(KERN_ERR @@ -421,7 +422,7 @@ int __init acpi_scan_nodes(unsigned long start, unsigned long end) } for_each_node_mask(i, nodes_parsed) - e820_register_active_regions(i, nodes[i].start >> PAGE_SHIFT, + memblock_x86_register_active_regions(i, nodes[i].start >> PAGE_SHIFT, nodes[i].end >> PAGE_SHIFT); /* for out of order entries in SRAT */ sort_node_map(); diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index 914f04695ce..b511f198691 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -44,6 +44,7 @@ #include #include #include +#include #include #include @@ -1735,7 +1736,7 @@ __init pgd_t *xen_setup_kernel_pagetable(pgd_t *pgd, __xen_write_cr3(true, __pa(pgd)); xen_mc_issue(PARAVIRT_LAZY_CPU); - reserve_early(__pa(xen_start_info->pt_base), + memblock_x86_reserve_range(__pa(xen_start_info->pt_base), __pa(xen_start_info->pt_base + xen_start_info->nr_pt_frames * PAGE_SIZE), "XEN PAGETABLES"); @@ -1773,7 +1774,7 @@ __init pgd_t *xen_setup_kernel_pagetable(pgd_t *pgd, pin_pagetable_pfn(MMUEXT_PIN_L3_TABLE, PFN_DOWN(__pa(swapper_pg_dir))); - reserve_early(__pa(xen_start_info->pt_base), + memblock_x86_reserve_range(__pa(xen_start_info->pt_base), __pa(xen_start_info->pt_base + xen_start_info->nr_pt_frames * PAGE_SIZE), "XEN PAGETABLES"); diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c index ad0047f47cd..2ac8f29f89c 100644 --- a/arch/x86/xen/setup.c +++ b/arch/x86/xen/setup.c @@ -8,6 +8,7 @@ #include #include #include +#include #include #include @@ -61,7 +62,7 @@ char * __init xen_memory_setup(void) * - xen_start_info * See comment above "struct start_info" in */ - reserve_early(__pa(xen_start_info->mfn_list), + memblock_x86_reserve_range(__pa(xen_start_info->mfn_list), __pa(xen_start_info->pt_base), "XEN START INFO"); diff --git a/mm/bootmem.c b/mm/bootmem.c index fda01a2c31a..13b0caa9793 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -436,7 +436,7 @@ void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, { #ifdef CONFIG_NO_BOOTMEM kmemleak_free_part(__va(physaddr), size); - free_early(physaddr, physaddr + size); + memblock_x86_free_range(physaddr, physaddr + size); #else unsigned long start, end; @@ -462,7 +462,7 @@ void __init free_bootmem(unsigned long addr, unsigned long size) { #ifdef CONFIG_NO_BOOTMEM kmemleak_free_part(__va(addr), size); - free_early(addr, addr + size); + memblock_x86_free_range(addr, addr + size); #else unsigned long start, end; -- cgit v1.2.3-70-g09d2 From a587d2daebcd2bc159d4348b6a7b028950a6d803 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 25 Aug 2010 13:39:18 -0700 Subject: x86: Remove not used early_res code and some functions in e820.c that are not used anymore Signed-off-by: Yinghai Lu Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/e820.h | 14 -- arch/x86/kernel/e820.c | 52 ---- include/linux/early_res.h | 23 -- kernel/Makefile | 1 - kernel/early_res.c | 590 -------------------------------------------- 5 files changed, 680 deletions(-) delete mode 100644 include/linux/early_res.h delete mode 100644 kernel/early_res.c (limited to 'arch/x86') diff --git a/arch/x86/include/asm/e820.h b/arch/x86/include/asm/e820.h index 388fed29146..718646384e0 100644 --- a/arch/x86/include/asm/e820.h +++ b/arch/x86/include/asm/e820.h @@ -112,31 +112,17 @@ static inline void early_memtest(unsigned long start, unsigned long end) } #endif -extern unsigned long end_user_pfn; - -extern u64 find_e820_area(u64 start, u64 end, u64 size, u64 align); -extern u64 find_e820_area_size(u64 start, u64 *sizep, u64 align); -extern u64 early_reserve_e820(u64 startt, u64 sizet, u64 align); - extern unsigned long e820_end_of_ram_pfn(void); extern unsigned long e820_end_of_low_ram_pfn(void); -extern void e820_register_active_regions(int nid, unsigned long start_pfn, - unsigned long end_pfn); -extern u64 e820_hole_size(u64 start, u64 end); - extern u64 early_reserve_e820(u64 startt, u64 sizet, u64 align); void memblock_x86_fill(void); - extern void finish_e820_parsing(void); extern void e820_reserve_resources(void); extern void e820_reserve_resources_late(void); extern void setup_memory_map(void); extern char *default_machine_specific_memory_setup(void); -void reserve_early(u64 start, u64 end, char *name); -void free_early(u64 start, u64 end); - /* * Returns true iff the specified range [s,e) is completely contained inside * the ISA region. diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index a9221d18a5e..d5fd89462d7 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -738,32 +738,6 @@ static int __init e820_mark_nvs_memory(void) core_initcall(e820_mark_nvs_memory); #endif -/* - * Find a free area with specified alignment in a specific range. - */ -u64 __init find_e820_area(u64 start, u64 end, u64 size, u64 align) -{ - u64 mem = memblock_find_in_range(start, end, size, align); - - if (mem == MEMBLOCK_ERROR) - return -1ULL; - - return mem; -} - -/* - * Find next free range after *start - */ -u64 __init find_e820_area_size(u64 start, u64 *sizep, u64 align) -{ - u64 mem = memblock_x86_find_in_range_size(start, sizep, align); - - if (mem == MEMBLOCK_ERROR) - return -1ULL - - return mem; -} - /* * pre allocated 4k and reserved it in memblock and e820_saved */ @@ -856,32 +830,6 @@ unsigned long __init e820_end_of_low_ram_pfn(void) return e820_end_pfn(1UL<<(32 - PAGE_SHIFT), E820_RAM); } -/* Walk the e820 map and register active regions within a node */ -void __init e820_register_active_regions(int nid, unsigned long start_pfn, - unsigned long last_pfn) -{ - memblock_x86_register_active_regions(nid, start_pfn, last_pfn); -} - -/* - * Find the hole size (in bytes) in the memory range. - * @start: starting address of the memory range to scan - * @end: ending address of the memory range to scan - */ -u64 __init e820_hole_size(u64 start, u64 end) -{ - return memblock_x86_hole_size(start, end); -} - -void reserve_early(u64 start, u64 end, char *name) -{ - memblock_x86_reserve_range(start, end, name); -} -void free_early(u64 start, u64 end) -{ - memblock_x86_free_range(start, end); -} - static void early_panic(char *msg) { early_printk(msg); diff --git a/include/linux/early_res.h b/include/linux/early_res.h deleted file mode 100644 index 29c09f57a13..00000000000 --- a/include/linux/early_res.h +++ /dev/null @@ -1,23 +0,0 @@ -#ifndef _LINUX_EARLY_RES_H -#define _LINUX_EARLY_RES_H -#ifdef __KERNEL__ - -extern void reserve_early(u64 start, u64 end, char *name); -extern void reserve_early_overlap_ok(u64 start, u64 end, char *name); -extern void free_early(u64 start, u64 end); -void free_early_partial(u64 start, u64 end); -extern void early_res_to_bootmem(u64 start, u64 end); - -void reserve_early_without_check(u64 start, u64 end, char *name); -u64 find_early_area(u64 ei_start, u64 ei_last, u64 start, u64 end, - u64 size, u64 align); -u64 find_early_area_size(u64 ei_start, u64 ei_last, u64 start, - u64 *sizep, u64 align); -u64 find_fw_memmap_area(u64 start, u64 end, u64 size, u64 align); -u64 get_max_mapped(void); -#include -int get_free_all_memory_range(struct range **rangep, int nodeid); - -#endif /* __KERNEL__ */ - -#endif /* _LINUX_EARLY_RES_H */ diff --git a/kernel/Makefile b/kernel/Makefile index 057472fbc27..80e61c3e44a 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -11,7 +11,6 @@ obj-y = sched.o fork.o exec_domain.o panic.o printk.o \ hrtimer.o rwsem.o nsproxy.o srcu.o semaphore.o \ notifier.o ksysfs.o pm_qos_params.o sched_clock.o cred.o \ async.o range.o -obj-$(CONFIG_HAVE_EARLY_RES) += early_res.o obj-y += groups.o ifdef CONFIG_FUNCTION_TRACER diff --git a/kernel/early_res.c b/kernel/early_res.c deleted file mode 100644 index 7bfae887f21..00000000000 --- a/kernel/early_res.c +++ /dev/null @@ -1,590 +0,0 @@ -/* - * early_res, could be used to replace bootmem - */ -#include -#include -#include -#include -#include -#include -#include -#include - -/* - * Early reserved memory areas. - */ -/* - * need to make sure this one is bigger enough before - * find_fw_memmap_area could be used - */ -#define MAX_EARLY_RES_X 32 - -struct early_res { - u64 start, end; - char name[15]; - char overlap_ok; -}; -static struct early_res early_res_x[MAX_EARLY_RES_X] __initdata; - -static int max_early_res __initdata = MAX_EARLY_RES_X; -static struct early_res *early_res __initdata = &early_res_x[0]; -static int early_res_count __initdata; - -static int __init find_overlapped_early(u64 start, u64 end) -{ - int i; - struct early_res *r; - - for (i = 0; i < max_early_res && early_res[i].end; i++) { - r = &early_res[i]; - if (end > r->start && start < r->end) - break; - } - - return i; -} - -/* - * Drop the i-th range from the early reservation map, - * by copying any higher ranges down one over it, and - * clearing what had been the last slot. - */ -static void __init drop_range(int i) -{ - int j; - - for (j = i + 1; j < max_early_res && early_res[j].end; j++) - ; - - memmove(&early_res[i], &early_res[i + 1], - (j - 1 - i) * sizeof(struct early_res)); - - early_res[j - 1].end = 0; - early_res_count--; -} - -static void __init drop_range_partial(int i, u64 start, u64 end) -{ - u64 common_start, common_end; - u64 old_start, old_end; - - old_start = early_res[i].start; - old_end = early_res[i].end; - common_start = max(old_start, start); - common_end = min(old_end, end); - - /* no overlap ? */ - if (common_start >= common_end) - return; - - if (old_start < common_start) { - /* make head segment */ - early_res[i].end = common_start; - if (old_end > common_end) { - char name[15]; - - /* - * Save a local copy of the name, since the - * early_res array could get resized inside - * reserve_early_without_check() -> - * __check_and_double_early_res(), which would - * make the current name pointer invalid. - */ - strncpy(name, early_res[i].name, - sizeof(early_res[i].name) - 1); - /* add another for left over on tail */ - reserve_early_without_check(common_end, old_end, name); - } - return; - } else { - if (old_end > common_end) { - /* reuse the entry for tail left */ - early_res[i].start = common_end; - return; - } - /* all covered */ - drop_range(i); - } -} - -/* - * Split any existing ranges that: - * 1) are marked 'overlap_ok', and - * 2) overlap with the stated range [start, end) - * into whatever portion (if any) of the existing range is entirely - * below or entirely above the stated range. Drop the portion - * of the existing range that overlaps with the stated range, - * which will allow the caller of this routine to then add that - * stated range without conflicting with any existing range. - */ -static void __init drop_overlaps_that_are_ok(u64 start, u64 end) -{ - int i; - struct early_res *r; - u64 lower_start, lower_end; - u64 upper_start, upper_end; - char name[15]; - - for (i = 0; i < max_early_res && early_res[i].end; i++) { - r = &early_res[i]; - - /* Continue past non-overlapping ranges */ - if (end <= r->start || start >= r->end) - continue; - - /* - * Leave non-ok overlaps as is; let caller - * panic "Overlapping early reservations" - * when it hits this overlap. - */ - if (!r->overlap_ok) - return; - - /* - * We have an ok overlap. We will drop it from the early - * reservation map, and add back in any non-overlapping - * portions (lower or upper) as separate, overlap_ok, - * non-overlapping ranges. - */ - - /* 1. Note any non-overlapping (lower or upper) ranges. */ - strncpy(name, r->name, sizeof(name) - 1); - - lower_start = lower_end = 0; - upper_start = upper_end = 0; - if (r->start < start) { - lower_start = r->start; - lower_end = start; - } - if (r->end > end) { - upper_start = end; - upper_end = r->end; - } - - /* 2. Drop the original ok overlapping range */ - drop_range(i); - - i--; /* resume for-loop on copied down entry */ - - /* 3. Add back in any non-overlapping ranges. */ - if (lower_end) - reserve_early_overlap_ok(lower_start, lower_end, name); - if (upper_end) - reserve_early_overlap_ok(upper_start, upper_end, name); - } -} - -static void __init __reserve_early(u64 start, u64 end, char *name, - int overlap_ok) -{ - int i; - struct early_res *r; - - i = find_overlapped_early(start, end); - if (i >= max_early_res) - panic("Too many early reservations"); - r = &early_res[i]; - if (r->end) - panic("Overlapping early reservations " - "%llx-%llx %s to %llx-%llx %s\n", - start, end - 1, name ? name : "", r->start, - r->end - 1, r->name); - r->start = start; - r->end = end; - r->overlap_ok = overlap_ok; - if (name) - strncpy(r->name, name, sizeof(r->name) - 1); - early_res_count++; -} - -/* - * A few early reservtations come here. - * - * The 'overlap_ok' in the name of this routine does -not- mean it - * is ok for these reservations to overlap an earlier reservation. - * Rather it means that it is ok for subsequent reservations to - * overlap this one. - * - * Use this entry point to reserve early ranges when you are doing - * so out of "Paranoia", reserving perhaps more memory than you need, - * just in case, and don't mind a subsequent overlapping reservation - * that is known to be needed. - * - * The drop_overlaps_that_are_ok() call here isn't really needed. - * It would be needed if we had two colliding 'overlap_ok' - * reservations, so that the second such would not panic on the - * overlap with the first. We don't have any such as of this - * writing, but might as well tolerate such if it happens in - * the future. - */ -void __init reserve_early_overlap_ok(u64 start, u64 end, char *name) -{ - drop_overlaps_that_are_ok(start, end); - __reserve_early(start, end, name, 1); -} - -static void __init __check_and_double_early_res(u64 ex_start, u64 ex_end) -{ - u64 start, end, size, mem; - struct early_res *new; - - /* do we have enough slots left ? */ - if ((max_early_res - early_res_count) > max(max_early_res/8, 2)) - return; - - /* double it */ - mem = -1ULL; - size = sizeof(struct early_res) * max_early_res * 2; - if (early_res == early_res_x) - start = 0; - else - start = early_res[0].end; - end = ex_start; - if (start + size < end) - mem = find_fw_memmap_area(start, end, size, - sizeof(struct early_res)); - if (mem == -1ULL) { - start = ex_end; - end = get_max_mapped(); - if (start + size < end) - mem = find_fw_memmap_area(start, end, size, - sizeof(struct early_res)); - } - if (mem == -1ULL) - panic("can not find more space for early_res array"); - - new = __va(mem); - /* save the first one for own */ - new[0].start = mem; - new[0].end = mem + size; - new[0].overlap_ok = 0; - /* copy old to new */ - if (early_res == early_res_x) { - memcpy(&new[1], &early_res[0], - sizeof(struct early_res) * max_early_res); - memset(&new[max_early_res+1], 0, - sizeof(struct early_res) * (max_early_res - 1)); - early_res_count++; - } else { - memcpy(&new[1], &early_res[1], - sizeof(struct early_res) * (max_early_res - 1)); - memset(&new[max_early_res], 0, - sizeof(struct early_res) * max_early_res); - } - memset(&early_res[0], 0, sizeof(struct early_res) * max_early_res); - early_res = new; - max_early_res *= 2; - printk(KERN_DEBUG "early_res array is doubled to %d at [%llx - %llx]\n", - max_early_res, mem, mem + size - 1); -} - -/* - * Most early reservations come here. - * - * We first have drop_overlaps_that_are_ok() drop any pre-existing - * 'overlap_ok' ranges, so that we can then reserve this memory - * range without risk of panic'ing on an overlapping overlap_ok - * early reservation. - */ -void __init reserve_early(u64 start, u64 end, char *name) -{ - if (start >= end) - return; - - __check_and_double_early_res(start, end); - - drop_overlaps_that_are_ok(start, end); - __reserve_early(start, end, name, 0); -} - -void __init reserve_early_without_check(u64 start, u64 end, char *name) -{ - struct early_res *r; - - if (start >= end) - return; - - __check_and_double_early_res(start, end); - - r = &early_res[early_res_count]; - - r->start = start; - r->end = end; - r->overlap_ok = 0; - if (name) - strncpy(r->name, name, sizeof(r->name) - 1); - early_res_count++; -} - -void __init free_early(u64 start, u64 end) -{ - struct early_res *r; - int i; - - kmemleak_free_part(__va(start), end - start); - - i = find_overlapped_early(start, end); - r = &early_res[i]; - if (i >= max_early_res || r->end != end || r->start != start) - panic("free_early on not reserved area: %llx-%llx!", - start, end - 1); - - drop_range(i); -} - -void __init free_early_partial(u64 start, u64 end) -{ - struct early_res *r; - int i; - - kmemleak_free_part(__va(start), end - start); - - if (start == end) - return; - - if (WARN_ONCE(start > end, " wrong range [%#llx, %#llx]\n", start, end)) - return; - -try_next: - i = find_overlapped_early(start, end); - if (i >= max_early_res) - return; - - r = &early_res[i]; - /* hole ? */ - if (r->end >= end && r->start <= start) { - drop_range_partial(i, start, end); - return; - } - - drop_range_partial(i, start, end); - goto try_next; -} - -#ifdef CONFIG_NO_BOOTMEM -static void __init subtract_early_res(struct range *range, int az) -{ - int i, count; - u64 final_start, final_end; - int idx = 0; - - count = 0; - for (i = 0; i < max_early_res && early_res[i].end; i++) - count++; - - /* need to skip first one ?*/ - if (early_res != early_res_x) - idx = 1; - -#define DEBUG_PRINT_EARLY_RES 1 - -#if DEBUG_PRINT_EARLY_RES - printk(KERN_INFO "Subtract (%d early reservations)\n", count); -#endif - for (i = idx; i < count; i++) { - struct early_res *r = &early_res[i]; -#if DEBUG_PRINT_EARLY_RES - printk(KERN_INFO " #%d [%010llx - %010llx] %15s\n", i, - r->start, r->end, r->name); -#endif - final_start = PFN_DOWN(r->start); - final_end = PFN_UP(r->end); - if (final_start >= final_end) - continue; - subtract_range(range, az, final_start, final_end); - } - -} - -int __init get_free_all_memory_range(struct range **rangep, int nodeid) -{ - int i, count; - u64 start = 0, end; - u64 size; - u64 mem; - struct range *range; - int nr_range; - - count = 0; - for (i = 0; i < max_early_res && early_res[i].end; i++) - count++; - - count *= 2; - - size = sizeof(struct range) * count; - end = get_max_mapped(); -#ifdef MAX_DMA32_PFN - if (end > (MAX_DMA32_PFN << PAGE_SHIFT)) - start = MAX_DMA32_PFN << PAGE_SHIFT; -#endif - mem = find_fw_memmap_area(start, end, size, sizeof(struct range)); - if (mem == -1ULL) - panic("can not find more space for range free"); - - range = __va(mem); - /* use early_node_map[] and early_res to get range array at first */ - memset(range, 0, size); - nr_range = 0; - - /* need to go over early_node_map to find out good range for node */ - nr_range = add_from_early_node_map(range, count, nr_range, nodeid); -#ifdef CONFIG_X86_32 - subtract_range(range, count, max_low_pfn, -1ULL); -#endif - subtract_early_res(range, count); - nr_range = clean_sort_range(range, count); - - /* need to clear it ? */ - if (nodeid == MAX_NUMNODES) { - memset(&early_res[0], 0, - sizeof(struct early_res) * max_early_res); - early_res = NULL; - max_early_res = 0; - } - - *rangep = range; - return nr_range; -} -#else -void __init early_res_to_bootmem(u64 start, u64 end) -{ - int i, count; - u64 final_start, final_end; - int idx = 0; - - count = 0; - for (i = 0; i < max_early_res && early_res[i].end; i++) - count++; - - /* need to skip first one ?*/ - if (early_res != early_res_x) - idx = 1; - - printk(KERN_INFO "(%d/%d early reservations) ==> bootmem [%010llx - %010llx]\n", - count - idx, max_early_res, start, end); - for (i = idx; i < count; i++) { - struct early_res *r = &early_res[i]; - printk(KERN_INFO " #%d [%010llx - %010llx] %16s", i, - r->start, r->end, r->name); - final_start = max(start, r->start); - final_end = min(end, r->end); - if (final_start >= final_end) { - printk(KERN_CONT "\n"); - continue; - } - printk(KERN_CONT " ==> [%010llx - %010llx]\n", - final_start, final_end); - reserve_bootmem_generic(final_start, final_end - final_start, - BOOTMEM_DEFAULT); - } - /* clear them */ - memset(&early_res[0], 0, sizeof(struct early_res) * max_early_res); - early_res = NULL; - max_early_res = 0; - early_res_count = 0; -} -#endif - -/* Check for already reserved areas */ -static inline int __init bad_addr(u64 *addrp, u64 size, u64 align) -{ - int i; - u64 addr = *addrp; - int changed = 0; - struct early_res *r; -again: - i = find_overlapped_early(addr, addr + size); - r = &early_res[i]; - if (i < max_early_res && r->end) { - *addrp = addr = round_up(r->end, align); - changed = 1; - goto again; - } - return changed; -} - -/* Check for already reserved areas */ -static inline int __init bad_addr_size(u64 *addrp, u64 *sizep, u64 align) -{ - int i; - u64 addr = *addrp, last; - u64 size = *sizep; - int changed = 0; -again: - last = addr + size; - for (i = 0; i < max_early_res && early_res[i].end; i++) { - struct early_res *r = &early_res[i]; - if (last > r->start && addr < r->start) { - size = r->start - addr; - changed = 1; - goto again; - } - if (last > r->end && addr < r->end) { - addr = round_up(r->end, align); - size = last - addr; - changed = 1; - goto again; - } - if (last <= r->end && addr >= r->start) { - (*sizep)++; - return 0; - } - } - if (changed) { - *addrp = addr; - *sizep = size; - } - return changed; -} - -/* - * Find a free area with specified alignment in a specific range. - * only with the area.between start to end is active range from early_node_map - * so they are good as RAM - */ -u64 __init find_early_area(u64 ei_start, u64 ei_last, u64 start, u64 end, - u64 size, u64 align) -{ - u64 addr, last; - - addr = round_up(ei_start, align); - if (addr < start) - addr = round_up(start, align); - if (addr >= ei_last) - goto out; - while (bad_addr(&addr, size, align) && addr+size <= ei_last) - ; - last = addr + size; - if (last > ei_last) - goto out; - if (last > end) - goto out; - - return addr; - -out: - return -1ULL; -} - -u64 __init find_early_area_size(u64 ei_start, u64 ei_last, u64 start, - u64 *sizep, u64 align) -{ - u64 addr, last; - - addr = round_up(ei_start, align); - if (addr < start) - addr = round_up(start, align); - if (addr >= ei_last) - goto out; - *sizep = ei_last - addr; - while (bad_addr_size(&addr, sizep, align) && addr + *sizep <= ei_last) - ; - last = addr + *sizep; - if (last > ei_last) - goto out; - - return addr; - -out: - return -1ULL; -} -- cgit v1.2.3-70-g09d2 From 6f2a75369e7561e800d86927ecd83c970996b21f Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 25 Aug 2010 13:39:18 -0700 Subject: x86, memblock: Use memblock_memory_size()/memblock_free_memory_size() to get correct dma_reserve memblock_memory_size() will return memory size in memblock.memory.region. memblock_free_memory_size() will return free memory size in memblock.memory.region. So We can get exact reseved size in specified range. Set the size right after initmem_init(), because later bootmem API will get area above 16M. (except some fallback). Later after we remove the bootmem, We could call that just before paging_init(). Signed-off-by: Yinghai Lu Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/e820.h | 2 ++ arch/x86/kernel/e820.c | 16 ++++++++++++++++ arch/x86/kernel/setup.c | 1 + arch/x86/mm/init_64.c | 7 ------- 4 files changed, 19 insertions(+), 7 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/e820.h b/arch/x86/include/asm/e820.h index 718646384e0..5be1542fbfa 100644 --- a/arch/x86/include/asm/e820.h +++ b/arch/x86/include/asm/e820.h @@ -117,6 +117,8 @@ extern unsigned long e820_end_of_low_ram_pfn(void); extern u64 early_reserve_e820(u64 startt, u64 sizet, u64 align); void memblock_x86_fill(void); +void memblock_find_dma_reserve(void); + extern void finish_e820_parsing(void); extern void e820_reserve_resources(void); extern void e820_reserve_resources_late(void); diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index d5fd89462d7..0c2b7ef7a34 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -1105,3 +1105,19 @@ void __init memblock_x86_fill(void) memblock_analyze(); memblock_dump_all(); } + +void __init memblock_find_dma_reserve(void) +{ +#ifdef CONFIG_X86_64 + u64 free_size_pfn; + u64 mem_size_pfn; + /* + * need to find out used area below MAX_DMA_PFN + * need to use memblock to get free size in [0, MAX_DMA_PFN] + * at first, and assume boot_mem will not take below MAX_DMA_PFN + */ + mem_size_pfn = memblock_x86_memory_in_range(0, MAX_DMA_PFN << PAGE_SHIFT) >> PAGE_SHIFT; + free_size_pfn = memblock_x86_free_memory_in_range(0, MAX_DMA_PFN << PAGE_SHIFT) >> PAGE_SHIFT; + set_dma_reserve(mem_size_pfn - free_size_pfn); +#endif +} diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index a4f01733e87..924c8f78e98 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -1013,6 +1013,7 @@ void __init setup_arch(char **cmdline_p) #endif initmem_init(0, max_pfn, acpi, k8); + memblock_find_dma_reserve(); #ifndef CONFIG_NO_BOOTMEM memblock_x86_to_bootmem(0, max_low_pfn< #include -static unsigned long dma_reserve __initdata; - static int __init parse_direct_gbpages_off(char *arg) { direct_gbpages = 0; @@ -821,11 +819,6 @@ int __init reserve_bootmem_generic(unsigned long phys, unsigned long len, reserve_bootmem(phys, len, flags); - if (phys+len <= MAX_DMA_PFN*PAGE_SIZE) { - dma_reserve += len / PAGE_SIZE; - set_dma_reserve(dma_reserve); - } - return 0; } #endif -- cgit v1.2.3-70-g09d2 From 774ea0bcb27f57b6fd521b3b6c43237782fed4b9 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 25 Aug 2010 13:39:18 -0700 Subject: x86: Remove old bootmem code Requested by Ingo, Thomas and HPA. The old bootmem code is no longer necessary, and the transition is complete. Remove it. Signed-off-by: Yinghai Lu Signed-off-by: H. Peter Anvin --- arch/x86/Kconfig | 10 +-------- arch/x86/kernel/setup.c | 4 ---- arch/x86/mm/init_32.c | 56 ------------------------------------------------- arch/x86/mm/init_64.c | 41 ------------------------------------ arch/x86/mm/memblock.c | 29 ------------------------- arch/x86/mm/numa_32.c | 3 --- arch/x86/mm/numa_64.c | 47 ----------------------------------------- 7 files changed, 1 insertion(+), 189 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 542bb2610cb..ce07615f1cd 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -585,15 +585,7 @@ config PARAVIRT_DEBUG a paravirt_op is missing when it is called. config NO_BOOTMEM - default y - bool "Disable Bootmem code" - ---help--- - Use memblock directly instead of bootmem before slab is ready. - - allocator (buddy) [generic] - - early allocator (bootmem) [generic] - - very early allocator (memblock) [some generic] - - very very early allocator (early brk model) [x86] - So reduce one layer between early allocator to final allocator + def_bool y config MEMTEST bool "Memtest" diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 924c8f78e98..1d114ff6a07 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -1014,10 +1014,6 @@ void __init setup_arch(char **cmdline_p) initmem_init(0, max_pfn, acpi, k8); memblock_find_dma_reserve(); -#ifndef CONFIG_NO_BOOTMEM - memblock_x86_to_bootmem(0, max_low_pfn<> PAGE_SHIFT, - start_pfn, end_pfn); - printk(KERN_INFO " node %d low ram: %08lx - %08lx\n", - nodeid, start_pfn< max_low_pfn) - continue; - if (end_pfn > max_low_pfn) - end_pfn = max_low_pfn; -#else - start_pfn = 0; - end_pfn = max_low_pfn; -#endif - bootmap = setup_node_bootmem(nodeid, start_pfn, end_pfn, - bootmap); - } -#endif - after_bootmem = 1; } diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index d6d408467c4..690b8d13971 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -572,23 +572,7 @@ kernel_physical_mapping_init(unsigned long start, void __init initmem_init(unsigned long start_pfn, unsigned long end_pfn, int acpi, int k8) { -#ifndef CONFIG_NO_BOOTMEM - unsigned long bootmap_size, bootmap; - - bootmap_size = bootmem_bootmap_pages(end_pfn)<> PAGE_SHIFT, - 0, end_pfn); memblock_x86_register_active_regions(0, start_pfn, end_pfn); - free_bootmem_with_active_regions(0, end_pfn); -#else - memblock_x86_register_active_regions(0, start_pfn, end_pfn); -#endif } #endif @@ -798,31 +782,6 @@ void mark_rodata_ro(void) #endif -#ifndef CONFIG_NO_BOOTMEM -int __init reserve_bootmem_generic(unsigned long phys, unsigned long len, - int flags) -{ - unsigned long pfn = phys >> PAGE_SHIFT; - - if (pfn >= max_pfn) { - /* - * This can happen with kdump kernels when accessing - * firmware tables: - */ - if (pfn < max_pfn_mapped) - return -EFAULT; - - printk(KERN_ERR "reserve_bootmem: illegal reserve %lx %lu\n", - phys, len); - return -EFAULT; - } - - reserve_bootmem(phys, len, flags); - - return 0; -} -#endif - int kern_addr_valid(unsigned long addr) { unsigned long above = ((long)addr) >> __VIRTUAL_MASK_SHIFT; diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index aaff3932588..50ecbc59757 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -109,7 +109,6 @@ static __init struct range *find_range_array(int count) return range; } -#ifdef CONFIG_NO_BOOTMEM static void __init memblock_x86_subtract_reserved(struct range *range, int az) { u64 final_start, final_end; @@ -182,34 +181,6 @@ int __init get_free_all_memory_range(struct range **rangep, int nodeid) *rangep = range; return nr_range; } -#else -void __init memblock_x86_to_bootmem(u64 start, u64 end) -{ - int count; - u64 final_start, final_end; - struct memblock_region *r; - - /* Take out region array itself */ - memblock_free_reserved_regions(); - - count = memblock.reserved.cnt; - memblock_dbg("(%d early reservations) ==> bootmem [%#010llx-%#010llx]\n", count, start, end - 1); - for_each_memblock(reserved, r) { - memblock_dbg(" [%#010llx-%#010llx] ", (u64)r->base, (u64)r->base + r->size - 1); - final_start = max(start, r->base); - final_end = min(end, r->base + r->size); - if (final_start >= final_end) { - memblock_dbg("\n"); - continue; - } - memblock_dbg(" ==> [%#010llx-%#010llx]\n", final_start, final_end - 1); - reserve_bootmem_generic(final_start, final_end - final_start, BOOTMEM_DEFAULT); - } - - /* Put region array back ? */ - memblock_reserve_reserved_regions(); -} -#endif static u64 __init __memblock_x86_memory_in_range(u64 addr, u64 limit, bool get_free) { diff --git a/arch/x86/mm/numa_32.c b/arch/x86/mm/numa_32.c index ddf9730b206..70ddeb75ba2 100644 --- a/arch/x86/mm/numa_32.c +++ b/arch/x86/mm/numa_32.c @@ -420,9 +420,6 @@ void __init initmem_init(unsigned long start_pfn, unsigned long end_pfn, for_each_online_node(nid) { memset(NODE_DATA(nid), 0, sizeof(struct pglist_data)); NODE_DATA(nid)->node_id = nid; -#ifndef CONFIG_NO_BOOTMEM - NODE_DATA(nid)->bdata = &bootmem_node_data[nid]; -#endif } setup_bootmem_allocator(); diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c index 984b1ff7db4..aef0ff74f7d 100644 --- a/arch/x86/mm/numa_64.c +++ b/arch/x86/mm/numa_64.c @@ -199,10 +199,6 @@ setup_node_bootmem(int nodeid, unsigned long start, unsigned long end) unsigned long start_pfn, last_pfn, nodedata_phys; const int pgdat_size = roundup(sizeof(pg_data_t), PAGE_SIZE); int nid; -#ifndef CONFIG_NO_BOOTMEM - unsigned long bootmap_start, bootmap_pages, bootmap_size; - void *bootmap; -#endif if (!end) return; @@ -239,47 +235,6 @@ setup_node_bootmem(int nodeid, unsigned long start, unsigned long end) NODE_DATA(nodeid)->node_start_pfn = start_pfn; NODE_DATA(nodeid)->node_spanned_pages = last_pfn - start_pfn; -#ifndef CONFIG_NO_BOOTMEM - NODE_DATA(nodeid)->bdata = &bootmem_node_data[nodeid]; - - /* - * Find a place for the bootmem map - * nodedata_phys could be on other nodes by alloc_bootmem, - * so need to sure bootmap_start not to be small, otherwise - * early_node_mem will get that with memblock_find_in_range instead - * of alloc_bootmem, that could clash with reserved range - */ - bootmap_pages = bootmem_bootmap_pages(last_pfn - start_pfn); - bootmap_start = roundup(nodedata_phys + pgdat_size, PAGE_SIZE); - /* - * SMP_CACHE_BYTES could be enough, but init_bootmem_node like - * to use that to align to PAGE_SIZE - */ - bootmap = early_node_mem(nodeid, bootmap_start, end, - bootmap_pages<> PAGE_SHIFT, - start_pfn, last_pfn); - - printk(KERN_INFO " bootmap [%016lx - %016lx] pages %lx\n", - bootmap_start, bootmap_start + bootmap_size - 1, - bootmap_pages); - nid = phys_to_nid(bootmap_start); - if (nid != nodeid) - printk(KERN_INFO " bootmap(%d) on node %d\n", nodeid, nid); - - free_bootmem_with_active_regions(nodeid, end); -#endif - node_set_online(nodeid); } @@ -704,9 +659,7 @@ unsigned long __init numa_free_all_bootmem(void) for_each_online_node(i) pages += free_all_bootmem_node(NODE_DATA(i)); -#ifdef CONFIG_NO_BOOTMEM pages += free_all_memory_core_early(MAX_NUMNODES); -#endif return pages; } -- cgit v1.2.3-70-g09d2 From 6f44d0337cc54a46e83b4c8a6195607e78fff71d Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Fri, 27 Aug 2010 14:19:33 -0400 Subject: x86, doc: Adding comments about .iommu_table and its neighbors. Updating the linker section with comments about .iommu_table and some other ones that I know of. CC: Sam Ravnborg CC: H. Peter Anvin CC: Fujita Tomonori CC: Thomas Gleixner CC: Ingo Molnar Signed-off-by: Konrad Rzeszutek Wilk LKML-Reference: <1282933173-19960-1-git-send-email-konrad.wilk@oracle.com> Signed-off-by: H. Peter Anvin --- arch/x86/kernel/vmlinux.lds.S | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S index b92e040466c..3f07c370f76 100644 --- a/arch/x86/kernel/vmlinux.lds.S +++ b/arch/x86/kernel/vmlinux.lds.S @@ -242,6 +242,12 @@ SECTIONS __x86_cpu_dev_end = .; } + /* + * start address and size of operations which during runtime + * can be patched with virtualization friendly instructions or + * baremetal native ones. Think page table operations. + * Details in paravirt_types.h + */ . = ALIGN(8); .parainstructions : AT(ADDR(.parainstructions) - LOAD_OFFSET) { __parainstructions = .; @@ -249,6 +255,11 @@ SECTIONS __parainstructions_end = .; } + /* + * struct alt_inst entries. From the header (alternative.h): + * "Alternative instructions for different CPU types or capabilities" + * Think locking instructions on spinlocks. + */ . = ALIGN(8); .altinstructions : AT(ADDR(.altinstructions) - LOAD_OFFSET) { __alt_instructions = .; @@ -256,10 +267,21 @@ SECTIONS __alt_instructions_end = .; } + /* + * And here are the replacement instructions. The linker sticks + * them as binary blobs. The .altinstructions has enough data to + * get the address and the length of them to patch the kernel safely. + */ .altinstr_replacement : AT(ADDR(.altinstr_replacement) - LOAD_OFFSET) { *(.altinstr_replacement) } + /* + * struct iommu_table_entry entries are injected in this section. + * It is an array of IOMMUs which during run time gets sorted depending + * on its dependency order. After rootfs_initcall is complete + * this section can be safely removed. + */ .iommu_table : AT(ADDR(.iommu_table) - LOAD_OFFSET) { __iommu_table = .; *(.iommu_table) -- cgit v1.2.3-70-g09d2 From 9fbaf49c7f717740002d49eee1bbd03d89d8766a Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Sat, 28 Aug 2010 17:41:03 +0200 Subject: x86, kmemcheck: Remove double test The opcodes 0x2e and 0x3e are tested for in the first Group 2 line as well. The sematic match that finds this problem is as follows: (http://coccinelle.lip6.fr/) // @expression@ expression E; @@ ( * E || ... || E | * E && ... && E ) // Signed-off-by: Julia Lawall Reviewed-by: Pekka Enberg Cc: Vegard Nossum LKML-Reference: <1283010066-20935-5-git-send-email-julia@diku.dk> Signed-off-by: Ingo Molnar --- arch/x86/mm/kmemcheck/opcode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/kmemcheck/opcode.c b/arch/x86/mm/kmemcheck/opcode.c index 63c19e27aa6..324aa3f0723 100644 --- a/arch/x86/mm/kmemcheck/opcode.c +++ b/arch/x86/mm/kmemcheck/opcode.c @@ -9,7 +9,7 @@ static bool opcode_is_prefix(uint8_t b) b == 0xf0 || b == 0xf2 || b == 0xf3 /* Group 2 */ || b == 0x2e || b == 0x36 || b == 0x3e || b == 0x26 - || b == 0x64 || b == 0x65 || b == 0x2e || b == 0x3e + || b == 0x64 || b == 0x65 /* Group 3 */ || b == 0x66 /* Group 4 */ -- cgit v1.2.3-70-g09d2 From 7ac41ccf47d82569d26f34beab1dec92cc3b6347 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Mon, 30 Aug 2010 14:10:02 -0400 Subject: x86, iommu: Fix IOMMU_INIT alignment rules This boot crash was observed: DMA-API: preallocated 32768 debug entries DMA-API: debugging enabled by kernel config BUG: unable to handle kernel paging request at 19da8955 IP: [] 0xf4ffffff *pde = 00000000 The crux of the failure was that even if we did not use any of the .iommu_table section, the linker would still insert it in the vmlinux file. This patch fixes that and also fixes the runtime crash where we would try to access the array. Reported-by: Ingo Molnar Signed-off-by: Konrad Rzeszutek Wilk Cc: Joerg Roedel Cc: FUJITA Tomonori LKML-Reference: <1283191802-25086-1-git-send-email-konrad.wilk@oracle.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/vmlinux.lds.S | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S index 3f07c370f76..38e2b67807e 100644 --- a/arch/x86/kernel/vmlinux.lds.S +++ b/arch/x86/kernel/vmlinux.lds.S @@ -285,10 +285,9 @@ SECTIONS .iommu_table : AT(ADDR(.iommu_table) - LOAD_OFFSET) { __iommu_table = .; *(.iommu_table) - . = ALIGN(8); __iommu_table_end = .; } - + . = ALIGN(8); /* * .exit.text is discard at runtime, not link time, to deal with * references from .altinstructions and .eh_frame -- cgit v1.2.3-70-g09d2 From c9cf4a019cff198ee5638323e3b0ee18886467e8 Mon Sep 17 00:00:00 2001 From: Cyrill Gorcunov Date: Wed, 25 Aug 2010 22:23:34 +0400 Subject: perf, x86, Pentium4: Add RAW events verification Implements verification of - Bits of ESCR EventMask field (meaningful bits in field are hardware predefined and others bits should be set to zero) - INSTR_COMPLETED event (it is available on predefined cpu model only) - Thread shared events (they should be guarded by "perf_event_paranoid" sysctl due to security reason). The side effect of this action is that PERF_COUNT_HW_BUS_CYCLES become a "paranoid" general event. Signed-off-by: Cyrill Gorcunov Tested-by: Lin Ming Cc: Frederic Weisbecker Cc: Peter Zijlstra LKML-Reference: <20100825182334.GB14874@lenovo> Signed-off-by: Ingo Molnar --- arch/x86/include/asm/perf_event_p4.h | 52 +++---- arch/x86/kernel/cpu/perf_event_p4.c | 292 ++++++++++++++++++++++++++++++++--- 2 files changed, 290 insertions(+), 54 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/perf_event_p4.h b/arch/x86/include/asm/perf_event_p4.h index def500776b1..a70cd216be5 100644 --- a/arch/x86/include/asm/perf_event_p4.h +++ b/arch/x86/include/asm/perf_event_p4.h @@ -36,19 +36,6 @@ #define P4_ESCR_EMASK(v) ((v) << P4_ESCR_EVENTMASK_SHIFT) #define P4_ESCR_TAG(v) ((v) << P4_ESCR_TAG_SHIFT) -/* Non HT mask */ -#define P4_ESCR_MASK \ - (P4_ESCR_EVENT_MASK | \ - P4_ESCR_EVENTMASK_MASK | \ - P4_ESCR_TAG_MASK | \ - P4_ESCR_TAG_ENABLE | \ - P4_ESCR_T0_OS | \ - P4_ESCR_T0_USR) - -/* HT mask */ -#define P4_ESCR_MASK_HT \ - (P4_ESCR_MASK | P4_ESCR_T1_OS | P4_ESCR_T1_USR) - #define P4_CCCR_OVF 0x80000000U #define P4_CCCR_CASCADE 0x40000000U #define P4_CCCR_OVF_PMI_T0 0x04000000U @@ -70,23 +57,6 @@ #define P4_CCCR_THRESHOLD(v) ((v) << P4_CCCR_THRESHOLD_SHIFT) #define P4_CCCR_ESEL(v) ((v) << P4_CCCR_ESCR_SELECT_SHIFT) -/* Non HT mask */ -#define P4_CCCR_MASK \ - (P4_CCCR_OVF | \ - P4_CCCR_CASCADE | \ - P4_CCCR_OVF_PMI_T0 | \ - P4_CCCR_FORCE_OVF | \ - P4_CCCR_EDGE | \ - P4_CCCR_THRESHOLD_MASK | \ - P4_CCCR_COMPLEMENT | \ - P4_CCCR_COMPARE | \ - P4_CCCR_ESCR_SELECT_MASK | \ - P4_CCCR_ENABLE) - -/* HT mask */ -#define P4_CCCR_MASK_HT \ - (P4_CCCR_MASK | P4_CCCR_OVF_PMI_T1 | P4_CCCR_THREAD_ANY) - #define P4_GEN_ESCR_EMASK(class, name, bit) \ class##__##name = ((1 << bit) << P4_ESCR_EVENTMASK_SHIFT) #define P4_ESCR_EMASK_BIT(class, name) class##__##name @@ -127,6 +97,28 @@ #define P4_CONFIG_HT_SHIFT 63 #define P4_CONFIG_HT (1ULL << P4_CONFIG_HT_SHIFT) +/* + * The bits we allow to pass for RAW events + */ +#define P4_CONFIG_MASK_ESCR \ + P4_ESCR_EVENT_MASK | \ + P4_ESCR_EVENTMASK_MASK | \ + P4_ESCR_TAG_MASK | \ + P4_ESCR_TAG_ENABLE + +#define P4_CONFIG_MASK_CCCR \ + P4_CCCR_EDGE | \ + P4_CCCR_THRESHOLD_MASK | \ + P4_CCCR_COMPLEMENT | \ + P4_CCCR_COMPARE | \ + P4_CCCR_THREAD_ANY | \ + P4_CCCR_RESERVED + +/* some dangerous bits are reserved for kernel internals */ +#define P4_CONFIG_MASK \ + (p4_config_pack_escr(P4_CONFIG_MASK_ESCR)) | \ + (p4_config_pack_cccr(P4_CONFIG_MASK_CCCR)) + static inline bool p4_is_event_cascaded(u64 config) { u32 cccr = p4_config_unpack_cccr(config); diff --git a/arch/x86/kernel/cpu/perf_event_p4.c b/arch/x86/kernel/cpu/perf_event_p4.c index 7e578e9cc58..4fe62c7165c 100644 --- a/arch/x86/kernel/cpu/perf_event_p4.c +++ b/arch/x86/kernel/cpu/perf_event_p4.c @@ -18,6 +18,8 @@ struct p4_event_bind { unsigned int opcode; /* Event code and ESCR selector */ unsigned int escr_msr[2]; /* ESCR MSR for this event */ + unsigned int escr_emask; /* valid ESCR EventMask bits */ + unsigned int shared; /* event is shared across threads */ char cntr[2][P4_CNTR_LIMIT]; /* counter index (offset), -1 on abscence */ }; @@ -66,231 +68,435 @@ static struct p4_event_bind p4_event_bind_map[] = { [P4_EVENT_TC_DELIVER_MODE] = { .opcode = P4_OPCODE(P4_EVENT_TC_DELIVER_MODE), .escr_msr = { MSR_P4_TC_ESCR0, MSR_P4_TC_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, DD) | + P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, DB) | + P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, DI) | + P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, BD) | + P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, BB) | + P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, BI) | + P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, ID), + .shared = 1, .cntr = { {4, 5, -1}, {6, 7, -1} }, }, [P4_EVENT_BPU_FETCH_REQUEST] = { .opcode = P4_OPCODE(P4_EVENT_BPU_FETCH_REQUEST), .escr_msr = { MSR_P4_BPU_ESCR0, MSR_P4_BPU_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_BPU_FETCH_REQUEST, TCMISS), .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_ITLB_REFERENCE] = { .opcode = P4_OPCODE(P4_EVENT_ITLB_REFERENCE), .escr_msr = { MSR_P4_ITLB_ESCR0, MSR_P4_ITLB_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_ITLB_REFERENCE, HIT) | + P4_ESCR_EMASK_BIT(P4_EVENT_ITLB_REFERENCE, MISS) | + P4_ESCR_EMASK_BIT(P4_EVENT_ITLB_REFERENCE, HIT_UK), .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_MEMORY_CANCEL] = { .opcode = P4_OPCODE(P4_EVENT_MEMORY_CANCEL), .escr_msr = { MSR_P4_DAC_ESCR0, MSR_P4_DAC_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_CANCEL, ST_RB_FULL) | + P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_CANCEL, 64K_CONF), .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_MEMORY_COMPLETE] = { .opcode = P4_OPCODE(P4_EVENT_MEMORY_COMPLETE), .escr_msr = { MSR_P4_SAAT_ESCR0 , MSR_P4_SAAT_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_COMPLETE, LSC) | + P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_COMPLETE, SSC), .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_LOAD_PORT_REPLAY] = { .opcode = P4_OPCODE(P4_EVENT_LOAD_PORT_REPLAY), .escr_msr = { MSR_P4_SAAT_ESCR0, MSR_P4_SAAT_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_LOAD_PORT_REPLAY, SPLIT_LD), .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_STORE_PORT_REPLAY] = { .opcode = P4_OPCODE(P4_EVENT_STORE_PORT_REPLAY), .escr_msr = { MSR_P4_SAAT_ESCR0 , MSR_P4_SAAT_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_STORE_PORT_REPLAY, SPLIT_ST), .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_MOB_LOAD_REPLAY] = { .opcode = P4_OPCODE(P4_EVENT_MOB_LOAD_REPLAY), .escr_msr = { MSR_P4_MOB_ESCR0, MSR_P4_MOB_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, NO_STA) | + P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, NO_STD) | + P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, PARTIAL_DATA) | + P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, UNALGN_ADDR), .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_PAGE_WALK_TYPE] = { .opcode = P4_OPCODE(P4_EVENT_PAGE_WALK_TYPE), .escr_msr = { MSR_P4_PMH_ESCR0, MSR_P4_PMH_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_PAGE_WALK_TYPE, DTMISS) | + P4_ESCR_EMASK_BIT(P4_EVENT_PAGE_WALK_TYPE, ITMISS), + .shared = 1, .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_BSQ_CACHE_REFERENCE] = { .opcode = P4_OPCODE(P4_EVENT_BSQ_CACHE_REFERENCE), .escr_msr = { MSR_P4_BSU_ESCR0, MSR_P4_BSU_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITS) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITM) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITS) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITM) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_MISS) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_MISS) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, WR_2ndL_MISS), .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_IOQ_ALLOCATION] = { .opcode = P4_OPCODE(P4_EVENT_IOQ_ALLOCATION), .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, DEFAULT) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, ALL_READ) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, ALL_WRITE) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_UC) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WC) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WT) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WP) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WB) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, OWN) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, OTHER) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, PREFETCH), .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_IOQ_ACTIVE_ENTRIES] = { /* shared ESCR */ .opcode = P4_OPCODE(P4_EVENT_IOQ_ACTIVE_ENTRIES), .escr_msr = { MSR_P4_FSB_ESCR1, MSR_P4_FSB_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, DEFAULT) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, ALL_READ) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, ALL_WRITE) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_UC) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WC) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WT) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WP) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WB) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, OWN) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, OTHER) | + P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, PREFETCH), .cntr = { {2, -1, -1}, {3, -1, -1} }, }, [P4_EVENT_FSB_DATA_ACTIVITY] = { .opcode = P4_OPCODE(P4_EVENT_FSB_DATA_ACTIVITY), .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_DRV) | + P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_OWN) | + P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_OTHER) | + P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_DRV) | + P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_OWN) | + P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_OTHER), + .shared = 1, .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_BSQ_ALLOCATION] = { /* shared ESCR, broken CCCR1 */ .opcode = P4_OPCODE(P4_EVENT_BSQ_ALLOCATION), .escr_msr = { MSR_P4_BSU_ESCR0, MSR_P4_BSU_ESCR0 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_TYPE0) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_TYPE1) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_LEN0) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_LEN1) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_IO_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_LOCK_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_CACHE_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_SPLIT_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_DEM_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_ORD_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE0) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE1) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE2), .cntr = { {0, -1, -1}, {1, -1, -1} }, }, [P4_EVENT_BSQ_ACTIVE_ENTRIES] = { /* shared ESCR */ .opcode = P4_OPCODE(P4_EVENT_BSQ_ACTIVE_ENTRIES), .escr_msr = { MSR_P4_BSU_ESCR1 , MSR_P4_BSU_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_TYPE0) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_TYPE1) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LEN0) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LEN1) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_IO_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LOCK_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_CACHE_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_SPLIT_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_DEM_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_ORD_TYPE) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE0) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE1) | + P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE2), .cntr = { {2, -1, -1}, {3, -1, -1} }, }, [P4_EVENT_SSE_INPUT_ASSIST] = { .opcode = P4_OPCODE(P4_EVENT_SSE_INPUT_ASSIST), .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_SSE_INPUT_ASSIST, ALL), + .shared = 1, .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_PACKED_SP_UOP] = { .opcode = P4_OPCODE(P4_EVENT_PACKED_SP_UOP), .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_PACKED_SP_UOP, ALL), + .shared = 1, .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_PACKED_DP_UOP] = { .opcode = P4_OPCODE(P4_EVENT_PACKED_DP_UOP), .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_PACKED_DP_UOP, ALL), + .shared = 1, .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_SCALAR_SP_UOP] = { .opcode = P4_OPCODE(P4_EVENT_SCALAR_SP_UOP), .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_SCALAR_SP_UOP, ALL), + .shared = 1, .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_SCALAR_DP_UOP] = { .opcode = P4_OPCODE(P4_EVENT_SCALAR_DP_UOP), .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_SCALAR_DP_UOP, ALL), + .shared = 1, .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_64BIT_MMX_UOP] = { .opcode = P4_OPCODE(P4_EVENT_64BIT_MMX_UOP), .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_64BIT_MMX_UOP, ALL), + .shared = 1, .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_128BIT_MMX_UOP] = { .opcode = P4_OPCODE(P4_EVENT_128BIT_MMX_UOP), .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_128BIT_MMX_UOP, ALL), + .shared = 1, .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_X87_FP_UOP] = { .opcode = P4_OPCODE(P4_EVENT_X87_FP_UOP), .escr_msr = { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_X87_FP_UOP, ALL), + .shared = 1, .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_TC_MISC] = { .opcode = P4_OPCODE(P4_EVENT_TC_MISC), .escr_msr = { MSR_P4_TC_ESCR0, MSR_P4_TC_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_TC_MISC, FLUSH), .cntr = { {4, 5, -1}, {6, 7, -1} }, }, [P4_EVENT_GLOBAL_POWER_EVENTS] = { .opcode = P4_OPCODE(P4_EVENT_GLOBAL_POWER_EVENTS), .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_GLOBAL_POWER_EVENTS, RUNNING), .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_TC_MS_XFER] = { .opcode = P4_OPCODE(P4_EVENT_TC_MS_XFER), .escr_msr = { MSR_P4_MS_ESCR0, MSR_P4_MS_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_TC_MS_XFER, CISC), .cntr = { {4, 5, -1}, {6, 7, -1} }, }, [P4_EVENT_UOP_QUEUE_WRITES] = { .opcode = P4_OPCODE(P4_EVENT_UOP_QUEUE_WRITES), .escr_msr = { MSR_P4_MS_ESCR0, MSR_P4_MS_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_UOP_QUEUE_WRITES, FROM_TC_BUILD) | + P4_ESCR_EMASK_BIT(P4_EVENT_UOP_QUEUE_WRITES, FROM_TC_DELIVER) | + P4_ESCR_EMASK_BIT(P4_EVENT_UOP_QUEUE_WRITES, FROM_ROM), .cntr = { {4, 5, -1}, {6, 7, -1} }, }, [P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE] = { .opcode = P4_OPCODE(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE), .escr_msr = { MSR_P4_TBPU_ESCR0 , MSR_P4_TBPU_ESCR0 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, CONDITIONAL) | + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, CALL) | + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, RETURN) | + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, INDIRECT), .cntr = { {4, 5, -1}, {6, 7, -1} }, }, [P4_EVENT_RETIRED_BRANCH_TYPE] = { .opcode = P4_OPCODE(P4_EVENT_RETIRED_BRANCH_TYPE), .escr_msr = { MSR_P4_TBPU_ESCR0 , MSR_P4_TBPU_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, CONDITIONAL) | + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, CALL) | + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, RETURN) | + P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, INDIRECT), .cntr = { {4, 5, -1}, {6, 7, -1} }, }, [P4_EVENT_RESOURCE_STALL] = { .opcode = P4_OPCODE(P4_EVENT_RESOURCE_STALL), .escr_msr = { MSR_P4_ALF_ESCR0, MSR_P4_ALF_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_RESOURCE_STALL, SBFULL), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_WC_BUFFER] = { .opcode = P4_OPCODE(P4_EVENT_WC_BUFFER), .escr_msr = { MSR_P4_DAC_ESCR0, MSR_P4_DAC_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_WC_BUFFER, WCB_EVICTS) | + P4_ESCR_EMASK_BIT(P4_EVENT_WC_BUFFER, WCB_FULL_EVICTS), + .shared = 1, .cntr = { {8, 9, -1}, {10, 11, -1} }, }, [P4_EVENT_B2B_CYCLES] = { .opcode = P4_OPCODE(P4_EVENT_B2B_CYCLES), .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .escr_emask = 0, .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_BNR] = { .opcode = P4_OPCODE(P4_EVENT_BNR), .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .escr_emask = 0, .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_SNOOP] = { .opcode = P4_OPCODE(P4_EVENT_SNOOP), .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .escr_emask = 0, .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_RESPONSE] = { .opcode = P4_OPCODE(P4_EVENT_RESPONSE), .escr_msr = { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 }, + .escr_emask = 0, .cntr = { {0, -1, -1}, {2, -1, -1} }, }, [P4_EVENT_FRONT_END_EVENT] = { .opcode = P4_OPCODE(P4_EVENT_FRONT_END_EVENT), .escr_msr = { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_FRONT_END_EVENT, NBOGUS) | + P4_ESCR_EMASK_BIT(P4_EVENT_FRONT_END_EVENT, BOGUS), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_EXECUTION_EVENT] = { .opcode = P4_OPCODE(P4_EVENT_EXECUTION_EVENT), .escr_msr = { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS0) | + P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS1) | + P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS2) | + P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS3) | + P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS0) | + P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS1) | + P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS2) | + P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS3), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_REPLAY_EVENT] = { .opcode = P4_OPCODE(P4_EVENT_REPLAY_EVENT), .escr_msr = { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_REPLAY_EVENT, NBOGUS) | + P4_ESCR_EMASK_BIT(P4_EVENT_REPLAY_EVENT, BOGUS), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_INSTR_RETIRED] = { .opcode = P4_OPCODE(P4_EVENT_INSTR_RETIRED), .escr_msr = { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, NBOGUSNTAG) | + P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, NBOGUSTAG) | + P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, BOGUSNTAG) | + P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, BOGUSTAG), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_UOPS_RETIRED] = { .opcode = P4_OPCODE(P4_EVENT_UOPS_RETIRED), .escr_msr = { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_UOPS_RETIRED, NBOGUS) | + P4_ESCR_EMASK_BIT(P4_EVENT_UOPS_RETIRED, BOGUS), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_UOP_TYPE] = { .opcode = P4_OPCODE(P4_EVENT_UOP_TYPE), .escr_msr = { MSR_P4_RAT_ESCR0, MSR_P4_RAT_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_UOP_TYPE, TAGLOADS) | + P4_ESCR_EMASK_BIT(P4_EVENT_UOP_TYPE, TAGSTORES), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_BRANCH_RETIRED] = { .opcode = P4_OPCODE(P4_EVENT_BRANCH_RETIRED), .escr_msr = { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMNP) | + P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMNM) | + P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMTP) | + P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMTM), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_MISPRED_BRANCH_RETIRED] = { .opcode = P4_OPCODE(P4_EVENT_MISPRED_BRANCH_RETIRED), .escr_msr = { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_MISPRED_BRANCH_RETIRED, NBOGUS), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_X87_ASSIST] = { .opcode = P4_OPCODE(P4_EVENT_X87_ASSIST), .escr_msr = { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, FPSU) | + P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, FPSO) | + P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, POAO) | + P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, POAU) | + P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, PREA), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_MACHINE_CLEAR] = { .opcode = P4_OPCODE(P4_EVENT_MACHINE_CLEAR), .escr_msr = { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_MACHINE_CLEAR, CLEAR) | + P4_ESCR_EMASK_BIT(P4_EVENT_MACHINE_CLEAR, MOCLEAR) | + P4_ESCR_EMASK_BIT(P4_EVENT_MACHINE_CLEAR, SMCLEAR), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_INSTR_COMPLETED] = { .opcode = P4_OPCODE(P4_EVENT_INSTR_COMPLETED), .escr_msr = { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 }, + .escr_emask = + P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_COMPLETED, NBOGUS) | + P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_COMPLETED, BOGUS), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, }; @@ -428,29 +634,73 @@ static u64 p4_pmu_event_map(int hw_event) return config; } +/* check cpu model specifics */ +static bool p4_event_match_cpu_model(unsigned int event_idx) +{ + /* INSTR_COMPLETED event only exist for model 3, 4, 6 (Prescott) */ + if (event_idx == P4_EVENT_INSTR_COMPLETED) { + if (boot_cpu_data.x86_model != 3 && + boot_cpu_data.x86_model != 4 && + boot_cpu_data.x86_model != 6) + return false; + } + + /* + * For info + * - IQ_ESCR0, IQ_ESCR1 only for models 1 and 2 + */ + + return true; +} + static int p4_validate_raw_event(struct perf_event *event) { - unsigned int v; + unsigned int v, emask; - /* user data may have out-of-bound event index */ + /* User data may have out-of-bound event index */ v = p4_config_unpack_event(event->attr.config); - if (v >= ARRAY_SIZE(p4_event_bind_map)) { - pr_warning("P4 PMU: Unknown event code: %d\n", v); + if (v >= ARRAY_SIZE(p4_event_bind_map)) + return -EINVAL; + + /* It may be unsupported: */ + if (!p4_event_match_cpu_model(v)) return -EINVAL; + + /* + * NOTE: P4_CCCR_THREAD_ANY has not the same meaning as + * in Architectural Performance Monitoring, it means not + * on _which_ logical cpu to count but rather _when_, ie it + * depends on logical cpu state -- count event if one cpu active, + * none, both or any, so we just allow user to pass any value + * desired. + * + * In turn we always set Tx_OS/Tx_USR bits bound to logical + * cpu without their propagation to another cpu + */ + + /* + * if an event is shared accross the logical threads + * the user needs special permissions to be able to use it + */ + if (p4_event_bind_map[v].shared) { + if (perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN)) + return -EACCES; } + /* ESCR EventMask bits may be invalid */ + emask = p4_config_unpack_escr(event->attr.config) & P4_ESCR_EVENTMASK_MASK; + if (emask & ~p4_event_bind_map[v].escr_emask) + return -EINVAL; + /* - * it may have some screwed PEBS bits + * it may have some invalid PEBS bits */ - if (p4_config_pebs_has(event->attr.config, P4_PEBS_CONFIG_ENABLE)) { - pr_warning("P4 PMU: PEBS are not supported yet\n"); + if (p4_config_pebs_has(event->attr.config, P4_PEBS_CONFIG_ENABLE)) return -EINVAL; - } + v = p4_config_unpack_metric(event->attr.config); - if (v >= ARRAY_SIZE(p4_pebs_bind_map)) { - pr_warning("P4 PMU: Unknown metric code: %d\n", v); + if (v >= ARRAY_SIZE(p4_pebs_bind_map)) return -EINVAL; - } return 0; } @@ -478,27 +728,21 @@ static int p4_hw_config(struct perf_event *event) if (event->attr.type == PERF_TYPE_RAW) { + /* + * Clear bits we reserve to be managed by kernel itself + * and never allowed from a user space + */ + event->attr.config &= P4_CONFIG_MASK; + rc = p4_validate_raw_event(event); if (rc) goto out; /* - * We don't control raw events so it's up to the caller - * to pass sane values (and we don't count the thread number - * on HT machine but allow HT-compatible specifics to be - * passed on) - * * Note that for RAW events we allow user to use P4_CCCR_RESERVED * bits since we keep additional info here (for cache events and etc) - * - * XXX: HT wide things should check perf_paranoid_cpu() && - * CAP_SYS_ADMIN */ - event->hw.config |= event->attr.config & - (p4_config_pack_escr(P4_ESCR_MASK_HT) | - p4_config_pack_cccr(P4_CCCR_MASK_HT | P4_CCCR_RESERVED)); - - event->hw.config &= ~P4_CCCR_FORCE_OVF; + event->hw.config |= event->attr.config; } rc = x86_setup_perfctr(event); -- cgit v1.2.3-70-g09d2 From e6b04b6b5a3182ae36cf9a69f1aaaee432edc8ad Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 2 Sep 2010 13:52:45 +0100 Subject: x86-64: Fix unwind annotations in syscall stubs With the return address removed from the stack, these should really refer to their caller's register state. Signed-off-by: Jan Beulich Acked-by: Alexander van Heukelum LKML-Reference: <4C7FBA3D0200007800013F61@vpn.id2.novell.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/entry_64.S | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index 17be5ec7cbb..16aeff0c315 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -714,9 +714,8 @@ END(ptregscall_common) ENTRY(stub_execve) CFI_STARTPROC - popq %r11 - CFI_ADJUST_CFA_OFFSET -8 - CFI_REGISTER rip, r11 + addq $8, %rsp + PARTIAL_FRAME 0 SAVE_REST FIXUP_TOP_OF_STACK %r11 movq %rsp, %rcx @@ -735,7 +734,7 @@ END(stub_execve) ENTRY(stub_rt_sigreturn) CFI_STARTPROC addq $8, %rsp - CFI_ADJUST_CFA_OFFSET -8 + PARTIAL_FRAME 0 SAVE_REST movq %rsp,%rdi FIXUP_TOP_OF_STACK %r11 @@ -1445,7 +1444,6 @@ error_swapgs: error_sti: TRACE_IRQS_OFF ret - CFI_ENDPROC /* * There are two places in the kernel that can potentially fault with @@ -1470,6 +1468,7 @@ bstep_iret: /* Fix truncated RIP */ movq %rcx,RIP+8(%rsp) jmp error_swapgs + CFI_ENDPROC END(error_entry) -- cgit v1.2.3-70-g09d2 From 1f130a783a796f147b080c594488b566c86007d0 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 2 Sep 2010 13:54:32 +0100 Subject: x86-64: Adjust frame type at paranoid_exit: As this isn't an exception or interrupt entry point, it doesn't have any of the hardware provide frame layouts active. Signed-off-by: Jan Beulich Acked-by: Alexander van Heukelum LKML-Reference: <4C7FBAA80200007800013F67@vpn.id2.novell.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/entry_64.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index 16aeff0c315..64dfe3045c1 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -1367,7 +1367,7 @@ paranoidzeroentry machine_check *machine_check_vector(%rip) /* ebx: no swapgs flag */ ENTRY(paranoid_exit) - INTR_FRAME + DEFAULT_FRAME DISABLE_INTERRUPTS(CLBR_NONE) TRACE_IRQS_OFF testl %ebx,%ebx /* swapgs needed? */ -- cgit v1.2.3-70-g09d2 From b1cccb1bb01dc1cb89f58723a58c3d4988d44d94 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 2 Sep 2010 13:55:11 +0100 Subject: x86-64: Use symbolics instead of raw numbers in entry_64.S ... making the code a little less fragile. Also use pushq_cfi instead of raw CFI annotations in two more places, and add two missing annotations after stack pointer adjustments which got modified here anyway. Signed-off-by: Jan Beulich Acked-by: Alexander van Heukelum LKML-Reference: <4C7FBACF0200007800013F6A@vpn.id2.novell.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/entry_64.S | 32 ++++++++++++++++---------------- 1 file changed, 16 insertions(+), 16 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index 64dfe3045c1..6f305830c80 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -795,8 +795,8 @@ END(interrupt) /* 0(%rsp): ~(interrupt number) */ .macro interrupt func - subq $10*8, %rsp - CFI_ADJUST_CFA_OFFSET 10*8 + subq $ORIG_RAX-ARGOFFSET+8, %rsp + CFI_ADJUST_CFA_OFFSET ORIG_RAX-ARGOFFSET+8 call save_args PARTIAL_FRAME 0 call \func @@ -1035,8 +1035,8 @@ ENTRY(\sym) INTR_FRAME PARAVIRT_ADJUST_EXCEPTION_FRAME pushq_cfi $-1 /* ORIG_RAX: no syscall to restart */ - subq $15*8,%rsp - CFI_ADJUST_CFA_OFFSET 15*8 + subq $ORIG_RAX-R15, %rsp + CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15 call error_entry DEFAULT_FRAME 0 movq %rsp,%rdi /* pt_regs pointer */ @@ -1051,9 +1051,9 @@ END(\sym) ENTRY(\sym) INTR_FRAME PARAVIRT_ADJUST_EXCEPTION_FRAME - pushq $-1 /* ORIG_RAX: no syscall to restart */ - CFI_ADJUST_CFA_OFFSET 8 - subq $15*8, %rsp + pushq_cfi $-1 /* ORIG_RAX: no syscall to restart */ + subq $ORIG_RAX-R15, %rsp + CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15 call save_paranoid TRACE_IRQS_OFF movq %rsp,%rdi /* pt_regs pointer */ @@ -1069,9 +1069,9 @@ END(\sym) ENTRY(\sym) INTR_FRAME PARAVIRT_ADJUST_EXCEPTION_FRAME - pushq $-1 /* ORIG_RAX: no syscall to restart */ - CFI_ADJUST_CFA_OFFSET 8 - subq $15*8, %rsp + pushq_cfi $-1 /* ORIG_RAX: no syscall to restart */ + subq $ORIG_RAX-R15, %rsp + CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15 call save_paranoid TRACE_IRQS_OFF movq %rsp,%rdi /* pt_regs pointer */ @@ -1088,8 +1088,8 @@ END(\sym) ENTRY(\sym) XCPT_FRAME PARAVIRT_ADJUST_EXCEPTION_FRAME - subq $15*8,%rsp - CFI_ADJUST_CFA_OFFSET 15*8 + subq $ORIG_RAX-R15, %rsp + CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15 call error_entry DEFAULT_FRAME 0 movq %rsp,%rdi /* pt_regs pointer */ @@ -1106,8 +1106,8 @@ END(\sym) ENTRY(\sym) XCPT_FRAME PARAVIRT_ADJUST_EXCEPTION_FRAME - subq $15*8,%rsp - CFI_ADJUST_CFA_OFFSET 15*8 + subq $ORIG_RAX-R15, %rsp + CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15 call save_paranoid DEFAULT_FRAME 0 TRACE_IRQS_OFF @@ -1497,8 +1497,8 @@ ENTRY(nmi) INTR_FRAME PARAVIRT_ADJUST_EXCEPTION_FRAME pushq_cfi $-1 - subq $15*8, %rsp - CFI_ADJUST_CFA_OFFSET 15*8 + subq $ORIG_RAX-R15, %rsp + CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15 call save_paranoid DEFAULT_FRAME 0 /* paranoidentry do_nmi, 0; without TRACE_IRQS_OFF */ -- cgit v1.2.3-70-g09d2 From a34107b5577968dc53cf9c2195c7c2d4a2caf9ce Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 2 Sep 2010 14:04:16 +0100 Subject: i386: Add unwind directives to syscall ptregs stubs When these stubs are actual functions (i.e. having a return instruction) and have stack manipulation instructions in them, they should also be annotated to allow unwinding through them. Signed-off-by: Jan Beulich Acked-by: Alexander van Heukelum LKML-Reference: <4C7FBCF00200007800013F99@vpn.id2.novell.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/entry_32.S | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S index 227d00920d2..d9b950ee559 100644 --- a/arch/x86/kernel/entry_32.S +++ b/arch/x86/kernel/entry_32.S @@ -750,14 +750,18 @@ ptregs_##name: \ #define PTREGSCALL3(name) \ ALIGN; \ ptregs_##name: \ + CFI_STARTPROC; \ leal 4(%esp),%eax; \ - pushl %eax; \ + pushl_cfi %eax; \ movl PT_EDX(%eax),%ecx; \ movl PT_ECX(%eax),%edx; \ movl PT_EBX(%eax),%eax; \ call sys_##name; \ addl $4,%esp; \ - ret + CFI_ADJUST_CFA_OFFSET -4; \ + ret; \ + CFI_ENDPROC; \ +ENDPROC(ptregs_##name) PTREGSCALL1(iopl) PTREGSCALL0(fork) @@ -772,15 +776,19 @@ PTREGSCALL1(vm86old) /* Clone is an oddball. The 4th arg is in %edi */ ALIGN; ptregs_clone: + CFI_STARTPROC leal 4(%esp),%eax - pushl %eax - pushl PT_EDI(%eax) + pushl_cfi %eax + pushl_cfi PT_EDI(%eax) movl PT_EDX(%eax),%ecx movl PT_ECX(%eax),%edx movl PT_EBX(%eax),%eax call sys_clone addl $8,%esp + CFI_ADJUST_CFA_OFFSET -8 ret + CFI_ENDPROC +ENDPROC(ptregs_clone) .macro FIXUP_ESPFIX_STACK /* -- cgit v1.2.3-70-g09d2 From df5d1874ce1a1f0e0eceff4fa3a9d45620243a68 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 2 Sep 2010 14:07:16 +0100 Subject: x86: Use {push,pop}{l,q}_cfi in more places ... plus additionally introduce {push,pop}f{l,q}_cfi. All in the hope that the code becomes better readable this way (it gets quite a bit smaller in any case). Signed-off-by: Jan Beulich Acked-by: Alexander van Heukelum LKML-Reference: <4C7FBDA40200007800013FAF@vpn.id2.novell.com> Signed-off-by: Ingo Molnar --- arch/x86/include/asm/dwarf2.h | 20 +++ arch/x86/kernel/entry_32.S | 294 ++++++++++++++---------------------------- arch/x86/kernel/entry_64.S | 65 ++++------ 3 files changed, 141 insertions(+), 238 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/dwarf2.h b/arch/x86/include/asm/dwarf2.h index 733f7e91e7a..32609919931 100644 --- a/arch/x86/include/asm/dwarf2.h +++ b/arch/x86/include/asm/dwarf2.h @@ -89,6 +89,16 @@ CFI_ADJUST_CFA_OFFSET -8 .endm + .macro pushfq_cfi + pushfq + CFI_ADJUST_CFA_OFFSET 8 + .endm + + .macro popfq_cfi + popfq + CFI_ADJUST_CFA_OFFSET -8 + .endm + .macro movq_cfi reg offset=0 movq %\reg, \offset(%rsp) CFI_REL_OFFSET \reg, \offset @@ -109,6 +119,16 @@ CFI_ADJUST_CFA_OFFSET -4 .endm + .macro pushfl_cfi + pushfl + CFI_ADJUST_CFA_OFFSET 4 + .endm + + .macro popfl_cfi + popfl + CFI_ADJUST_CFA_OFFSET -4 + .endm + .macro movl_cfi reg offset=0 movl %\reg, \offset(%esp) CFI_REL_OFFSET \reg, \offset diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S index d9b950ee559..9fb188d7bc7 100644 --- a/arch/x86/kernel/entry_32.S +++ b/arch/x86/kernel/entry_32.S @@ -115,8 +115,7 @@ /* unfortunately push/pop can't be no-op */ .macro PUSH_GS - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 .endm .macro POP_GS pop=0 addl $(4 + \pop), %esp @@ -140,14 +139,12 @@ #else /* CONFIG_X86_32_LAZY_GS */ .macro PUSH_GS - pushl %gs - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %gs /*CFI_REL_OFFSET gs, 0*/ .endm .macro POP_GS pop=0 -98: popl %gs - CFI_ADJUST_CFA_OFFSET -4 +98: popl_cfi %gs /*CFI_RESTORE gs*/ .if \pop <> 0 add $\pop, %esp @@ -195,35 +192,25 @@ .macro SAVE_ALL cld PUSH_GS - pushl %fs - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %fs /*CFI_REL_OFFSET fs, 0;*/ - pushl %es - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %es /*CFI_REL_OFFSET es, 0;*/ - pushl %ds - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ds /*CFI_REL_OFFSET ds, 0;*/ - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %eax CFI_REL_OFFSET eax, 0 - pushl %ebp - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ebp CFI_REL_OFFSET ebp, 0 - pushl %edi - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %edi CFI_REL_OFFSET edi, 0 - pushl %esi - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %esi CFI_REL_OFFSET esi, 0 - pushl %edx - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %edx CFI_REL_OFFSET edx, 0 - pushl %ecx - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ecx CFI_REL_OFFSET ecx, 0 - pushl %ebx - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ebx CFI_REL_OFFSET ebx, 0 movl $(__USER_DS), %edx movl %edx, %ds @@ -234,39 +221,29 @@ .endm .macro RESTORE_INT_REGS - popl %ebx - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %ebx CFI_RESTORE ebx - popl %ecx - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %ecx CFI_RESTORE ecx - popl %edx - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %edx CFI_RESTORE edx - popl %esi - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %esi CFI_RESTORE esi - popl %edi - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %edi CFI_RESTORE edi - popl %ebp - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %ebp CFI_RESTORE ebp - popl %eax - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %eax CFI_RESTORE eax .endm .macro RESTORE_REGS pop=0 RESTORE_INT_REGS -1: popl %ds - CFI_ADJUST_CFA_OFFSET -4 +1: popl_cfi %ds /*CFI_RESTORE ds;*/ -2: popl %es - CFI_ADJUST_CFA_OFFSET -4 +2: popl_cfi %es /*CFI_RESTORE es;*/ -3: popl %fs - CFI_ADJUST_CFA_OFFSET -4 +3: popl_cfi %fs /*CFI_RESTORE fs;*/ POP_GS \pop .pushsection .fixup, "ax" @@ -320,16 +297,12 @@ ENTRY(ret_from_fork) CFI_STARTPROC - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %eax call schedule_tail GET_THREAD_INFO(%ebp) - popl %eax - CFI_ADJUST_CFA_OFFSET -4 - pushl $0x0202 # Reset kernel eflags - CFI_ADJUST_CFA_OFFSET 4 - popfl - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %eax + pushl_cfi $0x0202 # Reset kernel eflags + popfl_cfi jmp syscall_exit CFI_ENDPROC END(ret_from_fork) @@ -409,29 +382,23 @@ sysenter_past_esp: * enough kernel state to call TRACE_IRQS_OFF can be called - but * we immediately enable interrupts at that point anyway. */ - pushl $(__USER_DS) - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $(__USER_DS) /*CFI_REL_OFFSET ss, 0*/ - pushl %ebp - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ebp CFI_REL_OFFSET esp, 0 - pushfl + pushfl_cfi orl $X86_EFLAGS_IF, (%esp) - CFI_ADJUST_CFA_OFFSET 4 - pushl $(__USER_CS) - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $(__USER_CS) /*CFI_REL_OFFSET cs, 0*/ /* * Push current_thread_info()->sysenter_return to the stack. * A tiny bit of offset fixup is necessary - 4*4 means the 4 words * pushed above; +8 corresponds to copy_thread's esp0 setting. */ - pushl (TI_sysenter_return-THREAD_SIZE+8+4*4)(%esp) - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi (TI_sysenter_return-THREAD_SIZE+8+4*4)(%esp) CFI_REL_OFFSET eip, 0 - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %eax SAVE_ALL ENABLE_INTERRUPTS(CLBR_NONE) @@ -486,8 +453,7 @@ sysenter_audit: movl %eax,%edx /* 2nd arg: syscall number */ movl $AUDIT_ARCH_I386,%eax /* 1st arg: audit arch */ call audit_syscall_entry - pushl %ebx - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ebx movl PT_EAX(%esp),%eax /* reload syscall number */ jmp sysenter_do_call @@ -529,8 +495,7 @@ ENDPROC(ia32_sysenter_target) # system call handler stub ENTRY(system_call) RING0_INT_FRAME # can't unwind into user space anyway - pushl %eax # save orig_eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %eax # save orig_eax SAVE_ALL GET_THREAD_INFO(%ebp) # system call tracing in operation / emulation @@ -566,7 +531,6 @@ restore_all_notrace: je ldt_ss # returning to user-space with LDT SS restore_nocheck: RESTORE_REGS 4 # skip orig_eax/error_code - CFI_ADJUST_CFA_OFFSET -4 irq_return: INTERRUPT_RETURN .section .fixup,"ax" @@ -619,10 +583,8 @@ ldt_ss: shr $16, %edx mov %dl, GDT_ESPFIX_SS + 4 /* bits 16..23 */ mov %dh, GDT_ESPFIX_SS + 7 /* bits 24..31 */ - pushl $__ESPFIX_SS - CFI_ADJUST_CFA_OFFSET 4 - push %eax /* new kernel esp */ - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $__ESPFIX_SS + pushl_cfi %eax /* new kernel esp */ /* Disable interrupts, but do not irqtrace this section: we * will soon execute iret and the tracer was already set to * the irqstate after the iret */ @@ -666,11 +628,9 @@ work_notifysig: # deal with pending signals and ALIGN work_notifysig_v86: - pushl %ecx # save ti_flags for do_notify_resume - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ecx # save ti_flags for do_notify_resume call save_v86_state # %eax contains pt_regs pointer - popl %ecx - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %ecx movl %eax, %esp #else movl %esp, %eax @@ -803,10 +763,8 @@ ENDPROC(ptregs_clone) mov GDT_ESPFIX_SS + 7, %ah /* bits 24..31 */ shl $16, %eax addl %esp, %eax /* the adjusted stack pointer */ - pushl $__KERNEL_DS - CFI_ADJUST_CFA_OFFSET 4 - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $__KERNEL_DS + pushl_cfi %eax lss (%esp), %esp /* switch to the normal stack segment */ CFI_ADJUST_CFA_OFFSET -8 .endm @@ -843,8 +801,7 @@ vector=FIRST_EXTERNAL_VECTOR .if vector <> FIRST_EXTERNAL_VECTOR CFI_ADJUST_CFA_OFFSET -4 .endif -1: pushl $(~vector+0x80) /* Note: always in signed byte range */ - CFI_ADJUST_CFA_OFFSET 4 +1: pushl_cfi $(~vector+0x80) /* Note: always in signed byte range */ .if ((vector-FIRST_EXTERNAL_VECTOR)%7) <> 6 jmp 2f .endif @@ -884,8 +841,7 @@ ENDPROC(common_interrupt) #define BUILD_INTERRUPT3(name, nr, fn) \ ENTRY(name) \ RING0_INT_FRAME; \ - pushl $~(nr); \ - CFI_ADJUST_CFA_OFFSET 4; \ + pushl_cfi $~(nr); \ SAVE_ALL; \ TRACE_IRQS_OFF \ movl %esp,%eax; \ @@ -901,21 +857,18 @@ ENDPROC(name) ENTRY(coprocessor_error) RING0_INT_FRAME - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 - pushl $do_coprocessor_error - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 + pushl_cfi $do_coprocessor_error jmp error_code CFI_ENDPROC END(coprocessor_error) ENTRY(simd_coprocessor_error) RING0_INT_FRAME - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 #ifdef CONFIG_X86_INVD_BUG /* AMD 486 bug: invd from userspace calls exception 19 instead of #GP */ -661: pushl $do_general_protection +661: pushl_cfi $do_general_protection 662: .section .altinstructions,"a" .balign 4 @@ -930,19 +883,16 @@ ENTRY(simd_coprocessor_error) 664: .previous #else - pushl $do_simd_coprocessor_error + pushl_cfi $do_simd_coprocessor_error #endif - CFI_ADJUST_CFA_OFFSET 4 jmp error_code CFI_ENDPROC END(simd_coprocessor_error) ENTRY(device_not_available) RING0_INT_FRAME - pushl $-1 # mark this as an int - CFI_ADJUST_CFA_OFFSET 4 - pushl $do_device_not_available - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $-1 # mark this as an int + pushl_cfi $do_device_not_available jmp error_code CFI_ENDPROC END(device_not_available) @@ -964,82 +914,68 @@ END(native_irq_enable_sysexit) ENTRY(overflow) RING0_INT_FRAME - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 - pushl $do_overflow - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 + pushl_cfi $do_overflow jmp error_code CFI_ENDPROC END(overflow) ENTRY(bounds) RING0_INT_FRAME - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 - pushl $do_bounds - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 + pushl_cfi $do_bounds jmp error_code CFI_ENDPROC END(bounds) ENTRY(invalid_op) RING0_INT_FRAME - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 - pushl $do_invalid_op - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 + pushl_cfi $do_invalid_op jmp error_code CFI_ENDPROC END(invalid_op) ENTRY(coprocessor_segment_overrun) RING0_INT_FRAME - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 - pushl $do_coprocessor_segment_overrun - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 + pushl_cfi $do_coprocessor_segment_overrun jmp error_code CFI_ENDPROC END(coprocessor_segment_overrun) ENTRY(invalid_TSS) RING0_EC_FRAME - pushl $do_invalid_TSS - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $do_invalid_TSS jmp error_code CFI_ENDPROC END(invalid_TSS) ENTRY(segment_not_present) RING0_EC_FRAME - pushl $do_segment_not_present - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $do_segment_not_present jmp error_code CFI_ENDPROC END(segment_not_present) ENTRY(stack_segment) RING0_EC_FRAME - pushl $do_stack_segment - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $do_stack_segment jmp error_code CFI_ENDPROC END(stack_segment) ENTRY(alignment_check) RING0_EC_FRAME - pushl $do_alignment_check - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $do_alignment_check jmp error_code CFI_ENDPROC END(alignment_check) ENTRY(divide_error) RING0_INT_FRAME - pushl $0 # no error code - CFI_ADJUST_CFA_OFFSET 4 - pushl $do_divide_error - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 # no error code + pushl_cfi $do_divide_error jmp error_code CFI_ENDPROC END(divide_error) @@ -1047,10 +983,8 @@ END(divide_error) #ifdef CONFIG_X86_MCE ENTRY(machine_check) RING0_INT_FRAME - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 - pushl machine_check_vector - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 + pushl_cfi machine_check_vector jmp error_code CFI_ENDPROC END(machine_check) @@ -1058,10 +992,8 @@ END(machine_check) ENTRY(spurious_interrupt_bug) RING0_INT_FRAME - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 - pushl $do_spurious_interrupt_bug - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 + pushl_cfi $do_spurious_interrupt_bug jmp error_code CFI_ENDPROC END(spurious_interrupt_bug) @@ -1092,8 +1024,7 @@ ENTRY(xen_sysenter_target) ENTRY(xen_hypervisor_callback) CFI_STARTPROC - pushl $0 - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $0 SAVE_ALL TRACE_IRQS_OFF @@ -1129,23 +1060,20 @@ ENDPROC(xen_hypervisor_callback) # We distinguish between categories by maintaining a status value in EAX. ENTRY(xen_failsafe_callback) CFI_STARTPROC - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %eax movl $1,%eax 1: mov 4(%esp),%ds 2: mov 8(%esp),%es 3: mov 12(%esp),%fs 4: mov 16(%esp),%gs testl %eax,%eax - popl %eax - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %eax lea 16(%esp),%esp CFI_ADJUST_CFA_OFFSET -16 jz 5f addl $16,%esp jmp iret_exc # EAX != 0 => Category 2 (Bad IRET) -5: pushl $0 # EAX == 0 => Category 1 (Bad segment) - CFI_ADJUST_CFA_OFFSET 4 +5: pushl_cfi $0 # EAX == 0 => Category 1 (Bad segment) SAVE_ALL jmp ret_from_exception CFI_ENDPROC @@ -1295,40 +1223,29 @@ syscall_table_size=(.-sys_call_table) ENTRY(page_fault) RING0_EC_FRAME - pushl $do_page_fault - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $do_page_fault ALIGN error_code: /* the function address is in %gs's slot on the stack */ - pushl %fs - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %fs /*CFI_REL_OFFSET fs, 0*/ - pushl %es - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %es /*CFI_REL_OFFSET es, 0*/ - pushl %ds - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ds /*CFI_REL_OFFSET ds, 0*/ - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %eax CFI_REL_OFFSET eax, 0 - pushl %ebp - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ebp CFI_REL_OFFSET ebp, 0 - pushl %edi - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %edi CFI_REL_OFFSET edi, 0 - pushl %esi - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %esi CFI_REL_OFFSET esi, 0 - pushl %edx - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %edx CFI_REL_OFFSET edx, 0 - pushl %ecx - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ecx CFI_REL_OFFSET ecx, 0 - pushl %ebx - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ebx CFI_REL_OFFSET ebx, 0 cld movl $(__KERNEL_PERCPU), %ecx @@ -1370,12 +1287,9 @@ END(page_fault) movl TSS_sysenter_sp0 + \offset(%esp), %esp CFI_DEF_CFA esp, 0 CFI_UNDEFINED eip - pushfl - CFI_ADJUST_CFA_OFFSET 4 - pushl $__KERNEL_CS - CFI_ADJUST_CFA_OFFSET 4 - pushl $sysenter_past_esp - CFI_ADJUST_CFA_OFFSET 4 + pushfl_cfi + pushl_cfi $__KERNEL_CS + pushl_cfi $sysenter_past_esp CFI_REL_OFFSET eip, 0 .endm @@ -1385,8 +1299,7 @@ ENTRY(debug) jne debug_stack_correct FIX_STACK 12, debug_stack_correct, debug_esp_fix_insn debug_stack_correct: - pushl $-1 # mark this as an int - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $-1 # mark this as an int SAVE_ALL TRACE_IRQS_OFF xorl %edx,%edx # error code 0 @@ -1406,32 +1319,27 @@ END(debug) */ ENTRY(nmi) RING0_INT_FRAME - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %eax movl %ss, %eax cmpw $__ESPFIX_SS, %ax - popl %eax - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %eax je nmi_espfix_stack cmpl $ia32_sysenter_target,(%esp) je nmi_stack_fixup - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %eax movl %esp,%eax /* Do not access memory above the end of our stack page, * it might not exist. */ andl $(THREAD_SIZE-1),%eax cmpl $(THREAD_SIZE-20),%eax - popl %eax - CFI_ADJUST_CFA_OFFSET -4 + popl_cfi %eax jae nmi_stack_correct cmpl $ia32_sysenter_target,12(%esp) je nmi_debug_stack_check nmi_stack_correct: /* We have a RING0_INT_FRAME here */ - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %eax SAVE_ALL xorl %edx,%edx # zero error code movl %esp,%eax # pt_regs pointer @@ -1460,18 +1368,14 @@ nmi_espfix_stack: * * create the pointer to lss back */ - pushl %ss - CFI_ADJUST_CFA_OFFSET 4 - pushl %esp - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %ss + pushl_cfi %esp addl $4, (%esp) /* copy the iret frame of 12 bytes */ .rept 3 - pushl 16(%esp) - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi 16(%esp) .endr - pushl %eax - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi %eax SAVE_ALL FIXUP_ESPFIX_STACK # %eax == %esp xorl %edx,%edx # zero error code @@ -1485,8 +1389,7 @@ END(nmi) ENTRY(int3) RING0_INT_FRAME - pushl $-1 # mark this as an int - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $-1 # mark this as an int SAVE_ALL TRACE_IRQS_OFF xorl %edx,%edx # zero error code @@ -1498,8 +1401,7 @@ END(int3) ENTRY(general_protection) RING0_EC_FRAME - pushl $do_general_protection - CFI_ADJUST_CFA_OFFSET 4 + pushl_cfi $do_general_protection jmp error_code CFI_ENDPROC END(general_protection) diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index 6f305830c80..8851a2bb8c0 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -213,23 +213,17 @@ ENDPROC(native_usergs_sysret64) .macro FAKE_STACK_FRAME child_rip /* push in order ss, rsp, eflags, cs, rip */ xorl %eax, %eax - pushq $__KERNEL_DS /* ss */ - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi $__KERNEL_DS /* ss */ /*CFI_REL_OFFSET ss,0*/ - pushq %rax /* rsp */ - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi %rax /* rsp */ CFI_REL_OFFSET rsp,0 - pushq $X86_EFLAGS_IF /* eflags - interrupts on */ - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi $X86_EFLAGS_IF /* eflags - interrupts on */ /*CFI_REL_OFFSET rflags,0*/ - pushq $__KERNEL_CS /* cs */ - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi $__KERNEL_CS /* cs */ /*CFI_REL_OFFSET cs,0*/ - pushq \child_rip /* rip */ - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi \child_rip /* rip */ CFI_REL_OFFSET rip,0 - pushq %rax /* orig rax */ - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi %rax /* orig rax */ .endm .macro UNFAKE_STACK_FRAME @@ -398,10 +392,8 @@ ENTRY(ret_from_fork) LOCK ; btr $TIF_FORK,TI_flags(%r8) - push kernel_eflags(%rip) - CFI_ADJUST_CFA_OFFSET 8 - popf # reset kernel eflags - CFI_ADJUST_CFA_OFFSET -8 + pushq_cfi kernel_eflags(%rip) + popfq_cfi # reset kernel eflags call schedule_tail # rdi: 'prev' task parameter @@ -521,11 +513,9 @@ sysret_careful: jnc sysret_signal TRACE_IRQS_ON ENABLE_INTERRUPTS(CLBR_NONE) - pushq %rdi - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi %rdi call schedule - popq %rdi - CFI_ADJUST_CFA_OFFSET -8 + popq_cfi %rdi jmp sysret_check /* Handle a signal */ @@ -634,11 +624,9 @@ int_careful: jnc int_very_careful TRACE_IRQS_ON ENABLE_INTERRUPTS(CLBR_NONE) - pushq %rdi - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi %rdi call schedule - popq %rdi - CFI_ADJUST_CFA_OFFSET -8 + popq_cfi %rdi DISABLE_INTERRUPTS(CLBR_NONE) TRACE_IRQS_OFF jmp int_with_check @@ -652,12 +640,10 @@ int_check_syscall_exit_work: /* Check for syscall exit trace */ testl $_TIF_WORK_SYSCALL_EXIT,%edx jz int_signal - pushq %rdi - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi %rdi leaq 8(%rsp),%rdi # &ptregs -> arg1 call syscall_trace_leave - popq %rdi - CFI_ADJUST_CFA_OFFSET -8 + popq_cfi %rdi andl $~(_TIF_WORK_SYSCALL_EXIT|_TIF_SYSCALL_EMU),%edi jmp int_restore_rest @@ -765,8 +751,7 @@ vector=FIRST_EXTERNAL_VECTOR .if vector <> FIRST_EXTERNAL_VECTOR CFI_ADJUST_CFA_OFFSET -8 .endif -1: pushq $(~vector+0x80) /* Note: always in signed byte range */ - CFI_ADJUST_CFA_OFFSET 8 +1: pushq_cfi $(~vector+0x80) /* Note: always in signed byte range */ .if ((vector-FIRST_EXTERNAL_VECTOR)%7) <> 6 jmp 2f .endif @@ -821,6 +806,7 @@ ret_from_intr: TRACE_IRQS_OFF decl PER_CPU_VAR(irq_count) leaveq + CFI_RESTORE rbp CFI_DEF_CFA_REGISTER rsp CFI_ADJUST_CFA_OFFSET -8 exit_intr: @@ -902,11 +888,9 @@ retint_careful: jnc retint_signal TRACE_IRQS_ON ENABLE_INTERRUPTS(CLBR_NONE) - pushq %rdi - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi %rdi call schedule - popq %rdi - CFI_ADJUST_CFA_OFFSET -8 + popq_cfi %rdi GET_THREAD_INFO(%rcx) DISABLE_INTERRUPTS(CLBR_NONE) TRACE_IRQS_OFF @@ -955,8 +939,7 @@ END(common_interrupt) .macro apicinterrupt num sym do_sym ENTRY(\sym) INTR_FRAME - pushq $~(\num) - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi $~(\num) interrupt \do_sym jmp ret_from_intr CFI_ENDPROC @@ -1138,16 +1121,14 @@ zeroentry simd_coprocessor_error do_simd_coprocessor_error /* edi: new selector */ ENTRY(native_load_gs_index) CFI_STARTPROC - pushf - CFI_ADJUST_CFA_OFFSET 8 + pushfq_cfi DISABLE_INTERRUPTS(CLBR_ANY & ~CLBR_RDI) SWAPGS gs_change: movl %edi,%gs 2: mfence /* workaround */ SWAPGS - popf - CFI_ADJUST_CFA_OFFSET -8 + popfq_cfi ret CFI_ENDPROC END(native_load_gs_index) @@ -1214,8 +1195,7 @@ END(kernel_execve) /* Call softirq on interrupt stack. Interrupts are off. */ ENTRY(call_softirq) CFI_STARTPROC - push %rbp - CFI_ADJUST_CFA_OFFSET 8 + pushq_cfi %rbp CFI_REL_OFFSET rbp,0 mov %rsp,%rbp CFI_DEF_CFA_REGISTER rbp @@ -1224,6 +1204,7 @@ ENTRY(call_softirq) push %rbp # backlink for old unwinder call __do_softirq leaveq + CFI_RESTORE rbp CFI_DEF_CFA_REGISTER rsp CFI_ADJUST_CFA_OFFSET -8 decl PER_CPU_VAR(irq_count) -- cgit v1.2.3-70-g09d2 From 7fe977dab356fbd7e86aa10bf83891761107c57c Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 2 Sep 2010 14:01:58 +0100 Subject: i386: Make kernel_execve() suitable for stack unwinding The explicit saving and restoring of %ebx was confusing stack unwind data consumers, and it is plain unnecessary to do this within the asm(), since that was only introduced for PIC user mode consumers of the original _syscall3() macro this was derived from. Signed-off-by: Jan Beulich Cc: Arnd Bergmann LKML-Reference: <4C7FBC660200007800013F95@vpn.id2.novell.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/sys_i386_32.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/sys_i386_32.c b/arch/x86/kernel/sys_i386_32.c index d5e06624e34..0b0cb5fede1 100644 --- a/arch/x86/kernel/sys_i386_32.c +++ b/arch/x86/kernel/sys_i386_32.c @@ -33,8 +33,8 @@ int kernel_execve(const char *filename, const char *const envp[]) { long __res; - asm volatile ("push %%ebx ; movl %2,%%ebx ; int $0x80 ; pop %%ebx" + asm volatile ("int $0x80" : "=a" (__res) - : "0" (__NR_execve), "ri" (filename), "c" (argv), "d" (envp) : "memory"); + : "0" (__NR_execve), "b" (filename), "c" (argv), "d" (envp) : "memory"); return __res; } -- cgit v1.2.3-70-g09d2 From 1c5f50ee347daea013671f718b70cd6bf497bef9 Mon Sep 17 00:00:00 2001 From: Wu Fengguang Date: Fri, 3 Sep 2010 17:04:07 +0800 Subject: x86, mm: fix uninitialized addr in kernel_physical_mapping_init() This re-adds the lost chunk in commit 9b861528a80. Reported-by: Stephen Rothwell Signed-off-by: Wu Fengguang Cc: Peter Zijlstra Cc: Haicheng Li Cc: Andi Kleen LKML-Reference: <20100903090407.GA19771@localhost> Signed-off-by: Ingo Molnar --- arch/x86/mm/init_64.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86') diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 64e7bc2ded9..74f0f358b07 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -570,6 +570,7 @@ kernel_physical_mapping_init(unsigned long start, start = (unsigned long)__va(start); end = (unsigned long)__va(end); + addr = start; for (; start < end; start = next) { pgd_t *pgd = pgd_offset_k(start); -- cgit v1.2.3-70-g09d2 From 57ab43e33122ffdc2eebca5d6de035699f0a8c06 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Fri, 3 Sep 2010 18:39:39 +0200 Subject: x86, GART: Remove superfluous AMD64_GARTEN There is a GARTEN so use that and drop the duplicate. Signed-off-by: Borislav Petkov Cc: Dave Airlie Cc: FUJITA Tomonori LKML-Reference: <1283531981-7495-2-git-send-email-bp@amd64.org> Signed-off-by: Ingo Molnar --- arch/x86/include/asm/gart.h | 1 - arch/x86/kernel/aperture_64.c | 4 ++-- drivers/char/agp/amd64-agp.c | 2 +- 3 files changed, 3 insertions(+), 4 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/gart.h b/arch/x86/include/asm/gart.h index 4ac5b0f33fc..fba0a72c4cc 100644 --- a/arch/x86/include/asm/gart.h +++ b/arch/x86/include/asm/gart.h @@ -27,7 +27,6 @@ extern int fix_aperture; #define AMD64_GARTAPERTUREBASE 0x94 #define AMD64_GARTTABLEBASE 0x98 #define AMD64_GARTCACHECTL 0x9c -#define AMD64_GARTEN (1<<0) #ifdef CONFIG_GART_IOMMU extern int gart_iommu_aperture; diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c index a2e0caf26e1..6fabd406aa7 100644 --- a/arch/x86/kernel/aperture_64.c +++ b/arch/x86/kernel/aperture_64.c @@ -307,7 +307,7 @@ void __init early_gart_iommu_check(void) continue; ctl = read_pci_config(bus, slot, 3, AMD64_GARTAPERTURECTL); - aper_enabled = ctl & AMD64_GARTEN; + aper_enabled = ctl & GARTEN; aper_order = (ctl >> 1) & 7; aper_size = (32 * 1024 * 1024) << aper_order; aper_base = read_pci_config(bus, slot, 3, AMD64_GARTAPERTUREBASE) & 0x7fff; @@ -362,7 +362,7 @@ void __init early_gart_iommu_check(void) continue; ctl = read_pci_config(bus, slot, 3, AMD64_GARTAPERTURECTL); - ctl &= ~AMD64_GARTEN; + ctl &= ~GARTEN; write_pci_config(bus, slot, 3, AMD64_GARTAPERTURECTL, ctl); } } diff --git a/drivers/char/agp/amd64-agp.c b/drivers/char/agp/amd64-agp.c index 70312da4c96..bfe372b3d9d 100644 --- a/drivers/char/agp/amd64-agp.c +++ b/drivers/char/agp/amd64-agp.c @@ -199,7 +199,7 @@ static void amd64_cleanup(void) struct pci_dev *dev = k8_northbridges[i]; /* disable gart translation */ pci_read_config_dword(dev, AMD64_GARTAPERTURECTL, &tmp); - tmp &= ~AMD64_GARTEN; + tmp &= ~GARTEN; pci_write_config_dword(dev, AMD64_GARTAPERTURECTL, tmp); } } -- cgit v1.2.3-70-g09d2 From 260133ab658bd2b80e07832a878e00405e19ff43 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Fri, 3 Sep 2010 18:39:40 +0200 Subject: x86, GART: Disable GART table walk probes Current code tramples over bit F3x90[6] which can be used to disable GART table walk probes. However, this bit should be set for performance reasons (speed up GART table walks). We are allowed to do that since we put GART tables in UC memory later anyway. Make it so. Signed-off-by: Borislav Petkov Cc: Dave Airlie Cc: FUJITA Tomonori LKML-Reference: <1283531981-7495-3-git-send-email-bp@amd64.org> Signed-off-by: Ingo Molnar --- arch/x86/include/asm/gart.h | 14 ++++++++++++++ arch/x86/kernel/aperture_64.c | 14 ++++++++------ arch/x86/kernel/pci-gart_64.c | 2 +- drivers/char/agp/amd64-agp.c | 2 +- 4 files changed, 24 insertions(+), 8 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/gart.h b/arch/x86/include/asm/gart.h index fba0a72c4cc..bf357f9b25f 100644 --- a/arch/x86/include/asm/gart.h +++ b/arch/x86/include/asm/gart.h @@ -17,6 +17,7 @@ extern int fix_aperture; #define GARTEN (1<<0) #define DISGARTCPU (1<<4) #define DISGARTIO (1<<5) +#define DISTLBWALKPRB (1<<6) /* GART cache control register bits. */ #define INVGART (1<<0) @@ -56,6 +57,19 @@ static inline void gart_iommu_hole_init(void) extern int agp_amd64_init(void); +static inline void gart_set_size_and_enable(struct pci_dev *dev, u32 order) +{ + u32 ctl; + + /* + * Don't enable translation but enable GART IO and CPU accesses. + * Also, set DISTLBWALKPRB since GART tables memory is UC. + */ + ctl = DISTLBWALKPRB | order << 1; + + pci_write_config_dword(dev, AMD64_GARTAPERTURECTL, ctl); +} + static inline void enable_gart_translation(struct pci_dev *dev, u64 addr) { u32 tmp, ctl; diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c index 6fabd406aa7..c9cb1736844 100644 --- a/arch/x86/kernel/aperture_64.c +++ b/arch/x86/kernel/aperture_64.c @@ -505,8 +505,13 @@ out: /* Fix up the north bridges */ for (i = 0; i < ARRAY_SIZE(bus_dev_ranges); i++) { - int bus; - int dev_base, dev_limit; + int bus, dev_base, dev_limit; + + /* + * Don't enable translation yet but enable GART IO and CPU + * accesses and set DISTLBWALKPRB since GART table memory is UC. + */ + u32 ctl = DISTLBWALKPRB | aper_order << 1; bus = bus_dev_ranges[i].bus; dev_base = bus_dev_ranges[i].dev_base; @@ -515,10 +520,7 @@ out: if (!early_is_k8_nb(read_pci_config(bus, slot, 3, 0x00))) continue; - /* Don't enable translation yet. That is done later. - Assume this BIOS didn't initialise the GART so - just overwrite all previous bits */ - write_pci_config(bus, slot, 3, AMD64_GARTAPERTURECTL, aper_order << 1); + write_pci_config(bus, slot, 3, AMD64_GARTAPERTURECTL, ctl); write_pci_config(bus, slot, 3, AMD64_GARTAPERTUREBASE, aper_alloc >> 25); } } diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index 0f7f130caa6..6015ee13e22 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -601,7 +601,7 @@ static void gart_fixup_northbridges(struct sys_device *dev) * Don't enable translations just yet. That is the next * step. Restore the pre-suspend aperture settings. */ - pci_write_config_dword(dev, AMD64_GARTAPERTURECTL, aperture_order << 1); + gart_set_size_and_enable(dev, aperture_order); pci_write_config_dword(dev, AMD64_GARTAPERTUREBASE, aperture_alloc >> 25); } } diff --git a/drivers/char/agp/amd64-agp.c b/drivers/char/agp/amd64-agp.c index bfe372b3d9d..564808a5c3c 100644 --- a/drivers/char/agp/amd64-agp.c +++ b/drivers/char/agp/amd64-agp.c @@ -313,7 +313,7 @@ static __devinit int fix_northbridge(struct pci_dev *nb, struct pci_dev *agp, if (order < 0 || !agp_aperture_valid(aper, (32*1024*1024)<> 25); return 0; -- cgit v1.2.3-70-g09d2 From d9fadd7ba99a67030783a212bcb17d11f0678433 Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Thu, 2 Sep 2010 15:37:10 +0200 Subject: x86, AMD: Remove needless CPU family check (for L3 cache info) Old 32-bit AMD CPUs (all w/o L3 cache) should always return 0 for cpuid_edx(0x80000006). For unknown reason the 32-bit implementation differed from the 64-bit implementation. See commit 67cddd94799 ("i386: Add L3 cache support to AMD CPUID4 emulation"). The current check is the result of the x86 merge. Signed-off-by: Andreas Herrmann Cc: Andi Kleen LKML-Reference: <20100902133710.GA5449@loge.amd.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/amd.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index fc563fabde6..0f0ace5d7db 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -540,7 +540,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) #endif if (c->extended_cpuid_level >= 0x80000006) { - if ((c->x86 >= 0x0f) && (cpuid_edx(0x80000006) & 0xf000)) + if (cpuid_edx(0x80000006) & 0xf000) num_cache_leaves = 4; else num_cache_leaves = 3; -- cgit v1.2.3-70-g09d2 From 592091c0e21655bfbdf68741dd5a920c2ac2bbe6 Mon Sep 17 00:00:00 2001 From: Jin Dongming Date: Tue, 31 Aug 2010 09:13:33 +0900 Subject: therm_throt.c: Trivial printk message fix for a unsuitable abbreviation of 'thermal' In unexpected_thermal_interrupt(), "LVT TMR interrupt" is used in error message. I don't think TMR is a suitable abbreviation for thermal. 1.TMR has been used in IA32 Architectures Software Developer's Manual, and is the abbreviation for Trigger Mode Register. 2.There is not an standard abbreviation "TMR" defined for thermal in IA32 Architectures Software Developer's Manual. 3.Though we could understand it as Thermal Monitor Register, it is easy to be misunderstood as a *TIMER* interrupt also. I think this patch will fix it. Signed-off-by: Jin Dongming Reviewed-by: Jean Delvare Cc: Brown Len Cc: Hidetoshi Seto Cc: Fenghua Yu LKML-Reference: <4C7C492D.5020704@np.css.fujitsu.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/mcheck/therm_throt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c index c2a8b26d4fe..1d0f743c9d6 100644 --- a/arch/x86/kernel/cpu/mcheck/therm_throt.c +++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c @@ -348,7 +348,7 @@ static void intel_thermal_interrupt(void) static void unexpected_thermal_interrupt(void) { - printk(KERN_ERR "CPU%d: Unexpected LVT TMR interrupt!\n", + printk(KERN_ERR "CPU%d: Unexpected LVT thermal interrupt!\n", smp_processor_id()); add_taint(TAINT_MACHINE_CHECK); } -- cgit v1.2.3-70-g09d2 From fe8e0c25cad28e8858ecfa5863333c70685a6811 Mon Sep 17 00:00:00 2001 From: Alexander van Heukelum Date: Mon, 6 Sep 2010 20:53:42 +0200 Subject: x86, 32-bit: Align percpu area and irq stacks to THREAD_SIZE The irq stacks, located in the percpu-area, need to be THREAD_SIZE aligned. Add the infrastucture to align percpu variables to larger-than-pagesize amounts within the percpu area, and use it to specify the alignment for the irq stacks. Also align the percpu area itself to THREAD_SIZE. This should make irq stacks work with 8K THREAD_SIZE. Signed-off-by: Alexander van Heukelum Cc: Tejun Heo Cc: hch@lst.de LKML-Reference: <1283799222.15941.1393621887@webmail.messagingengine.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/irq_32.c | 4 ++-- arch/x86/kernel/vmlinux.lds.S | 2 +- include/linux/percpu-defs.h | 12 ++++++++++++ 3 files changed, 15 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c index 3b5609f54c4..50fbbe60e50 100644 --- a/arch/x86/kernel/irq_32.c +++ b/arch/x86/kernel/irq_32.c @@ -60,8 +60,8 @@ union irq_ctx { static DEFINE_PER_CPU(union irq_ctx *, hardirq_ctx); static DEFINE_PER_CPU(union irq_ctx *, softirq_ctx); -static DEFINE_PER_CPU_PAGE_ALIGNED(union irq_ctx, hardirq_stack); -static DEFINE_PER_CPU_PAGE_ALIGNED(union irq_ctx, softirq_stack); +static DEFINE_PER_CPU_MULTIPAGE_ALIGNED(union irq_ctx, hardirq_stack, THREAD_SIZE); +static DEFINE_PER_CPU_MULTIPAGE_ALIGNED(union irq_ctx, softirq_stack, THREAD_SIZE); static void call_on_stack(void *func, void *stack) { diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S index d0bb52296fa..bb899475355 100644 --- a/arch/x86/kernel/vmlinux.lds.S +++ b/arch/x86/kernel/vmlinux.lds.S @@ -273,7 +273,7 @@ SECTIONS } #if !defined(CONFIG_X86_64) || !defined(CONFIG_SMP) - PERCPU(PAGE_SIZE) + PERCPU(THREAD_SIZE) #endif . = ALIGN(PAGE_SIZE); diff --git a/include/linux/percpu-defs.h b/include/linux/percpu-defs.h index ce2dc655cd1..ab20d119a85 100644 --- a/include/linux/percpu-defs.h +++ b/include/linux/percpu-defs.h @@ -138,6 +138,18 @@ DEFINE_PER_CPU_SECTION(type, name, "..page_aligned") \ __aligned(PAGE_SIZE) +/* + * Declaration/definition used for large per-CPU variables that must be + * aligned to something larger than the pagesize. + */ +#define DECLARE_PER_CPU_MULTIPAGE_ALIGNED(type, name, size) \ + DECLARE_PER_CPU_SECTION(type, name, "..page_aligned") \ + __aligned(size) + +#define DEFINE_PER_CPU_MULTIPAGE_ALIGNED(type, name, size) \ + DEFINE_PER_CPU_SECTION(type, name, "..page_aligned") \ + __aligned(size) + /* * Intermodule exports for per-CPU variables. sparse forgets about * address space across EXPORT_SYMBOL(), change EXPORT_SYMBOL() to -- cgit v1.2.3-70-g09d2 From 0f1cf415f00286a38f5ce35b459342dbfc895b50 Mon Sep 17 00:00:00 2001 From: Christian Dietrich Date: Mon, 6 Sep 2010 16:36:18 +0200 Subject: x86: Remove unnecessary #ifdef ACPI/X86_IO_ACPI The ACPI/X86_IO_ACPI ifdef isn't necessary at this point, because it is checked in an outer ifdef level already and has no effect here. Cleanup only, no functional effect. Signed-off-by: Christian Dietrich Acked-by: Borislav Petkov Cc: vamos-dev@i4.informatik.uni-erlangen.de LKML-Reference: Signed-off-by: Ingo Molnar --- arch/x86/kernel/early-quirks.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c index e5cc7e82e60..e4bd78c160d 100644 --- a/arch/x86/kernel/early-quirks.c +++ b/arch/x86/kernel/early-quirks.c @@ -97,7 +97,6 @@ static void __init nvidia_bugs(int num, int slot, int func) } -#if defined(CONFIG_ACPI) && defined(CONFIG_X86_IO_APIC) #if defined(CONFIG_ACPI) && defined(CONFIG_X86_IO_APIC) static u32 __init ati_ixp4x0_rev(int num, int slot, int func) { @@ -116,7 +115,6 @@ static u32 __init ati_ixp4x0_rev(int num, int slot, int func) d &= 0xff; return d; } -#endif static void __init ati_bugs(int num, int slot, int func) { -- cgit v1.2.3-70-g09d2 From 7ef8aa72ab176e0288f363d1247079732c5d5792 Mon Sep 17 00:00:00 2001 From: Andre Przywara Date: Mon, 6 Sep 2010 15:14:17 +0200 Subject: x86, cpu: Fix renamed, not-yet-shipping AMD CPUID feature bit The AMD SSE5 feature set as-it has been replaced by some extensions to the AVX instruction set. Thus the bit formerly advertised as SSE5 is re-used for one of these extensions (XOP). Although this changes the /proc/cpuinfo output, it is not user visible, as there are no CPUs (yet) having this feature. To avoid confusion this should be added to the stable series, too. Cc: stable@kernel.org [.32.x .34.x, .35.x] Signed-off-by: Andre Przywara LKML-Reference: <1283778860-26843-2-git-send-email-andre.przywara@amd.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/cpufeature.h | 2 +- arch/x86/kvm/x86.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h index 781a50b29a4..c9c73d8dedb 100644 --- a/arch/x86/include/asm/cpufeature.h +++ b/arch/x86/include/asm/cpufeature.h @@ -152,7 +152,7 @@ #define X86_FEATURE_3DNOWPREFETCH (6*32+ 8) /* 3DNow prefetch instructions */ #define X86_FEATURE_OSVW (6*32+ 9) /* OS Visible Workaround */ #define X86_FEATURE_IBS (6*32+10) /* Instruction Based Sampling */ -#define X86_FEATURE_SSE5 (6*32+11) /* SSE-5 */ +#define X86_FEATURE_XOP (6*32+11) /* extended AVX instructions */ #define X86_FEATURE_SKINIT (6*32+12) /* SKINIT/STGI instructions */ #define X86_FEATURE_WDT (6*32+13) /* Watchdog timer */ #define X86_FEATURE_NODEID_MSR (6*32+19) /* NodeId MSR */ diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 3a09c625d52..dd54779ccbe 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -1996,7 +1996,7 @@ static void do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, const u32 kvm_supported_word6_x86_features = F(LAHF_LM) | F(CMP_LEGACY) | F(SVM) | 0 /* ExtApicSpace */ | F(CR8_LEGACY) | F(ABM) | F(SSE4A) | F(MISALIGNSSE) | - F(3DNOWPREFETCH) | 0 /* OSVW */ | 0 /* IBS */ | F(SSE5) | + F(3DNOWPREFETCH) | 0 /* OSVW */ | 0 /* IBS */ | F(XOP) | 0 /* SKINIT */ | 0 /* WDT */; /* all calls to cpuid_count() should be made on the same cpu */ -- cgit v1.2.3-70-g09d2 From 33ed82fb6c5f032151f7e9f1ac7b667f78f426b8 Mon Sep 17 00:00:00 2001 From: Andre Przywara Date: Mon, 6 Sep 2010 15:14:18 +0200 Subject: x86, cpu: Update AMD CPUID feature bits AMD's public CPUID specification has been updated and some bits have got names. Add them to properly describe new CPU features. Signed-off-by: Andre Przywara LKML-Reference: <1283778860-26843-3-git-send-email-andre.przywara@amd.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/cpufeature.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h index c9c73d8dedb..341835df789 100644 --- a/arch/x86/include/asm/cpufeature.h +++ b/arch/x86/include/asm/cpufeature.h @@ -155,7 +155,11 @@ #define X86_FEATURE_XOP (6*32+11) /* extended AVX instructions */ #define X86_FEATURE_SKINIT (6*32+12) /* SKINIT/STGI instructions */ #define X86_FEATURE_WDT (6*32+13) /* Watchdog timer */ +#define X86_FEATURE_LWP (6*32+15) /* Light Weight Profiling */ +#define X86_FEATURE_FMA4 (6*32+16) /* 4 operands MAC instructions */ #define X86_FEATURE_NODEID_MSR (6*32+19) /* NodeId MSR */ +#define X86_FEATURE_TBM (6*32+21) /* trailing bit manipulations */ +#define X86_FEATURE_TOPOEXT (6*32+22) /* topology extensions CPUID leafs */ /* * Auxiliary flags: Linux defined - For features scattered in various -- cgit v1.2.3-70-g09d2 From 6d886fd042634c0d3312bace63a5d0c541b721dc Mon Sep 17 00:00:00 2001 From: Andre Przywara Date: Mon, 6 Sep 2010 15:14:19 +0200 Subject: x86, cpu: Fix allowed CPUID bits for KVM guests The AMD extensions to AVX (FMA4, XOP) work on the same YMM register set as AVX, so they are safe for guests to use, as long as AVX itself is allowed. Add F16C and AES on the way for the same reasons. Signed-off-by: Andre Przywara LKML-Reference: <1283778860-26843-4-git-send-email-andre.przywara@amd.com> Acked-by: Avi Kivity Signed-off-by: H. Peter Anvin --- arch/x86/kvm/x86.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index dd54779ccbe..6c2ecf0a806 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -1991,13 +1991,14 @@ static void do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, 0 /* Reserved */ | F(CX16) | 0 /* xTPR Update, PDCM */ | 0 /* Reserved, DCA */ | F(XMM4_1) | F(XMM4_2) | F(X2APIC) | F(MOVBE) | F(POPCNT) | - 0 /* Reserved, AES */ | F(XSAVE) | 0 /* OSXSAVE */ | F(AVX); + 0 /* Reserved*/ | F(AES) | F(XSAVE) | 0 /* OSXSAVE */ | F(AVX) | + F(F16C); /* cpuid 0x80000001.ecx */ const u32 kvm_supported_word6_x86_features = F(LAHF_LM) | F(CMP_LEGACY) | F(SVM) | 0 /* ExtApicSpace */ | F(CR8_LEGACY) | F(ABM) | F(SSE4A) | F(MISALIGNSSE) | F(3DNOWPREFETCH) | 0 /* OSVW */ | 0 /* IBS */ | F(XOP) | - 0 /* SKINIT */ | 0 /* WDT */; + 0 /* SKINIT, WDT, LWP */ | F(FMA4) | F(TBM); /* all calls to cpuid_count() should be made on the same cpu */ get_cpu(); -- cgit v1.2.3-70-g09d2 From aeb9c7d618264dcf6eea39142fefee096c3b09e2 Mon Sep 17 00:00:00 2001 From: Andre Przywara Date: Mon, 6 Sep 2010 15:14:20 +0200 Subject: x86, kvm: add new AMD SVM feature bits The recently updated CPUID specification names new SVM feature bits. Add them to the list of reported features. Signed-off-by: Andre Przywara LKML-Reference: <1283778860-26843-5-git-send-email-andre.przywara@amd.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/cpufeature.h | 7 +++++++ arch/x86/kernel/cpu/scattered.c | 6 ++++++ 2 files changed, 13 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h index 341835df789..bffeab7eab9 100644 --- a/arch/x86/include/asm/cpufeature.h +++ b/arch/x86/include/asm/cpufeature.h @@ -183,6 +183,13 @@ #define X86_FEATURE_LBRV (8*32+ 6) /* AMD LBR Virtualization support */ #define X86_FEATURE_SVML (8*32+ 7) /* "svm_lock" AMD SVM locking MSR */ #define X86_FEATURE_NRIPS (8*32+ 8) /* "nrip_save" AMD SVM next_rip save */ +#define X86_FEATURE_TSCRATEMSR (8*32+ 9) /* "tsc_scale" AMD TSC scaling support */ +#define X86_FEATURE_VMCBCLEAN (8*32+10) /* "vmcb_clean" AMD VMCB clean bits support */ +#define X86_FEATURE_FLUSHBYASID (8*32+11) /* AMD flush-by-ASID support */ +#define X86_FEATURE_DECODEASSISTS (8*32+12) /* AMD Decode Assists support */ +#define X86_FEATURE_PAUSEFILTER (8*32+13) /* AMD filtered pause intercept */ +#define X86_FEATURE_PFTHRESHOLD (8*32+14) /* AMD pause filter threshold */ + /* Intel-defined CPU features, CPUID level 0x00000007:0 (ebx), word 9 */ #define X86_FEATURE_FSGSBASE (9*32+ 0) /* {RD/WR}{FS/GS}BASE instructions*/ diff --git a/arch/x86/kernel/cpu/scattered.c b/arch/x86/kernel/cpu/scattered.c index 34b4dad6f0b..2c77931473f 100644 --- a/arch/x86/kernel/cpu/scattered.c +++ b/arch/x86/kernel/cpu/scattered.c @@ -43,6 +43,12 @@ void __cpuinit init_scattered_cpuid_features(struct cpuinfo_x86 *c) { X86_FEATURE_LBRV, CR_EDX, 1, 0x8000000a, 0 }, { X86_FEATURE_SVML, CR_EDX, 2, 0x8000000a, 0 }, { X86_FEATURE_NRIPS, CR_EDX, 3, 0x8000000a, 0 }, + { X86_FEATURE_TSCRATEMSR, CR_EDX, 4, 0x8000000a, 0 }, + { X86_FEATURE_VMCBCLEAN, CR_EDX, 5, 0x8000000a, 0 }, + { X86_FEATURE_FLUSHBYASID, CR_EDX, 6, 0x8000000a, 0 }, + { X86_FEATURE_DECODEASSISTS, CR_EDX, 7, 0x8000000a, 0 }, + { X86_FEATURE_PAUSEFILTER, CR_EDX,10, 0x8000000a, 0 }, + { X86_FEATURE_PFTHRESHOLD, CR_EDX,12, 0x8000000a, 0 }, { 0, 0, 0, 0, 0 } }; -- cgit v1.2.3-70-g09d2 From 37a2f9f30a360fb03522d15c85c78265ccd80287 Mon Sep 17 00:00:00 2001 From: Cliff Wickman Date: Wed, 8 Sep 2010 10:14:27 -0500 Subject: x86, kdump: Change copy_oldmem_page() to use cached addressing The copy of /proc/vmcore to a user buffer proceeds much faster if the kernel addresses memory as cached. With this patch we have seen an increase in transfer rate from less than 15MB/s to 80-460MB/s, depending on size of the transfer. This makes a big difference in time needed to save a system dump. Signed-off-by: Cliff Wickman Acked-by: "Eric W. Biederman" Cc: kexec@lists.infradead.org Cc: # as far back as it would apply LKML-Reference: Signed-off-by: Ingo Molnar --- arch/x86/kernel/crash_dump_64.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/crash_dump_64.c b/arch/x86/kernel/crash_dump_64.c index 045b36cada6..bf43188ca65 100644 --- a/arch/x86/kernel/crash_dump_64.c +++ b/arch/x86/kernel/crash_dump_64.c @@ -34,7 +34,7 @@ ssize_t copy_oldmem_page(unsigned long pfn, char *buf, if (!csize) return 0; - vaddr = ioremap(pfn << PAGE_SHIFT, PAGE_SIZE); + vaddr = ioremap_cache(pfn << PAGE_SHIFT, PAGE_SIZE); if (!vaddr) return -ENOMEM; -- cgit v1.2.3-70-g09d2 From 51b0fe39549a04858001922919ab355dee9bdfcf Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 11 Jun 2010 13:35:57 +0200 Subject: perf: Deconstify struct pmu sed -ie 's/const struct pmu\>/struct pmu/g' `git grep -l "const struct pmu\>"` Signed-off-by: Peter Zijlstra Cc: paulus Cc: stephane eranian Cc: Robert Richter Cc: Will Deacon Cc: Paul Mundt Cc: Frederic Weisbecker Cc: Cyrill Gorcunov Cc: Lin Ming Cc: Yanmin Cc: Deng-Cheng Zhu Cc: David Miller Cc: Michael Cree LKML-Reference: Signed-off-by: Ingo Molnar --- arch/alpha/kernel/perf_event.c | 4 ++-- arch/arm/kernel/perf_event.c | 2 +- arch/powerpc/kernel/perf_event.c | 8 ++++---- arch/powerpc/kernel/perf_event_fsl_emb.c | 2 +- arch/sh/kernel/perf_event.c | 4 ++-- arch/sparc/kernel/perf_event.c | 10 +++++----- arch/x86/kernel/cpu/perf_event.c | 14 +++++++------- include/linux/perf_event.h | 10 +++++----- kernel/perf_event.c | 26 +++++++++++++------------- 9 files changed, 40 insertions(+), 40 deletions(-) (limited to 'arch/x86') diff --git a/arch/alpha/kernel/perf_event.c b/arch/alpha/kernel/perf_event.c index 51c39fa4169..56fa4159038 100644 --- a/arch/alpha/kernel/perf_event.c +++ b/arch/alpha/kernel/perf_event.c @@ -642,7 +642,7 @@ static int __hw_perf_event_init(struct perf_event *event) return 0; } -static const struct pmu pmu = { +static struct pmu pmu = { .enable = alpha_pmu_enable, .disable = alpha_pmu_disable, .read = alpha_pmu_read, @@ -653,7 +653,7 @@ static const struct pmu pmu = { /* * Main entry point to initialise a HW performance event. */ -const struct pmu *hw_perf_event_init(struct perf_event *event) +struct pmu *hw_perf_event_init(struct perf_event *event) { int err; diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index 64ca8c3ab94..0671e92c511 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -491,7 +491,7 @@ __hw_perf_event_init(struct perf_event *event) return err; } -const struct pmu * +struct pmu * hw_perf_event_init(struct perf_event *event) { int err = 0; diff --git a/arch/powerpc/kernel/perf_event.c b/arch/powerpc/kernel/perf_event.c index d301a30445e..5f78681ad90 100644 --- a/arch/powerpc/kernel/perf_event.c +++ b/arch/powerpc/kernel/perf_event.c @@ -857,7 +857,7 @@ static void power_pmu_unthrottle(struct perf_event *event) * Set the flag to make pmu::enable() not perform the * schedulability test, it will be performed at commit time */ -void power_pmu_start_txn(const struct pmu *pmu) +void power_pmu_start_txn(struct pmu *pmu) { struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events); @@ -870,7 +870,7 @@ void power_pmu_start_txn(const struct pmu *pmu) * Clear the flag and pmu::enable() will perform the * schedulability test. */ -void power_pmu_cancel_txn(const struct pmu *pmu) +void power_pmu_cancel_txn(struct pmu *pmu) { struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events); @@ -882,7 +882,7 @@ void power_pmu_cancel_txn(const struct pmu *pmu) * Perform the group schedulability test as a whole * Return 0 if success */ -int power_pmu_commit_txn(const struct pmu *pmu) +int power_pmu_commit_txn(struct pmu *pmu) { struct cpu_hw_events *cpuhw; long i, n; @@ -1014,7 +1014,7 @@ static int hw_perf_cache_event(u64 config, u64 *eventp) return 0; } -const struct pmu *hw_perf_event_init(struct perf_event *event) +struct pmu *hw_perf_event_init(struct perf_event *event) { u64 ev; unsigned long flags; diff --git a/arch/powerpc/kernel/perf_event_fsl_emb.c b/arch/powerpc/kernel/perf_event_fsl_emb.c index 1ba45471ae4..d7619b5e7a6 100644 --- a/arch/powerpc/kernel/perf_event_fsl_emb.c +++ b/arch/powerpc/kernel/perf_event_fsl_emb.c @@ -428,7 +428,7 @@ static int hw_perf_cache_event(u64 config, u64 *eventp) return 0; } -const struct pmu *hw_perf_event_init(struct perf_event *event) +struct pmu *hw_perf_event_init(struct perf_event *event) { u64 ev; struct perf_event *events[MAX_HWEVENTS]; diff --git a/arch/sh/kernel/perf_event.c b/arch/sh/kernel/perf_event.c index 7a3dc356725..395572c94c6 100644 --- a/arch/sh/kernel/perf_event.c +++ b/arch/sh/kernel/perf_event.c @@ -257,13 +257,13 @@ static void sh_pmu_read(struct perf_event *event) sh_perf_event_update(event, &event->hw, event->hw.idx); } -static const struct pmu pmu = { +static struct pmu pmu = { .enable = sh_pmu_enable, .disable = sh_pmu_disable, .read = sh_pmu_read, }; -const struct pmu *hw_perf_event_init(struct perf_event *event) +struct pmu *hw_perf_event_init(struct perf_event *event) { int err = __hw_perf_event_init(event); if (unlikely(err)) { diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c index 4bc40293857..481b894a501 100644 --- a/arch/sparc/kernel/perf_event.c +++ b/arch/sparc/kernel/perf_event.c @@ -1099,7 +1099,7 @@ static int __hw_perf_event_init(struct perf_event *event) * Set the flag to make pmu::enable() not perform the * schedulability test, it will be performed at commit time */ -static void sparc_pmu_start_txn(const struct pmu *pmu) +static void sparc_pmu_start_txn(struct pmu *pmu) { struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events); @@ -1111,7 +1111,7 @@ static void sparc_pmu_start_txn(const struct pmu *pmu) * Clear the flag and pmu::enable() will perform the * schedulability test. */ -static void sparc_pmu_cancel_txn(const struct pmu *pmu) +static void sparc_pmu_cancel_txn(struct pmu *pmu) { struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events); @@ -1123,7 +1123,7 @@ static void sparc_pmu_cancel_txn(const struct pmu *pmu) * Perform the group schedulability test as a whole * Return 0 if success */ -static int sparc_pmu_commit_txn(const struct pmu *pmu) +static int sparc_pmu_commit_txn(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); int n; @@ -1142,7 +1142,7 @@ static int sparc_pmu_commit_txn(const struct pmu *pmu) return 0; } -static const struct pmu pmu = { +static struct pmu pmu = { .enable = sparc_pmu_enable, .disable = sparc_pmu_disable, .read = sparc_pmu_read, @@ -1152,7 +1152,7 @@ static const struct pmu pmu = { .commit_txn = sparc_pmu_commit_txn, }; -const struct pmu *hw_perf_event_init(struct perf_event *event) +struct pmu *hw_perf_event_init(struct perf_event *event) { int err = __hw_perf_event_init(event); diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index de6569c04cd..fdd97f2e996 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -618,7 +618,7 @@ static void x86_pmu_enable_all(int added) } } -static const struct pmu pmu; +static struct pmu pmu; static inline int is_x86_event(struct perf_event *event) { @@ -1427,7 +1427,7 @@ static inline void x86_pmu_read(struct perf_event *event) * Set the flag to make pmu::enable() not perform the * schedulability test, it will be performed at commit time */ -static void x86_pmu_start_txn(const struct pmu *pmu) +static void x86_pmu_start_txn(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -1440,7 +1440,7 @@ static void x86_pmu_start_txn(const struct pmu *pmu) * Clear the flag and pmu::enable() will perform the * schedulability test. */ -static void x86_pmu_cancel_txn(const struct pmu *pmu) +static void x86_pmu_cancel_txn(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -1457,7 +1457,7 @@ static void x86_pmu_cancel_txn(const struct pmu *pmu) * Perform the group schedulability test as a whole * Return 0 if success */ -static int x86_pmu_commit_txn(const struct pmu *pmu) +static int x86_pmu_commit_txn(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); int assign[X86_PMC_IDX_MAX]; @@ -1483,7 +1483,7 @@ static int x86_pmu_commit_txn(const struct pmu *pmu) return 0; } -static const struct pmu pmu = { +static struct pmu pmu = { .enable = x86_pmu_enable, .disable = x86_pmu_disable, .start = x86_pmu_start, @@ -1569,9 +1569,9 @@ out: return ret; } -const struct pmu *hw_perf_event_init(struct perf_event *event) +struct pmu *hw_perf_event_init(struct perf_event *event) { - const struct pmu *tmp; + struct pmu *tmp; int err; err = __hw_perf_event_init(event); diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 000610c4de7..09d048b5211 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -578,19 +578,19 @@ struct pmu { * Start the transaction, after this ->enable() doesn't need * to do schedulability tests. */ - void (*start_txn) (const struct pmu *pmu); + void (*start_txn) (struct pmu *pmu); /* * If ->start_txn() disabled the ->enable() schedulability test * then ->commit_txn() is required to perform one. On success * the transaction is closed. On error the transaction is kept * open until ->cancel_txn() is called. */ - int (*commit_txn) (const struct pmu *pmu); + int (*commit_txn) (struct pmu *pmu); /* * Will cancel the transaction, assumes ->disable() is called for * each successfull ->enable() during the transaction. */ - void (*cancel_txn) (const struct pmu *pmu); + void (*cancel_txn) (struct pmu *pmu); }; /** @@ -669,7 +669,7 @@ struct perf_event { int nr_siblings; int group_flags; struct perf_event *group_leader; - const struct pmu *pmu; + struct pmu *pmu; enum perf_event_active_state state; unsigned int attach_state; @@ -849,7 +849,7 @@ struct perf_output_handle { */ extern int perf_max_events; -extern const struct pmu *hw_perf_event_init(struct perf_event *event); +extern struct pmu *hw_perf_event_init(struct perf_event *event); extern void perf_event_task_sched_in(struct task_struct *task); extern void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next); diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 2d74f31220a..fb46fd13f31 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -75,7 +75,7 @@ static DEFINE_SPINLOCK(perf_resource_lock); /* * Architecture provided APIs - weak aliases: */ -extern __weak const struct pmu *hw_perf_event_init(struct perf_event *event) +extern __weak struct pmu *hw_perf_event_init(struct perf_event *event) { return NULL; } @@ -691,7 +691,7 @@ group_sched_in(struct perf_event *group_event, struct perf_event_context *ctx) { struct perf_event *event, *partial_group = NULL; - const struct pmu *pmu = group_event->pmu; + struct pmu *pmu = group_event->pmu; bool txn = false; if (group_event->state == PERF_EVENT_STATE_OFF) @@ -4501,7 +4501,7 @@ static int perf_swevent_int(struct perf_event *event) return 0; } -static const struct pmu perf_ops_generic = { +static struct pmu perf_ops_generic = { .enable = perf_swevent_enable, .disable = perf_swevent_disable, .start = perf_swevent_int, @@ -4614,7 +4614,7 @@ static void cpu_clock_perf_event_read(struct perf_event *event) cpu_clock_perf_event_update(event); } -static const struct pmu perf_ops_cpu_clock = { +static struct pmu perf_ops_cpu_clock = { .enable = cpu_clock_perf_event_enable, .disable = cpu_clock_perf_event_disable, .read = cpu_clock_perf_event_read, @@ -4671,7 +4671,7 @@ static void task_clock_perf_event_read(struct perf_event *event) task_clock_perf_event_update(event, time); } -static const struct pmu perf_ops_task_clock = { +static struct pmu perf_ops_task_clock = { .enable = task_clock_perf_event_enable, .disable = task_clock_perf_event_disable, .read = task_clock_perf_event_read, @@ -4785,7 +4785,7 @@ static int swevent_hlist_get(struct perf_event *event) #ifdef CONFIG_EVENT_TRACING -static const struct pmu perf_ops_tracepoint = { +static struct pmu perf_ops_tracepoint = { .enable = perf_trace_enable, .disable = perf_trace_disable, .start = perf_swevent_int, @@ -4849,7 +4849,7 @@ static void tp_perf_event_destroy(struct perf_event *event) perf_trace_destroy(event); } -static const struct pmu *tp_perf_event_init(struct perf_event *event) +static struct pmu *tp_perf_event_init(struct perf_event *event) { int err; @@ -4896,7 +4896,7 @@ static void perf_event_free_filter(struct perf_event *event) #else -static const struct pmu *tp_perf_event_init(struct perf_event *event) +static struct pmu *tp_perf_event_init(struct perf_event *event) { return NULL; } @@ -4918,7 +4918,7 @@ static void bp_perf_event_destroy(struct perf_event *event) release_bp_slot(event); } -static const struct pmu *bp_perf_event_init(struct perf_event *bp) +static struct pmu *bp_perf_event_init(struct perf_event *bp) { int err; @@ -4942,7 +4942,7 @@ void perf_bp_event(struct perf_event *bp, void *data) perf_swevent_add(bp, 1, 1, &sample, regs); } #else -static const struct pmu *bp_perf_event_init(struct perf_event *bp) +static struct pmu *bp_perf_event_init(struct perf_event *bp) { return NULL; } @@ -4964,9 +4964,9 @@ static void sw_perf_event_destroy(struct perf_event *event) swevent_hlist_put(event); } -static const struct pmu *sw_perf_event_init(struct perf_event *event) +static struct pmu *sw_perf_event_init(struct perf_event *event) { - const struct pmu *pmu = NULL; + struct pmu *pmu = NULL; u64 event_id = event->attr.config; /* @@ -5028,7 +5028,7 @@ perf_event_alloc(struct perf_event_attr *attr, perf_overflow_handler_t overflow_handler, gfp_t gfpflags) { - const struct pmu *pmu; + struct pmu *pmu; struct perf_event *event; struct hw_perf_event *hwc; long err; -- cgit v1.2.3-70-g09d2 From b0a873ebbf87bf38bf70b5e39a7cadc96099fa13 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 11 Jun 2010 13:35:08 +0200 Subject: perf: Register PMU implementations Simple registration interface for struct pmu, this provides the infrastructure for removing all the weak functions. Signed-off-by: Peter Zijlstra Cc: paulus Cc: stephane eranian Cc: Robert Richter Cc: Will Deacon Cc: Paul Mundt Cc: Frederic Weisbecker Cc: Cyrill Gorcunov Cc: Lin Ming Cc: Yanmin Cc: Deng-Cheng Zhu Cc: David Miller Cc: Michael Cree LKML-Reference: Signed-off-by: Ingo Molnar --- arch/alpha/kernel/perf_event.c | 37 +- arch/arm/kernel/perf_event.c | 38 +- arch/powerpc/kernel/perf_event.c | 46 +-- arch/powerpc/kernel/perf_event_fsl_emb.c | 37 +- arch/sh/kernel/perf_event.c | 35 +- arch/sparc/kernel/perf_event.c | 29 +- arch/x86/kernel/cpu/perf_event.c | 45 ++- include/linux/perf_event.h | 10 +- kernel/hw_breakpoint.c | 35 +- kernel/perf_event.c | 588 +++++++++++++++---------------- 10 files changed, 488 insertions(+), 412 deletions(-) (limited to 'arch/x86') diff --git a/arch/alpha/kernel/perf_event.c b/arch/alpha/kernel/perf_event.c index 56fa4159038..19660b5c298 100644 --- a/arch/alpha/kernel/perf_event.c +++ b/arch/alpha/kernel/perf_event.c @@ -642,34 +642,39 @@ static int __hw_perf_event_init(struct perf_event *event) return 0; } -static struct pmu pmu = { - .enable = alpha_pmu_enable, - .disable = alpha_pmu_disable, - .read = alpha_pmu_read, - .unthrottle = alpha_pmu_unthrottle, -}; - - /* * Main entry point to initialise a HW performance event. */ -struct pmu *hw_perf_event_init(struct perf_event *event) +static int alpha_pmu_event_init(struct perf_event *event) { int err; + switch (event->attr.type) { + case PERF_TYPE_RAW: + case PERF_TYPE_HARDWARE: + case PERF_TYPE_HW_CACHE: + break; + + default: + return -ENOENT; + } + if (!alpha_pmu) - return ERR_PTR(-ENODEV); + return -ENODEV; /* Do the real initialisation work. */ err = __hw_perf_event_init(event); - if (err) - return ERR_PTR(err); - - return &pmu; + return err; } - +static struct pmu pmu = { + .event_init = alpha_pmu_event_init, + .enable = alpha_pmu_enable, + .disable = alpha_pmu_disable, + .read = alpha_pmu_read, + .unthrottle = alpha_pmu_unthrottle, +}; /* * Main entry point - enable HW performance counters. @@ -838,5 +843,7 @@ void __init init_hw_perf_events(void) /* And set up PMU specification */ alpha_pmu = &ev67_pmu; perf_max_events = alpha_pmu->num_pmcs; + + perf_pmu_register(&pmu); } diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index 0671e92c511..f62f9db35db 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -306,12 +306,7 @@ out: return err; } -static struct pmu pmu = { - .enable = armpmu_enable, - .disable = armpmu_disable, - .unthrottle = armpmu_unthrottle, - .read = armpmu_read, -}; +static struct pmu pmu; static int validate_event(struct cpu_hw_events *cpuc, @@ -491,20 +486,29 @@ __hw_perf_event_init(struct perf_event *event) return err; } -struct pmu * -hw_perf_event_init(struct perf_event *event) +static int armpmu_event_init(struct perf_event *event) { int err = 0; + switch (event->attr.type) { + case PERF_TYPE_RAW: + case PERF_TYPE_HARDWARE: + case PERF_TYPE_HW_CACHE: + break; + + default: + return -ENOENT; + } + if (!armpmu) - return ERR_PTR(-ENODEV); + return -ENODEV; event->destroy = hw_perf_event_destroy; if (!atomic_inc_not_zero(&active_events)) { if (atomic_read(&active_events) > perf_max_events) { atomic_dec(&active_events); - return ERR_PTR(-ENOSPC); + return -ENOSPC; } mutex_lock(&pmu_reserve_mutex); @@ -518,15 +522,23 @@ hw_perf_event_init(struct perf_event *event) } if (err) - return ERR_PTR(err); + return err; err = __hw_perf_event_init(event); if (err) hw_perf_event_destroy(event); - return err ? ERR_PTR(err) : &pmu; + return err; } +static struct pmu pmu = { + .event_init = armpmu_event_init, + .enable = armpmu_enable, + .disable = armpmu_disable, + .unthrottle = armpmu_unthrottle, + .read = armpmu_read, +}; + void hw_perf_enable(void) { @@ -2994,6 +3006,8 @@ init_hw_perf_events(void) perf_max_events = -1; } + perf_pmu_register(&pmu); + return 0; } arch_initcall(init_hw_perf_events); diff --git a/arch/powerpc/kernel/perf_event.c b/arch/powerpc/kernel/perf_event.c index 5f78681ad90..19131b2614b 100644 --- a/arch/powerpc/kernel/perf_event.c +++ b/arch/powerpc/kernel/perf_event.c @@ -904,16 +904,6 @@ int power_pmu_commit_txn(struct pmu *pmu) return 0; } -struct pmu power_pmu = { - .enable = power_pmu_enable, - .disable = power_pmu_disable, - .read = power_pmu_read, - .unthrottle = power_pmu_unthrottle, - .start_txn = power_pmu_start_txn, - .cancel_txn = power_pmu_cancel_txn, - .commit_txn = power_pmu_commit_txn, -}; - /* * Return 1 if we might be able to put event on a limited PMC, * or 0 if not. @@ -1014,7 +1004,7 @@ static int hw_perf_cache_event(u64 config, u64 *eventp) return 0; } -struct pmu *hw_perf_event_init(struct perf_event *event) +static int power_pmu_event_init(struct perf_event *event) { u64 ev; unsigned long flags; @@ -1026,25 +1016,27 @@ struct pmu *hw_perf_event_init(struct perf_event *event) struct cpu_hw_events *cpuhw; if (!ppmu) - return ERR_PTR(-ENXIO); + return -ENOENT; + switch (event->attr.type) { case PERF_TYPE_HARDWARE: ev = event->attr.config; if (ev >= ppmu->n_generic || ppmu->generic_events[ev] == 0) - return ERR_PTR(-EOPNOTSUPP); + return -EOPNOTSUPP; ev = ppmu->generic_events[ev]; break; case PERF_TYPE_HW_CACHE: err = hw_perf_cache_event(event->attr.config, &ev); if (err) - return ERR_PTR(err); + return err; break; case PERF_TYPE_RAW: ev = event->attr.config; break; default: - return ERR_PTR(-EINVAL); + return -ENOENT; } + event->hw.config_base = ev; event->hw.idx = 0; @@ -1081,7 +1073,7 @@ struct pmu *hw_perf_event_init(struct perf_event *event) */ ev = normal_pmc_alternative(ev, flags); if (!ev) - return ERR_PTR(-EINVAL); + return -EINVAL; } } @@ -1095,19 +1087,19 @@ struct pmu *hw_perf_event_init(struct perf_event *event) n = collect_events(event->group_leader, ppmu->n_counter - 1, ctrs, events, cflags); if (n < 0) - return ERR_PTR(-EINVAL); + return -EINVAL; } events[n] = ev; ctrs[n] = event; cflags[n] = flags; if (check_excludes(ctrs, cflags, n, 1)) - return ERR_PTR(-EINVAL); + return -EINVAL; cpuhw = &get_cpu_var(cpu_hw_events); err = power_check_constraints(cpuhw, events, cflags, n + 1); put_cpu_var(cpu_hw_events); if (err) - return ERR_PTR(-EINVAL); + return -EINVAL; event->hw.config = events[n]; event->hw.event_base = cflags[n]; @@ -1132,11 +1124,20 @@ struct pmu *hw_perf_event_init(struct perf_event *event) } event->destroy = hw_perf_event_destroy; - if (err) - return ERR_PTR(err); - return &power_pmu; + return err; } +struct pmu power_pmu = { + .event_init = power_pmu_event_init, + .enable = power_pmu_enable, + .disable = power_pmu_disable, + .read = power_pmu_read, + .unthrottle = power_pmu_unthrottle, + .start_txn = power_pmu_start_txn, + .cancel_txn = power_pmu_cancel_txn, + .commit_txn = power_pmu_commit_txn, +}; + /* * A counter has overflowed; update its count and record * things if requested. Note that interrupts are hard-disabled @@ -1342,6 +1343,7 @@ int register_power_pmu(struct power_pmu *pmu) freeze_events_kernel = MMCR0_FCHV; #endif /* CONFIG_PPC64 */ + perf_pmu_register(&power_pmu); perf_cpu_notifier(power_pmu_notifier); return 0; diff --git a/arch/powerpc/kernel/perf_event_fsl_emb.c b/arch/powerpc/kernel/perf_event_fsl_emb.c index d7619b5e7a6..ea6a804e43f 100644 --- a/arch/powerpc/kernel/perf_event_fsl_emb.c +++ b/arch/powerpc/kernel/perf_event_fsl_emb.c @@ -378,13 +378,6 @@ static void fsl_emb_pmu_unthrottle(struct perf_event *event) local_irq_restore(flags); } -static struct pmu fsl_emb_pmu = { - .enable = fsl_emb_pmu_enable, - .disable = fsl_emb_pmu_disable, - .read = fsl_emb_pmu_read, - .unthrottle = fsl_emb_pmu_unthrottle, -}; - /* * Release the PMU if this is the last perf_event. */ @@ -428,7 +421,7 @@ static int hw_perf_cache_event(u64 config, u64 *eventp) return 0; } -struct pmu *hw_perf_event_init(struct perf_event *event) +static int fsl_emb_pmu_event_init(struct perf_event *event) { u64 ev; struct perf_event *events[MAX_HWEVENTS]; @@ -441,14 +434,14 @@ struct pmu *hw_perf_event_init(struct perf_event *event) case PERF_TYPE_HARDWARE: ev = event->attr.config; if (ev >= ppmu->n_generic || ppmu->generic_events[ev] == 0) - return ERR_PTR(-EOPNOTSUPP); + return -EOPNOTSUPP; ev = ppmu->generic_events[ev]; break; case PERF_TYPE_HW_CACHE: err = hw_perf_cache_event(event->attr.config, &ev); if (err) - return ERR_PTR(err); + return err; break; case PERF_TYPE_RAW: @@ -456,12 +449,12 @@ struct pmu *hw_perf_event_init(struct perf_event *event) break; default: - return ERR_PTR(-EINVAL); + return -ENOENT; } event->hw.config = ppmu->xlate_event(ev); if (!(event->hw.config & FSL_EMB_EVENT_VALID)) - return ERR_PTR(-EINVAL); + return -EINVAL; /* * If this is in a group, check if it can go on with all the @@ -473,7 +466,7 @@ struct pmu *hw_perf_event_init(struct perf_event *event) n = collect_events(event->group_leader, ppmu->n_counter - 1, events); if (n < 0) - return ERR_PTR(-EINVAL); + return -EINVAL; } if (event->hw.config & FSL_EMB_EVENT_RESTRICTED) { @@ -484,7 +477,7 @@ struct pmu *hw_perf_event_init(struct perf_event *event) } if (num_restricted >= ppmu->n_restricted) - return ERR_PTR(-EINVAL); + return -EINVAL; } event->hw.idx = -1; @@ -497,7 +490,7 @@ struct pmu *hw_perf_event_init(struct perf_event *event) if (event->attr.exclude_kernel) event->hw.config_base |= PMLCA_FCS; if (event->attr.exclude_idle) - return ERR_PTR(-ENOTSUPP); + return -ENOTSUPP; event->hw.last_period = event->hw.sample_period; local64_set(&event->hw.period_left, event->hw.last_period); @@ -523,11 +516,17 @@ struct pmu *hw_perf_event_init(struct perf_event *event) } event->destroy = hw_perf_event_destroy; - if (err) - return ERR_PTR(err); - return &fsl_emb_pmu; + return err; } +static struct pmu fsl_emb_pmu = { + .event_init = fsl_emb_pmu_event_init, + .enable = fsl_emb_pmu_enable, + .disable = fsl_emb_pmu_disable, + .read = fsl_emb_pmu_read, + .unthrottle = fsl_emb_pmu_unthrottle, +}; + /* * A counter has overflowed; update its count and record * things if requested. Note that interrupts are hard-disabled @@ -651,5 +650,7 @@ int register_fsl_emb_pmu(struct fsl_emb_pmu *pmu) pr_info("%s performance monitor hardware support registered\n", pmu->name); + perf_pmu_register(&fsl_emb_pmu); + return 0; } diff --git a/arch/sh/kernel/perf_event.c b/arch/sh/kernel/perf_event.c index 395572c94c6..8cb206597e0 100644 --- a/arch/sh/kernel/perf_event.c +++ b/arch/sh/kernel/perf_event.c @@ -257,26 +257,38 @@ static void sh_pmu_read(struct perf_event *event) sh_perf_event_update(event, &event->hw, event->hw.idx); } -static struct pmu pmu = { - .enable = sh_pmu_enable, - .disable = sh_pmu_disable, - .read = sh_pmu_read, -}; - -struct pmu *hw_perf_event_init(struct perf_event *event) +static int sh_pmu_event_init(struct perf_event *event) { - int err = __hw_perf_event_init(event); + int err; + + switch (event->attr.type) { + case PERF_TYPE_RAW: + case PERF_TYPE_HW_CACHE: + case PERF_TYPE_HARDWARE: + err = __hw_perf_event_init(event); + break; + + default: + return -ENOENT; + } + if (unlikely(err)) { if (event->destroy) event->destroy(event); - return ERR_PTR(err); } - return &pmu; + return err; } +static struct pmu pmu = { + .event_init = sh_pmu_event_init, + .enable = sh_pmu_enable, + .disable = sh_pmu_disable, + .read = sh_pmu_read, +}; + static void sh_pmu_setup(int cpu) -{ + struct cpu_hw_events *cpuhw = &per_cpu(cpu_hw_events, cpu); memset(cpuhw, 0, sizeof(struct cpu_hw_events)); @@ -325,6 +337,7 @@ int __cpuinit register_sh_pmu(struct sh_pmu *pmu) WARN_ON(pmu->num_events > MAX_HWEVENTS); + perf_pmu_register(&pmu); perf_cpu_notifier(sh_pmu_notifier); return 0; } diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c index 481b894a501..bed4327f5a7 100644 --- a/arch/sparc/kernel/perf_event.c +++ b/arch/sparc/kernel/perf_event.c @@ -1025,7 +1025,7 @@ out: return ret; } -static int __hw_perf_event_init(struct perf_event *event) +static int sparc_pmu_event_init(struct perf_event *event) { struct perf_event_attr *attr = &event->attr; struct perf_event *evts[MAX_HWEVENTS]; @@ -1038,17 +1038,27 @@ static int __hw_perf_event_init(struct perf_event *event) if (atomic_read(&nmi_active) < 0) return -ENODEV; - if (attr->type == PERF_TYPE_HARDWARE) { + switch (attr->type) { + case PERF_TYPE_HARDWARE: if (attr->config >= sparc_pmu->max_events) return -EINVAL; pmap = sparc_pmu->event_map(attr->config); - } else if (attr->type == PERF_TYPE_HW_CACHE) { + break; + + case PERF_TYPE_HW_CACHE: pmap = sparc_map_cache_event(attr->config); if (IS_ERR(pmap)) return PTR_ERR(pmap); - } else + break; + + case PERF_TYPE_RAW: return -EOPNOTSUPP; + default: + return -ENOENT; + + } + /* We save the enable bits in the config_base. */ hwc->config_base = sparc_pmu->irq_bit; if (!attr->exclude_user) @@ -1143,6 +1153,7 @@ static int sparc_pmu_commit_txn(struct pmu *pmu) } static struct pmu pmu = { + .event_init = sparc_pmu_event_init, .enable = sparc_pmu_enable, .disable = sparc_pmu_disable, .read = sparc_pmu_read, @@ -1152,15 +1163,6 @@ static struct pmu pmu = { .commit_txn = sparc_pmu_commit_txn, }; -struct pmu *hw_perf_event_init(struct perf_event *event) -{ - int err = __hw_perf_event_init(event); - - if (err) - return ERR_PTR(err); - return &pmu; -} - void perf_event_print_debug(void) { unsigned long flags; @@ -1280,6 +1282,7 @@ void __init init_hw_perf_events(void) /* All sparc64 PMUs currently have 2 events. */ perf_max_events = 2; + perf_pmu_register(&pmu); register_die_notifier(&perf_event_nmi_notifier); } diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index fdd97f2e996..2c89264ee79 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -530,7 +530,7 @@ static int x86_pmu_hw_config(struct perf_event *event) /* * Setup the hardware configuration for a given attr_type */ -static int __hw_perf_event_init(struct perf_event *event) +static int __x86_pmu_event_init(struct perf_event *event) { int err; @@ -1414,6 +1414,7 @@ void __init init_hw_perf_events(void) pr_info("... fixed-purpose events: %d\n", x86_pmu.num_counters_fixed); pr_info("... event mask: %016Lx\n", x86_pmu.intel_ctrl); + perf_pmu_register(&pmu); perf_cpu_notifier(x86_pmu_notifier); } @@ -1483,18 +1484,6 @@ static int x86_pmu_commit_txn(struct pmu *pmu) return 0; } -static struct pmu pmu = { - .enable = x86_pmu_enable, - .disable = x86_pmu_disable, - .start = x86_pmu_start, - .stop = x86_pmu_stop, - .read = x86_pmu_read, - .unthrottle = x86_pmu_unthrottle, - .start_txn = x86_pmu_start_txn, - .cancel_txn = x86_pmu_cancel_txn, - .commit_txn = x86_pmu_commit_txn, -}; - /* * validate that we can schedule this event */ @@ -1569,12 +1558,22 @@ out: return ret; } -struct pmu *hw_perf_event_init(struct perf_event *event) +int x86_pmu_event_init(struct perf_event *event) { struct pmu *tmp; int err; - err = __hw_perf_event_init(event); + switch (event->attr.type) { + case PERF_TYPE_RAW: + case PERF_TYPE_HARDWARE: + case PERF_TYPE_HW_CACHE: + break; + + default: + return -ENOENT; + } + + err = __x86_pmu_event_init(event); if (!err) { /* * we temporarily connect event to its pmu @@ -1594,12 +1593,24 @@ struct pmu *hw_perf_event_init(struct perf_event *event) if (err) { if (event->destroy) event->destroy(event); - return ERR_PTR(err); } - return &pmu; + return err; } +static struct pmu pmu = { + .event_init = x86_pmu_event_init, + .enable = x86_pmu_enable, + .disable = x86_pmu_disable, + .start = x86_pmu_start, + .stop = x86_pmu_stop, + .read = x86_pmu_read, + .unthrottle = x86_pmu_unthrottle, + .start_txn = x86_pmu_start_txn, + .cancel_txn = x86_pmu_cancel_txn, + .commit_txn = x86_pmu_commit_txn, +}; + /* * callchain support */ diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 09d048b5211..ab72f56eb37 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -561,6 +561,13 @@ struct perf_event; * struct pmu - generic performance monitoring unit */ struct pmu { + struct list_head entry; + + /* + * Should return -ENOENT when the @event doesn't match this pmu + */ + int (*event_init) (struct perf_event *event); + int (*enable) (struct perf_event *event); void (*disable) (struct perf_event *event); int (*start) (struct perf_event *event); @@ -849,7 +856,8 @@ struct perf_output_handle { */ extern int perf_max_events; -extern struct pmu *hw_perf_event_init(struct perf_event *event); +extern int perf_pmu_register(struct pmu *pmu); +extern void perf_pmu_unregister(struct pmu *pmu); extern void perf_event_task_sched_in(struct task_struct *task); extern void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next); diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c index d71a987fd2b..e9c5cfa1fd2 100644 --- a/kernel/hw_breakpoint.c +++ b/kernel/hw_breakpoint.c @@ -565,6 +565,34 @@ static struct notifier_block hw_breakpoint_exceptions_nb = { .priority = 0x7fffffff }; +static void bp_perf_event_destroy(struct perf_event *event) +{ + release_bp_slot(event); +} + +static int hw_breakpoint_event_init(struct perf_event *bp) +{ + int err; + + if (bp->attr.type != PERF_TYPE_BREAKPOINT) + return -ENOENT; + + err = register_perf_hw_breakpoint(bp); + if (err) + return err; + + bp->destroy = bp_perf_event_destroy; + + return 0; +} + +static struct pmu perf_breakpoint = { + .event_init = hw_breakpoint_event_init, + .enable = arch_install_hw_breakpoint, + .disable = arch_uninstall_hw_breakpoint, + .read = hw_breakpoint_pmu_read, +}; + static int __init init_hw_breakpoint(void) { unsigned int **task_bp_pinned; @@ -586,6 +614,8 @@ static int __init init_hw_breakpoint(void) constraints_initialized = 1; + perf_pmu_register(&perf_breakpoint); + return register_die_notifier(&hw_breakpoint_exceptions_nb); err_alloc: @@ -601,8 +631,3 @@ static int __init init_hw_breakpoint(void) core_initcall(init_hw_breakpoint); -struct pmu perf_ops_bp = { - .enable = arch_install_hw_breakpoint, - .disable = arch_uninstall_hw_breakpoint, - .read = hw_breakpoint_pmu_read, -}; diff --git a/kernel/perf_event.c b/kernel/perf_event.c index fb46fd13f31..288ce43de57 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -31,7 +31,6 @@ #include #include #include -#include #include @@ -72,14 +71,6 @@ static atomic64_t perf_event_id; */ static DEFINE_SPINLOCK(perf_resource_lock); -/* - * Architecture provided APIs - weak aliases: - */ -extern __weak struct pmu *hw_perf_event_init(struct perf_event *event) -{ - return NULL; -} - void __weak hw_perf_disable(void) { barrier(); } void __weak hw_perf_enable(void) { barrier(); } @@ -4501,182 +4492,6 @@ static int perf_swevent_int(struct perf_event *event) return 0; } -static struct pmu perf_ops_generic = { - .enable = perf_swevent_enable, - .disable = perf_swevent_disable, - .start = perf_swevent_int, - .stop = perf_swevent_void, - .read = perf_swevent_read, - .unthrottle = perf_swevent_void, /* hwc->interrupts already reset */ -}; - -/* - * hrtimer based swevent callback - */ - -static enum hrtimer_restart perf_swevent_hrtimer(struct hrtimer *hrtimer) -{ - enum hrtimer_restart ret = HRTIMER_RESTART; - struct perf_sample_data data; - struct pt_regs *regs; - struct perf_event *event; - u64 period; - - event = container_of(hrtimer, struct perf_event, hw.hrtimer); - event->pmu->read(event); - - perf_sample_data_init(&data, 0); - data.period = event->hw.last_period; - regs = get_irq_regs(); - - if (regs && !perf_exclude_event(event, regs)) { - if (!(event->attr.exclude_idle && current->pid == 0)) - if (perf_event_overflow(event, 0, &data, regs)) - ret = HRTIMER_NORESTART; - } - - period = max_t(u64, 10000, event->hw.sample_period); - hrtimer_forward_now(hrtimer, ns_to_ktime(period)); - - return ret; -} - -static void perf_swevent_start_hrtimer(struct perf_event *event) -{ - struct hw_perf_event *hwc = &event->hw; - - hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); - hwc->hrtimer.function = perf_swevent_hrtimer; - if (hwc->sample_period) { - u64 period; - - if (hwc->remaining) { - if (hwc->remaining < 0) - period = 10000; - else - period = hwc->remaining; - hwc->remaining = 0; - } else { - period = max_t(u64, 10000, hwc->sample_period); - } - __hrtimer_start_range_ns(&hwc->hrtimer, - ns_to_ktime(period), 0, - HRTIMER_MODE_REL, 0); - } -} - -static void perf_swevent_cancel_hrtimer(struct perf_event *event) -{ - struct hw_perf_event *hwc = &event->hw; - - if (hwc->sample_period) { - ktime_t remaining = hrtimer_get_remaining(&hwc->hrtimer); - hwc->remaining = ktime_to_ns(remaining); - - hrtimer_cancel(&hwc->hrtimer); - } -} - -/* - * Software event: cpu wall time clock - */ - -static void cpu_clock_perf_event_update(struct perf_event *event) -{ - int cpu = raw_smp_processor_id(); - s64 prev; - u64 now; - - now = cpu_clock(cpu); - prev = local64_xchg(&event->hw.prev_count, now); - local64_add(now - prev, &event->count); -} - -static int cpu_clock_perf_event_enable(struct perf_event *event) -{ - struct hw_perf_event *hwc = &event->hw; - int cpu = raw_smp_processor_id(); - - local64_set(&hwc->prev_count, cpu_clock(cpu)); - perf_swevent_start_hrtimer(event); - - return 0; -} - -static void cpu_clock_perf_event_disable(struct perf_event *event) -{ - perf_swevent_cancel_hrtimer(event); - cpu_clock_perf_event_update(event); -} - -static void cpu_clock_perf_event_read(struct perf_event *event) -{ - cpu_clock_perf_event_update(event); -} - -static struct pmu perf_ops_cpu_clock = { - .enable = cpu_clock_perf_event_enable, - .disable = cpu_clock_perf_event_disable, - .read = cpu_clock_perf_event_read, -}; - -/* - * Software event: task time clock - */ - -static void task_clock_perf_event_update(struct perf_event *event, u64 now) -{ - u64 prev; - s64 delta; - - prev = local64_xchg(&event->hw.prev_count, now); - delta = now - prev; - local64_add(delta, &event->count); -} - -static int task_clock_perf_event_enable(struct perf_event *event) -{ - struct hw_perf_event *hwc = &event->hw; - u64 now; - - now = event->ctx->time; - - local64_set(&hwc->prev_count, now); - - perf_swevent_start_hrtimer(event); - - return 0; -} - -static void task_clock_perf_event_disable(struct perf_event *event) -{ - perf_swevent_cancel_hrtimer(event); - task_clock_perf_event_update(event, event->ctx->time); - -} - -static void task_clock_perf_event_read(struct perf_event *event) -{ - u64 time; - - if (!in_nmi()) { - update_context_time(event->ctx); - time = event->ctx->time; - } else { - u64 now = perf_clock(); - u64 delta = now - event->ctx->timestamp; - time = event->ctx->time + delta; - } - - task_clock_perf_event_update(event, time); -} - -static struct pmu perf_ops_task_clock = { - .enable = task_clock_perf_event_enable, - .disable = task_clock_perf_event_disable, - .read = task_clock_perf_event_read, -}; - /* Deref the hlist from the update side */ static inline struct swevent_hlist * swevent_hlist_deref(struct perf_cpu_context *cpuctx) @@ -4783,17 +4598,63 @@ static int swevent_hlist_get(struct perf_event *event) return err; } -#ifdef CONFIG_EVENT_TRACING +atomic_t perf_swevent_enabled[PERF_COUNT_SW_MAX]; -static struct pmu perf_ops_tracepoint = { - .enable = perf_trace_enable, - .disable = perf_trace_disable, +static void sw_perf_event_destroy(struct perf_event *event) +{ + u64 event_id = event->attr.config; + + WARN_ON(event->parent); + + atomic_dec(&perf_swevent_enabled[event_id]); + swevent_hlist_put(event); +} + +static int perf_swevent_init(struct perf_event *event) +{ + int event_id = event->attr.config; + + if (event->attr.type != PERF_TYPE_SOFTWARE) + return -ENOENT; + + switch (event_id) { + case PERF_COUNT_SW_CPU_CLOCK: + case PERF_COUNT_SW_TASK_CLOCK: + return -ENOENT; + + default: + break; + } + + if (event_id > PERF_COUNT_SW_MAX) + return -ENOENT; + + if (!event->parent) { + int err; + + err = swevent_hlist_get(event); + if (err) + return err; + + atomic_inc(&perf_swevent_enabled[event_id]); + event->destroy = sw_perf_event_destroy; + } + + return 0; +} + +static struct pmu perf_swevent = { + .event_init = perf_swevent_init, + .enable = perf_swevent_enable, + .disable = perf_swevent_disable, .start = perf_swevent_int, .stop = perf_swevent_void, .read = perf_swevent_read, - .unthrottle = perf_swevent_void, + .unthrottle = perf_swevent_void, /* hwc->interrupts already reset */ }; +#ifdef CONFIG_EVENT_TRACING + static int perf_tp_filter_match(struct perf_event *event, struct perf_sample_data *data) { @@ -4849,10 +4710,13 @@ static void tp_perf_event_destroy(struct perf_event *event) perf_trace_destroy(event); } -static struct pmu *tp_perf_event_init(struct perf_event *event) +static int perf_tp_event_init(struct perf_event *event) { int err; + if (event->attr.type != PERF_TYPE_TRACEPOINT) + return -ENOENT; + /* * Raw tracepoint data is a severe data leak, only allow root to * have these. @@ -4860,15 +4724,30 @@ static struct pmu *tp_perf_event_init(struct perf_event *event) if ((event->attr.sample_type & PERF_SAMPLE_RAW) && perf_paranoid_tracepoint_raw() && !capable(CAP_SYS_ADMIN)) - return ERR_PTR(-EPERM); + return -EPERM; err = perf_trace_init(event); if (err) - return NULL; + return err; event->destroy = tp_perf_event_destroy; - return &perf_ops_tracepoint; + return 0; +} + +static struct pmu perf_tracepoint = { + .event_init = perf_tp_event_init, + .enable = perf_trace_enable, + .disable = perf_trace_disable, + .start = perf_swevent_int, + .stop = perf_swevent_void, + .read = perf_swevent_read, + .unthrottle = perf_swevent_void, +}; + +static inline void perf_tp_register(void) +{ + perf_pmu_register(&perf_tracepoint); } static int perf_event_set_filter(struct perf_event *event, void __user *arg) @@ -4896,9 +4775,8 @@ static void perf_event_free_filter(struct perf_event *event) #else -static struct pmu *tp_perf_event_init(struct perf_event *event) +static inline void perf_tp_register(void) { - return NULL; } static int perf_event_set_filter(struct perf_event *event, void __user *arg) @@ -4913,105 +4791,247 @@ static void perf_event_free_filter(struct perf_event *event) #endif /* CONFIG_EVENT_TRACING */ #ifdef CONFIG_HAVE_HW_BREAKPOINT -static void bp_perf_event_destroy(struct perf_event *event) +void perf_bp_event(struct perf_event *bp, void *data) { - release_bp_slot(event); + struct perf_sample_data sample; + struct pt_regs *regs = data; + + perf_sample_data_init(&sample, bp->attr.bp_addr); + + if (!perf_exclude_event(bp, regs)) + perf_swevent_add(bp, 1, 1, &sample, regs); } +#endif + +/* + * hrtimer based swevent callback + */ -static struct pmu *bp_perf_event_init(struct perf_event *bp) +static enum hrtimer_restart perf_swevent_hrtimer(struct hrtimer *hrtimer) { - int err; + enum hrtimer_restart ret = HRTIMER_RESTART; + struct perf_sample_data data; + struct pt_regs *regs; + struct perf_event *event; + u64 period; - err = register_perf_hw_breakpoint(bp); - if (err) - return ERR_PTR(err); + event = container_of(hrtimer, struct perf_event, hw.hrtimer); + event->pmu->read(event); + + perf_sample_data_init(&data, 0); + data.period = event->hw.last_period; + regs = get_irq_regs(); + + if (regs && !perf_exclude_event(event, regs)) { + if (!(event->attr.exclude_idle && current->pid == 0)) + if (perf_event_overflow(event, 0, &data, regs)) + ret = HRTIMER_NORESTART; + } - bp->destroy = bp_perf_event_destroy; + period = max_t(u64, 10000, event->hw.sample_period); + hrtimer_forward_now(hrtimer, ns_to_ktime(period)); - return &perf_ops_bp; + return ret; } -void perf_bp_event(struct perf_event *bp, void *data) +static void perf_swevent_start_hrtimer(struct perf_event *event) { - struct perf_sample_data sample; - struct pt_regs *regs = data; + struct hw_perf_event *hwc = &event->hw; - perf_sample_data_init(&sample, bp->attr.bp_addr); + hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + hwc->hrtimer.function = perf_swevent_hrtimer; + if (hwc->sample_period) { + u64 period; - if (!perf_exclude_event(bp, regs)) - perf_swevent_add(bp, 1, 1, &sample, regs); + if (hwc->remaining) { + if (hwc->remaining < 0) + period = 10000; + else + period = hwc->remaining; + hwc->remaining = 0; + } else { + period = max_t(u64, 10000, hwc->sample_period); + } + __hrtimer_start_range_ns(&hwc->hrtimer, + ns_to_ktime(period), 0, + HRTIMER_MODE_REL, 0); + } } -#else -static struct pmu *bp_perf_event_init(struct perf_event *bp) + +static void perf_swevent_cancel_hrtimer(struct perf_event *event) { - return NULL; + struct hw_perf_event *hwc = &event->hw; + + if (hwc->sample_period) { + ktime_t remaining = hrtimer_get_remaining(&hwc->hrtimer); + hwc->remaining = ktime_to_ns(remaining); + + hrtimer_cancel(&hwc->hrtimer); + } } -void perf_bp_event(struct perf_event *bp, void *regs) +/* + * Software event: cpu wall time clock + */ + +static void cpu_clock_event_update(struct perf_event *event) { + int cpu = raw_smp_processor_id(); + s64 prev; + u64 now; + + now = cpu_clock(cpu); + prev = local64_xchg(&event->hw.prev_count, now); + local64_add(now - prev, &event->count); } -#endif -atomic_t perf_swevent_enabled[PERF_COUNT_SW_MAX]; +static int cpu_clock_event_enable(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + int cpu = raw_smp_processor_id(); -static void sw_perf_event_destroy(struct perf_event *event) + local64_set(&hwc->prev_count, cpu_clock(cpu)); + perf_swevent_start_hrtimer(event); + + return 0; +} + +static void cpu_clock_event_disable(struct perf_event *event) { - u64 event_id = event->attr.config; + perf_swevent_cancel_hrtimer(event); + cpu_clock_event_update(event); +} - WARN_ON(event->parent); +static void cpu_clock_event_read(struct perf_event *event) +{ + cpu_clock_event_update(event); +} - atomic_dec(&perf_swevent_enabled[event_id]); - swevent_hlist_put(event); +static int cpu_clock_event_init(struct perf_event *event) +{ + if (event->attr.type != PERF_TYPE_SOFTWARE) + return -ENOENT; + + if (event->attr.config != PERF_COUNT_SW_CPU_CLOCK) + return -ENOENT; + + return 0; } -static struct pmu *sw_perf_event_init(struct perf_event *event) +static struct pmu perf_cpu_clock = { + .event_init = cpu_clock_event_init, + .enable = cpu_clock_event_enable, + .disable = cpu_clock_event_disable, + .read = cpu_clock_event_read, +}; + +/* + * Software event: task time clock + */ + +static void task_clock_event_update(struct perf_event *event, u64 now) { - struct pmu *pmu = NULL; - u64 event_id = event->attr.config; + u64 prev; + s64 delta; - /* - * Software events (currently) can't in general distinguish - * between user, kernel and hypervisor events. - * However, context switches and cpu migrations are considered - * to be kernel events, and page faults are never hypervisor - * events. - */ - switch (event_id) { - case PERF_COUNT_SW_CPU_CLOCK: - pmu = &perf_ops_cpu_clock; + prev = local64_xchg(&event->hw.prev_count, now); + delta = now - prev; + local64_add(delta, &event->count); +} - break; - case PERF_COUNT_SW_TASK_CLOCK: - /* - * If the user instantiates this as a per-cpu event, - * use the cpu_clock event instead. - */ - if (event->ctx->task) - pmu = &perf_ops_task_clock; - else - pmu = &perf_ops_cpu_clock; +static int task_clock_event_enable(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + u64 now; - break; - case PERF_COUNT_SW_PAGE_FAULTS: - case PERF_COUNT_SW_PAGE_FAULTS_MIN: - case PERF_COUNT_SW_PAGE_FAULTS_MAJ: - case PERF_COUNT_SW_CONTEXT_SWITCHES: - case PERF_COUNT_SW_CPU_MIGRATIONS: - case PERF_COUNT_SW_ALIGNMENT_FAULTS: - case PERF_COUNT_SW_EMULATION_FAULTS: - if (!event->parent) { - int err; - - err = swevent_hlist_get(event); - if (err) - return ERR_PTR(err); + now = event->ctx->time; - atomic_inc(&perf_swevent_enabled[event_id]); - event->destroy = sw_perf_event_destroy; + local64_set(&hwc->prev_count, now); + + perf_swevent_start_hrtimer(event); + + return 0; +} + +static void task_clock_event_disable(struct perf_event *event) +{ + perf_swevent_cancel_hrtimer(event); + task_clock_event_update(event, event->ctx->time); + +} + +static void task_clock_event_read(struct perf_event *event) +{ + u64 time; + + if (!in_nmi()) { + update_context_time(event->ctx); + time = event->ctx->time; + } else { + u64 now = perf_clock(); + u64 delta = now - event->ctx->timestamp; + time = event->ctx->time + delta; + } + + task_clock_event_update(event, time); +} + +static int task_clock_event_init(struct perf_event *event) +{ + if (event->attr.type != PERF_TYPE_SOFTWARE) + return -ENOENT; + + if (event->attr.config != PERF_COUNT_SW_TASK_CLOCK) + return -ENOENT; + + return 0; +} + +static struct pmu perf_task_clock = { + .event_init = task_clock_event_init, + .enable = task_clock_event_enable, + .disable = task_clock_event_disable, + .read = task_clock_event_read, +}; + +static LIST_HEAD(pmus); +static DEFINE_MUTEX(pmus_lock); +static struct srcu_struct pmus_srcu; + +int perf_pmu_register(struct pmu *pmu) +{ + mutex_lock(&pmus_lock); + list_add_rcu(&pmu->entry, &pmus); + mutex_unlock(&pmus_lock); + + return 0; +} + +void perf_pmu_unregister(struct pmu *pmu) +{ + mutex_lock(&pmus_lock); + list_del_rcu(&pmu->entry); + mutex_unlock(&pmus_lock); + + synchronize_srcu(&pmus_srcu); +} + +struct pmu *perf_init_event(struct perf_event *event) +{ + struct pmu *pmu = NULL; + int idx; + + idx = srcu_read_lock(&pmus_srcu); + list_for_each_entry_rcu(pmu, &pmus, entry) { + int ret = pmu->event_init(event); + if (!ret) + break; + if (ret != -ENOENT) { + pmu = ERR_PTR(ret); + break; } - pmu = &perf_ops_generic; - break; } + srcu_read_unlock(&pmus_srcu, idx); return pmu; } @@ -5092,29 +5112,8 @@ perf_event_alloc(struct perf_event_attr *attr, if (attr->inherit && (attr->read_format & PERF_FORMAT_GROUP)) goto done; - switch (attr->type) { - case PERF_TYPE_RAW: - case PERF_TYPE_HARDWARE: - case PERF_TYPE_HW_CACHE: - pmu = hw_perf_event_init(event); - break; - - case PERF_TYPE_SOFTWARE: - pmu = sw_perf_event_init(event); - break; - - case PERF_TYPE_TRACEPOINT: - pmu = tp_perf_event_init(event); - break; + pmu = perf_init_event(event); - case PERF_TYPE_BREAKPOINT: - pmu = bp_perf_event_init(event); - break; - - - default: - break; - } done: err = 0; if (!pmu) @@ -5979,22 +5978,15 @@ perf_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) return NOTIFY_OK; } -/* - * This has to have a higher priority than migration_notifier in sched.c. - */ -static struct notifier_block __cpuinitdata perf_cpu_nb = { - .notifier_call = perf_cpu_notify, - .priority = 20, -}; - void __init perf_event_init(void) { perf_event_init_all_cpus(); - perf_cpu_notify(&perf_cpu_nb, (unsigned long)CPU_UP_PREPARE, - (void *)(long)smp_processor_id()); - perf_cpu_notify(&perf_cpu_nb, (unsigned long)CPU_ONLINE, - (void *)(long)smp_processor_id()); - register_cpu_notifier(&perf_cpu_nb); + init_srcu_struct(&pmus_srcu); + perf_pmu_register(&perf_swevent); + perf_pmu_register(&perf_cpu_clock); + perf_pmu_register(&perf_task_clock); + perf_tp_register(); + perf_cpu_notifier(perf_cpu_notify); } static ssize_t perf_show_reserve_percpu(struct sysdev_class *class, -- cgit v1.2.3-70-g09d2 From 24cd7f54a0d47e1d5b3de29e2456bfbd2d8447b7 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 11 Jun 2010 17:32:03 +0200 Subject: perf: Reduce perf_disable() usage Since the current perf_disable() usage is only an optimization, remove it for now. This eases the removal of the __weak hw_perf_enable() interface. Signed-off-by: Peter Zijlstra Cc: paulus Cc: stephane eranian Cc: Robert Richter Cc: Will Deacon Cc: Paul Mundt Cc: Frederic Weisbecker Cc: Cyrill Gorcunov Cc: Lin Ming Cc: Yanmin Cc: Deng-Cheng Zhu Cc: David Miller Cc: Michael Cree LKML-Reference: Signed-off-by: Ingo Molnar --- arch/arm/kernel/perf_event.c | 3 +++ arch/powerpc/kernel/perf_event.c | 3 +++ arch/powerpc/kernel/perf_event_fsl_emb.c | 8 +++++-- arch/sh/kernel/perf_event.c | 11 +++++++--- arch/sparc/kernel/perf_event.c | 3 +++ arch/x86/kernel/cpu/perf_event.c | 22 ++++++++++++------- include/linux/perf_event.h | 20 ++++++++--------- kernel/perf_event.c | 37 +------------------------------- 8 files changed, 48 insertions(+), 59 deletions(-) (limited to 'arch/x86') diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index f62f9db35db..afc92c580d1 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -277,6 +277,8 @@ armpmu_enable(struct perf_event *event) int idx; int err = 0; + perf_disable(); + /* If we don't have a space for the counter then finish early. */ idx = armpmu->get_event_idx(cpuc, hwc); if (idx < 0) { @@ -303,6 +305,7 @@ armpmu_enable(struct perf_event *event) perf_event_update_userpage(event); out: + perf_enable(); return err; } diff --git a/arch/powerpc/kernel/perf_event.c b/arch/powerpc/kernel/perf_event.c index 19131b2614b..c1408821dbc 100644 --- a/arch/powerpc/kernel/perf_event.c +++ b/arch/powerpc/kernel/perf_event.c @@ -861,6 +861,7 @@ void power_pmu_start_txn(struct pmu *pmu) { struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events); + perf_disable(); cpuhw->group_flag |= PERF_EVENT_TXN; cpuhw->n_txn_start = cpuhw->n_events; } @@ -875,6 +876,7 @@ void power_pmu_cancel_txn(struct pmu *pmu) struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events); cpuhw->group_flag &= ~PERF_EVENT_TXN; + perf_enable(); } /* @@ -901,6 +903,7 @@ int power_pmu_commit_txn(struct pmu *pmu) cpuhw->event[i]->hw.config = cpuhw->events[i]; cpuhw->group_flag &= ~PERF_EVENT_TXN; + perf_enable(); return 0; } diff --git a/arch/powerpc/kernel/perf_event_fsl_emb.c b/arch/powerpc/kernel/perf_event_fsl_emb.c index ea6a804e43f..9bc84a7fd90 100644 --- a/arch/powerpc/kernel/perf_event_fsl_emb.c +++ b/arch/powerpc/kernel/perf_event_fsl_emb.c @@ -262,7 +262,7 @@ static int collect_events(struct perf_event *group, int max_count, return n; } -/* perf must be disabled, context locked on entry */ +/* context locked on entry */ static int fsl_emb_pmu_enable(struct perf_event *event) { struct cpu_hw_events *cpuhw; @@ -271,6 +271,7 @@ static int fsl_emb_pmu_enable(struct perf_event *event) u64 val; int i; + perf_disable(); cpuhw = &get_cpu_var(cpu_hw_events); if (event->hw.config & FSL_EMB_EVENT_RESTRICTED) @@ -310,15 +311,17 @@ static int fsl_emb_pmu_enable(struct perf_event *event) ret = 0; out: put_cpu_var(cpu_hw_events); + perf_enable(); return ret; } -/* perf must be disabled, context locked on entry */ +/* context locked on entry */ static void fsl_emb_pmu_disable(struct perf_event *event) { struct cpu_hw_events *cpuhw; int i = event->hw.idx; + perf_disable(); if (i < 0) goto out; @@ -346,6 +349,7 @@ static void fsl_emb_pmu_disable(struct perf_event *event) cpuhw->n_events--; out: + perf_enable(); put_cpu_var(cpu_hw_events); } diff --git a/arch/sh/kernel/perf_event.c b/arch/sh/kernel/perf_event.c index 8cb206597e0..d042989ceb4 100644 --- a/arch/sh/kernel/perf_event.c +++ b/arch/sh/kernel/perf_event.c @@ -230,11 +230,14 @@ static int sh_pmu_enable(struct perf_event *event) struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct hw_perf_event *hwc = &event->hw; int idx = hwc->idx; + int ret = -EAGAIN; + + perf_disable(); if (test_and_set_bit(idx, cpuc->used_mask)) { idx = find_first_zero_bit(cpuc->used_mask, sh_pmu->num_events); if (idx == sh_pmu->num_events) - return -EAGAIN; + goto out; set_bit(idx, cpuc->used_mask); hwc->idx = idx; @@ -248,8 +251,10 @@ static int sh_pmu_enable(struct perf_event *event) sh_pmu->enable(hwc, idx); perf_event_update_userpage(event); - - return 0; + ret = 0; +out: + perf_enable(); + return ret; } static void sh_pmu_read(struct perf_event *event) diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c index bed4327f5a7..d0131deeeaf 100644 --- a/arch/sparc/kernel/perf_event.c +++ b/arch/sparc/kernel/perf_event.c @@ -1113,6 +1113,7 @@ static void sparc_pmu_start_txn(struct pmu *pmu) { struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events); + perf_disable(); cpuhw->group_flag |= PERF_EVENT_TXN; } @@ -1126,6 +1127,7 @@ static void sparc_pmu_cancel_txn(struct pmu *pmu) struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events); cpuhw->group_flag &= ~PERF_EVENT_TXN; + perf_enable(); } /* @@ -1149,6 +1151,7 @@ static int sparc_pmu_commit_txn(struct pmu *pmu) return -EAGAIN; cpuc->group_flag &= ~PERF_EVENT_TXN; + perf_enable(); return 0; } diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 2c89264ee79..846070ce49c 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -969,10 +969,11 @@ static int x86_pmu_enable(struct perf_event *event) hwc = &event->hw; + perf_disable(); n0 = cpuc->n_events; - n = collect_events(cpuc, event, false); - if (n < 0) - return n; + ret = n = collect_events(cpuc, event, false); + if (ret < 0) + goto out; /* * If group events scheduling transaction was started, @@ -980,23 +981,26 @@ static int x86_pmu_enable(struct perf_event *event) * at commit time(->commit_txn) as a whole */ if (cpuc->group_flag & PERF_EVENT_TXN) - goto out; + goto done_collect; ret = x86_pmu.schedule_events(cpuc, n, assign); if (ret) - return ret; + goto out; /* * copy new assignment, now we know it is possible * will be used by hw_perf_enable() */ memcpy(cpuc->assign, assign, n*sizeof(int)); -out: +done_collect: cpuc->n_events = n; cpuc->n_added += n - n0; cpuc->n_txn += n - n0; - return 0; + ret = 0; +out: + perf_enable(); + return ret; } static int x86_pmu_start(struct perf_event *event) @@ -1432,6 +1436,7 @@ static void x86_pmu_start_txn(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + perf_disable(); cpuc->group_flag |= PERF_EVENT_TXN; cpuc->n_txn = 0; } @@ -1451,6 +1456,7 @@ static void x86_pmu_cancel_txn(struct pmu *pmu) */ cpuc->n_added -= cpuc->n_txn; cpuc->n_events -= cpuc->n_txn; + perf_enable(); } /* @@ -1480,7 +1486,7 @@ static int x86_pmu_commit_txn(struct pmu *pmu) memcpy(cpuc->assign, assign, n*sizeof(int)); cpuc->group_flag &= ~PERF_EVENT_TXN; - + perf_enable(); return 0; } diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index ab72f56eb37..243286a8ded 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -564,26 +564,26 @@ struct pmu { struct list_head entry; /* - * Should return -ENOENT when the @event doesn't match this pmu + * Should return -ENOENT when the @event doesn't match this PMU. */ int (*event_init) (struct perf_event *event); - int (*enable) (struct perf_event *event); + int (*enable) (struct perf_event *event); void (*disable) (struct perf_event *event); - int (*start) (struct perf_event *event); + int (*start) (struct perf_event *event); void (*stop) (struct perf_event *event); void (*read) (struct perf_event *event); void (*unthrottle) (struct perf_event *event); /* - * Group events scheduling is treated as a transaction, add group - * events as a whole and perform one schedulability test. If the test - * fails, roll back the whole group + * Group events scheduling is treated as a transaction, add + * group events as a whole and perform one schedulability test. + * If the test fails, roll back the whole group */ /* - * Start the transaction, after this ->enable() doesn't need - * to do schedulability tests. + * Start the transaction, after this ->enable() doesn't need to + * do schedulability tests. */ void (*start_txn) (struct pmu *pmu); /* @@ -594,8 +594,8 @@ struct pmu { */ int (*commit_txn) (struct pmu *pmu); /* - * Will cancel the transaction, assumes ->disable() is called for - * each successfull ->enable() during the transaction. + * Will cancel the transaction, assumes ->disable() is called + * for each successfull ->enable() during the transaction. */ void (*cancel_txn) (struct pmu *pmu); }; diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 149ca18371b..9a98ce95356 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -478,11 +478,6 @@ static void __perf_event_remove_from_context(void *info) return; raw_spin_lock(&ctx->lock); - /* - * Protect the list operation against NMI by disabling the - * events on a global level. - */ - perf_disable(); event_sched_out(event, cpuctx, ctx); @@ -498,7 +493,6 @@ static void __perf_event_remove_from_context(void *info) perf_max_events - perf_reserved_percpu); } - perf_enable(); raw_spin_unlock(&ctx->lock); } @@ -803,12 +797,6 @@ static void __perf_install_in_context(void *info) ctx->is_active = 1; update_context_time(ctx); - /* - * Protect the list operation against NMI by disabling the - * events on a global level. NOP for non NMI based events. - */ - perf_disable(); - add_event_to_ctx(event, ctx); if (event->cpu != -1 && event->cpu != smp_processor_id()) @@ -850,8 +838,6 @@ static void __perf_install_in_context(void *info) cpuctx->max_pertask--; unlock: - perf_enable(); - raw_spin_unlock(&ctx->lock); } @@ -972,12 +958,10 @@ static void __perf_event_enable(void *info) if (!group_can_go_on(event, cpuctx, 1)) { err = -EEXIST; } else { - perf_disable(); if (event == leader) err = group_sched_in(event, cpuctx, ctx); else err = event_sched_in(event, cpuctx, ctx); - perf_enable(); } if (err) { @@ -1090,9 +1074,8 @@ static void ctx_sched_out(struct perf_event_context *ctx, goto out; update_context_time(ctx); - perf_disable(); if (!ctx->nr_active) - goto out_enable; + goto out; if (event_type & EVENT_PINNED) { list_for_each_entry(event, &ctx->pinned_groups, group_entry) @@ -1103,9 +1086,6 @@ static void ctx_sched_out(struct perf_event_context *ctx, list_for_each_entry(event, &ctx->flexible_groups, group_entry) group_sched_out(event, cpuctx, ctx); } - - out_enable: - perf_enable(); out: raw_spin_unlock(&ctx->lock); } @@ -1364,8 +1344,6 @@ ctx_sched_in(struct perf_event_context *ctx, ctx->timestamp = perf_clock(); - perf_disable(); - /* * First go through the list and put on any pinned groups * in order to give them the best chance of going on. @@ -1377,7 +1355,6 @@ ctx_sched_in(struct perf_event_context *ctx, if (event_type & EVENT_FLEXIBLE) ctx_flexible_sched_in(ctx, cpuctx); - perf_enable(); out: raw_spin_unlock(&ctx->lock); } @@ -1425,8 +1402,6 @@ void perf_event_task_sched_in(struct task_struct *task) if (cpuctx->task_ctx == ctx) return; - perf_disable(); - /* * We want to keep the following priority order: * cpu pinned (that don't need to move), task pinned, @@ -1439,8 +1414,6 @@ void perf_event_task_sched_in(struct task_struct *task) ctx_sched_in(ctx, cpuctx, EVENT_FLEXIBLE); cpuctx->task_ctx = ctx; - - perf_enable(); } #define MAX_INTERRUPTS (~0ULL) @@ -1555,11 +1528,9 @@ static void perf_adjust_period(struct perf_event *event, u64 nsec, u64 count) hwc->sample_period = sample_period; if (local64_read(&hwc->period_left) > 8*sample_period) { - perf_disable(); perf_event_stop(event); local64_set(&hwc->period_left, 0); perf_event_start(event); - perf_enable(); } } @@ -1588,15 +1559,12 @@ static void perf_ctx_adjust_freq(struct perf_event_context *ctx) */ if (interrupts == MAX_INTERRUPTS) { perf_log_throttle(event, 1); - perf_disable(); event->pmu->unthrottle(event); - perf_enable(); } if (!event->attr.freq || !event->attr.sample_freq) continue; - perf_disable(); event->pmu->read(event); now = local64_read(&event->count); delta = now - hwc->freq_count_stamp; @@ -1604,7 +1572,6 @@ static void perf_ctx_adjust_freq(struct perf_event_context *ctx) if (delta > 0) perf_adjust_period(event, TICK_NSEC, delta); - perf_enable(); } raw_spin_unlock(&ctx->lock); } @@ -1647,7 +1614,6 @@ void perf_event_task_tick(struct task_struct *curr) if (!rotate) return; - perf_disable(); cpu_ctx_sched_out(cpuctx, EVENT_FLEXIBLE); if (ctx) task_ctx_sched_out(ctx, EVENT_FLEXIBLE); @@ -1659,7 +1625,6 @@ void perf_event_task_tick(struct task_struct *curr) cpu_ctx_sched_in(cpuctx, EVENT_FLEXIBLE); if (ctx) task_ctx_sched_in(curr, EVENT_FLEXIBLE); - perf_enable(); } static int event_enable_on_exec(struct perf_event *event, -- cgit v1.2.3-70-g09d2 From 33696fc0d141bbbcb12f75b69608ea83282e3117 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 14 Jun 2010 08:49:00 +0200 Subject: perf: Per PMU disable Changes perf_disable() into perf_pmu_disable(). Signed-off-by: Peter Zijlstra Cc: paulus Cc: stephane eranian Cc: Robert Richter Cc: Will Deacon Cc: Paul Mundt Cc: Frederic Weisbecker Cc: Cyrill Gorcunov Cc: Lin Ming Cc: Yanmin Cc: Deng-Cheng Zhu Cc: David Miller Cc: Michael Cree LKML-Reference: Signed-off-by: Ingo Molnar --- arch/alpha/kernel/perf_event.c | 30 +++++++++++++------------ arch/arm/kernel/perf_event.c | 28 +++++++++++------------ arch/powerpc/kernel/perf_event.c | 24 +++++++++++--------- arch/powerpc/kernel/perf_event_fsl_emb.c | 18 ++++++++------- arch/sh/kernel/perf_event.c | 38 +++++++++++++++++--------------- arch/sparc/kernel/perf_event.c | 20 +++++++++-------- arch/x86/kernel/cpu/perf_event.c | 16 ++++++++------ include/linux/perf_event.h | 13 ++++++----- kernel/perf_event.c | 31 ++++++++++++++++---------- 9 files changed, 119 insertions(+), 99 deletions(-) (limited to 'arch/x86') diff --git a/arch/alpha/kernel/perf_event.c b/arch/alpha/kernel/perf_event.c index 19660b5c298..3e260731f8e 100644 --- a/arch/alpha/kernel/perf_event.c +++ b/arch/alpha/kernel/perf_event.c @@ -435,7 +435,7 @@ static int alpha_pmu_enable(struct perf_event *event) * nevertheless we disable the PMCs first to enable a potential * final PMI to occur before we disable interrupts. */ - perf_disable(); + perf_pmu_disable(event->pmu); local_irq_save(flags); /* Default to error to be returned */ @@ -456,7 +456,7 @@ static int alpha_pmu_enable(struct perf_event *event) } local_irq_restore(flags); - perf_enable(); + perf_pmu_enable(event->pmu); return ret; } @@ -474,7 +474,7 @@ static void alpha_pmu_disable(struct perf_event *event) unsigned long flags; int j; - perf_disable(); + perf_pmu_disable(event->pmu); local_irq_save(flags); for (j = 0; j < cpuc->n_events; j++) { @@ -502,7 +502,7 @@ static void alpha_pmu_disable(struct perf_event *event) } local_irq_restore(flags); - perf_enable(); + perf_pmu_enable(event->pmu); } @@ -668,18 +668,10 @@ static int alpha_pmu_event_init(struct perf_event *event) return err; } -static struct pmu pmu = { - .event_init = alpha_pmu_event_init, - .enable = alpha_pmu_enable, - .disable = alpha_pmu_disable, - .read = alpha_pmu_read, - .unthrottle = alpha_pmu_unthrottle, -}; - /* * Main entry point - enable HW performance counters. */ -void hw_perf_enable(void) +static void alpha_pmu_pmu_enable(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -705,7 +697,7 @@ void hw_perf_enable(void) * Main entry point - disable HW performance counters. */ -void hw_perf_disable(void) +static void alpha_pmu_pmu_disable(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -718,6 +710,16 @@ void hw_perf_disable(void) wrperfmon(PERFMON_CMD_DISABLE, cpuc->idx_mask); } +static struct pmu pmu = { + .pmu_enable = alpha_pmu_pmu_enable, + .pmu_disable = alpha_pmu_pmu_disable, + .event_init = alpha_pmu_event_init, + .enable = alpha_pmu_enable, + .disable = alpha_pmu_disable, + .read = alpha_pmu_read, + .unthrottle = alpha_pmu_unthrottle, +}; + /* * Main entry point - don't know when this is called but it diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index afc92c580d1..3343f3f4b97 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -277,7 +277,7 @@ armpmu_enable(struct perf_event *event) int idx; int err = 0; - perf_disable(); + perf_pmu_disable(event->pmu); /* If we don't have a space for the counter then finish early. */ idx = armpmu->get_event_idx(cpuc, hwc); @@ -305,7 +305,7 @@ armpmu_enable(struct perf_event *event) perf_event_update_userpage(event); out: - perf_enable(); + perf_pmu_enable(event->pmu); return err; } @@ -534,16 +534,7 @@ static int armpmu_event_init(struct perf_event *event) return err; } -static struct pmu pmu = { - .event_init = armpmu_event_init, - .enable = armpmu_enable, - .disable = armpmu_disable, - .unthrottle = armpmu_unthrottle, - .read = armpmu_read, -}; - -void -hw_perf_enable(void) +static void armpmu_pmu_enable(struct pmu *pmu) { /* Enable all of the perf events on hardware. */ int idx; @@ -564,13 +555,22 @@ hw_perf_enable(void) armpmu->start(); } -void -hw_perf_disable(void) +static void armpmu_pmu_disable(struct pmu *pmu) { if (armpmu) armpmu->stop(); } +static struct pmu pmu = { + .pmu_enable = armpmu_pmu_enable, + .pmu_disable= armpmu_pmu_disable, + .event_init = armpmu_event_init, + .enable = armpmu_enable, + .disable = armpmu_disable, + .unthrottle = armpmu_unthrottle, + .read = armpmu_read, +}; + /* * ARMv6 Performance counter handling code. * diff --git a/arch/powerpc/kernel/perf_event.c b/arch/powerpc/kernel/perf_event.c index c1408821dbc..deb84bbcb0e 100644 --- a/arch/powerpc/kernel/perf_event.c +++ b/arch/powerpc/kernel/perf_event.c @@ -517,7 +517,7 @@ static void write_mmcr0(struct cpu_hw_events *cpuhw, unsigned long mmcr0) * Disable all events to prevent PMU interrupts and to allow * events to be added or removed. */ -void hw_perf_disable(void) +static void power_pmu_pmu_disable(struct pmu *pmu) { struct cpu_hw_events *cpuhw; unsigned long flags; @@ -565,7 +565,7 @@ void hw_perf_disable(void) * If we were previously disabled and events were added, then * put the new config on the PMU. */ -void hw_perf_enable(void) +static void power_pmu_pmu_enable(struct pmu *pmu) { struct perf_event *event; struct cpu_hw_events *cpuhw; @@ -735,7 +735,7 @@ static int power_pmu_enable(struct perf_event *event) int ret = -EAGAIN; local_irq_save(flags); - perf_disable(); + perf_pmu_disable(event->pmu); /* * Add the event to the list (if there is room) @@ -769,7 +769,7 @@ nocheck: ret = 0; out: - perf_enable(); + perf_pmu_enable(event->pmu); local_irq_restore(flags); return ret; } @@ -784,7 +784,7 @@ static void power_pmu_disable(struct perf_event *event) unsigned long flags; local_irq_save(flags); - perf_disable(); + perf_pmu_disable(event->pmu); power_pmu_read(event); @@ -821,7 +821,7 @@ static void power_pmu_disable(struct perf_event *event) cpuhw->mmcr[0] &= ~(MMCR0_PMXE | MMCR0_FCECE); } - perf_enable(); + perf_pmu_enable(event->pmu); local_irq_restore(flags); } @@ -837,7 +837,7 @@ static void power_pmu_unthrottle(struct perf_event *event) if (!event->hw.idx || !event->hw.sample_period) return; local_irq_save(flags); - perf_disable(); + perf_pmu_disable(event->pmu); power_pmu_read(event); left = event->hw.sample_period; event->hw.last_period = left; @@ -848,7 +848,7 @@ static void power_pmu_unthrottle(struct perf_event *event) local64_set(&event->hw.prev_count, val); local64_set(&event->hw.period_left, left); perf_event_update_userpage(event); - perf_enable(); + perf_pmu_enable(event->pmu); local_irq_restore(flags); } @@ -861,7 +861,7 @@ void power_pmu_start_txn(struct pmu *pmu) { struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events); - perf_disable(); + perf_pmu_disable(pmu); cpuhw->group_flag |= PERF_EVENT_TXN; cpuhw->n_txn_start = cpuhw->n_events; } @@ -876,7 +876,7 @@ void power_pmu_cancel_txn(struct pmu *pmu) struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events); cpuhw->group_flag &= ~PERF_EVENT_TXN; - perf_enable(); + perf_pmu_enable(pmu); } /* @@ -903,7 +903,7 @@ int power_pmu_commit_txn(struct pmu *pmu) cpuhw->event[i]->hw.config = cpuhw->events[i]; cpuhw->group_flag &= ~PERF_EVENT_TXN; - perf_enable(); + perf_pmu_enable(pmu); return 0; } @@ -1131,6 +1131,8 @@ static int power_pmu_event_init(struct perf_event *event) } struct pmu power_pmu = { + .pmu_enable = power_pmu_pmu_enable, + .pmu_disable = power_pmu_pmu_disable, .event_init = power_pmu_event_init, .enable = power_pmu_enable, .disable = power_pmu_disable, diff --git a/arch/powerpc/kernel/perf_event_fsl_emb.c b/arch/powerpc/kernel/perf_event_fsl_emb.c index 9bc84a7fd90..84b1974c628 100644 --- a/arch/powerpc/kernel/perf_event_fsl_emb.c +++ b/arch/powerpc/kernel/perf_event_fsl_emb.c @@ -177,7 +177,7 @@ static void fsl_emb_pmu_read(struct perf_event *event) * Disable all events to prevent PMU interrupts and to allow * events to be added or removed. */ -void hw_perf_disable(void) +static void fsl_emb_pmu_pmu_disable(struct pmu *pmu) { struct cpu_hw_events *cpuhw; unsigned long flags; @@ -216,7 +216,7 @@ void hw_perf_disable(void) * If we were previously disabled and events were added, then * put the new config on the PMU. */ -void hw_perf_enable(void) +static void fsl_emb_pmu_pmu_enable(struct pmu *pmu) { struct cpu_hw_events *cpuhw; unsigned long flags; @@ -271,7 +271,7 @@ static int fsl_emb_pmu_enable(struct perf_event *event) u64 val; int i; - perf_disable(); + perf_pmu_disable(event->pmu); cpuhw = &get_cpu_var(cpu_hw_events); if (event->hw.config & FSL_EMB_EVENT_RESTRICTED) @@ -311,7 +311,7 @@ static int fsl_emb_pmu_enable(struct perf_event *event) ret = 0; out: put_cpu_var(cpu_hw_events); - perf_enable(); + perf_pmu_enable(event->pmu); return ret; } @@ -321,7 +321,7 @@ static void fsl_emb_pmu_disable(struct perf_event *event) struct cpu_hw_events *cpuhw; int i = event->hw.idx; - perf_disable(); + perf_pmu_disable(event->pmu); if (i < 0) goto out; @@ -349,7 +349,7 @@ static void fsl_emb_pmu_disable(struct perf_event *event) cpuhw->n_events--; out: - perf_enable(); + perf_pmu_enable(event->pmu); put_cpu_var(cpu_hw_events); } @@ -367,7 +367,7 @@ static void fsl_emb_pmu_unthrottle(struct perf_event *event) if (event->hw.idx < 0 || !event->hw.sample_period) return; local_irq_save(flags); - perf_disable(); + perf_pmu_disable(event->pmu); fsl_emb_pmu_read(event); left = event->hw.sample_period; event->hw.last_period = left; @@ -378,7 +378,7 @@ static void fsl_emb_pmu_unthrottle(struct perf_event *event) local64_set(&event->hw.prev_count, val); local64_set(&event->hw.period_left, left); perf_event_update_userpage(event); - perf_enable(); + perf_pmu_enable(event->pmu); local_irq_restore(flags); } @@ -524,6 +524,8 @@ static int fsl_emb_pmu_event_init(struct perf_event *event) } static struct pmu fsl_emb_pmu = { + .pmu_enable = fsl_emb_pmu_pmu_enable, + .pmu_disable = fsl_emb_pmu_pmu_disable, .event_init = fsl_emb_pmu_event_init, .enable = fsl_emb_pmu_enable, .disable = fsl_emb_pmu_disable, diff --git a/arch/sh/kernel/perf_event.c b/arch/sh/kernel/perf_event.c index d042989ceb4..4bbe19058a5 100644 --- a/arch/sh/kernel/perf_event.c +++ b/arch/sh/kernel/perf_event.c @@ -232,7 +232,7 @@ static int sh_pmu_enable(struct perf_event *event) int idx = hwc->idx; int ret = -EAGAIN; - perf_disable(); + perf_pmu_disable(event->pmu); if (test_and_set_bit(idx, cpuc->used_mask)) { idx = find_first_zero_bit(cpuc->used_mask, sh_pmu->num_events); @@ -253,7 +253,7 @@ static int sh_pmu_enable(struct perf_event *event) perf_event_update_userpage(event); ret = 0; out: - perf_enable(); + perf_pmu_enable(event->pmu); return ret; } @@ -285,7 +285,25 @@ static int sh_pmu_event_init(struct perf_event *event) return err; } +static void sh_pmu_pmu_enable(struct pmu *pmu) +{ + if (!sh_pmu_initialized()) + return; + + sh_pmu->enable_all(); +} + +static void sh_pmu_pmu_disable(struct pmu *pmu) +{ + if (!sh_pmu_initialized()) + return; + + sh_pmu->disable_all(); +} + static struct pmu pmu = { + .pmu_enable = sh_pmu_pmu_enable, + .pmu_disable = sh_pmu_pmu_disable, .event_init = sh_pmu_event_init, .enable = sh_pmu_enable, .disable = sh_pmu_disable, @@ -316,22 +334,6 @@ sh_pmu_notifier(struct notifier_block *self, unsigned long action, void *hcpu) return NOTIFY_OK; } -void hw_perf_enable(void) -{ - if (!sh_pmu_initialized()) - return; - - sh_pmu->enable_all(); -} - -void hw_perf_disable(void) -{ - if (!sh_pmu_initialized()) - return; - - sh_pmu->disable_all(); -} - int __cpuinit register_sh_pmu(struct sh_pmu *pmu) { if (sh_pmu) diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c index d0131deeeaf..37cae676536 100644 --- a/arch/sparc/kernel/perf_event.c +++ b/arch/sparc/kernel/perf_event.c @@ -664,7 +664,7 @@ out: return pcr; } -void hw_perf_enable(void) +static void sparc_pmu_pmu_enable(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); u64 pcr; @@ -691,7 +691,7 @@ void hw_perf_enable(void) pcr_ops->write(cpuc->pcr); } -void hw_perf_disable(void) +static void sparc_pmu_pmu_disable(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); u64 val; @@ -718,7 +718,7 @@ static void sparc_pmu_disable(struct perf_event *event) int i; local_irq_save(flags); - perf_disable(); + perf_pmu_disable(event->pmu); for (i = 0; i < cpuc->n_events; i++) { if (event == cpuc->event[i]) { @@ -748,7 +748,7 @@ static void sparc_pmu_disable(struct perf_event *event) } } - perf_enable(); + perf_pmu_enable(event->pmu); local_irq_restore(flags); } @@ -991,7 +991,7 @@ static int sparc_pmu_enable(struct perf_event *event) unsigned long flags; local_irq_save(flags); - perf_disable(); + perf_pmu_disable(event->pmu); n0 = cpuc->n_events; if (n0 >= perf_max_events) @@ -1020,7 +1020,7 @@ nocheck: ret = 0; out: - perf_enable(); + perf_pmu_enable(event->pmu); local_irq_restore(flags); return ret; } @@ -1113,7 +1113,7 @@ static void sparc_pmu_start_txn(struct pmu *pmu) { struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events); - perf_disable(); + perf_pmu_disable(pmu); cpuhw->group_flag |= PERF_EVENT_TXN; } @@ -1127,7 +1127,7 @@ static void sparc_pmu_cancel_txn(struct pmu *pmu) struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events); cpuhw->group_flag &= ~PERF_EVENT_TXN; - perf_enable(); + perf_pmu_enable(pmu); } /* @@ -1151,11 +1151,13 @@ static int sparc_pmu_commit_txn(struct pmu *pmu) return -EAGAIN; cpuc->group_flag &= ~PERF_EVENT_TXN; - perf_enable(); + perf_pmu_enable(pmu); return 0; } static struct pmu pmu = { + .pmu_enable = sparc_pmu_pmu_enable, + .pmu_disable = sparc_pmu_pmu_disable, .event_init = sparc_pmu_event_init, .enable = sparc_pmu_enable, .disable = sparc_pmu_disable, diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 846070ce49c..79705ac4501 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -583,7 +583,7 @@ static void x86_pmu_disable_all(void) } } -void hw_perf_disable(void) +static void x86_pmu_pmu_disable(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -803,7 +803,7 @@ static inline int match_prev_assignment(struct hw_perf_event *hwc, static int x86_pmu_start(struct perf_event *event); static void x86_pmu_stop(struct perf_event *event); -void hw_perf_enable(void) +static void x86_pmu_pmu_enable(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct perf_event *event; @@ -969,7 +969,7 @@ static int x86_pmu_enable(struct perf_event *event) hwc = &event->hw; - perf_disable(); + perf_pmu_disable(event->pmu); n0 = cpuc->n_events; ret = n = collect_events(cpuc, event, false); if (ret < 0) @@ -999,7 +999,7 @@ done_collect: ret = 0; out: - perf_enable(); + perf_pmu_enable(event->pmu); return ret; } @@ -1436,7 +1436,7 @@ static void x86_pmu_start_txn(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - perf_disable(); + perf_pmu_disable(pmu); cpuc->group_flag |= PERF_EVENT_TXN; cpuc->n_txn = 0; } @@ -1456,7 +1456,7 @@ static void x86_pmu_cancel_txn(struct pmu *pmu) */ cpuc->n_added -= cpuc->n_txn; cpuc->n_events -= cpuc->n_txn; - perf_enable(); + perf_pmu_enable(pmu); } /* @@ -1486,7 +1486,7 @@ static int x86_pmu_commit_txn(struct pmu *pmu) memcpy(cpuc->assign, assign, n*sizeof(int)); cpuc->group_flag &= ~PERF_EVENT_TXN; - perf_enable(); + perf_pmu_enable(pmu); return 0; } @@ -1605,6 +1605,8 @@ int x86_pmu_event_init(struct perf_event *event) } static struct pmu pmu = { + .pmu_enable = x86_pmu_pmu_enable, + .pmu_disable = x86_pmu_pmu_disable, .event_init = x86_pmu_event_init, .enable = x86_pmu_enable, .disable = x86_pmu_disable, diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 243286a8ded..6abf103fb7f 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -563,6 +563,11 @@ struct perf_event; struct pmu { struct list_head entry; + int *pmu_disable_count; + + void (*pmu_enable) (struct pmu *pmu); + void (*pmu_disable) (struct pmu *pmu); + /* * Should return -ENOENT when the @event doesn't match this PMU. */ @@ -868,10 +873,8 @@ extern void perf_event_free_task(struct task_struct *task); extern void set_perf_event_pending(void); extern void perf_event_do_pending(void); extern void perf_event_print_debug(void); -extern void __perf_disable(void); -extern bool __perf_enable(void); -extern void perf_disable(void); -extern void perf_enable(void); +extern void perf_pmu_disable(struct pmu *pmu); +extern void perf_pmu_enable(struct pmu *pmu); extern int perf_event_task_disable(void); extern int perf_event_task_enable(void); extern void perf_event_update_userpage(struct perf_event *event); @@ -1056,8 +1059,6 @@ static inline void perf_event_exit_task(struct task_struct *child) { } static inline void perf_event_free_task(struct task_struct *task) { } static inline void perf_event_do_pending(void) { } static inline void perf_event_print_debug(void) { } -static inline void perf_disable(void) { } -static inline void perf_enable(void) { } static inline int perf_event_task_disable(void) { return -EINVAL; } static inline int perf_event_task_enable(void) { return -EINVAL; } diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 9a98ce95356..5ed0c06765b 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -71,23 +71,20 @@ static atomic64_t perf_event_id; */ static DEFINE_SPINLOCK(perf_resource_lock); -void __weak hw_perf_disable(void) { barrier(); } -void __weak hw_perf_enable(void) { barrier(); } - void __weak perf_event_print_debug(void) { } -static DEFINE_PER_CPU(int, perf_disable_count); - -void perf_disable(void) +void perf_pmu_disable(struct pmu *pmu) { - if (!__get_cpu_var(perf_disable_count)++) - hw_perf_disable(); + int *count = this_cpu_ptr(pmu->pmu_disable_count); + if (!(*count)++) + pmu->pmu_disable(pmu); } -void perf_enable(void) +void perf_pmu_enable(struct pmu *pmu) { - if (!--__get_cpu_var(perf_disable_count)) - hw_perf_enable(); + int *count = this_cpu_ptr(pmu->pmu_disable_count); + if (!--(*count)) + pmu->pmu_enable(pmu); } static void get_ctx(struct perf_event_context *ctx) @@ -4970,11 +4967,19 @@ static struct srcu_struct pmus_srcu; int perf_pmu_register(struct pmu *pmu) { + int ret; + mutex_lock(&pmus_lock); + ret = -ENOMEM; + pmu->pmu_disable_count = alloc_percpu(int); + if (!pmu->pmu_disable_count) + goto unlock; list_add_rcu(&pmu->entry, &pmus); + ret = 0; +unlock: mutex_unlock(&pmus_lock); - return 0; + return ret; } void perf_pmu_unregister(struct pmu *pmu) @@ -4984,6 +4989,8 @@ void perf_pmu_unregister(struct pmu *pmu) mutex_unlock(&pmus_lock); synchronize_srcu(&pmus_srcu); + + free_percpu(pmu->pmu_disable_count); } struct pmu *perf_init_event(struct perf_event *event) -- cgit v1.2.3-70-g09d2 From a4eaf7f14675cb512d69f0c928055e73d0c6d252 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 16 Jun 2010 14:37:10 +0200 Subject: perf: Rework the PMU methods Replace pmu::{enable,disable,start,stop,unthrottle} with pmu::{add,del,start,stop}, all of which take a flags argument. The new interface extends the capability to stop a counter while keeping it scheduled on the PMU. We replace the throttled state with the generic stopped state. This also allows us to efficiently stop/start counters over certain code paths (like IRQ handlers). It also allows scheduling a counter without it starting, allowing for a generic frozen state (useful for rotating stopped counters). The stopped state is implemented in two different ways, depending on how the architecture implemented the throttled state: 1) We disable the counter: a) the pmu has per-counter enable bits, we flip that b) we program a NOP event, preserving the counter state 2) We store the counter state and ignore all read/overflow events Signed-off-by: Peter Zijlstra Cc: paulus Cc: stephane eranian Cc: Robert Richter Cc: Will Deacon Cc: Paul Mundt Cc: Frederic Weisbecker Cc: Cyrill Gorcunov Cc: Lin Ming Cc: Yanmin Cc: Deng-Cheng Zhu Cc: David Miller Cc: Michael Cree LKML-Reference: Signed-off-by: Ingo Molnar --- arch/alpha/kernel/perf_event.c | 71 +++++++++++---- arch/arm/kernel/perf_event.c | 96 ++++++++++++-------- arch/powerpc/kernel/perf_event.c | 105 ++++++++++++++-------- arch/powerpc/kernel/perf_event_fsl_emb.c | 107 ++++++++++++++--------- arch/sh/kernel/perf_event.c | 75 +++++++++++----- arch/sparc/kernel/perf_event.c | 109 ++++++++++++++--------- arch/x86/kernel/cpu/perf_event.c | 106 ++++++++++++---------- arch/x86/kernel/cpu/perf_event_intel.c | 2 +- arch/x86/kernel/cpu/perf_event_intel_ds.c | 2 +- include/linux/ftrace_event.h | 4 +- include/linux/perf_event.h | 54 +++++++++--- kernel/hw_breakpoint.c | 29 ++++++- kernel/perf_event.c | 140 +++++++++++++++--------------- kernel/trace/trace_event_perf.c | 7 +- 14 files changed, 576 insertions(+), 331 deletions(-) (limited to 'arch/x86') diff --git a/arch/alpha/kernel/perf_event.c b/arch/alpha/kernel/perf_event.c index 3e260731f8e..380ef02d557 100644 --- a/arch/alpha/kernel/perf_event.c +++ b/arch/alpha/kernel/perf_event.c @@ -307,7 +307,7 @@ again: new_raw_count) != prev_raw_count) goto again; - delta = (new_raw_count - (prev_raw_count & alpha_pmu->pmc_count_mask[idx])) + ovf; + delta = (new_raw_count - (prev_raw_count & alpha_pmu->pmc_count_mask[idx])) + ovf; /* It is possible on very rare occasions that the PMC has overflowed * but the interrupt is yet to come. Detect and fix this situation. @@ -402,14 +402,13 @@ static void maybe_change_configuration(struct cpu_hw_events *cpuc) struct hw_perf_event *hwc = &pe->hw; int idx = hwc->idx; - if (cpuc->current_idx[j] != PMC_NO_INDEX) { - cpuc->idx_mask |= (1<current_idx[j]); - continue; + if (cpuc->current_idx[j] == PMC_NO_INDEX) { + alpha_perf_event_set_period(pe, hwc, idx); + cpuc->current_idx[j] = idx; } - alpha_perf_event_set_period(pe, hwc, idx); - cpuc->current_idx[j] = idx; - cpuc->idx_mask |= (1<current_idx[j]); + if (!(hwc->state & PERF_HES_STOPPED)) + cpuc->idx_mask |= (1<current_idx[j]); } cpuc->config = cpuc->event[0]->hw.config_base; } @@ -420,7 +419,7 @@ static void maybe_change_configuration(struct cpu_hw_events *cpuc) * - this function is called from outside this module via the pmu struct * returned from perf event initialisation. */ -static int alpha_pmu_enable(struct perf_event *event) +static int alpha_pmu_add(struct perf_event *event, int flags) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); int n0; @@ -455,6 +454,10 @@ static int alpha_pmu_enable(struct perf_event *event) } } + hwc->state = PERF_HES_UPTODATE; + if (!(flags & PERF_EF_START)) + hwc->state |= PERF_HES_STOPPED; + local_irq_restore(flags); perf_pmu_enable(event->pmu); @@ -467,7 +470,7 @@ static int alpha_pmu_enable(struct perf_event *event) * - this function is called from outside this module via the pmu struct * returned from perf event initialisation. */ -static void alpha_pmu_disable(struct perf_event *event) +static void alpha_pmu_del(struct perf_event *event, int flags) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct hw_perf_event *hwc = &event->hw; @@ -514,13 +517,44 @@ static void alpha_pmu_read(struct perf_event *event) } -static void alpha_pmu_unthrottle(struct perf_event *event) +static void alpha_pmu_stop(struct perf_event *event, int flags) { struct hw_perf_event *hwc = &event->hw; struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + if (!(hwc->state & PERF_HES_STOPPED)) { + cpuc->idx_mask &= !(1UL<idx); + hwc->state |= PERF_HES_STOPPED; + } + + if ((flags & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) { + alpha_perf_event_update(event, hwc, hwc->idx, 0); + hwc->state |= PERF_HES_UPTODATE; + } + + if (cpuc->enabled) + wrperfmon(PERFMON_CMD_ENABLE, (1UL<idx)); +} + + +static void alpha_pmu_start(struct perf_event *event, int flags) +{ + struct hw_perf_event *hwc = &event->hw; + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + + if (WARN_ON_ONCE(!(hwc->state & PERF_HES_STOPPED))) + return; + + if (flags & PERF_EF_RELOAD) { + WARN_ON_ONCE(!(hwc->state & PERF_HES_UPTODATE)); + alpha_perf_event_set_period(event, hwc, hwc->idx); + } + + hwc->state = 0; + cpuc->idx_mask |= 1UL<idx; - wrperfmon(PERFMON_CMD_ENABLE, (1UL<idx)); + if (cpuc->enabled) + wrperfmon(PERFMON_CMD_ENABLE, (1UL<idx)); } @@ -671,7 +705,7 @@ static int alpha_pmu_event_init(struct perf_event *event) /* * Main entry point - enable HW performance counters. */ -static void alpha_pmu_pmu_enable(struct pmu *pmu) +static void alpha_pmu_enable(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -697,7 +731,7 @@ static void alpha_pmu_pmu_enable(struct pmu *pmu) * Main entry point - disable HW performance counters. */ -static void alpha_pmu_pmu_disable(struct pmu *pmu) +static void alpha_pmu_disable(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -711,13 +745,14 @@ static void alpha_pmu_pmu_disable(struct pmu *pmu) } static struct pmu pmu = { - .pmu_enable = alpha_pmu_pmu_enable, - .pmu_disable = alpha_pmu_pmu_disable, + .pmu_enable = alpha_pmu_enable, + .pmu_disable = alpha_pmu_disable, .event_init = alpha_pmu_event_init, - .enable = alpha_pmu_enable, - .disable = alpha_pmu_disable, + .add = alpha_pmu_add, + .del = alpha_pmu_del, + .start = alpha_pmu_start, + .stop = alpha_pmu_stop, .read = alpha_pmu_read, - .unthrottle = alpha_pmu_unthrottle, }; diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index 3343f3f4b97..448cfa6b3ef 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -221,46 +221,56 @@ again: } static void -armpmu_disable(struct perf_event *event) +armpmu_read(struct perf_event *event) { - struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct hw_perf_event *hwc = &event->hw; - int idx = hwc->idx; - - WARN_ON(idx < 0); - - clear_bit(idx, cpuc->active_mask); - armpmu->disable(hwc, idx); - - barrier(); - armpmu_event_update(event, hwc, idx); - cpuc->events[idx] = NULL; - clear_bit(idx, cpuc->used_mask); + /* Don't read disabled counters! */ + if (hwc->idx < 0) + return; - perf_event_update_userpage(event); + armpmu_event_update(event, hwc, hwc->idx); } static void -armpmu_read(struct perf_event *event) +armpmu_stop(struct perf_event *event, int flags) { struct hw_perf_event *hwc = &event->hw; - /* Don't read disabled counters! */ - if (hwc->idx < 0) + if (!armpmu) return; - armpmu_event_update(event, hwc, hwc->idx); + /* + * ARM pmu always has to update the counter, so ignore + * PERF_EF_UPDATE, see comments in armpmu_start(). + */ + if (!(hwc->state & PERF_HES_STOPPED)) { + armpmu->disable(hwc, hwc->idx); + barrier(); /* why? */ + armpmu_event_update(event, hwc, hwc->idx); + hwc->state |= PERF_HES_STOPPED | PERF_HES_UPTODATE; + } } static void -armpmu_unthrottle(struct perf_event *event) +armpmu_start(struct perf_event *event, int flags) { struct hw_perf_event *hwc = &event->hw; + if (!armpmu) + return; + + /* + * ARM pmu always has to reprogram the period, so ignore + * PERF_EF_RELOAD, see the comment below. + */ + if (flags & PERF_EF_RELOAD) + WARN_ON_ONCE(!(hwc->state & PERF_HES_UPTODATE)); + + hwc->state = 0; /* * Set the period again. Some counters can't be stopped, so when we - * were throttled we simply disabled the IRQ source and the counter + * were stopped we simply disabled the IRQ source and the counter * may have been left counting. If we don't do this step then we may * get an interrupt too soon or *way* too late if the overflow has * happened since disabling. @@ -269,8 +279,25 @@ armpmu_unthrottle(struct perf_event *event) armpmu->enable(hwc, hwc->idx); } +static void +armpmu_del(struct perf_event *event, int flags) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + struct hw_perf_event *hwc = &event->hw; + int idx = hwc->idx; + + WARN_ON(idx < 0); + + clear_bit(idx, cpuc->active_mask); + armpmu_stop(event, PERF_EF_UPDATE); + cpuc->events[idx] = NULL; + clear_bit(idx, cpuc->used_mask); + + perf_event_update_userpage(event); +} + static int -armpmu_enable(struct perf_event *event) +armpmu_add(struct perf_event *event, int flags) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct hw_perf_event *hwc = &event->hw; @@ -295,11 +322,9 @@ armpmu_enable(struct perf_event *event) cpuc->events[idx] = event; set_bit(idx, cpuc->active_mask); - /* Set the period for the event. */ - armpmu_event_set_period(event, hwc, idx); - - /* Enable the event. */ - armpmu->enable(hwc, idx); + hwc->state = PERF_HES_STOPPED | PERF_HES_UPTODATE; + if (flags & PERF_EF_START) + armpmu_start(event, PERF_EF_RELOAD); /* Propagate our changes to the userspace mapping. */ perf_event_update_userpage(event); @@ -534,7 +559,7 @@ static int armpmu_event_init(struct perf_event *event) return err; } -static void armpmu_pmu_enable(struct pmu *pmu) +static void armpmu_enable(struct pmu *pmu) { /* Enable all of the perf events on hardware. */ int idx; @@ -555,20 +580,21 @@ static void armpmu_pmu_enable(struct pmu *pmu) armpmu->start(); } -static void armpmu_pmu_disable(struct pmu *pmu) +static void armpmu_disable(struct pmu *pmu) { if (armpmu) armpmu->stop(); } static struct pmu pmu = { - .pmu_enable = armpmu_pmu_enable, - .pmu_disable= armpmu_pmu_disable, - .event_init = armpmu_event_init, - .enable = armpmu_enable, - .disable = armpmu_disable, - .unthrottle = armpmu_unthrottle, - .read = armpmu_read, + .pmu_enable = armpmu_enable, + .pmu_disable = armpmu_disable, + .event_init = armpmu_event_init, + .add = armpmu_add, + .del = armpmu_del, + .start = armpmu_start, + .stop = armpmu_stop, + .read = armpmu_read, }; /* diff --git a/arch/powerpc/kernel/perf_event.c b/arch/powerpc/kernel/perf_event.c index deb84bbcb0e..9cb4924b6c0 100644 --- a/arch/powerpc/kernel/perf_event.c +++ b/arch/powerpc/kernel/perf_event.c @@ -402,6 +402,9 @@ static void power_pmu_read(struct perf_event *event) { s64 val, delta, prev; + if (event->hw.state & PERF_HES_STOPPED) + return; + if (!event->hw.idx) return; /* @@ -517,7 +520,7 @@ static void write_mmcr0(struct cpu_hw_events *cpuhw, unsigned long mmcr0) * Disable all events to prevent PMU interrupts and to allow * events to be added or removed. */ -static void power_pmu_pmu_disable(struct pmu *pmu) +static void power_pmu_disable(struct pmu *pmu) { struct cpu_hw_events *cpuhw; unsigned long flags; @@ -565,7 +568,7 @@ static void power_pmu_pmu_disable(struct pmu *pmu) * If we were previously disabled and events were added, then * put the new config on the PMU. */ -static void power_pmu_pmu_enable(struct pmu *pmu) +static void power_pmu_enable(struct pmu *pmu) { struct perf_event *event; struct cpu_hw_events *cpuhw; @@ -672,6 +675,8 @@ static void power_pmu_pmu_enable(struct pmu *pmu) } local64_set(&event->hw.prev_count, val); event->hw.idx = idx; + if (event->hw.state & PERF_HES_STOPPED) + val = 0; write_pmc(idx, val); perf_event_update_userpage(event); } @@ -727,7 +732,7 @@ static int collect_events(struct perf_event *group, int max_count, * re-enable the PMU in order to get hw_perf_enable to do the * actual work of reconfiguring the PMU. */ -static int power_pmu_enable(struct perf_event *event) +static int power_pmu_add(struct perf_event *event, int ef_flags) { struct cpu_hw_events *cpuhw; unsigned long flags; @@ -749,6 +754,9 @@ static int power_pmu_enable(struct perf_event *event) cpuhw->events[n0] = event->hw.config; cpuhw->flags[n0] = event->hw.event_base; + if (!(ef_flags & PERF_EF_START)) + event->hw.state = PERF_HES_STOPPED | PERF_HES_UPTODATE; + /* * If group events scheduling transaction was started, * skip the schedulability test here, it will be peformed @@ -777,7 +785,7 @@ nocheck: /* * Remove a event from the PMU. */ -static void power_pmu_disable(struct perf_event *event) +static void power_pmu_del(struct perf_event *event, int ef_flags) { struct cpu_hw_events *cpuhw; long i; @@ -826,27 +834,53 @@ static void power_pmu_disable(struct perf_event *event) } /* - * Re-enable interrupts on a event after they were throttled - * because they were coming too fast. + * POWER-PMU does not support disabling individual counters, hence + * program their cycle counter to their max value and ignore the interrupts. */ -static void power_pmu_unthrottle(struct perf_event *event) + +static void power_pmu_start(struct perf_event *event, int ef_flags) { - s64 val, left; unsigned long flags; + s64 left; if (!event->hw.idx || !event->hw.sample_period) return; + + if (!(event->hw.state & PERF_HES_STOPPED)) + return; + + if (ef_flags & PERF_EF_RELOAD) + WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE)); + + local_irq_save(flags); + perf_pmu_disable(event->pmu); + + event->hw.state = 0; + left = local64_read(&event->hw.period_left); + write_pmc(event->hw.idx, left); + + perf_event_update_userpage(event); + perf_pmu_enable(event->pmu); + local_irq_restore(flags); +} + +static void power_pmu_stop(struct perf_event *event, int ef_flags) +{ + unsigned long flags; + + if (!event->hw.idx || !event->hw.sample_period) + return; + + if (event->hw.state & PERF_HES_STOPPED) + return; + local_irq_save(flags); perf_pmu_disable(event->pmu); + power_pmu_read(event); - left = event->hw.sample_period; - event->hw.last_period = left; - val = 0; - if (left < 0x80000000L) - val = 0x80000000L - left; - write_pmc(event->hw.idx, val); - local64_set(&event->hw.prev_count, val); - local64_set(&event->hw.period_left, left); + event->hw.state |= PERF_HES_STOPPED | PERF_HES_UPTODATE; + write_pmc(event->hw.idx, 0); + perf_event_update_userpage(event); perf_pmu_enable(event->pmu); local_irq_restore(flags); @@ -1131,13 +1165,14 @@ static int power_pmu_event_init(struct perf_event *event) } struct pmu power_pmu = { - .pmu_enable = power_pmu_pmu_enable, - .pmu_disable = power_pmu_pmu_disable, + .pmu_enable = power_pmu_enable, + .pmu_disable = power_pmu_disable, .event_init = power_pmu_event_init, - .enable = power_pmu_enable, - .disable = power_pmu_disable, + .add = power_pmu_add, + .del = power_pmu_del, + .start = power_pmu_start, + .stop = power_pmu_stop, .read = power_pmu_read, - .unthrottle = power_pmu_unthrottle, .start_txn = power_pmu_start_txn, .cancel_txn = power_pmu_cancel_txn, .commit_txn = power_pmu_commit_txn, @@ -1155,6 +1190,11 @@ static void record_and_restart(struct perf_event *event, unsigned long val, s64 prev, delta, left; int record = 0; + if (event->hw.state & PERF_HES_STOPPED) { + write_pmc(event->hw.idx, 0); + return; + } + /* we don't have to worry about interrupts here */ prev = local64_read(&event->hw.prev_count); delta = (val - prev) & 0xfffffffful; @@ -1177,6 +1217,11 @@ static void record_and_restart(struct perf_event *event, unsigned long val, val = 0x80000000LL - left; } + write_pmc(event->hw.idx, val); + local64_set(&event->hw.prev_count, val); + local64_set(&event->hw.period_left, left); + perf_event_update_userpage(event); + /* * Finally record data if requested. */ @@ -1189,23 +1234,9 @@ static void record_and_restart(struct perf_event *event, unsigned long val, if (event->attr.sample_type & PERF_SAMPLE_ADDR) perf_get_data_addr(regs, &data.addr); - if (perf_event_overflow(event, nmi, &data, regs)) { - /* - * Interrupts are coming too fast - throttle them - * by setting the event to 0, so it will be - * at least 2^30 cycles until the next interrupt - * (assuming each event counts at most 2 counts - * per cycle). - */ - val = 0; - left = ~0ULL >> 1; - } + if (perf_event_overflow(event, nmi, &data, regs)) + power_pmu_stop(event, 0); } - - write_pmc(event->hw.idx, val); - local64_set(&event->hw.prev_count, val); - local64_set(&event->hw.period_left, left); - perf_event_update_userpage(event); } /* diff --git a/arch/powerpc/kernel/perf_event_fsl_emb.c b/arch/powerpc/kernel/perf_event_fsl_emb.c index 84b1974c628..7ecca59ddf7 100644 --- a/arch/powerpc/kernel/perf_event_fsl_emb.c +++ b/arch/powerpc/kernel/perf_event_fsl_emb.c @@ -156,6 +156,9 @@ static void fsl_emb_pmu_read(struct perf_event *event) { s64 val, delta, prev; + if (event->hw.state & PERF_HES_STOPPED) + return; + /* * Performance monitor interrupts come even when interrupts * are soft-disabled, as long as interrupts are hard-enabled. @@ -177,7 +180,7 @@ static void fsl_emb_pmu_read(struct perf_event *event) * Disable all events to prevent PMU interrupts and to allow * events to be added or removed. */ -static void fsl_emb_pmu_pmu_disable(struct pmu *pmu) +static void fsl_emb_pmu_disable(struct pmu *pmu) { struct cpu_hw_events *cpuhw; unsigned long flags; @@ -216,7 +219,7 @@ static void fsl_emb_pmu_pmu_disable(struct pmu *pmu) * If we were previously disabled and events were added, then * put the new config on the PMU. */ -static void fsl_emb_pmu_pmu_enable(struct pmu *pmu) +static void fsl_emb_pmu_enable(struct pmu *pmu) { struct cpu_hw_events *cpuhw; unsigned long flags; @@ -263,7 +266,7 @@ static int collect_events(struct perf_event *group, int max_count, } /* context locked on entry */ -static int fsl_emb_pmu_enable(struct perf_event *event) +static int fsl_emb_pmu_add(struct perf_event *event, int flags) { struct cpu_hw_events *cpuhw; int ret = -EAGAIN; @@ -302,6 +305,12 @@ static int fsl_emb_pmu_enable(struct perf_event *event) val = 0x80000000L - left; } local64_set(&event->hw.prev_count, val); + + if (!(flags & PERF_EF_START)) { + event->hw.state = PERF_HES_STOPPED | PERF_HES_UPTODATE; + val = 0; + } + write_pmc(i, val); perf_event_update_userpage(event); @@ -316,7 +325,7 @@ static int fsl_emb_pmu_enable(struct perf_event *event) } /* context locked on entry */ -static void fsl_emb_pmu_disable(struct perf_event *event) +static void fsl_emb_pmu_del(struct perf_event *event, int flags) { struct cpu_hw_events *cpuhw; int i = event->hw.idx; @@ -353,30 +362,49 @@ static void fsl_emb_pmu_disable(struct perf_event *event) put_cpu_var(cpu_hw_events); } -/* - * Re-enable interrupts on a event after they were throttled - * because they were coming too fast. - * - * Context is locked on entry, but perf is not disabled. - */ -static void fsl_emb_pmu_unthrottle(struct perf_event *event) +static void fsl_emb_pmu_start(struct perf_event *event, int ef_flags) +{ + unsigned long flags; + s64 left; + + if (event->hw.idx < 0 || !event->hw.sample_period) + return; + + if (!(event->hw.state & PERF_HES_STOPPED)) + return; + + if (ef_flags & PERF_EF_RELOAD) + WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE)); + + local_irq_save(flags); + perf_pmu_disable(event->pmu); + + event->hw.state = 0; + left = local64_read(&event->hw.period_left); + write_pmc(event->hw.idx, left); + + perf_event_update_userpage(event); + perf_pmu_enable(event->pmu); + local_irq_restore(flags); +} + +static void fsl_emb_pmu_stop(struct perf_event *event, int ef_flags) { - s64 val, left; unsigned long flags; if (event->hw.idx < 0 || !event->hw.sample_period) return; + + if (event->hw.state & PERF_HES_STOPPED) + return; + local_irq_save(flags); perf_pmu_disable(event->pmu); + fsl_emb_pmu_read(event); - left = event->hw.sample_period; - event->hw.last_period = left; - val = 0; - if (left < 0x80000000L) - val = 0x80000000L - left; - write_pmc(event->hw.idx, val); - local64_set(&event->hw.prev_count, val); - local64_set(&event->hw.period_left, left); + event->hw.state |= PERF_HES_STOPPED | PERF_HES_UPTODATE; + write_pmc(event->hw.idx, 0); + perf_event_update_userpage(event); perf_pmu_enable(event->pmu); local_irq_restore(flags); @@ -524,13 +552,14 @@ static int fsl_emb_pmu_event_init(struct perf_event *event) } static struct pmu fsl_emb_pmu = { - .pmu_enable = fsl_emb_pmu_pmu_enable, - .pmu_disable = fsl_emb_pmu_pmu_disable, + .pmu_enable = fsl_emb_pmu_enable, + .pmu_disable = fsl_emb_pmu_disable, .event_init = fsl_emb_pmu_event_init, - .enable = fsl_emb_pmu_enable, - .disable = fsl_emb_pmu_disable, + .add = fsl_emb_pmu_add, + .del = fsl_emb_pmu_del, + .start = fsl_emb_pmu_start, + .stop = fsl_emb_pmu_stop, .read = fsl_emb_pmu_read, - .unthrottle = fsl_emb_pmu_unthrottle, }; /* @@ -545,6 +574,11 @@ static void record_and_restart(struct perf_event *event, unsigned long val, s64 prev, delta, left; int record = 0; + if (event->hw.state & PERF_HES_STOPPED) { + write_pmc(event->hw.idx, 0); + return; + } + /* we don't have to worry about interrupts here */ prev = local64_read(&event->hw.prev_count); delta = (val - prev) & 0xfffffffful; @@ -567,6 +601,11 @@ static void record_and_restart(struct perf_event *event, unsigned long val, val = 0x80000000LL - left; } + write_pmc(event->hw.idx, val); + local64_set(&event->hw.prev_count, val); + local64_set(&event->hw.period_left, left); + perf_event_update_userpage(event); + /* * Finally record data if requested. */ @@ -576,23 +615,9 @@ static void record_and_restart(struct perf_event *event, unsigned long val, perf_sample_data_init(&data, 0); data.period = event->hw.last_period; - if (perf_event_overflow(event, nmi, &data, regs)) { - /* - * Interrupts are coming too fast - throttle them - * by setting the event to 0, so it will be - * at least 2^30 cycles until the next interrupt - * (assuming each event counts at most 2 counts - * per cycle). - */ - val = 0; - left = ~0ULL >> 1; - } + if (perf_event_overflow(event, nmi, &data, regs)) + fsl_emb_pmu_stop(event, 0); } - - write_pmc(event->hw.idx, val); - local64_set(&event->hw.prev_count, val); - local64_set(&event->hw.period_left, left); - perf_event_update_userpage(event); } static void perf_event_interrupt(struct pt_regs *regs) diff --git a/arch/sh/kernel/perf_event.c b/arch/sh/kernel/perf_event.c index 4bbe19058a5..cf39c487346 100644 --- a/arch/sh/kernel/perf_event.c +++ b/arch/sh/kernel/perf_event.c @@ -206,26 +206,52 @@ again: local64_add(delta, &event->count); } -static void sh_pmu_disable(struct perf_event *event) +static void sh_pmu_stop(struct perf_event *event, int flags) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct hw_perf_event *hwc = &event->hw; int idx = hwc->idx; - clear_bit(idx, cpuc->active_mask); - sh_pmu->disable(hwc, idx); + if (!(event->hw.state & PERF_HES_STOPPED)) { + sh_pmu->disable(hwc, idx); + cpuc->events[idx] = NULL; + event->hw.state |= PERF_HES_STOPPED; + } - barrier(); + if ((flags & PERF_EF_UPDATE) && !(event->hw.state & PERF_HES_UPTODATE)) { + sh_perf_event_update(event, &event->hw, idx); + event->hw.state |= PERF_HES_UPTODATE; + } +} - sh_perf_event_update(event, &event->hw, idx); +static void sh_pmu_start(struct perf_event *event, int flags) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + struct hw_perf_event *hwc = &event->hw; + int idx = hwc->idx; - cpuc->events[idx] = NULL; - clear_bit(idx, cpuc->used_mask); + if (WARN_ON_ONCE(idx == -1)) + return; + + if (flags & PERF_EF_RELOAD) + WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE)); + + cpuc->events[idx] = event; + event->hw.state = 0; + sh_pmu->enable(hwc, idx); +} + +static void sh_pmu_del(struct perf_event *event, int flags) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + + sh_pmu_stop(event, PERF_EF_UPDATE); + __clear_bit(event->hw.idx, cpuc->used_mask); perf_event_update_userpage(event); } -static int sh_pmu_enable(struct perf_event *event) +static int sh_pmu_add(struct perf_event *event, int flags) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct hw_perf_event *hwc = &event->hw; @@ -234,21 +260,20 @@ static int sh_pmu_enable(struct perf_event *event) perf_pmu_disable(event->pmu); - if (test_and_set_bit(idx, cpuc->used_mask)) { + if (__test_and_set_bit(idx, cpuc->used_mask)) { idx = find_first_zero_bit(cpuc->used_mask, sh_pmu->num_events); if (idx == sh_pmu->num_events) goto out; - set_bit(idx, cpuc->used_mask); + __set_bit(idx, cpuc->used_mask); hwc->idx = idx; } sh_pmu->disable(hwc, idx); - cpuc->events[idx] = event; - set_bit(idx, cpuc->active_mask); - - sh_pmu->enable(hwc, idx); + event->hw.state = PERF_HES_UPTODATE | PERF_HES_STOPPED; + if (flags & PERF_EF_START) + sh_pmu_start(event, PERF_EF_RELOAD); perf_event_update_userpage(event); ret = 0; @@ -285,7 +310,7 @@ static int sh_pmu_event_init(struct perf_event *event) return err; } -static void sh_pmu_pmu_enable(struct pmu *pmu) +static void sh_pmu_enable(struct pmu *pmu) { if (!sh_pmu_initialized()) return; @@ -293,7 +318,7 @@ static void sh_pmu_pmu_enable(struct pmu *pmu) sh_pmu->enable_all(); } -static void sh_pmu_pmu_disable(struct pmu *pmu) +static void sh_pmu_disable(struct pmu *pmu) { if (!sh_pmu_initialized()) return; @@ -302,11 +327,13 @@ static void sh_pmu_pmu_disable(struct pmu *pmu) } static struct pmu pmu = { - .pmu_enable = sh_pmu_pmu_enable, - .pmu_disable = sh_pmu_pmu_disable, + .pmu_enable = sh_pmu_enable, + .pmu_disable = sh_pmu_disable, .event_init = sh_pmu_event_init, - .enable = sh_pmu_enable, - .disable = sh_pmu_disable, + .add = sh_pmu_add, + .del = sh_pmu_del, + .start = sh_pmu_start, + .stop = sh_pmu_stop, .read = sh_pmu_read, }; @@ -334,15 +361,15 @@ sh_pmu_notifier(struct notifier_block *self, unsigned long action, void *hcpu) return NOTIFY_OK; } -int __cpuinit register_sh_pmu(struct sh_pmu *pmu) +int __cpuinit register_sh_pmu(struct sh_pmu *_pmu) { if (sh_pmu) return -EBUSY; - sh_pmu = pmu; + sh_pmu = _pmu; - pr_info("Performance Events: %s support registered\n", pmu->name); + pr_info("Performance Events: %s support registered\n", _pmu->name); - WARN_ON(pmu->num_events > MAX_HWEVENTS); + WARN_ON(_pmu->num_events > MAX_HWEVENTS); perf_pmu_register(&pmu); perf_cpu_notifier(sh_pmu_notifier); diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c index 37cae676536..516be2314b5 100644 --- a/arch/sparc/kernel/perf_event.c +++ b/arch/sparc/kernel/perf_event.c @@ -658,13 +658,16 @@ static u64 maybe_change_configuration(struct cpu_hw_events *cpuc, u64 pcr) enc = perf_event_get_enc(cpuc->events[i]); pcr &= ~mask_for_index(idx); - pcr |= event_encoding(enc, idx); + if (hwc->state & PERF_HES_STOPPED) + pcr |= nop_for_index(idx); + else + pcr |= event_encoding(enc, idx); } out: return pcr; } -static void sparc_pmu_pmu_enable(struct pmu *pmu) +static void sparc_pmu_enable(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); u64 pcr; @@ -691,7 +694,7 @@ static void sparc_pmu_pmu_enable(struct pmu *pmu) pcr_ops->write(cpuc->pcr); } -static void sparc_pmu_pmu_disable(struct pmu *pmu) +static void sparc_pmu_disable(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); u64 val; @@ -710,10 +713,53 @@ static void sparc_pmu_pmu_disable(struct pmu *pmu) pcr_ops->write(cpuc->pcr); } -static void sparc_pmu_disable(struct perf_event *event) +static int active_event_index(struct cpu_hw_events *cpuc, + struct perf_event *event) +{ + int i; + + for (i = 0; i < cpuc->n_events; i++) { + if (cpuc->event[i] == event) + break; + } + BUG_ON(i == cpuc->n_events); + return cpuc->current_idx[i]; +} + +static void sparc_pmu_start(struct perf_event *event, int flags) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + int idx = active_event_index(cpuc, event); + + if (flags & PERF_EF_RELOAD) { + WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE)); + sparc_perf_event_set_period(event, &event->hw, idx); + } + + event->hw.state = 0; + + sparc_pmu_enable_event(cpuc, &event->hw, idx); +} + +static void sparc_pmu_stop(struct perf_event *event, int flags) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + int idx = active_event_index(cpuc, event); + + if (!(event->hw.state & PERF_HES_STOPPED)) { + sparc_pmu_disable_event(cpuc, &event->hw, idx); + event->hw.state |= PERF_HES_STOPPED; + } + + if (!(event->hw.state & PERF_HES_UPTODATE) && (flags & PERF_EF_UPDATE)) { + sparc_perf_event_update(event, &event->hw, idx); + event->hw.state |= PERF_HES_UPTODATE; + } +} + +static void sparc_pmu_del(struct perf_event *event, int _flags) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - struct hw_perf_event *hwc = &event->hw; unsigned long flags; int i; @@ -722,7 +768,10 @@ static void sparc_pmu_disable(struct perf_event *event) for (i = 0; i < cpuc->n_events; i++) { if (event == cpuc->event[i]) { - int idx = cpuc->current_idx[i]; + /* Absorb the final count and turn off the + * event. + */ + sparc_pmu_stop(event, PERF_EF_UPDATE); /* Shift remaining entries down into * the existing slot. @@ -734,13 +783,6 @@ static void sparc_pmu_disable(struct perf_event *event) cpuc->current_idx[i]; } - /* Absorb the final count and turn off the - * event. - */ - sparc_pmu_disable_event(cpuc, hwc, idx); - barrier(); - sparc_perf_event_update(event, hwc, idx); - perf_event_update_userpage(event); cpuc->n_events--; @@ -752,19 +794,6 @@ static void sparc_pmu_disable(struct perf_event *event) local_irq_restore(flags); } -static int active_event_index(struct cpu_hw_events *cpuc, - struct perf_event *event) -{ - int i; - - for (i = 0; i < cpuc->n_events; i++) { - if (cpuc->event[i] == event) - break; - } - BUG_ON(i == cpuc->n_events); - return cpuc->current_idx[i]; -} - static void sparc_pmu_read(struct perf_event *event) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -774,15 +803,6 @@ static void sparc_pmu_read(struct perf_event *event) sparc_perf_event_update(event, hwc, idx); } -static void sparc_pmu_unthrottle(struct perf_event *event) -{ - struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - int idx = active_event_index(cpuc, event); - struct hw_perf_event *hwc = &event->hw; - - sparc_pmu_enable_event(cpuc, hwc, idx); -} - static atomic_t active_events = ATOMIC_INIT(0); static DEFINE_MUTEX(pmc_grab_mutex); @@ -984,7 +1004,7 @@ static int collect_events(struct perf_event *group, int max_count, return n; } -static int sparc_pmu_enable(struct perf_event *event) +static int sparc_pmu_add(struct perf_event *event, int ef_flags) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); int n0, ret = -EAGAIN; @@ -1001,6 +1021,10 @@ static int sparc_pmu_enable(struct perf_event *event) cpuc->events[n0] = event->hw.event_base; cpuc->current_idx[n0] = PIC_NO_INDEX; + event->hw.state = PERF_HES_UPTODATE; + if (!(ef_flags & PERF_EF_START)) + event->hw.state |= PERF_HES_STOPPED; + /* * If group events scheduling transaction was started, * skip the schedulability test here, it will be peformed @@ -1156,13 +1180,14 @@ static int sparc_pmu_commit_txn(struct pmu *pmu) } static struct pmu pmu = { - .pmu_enable = sparc_pmu_pmu_enable, - .pmu_disable = sparc_pmu_pmu_disable, + .pmu_enable = sparc_pmu_enable, + .pmu_disable = sparc_pmu_disable, .event_init = sparc_pmu_event_init, - .enable = sparc_pmu_enable, - .disable = sparc_pmu_disable, + .add = sparc_pmu_add, + .del = sparc_pmu_del, + .start = sparc_pmu_start, + .stop = sparc_pmu_stop, .read = sparc_pmu_read, - .unthrottle = sparc_pmu_unthrottle, .start_txn = sparc_pmu_start_txn, .cancel_txn = sparc_pmu_cancel_txn, .commit_txn = sparc_pmu_commit_txn, @@ -1243,7 +1268,7 @@ static int __kprobes perf_event_nmi_handler(struct notifier_block *self, continue; if (perf_event_overflow(event, 1, &data, regs)) - sparc_pmu_disable_event(cpuc, hwc, idx); + sparc_pmu_stop(event, 0); } return NOTIFY_STOP; diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 79705ac4501..dd6fec71067 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -583,7 +583,7 @@ static void x86_pmu_disable_all(void) } } -static void x86_pmu_pmu_disable(struct pmu *pmu) +static void x86_pmu_disable(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -800,10 +800,10 @@ static inline int match_prev_assignment(struct hw_perf_event *hwc, hwc->last_tag == cpuc->tags[i]; } -static int x86_pmu_start(struct perf_event *event); -static void x86_pmu_stop(struct perf_event *event); +static void x86_pmu_start(struct perf_event *event, int flags); +static void x86_pmu_stop(struct perf_event *event, int flags); -static void x86_pmu_pmu_enable(struct pmu *pmu) +static void x86_pmu_enable(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct perf_event *event; @@ -839,7 +839,14 @@ static void x86_pmu_pmu_enable(struct pmu *pmu) match_prev_assignment(hwc, cpuc, i)) continue; - x86_pmu_stop(event); + /* + * Ensure we don't accidentally enable a stopped + * counter simply because we rescheduled. + */ + if (hwc->state & PERF_HES_STOPPED) + hwc->state |= PERF_HES_ARCH; + + x86_pmu_stop(event, PERF_EF_UPDATE); } for (i = 0; i < cpuc->n_events; i++) { @@ -851,7 +858,10 @@ static void x86_pmu_pmu_enable(struct pmu *pmu) else if (i < n_running) continue; - x86_pmu_start(event); + if (hwc->state & PERF_HES_ARCH) + continue; + + x86_pmu_start(event, PERF_EF_RELOAD); } cpuc->n_added = 0; perf_events_lapic_init(); @@ -952,15 +962,12 @@ static void x86_pmu_enable_event(struct perf_event *event) } /* - * activate a single event + * Add a single event to the PMU. * * The event is added to the group of enabled events * but only if it can be scehduled with existing events. - * - * Called with PMU disabled. If successful and return value 1, - * then guaranteed to call perf_enable() and hw_perf_enable() */ -static int x86_pmu_enable(struct perf_event *event) +static int x86_pmu_add(struct perf_event *event, int flags) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct hw_perf_event *hwc; @@ -975,10 +982,14 @@ static int x86_pmu_enable(struct perf_event *event) if (ret < 0) goto out; + hwc->state = PERF_HES_UPTODATE | PERF_HES_STOPPED; + if (!(flags & PERF_EF_START)) + hwc->state |= PERF_HES_ARCH; + /* * If group events scheduling transaction was started, * skip the schedulability test here, it will be peformed - * at commit time(->commit_txn) as a whole + * at commit time (->commit_txn) as a whole */ if (cpuc->group_flag & PERF_EVENT_TXN) goto done_collect; @@ -1003,27 +1014,28 @@ out: return ret; } -static int x86_pmu_start(struct perf_event *event) +static void x86_pmu_start(struct perf_event *event, int flags) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); int idx = event->hw.idx; - if (idx == -1) - return -EAGAIN; + if (WARN_ON_ONCE(!(event->hw.state & PERF_HES_STOPPED))) + return; + + if (WARN_ON_ONCE(idx == -1)) + return; + + if (flags & PERF_EF_RELOAD) { + WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE)); + x86_perf_event_set_period(event); + } + + event->hw.state = 0; - x86_perf_event_set_period(event); cpuc->events[idx] = event; __set_bit(idx, cpuc->active_mask); x86_pmu.enable(event); perf_event_update_userpage(event); - - return 0; -} - -static void x86_pmu_unthrottle(struct perf_event *event) -{ - int ret = x86_pmu_start(event); - WARN_ON_ONCE(ret); } void perf_event_print_debug(void) @@ -1080,27 +1092,29 @@ void perf_event_print_debug(void) local_irq_restore(flags); } -static void x86_pmu_stop(struct perf_event *event) +static void x86_pmu_stop(struct perf_event *event, int flags) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct hw_perf_event *hwc = &event->hw; - int idx = hwc->idx; - - if (!__test_and_clear_bit(idx, cpuc->active_mask)) - return; - - x86_pmu.disable(event); - /* - * Drain the remaining delta count out of a event - * that we are disabling: - */ - x86_perf_event_update(event); + if (__test_and_clear_bit(hwc->idx, cpuc->active_mask)) { + x86_pmu.disable(event); + cpuc->events[hwc->idx] = NULL; + WARN_ON_ONCE(hwc->state & PERF_HES_STOPPED); + hwc->state |= PERF_HES_STOPPED; + } - cpuc->events[idx] = NULL; + if ((flags & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) { + /* + * Drain the remaining delta count out of a event + * that we are disabling: + */ + x86_perf_event_update(event); + hwc->state |= PERF_HES_UPTODATE; + } } -static void x86_pmu_disable(struct perf_event *event) +static void x86_pmu_del(struct perf_event *event, int flags) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); int i; @@ -1113,7 +1127,7 @@ static void x86_pmu_disable(struct perf_event *event) if (cpuc->group_flag & PERF_EVENT_TXN) return; - x86_pmu_stop(event); + x86_pmu_stop(event, PERF_EF_UPDATE); for (i = 0; i < cpuc->n_events; i++) { if (event == cpuc->event_list[i]) { @@ -1165,7 +1179,7 @@ static int x86_pmu_handle_irq(struct pt_regs *regs) continue; if (perf_event_overflow(event, 1, &data, regs)) - x86_pmu_stop(event); + x86_pmu_stop(event, 0); } if (handled) @@ -1605,15 +1619,17 @@ int x86_pmu_event_init(struct perf_event *event) } static struct pmu pmu = { - .pmu_enable = x86_pmu_pmu_enable, - .pmu_disable = x86_pmu_pmu_disable, + .pmu_enable = x86_pmu_enable, + .pmu_disable = x86_pmu_disable, + .event_init = x86_pmu_event_init, - .enable = x86_pmu_enable, - .disable = x86_pmu_disable, + + .add = x86_pmu_add, + .del = x86_pmu_del, .start = x86_pmu_start, .stop = x86_pmu_stop, .read = x86_pmu_read, - .unthrottle = x86_pmu_unthrottle, + .start_txn = x86_pmu_start_txn, .cancel_txn = x86_pmu_cancel_txn, .commit_txn = x86_pmu_commit_txn, diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index ee05c90012d..82395f2378e 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -763,7 +763,7 @@ again: data.period = event->hw.last_period; if (perf_event_overflow(event, 1, &data, regs)) - x86_pmu_stop(event); + x86_pmu_stop(event, 0); } /* diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c index 18018d1311c..9893a2f77b7 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_ds.c +++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c @@ -491,7 +491,7 @@ static void __intel_pmu_pebs_event(struct perf_event *event, regs.flags &= ~PERF_EFLAGS_EXACT; if (perf_event_overflow(event, 1, &data, ®s)) - x86_pmu_stop(event); + x86_pmu_stop(event, 0); } static void intel_pmu_drain_pebs_core(struct pt_regs *iregs) diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h index 5f8ad7bec63..8beabb958f6 100644 --- a/include/linux/ftrace_event.h +++ b/include/linux/ftrace_event.h @@ -252,8 +252,8 @@ DECLARE_PER_CPU(struct pt_regs, perf_trace_regs); extern int perf_trace_init(struct perf_event *event); extern void perf_trace_destroy(struct perf_event *event); -extern int perf_trace_enable(struct perf_event *event); -extern void perf_trace_disable(struct perf_event *event); +extern int perf_trace_add(struct perf_event *event, int flags); +extern void perf_trace_del(struct perf_event *event, int flags); extern int ftrace_profile_set_filter(struct perf_event *event, int event_id, char *filter_str); extern void ftrace_profile_free_filter(struct perf_event *event); diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 8cafa15af60..402073c6166 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -538,6 +538,7 @@ struct hw_perf_event { }; #endif }; + int state; local64_t prev_count; u64 sample_period; u64 last_period; @@ -549,6 +550,13 @@ struct hw_perf_event { #endif }; +/* + * hw_perf_event::state flags + */ +#define PERF_HES_STOPPED 0x01 /* the counter is stopped */ +#define PERF_HES_UPTODATE 0x02 /* event->count up-to-date */ +#define PERF_HES_ARCH 0x04 + struct perf_event; /* @@ -564,42 +572,62 @@ struct pmu { int *pmu_disable_count; + /* + * Fully disable/enable this PMU, can be used to protect from the PMI + * as well as for lazy/batch writing of the MSRs. + */ void (*pmu_enable) (struct pmu *pmu); /* optional */ void (*pmu_disable) (struct pmu *pmu); /* optional */ /* + * Try and initialize the event for this PMU. * Should return -ENOENT when the @event doesn't match this PMU. */ int (*event_init) (struct perf_event *event); - int (*enable) (struct perf_event *event); - void (*disable) (struct perf_event *event); - int (*start) (struct perf_event *event); - void (*stop) (struct perf_event *event); +#define PERF_EF_START 0x01 /* start the counter when adding */ +#define PERF_EF_RELOAD 0x02 /* reload the counter when starting */ +#define PERF_EF_UPDATE 0x04 /* update the counter when stopping */ + + /* + * Adds/Removes a counter to/from the PMU, can be done inside + * a transaction, see the ->*_txn() methods. + */ + int (*add) (struct perf_event *event, int flags); + void (*del) (struct perf_event *event, int flags); + + /* + * Starts/Stops a counter present on the PMU. The PMI handler + * should stop the counter when perf_event_overflow() returns + * !0. ->start() will be used to continue. + */ + void (*start) (struct perf_event *event, int flags); + void (*stop) (struct perf_event *event, int flags); + + /* + * Updates the counter value of the event. + */ void (*read) (struct perf_event *event); - void (*unthrottle) (struct perf_event *event); /* * Group events scheduling is treated as a transaction, add * group events as a whole and perform one schedulability test. * If the test fails, roll back the whole group - */ - - /* - * Start the transaction, after this ->enable() doesn't need to + * + * Start the transaction, after this ->add() doesn't need to * do schedulability tests. */ void (*start_txn) (struct pmu *pmu); /* optional */ /* - * If ->start_txn() disabled the ->enable() schedulability test + * If ->start_txn() disabled the ->add() schedulability test * then ->commit_txn() is required to perform one. On success * the transaction is closed. On error the transaction is kept * open until ->cancel_txn() is called. */ int (*commit_txn) (struct pmu *pmu); /* optional */ /* - * Will cancel the transaction, assumes ->disable() is called - * for each successfull ->enable() during the transaction. + * Will cancel the transaction, assumes ->del() is called + * for each successfull ->add() during the transaction. */ void (*cancel_txn) (struct pmu *pmu); /* optional */ }; @@ -680,7 +708,7 @@ struct perf_event { int nr_siblings; int group_flags; struct perf_event *group_leader; - struct pmu *pmu; + struct pmu *pmu; enum perf_event_active_state state; unsigned int attach_state; diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c index e9c5cfa1fd2..6f150095caf 100644 --- a/kernel/hw_breakpoint.c +++ b/kernel/hw_breakpoint.c @@ -586,10 +586,35 @@ static int hw_breakpoint_event_init(struct perf_event *bp) return 0; } +static int hw_breakpoint_add(struct perf_event *bp, int flags) +{ + if (!(flags & PERF_EF_START)) + bp->hw.state = PERF_HES_STOPPED; + + return arch_install_hw_breakpoint(bp); +} + +static void hw_breakpoint_del(struct perf_event *bp, int flags) +{ + arch_uninstall_hw_breakpoint(bp); +} + +static void hw_breakpoint_start(struct perf_event *bp, int flags) +{ + bp->hw.state = 0; +} + +static void hw_breakpoint_stop(struct perf_event *bp, int flags) +{ + bp->hw.state = PERF_HES_STOPPED; +} + static struct pmu perf_breakpoint = { .event_init = hw_breakpoint_event_init, - .enable = arch_install_hw_breakpoint, - .disable = arch_uninstall_hw_breakpoint, + .add = hw_breakpoint_add, + .del = hw_breakpoint_del, + .start = hw_breakpoint_start, + .stop = hw_breakpoint_stop, .read = hw_breakpoint_pmu_read, }; diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 1a6cdbf0d09..3bace4fd035 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -424,7 +424,7 @@ event_sched_out(struct perf_event *event, event->state = PERF_EVENT_STATE_OFF; } event->tstamp_stopped = ctx->time; - event->pmu->disable(event); + event->pmu->del(event, 0); event->oncpu = -1; if (!is_software_event(event)) @@ -649,7 +649,7 @@ event_sched_in(struct perf_event *event, */ smp_wmb(); - if (event->pmu->enable(event)) { + if (event->pmu->add(event, PERF_EF_START)) { event->state = PERF_EVENT_STATE_INACTIVE; event->oncpu = -1; return -EAGAIN; @@ -1482,22 +1482,6 @@ do { \ return div64_u64(dividend, divisor); } -static void perf_event_stop(struct perf_event *event) -{ - if (!event->pmu->stop) - return event->pmu->disable(event); - - return event->pmu->stop(event); -} - -static int perf_event_start(struct perf_event *event) -{ - if (!event->pmu->start) - return event->pmu->enable(event); - - return event->pmu->start(event); -} - static void perf_adjust_period(struct perf_event *event, u64 nsec, u64 count) { struct hw_perf_event *hwc = &event->hw; @@ -1517,9 +1501,9 @@ static void perf_adjust_period(struct perf_event *event, u64 nsec, u64 count) hwc->sample_period = sample_period; if (local64_read(&hwc->period_left) > 8*sample_period) { - perf_event_stop(event); + event->pmu->stop(event, PERF_EF_UPDATE); local64_set(&hwc->period_left, 0); - perf_event_start(event); + event->pmu->start(event, PERF_EF_RELOAD); } } @@ -1548,7 +1532,7 @@ static void perf_ctx_adjust_freq(struct perf_event_context *ctx) */ if (interrupts == MAX_INTERRUPTS) { perf_log_throttle(event, 1); - event->pmu->unthrottle(event); + event->pmu->start(event, 0); } if (!event->attr.freq || !event->attr.sample_freq) @@ -2506,6 +2490,9 @@ int perf_event_task_disable(void) static int perf_event_index(struct perf_event *event) { + if (event->hw.state & PERF_HES_STOPPED) + return 0; + if (event->state != PERF_EVENT_STATE_ACTIVE) return 0; @@ -4120,8 +4107,6 @@ static int __perf_event_overflow(struct perf_event *event, int nmi, struct hw_perf_event *hwc = &event->hw; int ret = 0; - throttle = (throttle && event->pmu->unthrottle != NULL); - if (!throttle) { hwc->interrupts++; } else { @@ -4246,7 +4231,7 @@ static void perf_swevent_overflow(struct perf_event *event, u64 overflow, } } -static void perf_swevent_add(struct perf_event *event, u64 nr, +static void perf_swevent_event(struct perf_event *event, u64 nr, int nmi, struct perf_sample_data *data, struct pt_regs *regs) { @@ -4272,6 +4257,9 @@ static void perf_swevent_add(struct perf_event *event, u64 nr, static int perf_exclude_event(struct perf_event *event, struct pt_regs *regs) { + if (event->hw.state & PERF_HES_STOPPED) + return 0; + if (regs) { if (event->attr.exclude_user && user_mode(regs)) return 1; @@ -4371,7 +4359,7 @@ static void do_perf_sw_event(enum perf_type_id type, u32 event_id, hlist_for_each_entry_rcu(event, node, head, hlist_entry) { if (perf_swevent_match(event, type, event_id, data, regs)) - perf_swevent_add(event, nr, nmi, data, regs); + perf_swevent_event(event, nr, nmi, data, regs); } end: rcu_read_unlock(); @@ -4415,7 +4403,7 @@ static void perf_swevent_read(struct perf_event *event) { } -static int perf_swevent_enable(struct perf_event *event) +static int perf_swevent_add(struct perf_event *event, int flags) { struct hw_perf_event *hwc = &event->hw; struct perf_cpu_context *cpuctx; @@ -4428,6 +4416,8 @@ static int perf_swevent_enable(struct perf_event *event) perf_swevent_set_period(event); } + hwc->state = !(flags & PERF_EF_START); + head = find_swevent_head(cpuctx, event); if (WARN_ON_ONCE(!head)) return -EINVAL; @@ -4437,18 +4427,19 @@ static int perf_swevent_enable(struct perf_event *event) return 0; } -static void perf_swevent_disable(struct perf_event *event) +static void perf_swevent_del(struct perf_event *event, int flags) { hlist_del_rcu(&event->hlist_entry); } -static void perf_swevent_void(struct perf_event *event) +static void perf_swevent_start(struct perf_event *event, int flags) { + event->hw.state = 0; } -static int perf_swevent_int(struct perf_event *event) +static void perf_swevent_stop(struct perf_event *event, int flags) { - return 0; + event->hw.state = PERF_HES_STOPPED; } /* Deref the hlist from the update side */ @@ -4604,12 +4595,11 @@ static int perf_swevent_init(struct perf_event *event) static struct pmu perf_swevent = { .event_init = perf_swevent_init, - .enable = perf_swevent_enable, - .disable = perf_swevent_disable, - .start = perf_swevent_int, - .stop = perf_swevent_void, + .add = perf_swevent_add, + .del = perf_swevent_del, + .start = perf_swevent_start, + .stop = perf_swevent_stop, .read = perf_swevent_read, - .unthrottle = perf_swevent_void, /* hwc->interrupts already reset */ }; #ifdef CONFIG_EVENT_TRACING @@ -4657,7 +4647,7 @@ void perf_tp_event(u64 addr, u64 count, void *record, int entry_size, hlist_for_each_entry_rcu(event, node, head, hlist_entry) { if (perf_tp_event_match(event, &data, regs)) - perf_swevent_add(event, count, 1, &data, regs); + perf_swevent_event(event, count, 1, &data, regs); } perf_swevent_put_recursion_context(rctx); @@ -4696,12 +4686,11 @@ static int perf_tp_event_init(struct perf_event *event) static struct pmu perf_tracepoint = { .event_init = perf_tp_event_init, - .enable = perf_trace_enable, - .disable = perf_trace_disable, - .start = perf_swevent_int, - .stop = perf_swevent_void, + .add = perf_trace_add, + .del = perf_trace_del, + .start = perf_swevent_start, + .stop = perf_swevent_stop, .read = perf_swevent_read, - .unthrottle = perf_swevent_void, }; static inline void perf_tp_register(void) @@ -4757,8 +4746,8 @@ void perf_bp_event(struct perf_event *bp, void *data) perf_sample_data_init(&sample, bp->attr.bp_addr); - if (!perf_exclude_event(bp, regs)) - perf_swevent_add(bp, 1, 1, &sample, regs); + if (!bp->hw.state && !perf_exclude_event(bp, regs)) + perf_swevent_event(bp, 1, 1, &sample, regs); } #endif @@ -4834,32 +4823,39 @@ static void perf_swevent_cancel_hrtimer(struct perf_event *event) static void cpu_clock_event_update(struct perf_event *event) { - int cpu = raw_smp_processor_id(); s64 prev; u64 now; - now = cpu_clock(cpu); + now = local_clock(); prev = local64_xchg(&event->hw.prev_count, now); local64_add(now - prev, &event->count); } -static int cpu_clock_event_enable(struct perf_event *event) +static void cpu_clock_event_start(struct perf_event *event, int flags) { - struct hw_perf_event *hwc = &event->hw; - int cpu = raw_smp_processor_id(); - - local64_set(&hwc->prev_count, cpu_clock(cpu)); + local64_set(&event->hw.prev_count, local_clock()); perf_swevent_start_hrtimer(event); - - return 0; } -static void cpu_clock_event_disable(struct perf_event *event) +static void cpu_clock_event_stop(struct perf_event *event, int flags) { perf_swevent_cancel_hrtimer(event); cpu_clock_event_update(event); } +static int cpu_clock_event_add(struct perf_event *event, int flags) +{ + if (flags & PERF_EF_START) + cpu_clock_event_start(event, flags); + + return 0; +} + +static void cpu_clock_event_del(struct perf_event *event, int flags) +{ + cpu_clock_event_stop(event, flags); +} + static void cpu_clock_event_read(struct perf_event *event) { cpu_clock_event_update(event); @@ -4878,8 +4874,10 @@ static int cpu_clock_event_init(struct perf_event *event) static struct pmu perf_cpu_clock = { .event_init = cpu_clock_event_init, - .enable = cpu_clock_event_enable, - .disable = cpu_clock_event_disable, + .add = cpu_clock_event_add, + .del = cpu_clock_event_del, + .start = cpu_clock_event_start, + .stop = cpu_clock_event_stop, .read = cpu_clock_event_read, }; @@ -4897,25 +4895,29 @@ static void task_clock_event_update(struct perf_event *event, u64 now) local64_add(delta, &event->count); } -static int task_clock_event_enable(struct perf_event *event) +static void task_clock_event_start(struct perf_event *event, int flags) { - struct hw_perf_event *hwc = &event->hw; - u64 now; - - now = event->ctx->time; - - local64_set(&hwc->prev_count, now); - + local64_set(&event->hw.prev_count, event->ctx->time); perf_swevent_start_hrtimer(event); - - return 0; } -static void task_clock_event_disable(struct perf_event *event) +static void task_clock_event_stop(struct perf_event *event, int flags) { perf_swevent_cancel_hrtimer(event); task_clock_event_update(event, event->ctx->time); +} + +static int task_clock_event_add(struct perf_event *event, int flags) +{ + if (flags & PERF_EF_START) + task_clock_event_start(event, flags); + return 0; +} + +static void task_clock_event_del(struct perf_event *event, int flags) +{ + task_clock_event_stop(event, PERF_EF_UPDATE); } static void task_clock_event_read(struct perf_event *event) @@ -4947,8 +4949,10 @@ static int task_clock_event_init(struct perf_event *event) static struct pmu perf_task_clock = { .event_init = task_clock_event_init, - .enable = task_clock_event_enable, - .disable = task_clock_event_disable, + .add = task_clock_event_add, + .del = task_clock_event_del, + .start = task_clock_event_start, + .stop = task_clock_event_stop, .read = task_clock_event_read, }; diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c index f3bbcd1c90c..39c059ca670 100644 --- a/kernel/trace/trace_event_perf.c +++ b/kernel/trace/trace_event_perf.c @@ -101,7 +101,7 @@ int perf_trace_init(struct perf_event *p_event) return ret; } -int perf_trace_enable(struct perf_event *p_event) +int perf_trace_add(struct perf_event *p_event, int flags) { struct ftrace_event_call *tp_event = p_event->tp_event; struct hlist_head __percpu *pcpu_list; @@ -111,13 +111,16 @@ int perf_trace_enable(struct perf_event *p_event) if (WARN_ON_ONCE(!pcpu_list)) return -EINVAL; + if (!(flags & PERF_EF_START)) + p_event->hw.state = PERF_HES_STOPPED; + list = this_cpu_ptr(pcpu_list); hlist_add_head_rcu(&p_event->hlist_entry, list); return 0; } -void perf_trace_disable(struct perf_event *p_event) +void perf_trace_del(struct perf_event *p_event, int flags) { hlist_del_rcu(&p_event->hlist_entry); } -- cgit v1.2.3-70-g09d2 From 15ac9a395a753cb28c674e7ea80386ffdff21785 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 6 Sep 2010 15:51:45 +0200 Subject: perf: Remove the sysfs bits Neither the overcommit nor the reservation sysfs parameter were actually working, remove them as they'll only get in the way. Signed-off-by: Peter Zijlstra Cc: paulus LKML-Reference: Signed-off-by: Ingo Molnar --- arch/alpha/kernel/perf_event.c | 3 +- arch/arm/kernel/perf_event.c | 9 +-- arch/sparc/kernel/perf_event.c | 9 +-- arch/x86/kernel/cpu/perf_event.c | 1 - include/linux/perf_event.h | 6 -- kernel/perf_event.c | 124 --------------------------------------- 6 files changed, 5 insertions(+), 147 deletions(-) (limited to 'arch/x86') diff --git a/arch/alpha/kernel/perf_event.c b/arch/alpha/kernel/perf_event.c index 380ef02d557..9bb8c024080 100644 --- a/arch/alpha/kernel/perf_event.c +++ b/arch/alpha/kernel/perf_event.c @@ -808,7 +808,7 @@ static void alpha_perf_event_irq_handler(unsigned long la_ptr, wrperfmon(PERFMON_CMD_DISABLE, cpuc->idx_mask); /* la_ptr is the counter that overflowed. */ - if (unlikely(la_ptr >= perf_max_events)) { + if (unlikely(la_ptr >= alpha_pmu->num_pmcs)) { /* This should never occur! */ irq_err_count++; pr_warning("PMI: silly index %ld\n", la_ptr); @@ -879,7 +879,6 @@ void __init init_hw_perf_events(void) /* And set up PMU specification */ alpha_pmu = &ev67_pmu; - perf_max_events = alpha_pmu->num_pmcs; perf_pmu_register(&pmu); } diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index 448cfa6b3ef..45d6a35217c 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -534,7 +534,7 @@ static int armpmu_event_init(struct perf_event *event) event->destroy = hw_perf_event_destroy; if (!atomic_inc_not_zero(&active_events)) { - if (atomic_read(&active_events) > perf_max_events) { + if (atomic_read(&active_events) > armpmu.num_events) { atomic_dec(&active_events); return -ENOSPC; } @@ -2974,14 +2974,12 @@ init_hw_perf_events(void) armpmu = &armv6pmu; memcpy(armpmu_perf_cache_map, armv6_perf_cache_map, sizeof(armv6_perf_cache_map)); - perf_max_events = armv6pmu.num_events; break; case 0xB020: /* ARM11mpcore */ armpmu = &armv6mpcore_pmu; memcpy(armpmu_perf_cache_map, armv6mpcore_perf_cache_map, sizeof(armv6mpcore_perf_cache_map)); - perf_max_events = armv6mpcore_pmu.num_events; break; case 0xC080: /* Cortex-A8 */ armv7pmu.id = ARM_PERF_PMU_ID_CA8; @@ -2993,7 +2991,6 @@ init_hw_perf_events(void) /* Reset PMNC and read the nb of CNTx counters supported */ armv7pmu.num_events = armv7_reset_read_pmnc(); - perf_max_events = armv7pmu.num_events; break; case 0xC090: /* Cortex-A9 */ armv7pmu.id = ARM_PERF_PMU_ID_CA9; @@ -3005,7 +3002,6 @@ init_hw_perf_events(void) /* Reset PMNC and read the nb of CNTx counters supported */ armv7pmu.num_events = armv7_reset_read_pmnc(); - perf_max_events = armv7pmu.num_events; break; } /* Intel CPUs [xscale]. */ @@ -3016,13 +3012,11 @@ init_hw_perf_events(void) armpmu = &xscale1pmu; memcpy(armpmu_perf_cache_map, xscale_perf_cache_map, sizeof(xscale_perf_cache_map)); - perf_max_events = xscale1pmu.num_events; break; case 2: armpmu = &xscale2pmu; memcpy(armpmu_perf_cache_map, xscale_perf_cache_map, sizeof(xscale_perf_cache_map)); - perf_max_events = xscale2pmu.num_events; break; } } @@ -3032,7 +3026,6 @@ init_hw_perf_events(void) arm_pmu_names[armpmu->id], armpmu->num_events); } else { pr_info("no hardware support available\n"); - perf_max_events = -1; } perf_pmu_register(&pmu); diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c index 516be2314b5..f9a70675936 100644 --- a/arch/sparc/kernel/perf_event.c +++ b/arch/sparc/kernel/perf_event.c @@ -897,7 +897,7 @@ static int sparc_check_constraints(struct perf_event **evts, if (!n_ev) return 0; - if (n_ev > perf_max_events) + if (n_ev > MAX_HWEVENTS) return -1; msk0 = perf_event_get_msk(events[0]); @@ -1014,7 +1014,7 @@ static int sparc_pmu_add(struct perf_event *event, int ef_flags) perf_pmu_disable(event->pmu); n0 = cpuc->n_events; - if (n0 >= perf_max_events) + if (n0 >= MAX_HWEVENTS) goto out; cpuc->event[n0] = event; @@ -1097,7 +1097,7 @@ static int sparc_pmu_event_init(struct perf_event *event) n = 0; if (event->group_leader != event) { n = collect_events(event->group_leader, - perf_max_events - 1, + MAX_HWEVENTS - 1, evts, events, current_idx_dmy); if (n < 0) return -EINVAL; @@ -1309,9 +1309,6 @@ void __init init_hw_perf_events(void) pr_cont("Supported PMU type is '%s'\n", sparc_pmu_type); - /* All sparc64 PMUs currently have 2 events. */ - perf_max_events = 2; - perf_pmu_register(&pmu); register_die_notifier(&perf_event_nmi_notifier); } diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index dd6fec71067..0fb17050360 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -1396,7 +1396,6 @@ void __init init_hw_perf_events(void) x86_pmu.num_counters = X86_PMC_MAX_GENERIC; } x86_pmu.intel_ctrl = (1 << x86_pmu.num_counters) - 1; - perf_max_events = x86_pmu.num_counters; if (x86_pmu.num_counters_fixed > X86_PMC_MAX_FIXED) { WARN(1, KERN_ERR "hw perf events fixed %d > max(%d), clipping!", diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 402073c6166..b22176d3ebd 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -860,7 +860,6 @@ struct perf_cpu_context { struct perf_event_context ctx; struct perf_event_context *task_ctx; int active_oncpu; - int max_pertask; int exclusive; struct swevent_hlist *swevent_hlist; struct mutex hlist_mutex; @@ -883,11 +882,6 @@ struct perf_output_handle { #ifdef CONFIG_PERF_EVENTS -/* - * Set by architecture code: - */ -extern int perf_max_events; - extern int perf_pmu_register(struct pmu *pmu); extern void perf_pmu_unregister(struct pmu *pmu); diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 3bace4fd035..8462e69409a 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -39,10 +39,6 @@ */ static DEFINE_PER_CPU(struct perf_cpu_context, perf_cpu_context); -int perf_max_events __read_mostly = 1; -static int perf_reserved_percpu __read_mostly; -static int perf_overcommit __read_mostly = 1; - static atomic_t nr_events __read_mostly; static atomic_t nr_mmap_events __read_mostly; static atomic_t nr_comm_events __read_mostly; @@ -66,11 +62,6 @@ int sysctl_perf_event_sample_rate __read_mostly = 100000; static atomic64_t perf_event_id; -/* - * Lock for (sysadmin-configurable) event reservations: - */ -static DEFINE_SPINLOCK(perf_resource_lock); - void __weak perf_event_print_debug(void) { } void perf_pmu_disable(struct pmu *pmu) @@ -480,16 +471,6 @@ static void __perf_event_remove_from_context(void *info) list_del_event(event, ctx); - if (!ctx->task) { - /* - * Allow more per task events with respect to the - * reservation: - */ - cpuctx->max_pertask = - min(perf_max_events - ctx->nr_events, - perf_max_events - perf_reserved_percpu); - } - raw_spin_unlock(&ctx->lock); } @@ -823,9 +804,6 @@ static void __perf_install_in_context(void *info) } } - if (!err && !ctx->task && cpuctx->max_pertask) - cpuctx->max_pertask--; - unlock: raw_spin_unlock(&ctx->lock); } @@ -5930,10 +5908,6 @@ static void __cpuinit perf_event_init_cpu(int cpu) cpuctx = &per_cpu(perf_cpu_context, cpu); - spin_lock(&perf_resource_lock); - cpuctx->max_pertask = perf_max_events - perf_reserved_percpu; - spin_unlock(&perf_resource_lock); - mutex_lock(&cpuctx->hlist_mutex); if (cpuctx->hlist_refcount > 0) { struct swevent_hlist *hlist; @@ -6008,101 +5982,3 @@ void __init perf_event_init(void) perf_tp_register(); perf_cpu_notifier(perf_cpu_notify); } - -static ssize_t perf_show_reserve_percpu(struct sysdev_class *class, - struct sysdev_class_attribute *attr, - char *buf) -{ - return sprintf(buf, "%d\n", perf_reserved_percpu); -} - -static ssize_t -perf_set_reserve_percpu(struct sysdev_class *class, - struct sysdev_class_attribute *attr, - const char *buf, - size_t count) -{ - struct perf_cpu_context *cpuctx; - unsigned long val; - int err, cpu, mpt; - - err = strict_strtoul(buf, 10, &val); - if (err) - return err; - if (val > perf_max_events) - return -EINVAL; - - spin_lock(&perf_resource_lock); - perf_reserved_percpu = val; - for_each_online_cpu(cpu) { - cpuctx = &per_cpu(perf_cpu_context, cpu); - raw_spin_lock_irq(&cpuctx->ctx.lock); - mpt = min(perf_max_events - cpuctx->ctx.nr_events, - perf_max_events - perf_reserved_percpu); - cpuctx->max_pertask = mpt; - raw_spin_unlock_irq(&cpuctx->ctx.lock); - } - spin_unlock(&perf_resource_lock); - - return count; -} - -static ssize_t perf_show_overcommit(struct sysdev_class *class, - struct sysdev_class_attribute *attr, - char *buf) -{ - return sprintf(buf, "%d\n", perf_overcommit); -} - -static ssize_t -perf_set_overcommit(struct sysdev_class *class, - struct sysdev_class_attribute *attr, - const char *buf, size_t count) -{ - unsigned long val; - int err; - - err = strict_strtoul(buf, 10, &val); - if (err) - return err; - if (val > 1) - return -EINVAL; - - spin_lock(&perf_resource_lock); - perf_overcommit = val; - spin_unlock(&perf_resource_lock); - - return count; -} - -static SYSDEV_CLASS_ATTR( - reserve_percpu, - 0644, - perf_show_reserve_percpu, - perf_set_reserve_percpu - ); - -static SYSDEV_CLASS_ATTR( - overcommit, - 0644, - perf_show_overcommit, - perf_set_overcommit - ); - -static struct attribute *perfclass_attrs[] = { - &attr_reserve_percpu.attr, - &attr_overcommit.attr, - NULL -}; - -static struct attribute_group perfclass_attr_group = { - .attrs = perfclass_attrs, - .name = "perf_events", -}; - -static int __init perf_event_sysfs_init(void) -{ - return sysfs_create_group(&cpu_sysdev_class.kset.kobj, - &perfclass_attr_group); -} -device_initcall(perf_event_sysfs_init); -- cgit v1.2.3-70-g09d2 From 2df7a6e9e8e67c19e5fe2eac3f2d2223b7bb4a7b Mon Sep 17 00:00:00 2001 From: Brian Gerst Date: Fri, 3 Sep 2010 21:17:08 -0400 Subject: x86: Use correct type for %cr4 %cr4 is 64-bit in 64-bit mode (although the upper 32-bits are currently reserved). Use unsigned long for the temporary variable to get the right size. Signed-off-by: Brian Gerst Acked-by: Pekka Enberg Cc: Suresh Siddha LKML-Reference: <1283563039-3466-2-git-send-email-brgerst@gmail.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/processor.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index 325b7bdbeba..396b80f2d87 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -602,7 +602,7 @@ extern unsigned long mmu_cr4_features; static inline void set_in_cr4(unsigned long mask) { - unsigned cr4; + unsigned long cr4; mmu_cr4_features |= mask; cr4 = read_cr4(); @@ -612,7 +612,7 @@ static inline void set_in_cr4(unsigned long mask) static inline void clear_in_cr4(unsigned long mask) { - unsigned cr4; + unsigned long cr4; mmu_cr4_features &= ~mask; cr4 = read_cr4(); -- cgit v1.2.3-70-g09d2 From 6ac8bac2684235f4caf22a410549c582aa7327d6 Mon Sep 17 00:00:00 2001 From: Brian Gerst Date: Fri, 3 Sep 2010 21:17:09 -0400 Subject: x86, fpu: Merge fpu_init() Make fpu_init() handle 32-bit setup. Signed-off-by: Brian Gerst Acked-by: Pekka Enberg Cc: Suresh Siddha LKML-Reference: <1283563039-3466-3-git-send-email-brgerst@gmail.com> Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/common.c | 7 ------- arch/x86/kernel/i387.c | 30 ++++++++++++++---------------- arch/x86/kernel/traps.c | 12 ------------ 3 files changed, 14 insertions(+), 35 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 490dac63c2d..f9e23e8639e 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -1264,13 +1264,6 @@ void __cpuinit cpu_init(void) clear_all_debug_regs(); dbg_restore_debug_regs(); - /* - * Force FPU initialization: - */ - current_thread_info()->status = 0; - clear_used_math(); - mxcsr_feature_mask_init(); - fpu_init(); xsave_init(); } diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index 2605c50b11d..82166519497 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -85,7 +85,6 @@ static void __cpuinit init_thread_xstate(void) #endif } -#ifdef CONFIG_X86_64 /* * Called at bootup to set up the initial FPU state that is later cloned * into all processes. @@ -93,12 +92,21 @@ static void __cpuinit init_thread_xstate(void) void __cpuinit fpu_init(void) { - unsigned long oldcr0 = read_cr0(); - - set_in_cr4(X86_CR4_OSFXSR); - set_in_cr4(X86_CR4_OSXMMEXCPT); + unsigned long cr0; + unsigned long cr4_mask = 0; - write_cr0(oldcr0 & ~(X86_CR0_TS|X86_CR0_EM)); /* clear TS and EM */ + if (cpu_has_fxsr) + cr4_mask |= X86_CR4_OSFXSR; + if (cpu_has_xmm) + cr4_mask |= X86_CR4_OSXMMEXCPT; + if (cr4_mask) + set_in_cr4(cr4_mask); + + cr0 = read_cr0(); + cr0 &= ~(X86_CR0_TS|X86_CR0_EM); /* clear TS and EM */ + if (!HAVE_HWFP) + cr0 |= X86_CR0_EM; + write_cr0(cr0); if (!smp_processor_id()) init_thread_xstate(); @@ -109,16 +117,6 @@ void __cpuinit fpu_init(void) clear_used_math(); } -#else /* CONFIG_X86_64 */ - -void __cpuinit fpu_init(void) -{ - if (!smp_processor_id()) - init_thread_xstate(); -} - -#endif /* CONFIG_X86_32 */ - void fpu_finit(struct fpu *fpu) { #ifdef CONFIG_X86_32 diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index 60788dee0f8..d0029eb5858 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -881,18 +881,6 @@ void __init trap_init(void) #endif #ifdef CONFIG_X86_32 - if (cpu_has_fxsr) { - printk(KERN_INFO "Enabling fast FPU save and restore... "); - set_in_cr4(X86_CR4_OSFXSR); - printk("done.\n"); - } - if (cpu_has_xmm) { - printk(KERN_INFO - "Enabling unmasked SIMD FPU exception support... "); - set_in_cr4(X86_CR4_OSXMMEXCPT); - printk("done.\n"); - } - set_system_trap_gate(SYSCALL_VECTOR, &system_call); set_bit(SYSCALL_VECTOR, used_vectors); #endif -- cgit v1.2.3-70-g09d2 From 51115d4d45700fc7c08306f7ba6e68551f526ae5 Mon Sep 17 00:00:00 2001 From: Brian Gerst Date: Fri, 3 Sep 2010 21:17:10 -0400 Subject: x86, fpu: Merge tolerant_fwait() Commit e2e75c91 merged the math exception handler, allowing both 32-bit and 64-bit to handle math exceptions from kernel mode. Switch to using the 64-bit version of tolerant_fwait() without fnclex, which simply ignores the exception if one is still pending from userspace. Signed-off-by: Brian Gerst Acked-by: Pekka Enberg Cc: Suresh Siddha LKML-Reference: <1283563039-3466-4-git-send-email-brgerst@gmail.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/i387.h | 19 ++++--------------- 1 file changed, 4 insertions(+), 15 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h index a73a8d5a5e6..5d8f9a79fa5 100644 --- a/arch/x86/include/asm/i387.h +++ b/arch/x86/include/asm/i387.h @@ -77,15 +77,6 @@ static inline void sanitize_i387_state(struct task_struct *tsk) } #ifdef CONFIG_X86_64 - -/* Ignore delayed exceptions from user space */ -static inline void tolerant_fwait(void) -{ - asm volatile("1: fwait\n" - "2:\n" - _ASM_EXTABLE(1b, 2b)); -} - static inline int fxrstor_checking(struct i387_fxsave_struct *fx) { int err; @@ -220,11 +211,6 @@ extern void finit_soft_fpu(struct i387_soft_struct *soft); static inline void finit_soft_fpu(struct i387_soft_struct *soft) {} #endif -static inline void tolerant_fwait(void) -{ - asm volatile("fnclex ; fwait"); -} - /* perform fxrstor iff the processor has extended states, otherwise frstor */ static inline int fxrstor_checking(struct i387_fxsave_struct *fx) { @@ -344,7 +330,10 @@ static inline void __unlazy_fpu(struct task_struct *tsk) static inline void __clear_fpu(struct task_struct *tsk) { if (task_thread_info(tsk)->status & TS_USEDFPU) { - tolerant_fwait(); + /* Ignore delayed exceptions from user space */ + asm volatile("1: fwait\n" + "2:\n" + _ASM_EXTABLE(1b, 2b)); task_thread_info(tsk)->status &= ~TS_USEDFPU; stts(); } -- cgit v1.2.3-70-g09d2 From bfd946cb891800d408decaae268a3480775178a3 Mon Sep 17 00:00:00 2001 From: Brian Gerst Date: Fri, 3 Sep 2010 21:17:11 -0400 Subject: x86, fpu: Merge __save_init_fpu() __save_init_fpu() is identical for 32-bit and 64-bit. Signed-off-by: Brian Gerst Acked-by: Pekka Enberg Cc: Suresh Siddha LKML-Reference: <1283563039-3466-5-git-send-email-brgerst@gmail.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/i387.h | 11 ++--------- 1 file changed, 2 insertions(+), 9 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h index 5d8f9a79fa5..88065e3a03c 100644 --- a/arch/x86/include/asm/i387.h +++ b/arch/x86/include/asm/i387.h @@ -197,12 +197,6 @@ static inline void fpu_save_init(struct fpu *fpu) fpu_clear(fpu); } -static inline void __save_init_fpu(struct task_struct *tsk) -{ - fpu_save_init(&tsk->thread.fpu); - task_thread_info(tsk)->status &= ~TS_USEDFPU; -} - #else /* CONFIG_X86_32 */ #ifdef CONFIG_MATH_EMULATION @@ -285,15 +279,14 @@ end: ; } +#endif /* CONFIG_X86_64 */ + static inline void __save_init_fpu(struct task_struct *tsk) { fpu_save_init(&tsk->thread.fpu); task_thread_info(tsk)->status &= ~TS_USEDFPU; } - -#endif /* CONFIG_X86_64 */ - static inline int fpu_fxrstor_checking(struct fpu *fpu) { return fxrstor_checking(&fpu->state->fxsave); -- cgit v1.2.3-70-g09d2 From a4d4fbc7735bba6654b20f859135f9d3f8fe7f76 Mon Sep 17 00:00:00 2001 From: Brian Gerst Date: Fri, 3 Sep 2010 21:17:12 -0400 Subject: x86-64, fpu: Disable preemption when using TS_USEDFPU Consolidates code and fixes the below race for 64-bit. commit 9fa2f37bfeb798728241cc4a19578ce6e4258f25 Author: torvalds Date: Tue Sep 2 07:37:25 2003 +0000 Be a lot more careful about TS_USEDFPU and preemption We had some races where we testecd (or set) TS_USEDFPU together with sequences that depended on the setting (like clearing or setting the TS flag in %cr0) and we could be preempted in between, which screws up the FPU state, since preemption will itself change USEDFPU and the TS flag. This makes it a lot more explicit: the "internal" low-level FPU functions ("__xxxx_fpu()") all require preemption to be disabled, and the exported "real" functions will make sure that is the case. One case - in __switch_to() - was switched to the non-preempt-safe internal version, since the scheduler itself has already disabled preemption. BKrev: 3f5448b5WRiQuyzAlbajs3qoQjSobw Signed-off-by: Brian Gerst Acked-by: Pekka Enberg Cc: Suresh Siddha LKML-Reference: <1283563039-3466-6-git-send-email-brgerst@gmail.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/i387.h | 15 --------------- arch/x86/kernel/process_64.c | 2 +- 2 files changed, 1 insertion(+), 16 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h index 88065e3a03c..8b40a8379cc 100644 --- a/arch/x86/include/asm/i387.h +++ b/arch/x86/include/asm/i387.h @@ -387,19 +387,6 @@ static inline void irq_ts_restore(int TS_state) stts(); } -#ifdef CONFIG_X86_64 - -static inline void save_init_fpu(struct task_struct *tsk) -{ - __save_init_fpu(tsk); - stts(); -} - -#define unlazy_fpu __unlazy_fpu -#define clear_fpu __clear_fpu - -#else /* CONFIG_X86_32 */ - /* * These disable preemption on their own and are safe */ @@ -425,8 +412,6 @@ static inline void clear_fpu(struct task_struct *tsk) preempt_enable(); } -#endif /* CONFIG_X86_64 */ - /* * i387 state interaction */ diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 3d9ea531ddd..b3d7a3a04f3 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -424,7 +424,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) load_TLS(next, cpu); /* Must be after DS reload */ - unlazy_fpu(prev_p); + __unlazy_fpu(prev_p); /* Make sure cpu is ready for new context */ if (preload_fpu) -- cgit v1.2.3-70-g09d2 From 10c11f304986a1f84201c2261a428701f9d2dffc Mon Sep 17 00:00:00 2001 From: Brian Gerst Date: Fri, 3 Sep 2010 21:17:13 -0400 Subject: x86-64, fpu: Fix %cs value in convert_from_fxsr() While %ds still contains the userspace selector, %cs is KERNEL_CS at this point. Always get %cs from pt_regs even for the current task. Signed-off-by: Brian Gerst Acked-by: Pekka Enberg Cc: Suresh Siddha LKML-Reference: <1283563039-3466-7-git-send-email-brgerst@gmail.com> Signed-off-by: H. Peter Anvin --- arch/x86/kernel/i387.c | 18 ++++++++---------- 1 file changed, 8 insertions(+), 10 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index 82166519497..f3775f5ef4a 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -389,19 +389,17 @@ convert_from_fxsr(struct user_i387_ia32_struct *env, struct task_struct *tsk) #ifdef CONFIG_X86_64 env->fip = fxsave->rip; env->foo = fxsave->rdp; + /* + * should be actually ds/cs at fpu exception time, but + * that information is not available in 64bit mode. + */ + env->fcs = task_pt_regs(tsk)->cs; if (tsk == current) { - /* - * should be actually ds/cs at fpu exception time, but - * that information is not available in 64bit mode. - */ - asm("mov %%ds, %[fos]" : [fos] "=r" (env->fos)); - asm("mov %%cs, %[fcs]" : [fcs] "=r" (env->fcs)); + savesegment(ds, env->fos); } else { - struct pt_regs *regs = task_pt_regs(tsk); - - env->fos = 0xffff0000 | tsk->thread.ds; - env->fcs = regs->cs; + env->fos = tsk->thread.ds; } + env->fos |= 0xffff0000; #else env->fip = fxsave->fip; env->fcs = (u16) fxsave->fcs | ((u32) fxsave->fop << 16); -- cgit v1.2.3-70-g09d2 From 820241356d6aa9a895fc10def15794a5a5bfcd98 Mon Sep 17 00:00:00 2001 From: Brian Gerst Date: Fri, 3 Sep 2010 21:17:14 -0400 Subject: x86-64, fpu: Simplify constraints for fxsave/fxtstor Use the "R" constraint (legacy register) instead of listing all the possible registers. Clean up the comments as well. Signed-off-by: Brian Gerst Acked-by: Pekka Enberg Cc: Suresh Siddha LKML-Reference: <1283563039-3466-8-git-send-email-brgerst@gmail.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/i387.h | 44 +++++++++++++++++--------------------------- 1 file changed, 17 insertions(+), 27 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h index 8b40a8379cc..768fcb25900 100644 --- a/arch/x86/include/asm/i387.h +++ b/arch/x86/include/asm/i387.h @@ -81,6 +81,7 @@ static inline int fxrstor_checking(struct i387_fxsave_struct *fx) { int err; + /* See comment in fxsave() below. */ asm volatile("1: rex64/fxrstor (%[fx])\n\t" "2:\n" ".section .fixup,\"ax\"\n" @@ -89,11 +90,7 @@ static inline int fxrstor_checking(struct i387_fxsave_struct *fx) ".previous\n" _ASM_EXTABLE(1b, 3b) : [err] "=r" (err) -#if 0 /* See comment in fxsave() below. */ - : [fx] "r" (fx), "m" (*fx), "0" (0)); -#else - : [fx] "cdaSDb" (fx), "m" (*fx), "0" (0)); -#endif + : [fx] "R" (fx), "m" (*fx), "0" (0)); return err; } @@ -140,6 +137,7 @@ static inline int fxsave_user(struct i387_fxsave_struct __user *fx) if (unlikely(err)) return -EFAULT; + /* See comment in fxsave() below. */ asm volatile("1: rex64/fxsave (%[fx])\n\t" "2:\n" ".section .fixup,\"ax\"\n" @@ -148,11 +146,7 @@ static inline int fxsave_user(struct i387_fxsave_struct __user *fx) ".previous\n" _ASM_EXTABLE(1b, 3b) : [err] "=r" (err), "=m" (*fx) -#if 0 /* See comment in fxsave() below. */ - : [fx] "r" (fx), "0" (0)); -#else - : [fx] "cdaSDb" (fx), "0" (0)); -#endif + : [fx] "R" (fx), "0" (0)); if (unlikely(err) && __clear_user(fx, sizeof(struct i387_fxsave_struct))) err = -EFAULT; @@ -165,26 +159,22 @@ static inline void fpu_fxsave(struct fpu *fpu) /* Using "rex64; fxsave %0" is broken because, if the memory operand uses any extended registers for addressing, a second REX prefix will be generated (to the assembler, rex64 followed by semicolon - is a separate instruction), and hence the 64-bitness is lost. */ -#if 0 - /* Using "fxsaveq %0" would be the ideal choice, but is only supported - starting with gas 2.16. */ - __asm__ __volatile__("fxsaveq %0" - : "=m" (fpu->state->fxsave)); -#elif 0 - /* Using, as a workaround, the properly prefixed form below isn't + is a separate instruction), and hence the 64-bitness is lost. + Using "fxsaveq %0" would be the ideal choice, but is only supported + starting with gas 2.16. + asm volatile("fxsaveq %0" + : "=m" (fpu->state->fxsave)); + Using, as a workaround, the properly prefixed form below isn't accepted by any binutils version so far released, complaining that the same type of prefix is used twice if an extended register is - needed for addressing (fix submitted to mainline 2005-11-21). */ - __asm__ __volatile__("rex64/fxsave %0" - : "=m" (fpu->state->fxsave)); -#else - /* This, however, we can work around by forcing the compiler to select + needed for addressing (fix submitted to mainline 2005-11-21). + asm volatile("rex64/fxsave %0" + : "=m" (fpu->state->fxsave)); + This, however, we can work around by forcing the compiler to select an addressing mode that doesn't require extended registers. */ - __asm__ __volatile__("rex64/fxsave (%1)" - : "=m" (fpu->state->fxsave) - : "cdaSDb" (&fpu->state->fxsave)); -#endif + asm volatile("rex64/fxsave (%[fx])" + : "=m" (fpu->state->fxsave) + : [fx] "R" (&fpu->state->fxsave)); } static inline void fpu_save_init(struct fpu *fpu) -- cgit v1.2.3-70-g09d2 From a334fe43d85f570ae907acf988a053c5eff78d6e Mon Sep 17 00:00:00 2001 From: Brian Gerst Date: Fri, 3 Sep 2010 21:17:15 -0400 Subject: x86-32, fpu: Remove math_emulate stub check_fpu() in bugs.c halts boot if no FPU is found and math emulation isn't enabled. Therefore this stub will never be used. Signed-off-by: Brian Gerst Acked-by: Pekka Enberg Cc: Suresh Siddha LKML-Reference: <1283563039-3466-9-git-send-email-brgerst@gmail.com> Signed-off-by: H. Peter Anvin --- arch/x86/kernel/traps.c | 23 ++++++----------------- 1 file changed, 6 insertions(+), 17 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index d0029eb5858..d43968503dd 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -776,21 +776,10 @@ asmlinkage void math_state_restore(void) } EXPORT_SYMBOL_GPL(math_state_restore); -#ifndef CONFIG_MATH_EMULATION -void math_emulate(struct math_emu_info *info) -{ - printk(KERN_EMERG - "math-emulation not enabled and no coprocessor found.\n"); - printk(KERN_EMERG "killing %s.\n", current->comm); - force_sig(SIGFPE, current); - schedule(); -} -#endif /* CONFIG_MATH_EMULATION */ - dotraplinkage void __kprobes do_device_not_available(struct pt_regs *regs, long error_code) { -#ifdef CONFIG_X86_32 +#ifdef CONFIG_MATH_EMULATION if (read_cr0() & X86_CR0_EM) { struct math_emu_info info = { }; @@ -798,12 +787,12 @@ do_device_not_available(struct pt_regs *regs, long error_code) info.regs = regs; math_emulate(&info); - } else { - math_state_restore(); /* interrupts still off */ - conditional_sti(regs); + return; } -#else - math_state_restore(); +#endif + math_state_restore(); /* interrupts still off */ +#ifdef CONFIG_X86_32 + conditional_sti(regs); #endif } -- cgit v1.2.3-70-g09d2 From 8eb91a577d7763d21628f6761045328784b1911c Mon Sep 17 00:00:00 2001 From: Brian Gerst Date: Fri, 3 Sep 2010 21:17:16 -0400 Subject: x86, fpu: Remove unnecessary ifdefs from i387 code. Remove ifdefs for code that the compiler can optimize away on 64-bit. Signed-off-by: Brian Gerst Acked-by: Pekka Enberg Cc: Suresh Siddha LKML-Reference: <1283563039-3466-10-git-send-email-brgerst@gmail.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/i387.h | 12 ++++++------ arch/x86/kernel/i387.c | 4 ---- 2 files changed, 6 insertions(+), 10 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h index 768fcb25900..42b507e16d1 100644 --- a/arch/x86/include/asm/i387.h +++ b/arch/x86/include/asm/i387.h @@ -55,6 +55,12 @@ extern int save_i387_xstate_ia32(void __user *buf); extern int restore_i387_xstate_ia32(void __user *buf); #endif +#ifdef CONFIG_MATH_EMULATION +extern void finit_soft_fpu(struct i387_soft_struct *soft); +#else +static inline void finit_soft_fpu(struct i387_soft_struct *soft) {} +#endif + #define X87_FSW_ES (1 << 7) /* Exception Summary */ static __always_inline __pure bool use_xsaveopt(void) @@ -189,12 +195,6 @@ static inline void fpu_save_init(struct fpu *fpu) #else /* CONFIG_X86_32 */ -#ifdef CONFIG_MATH_EMULATION -extern void finit_soft_fpu(struct i387_soft_struct *soft); -#else -static inline void finit_soft_fpu(struct i387_soft_struct *soft) {} -#endif - /* perform fxrstor iff the processor has extended states, otherwise frstor */ static inline int fxrstor_checking(struct i387_fxsave_struct *fx) { diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index f3775f5ef4a..e795e360bd5 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -79,10 +79,8 @@ static void __cpuinit init_thread_xstate(void) if (cpu_has_fxsr) xstate_size = sizeof(struct i387_fxsave_struct); -#ifdef CONFIG_X86_32 else xstate_size = sizeof(struct i387_fsave_struct); -#endif } /* @@ -119,12 +117,10 @@ void __cpuinit fpu_init(void) void fpu_finit(struct fpu *fpu) { -#ifdef CONFIG_X86_32 if (!HAVE_HWFP) { finit_soft_fpu(&fpu->state->soft); return; } -#endif if (cpu_has_fxsr) { struct i387_fxsave_struct *fx = &fpu->state->fxsave; -- cgit v1.2.3-70-g09d2 From eec73f813ab0954253e5e2168119c4555f83f07d Mon Sep 17 00:00:00 2001 From: Brian Gerst Date: Fri, 3 Sep 2010 21:17:17 -0400 Subject: x86, fpu: Remove PSHUFB_XMM5_* macros The PSHUFB_XMM5_* macros are no longer used. Signed-off-by: Brian Gerst Acked-by: Pekka Enberg Cc: Suresh Siddha LKML-Reference: <1283563039-3466-11-git-send-email-brgerst@gmail.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/i387.h | 3 --- 1 file changed, 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h index 42b507e16d1..907967e4fa8 100644 --- a/arch/x86/include/asm/i387.h +++ b/arch/x86/include/asm/i387.h @@ -465,7 +465,4 @@ extern void fpu_finit(struct fpu *fpu); #endif /* __ASSEMBLY__ */ -#define PSHUFB_XMM5_XMM0 .byte 0x66, 0x0f, 0x38, 0x00, 0xc5 -#define PSHUFB_XMM5_XMM6 .byte 0x66, 0x0f, 0x38, 0x00, 0xf5 - #endif /* _ASM_X86_I387_H */ -- cgit v1.2.3-70-g09d2 From 58a992b9cbaf449aeebd3575c3695a9eb5d95b5e Mon Sep 17 00:00:00 2001 From: Brian Gerst Date: Fri, 3 Sep 2010 21:17:18 -0400 Subject: x86-32, fpu: Rewrite fpu_save_init() Rewrite fpu_save_init() to prepare for merging with 64-bit. Signed-off-by: Brian Gerst Acked-by: Pekka Enberg Cc: Suresh Siddha LKML-Reference: <1283563039-3466-12-git-send-email-brgerst@gmail.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/i387.h | 47 +++++++++++++++++++++------------------------ 1 file changed, 22 insertions(+), 25 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h index 907967e4fa8..b45abefb89f 100644 --- a/arch/x86/include/asm/i387.h +++ b/arch/x86/include/asm/i387.h @@ -73,6 +73,11 @@ static __always_inline __pure bool use_xsave(void) return static_cpu_has(X86_FEATURE_XSAVE); } +static __always_inline __pure bool use_fxsr(void) +{ + return static_cpu_has(X86_FEATURE_FXSR); +} + extern void __sanitize_i387_state(struct task_struct *); static inline void sanitize_i387_state(struct task_struct *tsk) @@ -211,6 +216,12 @@ static inline int fxrstor_checking(struct i387_fxsave_struct *fx) return 0; } +static inline void fpu_fxsave(struct fpu *fpu) +{ + asm volatile("fxsave %[fx]" + : [fx] "=m" (fpu->state->fxsave)); +} + /* We need a safe address that is cheap to find and that is already in L1 during context switch. The best choices are unfortunately different for UP and SMP */ @@ -226,36 +237,24 @@ static inline int fxrstor_checking(struct i387_fxsave_struct *fx) static inline void fpu_save_init(struct fpu *fpu) { if (use_xsave()) { - struct xsave_struct *xstate = &fpu->state->xsave; - struct i387_fxsave_struct *fx = &fpu->state->fxsave; - fpu_xsave(fpu); /* * xsave header may indicate the init state of the FP. */ - if (!(xstate->xsave_hdr.xstate_bv & XSTATE_FP)) - goto end; - - if (unlikely(fx->swd & X87_FSW_ES)) - asm volatile("fnclex"); - - /* - * we can do a simple return here or be paranoid :) - */ - goto clear_state; + if (!(fpu->state->xsave.xsave_hdr.xstate_bv & XSTATE_FP)) + return; + } else if (use_fxsr()) { + fpu_fxsave(fpu); + } else { + asm volatile("fsave %[fx]; fwait" + : [fx] "=m" (fpu->state->fsave)); + return; } - /* Use more nops than strictly needed in case the compiler - varies code */ - alternative_input( - "fnsave %[fx] ;fwait;" GENERIC_NOP8 GENERIC_NOP4, - "fxsave %[fx]\n" - "bt $7,%[fsw] ; jnc 1f ; fnclex\n1:", - X86_FEATURE_FXSR, - [fx] "m" (fpu->state->fxsave), - [fsw] "m" (fpu->state->fxsave.swd) : "memory"); -clear_state: + if (unlikely(fpu->state->fxsave.swd & X87_FSW_ES)) + asm volatile("fnclex"); + /* AMD K7/K8 CPUs don't save/restore FDP/FIP/FOP unless an exception is pending. Clear the x87 state here by setting it to fixed values. safe_address is a random variable that should be in L1 */ @@ -265,8 +264,6 @@ clear_state: "fildl %[addr]", /* set F?P to defined value */ X86_FEATURE_FXSAVE_LEAK, [addr] "m" (safe_address)); -end: - ; } #endif /* CONFIG_X86_64 */ -- cgit v1.2.3-70-g09d2 From b2b57fe053c9cf8b8af5a0e826a465996afed0ff Mon Sep 17 00:00:00 2001 From: Brian Gerst Date: Fri, 3 Sep 2010 21:17:19 -0400 Subject: x86, fpu: Merge fpu_save_init() Make 64-bit use the 32-bit version of fpu_save_init(). Remove unused clear_fpu_state(). Signed-off-by: Brian Gerst Acked-by: Pekka Enberg Cc: Suresh Siddha LKML-Reference: <1283563039-3466-13-git-send-email-brgerst@gmail.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/i387.h | 48 ++++----------------------------------------- 1 file changed, 4 insertions(+), 44 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h index b45abefb89f..70626ed96cb 100644 --- a/arch/x86/include/asm/i387.h +++ b/arch/x86/include/asm/i387.h @@ -105,36 +105,6 @@ static inline int fxrstor_checking(struct i387_fxsave_struct *fx) return err; } -/* AMD CPUs don't save/restore FDP/FIP/FOP unless an exception - is pending. Clear the x87 state here by setting it to fixed - values. The kernel data segment can be sometimes 0 and sometimes - new user value. Both should be ok. - Use the PDA as safe address because it should be already in L1. */ -static inline void fpu_clear(struct fpu *fpu) -{ - struct xsave_struct *xstate = &fpu->state->xsave; - struct i387_fxsave_struct *fx = &fpu->state->fxsave; - - /* - * xsave header may indicate the init state of the FP. - */ - if (use_xsave() && - !(xstate->xsave_hdr.xstate_bv & XSTATE_FP)) - return; - - if (unlikely(fx->swd & X87_FSW_ES)) - asm volatile("fnclex"); - alternative_input(ASM_NOP8 ASM_NOP2, - " emms\n" /* clear stack tags */ - " fildl %%gs:0", /* load to clear state */ - X86_FEATURE_FXSAVE_LEAK); -} - -static inline void clear_fpu_state(struct task_struct *tsk) -{ - fpu_clear(&tsk->thread.fpu); -} - static inline int fxsave_user(struct i387_fxsave_struct __user *fx) { int err; @@ -188,16 +158,6 @@ static inline void fpu_fxsave(struct fpu *fpu) : [fx] "R" (&fpu->state->fxsave)); } -static inline void fpu_save_init(struct fpu *fpu) -{ - if (use_xsave()) - fpu_xsave(fpu); - else - fpu_fxsave(fpu); - - fpu_clear(fpu); -} - #else /* CONFIG_X86_32 */ /* perform fxrstor iff the processor has extended states, otherwise frstor */ @@ -222,6 +182,8 @@ static inline void fpu_fxsave(struct fpu *fpu) : [fx] "=m" (fpu->state->fxsave)); } +#endif /* CONFIG_X86_64 */ + /* We need a safe address that is cheap to find and that is already in L1 during context switch. The best choices are unfortunately different for UP and SMP */ @@ -259,15 +221,13 @@ static inline void fpu_save_init(struct fpu *fpu) is pending. Clear the x87 state here by setting it to fixed values. safe_address is a random variable that should be in L1 */ alternative_input( - GENERIC_NOP8 GENERIC_NOP2, + ASM_NOP8 ASM_NOP2, "emms\n\t" /* clear stack tags */ - "fildl %[addr]", /* set F?P to defined value */ + "fildl %P[addr]", /* set F?P to defined value */ X86_FEATURE_FXSAVE_LEAK, [addr] "m" (safe_address)); } -#endif /* CONFIG_X86_64 */ - static inline void __save_init_fpu(struct task_struct *tsk) { fpu_save_init(&tsk->thread.fpu); -- cgit v1.2.3-70-g09d2 From db7829c6cc32f3c0c9a324118d743acb1abff081 Mon Sep 17 00:00:00 2001 From: Brian Gerst Date: Thu, 9 Sep 2010 18:17:26 +0200 Subject: x86, percpu: Optimize this_cpu_ptr Allow arches to implement __this_cpu_ptr, and provide an x86 version. Before: movq $foo, %rax movq %gs:this_cpu_off, %rdx addq %rdx, %rax After: movq $foo, %rax addq %gs:this_cpu_off, %rax The benefit is doing it in one less instruction and not clobbering a temporary register. tj: * Beefed up the comment a bit and renamed in-macro temp variable to match neighboring macros. * Folded fix for const pointer case found in linux-next. * Fixed sparse notation. Signed-off-by: Brian Gerst Signed-off-by: Tejun Heo --- arch/x86/include/asm/percpu.h | 14 ++++++++++++++ include/asm-generic/percpu.h | 9 +++++++-- 2 files changed, 21 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/percpu.h b/arch/x86/include/asm/percpu.h index cd28f9ad910..f899e01a8ac 100644 --- a/arch/x86/include/asm/percpu.h +++ b/arch/x86/include/asm/percpu.h @@ -47,6 +47,20 @@ #ifdef CONFIG_SMP #define __percpu_arg(x) "%%"__stringify(__percpu_seg)":%P" #x #define __my_cpu_offset percpu_read(this_cpu_off) + +/* + * Compared to the generic __my_cpu_offset version, the following + * saves one instruction and avoids clobbering a temp register. + */ +#define __this_cpu_ptr(ptr) \ +({ \ + unsigned long tcp_ptr__; \ + __verify_pcpu_ptr(ptr); \ + asm volatile("add " __percpu_arg(1) ", %0" \ + : "=r" (tcp_ptr__) \ + : "m" (this_cpu_off), "0" (ptr)); \ + (typeof(*(ptr)) __kernel __force *)tcp_ptr__; \ +}) #else #define __percpu_arg(x) "%P" #x #endif diff --git a/include/asm-generic/percpu.h b/include/asm-generic/percpu.h index 08923b68476..ec643115bb5 100644 --- a/include/asm-generic/percpu.h +++ b/include/asm-generic/percpu.h @@ -60,9 +60,14 @@ extern unsigned long __per_cpu_offset[NR_CPUS]; #define __raw_get_cpu_var(var) \ (*SHIFT_PERCPU_PTR(&(var), __my_cpu_offset)) -#define this_cpu_ptr(ptr) SHIFT_PERCPU_PTR(ptr, my_cpu_offset) +#ifndef __this_cpu_ptr #define __this_cpu_ptr(ptr) SHIFT_PERCPU_PTR(ptr, __my_cpu_offset) - +#endif +#ifdef CONFIG_DEBUG_PREEMPT +#define this_cpu_ptr(ptr) SHIFT_PERCPU_PTR(ptr, my_cpu_offset) +#else +#define this_cpu_ptr(ptr) __this_cpu_ptr(ptr) +#endif #ifdef CONFIG_HAVE_SETUP_PER_CPU_AREA extern void setup_per_cpu_areas(void); -- cgit v1.2.3-70-g09d2 From a7f07cfbaa1dd5bf9e615948f280c92e7928e6f7 Mon Sep 17 00:00:00 2001 From: Venkatesh Pallipadi Date: Fri, 10 Sep 2010 15:55:49 -0700 Subject: x86, mtrr: Refactor MTRR type overlap check code Move the MTRR type overlap check into a new function. No functional change in this patch. Just making it easier to add multiple region overlap check in the following patch. Signed-off-by: Venkatesh Pallipadi LKML-Reference: <1284159350-19841-2-git-send-email-venki@google.com> Reviewed-by: Suresh Siddha Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/mtrr/generic.c | 44 +++++++++++++++++++++++++------------- 1 file changed, 29 insertions(+), 15 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c index 7d28d7d0388..14f4f0c0329 100644 --- a/arch/x86/kernel/cpu/mtrr/generic.c +++ b/arch/x86/kernel/cpu/mtrr/generic.c @@ -64,6 +64,33 @@ static inline void k8_check_syscfg_dram_mod_en(void) } } +/* + * Check and return the effective type for MTRR-MTRR type overlap. + * Returns 1 if the effective type is UNCACHEABLE, else returns 0 + */ +static int check_type_overlap(u8 *prev, u8 *curr) +{ + if (*prev == MTRR_TYPE_UNCACHABLE || *curr == MTRR_TYPE_UNCACHABLE) { + *prev = MTRR_TYPE_UNCACHABLE; + *curr = MTRR_TYPE_UNCACHABLE; + return 1; + } + + if ((*prev == MTRR_TYPE_WRBACK && *curr == MTRR_TYPE_WRTHROUGH) || + (*prev == MTRR_TYPE_WRTHROUGH && *curr == MTRR_TYPE_WRBACK)) { + *prev = MTRR_TYPE_WRTHROUGH; + *curr = MTRR_TYPE_WRTHROUGH; + } + + if (*prev != *curr) { + *prev = MTRR_TYPE_UNCACHABLE; + *curr = MTRR_TYPE_UNCACHABLE; + return 1; + } + + return 0; +} + /* * Returns the effective MTRR type for the region * Error returns: @@ -138,21 +165,8 @@ u8 mtrr_type_lookup(u64 start, u64 end) continue; } - if (prev_match == MTRR_TYPE_UNCACHABLE || - curr_match == MTRR_TYPE_UNCACHABLE) { - return MTRR_TYPE_UNCACHABLE; - } - - if ((prev_match == MTRR_TYPE_WRBACK && - curr_match == MTRR_TYPE_WRTHROUGH) || - (prev_match == MTRR_TYPE_WRTHROUGH && - curr_match == MTRR_TYPE_WRBACK)) { - prev_match = MTRR_TYPE_WRTHROUGH; - curr_match = MTRR_TYPE_WRTHROUGH; - } - - if (prev_match != curr_match) - return MTRR_TYPE_UNCACHABLE; + if (check_type_overlap(&prev_match, &curr_match)) + return curr_match; } if (mtrr_tom2) { -- cgit v1.2.3-70-g09d2 From 351e5a703ad994405bd900da330823d3b4a372e0 Mon Sep 17 00:00:00 2001 From: Venkatesh Pallipadi Date: Fri, 10 Sep 2010 15:55:50 -0700 Subject: x86, mtrr: Support mtrr lookup for range spanning across MTRR range mtrr_type_lookup [start:end] looked up the resultant MTRR type for that range, based on fixed and all variable MTRR ranges. It did check for multiple MTRR var ranges overlapping [start:end] and returned the net type. However, if the [start:end] range spanned across any var MTRR range, mtrr_type_lookup would return an error return of 0xFE. This was based on typical usage of mtrr_type_lookup in PAT mapping, where region being mapped would not normally span across MTRR ranges and also trying to keep the code simple. Mark recently reported the problem with this limitation. When there are two continguous MTRR's of type "writeback" and if there is a memory mapping over a region starting in one MTRR range and ending in another MTRR range, such mapping will fallback to "uncached" due to the above limitation. Change below adds support for such lookups spanning multiple MTRR ranges. We now have a wrapper mtrr_type_lookup that dynamically splits such a region into smaller chunks that fit within one MTRR range and does a __mtrr_type_lookup on it and combine the results later. Reported-by: Mark Langsdorf Signed-off-by: Venkatesh Pallipadi LKML-Reference: <1284159350-19841-3-git-send-email-venki@google.com> Reviewed-by: Suresh Siddha Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/mtrr/generic.c | 84 ++++++++++++++++++++++++++++++++++---- 1 file changed, 77 insertions(+), 7 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c index 14f4f0c0329..9f27228ceff 100644 --- a/arch/x86/kernel/cpu/mtrr/generic.c +++ b/arch/x86/kernel/cpu/mtrr/generic.c @@ -64,6 +64,18 @@ static inline void k8_check_syscfg_dram_mod_en(void) } } +/* Get the size of contiguous MTRR range */ +static u64 get_mtrr_size(u64 mask) +{ + u64 size; + + mask >>= PAGE_SHIFT; + mask |= size_or_mask; + size = -mask; + size <<= PAGE_SHIFT; + return size; +} + /* * Check and return the effective type for MTRR-MTRR type overlap. * Returns 1 if the effective type is UNCACHEABLE, else returns 0 @@ -92,17 +104,19 @@ static int check_type_overlap(u8 *prev, u8 *curr) } /* - * Returns the effective MTRR type for the region - * Error returns: - * - 0xFE - when the range is "not entirely covered" by _any_ var range MTRR - * - 0xFF - when MTRR is not enabled + * Error/Semi-error returns: + * 0xFF - when MTRR is not enabled + * *repeat == 1 implies [start:end] spanned across MTRR range and type returned + * corresponds only to [start:*partial_end]. + * Caller has to lookup again for [*partial_end:end]. */ -u8 mtrr_type_lookup(u64 start, u64 end) +static u8 __mtrr_type_lookup(u64 start, u64 end, u64 *partial_end, int *repeat) { int i; u64 base, mask; u8 prev_match, curr_match; + *repeat = 0; if (!mtrr_state_set) return 0xFF; @@ -153,8 +167,34 @@ u8 mtrr_type_lookup(u64 start, u64 end) start_state = ((start & mask) == (base & mask)); end_state = ((end & mask) == (base & mask)); - if (start_state != end_state) - return 0xFE; + + if (start_state != end_state) { + /* + * We have start:end spanning across an MTRR. + * We split the region into + * either + * (start:mtrr_end) (mtrr_end:end) + * or + * (start:mtrr_start) (mtrr_start:end) + * depending on kind of overlap. + * Return the type for first region and a pointer to + * the start of second region so that caller will + * lookup again on the second region. + * Note: This way we handle multiple overlaps as well. + */ + if (start_state) + *partial_end = base + get_mtrr_size(mask); + else + *partial_end = base; + + if (unlikely(*partial_end <= start)) { + WARN_ON(1); + *partial_end = start + PAGE_SIZE; + } + + end = *partial_end - 1; /* end is inclusive */ + *repeat = 1; + } if ((start & mask) != (base & mask)) continue; @@ -180,6 +220,36 @@ u8 mtrr_type_lookup(u64 start, u64 end) return mtrr_state.def_type; } +/* + * Returns the effective MTRR type for the region + * Error return: + * 0xFF - when MTRR is not enabled + */ +u8 mtrr_type_lookup(u64 start, u64 end) +{ + u8 type, prev_type; + int repeat; + u64 partial_end; + + type = __mtrr_type_lookup(start, end, &partial_end, &repeat); + + /* + * Common path is with repeat = 0. + * However, we can have cases where [start:end] spans across some + * MTRR range. Do repeated lookups for that case here. + */ + while (repeat) { + prev_type = type; + start = partial_end; + type = __mtrr_type_lookup(start, end, &partial_end, &repeat); + + if (check_type_overlap(&prev_type, &type)) + return type; + } + + return type; +} + /* Get the MSR pair relating to a var range */ static void get_mtrr_var_range(unsigned int index, struct mtrr_var_range *vr) -- cgit v1.2.3-70-g09d2 From d0ed0c32662e756e7daf85e70a5a27a9c1111331 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Sat, 11 Sep 2010 22:10:54 -0700 Subject: x86: Remove pr_ uses of KERN_ Signed-off-by: Joe Perches Cc: Jiri Kosina LKML-Reference: Signed-off-by: Ingo Molnar --- arch/x86/kernel/apb_timer.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apb_timer.c b/arch/x86/kernel/apb_timer.c index 8dd77800ff5..4004417a6b5 100644 --- a/arch/x86/kernel/apb_timer.c +++ b/arch/x86/kernel/apb_timer.c @@ -398,7 +398,7 @@ static int apbt_cpuhp_notify(struct notifier_block *n, } break; default: - pr_debug(KERN_INFO "APBT notified %lu, no action\n", action); + pr_debug("APBT notified %lu, no action\n", action); } return NOTIFY_OK; } @@ -552,7 +552,7 @@ bad_count: pr_debug("APB CS going back %lx:%lx:%lx ", t2, last_read, t2 - last_read); bad_count_x3: - pr_debug(KERN_INFO "tripple check enforced\n"); + pr_debug("triple check enforced\n"); t0 = apbt_readl(phy_cs_timer_id, APBTMR_N_CURRENT_VALUE); udelay(1); -- cgit v1.2.3-70-g09d2 From b0b2072df3b544f56b90173c2cde7a374c51546b Mon Sep 17 00:00:00 2001 From: Stephane Eranian Date: Fri, 10 Sep 2010 13:28:01 +0200 Subject: perf_events: Fix BTS interrupt handling to avoid being dazed by NMI (v2) Fix a bug introduced with commit de725de and the change in the meaning of the return value of intel_pmu_handle_irq(). With the current code, when you are using the BTS, you get 'dazed by NMI' each time the BTS buffer fills up. BTS does interrupt on the PMU vector, thus NMI. You need to take this into account in the return value of the function. This version fixes initial patch which was missing changes to perf_event_intel_ds.c. Signed-off-by: Stephane Eranian Acked-by: Don Zickus Cc: peterz@infradead.org Cc: paulus@samba.org Cc: davem@davemloft.net Cc: fweisbec@gmail.com Cc: perfmon2-devel@lists.sf.net Cc: eranian@gmail.com Cc: robert.richter@amd.com LKML-Reference: <4c8a1686.aae9d80a.5aa4.5e35@mx.google.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event_intel.c | 6 +++--- arch/x86/kernel/cpu/perf_event_intel_ds.c | 11 ++++++----- 2 files changed, 9 insertions(+), 8 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index 82395f2378e..c8f5c088cad 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -713,18 +713,18 @@ static int intel_pmu_handle_irq(struct pt_regs *regs) struct cpu_hw_events *cpuc; int bit, loops; u64 status; - int handled = 0; + int handled; perf_sample_data_init(&data, 0); cpuc = &__get_cpu_var(cpu_hw_events); intel_pmu_disable_all(); - intel_pmu_drain_bts_buffer(); + handled = intel_pmu_drain_bts_buffer(); status = intel_pmu_get_status(); if (!status) { intel_pmu_enable_all(0); - return 0; + return handled; } loops = 0; diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c index 9893a2f77b7..4977f9c400e 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_ds.c +++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c @@ -214,7 +214,7 @@ static void intel_pmu_disable_bts(void) update_debugctlmsr(debugctlmsr); } -static void intel_pmu_drain_bts_buffer(void) +static int intel_pmu_drain_bts_buffer(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct debug_store *ds = cpuc->ds; @@ -231,16 +231,16 @@ static void intel_pmu_drain_bts_buffer(void) struct pt_regs regs; if (!event) - return; + return 0; if (!ds) - return; + return 0; at = (struct bts_record *)(unsigned long)ds->bts_buffer_base; top = (struct bts_record *)(unsigned long)ds->bts_index; if (top <= at) - return; + return 0; ds->bts_index = ds->bts_buffer_base; @@ -256,7 +256,7 @@ static void intel_pmu_drain_bts_buffer(void) perf_prepare_sample(&header, &data, event, ®s); if (perf_output_begin(&handle, event, header.size * (top - at), 1, 1)) - return; + return 1; for (; at < top; at++) { data.ip = at->from; @@ -270,6 +270,7 @@ static void intel_pmu_drain_bts_buffer(void) /* There's new data available. */ event->hw.interrupts++; event->pending_kill = POLL_IN; + return 1; } /* -- cgit v1.2.3-70-g09d2 From 6376b2297502e72255b7eb2893c6044ad5a7b5f4 Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Wed, 15 Sep 2010 10:04:28 +0900 Subject: kprobes: Make functions static Make following (internal) functions static to make sparse happier :-) * get_optimized_kprobe: only called from static functions * kretprobe_table_unlock: _lock function is static * kprobes_optinsn_template_holder: never called but holding asm code Signed-off-by: Namhyung Kim Acked-by: Masami Hiramatsu LKML-Reference: <1284512670-2369-4-git-send-email-namhyung@gmail.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/kprobes.c | 2 +- kernel/kprobes.c | 5 +++-- 2 files changed, 4 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c index 770ebfb349e..05c20a48841 100644 --- a/arch/x86/kernel/kprobes.c +++ b/arch/x86/kernel/kprobes.c @@ -1129,7 +1129,7 @@ static void __kprobes synthesize_set_arg1(kprobe_opcode_t *addr, *(unsigned long *)addr = val; } -void __kprobes kprobes_optinsn_template_holder(void) +static void __used __kprobes kprobes_optinsn_template_holder(void) { asm volatile ( ".global optprobe_template_entry\n" diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 1b0dbe06707..c53aad5d7e5 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -399,7 +399,7 @@ static inline int kprobe_optready(struct kprobe *p) * Return an optimized kprobe whose optimizing code replaces * instructions including addr (exclude breakpoint). */ -struct kprobe *__kprobes get_optimized_kprobe(unsigned long addr) +static struct kprobe *__kprobes get_optimized_kprobe(unsigned long addr) { int i; struct kprobe *p = NULL; @@ -857,7 +857,8 @@ void __kprobes kretprobe_hash_unlock(struct task_struct *tsk, spin_unlock_irqrestore(hlist_lock, *flags); } -void __kprobes kretprobe_table_unlock(unsigned long hash, unsigned long *flags) +static void __kprobes kretprobe_table_unlock(unsigned long hash, + unsigned long *flags) { spinlock_t *hlist_lock = kretprobe_table_lock_ptr(hash); spin_unlock_irqrestore(hlist_lock, *flags); -- cgit v1.2.3-70-g09d2 From 6abded71d730322df96c5b7f4ab952ffd5a0080d Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Wed, 15 Sep 2010 10:04:29 +0900 Subject: kprobes: Remove __dummy_buf Remove __dummy_buf which is needed for kallsyms_lookup only. use kallsysm_lookup_size_offset instead. Signed-off-by: Namhyung Kim Acked-by: Masami Hiramatsu LKML-Reference: <1284512670-2369-5-git-send-email-namhyung@gmail.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/kprobes.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c index 05c20a48841..e05952af5d2 100644 --- a/arch/x86/kernel/kprobes.c +++ b/arch/x86/kernel/kprobes.c @@ -230,9 +230,6 @@ static int recover_probed_instruction(kprobe_opcode_t *buf, unsigned long addr) return 0; } -/* Dummy buffers for kallsyms_lookup */ -static char __dummy_buf[KSYM_NAME_LEN]; - /* Check if paddr is at an instruction boundary */ static int __kprobes can_probe(unsigned long paddr) { @@ -241,7 +238,7 @@ static int __kprobes can_probe(unsigned long paddr) struct insn insn; kprobe_opcode_t buf[MAX_INSN_SIZE]; - if (!kallsyms_lookup(paddr, NULL, &offset, NULL, __dummy_buf)) + if (!kallsyms_lookup_size_offset(paddr, NULL, &offset)) return 0; /* Decode instructions */ @@ -1269,11 +1266,9 @@ static int __kprobes can_optimize(unsigned long paddr) unsigned long addr, size = 0, offset = 0; struct insn insn; kprobe_opcode_t buf[MAX_INSN_SIZE]; - /* Dummy buffers for lookup_symbol_attrs */ - static char __dummy_buf[KSYM_NAME_LEN]; /* Lookup symbol including addr */ - if (!kallsyms_lookup(paddr, &size, &offset, NULL, __dummy_buf)) + if (!kallsyms_lookup_size_offset(paddr, &size, &offset)) return 0; /* Check there is enough space for a relative jump. */ -- cgit v1.2.3-70-g09d2 From 892df7f81c31ce7f85778aa78094e8d1f19b8413 Mon Sep 17 00:00:00 2001 From: Udo van den Heuvel Date: Tue, 14 Sep 2010 07:15:08 +0200 Subject: x86: HPET force enable for CX700 / VIA Epia LT Allow using HPET with the hpet=force command line option on VIA EPIA CX700 systems. Signed-off-by: Udo van den Heuvel Cc: Robert Hancock LKML-Reference: <4C8F04DC.5060303@xs4all.nl> Signed-off-by: Thomas Gleixner --- arch/x86/kernel/quirks.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/quirks.c b/arch/x86/kernel/quirks.c index 939b9e98245..8bbe8c56916 100644 --- a/arch/x86/kernel/quirks.c +++ b/arch/x86/kernel/quirks.c @@ -344,6 +344,8 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8235, vt8237_force_enable_hpet); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8237, vt8237_force_enable_hpet); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_CX700, + vt8237_force_enable_hpet); static void ati_force_hpet_resume(void) { -- cgit v1.2.3-70-g09d2 From 75e3cfbed6f71a8f151dc6e413b6ce3c390030cb Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Fri, 27 Aug 2010 11:09:48 -0700 Subject: x86, intr-remap: Set redirection hint in the IRTE Currently the redirection hint in the interrupt-remapping table entry is set to 0, which means the remapped interrupt is directed to the processors listed in the destination. So in logical flat mode in the presence of intr-remapping, this results in a single interrupt multi-casted to multiple cpu's as specified by the destination bit mask. But what we really want is to send that interrupt to one of the cpus based on the lowest priority delivery mode. Set the redirection hint in the IRTE to '1' to indicate that we want the remapped interrupt to be directed to only one of the processors listed in the destination. This fixes the issue of same interrupt getting delivered to multiple cpu's in the logical flat mode in the presence of interrupt-remapping. While there is no functional issue observed with this behavior, this will impact performance of such configurations (<=8 cpu's using logical flat mode in the presence of interrupt-remapping) Signed-off-by: Suresh Siddha LKML-Reference: <20100827181049.013051492@sbsiddha-MOBL3.sc.intel.com> Cc: Weidong Han Cc: # [v2.6.32+] Signed-off-by: H. Peter Anvin --- arch/x86/kernel/apic/io_apic.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index f1efebaf551..90f8a75f548 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -1392,6 +1392,7 @@ int setup_ioapic_entry(int apic_id, int irq, irte.dlvry_mode = apic->irq_delivery_mode; irte.vector = vector; irte.dest_id = IRTE_DEST(destination); + irte.redir_hint = 1; /* Set source-id of interrupt request */ set_ioapic_sid(&irte, apic_id); @@ -3343,6 +3344,7 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, irte.dlvry_mode = apic->irq_delivery_mode; irte.vector = cfg->vector; irte.dest_id = IRTE_DEST(dest); + irte.redir_hint = 1; /* Set source-id of interrupt request */ if (pdev) -- cgit v1.2.3-70-g09d2 From 62a92f4c69cd1d9361ad8c16be1dd16e6821bc15 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Fri, 27 Aug 2010 11:09:49 -0700 Subject: x86, intr-remap: Remove IRTE setup duplicate code Remove IRTE setup duplicate code with prepare_irte(). Signed-off-by: Suresh Siddha LKML-Reference: <20100827181049.095067319@sbsiddha-MOBL3.sc.intel.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/irq_remapping.h | 27 +++++++++++++++++++++++++++ arch/x86/kernel/apic/io_apic.c | 27 ++------------------------- 2 files changed, 29 insertions(+), 25 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/irq_remapping.h b/arch/x86/include/asm/irq_remapping.h index f275e224450..8d841505344 100644 --- a/arch/x86/include/asm/irq_remapping.h +++ b/arch/x86/include/asm/irq_remapping.h @@ -3,4 +3,31 @@ #define IRTE_DEST(dest) ((x2apic_mode) ? dest : dest << 8) +#ifdef CONFIG_INTR_REMAP +static inline void prepare_irte(struct irte *irte, int vector, + unsigned int dest) +{ + memset(irte, 0, sizeof(*irte)); + + irte->present = 1; + irte->dst_mode = apic->irq_dest_mode; + /* + * Trigger mode in the IRTE will always be edge, and for IO-APIC, the + * actual level or edge trigger will be setup in the IO-APIC + * RTE. This will help simplify level triggered irq migration. + * For more details, see the comments (in io_apic.c) explainig IO-APIC + * irq migration in the presence of interrupt-remapping. + */ + irte->trigger_mode = 0; + irte->dlvry_mode = apic->irq_delivery_mode; + irte->vector = vector; + irte->dest_id = IRTE_DEST(dest); + irte->redir_hint = 1; +} +#else +static void prepare_irte(struct irte *irte, int vector, unsigned int dest) +{ +} +#endif + #endif /* _ASM_X86_IRQ_REMAPPING_H */ diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 90f8a75f548..e8c95a22614 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -1377,22 +1377,7 @@ int setup_ioapic_entry(int apic_id, int irq, if (index < 0) panic("Failed to allocate IRTE for ioapic %d\n", apic_id); - memset(&irte, 0, sizeof(irte)); - - irte.present = 1; - irte.dst_mode = apic->irq_dest_mode; - /* - * Trigger mode in the IRTE will always be edge, and the - * actual level or edge trigger will be setup in the IO-APIC - * RTE. This will help simplify level triggered irq migration. - * For more details, see the comments above explainig IO-APIC - * irq migration in the presence of interrupt-remapping. - */ - irte.trigger_mode = 0; - irte.dlvry_mode = apic->irq_delivery_mode; - irte.vector = vector; - irte.dest_id = IRTE_DEST(destination); - irte.redir_hint = 1; + prepare_irte(&irte, vector, destination); /* Set source-id of interrupt request */ set_ioapic_sid(&irte, apic_id); @@ -3336,15 +3321,7 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, ir_index = map_irq_to_irte_handle(irq, &sub_handle); BUG_ON(ir_index == -1); - memset (&irte, 0, sizeof(irte)); - - irte.present = 1; - irte.dst_mode = apic->irq_dest_mode; - irte.trigger_mode = 0; /* edge */ - irte.dlvry_mode = apic->irq_delivery_mode; - irte.vector = cfg->vector; - irte.dest_id = IRTE_DEST(dest); - irte.redir_hint = 1; + prepare_irte(&irte, cfg->vector, dest); /* Set source-id of interrupt request */ if (pdev) -- cgit v1.2.3-70-g09d2 From fa47f7e52874683a9659df2f1f143105f676dc0f Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Fri, 27 Aug 2010 11:09:50 -0700 Subject: x86, x2apic: Simplify apic init in SMP and UP builds Move enable_IR_x2apic() inside the default_setup_apic_routing(), and for SMP platforms, move the default_setup_apic_routing() after smp_sanity_check(). This cleans up the code that tries to avoid multiple calls to default_setup_apic_routing() when smp_sanity_check() fails (which goes through the APIC_init_uniprocessor() path). Signed-off-by: Suresh Siddha LKML-Reference: <20100827181049.173087246@sbsiddha-MOBL3.sc.intel.com> Signed-off-by: H. Peter Anvin --- arch/x86/kernel/apic/apic.c | 3 --- arch/x86/kernel/apic/probe_64.c | 3 +++ arch/x86/kernel/smpboot.c | 4 ++-- 3 files changed, 5 insertions(+), 5 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index e3b534cda49..8cf86fb3b4e 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -1665,10 +1665,7 @@ int __init APIC_init_uniprocessor(void) } #endif -#ifndef CONFIG_SMP - enable_IR_x2apic(); default_setup_apic_routing(); -#endif verify_local_APIC(); connect_bsp_APIC(); diff --git a/arch/x86/kernel/apic/probe_64.c b/arch/x86/kernel/apic/probe_64.c index 83e9be4778e..f9e4e6a5407 100644 --- a/arch/x86/kernel/apic/probe_64.c +++ b/arch/x86/kernel/apic/probe_64.c @@ -54,6 +54,9 @@ static int apicid_phys_pkg_id(int initial_apic_id, int index_msb) */ void __init default_setup_apic_routing(void) { + + enable_IR_x2apic(); + #ifdef CONFIG_X86_X2APIC if (x2apic_mode #ifdef CONFIG_X86_UV diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 8b3bfc4dd70..87a8c6b00f8 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -1109,8 +1109,6 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus) } set_cpu_sibling_map(0); - enable_IR_x2apic(); - default_setup_apic_routing(); if (smp_sanity_check(max_cpus) < 0) { printk(KERN_INFO "SMP disabled\n"); @@ -1118,6 +1116,8 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus) goto out; } + default_setup_apic_routing(); + preempt_disable(); if (read_apic_id() != boot_cpu_physical_apicid) { panic("Boot APIC ID in local APIC unexpected (%d vs %d)", -- cgit v1.2.3-70-g09d2 From 3ee48b6af49cf534ca2f481ecc484b156a41451d Mon Sep 17 00:00:00 2001 From: Cliff Wickman Date: Thu, 16 Sep 2010 11:44:02 -0500 Subject: mm, x86: Saving vmcore with non-lazy freeing of vmas During the reading of /proc/vmcore the kernel is doing ioremap()/iounmap() repeatedly. And the buildup of un-flushed vm_area_struct's is causing a great deal of overhead. (rb_next() is chewing up most of that time). This solution is to provide function set_iounmap_nonlazy(). It causes a subsequent call to iounmap() to immediately purge the vma area (with try_purge_vmap_area_lazy()). With this patch we have seen the time for writing a 250MB compressed dump drop from 71 seconds to 44 seconds. Signed-off-by: Cliff Wickman Cc: Andrew Morton Cc: kexec@lists.infradead.org Cc: LKML-Reference: Signed-off-by: Ingo Molnar --- arch/x86/include/asm/io.h | 1 + arch/x86/kernel/crash_dump_64.c | 1 + mm/vmalloc.c | 9 +++++++++ 3 files changed, 11 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/io.h b/arch/x86/include/asm/io.h index 30a3e977612..6a45ec41ec2 100644 --- a/arch/x86/include/asm/io.h +++ b/arch/x86/include/asm/io.h @@ -206,6 +206,7 @@ static inline void __iomem *ioremap(resource_size_t offset, unsigned long size) extern void iounmap(volatile void __iomem *addr); +extern void set_iounmap_nonlazy(void); #ifdef __KERNEL__ diff --git a/arch/x86/kernel/crash_dump_64.c b/arch/x86/kernel/crash_dump_64.c index bf43188ca65..994828899e0 100644 --- a/arch/x86/kernel/crash_dump_64.c +++ b/arch/x86/kernel/crash_dump_64.c @@ -46,6 +46,7 @@ ssize_t copy_oldmem_page(unsigned long pfn, char *buf, } else memcpy(buf, vaddr + offset, csize); + set_iounmap_nonlazy(); iounmap(vaddr); return csize; } diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 6b8889da69a..d8087f0db50 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -516,6 +516,15 @@ static atomic_t vmap_lazy_nr = ATOMIC_INIT(0); /* for per-CPU blocks */ static void purge_fragmented_blocks_allcpus(void); +/* + * called before a call to iounmap() if the caller wants vm_area_struct's + * immediately freed. + */ +void set_iounmap_nonlazy(void) +{ + atomic_set(&vmap_lazy_nr, lazy_max_pages()+1); +} + /* * Purges all lazily-freed vmap areas. * -- cgit v1.2.3-70-g09d2 From 3518dd14ca888085797ca8d3a9e11c8ef9e7ae68 Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Fri, 17 Sep 2010 18:07:45 +0200 Subject: x86, cacheinfo: Fix dependency of AMD L3 CID L3 cache index disable code uses PCI accesses to AMD northbridge functions. Currently the code is #ifdef CONFIG_CPU_SUP_AMD. But it should be #if (defined(CONFIG_CPU_SUP_AMD) && defined(CONFIG_PCI)) which in the end is a dependency to K8_NB. Signed-off-by: Andreas Herrmann LKML-Reference: <20100917160744.GF4958@loge.amd.com> Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/intel_cacheinfo.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c index 898c2f4eab8..2521cdcb877 100644 --- a/arch/x86/kernel/cpu/intel_cacheinfo.c +++ b/arch/x86/kernel/cpu/intel_cacheinfo.c @@ -306,7 +306,7 @@ struct _cache_attr { ssize_t (*store)(struct _cpuid4_info *, const char *, size_t count); }; -#ifdef CONFIG_CPU_SUP_AMD +#ifdef CONFIG_K8_NB /* * L3 cache descriptors @@ -556,12 +556,12 @@ static struct _cache_attr cache_disable_0 = __ATTR(cache_disable_0, 0644, static struct _cache_attr cache_disable_1 = __ATTR(cache_disable_1, 0644, show_cache_disable_1, store_cache_disable_1); -#else /* CONFIG_CPU_SUP_AMD */ +#else /* CONFIG_K8_NB */ static void __cpuinit amd_check_l3_disable(struct _cpuid4_info_regs *this_leaf, int index) { }; -#endif /* CONFIG_CPU_SUP_AMD */ +#endif /* CONFIG_K8_NB */ static int __cpuinit cpuid4_cache_lookup_regs(int index, @@ -1000,7 +1000,7 @@ static struct attribute *default_attrs[] = { static struct attribute *default_l3_attrs[] = { DEFAULT_SYSFS_CACHE_ATTRS, -#ifdef CONFIG_CPU_SUP_AMD +#ifdef CONFIG_K8_NB &cache_disable_0.attr, &cache_disable_1.attr, #endif -- cgit v1.2.3-70-g09d2 From 900f9ac9f12dc3dd6fc8e33e16df172eafcaead6 Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Fri, 17 Sep 2010 18:02:54 +0200 Subject: x86, k8-gart: Decouple handling of garts and northbridges So far we only provide num_k8_northbridges. This is required in different areas (e.g. L3 cache index disable, GART). But not all AMD CPUs provide a GART. Thus it is useful to split off the GART handling from the generic caching of AMD northbridge misc devices. Signed-off-by: Andreas Herrmann LKML-Reference: <20100917160254.GC4958@loge.amd.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/k8.h | 13 +++++---- arch/x86/kernel/cpu/intel_cacheinfo.c | 4 +-- arch/x86/kernel/k8.c | 52 ++++++++++++++++++++--------------- arch/x86/kernel/pci-gart_64.c | 27 ++++++++++++------ drivers/char/agp/amd64-agp.c | 33 +++++++++++++++------- drivers/edac/amd64_edac.c | 2 +- 6 files changed, 82 insertions(+), 49 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/k8.h b/arch/x86/include/asm/k8.h index af00bd1d208..9cee145dcac 100644 --- a/arch/x86/include/asm/k8.h +++ b/arch/x86/include/asm/k8.h @@ -7,24 +7,27 @@ extern struct pci_device_id k8_nb_ids[]; struct bootnode; extern int early_is_k8_nb(u32 value); -extern struct pci_dev **k8_northbridges; -extern int num_k8_northbridges; extern int cache_k8_northbridges(void); extern void k8_flush_garts(void); extern int k8_get_nodes(struct bootnode *nodes); extern int k8_numa_init(unsigned long start_pfn, unsigned long end_pfn); extern int k8_scan_nodes(void); +struct k8_northbridge_info { + u16 num; + u8 gart_supported; + struct pci_dev **nb_misc; +}; +extern struct k8_northbridge_info k8_northbridges; + #ifdef CONFIG_K8_NB -extern int num_k8_northbridges; static inline struct pci_dev *node_to_k8_nb_misc(int node) { - return (node < num_k8_northbridges) ? k8_northbridges[node] : NULL; + return (node < k8_northbridges.num) ? k8_northbridges.nb_misc[node] : NULL; } #else -#define num_k8_northbridges 0 static inline struct pci_dev *node_to_k8_nb_misc(int node) { diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c index 2521cdcb877..6fdfb0b20f8 100644 --- a/arch/x86/kernel/cpu/intel_cacheinfo.c +++ b/arch/x86/kernel/cpu/intel_cacheinfo.c @@ -369,7 +369,7 @@ static void __cpuinit amd_check_l3_disable(struct _cpuid4_info_regs *this_leaf, return; /* not in virtualized environments */ - if (num_k8_northbridges == 0) + if (k8_northbridges.num == 0) return; /* @@ -377,7 +377,7 @@ static void __cpuinit amd_check_l3_disable(struct _cpuid4_info_regs *this_leaf, * never freed but this is done only on shutdown so it doesn't matter. */ if (!l3_caches) { - int size = num_k8_northbridges * sizeof(struct amd_l3_cache *); + int size = k8_northbridges.num * sizeof(struct amd_l3_cache *); l3_caches = kzalloc(size, GFP_ATOMIC); if (!l3_caches) diff --git a/arch/x86/kernel/k8.c b/arch/x86/kernel/k8.c index 0f7bc20cfcd..5de1b6b3963 100644 --- a/arch/x86/kernel/k8.c +++ b/arch/x86/kernel/k8.c @@ -10,9 +10,6 @@ #include #include -int num_k8_northbridges; -EXPORT_SYMBOL(num_k8_northbridges); - static u32 *flush_words; struct pci_device_id k8_nb_ids[] = { @@ -22,7 +19,7 @@ struct pci_device_id k8_nb_ids[] = { }; EXPORT_SYMBOL(k8_nb_ids); -struct pci_dev **k8_northbridges; +struct k8_northbridge_info k8_northbridges; EXPORT_SYMBOL(k8_northbridges); static struct pci_dev *next_k8_northbridge(struct pci_dev *dev) @@ -40,36 +37,44 @@ int cache_k8_northbridges(void) int i; struct pci_dev *dev; - if (num_k8_northbridges) + if (k8_northbridges.num) return 0; dev = NULL; while ((dev = next_k8_northbridge(dev)) != NULL) - num_k8_northbridges++; + k8_northbridges.num++; + + /* some CPU families (e.g. family 0x11) do not support GART */ + if (boot_cpu_data.x86 == 0xf || boot_cpu_data.x86 == 0x10) + k8_northbridges.gart_supported = 1; - k8_northbridges = kmalloc((num_k8_northbridges + 1) * sizeof(void *), - GFP_KERNEL); - if (!k8_northbridges) + k8_northbridges.nb_misc = kmalloc((k8_northbridges.num + 1) * + sizeof(void *), GFP_KERNEL); + if (!k8_northbridges.nb_misc) return -ENOMEM; - if (!num_k8_northbridges) { - k8_northbridges[0] = NULL; + if (!k8_northbridges.num) { + k8_northbridges.nb_misc[0] = NULL; return 0; } - flush_words = kmalloc(num_k8_northbridges * sizeof(u32), GFP_KERNEL); - if (!flush_words) { - kfree(k8_northbridges); - return -ENOMEM; + if (k8_northbridges.gart_supported) { + flush_words = kmalloc(k8_northbridges.num * sizeof(u32), + GFP_KERNEL); + if (!flush_words) { + kfree(k8_northbridges.nb_misc); + return -ENOMEM; + } } dev = NULL; i = 0; while ((dev = next_k8_northbridge(dev)) != NULL) { - k8_northbridges[i] = dev; - pci_read_config_dword(dev, 0x9c, &flush_words[i++]); + k8_northbridges.nb_misc[i] = dev; + if (k8_northbridges.gart_supported) + pci_read_config_dword(dev, 0x9c, &flush_words[i++]); } - k8_northbridges[i] = NULL; + k8_northbridges.nb_misc[i] = NULL; return 0; } EXPORT_SYMBOL_GPL(cache_k8_northbridges); @@ -93,22 +98,25 @@ void k8_flush_garts(void) unsigned long flags; static DEFINE_SPINLOCK(gart_lock); + if (!k8_northbridges.gart_supported) + return; + /* Avoid races between AGP and IOMMU. In theory it's not needed but I'm not sure if the hardware won't lose flush requests when another is pending. This whole thing is so expensive anyways that it doesn't matter to serialize more. -AK */ spin_lock_irqsave(&gart_lock, flags); flushed = 0; - for (i = 0; i < num_k8_northbridges; i++) { - pci_write_config_dword(k8_northbridges[i], 0x9c, + for (i = 0; i < k8_northbridges.num; i++) { + pci_write_config_dword(k8_northbridges.nb_misc[i], 0x9c, flush_words[i]|1); flushed++; } - for (i = 0; i < num_k8_northbridges; i++) { + for (i = 0; i < k8_northbridges.num; i++) { u32 w; /* Make sure the hardware actually executed the flush*/ for (;;) { - pci_read_config_dword(k8_northbridges[i], + pci_read_config_dword(k8_northbridges.nb_misc[i], 0x9c, &w); if (!(w & 1)) break; diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index 0f7f130caa6..8f214a2643f 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -560,8 +560,11 @@ static void enable_gart_translations(void) { int i; - for (i = 0; i < num_k8_northbridges; i++) { - struct pci_dev *dev = k8_northbridges[i]; + if (!k8_northbridges.gart_supported) + return; + + for (i = 0; i < k8_northbridges.num; i++) { + struct pci_dev *dev = k8_northbridges.nb_misc[i]; enable_gart_translation(dev, __pa(agp_gatt_table)); } @@ -592,10 +595,13 @@ static void gart_fixup_northbridges(struct sys_device *dev) if (!fix_up_north_bridges) return; + if (!k8_northbridges.gart_supported) + return; + pr_info("PCI-DMA: Restoring GART aperture settings\n"); - for (i = 0; i < num_k8_northbridges; i++) { - struct pci_dev *dev = k8_northbridges[i]; + for (i = 0; i < k8_northbridges.num; i++) { + struct pci_dev *dev = k8_northbridges.nb_misc[i]; /* * Don't enable translations just yet. That is the next @@ -649,8 +655,8 @@ static __init int init_k8_gatt(struct agp_kern_info *info) aper_size = aper_base = info->aper_size = 0; dev = NULL; - for (i = 0; i < num_k8_northbridges; i++) { - dev = k8_northbridges[i]; + for (i = 0; i < k8_northbridges.num; i++) { + dev = k8_northbridges.nb_misc[i]; new_aper_base = read_aperture(dev, &new_aper_size); if (!new_aper_base) goto nommu; @@ -718,10 +724,13 @@ static void gart_iommu_shutdown(void) if (!no_agp) return; - for (i = 0; i < num_k8_northbridges; i++) { + if (!k8_northbridges.gart_supported) + return; + + for (i = 0; i < k8_northbridges.num; i++) { u32 ctl; - dev = k8_northbridges[i]; + dev = k8_northbridges.nb_misc[i]; pci_read_config_dword(dev, AMD64_GARTAPERTURECTL, &ctl); ctl &= ~GARTEN; @@ -739,7 +748,7 @@ int __init gart_iommu_init(void) unsigned long scratch; long i; - if (num_k8_northbridges == 0) + if (!k8_northbridges.gart_supported) return 0; #ifndef CONFIG_AGP_AMD64 diff --git a/drivers/char/agp/amd64-agp.c b/drivers/char/agp/amd64-agp.c index 70312da4c96..bdf00d583d7 100644 --- a/drivers/char/agp/amd64-agp.c +++ b/drivers/char/agp/amd64-agp.c @@ -124,7 +124,7 @@ static int amd64_fetch_size(void) u32 temp; struct aper_size_info_32 *values; - dev = k8_northbridges[0]; + dev = k8_northbridges.nb_misc[0]; if (dev==NULL) return 0; @@ -181,10 +181,14 @@ static int amd_8151_configure(void) unsigned long gatt_bus = virt_to_phys(agp_bridge->gatt_table_real); int i; + if (!k8_northbridges.gart_supported) + return 0; + /* Configure AGP regs in each x86-64 host bridge. */ - for (i = 0; i < num_k8_northbridges; i++) { + for (i = 0; i < k8_northbridges.num; i++) { agp_bridge->gart_bus_addr = - amd64_configure(k8_northbridges[i], gatt_bus); + amd64_configure(k8_northbridges.nb_misc[i], + gatt_bus); } k8_flush_garts(); return 0; @@ -195,8 +199,12 @@ static void amd64_cleanup(void) { u32 tmp; int i; - for (i = 0; i < num_k8_northbridges; i++) { - struct pci_dev *dev = k8_northbridges[i]; + + if (!k8_northbridges.gart_supported) + return; + + for (i = 0; i < k8_northbridges.num; i++) { + struct pci_dev *dev = k8_northbridges.nb_misc[i]; /* disable gart translation */ pci_read_config_dword(dev, AMD64_GARTAPERTURECTL, &tmp); tmp &= ~AMD64_GARTEN; @@ -319,16 +327,19 @@ static __devinit int fix_northbridge(struct pci_dev *nb, struct pci_dev *agp, return 0; } -static __devinit int cache_nbs (struct pci_dev *pdev, u32 cap_ptr) +static __devinit int cache_nbs(struct pci_dev *pdev, u32 cap_ptr) { int i; if (cache_k8_northbridges() < 0) return -ENODEV; + if (!k8_northbridges.gart_supported) + return -ENODEV; + i = 0; - for (i = 0; i < num_k8_northbridges; i++) { - struct pci_dev *dev = k8_northbridges[i]; + for (i = 0; i < k8_northbridges.num; i++) { + struct pci_dev *dev = k8_northbridges.nb_misc[i]; if (fix_northbridge(dev, pdev, cap_ptr) < 0) { dev_err(&dev->dev, "no usable aperture found\n"); #ifdef __x86_64__ @@ -405,7 +416,8 @@ static int __devinit uli_agp_init(struct pci_dev *pdev) } /* shadow x86-64 registers into ULi registers */ - pci_read_config_dword (k8_northbridges[0], AMD64_GARTAPERTUREBASE, &httfea); + pci_read_config_dword (k8_northbridges.nb_misc[0], AMD64_GARTAPERTUREBASE, + &httfea); /* if x86-64 aperture base is beyond 4G, exit here */ if ((httfea & 0x7fff) >> (32 - 25)) { @@ -472,7 +484,8 @@ static int nforce3_agp_init(struct pci_dev *pdev) pci_write_config_dword(dev1, NVIDIA_X86_64_1_APSIZE, tmp); /* shadow x86-64 registers into NVIDIA registers */ - pci_read_config_dword (k8_northbridges[0], AMD64_GARTAPERTUREBASE, &apbase); + pci_read_config_dword (k8_northbridges.nb_misc[0], AMD64_GARTAPERTUREBASE, + &apbase); /* if x86-64 aperture base is beyond 4G, exit here */ if ( (apbase & 0x7fff) >> (32 - 25) ) { diff --git a/drivers/edac/amd64_edac.c b/drivers/edac/amd64_edac.c index e7d5d6b5dcf..5babf6f4805 100644 --- a/drivers/edac/amd64_edac.c +++ b/drivers/edac/amd64_edac.c @@ -2927,7 +2927,7 @@ static int __init amd64_edac_init(void) * to finish initialization of the MC instances. */ err = -ENODEV; - for (nb = 0; nb < num_k8_northbridges; nb++) { + for (nb = 0; nb < k8_northbridges.num; nb++) { if (!pvt_lookup[nb]) continue; -- cgit v1.2.3-70-g09d2 From bc83cccc761953f878088cdfa682de0970b5561f Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Fri, 17 Sep 2010 15:36:40 -0700 Subject: x86, mwait: Move mwait constants to a common header file We have MWAIT constants spread across three different .c files, for no good reason. Move them all into a common header file. Signed-off-by: H. Peter Anvin Reviewed-by: Arjan van de Ven Cc: Len Brown LKML-Reference: --- arch/x86/include/asm/mwait.h | 15 +++++++++++++++ arch/x86/kernel/acpi/cstate.c | 11 +---------- drivers/acpi/acpi_pad.c | 7 +------ drivers/idle/intel_idle.c | 9 +-------- 4 files changed, 18 insertions(+), 24 deletions(-) create mode 100644 arch/x86/include/asm/mwait.h (limited to 'arch/x86') diff --git a/arch/x86/include/asm/mwait.h b/arch/x86/include/asm/mwait.h new file mode 100644 index 00000000000..bcdff997668 --- /dev/null +++ b/arch/x86/include/asm/mwait.h @@ -0,0 +1,15 @@ +#ifndef _ASM_X86_MWAIT_H +#define _ASM_X86_MWAIT_H + +#define MWAIT_SUBSTATE_MASK 0xf +#define MWAIT_CSTATE_MASK 0xf +#define MWAIT_SUBSTATE_SIZE 4 +#define MWAIT_MAX_NUM_CSTATES 8 + +#define CPUID_MWAIT_LEAF 5 +#define CPUID5_ECX_EXTENSIONS_SUPPORTED 0x1 +#define CPUID5_ECX_INTERRUPT_BREAK 0x2 + +#define MWAIT_ECX_INTERRUPT_BREAK 0x1 + +#endif /* _ASM_X86_MWAIT_H */ diff --git a/arch/x86/kernel/acpi/cstate.c b/arch/x86/kernel/acpi/cstate.c index fb7a5f052e2..bcc4adda7b9 100644 --- a/arch/x86/kernel/acpi/cstate.c +++ b/arch/x86/kernel/acpi/cstate.c @@ -13,6 +13,7 @@ #include #include +#include /* * Initialize bm_flags based on the CPU cache properties @@ -65,16 +66,6 @@ static struct cstate_entry *cpu_cstate_entry; /* per CPU ptr */ static short mwait_supported[ACPI_PROCESSOR_MAX_POWER]; -#define MWAIT_SUBSTATE_MASK (0xf) -#define MWAIT_CSTATE_MASK (0xf) -#define MWAIT_SUBSTATE_SIZE (4) - -#define CPUID_MWAIT_LEAF (5) -#define CPUID5_ECX_EXTENSIONS_SUPPORTED (0x1) -#define CPUID5_ECX_INTERRUPT_BREAK (0x2) - -#define MWAIT_ECX_INTERRUPT_BREAK (0x1) - #define NATIVE_CSTATE_BEYOND_HALT (2) static long acpi_processor_ffh_cstate_probe_cpu(void *_cx) diff --git a/drivers/acpi/acpi_pad.c b/drivers/acpi/acpi_pad.c index b76848c80be..ca2694d365e 100644 --- a/drivers/acpi/acpi_pad.c +++ b/drivers/acpi/acpi_pad.c @@ -30,18 +30,13 @@ #include #include #include +#include #define ACPI_PROCESSOR_AGGREGATOR_CLASS "acpi_pad" #define ACPI_PROCESSOR_AGGREGATOR_DEVICE_NAME "Processor Aggregator" #define ACPI_PROCESSOR_AGGREGATOR_NOTIFY 0x80 static DEFINE_MUTEX(isolated_cpus_lock); -#define MWAIT_SUBSTATE_MASK (0xf) -#define MWAIT_CSTATE_MASK (0xf) -#define MWAIT_SUBSTATE_SIZE (4) -#define CPUID_MWAIT_LEAF (5) -#define CPUID5_ECX_EXTENSIONS_SUPPORTED (0x1) -#define CPUID5_ECX_INTERRUPT_BREAK (0x2) static unsigned long power_saving_mwait_eax; static unsigned char tsc_detected_unstable; diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c index a10152bb142..065c8041c91 100755 --- a/drivers/idle/intel_idle.c +++ b/drivers/idle/intel_idle.c @@ -59,18 +59,11 @@ #include /* ktime_get_real() */ #include #include +#include #define INTEL_IDLE_VERSION "0.4" #define PREFIX "intel_idle: " -#define MWAIT_SUBSTATE_MASK (0xf) -#define MWAIT_CSTATE_MASK (0xf) -#define MWAIT_SUBSTATE_SIZE (4) -#define MWAIT_MAX_NUM_CSTATES 8 -#define CPUID_MWAIT_LEAF (5) -#define CPUID5_ECX_EXTENSIONS_SUPPORTED (0x1) -#define CPUID5_ECX_INTERRUPT_BREAK (0x2) - static struct cpuidle_driver intel_idle_driver = { .name = "intel_idle", .owner = THIS_MODULE, -- cgit v1.2.3-70-g09d2 From ea53069231f9317062910d6e772cca4ce93de8c8 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Fri, 17 Sep 2010 15:39:11 -0700 Subject: x86, hotplug: Use mwait to offline a processor, fix the legacy case The code in native_play_dead() has a number of problems: 1. We should use MWAIT when available, to put ourselves into a deeper sleep state. 2. We use the existence of CLFLUSH to determine if WBINVD is safe, but that is totally bogus -- WBINVD is 486+, whereas CLFLUSH is a much later addition. 3. We should do WBINVD inside the loop, just in case of something like setting an A bit on page tables. Pointed out by Arjan van de Ven. This code is based in part of a previous patch by Venki Pallipadi, but unlike that patch this one keeps all the detection code local instead of pre-caching a bunch of information. We're shutting down the CPU; there is absolutely no hurry. This patch moves all the code to C and deletes the global wbinvd_halt() which is broken anyway. Originally-by: Venkatesh Pallipadi Signed-off-by: H. Peter Anvin Reviewed-by: Arjan van de Ven Cc: Len Brown Cc: Venkatesh Pallipadi Cc: Peter Zijlstra LKML-Reference: <20090522232230.162239000@intel.com> --- arch/x86/include/asm/processor.h | 23 --------------- arch/x86/kernel/smpboot.c | 63 +++++++++++++++++++++++++++++++++++++++- 2 files changed, 62 insertions(+), 24 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index 325b7bdbeba..f358241e412 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -764,29 +764,6 @@ extern unsigned long idle_halt; extern unsigned long idle_nomwait; extern bool c1e_detected; -/* - * on systems with caches, caches must be flashed as the absolute - * last instruction before going into a suspended halt. Otherwise, - * dirty data can linger in the cache and become stale on resume, - * leading to strange errors. - * - * perform a variety of operations to guarantee that the compiler - * will not reorder instructions. wbinvd itself is serializing - * so the processor will not reorder. - * - * Systems without cache can just go into halt. - */ -static inline void wbinvd_halt(void) -{ - mb(); - /* check for clflush to determine if wbinvd is legal */ - if (cpu_has_clflush) - asm volatile("cli; wbinvd; 1: hlt; jmp 1b" : : : "memory"); - else - while (1) - halt(); -} - extern void enable_sep_cpu(void); extern int sysenter_setup(void); diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 8b3bfc4dd70..07bf4233441 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -62,6 +62,7 @@ #include #include #include +#include #include #include #include @@ -1383,11 +1384,71 @@ void play_dead_common(void) local_irq_disable(); } +/* + * We need to flush the caches before going to sleep, lest we have + * dirty data in our caches when we come back up. + */ +static inline void mwait_play_dead(void) +{ + unsigned int eax, ebx, ecx, edx; + unsigned int highest_cstate = 0; + unsigned int highest_subcstate = 0; + int i; + + if (!cpu_has(¤t_cpu_data, X86_FEATURE_MWAIT)) + return; + if (current_cpu_data.cpuid_level < CPUID_MWAIT_LEAF) + return; + + eax = CPUID_MWAIT_LEAF; + ecx = 0; + native_cpuid(&eax, &ebx, &ecx, &edx); + + /* + * eax will be 0 if EDX enumeration is not valid. + * Initialized below to cstate, sub_cstate value when EDX is valid. + */ + if (!(ecx & CPUID5_ECX_EXTENSIONS_SUPPORTED)) { + eax = 0; + } else { + edx >>= MWAIT_SUBSTATE_SIZE; + for (i = 0; i < 7 && edx; i++, edx >>= MWAIT_SUBSTATE_SIZE) { + if (edx & MWAIT_SUBSTATE_MASK) { + highest_cstate = i; + highest_subcstate = edx & MWAIT_SUBSTATE_MASK; + } + } + eax = (highest_cstate << MWAIT_SUBSTATE_SIZE) | + (highest_subcstate - 1); + } + + while (1) { + mb(); + wbinvd(); + __monitor(¤t_thread_info()->flags, 0, 0); + mb(); + __mwait(eax, 0); + } +} + +static inline void hlt_play_dead(void) +{ + while (1) { + mb(); + if (current_cpu_data.x86 >= 4) + wbinvd(); + mb(); + native_halt(); + } +} + void native_play_dead(void) { play_dead_common(); tboot_shutdown(TB_SHUTDOWN_WFS); - wbinvd_halt(); + + mwait_play_dead(); /* Only returns on failure */ + hlt_play_dead(); } #else /* ... !CONFIG_HOTPLUG_CPU */ -- cgit v1.2.3-70-g09d2 From a68e5c94f7d3dd64fef34dd5d97e365cae4bb42a Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Fri, 17 Sep 2010 17:06:46 -0700 Subject: x86, hotplug: Move WBINVD back outside the play_dead loop On processors with hyperthreading, when only one thread is offlined the other thread can cause a spurious wakeup on the idled thread. We do not want to re-WBINVD when that happens. Ideally, we should simply skip WBINVD unless we're the last thread on a particular core to shut down, but there might be similar issues elsewhere in the system. Thus, revert to previous behavior of only WBINVD outside the loop. Partly as a result, remove the mb()'s around it: they are not necessary since wbinvd() is a serializing instruction, but they were intended to make sure the compiler didn't do any funny loop optimizations. Reported-by: Asit Mallick Signed-off-by: H. Peter Anvin Cc: Arjan van de Ven Cc: Len Brown Cc: Venkatesh Pallipadi Cc: Peter Zijlstra LKML-Reference: --- arch/x86/kernel/smpboot.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 07bf4233441..55c80ffb871 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -1422,9 +1422,9 @@ static inline void mwait_play_dead(void) (highest_subcstate - 1); } + wbinvd(); + while (1) { - mb(); - wbinvd(); __monitor(¤t_thread_info()->flags, 0, 0); mb(); __mwait(eax, 0); @@ -1433,11 +1433,10 @@ static inline void mwait_play_dead(void) static inline void hlt_play_dead(void) { + if (current_cpu_data.x86 >= 4) + wbinvd(); + while (1) { - mb(); - if (current_cpu_data.x86 >= 4) - wbinvd(); - mb(); native_halt(); } } -- cgit v1.2.3-70-g09d2 From 995bd3bb5c78f3ff71339803c0b8337ed36d64fb Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 15 Sep 2010 15:11:57 +0200 Subject: x86: Hpet: Avoid the comparator readback penalty Due to the overly intelligent design of HPETs, we need to workaround the problem that the compare value which we write is already behind the actual counter value at the point where the value hits the real compare register. This happens for two reasons: 1) We read out the counter, add the delta and write the result to the compare register. When a NMI or SMI hits between the read out and the write then the counter can be ahead of the event already 2) The write to the compare register is delayed by up to two HPET cycles in certain chipsets. We worked around this by reading back the compare register to make sure that the written value has hit the hardware. For certain ICH9+ chipsets this can require two readouts, as the first one can return the previous compare register value. That's bad performance wise for the normal case where the event is far enough in the future. As we already know that the write can be delayed by up to two cycles we can avoid the read back of the compare register completely if we make the decision whether the delta has elapsed already or not based on the following calculation: cmp = event - actual_count; If cmp is less than 8 HPET clock cycles, then we decide that the event has happened already and return -ETIME. That covers the above #1 and #2 problems which would cause a wait for HPET wraparound (~306 seconds). Signed-off-by: Thomas Gleixner Tested-by: Nix Tested-by: Artur Skawina Cc: Damien Wyart Tested-by: John Drescher Cc: Venkatesh Pallipadi Cc: Arjan van de Ven Cc: Andreas Herrmann Tested-by: Borislav Petkov Cc: Suresh Siddha LKML-Reference: --- arch/x86/kernel/hpet.c | 51 +++++++++++++++++++++----------------------------- 1 file changed, 21 insertions(+), 30 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index 410fdb3f193..0b568b30a4d 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -380,44 +380,35 @@ static int hpet_next_event(unsigned long delta, struct clock_event_device *evt, int timer) { u32 cnt; + s32 res; cnt = hpet_readl(HPET_COUNTER); cnt += (u32) delta; hpet_writel(cnt, HPET_Tn_CMP(timer)); /* - * We need to read back the CMP register on certain HPET - * implementations (ATI chipsets) which seem to delay the - * transfer of the compare register into the internal compare - * logic. With small deltas this might actually be too late as - * the counter could already be higher than the compare value - * at that point and we would wait for the next hpet interrupt - * forever. We found out that reading the CMP register back - * forces the transfer so we can rely on the comparison with - * the counter register below. If the read back from the - * compare register does not match the value we programmed - * then we might have a real hardware problem. We can not do - * much about it here, but at least alert the user/admin with - * a prominent warning. - * - * An erratum on some chipsets (ICH9,..), results in - * comparator read immediately following a write returning old - * value. Workaround for this is to read this value second - * time, when first read returns old value. - * - * In fact the write to the comparator register is delayed up - * to two HPET cycles so the workaround we tried to restrict - * the readback to those known to be borked ATI chipsets - * failed miserably. So we give up on optimizations forever - * and penalize all HPET incarnations unconditionally. + * HPETs are a complete disaster. The compare register is + * based on a equal comparison and neither provides a less + * than or equal functionality (which would require to take + * the wraparound into account) nor a simple count down event + * mode. Further the write to the comparator register is + * delayed internally up to two HPET clock cycles in certain + * chipsets (ATI, ICH9,10). We worked around that by reading + * back the compare register, but that required another + * workaround for ICH9,10 chips where the first readout after + * write can return the old stale value. We already have a + * minimum delta of 5us enforced, but a NMI or SMI hitting + * between the counter readout and the comparator write can + * move us behind that point easily. Now instead of reading + * the compare register back several times, we make the ETIME + * decision based on the following: Return ETIME if the + * counter value after the write is less than 8 HPET cycles + * away from the event or if the counter is already ahead of + * the event. */ - if (unlikely((u32)hpet_readl(HPET_Tn_CMP(timer)) != cnt)) { - if (hpet_readl(HPET_Tn_CMP(timer)) != cnt) - printk_once(KERN_WARNING - "hpet: compare register read back failed.\n"); - } + res = (s32)(cnt - hpet_readl(HPET_COUNTER)); - return (s32)(hpet_readl(HPET_COUNTER) - cnt) >= 0 ? -ETIME : 0; + return res < 8 ? -ETIME : 0; } static void hpet_legacy_set_mode(enum clock_event_mode mode, -- cgit v1.2.3-70-g09d2 From 23ac4ae827e6264e21b898f2cd3f601450aa02a6 Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Fri, 17 Sep 2010 18:03:43 +0200 Subject: x86, k8: Rename k8.[ch] to amd_nb.[ch] and CONFIG_K8_NB to CONFIG_AMD_NB The file names are somehow misleading as the code is not specific to AMD K8 CPUs anymore. The files accomodate code for other AMD CPU northbridges as well. Same is true for the config option which is valid for AMD CPU northbridges in general and not specific to K8. Signed-off-by: Andreas Herrmann LKML-Reference: <20100917160343.GD4958@loge.amd.com> Signed-off-by: H. Peter Anvin --- arch/x86/Kconfig | 4 +- arch/x86/include/asm/amd_nb.h | 39 +++++++++ arch/x86/include/asm/k8.h | 39 --------- arch/x86/kernel/Makefile | 2 +- arch/x86/kernel/amd_nb.c | 145 ++++++++++++++++++++++++++++++++++ arch/x86/kernel/aperture_64.c | 2 +- arch/x86/kernel/cpu/intel_cacheinfo.c | 10 +-- arch/x86/kernel/k8.c | 145 ---------------------------------- arch/x86/kernel/pci-gart_64.c | 2 +- arch/x86/kernel/setup.c | 2 +- arch/x86/mm/k8topology_64.c | 2 +- arch/x86/mm/numa_64.c | 2 +- drivers/char/agp/Kconfig | 2 +- drivers/char/agp/amd64-agp.c | 2 +- drivers/edac/Kconfig | 2 +- drivers/edac/amd64_edac.c | 2 +- 16 files changed, 201 insertions(+), 201 deletions(-) create mode 100644 arch/x86/include/asm/amd_nb.h delete mode 100644 arch/x86/include/asm/k8.h create mode 100644 arch/x86/kernel/amd_nb.c delete mode 100644 arch/x86/kernel/k8.c (limited to 'arch/x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index cea0cd9a316..7fd41f0d754 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -670,7 +670,7 @@ config GART_IOMMU bool "GART IOMMU support" if EMBEDDED default y select SWIOTLB - depends on X86_64 && PCI && K8_NB + depends on X86_64 && PCI && AMD_NB ---help--- Support for full DMA access of devices with 32bit memory access only on systems with more than 3GB. This is usually needed for USB, @@ -2076,7 +2076,7 @@ config OLPC_OPENFIRMWARE endif # X86_32 -config K8_NB +config AMD_NB def_bool y depends on CPU_SUP_AMD && PCI diff --git a/arch/x86/include/asm/amd_nb.h b/arch/x86/include/asm/amd_nb.h new file mode 100644 index 00000000000..c8517f81b21 --- /dev/null +++ b/arch/x86/include/asm/amd_nb.h @@ -0,0 +1,39 @@ +#ifndef _ASM_X86_AMD_NB_H +#define _ASM_X86_AMD_NB_H + +#include + +extern struct pci_device_id k8_nb_ids[]; +struct bootnode; + +extern int early_is_k8_nb(u32 value); +extern int cache_k8_northbridges(void); +extern void k8_flush_garts(void); +extern int k8_get_nodes(struct bootnode *nodes); +extern int k8_numa_init(unsigned long start_pfn, unsigned long end_pfn); +extern int k8_scan_nodes(void); + +struct k8_northbridge_info { + u16 num; + u8 gart_supported; + struct pci_dev **nb_misc; +}; +extern struct k8_northbridge_info k8_northbridges; + +#ifdef CONFIG_AMD_NB + +static inline struct pci_dev *node_to_k8_nb_misc(int node) +{ + return (node < k8_northbridges.num) ? k8_northbridges.nb_misc[node] : NULL; +} + +#else + +static inline struct pci_dev *node_to_k8_nb_misc(int node) +{ + return NULL; +} +#endif + + +#endif /* _ASM_X86_AMD_NB_H */ diff --git a/arch/x86/include/asm/k8.h b/arch/x86/include/asm/k8.h deleted file mode 100644 index 9cee145dcac..00000000000 --- a/arch/x86/include/asm/k8.h +++ /dev/null @@ -1,39 +0,0 @@ -#ifndef _ASM_X86_K8_H -#define _ASM_X86_K8_H - -#include - -extern struct pci_device_id k8_nb_ids[]; -struct bootnode; - -extern int early_is_k8_nb(u32 value); -extern int cache_k8_northbridges(void); -extern void k8_flush_garts(void); -extern int k8_get_nodes(struct bootnode *nodes); -extern int k8_numa_init(unsigned long start_pfn, unsigned long end_pfn); -extern int k8_scan_nodes(void); - -struct k8_northbridge_info { - u16 num; - u8 gart_supported; - struct pci_dev **nb_misc; -}; -extern struct k8_northbridge_info k8_northbridges; - -#ifdef CONFIG_K8_NB - -static inline struct pci_dev *node_to_k8_nb_misc(int node) -{ - return (node < k8_northbridges.num) ? k8_northbridges.nb_misc[node] : NULL; -} - -#else - -static inline struct pci_dev *node_to_k8_nb_misc(int node) -{ - return NULL; -} -#endif - - -#endif /* _ASM_X86_K8_H */ diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 0925676266b..25dc82d3e76 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -87,7 +87,7 @@ obj-$(CONFIG_EARLY_PRINTK) += early_printk.o obj-$(CONFIG_HPET_TIMER) += hpet.o obj-$(CONFIG_APB_TIMER) += apb_timer.o -obj-$(CONFIG_K8_NB) += k8.o +obj-$(CONFIG_AMD_NB) += amd_nb.o obj-$(CONFIG_DEBUG_RODATA_TEST) += test_rodata.o obj-$(CONFIG_DEBUG_NX_TEST) += test_nx.o diff --git a/arch/x86/kernel/amd_nb.c b/arch/x86/kernel/amd_nb.c new file mode 100644 index 00000000000..4ffc38df7ac --- /dev/null +++ b/arch/x86/kernel/amd_nb.c @@ -0,0 +1,145 @@ +/* + * Shared support code for AMD K8 northbridges and derivates. + * Copyright 2006 Andi Kleen, SUSE Labs. Subject to GPLv2. + */ +#include +#include +#include +#include +#include +#include +#include + +static u32 *flush_words; + +struct pci_device_id k8_nb_ids[] = { + { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_K8_NB_MISC) }, + { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_10H_NB_MISC) }, + {} +}; +EXPORT_SYMBOL(k8_nb_ids); + +struct k8_northbridge_info k8_northbridges; +EXPORT_SYMBOL(k8_northbridges); + +static struct pci_dev *next_k8_northbridge(struct pci_dev *dev) +{ + do { + dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev); + if (!dev) + break; + } while (!pci_match_id(&k8_nb_ids[0], dev)); + return dev; +} + +int cache_k8_northbridges(void) +{ + int i; + struct pci_dev *dev; + + if (k8_northbridges.num) + return 0; + + dev = NULL; + while ((dev = next_k8_northbridge(dev)) != NULL) + k8_northbridges.num++; + + /* some CPU families (e.g. family 0x11) do not support GART */ + if (boot_cpu_data.x86 == 0xf || boot_cpu_data.x86 == 0x10) + k8_northbridges.gart_supported = 1; + + k8_northbridges.nb_misc = kmalloc((k8_northbridges.num + 1) * + sizeof(void *), GFP_KERNEL); + if (!k8_northbridges.nb_misc) + return -ENOMEM; + + if (!k8_northbridges.num) { + k8_northbridges.nb_misc[0] = NULL; + return 0; + } + + if (k8_northbridges.gart_supported) { + flush_words = kmalloc(k8_northbridges.num * sizeof(u32), + GFP_KERNEL); + if (!flush_words) { + kfree(k8_northbridges.nb_misc); + return -ENOMEM; + } + } + + dev = NULL; + i = 0; + while ((dev = next_k8_northbridge(dev)) != NULL) { + k8_northbridges.nb_misc[i] = dev; + if (k8_northbridges.gart_supported) + pci_read_config_dword(dev, 0x9c, &flush_words[i++]); + } + k8_northbridges.nb_misc[i] = NULL; + return 0; +} +EXPORT_SYMBOL_GPL(cache_k8_northbridges); + +/* Ignores subdevice/subvendor but as far as I can figure out + they're useless anyways */ +int __init early_is_k8_nb(u32 device) +{ + struct pci_device_id *id; + u32 vendor = device & 0xffff; + device >>= 16; + for (id = k8_nb_ids; id->vendor; id++) + if (vendor == id->vendor && device == id->device) + return 1; + return 0; +} + +void k8_flush_garts(void) +{ + int flushed, i; + unsigned long flags; + static DEFINE_SPINLOCK(gart_lock); + + if (!k8_northbridges.gart_supported) + return; + + /* Avoid races between AGP and IOMMU. In theory it's not needed + but I'm not sure if the hardware won't lose flush requests + when another is pending. This whole thing is so expensive anyways + that it doesn't matter to serialize more. -AK */ + spin_lock_irqsave(&gart_lock, flags); + flushed = 0; + for (i = 0; i < k8_northbridges.num; i++) { + pci_write_config_dword(k8_northbridges.nb_misc[i], 0x9c, + flush_words[i]|1); + flushed++; + } + for (i = 0; i < k8_northbridges.num; i++) { + u32 w; + /* Make sure the hardware actually executed the flush*/ + for (;;) { + pci_read_config_dword(k8_northbridges.nb_misc[i], + 0x9c, &w); + if (!(w & 1)) + break; + cpu_relax(); + } + } + spin_unlock_irqrestore(&gart_lock, flags); + if (!flushed) + printk("nothing to flush?\n"); +} +EXPORT_SYMBOL_GPL(k8_flush_garts); + +static __init int init_k8_nbs(void) +{ + int err = 0; + + err = cache_k8_northbridges(); + + if (err < 0) + printk(KERN_NOTICE "K8 NB: Cannot enumerate AMD northbridges.\n"); + + return err; +} + +/* This has to go after the PCI subsystem */ +fs_initcall(init_k8_nbs); diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c index a2e0caf26e1..e91e042a5c8 100644 --- a/arch/x86/kernel/aperture_64.c +++ b/arch/x86/kernel/aperture_64.c @@ -27,7 +27,7 @@ #include #include #include -#include +#include #include int gart_iommu_aperture; diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c index 6fdfb0b20f8..12cd823c8d0 100644 --- a/arch/x86/kernel/cpu/intel_cacheinfo.c +++ b/arch/x86/kernel/cpu/intel_cacheinfo.c @@ -17,7 +17,7 @@ #include #include -#include +#include #include #define LVL_1_INST 1 @@ -306,7 +306,7 @@ struct _cache_attr { ssize_t (*store)(struct _cpuid4_info *, const char *, size_t count); }; -#ifdef CONFIG_K8_NB +#ifdef CONFIG_AMD_NB /* * L3 cache descriptors @@ -556,12 +556,12 @@ static struct _cache_attr cache_disable_0 = __ATTR(cache_disable_0, 0644, static struct _cache_attr cache_disable_1 = __ATTR(cache_disable_1, 0644, show_cache_disable_1, store_cache_disable_1); -#else /* CONFIG_K8_NB */ +#else /* CONFIG_AMD_NB */ static void __cpuinit amd_check_l3_disable(struct _cpuid4_info_regs *this_leaf, int index) { }; -#endif /* CONFIG_K8_NB */ +#endif /* CONFIG_AMD_NB */ static int __cpuinit cpuid4_cache_lookup_regs(int index, @@ -1000,7 +1000,7 @@ static struct attribute *default_attrs[] = { static struct attribute *default_l3_attrs[] = { DEFAULT_SYSFS_CACHE_ATTRS, -#ifdef CONFIG_K8_NB +#ifdef CONFIG_AMD_NB &cache_disable_0.attr, &cache_disable_1.attr, #endif diff --git a/arch/x86/kernel/k8.c b/arch/x86/kernel/k8.c deleted file mode 100644 index 5de1b6b3963..00000000000 --- a/arch/x86/kernel/k8.c +++ /dev/null @@ -1,145 +0,0 @@ -/* - * Shared support code for AMD K8 northbridges and derivates. - * Copyright 2006 Andi Kleen, SUSE Labs. Subject to GPLv2. - */ -#include -#include -#include -#include -#include -#include -#include - -static u32 *flush_words; - -struct pci_device_id k8_nb_ids[] = { - { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_K8_NB_MISC) }, - { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_10H_NB_MISC) }, - {} -}; -EXPORT_SYMBOL(k8_nb_ids); - -struct k8_northbridge_info k8_northbridges; -EXPORT_SYMBOL(k8_northbridges); - -static struct pci_dev *next_k8_northbridge(struct pci_dev *dev) -{ - do { - dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev); - if (!dev) - break; - } while (!pci_match_id(&k8_nb_ids[0], dev)); - return dev; -} - -int cache_k8_northbridges(void) -{ - int i; - struct pci_dev *dev; - - if (k8_northbridges.num) - return 0; - - dev = NULL; - while ((dev = next_k8_northbridge(dev)) != NULL) - k8_northbridges.num++; - - /* some CPU families (e.g. family 0x11) do not support GART */ - if (boot_cpu_data.x86 == 0xf || boot_cpu_data.x86 == 0x10) - k8_northbridges.gart_supported = 1; - - k8_northbridges.nb_misc = kmalloc((k8_northbridges.num + 1) * - sizeof(void *), GFP_KERNEL); - if (!k8_northbridges.nb_misc) - return -ENOMEM; - - if (!k8_northbridges.num) { - k8_northbridges.nb_misc[0] = NULL; - return 0; - } - - if (k8_northbridges.gart_supported) { - flush_words = kmalloc(k8_northbridges.num * sizeof(u32), - GFP_KERNEL); - if (!flush_words) { - kfree(k8_northbridges.nb_misc); - return -ENOMEM; - } - } - - dev = NULL; - i = 0; - while ((dev = next_k8_northbridge(dev)) != NULL) { - k8_northbridges.nb_misc[i] = dev; - if (k8_northbridges.gart_supported) - pci_read_config_dword(dev, 0x9c, &flush_words[i++]); - } - k8_northbridges.nb_misc[i] = NULL; - return 0; -} -EXPORT_SYMBOL_GPL(cache_k8_northbridges); - -/* Ignores subdevice/subvendor but as far as I can figure out - they're useless anyways */ -int __init early_is_k8_nb(u32 device) -{ - struct pci_device_id *id; - u32 vendor = device & 0xffff; - device >>= 16; - for (id = k8_nb_ids; id->vendor; id++) - if (vendor == id->vendor && device == id->device) - return 1; - return 0; -} - -void k8_flush_garts(void) -{ - int flushed, i; - unsigned long flags; - static DEFINE_SPINLOCK(gart_lock); - - if (!k8_northbridges.gart_supported) - return; - - /* Avoid races between AGP and IOMMU. In theory it's not needed - but I'm not sure if the hardware won't lose flush requests - when another is pending. This whole thing is so expensive anyways - that it doesn't matter to serialize more. -AK */ - spin_lock_irqsave(&gart_lock, flags); - flushed = 0; - for (i = 0; i < k8_northbridges.num; i++) { - pci_write_config_dword(k8_northbridges.nb_misc[i], 0x9c, - flush_words[i]|1); - flushed++; - } - for (i = 0; i < k8_northbridges.num; i++) { - u32 w; - /* Make sure the hardware actually executed the flush*/ - for (;;) { - pci_read_config_dword(k8_northbridges.nb_misc[i], - 0x9c, &w); - if (!(w & 1)) - break; - cpu_relax(); - } - } - spin_unlock_irqrestore(&gart_lock, flags); - if (!flushed) - printk("nothing to flush?\n"); -} -EXPORT_SYMBOL_GPL(k8_flush_garts); - -static __init int init_k8_nbs(void) -{ - int err = 0; - - err = cache_k8_northbridges(); - - if (err < 0) - printk(KERN_NOTICE "K8 NB: Cannot enumerate AMD northbridges.\n"); - - return err; -} - -/* This has to go after the PCI subsystem */ -fs_initcall(init_k8_nbs); diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index 8f214a2643f..67e5665ce42 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -39,7 +39,7 @@ #include #include #include -#include +#include #include static unsigned long iommu_bus_base; /* GART remapping area (physical) */ diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index c3a4fbb2b99..77eea0362a4 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -107,7 +107,7 @@ #include #include #include -#include +#include #ifdef CONFIG_X86_64 #include #endif diff --git a/arch/x86/mm/k8topology_64.c b/arch/x86/mm/k8topology_64.c index 970ed579d4e..ab75b181812 100644 --- a/arch/x86/mm/k8topology_64.c +++ b/arch/x86/mm/k8topology_64.c @@ -22,7 +22,7 @@ #include #include #include -#include +#include static struct bootnode __initdata nodes[8]; static nodemask_t __initdata nodes_parsed = NODE_MASK_NONE; diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c index a7bcc23ef96..4962f1aeda6 100644 --- a/arch/x86/mm/numa_64.c +++ b/arch/x86/mm/numa_64.c @@ -18,7 +18,7 @@ #include #include #include -#include +#include struct pglist_data *node_data[MAX_NUMNODES] __read_mostly; EXPORT_SYMBOL(node_data); diff --git a/drivers/char/agp/Kconfig b/drivers/char/agp/Kconfig index 4b66c69eaf5..5ddf67e76f8 100644 --- a/drivers/char/agp/Kconfig +++ b/drivers/char/agp/Kconfig @@ -57,7 +57,7 @@ config AGP_AMD config AGP_AMD64 tristate "AMD Opteron/Athlon64 on-CPU GART support" - depends on AGP && X86 && K8_NB + depends on AGP && X86 && AMD_NB help This option gives you AGP support for the GLX component of X using the on-CPU northbridge of the AMD Athlon64/Opteron CPUs. diff --git a/drivers/char/agp/amd64-agp.c b/drivers/char/agp/amd64-agp.c index bdf00d583d7..4d6087c9ec9 100644 --- a/drivers/char/agp/amd64-agp.c +++ b/drivers/char/agp/amd64-agp.c @@ -15,7 +15,7 @@ #include #include /* PAGE_SIZE */ #include -#include +#include #include #include "agp.h" diff --git a/drivers/edac/Kconfig b/drivers/edac/Kconfig index 70bb350de99..734e2e06237 100644 --- a/drivers/edac/Kconfig +++ b/drivers/edac/Kconfig @@ -66,7 +66,7 @@ config EDAC_MCE config EDAC_AMD64 tristate "AMD64 (Opteron, Athlon64) K8, F10h, F11h" - depends on EDAC_MM_EDAC && K8_NB && X86_64 && PCI && EDAC_DECODE_MCE + depends on EDAC_MM_EDAC && AMD_NB && X86_64 && PCI && EDAC_DECODE_MCE help Support for error detection and correction on the AMD 64 Families of Memory Controllers (K8, F10h and F11h) diff --git a/drivers/edac/amd64_edac.c b/drivers/edac/amd64_edac.c index 5babf6f4805..09fcc528232 100644 --- a/drivers/edac/amd64_edac.c +++ b/drivers/edac/amd64_edac.c @@ -1,5 +1,5 @@ #include "amd64_edac.h" -#include +#include static struct edac_pci_ctl_info *amd64_ctl_pci; -- cgit v1.2.3-70-g09d2 From f49aa448561fe9215f43405cac6f31eb86317792 Mon Sep 17 00:00:00 2001 From: Jason Baron Date: Fri, 17 Sep 2010 11:08:51 -0400 Subject: jump label: Make dynamic no-op selection available outside of ftrace Move Steve's code for finding the best 5-byte no-op from ftrace.c to alternative.c. The idea is that other consumers (in this case jump label) want to make use of that code. Signed-off-by: Jason Baron LKML-Reference: <96259ae74172dcac99c0020c249743c523a92e18.1284733808.git.jbaron@redhat.com> Signed-off-by: Steven Rostedt --- arch/x86/include/asm/alternative.h | 8 +++++ arch/x86/kernel/alternative.c | 64 ++++++++++++++++++++++++++++++++++++++ arch/x86/kernel/ftrace.c | 63 +------------------------------------ arch/x86/kernel/setup.c | 6 ++++ 4 files changed, 79 insertions(+), 62 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/alternative.h b/arch/x86/include/asm/alternative.h index bc6abb7bc7e..27a35b68918 100644 --- a/arch/x86/include/asm/alternative.h +++ b/arch/x86/include/asm/alternative.h @@ -180,4 +180,12 @@ static inline void apply_paravirt(struct paravirt_patch_site *start, extern void *text_poke(void *addr, const void *opcode, size_t len); extern void *text_poke_smp(void *addr, const void *opcode, size_t len); +#if defined(CONFIG_DYNAMIC_FTRACE) +#define IDEAL_NOP_SIZE_5 5 +extern unsigned char ideal_nop5[IDEAL_NOP_SIZE_5]; +extern void arch_init_ideal_nop5(void); +#else +static inline void arch_init_ideal_nop5(void) {} +#endif + #endif /* _ASM_X86_ALTERNATIVE_H */ diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index f65ab8b014c..1849d8036ee 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -641,3 +641,67 @@ void *__kprobes text_poke_smp(void *addr, const void *opcode, size_t len) return addr; } +#if defined(CONFIG_DYNAMIC_FTRACE) + +unsigned char ideal_nop5[IDEAL_NOP_SIZE_5]; + +void __init arch_init_ideal_nop5(void) +{ + extern const unsigned char ftrace_test_p6nop[]; + extern const unsigned char ftrace_test_nop5[]; + extern const unsigned char ftrace_test_jmp[]; + int faulted = 0; + + /* + * There is no good nop for all x86 archs. + * We will default to using the P6_NOP5, but first we + * will test to make sure that the nop will actually + * work on this CPU. If it faults, we will then + * go to a lesser efficient 5 byte nop. If that fails + * we then just use a jmp as our nop. This isn't the most + * efficient nop, but we can not use a multi part nop + * since we would then risk being preempted in the middle + * of that nop, and if we enabled tracing then, it might + * cause a system crash. + * + * TODO: check the cpuid to determine the best nop. + */ + asm volatile ( + "ftrace_test_jmp:" + "jmp ftrace_test_p6nop\n" + "nop\n" + "nop\n" + "nop\n" /* 2 byte jmp + 3 bytes */ + "ftrace_test_p6nop:" + P6_NOP5 + "jmp 1f\n" + "ftrace_test_nop5:" + ".byte 0x66,0x66,0x66,0x66,0x90\n" + "1:" + ".section .fixup, \"ax\"\n" + "2: movl $1, %0\n" + " jmp ftrace_test_nop5\n" + "3: movl $2, %0\n" + " jmp 1b\n" + ".previous\n" + _ASM_EXTABLE(ftrace_test_p6nop, 2b) + _ASM_EXTABLE(ftrace_test_nop5, 3b) + : "=r"(faulted) : "0" (faulted)); + + switch (faulted) { + case 0: + pr_info("converting mcount calls to 0f 1f 44 00 00\n"); + memcpy(ideal_nop5, ftrace_test_p6nop, IDEAL_NOP_SIZE_5); + break; + case 1: + pr_info("converting mcount calls to 66 66 66 66 90\n"); + memcpy(ideal_nop5, ftrace_test_nop5, IDEAL_NOP_SIZE_5); + break; + case 2: + pr_info("converting mcount calls to jmp . + 5\n"); + memcpy(ideal_nop5, ftrace_test_jmp, IDEAL_NOP_SIZE_5); + break; + } + +} +#endif diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c index cd37469b54e..3afb33f14d2 100644 --- a/arch/x86/kernel/ftrace.c +++ b/arch/x86/kernel/ftrace.c @@ -257,14 +257,9 @@ do_ftrace_mod_code(unsigned long ip, void *new_code) return mod_code_status; } - - - -static unsigned char ftrace_nop[MCOUNT_INSN_SIZE]; - static unsigned char *ftrace_nop_replace(void) { - return ftrace_nop; + return ideal_nop5; } static int @@ -338,62 +333,6 @@ int ftrace_update_ftrace_func(ftrace_func_t func) int __init ftrace_dyn_arch_init(void *data) { - extern const unsigned char ftrace_test_p6nop[]; - extern const unsigned char ftrace_test_nop5[]; - extern const unsigned char ftrace_test_jmp[]; - int faulted = 0; - - /* - * There is no good nop for all x86 archs. - * We will default to using the P6_NOP5, but first we - * will test to make sure that the nop will actually - * work on this CPU. If it faults, we will then - * go to a lesser efficient 5 byte nop. If that fails - * we then just use a jmp as our nop. This isn't the most - * efficient nop, but we can not use a multi part nop - * since we would then risk being preempted in the middle - * of that nop, and if we enabled tracing then, it might - * cause a system crash. - * - * TODO: check the cpuid to determine the best nop. - */ - asm volatile ( - "ftrace_test_jmp:" - "jmp ftrace_test_p6nop\n" - "nop\n" - "nop\n" - "nop\n" /* 2 byte jmp + 3 bytes */ - "ftrace_test_p6nop:" - P6_NOP5 - "jmp 1f\n" - "ftrace_test_nop5:" - ".byte 0x66,0x66,0x66,0x66,0x90\n" - "1:" - ".section .fixup, \"ax\"\n" - "2: movl $1, %0\n" - " jmp ftrace_test_nop5\n" - "3: movl $2, %0\n" - " jmp 1b\n" - ".previous\n" - _ASM_EXTABLE(ftrace_test_p6nop, 2b) - _ASM_EXTABLE(ftrace_test_nop5, 3b) - : "=r"(faulted) : "0" (faulted)); - - switch (faulted) { - case 0: - pr_info("converting mcount calls to 0f 1f 44 00 00\n"); - memcpy(ftrace_nop, ftrace_test_p6nop, MCOUNT_INSN_SIZE); - break; - case 1: - pr_info("converting mcount calls to 66 66 66 66 90\n"); - memcpy(ftrace_nop, ftrace_test_nop5, MCOUNT_INSN_SIZE); - break; - case 2: - pr_info("converting mcount calls to jmp . + 5\n"); - memcpy(ftrace_nop, ftrace_test_jmp, MCOUNT_INSN_SIZE); - break; - } - /* The return code is retured via data */ *(unsigned long *)data = 0; diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index c3a4fbb2b99..00e167870f7 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -112,6 +112,7 @@ #include #endif #include +#include /* * end_pfn only includes RAM, while max_pfn_mapped includes all e820 entries. @@ -726,6 +727,7 @@ void __init setup_arch(char **cmdline_p) { int acpi = 0; int k8 = 0; + unsigned long flags; #ifdef CONFIG_X86_32 memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data)); @@ -1071,6 +1073,10 @@ void __init setup_arch(char **cmdline_p) x86_init.oem.banner(); mcheck_init(); + + local_irq_save(flags); + arch_init_ideal_nop5(); + local_irq_restore(flags); } #ifdef CONFIG_X86_32 -- cgit v1.2.3-70-g09d2 From fa6f2cc77081792e4edca9168420a3422299ef15 Mon Sep 17 00:00:00 2001 From: Jason Baron Date: Fri, 17 Sep 2010 11:08:56 -0400 Subject: jump label: Make text_poke_early() globally visible Make text_poke_early available outside of alternative.c. The jump label patchset wants to make use of it in order to set up the optimal no-op sequences at run-time. Signed-off-by: Jason Baron LKML-Reference: <04cfddf2ba77bcabfc3e524f1849d871d6a1cf9d.1284733808.git.jbaron@redhat.com> Signed-off-by: Steven Rostedt --- arch/x86/include/asm/alternative.h | 2 ++ arch/x86/kernel/alternative.c | 4 ++-- 2 files changed, 4 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/alternative.h b/arch/x86/include/asm/alternative.h index 27a35b68918..634bf782dca 100644 --- a/arch/x86/include/asm/alternative.h +++ b/arch/x86/include/asm/alternative.h @@ -160,6 +160,8 @@ static inline void apply_paravirt(struct paravirt_patch_site *start, #define __parainstructions_end NULL #endif +extern void *text_poke_early(void *addr, const void *opcode, size_t len); + /* * Clear and restore the kernel write-protection flag on the local CPU. * Allows the kernel to edit read-only pages. diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index 1849d8036ee..083bd010d92 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -195,7 +195,7 @@ static void __init_or_module add_nops(void *insns, unsigned int len) extern struct alt_instr __alt_instructions[], __alt_instructions_end[]; extern s32 __smp_locks[], __smp_locks_end[]; -static void *text_poke_early(void *addr, const void *opcode, size_t len); +void *text_poke_early(void *addr, const void *opcode, size_t len); /* Replace instructions with better alternatives for this CPU type. This runs before SMP is initialized to avoid SMP problems with @@ -522,7 +522,7 @@ void __init alternative_instructions(void) * instructions. And on the local CPU you need to be protected again NMI or MCE * handlers seeing an inconsistent instruction while you patch. */ -static void *__init_or_module text_poke_early(void *addr, const void *opcode, +void *__init_or_module text_poke_early(void *addr, const void *opcode, size_t len) { unsigned long flags; -- cgit v1.2.3-70-g09d2 From ce5f68246bf2385d6174856708d0b746dc378f20 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Mon, 20 Sep 2010 13:04:45 -0700 Subject: x86, hotplug: In the MWAIT case of play_dead, CLFLUSH the cache line When we're using MWAIT for play_dead, explicitly CLFLUSH the cache line before executing MONITOR. This is a potential workaround for the Xeon 7400 erratum AAI65 after having a spurious wakeup and returning around the loop. "Potential" here because it is not certain that that erratum could actually trigger; however, the CLFLUSH should be harmless. Signed-off-by: H. Peter Anvin Acked-by: Venkatesh Pallipadi Cc: Asit Mallick Cc: Arjan van de Ven Cc: Len Brown --- arch/x86/kernel/smpboot.c | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 55c80ffb871..fdccfe9dc63 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -1394,9 +1394,12 @@ static inline void mwait_play_dead(void) unsigned int highest_cstate = 0; unsigned int highest_subcstate = 0; int i; + void *mwait_ptr; if (!cpu_has(¤t_cpu_data, X86_FEATURE_MWAIT)) return; + if (!cpu_has(¤t_cpu_data, X86_FEATURE_CLFLSH)) + return; if (current_cpu_data.cpuid_level < CPUID_MWAIT_LEAF) return; @@ -1422,10 +1425,25 @@ static inline void mwait_play_dead(void) (highest_subcstate - 1); } + /* + * This should be a memory location in a cache line which is + * unlikely to be touched by other processors. The actual + * content is immaterial as it is not actually modified in any way. + */ + mwait_ptr = ¤t_thread_info()->flags; + wbinvd(); while (1) { - __monitor(¤t_thread_info()->flags, 0, 0); + /* + * The CLFLUSH is a workaround for erratum AAI65 for + * the Xeon 7400 series. It's not clear it is actually + * needed, but it should be harmless in either case. + * The WBINVD is insufficient due to the spurious-wakeup + * case where we return around the loop. + */ + clflush(mwait_ptr); + __monitor(mwait_ptr, 0, 0); mb(); __mwait(eax, 0); } -- cgit v1.2.3-70-g09d2 From c2b9ff24a0df649d4d40947878b5b5ac39c7299e Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Mon, 20 Sep 2010 18:01:46 -0700 Subject: x86, cpu: Re-run get_cpu_cap() after adjusting the CPUID level At least on Intel, adjusting the max CPUID level can expose new CPUID features, so we need to re-run get_cpu_cap() after changing the CPUID level. Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/common.c | 2 +- arch/x86/kernel/cpu/cpu.h | 1 + arch/x86/kernel/cpu/intel.c | 1 + 3 files changed, 3 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 490dac63c2d..f2f9ac7da25 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -545,7 +545,7 @@ void __cpuinit cpu_detect(struct cpuinfo_x86 *c) } } -static void __cpuinit get_cpu_cap(struct cpuinfo_x86 *c) +void __cpuinit get_cpu_cap(struct cpuinfo_x86 *c) { u32 tfms, xlvl; u32 ebx; diff --git a/arch/x86/kernel/cpu/cpu.h b/arch/x86/kernel/cpu/cpu.h index 3624e8a0f71..c16456bc11a 100644 --- a/arch/x86/kernel/cpu/cpu.h +++ b/arch/x86/kernel/cpu/cpu.h @@ -32,6 +32,7 @@ struct cpu_dev { extern const struct cpu_dev *const __x86_cpu_dev_start[], *const __x86_cpu_dev_end[]; +extern void get_cpu_cap(struct cpuinfo_x86 *c); extern void cpu_detect_cache_sizes(struct cpuinfo_x86 *c); #endif diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 85f69cdeae1..b4389441efb 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -39,6 +39,7 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) misc_enable &= ~MSR_IA32_MISC_ENABLE_LIMIT_CPUID; wrmsrl(MSR_IA32_MISC_ENABLE, misc_enable); c->cpuid_level = cpuid_eax(0); + get_cpu_cap(c); } } -- cgit v1.2.3-70-g09d2 From bb7ab785ad05a97a2c9ffb3a06547ed39f3133e8 Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Tue, 21 Sep 2010 03:26:35 -0400 Subject: oprofile: Add Support for Intel CPU Family 6 / Model 29 This patch adds CPU type detection for dunnington processor (Family 6 / Model 29) to be identified as core 2 family cpu type (wikipedia source). I tested oprofile on Intel(R) Xeon(R) CPU E7440 reporting itself as model 29, and it runs without an issue. Spec: http://www.intel.com/Assets/en_US/PDF/specupdate/320336.pdf Signed-off-by: Jiri Olsa Acked-by: Andi Kleen Cc: stable@kernel.org Signed-off-by: Robert Richter --- arch/x86/oprofile/nmi_int.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86') diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c index 009b819f48d..f1575c9a257 100644 --- a/arch/x86/oprofile/nmi_int.c +++ b/arch/x86/oprofile/nmi_int.c @@ -674,6 +674,7 @@ static int __init ppro_init(char **cpu_type) case 0x0f: case 0x16: case 0x17: + case 0x1d: *cpu_type = "i386/core_2"; break; case 0x1a: -- cgit v1.2.3-70-g09d2 From bf5438fca2950b03c21ad868090cc1a8fcd49536 Mon Sep 17 00:00:00 2001 From: Jason Baron Date: Fri, 17 Sep 2010 11:09:00 -0400 Subject: jump label: Base patch for jump label base patch to implement 'jump labeling'. Based on a new 'asm goto' inline assembly gcc mechanism, we can now branch to labels from an 'asm goto' statment. This allows us to create a 'no-op' fastpath, which can subsequently be patched with a jump to the slowpath code. This is useful for code which might be rarely used, but which we'd like to be able to call, if needed. Tracepoints are the current usecase that these are being implemented for. Acked-by: David S. Miller Signed-off-by: Jason Baron LKML-Reference: [ cleaned up some formating ] Signed-off-by: Steven Rostedt --- Makefile | 5 + arch/Kconfig | 3 + arch/x86/include/asm/alternative.h | 3 +- arch/x86/kernel/alternative.c | 2 +- include/asm-generic/vmlinux.lds.h | 10 ++ include/linux/jump_label.h | 58 +++++++ include/linux/module.h | 5 +- kernel/Makefile | 2 +- kernel/jump_label.c | 346 +++++++++++++++++++++++++++++++++++++ kernel/kprobes.c | 1 + kernel/module.c | 6 + scripts/gcc-goto.sh | 5 + 12 files changed, 442 insertions(+), 4 deletions(-) create mode 100644 include/linux/jump_label.h create mode 100644 kernel/jump_label.c create mode 100644 scripts/gcc-goto.sh (limited to 'arch/x86') diff --git a/Makefile b/Makefile index 92ab33f16cf..a906378d505 100644 --- a/Makefile +++ b/Makefile @@ -591,6 +591,11 @@ KBUILD_CFLAGS += $(call cc-option,-fno-strict-overflow) # conserve stack if available KBUILD_CFLAGS += $(call cc-option,-fconserve-stack) +# check for 'asm goto' +ifeq ($(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-goto.sh $(CC)), y) + KBUILD_CFLAGS += -DCC_HAVE_ASM_GOTO +endif + # Add user supplied CPPFLAGS, AFLAGS and CFLAGS as the last assignments # But warn user when we do so warn-assign = \ diff --git a/arch/Kconfig b/arch/Kconfig index 4877a8c8ee1..1462d8492d8 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -158,4 +158,7 @@ config HAVE_PERF_EVENTS_NMI subsystem. Also has support for calculating CPU cycle events to determine how many clock cycles in a given period. +config HAVE_ARCH_JUMP_LABEL + bool + source "kernel/gcov/Kconfig" diff --git a/arch/x86/include/asm/alternative.h b/arch/x86/include/asm/alternative.h index 634bf782dca..76561d20ea2 100644 --- a/arch/x86/include/asm/alternative.h +++ b/arch/x86/include/asm/alternative.h @@ -4,6 +4,7 @@ #include #include #include +#include #include /* @@ -182,7 +183,7 @@ extern void *text_poke_early(void *addr, const void *opcode, size_t len); extern void *text_poke(void *addr, const void *opcode, size_t len); extern void *text_poke_smp(void *addr, const void *opcode, size_t len); -#if defined(CONFIG_DYNAMIC_FTRACE) +#if defined(CONFIG_DYNAMIC_FTRACE) || defined(HAVE_JUMP_LABEL) #define IDEAL_NOP_SIZE_5 5 extern unsigned char ideal_nop5[IDEAL_NOP_SIZE_5]; extern void arch_init_ideal_nop5(void); diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index 083bd010d92..cb0e6d385f6 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -641,7 +641,7 @@ void *__kprobes text_poke_smp(void *addr, const void *opcode, size_t len) return addr; } -#if defined(CONFIG_DYNAMIC_FTRACE) +#if defined(CONFIG_DYNAMIC_FTRACE) || defined(HAVE_JUMP_LABEL) unsigned char ideal_nop5[IDEAL_NOP_SIZE_5]; diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index 8a92a170fb7..ef2af9948ea 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -220,6 +220,8 @@ \ BUG_TABLE \ \ + JUMP_TABLE \ + \ /* PCI quirks */ \ .pci_fixup : AT(ADDR(.pci_fixup) - LOAD_OFFSET) { \ VMLINUX_SYMBOL(__start_pci_fixups_early) = .; \ @@ -563,6 +565,14 @@ #define BUG_TABLE #endif +#define JUMP_TABLE \ + . = ALIGN(8); \ + __jump_table : AT(ADDR(__jump_table) - LOAD_OFFSET) { \ + VMLINUX_SYMBOL(__start___jump_table) = .; \ + *(__jump_table) \ + VMLINUX_SYMBOL(__stop___jump_table) = .; \ + } + #ifdef CONFIG_PM_TRACE #define TRACEDATA \ . = ALIGN(4); \ diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h new file mode 100644 index 00000000000..de58656d28e --- /dev/null +++ b/include/linux/jump_label.h @@ -0,0 +1,58 @@ +#ifndef _LINUX_JUMP_LABEL_H +#define _LINUX_JUMP_LABEL_H + +#if defined(CC_HAVE_ASM_GOTO) && defined(CONFIG_HAVE_ARCH_JUMP_LABEL) +# include +# define HAVE_JUMP_LABEL +#endif + +enum jump_label_type { + JUMP_LABEL_ENABLE, + JUMP_LABEL_DISABLE +}; + +struct module; + +#ifdef HAVE_JUMP_LABEL + +extern struct jump_entry __start___jump_table[]; +extern struct jump_entry __stop___jump_table[]; + +extern void arch_jump_label_transform(struct jump_entry *entry, + enum jump_label_type type); +extern void jump_label_update(unsigned long key, enum jump_label_type type); +extern void jump_label_apply_nops(struct module *mod); +extern void arch_jump_label_text_poke_early(jump_label_t addr); + +#define enable_jump_label(key) \ + jump_label_update((unsigned long)key, JUMP_LABEL_ENABLE); + +#define disable_jump_label(key) \ + jump_label_update((unsigned long)key, JUMP_LABEL_DISABLE); + +#else + +#define JUMP_LABEL(key, label) \ +do { \ + if (unlikely(*key)) \ + goto label; \ +} while (0) + +#define enable_jump_label(cond_var) \ +do { \ + *(cond_var) = 1; \ +} while (0) + +#define disable_jump_label(cond_var) \ +do { \ + *(cond_var) = 0; \ +} while (0) + +static inline int jump_label_apply_nops(struct module *mod) +{ + return 0; +} + +#endif + +#endif diff --git a/include/linux/module.h b/include/linux/module.h index 8a6b9fdc7ff..403ac26023c 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -350,7 +350,10 @@ struct module struct tracepoint *tracepoints; unsigned int num_tracepoints; #endif - +#ifdef HAVE_JUMP_LABEL + struct jump_entry *jump_entries; + unsigned int num_jump_entries; +#endif #ifdef CONFIG_TRACING const char **trace_bprintk_fmt_start; unsigned int num_trace_bprintk_fmt; diff --git a/kernel/Makefile b/kernel/Makefile index 0b72d1a74be..d52b473c99a 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -10,7 +10,7 @@ obj-y = sched.o fork.o exec_domain.o panic.o printk.o \ kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o mutex.o \ hrtimer.o rwsem.o nsproxy.o srcu.o semaphore.o \ notifier.o ksysfs.o pm_qos_params.o sched_clock.o cred.o \ - async.o range.o + async.o range.o jump_label.o obj-$(CONFIG_HAVE_EARLY_RES) += early_res.o obj-y += groups.o diff --git a/kernel/jump_label.c b/kernel/jump_label.c new file mode 100644 index 00000000000..460fd40112b --- /dev/null +++ b/kernel/jump_label.c @@ -0,0 +1,346 @@ +/* + * jump label support + * + * Copyright (C) 2009 Jason Baron + * + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#ifdef HAVE_JUMP_LABEL + +#define JUMP_LABEL_HASH_BITS 6 +#define JUMP_LABEL_TABLE_SIZE (1 << JUMP_LABEL_HASH_BITS) +static struct hlist_head jump_label_table[JUMP_LABEL_TABLE_SIZE]; + +/* mutex to protect coming/going of the the jump_label table */ +static DEFINE_MUTEX(jump_label_mutex); + +struct jump_label_entry { + struct hlist_node hlist; + struct jump_entry *table; + int nr_entries; + /* hang modules off here */ + struct hlist_head modules; + unsigned long key; +}; + +struct jump_label_module_entry { + struct hlist_node hlist; + struct jump_entry *table; + int nr_entries; + struct module *mod; +}; + +static int jump_label_cmp(const void *a, const void *b) +{ + const struct jump_entry *jea = a; + const struct jump_entry *jeb = b; + + if (jea->key < jeb->key) + return -1; + + if (jea->key > jeb->key) + return 1; + + return 0; +} + +static void +sort_jump_label_entries(struct jump_entry *start, struct jump_entry *stop) +{ + unsigned long size; + + size = (((unsigned long)stop - (unsigned long)start) + / sizeof(struct jump_entry)); + sort(start, size, sizeof(struct jump_entry), jump_label_cmp, NULL); +} + +static struct jump_label_entry *get_jump_label_entry(jump_label_t key) +{ + struct hlist_head *head; + struct hlist_node *node; + struct jump_label_entry *e; + u32 hash = jhash((void *)&key, sizeof(jump_label_t), 0); + + head = &jump_label_table[hash & (JUMP_LABEL_TABLE_SIZE - 1)]; + hlist_for_each_entry(e, node, head, hlist) { + if (key == e->key) + return e; + } + return NULL; +} + +static struct jump_label_entry * +add_jump_label_entry(jump_label_t key, int nr_entries, struct jump_entry *table) +{ + struct hlist_head *head; + struct jump_label_entry *e; + u32 hash; + + e = get_jump_label_entry(key); + if (e) + return ERR_PTR(-EEXIST); + + e = kmalloc(sizeof(struct jump_label_entry), GFP_KERNEL); + if (!e) + return ERR_PTR(-ENOMEM); + + hash = jhash((void *)&key, sizeof(jump_label_t), 0); + head = &jump_label_table[hash & (JUMP_LABEL_TABLE_SIZE - 1)]; + e->key = key; + e->table = table; + e->nr_entries = nr_entries; + INIT_HLIST_HEAD(&(e->modules)); + hlist_add_head(&e->hlist, head); + return e; +} + +static int +build_jump_label_hashtable(struct jump_entry *start, struct jump_entry *stop) +{ + struct jump_entry *iter, *iter_begin; + struct jump_label_entry *entry; + int count; + + sort_jump_label_entries(start, stop); + iter = start; + while (iter < stop) { + entry = get_jump_label_entry(iter->key); + if (!entry) { + iter_begin = iter; + count = 0; + while ((iter < stop) && + (iter->key == iter_begin->key)) { + iter++; + count++; + } + entry = add_jump_label_entry(iter_begin->key, + count, iter_begin); + if (IS_ERR(entry)) + return PTR_ERR(entry); + } else { + WARN_ONCE(1, KERN_ERR "build_jump_hashtable: unexpected entry!\n"); + return -1; + } + } + return 0; +} + +/*** + * jump_label_update - update jump label text + * @key - key value associated with a a jump label + * @type - enum set to JUMP_LABEL_ENABLE or JUMP_LABEL_DISABLE + * + * Will enable/disable the jump for jump label @key, depending on the + * value of @type. + * + */ + +void jump_label_update(unsigned long key, enum jump_label_type type) +{ + struct jump_entry *iter; + struct jump_label_entry *entry; + struct hlist_node *module_node; + struct jump_label_module_entry *e_module; + int count; + + mutex_lock(&jump_label_mutex); + entry = get_jump_label_entry((jump_label_t)key); + if (entry) { + count = entry->nr_entries; + iter = entry->table; + while (count--) { + if (kernel_text_address(iter->code)) + arch_jump_label_transform(iter, type); + iter++; + } + /* eanble/disable jump labels in modules */ + hlist_for_each_entry(e_module, module_node, &(entry->modules), + hlist) { + count = e_module->nr_entries; + iter = e_module->table; + while (count--) { + if (kernel_text_address(iter->code)) + arch_jump_label_transform(iter, type); + iter++; + } + } + } + mutex_unlock(&jump_label_mutex); +} + +static __init int init_jump_label(void) +{ + int ret; + struct jump_entry *iter_start = __start___jump_table; + struct jump_entry *iter_stop = __stop___jump_table; + struct jump_entry *iter; + + mutex_lock(&jump_label_mutex); + ret = build_jump_label_hashtable(__start___jump_table, + __stop___jump_table); + iter = iter_start; + while (iter < iter_stop) { + arch_jump_label_text_poke_early(iter->code); + iter++; + } + mutex_unlock(&jump_label_mutex); + return ret; +} +early_initcall(init_jump_label); + +#ifdef CONFIG_MODULES + +static struct jump_label_module_entry * +add_jump_label_module_entry(struct jump_label_entry *entry, + struct jump_entry *iter_begin, + int count, struct module *mod) +{ + struct jump_label_module_entry *e; + + e = kmalloc(sizeof(struct jump_label_module_entry), GFP_KERNEL); + if (!e) + return ERR_PTR(-ENOMEM); + e->mod = mod; + e->nr_entries = count; + e->table = iter_begin; + hlist_add_head(&e->hlist, &entry->modules); + return e; +} + +static int add_jump_label_module(struct module *mod) +{ + struct jump_entry *iter, *iter_begin; + struct jump_label_entry *entry; + struct jump_label_module_entry *module_entry; + int count; + + /* if the module doesn't have jump label entries, just return */ + if (!mod->num_jump_entries) + return 0; + + sort_jump_label_entries(mod->jump_entries, + mod->jump_entries + mod->num_jump_entries); + iter = mod->jump_entries; + while (iter < mod->jump_entries + mod->num_jump_entries) { + entry = get_jump_label_entry(iter->key); + iter_begin = iter; + count = 0; + while ((iter < mod->jump_entries + mod->num_jump_entries) && + (iter->key == iter_begin->key)) { + iter++; + count++; + } + if (!entry) { + entry = add_jump_label_entry(iter_begin->key, 0, NULL); + if (IS_ERR(entry)) + return PTR_ERR(entry); + } + module_entry = add_jump_label_module_entry(entry, iter_begin, + count, mod); + if (IS_ERR(module_entry)) + return PTR_ERR(module_entry); + } + return 0; +} + +static void remove_jump_label_module(struct module *mod) +{ + struct hlist_head *head; + struct hlist_node *node, *node_next, *module_node, *module_node_next; + struct jump_label_entry *e; + struct jump_label_module_entry *e_module; + int i; + + /* if the module doesn't have jump label entries, just return */ + if (!mod->num_jump_entries) + return; + + for (i = 0; i < JUMP_LABEL_TABLE_SIZE; i++) { + head = &jump_label_table[i]; + hlist_for_each_entry_safe(e, node, node_next, head, hlist) { + hlist_for_each_entry_safe(e_module, module_node, + module_node_next, + &(e->modules), hlist) { + if (e_module->mod == mod) { + hlist_del(&e_module->hlist); + kfree(e_module); + } + } + if (hlist_empty(&e->modules) && (e->nr_entries == 0)) { + hlist_del(&e->hlist); + kfree(e); + } + } + } +} + +static int +jump_label_module_notify(struct notifier_block *self, unsigned long val, + void *data) +{ + struct module *mod = data; + int ret = 0; + + switch (val) { + case MODULE_STATE_COMING: + mutex_lock(&jump_label_mutex); + ret = add_jump_label_module(mod); + if (ret) + remove_jump_label_module(mod); + mutex_unlock(&jump_label_mutex); + break; + case MODULE_STATE_GOING: + mutex_lock(&jump_label_mutex); + remove_jump_label_module(mod); + mutex_unlock(&jump_label_mutex); + break; + } + return ret; +} + +/*** + * apply_jump_label_nops - patch module jump labels with arch_get_jump_label_nop() + * @mod: module to patch + * + * Allow for run-time selection of the optimal nops. Before the module + * loads patch these with arch_get_jump_label_nop(), which is specified by + * the arch specific jump label code. + */ +void jump_label_apply_nops(struct module *mod) +{ + struct jump_entry *iter; + + /* if the module doesn't have jump label entries, just return */ + if (!mod->num_jump_entries) + return; + + iter = mod->jump_entries; + while (iter < mod->jump_entries + mod->num_jump_entries) { + arch_jump_label_text_poke_early(iter->code); + iter++; + } +} + +struct notifier_block jump_label_module_nb = { + .notifier_call = jump_label_module_notify, + .priority = 0, +}; + +static __init int init_jump_label_module(void) +{ + return register_module_notifier(&jump_label_module_nb); +} +early_initcall(init_jump_label_module); + +#endif /* CONFIG_MODULES */ + +#endif diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 6dd5359e1f0..18904e42a91 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -47,6 +47,7 @@ #include #include #include +#include #include #include diff --git a/kernel/module.c b/kernel/module.c index d0b5f8db11b..eba134157ef 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -55,6 +55,7 @@ #include #include #include +#include #define CREATE_TRACE_POINTS #include @@ -2308,6 +2309,11 @@ static void find_module_sections(struct module *mod, struct load_info *info) sizeof(*mod->tracepoints), &mod->num_tracepoints); #endif +#ifdef HAVE_JUMP_LABEL + mod->jump_entries = section_objs(info, "__jump_table", + sizeof(*mod->jump_entries), + &mod->num_jump_entries); +#endif #ifdef CONFIG_EVENT_TRACING mod->trace_events = section_objs(info, "_ftrace_events", sizeof(*mod->trace_events), diff --git a/scripts/gcc-goto.sh b/scripts/gcc-goto.sh new file mode 100644 index 00000000000..8e82424be7a --- /dev/null +++ b/scripts/gcc-goto.sh @@ -0,0 +1,5 @@ +#!/bin/sh +# Test for gcc 'asm goto' suport +# Copyright (C) 2010, Jason Baron + +echo "int main(void) { entry: asm goto (\"\"::::entry); return 0; }" | $1 -x c - -c -o /dev/null >/dev/null 2>&1 && echo "y" -- cgit v1.2.3-70-g09d2 From 4c3ef6d79328c0e23ade60cbfc8d496123a6855c Mon Sep 17 00:00:00 2001 From: Jason Baron Date: Fri, 17 Sep 2010 11:09:08 -0400 Subject: jump label: Add jump_label_text_reserved() to reserve jump points Add a jump_label_text_reserved(void *start, void *end), so that other pieces of code that want to modify kernel text, can first verify that jump label has not reserved the instruction. Acked-by: Masami Hiramatsu Signed-off-by: Jason Baron LKML-Reference: <06236663a3a7b1c1f13576bb9eccb6d9c17b7bfe.1284733808.git.jbaron@redhat.com> Signed-off-by: Steven Rostedt --- arch/x86/kernel/kprobes.c | 3 +- include/linux/jump_label.h | 8 ++++- kernel/jump_label.c | 83 ++++++++++++++++++++++++++++++++++++++++++++++ kernel/kprobes.c | 3 +- 4 files changed, 94 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c index e05952af5d2..1cbd54c0df9 100644 --- a/arch/x86/kernel/kprobes.c +++ b/arch/x86/kernel/kprobes.c @@ -1218,7 +1218,8 @@ static int __kprobes copy_optimized_instructions(u8 *dest, u8 *src) } /* Check whether the address range is reserved */ if (ftrace_text_reserved(src, src + len - 1) || - alternatives_text_reserved(src, src + len - 1)) + alternatives_text_reserved(src, src + len - 1) || + jump_label_text_reserved(src, src + len - 1)) return -EBUSY; return len; diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h index de58656d28e..b72cd9f92c2 100644 --- a/include/linux/jump_label.h +++ b/include/linux/jump_label.h @@ -20,9 +20,10 @@ extern struct jump_entry __stop___jump_table[]; extern void arch_jump_label_transform(struct jump_entry *entry, enum jump_label_type type); +extern void arch_jump_label_text_poke_early(jump_label_t addr); extern void jump_label_update(unsigned long key, enum jump_label_type type); extern void jump_label_apply_nops(struct module *mod); -extern void arch_jump_label_text_poke_early(jump_label_t addr); +extern int jump_label_text_reserved(void *start, void *end); #define enable_jump_label(key) \ jump_label_update((unsigned long)key, JUMP_LABEL_ENABLE); @@ -53,6 +54,11 @@ static inline int jump_label_apply_nops(struct module *mod) return 0; } +static inline int jump_label_text_reserved(void *start, void *end) +{ + return 0; +} + #endif #endif diff --git a/kernel/jump_label.c b/kernel/jump_label.c index 460fd40112b..7be868bf25c 100644 --- a/kernel/jump_label.c +++ b/kernel/jump_label.c @@ -177,6 +177,89 @@ void jump_label_update(unsigned long key, enum jump_label_type type) mutex_unlock(&jump_label_mutex); } +static int addr_conflict(struct jump_entry *entry, void *start, void *end) +{ + if (entry->code <= (unsigned long)end && + entry->code + JUMP_LABEL_NOP_SIZE > (unsigned long)start) + return 1; + + return 0; +} + +#ifdef CONFIG_MODULES + +static int module_conflict(void *start, void *end) +{ + struct hlist_head *head; + struct hlist_node *node, *node_next, *module_node, *module_node_next; + struct jump_label_entry *e; + struct jump_label_module_entry *e_module; + struct jump_entry *iter; + int i, count; + int conflict = 0; + + for (i = 0; i < JUMP_LABEL_TABLE_SIZE; i++) { + head = &jump_label_table[i]; + hlist_for_each_entry_safe(e, node, node_next, head, hlist) { + hlist_for_each_entry_safe(e_module, module_node, + module_node_next, + &(e->modules), hlist) { + count = e_module->nr_entries; + iter = e_module->table; + while (count--) { + if (addr_conflict(iter, start, end)) { + conflict = 1; + goto out; + } + iter++; + } + } + } + } +out: + return conflict; +} + +#endif + +/*** + * jump_label_text_reserved - check if addr range is reserved + * @start: start text addr + * @end: end text addr + * + * checks if the text addr located between @start and @end + * overlaps with any of the jump label patch addresses. Code + * that wants to modify kernel text should first verify that + * it does not overlap with any of the jump label addresses. + * + * returns 1 if there is an overlap, 0 otherwise + */ +int jump_label_text_reserved(void *start, void *end) +{ + struct jump_entry *iter; + struct jump_entry *iter_start = __start___jump_table; + struct jump_entry *iter_stop = __start___jump_table; + int conflict = 0; + + mutex_lock(&jump_label_mutex); + iter = iter_start; + while (iter < iter_stop) { + if (addr_conflict(iter, start, end)) { + conflict = 1; + goto out; + } + iter++; + } + + /* now check modules */ +#ifdef CONFIG_MODULES + conflict = module_conflict(start, end); +#endif +out: + mutex_unlock(&jump_label_mutex); + return conflict; +} + static __init int init_jump_label(void) { int ret; diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 18904e42a91..ec4210c6501 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -1147,7 +1147,8 @@ int __kprobes register_kprobe(struct kprobe *p) preempt_disable(); if (!kernel_text_address((unsigned long) p->addr) || in_kprobes_functions((unsigned long) p->addr) || - ftrace_text_reserved(p->addr, p->addr)) { + ftrace_text_reserved(p->addr, p->addr) || + jump_label_text_reserved(p->addr, p->addr)) { preempt_enable(); return -EINVAL; } -- cgit v1.2.3-70-g09d2 From d9f5ab7b1c0a520867af389bab5d5fcdbd0e407e Mon Sep 17 00:00:00 2001 From: Jason Baron Date: Fri, 17 Sep 2010 11:09:22 -0400 Subject: jump label: x86 support add x86 support for jump label. I'm keeping this patch separate so its clear to arch maintainers what was required for x86 support this new feature. Hopefully, it wouldn't be too painful for other archs. Signed-off-by: Jason Baron LKML-Reference: [ cleaned up some formatting ] Signed-off-by: Steven Rostedt --- arch/x86/Kconfig | 1 + arch/x86/include/asm/jump_label.h | 47 ++++++++++++++++++++++++++++++++++++ arch/x86/kernel/Makefile | 2 +- arch/x86/kernel/jump_label.c | 50 +++++++++++++++++++++++++++++++++++++++ arch/x86/kernel/module.c | 3 +++ 5 files changed, 102 insertions(+), 1 deletion(-) create mode 100644 arch/x86/include/asm/jump_label.h create mode 100644 arch/x86/kernel/jump_label.c (limited to 'arch/x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index cea0cd9a316..afcd6632c94 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -59,6 +59,7 @@ config X86 select ANON_INODES select HAVE_ARCH_KMEMCHECK select HAVE_USER_RETURN_NOTIFIER + select HAVE_ARCH_JUMP_LABEL if !CC_OPTIMIZE_FOR_SIZE config INSTRUCTION_DECODER def_bool (KPROBES || PERF_EVENTS) diff --git a/arch/x86/include/asm/jump_label.h b/arch/x86/include/asm/jump_label.h new file mode 100644 index 00000000000..b4a2cb40337 --- /dev/null +++ b/arch/x86/include/asm/jump_label.h @@ -0,0 +1,47 @@ +#ifndef _ASM_X86_JUMP_LABEL_H +#define _ASM_X86_JUMP_LABEL_H + +#ifdef __KERNEL__ + +#include +#include + +#define JUMP_LABEL_NOP_SIZE 5 + +# define JUMP_LABEL_INITIAL_NOP ".byte 0xe9 \n\t .long 0\n\t" + +# define JUMP_LABEL(key, label) \ + do { \ + asm goto("1:" \ + JUMP_LABEL_INITIAL_NOP \ + ".pushsection __jump_table, \"a\" \n\t"\ + _ASM_PTR "1b, %l[" #label "], %c0 \n\t" \ + ".popsection \n\t" \ + : : "i" (key) : : label); \ + } while (0) + +#endif /* __KERNEL__ */ + +#ifdef CONFIG_X86_64 + +typedef u64 jump_label_t; + +struct jump_entry { + jump_label_t code; + jump_label_t target; + jump_label_t key; +}; + +#else + +typedef u32 jump_label_t; + +struct jump_entry { + jump_label_t code; + jump_label_t target; + jump_label_t key; +}; + +#endif + +#endif diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 0925676266b..24fa1718ddb 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -32,7 +32,7 @@ GCOV_PROFILE_paravirt.o := n obj-y := process_$(BITS).o signal.o entry_$(BITS).o obj-y += traps.o irq.o irq_$(BITS).o dumpstack_$(BITS).o obj-y += time.o ioport.o ldt.o dumpstack.o -obj-y += setup.o x86_init.o i8259.o irqinit.o +obj-y += setup.o x86_init.o i8259.o irqinit.o jump_label.o obj-$(CONFIG_X86_VISWS) += visws_quirks.o obj-$(CONFIG_X86_32) += probe_roms_32.o obj-$(CONFIG_X86_32) += sys_i386_32.o i386_ksyms_32.o diff --git a/arch/x86/kernel/jump_label.c b/arch/x86/kernel/jump_label.c new file mode 100644 index 00000000000..961b6b30ba9 --- /dev/null +++ b/arch/x86/kernel/jump_label.c @@ -0,0 +1,50 @@ +/* + * jump label x86 support + * + * Copyright (C) 2009 Jason Baron + * + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#ifdef HAVE_JUMP_LABEL + +union jump_code_union { + char code[JUMP_LABEL_NOP_SIZE]; + struct { + char jump; + int offset; + } __attribute__((packed)); +}; + +void arch_jump_label_transform(struct jump_entry *entry, + enum jump_label_type type) +{ + union jump_code_union code; + + if (type == JUMP_LABEL_ENABLE) { + code.jump = 0xe9; + code.offset = entry->target - + (entry->code + JUMP_LABEL_NOP_SIZE); + } else + memcpy(&code, ideal_nop5, JUMP_LABEL_NOP_SIZE); + get_online_cpus(); + mutex_lock(&text_mutex); + text_poke_smp((void *)entry->code, &code, JUMP_LABEL_NOP_SIZE); + mutex_unlock(&text_mutex); + put_online_cpus(); +} + +void arch_jump_label_text_poke_early(jump_label_t addr) +{ + text_poke_early((void *)addr, ideal_nop5, JUMP_LABEL_NOP_SIZE); +} + +#endif diff --git a/arch/x86/kernel/module.c b/arch/x86/kernel/module.c index e0bc186d750..5399f58de7e 100644 --- a/arch/x86/kernel/module.c +++ b/arch/x86/kernel/module.c @@ -239,6 +239,9 @@ int module_finalize(const Elf_Ehdr *hdr, apply_paravirt(pseg, pseg + para->sh_size); } + /* make jump label nops */ + jump_label_apply_nops(me); + return module_bug_finalize(hdr, sechdrs, me); } -- cgit v1.2.3-70-g09d2 From 95fccd465eefb3d6bf80dae0496607b534d38313 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Wed, 22 Sep 2010 17:37:43 -0400 Subject: jump label: Remove duplicate structure for x86 The structure in the x86 jump label code uses the typedef jump_label_t, which is defined by the #ifdef arch type. The structure does not need to be duplicated there. Signed-off-by: Steven Rostedt --- arch/x86/include/asm/jump_label.h | 12 +----------- 1 file changed, 1 insertion(+), 11 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/jump_label.h b/arch/x86/include/asm/jump_label.h index b4a2cb40337..f52d42e8058 100644 --- a/arch/x86/include/asm/jump_label.h +++ b/arch/x86/include/asm/jump_label.h @@ -23,18 +23,10 @@ #endif /* __KERNEL__ */ #ifdef CONFIG_X86_64 - typedef u64 jump_label_t; - -struct jump_entry { - jump_label_t code; - jump_label_t target; - jump_label_t key; -}; - #else - typedef u32 jump_label_t; +#endif struct jump_entry { jump_label_t code; @@ -43,5 +35,3 @@ struct jump_entry { }; #endif - -#endif -- cgit v1.2.3-70-g09d2 From 234bb549eea16ec7d5207ba747fb8dbf489e64c1 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 2 Sep 2010 13:46:34 +0100 Subject: x86, cleanups: Use clear_page/copy_page rather than memset/memcpy When operating on whole pages, use clear_page() and copy_page() in favor of memset() and memcpy(); after all that's what they are intended for. Signed-off-by: Jan Beulich LKML-Reference: <4C7FB8CA0200007800013F51@vpn.id2.novell.com> Cc: Andrew Morton Signed-off-by: H. Peter Anvin --- arch/x86/kernel/machine_kexec_64.c | 4 ++-- arch/x86/kvm/lapic.c | 3 +-- arch/x86/mm/init_32.c | 4 ++-- arch/x86/mm/init_64.c | 2 +- 4 files changed, 6 insertions(+), 7 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/machine_kexec_64.c b/arch/x86/kernel/machine_kexec_64.c index 035c8c52918..b3ea9db39db 100644 --- a/arch/x86/kernel/machine_kexec_64.c +++ b/arch/x86/kernel/machine_kexec_64.c @@ -36,7 +36,7 @@ static int init_one_level2_page(struct kimage *image, pgd_t *pgd, if (!page) goto out; pud = (pud_t *)page_address(page); - memset(pud, 0, PAGE_SIZE); + clear_page(pud); set_pgd(pgd, __pgd(__pa(pud) | _KERNPG_TABLE)); } pud = pud_offset(pgd, addr); @@ -45,7 +45,7 @@ static int init_one_level2_page(struct kimage *image, pgd_t *pgd, if (!page) goto out; pmd = (pmd_t *)page_address(page); - memset(pmd, 0, PAGE_SIZE); + clear_page(pmd); set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE)); } pmd = pmd_offset(pud, addr); diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index 77d8c0f4817..22b06f7660f 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -1056,14 +1056,13 @@ int kvm_create_lapic(struct kvm_vcpu *vcpu) vcpu->arch.apic = apic; - apic->regs_page = alloc_page(GFP_KERNEL); + apic->regs_page = alloc_page(GFP_KERNEL|__GFP_ZERO); if (apic->regs_page == NULL) { printk(KERN_ERR "malloc apic regs error for vcpu %x\n", vcpu->vcpu_id); goto nomem_free_apic; } apic->regs = page_address(apic->regs_page); - memset(apic->regs, 0, PAGE_SIZE); apic->vcpu = vcpu; hrtimer_init(&apic->lapic_timer.timer, CLOCK_MONOTONIC, diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index bca79091b9d..558f2d33207 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -67,7 +67,7 @@ static __init void *alloc_low_page(void) panic("alloc_low_page: ran out of memory"); adr = __va(pfn * PAGE_SIZE); - memset(adr, 0, PAGE_SIZE); + clear_page(adr); return adr; } @@ -558,7 +558,7 @@ char swsusp_pg_dir[PAGE_SIZE] static inline void save_pg_dir(void) { - memcpy(swsusp_pg_dir, swapper_pg_dir, PAGE_SIZE); + copy_page(swsusp_pg_dir, swapper_pg_dir); } #else /* !CONFIG_ACPI_SLEEP */ static inline void save_pg_dir(void) diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 9a6674689a2..7c48ad4faca 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -293,7 +293,7 @@ static __ref void *alloc_low_page(unsigned long *phys) panic("alloc_low_page: ran out of memory"); adr = early_memremap(pfn * PAGE_SIZE, PAGE_SIZE); - memset(adr, 0, PAGE_SIZE); + clear_page(adr); *phys = pfn * PAGE_SIZE; return adr; } -- cgit v1.2.3-70-g09d2 From 46eb3b64dddd20f44e76b08676fa642dd374bf1d Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Wed, 22 Sep 2010 23:10:23 -0400 Subject: jump label/x86/sparc64: Remove !CC_OPTIMIZE_FOR_SIZE config conditions The !CC_OPTIMIZE_FOR_SIZE was added to enable the jump label functionality because Jason noticed that the gcc option would not optimize the labels and may even hurt performance. But this is a gcc problem not a kernel one. Removing this condition should add motivation to the gcc developers to actually fix it. Cc: Jason Baron Acked-by: David S. Miller Signed-off-by: Steven Rostedt --- arch/sparc/Kconfig | 2 +- arch/x86/Kconfig | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig index a81b04e07fe..9212cd42a83 100644 --- a/arch/sparc/Kconfig +++ b/arch/sparc/Kconfig @@ -30,7 +30,7 @@ config SPARC select PERF_USE_VMALLOC select HAVE_DMA_ATTRS select HAVE_DMA_API_DEBUG - select HAVE_ARCH_JUMP_LABEL if !CC_OPTIMIZE_FOR_SIZE + select HAVE_ARCH_JUMP_LABEL config SPARC32 def_bool !64BIT diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index afcd6632c94..b431a0824a9 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -59,7 +59,7 @@ config X86 select ANON_INODES select HAVE_ARCH_KMEMCHECK select HAVE_USER_RETURN_NOTIFIER - select HAVE_ARCH_JUMP_LABEL if !CC_OPTIMIZE_FOR_SIZE + select HAVE_ARCH_JUMP_LABEL config INSTRUCTION_DECODER def_bool (KPROBES || PERF_EVENTS) -- cgit v1.2.3-70-g09d2 From 6554287b1de0448f1e02e200d02b43914e997d15 Mon Sep 17 00:00:00 2001 From: Bart Oldeman Date: Thu, 23 Sep 2010 13:16:58 -0400 Subject: x86, vm86: Fix preemption bug for int1 debug and int3 breakpoint handlers. Impact: fix kernel bug such as: BUG: scheduling while atomic: dosemu.bin/19680/0x00000004 See also Ubuntu bug 455067 at https://bugs.launchpad.net/ubuntu/+source/linux/+bug/455067 Commits 4915a35e35a037254550a2ba9f367a812bc37d40 ("Use preempt_conditional_sti/cli in do_int3, like on x86_64.") and 3d2a71a596bd9c761c8487a2178e95f8a61da083 ("x86, traps: converge do_debug handlers") started disabling preemption in int1 and int3 handlers on i386. The problem with vm86 is that the call to handle_vm86_trap() may jump straight to entry_32.S and never returns so preempt is never enabled again, and there is an imbalance in the preempt count. Commit be716615fe596ee117292dc615e95f707fb67fd1 ("x86, vm86: fix preemption bug"), which was later (accidentally?) reverted by commit 08d68323d1f0c34452e614263b212ca556dae47f ("hw-breakpoints: modifying generic debug exception to use thread-specific debug registers") fixed the problem for debug exceptions but not for breakpoints. There are three solutions to this problem. 1. Reenable preemption before calling handle_vm86_trap(). This was the approach that was later reverted. 2. Do not disable preemption for i386 in breakpoint and debug handlers. This was the situation before October 2008. As far as I understand preemption only needs to be disabled on x86_64 because a seperate stack is used, but it's nice to have things work the same way on i386 and x86_64. 3. Let handle_vm86_trap() return instead of jumping to assembly code. By setting a flag in _TIF_WORK_MASK, either TIF_IRET or TIF_NOTIFY_RESUME, the code in entry_32.S is instructed to return to 32 bit mode from V86 mode. The logic in entry_32.S was already present to handle signals. (I chose TIF_IRET because it's slightly more efficient in do_notify_resume() in signal.c, but in fact TIF_IRET can probably be replaced by TIF_NOTIFY_RESUME everywhere.) I'm submitting approach 3, because I believe it is the most elegant and prevents future confusion. Still, an obvious preempt_conditional_cli(regs); is necessary in traps.c to correct the bug. [ hpa: This is technically a regression, but because: 1. the regression is so old, 2. the patch seems relatively high risk, justifying more testing, and 3. we're late in the 2.6.36-rc cycle, I'm queuing it up for the 2.6.37 merge window. It might, however, justify as a -stable backport at a latter time, hence Cc: stable. ] Signed-off-by: Bart Oldeman LKML-Reference: Cc: Frederic Weisbecker Cc: K.Prasad Cc: Alan Stern Cc: Alexander van Heukelum Cc: Signed-off-by: H. Peter Anvin --- arch/x86/kernel/traps.c | 1 + arch/x86/kernel/vm86_32.c | 10 ++++++++-- 2 files changed, 9 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index 60788dee0f8..9f4edeb2132 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -575,6 +575,7 @@ dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code) if (regs->flags & X86_VM_MASK) { handle_vm86_trap((struct kernel_vm86_regs *) regs, error_code, 1); + preempt_conditional_cli(regs); return; } diff --git a/arch/x86/kernel/vm86_32.c b/arch/x86/kernel/vm86_32.c index 5ffb5622f79..61fb9851962 100644 --- a/arch/x86/kernel/vm86_32.c +++ b/arch/x86/kernel/vm86_32.c @@ -551,8 +551,14 @@ cannot_handle: int handle_vm86_trap(struct kernel_vm86_regs *regs, long error_code, int trapno) { if (VMPI.is_vm86pus) { - if ((trapno == 3) || (trapno == 1)) - return_to_32bit(regs, VM86_TRAP + (trapno << 8)); + if ((trapno == 3) || (trapno == 1)) { + KVM86->regs32->ax = VM86_TRAP + (trapno << 8); + /* setting this flag forces the code in entry_32.S to + call save_v86_state() and change the stack pointer + to KVM86->regs32 */ + set_thread_flag(TIF_IRET); + return 0; + } do_int(regs, trapno, (unsigned char __user *) (regs->pt.ss << 4), SP(regs)); return 0; } -- cgit v1.2.3-70-g09d2 From 76fb657017588a0912f0d1d140cb807446e4ef05 Mon Sep 17 00:00:00 2001 From: Daniel Drake Date: Thu, 23 Sep 2010 17:28:04 +0100 Subject: x86, olpc: Only enable PCI configuration type override on XO-1 This configuration type override is for XO-1 only and must not happen on XO-1.5. Signed-off-by: Daniel Drake LKML-Reference: <20100923162805.0F6549D401B@zog.reactivated.net> Cc: Andres Solomon Cc: Grant Likely Signed-off-by: H. Peter Anvin --- arch/x86/Kconfig | 2 +- arch/x86/kernel/olpc.c | 6 ++++-- arch/x86/pci/olpc.c | 2 +- 3 files changed, 6 insertions(+), 4 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index cea0cd9a316..0ed4c9bfcd1 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -1900,7 +1900,7 @@ config PCI_GODIRECT bool "Direct" config PCI_GOOLPC - bool "OLPC" + bool "OLPC XO-1" depends on OLPC config PCI_GOANY diff --git a/arch/x86/kernel/olpc.c b/arch/x86/kernel/olpc.c index 0e0cdde519b..635888cf050 100644 --- a/arch/x86/kernel/olpc.c +++ b/arch/x86/kernel/olpc.c @@ -242,8 +242,10 @@ static int __init olpc_init(void) (unsigned char *) &olpc_platform_info.ecver, 1); #ifdef CONFIG_PCI_OLPC - /* If the VSA exists let it emulate PCI, if not emulate in kernel */ - if (!cs5535_has_vsa2()) + /* If the VSA exists let it emulate PCI, if not emulate in kernel. + * XO-1 only. */ + if (olpc_platform_info.boardrev < olpc_board_pre(0xd0) && + !cs5535_has_vsa2()) x86_init.pci.arch_init = pci_olpc_init; #endif diff --git a/arch/x86/pci/olpc.c b/arch/x86/pci/olpc.c index b34815408f5..13700ec8e2e 100644 --- a/arch/x86/pci/olpc.c +++ b/arch/x86/pci/olpc.c @@ -304,7 +304,7 @@ static struct pci_raw_ops pci_olpc_conf = { int __init pci_olpc_init(void) { - printk(KERN_INFO "PCI: Using configuration type OLPC\n"); + printk(KERN_INFO "PCI: Using configuration type OLPC XO-1\n"); raw_pci_ops = &pci_olpc_conf; is_lx = is_geode_lx(); return 0; -- cgit v1.2.3-70-g09d2 From 3e3c486012a3dbb113c0ca15ee265d309d77aea9 Mon Sep 17 00:00:00 2001 From: Daniel Drake Date: Thu, 23 Sep 2010 17:28:46 +0100 Subject: x86, olpc: Rework BIOS signature check The XO-1.5 laptop is not currently detected as an OLPC machine because it fails this XO-1-centric check. Now that we have OLPC OFW support in the kernel, a more sensible check is to see if we found OFW during boot and check the architecture property. Also remove a now-meaningless codepath, as we're always going to have OFW support with OLPC. Signed-off-by: Daniel Drake LKML-Reference: <20100923162846.D8D409D401B@zog.reactivated.net> Cc: Andres Salomon Cc: Grant Likely Signed-off-by: H. Peter Anvin --- arch/x86/Kconfig | 3 ++- arch/x86/include/asm/olpc_ofw.h | 4 +++ arch/x86/kernel/olpc.c | 58 ++++++++++++++++++----------------------- arch/x86/kernel/olpc_ofw.c | 6 +++++ 4 files changed, 37 insertions(+), 34 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 0ed4c9bfcd1..c2552557134 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -2061,6 +2061,7 @@ config SCx200HR_TIMER config OLPC bool "One Laptop Per Child support" select GPIOLIB + select OLPC_OPENFIRMWARE ---help--- Add support for detecting the unique features of the OLPC XO hardware. @@ -2068,7 +2069,7 @@ config OLPC config OLPC_OPENFIRMWARE bool "Support for OLPC's Open Firmware" depends on !X86_64 && !X86_PAE - default y if OLPC + default n help This option adds support for the implementation of Open Firmware that is used on the OLPC XO-1 Children's Machine. diff --git a/arch/x86/include/asm/olpc_ofw.h b/arch/x86/include/asm/olpc_ofw.h index 08fde475cb3..2a8478140bb 100644 --- a/arch/x86/include/asm/olpc_ofw.h +++ b/arch/x86/include/asm/olpc_ofw.h @@ -21,10 +21,14 @@ extern void olpc_ofw_detect(void); /* install OFW's pde permanently into the kernel's pgtable */ extern void setup_olpc_ofw_pgd(void); +/* check if OFW was detected during boot */ +extern bool olpc_ofw_present(void); + #else /* !CONFIG_OLPC_OPENFIRMWARE */ static inline void olpc_ofw_detect(void) { } static inline void setup_olpc_ofw_pgd(void) { } +static inline bool olpc_ofw_present(void) { return false; } #endif /* !CONFIG_OLPC_OPENFIRMWARE */ diff --git a/arch/x86/kernel/olpc.c b/arch/x86/kernel/olpc.c index 635888cf050..37c49934c78 100644 --- a/arch/x86/kernel/olpc.c +++ b/arch/x86/kernel/olpc.c @@ -183,8 +183,21 @@ err: } EXPORT_SYMBOL_GPL(olpc_ec_cmd); -#ifdef CONFIG_OLPC_OPENFIRMWARE -static void __init platform_detect(void) +static bool __init check_ofw_architecture(void) +{ + size_t propsize; + char olpc_arch[5]; + const void *args[] = { NULL, "architecture", olpc_arch, (void *)5 }; + void *res[] = { &propsize }; + + if (olpc_ofw("getprop", args, res)) { + printk(KERN_ERR "ofw: getprop call failed!\n"); + return false; + } + return propsize == 5 && strncmp("OLPC", olpc_arch, 5) == 0; +} + +static u32 __init get_board_revision(void) { size_t propsize; __be32 rev; @@ -193,46 +206,27 @@ static void __init platform_detect(void) if (olpc_ofw("getprop", args, res) || propsize != 4) { printk(KERN_ERR "ofw: getprop call failed!\n"); - rev = cpu_to_be32(0); + return cpu_to_be32(0); } - olpc_platform_info.boardrev = be32_to_cpu(rev); + return be32_to_cpu(rev); } -#else -static void __init platform_detect(void) + +static bool __init platform_detect(void) { - /* stopgap until OFW support is added to the kernel */ - olpc_platform_info.boardrev = olpc_board(0xc2); + if (!check_ofw_architecture()) + return false; + olpc_platform_info.flags |= OLPC_F_PRESENT; + olpc_platform_info.boardrev = get_board_revision(); + return true; } -#endif static int __init olpc_init(void) { - unsigned char *romsig; - - /* The ioremap check is dangerous; limit what we run it on */ - if (!is_geode() || cs5535_has_vsa2()) + if (!olpc_ofw_present() || !platform_detect()) return 0; spin_lock_init(&ec_lock); - romsig = ioremap(0xffffffc0, 16); - if (!romsig) - return 0; - - if (strncmp(romsig, "CL1 Q", 7)) - goto unmap; - if (strncmp(romsig+6, romsig+13, 3)) { - printk(KERN_INFO "OLPC BIOS signature looks invalid. " - "Assuming not OLPC\n"); - goto unmap; - } - - printk(KERN_INFO "OLPC board with OpenFirmware %.16s\n", romsig); - olpc_platform_info.flags |= OLPC_F_PRESENT; - - /* get the platform revision */ - platform_detect(); - /* assume B1 and above models always have a DCON */ if (olpc_board_at_least(olpc_board(0xb1))) olpc_platform_info.flags |= OLPC_F_DCON; @@ -254,8 +248,6 @@ static int __init olpc_init(void) olpc_platform_info.boardrev >> 4, olpc_platform_info.ecver); -unmap: - iounmap(romsig); return 0; } diff --git a/arch/x86/kernel/olpc_ofw.c b/arch/x86/kernel/olpc_ofw.c index 3218aa71ab5..78732046437 100644 --- a/arch/x86/kernel/olpc_ofw.c +++ b/arch/x86/kernel/olpc_ofw.c @@ -74,6 +74,12 @@ int __olpc_ofw(const char *name, int nr_args, const void **args, int nr_res, } EXPORT_SYMBOL_GPL(__olpc_ofw); +bool olpc_ofw_present(void) +{ + return olpc_ofw_cif != NULL; +} +EXPORT_SYMBOL_GPL(olpc_ofw_present); + /* OFW cif _should_ be above this address */ #define OFW_MIN 0xff000000 -- cgit v1.2.3-70-g09d2 From 3b4b682becdfa9f42321aa024d5cc84f71f06d8c Mon Sep 17 00:00:00 2001 From: Ma Ling Date: Fri, 17 Sep 2010 03:12:40 +0800 Subject: x86, mem: Optimize memmove for small size and unaligned cases movs instruction will combine data to accelerate moving data, however we need to concern two cases about it. 1. movs instruction need long lantency to startup, so here we use general mov instruction to copy data. 2. movs instruction is not good for unaligned case, even if src offset is 0x10, dest offset is 0x0, we avoid and handle the case by general mov instruction. Signed-off-by: Ma Ling LKML-Reference: <1284664360-6138-1-git-send-email-ling.ma@intel.com> Signed-off-by: H. Peter Anvin --- arch/x86/lib/memcpy_32.c | 213 ++++++++++++++++++++++++++++++++++++------- arch/x86/lib/memmove_64.c | 225 ++++++++++++++++++++++++++++++++++++---------- 2 files changed, 362 insertions(+), 76 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/lib/memcpy_32.c b/arch/x86/lib/memcpy_32.c index 81130d477ee..b908a59eccf 100644 --- a/arch/x86/lib/memcpy_32.c +++ b/arch/x86/lib/memcpy_32.c @@ -22,36 +22,187 @@ EXPORT_SYMBOL(memset); void *memmove(void *dest, const void *src, size_t n) { - int d0, d1, d2; - - if (dest < src) { - if ((dest + n) < src) - return memcpy(dest, src, n); - else - __asm__ __volatile__( - "rep\n\t" - "movsb\n\t" - : "=&c" (d0), "=&S" (d1), "=&D" (d2) - :"0" (n), - "1" (src), - "2" (dest) - :"memory"); - } else { - if((src + n) < dest) - return memcpy(dest, src, n); - else - __asm__ __volatile__( - "std\n\t" - "rep\n\t" - "movsb\n\t" - "cld" - : "=&c" (d0), "=&S" (d1), "=&D" (d2) - :"0" (n), - "1" (n-1+src), - "2" (n-1+dest) - :"memory"); - } - - return dest; + int d0,d1,d2,d3,d4,d5; + char *ret = dest; + + __asm__ __volatile__( + /* Handle more 16bytes in loop */ + "cmp $0x10, %0\n\t" + "jb 1f\n\t" + + /* Decide forward/backward copy mode */ + "cmp %2, %1\n\t" + "jb 2f\n\t" + + /* + * movs instruction have many startup latency + * so we handle small size by general register. + */ + "cmp $680, %0\n\t" + "jb 3f\n\t" + /* + * movs instruction is only good for aligned case. + */ + "mov %1, %3\n\t" + "xor %2, %3\n\t" + "and $0xff, %3\n\t" + "jz 4f\n\t" + "3:\n\t" + "sub $0x10, %0\n\t" + + /* + * We gobble 16byts forward in each loop. + */ + "3:\n\t" + "sub $0x10, %0\n\t" + "mov 0*4(%1), %3\n\t" + "mov 1*4(%1), %4\n\t" + "mov %3, 0*4(%2)\n\t" + "mov %4, 1*4(%2)\n\t" + "mov 2*4(%1), %3\n\t" + "mov 3*4(%1), %4\n\t" + "mov %3, 2*4(%2)\n\t" + "mov %4, 3*4(%2)\n\t" + "lea 0x10(%1), %1\n\t" + "lea 0x10(%2), %2\n\t" + "jae 3b\n\t" + "add $0x10, %0\n\t" + "jmp 1f\n\t" + + /* + * Handle data forward by movs. + */ + ".p2align 4\n\t" + "4:\n\t" + "mov -4(%1, %0), %3\n\t" + "lea -4(%2, %0), %4\n\t" + "shr $2, %0\n\t" + "rep movsl\n\t" + "mov %3, (%4)\n\t" + "jmp 11f\n\t" + /* + * Handle data backward by movs. + */ + ".p2align 4\n\t" + "6:\n\t" + "mov (%1), %3\n\t" + "mov %2, %4\n\t" + "lea -4(%1, %0), %1\n\t" + "lea -4(%2, %0), %2\n\t" + "shr $2, %0\n\t" + "std\n\t" + "rep movsl\n\t" + "mov %3,(%4)\n\t" + "cld\n\t" + "jmp 11f\n\t" + + /* + * Start to prepare for backward copy. + */ + ".p2align 4\n\t" + "2:\n\t" + "cmp $680, %0\n\t" + "jb 5f\n\t" + "mov %1, %3\n\t" + "xor %2, %3\n\t" + "and $0xff, %3\n\t" + "jz 6b\n\t" + + /* + * Calculate copy position to tail. + */ + "5:\n\t" + "add %0, %1\n\t" + "add %0, %2\n\t" + "sub $0x10, %0\n\t" + + /* + * We gobble 16byts backward in each loop. + */ + "7:\n\t" + "sub $0x10, %0\n\t" + + "mov -1*4(%1), %3\n\t" + "mov -2*4(%1), %4\n\t" + "mov %3, -1*4(%2)\n\t" + "mov %4, -2*4(%2)\n\t" + "mov -3*4(%1), %3\n\t" + "mov -4*4(%1), %4\n\t" + "mov %3, -3*4(%2)\n\t" + "mov %4, -4*4(%2)\n\t" + "lea -0x10(%1), %1\n\t" + "lea -0x10(%2), %2\n\t" + "jae 7b\n\t" + /* + * Calculate copy position to head. + */ + "add $0x10, %0\n\t" + "sub %0, %1\n\t" + "sub %0, %2\n\t" + + /* + * Move data from 8 bytes to 15 bytes. + */ + ".p2align 4\n\t" + "1:\n\t" + "cmp $8, %0\n\t" + "jb 8f\n\t" + "mov 0*4(%1), %3\n\t" + "mov 1*4(%1), %4\n\t" + "mov -2*4(%1, %0), %5\n\t" + "mov -1*4(%1, %0), %1\n\t" + + "mov %3, 0*4(%2)\n\t" + "mov %4, 1*4(%2)\n\t" + "mov %5, -2*4(%2, %0)\n\t" + "mov %1, -1*4(%2, %0)\n\t" + "jmp 11f\n\t" + + /* + * Move data from 4 bytes to 7 bytes. + */ + ".p2align 4\n\t" + "8:\n\t" + "cmp $4, %0\n\t" + "jb 9f\n\t" + "mov 0*4(%1), %3\n\t" + "mov -1*4(%1, %0), %4\n\t" + "mov %3, 0*4(%2)\n\t" + "mov %4, -1*4(%2, %0)\n\t" + "jmp 11f\n\t" + + /* + * Move data from 2 bytes to 3 bytes. + */ + ".p2align 4\n\t" + "9:\n\t" + "cmp $2, %0\n\t" + "jb 10f\n\t" + "movw 0*2(%1), %%dx\n\t" + "movw -1*2(%1, %0), %%bx\n\t" + "movw %%dx, 0*2(%2)\n\t" + "movw %%bx, -1*2(%2, %0)\n\t" + "jmp 11f\n\t" + + /* + * Move data for 1 byte. + */ + ".p2align 4\n\t" + "10:\n\t" + "cmp $1, %0\n\t" + "jb 11f\n\t" + "movb (%1), %%cl\n\t" + "movb %%cl, (%2)\n\t" + ".p2align 4\n\t" + "11:" + : "=&c" (d0), "=&S" (d1), "=&D" (d2), + "=r" (d3),"=r" (d4), "=r"(d5) + :"0" (n), + "1" (src), + "2" (dest) + :"memory"); + + return ret; + } EXPORT_SYMBOL(memmove); diff --git a/arch/x86/lib/memmove_64.c b/arch/x86/lib/memmove_64.c index ecacc4b3d9e..6d0f0ec41b3 100644 --- a/arch/x86/lib/memmove_64.c +++ b/arch/x86/lib/memmove_64.c @@ -8,50 +8,185 @@ #undef memmove void *memmove(void *dest, const void *src, size_t count) { - unsigned long d0, d1, d2, d3; - if (dest < src) { - if ((dest + count) < src) - return memcpy(dest, src, count); - else - __asm__ __volatile__( - "movq %0, %3\n\t" - "shr $3, %0\n\t" - "andq $7, %3\n\t" - "rep\n\t" - "movsq\n\t" - "movq %3, %0\n\t" - "rep\n\t" - "movsb" - : "=&c" (d0), "=&S" (d1), "=&D" (d2), "=r" (d3) - :"0" (count), - "1" (src), - "2" (dest) - :"memory"); - } else { - if((src + count) < dest) - return memcpy(dest, src, count); - else - __asm__ __volatile__( - "movq %0, %3\n\t" - "lea -8(%1, %0), %1\n\t" - "lea -8(%2, %0), %2\n\t" - "shr $3, %0\n\t" - "andq $7, %3\n\t" - "std\n\t" - "rep\n\t" - "movsq\n\t" - "lea 7(%1), %1\n\t" - "lea 7(%2), %2\n\t" - "movq %3, %0\n\t" - "rep\n\t" - "movsb\n\t" - "cld" - : "=&c" (d0), "=&S" (d1), "=&D" (d2), "=r" (d3) - :"0" (count), - "1" (src), - "2" (dest) - :"memory"); - } - return dest; + unsigned long d0,d1,d2,d3,d4,d5,d6,d7; + char *ret; + + __asm__ __volatile__( + /* Handle more 32bytes in loop */ + "mov %2, %3\n\t" + "cmp $0x20, %0\n\t" + "jb 1f\n\t" + + /* Decide forward/backward copy mode */ + "cmp %2, %1\n\t" + "jb 2f\n\t" + + /* + * movsq instruction have many startup latency + * so we handle small size by general register. + */ + "cmp $680, %0\n\t" + "jb 3f\n\t" + /* + * movsq instruction is only good for aligned case. + */ + "cmpb %%dil, %%sil\n\t" + "je 4f\n\t" + "3:\n\t" + "sub $0x20, %0\n\t" + /* + * We gobble 32byts forward in each loop. + */ + "5:\n\t" + "sub $0x20, %0\n\t" + "movq 0*8(%1), %4\n\t" + "movq 1*8(%1), %5\n\t" + "movq 2*8(%1), %6\n\t" + "movq 3*8(%1), %7\n\t" + "leaq 4*8(%1), %1\n\t" + + "movq %4, 0*8(%2)\n\t" + "movq %5, 1*8(%2)\n\t" + "movq %6, 2*8(%2)\n\t" + "movq %7, 3*8(%2)\n\t" + "leaq 4*8(%2), %2\n\t" + "jae 5b\n\t" + "addq $0x20, %0\n\t" + "jmp 1f\n\t" + /* + * Handle data forward by movsq. + */ + ".p2align 4\n\t" + "4:\n\t" + "movq %0, %8\n\t" + "movq -8(%1, %0), %4\n\t" + "lea -8(%2, %0), %5\n\t" + "shrq $3, %8\n\t" + "rep movsq\n\t" + "movq %4, (%5)\n\t" + "jmp 13f\n\t" + /* + * Handle data backward by movsq. + */ + ".p2align 4\n\t" + "7:\n\t" + "movq %0, %8\n\t" + "movq (%1), %4\n\t" + "movq %2, %5\n\t" + "leaq -8(%1, %0), %1\n\t" + "leaq -8(%2, %0), %2\n\t" + "shrq $3, %8\n\t" + "std\n\t" + "rep movsq\n\t" + "cld\n\t" + "movq %4, (%5)\n\t" + "jmp 13f\n\t" + + /* + * Start to prepare for backward copy. + */ + ".p2align 4\n\t" + "2:\n\t" + "cmp $680, %0\n\t" + "jb 6f \n\t" + "cmp %%dil, %%sil\n\t" + "je 7b \n\t" + "6:\n\t" + /* + * Calculate copy position to tail. + */ + "addq %0, %1\n\t" + "addq %0, %2\n\t" + "subq $0x20, %0\n\t" + /* + * We gobble 32byts backward in each loop. + */ + "8:\n\t" + "subq $0x20, %0\n\t" + "movq -1*8(%1), %4\n\t" + "movq -2*8(%1), %5\n\t" + "movq -3*8(%1), %6\n\t" + "movq -4*8(%1), %7\n\t" + "leaq -4*8(%1), %1\n\t" + + "movq %4, -1*8(%2)\n\t" + "movq %5, -2*8(%2)\n\t" + "movq %6, -3*8(%2)\n\t" + "movq %7, -4*8(%2)\n\t" + "leaq -4*8(%2), %2\n\t" + "jae 8b\n\t" + /* + * Calculate copy position to head. + */ + "addq $0x20, %0\n\t" + "subq %0, %1\n\t" + "subq %0, %2\n\t" + "1:\n\t" + "cmpq $16, %0\n\t" + "jb 9f\n\t" + /* + * Move data from 16 bytes to 31 bytes. + */ + "movq 0*8(%1), %4\n\t" + "movq 1*8(%1), %5\n\t" + "movq -2*8(%1, %0), %6\n\t" + "movq -1*8(%1, %0), %7\n\t" + "movq %4, 0*8(%2)\n\t" + "movq %5, 1*8(%2)\n\t" + "movq %6, -2*8(%2, %0)\n\t" + "movq %7, -1*8(%2, %0)\n\t" + "jmp 13f\n\t" + ".p2align 4\n\t" + "9:\n\t" + "cmpq $8, %0\n\t" + "jb 10f\n\t" + /* + * Move data from 8 bytes to 15 bytes. + */ + "movq 0*8(%1), %4\n\t" + "movq -1*8(%1, %0), %5\n\t" + "movq %4, 0*8(%2)\n\t" + "movq %5, -1*8(%2, %0)\n\t" + "jmp 13f\n\t" + "10:\n\t" + "cmpq $4, %0\n\t" + "jb 11f\n\t" + /* + * Move data from 4 bytes to 7 bytes. + */ + "movl (%1), %4d\n\t" + "movl -4(%1, %0), %5d\n\t" + "movl %4d, (%2)\n\t" + "movl %5d, -4(%2, %0)\n\t" + "jmp 13f\n\t" + "11:\n\t" + "cmp $2, %0\n\t" + "jb 12f\n\t" + /* + * Move data from 2 bytes to 3 bytes. + */ + "movw (%1), %4w\n\t" + "movw -2(%1, %0), %5w\n\t" + "movw %4w, (%2)\n\t" + "movw %5w, -2(%2, %0)\n\t" + "jmp 13f\n\t" + "12:\n\t" + "cmp $1, %0\n\t" + "jb 13f\n\t" + /* + * Move data for 1 byte. + */ + "movb (%1), %4b\n\t" + "movb %4b, (%2)\n\t" + "13:\n\t" + : "=&d" (d0), "=&S" (d1), "=&D" (d2), "=&a" (ret) , + "=r"(d3), "=r"(d4), "=r"(d5), "=r"(d6), "=&c" (d7) + :"0" (count), + "1" (src), + "2" (dest) + :"memory"); + + return ret; + } EXPORT_SYMBOL(memmove); -- cgit v1.2.3-70-g09d2 From d900329e20f4476db6461752accebcf7935a8055 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Tue, 28 Sep 2010 15:35:01 -0700 Subject: x86, cpu: After uncapping CPUID, re-run CPU feature detection After uncapping the CPUID level, we need to also re-run the CPU feature detection code. This resolves kernel bugzilla 16322. Reported-by: boris64 Cc: v2.6.29..2.6.35 LKML-Reference: Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/common.c | 2 +- arch/x86/kernel/cpu/cpu.h | 1 + arch/x86/kernel/cpu/intel.c | 1 + 3 files changed, 3 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 490dac63c2d..f2f9ac7da25 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -545,7 +545,7 @@ void __cpuinit cpu_detect(struct cpuinfo_x86 *c) } } -static void __cpuinit get_cpu_cap(struct cpuinfo_x86 *c) +void __cpuinit get_cpu_cap(struct cpuinfo_x86 *c) { u32 tfms, xlvl; u32 ebx; diff --git a/arch/x86/kernel/cpu/cpu.h b/arch/x86/kernel/cpu/cpu.h index 3624e8a0f71..f668bb1f7d4 100644 --- a/arch/x86/kernel/cpu/cpu.h +++ b/arch/x86/kernel/cpu/cpu.h @@ -33,5 +33,6 @@ extern const struct cpu_dev *const __x86_cpu_dev_start[], *const __x86_cpu_dev_end[]; extern void cpu_detect_cache_sizes(struct cpuinfo_x86 *c); +extern void get_cpu_cap(struct cpuinfo_x86 *c); #endif diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 85f69cdeae1..b4389441efb 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -39,6 +39,7 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) misc_enable &= ~MSR_IA32_MISC_ENABLE_LIMIT_CPUID; wrmsrl(MSR_IA32_MISC_ENABLE, misc_enable); c->cpuid_level = cpuid_eax(0); + get_cpu_cap(c); } } -- cgit v1.2.3-70-g09d2 From bd126b23a2f30c3c7d268db2b96866923eb732a5 Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Sun, 8 Aug 2010 02:17:29 +0900 Subject: ACPI: add missing __percpu markup in arch/x86/kernel/acpi/cstate.c cpu_cstate_entry is a percpu pointer but was missing __percpu markup. Signed-off-by: Namhyung Kim Acked-by: Tejun Heo Signed-off-by: Len Brown --- arch/x86/kernel/acpi/cstate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/acpi/cstate.c b/arch/x86/kernel/acpi/cstate.c index fb7a5f052e2..fb16f17e59b 100644 --- a/arch/x86/kernel/acpi/cstate.c +++ b/arch/x86/kernel/acpi/cstate.c @@ -61,7 +61,7 @@ struct cstate_entry { unsigned int ecx; } states[ACPI_PROCESSOR_MAX_POWER]; }; -static struct cstate_entry *cpu_cstate_entry; /* per CPU ptr */ +static struct cstate_entry __percpu *cpu_cstate_entry; /* per CPU ptr */ static short mwait_supported[ACPI_PROCESSOR_MAX_POWER]; -- cgit v1.2.3-70-g09d2 From b365a85c68161ea5db5476eb8845a91ceb1777ea Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 29 Sep 2010 10:41:05 +0200 Subject: x86, UV: Use allocated buffer in tlb_uv.c:tunables_read() The original code didn't check that the value returned from snprintf() was less than the size of the buffer. Although it didn't cause a runtime bug in this case, it makes the static checkers complain. Andrew Morton suggested a dynamically sized buffer would be cleaner. Suggested-by: Andrew Morton Signed-off-by: Dan Carpenter Cc: Cliff Wickman Cc: Jack Steiner Cc: Robin Holt LKML-Reference: <20100929083118.GA6376@bicker> Signed-off-by: Ingo Molnar --- arch/x86/kernel/tlb_uv.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c index 312ef029281..33e77e4f7c8 100644 --- a/arch/x86/kernel/tlb_uv.c +++ b/arch/x86/kernel/tlb_uv.c @@ -1001,10 +1001,10 @@ static int uv_ptc_seq_show(struct seq_file *file, void *data) static ssize_t tunables_read(struct file *file, char __user *userbuf, size_t count, loff_t *ppos) { - char buf[300]; + char *buf; int ret; - ret = snprintf(buf, 300, "%s %s %s\n%d %d %d %d %d %d %d %d %d\n", + buf = kasprintf(GFP_KERNEL, "%s %s %s\n%d %d %d %d %d %d %d %d %d\n", "max_bau_concurrent plugged_delay plugsb4reset", "timeoutsb4reset ipi_reset_limit complete_threshold", "congested_response_us congested_reps congested_period", @@ -1012,7 +1012,12 @@ static ssize_t tunables_read(struct file *file, char __user *userbuf, timeoutsb4reset, ipi_reset_limit, complete_threshold, congested_response_us, congested_reps, congested_period); - return simple_read_from_buffer(userbuf, count, ppos, buf, ret); + if (!buf) + return -ENOMEM; + + ret = simple_read_from_buffer(userbuf, count, ppos, buf, strlen(buf)); + kfree(buf); + return ret; } /* -- cgit v1.2.3-70-g09d2 From 03e22198d2379ffa746c9ea332fbb1f094f9423b Mon Sep 17 00:00:00 2001 From: Cyrill Gorcunov Date: Wed, 29 Sep 2010 23:01:38 -0400 Subject: perf, x86: Handle in flight NMIs on P4 platform Stephane reported we've forgot to guard the P4 platform against spurious in-flight performance IRQs. Fix it. This fixes potential spurious 'dazed and confused' NMI messages. Reported-by: Stephane Eranian Signed-off-by: Cyrill Gorcunov Signed-off-by: Don Zickus Cc: fweisbec@gmail.com Cc: peterz@infradead.org Cc: Robert Richter Cc: Lin Ming LKML-Reference: <1285815698-4298-1-git-send-email-dzickus@redhat.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event_p4.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/perf_event_p4.c b/arch/x86/kernel/cpu/perf_event_p4.c index b560db3305b..24901517399 100644 --- a/arch/x86/kernel/cpu/perf_event_p4.c +++ b/arch/x86/kernel/cpu/perf_event_p4.c @@ -660,8 +660,12 @@ static int p4_pmu_handle_irq(struct pt_regs *regs) for (idx = 0; idx < x86_pmu.num_counters; idx++) { int overflow; - if (!test_bit(idx, cpuc->active_mask)) + if (!test_bit(idx, cpuc->active_mask)) { + /* catch in-flight IRQs */ + if (__test_and_clear_bit(idx, cpuc->running)) + handled++; continue; + } event = cpuc->events[idx]; hwc = &event->hw; -- cgit v1.2.3-70-g09d2 From 86cf1474948a91c67b3dd58f8b360006c82e0068 Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Fri, 13 Aug 2010 23:00:11 +0900 Subject: [CPUFREQ] acpi-cpufreq: add missing __percpu markup acpi_perf_data is a percpu pointer but was missing __percpu markup. Add it. Signed-off-by: Namhyung Kim Acked-by: Tejun Heo Signed-off-by: Dave Jones --- arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c index 246cd3afbb5..cd8da247dda 100644 --- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c +++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c @@ -72,7 +72,7 @@ struct acpi_cpufreq_data { static DEFINE_PER_CPU(struct acpi_cpufreq_data *, acfreq_data); /* acpi_perf_data is a pointer to percpu data. */ -static struct acpi_processor_performance *acpi_perf_data; +static struct acpi_processor_performance __percpu *acpi_perf_data; static struct cpufreq_driver acpi_cpufreq_driver; -- cgit v1.2.3-70-g09d2 From 3682930623f63c693845d9620c6bcdf5598c9bbb Mon Sep 17 00:00:00 2001 From: Pekka Enberg Date: Thu, 30 Sep 2010 22:57:33 +0300 Subject: [CPUFREQ] Fix memory leaks in pcc_cpufreq_do_osc If acpi_evaluate_object() function call doesn't fail, we must kfree() output.buffer before returning from pcc_cpufreq_do_osc(). Signed-off-by: Pekka Enberg Acked-by: David Rientjes Signed-off-by: Dave Jones --- arch/x86/kernel/cpu/cpufreq/pcc-cpufreq.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/cpufreq/pcc-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/pcc-cpufreq.c index 994230d4dc4..4f6f679f279 100644 --- a/arch/x86/kernel/cpu/cpufreq/pcc-cpufreq.c +++ b/arch/x86/kernel/cpu/cpufreq/pcc-cpufreq.c @@ -368,16 +368,22 @@ static int __init pcc_cpufreq_do_osc(acpi_handle *handle) return -ENODEV; out_obj = output.pointer; - if (out_obj->type != ACPI_TYPE_BUFFER) - return -ENODEV; + if (out_obj->type != ACPI_TYPE_BUFFER) { + ret = -ENODEV; + goto out_free; + } errors = *((u32 *)out_obj->buffer.pointer) & ~(1 << 0); - if (errors) - return -ENODEV; + if (errors) { + ret = -ENODEV; + goto out_free; + } supported = *((u32 *)(out_obj->buffer.pointer + 4)); - if (!(supported & 0x1)) - return -ENODEV; + if (!(supported & 0x1)) { + ret = -ENODEV; + goto out_free; + } out_free: kfree(output.pointer); -- cgit v1.2.3-70-g09d2 From 1cf180c94e9166cda083ff65333883ab3648e852 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 28 Sep 2010 20:57:19 +0200 Subject: x86, irq: Plug memory leak in sparse irq free_irq_cfg() is not freeing the cpumask_vars in irq_cfg. Fixing this triggers a use after free caused by the fact that copying struct irq_cfg is done with memcpy, which copies the pointer not the cpumask. Fix both places. Signed-off-by: Thomas Gleixner Cc: Yinghai Lu LKML-Reference: Signed-off-by: Thomas Gleixner Cc: stable@kernel.org Signed-off-by: H. Peter Anvin --- arch/x86/kernel/apic/io_apic.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index f1efebaf551..5c5b8f3dddb 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -306,14 +306,19 @@ void arch_init_copy_chip_data(struct irq_desc *old_desc, old_cfg = old_desc->chip_data; - memcpy(cfg, old_cfg, sizeof(struct irq_cfg)); + cfg->vector = old_cfg->vector; + cfg->move_in_progress = old_cfg->move_in_progress; + cpumask_copy(cfg->domain, old_cfg->domain); + cpumask_copy(cfg->old_domain, old_cfg->old_domain); init_copy_irq_2_pin(old_cfg, cfg, node); } -static void free_irq_cfg(struct irq_cfg *old_cfg) +static void free_irq_cfg(struct irq_cfg *cfg) { - kfree(old_cfg); + free_cpumask_var(cfg->domain); + free_cpumask_var(cfg->old_domain); + kfree(cfg); } void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc) -- cgit v1.2.3-70-g09d2 From 021989622810b02aab4b24f91e1f5ada2b654579 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 28 Sep 2010 23:20:23 +0200 Subject: x86, hpet: Fix bogus error check in hpet_assign_irq() create_irq() returns -1 if the interrupt allocation failed, but the code checks for irq == 0. Use create_irq_nr() instead. Signed-off-by: Thomas Gleixner Cc: Venkatesh Pallipadi LKML-Reference: Cc: stable@kernel.org Signed-off-by: H. Peter Anvin --- arch/x86/kernel/hpet.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index 410fdb3f193..7494999141b 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -506,7 +506,7 @@ static int hpet_assign_irq(struct hpet_dev *dev) { unsigned int irq; - irq = create_irq(); + irq = create_irq_nr(0, -1); if (!irq) return -EINVAL; -- cgit v1.2.3-70-g09d2 From 40c6b3cb64cd1d02322df5f729cca25084580f40 Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Wed, 29 Sep 2010 10:46:46 -0400 Subject: oprofile, x86: Using struct stack_frame for 64bit processes dump Removing unnecessary struct frame_head and replacing it with struct stack_frame. The struct stack_frame is already defined and used in other places in kernel, so there's no reason to define new structure. Signed-off-by: Jiri Olsa Signed-off-by: Robert Richter --- arch/x86/oprofile/backtrace.c | 19 +++++++------------ 1 file changed, 7 insertions(+), 12 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/oprofile/backtrace.c b/arch/x86/oprofile/backtrace.c index 3855096c59b..d640a86198b 100644 --- a/arch/x86/oprofile/backtrace.c +++ b/arch/x86/oprofile/backtrace.c @@ -48,35 +48,30 @@ static struct stacktrace_ops backtrace_ops = { .walk_stack = print_context_stack, }; -struct frame_head { - struct frame_head *bp; - unsigned long ret; -} __attribute__((packed)); - -static struct frame_head *dump_user_backtrace(struct frame_head *head) +static struct stack_frame *dump_user_backtrace(struct stack_frame *head) { - struct frame_head bufhead[2]; + struct stack_frame bufhead[2]; - /* Also check accessibility of one struct frame_head beyond */ + /* Also check accessibility of one struct stack_frame beyond */ if (!access_ok(VERIFY_READ, head, sizeof(bufhead))) return NULL; if (__copy_from_user_inatomic(bufhead, head, sizeof(bufhead))) return NULL; - oprofile_add_trace(bufhead[0].ret); + oprofile_add_trace(bufhead[0].return_address); /* frame pointers should strictly progress back up the stack * (towards higher addresses) */ - if (head >= bufhead[0].bp) + if (head >= bufhead[0].next_frame) return NULL; - return bufhead[0].bp; + return bufhead[0].next_frame; } void x86_backtrace(struct pt_regs * const regs, unsigned int depth) { - struct frame_head *head = (struct frame_head *)frame_pointer(regs); + struct stack_frame *head = (struct stack_frame *)frame_pointer(regs); if (!user_mode_vm(regs)) { unsigned long stack = kernel_stack_pointer(regs); -- cgit v1.2.3-70-g09d2 From f6dedecc37164a58bb80ae2ed9d204669ffc4850 Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Wed, 29 Sep 2010 10:46:47 -0400 Subject: oprofile, x86: Adding backtrace dump for 32bit process in compat mode This patch implements the oprofile backtrace generation for 32 bit applications running in the 64bit environment (compat mode). With this change it's possible to get backtrace for 32bits applications under the 64bits environment using oprofile's callgraph options. opcontrol --setup -c ... opreport -l -cg ... Signed-off-by: Jiri Olsa Signed-off-by: Robert Richter --- arch/x86/oprofile/backtrace.c | 53 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 53 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/oprofile/backtrace.c b/arch/x86/oprofile/backtrace.c index d640a86198b..2d49d4e19a3 100644 --- a/arch/x86/oprofile/backtrace.c +++ b/arch/x86/oprofile/backtrace.c @@ -14,6 +14,7 @@ #include #include #include +#include static void backtrace_warning_symbol(void *data, char *msg, unsigned long symbol) @@ -48,6 +49,55 @@ static struct stacktrace_ops backtrace_ops = { .walk_stack = print_context_stack, }; +#ifdef CONFIG_COMPAT +static struct stack_frame_ia32 * +dump_user_backtrace_32(struct stack_frame_ia32 *head) +{ + struct stack_frame_ia32 bufhead[2]; + struct stack_frame_ia32 *fp; + + /* Also check accessibility of one struct frame_head beyond */ + if (!access_ok(VERIFY_READ, head, sizeof(bufhead))) + return NULL; + if (__copy_from_user_inatomic(bufhead, head, sizeof(bufhead))) + return NULL; + + fp = (struct stack_frame_ia32 *) compat_ptr(bufhead[0].next_frame); + + oprofile_add_trace(bufhead[0].return_address); + + /* frame pointers should strictly progress back up the stack + * (towards higher addresses) */ + if (head >= fp) + return NULL; + + return fp; +} + +static inline int +x86_backtrace_32(struct pt_regs * const regs, unsigned int depth) +{ + struct stack_frame_ia32 *head; + + /* User process is 32-bit */ + if (!current || !test_thread_flag(TIF_IA32)) + return 0; + + head = (struct stack_frame_ia32 *) regs->bp; + while (depth-- && head) + head = dump_user_backtrace_32(head); + + return 1; +} + +#else +static inline int +x86_backtrace_32(struct pt_regs * const regs, unsigned int depth) +{ + return 0; +} +#endif /* CONFIG_COMPAT */ + static struct stack_frame *dump_user_backtrace(struct stack_frame *head) { struct stack_frame bufhead[2]; @@ -81,6 +131,9 @@ x86_backtrace(struct pt_regs * const regs, unsigned int depth) return; } + if (x86_backtrace_32(regs, depth)) + return; + while (depth-- && head) head = dump_user_backtrace(head); } -- cgit v1.2.3-70-g09d2 From 5140434d5f82f2e2119926272ada2e9731ec04f1 Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Thu, 30 Sep 2010 18:55:47 +0200 Subject: oprofile, x86: Simplify init/exit functions Now, that we only call the exit function if init succeeds with commit: 979048e oprofile: don't call arch exit code from init code on failure we can simplify the x86 init/exit functions too. Variable using_nmi becomes obsolete. Signed-off-by: Robert Richter --- arch/x86/oprofile/nmi_int.c | 9 +-------- 1 file changed, 1 insertion(+), 8 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c index f1575c9a257..bd1489c3ce0 100644 --- a/arch/x86/oprofile/nmi_int.c +++ b/arch/x86/oprofile/nmi_int.c @@ -695,9 +695,6 @@ static int __init ppro_init(char **cpu_type) return 1; } -/* in order to get sysfs right */ -static int using_nmi; - int __init op_nmi_init(struct oprofile_operations *ops) { __u8 vendor = boot_cpu_data.x86_vendor; @@ -705,8 +702,6 @@ int __init op_nmi_init(struct oprofile_operations *ops) char *cpu_type = NULL; int ret = 0; - using_nmi = 0; - if (!cpu_has_apic) return -ENODEV; @@ -790,13 +785,11 @@ int __init op_nmi_init(struct oprofile_operations *ops) if (ret) return ret; - using_nmi = 1; printk(KERN_INFO "oprofile: using NMI interrupt.\n"); return 0; } void op_nmi_exit(void) { - if (using_nmi) - exit_sysfs(); + exit_sysfs(); } -- cgit v1.2.3-70-g09d2 From 3fdbf004c1706480a7c7fac3c9d836fa6df20d7d Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Thu, 30 Sep 2010 14:32:35 +0200 Subject: x86, mtrr: Assume SYS_CFG[Tom2ForceMemTypeWB] exists on all future AMD CPUs Instead of adapting the CPU family check in amd_special_default_mtrr() for each new CPU family assume that all new AMD CPUs support the necessary bits in SYS_CFG MSR. Tom2Enabled is architectural (defined in APM Vol.2). Tom2ForceMemTypeWB is defined in all BKDGs starting with K8 NPT. In pre K8-NPT BKDG this bit is reserved (read as zero). W/o this adaption Linux would unnecessarily complain about bad MTRR settings on every new AMD CPU family, e.g. [ 0.000000] WARNING: BIOS bug: CPU MTRRs don't cover all of memory, losing 4863MB of RAM. Cc: stable@kernel.org # .32.x, .35.x Signed-off-by: Andreas Herrmann LKML-Reference: <20100930123235.GB20545@loge.amd.com> Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/mtrr/cleanup.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/mtrr/cleanup.c b/arch/x86/kernel/cpu/mtrr/cleanup.c index c5f59d07142..ac140c7be39 100644 --- a/arch/x86/kernel/cpu/mtrr/cleanup.c +++ b/arch/x86/kernel/cpu/mtrr/cleanup.c @@ -827,7 +827,7 @@ int __init amd_special_default_mtrr(void) if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD) return 0; - if (boot_cpu_data.x86 < 0xf || boot_cpu_data.x86 > 0x11) + if (boot_cpu_data.x86 < 0xf) return 0; /* In case some hypervisor doesn't pass SYSCFG through: */ if (rdmsr_safe(MSR_K8_SYSCFG, &l, &h) < 0) -- cgit v1.2.3-70-g09d2 From 420b13b60a3e5c5dcc6ec290e131cf5fbc603d94 Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Thu, 30 Sep 2010 14:33:58 +0200 Subject: x86, nmi: Support NMI watchdog on newer AMD CPU families CPU families 0x12, 0x14 and 0x15 support this functionality. Signed-off-by: Andreas Herrmann LKML-Reference: <20100930123357.GC20545@loge.amd.com> Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/perfctr-watchdog.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c index fb329e9f849..d9f4ff8fcd6 100644 --- a/arch/x86/kernel/cpu/perfctr-watchdog.c +++ b/arch/x86/kernel/cpu/perfctr-watchdog.c @@ -700,11 +700,10 @@ static void probe_nmi_watchdog(void) { switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_AMD: - if (boot_cpu_data.x86 != 6 && boot_cpu_data.x86 != 15 && - boot_cpu_data.x86 != 16 && boot_cpu_data.x86 != 17) - return; - wd_ops = &k7_wd_ops; - break; + if (boot_cpu_data.x86 == 6 || + (boot_cpu_data.x86 >= 0xf && boot_cpu_data.x86 <= 0x15)) + wd_ops = &k7_wd_ops; + return; case X86_VENDOR_INTEL: /* Work around where perfctr1 doesn't have a working enable * bit as described in the following errata: -- cgit v1.2.3-70-g09d2 From 23588c38a84c9175c6668789b64ffba4651e5c6a Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Thu, 30 Sep 2010 14:36:28 +0200 Subject: x86, amd: Add support for CPUID topology extension of AMD CPUs Node information (ID, number of internal nodes) is provided via CPUID Fn8000_001e_ECX. See AMD CPUID Specification (Publication # 25481, Revision 2.34, September 2010). Signed-off-by: Andreas Herrmann LKML-Reference: <20100930123628.GD20545@loge.amd.com> Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/amd.c | 50 ++++++++++++++++++++++++----------------------- 1 file changed, 26 insertions(+), 24 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 0f0ace5d7db..7e6a37d2425 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -253,37 +253,41 @@ static int __cpuinit nearby_node(int apicid) #endif /* - * Fixup core topology information for AMD multi-node processors. - * Assumption: Number of cores in each internal node is the same. + * Fixup core topology information for + * (1) AMD multi-node processors + * Assumption: Number of cores in each internal node is the same. */ #ifdef CONFIG_X86_HT -static void __cpuinit amd_fixup_dcm(struct cpuinfo_x86 *c) +static void __cpuinit amd_get_topology(struct cpuinfo_x86 *c) { - unsigned long long value; u32 nodes, cores_per_node; + u8 node_id; + unsigned long long value; int cpu = smp_processor_id(); - if (!cpu_has(c, X86_FEATURE_NODEID_MSR)) - return; - - /* fixup topology information only once for a core */ - if (cpu_has(c, X86_FEATURE_AMD_DCM)) + /* get information required for multi-node processors */ + if (cpu_has(c, X86_FEATURE_TOPOEXT)) { + value = cpuid_ecx(0x8000001e); + nodes = ((value >> 8) & 7) + 1; + node_id = value & 7; + } else if (cpu_has(c, X86_FEATURE_NODEID_MSR)) { + rdmsrl(MSR_FAM10H_NODE_ID, value); + nodes = ((value >> 3) & 7) + 1; + node_id = value & 7; + } else return; - rdmsrl(MSR_FAM10H_NODE_ID, value); + /* fixup multi-node processor information */ + if (nodes > 1) { + set_cpu_cap(c, X86_FEATURE_AMD_DCM); + cores_per_node = c->x86_max_cores / nodes; - nodes = ((value >> 3) & 7) + 1; - if (nodes == 1) - return; - - set_cpu_cap(c, X86_FEATURE_AMD_DCM); - cores_per_node = c->x86_max_cores / nodes; + /* store NodeID, use llc_shared_map to store sibling info */ + per_cpu(cpu_llc_id, cpu) = node_id; - /* store NodeID, use llc_shared_map to store sibling info */ - per_cpu(cpu_llc_id, cpu) = value & 7; - - /* fixup core id to be in range from 0 to (cores_per_node - 1) */ - c->cpu_core_id = c->cpu_core_id % cores_per_node; + /* core id to be in range from 0 to (cores_per_node - 1) */ + c->cpu_core_id = c->cpu_core_id % cores_per_node; + } } #endif @@ -304,9 +308,7 @@ static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c) c->phys_proc_id = c->initial_apicid >> bits; /* use socket ID also for last level cache */ per_cpu(cpu_llc_id, cpu) = c->phys_proc_id; - /* fixup topology information on multi-node processors */ - if ((c->x86 == 0x10) && (c->x86_model == 9)) - amd_fixup_dcm(c); + amd_get_topology(c); #endif } -- cgit v1.2.3-70-g09d2 From 6057b4d331f19a3ea51aec463ea7839c128b3227 Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Thu, 30 Sep 2010 14:38:57 +0200 Subject: x86, amd: Extract compute unit information for AMD CPUs Get compute unit information from CPUID Fn8000_001E_EBX. (See AMD CPUID Specification - publication # 25481, revision 2.34, September 2010.) Note that each core on a compute unit still has a core_id of its own. Signed-off-by: Andreas Herrmann LKML-Reference: <20100930123857.GE20545@loge.amd.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/processor.h | 2 ++ arch/x86/kernel/cpu/amd.c | 20 +++++++++++++++----- 2 files changed, 17 insertions(+), 5 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index 325b7bdbeba..69e80c2ec6c 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -110,6 +110,8 @@ struct cpuinfo_x86 { u16 phys_proc_id; /* Core id: */ u16 cpu_core_id; + /* Compute unit id */ + u8 compute_unit_id; /* Index into per_cpu list: */ u16 cpu_index; #endif diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 7e6a37d2425..70168ab88b7 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -256,21 +256,29 @@ static int __cpuinit nearby_node(int apicid) * Fixup core topology information for * (1) AMD multi-node processors * Assumption: Number of cores in each internal node is the same. + * (2) AMD processors supporting compute units */ #ifdef CONFIG_X86_HT static void __cpuinit amd_get_topology(struct cpuinfo_x86 *c) { - u32 nodes, cores_per_node; + u32 nodes; u8 node_id; - unsigned long long value; int cpu = smp_processor_id(); /* get information required for multi-node processors */ if (cpu_has(c, X86_FEATURE_TOPOEXT)) { - value = cpuid_ecx(0x8000001e); - nodes = ((value >> 8) & 7) + 1; - node_id = value & 7; + u32 eax, ebx, ecx, edx; + + cpuid(0x8000001e, &eax, &ebx, &ecx, &edx); + nodes = ((ecx >> 8) & 7) + 1; + node_id = ecx & 7; + + /* get compute unit information */ + smp_num_siblings = ((ebx >> 8) & 3) + 1; + c->compute_unit_id = ebx & 0xff; } else if (cpu_has(c, X86_FEATURE_NODEID_MSR)) { + u64 value; + rdmsrl(MSR_FAM10H_NODE_ID, value); nodes = ((value >> 3) & 7) + 1; node_id = value & 7; @@ -279,6 +287,8 @@ static void __cpuinit amd_get_topology(struct cpuinfo_x86 *c) /* fixup multi-node processor information */ if (nodes > 1) { + u32 cores_per_node; + set_cpu_cap(c, X86_FEATURE_AMD_DCM); cores_per_node = c->x86_max_cores / nodes; -- cgit v1.2.3-70-g09d2 From d4fbe4f03557e1fd4d9bbb3a1957aad560f39e96 Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Thu, 30 Sep 2010 14:41:56 +0200 Subject: x86, amd: Use compute unit information to determine thread siblings This information is vital for different load balancing policies. Signed-off-by: Andreas Herrmann LKML-Reference: <20100930124156.GF20545@loge.amd.com> Signed-off-by: H. Peter Anvin --- arch/x86/kernel/smpboot.c | 28 ++++++++++++++++++++-------- 1 file changed, 20 insertions(+), 8 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 8b3bfc4dd70..bc2cc444844 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -397,6 +397,19 @@ void __cpuinit smp_store_cpu_info(int id) identify_secondary_cpu(c); } +static void __cpuinit link_thread_siblings(int cpu1, int cpu2) +{ + struct cpuinfo_x86 *c1 = &cpu_data(cpu1); + struct cpuinfo_x86 *c2 = &cpu_data(cpu2); + + cpumask_set_cpu(cpu1, cpu_sibling_mask(cpu2)); + cpumask_set_cpu(cpu2, cpu_sibling_mask(cpu1)); + cpumask_set_cpu(cpu1, cpu_core_mask(cpu2)); + cpumask_set_cpu(cpu2, cpu_core_mask(cpu1)); + cpumask_set_cpu(cpu1, c2->llc_shared_map); + cpumask_set_cpu(cpu2, c1->llc_shared_map); +} + void __cpuinit set_cpu_sibling_map(int cpu) { @@ -409,14 +422,13 @@ void __cpuinit set_cpu_sibling_map(int cpu) for_each_cpu(i, cpu_sibling_setup_mask) { struct cpuinfo_x86 *o = &cpu_data(i); - if (c->phys_proc_id == o->phys_proc_id && - c->cpu_core_id == o->cpu_core_id) { - cpumask_set_cpu(i, cpu_sibling_mask(cpu)); - cpumask_set_cpu(cpu, cpu_sibling_mask(i)); - cpumask_set_cpu(i, cpu_core_mask(cpu)); - cpumask_set_cpu(cpu, cpu_core_mask(i)); - cpumask_set_cpu(i, c->llc_shared_map); - cpumask_set_cpu(cpu, o->llc_shared_map); + if (cpu_has(c, X86_FEATURE_TOPOEXT)) { + if (c->phys_proc_id == o->phys_proc_id && + c->compute_unit_id == o->compute_unit_id) + link_thread_siblings(cpu, i); + } else if (c->phys_proc_id == o->phys_proc_id && + c->cpu_core_id == o->cpu_core_id) { + link_thread_siblings(cpu, i); } } } else { -- cgit v1.2.3-70-g09d2 From 5c80cc78de46aef6cd5e714208da05c3f7f548f8 Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Thu, 30 Sep 2010 14:43:16 +0200 Subject: x86, amd_nb: Enable GART support for AMD family 0x15 CPUs AMD CPU family 0x15 still supports GART for compatibility reasons. Signed-off-by: Andreas Herrmann LKML-Reference: <20100930124316.GG20545@loge.amd.com> Signed-off-by: H. Peter Anvin --- arch/x86/kernel/amd_nb.c | 4 +++- include/linux/pci_ids.h | 1 + 2 files changed, 4 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/amd_nb.c b/arch/x86/kernel/amd_nb.c index 4ffc38df7ac..8f6463d8ed0 100644 --- a/arch/x86/kernel/amd_nb.c +++ b/arch/x86/kernel/amd_nb.c @@ -15,6 +15,7 @@ static u32 *flush_words; struct pci_device_id k8_nb_ids[] = { { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_K8_NB_MISC) }, { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_10H_NB_MISC) }, + { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_15H_NB_MISC) }, {} }; EXPORT_SYMBOL(k8_nb_ids); @@ -45,7 +46,8 @@ int cache_k8_northbridges(void) k8_northbridges.num++; /* some CPU families (e.g. family 0x11) do not support GART */ - if (boot_cpu_data.x86 == 0xf || boot_cpu_data.x86 == 0x10) + if (boot_cpu_data.x86 == 0xf || boot_cpu_data.x86 == 0x10 || + boot_cpu_data.x86 == 0x15) k8_northbridges.gart_supported = 1; k8_northbridges.nb_misc = kmalloc((k8_northbridges.num + 1) * diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 10d33309e9a..edc0279be72 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -514,6 +514,7 @@ #define PCI_DEVICE_ID_AMD_11H_NB_DRAM 0x1302 #define PCI_DEVICE_ID_AMD_11H_NB_MISC 0x1303 #define PCI_DEVICE_ID_AMD_11H_NB_LINK 0x1304 +#define PCI_DEVICE_ID_AMD_15H_NB_MISC 0x1603 #define PCI_DEVICE_ID_AMD_LANCE 0x2000 #define PCI_DEVICE_ID_AMD_LANCE_HOME 0x2001 #define PCI_DEVICE_ID_AMD_SCSI 0x2020 -- cgit v1.2.3-70-g09d2 From 3bb9808e99bcc36eecb8e082bf70efb2a0bcdcb7 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 27 Sep 2010 12:46:02 +0000 Subject: x86: Use genirq Kconfig Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra LKML-Reference: <20100927121843.314600915@linutronix.de> Reviewed-by: H. Peter Anvin Reviewed-by: Ingo Molnar --- arch/x86/Kconfig | 36 +++++------------------------------- 1 file changed, 5 insertions(+), 31 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index cea0cd9a316..3ec657f7ee7 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -59,6 +59,11 @@ config X86 select ANON_INODES select HAVE_ARCH_KMEMCHECK select HAVE_USER_RETURN_NOTIFIER + select HAVE_GENERIC_HARDIRQS + select HAVE_SPARSE_IRQ + select NUMA_IRQ_DESC if (SPARSE_IRQ && NUMA) + select GENERIC_IRQ_PROBE + select GENERIC_PENDING_IRQ if SMP config INSTRUCTION_DECODER def_bool (KPROBES || PERF_EVENTS) @@ -200,20 +205,6 @@ config HAVE_INTEL_TXT def_bool y depends on EXPERIMENTAL && DMAR && ACPI -# Use the generic interrupt handling code in kernel/irq/: -config GENERIC_HARDIRQS - def_bool y - -config GENERIC_HARDIRQS_NO__DO_IRQ - def_bool y - -config GENERIC_IRQ_PROBE - def_bool y - -config GENERIC_PENDING_IRQ - def_bool y - depends on GENERIC_HARDIRQS && SMP - config USE_GENERIC_SMP_HELPERS def_bool y depends on SMP @@ -296,23 +287,6 @@ config X86_X2APIC If you don't know what to do here, say N. -config SPARSE_IRQ - bool "Support sparse irq numbering" - depends on PCI_MSI || HT_IRQ - ---help--- - This enables support for sparse irqs. This is useful for distro - kernels that want to define a high CONFIG_NR_CPUS value but still - want to have low kernel memory footprint on smaller machines. - - ( Sparse IRQs can also be beneficial on NUMA boxes, as they spread - out the irq_desc[] array in a more NUMA-friendly way. ) - - If you don't know what to do here, say N. - -config NUMA_IRQ_DESC - def_bool y - depends on SPARSE_IRQ && NUMA - config X86_MPPARSE bool "Enable MPS table" if ACPI default y -- cgit v1.2.3-70-g09d2 From 366d4a43b1a7a861c356a0e407c4f03f454d42ea Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Mon, 4 Oct 2010 09:31:27 +0200 Subject: x86, cpu: Fix X86_FEATURE_NOPL ba0593bf553c450a03dbc5f8c1f0ff58b778a0c8 cleared the aforementioned cpuid bit only on 32-bit due to various problems with Virtual PC. This somehow got lost during the 32- + 64-bit merge so restore the feature bit on 64-bit. For that, set it explicitly for non-constant arguments of cpu_has(). Update comment for future reference. Signed-off-by: Borislav Petkov LKML-Reference: <20101004073127.GA20305@liondog.tnic> Cc: Ryan O'Neill Cc: Linus Torvalds Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/common.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index f2f9ac7da25..927e630aeaf 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -704,16 +704,21 @@ void __init early_cpu_init(void) } /* - * The NOPL instruction is supposed to exist on all CPUs with - * family >= 6; unfortunately, that's not true in practice because - * of early VIA chips and (more importantly) broken virtualizers that - * are not easy to detect. In the latter case it doesn't even *fail* - * reliably, so probing for it doesn't even work. Disable it completely + * The NOPL instruction is supposed to exist on all CPUs of family >= 6; + * unfortunately, that's not true in practice because of early VIA + * chips and (more importantly) broken virtualizers that are not easy + * to detect. In the latter case it doesn't even *fail* reliably, so + * probing for it doesn't even work. Disable it completely on 32-bit * unless we can find a reliable way to detect all the broken cases. + * Enable it explicitly on 64-bit for non-constant inputs of cpu_has(). */ static void __cpuinit detect_nopl(struct cpuinfo_x86 *c) { +#ifdef CONFIG_X86_32 clear_cpu_cap(c, X86_FEATURE_NOPL); +#else + set_cpu_cap(c, X86_FEATURE_NOPL); +#endif } static void __cpuinit generic_identify(struct cpuinfo_x86 *c) -- cgit v1.2.3-70-g09d2 From c62f981f9309486ba5546edbb5925f71e441fa65 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Fri, 1 Oct 2010 14:18:47 -0700 Subject: perf, gcc-4.6: Fix set but unused variable Just dead code I believe. Signed-off-by: Andi Kleen Signed-off-by: Andrew Morton Cc: andi@firstfloor.org Cc: Peter Zijlstra Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 6526a86616a..e2513f26ba8 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -1152,7 +1152,6 @@ static int x86_pmu_handle_irq(struct pt_regs *regs) struct perf_sample_data data; struct cpu_hw_events *cpuc; struct perf_event *event; - struct hw_perf_event *hwc; int idx, handled = 0; u64 val; @@ -1173,7 +1172,6 @@ static int x86_pmu_handle_irq(struct pt_regs *regs) } event = cpuc->events[idx]; - hwc = &event->hw; val = x86_perf_event_update(event); if (val & (1ULL << (x86_pmu.cntval_bits - 1))) -- cgit v1.2.3-70-g09d2 From 31e7e931cdc27f76dc68444edc4df1c0d1bfa6cc Mon Sep 17 00:00:00 2001 From: Stefano Stabellini Date: Fri, 1 Oct 2010 17:35:46 +0100 Subject: xen: do not initialize PV timers on HVM if !xen_have_vector_callback if !xen_have_vector_callback do not initialize PV timer unconditionally because we still don't know how many cpus are available and if there is more than one we won't be able to receive the timer interrupts on cpu > 0. This patch fixes an hang at boot when Xen does not support vector callbacks and the guest has multiple vcpus. Signed-off-by: Stefano Stabellini Acked-by: Jeremy Fitzhardinge --- arch/x86/xen/time.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c index 1a5353a753f..b2bb5aa3b05 100644 --- a/arch/x86/xen/time.c +++ b/arch/x86/xen/time.c @@ -489,8 +489,9 @@ static void xen_hvm_setup_cpu_clockevents(void) __init void xen_hvm_init_time_ops(void) { /* vector callback is needed otherwise we cannot receive interrupts - * on cpu > 0 */ - if (!xen_have_vector_callback && num_present_cpus() > 1) + * on cpu > 0 and at this point we don't know how many cpus are + * available */ + if (!xen_have_vector_callback) return; if (!xen_feature(XENFEAT_hvm_safe_pvclock)) { printk(KERN_INFO "Xen doesn't support pvclock on HVM," -- cgit v1.2.3-70-g09d2 From 5336377d6225959624146629ce3fc88ee8ecda3d Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 5 Oct 2010 11:29:27 -0700 Subject: modules: Fix module_bug_list list corruption race With all the recent module loading cleanups, we've minimized the code that sits under module_mutex, fixing various deadlocks and making it possible to do most of the module loading in parallel. However, that whole conversion totally missed the rather obscure code that adds a new module to the list for BUG() handling. That code was doubly obscure because (a) the code itself lives in lib/bugs.c (for dubious reasons) and (b) it gets called from the architecture-specific "module_finalize()" rather than from generic code. Calling it from arch-specific code makes no sense what-so-ever to begin with, and is now actively wrong since that code isn't protected by the module loading lock any more. So this commit moves the "module_bug_{finalize,cleanup}()" calls away from the arch-specific code, and into the generic code - and in the process protects it with the module_mutex so that the list operations are now safe. Future fixups: - move the module list handling code into kernel/module.c where it belongs. - get rid of 'module_bug_list' and just use the regular list of modules (called 'modules' - imagine that) that we already create and maintain for other reasons. Reported-and-tested-by: Thomas Gleixner Cc: Rusty Russell Cc: Adrian Bunk Cc: Andrew Morton Cc: stable@kernel.org Signed-off-by: Linus Torvalds --- arch/avr32/kernel/module.c | 3 +-- arch/h8300/kernel/module.c | 3 +-- arch/mn10300/kernel/module.c | 3 +-- arch/parisc/kernel/module.c | 3 +-- arch/powerpc/kernel/module.c | 5 ----- arch/s390/kernel/module.c | 3 +-- arch/sh/kernel/module.c | 2 -- arch/x86/kernel/module.c | 3 +-- include/linux/module.h | 5 ++--- kernel/module.c | 4 ++++ lib/bug.c | 6 ++---- 11 files changed, 14 insertions(+), 26 deletions(-) (limited to 'arch/x86') diff --git a/arch/avr32/kernel/module.c b/arch/avr32/kernel/module.c index 98f94d041d9..a727f54d64d 100644 --- a/arch/avr32/kernel/module.c +++ b/arch/avr32/kernel/module.c @@ -314,10 +314,9 @@ int module_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs, vfree(module->arch.syminfo); module->arch.syminfo = NULL; - return module_bug_finalize(hdr, sechdrs, module); + return 0; } void module_arch_cleanup(struct module *module) { - module_bug_cleanup(module); } diff --git a/arch/h8300/kernel/module.c b/arch/h8300/kernel/module.c index 0865e291c20..db4953dc4e1 100644 --- a/arch/h8300/kernel/module.c +++ b/arch/h8300/kernel/module.c @@ -112,10 +112,9 @@ int module_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs, struct module *me) { - return module_bug_finalize(hdr, sechdrs, me); + return 0; } void module_arch_cleanup(struct module *mod) { - module_bug_cleanup(mod); } diff --git a/arch/mn10300/kernel/module.c b/arch/mn10300/kernel/module.c index 6aea7fd7699..196a111e2e2 100644 --- a/arch/mn10300/kernel/module.c +++ b/arch/mn10300/kernel/module.c @@ -206,7 +206,7 @@ int module_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs, struct module *me) { - return module_bug_finalize(hdr, sechdrs, me); + return 0; } /* @@ -214,5 +214,4 @@ int module_finalize(const Elf_Ehdr *hdr, */ void module_arch_cleanup(struct module *mod) { - module_bug_cleanup(mod); } diff --git a/arch/parisc/kernel/module.c b/arch/parisc/kernel/module.c index 159a2b81e90..6e81bb596e5 100644 --- a/arch/parisc/kernel/module.c +++ b/arch/parisc/kernel/module.c @@ -941,11 +941,10 @@ int module_finalize(const Elf_Ehdr *hdr, nsyms = newptr - (Elf_Sym *)symhdr->sh_addr; DEBUGP("NEW num_symtab %lu\n", nsyms); symhdr->sh_size = nsyms * sizeof(Elf_Sym); - return module_bug_finalize(hdr, sechdrs, me); + return 0; } void module_arch_cleanup(struct module *mod) { deregister_unwind_table(mod); - module_bug_cleanup(mod); } diff --git a/arch/powerpc/kernel/module.c b/arch/powerpc/kernel/module.c index 477c663e014..4ef93ae2235 100644 --- a/arch/powerpc/kernel/module.c +++ b/arch/powerpc/kernel/module.c @@ -65,10 +65,6 @@ int module_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sect; int err; - err = module_bug_finalize(hdr, sechdrs, me); - if (err) - return err; - /* Apply feature fixups */ sect = find_section(hdr, sechdrs, "__ftr_fixup"); if (sect != NULL) @@ -101,5 +97,4 @@ int module_finalize(const Elf_Ehdr *hdr, void module_arch_cleanup(struct module *mod) { - module_bug_cleanup(mod); } diff --git a/arch/s390/kernel/module.c b/arch/s390/kernel/module.c index 22cfd634c35..f7167ee4604 100644 --- a/arch/s390/kernel/module.c +++ b/arch/s390/kernel/module.c @@ -407,10 +407,9 @@ int module_finalize(const Elf_Ehdr *hdr, { vfree(me->arch.syminfo); me->arch.syminfo = NULL; - return module_bug_finalize(hdr, sechdrs, me); + return 0; } void module_arch_cleanup(struct module *mod) { - module_bug_cleanup(mod); } diff --git a/arch/sh/kernel/module.c b/arch/sh/kernel/module.c index 43adddfe4c0..ae0be697a89 100644 --- a/arch/sh/kernel/module.c +++ b/arch/sh/kernel/module.c @@ -149,13 +149,11 @@ int module_finalize(const Elf_Ehdr *hdr, int ret = 0; ret |= module_dwarf_finalize(hdr, sechdrs, me); - ret |= module_bug_finalize(hdr, sechdrs, me); return ret; } void module_arch_cleanup(struct module *mod) { - module_bug_cleanup(mod); module_dwarf_cleanup(mod); } diff --git a/arch/x86/kernel/module.c b/arch/x86/kernel/module.c index e0bc186d750..1c355c55096 100644 --- a/arch/x86/kernel/module.c +++ b/arch/x86/kernel/module.c @@ -239,11 +239,10 @@ int module_finalize(const Elf_Ehdr *hdr, apply_paravirt(pseg, pseg + para->sh_size); } - return module_bug_finalize(hdr, sechdrs, me); + return 0; } void module_arch_cleanup(struct module *mod) { alternatives_smp_module_del(mod); - module_bug_cleanup(mod); } diff --git a/include/linux/module.h b/include/linux/module.h index 8a6b9fdc7ff..aace066bad8 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -686,17 +686,16 @@ extern int module_sysfs_initialized; #ifdef CONFIG_GENERIC_BUG -int module_bug_finalize(const Elf_Ehdr *, const Elf_Shdr *, +void module_bug_finalize(const Elf_Ehdr *, const Elf_Shdr *, struct module *); void module_bug_cleanup(struct module *); #else /* !CONFIG_GENERIC_BUG */ -static inline int module_bug_finalize(const Elf_Ehdr *hdr, +static inline void module_bug_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs, struct module *mod) { - return 0; } static inline void module_bug_cleanup(struct module *mod) {} #endif /* CONFIG_GENERIC_BUG */ diff --git a/kernel/module.c b/kernel/module.c index d0b5f8db11b..ccd64199184 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -1537,6 +1537,7 @@ static int __unlink_module(void *_mod) { struct module *mod = _mod; list_del(&mod->list); + module_bug_cleanup(mod); return 0; } @@ -2625,6 +2626,7 @@ static struct module *load_module(void __user *umod, if (err < 0) goto ddebug; + module_bug_finalize(info.hdr, info.sechdrs, mod); list_add_rcu(&mod->list, &modules); mutex_unlock(&module_mutex); @@ -2650,6 +2652,8 @@ static struct module *load_module(void __user *umod, mutex_lock(&module_mutex); /* Unlink carefully: kallsyms could be walking list. */ list_del_rcu(&mod->list); + module_bug_cleanup(mod); + ddebug: if (!mod->taints) dynamic_debug_remove(info.debug); diff --git a/lib/bug.c b/lib/bug.c index 7cdfad88128..19552096d16 100644 --- a/lib/bug.c +++ b/lib/bug.c @@ -72,8 +72,8 @@ static const struct bug_entry *module_find_bug(unsigned long bugaddr) return NULL; } -int module_bug_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs, - struct module *mod) +void module_bug_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs, + struct module *mod) { char *secstrings; unsigned int i; @@ -97,8 +97,6 @@ int module_bug_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs, * could potentially lead to deadlock and thus be counter-productive. */ list_add(&mod->bug_list, &module_bug_list); - - return 0; } void module_bug_cleanup(struct module *mod) -- cgit v1.2.3-70-g09d2 From 9f4c13964b58608fbce05540743281ea3146c0e8 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Tue, 5 Oct 2010 16:05:14 -0700 Subject: x86, memblock: Fix crashkernel allocation Cai Qian found crashkernel is broken with the x86 memblock changes. 1. crashkernel=128M@32M always reported that range is used, even if the first kernel is small and does not usethat range 2. we always got following report when using "kexec -p" Could not find a free area of memory of a000 bytes... locate_hole failed The root cause is that generic memblock_find_in_range() will try to allocate from the top of the range, whereas the kexec code was written assuming that allocation was always near the bottom and that it could blindly extend memory upward. Unfortunately the kexec code doesn't have a system for requesting the range that it really needs, so this is subject to probabilistic failures. This patch hacks around the problem by limiting the target range heuristically to below the traditional bzImage max range. This number is arbitrary and not always correct, and a much better result would be obtained by having kexec communicate this number based on the kernel header information and any appropriate command line options. Reported-and-Bisected-by: CAI Qian Signed-off-by: Yinghai Lu LKML-Reference: <4CABAF2A.5090501@kernel.org> Cc: Vivek Goyal Signed-off-by: H. Peter Anvin --- arch/x86/kernel/setup.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index bf89e0a59b8..b11a238b2e3 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -502,6 +502,7 @@ static inline unsigned long long get_total_mem(void) return total << PAGE_SHIFT; } +#define DEFAULT_BZIMAGE_ADDR_MAX 0x37FFFFFF static void __init reserve_crashkernel(void) { unsigned long long total_mem; @@ -519,8 +520,12 @@ static void __init reserve_crashkernel(void) if (crash_base <= 0) { const unsigned long long alignment = 16<<20; /* 16M */ - crash_base = memblock_find_in_range(alignment, ULONG_MAX, crash_size, - alignment); + /* + * kexec want bzImage is below DEFAULT_BZIMAGE_ADDR_MAX + */ + crash_base = memblock_find_in_range(alignment, + DEFAULT_BZIMAGE_ADDR_MAX, crash_size, alignment); + if (crash_base == MEMBLOCK_ERROR) { pr_info("crashkernel reservation failed - No suitable area found.\n"); return; @@ -528,8 +533,8 @@ static void __init reserve_crashkernel(void) } else { unsigned long long start; - start = memblock_find_in_range(crash_base, ULONG_MAX, crash_size, - 1<<20); + start = memblock_find_in_range(crash_base, + crash_base + crash_size, crash_size, 1<<20); if (start != crash_base) { pr_info("crashkernel reservation failed - memory is in use.\n"); return; -- cgit v1.2.3-70-g09d2 From 1d931264af0f10649b35afa8fbd2e169da51ac08 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Tue, 5 Oct 2010 16:15:15 -0700 Subject: x86-32, memblock: Make add_highpages honor early reserved ranges Originally the only early reserved range that is overlapped with high pages is "KVA RAM", but we already do remove that from the active ranges. However, It turns out Xen could have that kind of overlapping to support memory ballooning.x So we need to make add_highpage_with_active_regions() to subtract memblock reserved just like low ram; this is the proper design anyway. In this patch, refactering get_freel_all_memory_range() to make it can be used by add_highpage_with_active_regions(). Also we don't need to remove "KVA RAM" from active ranges. Signed-off-by: Yinghai Lu LKML-Reference: <4CABB183.1040607@kernel.org> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/memblock.h | 2 ++ arch/x86/mm/init_32.c | 53 +++++++++++++---------------------------- arch/x86/mm/memblock.c | 19 +++++++++++---- arch/x86/mm/numa_32.c | 2 -- 4 files changed, 33 insertions(+), 43 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/memblock.h b/arch/x86/include/asm/memblock.h index 2c304bb6e07..19ae14ba697 100644 --- a/arch/x86/include/asm/memblock.h +++ b/arch/x86/include/asm/memblock.h @@ -9,6 +9,8 @@ void memblock_x86_to_bootmem(u64 start, u64 end); void memblock_x86_reserve_range(u64 start, u64 end, char *name); void memblock_x86_free_range(u64 start, u64 end); struct range; +int __get_free_all_memory_range(struct range **range, int nodeid, + unsigned long start_pfn, unsigned long end_pfn); int get_free_all_memory_range(struct range **rangep, int nodeid); void memblock_x86_register_active_regions(int nid, unsigned long start_pfn, diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index c2385d7ae31..85467099d6d 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -423,49 +423,28 @@ static void __init add_one_highpage_init(struct page *page) totalhigh_pages++; } -struct add_highpages_data { - unsigned long start_pfn; - unsigned long end_pfn; -}; - -static int __init add_highpages_work_fn(unsigned long start_pfn, - unsigned long end_pfn, void *datax) +void __init add_highpages_with_active_regions(int nid, + unsigned long start_pfn, unsigned long end_pfn) { - int node_pfn; - struct page *page; - unsigned long final_start_pfn, final_end_pfn; - struct add_highpages_data *data; + struct range *range; + int nr_range; + int i; - data = (struct add_highpages_data *)datax; + nr_range = __get_free_all_memory_range(&range, nid, start_pfn, end_pfn); - final_start_pfn = max(start_pfn, data->start_pfn); - final_end_pfn = min(end_pfn, data->end_pfn); - if (final_start_pfn >= final_end_pfn) - return 0; + for (i = 0; i < nr_range; i++) { + struct page *page; + int node_pfn; - for (node_pfn = final_start_pfn; node_pfn < final_end_pfn; - node_pfn++) { - if (!pfn_valid(node_pfn)) - continue; - page = pfn_to_page(node_pfn); - add_one_highpage_init(page); + for (node_pfn = range[i].start; node_pfn < range[i].end; + node_pfn++) { + if (!pfn_valid(node_pfn)) + continue; + page = pfn_to_page(node_pfn); + add_one_highpage_init(page); + } } - - return 0; - } - -void __init add_highpages_with_active_regions(int nid, unsigned long start_pfn, - unsigned long end_pfn) -{ - struct add_highpages_data data; - - data.start_pfn = start_pfn; - data.end_pfn = end_pfn; - - work_with_active_regions(nid, add_highpages_work_fn, &data); -} - #else static inline void permanent_kmaps_init(pgd_t *pgd_base) { diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index 50ecbc59757..fd7a0404945 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -156,7 +156,8 @@ static int __init count_early_node_map(int nodeid) return data.nr; } -int __init get_free_all_memory_range(struct range **rangep, int nodeid) +int __init __get_free_all_memory_range(struct range **rangep, int nodeid, + unsigned long start_pfn, unsigned long end_pfn) { int count; struct range *range; @@ -172,9 +173,9 @@ int __init get_free_all_memory_range(struct range **rangep, int nodeid) * at first */ nr_range = add_from_early_node_map(range, count, nr_range, nodeid); -#ifdef CONFIG_X86_32 - subtract_range(range, count, max_low_pfn, -1ULL); -#endif + subtract_range(range, count, 0, start_pfn); + subtract_range(range, count, end_pfn, -1ULL); + memblock_x86_subtract_reserved(range, count); nr_range = clean_sort_range(range, count); @@ -182,6 +183,16 @@ int __init get_free_all_memory_range(struct range **rangep, int nodeid) return nr_range; } +int __init get_free_all_memory_range(struct range **rangep, int nodeid) +{ + unsigned long end_pfn = -1UL; + +#ifdef CONFIG_X86_32 + end_pfn = max_low_pfn; +#endif + return __get_free_all_memory_range(rangep, nodeid, 0, end_pfn); +} + static u64 __init __memblock_x86_memory_in_range(u64 addr, u64 limit, bool get_free) { int i, count; diff --git a/arch/x86/mm/numa_32.c b/arch/x86/mm/numa_32.c index 70ddeb75ba2..84a3e4c9f27 100644 --- a/arch/x86/mm/numa_32.c +++ b/arch/x86/mm/numa_32.c @@ -326,8 +326,6 @@ static __init unsigned long calculate_numa_remap_pages(void) "KVA RAM"); node_remap_start_pfn[nid] = node_kva_final>>PAGE_SHIFT; - remove_active_range(nid, node_remap_start_pfn[nid], - node_remap_start_pfn[nid] + size); } printk(KERN_INFO "Reserving total of %lx pages for numa KVA remap\n", reserve_pages); -- cgit v1.2.3-70-g09d2 From 16c36f743bf8481d0ba40a6de0af11736095d7cf Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Mon, 4 Oct 2010 14:58:04 -0700 Subject: x86, memblock: Remove __memblock_x86_find_in_range_size() Fold it into memblock_x86_find_in_range(), and change bad_addr_size() to check_reserve_memblock(). So whole memblock_x86_find_in_range_size() code is more readable. Signed-off-by: Yinghai Lu LKML-Reference: <4CAA4DEC.4000401@kernel.org> Signed-off-by: H. Peter Anvin --- arch/x86/mm/memblock.c | 39 +++++++++++---------------------------- 1 file changed, 11 insertions(+), 28 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index fd7a0404945..aa1169392b8 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -8,7 +8,7 @@ #include /* Check for already reserved areas */ -static inline bool __init bad_addr_size(u64 *addrp, u64 *sizep, u64 align) +static bool __init check_with_memblock_reserved_size(u64 *addrp, u64 *sizep, u64 align) { struct memblock_region *r; u64 addr = *addrp, last; @@ -30,7 +30,7 @@ again: goto again; } if (last <= (r->base + r->size) && addr >= r->base) { - (*sizep)++; + *sizep = 0; return false; } } @@ -41,29 +41,6 @@ again: return changed; } -static u64 __init __memblock_x86_find_in_range_size(u64 ei_start, u64 ei_last, u64 start, - u64 *sizep, u64 align) -{ - u64 addr, last; - - addr = round_up(ei_start, align); - if (addr < start) - addr = round_up(start, align); - if (addr >= ei_last) - goto out; - *sizep = ei_last - addr; - while (bad_addr_size(&addr, sizep, align) && addr + *sizep <= ei_last) - ; - last = addr + *sizep; - if (last > ei_last) - goto out; - - return addr; - -out: - return MEMBLOCK_ERROR; -} - /* * Find next free range after start, and size is returned in *sizep */ @@ -76,10 +53,16 @@ u64 __init memblock_x86_find_in_range_size(u64 start, u64 *sizep, u64 align) u64 ei_last = ei_start + r->size; u64 addr; - addr = __memblock_x86_find_in_range_size(ei_start, ei_last, start, - sizep, align); + addr = round_up(ei_start, align); + if (addr < start) + addr = round_up(start, align); + if (addr >= ei_last) + continue; + *sizep = ei_last - addr; + while (check_with_memblock_reserved_size(&addr, sizep, align)) + ; - if (addr != MEMBLOCK_ERROR) + if (*sizep) return addr; } -- cgit v1.2.3-70-g09d2 From 161b0275e2311b8bd9609d5f32e2b703cf5d70a8 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 2 Sep 2010 19:35:22 -0700 Subject: x86, mm: Add RESERVE_BRK_ARRAY() helper This is useful when converting static arrays into boot-time brk allocated objects. Signed-off-by: Jeremy Fitzhardinge LKML-Reference: <4C805EEA.1080205@goop.org> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/setup.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/setup.h b/arch/x86/include/asm/setup.h index ef292c792d7..d6763b139a8 100644 --- a/arch/x86/include/asm/setup.h +++ b/arch/x86/include/asm/setup.h @@ -93,6 +93,11 @@ void *extend_brk(size_t size, size_t align); : : "i" (sz)); \ } +/* Helper for reserving space for arrays of things */ +#define RESERVE_BRK_ARRAY(type, name, entries) \ + type *name; \ + RESERVE_BRK(name, sizeof(type) * entries) + #ifdef __i386__ void __init i386_start_kernel(void); -- cgit v1.2.3-70-g09d2 From df9ee29270c11dba7d0fe0b83ce47a4d8e8d2101 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 7 Oct 2010 14:08:55 +0100 Subject: Fix IRQ flag handling naming Fix the IRQ flag handling naming. In linux/irqflags.h under one configuration, it maps: local_irq_enable() -> raw_local_irq_enable() local_irq_disable() -> raw_local_irq_disable() local_irq_save() -> raw_local_irq_save() ... and under the other configuration, it maps: raw_local_irq_enable() -> local_irq_enable() raw_local_irq_disable() -> local_irq_disable() raw_local_irq_save() -> local_irq_save() ... This is quite confusing. There should be one set of names expected of the arch, and this should be wrapped to give another set of names that are expected by users of this facility. Change this to have the arch provide: flags = arch_local_save_flags() flags = arch_local_irq_save() arch_local_irq_restore(flags) arch_local_irq_disable() arch_local_irq_enable() arch_irqs_disabled_flags(flags) arch_irqs_disabled() arch_safe_halt() Then linux/irqflags.h wraps these to provide: raw_local_save_flags(flags) raw_local_irq_save(flags) raw_local_irq_restore(flags) raw_local_irq_disable() raw_local_irq_enable() raw_irqs_disabled_flags(flags) raw_irqs_disabled() raw_safe_halt() with type checking on the flags 'arguments', and then wraps those to provide: local_save_flags(flags) local_irq_save(flags) local_irq_restore(flags) local_irq_disable() local_irq_enable() irqs_disabled_flags(flags) irqs_disabled() safe_halt() with tracing included if enabled. The arch functions can now all be inline functions rather than some of them having to be macros. Signed-off-by: David Howells [X86, FRV, MN10300] Signed-off-by: Chris Metcalf [Tile] Signed-off-by: Michal Simek [Microblaze] Tested-by: Catalin Marinas [ARM] Acked-by: Thomas Gleixner Acked-by: Haavard Skinnemoen [AVR] Acked-by: Tony Luck [IA-64] Acked-by: Hirokazu Takata [M32R] Acked-by: Greg Ungerer [M68K/M68KNOMMU] Acked-by: Ralf Baechle [MIPS] Acked-by: Kyle McMartin [PA-RISC] Acked-by: Paul Mackerras [PowerPC] Acked-by: Martin Schwidefsky [S390] Acked-by: Chen Liqin [Score] Acked-by: Matt Fleming [SH] Acked-by: David S. Miller [Sparc] Acked-by: Chris Zankel [Xtensa] Reviewed-by: Richard Henderson [Alpha] Reviewed-by: Yoshinori Sato [H8300] Cc: starvik@axis.com [CRIS] Cc: jesper.nilsson@axis.com [CRIS] Cc: linux-cris-kernel@axis.com --- arch/alpha/include/asm/irqflags.h | 67 ++++++++++ arch/alpha/include/asm/system.h | 28 ----- arch/arm/include/asm/irqflags.h | 145 +++++++++++++--------- arch/avr32/include/asm/irqflags.h | 29 ++--- arch/blackfin/include/asm/irqflags.h | 12 -- arch/blackfin/kernel/trace.c | 1 + arch/cris/include/arch-v10/arch/irqflags.h | 45 +++++++ arch/cris/include/arch-v10/arch/system.h | 16 --- arch/cris/include/arch-v32/arch/irqflags.h | 46 +++++++ arch/cris/include/arch-v32/arch/system.h | 22 ---- arch/cris/include/asm/irqflags.h | 1 + arch/cris/include/asm/system.h | 1 + arch/frv/include/asm/irqflags.h | 158 +++++++++++++++++++++++ arch/frv/include/asm/system.h | 136 -------------------- arch/h8300/include/asm/irqflags.h | 43 +++++++ arch/h8300/include/asm/system.h | 24 +--- arch/ia64/include/asm/irqflags.h | 94 ++++++++++++++ arch/ia64/include/asm/system.h | 76 ------------ arch/m32r/include/asm/irqflags.h | 104 ++++++++++++++++ arch/m32r/include/asm/system.h | 66 +--------- arch/m68k/include/asm/entry_no.h | 2 +- arch/m68k/include/asm/irqflags.h | 76 ++++++++++++ arch/m68k/include/asm/system_mm.h | 25 +--- arch/m68k/include/asm/system_no.h | 57 +-------- arch/m68knommu/kernel/asm-offsets.c | 2 - arch/m68knommu/platform/coldfire/head.S | 1 + arch/microblaze/include/asm/irqflags.h | 193 +++++++++++++++-------------- arch/mips/include/asm/irqflags.h | 53 ++++---- arch/mips/kernel/smtc.c | 4 +- arch/mn10300/include/asm/irqflags.h | 123 ++++++++++++++++++ arch/mn10300/include/asm/system.h | 109 +--------------- arch/mn10300/kernel/entry.S | 1 + arch/parisc/include/asm/irqflags.h | 46 +++++++ arch/parisc/include/asm/system.h | 19 +-- arch/powerpc/include/asm/hw_irq.h | 113 ++++++++++------- arch/powerpc/include/asm/irqflags.h | 2 +- arch/powerpc/kernel/exceptions-64s.S | 4 +- arch/powerpc/kernel/irq.c | 4 +- arch/s390/include/asm/irqflags.h | 51 ++++---- arch/s390/include/asm/system.h | 2 +- arch/s390/kernel/mem_detect.c | 4 +- arch/s390/mm/init.c | 3 +- arch/s390/mm/maccess.c | 4 +- arch/score/include/asm/irqflags.h | 187 +++++++++++++++------------- arch/sh/include/asm/irqflags.h | 4 +- arch/sh/kernel/irq_32.c | 12 +- arch/sparc/include/asm/irqflags_32.h | 35 +++--- arch/sparc/include/asm/irqflags_64.h | 29 ++--- arch/sparc/kernel/irq_32.c | 13 +- arch/sparc/prom/p1275.c | 2 +- arch/tile/include/asm/irqflags.h | 36 +++--- arch/x86/include/asm/irqflags.h | 32 ++--- arch/x86/include/asm/paravirt.h | 16 +-- arch/x86/xen/spinlock.c | 2 +- arch/xtensa/include/asm/irqflags.h | 58 +++++++++ arch/xtensa/include/asm/system.h | 33 +---- drivers/s390/char/sclp.c | 2 +- include/asm-generic/atomic.h | 5 +- include/asm-generic/cmpxchg-local.h | 1 + include/asm-generic/hardirq.h | 1 - include/asm-generic/irqflags.h | 52 ++++---- include/linux/irqflags.h | 107 +++++++++------- include/linux/spinlock.h | 1 + 63 files changed, 1482 insertions(+), 1158 deletions(-) create mode 100644 arch/alpha/include/asm/irqflags.h create mode 100644 arch/cris/include/arch-v10/arch/irqflags.h create mode 100644 arch/cris/include/arch-v32/arch/irqflags.h create mode 100644 arch/cris/include/asm/irqflags.h create mode 100644 arch/frv/include/asm/irqflags.h create mode 100644 arch/h8300/include/asm/irqflags.h create mode 100644 arch/ia64/include/asm/irqflags.h create mode 100644 arch/m32r/include/asm/irqflags.h create mode 100644 arch/m68k/include/asm/irqflags.h create mode 100644 arch/mn10300/include/asm/irqflags.h create mode 100644 arch/parisc/include/asm/irqflags.h create mode 100644 arch/xtensa/include/asm/irqflags.h (limited to 'arch/x86') diff --git a/arch/alpha/include/asm/irqflags.h b/arch/alpha/include/asm/irqflags.h new file mode 100644 index 00000000000..299bbc7e9d7 --- /dev/null +++ b/arch/alpha/include/asm/irqflags.h @@ -0,0 +1,67 @@ +#ifndef __ALPHA_IRQFLAGS_H +#define __ALPHA_IRQFLAGS_H + +#include + +#define IPL_MIN 0 +#define IPL_SW0 1 +#define IPL_SW1 2 +#define IPL_DEV0 3 +#define IPL_DEV1 4 +#define IPL_TIMER 5 +#define IPL_PERF 6 +#define IPL_POWERFAIL 6 +#define IPL_MCHECK 7 +#define IPL_MAX 7 + +#ifdef CONFIG_ALPHA_BROKEN_IRQ_MASK +#undef IPL_MIN +#define IPL_MIN __min_ipl +extern int __min_ipl; +#endif + +#define getipl() (rdps() & 7) +#define setipl(ipl) ((void) swpipl(ipl)) + +static inline unsigned long arch_local_save_flags(void) +{ + return rdps(); +} + +static inline void arch_local_irq_disable(void) +{ + setipl(IPL_MAX); + barrier(); +} + +static inline unsigned long arch_local_irq_save(void) +{ + unsigned long flags = swpipl(IPL_MAX); + barrier(); + return flags; +} + +static inline void arch_local_irq_enable(void) +{ + barrier(); + setipl(IPL_MIN); +} + +static inline void arch_local_irq_restore(unsigned long flags) +{ + barrier(); + setipl(flags); + barrier(); +} + +static inline bool arch_irqs_disabled_flags(unsigned long flags) +{ + return flags == IPL_MAX; +} + +static inline bool arch_irqs_disabled(void) +{ + return arch_irqs_disabled_flags(getipl()); +} + +#endif /* __ALPHA_IRQFLAGS_H */ diff --git a/arch/alpha/include/asm/system.h b/arch/alpha/include/asm/system.h index 5aa40cca4f2..9f78e693463 100644 --- a/arch/alpha/include/asm/system.h +++ b/arch/alpha/include/asm/system.h @@ -259,34 +259,6 @@ __CALL_PAL_RW2(wrperfmon, unsigned long, unsigned long, unsigned long); __CALL_PAL_W1(wrusp, unsigned long); __CALL_PAL_W1(wrvptptr, unsigned long); -#define IPL_MIN 0 -#define IPL_SW0 1 -#define IPL_SW1 2 -#define IPL_DEV0 3 -#define IPL_DEV1 4 -#define IPL_TIMER 5 -#define IPL_PERF 6 -#define IPL_POWERFAIL 6 -#define IPL_MCHECK 7 -#define IPL_MAX 7 - -#ifdef CONFIG_ALPHA_BROKEN_IRQ_MASK -#undef IPL_MIN -#define IPL_MIN __min_ipl -extern int __min_ipl; -#endif - -#define getipl() (rdps() & 7) -#define setipl(ipl) ((void) swpipl(ipl)) - -#define local_irq_disable() do { setipl(IPL_MAX); barrier(); } while(0) -#define local_irq_enable() do { barrier(); setipl(IPL_MIN); } while(0) -#define local_save_flags(flags) ((flags) = rdps()) -#define local_irq_save(flags) do { (flags) = swpipl(IPL_MAX); barrier(); } while(0) -#define local_irq_restore(flags) do { barrier(); setipl(flags); barrier(); } while(0) - -#define irqs_disabled() (getipl() == IPL_MAX) - /* * TB routines.. */ diff --git a/arch/arm/include/asm/irqflags.h b/arch/arm/include/asm/irqflags.h index 6d09974e664..1e6cca55c75 100644 --- a/arch/arm/include/asm/irqflags.h +++ b/arch/arm/include/asm/irqflags.h @@ -10,66 +10,85 @@ */ #if __LINUX_ARM_ARCH__ >= 6 -#define raw_local_irq_save(x) \ - ({ \ - __asm__ __volatile__( \ - "mrs %0, cpsr @ local_irq_save\n" \ - "cpsid i" \ - : "=r" (x) : : "memory", "cc"); \ - }) +static inline unsigned long arch_local_irq_save(void) +{ + unsigned long flags; + + asm volatile( + " mrs %0, cpsr @ arch_local_irq_save\n" + " cpsid i" + : "=r" (flags) : : "memory", "cc"); + return flags; +} + +static inline void arch_local_irq_enable(void) +{ + asm volatile( + " cpsie i @ arch_local_irq_enable" + : + : + : "memory", "cc"); +} + +static inline void arch_local_irq_disable(void) +{ + asm volatile( + " cpsid i @ arch_local_irq_disable" + : + : + : "memory", "cc"); +} -#define raw_local_irq_enable() __asm__("cpsie i @ __sti" : : : "memory", "cc") -#define raw_local_irq_disable() __asm__("cpsid i @ __cli" : : : "memory", "cc") #define local_fiq_enable() __asm__("cpsie f @ __stf" : : : "memory", "cc") #define local_fiq_disable() __asm__("cpsid f @ __clf" : : : "memory", "cc") - #else /* * Save the current interrupt enable state & disable IRQs */ -#define raw_local_irq_save(x) \ - ({ \ - unsigned long temp; \ - (void) (&temp == &x); \ - __asm__ __volatile__( \ - "mrs %0, cpsr @ local_irq_save\n" \ -" orr %1, %0, #128\n" \ -" msr cpsr_c, %1" \ - : "=r" (x), "=r" (temp) \ - : \ - : "memory", "cc"); \ - }) - +static inline unsigned long arch_local_irq_save(void) +{ + unsigned long flags, temp; + + asm volatile( + " mrs %0, cpsr @ arch_local_irq_save\n" + " orr %1, %0, #128\n" + " msr cpsr_c, %1" + : "=r" (flags), "=r" (temp) + : + : "memory", "cc"); + return flags; +} + /* * Enable IRQs */ -#define raw_local_irq_enable() \ - ({ \ - unsigned long temp; \ - __asm__ __volatile__( \ - "mrs %0, cpsr @ local_irq_enable\n" \ -" bic %0, %0, #128\n" \ -" msr cpsr_c, %0" \ - : "=r" (temp) \ - : \ - : "memory", "cc"); \ - }) +static inline void arch_local_irq_enable(void) +{ + unsigned long temp; + asm volatile( + " mrs %0, cpsr @ arch_local_irq_enable\n" + " bic %0, %0, #128\n" + " msr cpsr_c, %0" + : "=r" (temp) + : + : "memory", "cc"); +} /* * Disable IRQs */ -#define raw_local_irq_disable() \ - ({ \ - unsigned long temp; \ - __asm__ __volatile__( \ - "mrs %0, cpsr @ local_irq_disable\n" \ -" orr %0, %0, #128\n" \ -" msr cpsr_c, %0" \ - : "=r" (temp) \ - : \ - : "memory", "cc"); \ - }) +static inline void arch_local_irq_disable(void) +{ + unsigned long temp; + asm volatile( + " mrs %0, cpsr @ arch_local_irq_disable\n" + " orr %0, %0, #128\n" + " msr cpsr_c, %0" + : "=r" (temp) + : + : "memory", "cc"); +} /* * Enable FIQs @@ -106,27 +125,31 @@ /* * Save the current interrupt enable state. */ -#define raw_local_save_flags(x) \ - ({ \ - __asm__ __volatile__( \ - "mrs %0, cpsr @ local_save_flags" \ - : "=r" (x) : : "memory", "cc"); \ - }) +static inline unsigned long arch_local_save_flags(void) +{ + unsigned long flags; + asm volatile( + " mrs %0, cpsr @ local_save_flags" + : "=r" (flags) : : "memory", "cc"); + return flags; +} /* * restore saved IRQ & FIQ state */ -#define raw_local_irq_restore(x) \ - __asm__ __volatile__( \ - "msr cpsr_c, %0 @ local_irq_restore\n" \ - : \ - : "r" (x) \ - : "memory", "cc") +static inline void arch_local_irq_restore(unsigned long flags) +{ + asm volatile( + " msr cpsr_c, %0 @ local_irq_restore" + : + : "r" (flags) + : "memory", "cc"); +} -#define raw_irqs_disabled_flags(flags) \ -({ \ - (int)((flags) & PSR_I_BIT); \ -}) +static inline int arch_irqs_disabled_flags(unsigned long flags) +{ + return flags & PSR_I_BIT; +} #endif #endif diff --git a/arch/avr32/include/asm/irqflags.h b/arch/avr32/include/asm/irqflags.h index 93570daac38..006e9487372 100644 --- a/arch/avr32/include/asm/irqflags.h +++ b/arch/avr32/include/asm/irqflags.h @@ -8,16 +8,14 @@ #ifndef __ASM_AVR32_IRQFLAGS_H #define __ASM_AVR32_IRQFLAGS_H +#include #include -static inline unsigned long __raw_local_save_flags(void) +static inline unsigned long arch_local_save_flags(void) { return sysreg_read(SR); } -#define raw_local_save_flags(x) \ - do { (x) = __raw_local_save_flags(); } while (0) - /* * This will restore ALL status register flags, not only the interrupt * mask flag. @@ -25,44 +23,39 @@ static inline unsigned long __raw_local_save_flags(void) * The empty asm statement informs the compiler of this fact while * also serving as a barrier. */ -static inline void raw_local_irq_restore(unsigned long flags) +static inline void arch_local_irq_restore(unsigned long flags) { sysreg_write(SR, flags); asm volatile("" : : : "memory", "cc"); } -static inline void raw_local_irq_disable(void) +static inline void arch_local_irq_disable(void) { asm volatile("ssrf %0" : : "n"(SYSREG_GM_OFFSET) : "memory"); } -static inline void raw_local_irq_enable(void) +static inline void arch_local_irq_enable(void) { asm volatile("csrf %0" : : "n"(SYSREG_GM_OFFSET) : "memory"); } -static inline int raw_irqs_disabled_flags(unsigned long flags) +static inline bool arch_irqs_disabled_flags(unsigned long flags) { return (flags & SYSREG_BIT(GM)) != 0; } -static inline int raw_irqs_disabled(void) +static inline bool arch_irqs_disabled(void) { - unsigned long flags = __raw_local_save_flags(); - - return raw_irqs_disabled_flags(flags); + return arch_irqs_disabled_flags(arch_local_save_flags()); } -static inline unsigned long __raw_local_irq_save(void) +static inline unsigned long arch_local_irq_save(void) { - unsigned long flags = __raw_local_save_flags(); + unsigned long flags = arch_local_save_flags(); - raw_local_irq_disable(); + arch_local_irq_disable(); return flags; } -#define raw_local_irq_save(flags) \ - do { (flags) = __raw_local_irq_save(); } while (0) - #endif /* __ASM_AVR32_IRQFLAGS_H */ diff --git a/arch/blackfin/include/asm/irqflags.h b/arch/blackfin/include/asm/irqflags.h index 994d7679101..41c4d70544e 100644 --- a/arch/blackfin/include/asm/irqflags.h +++ b/arch/blackfin/include/asm/irqflags.h @@ -218,16 +218,4 @@ static inline void hard_local_irq_restore(unsigned long flags) #endif /* !CONFIG_IPIPE */ - -/* - * Raw interface to linux/irqflags.h. - */ -#define raw_local_save_flags(flags) do { (flags) = arch_local_save_flags(); } while (0) -#define raw_local_irq_save(flags) do { (flags) = arch_local_irq_save(); } while (0) -#define raw_local_irq_restore(flags) arch_local_irq_restore(flags) -#define raw_local_irq_enable() arch_local_irq_enable() -#define raw_local_irq_disable() arch_local_irq_disable() -#define raw_irqs_disabled_flags(flags) arch_irqs_disabled_flags(flags) -#define raw_irqs_disabled() arch_irqs_disabled() - #endif diff --git a/arch/blackfin/kernel/trace.c b/arch/blackfin/kernel/trace.c index 59fcdf6b013..05b550891ce 100644 --- a/arch/blackfin/kernel/trace.c +++ b/arch/blackfin/kernel/trace.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include #include diff --git a/arch/cris/include/arch-v10/arch/irqflags.h b/arch/cris/include/arch-v10/arch/irqflags.h new file mode 100644 index 00000000000..75ef1899124 --- /dev/null +++ b/arch/cris/include/arch-v10/arch/irqflags.h @@ -0,0 +1,45 @@ +#ifndef __ASM_CRIS_ARCH_IRQFLAGS_H +#define __ASM_CRIS_ARCH_IRQFLAGS_H + +#include + +static inline unsigned long arch_local_save_flags(void) +{ + unsigned long flags; + asm volatile("move $ccr,%0" : "=rm" (flags) : : "memory"); + return flags; +} + +static inline void arch_local_irq_disable(void) +{ + asm volatile("di" : : : "memory"); +} + +static inline void arch_local_irq_enable(void) +{ + asm volatile("ei" : : : "memory"); +} + +static inline unsigned long arch_local_irq_save(void) +{ + unsigned long flags = arch_local_save_flags(); + arch_local_irq_disable(); + return flags; +} + +static inline void arch_local_irq_restore(unsigned long flags) +{ + asm volatile("move %0,$ccr" : : "rm" (flags) : "memory"); +} + +static inline bool arch_irqs_disabled_flags(unsigned long flags) +{ + return !(flags & (1 << 5)); +} + +static inline bool arch_irqs_disabled(void) +{ + return arch_irqs_disabled_flags(arch_local_save_flags()); +} + +#endif /* __ASM_CRIS_ARCH_IRQFLAGS_H */ diff --git a/arch/cris/include/arch-v10/arch/system.h b/arch/cris/include/arch-v10/arch/system.h index 4a9cd36c9e1..935fde34aa1 100644 --- a/arch/cris/include/arch-v10/arch/system.h +++ b/arch/cris/include/arch-v10/arch/system.h @@ -44,20 +44,4 @@ static inline unsigned long _get_base(char * addr) struct __xchg_dummy { unsigned long a[100]; }; #define __xg(x) ((struct __xchg_dummy *)(x)) -/* interrupt control.. */ -#define local_save_flags(x) __asm__ __volatile__ ("move $ccr,%0" : "=rm" (x) : : "memory"); -#define local_irq_restore(x) __asm__ __volatile__ ("move %0,$ccr" : : "rm" (x) : "memory"); -#define local_irq_disable() __asm__ __volatile__ ( "di" : : :"memory"); -#define local_irq_enable() __asm__ __volatile__ ( "ei" : : :"memory"); - -#define irqs_disabled() \ -({ \ - unsigned long flags; \ - local_save_flags(flags); \ - !(flags & (1<<5)); \ -}) - -/* For spinlocks etc */ -#define local_irq_save(x) __asm__ __volatile__ ("move $ccr,%0\n\tdi" : "=rm" (x) : : "memory"); - #endif diff --git a/arch/cris/include/arch-v32/arch/irqflags.h b/arch/cris/include/arch-v32/arch/irqflags.h new file mode 100644 index 00000000000..041851f8ec6 --- /dev/null +++ b/arch/cris/include/arch-v32/arch/irqflags.h @@ -0,0 +1,46 @@ +#ifndef __ASM_CRIS_ARCH_IRQFLAGS_H +#define __ASM_CRIS_ARCH_IRQFLAGS_H + +#include +#include + +static inline unsigned long arch_local_save_flags(void) +{ + unsigned long flags; + asm volatile("move $ccs,%0" : "=rm" (flags) : : "memory"); + return flags; +} + +static inline void arch_local_irq_disable(void) +{ + asm volatile("di" : : : "memory"); +} + +static inline void arch_local_irq_enable(void) +{ + asm volatile("ei" : : : "memory"); +} + +static inline unsigned long arch_local_irq_save(void) +{ + unsigned long flags = arch_local_save_flags(); + arch_local_irq_disable(); + return flags; +} + +static inline void arch_local_irq_restore(unsigned long flags) +{ + asm volatile("move %0,$ccs" : : "rm" (flags) : "memory"); +} + +static inline bool arch_irqs_disabled_flags(unsigned long flags) +{ + return !(flags & (1 << I_CCS_BITNR)); +} + +static inline bool arch_irqs_disabled(void) +{ + return arch_irqs_disabled_flags(arch_local_save_flags()); +} + +#endif /* __ASM_CRIS_ARCH_IRQFLAGS_H */ diff --git a/arch/cris/include/arch-v32/arch/system.h b/arch/cris/include/arch-v32/arch/system.h index 6ca90f1f110..76cea99eaa6 100644 --- a/arch/cris/include/arch-v32/arch/system.h +++ b/arch/cris/include/arch-v32/arch/system.h @@ -44,26 +44,4 @@ static inline unsigned long rdsp(void) struct __xchg_dummy { unsigned long a[100]; }; #define __xg(x) ((struct __xchg_dummy *)(x)) -/* Used for interrupt control. */ -#define local_save_flags(x) \ - __asm__ __volatile__ ("move $ccs, %0" : "=rm" (x) : : "memory"); - -#define local_irq_restore(x) \ - __asm__ __volatile__ ("move %0, $ccs" : : "rm" (x) : "memory"); - -#define local_irq_disable() __asm__ __volatile__ ("di" : : : "memory"); -#define local_irq_enable() __asm__ __volatile__ ("ei" : : : "memory"); - -#define irqs_disabled() \ -({ \ - unsigned long flags; \ - \ - local_save_flags(flags);\ - !(flags & (1 << I_CCS_BITNR)); \ -}) - -/* Used for spinlocks, etc. */ -#define local_irq_save(x) \ - __asm__ __volatile__ ("move $ccs, %0\n\tdi" : "=rm" (x) : : "memory"); - #endif /* _ASM_CRIS_ARCH_SYSTEM_H */ diff --git a/arch/cris/include/asm/irqflags.h b/arch/cris/include/asm/irqflags.h new file mode 100644 index 00000000000..943ba5ca6d2 --- /dev/null +++ b/arch/cris/include/asm/irqflags.h @@ -0,0 +1 @@ +#include diff --git a/arch/cris/include/asm/system.h b/arch/cris/include/asm/system.h index 8657b084a92..ea10592f7d7 100644 --- a/arch/cris/include/asm/system.h +++ b/arch/cris/include/asm/system.h @@ -1,6 +1,7 @@ #ifndef __ASM_CRIS_SYSTEM_H #define __ASM_CRIS_SYSTEM_H +#include #include /* the switch_to macro calls resume, an asm function in entry.S which does the actual diff --git a/arch/frv/include/asm/irqflags.h b/arch/frv/include/asm/irqflags.h new file mode 100644 index 00000000000..82f0b5363f4 --- /dev/null +++ b/arch/frv/include/asm/irqflags.h @@ -0,0 +1,158 @@ +/* FR-V interrupt handling + * + * Copyright (C) 2010 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public Licence + * as published by the Free Software Foundation; either version + * 2 of the Licence, or (at your option) any later version. + */ + +#ifndef _ASM_IRQFLAGS_H +#define _ASM_IRQFLAGS_H + +/* + * interrupt flag manipulation + * - use virtual interrupt management since touching the PSR is slow + * - ICC2.Z: T if interrupts virtually disabled + * - ICC2.C: F if interrupts really disabled + * - if Z==1 upon interrupt: + * - C is set to 0 + * - interrupts are really disabled + * - entry.S returns immediately + * - uses TIHI (TRAP if Z==0 && C==0) #2 to really reenable interrupts + * - if taken, the trap: + * - sets ICC2.C + * - enables interrupts + */ +static inline void arch_local_irq_disable(void) +{ + /* set Z flag, but don't change the C flag */ + asm volatile(" andcc gr0,gr0,gr0,icc2 \n" + : + : + : "memory", "icc2" + ); +} + +static inline void arch_local_irq_enable(void) +{ + /* clear Z flag and then test the C flag */ + asm volatile(" oricc gr0,#1,gr0,icc2 \n" + " tihi icc2,gr0,#2 \n" + : + : + : "memory", "icc2" + ); +} + +static inline unsigned long arch_local_save_flags(void) +{ + unsigned long flags; + + asm volatile("movsg ccr,%0" + : "=r"(flags) + : + : "memory"); + + /* shift ICC2.Z to bit 0 */ + flags >>= 26; + + /* make flags 1 if interrupts disabled, 0 otherwise */ + return flags & 1UL; + +} + +static inline unsigned long arch_local_irq_save(void) +{ + unsigned long flags = arch_local_save_flags(); + arch_local_irq_disable(); + return flags; +} + +static inline void arch_local_irq_restore(unsigned long flags) +{ + /* load the Z flag by turning 1 if disabled into 0 if disabled + * and thus setting the Z flag but not the C flag */ + asm volatile(" xoricc %0,#1,gr0,icc2 \n" + /* then trap if Z=0 and C=0 */ + " tihi icc2,gr0,#2 \n" + : + : "r"(flags) + : "memory", "icc2" + ); + +} + +static inline bool arch_irqs_disabled_flags(unsigned long flags) +{ + return flags; +} + +static inline bool arch_irqs_disabled(void) +{ + return arch_irqs_disabled_flags(arch_local_save_flags()); +} + +/* + * real interrupt flag manipulation + */ +#define __arch_local_irq_disable() \ +do { \ + unsigned long psr; \ + asm volatile(" movsg psr,%0 \n" \ + " andi %0,%2,%0 \n" \ + " ori %0,%1,%0 \n" \ + " movgs %0,psr \n" \ + : "=r"(psr) \ + : "i" (PSR_PIL_14), "i" (~PSR_PIL) \ + : "memory"); \ +} while (0) + +#define __arch_local_irq_enable() \ +do { \ + unsigned long psr; \ + asm volatile(" movsg psr,%0 \n" \ + " andi %0,%1,%0 \n" \ + " movgs %0,psr \n" \ + : "=r"(psr) \ + : "i" (~PSR_PIL) \ + : "memory"); \ +} while (0) + +#define __arch_local_save_flags(flags) \ +do { \ + typecheck(unsigned long, flags); \ + asm("movsg psr,%0" \ + : "=r"(flags) \ + : \ + : "memory"); \ +} while (0) + +#define __arch_local_irq_save(flags) \ +do { \ + unsigned long npsr; \ + typecheck(unsigned long, flags); \ + asm volatile(" movsg psr,%0 \n" \ + " andi %0,%3,%1 \n" \ + " ori %1,%2,%1 \n" \ + " movgs %1,psr \n" \ + : "=r"(flags), "=r"(npsr) \ + : "i" (PSR_PIL_14), "i" (~PSR_PIL) \ + : "memory"); \ +} while (0) + +#define __arch_local_irq_restore(flags) \ +do { \ + typecheck(unsigned long, flags); \ + asm volatile(" movgs %0,psr \n" \ + : \ + : "r" (flags) \ + : "memory"); \ +} while (0) + +#define __arch_irqs_disabled() \ + ((__get_PSR() & PSR_PIL) >= PSR_PIL_14) + +#endif /* _ASM_IRQFLAGS_H */ diff --git a/arch/frv/include/asm/system.h b/arch/frv/include/asm/system.h index efd22d9077a..0a6d8d9ca45 100644 --- a/arch/frv/include/asm/system.h +++ b/arch/frv/include/asm/system.h @@ -36,142 +36,6 @@ do { \ mb(); \ } while(0) -/* - * interrupt flag manipulation - * - use virtual interrupt management since touching the PSR is slow - * - ICC2.Z: T if interrupts virtually disabled - * - ICC2.C: F if interrupts really disabled - * - if Z==1 upon interrupt: - * - C is set to 0 - * - interrupts are really disabled - * - entry.S returns immediately - * - uses TIHI (TRAP if Z==0 && C==0) #2 to really reenable interrupts - * - if taken, the trap: - * - sets ICC2.C - * - enables interrupts - */ -#define local_irq_disable() \ -do { \ - /* set Z flag, but don't change the C flag */ \ - asm volatile(" andcc gr0,gr0,gr0,icc2 \n" \ - : \ - : \ - : "memory", "icc2" \ - ); \ -} while(0) - -#define local_irq_enable() \ -do { \ - /* clear Z flag and then test the C flag */ \ - asm volatile(" oricc gr0,#1,gr0,icc2 \n" \ - " tihi icc2,gr0,#2 \n" \ - : \ - : \ - : "memory", "icc2" \ - ); \ -} while(0) - -#define local_save_flags(flags) \ -do { \ - typecheck(unsigned long, flags); \ - asm volatile("movsg ccr,%0" \ - : "=r"(flags) \ - : \ - : "memory"); \ - \ - /* shift ICC2.Z to bit 0 */ \ - flags >>= 26; \ - \ - /* make flags 1 if interrupts disabled, 0 otherwise */ \ - flags &= 1UL; \ -} while(0) - -#define irqs_disabled() \ - ({unsigned long flags; local_save_flags(flags); !!flags; }) - -#define local_irq_save(flags) \ -do { \ - typecheck(unsigned long, flags); \ - local_save_flags(flags); \ - local_irq_disable(); \ -} while(0) - -#define local_irq_restore(flags) \ -do { \ - typecheck(unsigned long, flags); \ - \ - /* load the Z flag by turning 1 if disabled into 0 if disabled \ - * and thus setting the Z flag but not the C flag */ \ - asm volatile(" xoricc %0,#1,gr0,icc2 \n" \ - /* then test Z=0 and C=0 */ \ - " tihi icc2,gr0,#2 \n" \ - : \ - : "r"(flags) \ - : "memory", "icc2" \ - ); \ - \ -} while(0) - -/* - * real interrupt flag manipulation - */ -#define __local_irq_disable() \ -do { \ - unsigned long psr; \ - asm volatile(" movsg psr,%0 \n" \ - " andi %0,%2,%0 \n" \ - " ori %0,%1,%0 \n" \ - " movgs %0,psr \n" \ - : "=r"(psr) \ - : "i" (PSR_PIL_14), "i" (~PSR_PIL) \ - : "memory"); \ -} while(0) - -#define __local_irq_enable() \ -do { \ - unsigned long psr; \ - asm volatile(" movsg psr,%0 \n" \ - " andi %0,%1,%0 \n" \ - " movgs %0,psr \n" \ - : "=r"(psr) \ - : "i" (~PSR_PIL) \ - : "memory"); \ -} while(0) - -#define __local_save_flags(flags) \ -do { \ - typecheck(unsigned long, flags); \ - asm("movsg psr,%0" \ - : "=r"(flags) \ - : \ - : "memory"); \ -} while(0) - -#define __local_irq_save(flags) \ -do { \ - unsigned long npsr; \ - typecheck(unsigned long, flags); \ - asm volatile(" movsg psr,%0 \n" \ - " andi %0,%3,%1 \n" \ - " ori %1,%2,%1 \n" \ - " movgs %1,psr \n" \ - : "=r"(flags), "=r"(npsr) \ - : "i" (PSR_PIL_14), "i" (~PSR_PIL) \ - : "memory"); \ -} while(0) - -#define __local_irq_restore(flags) \ -do { \ - typecheck(unsigned long, flags); \ - asm volatile(" movgs %0,psr \n" \ - : \ - : "r" (flags) \ - : "memory"); \ -} while(0) - -#define __irqs_disabled() \ - ((__get_PSR() & PSR_PIL) >= PSR_PIL_14) - /* * Force strict CPU ordering. */ diff --git a/arch/h8300/include/asm/irqflags.h b/arch/h8300/include/asm/irqflags.h new file mode 100644 index 00000000000..9617cd57aeb --- /dev/null +++ b/arch/h8300/include/asm/irqflags.h @@ -0,0 +1,43 @@ +#ifndef _H8300_IRQFLAGS_H +#define _H8300_IRQFLAGS_H + +static inline unsigned long arch_local_save_flags(void) +{ + unsigned long flags; + asm volatile ("stc ccr,%w0" : "=r" (flags)); + return flags; +} + +static inline void arch_local_irq_disable(void) +{ + asm volatile ("orc #0x80,ccr" : : : "memory"); +} + +static inline void arch_local_irq_enable(void) +{ + asm volatile ("andc #0x7f,ccr" : : : "memory"); +} + +static inline unsigned long arch_local_irq_save(void) +{ + unsigned long flags = arch_local_save_flags(); + arch_local_irq_disable(); + return flags; +} + +static inline void arch_local_irq_restore(unsigned long flags) +{ + asm volatile ("ldc %w0,ccr" : : "r" (flags) : "memory"); +} + +static inline bool arch_irqs_disabled_flags(unsigned long flags) +{ + return (flags & 0x80) == 0x80; +} + +static inline bool arch_irqs_disabled(void) +{ + return arch_irqs_disabled_flags(arch_local_save_flags()); +} + +#endif /* _H8300_IRQFLAGS_H */ diff --git a/arch/h8300/include/asm/system.h b/arch/h8300/include/asm/system.h index 16bf1560ff6..2c2382e50d9 100644 --- a/arch/h8300/include/asm/system.h +++ b/arch/h8300/include/asm/system.h @@ -2,6 +2,7 @@ #define _H8300_SYSTEM_H #include +#include struct pt_regs; @@ -51,31 +52,8 @@ asmlinkage void resume(void); (last) = _last; \ } -#define __sti() asm volatile ("andc #0x7f,ccr") -#define __cli() asm volatile ("orc #0x80,ccr") - -#define __save_flags(x) \ - asm volatile ("stc ccr,%w0":"=r" (x)) - -#define __restore_flags(x) \ - asm volatile ("ldc %w0,ccr": :"r" (x)) - -#define irqs_disabled() \ -({ \ - unsigned char flags; \ - __save_flags(flags); \ - ((flags & 0x80) == 0x80); \ -}) - #define iret() __asm__ __volatile__ ("rte": : :"memory", "sp", "cc") -/* For spinlocks etc */ -#define local_irq_disable() __cli() -#define local_irq_enable() __sti() -#define local_irq_save(x) ({ __save_flags(x); local_irq_disable(); }) -#define local_irq_restore(x) __restore_flags(x) -#define local_save_flags(x) __save_flags(x) - /* * Force strict CPU ordering. * Not really required on H8... diff --git a/arch/ia64/include/asm/irqflags.h b/arch/ia64/include/asm/irqflags.h new file mode 100644 index 00000000000..f82d6be2ecd --- /dev/null +++ b/arch/ia64/include/asm/irqflags.h @@ -0,0 +1,94 @@ +/* + * IRQ flags defines. + * + * Copyright (C) 1998-2003 Hewlett-Packard Co + * David Mosberger-Tang + * Copyright (C) 1999 Asit Mallick + * Copyright (C) 1999 Don Dugger + */ + +#ifndef _ASM_IA64_IRQFLAGS_H +#define _ASM_IA64_IRQFLAGS_H + +#ifdef CONFIG_IA64_DEBUG_IRQ +extern unsigned long last_cli_ip; +static inline void arch_maybe_save_ip(unsigned long flags) +{ + if (flags & IA64_PSR_I) + last_cli_ip = ia64_getreg(_IA64_REG_IP); +} +#else +#define arch_maybe_save_ip(flags) do {} while (0) +#endif + +/* + * - clearing psr.i is implicitly serialized (visible by next insn) + * - setting psr.i requires data serialization + * - we need a stop-bit before reading PSR because we sometimes + * write a floating-point register right before reading the PSR + * and that writes to PSR.mfl + */ + +static inline unsigned long arch_local_save_flags(void) +{ + ia64_stop(); +#ifdef CONFIG_PARAVIRT + return ia64_get_psr_i(); +#else + return ia64_getreg(_IA64_REG_PSR); +#endif +} + +static inline unsigned long arch_local_irq_save(void) +{ + unsigned long flags = arch_local_save_flags(); + + ia64_stop(); + ia64_rsm(IA64_PSR_I); + arch_maybe_save_ip(flags); + return flags; +} + +static inline void arch_local_irq_disable(void) +{ +#ifdef CONFIG_IA64_DEBUG_IRQ + arch_local_irq_save(); +#else + ia64_stop(); + ia64_rsm(IA64_PSR_I); +#endif +} + +static inline void arch_local_irq_enable(void) +{ + ia64_stop(); + ia64_ssm(IA64_PSR_I); + ia64_srlz_d(); +} + +static inline void arch_local_irq_restore(unsigned long flags) +{ +#ifdef CONFIG_IA64_DEBUG_IRQ + unsigned long old_psr = arch_local_save_flags(); +#endif + ia64_intrin_local_irq_restore(flags & IA64_PSR_I); + arch_maybe_save_ip(old_psr & ~flags); +} + +static inline bool arch_irqs_disabled_flags(unsigned long flags) +{ + return (flags & IA64_PSR_I) == 0; +} + +static inline bool arch_irqs_disabled(void) +{ + return arch_irqs_disabled_flags(arch_local_save_flags()); +} + +static inline void arch_safe_halt(void) +{ + ia64_pal_halt_light(); /* PAL_HALT_LIGHT */ +} + + +#endif /* _ASM_IA64_IRQFLAGS_H */ diff --git a/arch/ia64/include/asm/system.h b/arch/ia64/include/asm/system.h index 9f342a574ce..2feb7f64c03 100644 --- a/arch/ia64/include/asm/system.h +++ b/arch/ia64/include/asm/system.h @@ -107,87 +107,11 @@ extern struct ia64_boot_param { */ #define set_mb(var, value) do { (var) = (value); mb(); } while (0) -#define safe_halt() ia64_pal_halt_light() /* PAL_HALT_LIGHT */ - /* * The group barrier in front of the rsm & ssm are necessary to ensure * that none of the previous instructions in the same group are * affected by the rsm/ssm. */ -/* For spinlocks etc */ - -/* - * - clearing psr.i is implicitly serialized (visible by next insn) - * - setting psr.i requires data serialization - * - we need a stop-bit before reading PSR because we sometimes - * write a floating-point register right before reading the PSR - * and that writes to PSR.mfl - */ -#ifdef CONFIG_PARAVIRT -#define __local_save_flags() ia64_get_psr_i() -#else -#define __local_save_flags() ia64_getreg(_IA64_REG_PSR) -#endif - -#define __local_irq_save(x) \ -do { \ - ia64_stop(); \ - (x) = __local_save_flags(); \ - ia64_stop(); \ - ia64_rsm(IA64_PSR_I); \ -} while (0) - -#define __local_irq_disable() \ -do { \ - ia64_stop(); \ - ia64_rsm(IA64_PSR_I); \ -} while (0) - -#define __local_irq_restore(x) ia64_intrin_local_irq_restore((x) & IA64_PSR_I) - -#ifdef CONFIG_IA64_DEBUG_IRQ - - extern unsigned long last_cli_ip; - -# define __save_ip() last_cli_ip = ia64_getreg(_IA64_REG_IP) - -# define local_irq_save(x) \ -do { \ - unsigned long __psr; \ - \ - __local_irq_save(__psr); \ - if (__psr & IA64_PSR_I) \ - __save_ip(); \ - (x) = __psr; \ -} while (0) - -# define local_irq_disable() do { unsigned long __x; local_irq_save(__x); } while (0) - -# define local_irq_restore(x) \ -do { \ - unsigned long __old_psr, __psr = (x); \ - \ - local_save_flags(__old_psr); \ - __local_irq_restore(__psr); \ - if ((__old_psr & IA64_PSR_I) && !(__psr & IA64_PSR_I)) \ - __save_ip(); \ -} while (0) - -#else /* !CONFIG_IA64_DEBUG_IRQ */ -# define local_irq_save(x) __local_irq_save(x) -# define local_irq_disable() __local_irq_disable() -# define local_irq_restore(x) __local_irq_restore(x) -#endif /* !CONFIG_IA64_DEBUG_IRQ */ - -#define local_irq_enable() ({ ia64_stop(); ia64_ssm(IA64_PSR_I); ia64_srlz_d(); }) -#define local_save_flags(flags) ({ ia64_stop(); (flags) = __local_save_flags(); }) - -#define irqs_disabled() \ -({ \ - unsigned long __ia64_id_flags; \ - local_save_flags(__ia64_id_flags); \ - (__ia64_id_flags & IA64_PSR_I) == 0; \ -}) #ifdef __KERNEL__ diff --git a/arch/m32r/include/asm/irqflags.h b/arch/m32r/include/asm/irqflags.h new file mode 100644 index 00000000000..1f92d29982a --- /dev/null +++ b/arch/m32r/include/asm/irqflags.h @@ -0,0 +1,104 @@ +/* + * 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) 2001 Hiroyuki Kondo, Hirokazu Takata, and Hitoshi Yamamoto + * Copyright (C) 2004, 2006 Hirokazu Takata + */ + +#ifndef _ASM_M32R_IRQFLAGS_H +#define _ASM_M32R_IRQFLAGS_H + +#include + +static inline unsigned long arch_local_save_flags(void) +{ + unsigned long flags; + asm volatile("mvfc %0,psw" : "=r"(flags)); + return flags; +} + +static inline void arch_local_irq_disable(void) +{ +#if !defined(CONFIG_CHIP_M32102) && !defined(CONFIG_CHIP_M32104) + asm volatile ( + "clrpsw #0x40 -> nop" + : : : "memory"); +#else + unsigned long tmpreg0, tmpreg1; + asm volatile ( + "ld24 %0, #0 ; Use 32-bit insn. \n\t" + "mvfc %1, psw ; No interrupt can be accepted here. \n\t" + "mvtc %0, psw \n\t" + "and3 %0, %1, #0xffbf \n\t" + "mvtc %0, psw \n\t" + : "=&r" (tmpreg0), "=&r" (tmpreg1) + : + : "cbit", "memory"); +#endif +} + +static inline void arch_local_irq_enable(void) +{ +#if !defined(CONFIG_CHIP_M32102) && !defined(CONFIG_CHIP_M32104) + asm volatile ( + "setpsw #0x40 -> nop" + : : : "memory"); +#else + unsigned long tmpreg; + asm volatile ( + "mvfc %0, psw; \n\t" + "or3 %0, %0, #0x0040; \n\t" + "mvtc %0, psw; \n\t" + : "=&r" (tmpreg) + : + : "cbit", "memory"); +#endif +} + +static inline unsigned long arch_local_irq_save(void) +{ + unsigned long flags; + +#if !(defined(CONFIG_CHIP_M32102) || defined(CONFIG_CHIP_M32104)) + asm volatile ( + "mvfc %0, psw; \n\t" + "clrpsw #0x40 -> nop; \n\t" + : "=r" (flags) + : + : "memory"); +#else + unsigned long tmpreg; + asm volatile ( + "ld24 %1, #0 \n\t" + "mvfc %0, psw \n\t" + "mvtc %1, psw \n\t" + "and3 %1, %0, #0xffbf \n\t" + "mvtc %1, psw \n\t" + : "=r" (flags), "=&r" (tmpreg) + : + : "cbit", "memory"); +#endif + return flags; +} + +static inline void arch_local_irq_restore(unsigned long flags) +{ + asm volatile("mvtc %0,psw" + : + : "r" (flags) + : "cbit", "memory"); +} + +static inline bool arch_irqs_disabled_flags(unsigned long flags) +{ + return !(flags & 0x40); +} + +static inline bool arch_irqs_disabled(void) +{ + return arch_irqs_disabled_flags(arch_local_save_flags()); +} + +#endif /* _ASM_M32R_IRQFLAGS_H */ diff --git a/arch/m32r/include/asm/system.h b/arch/m32r/include/asm/system.h index c980f5ba8de..13c46794ccb 100644 --- a/arch/m32r/include/asm/system.h +++ b/arch/m32r/include/asm/system.h @@ -11,6 +11,7 @@ */ #include +#include #include #ifdef __KERNEL__ @@ -54,71 +55,6 @@ ); \ } while(0) -/* Interrupt Control */ -#if !defined(CONFIG_CHIP_M32102) && !defined(CONFIG_CHIP_M32104) -#define local_irq_enable() \ - __asm__ __volatile__ ("setpsw #0x40 -> nop": : :"memory") -#define local_irq_disable() \ - __asm__ __volatile__ ("clrpsw #0x40 -> nop": : :"memory") -#else /* CONFIG_CHIP_M32102 || CONFIG_CHIP_M32104 */ -static inline void local_irq_enable(void) -{ - unsigned long tmpreg; - __asm__ __volatile__( - "mvfc %0, psw; \n\t" - "or3 %0, %0, #0x0040; \n\t" - "mvtc %0, psw; \n\t" - : "=&r" (tmpreg) : : "cbit", "memory"); -} - -static inline void local_irq_disable(void) -{ - unsigned long tmpreg0, tmpreg1; - __asm__ __volatile__( - "ld24 %0, #0 ; Use 32-bit insn. \n\t" - "mvfc %1, psw ; No interrupt can be accepted here. \n\t" - "mvtc %0, psw \n\t" - "and3 %0, %1, #0xffbf \n\t" - "mvtc %0, psw \n\t" - : "=&r" (tmpreg0), "=&r" (tmpreg1) : : "cbit", "memory"); -} -#endif /* CONFIG_CHIP_M32102 || CONFIG_CHIP_M32104 */ - -#define local_save_flags(x) \ - __asm__ __volatile__("mvfc %0,psw" : "=r"(x) : /* no input */) - -#define local_irq_restore(x) \ - __asm__ __volatile__("mvtc %0,psw" : /* no outputs */ \ - : "r" (x) : "cbit", "memory") - -#if !(defined(CONFIG_CHIP_M32102) || defined(CONFIG_CHIP_M32104)) -#define local_irq_save(x) \ - __asm__ __volatile__( \ - "mvfc %0, psw; \n\t" \ - "clrpsw #0x40 -> nop; \n\t" \ - : "=r" (x) : /* no input */ : "memory") -#else /* CONFIG_CHIP_M32102 || CONFIG_CHIP_M32104 */ -#define local_irq_save(x) \ - ({ \ - unsigned long tmpreg; \ - __asm__ __volatile__( \ - "ld24 %1, #0 \n\t" \ - "mvfc %0, psw \n\t" \ - "mvtc %1, psw \n\t" \ - "and3 %1, %0, #0xffbf \n\t" \ - "mvtc %1, psw \n\t" \ - : "=r" (x), "=&r" (tmpreg) \ - : : "cbit", "memory"); \ - }) -#endif /* CONFIG_CHIP_M32102 || CONFIG_CHIP_M32104 */ - -#define irqs_disabled() \ - ({ \ - unsigned long flags; \ - local_save_flags(flags); \ - !(flags & 0x40); \ - }) - #define nop() __asm__ __volatile__ ("nop" : : ) #define xchg(ptr, x) \ diff --git a/arch/m68k/include/asm/entry_no.h b/arch/m68k/include/asm/entry_no.h index 907ed03d792..80e41492aa2 100644 --- a/arch/m68k/include/asm/entry_no.h +++ b/arch/m68k/include/asm/entry_no.h @@ -28,7 +28,7 @@ * M68K COLDFIRE */ -#define ALLOWINT 0xf8ff +#define ALLOWINT (~0x700) #ifdef __ASSEMBLY__ diff --git a/arch/m68k/include/asm/irqflags.h b/arch/m68k/include/asm/irqflags.h new file mode 100644 index 00000000000..4a5b284a155 --- /dev/null +++ b/arch/m68k/include/asm/irqflags.h @@ -0,0 +1,76 @@ +#ifndef _M68K_IRQFLAGS_H +#define _M68K_IRQFLAGS_H + +#include +#include +#include +#include +#include + +static inline unsigned long arch_local_save_flags(void) +{ + unsigned long flags; + asm volatile ("movew %%sr,%0" : "=d" (flags) : : "memory"); + return flags; +} + +static inline void arch_local_irq_disable(void) +{ +#ifdef CONFIG_COLDFIRE + asm volatile ( + "move %/sr,%%d0 \n\t" + "ori.l #0x0700,%%d0 \n\t" + "move %%d0,%/sr \n" + : /* no outputs */ + : + : "cc", "%d0", "memory"); +#else + asm volatile ("oriw #0x0700,%%sr" : : : "memory"); +#endif +} + +static inline void arch_local_irq_enable(void) +{ +#if defined(CONFIG_COLDFIRE) + asm volatile ( + "move %/sr,%%d0 \n\t" + "andi.l #0xf8ff,%%d0 \n\t" + "move %%d0,%/sr \n" + : /* no outputs */ + : + : "cc", "%d0", "memory"); +#else +# if defined(CONFIG_MMU) + if (MACH_IS_Q40 || !hardirq_count()) +# endif + asm volatile ( + "andiw %0,%%sr" + : + : "i" (ALLOWINT) + : "memory"); +#endif +} + +static inline unsigned long arch_local_irq_save(void) +{ + unsigned long flags = arch_local_save_flags(); + arch_local_irq_disable(); + return flags; +} + +static inline void arch_local_irq_restore(unsigned long flags) +{ + asm volatile ("movew %0,%%sr" : : "d" (flags) : "memory"); +} + +static inline bool arch_irqs_disabled_flags(unsigned long flags) +{ + return (flags & ~ALLOWINT) != 0; +} + +static inline bool arch_irqs_disabled(void) +{ + return arch_irqs_disabled_flags(arch_local_save_flags()); +} + +#endif /* _M68K_IRQFLAGS_H */ diff --git a/arch/m68k/include/asm/system_mm.h b/arch/m68k/include/asm/system_mm.h index dbb6515ffd5..12053c44ccc 100644 --- a/arch/m68k/include/asm/system_mm.h +++ b/arch/m68k/include/asm/system_mm.h @@ -3,6 +3,7 @@ #include #include +#include #include #include @@ -62,30 +63,6 @@ asmlinkage void resume(void); #define smp_wmb() barrier() #define smp_read_barrier_depends() ((void)0) -/* interrupt control.. */ -#if 0 -#define local_irq_enable() asm volatile ("andiw %0,%%sr": : "i" (ALLOWINT) : "memory") -#else -#include -#define local_irq_enable() ({ \ - if (MACH_IS_Q40 || !hardirq_count()) \ - asm volatile ("andiw %0,%%sr": : "i" (ALLOWINT) : "memory"); \ -}) -#endif -#define local_irq_disable() asm volatile ("oriw #0x0700,%%sr": : : "memory") -#define local_save_flags(x) asm volatile ("movew %%sr,%0":"=d" (x) : : "memory") -#define local_irq_restore(x) asm volatile ("movew %0,%%sr": :"d" (x) : "memory") - -static inline int irqs_disabled(void) -{ - unsigned long flags; - local_save_flags(flags); - return flags & ~ALLOWINT; -} - -/* For spinlocks etc */ -#define local_irq_save(x) ({ local_save_flags(x); local_irq_disable(); }) - #define xchg(ptr,x) ((__typeof__(*(ptr)))__xchg((unsigned long)(x),(ptr),sizeof(*(ptr)))) struct __xchg_dummy { unsigned long a[100]; }; diff --git a/arch/m68k/include/asm/system_no.h b/arch/m68k/include/asm/system_no.h index 3c0718d7439..20126c09794 100644 --- a/arch/m68k/include/asm/system_no.h +++ b/arch/m68k/include/asm/system_no.h @@ -2,6 +2,7 @@ #define _M68KNOMMU_SYSTEM_H #include +#include #include #include @@ -46,54 +47,6 @@ asmlinkage void resume(void); (last) = _last; \ } -#ifdef CONFIG_COLDFIRE -#define local_irq_enable() __asm__ __volatile__ ( \ - "move %/sr,%%d0\n\t" \ - "andi.l #0xf8ff,%%d0\n\t" \ - "move %%d0,%/sr\n" \ - : /* no outputs */ \ - : \ - : "cc", "%d0", "memory") -#define local_irq_disable() __asm__ __volatile__ ( \ - "move %/sr,%%d0\n\t" \ - "ori.l #0x0700,%%d0\n\t" \ - "move %%d0,%/sr\n" \ - : /* no outputs */ \ - : \ - : "cc", "%d0", "memory") -/* For spinlocks etc */ -#define local_irq_save(x) __asm__ __volatile__ ( \ - "movew %%sr,%0\n\t" \ - "movew #0x0700,%%d0\n\t" \ - "or.l %0,%%d0\n\t" \ - "movew %%d0,%/sr" \ - : "=d" (x) \ - : \ - : "cc", "%d0", "memory") -#else - -/* portable version */ /* FIXME - see entry.h*/ -#define ALLOWINT 0xf8ff - -#define local_irq_enable() asm volatile ("andiw %0,%%sr": : "i" (ALLOWINT) : "memory") -#define local_irq_disable() asm volatile ("oriw #0x0700,%%sr": : : "memory") -#endif - -#define local_save_flags(x) asm volatile ("movew %%sr,%0":"=d" (x) : : "memory") -#define local_irq_restore(x) asm volatile ("movew %0,%%sr": :"d" (x) : "memory") - -/* For spinlocks etc */ -#ifndef local_irq_save -#define local_irq_save(x) do { local_save_flags(x); local_irq_disable(); } while (0) -#endif - -#define irqs_disabled() \ -({ \ - unsigned long flags; \ - local_save_flags(flags); \ - ((flags & 0x0700) == 0x0700); \ -}) - #define iret() __asm__ __volatile__ ("rte": : :"memory", "sp", "cc") /* @@ -206,12 +159,4 @@ static inline unsigned long __xchg(unsigned long x, volatile void * ptr, int siz #define arch_align_stack(x) (x) -static inline int irqs_disabled_flags(unsigned long flags) -{ - if (flags & 0x0700) - return 0; - else - return 1; -} - #endif /* _M68KNOMMU_SYSTEM_H */ diff --git a/arch/m68knommu/kernel/asm-offsets.c b/arch/m68knommu/kernel/asm-offsets.c index 9a8876f715d..24335022fa2 100644 --- a/arch/m68knommu/kernel/asm-offsets.c +++ b/arch/m68knommu/kernel/asm-offsets.c @@ -74,8 +74,6 @@ int main(void) DEFINE(PT_PTRACED, PT_PTRACED); - DEFINE(THREAD_SIZE, THREAD_SIZE); - /* Offsets in thread_info structure */ DEFINE(TI_TASK, offsetof(struct thread_info, task)); DEFINE(TI_EXECDOMAIN, offsetof(struct thread_info, exec_domain)); diff --git a/arch/m68knommu/platform/coldfire/head.S b/arch/m68knommu/platform/coldfire/head.S index 4b91aa24eb0..0b2d7c7adf7 100644 --- a/arch/m68knommu/platform/coldfire/head.S +++ b/arch/m68knommu/platform/coldfire/head.S @@ -15,6 +15,7 @@ #include #include #include +#include /*****************************************************************************/ diff --git a/arch/microblaze/include/asm/irqflags.h b/arch/microblaze/include/asm/irqflags.h index 2c38c6d8017..5fd31905775 100644 --- a/arch/microblaze/include/asm/irqflags.h +++ b/arch/microblaze/include/asm/irqflags.h @@ -9,103 +9,114 @@ #ifndef _ASM_MICROBLAZE_IRQFLAGS_H #define _ASM_MICROBLAZE_IRQFLAGS_H -#include +#include #include -# if CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR - -# define raw_local_irq_save(flags) \ - do { \ - asm volatile (" msrclr %0, %1; \ - nop;" \ - : "=r"(flags) \ - : "i"(MSR_IE) \ - : "memory"); \ - } while (0) - -# define raw_local_irq_disable() \ - do { \ - asm volatile (" msrclr r0, %0; \ - nop;" \ - : \ - : "i"(MSR_IE) \ - : "memory"); \ - } while (0) - -# define raw_local_irq_enable() \ - do { \ - asm volatile (" msrset r0, %0; \ - nop;" \ - : \ - : "i"(MSR_IE) \ - : "memory"); \ - } while (0) - -# else /* CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR == 0 */ - -# define raw_local_irq_save(flags) \ - do { \ - register unsigned tmp; \ - asm volatile (" mfs %0, rmsr; \ - nop; \ - andi %1, %0, %2; \ - mts rmsr, %1; \ - nop;" \ - : "=r"(flags), "=r" (tmp) \ - : "i"(~MSR_IE) \ - : "memory"); \ - } while (0) - -# define raw_local_irq_disable() \ - do { \ - register unsigned tmp; \ - asm volatile (" mfs %0, rmsr; \ - nop; \ - andi %0, %0, %1; \ - mts rmsr, %0; \ - nop;" \ - : "=r"(tmp) \ - : "i"(~MSR_IE) \ - : "memory"); \ - } while (0) - -# define raw_local_irq_enable() \ - do { \ - register unsigned tmp; \ - asm volatile (" mfs %0, rmsr; \ - nop; \ - ori %0, %0, %1; \ - mts rmsr, %0; \ - nop;" \ - : "=r"(tmp) \ - : "i"(MSR_IE) \ - : "memory"); \ - } while (0) - -# endif /* CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR */ - -#define raw_local_irq_restore(flags) \ - do { \ - asm volatile (" mts rmsr, %0; \ - nop;" \ - : \ - : "r"(flags) \ - : "memory"); \ - } while (0) - -static inline unsigned long get_msr(void) +#ifdef CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR + +static inline unsigned long arch_local_irq_save(void) +{ + unsigned long flags; + asm volatile(" msrclr %0, %1 \n" + " nop \n" + : "=r"(flags) + : "i"(MSR_IE) + : "memory"); + return flags; +} + +static inline void arch_local_irq_disable(void) +{ + /* this uses r0 without declaring it - is that correct? */ + asm volatile(" msrclr r0, %0 \n" + " nop \n" + : + : "i"(MSR_IE) + : "memory"); +} + +static inline void arch_local_irq_enable(void) +{ + /* this uses r0 without declaring it - is that correct? */ + asm volatile(" msrset r0, %0 \n" + " nop \n" + : + : "i"(MSR_IE) + : "memory"); +} + +#else /* !CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR */ + +static inline unsigned long arch_local_irq_save(void) +{ + unsigned long flags, tmp; + asm volatile (" mfs %0, rmsr \n" + " nop \n" + " andi %1, %0, %2 \n" + " mts rmsr, %1 \n" + " nop \n" + : "=r"(flags), "=r"(tmp) + : "i"(~MSR_IE) + : "memory"); + return flags; +} + +static inline void arch_local_irq_disable(void) +{ + unsigned long tmp; + asm volatile(" mfs %0, rmsr \n" + " nop \n" + " andi %0, %0, %1 \n" + " mts rmsr, %0 \n" + " nop \n" + : "=r"(tmp) + : "i"(~MSR_IE) + : "memory"); +} + +static inline void arch_local_irq_enable(void) +{ + unsigned long tmp; + asm volatile(" mfs %0, rmsr \n" + " nop \n" + " ori %0, %0, %1 \n" + " mts rmsr, %0 \n" + " nop \n" + : "=r"(tmp) + : "i"(MSR_IE) + : "memory"); +} + +#endif /* CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR */ + +static inline unsigned long arch_local_save_flags(void) { unsigned long flags; - asm volatile (" mfs %0, rmsr; \ - nop;" \ - : "=r"(flags) \ - : \ - : "memory"); \ + asm volatile(" mfs %0, rmsr \n" + " nop \n" + : "=r"(flags) + : + : "memory"); return flags; } -#define raw_local_save_flags(flags) ((flags) = get_msr()) -#define raw_irqs_disabled() ((get_msr() & MSR_IE) == 0) -#define raw_irqs_disabled_flags(flags) ((flags & MSR_IE) == 0) +static inline void arch_local_irq_restore(unsigned long flags) +{ + asm volatile(" mts rmsr, %0 \n" + " nop \n" + : + : "r"(flags) + : "memory"); +} + +static inline bool arch_irqs_disabled_flags(unsigned long flags) +{ + return (flags & MSR_IE) == 0; +} + +static inline bool arch_irqs_disabled(void) +{ + return arch_irqs_disabled_flags(arch_local_save_flags()); +} #endif /* _ASM_MICROBLAZE_IRQFLAGS_H */ diff --git a/arch/mips/include/asm/irqflags.h b/arch/mips/include/asm/irqflags.h index 701ec0ba8fa..9ef3b0d1789 100644 --- a/arch/mips/include/asm/irqflags.h +++ b/arch/mips/include/asm/irqflags.h @@ -17,7 +17,7 @@ #include __asm__( - " .macro raw_local_irq_enable \n" + " .macro arch_local_irq_enable \n" " .set push \n" " .set reorder \n" " .set noat \n" @@ -40,7 +40,7 @@ __asm__( extern void smtc_ipi_replay(void); -static inline void raw_local_irq_enable(void) +static inline void arch_local_irq_enable(void) { #ifdef CONFIG_MIPS_MT_SMTC /* @@ -50,7 +50,7 @@ static inline void raw_local_irq_enable(void) smtc_ipi_replay(); #endif __asm__ __volatile__( - "raw_local_irq_enable" + "arch_local_irq_enable" : /* no outputs */ : /* no inputs */ : "memory"); @@ -76,7 +76,7 @@ static inline void raw_local_irq_enable(void) * Workaround: mask EXL bit of the result or place a nop before mfc0. */ __asm__( - " .macro raw_local_irq_disable\n" + " .macro arch_local_irq_disable\n" " .set push \n" " .set noat \n" #ifdef CONFIG_MIPS_MT_SMTC @@ -97,17 +97,17 @@ __asm__( " .set pop \n" " .endm \n"); -static inline void raw_local_irq_disable(void) +static inline void arch_local_irq_disable(void) { __asm__ __volatile__( - "raw_local_irq_disable" + "arch_local_irq_disable" : /* no outputs */ : /* no inputs */ : "memory"); } __asm__( - " .macro raw_local_save_flags flags \n" + " .macro arch_local_save_flags flags \n" " .set push \n" " .set reorder \n" #ifdef CONFIG_MIPS_MT_SMTC @@ -118,13 +118,15 @@ __asm__( " .set pop \n" " .endm \n"); -#define raw_local_save_flags(x) \ -__asm__ __volatile__( \ - "raw_local_save_flags %0" \ - : "=r" (x)) +static inline unsigned long arch_local_save_flags(void) +{ + unsigned long flags; + asm volatile("arch_local_save_flags %0" : "=r" (flags)); + return flags; +} __asm__( - " .macro raw_local_irq_save result \n" + " .macro arch_local_irq_save result \n" " .set push \n" " .set reorder \n" " .set noat \n" @@ -148,15 +150,18 @@ __asm__( " .set pop \n" " .endm \n"); -#define raw_local_irq_save(x) \ -__asm__ __volatile__( \ - "raw_local_irq_save\t%0" \ - : "=r" (x) \ - : /* no inputs */ \ - : "memory") +static inline unsigned long arch_local_irq_save(void) +{ + unsigned long flags; + asm volatile("arch_local_irq_save\t%0" + : "=r" (flags) + : /* no inputs */ + : "memory"); + return flags; +} __asm__( - " .macro raw_local_irq_restore flags \n" + " .macro arch_local_irq_restore flags \n" " .set push \n" " .set noreorder \n" " .set noat \n" @@ -196,7 +201,7 @@ __asm__( " .endm \n"); -static inline void raw_local_irq_restore(unsigned long flags) +static inline void arch_local_irq_restore(unsigned long flags) { unsigned long __tmp1; @@ -211,24 +216,24 @@ static inline void raw_local_irq_restore(unsigned long flags) #endif __asm__ __volatile__( - "raw_local_irq_restore\t%0" + "arch_local_irq_restore\t%0" : "=r" (__tmp1) : "0" (flags) : "memory"); } -static inline void __raw_local_irq_restore(unsigned long flags) +static inline void __arch_local_irq_restore(unsigned long flags) { unsigned long __tmp1; __asm__ __volatile__( - "raw_local_irq_restore\t%0" + "arch_local_irq_restore\t%0" : "=r" (__tmp1) : "0" (flags) : "memory"); } -static inline int raw_irqs_disabled_flags(unsigned long flags) +static inline int arch_irqs_disabled_flags(unsigned long flags) { #ifdef CONFIG_MIPS_MT_SMTC /* diff --git a/arch/mips/kernel/smtc.c b/arch/mips/kernel/smtc.c index cfeb2c15589..39c08254b0f 100644 --- a/arch/mips/kernel/smtc.c +++ b/arch/mips/kernel/smtc.c @@ -1038,7 +1038,7 @@ void deferred_smtc_ipi(void) * but it's more efficient, given that we're already * running down the IPI queue. */ - __raw_local_irq_restore(flags); + __arch_local_irq_restore(flags); } } @@ -1190,7 +1190,7 @@ void smtc_ipi_replay(void) /* ** But use a raw restore here to avoid recursion. */ - __raw_local_irq_restore(flags); + __arch_local_irq_restore(flags); if (pipi) { self_ipi(pipi); diff --git a/arch/mn10300/include/asm/irqflags.h b/arch/mn10300/include/asm/irqflags.h new file mode 100644 index 00000000000..5e529a117cb --- /dev/null +++ b/arch/mn10300/include/asm/irqflags.h @@ -0,0 +1,123 @@ +/* MN10300 IRQ flag handling + * + * Copyright (C) 2010 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public Licence + * as published by the Free Software Foundation; either version + * 2 of the Licence, or (at your option) any later version. + */ + +#ifndef _ASM_IRQFLAGS_H +#define _ASM_IRQFLAGS_H + +#include + +/* + * interrupt control + * - "disabled": run in IM1/2 + * - level 0 - GDB stub + * - level 1 - virtual serial DMA (if present) + * - level 5 - normal interrupt priority + * - level 6 - timer interrupt + * - "enabled": run in IM7 + */ +#ifdef CONFIG_MN10300_TTYSM +#define MN10300_CLI_LEVEL EPSW_IM_2 +#else +#define MN10300_CLI_LEVEL EPSW_IM_1 +#endif + +#ifndef __ASSEMBLY__ + +static inline unsigned long arch_local_save_flags(void) +{ + unsigned long flags; + + asm volatile("mov epsw,%0" : "=d"(flags)); + return flags; +} + +static inline void arch_local_irq_disable(void) +{ + asm volatile( + " and %0,epsw \n" + " or %1,epsw \n" + " nop \n" + " nop \n" + " nop \n" + : + : "i"(~EPSW_IM), "i"(EPSW_IE | MN10300_CLI_LEVEL) + : "memory"); +} + +static inline unsigned long arch_local_irq_save(void) +{ + unsigned long flags; + + flags = arch_local_save_flags(); + arch_local_irq_disable(); + return flags; +} + +/* + * we make sure arch_irq_enable() doesn't cause priority inversion + */ +extern unsigned long __mn10300_irq_enabled_epsw; + +static inline void arch_local_irq_enable(void) +{ + unsigned long tmp; + + asm volatile( + " mov epsw,%0 \n" + " and %1,%0 \n" + " or %2,%0 \n" + " mov %0,epsw \n" + : "=&d"(tmp) + : "i"(~EPSW_IM), "r"(__mn10300_irq_enabled_epsw) + : "memory"); +} + +static inline void arch_local_irq_restore(unsigned long flags) +{ + asm volatile( + " mov %0,epsw \n" + " nop \n" + " nop \n" + " nop \n" + : + : "d"(flags) + : "memory", "cc"); +} + +static inline bool arch_irqs_disabled_flags(unsigned long flags) +{ + return (flags & EPSW_IM) <= MN10300_CLI_LEVEL; +} + +static inline bool arch_irqs_disabled(void) +{ + return arch_irqs_disabled_flags(arch_local_save_flags()); +} + +/* + * Hook to save power by halting the CPU + * - called from the idle loop + * - must reenable interrupts (which takes three instruction cycles to complete) + */ +static inline void arch_safe_halt(void) +{ + asm volatile( + " or %0,epsw \n" + " nop \n" + " nop \n" + " bset %2,(%1) \n" + : + : "i"(EPSW_IE|EPSW_IM), "n"(&CPUM), "i"(CPUM_SLEEP) + : "cc"); +} + +#endif /* __ASSEMBLY__ */ +#endif /* _ASM_IRQFLAGS_H */ diff --git a/arch/mn10300/include/asm/system.h b/arch/mn10300/include/asm/system.h index 3636c054dcd..9f7c7e17c01 100644 --- a/arch/mn10300/include/asm/system.h +++ b/arch/mn10300/include/asm/system.h @@ -17,6 +17,7 @@ #ifndef __ASSEMBLY__ #include +#include struct task_struct; struct thread_struct; @@ -79,114 +80,6 @@ do { \ #define read_barrier_depends() do {} while (0) #define smp_read_barrier_depends() do {} while (0) -/*****************************************************************************/ -/* - * interrupt control - * - "disabled": run in IM1/2 - * - level 0 - GDB stub - * - level 1 - virtual serial DMA (if present) - * - level 5 - normal interrupt priority - * - level 6 - timer interrupt - * - "enabled": run in IM7 - */ -#ifdef CONFIG_MN10300_TTYSM -#define MN10300_CLI_LEVEL EPSW_IM_2 -#else -#define MN10300_CLI_LEVEL EPSW_IM_1 -#endif - -#define local_save_flags(x) \ -do { \ - typecheck(unsigned long, x); \ - asm volatile( \ - " mov epsw,%0 \n" \ - : "=d"(x) \ - ); \ -} while (0) - -#define local_irq_disable() \ -do { \ - asm volatile( \ - " and %0,epsw \n" \ - " or %1,epsw \n" \ - " nop \n" \ - " nop \n" \ - " nop \n" \ - : \ - : "i"(~EPSW_IM), "i"(EPSW_IE | MN10300_CLI_LEVEL) \ - ); \ -} while (0) - -#define local_irq_save(x) \ -do { \ - local_save_flags(x); \ - local_irq_disable(); \ -} while (0) - -/* - * we make sure local_irq_enable() doesn't cause priority inversion - */ -#ifndef __ASSEMBLY__ - -extern unsigned long __mn10300_irq_enabled_epsw; - -#endif - -#define local_irq_enable() \ -do { \ - unsigned long tmp; \ - \ - asm volatile( \ - " mov epsw,%0 \n" \ - " and %1,%0 \n" \ - " or %2,%0 \n" \ - " mov %0,epsw \n" \ - : "=&d"(tmp) \ - : "i"(~EPSW_IM), "r"(__mn10300_irq_enabled_epsw) \ - : "cc" \ - ); \ -} while (0) - -#define local_irq_restore(x) \ -do { \ - typecheck(unsigned long, x); \ - asm volatile( \ - " mov %0,epsw \n" \ - " nop \n" \ - " nop \n" \ - " nop \n" \ - : \ - : "d"(x) \ - : "memory", "cc" \ - ); \ -} while (0) - -#define irqs_disabled() \ -({ \ - unsigned long flags; \ - local_save_flags(flags); \ - (flags & EPSW_IM) <= MN10300_CLI_LEVEL; \ -}) - -/* hook to save power by halting the CPU - * - called from the idle loop - * - must reenable interrupts (which takes three instruction cycles to complete) - */ -#define safe_halt() \ -do { \ - asm volatile(" or %0,epsw \n" \ - " nop \n" \ - " nop \n" \ - " bset %2,(%1) \n" \ - : \ - : "i"(EPSW_IE|EPSW_IM), "n"(&CPUM), "i"(CPUM_SLEEP)\ - : "cc" \ - ); \ -} while (0) - -#define STI or EPSW_IE|EPSW_IM,epsw -#define CLI and ~EPSW_IM,epsw; or EPSW_IE|MN10300_CLI_LEVEL,epsw; nop; nop; nop - /*****************************************************************************/ /* * MN10300 doesn't actually have an exchange instruction diff --git a/arch/mn10300/kernel/entry.S b/arch/mn10300/kernel/entry.S index d9ed5a15c54..3d394b4eefb 100644 --- a/arch/mn10300/kernel/entry.S +++ b/arch/mn10300/kernel/entry.S @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include diff --git a/arch/parisc/include/asm/irqflags.h b/arch/parisc/include/asm/irqflags.h new file mode 100644 index 00000000000..34f9cb9b475 --- /dev/null +++ b/arch/parisc/include/asm/irqflags.h @@ -0,0 +1,46 @@ +#ifndef __PARISC_IRQFLAGS_H +#define __PARISC_IRQFLAGS_H + +#include +#include + +static inline unsigned long arch_local_save_flags(void) +{ + unsigned long flags; + asm volatile("ssm 0, %0" : "=r" (flags) : : "memory"); + return flags; +} + +static inline void arch_local_irq_disable(void) +{ + asm volatile("rsm %0,%%r0\n" : : "i" (PSW_I) : "memory"); +} + +static inline void arch_local_irq_enable(void) +{ + asm volatile("ssm %0,%%r0\n" : : "i" (PSW_I) : "memory"); +} + +static inline unsigned long arch_local_irq_save(void) +{ + unsigned long flags; + asm volatile("rsm %1,%0" : "=r" (flags) : "i" (PSW_I) : "memory"); + return flags; +} + +static inline void arch_local_irq_restore(unsigned long flags) +{ + asm volatile("mtsm %0" : : "r" (flags) : "memory"); +} + +static inline bool arch_irqs_disabled_flags(unsigned long flags) +{ + return (flags & PSW_I) == 0; +} + +static inline bool arch_irqs_disabled(void) +{ + return arch_irqs_disabled_flags(arch_local_save_flags()); +} + +#endif /* __PARISC_IRQFLAGS_H */ diff --git a/arch/parisc/include/asm/system.h b/arch/parisc/include/asm/system.h index 2ab4af58ecb..b19e63a8e84 100644 --- a/arch/parisc/include/asm/system.h +++ b/arch/parisc/include/asm/system.h @@ -1,7 +1,7 @@ #ifndef __PARISC_SYSTEM_H #define __PARISC_SYSTEM_H -#include +#include /* The program status word as bitfields. */ struct pa_psw { @@ -48,23 +48,6 @@ extern struct task_struct *_switch_to(struct task_struct *, struct task_struct * (last) = _switch_to(prev, next); \ } while(0) -/* interrupt control */ -#define local_save_flags(x) __asm__ __volatile__("ssm 0, %0" : "=r" (x) : : "memory") -#define local_irq_disable() __asm__ __volatile__("rsm %0,%%r0\n" : : "i" (PSW_I) : "memory" ) -#define local_irq_enable() __asm__ __volatile__("ssm %0,%%r0\n" : : "i" (PSW_I) : "memory" ) - -#define local_irq_save(x) \ - __asm__ __volatile__("rsm %1,%0" : "=r" (x) :"i" (PSW_I) : "memory" ) -#define local_irq_restore(x) \ - __asm__ __volatile__("mtsm %0" : : "r" (x) : "memory" ) - -#define irqs_disabled() \ -({ \ - unsigned long flags; \ - local_save_flags(flags); \ - (flags & PSW_I) == 0; \ -}) - #define mfctl(reg) ({ \ unsigned long cr; \ __asm__ __volatile__( \ diff --git a/arch/powerpc/include/asm/hw_irq.h b/arch/powerpc/include/asm/hw_irq.h index bd100fcf40d..ff08b70b36d 100644 --- a/arch/powerpc/include/asm/hw_irq.h +++ b/arch/powerpc/include/asm/hw_irq.h @@ -16,42 +16,57 @@ extern void timer_interrupt(struct pt_regs *); #ifdef CONFIG_PPC64 #include -static inline unsigned long local_get_flags(void) +static inline unsigned long arch_local_save_flags(void) { unsigned long flags; - __asm__ __volatile__("lbz %0,%1(13)" - : "=r" (flags) - : "i" (offsetof(struct paca_struct, soft_enabled))); + asm volatile( + "lbz %0,%1(13)" + : "=r" (flags) + : "i" (offsetof(struct paca_struct, soft_enabled))); return flags; } -static inline unsigned long raw_local_irq_disable(void) +static inline unsigned long arch_local_irq_disable(void) { unsigned long flags, zero; - __asm__ __volatile__("li %1,0; lbz %0,%2(13); stb %1,%2(13)" - : "=r" (flags), "=&r" (zero) - : "i" (offsetof(struct paca_struct, soft_enabled)) - : "memory"); + asm volatile( + "li %1,0; lbz %0,%2(13); stb %1,%2(13)" + : "=r" (flags), "=&r" (zero) + : "i" (offsetof(struct paca_struct, soft_enabled)) + : "memory"); return flags; } -extern void raw_local_irq_restore(unsigned long); +extern void arch_local_irq_restore(unsigned long); extern void iseries_handle_interrupts(void); -#define raw_local_irq_enable() raw_local_irq_restore(1) -#define raw_local_save_flags(flags) ((flags) = local_get_flags()) -#define raw_local_irq_save(flags) ((flags) = raw_local_irq_disable()) +static inline void arch_local_irq_enable(void) +{ + arch_local_irq_restore(1); +} + +static inline unsigned long arch_local_irq_save(void) +{ + return arch_local_irq_disable(); +} + +static inline bool arch_irqs_disabled_flags(unsigned long flags) +{ + return flags == 0; +} -#define raw_irqs_disabled() (local_get_flags() == 0) -#define raw_irqs_disabled_flags(flags) ((flags) == 0) +static inline bool arch_irqs_disabled(void) +{ + return arch_irqs_disabled_flags(arch_local_save_flags()); +} #ifdef CONFIG_PPC_BOOK3E -#define __hard_irq_enable() __asm__ __volatile__("wrteei 1": : :"memory"); -#define __hard_irq_disable() __asm__ __volatile__("wrteei 0": : :"memory"); +#define __hard_irq_enable() asm volatile("wrteei 1" : : : "memory"); +#define __hard_irq_disable() asm volatile("wrteei 0" : : : "memory"); #else #define __hard_irq_enable() __mtmsrd(mfmsr() | MSR_EE, 1) #define __hard_irq_disable() __mtmsrd(mfmsr() & ~MSR_EE, 1) @@ -64,64 +79,66 @@ extern void iseries_handle_interrupts(void); get_paca()->hard_enabled = 0; \ } while(0) -#else +#else /* CONFIG_PPC64 */ -#if defined(CONFIG_BOOKE) #define SET_MSR_EE(x) mtmsr(x) -#define raw_local_irq_restore(flags) __asm__ __volatile__("wrtee %0" : : "r" (flags) : "memory") + +static inline unsigned long arch_local_save_flags(void) +{ + return mfmsr(); +} + +static inline void arch_local_irq_restore(unsigned long flags) +{ +#if defined(CONFIG_BOOKE) + asm volatile("wrtee %0" : : "r" (flags) : "memory"); #else -#define SET_MSR_EE(x) mtmsr(x) -#define raw_local_irq_restore(flags) mtmsr(flags) + mtmsr(flags); #endif +} -static inline void raw_local_irq_disable(void) +static inline unsigned long arch_local_irq_save(void) { + unsigned long flags = arch_local_save_flags(); #ifdef CONFIG_BOOKE - __asm__ __volatile__("wrteei 0": : :"memory"); + asm volatile("wrteei 0" : : : "memory"); #else - unsigned long msr; - - msr = mfmsr(); - SET_MSR_EE(msr & ~MSR_EE); + SET_MSR_EE(flags & ~MSR_EE); #endif + return flags; } -static inline void raw_local_irq_enable(void) +static inline void arch_local_irq_disable(void) { #ifdef CONFIG_BOOKE - __asm__ __volatile__("wrteei 1": : :"memory"); + asm volatile("wrteei 0" : : : "memory"); #else - unsigned long msr; - - msr = mfmsr(); - SET_MSR_EE(msr | MSR_EE); + arch_local_irq_save(); #endif } -static inline void raw_local_irq_save_ptr(unsigned long *flags) +static inline void arch_local_irq_enable(void) { - unsigned long msr; - msr = mfmsr(); - *flags = msr; #ifdef CONFIG_BOOKE - __asm__ __volatile__("wrteei 0": : :"memory"); + asm volatile("wrteei 1" : : : "memory"); #else - SET_MSR_EE(msr & ~MSR_EE); + unsigned long msr = mfmsr(); + SET_MSR_EE(msr | MSR_EE); #endif } -#define raw_local_save_flags(flags) ((flags) = mfmsr()) -#define raw_local_irq_save(flags) raw_local_irq_save_ptr(&flags) -#define raw_irqs_disabled() ((mfmsr() & MSR_EE) == 0) -#define raw_irqs_disabled_flags(flags) (((flags) & MSR_EE) == 0) - -#define hard_irq_disable() raw_local_irq_disable() - -static inline int irqs_disabled_flags(unsigned long flags) +static inline bool arch_irqs_disabled_flags(unsigned long flags) { return (flags & MSR_EE) == 0; } +static inline bool arch_irqs_disabled(void) +{ + return arch_irqs_disabled_flags(arch_local_save_flags()); +} + +#define hard_irq_disable() arch_local_irq_disable() + #endif /* CONFIG_PPC64 */ /* diff --git a/arch/powerpc/include/asm/irqflags.h b/arch/powerpc/include/asm/irqflags.h index 5f68ecfdf51..b85d8ddbb66 100644 --- a/arch/powerpc/include/asm/irqflags.h +++ b/arch/powerpc/include/asm/irqflags.h @@ -6,7 +6,7 @@ #ifndef __ASSEMBLY__ /* - * Get definitions for raw_local_save_flags(x), etc. + * Get definitions for arch_local_save_flags(x), etc. */ #include diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S index f53029a0155..39b0c48872d 100644 --- a/arch/powerpc/kernel/exceptions-64s.S +++ b/arch/powerpc/kernel/exceptions-64s.S @@ -818,12 +818,12 @@ END_FW_FTR_SECTION_IFCLR(FW_FEATURE_ISERIES) /* * hash_page couldn't handle it, set soft interrupt enable back - * to what it was before the trap. Note that .raw_local_irq_restore + * to what it was before the trap. Note that .arch_local_irq_restore * handles any interrupts pending at this point. */ ld r3,SOFTE(r1) TRACE_AND_RESTORE_IRQ_PARTIAL(r3, 11f) - bl .raw_local_irq_restore + bl .arch_local_irq_restore b 11f /* We have a data breakpoint exception - handle it */ diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index 4a65386995d..1903290f546 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -116,7 +116,7 @@ static inline notrace void set_soft_enabled(unsigned long enable) : : "r" (enable), "i" (offsetof(struct paca_struct, soft_enabled))); } -notrace void raw_local_irq_restore(unsigned long en) +notrace void arch_local_irq_restore(unsigned long en) { /* * get_paca()->soft_enabled = en; @@ -192,7 +192,7 @@ notrace void raw_local_irq_restore(unsigned long en) __hard_irq_enable(); } -EXPORT_SYMBOL(raw_local_irq_restore); +EXPORT_SYMBOL(arch_local_irq_restore); #endif /* CONFIG_PPC64 */ static int show_other_interrupts(struct seq_file *p, int prec) diff --git a/arch/s390/include/asm/irqflags.h b/arch/s390/include/asm/irqflags.h index 15b3ac25389..865d6d891ac 100644 --- a/arch/s390/include/asm/irqflags.h +++ b/arch/s390/include/asm/irqflags.h @@ -8,8 +8,8 @@ #include -/* store then or system mask. */ -#define __raw_local_irq_stosm(__or) \ +/* store then OR system mask. */ +#define __arch_local_irq_stosm(__or) \ ({ \ unsigned long __mask; \ asm volatile( \ @@ -18,8 +18,8 @@ __mask; \ }) -/* store then and system mask. */ -#define __raw_local_irq_stnsm(__and) \ +/* store then AND system mask. */ +#define __arch_local_irq_stnsm(__and) \ ({ \ unsigned long __mask; \ asm volatile( \ @@ -29,39 +29,44 @@ }) /* set system mask. */ -#define __raw_local_irq_ssm(__mask) \ -({ \ - asm volatile("ssm %0" : : "Q" (__mask) : "memory"); \ -}) +static inline void __arch_local_irq_ssm(unsigned long flags) +{ + asm volatile("ssm %0" : : "Q" (flags) : "memory"); +} -/* interrupt control.. */ -static inline unsigned long raw_local_irq_enable(void) +static inline unsigned long arch_local_save_flags(void) { - return __raw_local_irq_stosm(0x03); + return __arch_local_irq_stosm(0x00); } -static inline unsigned long raw_local_irq_disable(void) +static inline unsigned long arch_local_irq_save(void) { - return __raw_local_irq_stnsm(0xfc); + return __arch_local_irq_stnsm(0xfc); } -#define raw_local_save_flags(x) \ -do { \ - typecheck(unsigned long, x); \ - (x) = __raw_local_irq_stosm(0x00); \ -} while (0) +static inline void arch_local_irq_disable(void) +{ + arch_local_irq_save(); +} -static inline void raw_local_irq_restore(unsigned long flags) +static inline void arch_local_irq_enable(void) { - __raw_local_irq_ssm(flags); + __arch_local_irq_stosm(0x03); } -static inline int raw_irqs_disabled_flags(unsigned long flags) +static inline void arch_local_irq_restore(unsigned long flags) +{ + __arch_local_irq_ssm(flags); +} + +static inline bool arch_irqs_disabled_flags(unsigned long flags) { return !(flags & (3UL << (BITS_PER_LONG - 8))); } -/* For spinlocks etc */ -#define raw_local_irq_save(x) ((x) = raw_local_irq_disable()) +static inline bool arch_irqs_disabled(void) +{ + return arch_irqs_disabled_flags(arch_local_save_flags()); +} #endif /* __ASM_IRQFLAGS_H */ diff --git a/arch/s390/include/asm/system.h b/arch/s390/include/asm/system.h index cef66210c84..8e8a50eeed9 100644 --- a/arch/s390/include/asm/system.h +++ b/arch/s390/include/asm/system.h @@ -399,7 +399,7 @@ static inline unsigned long __cmpxchg_local(volatile void *ptr, static inline void __set_psw_mask(unsigned long mask) { - __load_psw_mask(mask | (__raw_local_irq_stosm(0x00) & ~(-1UL >> 8))); + __load_psw_mask(mask | (arch_local_save_flags() & ~(-1UL >> 8))); } #define local_mcck_enable() __set_psw_mask(psw_kernel_bits) diff --git a/arch/s390/kernel/mem_detect.c b/arch/s390/kernel/mem_detect.c index 559af0d0787..0fbe4e32f7b 100644 --- a/arch/s390/kernel/mem_detect.c +++ b/arch/s390/kernel/mem_detect.c @@ -54,11 +54,11 @@ void detect_memory_layout(struct mem_chunk chunk[]) * right thing and we don't get scheduled away with low address * protection disabled. */ - flags = __raw_local_irq_stnsm(0xf8); + flags = __arch_local_irq_stnsm(0xf8); __ctl_store(cr0, 0, 0); __ctl_clear_bit(0, 28); find_memory_chunks(chunk); __ctl_load(cr0, 0, 0); - __raw_local_irq_ssm(flags); + arch_local_irq_restore(flags); } EXPORT_SYMBOL(detect_memory_layout); diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c index 30eb6d02ddb..94b8ba2ec85 100644 --- a/arch/s390/mm/init.c +++ b/arch/s390/mm/init.c @@ -50,7 +50,6 @@ EXPORT_SYMBOL(empty_zero_page); */ void __init paging_init(void) { - static const int ssm_mask = 0x04000000L; unsigned long max_zone_pfns[MAX_NR_ZONES]; unsigned long pgd_type; @@ -72,7 +71,7 @@ void __init paging_init(void) __ctl_load(S390_lowcore.kernel_asce, 1, 1); __ctl_load(S390_lowcore.kernel_asce, 7, 7); __ctl_load(S390_lowcore.kernel_asce, 13, 13); - __raw_local_irq_ssm(ssm_mask); + arch_local_irq_restore(4UL << (BITS_PER_LONG - 8)); atomic_set(&init_mm.context.attach_count, 1); diff --git a/arch/s390/mm/maccess.c b/arch/s390/mm/maccess.c index a8c2af8c650..71a4b0d34be 100644 --- a/arch/s390/mm/maccess.c +++ b/arch/s390/mm/maccess.c @@ -71,7 +71,7 @@ int memcpy_real(void *dest, void *src, size_t count) if (!count) return 0; - flags = __raw_local_irq_stnsm(0xf8UL); + flags = __arch_local_irq_stnsm(0xf8UL); asm volatile ( "0: mvcle %1,%2,0x0\n" "1: jo 0b\n" @@ -82,6 +82,6 @@ int memcpy_real(void *dest, void *src, size_t count) "+d" (_len2), "=m" (*((long *) dest)) : "m" (*((long *) src)) : "cc", "memory"); - __raw_local_irq_ssm(flags); + arch_local_irq_restore(flags); return rc; } diff --git a/arch/score/include/asm/irqflags.h b/arch/score/include/asm/irqflags.h index 690a6cae729..5c7563891e2 100644 --- a/arch/score/include/asm/irqflags.h +++ b/arch/score/include/asm/irqflags.h @@ -3,107 +3,118 @@ #ifndef __ASSEMBLY__ -#define raw_local_irq_save(x) \ -{ \ - __asm__ __volatile__( \ - "mfcr r8, cr0;" \ - "li r9, 0xfffffffe;" \ - "nop;" \ - "mv %0, r8;" \ - "and r8, r8, r9;" \ - "mtcr r8, cr0;" \ - "nop;" \ - "nop;" \ - "nop;" \ - "nop;" \ - "nop;" \ - : "=r" (x) \ - : \ - : "r8", "r9" \ - ); \ +#include + +static inline unsigned long arch_local_save_flags(void) +{ + unsigned long flags; + + asm volatile( + " mfcr r8, cr0 \n" + " nop \n" + " nop \n" + " mv %0, r8 \n" + " nop \n" + " nop \n" + " nop \n" + " nop \n" + " nop \n" + " ldi r9, 0x1 \n" + " and %0, %0, r9 \n" + : "=r" (flags) + : + : "r8", "r9"); + return flags; } -#define raw_local_irq_restore(x) \ -{ \ - __asm__ __volatile__( \ - "mfcr r8, cr0;" \ - "ldi r9, 0x1;" \ - "and %0, %0, r9;" \ - "or r8, r8, %0;" \ - "mtcr r8, cr0;" \ - "nop;" \ - "nop;" \ - "nop;" \ - "nop;" \ - "nop;" \ - : \ - : "r"(x) \ - : "r8", "r9" \ - ); \ +static inline unsigned long arch_local_irq_save(void) +{ + unsigned long flags + + asm volatile( + " mfcr r8, cr0 \n" + " li r9, 0xfffffffe \n" + " nop \n" + " mv %0, r8 \n" + " and r8, r8, r9 \n" + " mtcr r8, cr0 \n" + " nop \n" + " nop \n" + " nop \n" + " nop \n" + " nop \n" + : "=r" (flags) + : + : "r8", "r9", "memory"); + + return flags; } -#define raw_local_irq_enable(void) \ -{ \ - __asm__ __volatile__( \ - "mfcr\tr8,cr0;" \ - "nop;" \ - "nop;" \ - "ori\tr8,0x1;" \ - "mtcr\tr8,cr0;" \ - "nop;" \ - "nop;" \ - "nop;" \ - "nop;" \ - "nop;" \ - : \ - : \ - : "r8"); \ +static inline void arch_local_irq_restore(unsigned long flags) +{ + asm volatile( + " mfcr r8, cr0 \n" + " ldi r9, 0x1 \n" + " and %0, %0, r9 \n" + " or r8, r8, %0 \n" + " mtcr r8, cr0 \n" + " nop \n" + " nop \n" + " nop \n" + " nop \n" + " nop \n" + : + : "r"(flags) + : "r8", "r9", "memory"); } -#define raw_local_irq_disable(void) \ -{ \ - __asm__ __volatile__( \ - "mfcr\tr8,cr0;" \ - "nop;" \ - "nop;" \ - "srli\tr8,r8,1;" \ - "slli\tr8,r8,1;" \ - "mtcr\tr8,cr0;" \ - "nop;" \ - "nop;" \ - "nop;" \ - "nop;" \ - "nop;" \ - : \ - : \ - : "r8"); \ +static inline void arch_local_irq_enable(void) +{ + asm volatile( + " mfcr r8,cr0 \n" + " nop \n" + " nop \n" + " ori r8,0x1 \n" + " mtcr r8,cr0 \n" + " nop \n" + " nop \n" + " nop \n" + " nop \n" + " nop \n" + : + : + : "r8", "memory"); } -#define raw_local_save_flags(x) \ -{ \ - __asm__ __volatile__( \ - "mfcr r8, cr0;" \ - "nop;" \ - "nop;" \ - "mv %0, r8;" \ - "nop;" \ - "nop;" \ - "nop;" \ - "nop;" \ - "nop;" \ - "ldi r9, 0x1;" \ - "and %0, %0, r9;" \ - : "=r" (x) \ - : \ - : "r8", "r9" \ - ); \ +static inline void arch_local_irq_disable(void) +{ + asm volatile( + " mfcr r8,cr0 \n" + " nop \n" + " nop \n" + " srli r8,r8,1 \n" + " slli r8,r8,1 \n" + " mtcr r8,cr0 \n" + " nop \n" + " nop \n" + " nop \n" + " nop \n" + " nop \n" + : + : + : "r8", "memory"); } -static inline int raw_irqs_disabled_flags(unsigned long flags) +static inline bool arch_irqs_disabled_flags(unsigned long flags) { return !(flags & 1); } -#endif +static inline bool arch_irqs_disabled(void) +{ + return arch_irqs_disabled_flags(arch_local_save_flags()); +} + +#endif /* __ASSEMBLY__ */ #endif /* _ASM_SCORE_IRQFLAGS_H */ diff --git a/arch/sh/include/asm/irqflags.h b/arch/sh/include/asm/irqflags.h index a741153b41c..43b7608606c 100644 --- a/arch/sh/include/asm/irqflags.h +++ b/arch/sh/include/asm/irqflags.h @@ -1,8 +1,8 @@ #ifndef __ASM_SH_IRQFLAGS_H #define __ASM_SH_IRQFLAGS_H -#define RAW_IRQ_DISABLED 0xf0 -#define RAW_IRQ_ENABLED 0x00 +#define ARCH_IRQ_DISABLED 0xf0 +#define ARCH_IRQ_ENABLED 0x00 #include diff --git a/arch/sh/kernel/irq_32.c b/arch/sh/kernel/irq_32.c index e33ab15831f..e5a755be912 100644 --- a/arch/sh/kernel/irq_32.c +++ b/arch/sh/kernel/irq_32.c @@ -10,11 +10,11 @@ #include #include -void notrace raw_local_irq_restore(unsigned long flags) +void notrace arch_local_irq_restore(unsigned long flags) { unsigned long __dummy0, __dummy1; - if (flags == RAW_IRQ_DISABLED) { + if (flags == ARCH_IRQ_DISABLED) { __asm__ __volatile__ ( "stc sr, %0\n\t" "or #0xf0, %0\n\t" @@ -33,14 +33,14 @@ void notrace raw_local_irq_restore(unsigned long flags) #endif "ldc %0, sr\n\t" : "=&r" (__dummy0), "=r" (__dummy1) - : "1" (~RAW_IRQ_DISABLED) + : "1" (~ARCH_IRQ_DISABLED) : "memory" ); } } -EXPORT_SYMBOL(raw_local_irq_restore); +EXPORT_SYMBOL(arch_local_irq_restore); -unsigned long notrace __raw_local_save_flags(void) +unsigned long notrace arch_local_save_flags(void) { unsigned long flags; @@ -54,4 +54,4 @@ unsigned long notrace __raw_local_save_flags(void) return flags; } -EXPORT_SYMBOL(__raw_local_save_flags); +EXPORT_SYMBOL(arch_local_save_flags); diff --git a/arch/sparc/include/asm/irqflags_32.h b/arch/sparc/include/asm/irqflags_32.h index 0fca9d97d44..d4d0711de0f 100644 --- a/arch/sparc/include/asm/irqflags_32.h +++ b/arch/sparc/include/asm/irqflags_32.h @@ -5,33 +5,40 @@ * * This file gets included from lowlevel asm headers too, to provide * wrapped versions of the local_irq_*() APIs, based on the - * raw_local_irq_*() functions from the lowlevel headers. + * arch_local_irq_*() functions from the lowlevel headers. */ #ifndef _ASM_IRQFLAGS_H #define _ASM_IRQFLAGS_H #ifndef __ASSEMBLY__ -extern void raw_local_irq_restore(unsigned long); -extern unsigned long __raw_local_irq_save(void); -extern void raw_local_irq_enable(void); +#include -static inline unsigned long getipl(void) +extern void arch_local_irq_restore(unsigned long); +extern unsigned long arch_local_irq_save(void); +extern void arch_local_irq_enable(void); + +static inline unsigned long arch_local_save_flags(void) { - unsigned long retval; + unsigned long flags; + + asm volatile("rd %%psr, %0" : "=r" (flags)); + return flags; +} - __asm__ __volatile__("rd %%psr, %0" : "=r" (retval)); - return retval; +static inline void arch_local_irq_disable(void) +{ + arch_local_irq_save(); } -#define raw_local_save_flags(flags) ((flags) = getipl()) -#define raw_local_irq_save(flags) ((flags) = __raw_local_irq_save()) -#define raw_local_irq_disable() ((void) __raw_local_irq_save()) -#define raw_irqs_disabled() ((getipl() & PSR_PIL) != 0) +static inline bool arch_irqs_disabled_flags(unsigned long flags) +{ + return (flags & PSR_PIL) != 0; +} -static inline int raw_irqs_disabled_flags(unsigned long flags) +static inline bool arch_irqs_disabled(void) { - return ((flags & PSR_PIL) != 0); + return arch_irqs_disabled_flags(arch_local_save_flags()); } #endif /* (__ASSEMBLY__) */ diff --git a/arch/sparc/include/asm/irqflags_64.h b/arch/sparc/include/asm/irqflags_64.h index bfa1ea45b4c..aab969c82c2 100644 --- a/arch/sparc/include/asm/irqflags_64.h +++ b/arch/sparc/include/asm/irqflags_64.h @@ -5,7 +5,7 @@ * * This file gets included from lowlevel asm headers too, to provide * wrapped versions of the local_irq_*() APIs, based on the - * raw_local_irq_*() functions from the lowlevel headers. + * arch_local_irq_*() functions from the lowlevel headers. */ #ifndef _ASM_IRQFLAGS_H #define _ASM_IRQFLAGS_H @@ -14,7 +14,7 @@ #ifndef __ASSEMBLY__ -static inline unsigned long __raw_local_save_flags(void) +static inline unsigned long arch_local_save_flags(void) { unsigned long flags; @@ -26,10 +26,7 @@ static inline unsigned long __raw_local_save_flags(void) return flags; } -#define raw_local_save_flags(flags) \ - do { (flags) = __raw_local_save_flags(); } while (0) - -static inline void raw_local_irq_restore(unsigned long flags) +static inline void arch_local_irq_restore(unsigned long flags) { __asm__ __volatile__( "wrpr %0, %%pil" @@ -39,7 +36,7 @@ static inline void raw_local_irq_restore(unsigned long flags) ); } -static inline void raw_local_irq_disable(void) +static inline void arch_local_irq_disable(void) { __asm__ __volatile__( "wrpr %0, %%pil" @@ -49,7 +46,7 @@ static inline void raw_local_irq_disable(void) ); } -static inline void raw_local_irq_enable(void) +static inline void arch_local_irq_enable(void) { __asm__ __volatile__( "wrpr 0, %%pil" @@ -59,22 +56,17 @@ static inline void raw_local_irq_enable(void) ); } -static inline int raw_irqs_disabled_flags(unsigned long flags) +static inline int arch_irqs_disabled_flags(unsigned long flags) { return (flags > 0); } -static inline int raw_irqs_disabled(void) +static inline int arch_irqs_disabled(void) { - unsigned long flags = __raw_local_save_flags(); - - return raw_irqs_disabled_flags(flags); + return arch_irqs_disabled_flags(arch_local_save_flags()); } -/* - * For spinlocks, etc: - */ -static inline unsigned long __raw_local_irq_save(void) +static inline unsigned long arch_local_irq_save(void) { unsigned long flags, tmp; @@ -100,9 +92,6 @@ static inline unsigned long __raw_local_irq_save(void) return flags; } -#define raw_local_irq_save(flags) \ - do { (flags) = __raw_local_irq_save(); } while (0) - #endif /* (__ASSEMBLY__) */ #endif /* !(_ASM_IRQFLAGS_H) */ diff --git a/arch/sparc/kernel/irq_32.c b/arch/sparc/kernel/irq_32.c index e1af4372832..0116d8d10de 100644 --- a/arch/sparc/kernel/irq_32.c +++ b/arch/sparc/kernel/irq_32.c @@ -57,7 +57,7 @@ #define SMP_NOP2 #define SMP_NOP3 #endif /* SMP */ -unsigned long __raw_local_irq_save(void) +unsigned long arch_local_irq_save(void) { unsigned long retval; unsigned long tmp; @@ -74,8 +74,9 @@ unsigned long __raw_local_irq_save(void) return retval; } +EXPORT_SYMBOL(arch_local_irq_save); -void raw_local_irq_enable(void) +void arch_local_irq_enable(void) { unsigned long tmp; @@ -89,8 +90,9 @@ void raw_local_irq_enable(void) : "i" (PSR_PIL) : "memory"); } +EXPORT_SYMBOL(arch_local_irq_enable); -void raw_local_irq_restore(unsigned long old_psr) +void arch_local_irq_restore(unsigned long old_psr) { unsigned long tmp; @@ -105,10 +107,7 @@ void raw_local_irq_restore(unsigned long old_psr) : "i" (PSR_PIL), "r" (old_psr) : "memory"); } - -EXPORT_SYMBOL(__raw_local_irq_save); -EXPORT_SYMBOL(raw_local_irq_enable); -EXPORT_SYMBOL(raw_local_irq_restore); +EXPORT_SYMBOL(arch_local_irq_restore); /* * Dave Redman (djhr@tadpole.co.uk) diff --git a/arch/sparc/prom/p1275.c b/arch/sparc/prom/p1275.c index fa6e4e219b9..d9850c2b9bf 100644 --- a/arch/sparc/prom/p1275.c +++ b/arch/sparc/prom/p1275.c @@ -39,7 +39,7 @@ void p1275_cmd_direct(unsigned long *args) unsigned long flags; raw_local_save_flags(flags); - raw_local_irq_restore(PIL_NMI); + raw_local_irq_restore((unsigned long)PIL_NMI); raw_spin_lock(&prom_entry_lock); prom_world(1); diff --git a/arch/tile/include/asm/irqflags.h b/arch/tile/include/asm/irqflags.h index 45cf67c2f28..a11d4837ee4 100644 --- a/arch/tile/include/asm/irqflags.h +++ b/arch/tile/include/asm/irqflags.h @@ -103,55 +103,57 @@ DECLARE_PER_CPU(unsigned long long, interrupts_enabled_mask); #define INITIAL_INTERRUPTS_ENABLED INT_MASK(INT_MEM_ERROR) /* Disable interrupts. */ -#define raw_local_irq_disable() \ +#define arch_local_irq_disable() \ interrupt_mask_set_mask(LINUX_MASKABLE_INTERRUPTS) /* Disable all interrupts, including NMIs. */ -#define raw_local_irq_disable_all() \ +#define arch_local_irq_disable_all() \ interrupt_mask_set_mask(-1UL) /* Re-enable all maskable interrupts. */ -#define raw_local_irq_enable() \ +#define arch_local_irq_enable() \ interrupt_mask_reset_mask(__get_cpu_var(interrupts_enabled_mask)) /* Disable or enable interrupts based on flag argument. */ -#define raw_local_irq_restore(disabled) do { \ +#define arch_local_irq_restore(disabled) do { \ if (disabled) \ - raw_local_irq_disable(); \ + arch_local_irq_disable(); \ else \ - raw_local_irq_enable(); \ + arch_local_irq_enable(); \ } while (0) /* Return true if "flags" argument means interrupts are disabled. */ -#define raw_irqs_disabled_flags(flags) ((flags) != 0) +#define arch_irqs_disabled_flags(flags) ((flags) != 0) /* Return true if interrupts are currently disabled. */ -#define raw_irqs_disabled() interrupt_mask_check(INT_MEM_ERROR) +#define arch_irqs_disabled() interrupt_mask_check(INT_MEM_ERROR) /* Save whether interrupts are currently disabled. */ -#define raw_local_save_flags(flags) ((flags) = raw_irqs_disabled()) +#define arch_local_save_flags() arch_irqs_disabled() /* Save whether interrupts are currently disabled, then disable them. */ -#define raw_local_irq_save(flags) \ - do { raw_local_save_flags(flags); raw_local_irq_disable(); } while (0) +#define arch_local_irq_save() ({ \ + unsigned long __flags = arch_local_save_flags(); \ + arch_local_irq_disable(); \ + __flags; }) /* Prevent the given interrupt from being enabled next time we enable irqs. */ -#define raw_local_irq_mask(interrupt) \ +#define arch_local_irq_mask(interrupt) \ (__get_cpu_var(interrupts_enabled_mask) &= ~INT_MASK(interrupt)) /* Prevent the given interrupt from being enabled immediately. */ -#define raw_local_irq_mask_now(interrupt) do { \ - raw_local_irq_mask(interrupt); \ +#define arch_local_irq_mask_now(interrupt) do { \ + arch_local_irq_mask(interrupt); \ interrupt_mask_set(interrupt); \ } while (0) /* Allow the given interrupt to be enabled next time we enable irqs. */ -#define raw_local_irq_unmask(interrupt) \ +#define arch_local_irq_unmask(interrupt) \ (__get_cpu_var(interrupts_enabled_mask) |= INT_MASK(interrupt)) /* Allow the given interrupt to be enabled immediately, if !irqs_disabled. */ -#define raw_local_irq_unmask_now(interrupt) do { \ - raw_local_irq_unmask(interrupt); \ +#define arch_local_irq_unmask_now(interrupt) do { \ + arch_local_irq_unmask(interrupt); \ if (!irqs_disabled()) \ interrupt_mask_reset(interrupt); \ } while (0) diff --git a/arch/x86/include/asm/irqflags.h b/arch/x86/include/asm/irqflags.h index 9e2b952f810..5745ce8bf10 100644 --- a/arch/x86/include/asm/irqflags.h +++ b/arch/x86/include/asm/irqflags.h @@ -61,22 +61,22 @@ static inline void native_halt(void) #else #ifndef __ASSEMBLY__ -static inline unsigned long __raw_local_save_flags(void) +static inline unsigned long arch_local_save_flags(void) { return native_save_fl(); } -static inline void raw_local_irq_restore(unsigned long flags) +static inline void arch_local_irq_restore(unsigned long flags) { native_restore_fl(flags); } -static inline void raw_local_irq_disable(void) +static inline void arch_local_irq_disable(void) { native_irq_disable(); } -static inline void raw_local_irq_enable(void) +static inline void arch_local_irq_enable(void) { native_irq_enable(); } @@ -85,7 +85,7 @@ static inline void raw_local_irq_enable(void) * Used in the idle loop; sti takes one instruction cycle * to complete: */ -static inline void raw_safe_halt(void) +static inline void arch_safe_halt(void) { native_safe_halt(); } @@ -102,12 +102,10 @@ static inline void halt(void) /* * For spinlocks, etc: */ -static inline unsigned long __raw_local_irq_save(void) +static inline unsigned long arch_local_irq_save(void) { - unsigned long flags = __raw_local_save_flags(); - - raw_local_irq_disable(); - + unsigned long flags = arch_local_save_flags(); + arch_local_irq_disable(); return flags; } #else @@ -153,22 +151,16 @@ static inline unsigned long __raw_local_irq_save(void) #endif /* CONFIG_PARAVIRT */ #ifndef __ASSEMBLY__ -#define raw_local_save_flags(flags) \ - do { (flags) = __raw_local_save_flags(); } while (0) - -#define raw_local_irq_save(flags) \ - do { (flags) = __raw_local_irq_save(); } while (0) - -static inline int raw_irqs_disabled_flags(unsigned long flags) +static inline int arch_irqs_disabled_flags(unsigned long flags) { return !(flags & X86_EFLAGS_IF); } -static inline int raw_irqs_disabled(void) +static inline int arch_irqs_disabled(void) { - unsigned long flags = __raw_local_save_flags(); + unsigned long flags = arch_local_save_flags(); - return raw_irqs_disabled_flags(flags); + return arch_irqs_disabled_flags(flags); } #else diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h index 5653f43d90e..499954c530d 100644 --- a/arch/x86/include/asm/paravirt.h +++ b/arch/x86/include/asm/paravirt.h @@ -105,7 +105,7 @@ static inline void write_cr8(unsigned long x) } #endif -static inline void raw_safe_halt(void) +static inline void arch_safe_halt(void) { PVOP_VCALL0(pv_irq_ops.safe_halt); } @@ -829,32 +829,32 @@ static __always_inline void arch_spin_unlock(struct arch_spinlock *lock) #define __PV_IS_CALLEE_SAVE(func) \ ((struct paravirt_callee_save) { func }) -static inline unsigned long __raw_local_save_flags(void) +static inline unsigned long arch_local_save_flags(void) { return PVOP_CALLEE0(unsigned long, pv_irq_ops.save_fl); } -static inline void raw_local_irq_restore(unsigned long f) +static inline void arch_local_irq_restore(unsigned long f) { PVOP_VCALLEE1(pv_irq_ops.restore_fl, f); } -static inline void raw_local_irq_disable(void) +static inline void arch_local_irq_disable(void) { PVOP_VCALLEE0(pv_irq_ops.irq_disable); } -static inline void raw_local_irq_enable(void) +static inline void arch_local_irq_enable(void) { PVOP_VCALLEE0(pv_irq_ops.irq_enable); } -static inline unsigned long __raw_local_irq_save(void) +static inline unsigned long arch_local_irq_save(void) { unsigned long f; - f = __raw_local_save_flags(); - raw_local_irq_disable(); + f = arch_local_save_flags(); + arch_local_irq_disable(); return f; } diff --git a/arch/x86/xen/spinlock.c b/arch/x86/xen/spinlock.c index e0500646585..23e061b9327 100644 --- a/arch/x86/xen/spinlock.c +++ b/arch/x86/xen/spinlock.c @@ -224,7 +224,7 @@ static noinline int xen_spin_lock_slow(struct arch_spinlock *lock, bool irq_enab goto out; } - flags = __raw_local_save_flags(); + flags = arch_local_save_flags(); if (irq_enable) { ADD_STATS(taken_slow_irqenable, 1); raw_local_irq_enable(); diff --git a/arch/xtensa/include/asm/irqflags.h b/arch/xtensa/include/asm/irqflags.h new file mode 100644 index 00000000000..dae9a8bdcb1 --- /dev/null +++ b/arch/xtensa/include/asm/irqflags.h @@ -0,0 +1,58 @@ +/* + * Xtensa IRQ flags handling functions + * + * 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) 2001 - 2005 Tensilica Inc. + */ + +#ifndef _XTENSA_IRQFLAGS_H +#define _XTENSA_IRQFLAGS_H + +#include + +static inline unsigned long arch_local_save_flags(void) +{ + unsigned long flags; + asm volatile("rsr %0,"__stringify(PS) : "=a" (flags)); + return flags; +} + +static inline unsigned long arch_local_irq_save(void) +{ + unsigned long flags; + asm volatile("rsil %0, "__stringify(LOCKLEVEL) + : "=a" (flags) :: "memory"); + return flags; +} + +static inline void arch_local_irq_disable(void) +{ + arch_local_irq_save(); +} + +static inline void arch_local_irq_enable(void) +{ + unsigned long flags; + asm volatile("rsil %0, 0" : "=a" (flags) :: "memory"); +} + +static inline void arch_local_irq_restore(unsigned long flags) +{ + asm volatile("wsr %0, "__stringify(PS)" ; rsync" + :: "a" (flags) : "memory"); +} + +static inline bool arch_irqs_disabled_flags(unsigned long flags) +{ + return (flags & 0xf) != 0; +} + +static inline bool arch_irqs_disabled(void) +{ + return arch_irqs_disabled_flags(arch_local_save_flags()); +} + +#endif /* _XTENSA_IRQFLAGS_H */ diff --git a/arch/xtensa/include/asm/system.h b/arch/xtensa/include/asm/system.h index 62b1e8f3c13..1e7e09ab6cd 100644 --- a/arch/xtensa/include/asm/system.h +++ b/arch/xtensa/include/asm/system.h @@ -12,41 +12,10 @@ #define _XTENSA_SYSTEM_H #include +#include #include -/* interrupt control */ - -#define local_save_flags(x) \ - __asm__ __volatile__ ("rsr %0,"__stringify(PS) : "=a" (x)); -#define local_irq_restore(x) do { \ - __asm__ __volatile__ ("wsr %0, "__stringify(PS)" ; rsync" \ - :: "a" (x) : "memory"); } while(0); -#define local_irq_save(x) do { \ - __asm__ __volatile__ ("rsil %0, "__stringify(LOCKLEVEL) \ - : "=a" (x) :: "memory");} while(0); - -static inline void local_irq_disable(void) -{ - unsigned long flags; - __asm__ __volatile__ ("rsil %0, "__stringify(LOCKLEVEL) - : "=a" (flags) :: "memory"); -} -static inline void local_irq_enable(void) -{ - unsigned long flags; - __asm__ __volatile__ ("rsil %0, 0" : "=a" (flags) :: "memory"); - -} - -static inline int irqs_disabled(void) -{ - unsigned long flags; - local_save_flags(flags); - return flags & 0xf; -} - - #define smp_read_barrier_depends() do { } while(0) #define read_barrier_depends() do { } while(0) diff --git a/drivers/s390/char/sclp.c b/drivers/s390/char/sclp.c index f6d72e1f2a3..5707a80b96b 100644 --- a/drivers/s390/char/sclp.c +++ b/drivers/s390/char/sclp.c @@ -468,7 +468,7 @@ sclp_sync_wait(void) cr0_sync &= 0xffff00a0; cr0_sync |= 0x00000200; __ctl_load(cr0_sync, 0, 0); - __raw_local_irq_stosm(0x01); + __arch_local_irq_stosm(0x01); /* Loop until driver state indicates finished request */ while (sclp_running_state != sclp_running_state_idle) { /* Check for expired request timer */ diff --git a/include/asm-generic/atomic.h b/include/asm-generic/atomic.h index e53347fbf1d..fd57b8477fa 100644 --- a/include/asm-generic/atomic.h +++ b/include/asm-generic/atomic.h @@ -43,6 +43,7 @@ */ #define atomic_set(v, i) (((v)->counter) = (i)) +#include #include /** @@ -57,7 +58,7 @@ static inline int atomic_add_return(int i, atomic_t *v) unsigned long flags; int temp; - raw_local_irq_save(flags); /* Don't trace it in a irqsoff handler */ + raw_local_irq_save(flags); /* Don't trace it in an irqsoff handler */ temp = v->counter; temp += i; v->counter = temp; @@ -78,7 +79,7 @@ static inline int atomic_sub_return(int i, atomic_t *v) unsigned long flags; int temp; - raw_local_irq_save(flags); /* Don't trace it in a irqsoff handler */ + raw_local_irq_save(flags); /* Don't trace it in an irqsoff handler */ temp = v->counter; temp -= i; v->counter = temp; diff --git a/include/asm-generic/cmpxchg-local.h b/include/asm-generic/cmpxchg-local.h index b2ba2fc8829..2533fddd34a 100644 --- a/include/asm-generic/cmpxchg-local.h +++ b/include/asm-generic/cmpxchg-local.h @@ -2,6 +2,7 @@ #define __ASM_GENERIC_CMPXCHG_LOCAL_H #include +#include extern unsigned long wrong_size_cmpxchg(volatile void *ptr); diff --git a/include/asm-generic/hardirq.h b/include/asm-generic/hardirq.h index 62f59080e5c..c0771aa248c 100644 --- a/include/asm-generic/hardirq.h +++ b/include/asm-generic/hardirq.h @@ -3,7 +3,6 @@ #include #include -#include typedef struct { unsigned int __softirq_pending; diff --git a/include/asm-generic/irqflags.h b/include/asm-generic/irqflags.h index 9aebf618275..1f40d0024cf 100644 --- a/include/asm-generic/irqflags.h +++ b/include/asm-generic/irqflags.h @@ -5,68 +5,62 @@ * All architectures should implement at least the first two functions, * usually inline assembly will be the best way. */ -#ifndef RAW_IRQ_DISABLED -#define RAW_IRQ_DISABLED 0 -#define RAW_IRQ_ENABLED 1 +#ifndef ARCH_IRQ_DISABLED +#define ARCH_IRQ_DISABLED 0 +#define ARCH_IRQ_ENABLED 1 #endif /* read interrupt enabled status */ -#ifndef __raw_local_save_flags -unsigned long __raw_local_save_flags(void); +#ifndef arch_local_save_flags +unsigned long arch_local_save_flags(void); #endif /* set interrupt enabled status */ -#ifndef raw_local_irq_restore -void raw_local_irq_restore(unsigned long flags); +#ifndef arch_local_irq_restore +void arch_local_irq_restore(unsigned long flags); #endif /* get status and disable interrupts */ -#ifndef __raw_local_irq_save -static inline unsigned long __raw_local_irq_save(void) +#ifndef arch_local_irq_save +static inline unsigned long arch_local_irq_save(void) { unsigned long flags; - flags = __raw_local_save_flags(); - raw_local_irq_restore(RAW_IRQ_DISABLED); + flags = arch_local_save_flags(); + arch_local_irq_restore(ARCH_IRQ_DISABLED); return flags; } #endif /* test flags */ -#ifndef raw_irqs_disabled_flags -static inline int raw_irqs_disabled_flags(unsigned long flags) +#ifndef arch_irqs_disabled_flags +static inline int arch_irqs_disabled_flags(unsigned long flags) { - return flags == RAW_IRQ_DISABLED; + return flags == ARCH_IRQ_DISABLED; } #endif /* unconditionally enable interrupts */ -#ifndef raw_local_irq_enable -static inline void raw_local_irq_enable(void) +#ifndef arch_local_irq_enable +static inline void arch_local_irq_enable(void) { - raw_local_irq_restore(RAW_IRQ_ENABLED); + arch_local_irq_restore(ARCH_IRQ_ENABLED); } #endif /* unconditionally disable interrupts */ -#ifndef raw_local_irq_disable -static inline void raw_local_irq_disable(void) +#ifndef arch_local_irq_disable +static inline void arch_local_irq_disable(void) { - raw_local_irq_restore(RAW_IRQ_DISABLED); + arch_local_irq_restore(ARCH_IRQ_DISABLED); } #endif /* test hardware interrupt enable bit */ -#ifndef raw_irqs_disabled -static inline int raw_irqs_disabled(void) +#ifndef arch_irqs_disabled +static inline int arch_irqs_disabled(void) { - return raw_irqs_disabled_flags(__raw_local_save_flags()); + return arch_irqs_disabled_flags(arch_local_save_flags()); } #endif -#define raw_local_save_flags(flags) \ - do { (flags) = __raw_local_save_flags(); } while (0) - -#define raw_local_irq_save(flags) \ - do { (flags) = __raw_local_irq_save(); } while (0) - #endif /* __ASM_GENERIC_IRQFLAGS_H */ diff --git a/include/linux/irqflags.h b/include/linux/irqflags.h index 006bf45eae3..d176d658fe2 100644 --- a/include/linux/irqflags.h +++ b/include/linux/irqflags.h @@ -12,6 +12,7 @@ #define _LINUX_TRACE_IRQFLAGS_H #include +#include #ifdef CONFIG_TRACE_IRQFLAGS extern void trace_softirqs_on(unsigned long ip); @@ -52,17 +53,45 @@ # define start_critical_timings() do { } while (0) #endif -#ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT - -#include +/* + * Wrap the arch provided IRQ routines to provide appropriate checks. + */ +#define raw_local_irq_disable() arch_local_irq_disable() +#define raw_local_irq_enable() arch_local_irq_enable() +#define raw_local_irq_save(flags) \ + do { \ + typecheck(unsigned long, flags); \ + flags = arch_local_irq_save(); \ + } while (0) +#define raw_local_irq_restore(flags) \ + do { \ + typecheck(unsigned long, flags); \ + arch_local_irq_restore(flags); \ + } while (0) +#define raw_local_save_flags(flags) \ + do { \ + typecheck(unsigned long, flags); \ + flags = arch_local_save_flags(); \ + } while (0) +#define raw_irqs_disabled_flags(flags) \ + ({ \ + typecheck(unsigned long, flags); \ + arch_irqs_disabled_flags(flags); \ + }) +#define raw_irqs_disabled() (arch_irqs_disabled()) +#define raw_safe_halt() arch_safe_halt() +/* + * The local_irq_*() APIs are equal to the raw_local_irq*() + * if !TRACE_IRQFLAGS. + */ +#ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT #define local_irq_enable() \ do { trace_hardirqs_on(); raw_local_irq_enable(); } while (0) #define local_irq_disable() \ do { raw_local_irq_disable(); trace_hardirqs_off(); } while (0) #define local_irq_save(flags) \ do { \ - typecheck(unsigned long, flags); \ raw_local_irq_save(flags); \ trace_hardirqs_off(); \ } while (0) @@ -70,7 +99,6 @@ #define local_irq_restore(flags) \ do { \ - typecheck(unsigned long, flags); \ if (raw_irqs_disabled_flags(flags)) { \ raw_local_irq_restore(flags); \ trace_hardirqs_off(); \ @@ -79,51 +107,44 @@ raw_local_irq_restore(flags); \ } \ } while (0) -#else /* !CONFIG_TRACE_IRQFLAGS_SUPPORT */ -/* - * The local_irq_*() APIs are equal to the raw_local_irq*() - * if !TRACE_IRQFLAGS. - */ -# define raw_local_irq_disable() local_irq_disable() -# define raw_local_irq_enable() local_irq_enable() -# define raw_local_irq_save(flags) \ - do { \ - typecheck(unsigned long, flags); \ - local_irq_save(flags); \ - } while (0) -# define raw_local_irq_restore(flags) \ +#define local_save_flags(flags) \ do { \ - typecheck(unsigned long, flags); \ - local_irq_restore(flags); \ + raw_local_save_flags(flags); \ } while (0) -#endif /* CONFIG_TRACE_IRQFLAGS_SUPPORT */ -#ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT -#define safe_halt() \ - do { \ - trace_hardirqs_on(); \ - raw_safe_halt(); \ - } while (0) +#define irqs_disabled_flags(flags) \ + ({ \ + raw_irqs_disabled_flags(flags); \ + }) -#define local_save_flags(flags) \ - do { \ - typecheck(unsigned long, flags); \ - raw_local_save_flags(flags); \ +#define irqs_disabled() \ + ({ \ + unsigned long _flags; \ + raw_local_save_flags(_flags); \ + raw_irqs_disabled_flags(_flags); \ + }) + +#define safe_halt() \ + do { \ + trace_hardirqs_on(); \ + raw_safe_halt(); \ } while (0) -#define irqs_disabled() \ -({ \ - unsigned long _flags; \ - \ - raw_local_save_flags(_flags); \ - raw_irqs_disabled_flags(_flags); \ -}) -#define irqs_disabled_flags(flags) \ -({ \ - typecheck(unsigned long, flags); \ - raw_irqs_disabled_flags(flags); \ -}) +#else /* !CONFIG_TRACE_IRQFLAGS_SUPPORT */ + +#define local_irq_enable() do { raw_local_irq_enable(); } while (0) +#define local_irq_disable() do { raw_local_irq_disable(); } while (0) +#define local_irq_save(flags) \ + do { \ + raw_local_irq_save(flags); \ + } while (0) +#define local_irq_restore(flags) do { raw_local_irq_restore(flags); } while (0) +#define local_save_flags(flags) do { raw_local_save_flags(flags); } while (0) +#define irqs_disabled() (raw_irqs_disabled()) +#define irqs_disabled_flags(flags) (raw_irqs_disabled_flags(flags)) +#define safe_halt() do { raw_safe_halt(); } while (0) + #endif /* CONFIG_TRACE_IRQFLAGS_SUPPORT */ #endif diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index f8854655860..80e535897de 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -50,6 +50,7 @@ #include #include #include +#include #include #include #include -- cgit v1.2.3-70-g09d2 From a416e9e1dde0fbcf20cda59df284cc0dcf2aadc4 Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Wed, 29 Sep 2010 23:29:48 +0900 Subject: x86-32: Fix sparse warning for the __PHYSICAL_MASK calculation On 32-bit non-PAE system, cast to 'phys_addr_t' truncates value before subtraction. Subtracting before cast produce same result but remove following warnings from sparse: arch/x86/include/asm/pgtable_types.h:255:38: warning: cast truncates bits from constant value (100000000 becomes 0) arch/x86/include/asm/pgtable_types.h:270:38: warning: cast truncates bits from constant value (100000000 becomes 0) arch/x86/include/asm/pgtable.h:127:32: warning: cast truncates bits from constant value (100000000 becomes 0) arch/x86/include/asm/pgtable.h:132:32: warning: cast truncates bits from constant value (100000000 becomes 0) arch/x86/include/asm/pgtable.h:344:31: warning: cast truncates bits from constant value (100000000 becomes 0) 64-bit or PAE machines will not be affected by this change. Signed-off-by: Namhyung Kim LKML-Reference: <1285770588-14065-1-git-send-email-namhyung@gmail.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/page_types.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/page_types.h b/arch/x86/include/asm/page_types.h index a667f24c725..1df66211fd1 100644 --- a/arch/x86/include/asm/page_types.h +++ b/arch/x86/include/asm/page_types.h @@ -8,7 +8,7 @@ #define PAGE_SIZE (_AC(1,UL) << PAGE_SHIFT) #define PAGE_MASK (~(PAGE_SIZE-1)) -#define __PHYSICAL_MASK ((phys_addr_t)(1ULL << __PHYSICAL_MASK_SHIFT) - 1) +#define __PHYSICAL_MASK ((phys_addr_t)((1ULL << __PHYSICAL_MASK_SHIFT) - 1)) #define __VIRTUAL_MASK ((1UL << __VIRTUAL_MASK_SHIFT) - 1) /* Cast PAGE_MASK to a signed type so that it is sign-extended if -- cgit v1.2.3-70-g09d2 From 55572b293b3a5929e8c54bc91d14ae6264186bf6 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Thu, 7 Oct 2010 16:42:54 -0700 Subject: x86, mrst: A function in a header file needs to be marked "inline" A function in a header file needs to be explicitly marked "inline", or gcc will complain if it is not used. Signed-off-by: H. Peter Anvin Cc: Jacob Pan Cc: v2.6.36 LKML-Reference: <1274295685-6774-3-git-send-email-jacob.jun.pan@linux.intel.com> --- arch/x86/include/asm/mrst.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/mrst.h b/arch/x86/include/asm/mrst.h index 16350740edf..33fc2966beb 100644 --- a/arch/x86/include/asm/mrst.h +++ b/arch/x86/include/asm/mrst.h @@ -26,7 +26,7 @@ enum mrst_cpu_type { }; extern enum mrst_cpu_type __mrst_cpu_chip; -static enum mrst_cpu_type mrst_identify_cpu(void) +static inline enum mrst_cpu_type mrst_identify_cpu(void) { return __mrst_cpu_chip; } -- cgit v1.2.3-70-g09d2 From 68f4d5a00adaab33b136fce2c72d5c377b39b0b0 Mon Sep 17 00:00:00 2001 From: Zhao Yakui Date: Fri, 8 Oct 2010 09:47:33 +0800 Subject: x86, setup: Use string copy operation to optimze copy in kernel compression The kernel decompression code parses the ELF header and then copies the segment to the corresponding destination. Currently it uses slow byte-copy code. This patch makes it use the string copy operations instead. In the test the copy performance can be improved very significantly after using the string copy operation mechanism. 1. The copy time can be reduced from 150ms to 20ms on one Atom machine 2. The copy time can be reduced about 80% on another machine The time is reduced from 7ms to 1.5ms when using 32-bit kernel. The time is reduced from 10ms to 2ms when using 64-bit kernel. Signed-off-by: Zhao Yakui LKML-Reference: <1286502453-7043-1-git-send-email-yakui.zhao@intel.com> Signed-off-by: H. Peter Anvin --- arch/x86/boot/compressed/misc.c | 29 +++++++++++++++++++++++------ 1 file changed, 23 insertions(+), 6 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/boot/compressed/misc.c b/arch/x86/boot/compressed/misc.c index 8f7bef8e9ff..23f315c9f21 100644 --- a/arch/x86/boot/compressed/misc.c +++ b/arch/x86/boot/compressed/misc.c @@ -229,18 +229,35 @@ void *memset(void *s, int c, size_t n) ss[i] = c; return s; } - +#ifdef CONFIG_X86_32 void *memcpy(void *dest, const void *src, size_t n) { - int i; - const char *s = src; - char *d = dest; + int d0, d1, d2; + asm volatile( + "rep ; movsl\n\t" + "movl %4,%%ecx\n\t" + "rep ; movsb\n\t" + : "=&c" (d0), "=&D" (d1), "=&S" (d2) + : "0" (n >> 2), "g" (n & 3), "1" (dest), "2" (src) + : "memory"); - for (i = 0; i < n; i++) - d[i] = s[i]; return dest; } +#else +void *memcpy(void *dest, const void *src, size_t n) +{ + long d0, d1, d2; + asm volatile( + "rep ; movsq\n\t" + "movq %4,%%rcx\n\t" + "rep ; movsb\n\t" + : "=&c" (d0), "=&D" (d1), "=&S" (d2) + : "0" (n >> 3), "g" (n & 7), "1" (dest), "2" (src) + : "memory"); + return dest; +} +#endif static void error(char *x) { -- cgit v1.2.3-70-g09d2 From 5a47c7dae861c3ca3edf178546641909851bf715 Mon Sep 17 00:00:00 2001 From: Feng Tang Date: Mon, 13 Sep 2010 15:08:54 +0800 Subject: x86: Add two helper macros for fixed address mapping Sometimes fixmap will be used to map an physical address which is not PAGE align, so to use it we need first map it and then add the address offset to the mapped fixed address. These 2 new helpers are suggested by Ingo Molnar to make the process simpler. For a physicall address like "phys", a directly usable virtual address can be get by virt = (void *)set_fixmap_offset(fixed_idx, phys); or virt = (void *)set_fixmap_offset_nocache(fixed_idx, phys); (depends on whether the physical address is cachable or not). Signed-off-by: Feng Tang Cc: alan@linux.intel.com Cc: greg@kroah.com Cc: x86@kernel.org LKML-Reference: <1284361736-23011-3-git-send-email-feng.tang@intel.com> Signed-off-by: Ingo Molnar --- arch/x86/include/asm/fixmap.h | 15 +++++++++++++++ 1 file changed, 15 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/fixmap.h b/arch/x86/include/asm/fixmap.h index d07b44f7d1d..4d293dced62 100644 --- a/arch/x86/include/asm/fixmap.h +++ b/arch/x86/include/asm/fixmap.h @@ -214,5 +214,20 @@ static inline unsigned long virt_to_fix(const unsigned long vaddr) BUG_ON(vaddr >= FIXADDR_TOP || vaddr < FIXADDR_START); return __virt_to_fix(vaddr); } + +/* Return an pointer with offset calculated */ +static inline unsigned long __set_fixmap_offset(enum fixed_addresses idx, + phys_addr_t phys, pgprot_t flags) +{ + __set_fixmap(idx, phys, flags); + return fix_to_virt(idx) + (phys & (PAGE_SIZE - 1)); +} + +#define set_fixmap_offset(idx, phys) \ + __set_fixmap_offset(idx, phys, PAGE_KERNEL) + +#define set_fixmap_offset_nocache(idx, phys) \ + __set_fixmap_offset(idx, phys, PAGE_KERNEL_NOCACHE) + #endif /* !__ASSEMBLY__ */ #endif /* _ASM_X86_FIXMAP_H */ -- cgit v1.2.3-70-g09d2 From c20b5c3318fe45e4f33f01a91ccead645dfdf619 Mon Sep 17 00:00:00 2001 From: Feng Tang Date: Mon, 13 Sep 2010 15:08:55 +0800 Subject: x86, earlyprintk: Add earlyprintk for Intel Moorestown platform Intel Moorestown platform has a spi-uart device(Maxim3110), which connects to a Designware spi core controller. This patch will add early console function based on it. As it will be used long before Linux spi subsystem get initialised, we simply directly manipulate the spi controller's register to acheive the early console func. This is safe as it will be disabled when devices subsytem get initialised. To use it, user need enable CONFIG_X86_MRST_EARLY_PRINTK in kenrel config and add "earlyprintk=mrst" in kernel command line. Signed-off-by: Feng Tang Acked-by: Alan Cox Cc: greg@kroah.com LKML-Reference: <1284361736-23011-4-git-send-email-feng.tang@intel.com> Signed-off-by: Ingo Molnar --- arch/x86/Kconfig.debug | 4 + arch/x86/include/asm/mrst.h | 5 + arch/x86/kernel/Makefile | 1 + arch/x86/kernel/early_printk.c | 7 ++ arch/x86/kernel/early_printk_mrst.c | 232 ++++++++++++++++++++++++++++++++++++ 5 files changed, 249 insertions(+) create mode 100644 arch/x86/kernel/early_printk_mrst.c (limited to 'arch/x86') diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug index 75085080b63..e5bb96b10f1 100644 --- a/arch/x86/Kconfig.debug +++ b/arch/x86/Kconfig.debug @@ -43,6 +43,10 @@ config EARLY_PRINTK with klogd/syslogd or the X server. You should normally N here, unless you want to debug such a crash. +config EARLY_PRINTK_MRST + bool "Early printk for MRST platform support" + depends on EARLY_PRINTK && X86_MRST + config EARLY_PRINTK_DBGP bool "Early printk via EHCI debug port" depends on EARLY_PRINTK && PCI diff --git a/arch/x86/include/asm/mrst.h b/arch/x86/include/asm/mrst.h index 33fc2966beb..d0ea5bc505a 100644 --- a/arch/x86/include/asm/mrst.h +++ b/arch/x86/include/asm/mrst.h @@ -10,6 +10,9 @@ */ #ifndef _ASM_X86_MRST_H #define _ASM_X86_MRST_H + +#include + extern int pci_mrst_init(void); int __init sfi_parse_mrtc(struct sfi_table_header *table); @@ -42,4 +45,6 @@ extern enum mrst_timer_options mrst_timer_options; #define SFI_MTMR_MAX_NUM 8 #define SFI_MRTC_MAX 8 +extern struct console early_mrst_console; +extern void mrst_early_console_init(void); #endif /* _ASM_X86_MRST_H */ diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index fedf32a8c3e..3cd01d04613 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -85,6 +85,7 @@ obj-$(CONFIG_DOUBLEFAULT) += doublefault_32.o obj-$(CONFIG_KGDB) += kgdb.o obj-$(CONFIG_VM86) += vm86_32.o obj-$(CONFIG_EARLY_PRINTK) += early_printk.o +obj-$(CONFIG_EARLY_PRINTK_MRST) += early_printk_mrst.o obj-$(CONFIG_HPET_TIMER) += hpet.o obj-$(CONFIG_APB_TIMER) += apb_timer.o diff --git a/arch/x86/kernel/early_printk.c b/arch/x86/kernel/early_printk.c index fa99bae75ac..6082463768a 100644 --- a/arch/x86/kernel/early_printk.c +++ b/arch/x86/kernel/early_printk.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include @@ -238,6 +239,12 @@ static int __init setup_early_printk(char *buf) #ifdef CONFIG_HVC_XEN if (!strncmp(buf, "xen", 3)) early_console_register(&xenboot_console, keep); +#endif +#ifdef CONFIG_X86_MRST_EARLY_PRINTK + if (!strncmp(buf, "mrst", 4)) { + mrst_early_console_init(); + early_console_register(&early_mrst_console, keep); + } #endif buf++; } diff --git a/arch/x86/kernel/early_printk_mrst.c b/arch/x86/kernel/early_printk_mrst.c new file mode 100644 index 00000000000..05d27e1e2b6 --- /dev/null +++ b/arch/x86/kernel/early_printk_mrst.c @@ -0,0 +1,232 @@ +/* + * early_printk_mrst.c - spi-uart early printk for Intel Moorestown platform + * + * Copyright (c) 2008-2010, Intel Corporation + * + * 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; version 2 + * of the License. + */ + +#include +#include +#include +#include +#include + +#include +#include +#include + +#define MRST_SPI_TIMEOUT 0x200000 +#define MRST_REGBASE_SPI0 0xff128000 +#define MRST_REGBASE_SPI1 0xff128400 +#define MRST_CLK_SPI0_REG 0xff11d86c + +/* Bit fields in CTRLR0 */ +#define SPI_DFS_OFFSET 0 + +#define SPI_FRF_OFFSET 4 +#define SPI_FRF_SPI 0x0 +#define SPI_FRF_SSP 0x1 +#define SPI_FRF_MICROWIRE 0x2 +#define SPI_FRF_RESV 0x3 + +#define SPI_MODE_OFFSET 6 +#define SPI_SCPH_OFFSET 6 +#define SPI_SCOL_OFFSET 7 +#define SPI_TMOD_OFFSET 8 +#define SPI_TMOD_TR 0x0 /* xmit & recv */ +#define SPI_TMOD_TO 0x1 /* xmit only */ +#define SPI_TMOD_RO 0x2 /* recv only */ +#define SPI_TMOD_EPROMREAD 0x3 /* eeprom read mode */ + +#define SPI_SLVOE_OFFSET 10 +#define SPI_SRL_OFFSET 11 +#define SPI_CFS_OFFSET 12 + +/* Bit fields in SR, 7 bits */ +#define SR_MASK 0x7f /* cover 7 bits */ +#define SR_BUSY (1 << 0) +#define SR_TF_NOT_FULL (1 << 1) +#define SR_TF_EMPT (1 << 2) +#define SR_RF_NOT_EMPT (1 << 3) +#define SR_RF_FULL (1 << 4) +#define SR_TX_ERR (1 << 5) +#define SR_DCOL (1 << 6) + +struct dw_spi_reg { + u32 ctrl0; + u32 ctrl1; + u32 ssienr; + u32 mwcr; + u32 ser; + u32 baudr; + u32 txfltr; + u32 rxfltr; + u32 txflr; + u32 rxflr; + u32 sr; + u32 imr; + u32 isr; + u32 risr; + u32 txoicr; + u32 rxoicr; + u32 rxuicr; + u32 msticr; + u32 icr; + u32 dmacr; + u32 dmatdlr; + u32 dmardlr; + u32 idr; + u32 version; + + /* Currently operates as 32 bits, though only the low 16 bits matter */ + u32 dr; +} __packed; + +#define dw_readl(dw, name) __raw_readl(&(dw)->name) +#define dw_writel(dw, name, val) __raw_writel((val), &(dw)->name) + +/* Default use SPI0 register for mrst, we will detect Penwell and use SPI1 */ +static unsigned long mrst_spi_paddr = MRST_REGBASE_SPI0; + +static u32 *pclk_spi0; +/* Always contains an accessable address, start with 0 */ +static struct dw_spi_reg *pspi; + +static struct kmsg_dumper dw_dumper; +static int dumper_registered; + +static void dw_kmsg_dump(struct kmsg_dumper *dumper, + enum kmsg_dump_reason reason, + const char *s1, unsigned long l1, + const char *s2, unsigned long l2) +{ + int i; + + /* When run to this, we'd better re-init the HW */ + mrst_early_console_init(); + + for (i = 0; i < l1; i++) + early_mrst_console.write(&early_mrst_console, s1 + i, 1); + for (i = 0; i < l2; i++) + early_mrst_console.write(&early_mrst_console, s2 + i, 1); +} + +/* Set the ratio rate to 115200, 8n1, IRQ disabled */ +static void max3110_write_config(void) +{ + u16 config; + + config = 0xc001; + dw_writel(pspi, dr, config); +} + +/* Translate char to a eligible word and send to max3110 */ +static void max3110_write_data(char c) +{ + u16 data; + + data = 0x8000 | c; + dw_writel(pspi, dr, data); +} + +void mrst_early_console_init(void) +{ + u32 ctrlr0 = 0; + u32 spi0_cdiv; + u32 freq; /* Freqency info only need be searched once */ + + /* Base clk is 100 MHz, the actual clk = 100M / (clk_divider + 1) */ + pclk_spi0 = (void *)set_fixmap_offset_nocache(FIX_EARLYCON_MEM_BASE, + MRST_CLK_SPI0_REG); + spi0_cdiv = ((*pclk_spi0) & 0xe00) >> 9; + freq = 100000000 / (spi0_cdiv + 1); + + if (mrst_identify_cpu() == MRST_CPU_CHIP_PENWELL) + mrst_spi_paddr = MRST_REGBASE_SPI1; + + pspi = (void *)set_fixmap_offset_nocache(FIX_EARLYCON_MEM_BASE, + mrst_spi_paddr); + + /* Disable SPI controller */ + dw_writel(pspi, ssienr, 0); + + /* Set control param, 8 bits, transmit only mode */ + ctrlr0 = dw_readl(pspi, ctrl0); + + ctrlr0 &= 0xfcc0; + ctrlr0 |= 0xf | (SPI_FRF_SPI << SPI_FRF_OFFSET) + | (SPI_TMOD_TO << SPI_TMOD_OFFSET); + dw_writel(pspi, ctrl0, ctrlr0); + + /* + * Change the spi0 clk to comply with 115200 bps, use 100000 to + * calculate the clk dividor to make the clock a little slower + * than real baud rate. + */ + dw_writel(pspi, baudr, freq/100000); + + /* Disable all INT for early phase */ + dw_writel(pspi, imr, 0x0); + + /* Set the cs to spi-uart */ + dw_writel(pspi, ser, 0x2); + + /* Enable the HW, the last step for HW init */ + dw_writel(pspi, ssienr, 0x1); + + /* Set the default configuration */ + max3110_write_config(); + + /* Register the kmsg dumper */ + if (!dumper_registered) { + dw_dumper.dump = dw_kmsg_dump; + kmsg_dump_register(&dw_dumper); + dumper_registered = 1; + } +} + +/* Slave select should be called in the read/write function */ +static void early_mrst_spi_putc(char c) +{ + unsigned int timeout; + u32 sr; + + timeout = MRST_SPI_TIMEOUT; + /* Early putc needs to make sure the TX FIFO is not full */ + while (--timeout) { + sr = dw_readl(pspi, sr); + if (!(sr & SR_TF_NOT_FULL)) + cpu_relax(); + else + break; + } + + if (!timeout) + pr_warning("MRST earlycon: timed out\n"); + else + max3110_write_data(c); +} + +/* Early SPI only uses polling mode */ +static void early_mrst_spi_write(struct console *con, const char *str, unsigned n) +{ + int i; + + for (i = 0; i < n && *str; i++) { + if (*str == '\n') + early_mrst_spi_putc('\r'); + early_mrst_spi_putc(*str); + str++; + } +} + +struct console early_mrst_console = { + .name = "earlymrst", + .write = early_mrst_spi_write, + .flags = CON_PRINTBUFFER, + .index = -1, +}; -- cgit v1.2.3-70-g09d2 From 4d033556f1bfaa7604b951bfadd17aaf1b74e4c5 Mon Sep 17 00:00:00 2001 From: Feng Tang Date: Mon, 13 Sep 2010 15:08:56 +0800 Subject: x86, earlyprintk: Add hsu early console for Intel Medfield platform Intel Medfield platform has a high speed UART device, which could act as a early console. To enable early printk of HSU console, simply add "earlyprintk=hsu" in kernel command line. Currently we put the code in the early_printk_mrst.c as it is also for Intel MID platforms like the mrst early console Signed-off-by: Feng Tang Acked-by: Alan Cox Cc: greg@kroah.com LKML-Reference: <1284361736-23011-5-git-send-email-feng.tang@intel.com> Signed-off-by: Ingo Molnar --- arch/x86/include/asm/mrst.h | 3 ++ arch/x86/kernel/early_printk.c | 6 +++ arch/x86/kernel/early_printk_mrst.c | 89 ++++++++++++++++++++++++++++++++++++- 3 files changed, 97 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/mrst.h b/arch/x86/include/asm/mrst.h index d0ea5bc505a..4a711a684b1 100644 --- a/arch/x86/include/asm/mrst.h +++ b/arch/x86/include/asm/mrst.h @@ -47,4 +47,7 @@ extern enum mrst_timer_options mrst_timer_options; extern struct console early_mrst_console; extern void mrst_early_console_init(void); + +extern struct console early_hsu_console; +extern void hsu_early_console_init(void); #endif /* _ASM_X86_MRST_H */ diff --git a/arch/x86/kernel/early_printk.c b/arch/x86/kernel/early_printk.c index 6082463768a..4572f25f932 100644 --- a/arch/x86/kernel/early_printk.c +++ b/arch/x86/kernel/early_printk.c @@ -245,6 +245,12 @@ static int __init setup_early_printk(char *buf) mrst_early_console_init(); early_console_register(&early_mrst_console, keep); } + + if (!strncmp(buf, "hsu", 3)) { + hsu_early_console_init(); + early_console_register(&early_hsu_console, keep); + } + #endif buf++; } diff --git a/arch/x86/kernel/early_printk_mrst.c b/arch/x86/kernel/early_printk_mrst.c index 05d27e1e2b6..65df603622b 100644 --- a/arch/x86/kernel/early_printk_mrst.c +++ b/arch/x86/kernel/early_printk_mrst.c @@ -1,5 +1,5 @@ /* - * early_printk_mrst.c - spi-uart early printk for Intel Moorestown platform + * early_printk_mrst.c - early consoles for Intel MID platforms * * Copyright (c) 2008-2010, Intel Corporation * @@ -9,9 +9,19 @@ * of the License. */ +/* + * This file implements two early consoles named mrst and hsu. + * mrst is based on Maxim3110 spi-uart device, it exists in both + * Moorestown and Medfield platforms, while hsu is based on a High + * Speed UART device which only exists in the Medfield platform + */ + +#include +#include #include #include #include +#include #include #include @@ -230,3 +240,80 @@ struct console early_mrst_console = { .flags = CON_PRINTBUFFER, .index = -1, }; + +/* + * Following is the early console based on Medfield HSU (High + * Speed UART) device. + */ +#define HSU_PORT2_PADDR 0xffa28180 + +static void __iomem *phsu; + +void hsu_early_console_init(void) +{ + u8 lcr; + + phsu = (void *)set_fixmap_offset_nocache(FIX_EARLYCON_MEM_BASE, + HSU_PORT2_PADDR); + + /* Disable FIFO */ + writeb(0x0, phsu + UART_FCR); + + /* Set to default 115200 bps, 8n1 */ + lcr = readb(phsu + UART_LCR); + writeb((0x80 | lcr), phsu + UART_LCR); + writeb(0x18, phsu + UART_DLL); + writeb(lcr, phsu + UART_LCR); + writel(0x3600, phsu + UART_MUL*4); + + writeb(0x8, phsu + UART_MCR); + writeb(0x7, phsu + UART_FCR); + writeb(0x3, phsu + UART_LCR); + + /* Clear IRQ status */ + readb(phsu + UART_LSR); + readb(phsu + UART_RX); + readb(phsu + UART_IIR); + readb(phsu + UART_MSR); + + /* Enable FIFO */ + writeb(0x7, phsu + UART_FCR); +} + +#define BOTH_EMPTY (UART_LSR_TEMT | UART_LSR_THRE) + +static void early_hsu_putc(char ch) +{ + unsigned int timeout = 10000; /* 10ms */ + u8 status; + + while (--timeout) { + status = readb(phsu + UART_LSR); + if (status & BOTH_EMPTY) + break; + udelay(1); + } + + /* Only write the char when there was no timeout */ + if (timeout) + writeb(ch, phsu + UART_TX); +} + +static void early_hsu_write(struct console *con, const char *str, unsigned n) +{ + int i; + + for (i = 0; i < n && *str; i++) { + if (*str == '\n') + early_hsu_putc('\r'); + early_hsu_putc(*str); + str++; + } +} + +struct console early_hsu_console = { + .name = "earlyhsu", + .write = early_hsu_write, + .flags = CON_PRINTBUFFER, + .index = -1, +}; -- cgit v1.2.3-70-g09d2 From 286e5b97eb22baab9d9a41ca76c6b933a484252c Mon Sep 17 00:00:00 2001 From: Paul Fox Date: Fri, 1 Oct 2010 18:17:19 +0100 Subject: x86, olpc: Don't retry EC commands forever Avoids a potential infinite loop. It was observed once, during an EC hacking/debugging session - not in regular operation. Signed-off-by: Daniel Drake Cc: dilinger@queued.net Cc: Signed-off-by: Ingo Molnar --- arch/x86/kernel/olpc.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/olpc.c b/arch/x86/kernel/olpc.c index 37c49934c78..74726130259 100644 --- a/arch/x86/kernel/olpc.c +++ b/arch/x86/kernel/olpc.c @@ -114,6 +114,7 @@ int olpc_ec_cmd(unsigned char cmd, unsigned char *inbuf, size_t inlen, unsigned long flags; int ret = -EIO; int i; + int restarts = 0; spin_lock_irqsave(&ec_lock, flags); @@ -169,7 +170,9 @@ restart: if (wait_on_obf(0x6c, 1)) { printk(KERN_ERR "olpc-ec: timeout waiting for" " EC to provide data!\n"); - goto restart; + if (restarts++ < 10) + goto restart; + goto err; } outbuf[i] = inb(0x68); pr_devel("olpc-ec: received 0x%x\n", outbuf[i]); -- cgit v1.2.3-70-g09d2 From b62be8ea9db4048112219ff6d6ce5f183179d4dc Mon Sep 17 00:00:00 2001 From: Jin Dongming Date: Thu, 26 Aug 2010 17:29:05 +0900 Subject: x86, mce, therm_throt.c: Fix missing curly braces in error handling logic When the feature PTS is not supported by CPU, the sysfile package_power_limit_count for package should not be generated. This patch is used for fixing missing { and }. The patch is not complete as there are other error handling problems in this function - but that can wait until the merge window. Signed-off-by: Jin Dongming Reviewed-by: Fenghua Yu Acked-by: Jean Delvare Cc: Brown Len Cc: Guenter Roeck Cc: Hidetoshi Seto Cc: lm-sensors@lm-sensors.org LKML-Reference: <4C7625D1.4060201@np.css.fujitsu.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/mcheck/therm_throt.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c index d9368eeda30..169d8804a9f 100644 --- a/arch/x86/kernel/cpu/mcheck/therm_throt.c +++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c @@ -216,7 +216,7 @@ static __cpuinit int thermal_throttle_add_dev(struct sys_device *sys_dev, err = sysfs_add_file_to_group(&sys_dev->kobj, &attr_core_power_limit_count.attr, thermal_attr_group.name); - if (cpu_has(c, X86_FEATURE_PTS)) + if (cpu_has(c, X86_FEATURE_PTS)) { err = sysfs_add_file_to_group(&sys_dev->kobj, &attr_package_throttle_count.attr, thermal_attr_group.name); @@ -224,6 +224,7 @@ static __cpuinit int thermal_throttle_add_dev(struct sys_device *sys_dev, err = sysfs_add_file_to_group(&sys_dev->kobj, &attr_package_power_limit_count.attr, thermal_attr_group.name); + } return err; } -- cgit v1.2.3-70-g09d2 From 6e9636693373d938aa3b13427be3d212f172ac06 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Fri, 8 Oct 2010 14:53:48 -0400 Subject: x86, iommu: Update header comments with appropriate naming The header comments diverged a bit from the implementation. Lets re-sync them. Signed-off-by: Konrad Rzeszutek Wilk LKML-Reference: <1286564028-2352-3-git-send-email-konrad.wilk@oracle.com> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/iommu_table.h | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/iommu_table.h b/arch/x86/include/asm/iommu_table.h index df55a78888e..f229b13a5f3 100644 --- a/arch/x86/include/asm/iommu_table.h +++ b/arch/x86/include/asm/iommu_table.h @@ -1,4 +1,3 @@ - #ifndef _ASM_X86_IOMMU_TABLE_H #define _ASM_X86_IOMMU_TABLE_H @@ -60,7 +59,7 @@ struct iommu_table_entry { * and it will be run after the SWIOTLB and the other IOMMUs * that utilize this macro. If the IOMMU is detected (ie, the * detect routine returns a positive value), the other IOMMUs - * are also checked. You can use IOMMU_INIT_FINISH if you prefer + * are also checked. You can use IOMMU_INIT_POST_FINISH if you prefer * to stop detecting the other IOMMUs after yours has been detected. */ #define IOMMU_INIT_POST(_detect) \ @@ -80,9 +79,9 @@ struct iommu_table_entry { * d). Similar to the 'init', except that this gets called from pci_iommu_init * where we do have a memory allocator. * - * The _CONT vs the _EXIT differs in that the _CONT variant will + * The standard vs the _FINISH differs in that the _FINISH variant will * continue detecting other IOMMUs in the call list after the - * the detection routine returns a positive number. The _EXIT will + * the detection routine returns a positive number. The _FINISH will * stop the execution chain. Both will still call the 'init' and * 'late_init' functions if they are set. */ -- cgit v1.2.3-70-g09d2 From 708ff2a0097b02d32d375b66996661f36cd4d6d1 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Wed, 29 Sep 2010 18:08:50 +0900 Subject: bitops: make asm-generic/bitops/find.h more generic asm-generic/bitops/find.h has the extern declarations of find_next_bit() and find_next_zero_bit() and the macro definitions of find_first_bit() and find_first_zero_bit(). It is only usable by the architectures which enables CONFIG_GENERIC_FIND_NEXT_BIT and disables CONFIG_GENERIC_FIND_FIRST_BIT. x86 and tile enable both CONFIG_GENERIC_FIND_NEXT_BIT and CONFIG_GENERIC_FIND_FIRST_BIT. These architectures cannot include asm-generic/bitops/find.h in their asm/bitops.h. So ifdefed extern declarations of find_first_bit and find_first_zero_bit() are put in linux/bitops.h. This makes asm-generic/bitops/find.h usable by these architectures and use it. Also this change is needed for the forthcoming duplicated extern declarations cleanup. Signed-off-by: Akinobu Mita Signed-off-by: Arnd Bergmann Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Cc: x86@kernel.org Cc: Chris Metcalf --- arch/tile/include/asm/bitops.h | 1 + arch/x86/include/asm/bitops.h | 2 ++ include/asm-generic/bitops/find.h | 25 +++++++++++++++++++++++++ include/linux/bitops.h | 22 ---------------------- 4 files changed, 28 insertions(+), 22 deletions(-) (limited to 'arch/x86') diff --git a/arch/tile/include/asm/bitops.h b/arch/tile/include/asm/bitops.h index 6832b4be899..6d4f0ff2c68 100644 --- a/arch/tile/include/asm/bitops.h +++ b/arch/tile/include/asm/bitops.h @@ -120,6 +120,7 @@ static inline unsigned long __arch_hweight64(__u64 w) #include #include +#include #include #include #include diff --git a/arch/x86/include/asm/bitops.h b/arch/x86/include/asm/bitops.h index bafd80defa4..903683b07e4 100644 --- a/arch/x86/include/asm/bitops.h +++ b/arch/x86/include/asm/bitops.h @@ -440,6 +440,8 @@ static inline int fls(int x) #ifdef __KERNEL__ +#include + #include #define ARCH_HAS_FAST_MULTIPLIER 1 diff --git a/include/asm-generic/bitops/find.h b/include/asm-generic/bitops/find.h index 1914e974251..30afec0db7d 100644 --- a/include/asm-generic/bitops/find.h +++ b/include/asm-generic/bitops/find.h @@ -9,7 +9,32 @@ extern unsigned long find_next_zero_bit(const unsigned long *addr, unsigned long size, unsigned long offset); #endif +#ifdef CONFIG_GENERIC_FIND_FIRST_BIT + +/** + * find_first_bit - find the first set bit in a memory region + * @addr: The address to start the search at + * @size: The maximum size to search + * + * Returns the bit number of the first set bit. + */ +extern unsigned long find_first_bit(const unsigned long *addr, + unsigned long size); + +/** + * find_first_zero_bit - find the first cleared bit in a memory region + * @addr: The address to start the search at + * @size: The maximum size to search + * + * Returns the bit number of the first cleared bit. + */ +extern unsigned long find_first_zero_bit(const unsigned long *addr, + unsigned long size); +#else /* CONFIG_GENERIC_FIND_FIRST_BIT */ + #define find_first_bit(addr, size) find_next_bit((addr), (size), 0) #define find_first_zero_bit(addr, size) find_next_zero_bit((addr), (size), 0) +#endif /* CONFIG_GENERIC_FIND_FIRST_BIT */ + #endif /*_ASM_GENERIC_BITOPS_FIND_H_ */ diff --git a/include/linux/bitops.h b/include/linux/bitops.h index fc68053378c..adb0f113f57 100644 --- a/include/linux/bitops.h +++ b/include/linux/bitops.h @@ -136,28 +136,6 @@ static inline unsigned long __ffs64(u64 word) } #ifdef __KERNEL__ -#ifdef CONFIG_GENERIC_FIND_FIRST_BIT - -/** - * find_first_bit - find the first set bit in a memory region - * @addr: The address to start the search at - * @size: The maximum size to search - * - * Returns the bit number of the first set bit. - */ -extern unsigned long find_first_bit(const unsigned long *addr, - unsigned long size); - -/** - * find_first_zero_bit - find the first cleared bit in a memory region - * @addr: The address to start the search at - * @size: The maximum size to search - * - * Returns the bit number of the first cleared bit. - */ -extern unsigned long find_first_zero_bit(const unsigned long *addr, - unsigned long size); -#endif /* CONFIG_GENERIC_FIND_FIRST_BIT */ #ifdef CONFIG_GENERIC_FIND_LAST_BIT /** -- cgit v1.2.3-70-g09d2 From 6dcbfe4f0b4e17e289d56fa534b7ce5a6b7f63a3 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Fri, 8 Oct 2010 12:08:34 +0200 Subject: x86, AMD, MCE thresholding: Fix the MCi_MISCj iteration order This fixes possible cases of not collecting valid error info in the MCE error thresholding groups on F10h hardware. The current code contains a subtle problem of checking only the Valid bit of MSR0000_0413 (which is MC4_MISC0 - DRAM thresholding group) in its first iteration and breaking out if the bit is cleared. But (!), this MSR contains an offset value, BlkPtr[31:24], which points to the remaining MSRs in this thresholding group which might contain valid information too. But if we bail out only after we checked the valid bit in the first MSR and not the block pointer too, we miss that other information. The thing is, MC4_MISC0[BlkPtr] is not predicated on MCi_STATUS[MiscV] or MC4_MISC0[Valid] and should be checked prior to iterating over the MCI_MISCj thresholding group, irrespective of the MC4_MISC0[Valid] setting. Signed-off-by: Borislav Petkov Cc: Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/mcheck/mce_amd.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd.c b/arch/x86/kernel/cpu/mcheck/mce_amd.c index 5e975298fa8..39aaee5c1ab 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_amd.c +++ b/arch/x86/kernel/cpu/mcheck/mce_amd.c @@ -141,6 +141,7 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c) address = (low & MASK_BLKPTR_LO) >> 21; if (!address) break; + address += MCG_XBLK_ADDR; } else ++address; @@ -148,12 +149,8 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c) if (rdmsr_safe(address, &low, &high)) break; - if (!(high & MASK_VALID_HI)) { - if (block) - continue; - else - break; - } + if (!(high & MASK_VALID_HI)) + continue; if (!(high & MASK_CNTP_HI) || (high & MASK_LOCKED_HI)) -- cgit v1.2.3-70-g09d2 From 58877679fd393d3ef71aa383031ac7817561463d Mon Sep 17 00:00:00 2001 From: Zachary Amsden Date: Thu, 19 Aug 2010 22:07:18 -1000 Subject: KVM: x86: Fix SVM VMCB reset On reset, VMCB TSC should be set to zero. Instead, code was setting tsc_offset to zero, which passes through the underlying TSC. Signed-off-by: Zachary Amsden Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/svm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index bc5b9b8d4a3..12b502de136 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -766,7 +766,7 @@ static void init_vmcb(struct vcpu_svm *svm) control->iopm_base_pa = iopm_base; control->msrpm_base_pa = __pa(svm->msrpm); - control->tsc_offset = 0; + control->tsc_offset = 0-native_read_tsc(); control->int_ctl = V_INTR_MASKING_MASK; init_seg(&save->es); -- cgit v1.2.3-70-g09d2 From 47008cd887c1836bcadda123ba73e1863de7a6c4 Mon Sep 17 00:00:00 2001 From: Zachary Amsden Date: Thu, 19 Aug 2010 22:07:19 -1000 Subject: KVM: x86: Move TSC reset out of vmcb_init The VMCB is reset whenever we receive a startup IPI, so Linux is setting TSC back to zero happens very late in the boot process and destabilizing the TSC. Instead, just set TSC to zero once at VCPU creation time. Why the separate patch? So git-bisect is your friend. Signed-off-by: Zachary Amsden Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/svm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 12b502de136..81ed28cb36e 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -766,7 +766,6 @@ static void init_vmcb(struct vcpu_svm *svm) control->iopm_base_pa = iopm_base; control->msrpm_base_pa = __pa(svm->msrpm); - control->tsc_offset = 0-native_read_tsc(); control->int_ctl = V_INTR_MASKING_MASK; init_seg(&save->es); @@ -902,6 +901,7 @@ static struct kvm_vcpu *svm_create_vcpu(struct kvm *kvm, unsigned int id) svm->vmcb_pa = page_to_pfn(page) << PAGE_SHIFT; svm->asid_generation = 0; init_vmcb(svm); + svm->vmcb->control.tsc_offset = 0-native_read_tsc(); err = fx_init(&svm->vcpu); if (err) -- cgit v1.2.3-70-g09d2 From 73cf624d029d776a33d0a80c695485b3f9b36231 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Sun, 10 Oct 2010 19:52:15 -0700 Subject: x86, numa: For each node, register the memory blocks actually used Russ reported SGI UV is broken recently. He said: | The SRAT table shows that memory range is spread over two nodes. | | SRAT: Node 0 PXM 0 100000000-800000000 | SRAT: Node 1 PXM 1 800000000-1000000000 | SRAT: Node 0 PXM 0 1000000000-1080000000 | |Previously, the kernel early_node_map[] would show three entries |with the proper node. | |[ 0.000000] 0: 0x00100000 -> 0x00800000 |[ 0.000000] 1: 0x00800000 -> 0x01000000 |[ 0.000000] 0: 0x01000000 -> 0x01080000 | |The problem is recent community kernel early_node_map[] shows |only two entries with the node 0 entry overlapping the node 1 |entry. | | 0: 0x00100000 -> 0x01080000 | 1: 0x00800000 -> 0x01000000 After looking at the changelog, Found out that it has been broken for a while by following commit |commit 8716273caef7f55f39fe4fc6c69c5f9f197f41f1 |Author: David Rientjes |Date: Fri Sep 25 15:20:04 2009 -0700 | | x86: Export srat physical topology Before that commit, register_active_regions() is called for every SRAT memory entry right away. Use nodememblk_range[] instead of nodes[] in order to make sure we capture the actual memory blocks registered with each node. nodes[] contains an extended range which spans all memory regions associated with a node, but that does not mean that all the memory in between are included. Reported-by: Russ Anderson Tested-by: Russ Anderson Signed-off-by: Yinghai Lu LKML-Reference: <4CB27BDF.5000800@kernel.org> Acked-by: David Rientjes Cc: 2.6.33 .34 .35 .36 Signed-off-by: H. Peter Anvin --- arch/x86/mm/srat_64.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat_64.c index f9897f7a9ef..9c0d0d399c3 100644 --- a/arch/x86/mm/srat_64.c +++ b/arch/x86/mm/srat_64.c @@ -420,9 +420,11 @@ int __init acpi_scan_nodes(unsigned long start, unsigned long end) return -1; } - for_each_node_mask(i, nodes_parsed) - e820_register_active_regions(i, nodes[i].start >> PAGE_SHIFT, - nodes[i].end >> PAGE_SHIFT); + for (i = 0; i < num_node_memblks; i++) + e820_register_active_regions(memblk_nodeid[i], + node_memblk_range[i].start >> PAGE_SHIFT, + node_memblk_range[i].end >> PAGE_SHIFT); + /* for out of order entries in SRAT */ sort_node_map(); if (!nodes_cover_memory(nodes)) { -- cgit v1.2.3-70-g09d2 From 236260b90dd94516982ad67aa6f5449c4c37db7b Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Wed, 6 Oct 2010 15:52:29 -0700 Subject: memblock: Allow memblock_init to be called early The Xen setup code needs to call memblock_x86_reserve_range() very early, so allow it to initialize the memblock subsystem before doing so. The second memblock_init() is ignored. Signed-off-by: Jeremy Fitzhardinge Cc: Yinghai Lu Cc: Benjamin Herrenschmidt LKML-Reference: <4CACFDAD.3090900@goop.org> Signed-off-by: H. Peter Anvin --- arch/x86/xen/enlighten.c | 3 +++ mm/memblock.c | 6 ++++++ 2 files changed, 9 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 7d46c844141..63b83ceebd1 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include @@ -1183,6 +1184,8 @@ asmlinkage void __init xen_start_kernel(void) local_irq_disable(); early_boot_irqs_off(); + memblock_init(); + xen_raw_console_write("mapping kernel into physical memory\n"); pgd = xen_setup_kernel_pagetable(pgd, xen_start_info->nr_pages); diff --git a/mm/memblock.c b/mm/memblock.c index 9ad39690a2b..ae8b06c828c 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -752,6 +752,12 @@ void __init memblock_analyze(void) void __init memblock_init(void) { + static int init_done __initdata = 0; + + if (init_done) + return; + init_done = 1; + /* Hookup the initial arrays */ memblock.memory.regions = memblock_memory_init_regions; memblock.memory.max = INIT_MEMBLOCK_REGIONS; -- cgit v1.2.3-70-g09d2 From 50f2d7f682f9c0ed58191d0982fe77888d59d162 Mon Sep 17 00:00:00 2001 From: Nikanth Karthikesan Date: Thu, 30 Sep 2010 17:34:10 +0530 Subject: x86, numa: Assign CPUs to nodes in round-robin manner on fake NUMA commit d9c2d5ac6af87b4491bff107113aaf16f6c2b2d9 "x86, numa: Use near(er) online node instead of roundrobin for NUMA" changed NUMA initialization on Intel to choose the nearest online node or first node. Fake NUMA would be better of with round-robin initialization, instead of the all CPUS on first node. Change the choice of first node, back to round-robin. For testing NUMA kernel behaviour without cpusets and NUMA aware applications, it would be better to have cpus in different nodes, rather than all in a single node. With cpusets migration of tasks scenarios cannot not be tested. I guess having it round-robin shouldn't affect the use cases for all cpus on the first node. The code comments in arch/x86/mm/numa_64.c:759 indicate that this used to be the case, which was changed by commit d9c2d5ac6. It changed from roundrobin to nearer or first node. And I couldn't find any reason for this change in its changelog. Signed-off-by: Nikanth Karthikesan Cc: David Rientjes Signed-off-by: Andrew Morton --- arch/x86/kernel/cpu/intel.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index b4389441efb..6d61786c2dd 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -284,9 +284,7 @@ static void __cpuinit srat_detect_node(struct cpuinfo_x86 *c) /* Don't do the funky fallback heuristics the AMD version employs for now. */ node = apicid_to_node[apicid]; - if (node == NUMA_NO_NODE) - node = first_node(node_online_map); - else if (!node_online(node)) { + if (node == NUMA_NO_NODE || !node_online(node)) { /* reuse the value from init_cpu_to_node() */ node = cpu_to_node(cpu); } -- cgit v1.2.3-70-g09d2 From b683de2b3cb17bb10fa6fd4af614dc75b5749fe0 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 27 Sep 2010 20:55:03 +0200 Subject: genirq: Query arch for number of early descriptors sparse irq sets up NR_IRQS_LEGACY irq descriptors and archs then go ahead and allocate more. Use the unused return value of arch_probe_nr_irqs() to let the architecture return the number of early allocations. Fix up all users. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/arm/kernel/irq.c | 6 ++---- arch/sh/kernel/irq.c | 2 +- arch/x86/kernel/apic/io_apic.c | 2 +- include/linux/irq.h | 4 ++++ kernel/irq/irqdesc.c | 10 +++++----- kernel/softirq.c | 4 +++- 6 files changed, 16 insertions(+), 12 deletions(-) (limited to 'arch/x86') diff --git a/arch/arm/kernel/irq.c b/arch/arm/kernel/irq.c index c0d5c3b3a76..5456d11d6ae 100644 --- a/arch/arm/kernel/irq.c +++ b/arch/arm/kernel/irq.c @@ -157,10 +157,8 @@ void __init init_IRQ(void) struct irq_desc *desc; int irq; - for (irq = 0; irq < nr_irqs; irq++) { - desc = irq_to_desc_alloc_node(irq, 0); + for (irq = 0; irq < nr_irqs; irq++) desc->status |= IRQ_NOREQUEST | IRQ_NOPROBE; - } init_arch_irq(); } @@ -169,7 +167,7 @@ void __init init_IRQ(void) int __init arch_probe_nr_irqs(void) { nr_irqs = arch_nr_irqs ? arch_nr_irqs : NR_IRQS; - return 0; + return nr_irqs; } #endif diff --git a/arch/sh/kernel/irq.c b/arch/sh/kernel/irq.c index 257de1f0692..ae5bac39b89 100644 --- a/arch/sh/kernel/irq.c +++ b/arch/sh/kernel/irq.c @@ -290,7 +290,7 @@ void __init init_IRQ(void) int __init arch_probe_nr_irqs(void) { nr_irqs = sh_mv.mv_nr_irqs; - return 0; + return NR_IRQS_LEGACY; } #endif diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index f1efebaf551..5aee1d1a306 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -3880,7 +3880,7 @@ int __init arch_probe_nr_irqs(void) if (nr < nr_irqs) nr_irqs = nr; - return 0; + return NR_IRQS_LEGACY; } #endif diff --git a/include/linux/irq.h b/include/linux/irq.h index 096b74d5d0d..ef878823ee3 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -214,6 +214,10 @@ struct irq_chip { */ #include +#ifndef NR_IRQS_LEGACY +# define NR_IRQS_LEGACY 0 +#endif + #ifndef ARCH_IRQ_INIT_FLAGS # define ARCH_IRQ_INIT_FLAGS 0 #endif diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c index 7cbe4f93e2f..a1fbd1d347a 100644 --- a/kernel/irq/irqdesc.c +++ b/kernel/irq/irqdesc.c @@ -226,16 +226,16 @@ struct irq_desc * __ref irq_to_desc_alloc_node(unsigned int irq, int node) int __init early_irq_init(void) { - int i, node = first_online_node; + int i, initcnt, node = first_online_node; struct irq_desc *desc; init_irq_default_affinity(); - /* initialize nr_irqs based on nr_cpu_ids */ - arch_probe_nr_irqs(); - printk(KERN_INFO "NR_IRQS:%d nr_irqs:%d\n", NR_IRQS, nr_irqs); + /* Let arch update nr_irqs and return the nr of preallocated irqs */ + initcnt = arch_probe_nr_irqs(); + printk(KERN_INFO "NR_IRQS:%d nr_irqs:%d %d\n", NR_IRQS, nr_irqs, initcnt); - for (i = 0; i < NR_IRQS_LEGACY; i++) { + for (i = 0; i < initcnt; i++) { desc = alloc_desc(i, node); set_bit(i, allocated_irqs); irq_insert_desc(i, desc); diff --git a/kernel/softirq.c b/kernel/softirq.c index 07b4f1b1a73..14a7b80b2cc 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -886,9 +886,10 @@ int __init __weak early_irq_init(void) return 0; } +#ifdef CONFIG_GENERIC_HARDIRQS int __init __weak arch_probe_nr_irqs(void) { - return 0; + return NR_IRQS_LEGACY; } int __init __weak arch_early_irq_init(void) @@ -900,3 +901,4 @@ int __weak arch_init_chip_data(struct irq_desc *desc, int node) { return 0; } +#endif -- cgit v1.2.3-70-g09d2 From 1c9db52534a2c0e9776788cd34ccc193289fc18c Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 28 Sep 2010 16:46:51 +0200 Subject: pci: Convert msi to new irq_chip functions Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar Acked-by: Jesse Barnes Cc: Benjamin Herrenschmidt Cc: "David S. Miller" Cc: Tony Luck Cc: Russell King --- arch/arm/mach-iop13xx/msi.c | 8 ++++---- arch/ia64/kernel/msi_ia64.c | 4 ++-- arch/ia64/sn/kernel/msi_sn.c | 4 ++-- arch/powerpc/platforms/cell/axon_msi.c | 6 +++--- arch/powerpc/platforms/pseries/xics.c | 2 +- arch/powerpc/sysdev/fsl_msi.c | 4 ++-- arch/powerpc/sysdev/mpic_pasemi_msi.c | 22 +++++++++++----------- arch/powerpc/sysdev/mpic_u3msi.c | 18 +++++++++--------- arch/sparc/kernel/pci_msi.c | 8 ++++---- arch/x86/kernel/apic/io_apic.c | 8 ++++---- drivers/pci/msi.c | 14 +++++++------- include/linux/msi.h | 5 +++-- 12 files changed, 52 insertions(+), 51 deletions(-) (limited to 'arch/x86') diff --git a/arch/arm/mach-iop13xx/msi.c b/arch/arm/mach-iop13xx/msi.c index f34b0ed8063..7149fcc16c8 100644 --- a/arch/arm/mach-iop13xx/msi.c +++ b/arch/arm/mach-iop13xx/msi.c @@ -164,10 +164,10 @@ static void iop13xx_msi_nop(unsigned int irq) static struct irq_chip iop13xx_msi_chip = { .name = "PCI-MSI", .ack = iop13xx_msi_nop, - .enable = unmask_msi_irq, - .disable = mask_msi_irq, - .mask = mask_msi_irq, - .unmask = unmask_msi_irq, + .irq_enable = unmask_msi_irq, + .irq_disable = mask_msi_irq, + .irq_mask = mask_msi_irq, + .irq_unmask = unmask_msi_irq, }; int arch_setup_msi_irq(struct pci_dev *pdev, struct msi_desc *desc) diff --git a/arch/ia64/kernel/msi_ia64.c b/arch/ia64/kernel/msi_ia64.c index 4a746ea838f..dbb43424704 100644 --- a/arch/ia64/kernel/msi_ia64.c +++ b/arch/ia64/kernel/msi_ia64.c @@ -104,8 +104,8 @@ static int ia64_msi_retrigger_irq(unsigned int irq) */ static struct irq_chip ia64_msi_chip = { .name = "PCI-MSI", - .mask = mask_msi_irq, - .unmask = unmask_msi_irq, + .irq_mask = mask_msi_irq, + .irq_unmask = unmask_msi_irq, .ack = ia64_ack_msi_irq, #ifdef CONFIG_SMP .set_affinity = ia64_set_msi_irq_affinity, diff --git a/arch/ia64/sn/kernel/msi_sn.c b/arch/ia64/sn/kernel/msi_sn.c index 0c72dd46383..a5e500f0285 100644 --- a/arch/ia64/sn/kernel/msi_sn.c +++ b/arch/ia64/sn/kernel/msi_sn.c @@ -228,8 +228,8 @@ static int sn_msi_retrigger_irq(unsigned int irq) static struct irq_chip sn_msi_chip = { .name = "PCI-MSI", - .mask = mask_msi_irq, - .unmask = unmask_msi_irq, + .irq_mask = mask_msi_irq, + .irq_unmask = unmask_msi_irq, .ack = sn_ack_msi_irq, #ifdef CONFIG_SMP .set_affinity = sn_set_msi_irq_affinity, diff --git a/arch/powerpc/platforms/cell/axon_msi.c b/arch/powerpc/platforms/cell/axon_msi.c index 97085530aa6..e3e379c6caa 100644 --- a/arch/powerpc/platforms/cell/axon_msi.c +++ b/arch/powerpc/platforms/cell/axon_msi.c @@ -310,9 +310,9 @@ static void axon_msi_teardown_msi_irqs(struct pci_dev *dev) } static struct irq_chip msic_irq_chip = { - .mask = mask_msi_irq, - .unmask = unmask_msi_irq, - .shutdown = unmask_msi_irq, + .irq_mask = mask_msi_irq, + .irq_unmask = unmask_msi_irq, + .irq_shutdown = mask_msi_irq, .name = "AXON-MSI", }; diff --git a/arch/powerpc/platforms/pseries/xics.c b/arch/powerpc/platforms/pseries/xics.c index 93834b0d827..67e2c4bdac8 100644 --- a/arch/powerpc/platforms/pseries/xics.c +++ b/arch/powerpc/platforms/pseries/xics.c @@ -243,7 +243,7 @@ static unsigned int xics_startup(unsigned int virq) * at that level, so we do it here by hand. */ if (irq_to_desc(virq)->msi_desc) - unmask_msi_irq(virq); + unmask_msi_irq(irq_get_irq_data(virq)); /* unmask it */ xics_unmask_irq(virq); diff --git a/arch/powerpc/sysdev/fsl_msi.c b/arch/powerpc/sysdev/fsl_msi.c index 87991d3abba..bdbd896c89d 100644 --- a/arch/powerpc/sysdev/fsl_msi.c +++ b/arch/powerpc/sysdev/fsl_msi.c @@ -51,8 +51,8 @@ static void fsl_msi_end_irq(unsigned int virq) } static struct irq_chip fsl_msi_chip = { - .mask = mask_msi_irq, - .unmask = unmask_msi_irq, + .irq_mask = mask_msi_irq, + .irq_unmask = unmask_msi_irq, .ack = fsl_msi_end_irq, .name = "FSL-MSI", }; diff --git a/arch/powerpc/sysdev/mpic_pasemi_msi.c b/arch/powerpc/sysdev/mpic_pasemi_msi.c index 3b6a9a43718..320ad5a9a25 100644 --- a/arch/powerpc/sysdev/mpic_pasemi_msi.c +++ b/arch/powerpc/sysdev/mpic_pasemi_msi.c @@ -39,24 +39,24 @@ static struct mpic *msi_mpic; -static void mpic_pasemi_msi_mask_irq(unsigned int irq) +static void mpic_pasemi_msi_mask_irq(struct irq_data *data) { - pr_debug("mpic_pasemi_msi_mask_irq %d\n", irq); - mask_msi_irq(irq); - mpic_mask_irq(irq); + pr_debug("mpic_pasemi_msi_mask_irq %d\n", data->irq); + mask_msi_irq(data); + mpic_mask_irq(data->irq); } -static void mpic_pasemi_msi_unmask_irq(unsigned int irq) +static void mpic_pasemi_msi_unmask_irq(struct irq_data *data) { - pr_debug("mpic_pasemi_msi_unmask_irq %d\n", irq); - mpic_unmask_irq(irq); - unmask_msi_irq(irq); + pr_debug("mpic_pasemi_msi_unmask_irq %d\n", data->irq); + mpic_unmask_irq(data->irq); + unmask_msi_irq(data); } static struct irq_chip mpic_pasemi_msi_chip = { - .shutdown = mpic_pasemi_msi_mask_irq, - .mask = mpic_pasemi_msi_mask_irq, - .unmask = mpic_pasemi_msi_unmask_irq, + .irq_shutdown = mpic_pasemi_msi_mask_irq, + .irq_mask = mpic_pasemi_msi_mask_irq, + .irq_unmask = mpic_pasemi_msi_unmask_irq, .eoi = mpic_end_irq, .set_type = mpic_set_irq_type, .set_affinity = mpic_set_affinity, diff --git a/arch/powerpc/sysdev/mpic_u3msi.c b/arch/powerpc/sysdev/mpic_u3msi.c index bcbfe79c704..a2b028b4a20 100644 --- a/arch/powerpc/sysdev/mpic_u3msi.c +++ b/arch/powerpc/sysdev/mpic_u3msi.c @@ -23,22 +23,22 @@ /* A bit ugly, can we get this from the pci_dev somehow? */ static struct mpic *msi_mpic; -static void mpic_u3msi_mask_irq(unsigned int irq) +static void mpic_u3msi_mask_irq(struct irq_data *data) { - mask_msi_irq(irq); - mpic_mask_irq(irq); + mask_msi_irq(data); + mpic_mask_irq(data->irq); } -static void mpic_u3msi_unmask_irq(unsigned int irq) +static void mpic_u3msi_unmask_irq(struct irq_data *data) { - mpic_unmask_irq(irq); - unmask_msi_irq(irq); + mpic_unmask_irq(data->irq); + unmask_msi_irq(data); } static struct irq_chip mpic_u3msi_chip = { - .shutdown = mpic_u3msi_mask_irq, - .mask = mpic_u3msi_mask_irq, - .unmask = mpic_u3msi_unmask_irq, + .irq_shutdown = mpic_u3msi_mask_irq, + .irq_mask = mpic_u3msi_mask_irq, + .irq_unmask = mpic_u3msi_unmask_irq, .eoi = mpic_end_irq, .set_type = mpic_set_irq_type, .set_affinity = mpic_set_affinity, diff --git a/arch/sparc/kernel/pci_msi.c b/arch/sparc/kernel/pci_msi.c index 548b8ca9c21..b210416ace7 100644 --- a/arch/sparc/kernel/pci_msi.c +++ b/arch/sparc/kernel/pci_msi.c @@ -114,10 +114,10 @@ static void free_msi(struct pci_pbm_info *pbm, int msi_num) static struct irq_chip msi_irq = { .name = "PCI-MSI", - .mask = mask_msi_irq, - .unmask = unmask_msi_irq, - .enable = unmask_msi_irq, - .disable = mask_msi_irq, + .irq_mask = mask_msi_irq, + .irq_unmask = unmask_msi_irq, + .irq_enable = unmask_msi_irq, + .irq_disable = mask_msi_irq, /* XXX affinity XXX */ }; diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 7556eb7a1a4..b79938ff9bd 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -3441,8 +3441,8 @@ ir_set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask) */ static struct irq_chip msi_chip = { .name = "PCI-MSI", - .unmask = unmask_msi_irq, - .mask = mask_msi_irq, + .irq_unmask = unmask_msi_irq, + .irq_mask = mask_msi_irq, .ack = ack_apic_edge, #ifdef CONFIG_SMP .set_affinity = set_msi_irq_affinity, @@ -3452,8 +3452,8 @@ static struct irq_chip msi_chip = { static struct irq_chip msi_ir_chip = { .name = "IR-PCI-MSI", - .unmask = unmask_msi_irq, - .mask = mask_msi_irq, + .irq_unmask = unmask_msi_irq, + .irq_mask = mask_msi_irq, #ifdef CONFIG_INTR_REMAP .ack = ir_ack_apic_edge, #ifdef CONFIG_SMP diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c index 69b7be33b3a..55e0f9378df 100644 --- a/drivers/pci/msi.c +++ b/drivers/pci/msi.c @@ -170,27 +170,27 @@ static void msix_mask_irq(struct msi_desc *desc, u32 flag) desc->masked = __msix_mask_irq(desc, flag); } -static void msi_set_mask_bit(unsigned irq, u32 flag) +static void msi_set_mask_bit(struct irq_data *data, u32 flag) { - struct msi_desc *desc = get_irq_msi(irq); + struct msi_desc *desc = irq_data_get_msi(data); if (desc->msi_attrib.is_msix) { msix_mask_irq(desc, flag); readl(desc->mask_base); /* Flush write to device */ } else { - unsigned offset = irq - desc->dev->irq; + unsigned offset = data->irq - desc->dev->irq; msi_mask_irq(desc, 1 << offset, flag << offset); } } -void mask_msi_irq(unsigned int irq) +void mask_msi_irq(struct irq_data *data) { - msi_set_mask_bit(irq, 1); + msi_set_mask_bit(data, 1); } -void unmask_msi_irq(unsigned int irq) +void unmask_msi_irq(struct irq_data *data) { - msi_set_mask_bit(irq, 0); + msi_set_mask_bit(data, 0); } void read_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg) diff --git a/include/linux/msi.h b/include/linux/msi.h index 91b05c17185..329d17c395a 100644 --- a/include/linux/msi.h +++ b/include/linux/msi.h @@ -11,8 +11,9 @@ struct msi_msg { /* Helper functions */ struct irq_desc; -extern void mask_msi_irq(unsigned int irq); -extern void unmask_msi_irq(unsigned int irq); +struct irq_data; +extern void mask_msi_irq(struct irq_data *data); +extern void unmask_msi_irq(struct irq_data *data); extern void read_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg); extern void get_cached_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg); extern void write_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg); -- cgit v1.2.3-70-g09d2 From 39431acb1a4c464e62471cb3058b8ffffb9244db Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 28 Sep 2010 19:09:51 +0200 Subject: pci: Cleanup the irq_desc mess in msi Handing down irq_desc to msi just so that msi can access irq_desc.irq_data.msi_desc is a pretty stupid idea. The calling code can hand down a pointer to msi_desc so msi code does not need to know about the irq descriptor at all. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar Acked-by: Jesse Barnes --- arch/x86/kernel/apic/io_apic.c | 4 ++-- drivers/pci/msi.c | 24 +++++++++--------------- include/linux/msi.h | 8 ++++---- 3 files changed, 15 insertions(+), 21 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index b79938ff9bd..74bb027b517 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -3383,14 +3383,14 @@ static int set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask) cfg = desc->chip_data; - get_cached_msi_msg_desc(desc, &msg); + __get_cached_msi_msg(desc->irq_data.msi_desc, &msg); msg.data &= ~MSI_DATA_VECTOR_MASK; msg.data |= MSI_DATA_VECTOR(cfg->vector); msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK; msg.address_lo |= MSI_ADDR_DEST_ID(dest); - write_msi_msg_desc(desc, &msg); + __write_msi_msg(desc->irq_data.msi_desc, &msg); return 0; } diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c index 55e0f9378df..5fcf5aec680 100644 --- a/drivers/pci/msi.c +++ b/drivers/pci/msi.c @@ -193,10 +193,8 @@ void unmask_msi_irq(struct irq_data *data) msi_set_mask_bit(data, 0); } -void read_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg) +void __read_msi_msg(struct msi_desc *entry, struct msi_msg *msg) { - struct msi_desc *entry = get_irq_desc_msi(desc); - BUG_ON(entry->dev->current_state != PCI_D0); if (entry->msi_attrib.is_msix) { @@ -227,15 +225,13 @@ void read_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg) void read_msi_msg(unsigned int irq, struct msi_msg *msg) { - struct irq_desc *desc = irq_to_desc(irq); + struct msi_desc *entry = get_irq_msi(irq); - read_msi_msg_desc(desc, msg); + __read_msi_msg(entry, msg); } -void get_cached_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg) +void __get_cached_msi_msg(struct msi_desc *entry, struct msi_msg *msg) { - struct msi_desc *entry = get_irq_desc_msi(desc); - /* Assert that the cache is valid, assuming that * valid messages are not all-zeroes. */ BUG_ON(!(entry->msg.address_hi | entry->msg.address_lo | @@ -246,15 +242,13 @@ void get_cached_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg) void get_cached_msi_msg(unsigned int irq, struct msi_msg *msg) { - struct irq_desc *desc = irq_to_desc(irq); + struct msi_desc *entry = get_irq_msi(irq); - get_cached_msi_msg_desc(desc, msg); + __get_cached_msi_msg(entry, msg); } -void write_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg) +void __write_msi_msg(struct msi_desc *entry, struct msi_msg *msg) { - struct msi_desc *entry = get_irq_desc_msi(desc); - if (entry->dev->current_state != PCI_D0) { /* Don't touch the hardware now */ } else if (entry->msi_attrib.is_msix) { @@ -292,9 +286,9 @@ void write_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg) void write_msi_msg(unsigned int irq, struct msi_msg *msg) { - struct irq_desc *desc = irq_to_desc(irq); + struct msi_desc *entry = get_irq_msi(irq); - write_msi_msg_desc(desc, msg); + __write_msi_msg(entry, msg); } static void free_msi_irqs(struct pci_dev *dev) diff --git a/include/linux/msi.h b/include/linux/msi.h index 329d17c395a..05acced439a 100644 --- a/include/linux/msi.h +++ b/include/linux/msi.h @@ -10,13 +10,13 @@ struct msi_msg { }; /* Helper functions */ -struct irq_desc; struct irq_data; +struct msi_desc; extern void mask_msi_irq(struct irq_data *data); extern void unmask_msi_irq(struct irq_data *data); -extern void read_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg); -extern void get_cached_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg); -extern void write_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg); +extern void __read_msi_msg(struct msi_desc *entry, struct msi_msg *msg); +extern void __get_cached_msi_msg(struct msi_desc *entry, struct msi_msg *msg); +extern void __write_msi_msg(struct msi_desc *entry, struct msi_msg *msg); extern void read_msi_msg(unsigned int irq, struct msi_msg *msg); extern void get_cached_msi_msg(unsigned int irq, struct msi_msg *msg); extern void write_msi_msg(unsigned int irq, struct msi_msg *msg); -- cgit v1.2.3-70-g09d2 From 011d578fdadb64bcc1deedbb02216bfee6a9b4dc Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 28 Sep 2010 00:15:31 +0200 Subject: x86: Remove useless reinitialization of irq descriptors The descriptors are already initialized in exactly this way. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/kernel/irqinit.c | 17 ++++------------- 1 file changed, 4 insertions(+), 13 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/irqinit.c b/arch/x86/kernel/irqinit.c index 990ae7cfc57..a91ab503e24 100644 --- a/arch/x86/kernel/irqinit.c +++ b/arch/x86/kernel/irqinit.c @@ -100,6 +100,8 @@ int vector_used_by_percpu_irq(unsigned int vector) void __init init_ISA_irqs(void) { + struct irq_chip *chip = legacy_pic->chip; + const char *name = chip->name; int i; #if defined(CONFIG_X86_64) || defined(CONFIG_X86_LOCAL_APIC) @@ -107,19 +109,8 @@ void __init init_ISA_irqs(void) #endif legacy_pic->init(0); - /* - * 16 old-style INTA-cycle interrupts: - */ - for (i = 0; i < legacy_pic->nr_legacy_irqs; i++) { - struct irq_desc *desc = irq_to_desc(i); - - desc->status = IRQ_DISABLED; - desc->action = NULL; - desc->depth = 1; - - set_irq_chip_and_handler_name(i, &i8259A_chip, - handle_level_irq, "XT"); - } + for (i = 0; i < legacy_pic->nr_legacy_irqs; i++) + set_irq_chip_and_handler_name(i, chip, handle_level_irq, name); } void __init init_IRQ(void) -- cgit v1.2.3-70-g09d2 From a3c08e5d80c54e32423efbba113b02942c91f726 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 8 Oct 2010 20:24:58 +0200 Subject: x86: Convert irq_chip access to new functions Before moving the irq chips to the new functions, fixup direct callers. The cpu offline irq fixup code needs to become generic and archs need to honour the "force" flag as an indicator, but that's for later. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/kernel/irq.c | 24 +++++++++++++----------- 1 file changed, 13 insertions(+), 11 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c index 91fd0c70a18..d765bdc4807 100644 --- a/arch/x86/kernel/irq.c +++ b/arch/x86/kernel/irq.c @@ -159,7 +159,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_printf(p, "%*d: ", prec, i); for_each_online_cpu(j) seq_printf(p, "%10u ", kstat_irqs_cpu(i, j)); - seq_printf(p, " %8s", desc->chip->name); + seq_printf(p, " %8s", desc->irq_data.chip->name); seq_printf(p, "-%-8s", desc->name); if (action) { @@ -282,6 +282,7 @@ void fixup_irqs(void) unsigned int irq, vector; static int warned; struct irq_desc *desc; + struct irq_data *data; for_each_irq_desc(irq, desc) { int break_affinity = 0; @@ -296,7 +297,8 @@ void fixup_irqs(void) /* interrupt's are disabled at this point */ raw_spin_lock(&desc->lock); - affinity = desc->affinity; + data = &desc->irq_data; + affinity = data->affinity; if (!irq_has_action(irq) || cpumask_equal(affinity, cpu_online_mask)) { raw_spin_unlock(&desc->lock); @@ -315,16 +317,16 @@ void fixup_irqs(void) affinity = cpu_all_mask; } - if (!(desc->status & IRQ_MOVE_PCNTXT) && desc->chip->mask) - desc->chip->mask(irq); + if (!(desc->status & IRQ_MOVE_PCNTXT) && data->chip->irq_mask) + data->chip->irq_mask(data); - if (desc->chip->set_affinity) - desc->chip->set_affinity(irq, affinity); + if (data->chip->irq_set_affinity) + data->chip->irq_set_affinity(data, affinity, true); else if (!(warned++)) set_affinity = 0; - if (!(desc->status & IRQ_MOVE_PCNTXT) && desc->chip->unmask) - desc->chip->unmask(irq); + if (!(desc->status & IRQ_MOVE_PCNTXT) && data->chip->irq_unmask) + data->chip->irq_unmask(data); raw_spin_unlock(&desc->lock); @@ -355,10 +357,10 @@ void fixup_irqs(void) if (irr & (1 << (vector % 32))) { irq = __get_cpu_var(vector_irq)[vector]; - desc = irq_to_desc(irq); + data = irq_get_irq_data(irq); raw_spin_lock(&desc->lock); - if (desc->chip->retrigger) - desc->chip->retrigger(irq); + if (data->chip->irq_retrigger) + data->chip->irq_retrigger(data); raw_spin_unlock(&desc->lock); } } -- cgit v1.2.3-70-g09d2 From a5ef2e70405c8a9ee380b5ff33a008c75454791f Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 28 Sep 2010 11:11:10 +0200 Subject: x86: Sanitize apb timer interrupt handling Disable the interrupt in CPU_DEAD where it belongs. Remove the open coded irq_desc manipulation. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar Cc: Jacob Pan --- arch/x86/kernel/apb_timer.c | 54 ++++++++++++++++++++------------------------- 1 file changed, 24 insertions(+), 30 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apb_timer.c b/arch/x86/kernel/apb_timer.c index 08f75fb4f50..42a70a2accc 100644 --- a/arch/x86/kernel/apb_timer.c +++ b/arch/x86/kernel/apb_timer.c @@ -231,34 +231,6 @@ static void apbt_restart_clocksource(struct clocksource *cs) apbt_start_counter(phy_cs_timer_id); } -/* Setup IRQ routing via IOAPIC */ -#ifdef CONFIG_SMP -static void apbt_setup_irq(struct apbt_dev *adev) -{ - struct irq_chip *chip; - struct irq_desc *desc; - - /* timer0 irq has been setup early */ - if (adev->irq == 0) - return; - desc = irq_to_desc(adev->irq); - chip = get_irq_chip(adev->irq); - disable_irq(adev->irq); - desc->status |= IRQ_MOVE_PCNTXT; - irq_set_affinity(adev->irq, cpumask_of(adev->cpu)); - /* APB timer irqs are set up as mp_irqs, timer is edge triggerred */ - set_irq_chip_and_handler_name(adev->irq, chip, handle_edge_irq, "edge"); - enable_irq(adev->irq); - if (system_state == SYSTEM_BOOTING) - if (request_irq(adev->irq, apbt_interrupt_handler, - IRQF_TIMER | IRQF_DISABLED | IRQF_NOBALANCING, - adev->name, adev)) { - printk(KERN_ERR "Failed request IRQ for APBT%d\n", - adev->num); - } -} -#endif - static void apbt_enable_int(int n) { unsigned long ctrl = apbt_readl(n, APBTMR_N_CONTROL); @@ -334,6 +306,27 @@ static int __init apbt_clockevent_register(void) } #ifdef CONFIG_SMP + +static void apbt_setup_irq(struct apbt_dev *adev) +{ + /* timer0 irq has been setup early */ + if (adev->irq == 0) + return; + + if (system_state == SYSTEM_BOOTING) { + irq_modify_status(adev->irq, 0, IRQ_MOVE_PCNTXT); + /* APB timer irqs are set up as mp_irqs, timer is edge type */ + __set_irq_handler(adev->irq, handle_edge_irq, 0, "edge"); + if (request_irq(adev->irq, apbt_interrupt_handler, + IRQF_TIMER | IRQF_DISABLED | IRQF_NOBALANCING, + adev->name, adev)) { + printk(KERN_ERR "Failed request IRQ for APBT%d\n", + adev->num); + } + } else + enable_irq(adev->irq); +} + /* Should be called with per cpu */ void apbt_setup_secondary_clock(void) { @@ -389,10 +382,11 @@ static int apbt_cpuhp_notify(struct notifier_block *n, switch (action & 0xf) { case CPU_DEAD: + disable_irq(adev->irq); apbt_disable_int(cpu); - if (system_state == SYSTEM_RUNNING) + if (system_state == SYSTEM_RUNNING) { pr_debug("skipping APBT CPU %lu offline\n", cpu); - else if (adev) { + } else if (adev) { pr_debug("APBT clockevent for cpu %lu offline\n", cpu); free_irq(adev->irq, adev); } -- cgit v1.2.3-70-g09d2 From fe25c7fc2e036e1569faac8715a8aa5496cda78d Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 28 Sep 2010 14:57:24 +0200 Subject: x86: lguest: Convert to new irq chip functions Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar Cc: Rusty Russell --- arch/x86/lguest/boot.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c index 9d5f5584845..2d4e6fcac83 100644 --- a/arch/x86/lguest/boot.c +++ b/arch/x86/lguest/boot.c @@ -791,22 +791,22 @@ static void lguest_flush_tlb_kernel(void) * simple as setting a bit. We don't actually "ack" interrupts as such, we * just mask and unmask them. I wonder if we should be cleverer? */ -static void disable_lguest_irq(unsigned int irq) +static void disable_lguest_irq(struct irq_data *data) { - set_bit(irq, lguest_data.blocked_interrupts); + set_bit(data->irq, lguest_data.blocked_interrupts); } -static void enable_lguest_irq(unsigned int irq) +static void enable_lguest_irq(struct irq_data *data) { - clear_bit(irq, lguest_data.blocked_interrupts); + clear_bit(data->irq, lguest_data.blocked_interrupts); } /* This structure describes the lguest IRQ controller. */ static struct irq_chip lguest_irq_controller = { .name = "lguest", - .mask = disable_lguest_irq, - .mask_ack = disable_lguest_irq, - .unmask = enable_lguest_irq, + .irq_mask = disable_lguest_irq, + .irq_mask_ack = disable_lguest_irq, + .irq_unmask = enable_lguest_irq, }; /* -- cgit v1.2.3-70-g09d2 From 020dd984d7c0792e8234f2e4b4fb0534fe750f9d Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 28 Sep 2010 14:59:58 +0200 Subject: x86: Cleanup visws interrupt handling Remove the open coded access to irq_desc and convert to the new irq chip functions. Change the mask function of piix4_virtual_irq_type so we can use the generic irq handling function for the virtual interrupt instead of open coding it. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/kernel/visws_quirks.c | 140 +++++++++++++---------------------------- 1 file changed, 44 insertions(+), 96 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/visws_quirks.c b/arch/x86/kernel/visws_quirks.c index e680ea52db9..3371bd053b8 100644 --- a/arch/x86/kernel/visws_quirks.c +++ b/arch/x86/kernel/visws_quirks.c @@ -66,10 +66,7 @@ static void __init visws_time_init(void) } /* Replaces the default init_ISA_irqs in the generic setup */ -static void __init visws_pre_intr_init(void) -{ - init_VISWS_APIC_irqs(); -} +static void __init visws_pre_intr_init(void); /* Quirk for machine specific memory setup. */ @@ -429,67 +426,34 @@ static int is_co_apic(unsigned int irq) /* * This is the SGI Cobalt (IO-)APIC: */ - -static void enable_cobalt_irq(unsigned int irq) +static void enable_cobalt_irq(struct irq_data *data) { - co_apic_set(is_co_apic(irq), irq); + co_apic_set(is_co_apic(data->irq), data->irq); } -static void disable_cobalt_irq(unsigned int irq) +static void disable_cobalt_irq(struct irq_data *data) { - int entry = is_co_apic(irq); + int entry = is_co_apic(data->irq); co_apic_write(CO_APIC_LO(entry), CO_APIC_MASK); co_apic_read(CO_APIC_LO(entry)); } -/* - * "irq" really just serves to identify the device. Here is where we - * map this to the Cobalt APIC entry where it's physically wired. - * This is called via request_irq -> setup_irq -> irq_desc->startup() - */ -static unsigned int startup_cobalt_irq(unsigned int irq) +static void ack_cobalt_irq(struct irq_data *data) { unsigned long flags; - struct irq_desc *desc = irq_to_desc(irq); spin_lock_irqsave(&cobalt_lock, flags); - if ((desc->status & (IRQ_DISABLED | IRQ_INPROGRESS | IRQ_WAITING))) - desc->status &= ~(IRQ_DISABLED | IRQ_INPROGRESS | IRQ_WAITING); - enable_cobalt_irq(irq); - spin_unlock_irqrestore(&cobalt_lock, flags); - return 0; -} - -static void ack_cobalt_irq(unsigned int irq) -{ - unsigned long flags; - - spin_lock_irqsave(&cobalt_lock, flags); - disable_cobalt_irq(irq); + disable_cobalt_irq(data); apic_write(APIC_EOI, APIC_EIO_ACK); spin_unlock_irqrestore(&cobalt_lock, flags); } -static void end_cobalt_irq(unsigned int irq) -{ - unsigned long flags; - struct irq_desc *desc = irq_to_desc(irq); - - spin_lock_irqsave(&cobalt_lock, flags); - if (!(desc->status & (IRQ_DISABLED | IRQ_INPROGRESS))) - enable_cobalt_irq(irq); - spin_unlock_irqrestore(&cobalt_lock, flags); -} - static struct irq_chip cobalt_irq_type = { - .name = "Cobalt-APIC", - .startup = startup_cobalt_irq, - .shutdown = disable_cobalt_irq, - .enable = enable_cobalt_irq, - .disable = disable_cobalt_irq, - .ack = ack_cobalt_irq, - .end = end_cobalt_irq, + .name = "Cobalt-APIC", + .irq_enable = enable_cobalt_irq, + .irq_disable = disable_cobalt_irq, + .irq_ack = ack_cobalt_irq, }; @@ -503,35 +467,34 @@ static struct irq_chip cobalt_irq_type = { * interrupt controller type, and through a special virtual interrupt- * controller. Device drivers only see the virtual interrupt sources. */ -static unsigned int startup_piix4_master_irq(unsigned int irq) +static unsigned int startup_piix4_master_irq(struct irq_data *data) { legacy_pic->init(0); - - return startup_cobalt_irq(irq); + enable_cobalt_irq(data); } -static void end_piix4_master_irq(unsigned int irq) +static void end_piix4_master_irq(struct irq_data *data) { unsigned long flags; spin_lock_irqsave(&cobalt_lock, flags); - enable_cobalt_irq(irq); + enable_cobalt_irq(data); spin_unlock_irqrestore(&cobalt_lock, flags); } static struct irq_chip piix4_master_irq_type = { - .name = "PIIX4-master", - .startup = startup_piix4_master_irq, - .ack = ack_cobalt_irq, - .end = end_piix4_master_irq, + .name = "PIIX4-master", + .irq_startup = startup_piix4_master_irq, + .irq_ack = ack_cobalt_irq, }; +static void pii4_mask(struct irq_data *data) { } static struct irq_chip piix4_virtual_irq_type = { - .name = "PIIX4-virtual", + .name = "PIIX4-virtual", + .mask = pii4_mask, }; - /* * PIIX4-8259 master/virtual functions to handle interrupt requests * from legacy devices: floppy, parallel, serial, rtc. @@ -549,9 +512,8 @@ static struct irq_chip piix4_virtual_irq_type = { */ static irqreturn_t piix4_master_intr(int irq, void *dev_id) { - int realirq; - struct irq_desc *desc; unsigned long flags; + int realirq; raw_spin_lock_irqsave(&i8259A_lock, flags); @@ -592,18 +554,10 @@ static irqreturn_t piix4_master_intr(int irq, void *dev_id) raw_spin_unlock_irqrestore(&i8259A_lock, flags); - desc = irq_to_desc(realirq); - /* * handle this 'virtual interrupt' as a Cobalt one now. */ - kstat_incr_irqs_this_cpu(realirq, desc); - - if (likely(desc->action != NULL)) - handle_IRQ_event(realirq, desc->action); - - if (!(desc->status & IRQ_DISABLED)) - legacy_pic->chip->unmask(realirq); + generic_handle_irq(realirq); return IRQ_HANDLED; @@ -624,41 +578,35 @@ static struct irqaction cascade_action = { static inline void set_piix4_virtual_irq_type(void) { - piix4_virtual_irq_type.shutdown = i8259A_chip.mask; piix4_virtual_irq_type.enable = i8259A_chip.unmask; piix4_virtual_irq_type.disable = i8259A_chip.mask; + piix4_virtual_irq_type.unmask = i8259A_chip.unmask; } -void init_VISWS_APIC_irqs(void) +static void __init visws_pre_intr_init(void) { int i; - for (i = 0; i < CO_IRQ_APIC0 + CO_APIC_LAST + 1; i++) { - struct irq_desc *desc = irq_to_desc(i); - - desc->status = IRQ_DISABLED; - desc->action = 0; - desc->depth = 1; + set_piix4_virtual_irq_type(); - if (i == 0) { - desc->chip = &cobalt_irq_type; - } - else if (i == CO_IRQ_IDE0) { - desc->chip = &cobalt_irq_type; - } - else if (i == CO_IRQ_IDE1) { - desc->chip = &cobalt_irq_type; - } - else if (i == CO_IRQ_8259) { - desc->chip = &piix4_master_irq_type; - } - else if (i < CO_IRQ_APIC0) { - set_piix4_virtual_irq_type(); - desc->chip = &piix4_virtual_irq_type; - } - else if (IS_CO_APIC(i)) { - desc->chip = &cobalt_irq_type; - } + for (i = 0; i < CO_IRQ_APIC0 + CO_APIC_LAST + 1; i++) { + struct irq_chip *chip = NULL; + + if (i == 0) + chip = &cobalt_irq_type; + else if (i == CO_IRQ_IDE0) + chip = &cobalt_irq_type; + else if (i == CO_IRQ_IDE1) + >chip = &cobalt_irq_type; + else if (i == CO_IRQ_8259) + chip = &piix4_master_irq_type; + else if (i < CO_IRQ_APIC0) + chip = &piix4_virtual_irq_type; + else if (IS_CO_APIC(i)) + chip = &cobalt_irq_type; + + if (chip) + set_irq_chip(i, chip); } setup_irq(CO_IRQ_8259, &master_action); -- cgit v1.2.3-70-g09d2 From 4305df947ca1fd52867c8d56837a4e6b1e33167c Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 28 Sep 2010 15:01:33 +0200 Subject: x86: i8259: Convert to new irq_chip functions Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/include/asm/i8259.h | 2 ++ arch/x86/kernel/apic/io_apic.c | 20 +++++++------- arch/x86/kernel/apic/nmi.c | 2 +- arch/x86/kernel/i8259.c | 63 +++++++++++++++++++++--------------------- arch/x86/kernel/smpboot.c | 4 +-- 5 files changed, 47 insertions(+), 44 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/i8259.h b/arch/x86/include/asm/i8259.h index 1655147646a..a20365953bf 100644 --- a/arch/x86/include/asm/i8259.h +++ b/arch/x86/include/asm/i8259.h @@ -55,6 +55,8 @@ extern struct irq_chip i8259A_chip; struct legacy_pic { int nr_legacy_irqs; struct irq_chip *chip; + void (*mask)(unsigned int irq); + void (*unmask)(unsigned int irq); void (*mask_all)(void); void (*restore_mask)(void); void (*init)(int auto_eoi); diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 74bb027b517..e5ae2a22262 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -1459,7 +1459,7 @@ static void setup_IO_APIC_irq(int apic_id, int pin, unsigned int irq, struct irq ioapic_register_intr(irq, desc, trigger); if (irq < legacy_pic->nr_legacy_irqs) - legacy_pic->chip->mask(irq); + legacy_pic->mask(irq); ioapic_write_entry(apic_id, pin, entry); } @@ -2233,7 +2233,7 @@ static unsigned int startup_ioapic_irq(unsigned int irq) raw_spin_lock_irqsave(&ioapic_lock, flags); if (irq < legacy_pic->nr_legacy_irqs) { - legacy_pic->chip->mask(irq); + legacy_pic->mask(irq); if (legacy_pic->irq_pending(irq)) was_pending = 1; } @@ -2928,7 +2928,7 @@ static inline void __init check_timer(void) /* * get/set the timer IRQ vector: */ - legacy_pic->chip->mask(0); + legacy_pic->mask(0); assign_irq_vector(0, cfg, apic->target_cpus()); /* @@ -3000,7 +3000,7 @@ static inline void __init check_timer(void) if (timer_irq_works()) { if (nmi_watchdog == NMI_IO_APIC) { setup_nmi(); - legacy_pic->chip->unmask(0); + legacy_pic->unmask(0); } if (disable_timer_pin_1 > 0) clear_IO_APIC_pin(0, pin1); @@ -3023,14 +3023,14 @@ static inline void __init check_timer(void) */ replace_pin_at_irq_node(cfg, node, apic1, pin1, apic2, pin2); setup_timer_IRQ0_pin(apic2, pin2, cfg->vector); - legacy_pic->chip->unmask(0); + legacy_pic->unmask(0); if (timer_irq_works()) { apic_printk(APIC_QUIET, KERN_INFO "....... works.\n"); timer_through_8259 = 1; if (nmi_watchdog == NMI_IO_APIC) { - legacy_pic->chip->mask(0); + legacy_pic->mask(0); setup_nmi(); - legacy_pic->chip->unmask(0); + legacy_pic->unmask(0); } goto out; } @@ -3038,7 +3038,7 @@ static inline void __init check_timer(void) * Cleanup, just in case ... */ local_irq_disable(); - legacy_pic->chip->mask(0); + legacy_pic->mask(0); clear_IO_APIC_pin(apic2, pin2); apic_printk(APIC_QUIET, KERN_INFO "....... failed.\n"); } @@ -3057,14 +3057,14 @@ static inline void __init check_timer(void) lapic_register_intr(0, desc); apic_write(APIC_LVT0, APIC_DM_FIXED | cfg->vector); /* Fixed mode */ - legacy_pic->chip->unmask(0); + legacy_pic->unmask(0); if (timer_irq_works()) { apic_printk(APIC_QUIET, KERN_INFO "..... works.\n"); goto out; } local_irq_disable(); - legacy_pic->chip->mask(0); + legacy_pic->mask(0); apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED | cfg->vector); apic_printk(APIC_QUIET, KERN_INFO "..... failed.\n"); diff --git a/arch/x86/kernel/apic/nmi.c b/arch/x86/kernel/apic/nmi.c index a43f71cb30f..c90041ccb74 100644 --- a/arch/x86/kernel/apic/nmi.c +++ b/arch/x86/kernel/apic/nmi.c @@ -178,7 +178,7 @@ int __init check_nmi_watchdog(void) error: if (nmi_watchdog == NMI_IO_APIC) { if (!timer_through_8259) - legacy_pic->chip->mask(0); + legacy_pic->mask(0); on_each_cpu(__acpi_nmi_disable, NULL, 1); } diff --git a/arch/x86/kernel/i8259.c b/arch/x86/kernel/i8259.c index cafa7c80ac9..20757cb2efa 100644 --- a/arch/x86/kernel/i8259.c +++ b/arch/x86/kernel/i8259.c @@ -29,24 +29,10 @@ * plus some generic x86 specific things if generic specifics makes * any sense at all. */ +static void init_8259A(int auto_eoi); static int i8259A_auto_eoi; DEFINE_RAW_SPINLOCK(i8259A_lock); -static void mask_and_ack_8259A(unsigned int); -static void mask_8259A(void); -static void unmask_8259A(void); -static void disable_8259A_irq(unsigned int irq); -static void enable_8259A_irq(unsigned int irq); -static void init_8259A(int auto_eoi); -static int i8259A_irq_pending(unsigned int irq); - -struct irq_chip i8259A_chip = { - .name = "XT-PIC", - .mask = disable_8259A_irq, - .disable = disable_8259A_irq, - .unmask = enable_8259A_irq, - .mask_ack = mask_and_ack_8259A, -}; /* * 8259A PIC functions to handle ISA devices: @@ -68,7 +54,7 @@ unsigned int cached_irq_mask = 0xffff; */ unsigned long io_apic_irqs; -static void disable_8259A_irq(unsigned int irq) +static void mask_8259A_irq(unsigned int irq) { unsigned int mask = 1 << irq; unsigned long flags; @@ -82,7 +68,12 @@ static void disable_8259A_irq(unsigned int irq) raw_spin_unlock_irqrestore(&i8259A_lock, flags); } -static void enable_8259A_irq(unsigned int irq) +static void disable_8259A_irq(struct irq_data *data) +{ + mask_8259A_irq(data->irq); +} + +static void unmask_8259A_irq(unsigned int irq) { unsigned int mask = ~(1 << irq); unsigned long flags; @@ -96,6 +87,11 @@ static void enable_8259A_irq(unsigned int irq) raw_spin_unlock_irqrestore(&i8259A_lock, flags); } +static void enable_8259A_irq(struct irq_data *data) +{ + unmask_8259A_irq(data->irq); +} + static int i8259A_irq_pending(unsigned int irq) { unsigned int mask = 1<irq; unsigned int irqmask = 1 << irq; unsigned long flags; @@ -223,6 +220,14 @@ spurious_8259A_irq: } } +struct irq_chip i8259A_chip = { + .name = "XT-PIC", + .irq_mask = disable_8259A_irq, + .irq_disable = disable_8259A_irq, + .irq_unmask = enable_8259A_irq, + .irq_mask_ack = mask_and_ack_8259A, +}; + static char irq_trigger[2]; /** * ELCR registers (0x4d0, 0x4d1) control edge/level of IRQ @@ -342,9 +347,9 @@ static void init_8259A(int auto_eoi) * In AEOI mode we just have to mask the interrupt * when acking. */ - i8259A_chip.mask_ack = disable_8259A_irq; + i8259A_chip.irq_mask_ack = disable_8259A_irq; else - i8259A_chip.mask_ack = mask_and_ack_8259A; + i8259A_chip.irq_mask_ack = mask_and_ack_8259A; udelay(100); /* wait for 8259A to initialize */ @@ -363,14 +368,6 @@ static void init_8259A(int auto_eoi) static void legacy_pic_noop(void) { }; static void legacy_pic_uint_noop(unsigned int unused) { }; static void legacy_pic_int_noop(int unused) { }; - -static struct irq_chip dummy_pic_chip = { - .name = "dummy pic", - .mask = legacy_pic_uint_noop, - .unmask = legacy_pic_uint_noop, - .disable = legacy_pic_uint_noop, - .mask_ack = legacy_pic_uint_noop, -}; static int legacy_pic_irq_pending_noop(unsigned int irq) { return 0; @@ -378,7 +375,9 @@ static int legacy_pic_irq_pending_noop(unsigned int irq) struct legacy_pic null_legacy_pic = { .nr_legacy_irqs = 0, - .chip = &dummy_pic_chip, + .chip = &dummy_irq_chip, + .mask = legacy_pic_uint_noop, + .unmask = legacy_pic_uint_noop, .mask_all = legacy_pic_noop, .restore_mask = legacy_pic_noop, .init = legacy_pic_int_noop, @@ -389,7 +388,9 @@ struct legacy_pic null_legacy_pic = { struct legacy_pic default_legacy_pic = { .nr_legacy_irqs = NR_IRQS_LEGACY, .chip = &i8259A_chip, - .mask_all = mask_8259A, + .mask = mask_8259A_irq, + .unmask = unmask_8259A_irq, + .mask_all = mask_8259A, .restore_mask = unmask_8259A, .init = init_8259A, .irq_pending = i8259A_irq_pending, diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 87a8c6b00f8..864b386f6c0 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -324,9 +324,9 @@ notrace static void __cpuinit start_secondary(void *unused) check_tsc_sync_target(); if (nmi_watchdog == NMI_IO_APIC) { - legacy_pic->chip->mask(0); + legacy_pic->mask(0); enable_NMI_through_LVT0(); - legacy_pic->chip->unmask(0); + legacy_pic->unmask(0); } /* This must be done before setting cpu_online_mask */ -- cgit v1.2.3-70-g09d2 From d4eba29770244e7cc5e60c0977d73d84148a3d6d Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 24 Sep 2010 12:26:18 +0200 Subject: x86: Cleanup access to irq_data Fixup the open coded access to irq_desc->[handler_data|chip_data|msi-desc] Use the macros and inline functions for it. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/kernel/apic/io_apic.c | 63 +++++++++++++++++++++--------------------- arch/x86/kernel/uv_irq.c | 2 +- 2 files changed, 33 insertions(+), 32 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index e5ae2a22262..fa0d92a6db5 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -190,7 +190,7 @@ struct irq_cfg *irq_cfg(unsigned int irq) desc = irq_to_desc(irq); if (desc) - cfg = desc->chip_data; + cfg = get_irq_desc_chip_data(desc); return cfg; } @@ -219,10 +219,11 @@ int arch_init_chip_data(struct irq_desc *desc, int node) { struct irq_cfg *cfg; - cfg = desc->chip_data; + cfg = get_irq_desc_chip_data(desc); if (!cfg) { - desc->chip_data = get_one_free_irq_cfg(node); - if (!desc->chip_data) { + cfg = get_one_free_irq_cfg(node); + desc->chip_data = cfg; + if (!cfg) { printk(KERN_ERR "can not alloc irq_cfg\n"); BUG_ON(1); } @@ -325,8 +326,8 @@ void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc) { struct irq_cfg *old_cfg, *cfg; - old_cfg = old_desc->chip_data; - cfg = desc->chip_data; + old_cfg = get_irq_desc_chip_data(old_desc); + cfg = get_irq_desc_chip_data(desc); if (old_cfg == cfg) return; @@ -594,7 +595,7 @@ static void __mask_IO_APIC_irq(struct irq_cfg *cfg) static void mask_IO_APIC_irq_desc(struct irq_desc *desc) { - struct irq_cfg *cfg = desc->chip_data; + struct irq_cfg *cfg = get_irq_desc_chip_data(desc); unsigned long flags; BUG_ON(!cfg); @@ -606,7 +607,7 @@ static void mask_IO_APIC_irq_desc(struct irq_desc *desc) static void unmask_IO_APIC_irq_desc(struct irq_desc *desc) { - struct irq_cfg *cfg = desc->chip_data; + struct irq_cfg *cfg = get_irq_desc_chip_data(desc); unsigned long flags; raw_spin_lock_irqsave(&ioapic_lock, flags); @@ -1269,7 +1270,7 @@ void __setup_vector_irq(int cpu) raw_spin_lock(&vector_lock); /* Mark the inuse vectors */ for_each_irq_desc(irq, desc) { - cfg = desc->chip_data; + cfg = get_irq_desc_chip_data(desc); /* * If it is a legacy IRQ handled by the legacy PIC, this cpu @@ -1427,7 +1428,7 @@ static void setup_IO_APIC_irq(int apic_id, int pin, unsigned int irq, struct irq if (!IO_APIC_IRQ(irq)) return; - cfg = desc->chip_data; + cfg = get_irq_desc_chip_data(desc); /* * For legacy irqs, cfg->domain starts with cpu 0 for legacy @@ -1516,7 +1517,7 @@ static void __init setup_IO_APIC_irqs(void) printk(KERN_INFO "can not get irq_desc for %d\n", irq); continue; } - cfg = desc->chip_data; + cfg = get_irq_desc_chip_data(desc); add_pin_to_irq_node(cfg, node, apic_id, pin); /* * don't mark it in pin_programmed, so later acpi could @@ -1567,7 +1568,7 @@ void setup_IO_APIC_irq_extra(u32 gsi) return; } - cfg = desc->chip_data; + cfg = get_irq_desc_chip_data(desc); add_pin_to_irq_node(cfg, node, apic_id, pin); if (test_bit(pin, mp_ioapic_routing[apic_id].pin_programmed)) { @@ -1718,7 +1719,7 @@ __apicdebuginit(void) print_IO_APIC(void) for_each_irq_desc(irq, desc) { struct irq_pin_list *entry; - cfg = desc->chip_data; + cfg = get_irq_desc_chip_data(desc); if (!cfg) continue; entry = cfg->irq_2_pin; @@ -2323,7 +2324,7 @@ set_desc_affinity(struct irq_desc *desc, const struct cpumask *mask, return -1; irq = desc->irq; - cfg = desc->chip_data; + cfg = get_irq_desc_chip_data(desc); if (assign_irq_vector(irq, cfg, mask)) return -1; @@ -2343,7 +2344,7 @@ set_ioapic_affinity_irq_desc(struct irq_desc *desc, const struct cpumask *mask) int ret = -1; irq = desc->irq; - cfg = desc->chip_data; + cfg = get_irq_desc_chip_data(desc); raw_spin_lock_irqsave(&ioapic_lock, flags); ret = set_desc_affinity(desc, mask, &dest); @@ -2396,7 +2397,7 @@ migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask) if (get_irte(irq, &irte)) return ret; - cfg = desc->chip_data; + cfg = get_irq_desc_chip_data(desc); if (assign_irq_vector(irq, cfg, mask)) return ret; @@ -2500,7 +2501,7 @@ unlock: static void __irq_complete_move(struct irq_desc **descp, unsigned vector) { struct irq_desc *desc = *descp; - struct irq_cfg *cfg = desc->chip_data; + struct irq_cfg *cfg = get_irq_desc_chip_data(desc); unsigned me; if (likely(!cfg->move_in_progress)) @@ -2520,7 +2521,7 @@ static void irq_complete_move(struct irq_desc **descp) void irq_force_complete_move(int irq) { struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg = desc->chip_data; + struct irq_cfg *cfg = get_irq_desc_chip_data(desc); if (!cfg) return; @@ -2588,7 +2589,7 @@ static void eoi_ioapic_irq(struct irq_desc *desc) unsigned int irq; irq = desc->irq; - cfg = desc->chip_data; + cfg = get_irq_desc_chip_data(desc); raw_spin_lock_irqsave(&ioapic_lock, flags); __eoi_ioapic_irq(irq, cfg); @@ -2644,7 +2645,7 @@ static void ack_apic_level(unsigned int irq) * we use the above logic (mask+edge followed by unmask+level) from * Manfred Spraul to clear the remote IRR. */ - cfg = desc->chip_data; + cfg = get_irq_desc_chip_data(desc); i = cfg->vector; v = apic_read(APIC_TMR + ((i & ~0x1f) >> 1)); @@ -2695,7 +2696,7 @@ static void ack_apic_level(unsigned int irq) * accurate and is causing problems then it is a hardware bug * and you can go talk to the chipset vendor about it. */ - cfg = desc->chip_data; + cfg = get_irq_desc_chip_data(desc); if (!io_apic_level_ack_pending(cfg)) move_masked_irq(irq); unmask_IO_APIC_irq_desc(desc); @@ -2763,7 +2764,7 @@ static inline void init_IO_APIC_traps(void) * 0x80, because int 0x80 is hm, kind of importantish. ;) */ for_each_irq_desc(irq, desc) { - cfg = desc->chip_data; + cfg = get_irq_desc_chip_data(desc); if (IO_APIC_IRQ(irq) && cfg && !cfg->vector) { /* * Hmm.. We don't have an entry for this, @@ -2917,7 +2918,7 @@ int timer_through_8259 __initdata; static inline void __init check_timer(void) { struct irq_desc *desc = irq_to_desc(0); - struct irq_cfg *cfg = desc->chip_data; + struct irq_cfg *cfg = get_irq_desc_chip_data(desc); int node = cpu_to_node(0); int apic1, pin1, apic2, pin2; unsigned long flags; @@ -3250,13 +3251,13 @@ unsigned int create_irq_nr(unsigned int irq_want, int node) printk(KERN_INFO "can not get irq_desc for %d\n", new); continue; } - cfg_new = desc_new->chip_data; + cfg_new = get_irq_desc_chip_data(desc_new); if (cfg_new->vector != 0) continue; desc_new = move_irq_desc(desc_new, node); - cfg_new = desc_new->chip_data; + cfg_new = get_irq_desc_chip_data(desc_new); if (__assign_irq_vector(new, cfg_new, apic->target_cpus()) == 0) irq = new; @@ -3381,7 +3382,7 @@ static int set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask) if (set_desc_affinity(desc, mask, &dest)) return -1; - cfg = desc->chip_data; + cfg = get_irq_desc_chip_data(desc); __get_cached_msi_msg(desc->irq_data.msi_desc, &msg); @@ -3403,7 +3404,7 @@ static int ir_set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask) { struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg = desc->chip_data; + struct irq_cfg *cfg = get_irq_desc_chip_data(desc); unsigned int dest; struct irte irte; @@ -3595,7 +3596,7 @@ static int dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask) if (set_desc_affinity(desc, mask, &dest)) return -1; - cfg = desc->chip_data; + cfg = get_irq_desc_chip_data(desc); dmar_msi_read(irq, &msg); @@ -3650,7 +3651,7 @@ static int hpet_msi_set_affinity(unsigned int irq, const struct cpumask *mask) if (set_desc_affinity(desc, mask, &dest)) return -1; - cfg = desc->chip_data; + cfg = get_irq_desc_chip_data(desc); hpet_msi_read(irq, &msg); @@ -3756,7 +3757,7 @@ static int set_ht_irq_affinity(unsigned int irq, const struct cpumask *mask) if (set_desc_affinity(desc, mask, &dest)) return -1; - cfg = desc->chip_data; + cfg = get_irq_desc_chip_data(desc); target_ht_irq(irq, dest, cfg->vector); @@ -3903,7 +3904,7 @@ static int __io_apic_set_pci_routing(struct device *dev, int irq, * IRQs < 16 are already in the irq_2_pin[] map */ if (irq >= legacy_pic->nr_legacy_irqs) { - cfg = desc->chip_data; + cfg = get_irq_desc_chip_data(desc); if (add_pin_to_irq_node_nopanic(cfg, node, ioapic, pin)) { printk(KERN_INFO "can not add pin %d for irq %d\n", pin, irq); diff --git a/arch/x86/kernel/uv_irq.c b/arch/x86/kernel/uv_irq.c index 1132129db79..2233a42fb90 100644 --- a/arch/x86/kernel/uv_irq.c +++ b/arch/x86/kernel/uv_irq.c @@ -209,7 +209,7 @@ static void arch_disable_uv_irq(int mmr_pnode, unsigned long mmr_offset) static int uv_set_irq_affinity(unsigned int irq, const struct cpumask *mask) { struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg = desc->chip_data; + struct irq_cfg *cfg = get_irq_desc_chip_data(desc); unsigned int dest; unsigned long mmr_value; struct uv_IO_APIC_route_entry *entry; -- cgit v1.2.3-70-g09d2 From dd5f15e5cf104c9170b76ae3274f35b42a3e4161 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 28 Sep 2010 15:18:35 +0200 Subject: x86: Cleanup io_apic Sanitize functions. Remove irq_desc pointer magic. Preparatory patch for further cleanups. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/kernel/apic/io_apic.c | 109 ++++++++++++++++------------------------- 1 file changed, 42 insertions(+), 67 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index fa0d92a6db5..3c4dee8a9ef 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -572,11 +572,6 @@ static void __unmask_and_level_IO_APIC_irq(struct irq_pin_list *entry) IO_APIC_REDIR_LEVEL_TRIGGER, NULL); } -static void __unmask_IO_APIC_irq(struct irq_cfg *cfg) -{ - io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED, 0, NULL); -} - static void io_apic_sync(struct irq_pin_list *entry) { /* @@ -588,44 +583,41 @@ static void io_apic_sync(struct irq_pin_list *entry) readl(&io_apic->data); } -static void __mask_IO_APIC_irq(struct irq_cfg *cfg) +static void mask_ioapic(struct irq_cfg *cfg) { + unsigned long flags; + + raw_spin_lock_irqsave(&ioapic_lock, flags); io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, &io_apic_sync); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); } -static void mask_IO_APIC_irq_desc(struct irq_desc *desc) +static void mask_ioapic_irq(unsigned int irq) { - struct irq_cfg *cfg = get_irq_desc_chip_data(desc); - unsigned long flags; + struct irq_cfg *cfg = get_irq_chip_data(irq); - BUG_ON(!cfg); + mask_ioapic(cfg); +} - raw_spin_lock_irqsave(&ioapic_lock, flags); - __mask_IO_APIC_irq(cfg); - raw_spin_unlock_irqrestore(&ioapic_lock, flags); +static void __unmask_ioapic(struct irq_cfg *cfg) +{ + io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED, 0, NULL); } -static void unmask_IO_APIC_irq_desc(struct irq_desc *desc) +static void unmask_ioapic(struct irq_cfg *cfg) { - struct irq_cfg *cfg = get_irq_desc_chip_data(desc); unsigned long flags; raw_spin_lock_irqsave(&ioapic_lock, flags); - __unmask_IO_APIC_irq(cfg); + __unmask_ioapic(cfg); raw_spin_unlock_irqrestore(&ioapic_lock, flags); } -static void mask_IO_APIC_irq(unsigned int irq) +static void unmask_ioapic_irq(unsigned int irq) { - struct irq_desc *desc = irq_to_desc(irq); + struct irq_cfg *cfg = get_irq_chip_data(irq); - mask_IO_APIC_irq_desc(desc); -} -static void unmask_IO_APIC_irq(unsigned int irq) -{ - struct irq_desc *desc = irq_to_desc(irq); - - unmask_IO_APIC_irq_desc(desc); + unmask_ioapic(cfg); } static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin) @@ -2239,7 +2231,7 @@ static unsigned int startup_ioapic_irq(unsigned int irq) was_pending = 1; } cfg = irq_cfg(irq); - __unmask_IO_APIC_irq(cfg); + __unmask_ioapic(cfg); raw_spin_unlock_irqrestore(&ioapic_lock, flags); return was_pending; @@ -2498,10 +2490,8 @@ unlock: irq_exit(); } -static void __irq_complete_move(struct irq_desc **descp, unsigned vector) +static void __irq_complete_move(struct irq_cfg *cfg, unsigned vector) { - struct irq_desc *desc = *descp; - struct irq_cfg *cfg = get_irq_desc_chip_data(desc); unsigned me; if (likely(!cfg->move_in_progress)) @@ -2513,30 +2503,29 @@ static void __irq_complete_move(struct irq_desc **descp, unsigned vector) send_cleanup_vector(cfg); } -static void irq_complete_move(struct irq_desc **descp) +static void irq_complete_move(struct irq_cfg *cfg) { - __irq_complete_move(descp, ~get_irq_regs()->orig_ax); + __irq_complete_move(cfg, ~get_irq_regs()->orig_ax); } void irq_force_complete_move(int irq) { - struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg = get_irq_desc_chip_data(desc); + struct irq_cfg *cfg = get_irq_chip_data(irq); if (!cfg) return; - __irq_complete_move(&desc, cfg->vector); + __irq_complete_move(cfg, cfg->vector); } #else -static inline void irq_complete_move(struct irq_desc **descp) {} +static inline void irq_complete_move(struct irq_cfg *cfg) { } #endif static void ack_apic_edge(unsigned int irq) { - struct irq_desc *desc = irq_to_desc(irq); + struct irq_cfg *cfg = get_irq_chip_data(irq); - irq_complete_move(&desc); + irq_complete_move(cfg); move_native_irq(irq); ack_APIC_irq(); } @@ -2559,10 +2548,12 @@ atomic_t irq_mis_count; * Otherwise, we simulate the EOI message manually by changing the trigger * mode to edge and then back to level, with RTE being masked during this. */ -static void __eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg) +static void eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg) { struct irq_pin_list *entry; + unsigned long flags; + raw_spin_lock_irqsave(&ioapic_lock, flags); for_each_irq_pin(entry, cfg->irq_2_pin) { if (mp_ioapics[entry->apic].apicver >= 0x20) { /* @@ -2580,36 +2571,22 @@ static void __eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg) __unmask_and_level_IO_APIC_irq(entry); } } -} - -static void eoi_ioapic_irq(struct irq_desc *desc) -{ - struct irq_cfg *cfg; - unsigned long flags; - unsigned int irq; - - irq = desc->irq; - cfg = get_irq_desc_chip_data(desc); - - raw_spin_lock_irqsave(&ioapic_lock, flags); - __eoi_ioapic_irq(irq, cfg); raw_spin_unlock_irqrestore(&ioapic_lock, flags); } static void ack_apic_level(unsigned int irq) { struct irq_desc *desc = irq_to_desc(irq); + struct irq_cfg *cfg = get_irq_desc_chip_data(desc); + int i, do_unmask_irq = 0; unsigned long v; - int i; - struct irq_cfg *cfg; - int do_unmask_irq = 0; - irq_complete_move(&desc); + irq_complete_move(cfg); #ifdef CONFIG_GENERIC_PENDING_IRQ /* If we are moving the irq we need to mask it */ if (unlikely(desc->status & IRQ_MOVE_PENDING)) { do_unmask_irq = 1; - mask_IO_APIC_irq_desc(desc); + mask_ioapic(cfg); } #endif @@ -2645,7 +2622,6 @@ static void ack_apic_level(unsigned int irq) * we use the above logic (mask+edge followed by unmask+level) from * Manfred Spraul to clear the remote IRR. */ - cfg = get_irq_desc_chip_data(desc); i = cfg->vector; v = apic_read(APIC_TMR + ((i & ~0x1f) >> 1)); @@ -2665,7 +2641,7 @@ static void ack_apic_level(unsigned int irq) if (!(v & (1 << (i & 0x1f)))) { atomic_inc(&irq_mis_count); - eoi_ioapic_irq(desc); + eoi_ioapic_irq(irq, cfg); } /* Now we can move and renable the irq */ @@ -2696,10 +2672,9 @@ static void ack_apic_level(unsigned int irq) * accurate and is causing problems then it is a hardware bug * and you can go talk to the chipset vendor about it. */ - cfg = get_irq_desc_chip_data(desc); if (!io_apic_level_ack_pending(cfg)) move_masked_irq(irq); - unmask_IO_APIC_irq_desc(desc); + unmask_ioapic(cfg); } } @@ -2711,18 +2686,18 @@ static void ir_ack_apic_edge(unsigned int irq) static void ir_ack_apic_level(unsigned int irq) { - struct irq_desc *desc = irq_to_desc(irq); + struct irq_cfg *cfg = get_irq_chip_data(irq); ack_APIC_irq(); - eoi_ioapic_irq(desc); + eoi_ioapic_irq(irq, cfg); } #endif /* CONFIG_INTR_REMAP */ static struct irq_chip ioapic_chip __read_mostly = { .name = "IO-APIC", .startup = startup_ioapic_irq, - .mask = mask_IO_APIC_irq, - .unmask = unmask_IO_APIC_irq, + .mask = mask_ioapic_irq, + .unmask = unmask_ioapic_irq, .ack = ack_apic_edge, .eoi = ack_apic_level, #ifdef CONFIG_SMP @@ -2734,8 +2709,8 @@ static struct irq_chip ioapic_chip __read_mostly = { static struct irq_chip ir_ioapic_chip __read_mostly = { .name = "IR-IO-APIC", .startup = startup_ioapic_irq, - .mask = mask_IO_APIC_irq, - .unmask = unmask_IO_APIC_irq, + .mask = mask_ioapic_irq, + .unmask = unmask_ioapic_irq, #ifdef CONFIG_INTR_REMAP .ack = ir_ack_apic_edge, .eoi = ir_ack_apic_level, @@ -2996,7 +2971,7 @@ static inline void __init check_timer(void) int idx; idx = find_irq_entry(apic1, pin1, mp_INT); if (idx != -1 && irq_trigger(idx)) - unmask_IO_APIC_irq_desc(desc); + unmask_ioapic(cfg); } if (timer_irq_works()) { if (nmi_watchdog == NMI_IO_APIC) { -- cgit v1.2.3-70-g09d2 From 61a38ce3f59cdb4654e9444329195bd57c3baf74 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 28 Sep 2010 16:00:34 +0200 Subject: x86: io_apic: Convert startup to new irq_chip function Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/kernel/apic/io_apic.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 3c4dee8a9ef..5ced690b849 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -2218,11 +2218,10 @@ static int __init timer_irq_works(void) * an edge even if it isn't on the 8259A... */ -static unsigned int startup_ioapic_irq(unsigned int irq) +static unsigned int startup_ioapic_irq(struct irq_data *data) { - int was_pending = 0; + int was_pending = 0, irq = data->irq; unsigned long flags; - struct irq_cfg *cfg; raw_spin_lock_irqsave(&ioapic_lock, flags); if (irq < legacy_pic->nr_legacy_irqs) { @@ -2230,8 +2229,7 @@ static unsigned int startup_ioapic_irq(unsigned int irq) if (legacy_pic->irq_pending(irq)) was_pending = 1; } - cfg = irq_cfg(irq); - __unmask_ioapic(cfg); + __unmask_ioapic(data->chip_data); raw_spin_unlock_irqrestore(&ioapic_lock, flags); return was_pending; @@ -2695,7 +2693,7 @@ static void ir_ack_apic_level(unsigned int irq) static struct irq_chip ioapic_chip __read_mostly = { .name = "IO-APIC", - .startup = startup_ioapic_irq, + .irq_startup = startup_ioapic_irq, .mask = mask_ioapic_irq, .unmask = unmask_ioapic_irq, .ack = ack_apic_edge, @@ -2708,7 +2706,7 @@ static struct irq_chip ioapic_chip __read_mostly = { static struct irq_chip ir_ioapic_chip __read_mostly = { .name = "IR-IO-APIC", - .startup = startup_ioapic_irq, + .irq_startup = startup_ioapic_irq, .mask = mask_ioapic_irq, .unmask = unmask_ioapic_irq, #ifdef CONFIG_INTR_REMAP -- cgit v1.2.3-70-g09d2 From 90297c5fe71d32a2a0ead38bd8f6b1112a2e5ac0 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 28 Sep 2010 16:03:54 +0200 Subject: x86: ioapic: Convert mask to new irq_chip function Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/kernel/apic/io_apic.c | 95 +++++++++++++++++++----------------------- 1 file changed, 43 insertions(+), 52 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 5ced690b849..b8b013f0cfd 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -592,11 +592,9 @@ static void mask_ioapic(struct irq_cfg *cfg) raw_spin_unlock_irqrestore(&ioapic_lock, flags); } -static void mask_ioapic_irq(unsigned int irq) +static void mask_ioapic_irq(struct irq_data *data) { - struct irq_cfg *cfg = get_irq_chip_data(irq); - - mask_ioapic(cfg); + mask_ioapic(data->chip_data); } static void __unmask_ioapic(struct irq_cfg *cfg) @@ -613,11 +611,9 @@ static void unmask_ioapic(struct irq_cfg *cfg) raw_spin_unlock_irqrestore(&ioapic_lock, flags); } -static void unmask_ioapic_irq(unsigned int irq) +static void unmask_ioapic_irq(struct irq_data *data) { - struct irq_cfg *cfg = get_irq_chip_data(irq); - - unmask_ioapic(cfg); + unmask_ioapic(data->chip_data); } static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin) @@ -2235,10 +2231,9 @@ static unsigned int startup_ioapic_irq(struct irq_data *data) return was_pending; } -static int ioapic_retrigger_irq(unsigned int irq) +static int ioapic_retrigger_irq(struct irq_data *data) { - - struct irq_cfg *cfg = irq_cfg(irq); + struct irq_cfg *cfg = data->chip_data; unsigned long flags; raw_spin_lock_irqsave(&vector_lock, flags); @@ -2519,12 +2514,10 @@ void irq_force_complete_move(int irq) static inline void irq_complete_move(struct irq_cfg *cfg) { } #endif -static void ack_apic_edge(unsigned int irq) +static void ack_apic_edge(struct irq_data *data) { - struct irq_cfg *cfg = get_irq_chip_data(irq); - - irq_complete_move(cfg); - move_native_irq(irq); + irq_complete_move(data->chip_data); + move_native_irq(data->irq); ack_APIC_irq(); } @@ -2572,11 +2565,11 @@ static void eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg) raw_spin_unlock_irqrestore(&ioapic_lock, flags); } -static void ack_apic_level(unsigned int irq) +static void ack_apic_level(struct irq_data *data) { + struct irq_cfg *cfg = data->chip_data; + int i, do_unmask_irq = 0, irq = data->irq; struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg = get_irq_desc_chip_data(desc); - int i, do_unmask_irq = 0; unsigned long v; irq_complete_move(cfg); @@ -2677,46 +2670,44 @@ static void ack_apic_level(unsigned int irq) } #ifdef CONFIG_INTR_REMAP -static void ir_ack_apic_edge(unsigned int irq) +static void ir_ack_apic_edge(struct irq_data *data) { ack_APIC_irq(); } -static void ir_ack_apic_level(unsigned int irq) +static void ir_ack_apic_level(struct irq_data *data) { - struct irq_cfg *cfg = get_irq_chip_data(irq); - ack_APIC_irq(); - eoi_ioapic_irq(irq, cfg); + eoi_ioapic_irq(data->irq, data->chip_data); } #endif /* CONFIG_INTR_REMAP */ static struct irq_chip ioapic_chip __read_mostly = { .name = "IO-APIC", .irq_startup = startup_ioapic_irq, - .mask = mask_ioapic_irq, - .unmask = unmask_ioapic_irq, - .ack = ack_apic_edge, - .eoi = ack_apic_level, + .irq_mask = mask_ioapic_irq, + .irq_unmask = unmask_ioapic_irq, + .irq_ack = ack_apic_edge, + .irq_eoi = ack_apic_level, #ifdef CONFIG_SMP .set_affinity = set_ioapic_affinity_irq, #endif - .retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; static struct irq_chip ir_ioapic_chip __read_mostly = { .name = "IR-IO-APIC", .irq_startup = startup_ioapic_irq, - .mask = mask_ioapic_irq, - .unmask = unmask_ioapic_irq, + .irq_mask = mask_ioapic_irq, + .irq_unmask = unmask_ioapic_irq, #ifdef CONFIG_INTR_REMAP - .ack = ir_ack_apic_edge, - .eoi = ir_ack_apic_level, + .irq_ack = ir_ack_apic_edge, + .irq_eoi = ir_ack_apic_level, #ifdef CONFIG_SMP .set_affinity = set_ir_ioapic_affinity_irq, #endif #endif - .retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; static inline void init_IO_APIC_traps(void) @@ -2757,7 +2748,7 @@ static inline void init_IO_APIC_traps(void) * The local APIC irq-chip implementation: */ -static void mask_lapic_irq(unsigned int irq) +static void mask_lapic_irq(struct irq_data *data) { unsigned long v; @@ -2765,7 +2756,7 @@ static void mask_lapic_irq(unsigned int irq) apic_write(APIC_LVT0, v | APIC_LVT_MASKED); } -static void unmask_lapic_irq(unsigned int irq) +static void unmask_lapic_irq(struct irq_data *data) { unsigned long v; @@ -2773,16 +2764,16 @@ static void unmask_lapic_irq(unsigned int irq) apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED); } -static void ack_lapic_irq(unsigned int irq) +static void ack_lapic_irq(struct irq_data *data) { ack_APIC_irq(); } static struct irq_chip lapic_chip __read_mostly = { .name = "local-APIC", - .mask = mask_lapic_irq, - .unmask = unmask_lapic_irq, - .ack = ack_lapic_irq, + .irq_mask = mask_lapic_irq, + .irq_unmask = unmask_lapic_irq, + .irq_ack = ack_lapic_irq, }; static void lapic_register_intr(int irq, struct irq_desc *desc) @@ -3417,11 +3408,11 @@ static struct irq_chip msi_chip = { .name = "PCI-MSI", .irq_unmask = unmask_msi_irq, .irq_mask = mask_msi_irq, - .ack = ack_apic_edge, + .irq_ack = ack_apic_edge, #ifdef CONFIG_SMP .set_affinity = set_msi_irq_affinity, #endif - .retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; static struct irq_chip msi_ir_chip = { @@ -3429,12 +3420,12 @@ static struct irq_chip msi_ir_chip = { .irq_unmask = unmask_msi_irq, .irq_mask = mask_msi_irq, #ifdef CONFIG_INTR_REMAP - .ack = ir_ack_apic_edge, + .irq_ack = ir_ack_apic_edge, #ifdef CONFIG_SMP .set_affinity = ir_set_msi_irq_affinity, #endif #endif - .retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; /* @@ -3589,11 +3580,11 @@ static struct irq_chip dmar_msi_type = { .name = "DMAR_MSI", .unmask = dmar_msi_unmask, .mask = dmar_msi_mask, - .ack = ack_apic_edge, + .irq_ack = ack_apic_edge, #ifdef CONFIG_SMP .set_affinity = dmar_msi_set_affinity, #endif - .retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; int arch_setup_dmar_msi(unsigned int irq) @@ -3645,23 +3636,23 @@ static struct irq_chip ir_hpet_msi_type = { .unmask = hpet_msi_unmask, .mask = hpet_msi_mask, #ifdef CONFIG_INTR_REMAP - .ack = ir_ack_apic_edge, + .irq_ack = ir_ack_apic_edge, #ifdef CONFIG_SMP .set_affinity = ir_set_msi_irq_affinity, #endif #endif - .retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; static struct irq_chip hpet_msi_type = { .name = "HPET_MSI", .unmask = hpet_msi_unmask, .mask = hpet_msi_mask, - .ack = ack_apic_edge, + .irq_ack = ack_apic_edge, #ifdef CONFIG_SMP .set_affinity = hpet_msi_set_affinity, #endif - .retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; int arch_setup_hpet_msi(unsigned int irq, unsigned int id) @@ -3743,11 +3734,11 @@ static struct irq_chip ht_irq_chip = { .name = "PCI-HT", .mask = mask_ht_irq, .unmask = unmask_ht_irq, - .ack = ack_apic_edge, + .irq_ack = ack_apic_edge, #ifdef CONFIG_SMP .set_affinity = set_ht_irq_affinity, #endif - .retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) -- cgit v1.2.3-70-g09d2 From d0fbca8f9304d1760fdc906b35b06e89fbdbb51f Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 28 Sep 2010 16:18:39 +0200 Subject: x86: ioapic/hpet: Convert to new chip functions Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/include/asm/hpet.h | 10 ++++++---- arch/x86/kernel/apic/io_apic.c | 30 ++++++++++++++---------------- arch/x86/kernel/hpet.c | 16 ++++++---------- 3 files changed, 26 insertions(+), 30 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/hpet.h b/arch/x86/include/asm/hpet.h index 1d5c08a1bdf..2c392d663dc 100644 --- a/arch/x86/include/asm/hpet.h +++ b/arch/x86/include/asm/hpet.h @@ -74,10 +74,12 @@ extern void hpet_disable(void); extern unsigned int hpet_readl(unsigned int a); extern void force_hpet_resume(void); -extern void hpet_msi_unmask(unsigned int irq); -extern void hpet_msi_mask(unsigned int irq); -extern void hpet_msi_write(unsigned int irq, struct msi_msg *msg); -extern void hpet_msi_read(unsigned int irq, struct msi_msg *msg); +struct irq_data; +extern void hpet_msi_unmask(struct irq_data *data); +extern void hpet_msi_mask(struct irq_data *data); +struct hpet_dev; +extern void hpet_msi_write(struct hpet_dev *hdev, struct msi_msg *msg); +extern void hpet_msi_read(struct hpet_dev *hdev, struct msi_msg *msg); #ifdef CONFIG_PCI_MSI extern int arch_setup_hpet_msi(unsigned int irq, unsigned int id); diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index b8b013f0cfd..49aa857ff00 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -3605,26 +3605,25 @@ int arch_setup_dmar_msi(unsigned int irq) #ifdef CONFIG_HPET_TIMER #ifdef CONFIG_SMP -static int hpet_msi_set_affinity(unsigned int irq, const struct cpumask *mask) +static int hpet_msi_set_affinity(struct irq_data *data, + const struct cpumask *mask, bool force) { - struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg; + struct irq_desc *desc = irq_to_desc(data->irq); + struct irq_cfg *cfg = data->chip_data; struct msi_msg msg; unsigned int dest; if (set_desc_affinity(desc, mask, &dest)) return -1; - cfg = get_irq_desc_chip_data(desc); - - hpet_msi_read(irq, &msg); + hpet_msi_read(data->handler_data, &msg); msg.data &= ~MSI_DATA_VECTOR_MASK; msg.data |= MSI_DATA_VECTOR(cfg->vector); msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK; msg.address_lo |= MSI_ADDR_DEST_ID(dest); - hpet_msi_write(irq, &msg); + hpet_msi_write(data->handler_data, &msg); return 0; } @@ -3633,8 +3632,8 @@ static int hpet_msi_set_affinity(unsigned int irq, const struct cpumask *mask) static struct irq_chip ir_hpet_msi_type = { .name = "IR-HPET_MSI", - .unmask = hpet_msi_unmask, - .mask = hpet_msi_mask, + .irq_unmask = hpet_msi_unmask, + .irq_mask = hpet_msi_mask, #ifdef CONFIG_INTR_REMAP .irq_ack = ir_ack_apic_edge, #ifdef CONFIG_SMP @@ -3646,20 +3645,19 @@ static struct irq_chip ir_hpet_msi_type = { static struct irq_chip hpet_msi_type = { .name = "HPET_MSI", - .unmask = hpet_msi_unmask, - .mask = hpet_msi_mask, + .irq_unmask = hpet_msi_unmask, + .irq_mask = hpet_msi_mask, .irq_ack = ack_apic_edge, #ifdef CONFIG_SMP - .set_affinity = hpet_msi_set_affinity, + .irq_set_affinity = hpet_msi_set_affinity, #endif .irq_retrigger = ioapic_retrigger_irq, }; int arch_setup_hpet_msi(unsigned int irq, unsigned int id) { - int ret; struct msi_msg msg; - struct irq_desc *desc = irq_to_desc(irq); + int ret; if (intr_remapping_enabled) { struct intel_iommu *iommu = map_hpet_to_ir(id); @@ -3677,8 +3675,8 @@ int arch_setup_hpet_msi(unsigned int irq, unsigned int id) if (ret < 0) return ret; - hpet_msi_write(irq, &msg); - desc->status |= IRQ_MOVE_PCNTXT; + hpet_msi_write(get_irq_data(irq), &msg); + irq_set_status_flags(irq,IRQ_MOVE_PCNTXT); if (irq_remapped(irq)) set_irq_chip_and_handler_name(irq, &ir_hpet_msi_type, handle_edge_irq, "edge"); diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index 7494999141b..efaf906daf9 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -440,9 +440,9 @@ static int hpet_legacy_next_event(unsigned long delta, static DEFINE_PER_CPU(struct hpet_dev *, cpu_hpet_dev); static struct hpet_dev *hpet_devs; -void hpet_msi_unmask(unsigned int irq) +void hpet_msi_unmask(struct irq_data *data) { - struct hpet_dev *hdev = get_irq_data(irq); + struct hpet_dev *hdev = data->handler_data; unsigned int cfg; /* unmask it */ @@ -451,10 +451,10 @@ void hpet_msi_unmask(unsigned int irq) hpet_writel(cfg, HPET_Tn_CFG(hdev->num)); } -void hpet_msi_mask(unsigned int irq) +void hpet_msi_mask(struct irq_data *data) { + struct hpet_dev *hdev = data->handler_data; unsigned int cfg; - struct hpet_dev *hdev = get_irq_data(irq); /* mask it */ cfg = hpet_readl(HPET_Tn_CFG(hdev->num)); @@ -462,18 +462,14 @@ void hpet_msi_mask(unsigned int irq) hpet_writel(cfg, HPET_Tn_CFG(hdev->num)); } -void hpet_msi_write(unsigned int irq, struct msi_msg *msg) +void hpet_msi_write(struct hpet_dev *hdev, struct msi_msg *msg) { - struct hpet_dev *hdev = get_irq_data(irq); - hpet_writel(msg->data, HPET_Tn_ROUTE(hdev->num)); hpet_writel(msg->address_lo, HPET_Tn_ROUTE(hdev->num) + 4); } -void hpet_msi_read(unsigned int irq, struct msi_msg *msg) +void hpet_msi_read(struct hpet_dev *hdev, struct msi_msg *msg) { - struct hpet_dev *hdev = get_irq_data(irq); - msg->data = hpet_readl(HPET_Tn_ROUTE(hdev->num)); msg->address_lo = hpet_readl(HPET_Tn_ROUTE(hdev->num) + 4); msg->address_hi = 0; -- cgit v1.2.3-70-g09d2 From 5c2837fbaa609e615ef9a1c58a4cd26ce90be35b Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 28 Sep 2010 17:15:11 +0200 Subject: dmar: Convert to new irq chip functions Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar Acked-by: David Woodhouse --- arch/ia64/kernel/msi_ia64.c | 4 ++-- arch/x86/kernel/apic/io_apic.c | 4 ++-- drivers/pci/dmar.c | 8 ++++---- include/linux/dmar.h | 5 +++-- 4 files changed, 11 insertions(+), 10 deletions(-) (limited to 'arch/x86') diff --git a/arch/ia64/kernel/msi_ia64.c b/arch/ia64/kernel/msi_ia64.c index dbb43424704..00b19a416ea 100644 --- a/arch/ia64/kernel/msi_ia64.c +++ b/arch/ia64/kernel/msi_ia64.c @@ -160,8 +160,8 @@ static int dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask) static struct irq_chip dmar_msi_type = { .name = "DMAR_MSI", - .unmask = dmar_msi_unmask, - .mask = dmar_msi_mask, + .irq_unmask = dmar_msi_unmask, + .irq_mask = dmar_msi_mask, .ack = ia64_ack_msi_irq, #ifdef CONFIG_SMP .set_affinity = dmar_msi_set_affinity, diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 49aa857ff00..72b253ef310 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -3578,8 +3578,8 @@ static int dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask) static struct irq_chip dmar_msi_type = { .name = "DMAR_MSI", - .unmask = dmar_msi_unmask, - .mask = dmar_msi_mask, + .irq_unmask = dmar_msi_unmask, + .irq_mask = dmar_msi_mask, .irq_ack = ack_apic_edge, #ifdef CONFIG_SMP .set_affinity = dmar_msi_set_affinity, diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c index 0a19708074c..3de3a436a43 100644 --- a/drivers/pci/dmar.c +++ b/drivers/pci/dmar.c @@ -1221,9 +1221,9 @@ const char *dmar_get_fault_reason(u8 fault_reason, int *fault_type) } } -void dmar_msi_unmask(unsigned int irq) +void dmar_msi_unmask(struct irq_data *data) { - struct intel_iommu *iommu = get_irq_data(irq); + struct intel_iommu *iommu = irq_data_get_irq_data(data); unsigned long flag; /* unmask it */ @@ -1234,10 +1234,10 @@ void dmar_msi_unmask(unsigned int irq) spin_unlock_irqrestore(&iommu->register_lock, flag); } -void dmar_msi_mask(unsigned int irq) +void dmar_msi_mask(struct irq_data *data) { unsigned long flag; - struct intel_iommu *iommu = get_irq_data(irq); + struct intel_iommu *iommu = irq_data_get_irq_data(data); /* mask it */ spin_lock_irqsave(&iommu->register_lock, flag); diff --git a/include/linux/dmar.h b/include/linux/dmar.h index d7cecc90ed3..cb86aa1ca43 100644 --- a/include/linux/dmar.h +++ b/include/linux/dmar.h @@ -187,8 +187,9 @@ static inline int set_msi_sid(struct irte *irte, struct pci_dev *dev) /* Can't use the common MSI interrupt functions * since DMAR is not a pci device */ -extern void dmar_msi_unmask(unsigned int irq); -extern void dmar_msi_mask(unsigned int irq); +struct irq_data; +extern void dmar_msi_unmask(struct irq_data *data); +extern void dmar_msi_mask(struct irq_data *data); extern void dmar_msi_read(int irq, struct msi_msg *msg); extern void dmar_msi_write(int irq, struct msi_msg *msg); extern int dmar_set_interrupt(struct intel_iommu *iommu); -- cgit v1.2.3-70-g09d2 From e9f7ac664bfc36685a8eb3315ec21c067d0cee36 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 28 Sep 2010 17:22:09 +0200 Subject: ht: Convert to new irq_chip functions Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar Cc: Jesse Barnes --- arch/x86/kernel/apic/io_apic.c | 4 ++-- drivers/pci/htirq.c | 22 ++++++++-------------- include/linux/htirq.h | 5 +++-- 3 files changed, 13 insertions(+), 18 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 72b253ef310..5579f3f5943 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -3730,8 +3730,8 @@ static int set_ht_irq_affinity(unsigned int irq, const struct cpumask *mask) static struct irq_chip ht_irq_chip = { .name = "PCI-HT", - .mask = mask_ht_irq, - .unmask = unmask_ht_irq, + .irq_mask = mask_ht_irq, + .irq_unmask = unmask_ht_irq, .irq_ack = ack_apic_edge, #ifdef CONFIG_SMP .set_affinity = set_ht_irq_affinity, diff --git a/drivers/pci/htirq.c b/drivers/pci/htirq.c index 98abf8b9129..834842aa5bb 100644 --- a/drivers/pci/htirq.c +++ b/drivers/pci/htirq.c @@ -57,28 +57,22 @@ void fetch_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg) *msg = cfg->msg; } -void mask_ht_irq(unsigned int irq) +void mask_ht_irq(struct irq_data *data) { - struct ht_irq_cfg *cfg; - struct ht_irq_msg msg; - - cfg = get_irq_data(irq); + struct ht_irq_cfg *cfg = irq_data_get_irq_data(data); + struct ht_irq_msg msg = cfg->msg; - msg = cfg->msg; msg.address_lo |= 1; - write_ht_irq_msg(irq, &msg); + write_ht_irq_msg(data->irq, &msg); } -void unmask_ht_irq(unsigned int irq) +void unmask_ht_irq(struct irq_data *data) { - struct ht_irq_cfg *cfg; - struct ht_irq_msg msg; - - cfg = get_irq_data(irq); + struct ht_irq_cfg *cfg = irq_data_get_irq_data(data); + struct ht_irq_msg msg = cfg->msg; - msg = cfg->msg; msg.address_lo &= ~1; - write_ht_irq_msg(irq, &msg); + write_ht_irq_msg(data->irq, &msg); } /** diff --git a/include/linux/htirq.h b/include/linux/htirq.h index c96ea46737d..70a1dbbf209 100644 --- a/include/linux/htirq.h +++ b/include/linux/htirq.h @@ -9,8 +9,9 @@ struct ht_irq_msg { /* Helper functions.. */ void fetch_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg); void write_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg); -void mask_ht_irq(unsigned int irq); -void unmask_ht_irq(unsigned int irq); +struct irq_data; +void mask_ht_irq(struct irq_data *data); +void unmask_ht_irq(struct irq_data *data); /* The arch hook for getting things started */ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev); -- cgit v1.2.3-70-g09d2 From 60c69948e5b6357ac0d5ef2a2d0ce31c173c3c64 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 28 Sep 2010 17:28:38 +0200 Subject: x86: ioapic: Clean up the direct access to irq_desc Most of it is useless pseudo optimization. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/kernel/apic/io_apic.c | 79 +++++++++++++++--------------------------- 1 file changed, 28 insertions(+), 51 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 5579f3f5943..82c3c66e333 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -150,10 +150,7 @@ static struct irq_cfg irq_cfgx[NR_IRQS]; int __init arch_early_irq_init(void) { struct irq_cfg *cfg; - struct irq_desc *desc; - int count; - int node; - int i; + int count, node, i; if (!legacy_pic->nr_legacy_irqs) { nr_irqs_gsi = 0; @@ -165,8 +162,7 @@ int __init arch_early_irq_init(void) node = cpu_to_node(0); for (i = 0; i < count; i++) { - desc = irq_to_desc(i); - desc->chip_data = &cfg[i]; + set_irq_chip_data(i, &cfg[i]); zalloc_cpumask_var_node(&cfg[i].domain, GFP_NOWAIT, node); zalloc_cpumask_var_node(&cfg[i].old_domain, GFP_NOWAIT, node); /* @@ -185,14 +181,7 @@ int __init arch_early_irq_init(void) #ifdef CONFIG_SPARSE_IRQ struct irq_cfg *irq_cfg(unsigned int irq) { - struct irq_cfg *cfg = NULL; - struct irq_desc *desc; - - desc = irq_to_desc(irq); - if (desc) - cfg = get_irq_desc_chip_data(desc); - - return cfg; + return get_irq_chip_data(irq); } static struct irq_cfg *get_one_free_irq_cfg(int node) @@ -1316,17 +1305,17 @@ static inline int IO_APIC_irq_trigger(int irq) } #endif -static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long trigger) +static void ioapic_register_intr(unsigned int irq, unsigned long trigger) { if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) || trigger == IOAPIC_LEVEL) - desc->status |= IRQ_LEVEL; + irq_set_status_flags(irq, IRQ_LEVEL); else - desc->status &= ~IRQ_LEVEL; + irq_clear_status_flags(irq, IRQ_LEVEL); if (irq_remapped(irq)) { - desc->status |= IRQ_MOVE_PCNTXT; + irq_set_status_flags(irq, IRQ_MOVE_PCNTXT); if (trigger) set_irq_chip_and_handler_name(irq, &ir_ioapic_chip, handle_fasteoi_irq, @@ -1406,18 +1395,14 @@ int setup_ioapic_entry(int apic_id, int irq, return 0; } -static void setup_IO_APIC_irq(int apic_id, int pin, unsigned int irq, struct irq_desc *desc, - int trigger, int polarity) +static void setup_ioapic_irq(int apic_id, int pin, unsigned int irq, + struct irq_cfg *cfg, int trigger, int polarity) { - struct irq_cfg *cfg; struct IO_APIC_route_entry entry; unsigned int dest; if (!IO_APIC_IRQ(irq)) return; - - cfg = get_irq_desc_chip_data(desc); - /* * For legacy irqs, cfg->domain starts with cpu 0 for legacy * controllers like 8259. Now that IO-APIC can handle this irq, update @@ -1446,7 +1431,7 @@ static void setup_IO_APIC_irq(int apic_id, int pin, unsigned int irq, struct irq return; } - ioapic_register_intr(irq, desc, trigger); + ioapic_register_intr(irq, trigger); if (irq < legacy_pic->nr_legacy_irqs) legacy_pic->mask(irq); @@ -1511,8 +1496,8 @@ static void __init setup_IO_APIC_irqs(void) * don't mark it in pin_programmed, so later acpi could * set it correctly when irq < 16 */ - setup_IO_APIC_irq(apic_id, pin, irq, desc, - irq_trigger(idx), irq_polarity(idx)); + setup_ioapic_irq(apic_id, pin, irq, cfg, irq_trigger(idx), + irq_polarity(idx)); } if (notcon) @@ -1566,7 +1551,7 @@ void setup_IO_APIC_irq_extra(u32 gsi) } set_bit(pin, mp_ioapic_routing[apic_id].pin_programmed); - setup_IO_APIC_irq(apic_id, pin, irq, desc, + setup_ioapic_irq(apic_id, pin, irq, cfg, irq_trigger(idx), irq_polarity(idx)); } @@ -2776,9 +2761,9 @@ static struct irq_chip lapic_chip __read_mostly = { .irq_ack = ack_lapic_irq, }; -static void lapic_register_intr(int irq, struct irq_desc *desc) +static void lapic_register_intr(int irq) { - desc->status &= ~IRQ_LEVEL; + irq_clear_status_flags(irq, IRQ_LEVEL); set_irq_chip_and_handler_name(irq, &lapic_chip, handle_edge_irq, "edge"); } @@ -2881,8 +2866,7 @@ int timer_through_8259 __initdata; */ static inline void __init check_timer(void) { - struct irq_desc *desc = irq_to_desc(0); - struct irq_cfg *cfg = get_irq_desc_chip_data(desc); + struct irq_cfg *cfg = get_irq_chip_data(0); int node = cpu_to_node(0); int apic1, pin1, apic2, pin2; unsigned long flags; @@ -2952,7 +2936,7 @@ static inline void __init check_timer(void) add_pin_to_irq_node(cfg, node, apic1, pin1); setup_timer_IRQ0_pin(apic1, pin1, cfg->vector); } else { - /* for edge trigger, setup_IO_APIC_irq already + /* for edge trigger, setup_ioapic_irq already * leave it unmasked. * so only need to unmask if it is level-trigger * do we really have level trigger timer? @@ -3020,7 +3004,7 @@ static inline void __init check_timer(void) apic_printk(APIC_QUIET, KERN_INFO "...trying to set up timer as Virtual Wire IRQ...\n"); - lapic_register_intr(0, desc); + lapic_register_intr(0); apic_write(APIC_LVT0, APIC_DM_FIXED | cfg->vector); /* Fixed mode */ legacy_pic->unmask(0); @@ -3457,8 +3441,8 @@ static int msi_alloc_irte(struct pci_dev *dev, int irq, int nvec) static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq) { - int ret; struct msi_msg msg; + int ret; ret = msi_compose_msg(dev, irq, &msg, -1); if (ret < 0) @@ -3468,11 +3452,7 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq) write_msi_msg(irq, &msg); if (irq_remapped(irq)) { - struct irq_desc *desc = irq_to_desc(irq); - /* - * irq migration in process context - */ - desc->status |= IRQ_MOVE_PCNTXT; + irq_set_status_flags(irq, IRQ_MOVE_PCNTXT); set_irq_chip_and_handler_name(irq, &msi_ir_chip, handle_edge_irq, "edge"); } else set_irq_chip_and_handler_name(irq, &msi_chip, handle_edge_irq, "edge"); @@ -3484,13 +3464,10 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq) int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) { - unsigned int irq; - int ret, sub_handle; + int node, ret, sub_handle, index = 0; + unsigned int irq, irq_want; struct msi_desc *msidesc; - unsigned int irq_want; struct intel_iommu *iommu = NULL; - int index = 0; - int node; /* x86 doesn't support multiple MSI yet */ if (type == PCI_CAP_ID_MSI && nvec > 1) @@ -3676,7 +3653,7 @@ int arch_setup_hpet_msi(unsigned int irq, unsigned int id) return ret; hpet_msi_write(get_irq_data(irq), &msg); - irq_set_status_flags(irq,IRQ_MOVE_PCNTXT); + irq_set_status_flags(irq, IRQ_MOVE_PCNTXT); if (irq_remapped(irq)) set_irq_chip_and_handler_name(irq, &ir_hpet_msi_type, handle_edge_irq, "edge"); @@ -3862,11 +3839,12 @@ static int __io_apic_set_pci_routing(struct device *dev, int irq, trigger = irq_attr->trigger; polarity = irq_attr->polarity; + cfg = get_irq_desc_chip_data(desc); + /* * IRQs < 16 are already in the irq_2_pin[] map */ if (irq >= legacy_pic->nr_legacy_irqs) { - cfg = get_irq_desc_chip_data(desc); if (add_pin_to_irq_node_nopanic(cfg, node, ioapic, pin)) { printk(KERN_INFO "can not add pin %d for irq %d\n", pin, irq); @@ -3874,7 +3852,7 @@ static int __io_apic_set_pci_routing(struct device *dev, int irq, } } - setup_IO_APIC_irq(ioapic, pin, irq, desc, trigger, polarity); + setup_ioapic_irq(ioapic, pin, irq, cfg, trigger, polarity); return 0; } @@ -4258,13 +4236,12 @@ void __init mp_register_ioapic(int id, u32 address, u32 gsi_base) void __init pre_init_apic_IRQ0(void) { struct irq_cfg *cfg; - struct irq_desc *desc; printk(KERN_INFO "Early APIC setup for system timer0\n"); #ifndef CONFIG_SMP phys_cpu_present_map = physid_mask_of_physid(boot_cpu_physical_apicid); #endif - desc = irq_to_desc_alloc_node(0, 0); + irq_to_desc_alloc_node(0, 0); setup_local_APIC(); @@ -4272,5 +4249,5 @@ void __init pre_init_apic_IRQ0(void) add_pin_to_irq_node(cfg, 0, 0, 0); set_irq_chip_and_handler_name(0, &ioapic_chip, handle_edge_irq, "edge"); - setup_IO_APIC_irq(0, 0, 0, desc, 0, 0); + setup_ioapic_irq(0, 0, 0, cfg, 0, 0); } -- cgit v1.2.3-70-g09d2 From f7e909eae444ff733ecc5628af76d89c363ab480 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 8 Oct 2010 21:40:23 +0200 Subject: x86: Prepare the affinity common functions for taking struct irq_data * While at it rename it to sensible function names and fix the return value from unsigned to int for __ioapic_set_affinity (set_desc_affinity). Returning -1 in a function returning unsigned int is somewhat strange. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/include/asm/hw_irq.h | 6 ++-- arch/x86/kernel/apic/io_apic.c | 77 ++++++++++++++++-------------------------- arch/x86/kernel/uv_irq.c | 2 +- 3 files changed, 33 insertions(+), 52 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/hw_irq.h b/arch/x86/include/asm/hw_irq.h index 46c0fe05f23..76848f27b1a 100644 --- a/arch/x86/include/asm/hw_irq.h +++ b/arch/x86/include/asm/hw_irq.h @@ -95,9 +95,9 @@ extern struct irq_cfg *irq_cfg(unsigned int); extern int assign_irq_vector(int, struct irq_cfg *, const struct cpumask *); extern void send_cleanup_vector(struct irq_cfg *); -struct irq_desc; -extern unsigned int set_desc_affinity(struct irq_desc *, const struct cpumask *, - unsigned int *dest_id); +struct irq_data; +int __ioapic_set_affinity(struct irq_data *, const struct cpumask *, + unsigned int *dest_id); extern int IO_APIC_get_PCI_irq_vector(int bus, int devfn, int pin, struct io_apic_irq_attr *irq_attr); extern void setup_ioapic_dest(void); diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 82c3c66e333..60ca9a47087 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -2279,65 +2279,46 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq } /* - * Either sets desc->affinity to a valid value, and returns + * Either sets data->affinity to a valid value, and returns * ->cpu_mask_to_apicid of that in dest_id, or returns -1 and - * leaves desc->affinity untouched. + * leaves data->affinity untouched. */ -unsigned int -set_desc_affinity(struct irq_desc *desc, const struct cpumask *mask, - unsigned int *dest_id) +int __ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask, + unsigned int *dest_id) { - struct irq_cfg *cfg; - unsigned int irq; + struct irq_cfg *cfg = data->chip_data; if (!cpumask_intersects(mask, cpu_online_mask)) return -1; - irq = desc->irq; - cfg = get_irq_desc_chip_data(desc); - if (assign_irq_vector(irq, cfg, mask)) + if (assign_irq_vector(data->irq, data->chip_data, mask)) return -1; - cpumask_copy(desc->affinity, mask); + cpumask_copy(data->affinity, mask); - *dest_id = apic->cpu_mask_to_apicid_and(desc->affinity, cfg->domain); + *dest_id = apic->cpu_mask_to_apicid_and(mask, cfg->domain); return 0; } static int -set_ioapic_affinity_irq_desc(struct irq_desc *desc, const struct cpumask *mask) +ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask, + bool force) { - struct irq_cfg *cfg; + unsigned int dest, irq = data->irq; unsigned long flags; - unsigned int dest; - unsigned int irq; - int ret = -1; - - irq = desc->irq; - cfg = get_irq_desc_chip_data(desc); + int ret; raw_spin_lock_irqsave(&ioapic_lock, flags); - ret = set_desc_affinity(desc, mask, &dest); + ret = __ioapic_set_affinity(data, mask, &dest); if (!ret) { /* Only the high 8 bits are valid. */ dest = SET_APIC_LOGICAL_ID(dest); - __target_IO_APIC_irq(irq, dest, cfg); + __target_IO_APIC_irq(irq, dest, data->chip_data); } raw_spin_unlock_irqrestore(&ioapic_lock, flags); - return ret; } -static int -set_ioapic_affinity_irq(unsigned int irq, const struct cpumask *mask) -{ - struct irq_desc *desc; - - desc = irq_to_desc(irq); - - return set_ioapic_affinity_irq_desc(desc, mask); -} - #ifdef CONFIG_INTR_REMAP /* @@ -2668,16 +2649,16 @@ static void ir_ack_apic_level(struct irq_data *data) #endif /* CONFIG_INTR_REMAP */ static struct irq_chip ioapic_chip __read_mostly = { - .name = "IO-APIC", - .irq_startup = startup_ioapic_irq, - .irq_mask = mask_ioapic_irq, - .irq_unmask = unmask_ioapic_irq, - .irq_ack = ack_apic_edge, - .irq_eoi = ack_apic_level, + .name = "IO-APIC", + .irq_startup = startup_ioapic_irq, + .irq_mask = mask_ioapic_irq, + .irq_unmask = unmask_ioapic_irq, + .irq_ack = ack_apic_edge, + .irq_eoi = ack_apic_level, #ifdef CONFIG_SMP - .set_affinity = set_ioapic_affinity_irq, + .irq_set_affinity = ioapic_set_affinity, #endif - .irq_retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; static struct irq_chip ir_ioapic_chip __read_mostly = { @@ -3327,7 +3308,7 @@ static int set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask) struct msi_msg msg; unsigned int dest; - if (set_desc_affinity(desc, mask, &dest)) + if (__ioapic_set_affinity(&desc->irq_data, mask, &dest)) return -1; cfg = get_irq_desc_chip_data(desc); @@ -3359,7 +3340,7 @@ ir_set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask) if (get_irte(irq, &irte)) return -1; - if (set_desc_affinity(desc, mask, &dest)) + if (__ioapic_set_affinity(&desc->irq_data, mask, &dest)) return -1; irte.vector = cfg->vector; @@ -3534,7 +3515,7 @@ static int dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask) struct msi_msg msg; unsigned int dest; - if (set_desc_affinity(desc, mask, &dest)) + if (__ioapic_set_affinity(&desc->irq_data, mask, &dest)) return -1; cfg = get_irq_desc_chip_data(desc); @@ -3590,7 +3571,7 @@ static int hpet_msi_set_affinity(struct irq_data *data, struct msi_msg msg; unsigned int dest; - if (set_desc_affinity(desc, mask, &dest)) + if (__ioapic_set_affinity(&desc->irq_data, mask, &dest)) return -1; hpet_msi_read(data->handler_data, &msg); @@ -3693,7 +3674,7 @@ static int set_ht_irq_affinity(unsigned int irq, const struct cpumask *mask) struct irq_cfg *cfg; unsigned int dest; - if (set_desc_affinity(desc, mask, &dest)) + if (__ioapic_set_affinity(&desc->irq_data, mask, &dest)) return -1; cfg = get_irq_desc_chip_data(desc); @@ -4045,14 +4026,14 @@ void __init setup_ioapic_dest(void) */ if (desc->status & (IRQ_NO_BALANCING | IRQ_AFFINITY_SET)) - mask = desc->affinity; + mask = desc->irq_data.affinity; else mask = apic->target_cpus(); if (intr_remapping_enabled) set_ir_ioapic_affinity_irq_desc(desc, mask); else - set_ioapic_affinity_irq_desc(desc, mask); + ioapic_set_affinity(&desc->irq_data, mask, false); } } diff --git a/arch/x86/kernel/uv_irq.c b/arch/x86/kernel/uv_irq.c index 2233a42fb90..56a42e11ae0 100644 --- a/arch/x86/kernel/uv_irq.c +++ b/arch/x86/kernel/uv_irq.c @@ -216,7 +216,7 @@ static int uv_set_irq_affinity(unsigned int irq, const struct cpumask *mask) unsigned long mmr_offset; int mmr_pnode; - if (set_desc_affinity(desc, mask, &dest)) + if (__ioapic_set_affinity(&desc->irq_data, mask, &dest)) return -1; mmr_value = 0; -- cgit v1.2.3-70-g09d2 From 5346b2a78fa3b900da672928978475acdd4632dc Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 8 Oct 2010 21:49:03 +0200 Subject: x86: Convert msi affinity setting to new chip functions Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar Cc: Jesse Barnes --- arch/x86/kernel/apic/io_apic.c | 26 ++++++++++++-------------- 1 file changed, 12 insertions(+), 14 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 60ca9a47087..268c5450392 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -3301,26 +3301,24 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, } #ifdef CONFIG_SMP -static int set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask) +static int +msi_set_affinity(struct irq_data *data, const struct cpumask *mask, bool force) { - struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg; + struct irq_cfg *cfg = data->chip_data; struct msi_msg msg; unsigned int dest; - if (__ioapic_set_affinity(&desc->irq_data, mask, &dest)) + if (__ioapic_set_affinity(data, mask, &dest)) return -1; - cfg = get_irq_desc_chip_data(desc); - - __get_cached_msi_msg(desc->irq_data.msi_desc, &msg); + __get_cached_msi_msg(data->msi_desc, &msg); msg.data &= ~MSI_DATA_VECTOR_MASK; msg.data |= MSI_DATA_VECTOR(cfg->vector); msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK; msg.address_lo |= MSI_ADDR_DEST_ID(dest); - __write_msi_msg(desc->irq_data.msi_desc, &msg); + __write_msi_msg(data->msi_desc, &msg); return 0; } @@ -3370,14 +3368,14 @@ ir_set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask) * which implement the MSI or MSI-X Capability Structure. */ static struct irq_chip msi_chip = { - .name = "PCI-MSI", - .irq_unmask = unmask_msi_irq, - .irq_mask = mask_msi_irq, - .irq_ack = ack_apic_edge, + .name = "PCI-MSI", + .irq_unmask = unmask_msi_irq, + .irq_mask = mask_msi_irq, + .irq_ack = ack_apic_edge, #ifdef CONFIG_SMP - .set_affinity = set_msi_irq_affinity, + .irq_set_affinity = msi_set_affinity, #endif - .irq_retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; static struct irq_chip msi_ir_chip = { -- cgit v1.2.3-70-g09d2 From f19f5ecc920215decfea54f26e3eb14064506675 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 8 Oct 2010 21:50:22 +0200 Subject: x86: Convert remapped ioapic affinity setting to new irq chip function Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar Cc: Suresh Siddha --- arch/x86/kernel/apic/io_apic.c | 58 +++++++++++++++--------------------------- 1 file changed, 20 insertions(+), 38 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 268c5450392..49cc27d5658 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -2333,24 +2333,21 @@ ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask, * the interrupt-remapping table entry. */ static int -migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask) +ir_ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask, + bool force) { - struct irq_cfg *cfg; + struct irq_cfg *cfg = data->chip_data; + unsigned int dest, irq = data->irq; struct irte irte; - unsigned int dest; - unsigned int irq; - int ret = -1; if (!cpumask_intersects(mask, cpu_online_mask)) - return ret; + return -EINVAL; - irq = desc->irq; if (get_irte(irq, &irte)) - return ret; + return -EBUSY; - cfg = get_irq_desc_chip_data(desc); if (assign_irq_vector(irq, cfg, mask)) - return ret; + return -EBUSY; dest = apic->cpu_mask_to_apicid_and(cfg->domain, mask); @@ -2365,29 +2362,14 @@ migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask) if (cfg->move_in_progress) send_cleanup_vector(cfg); - cpumask_copy(desc->affinity, mask); - + cpumask_copy(data->affinity, mask); return 0; } -/* - * Migrates the IRQ destination in the process context. - */ -static int set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc, - const struct cpumask *mask) -{ - return migrate_ioapic_irq_desc(desc, mask); -} -static int set_ir_ioapic_affinity_irq(unsigned int irq, - const struct cpumask *mask) -{ - struct irq_desc *desc = irq_to_desc(irq); - - return set_ir_ioapic_affinity_irq_desc(desc, mask); -} #else -static inline int set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc, - const struct cpumask *mask) +static inline int +ir_ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask, + bool force) { return 0; } @@ -2662,18 +2644,18 @@ static struct irq_chip ioapic_chip __read_mostly = { }; static struct irq_chip ir_ioapic_chip __read_mostly = { - .name = "IR-IO-APIC", - .irq_startup = startup_ioapic_irq, - .irq_mask = mask_ioapic_irq, - .irq_unmask = unmask_ioapic_irq, + .name = "IR-IO-APIC", + .irq_startup = startup_ioapic_irq, + .irq_mask = mask_ioapic_irq, + .irq_unmask = unmask_ioapic_irq, #ifdef CONFIG_INTR_REMAP - .irq_ack = ir_ack_apic_edge, - .irq_eoi = ir_ack_apic_level, + .irq_ack = ir_ack_apic_edge, + .irq_eoi = ir_ack_apic_level, #ifdef CONFIG_SMP - .set_affinity = set_ir_ioapic_affinity_irq, + .irq_set_affinity = ir_ioapic_set_affinity, #endif #endif - .irq_retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; static inline void init_IO_APIC_traps(void) @@ -4029,7 +4011,7 @@ void __init setup_ioapic_dest(void) mask = apic->target_cpus(); if (intr_remapping_enabled) - set_ir_ioapic_affinity_irq_desc(desc, mask); + ir_ioapic_set_affinity(&desc->irq_data, mask, false); else ioapic_set_affinity(&desc->irq_data, mask, false); } -- cgit v1.2.3-70-g09d2 From b5d1c465794f521c352d9c1a33159750c9c3fa84 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 8 Oct 2010 22:15:49 +0200 Subject: x86: Convert remapped msi to new chip.irq_set_affinity function Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar Cc: Suresh Siddha Cc: Jesse Barnes --- arch/x86/kernel/apic/io_apic.c | 34 +++++++++++++++++----------------- 1 file changed, 17 insertions(+), 17 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 49cc27d5658..13f8e28ba4d 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -3310,17 +3310,17 @@ msi_set_affinity(struct irq_data *data, const struct cpumask *mask, bool force) * done in the process context using interrupt-remapping hardware. */ static int -ir_set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask) +ir_msi_set_affinity(struct irq_data *data, const struct cpumask *mask, + bool force) { - struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg = get_irq_desc_chip_data(desc); - unsigned int dest; + struct irq_cfg *cfg = data->chip_data; + unsigned int dest, irq = data->irq; struct irte irte; if (get_irte(irq, &irte)) return -1; - if (__ioapic_set_affinity(&desc->irq_data, mask, &dest)) + if (__ioapic_set_affinity(data, mask, &dest)) return -1; irte.vector = cfg->vector; @@ -3361,16 +3361,16 @@ static struct irq_chip msi_chip = { }; static struct irq_chip msi_ir_chip = { - .name = "IR-PCI-MSI", - .irq_unmask = unmask_msi_irq, - .irq_mask = mask_msi_irq, + .name = "IR-PCI-MSI", + .irq_unmask = unmask_msi_irq, + .irq_mask = mask_msi_irq, #ifdef CONFIG_INTR_REMAP - .irq_ack = ir_ack_apic_edge, + .irq_ack = ir_ack_apic_edge, #ifdef CONFIG_SMP - .set_affinity = ir_set_msi_irq_affinity, + .irq_set_affinity = ir_msi_set_affinity, #endif #endif - .irq_retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; /* @@ -3569,16 +3569,16 @@ static int hpet_msi_set_affinity(struct irq_data *data, #endif /* CONFIG_SMP */ static struct irq_chip ir_hpet_msi_type = { - .name = "IR-HPET_MSI", - .irq_unmask = hpet_msi_unmask, - .irq_mask = hpet_msi_mask, + .name = "IR-HPET_MSI", + .irq_unmask = hpet_msi_unmask, + .irq_mask = hpet_msi_mask, #ifdef CONFIG_INTR_REMAP - .irq_ack = ir_ack_apic_edge, + .irq_ack = ir_ack_apic_edge, #ifdef CONFIG_SMP - .set_affinity = ir_set_msi_irq_affinity, + .irq_set_affinity = ir_msi_set_affinity, #endif #endif - .irq_retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; static struct irq_chip hpet_msi_type = { -- cgit v1.2.3-70-g09d2 From fe52b2d25919eaa01c51651a664f4f2ba6bd2a11 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 8 Oct 2010 22:19:29 +0200 Subject: x86: Convert dmar affinity setting to new chip function Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar Cc: David Woodhouse --- arch/x86/kernel/apic/io_apic.c | 25 ++++++++++++------------- 1 file changed, 12 insertions(+), 13 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 13f8e28ba4d..6f8ac4c542b 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -3488,18 +3488,17 @@ void arch_teardown_msi_irq(unsigned int irq) #if defined (CONFIG_DMAR) || defined (CONFIG_INTR_REMAP) #ifdef CONFIG_SMP -static int dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask) +static int +dmar_msi_set_affinity(struct irq_data *data, const struct cpumask *mask, + bool force) { - struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg; + struct irq_cfg *cfg = data->chip_data; + unsigned int dest, irq = data->irq; struct msi_msg msg; - unsigned int dest; - if (__ioapic_set_affinity(&desc->irq_data, mask, &dest)) + if (__ioapic_set_affinity(data, mask, &dest)) return -1; - cfg = get_irq_desc_chip_data(desc); - dmar_msi_read(irq, &msg); msg.data &= ~MSI_DATA_VECTOR_MASK; @@ -3515,14 +3514,14 @@ static int dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask) #endif /* CONFIG_SMP */ static struct irq_chip dmar_msi_type = { - .name = "DMAR_MSI", - .irq_unmask = dmar_msi_unmask, - .irq_mask = dmar_msi_mask, - .irq_ack = ack_apic_edge, + .name = "DMAR_MSI", + .irq_unmask = dmar_msi_unmask, + .irq_mask = dmar_msi_mask, + .irq_ack = ack_apic_edge, #ifdef CONFIG_SMP - .set_affinity = dmar_msi_set_affinity, + .irq_set_affinity = dmar_msi_set_affinity, #endif - .irq_retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; int arch_setup_dmar_msi(unsigned int irq) -- cgit v1.2.3-70-g09d2 From 0e09ddf2d71aeff92ff8055ac7600b85c255ee85 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 8 Oct 2010 22:21:26 +0200 Subject: x86: Cleanup hpet affinity setting Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/kernel/apic/io_apic.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 6f8ac4c542b..0b9ec3cb311 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -3545,12 +3545,11 @@ int arch_setup_dmar_msi(unsigned int irq) static int hpet_msi_set_affinity(struct irq_data *data, const struct cpumask *mask, bool force) { - struct irq_desc *desc = irq_to_desc(data->irq); struct irq_cfg *cfg = data->chip_data; struct msi_msg msg; unsigned int dest; - if (__ioapic_set_affinity(&desc->irq_data, mask, &dest)) + if (__ioapic_set_affinity(data, mask, &dest)) return -1; hpet_msi_read(data->handler_data, &msg); -- cgit v1.2.3-70-g09d2 From be5b7bf73802a9391158d9fcc0bc6b07670c73a5 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 8 Oct 2010 22:31:46 +0200 Subject: x86: Convert ht set_affinity to new chip function Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar Cc: Jesse Barnes --- arch/x86/kernel/apic/io_apic.c | 25 +++++++++++-------------- 1 file changed, 11 insertions(+), 14 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 0b9ec3cb311..b144f7a9597 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -3646,33 +3646,30 @@ static void target_ht_irq(unsigned int irq, unsigned int dest, u8 vector) write_ht_irq_msg(irq, &msg); } -static int set_ht_irq_affinity(unsigned int irq, const struct cpumask *mask) +static int +ht_set_affinity(struct irq_data *data, const struct cpumask *mask, bool force) { - struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg; + struct irq_cfg *cfg = data->chip_data; unsigned int dest; - if (__ioapic_set_affinity(&desc->irq_data, mask, &dest)) + if (__ioapic_set_affinity(data, mask, &dest)) return -1; - cfg = get_irq_desc_chip_data(desc); - - target_ht_irq(irq, dest, cfg->vector); - + target_ht_irq(data->irq, dest, cfg->vector); return 0; } #endif static struct irq_chip ht_irq_chip = { - .name = "PCI-HT", - .irq_mask = mask_ht_irq, - .irq_unmask = unmask_ht_irq, - .irq_ack = ack_apic_edge, + .name = "PCI-HT", + .irq_mask = mask_ht_irq, + .irq_unmask = unmask_ht_irq, + .irq_ack = ack_apic_edge, #ifdef CONFIG_SMP - .set_affinity = set_ht_irq_affinity, + .irq_set_affinity = ht_set_affinity, #endif - .irq_retrigger = ioapic_retrigger_irq, + .irq_retrigger = ioapic_retrigger_irq, }; int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) -- cgit v1.2.3-70-g09d2 From 7e495529b62cf462eb2d8875fe15ca446b8e1f94 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 28 Sep 2010 23:31:50 +0200 Subject: x86: ioapic: Cleanup some more Cleanup after the irq_chip conversion a bit. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/kernel/apic/io_apic.c | 20 ++++++++------------ 1 file changed, 8 insertions(+), 12 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index b144f7a9597..43030995dcc 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -131,13 +131,9 @@ struct irq_pin_list { struct irq_pin_list *next; }; -static struct irq_pin_list *get_one_free_irq_2_pin(int node) +static struct irq_pin_list *alloc_irq_pin_list(int node) { - struct irq_pin_list *pin; - - pin = kzalloc_node(sizeof(*pin), GFP_ATOMIC, node); - - return pin; + return kzalloc_node(sizeof(struct irq_pin_list), GFP_ATOMIC, node); } /* irq_cfg is indexed by the sum of all RTEs in all I/O APICs. */ @@ -232,7 +228,7 @@ init_copy_irq_2_pin(struct irq_cfg *old_cfg, struct irq_cfg *cfg, int node) if (!old_entry) return; - entry = get_one_free_irq_2_pin(node); + entry = alloc_irq_pin_list(node); if (!entry) return; @@ -242,7 +238,7 @@ init_copy_irq_2_pin(struct irq_cfg *old_cfg, struct irq_cfg *cfg, int node) tail = entry; old_entry = old_entry->next; while (old_entry) { - entry = get_one_free_irq_2_pin(node); + entry = alloc_irq_pin_list(node); if (!entry) { entry = head; while (entry) { @@ -471,7 +467,7 @@ static void ioapic_mask_entry(int apic, int pin) * fast in the common case, and fast for shared ISA-space IRQs. */ static int -add_pin_to_irq_node_nopanic(struct irq_cfg *cfg, int node, int apic, int pin) +__add_pin_to_irq_node(struct irq_cfg *cfg, int node, int apic, int pin) { struct irq_pin_list **last, *entry; @@ -483,7 +479,7 @@ add_pin_to_irq_node_nopanic(struct irq_cfg *cfg, int node, int apic, int pin) last = &entry->next; } - entry = get_one_free_irq_2_pin(node); + entry = alloc_irq_pin_list(node); if (!entry) { printk(KERN_ERR "can not alloc irq_pin_list (%d,%d,%d)\n", node, apic, pin); @@ -498,7 +494,7 @@ add_pin_to_irq_node_nopanic(struct irq_cfg *cfg, int node, int apic, int pin) static void add_pin_to_irq_node(struct irq_cfg *cfg, int node, int apic, int pin) { - if (add_pin_to_irq_node_nopanic(cfg, node, apic, pin)) + if (__add_pin_to_irq_node(cfg, node, apic, pin)) panic("IO-APIC: failed to add irq-pin. Can not proceed\n"); } @@ -3801,7 +3797,7 @@ static int __io_apic_set_pci_routing(struct device *dev, int irq, * IRQs < 16 are already in the irq_2_pin[] map */ if (irq >= legacy_pic->nr_legacy_irqs) { - if (add_pin_to_irq_node_nopanic(cfg, node, ioapic, pin)) { + if (__add_pin_to_irq_node(cfg, node, ioapic, pin)) { printk(KERN_INFO "can not add pin %d for irq %d\n", pin, irq); return 0; -- cgit v1.2.3-70-g09d2 From 6e2fff50a5bd72a3f9e6f3ef6e9137efddb2d580 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 6 Oct 2010 22:07:03 +0200 Subject: x86: ioapic: Cleanup get_one_free_irq_cfg() Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/kernel/apic/io_apic.c | 23 +++++++++++------------ 1 file changed, 11 insertions(+), 12 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 43030995dcc..452f781a042 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -185,19 +185,18 @@ static struct irq_cfg *get_one_free_irq_cfg(int node) struct irq_cfg *cfg; cfg = kzalloc_node(sizeof(*cfg), GFP_ATOMIC, node); - if (cfg) { - if (!zalloc_cpumask_var_node(&cfg->domain, GFP_ATOMIC, node)) { - kfree(cfg); - cfg = NULL; - } else if (!zalloc_cpumask_var_node(&cfg->old_domain, - GFP_ATOMIC, node)) { - free_cpumask_var(cfg->domain); - kfree(cfg); - cfg = NULL; - } - } - + if (!cfg) + return NULL; + if (!zalloc_cpumask_var_node(&cfg->domain, GFP_ATOMIC, node)) + goto out_cfg; + if (!zalloc_cpumask_var_node(&cfg->old_domain, GFP_ATOMIC, node)) + goto out_domain; return cfg; +out_domain: + free_cpumask_var(cfg->domain); +out_cfg: + kfree(cfg); + return NULL; } int arch_init_chip_data(struct irq_desc *desc, int node) -- cgit v1.2.3-70-g09d2 From 08c33db6d044d9dc74ddf8d9ee3cb1fa3eca262b Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 6 Oct 2010 22:14:21 +0200 Subject: x86: Implement new allocator functions Implement new allocator functions which make use of the core changes. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/kernel/apic/io_apic.c | 54 ++++++++++++++++++++++++++++++++++++------ 1 file changed, 47 insertions(+), 7 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 452f781a042..065c5dc88b8 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -199,6 +199,13 @@ out_cfg: return NULL; } +static void free_irq_cfg(struct irq_cfg *cfg) +{ + free_cpumask_var(cfg->domain); + free_cpumask_var(cfg->old_domain); + kfree(cfg); +} + int arch_init_chip_data(struct irq_desc *desc, int node) { struct irq_cfg *cfg; @@ -299,13 +306,6 @@ void arch_init_copy_chip_data(struct irq_desc *old_desc, init_copy_irq_2_pin(old_cfg, cfg, node); } -static void free_irq_cfg(struct irq_cfg *cfg) -{ - free_cpumask_var(cfg->domain); - free_cpumask_var(cfg->old_domain); - kfree(cfg); -} - void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc) { struct irq_cfg *old_cfg, *cfg; @@ -325,13 +325,53 @@ void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc) /* end for move_irq_desc */ #else + struct irq_cfg *irq_cfg(unsigned int irq) { return irq < nr_irqs ? irq_cfgx + irq : NULL; } +static struct irq_cfg *get_one_free_irq_cfg(unsigned int irq, int node) +{ + return irq_cfgx + irq; +} + +static inline void free_irq_cfg(struct irq_cfg *cfg) { } + #endif +static struct irq_cfg *alloc_irq_and_cfg_at(unsigned int at, int node) +{ + int res = irq_alloc_desc_at(at, node); + struct irq_cfg *cfg; + + if (res < 0) { + if (res != -EEXIST) + return NULL; + cfg = get_irq_chip_data(at); + if (cfg) + return cfg; + } + + cfg = get_one_free_irq_cfg(node); + if (cfg) + set_irq_chip_data(at, cfg); + else + irq_free_desc(at); + return cfg; +} + +static int alloc_irq_from(unsigned int from, int node) +{ + return irq_alloc_desc_from(from, node); +} + +static void free_irq_at(unsigned int at, struct irq_cfg *cfg) +{ + free_irq_cfg(cfg); + irq_free_desc(at); +} + struct io_apic { unsigned int index; unsigned int unused[3]; -- cgit v1.2.3-70-g09d2 From f981a3dc1941035a108da1276c448de6b10ddac9 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 8 Oct 2010 10:44:21 +0200 Subject: x86: io_apic: Prepare alloc/free_irq_cfg() Rename the grossly misnamed get_one_free_irq_cfg() to alloc_irq_cfg(). Add a (not yet used) irq number argument to free_irq_cfg() Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/kernel/apic/io_apic.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 065c5dc88b8..06da8fe2647 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -180,7 +180,7 @@ struct irq_cfg *irq_cfg(unsigned int irq) return get_irq_chip_data(irq); } -static struct irq_cfg *get_one_free_irq_cfg(int node) +static struct irq_cfg *alloc_irq_cfg(unsigned int irq, int node) { struct irq_cfg *cfg; @@ -199,7 +199,7 @@ out_cfg: return NULL; } -static void free_irq_cfg(struct irq_cfg *cfg) +static void free_irq_cfg(unsigned int at, struct irq_cfg *cfg) { free_cpumask_var(cfg->domain); free_cpumask_var(cfg->old_domain); @@ -212,7 +212,7 @@ int arch_init_chip_data(struct irq_desc *desc, int node) cfg = get_irq_desc_chip_data(desc); if (!cfg) { - cfg = get_one_free_irq_cfg(node); + cfg = alloc_irq_cfg(desc->irq, node); desc->chip_data = cfg; if (!cfg) { printk(KERN_ERR "can not alloc irq_cfg\n"); @@ -289,7 +289,7 @@ void arch_init_copy_chip_data(struct irq_desc *old_desc, struct irq_cfg *cfg; struct irq_cfg *old_cfg; - cfg = get_one_free_irq_cfg(node); + cfg = alloc_irq_cfg(desc->irq, node); if (!cfg) return; @@ -318,7 +318,7 @@ void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc) if (old_cfg) { free_irq_2_pin(old_cfg, cfg); - free_irq_cfg(old_cfg); + free_irq_cfg(old_desc->irq, old_cfg); old_desc->chip_data = NULL; } } @@ -331,12 +331,12 @@ struct irq_cfg *irq_cfg(unsigned int irq) return irq < nr_irqs ? irq_cfgx + irq : NULL; } -static struct irq_cfg *get_one_free_irq_cfg(unsigned int irq, int node) +static struct irq_cfg *alloc_irq_cfg(unsigned int irq, int node) { return irq_cfgx + irq; } -static inline void free_irq_cfg(struct irq_cfg *cfg) { } +static inline void free_irq_cfg(unsigned int at, struct irq_cfg *cfg) { } #endif @@ -353,7 +353,7 @@ static struct irq_cfg *alloc_irq_and_cfg_at(unsigned int at, int node) return cfg; } - cfg = get_one_free_irq_cfg(node); + cfg = alloc_irq_cfg(at, node); if (cfg) set_irq_chip_data(at, cfg); else @@ -368,7 +368,7 @@ static int alloc_irq_from(unsigned int from, int node) static void free_irq_at(unsigned int at, struct irq_cfg *cfg) { - free_irq_cfg(cfg); + free_irq_cfg(at, cfg); irq_free_desc(at); } -- cgit v1.2.3-70-g09d2 From fe6dab4e79e82ec35879bfe1a8968b7d15ac0d91 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Fri, 8 Oct 2010 22:44:02 -0700 Subject: x86: Don't setup ioapic irq for sci twice The sparseirq rework triggered a warning in the iommu code, which was caused by setting up ioapic for ACPI irq 9 twice. This function is solely to handle interrupts which are on a secondary ioapic and outside the legacy irq range. Replace the sparse irq_to_desc check with a non ifdeffed version. [ tglx: Moved it before the ioapic sparse conversion and simplified the inverse logic ] Signed-off-by: Yinghai Lu LKML-Reference: <4CB00122.3030301@kernel.org> Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/kernel/apic/io_apic.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 06da8fe2647..5aae718a713 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -1565,11 +1565,11 @@ void setup_IO_APIC_irq_extra(u32 gsi) return; irq = pin_2_irq(idx, apic_id, pin); -#ifdef CONFIG_SPARSE_IRQ - desc = irq_to_desc(irq); - if (desc) + + /* Only handle the non legacy irqs on secondary ioapics */ + if (apic_id == 0 || irq < NR_IRQS_LEGACY) return; -#endif + desc = irq_to_desc_alloc_node(irq, node); if (!desc) { printk(KERN_INFO "can not get irq_desc for %d\n", irq); -- cgit v1.2.3-70-g09d2 From fbc6bff04a095e049be290ff6f6ac68839166bd6 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 28 Sep 2010 20:34:53 +0200 Subject: x86: ioapic: Cleanup sparse irq code Switch over to the new allocator and remove all the magic which was caused by the unability to destroy irq descriptors. Get rid of the create_irq_nr() loop for sparse and non sparse irq. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/Kconfig | 1 - arch/x86/kernel/apic/io_apic.c | 107 ++++++++++++++++++----------------------- 2 files changed, 48 insertions(+), 60 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 3ec657f7ee7..8cc510874e1 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -61,7 +61,6 @@ config X86 select HAVE_USER_RETURN_NOTIFIER select HAVE_GENERIC_HARDIRQS select HAVE_SPARSE_IRQ - select NUMA_IRQ_DESC if (SPARSE_IRQ && NUMA) select GENERIC_IRQ_PROBE select GENERIC_PENDING_IRQ if SMP diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 5aae718a713..ed340297571 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -157,6 +157,9 @@ int __init arch_early_irq_init(void) count = ARRAY_SIZE(irq_cfgx); node = cpu_to_node(0); + /* Make sure the legacy interrupts are marked in the bitmap */ + irq_reserve_irqs(0, legacy_pic->nr_legacy_irqs); + for (i = 0; i < count; i++) { set_irq_chip_data(i, &cfg[i]); zalloc_cpumask_var_node(&cfg[i].domain, GFP_NOWAIT, node); @@ -201,11 +204,15 @@ out_cfg: static void free_irq_cfg(unsigned int at, struct irq_cfg *cfg) { + if (!cfg) + return; + set_irq_chip_data(at, NULL); free_cpumask_var(cfg->domain); free_cpumask_var(cfg->old_domain); kfree(cfg); } +#if 0 int arch_init_chip_data(struct irq_desc *desc, int node) { struct irq_cfg *cfg; @@ -323,6 +330,7 @@ void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc) } } /* end for move_irq_desc */ +#endif #else @@ -1479,11 +1487,9 @@ static struct { static void __init setup_IO_APIC_irqs(void) { - int apic_id, pin, idx, irq; - int notcon = 0; - struct irq_desc *desc; - struct irq_cfg *cfg; + int apic_id, pin, idx, irq, notcon = 0; int node = cpu_to_node(0); + struct irq_cfg *cfg; apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n"); @@ -1520,12 +1526,10 @@ static void __init setup_IO_APIC_irqs(void) apic->multi_timer_check(apic_id, irq)) continue; - desc = irq_to_desc_alloc_node(irq, node); - if (!desc) { - printk(KERN_INFO "can not get irq_desc for %d\n", irq); + cfg = alloc_irq_and_cfg_at(irq, node); + if (!cfg) continue; - } - cfg = get_irq_desc_chip_data(desc); + add_pin_to_irq_node(cfg, node, apic_id, pin); /* * don't mark it in pin_programmed, so later acpi could @@ -1547,9 +1551,7 @@ static void __init setup_IO_APIC_irqs(void) */ void setup_IO_APIC_irq_extra(u32 gsi) { - int apic_id = 0, pin, idx, irq; - int node = cpu_to_node(0); - struct irq_desc *desc; + int apic_id = 0, pin, idx, irq, node = cpu_to_node(0); struct irq_cfg *cfg; /* @@ -1570,13 +1572,10 @@ void setup_IO_APIC_irq_extra(u32 gsi) if (apic_id == 0 || irq < NR_IRQS_LEGACY) return; - desc = irq_to_desc_alloc_node(irq, node); - if (!desc) { - printk(KERN_INFO "can not get irq_desc for %d\n", irq); + cfg = alloc_irq_and_cfg_at(irq, node); + if (!cfg) return; - } - cfg = get_irq_desc_chip_data(desc); add_pin_to_irq_node(cfg, node, apic_id, pin); if (test_bit(pin, mp_ioapic_routing[apic_id].pin_programmed)) { @@ -3177,44 +3176,37 @@ device_initcall(ioapic_init_sysfs); /* * Dynamic irq allocate and deallocation */ -unsigned int create_irq_nr(unsigned int irq_want, int node) +unsigned int create_irq_nr(unsigned int from, int node) { - /* Allocate an unused irq */ - unsigned int irq; - unsigned int new; + struct irq_cfg *cfg; unsigned long flags; - struct irq_cfg *cfg_new = NULL; - struct irq_desc *desc_new = NULL; - - irq = 0; - if (irq_want < nr_irqs_gsi) - irq_want = nr_irqs_gsi; - - raw_spin_lock_irqsave(&vector_lock, flags); - for (new = irq_want; new < nr_irqs; new++) { - desc_new = irq_to_desc_alloc_node(new, node); - if (!desc_new) { - printk(KERN_INFO "can not get irq_desc for %d\n", new); - continue; - } - cfg_new = get_irq_desc_chip_data(desc_new); - - if (cfg_new->vector != 0) - continue; + unsigned int ret = 0; + int irq; - desc_new = move_irq_desc(desc_new, node); - cfg_new = get_irq_desc_chip_data(desc_new); + if (from < nr_irqs_gsi) + from = nr_irqs_gsi; - if (__assign_irq_vector(new, cfg_new, apic->target_cpus()) == 0) - irq = new; - break; + irq = alloc_irq_from(from, node); + if (irq < 0) + return 0; + cfg = alloc_irq_cfg(irq, node); + if (!cfg) { + free_irq_at(irq, NULL); + return 0; } - raw_spin_unlock_irqrestore(&vector_lock, flags); - if (irq > 0) - dynamic_irq_init_keep_chip_data(irq); + raw_spin_lock_irqsave(&vector_lock, flags); + if (!__assign_irq_vector(irq, cfg, apic->target_cpus())) + ret = irq; + raw_spin_unlock_irqrestore(&vector_lock, flags); - return irq; + if (ret) { + set_irq_chip_data(irq, cfg); + irq_clear_status_flags(irq, IRQ_NOREQUEST); + } else { + free_irq_at(irq, cfg); + } + return ret; } int create_irq(void) @@ -3234,14 +3226,16 @@ int create_irq(void) void destroy_irq(unsigned int irq) { + struct irq_cfg *cfg = get_irq_chip_data(irq); unsigned long flags; - dynamic_irq_cleanup_keep_chip_data(irq); + irq_set_status_flags(irq, IRQ_NOREQUEST|IRQ_NOPROBE); free_irte(irq); raw_spin_lock_irqsave(&vector_lock, flags); - __clear_irq_vector(irq, get_irq_chip_data(irq)); + __clear_irq_vector(irq, cfg); raw_spin_unlock_irqrestore(&vector_lock, flags); + free_irq_at(irq, cfg); } /* @@ -3802,7 +3796,6 @@ int __init arch_probe_nr_irqs(void) static int __io_apic_set_pci_routing(struct device *dev, int irq, struct io_apic_irq_attr *irq_attr) { - struct irq_desc *desc; struct irq_cfg *cfg; int node; int ioapic, pin; @@ -3820,18 +3813,14 @@ static int __io_apic_set_pci_routing(struct device *dev, int irq, else node = cpu_to_node(0); - desc = irq_to_desc_alloc_node(irq, node); - if (!desc) { - printk(KERN_INFO "can not get irq_desc %d\n", irq); + cfg = alloc_irq_and_cfg_at(irq, node); + if (!cfg) return 0; - } pin = irq_attr->ioapic_pin; trigger = irq_attr->trigger; polarity = irq_attr->polarity; - cfg = get_irq_desc_chip_data(desc); - /* * IRQs < 16 are already in the irq_2_pin[] map */ @@ -4232,11 +4221,11 @@ void __init pre_init_apic_IRQ0(void) #ifndef CONFIG_SMP phys_cpu_present_map = physid_mask_of_physid(boot_cpu_physical_apicid); #endif - irq_to_desc_alloc_node(0, 0); + /* Make sure the irq descriptor is set up */ + cfg = alloc_irq_and_cfg_at(0, 0); setup_local_APIC(); - cfg = irq_cfg(0); add_pin_to_irq_node(cfg, 0, 0, 0); set_irq_chip_and_handler_name(0, &ioapic_chip, handle_edge_irq, "edge"); -- cgit v1.2.3-70-g09d2 From bc5fdf9f3aad37406b3c8d635a7940cd65de0c12 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 8 Oct 2010 10:40:53 +0200 Subject: x86: io_apic: Remove the now unused sparse_irq arch_* functions Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/kernel/apic/io_apic.c | 120 ----------------------------------------- 1 file changed, 120 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index ed340297571..6ff6bb883c5 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -212,126 +212,6 @@ static void free_irq_cfg(unsigned int at, struct irq_cfg *cfg) kfree(cfg); } -#if 0 -int arch_init_chip_data(struct irq_desc *desc, int node) -{ - struct irq_cfg *cfg; - - cfg = get_irq_desc_chip_data(desc); - if (!cfg) { - cfg = alloc_irq_cfg(desc->irq, node); - desc->chip_data = cfg; - if (!cfg) { - printk(KERN_ERR "can not alloc irq_cfg\n"); - BUG_ON(1); - } - } - - return 0; -} - -/* for move_irq_desc */ -static void -init_copy_irq_2_pin(struct irq_cfg *old_cfg, struct irq_cfg *cfg, int node) -{ - struct irq_pin_list *old_entry, *head, *tail, *entry; - - cfg->irq_2_pin = NULL; - old_entry = old_cfg->irq_2_pin; - if (!old_entry) - return; - - entry = alloc_irq_pin_list(node); - if (!entry) - return; - - entry->apic = old_entry->apic; - entry->pin = old_entry->pin; - head = entry; - tail = entry; - old_entry = old_entry->next; - while (old_entry) { - entry = alloc_irq_pin_list(node); - if (!entry) { - entry = head; - while (entry) { - head = entry->next; - kfree(entry); - entry = head; - } - /* still use the old one */ - return; - } - entry->apic = old_entry->apic; - entry->pin = old_entry->pin; - tail->next = entry; - tail = entry; - old_entry = old_entry->next; - } - - tail->next = NULL; - cfg->irq_2_pin = head; -} - -static void free_irq_2_pin(struct irq_cfg *old_cfg, struct irq_cfg *cfg) -{ - struct irq_pin_list *entry, *next; - - if (old_cfg->irq_2_pin == cfg->irq_2_pin) - return; - - entry = old_cfg->irq_2_pin; - - while (entry) { - next = entry->next; - kfree(entry); - entry = next; - } - old_cfg->irq_2_pin = NULL; -} - -void arch_init_copy_chip_data(struct irq_desc *old_desc, - struct irq_desc *desc, int node) -{ - struct irq_cfg *cfg; - struct irq_cfg *old_cfg; - - cfg = alloc_irq_cfg(desc->irq, node); - - if (!cfg) - return; - - desc->chip_data = cfg; - - old_cfg = old_desc->chip_data; - - cfg->vector = old_cfg->vector; - cfg->move_in_progress = old_cfg->move_in_progress; - cpumask_copy(cfg->domain, old_cfg->domain); - cpumask_copy(cfg->old_domain, old_cfg->old_domain); - - init_copy_irq_2_pin(old_cfg, cfg, node); -} - -void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc) -{ - struct irq_cfg *old_cfg, *cfg; - - old_cfg = get_irq_desc_chip_data(old_desc); - cfg = get_irq_desc_chip_data(desc); - - if (old_cfg == cfg) - return; - - if (old_cfg) { - free_irq_2_pin(old_cfg, cfg); - free_irq_cfg(old_desc->irq, old_cfg); - old_desc->chip_data = NULL; - } -} -/* end for move_irq_desc */ -#endif - #else struct irq_cfg *irq_cfg(unsigned int irq) -- cgit v1.2.3-70-g09d2 From 423f085952fd7253407cb92984cc2d495a564481 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sun, 10 Oct 2010 11:39:09 +0200 Subject: x86: Embedd irq_2_iommu into irq_cfg That interrupt remapping code is x86 specific and tied to the io_apic code. No need for separate allocator functions in the interrupt remapping code. This allows to simplify the code and irq_2_iommu is small (13 bytes on 64bit) so it's not a real problem even if interrupt remapping is runtime disabled. If it's compile time disabled the impact is zero. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar Acked-by: Suresh Siddha Cc: David Woodhouse Cc: Jesse Barnes --- arch/x86/include/asm/hw_irq.h | 10 ++++++++++ drivers/pci/intr_remapping.c | 7 ------- include/linux/dmar.h | 1 + 3 files changed, 11 insertions(+), 7 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/hw_irq.h b/arch/x86/include/asm/hw_irq.h index 76848f27b1a..e756c4bfed9 100644 --- a/arch/x86/include/asm/hw_irq.h +++ b/arch/x86/include/asm/hw_irq.h @@ -78,6 +78,13 @@ static inline void set_io_apic_irq_attr(struct io_apic_irq_attr *irq_attr, irq_attr->polarity = polarity; } +struct irq_2_iommu { + struct intel_iommu *iommu; + u16 irte_index; + u16 sub_handle; + u8 irte_mask; +}; + /* * This is performance-critical, we want to do it O(1) * @@ -89,6 +96,9 @@ struct irq_cfg { cpumask_var_t old_domain; u8 vector; u8 move_in_progress : 1; +#ifdef CONFIG_INTR_REMAP + struct irq_2_iommu irq_2_iommu; +#endif }; extern struct irq_cfg *irq_cfg(unsigned int); diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c index 343f7299c78..0f4691c5fab 100644 --- a/drivers/pci/intr_remapping.c +++ b/drivers/pci/intr_remapping.c @@ -46,13 +46,6 @@ static __init int setup_intremap(char *str) } early_param("intremap", setup_intremap); -struct irq_2_iommu { - struct intel_iommu *iommu; - u16 irte_index; - u16 sub_handle; - u8 irte_mask; -}; - #ifdef CONFIG_GENERIC_HARDIRQS static struct irq_2_iommu *irq_2_iommu(unsigned int irq) { diff --git a/include/linux/dmar.h b/include/linux/dmar.h index 200439ec7c4..4475f8cf7a6 100644 --- a/include/linux/dmar.h +++ b/include/linux/dmar.h @@ -106,6 +106,7 @@ struct irte { __u64 high; }; }; + #ifdef CONFIG_INTR_REMAP extern int intr_remapping_enabled; extern int intr_remapping_supported(void); -- cgit v1.2.3-70-g09d2 From 1a0730d6649113c820217387a011a17dd4aff3ad Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 11 Oct 2010 11:55:37 +0200 Subject: x86: Speed up the irq_remapped check in hot pathes irq_2_iommu is in struct irq_cfg, so we can do the irq_remapped check based on irq_cfg instead of going through a lookup function. That's especially interesting in the eoi_ioapic_irq() hotpath. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar Acked-by: Suresh Siddha Cc: David Woodhouse Cc: Jesse Barnes --- arch/x86/include/asm/irq_remapping.h | 8 ++++++++ arch/x86/kernel/apic/io_apic.c | 12 ++++++------ drivers/pci/intr_remapping.c | 7 ------- include/linux/dmar.h | 2 -- 4 files changed, 14 insertions(+), 15 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/irq_remapping.h b/arch/x86/include/asm/irq_remapping.h index 8d841505344..1c23360fb2d 100644 --- a/arch/x86/include/asm/irq_remapping.h +++ b/arch/x86/include/asm/irq_remapping.h @@ -24,10 +24,18 @@ static inline void prepare_irte(struct irte *irte, int vector, irte->dest_id = IRTE_DEST(dest); irte->redir_hint = 1; } +static inline bool irq_remapped(struct irq_cfg *cfg) +{ + return cfg->irq_2_iommu.iommu != NULL; +} #else static void prepare_irte(struct irte *irte, int vector, unsigned int dest) { } +static inline bool irq_remapped(struct irq_cfg *cfg) +{ + return false; +} #endif #endif /* _ASM_X86_IRQ_REMAPPING_H */ diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 6ff6bb883c5..1b8e8a10612 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -1237,7 +1237,7 @@ static void ioapic_register_intr(unsigned int irq, unsigned long trigger) else irq_clear_status_flags(irq, IRQ_LEVEL); - if (irq_remapped(irq)) { + if (irq_remapped(get_irq_chip_data(irq))) { irq_set_status_flags(irq, IRQ_MOVE_PCNTXT); if (trigger) set_irq_chip_and_handler_name(irq, &ir_ioapic_chip, @@ -2183,7 +2183,7 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq * With interrupt-remapping, destination information comes * from interrupt-remapping table entry. */ - if (!irq_remapped(irq)) + if (!irq_remapped(cfg)) io_apic_write(apic, 0x11 + pin*2, dest); reg = io_apic_read(apic, 0x10 + pin*2); reg &= ~IO_APIC_REDIR_VECTOR_MASK; @@ -2415,7 +2415,7 @@ static void eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg) * intr-remapping table entry. Hence for the io-apic * EOI we use the pin number. */ - if (irq_remapped(irq)) + if (irq_remapped(cfg)) io_apic_eoi(entry->apic, entry->pin); else io_apic_eoi(entry->apic, cfg->vector); @@ -3139,7 +3139,7 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, dest = apic->cpu_mask_to_apicid_and(cfg->domain, apic->target_cpus()); - if (irq_remapped(irq)) { + if (irq_remapped(get_irq_chip_data(irq))) { struct irte irte; int ir_index; u16 sub_handle; @@ -3321,7 +3321,7 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq) set_irq_msi(irq, msidesc); write_msi_msg(irq, &msg); - if (irq_remapped(irq)) { + if (irq_remapped(get_irq_chip_data(irq))) { irq_set_status_flags(irq, IRQ_MOVE_PCNTXT); set_irq_chip_and_handler_name(irq, &msi_ir_chip, handle_edge_irq, "edge"); } else @@ -3522,7 +3522,7 @@ int arch_setup_hpet_msi(unsigned int irq, unsigned int id) hpet_msi_write(get_irq_data(irq), &msg); irq_set_status_flags(irq, IRQ_MOVE_PCNTXT); - if (irq_remapped(irq)) + if (irq_remapped(get_irq_chip_data(irq))) set_irq_chip_and_handler_name(irq, &ir_hpet_msi_type, handle_edge_irq, "edge"); else diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c index a620b8bd8f4..ec87cd66f3e 100644 --- a/drivers/pci/intr_remapping.c +++ b/drivers/pci/intr_remapping.c @@ -54,13 +54,6 @@ static struct irq_2_iommu *irq_2_iommu(unsigned int irq) return cfg ? &cfg->irq_2_iommu : NULL; } -int irq_remapped(int irq) -{ - struct irq_2_iommu *irq_iommu = irq_2_iommu(irq); - - return irq_iommu ? irq_iommu->iommu != NULL : 0; -} - int get_irte(int irq, struct irte *entry) { struct irq_2_iommu *irq_iommu = irq_2_iommu(irq); diff --git a/include/linux/dmar.h b/include/linux/dmar.h index 4475f8cf7a6..51651b76d40 100644 --- a/include/linux/dmar.h +++ b/include/linux/dmar.h @@ -122,7 +122,6 @@ extern int set_irte_irq(int irq, struct intel_iommu *iommu, u16 index, extern int map_irq_to_irte_handle(int irq, u16 *sub_handle); extern int free_irte(int irq); -extern int irq_remapped(int irq); extern struct intel_iommu *map_dev_to_ir(struct pci_dev *dev); extern struct intel_iommu *map_ioapic_to_ir(int apic); extern struct intel_iommu *map_hpet_to_ir(u8 id); @@ -176,7 +175,6 @@ static inline int set_msi_sid(struct irte *irte, struct pci_dev *dev) return 0; } -#define irq_remapped(irq) (0) #define enable_intr_remapping(mode) (-1) #define disable_intr_remapping() (0) #define reenable_intr_remapping(mode) (0) -- cgit v1.2.3-70-g09d2 From 1a8ce7ff68d777195da2d340561bda610e533a64 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 4 Oct 2010 21:08:56 +0200 Subject: x86: Make io_apic.c local functions static No users outside of io_apic.c Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/include/asm/io_apic.h | 6 ------ arch/x86/kernel/apic/io_apic.c | 10 +++++----- 2 files changed, 5 insertions(+), 11 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/io_apic.h b/arch/x86/include/asm/io_apic.h index 9cb2edb87c2..c8be4566c3d 100644 --- a/arch/x86/include/asm/io_apic.h +++ b/arch/x86/include/asm/io_apic.h @@ -170,12 +170,6 @@ extern int restore_IO_APIC_setup(struct IO_APIC_route_entry **ioapic_entries); extern void probe_nr_irqs_gsi(void); -extern int setup_ioapic_entry(int apic, int irq, - struct IO_APIC_route_entry *entry, - unsigned int destination, int trigger, - int polarity, int vector, int pin); -extern void ioapic_write_entry(int apic, int pin, - struct IO_APIC_route_entry e); extern void setup_ioapic_ids_from_mpc(void); struct mp_ioapic_gsi{ diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 1b8e8a10612..0102543b564 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -364,7 +364,7 @@ __ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) io_apic_write(apic, 0x10 + 2*pin, eu.w1); } -void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) +static void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) { unsigned long flags; raw_spin_lock_irqsave(&ioapic_lock, flags); @@ -1259,10 +1259,10 @@ static void ioapic_register_intr(unsigned int irq, unsigned long trigger) handle_edge_irq, "edge"); } -int setup_ioapic_entry(int apic_id, int irq, - struct IO_APIC_route_entry *entry, - unsigned int destination, int trigger, - int polarity, int vector, int pin) +static int setup_ioapic_entry(int apic_id, int irq, + struct IO_APIC_route_entry *entry, + unsigned int destination, int trigger, + int polarity, int vector, int pin) { /* * add it to the IO-APIC irq-routing table: -- cgit v1.2.3-70-g09d2 From 48b2650196364e4ef124efb841b63c2326e4ccb2 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 30 Sep 2010 11:43:08 +0200 Subject: x86: uv: Clean up the direct access to irq_desc Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/include/asm/hw_irq.h | 1 - arch/x86/kernel/apic/io_apic.c | 2 +- arch/x86/kernel/uv_irq.c | 55 +++++++++++++++--------------------------- 3 files changed, 20 insertions(+), 38 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/hw_irq.h b/arch/x86/include/asm/hw_irq.h index e756c4bfed9..d5905fd8ba4 100644 --- a/arch/x86/include/asm/hw_irq.h +++ b/arch/x86/include/asm/hw_irq.h @@ -101,7 +101,6 @@ struct irq_cfg { #endif }; -extern struct irq_cfg *irq_cfg(unsigned int); extern int assign_irq_vector(int, struct irq_cfg *, const struct cpumask *); extern void send_cleanup_vector(struct irq_cfg *); diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 0102543b564..5193f201b91 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -178,7 +178,7 @@ int __init arch_early_irq_init(void) } #ifdef CONFIG_SPARSE_IRQ -struct irq_cfg *irq_cfg(unsigned int irq) +static struct irq_cfg *irq_cfg(unsigned int irq) { return get_irq_chip_data(irq); } diff --git a/arch/x86/kernel/uv_irq.c b/arch/x86/kernel/uv_irq.c index 56a42e11ae0..7b24460917d 100644 --- a/arch/x86/kernel/uv_irq.c +++ b/arch/x86/kernel/uv_irq.c @@ -28,34 +28,21 @@ struct uv_irq_2_mmr_pnode{ static spinlock_t uv_irq_lock; static struct rb_root uv_irq_root; -static int uv_set_irq_affinity(unsigned int, const struct cpumask *); +static int uv_set_irq_affinity(struct irq_data *, const struct cpumask *, bool); -static void uv_noop(unsigned int irq) -{ -} - -static unsigned int uv_noop_ret(unsigned int irq) -{ - return 0; -} +static void uv_noop(struct irq_data *data) { } -static void uv_ack_apic(unsigned int irq) +static void uv_ack_apic(struct irq_data *data) { ack_APIC_irq(); } static struct irq_chip uv_irq_chip = { - .name = "UV-CORE", - .startup = uv_noop_ret, - .shutdown = uv_noop, - .enable = uv_noop, - .disable = uv_noop, - .ack = uv_noop, - .mask = uv_noop, - .unmask = uv_noop, - .eoi = uv_ack_apic, - .end = uv_noop, - .set_affinity = uv_set_irq_affinity, + .name = "UV-CORE", + .irq_mask = uv_noop, + .irq_unmask = uv_noop, + .irq_eoi = uv_ack_apic, + .irq_set_affinity = uv_set_irq_affinity, }; /* @@ -144,26 +131,22 @@ arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade, unsigned long mmr_offset, int limit) { const struct cpumask *eligible_cpu = cpumask_of(cpu); - struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg; - int mmr_pnode; + struct irq_cfg *cfg = get_irq_chip_data(irq); unsigned long mmr_value; struct uv_IO_APIC_route_entry *entry; - int err; + int mmr_pnode, err; BUILD_BUG_ON(sizeof(struct uv_IO_APIC_route_entry) != sizeof(unsigned long)); - cfg = irq_cfg(irq); - err = assign_irq_vector(irq, cfg, eligible_cpu); if (err != 0) return err; if (limit == UV_AFFINITY_CPU) - desc->status |= IRQ_NO_BALANCING; + irq_set_status_flags(irq, IRQ_NO_BALANCING); else - desc->status |= IRQ_MOVE_PCNTXT; + irq_set_status_flags(irq, IRQ_MOVE_PCNTXT); set_irq_chip_and_handler_name(irq, &uv_irq_chip, handle_percpu_irq, irq_name); @@ -206,17 +189,17 @@ static void arch_disable_uv_irq(int mmr_pnode, unsigned long mmr_offset) uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value); } -static int uv_set_irq_affinity(unsigned int irq, const struct cpumask *mask) +static int +uv_set_irq_affinity(struct irq_data *data, const struct cpumask *mask, + bool force) { - struct irq_desc *desc = irq_to_desc(irq); - struct irq_cfg *cfg = get_irq_desc_chip_data(desc); + struct irq_cfg *cfg = data->chip_data; unsigned int dest; - unsigned long mmr_value; + unsigned long mmr_value, mmr_offset; struct uv_IO_APIC_route_entry *entry; - unsigned long mmr_offset; int mmr_pnode; - if (__ioapic_set_affinity(&desc->irq_data, mask, &dest)) + if (__ioapic_set_affinity(data, mask, &dest)) return -1; mmr_value = 0; @@ -231,7 +214,7 @@ static int uv_set_irq_affinity(unsigned int irq, const struct cpumask *mask) entry->dest = dest; /* Get previously stored MMR and pnode of hub sourcing interrupts */ - if (uv_irq_2_mmr_info(irq, &mmr_offset, &mmr_pnode)) + if (uv_irq_2_mmr_info(data->irq, &mmr_offset, &mmr_pnode)) return -1; uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value); -- cgit v1.2.3-70-g09d2 From ad9f43340f48c5f7a0a5ef7656986e23d06bf996 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 30 Sep 2010 11:26:43 +0200 Subject: x86: Use sane enumeration Instead of looping through all interrupts, use the bitmap lookup to find the next. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/kernel/apic/io_apic.c | 22 ++++++++++------------ 1 file changed, 10 insertions(+), 12 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 5193f201b91..057b0e13d1c 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -1160,7 +1160,6 @@ void __setup_vector_irq(int cpu) /* Initialize vector_irq on a new cpu */ int irq, vector; struct irq_cfg *cfg; - struct irq_desc *desc; /* * vector_lock will make sure that we don't run into irq vector @@ -1169,9 +1168,10 @@ void __setup_vector_irq(int cpu) */ raw_spin_lock(&vector_lock); /* Mark the inuse vectors */ - for_each_irq_desc(irq, desc) { - cfg = get_irq_desc_chip_data(desc); - + for_each_active_irq(irq) { + cfg = get_irq_chip_data(irq); + if (!cfg) + continue; /* * If it is a legacy IRQ handled by the legacy PIC, this cpu * will be part of the irq_cfg's domain. @@ -1516,7 +1516,6 @@ __apicdebuginit(void) print_IO_APIC(void) union IO_APIC_reg_03 reg_03; unsigned long flags; struct irq_cfg *cfg; - struct irq_desc *desc; unsigned int irq; printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries); @@ -1603,10 +1602,10 @@ __apicdebuginit(void) print_IO_APIC(void) } } printk(KERN_DEBUG "IRQ to pin mappings:\n"); - for_each_irq_desc(irq, desc) { + for_each_active_irq(irq) { struct irq_pin_list *entry; - cfg = get_irq_desc_chip_data(desc); + cfg = get_irq_chip_data(irq); if (!cfg) continue; entry = cfg->irq_2_pin; @@ -2574,9 +2573,8 @@ static struct irq_chip ir_ioapic_chip __read_mostly = { static inline void init_IO_APIC_traps(void) { - int irq; - struct irq_desc *desc; struct irq_cfg *cfg; + unsigned int irq; /* * NOTE! The local APIC isn't very good at handling @@ -2589,8 +2587,8 @@ static inline void init_IO_APIC_traps(void) * Also, we've got to be careful not to trash gate * 0x80, because int 0x80 is hm, kind of importantish. ;) */ - for_each_irq_desc(irq, desc) { - cfg = get_irq_desc_chip_data(desc); + for_each_active_irq(irq) { + cfg = get_irq_chip_data(irq); if (IO_APIC_IRQ(irq) && cfg && !cfg->vector) { /* * Hmm.. We don't have an entry for this, @@ -2601,7 +2599,7 @@ static inline void init_IO_APIC_traps(void) legacy_pic->make_irq(irq); else /* Strange. Oh, well.. */ - desc->chip = &no_irq_chip; + set_irq_chip(irq, &no_irq_chip); } } } -- cgit v1.2.3-70-g09d2 From c2f31c37b7303150ffcce53f6c6ed4ac62952b34 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 30 Sep 2010 12:19:03 +0200 Subject: x86: lguest: Use new irq allocator Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar Cc: Rusty Russell --- arch/x86/lguest/boot.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c index 2d4e6fcac83..73b1e1a1f48 100644 --- a/arch/x86/lguest/boot.c +++ b/arch/x86/lguest/boot.c @@ -838,12 +838,12 @@ static void __init lguest_init_IRQ(void) * rather than set them in lguest_init_IRQ we are called here every time an * lguest device needs an interrupt. * - * FIXME: irq_to_desc_alloc_node() can fail due to lack of memory, we should + * FIXME: irq_alloc_desc_at() can fail due to lack of memory, we should * pass that up! */ void lguest_setup_irq(unsigned int irq) { - irq_to_desc_alloc_node(irq, 0); + irq_alloc_desc_at(irq, 0); set_irq_chip_and_handler_name(irq, &lguest_irq_controller, handle_level_irq, "level"); } -- cgit v1.2.3-70-g09d2 From 2ee39065988d632b403f8470942b0b5edee3632b Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 6 Oct 2010 16:28:51 +0200 Subject: x86: Switch sparse_irq allocations to GFP_KERNEL No callers from atomic context (except boot) anymore. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/kernel/apic/io_apic.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 057b0e13d1c..20e47e04539 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -133,7 +133,7 @@ struct irq_pin_list { static struct irq_pin_list *alloc_irq_pin_list(int node) { - return kzalloc_node(sizeof(struct irq_pin_list), GFP_ATOMIC, node); + return kzalloc_node(sizeof(struct irq_pin_list), GFP_KERNEL, node); } /* irq_cfg is indexed by the sum of all RTEs in all I/O APICs. */ @@ -162,8 +162,8 @@ int __init arch_early_irq_init(void) for (i = 0; i < count; i++) { set_irq_chip_data(i, &cfg[i]); - zalloc_cpumask_var_node(&cfg[i].domain, GFP_NOWAIT, node); - zalloc_cpumask_var_node(&cfg[i].old_domain, GFP_NOWAIT, node); + zalloc_cpumask_var_node(&cfg[i].domain, GFP_KERNEL, node); + zalloc_cpumask_var_node(&cfg[i].old_domain, GFP_KERNEL, node); /* * For legacy IRQ's, start with assigning irq0 to irq15 to * IRQ0_VECTOR to IRQ15_VECTOR on cpu 0. @@ -187,12 +187,12 @@ static struct irq_cfg *alloc_irq_cfg(unsigned int irq, int node) { struct irq_cfg *cfg; - cfg = kzalloc_node(sizeof(*cfg), GFP_ATOMIC, node); + cfg = kzalloc_node(sizeof(*cfg), GFP_KERNEL, node); if (!cfg) return NULL; - if (!zalloc_cpumask_var_node(&cfg->domain, GFP_ATOMIC, node)) + if (!zalloc_cpumask_var_node(&cfg->domain, GFP_KERNEL, node)) goto out_cfg; - if (!zalloc_cpumask_var_node(&cfg->old_domain, GFP_ATOMIC, node)) + if (!zalloc_cpumask_var_node(&cfg->old_domain, GFP_KERNEL, node)) goto out_domain; return cfg; out_domain: @@ -595,14 +595,14 @@ struct IO_APIC_route_entry **alloc_ioapic_entries(void) struct IO_APIC_route_entry **ioapic_entries; ioapic_entries = kzalloc(sizeof(*ioapic_entries) * nr_ioapics, - GFP_ATOMIC); + GFP_KERNEL); if (!ioapic_entries) return 0; for (apic = 0; apic < nr_ioapics; apic++) { ioapic_entries[apic] = kzalloc(sizeof(struct IO_APIC_route_entry) * - nr_ioapic_registers[apic], GFP_ATOMIC); + nr_ioapic_registers[apic], GFP_KERNEL); if (!ioapic_entries[apic]) goto nomem; } -- cgit v1.2.3-70-g09d2 From bf1ebf007911d70a89de9a4a1b36d403e8eb064b Mon Sep 17 00:00:00 2001 From: Daniel Drake Date: Sun, 10 Oct 2010 10:40:32 +0100 Subject: x86, olpc: Add XO-1 poweroff support Add a pm_power_off handler for the OLPC XO-1 laptop. The driver can be built modular and follows the behaviour of the APM driver, setting pm_power_off to NULL on unload. However, the ability to unload the module will probably be removed (with a simple __module_get(THIS_MODULE)) if/when XO-1 suspend/resume support is added to this file at a later date. Signed-off-by: Daniel Drake LKML-Reference: <20101010094032.9AE669D401B@zog.reactivated.net> Signed-off-by: H. Peter Anvin --- arch/x86/Kconfig | 6 ++ arch/x86/kernel/Makefile | 1 + arch/x86/kernel/olpc-xo1.c | 140 +++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 147 insertions(+) create mode 100644 arch/x86/kernel/olpc-xo1.c (limited to 'arch/x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index c2552557134..0bcb8765b1c 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -2066,6 +2066,12 @@ config OLPC Add support for detecting the unique features of the OLPC XO hardware. +config OLPC_XO1 + tristate "OLPC XO-1 support" + depends on OLPC + ---help--- + Add support for non-essential features of the OLPC XO-1 laptop. + config OLPC_OPENFIRMWARE bool "Support for OLPC's Open Firmware" depends on !X86_64 && !X86_PAE diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 0925676266b..e9777791e39 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -104,6 +104,7 @@ obj-$(CONFIG_SCx200) += scx200.o scx200-y += scx200_32.o obj-$(CONFIG_OLPC) += olpc.o +obj-$(CONFIG_OLPC_XO1) += olpc-xo1.o obj-$(CONFIG_OLPC_OPENFIRMWARE) += olpc_ofw.o obj-$(CONFIG_X86_MRST) += mrst.o diff --git a/arch/x86/kernel/olpc-xo1.c b/arch/x86/kernel/olpc-xo1.c new file mode 100644 index 00000000000..f5442c03abc --- /dev/null +++ b/arch/x86/kernel/olpc-xo1.c @@ -0,0 +1,140 @@ +/* + * Support for features of the OLPC XO-1 laptop + * + * Copyright (C) 2010 One Laptop per Child + * Copyright (C) 2006 Red Hat, Inc. + * Copyright (C) 2006 Advanced Micro Devices, 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 Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include +#include +#include +#include +#include + +#include +#include + +#define DRV_NAME "olpc-xo1" + +#define PMS_BAR 4 +#define ACPI_BAR 5 + +/* PMC registers (PMS block) */ +#define PM_SCLK 0x10 +#define PM_IN_SLPCTL 0x20 +#define PM_WKXD 0x34 +#define PM_WKD 0x30 +#define PM_SSC 0x54 + +/* PM registers (ACPI block) */ +#define PM1_CNT 0x08 +#define PM_GPE0_STS 0x18 + +static unsigned long acpi_base; +static unsigned long pms_base; + +static void xo1_power_off(void) +{ + printk(KERN_INFO "OLPC XO-1 power off sequence...\n"); + + /* Enable all of these controls with 0 delay */ + outl(0x40000000, pms_base + PM_SCLK); + outl(0x40000000, pms_base + PM_IN_SLPCTL); + outl(0x40000000, pms_base + PM_WKXD); + outl(0x40000000, pms_base + PM_WKD); + + /* Clear status bits (possibly unnecessary) */ + outl(0x0002ffff, pms_base + PM_SSC); + outl(0xffffffff, acpi_base + PM_GPE0_STS); + + /* Write SLP_EN bit to start the machinery */ + outl(0x00002000, acpi_base + PM1_CNT); +} + +/* Read the base addresses from the PCI BAR info */ +static int __devinit setup_bases(struct pci_dev *pdev) +{ + int r; + + r = pci_enable_device_io(pdev); + if (r) { + dev_err(&pdev->dev, "can't enable device IO\n"); + return r; + } + + r = pci_request_region(pdev, ACPI_BAR, DRV_NAME); + if (r) { + dev_err(&pdev->dev, "can't alloc PCI BAR #%d\n", ACPI_BAR); + return r; + } + + r = pci_request_region(pdev, PMS_BAR, DRV_NAME); + if (r) { + dev_err(&pdev->dev, "can't alloc PCI BAR #%d\n", PMS_BAR); + pci_release_region(pdev, ACPI_BAR); + return r; + } + + acpi_base = pci_resource_start(pdev, ACPI_BAR); + pms_base = pci_resource_start(pdev, PMS_BAR); + + return 0; +} + +static int __devinit olpc_xo1_probe(struct platform_device *pdev) +{ + struct pci_dev *pcidev; + int r; + + pcidev = pci_get_device(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_CS5536_ISA, + NULL); + if (!pdev) + return -ENODEV; + + r = setup_bases(pcidev); + if (r) + return r; + + pm_power_off = xo1_power_off; + + printk(KERN_INFO "OLPC XO-1 support registered\n"); + return 0; +} + +static int __devexit olpc_xo1_remove(struct platform_device *pdev) +{ + pm_power_off = NULL; + return 0; +} + +static struct platform_driver olpc_xo1_driver = { + .driver = { + .name = DRV_NAME, + .owner = THIS_MODULE, + }, + .probe = olpc_xo1_probe, + .remove = __devexit_p(olpc_xo1_remove), +}; + +static int __init olpc_xo1_init(void) +{ + return platform_driver_register(&olpc_xo1_driver); +} + +static void __exit olpc_xo1_exit(void) +{ + platform_driver_unregister(&olpc_xo1_driver); +} + +MODULE_AUTHOR("Daniel Drake "); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("platform:olpc-xo1"); + +module_init(olpc_xo1_init); +module_exit(olpc_xo1_exit); -- cgit v1.2.3-70-g09d2 From 5bcd757f93cc713cf61bbeefceda7539d9afca55 Mon Sep 17 00:00:00 2001 From: Matthew Garrett Date: Mon, 4 Oct 2010 14:59:31 -0400 Subject: x86/amd-iommu: Reenable AMD IOMMU if it's mysteriously vanished over suspend AMD's reference BIOS code had a bug that could result in the firmware failing to reenable the iommu on resume. It transpires that this causes certain less than desirable behaviour when it comes to PCI accesses, to whit them ending up somewhere near Bristol when the more desirable outcome was Edinburgh. Sadness ensues, perhaps along with filesystem corruption. Let's make sure that it gets turned back on, and that we restore its configuration so decisions it makes bear some resemblance to those made by reasonable people rather than crack-addled lemurs who spent all your DMA on Thunderbird. Signed-off-by: Matthew Garrett Signed-off-by: Joerg Roedel --- arch/x86/include/asm/amd_iommu_types.h | 21 ++++-- arch/x86/kernel/amd_iommu_init.c | 122 +++++++++++++++++++++++++++++---- 2 files changed, 123 insertions(+), 20 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/amd_iommu_types.h b/arch/x86/include/asm/amd_iommu_types.h index 08616180dea..bdd20c8891e 100644 --- a/arch/x86/include/asm/amd_iommu_types.h +++ b/arch/x86/include/asm/amd_iommu_types.h @@ -416,13 +416,22 @@ struct amd_iommu { struct dma_ops_domain *default_dom; /* - * This array is required to work around a potential BIOS bug. - * The BIOS may miss to restore parts of the PCI configuration - * space when the system resumes from S3. The result is that the - * IOMMU does not execute commands anymore which leads to system - * failure. + * We can't rely on the BIOS to restore all values on reinit, so we + * need to stash them */ - u32 cache_cfg[4]; + + /* The iommu BAR */ + u32 stored_addr_lo; + u32 stored_addr_hi; + + /* + * Each iommu has 6 l1s, each of which is documented as having 0x12 + * registers + */ + u32 stored_l1[6][0x12]; + + /* The l2 indirect registers */ + u32 stored_l2[0x83]; }; /* diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index 5a170cbbbed..44710d86a43 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -194,6 +194,39 @@ static inline unsigned long tbl_size(int entry_size) return 1UL << shift; } +/* Access to l1 and l2 indexed register spaces */ + +static u32 iommu_read_l1(struct amd_iommu *iommu, u16 l1, u8 address) +{ + u32 val; + + pci_write_config_dword(iommu->dev, 0xf8, (address | l1 << 16)); + pci_read_config_dword(iommu->dev, 0xfc, &val); + return val; +} + +static void iommu_write_l1(struct amd_iommu *iommu, u16 l1, u8 address, u32 val) +{ + pci_write_config_dword(iommu->dev, 0xf8, (address | l1 << 16 | 1 << 31)); + pci_write_config_dword(iommu->dev, 0xfc, val); + pci_write_config_dword(iommu->dev, 0xf8, (address | l1 << 16)); +} + +static u32 iommu_read_l2(struct amd_iommu *iommu, u8 address) +{ + u32 val; + + pci_write_config_dword(iommu->dev, 0xf0, address); + pci_read_config_dword(iommu->dev, 0xf4, &val); + return val; +} + +static void iommu_write_l2(struct amd_iommu *iommu, u8 address, u32 val) +{ + pci_write_config_dword(iommu->dev, 0xf0, (address | 1 << 8)); + pci_write_config_dword(iommu->dev, 0xf4, val); +} + /**************************************************************************** * * AMD IOMMU MMIO register space handling functions @@ -619,6 +652,7 @@ static void __init init_iommu_from_pci(struct amd_iommu *iommu) { int cap_ptr = iommu->cap_ptr; u32 range, misc; + int i, j; pci_read_config_dword(iommu->dev, cap_ptr + MMIO_CAP_HDR_OFFSET, &iommu->cap); @@ -633,12 +667,29 @@ static void __init init_iommu_from_pci(struct amd_iommu *iommu) MMIO_GET_LD(range)); iommu->evt_msi_num = MMIO_MSI_NUM(misc); - if (is_rd890_iommu(iommu->dev)) { - pci_read_config_dword(iommu->dev, 0xf0, &iommu->cache_cfg[0]); - pci_read_config_dword(iommu->dev, 0xf4, &iommu->cache_cfg[1]); - pci_read_config_dword(iommu->dev, 0xf8, &iommu->cache_cfg[2]); - pci_read_config_dword(iommu->dev, 0xfc, &iommu->cache_cfg[3]); - } + if (!is_rd890_iommu(iommu->dev)) + return; + + /* + * Some rd890 systems may not be fully reconfigured by the BIOS, so + * it's necessary for us to store this information so it can be + * reprogrammed on resume + */ + + pci_read_config_dword(iommu->dev, iommu->cap_ptr + 4, + &iommu->stored_addr_lo); + pci_read_config_dword(iommu->dev, iommu->cap_ptr + 8, + &iommu->stored_addr_hi); + + /* Low bit locks writes to configuration space */ + iommu->stored_addr_lo &= ~1; + + for (i = 0; i < 6; i++) + for (j = 0; j < 0x12; j++) + iommu->stored_l1[i][j] = iommu_read_l1(iommu, i, j); + + for (i = 0; i < 0x83; i++) + iommu->stored_l2[i] = iommu_read_l2(iommu, i); } /* @@ -1127,14 +1178,53 @@ static void iommu_init_flags(struct amd_iommu *iommu) iommu_feature_enable(iommu, CONTROL_COHERENT_EN); } -static void iommu_apply_quirks(struct amd_iommu *iommu) +static void iommu_apply_resume_quirks(struct amd_iommu *iommu) { - if (is_rd890_iommu(iommu->dev)) { - pci_write_config_dword(iommu->dev, 0xf0, iommu->cache_cfg[0]); - pci_write_config_dword(iommu->dev, 0xf4, iommu->cache_cfg[1]); - pci_write_config_dword(iommu->dev, 0xf8, iommu->cache_cfg[2]); - pci_write_config_dword(iommu->dev, 0xfc, iommu->cache_cfg[3]); - } + int i, j; + u32 ioc_feature_control; + struct pci_dev *pdev = NULL; + + /* RD890 BIOSes may not have completely reconfigured the iommu */ + if (!is_rd890_iommu(iommu->dev)) + return; + + /* + * First, we need to ensure that the iommu is enabled. This is + * controlled by a register in the northbridge + */ + pdev = pci_get_bus_and_slot(iommu->dev->bus->number, PCI_DEVFN(0, 0)); + + if (!pdev) + return; + + /* Select Northbridge indirect register 0x75 and enable writing */ + pci_write_config_dword(pdev, 0x60, 0x75 | (1 << 7)); + pci_read_config_dword(pdev, 0x64, &ioc_feature_control); + + /* Enable the iommu */ + if (!(ioc_feature_control & 0x1)) + pci_write_config_dword(pdev, 0x64, ioc_feature_control | 1); + + pci_dev_put(pdev); + + /* Restore the iommu BAR */ + pci_write_config_dword(iommu->dev, iommu->cap_ptr + 4, + iommu->stored_addr_lo); + pci_write_config_dword(iommu->dev, iommu->cap_ptr + 8, + iommu->stored_addr_hi); + + /* Restore the l1 indirect regs for each of the 6 l1s */ + for (i = 0; i < 6; i++) + for (j = 0; j < 0x12; j++) + iommu_write_l1(iommu, i, j, iommu->stored_l1[i][j]); + + /* Restore the l2 indirect regs */ + for (i = 0; i < 0x83; i++) + iommu_write_l2(iommu, i, iommu->stored_l2[i]); + + /* Lock PCI setup registers */ + pci_write_config_dword(iommu->dev, iommu->cap_ptr + 4, + iommu->stored_addr_lo | 1); } /* @@ -1147,7 +1237,6 @@ static void enable_iommus(void) for_each_iommu(iommu) { iommu_disable(iommu); - iommu_apply_quirks(iommu); iommu_init_flags(iommu); iommu_set_device_table(iommu); iommu_enable_command_buffer(iommu); @@ -1173,6 +1262,11 @@ static void disable_iommus(void) static int amd_iommu_resume(struct sys_device *dev) { + struct amd_iommu *iommu; + + for_each_iommu(iommu) + iommu_apply_resume_quirks(iommu); + /* re-load the hardware */ enable_iommus(); -- cgit v1.2.3-70-g09d2 From 5d0d71569e671239ae0d905ced9b65cd843f99ee Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Wed, 13 Oct 2010 11:13:21 +0200 Subject: x86/amd-iommu: Update copyright headers This patch updates the copyright headers in all source files of the AMD IOMMU driver. Signed-off-by: Joerg Roedel --- arch/x86/include/asm/amd_iommu.h | 2 +- arch/x86/include/asm/amd_iommu_proto.h | 2 +- arch/x86/include/asm/amd_iommu_types.h | 2 +- arch/x86/kernel/amd_iommu.c | 2 +- arch/x86/kernel/amd_iommu_init.c | 2 +- 5 files changed, 5 insertions(+), 5 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/amd_iommu.h b/arch/x86/include/asm/amd_iommu.h index 5af2982133b..f16a2caca1e 100644 --- a/arch/x86/include/asm/amd_iommu.h +++ b/arch/x86/include/asm/amd_iommu.h @@ -1,5 +1,5 @@ /* - * Copyright (C) 2007-2009 Advanced Micro Devices, Inc. + * Copyright (C) 2007-2010 Advanced Micro Devices, Inc. * Author: Joerg Roedel * Leo Duran * diff --git a/arch/x86/include/asm/amd_iommu_proto.h b/arch/x86/include/asm/amd_iommu_proto.h index cb030374b90..916bc8111a0 100644 --- a/arch/x86/include/asm/amd_iommu_proto.h +++ b/arch/x86/include/asm/amd_iommu_proto.h @@ -1,5 +1,5 @@ /* - * Copyright (C) 2009 Advanced Micro Devices, Inc. + * Copyright (C) 2009-2010 Advanced Micro Devices, Inc. * Author: Joerg Roedel * * This program is free software; you can redistribute it and/or modify it diff --git a/arch/x86/include/asm/amd_iommu_types.h b/arch/x86/include/asm/amd_iommu_types.h index bdd20c8891e..e3509fc303b 100644 --- a/arch/x86/include/asm/amd_iommu_types.h +++ b/arch/x86/include/asm/amd_iommu_types.h @@ -1,5 +1,5 @@ /* - * Copyright (C) 2007-2009 Advanced Micro Devices, Inc. + * Copyright (C) 2007-2010 Advanced Micro Devices, Inc. * Author: Joerg Roedel * Leo Duran * diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index 679b6450382..d2fdb0826df 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2007-2009 Advanced Micro Devices, Inc. + * Copyright (C) 2007-2010 Advanced Micro Devices, Inc. * Author: Joerg Roedel * Leo Duran * diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index 44710d86a43..3cb482e123d 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2007-2009 Advanced Micro Devices, Inc. + * Copyright (C) 2007-2010 Advanced Micro Devices, Inc. * Author: Joerg Roedel * Leo Duran * -- cgit v1.2.3-70-g09d2 From 447b1d43ded08c11f4f3d344b4e07e6dcddbef95 Mon Sep 17 00:00:00 2001 From: Daniel Drake Date: Wed, 13 Oct 2010 19:10:42 +0100 Subject: x86, olpc: Register XO-1 platform devices The upcoming XO-1 rfkill driver (for drivers/platform/x86) will register itself with the name "xo1-rfkill", and the already-merged XO-1 poweroff code uses name "olpc-xo1" Add the necessary mechanics so that these devices are properly initialized on XO-1 laptops. Signed-off-by: Daniel Drake LKML-Reference: <20101013181042.90C8F9D401B@zog.reactivated.net> Cc: Matthew Garrett Signed-off-by: H. Peter Anvin --- arch/x86/kernel/olpc.c | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/olpc.c b/arch/x86/kernel/olpc.c index 74726130259..edaf3fe8dc5 100644 --- a/arch/x86/kernel/olpc.c +++ b/arch/x86/kernel/olpc.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include @@ -223,8 +224,25 @@ static bool __init platform_detect(void) return true; } +static int __init add_xo1_platform_devices(void) +{ + struct platform_device *pdev; + + pdev = platform_device_register_simple("xo1-rfkill", -1, NULL, 0); + if (IS_ERR(pdev)) + return PTR_ERR(pdev); + + pdev = platform_device_register_simple("olpc-xo1", -1, NULL, 0); + if (IS_ERR(pdev)) + return PTR_ERR(pdev); + + return 0; +} + static int __init olpc_init(void) { + int r = 0; + if (!olpc_ofw_present() || !platform_detect()) return 0; @@ -251,6 +269,12 @@ static int __init olpc_init(void) olpc_platform_info.boardrev >> 4, olpc_platform_info.ecver); + if (olpc_platform_info.boardrev < olpc_board_pre(0xd0)) { /* XO-1 */ + r = add_xo1_platform_devices(); + if (r) + return r; + } + return 0; } -- cgit v1.2.3-70-g09d2 From d7acb92fea932ad2e7846480aeacddc2c03c8485 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Wed, 13 Oct 2010 16:00:29 -0700 Subject: x86-64, asm: If the assembler supports fxsave64, use it Kbuild allows for us to probe for the existence of specific constructs in the assembler, use them to find out if we can use fxsave64 and permit the compiler to generate better code. Signed-off-by: H. Peter Anvin --- arch/x86/Makefile | 8 ++++++-- arch/x86/include/asm/i387.h | 2 +- 2 files changed, 7 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/Makefile b/arch/x86/Makefile index 8aa1b59b907..40668a96ca0 100644 --- a/arch/x86/Makefile +++ b/arch/x86/Makefile @@ -96,8 +96,12 @@ cfi := $(call as-instr,.cfi_startproc\n.cfi_rel_offset $(sp-y)$(comma)0\n.cfi_en # is .cfi_signal_frame supported too? cfi-sigframe := $(call as-instr,.cfi_startproc\n.cfi_signal_frame\n.cfi_endproc,-DCONFIG_AS_CFI_SIGNAL_FRAME=1) cfi-sections := $(call as-instr,.cfi_sections .debug_frame,-DCONFIG_AS_CFI_SECTIONS=1) -KBUILD_AFLAGS += $(cfi) $(cfi-sigframe) $(cfi-sections) -KBUILD_CFLAGS += $(cfi) $(cfi-sigframe) $(cfi-sections) + +# does binutils support specific instructions? +asinstr := $(call as-instr,fxsaveq (%rax),-DCONFIG_AS_FXSAVEQ=1) + +KBUILD_AFLAGS += $(cfi) $(cfi-sigframe) $(cfi-sections) $(asinstr) +KBUILD_CFLAGS += $(cfi) $(cfi-sigframe) $(cfi-sections) $(asinstr) LDFLAGS := -m elf_$(UTS_MACHINE) diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h index a73a8d5a5e6..70f105b352e 100644 --- a/arch/x86/include/asm/i387.h +++ b/arch/x86/include/asm/i387.h @@ -175,7 +175,7 @@ static inline void fpu_fxsave(struct fpu *fpu) uses any extended registers for addressing, a second REX prefix will be generated (to the assembler, rex64 followed by semicolon is a separate instruction), and hence the 64-bitness is lost. */ -#if 0 +#ifdef CONFIG_AS_FXSAVEQ /* Using "fxsaveq %0" would be the ideal choice, but is only supported starting with gas 2.16. */ __asm__ __volatile__("fxsaveq %0" -- cgit v1.2.3-70-g09d2 From fef5ba797991f9335bcfc295942b684f9bf613a1 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Wed, 13 Oct 2010 16:02:24 -0700 Subject: xen: Cope with unmapped pages when initializing kernel pagetable Xen requires that all pages containing pagetable entries to be mapped read-only. If pages used for the initial pagetable are already mapped then we can change the mapping to RO. However, if they are initially unmapped, we need to make sure that when they are later mapped, they are also mapped RO. We do this by knowing that the kernel pagetable memory is pre-allocated in the range e820_table_start - e820_table_end, so any pfn within this range should be mapped read-only. However, the pagetable setup code early_ioremaps the pages to write their entries, so we must make sure that mappings created in the early_ioremap fixmap area are mapped RW. (Those mappings are removed before the pages are presented to Xen as pagetable pages.) Signed-off-by: Jeremy Fitzhardinge LKML-Reference: <4CB63A80.8060702@goop.org> Cc: Yinghai Lu Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/io.h | 1 + arch/x86/mm/ioremap.c | 5 +++++ arch/x86/xen/mmu.c | 26 ++++++++++++++++++-------- 3 files changed, 24 insertions(+), 8 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/io.h b/arch/x86/include/asm/io.h index 30a3e977612..66aee6c4123 100644 --- a/arch/x86/include/asm/io.h +++ b/arch/x86/include/asm/io.h @@ -348,6 +348,7 @@ extern void __iomem *early_memremap(resource_size_t phys_addr, unsigned long size); extern void early_iounmap(void __iomem *addr, unsigned long size); extern void fixup_early_ioremap(void); +extern bool is_early_ioremap_ptep(pte_t *ptep); #define IO_SPACE_LIMIT 0xffff diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index 3ba6e0608c5..0369843511d 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -362,6 +362,11 @@ static inline pte_t * __init early_ioremap_pte(unsigned long addr) return &bm_pte[pte_index(addr)]; } +bool __init is_early_ioremap_ptep(pte_t *ptep) +{ + return ptep >= &bm_pte[0] && ptep < &bm_pte[PAGE_SIZE/sizeof(pte_t)]; +} + static unsigned long slot_virt[FIX_BTMAPS_SLOTS] __initdata; void __init early_ioremap_init(void) diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index 4fe04ac0bae..7d55e9ee3a7 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -56,6 +56,7 @@ #include #include #include +#include #include #include @@ -360,7 +361,8 @@ void make_lowmem_page_readonly(void *vaddr) unsigned int level; pte = lookup_address(address, &level); - BUG_ON(pte == NULL); + if (pte == NULL) + return; /* vaddr missing */ ptev = pte_wrprotect(*pte); @@ -375,7 +377,8 @@ void make_lowmem_page_readwrite(void *vaddr) unsigned int level; pte = lookup_address(address, &level); - BUG_ON(pte == NULL); + if (pte == NULL) + return; /* vaddr missing */ ptev = pte_mkwrite(*pte); @@ -1509,13 +1512,25 @@ static void xen_pgd_free(struct mm_struct *mm, pgd_t *pgd) #endif } -#ifdef CONFIG_X86_32 static __init pte_t mask_rw_pte(pte_t *ptep, pte_t pte) { + unsigned long pfn = pte_pfn(pte); + +#ifdef CONFIG_X86_32 /* If there's an existing pte, then don't allow _PAGE_RW to be set */ if (pte_val_ma(*ptep) & _PAGE_PRESENT) pte = __pte_ma(((pte_val_ma(*ptep) & _PAGE_RW) | ~_PAGE_RW) & pte_val_ma(pte)); +#endif + + /* + * If the new pfn is within the range of the newly allocated + * kernel pagetable, and it isn't being mapped into an + * early_ioremap fixmap slot, make sure it is RO. + */ + if (!is_early_ioremap_ptep(ptep) && + pfn >= e820_table_start && pfn < e820_table_end) + pte = pte_wrprotect(pte); return pte; } @@ -1528,7 +1543,6 @@ static __init void xen_set_pte_init(pte_t *ptep, pte_t pte) xen_set_pte(ptep, pte); } -#endif static void pin_pagetable_pfn(unsigned cmd, unsigned long pfn) { @@ -1973,11 +1987,7 @@ static const struct pv_mmu_ops xen_mmu_ops __initdata = { .alloc_pmd_clone = paravirt_nop, .release_pmd = xen_release_pmd_init, -#ifdef CONFIG_X86_64 - .set_pte = xen_set_pte, -#else .set_pte = xen_set_pte_init, -#endif .set_pte_at = xen_set_pte_at, .set_pmd = xen_set_pmd_hyper, -- cgit v1.2.3-70-g09d2 From 3caa37519ccbb200c7fbbf6ff4fb306a30f29425 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Thu, 14 Oct 2010 12:10:36 +0900 Subject: x86: Use __stop_machine() in text_poke_smp() Use __stop_machine() in text_poke_smp() because the caller must get online_cpus before calling text_poke_smp(), but stop_machine() do it again. We don't need it. Signed-off-by: Masami Hiramatsu Cc: Rusty Russell Cc: Ananth N Mavinakayanahalli Cc: 2nddept-manager@sdl.hitachi.co.jp Cc: Mathieu Desnoyers LKML-Reference: <20101014031036.4100.83989.stgit@ltc236.sdl.hitachi.co.jp> Signed-off-by: Ingo Molnar --- arch/x86/kernel/alternative.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index cb0e6d385f6..a36bb90aef5 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -637,7 +637,8 @@ void *__kprobes text_poke_smp(void *addr, const void *opcode, size_t len) tpp.len = len; atomic_set(&stop_machine_first, 1); wrote_text = 0; - stop_machine(stop_machine_text_poke, (void *)&tpp, NULL); + /* Use __stop_machine() because the caller already got online_cpus. */ + __stop_machine(stop_machine_text_poke, (void *)&tpp, NULL); return addr; } -- cgit v1.2.3-70-g09d2 From 3cba11d32bb4b24c3ba257043595772df4b9c7b5 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Thu, 14 Oct 2010 12:10:42 +0900 Subject: kconfig/x86: Add HAVE_TEXT_POKE_SMP config for stop_machine dependency Since the text_poke_smp() definately depends on actual stop_machine() on smp, add that dependency to Kconfig. Signed-off-by: Masami Hiramatsu Cc: Rusty Russell Cc: Ananth N Mavinakayanahalli Cc: 2nddept-manager@sdl.hitachi.co.jp Cc: Mathieu Desnoyers LKML-Reference: <20101014031042.4100.90877.stgit@ltc236.sdl.hitachi.co.jp> Signed-off-by: Ingo Molnar --- arch/x86/Kconfig | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index b431a0824a9..c14d8b4d2f7 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -60,6 +60,7 @@ config X86 select HAVE_ARCH_KMEMCHECK select HAVE_USER_RETURN_NOTIFIER select HAVE_ARCH_JUMP_LABEL + select HAVE_TEXT_POKE_SMP config INSTRUCTION_DECODER def_bool (KPROBES || PERF_EVENTS) @@ -2126,6 +2127,10 @@ config HAVE_ATOMIC_IOMAP def_bool y depends on X86_32 +config HAVE_TEXT_POKE_SMP + bool + select STOP_MACHINE if SMP + source "net/Kconfig" source "drivers/Kconfig" -- cgit v1.2.3-70-g09d2 From 67e87f0a1c5cbc750f81ebf6a128e8ff6f4376cc Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Wed, 13 Oct 2010 16:34:15 -0700 Subject: x86-64: Only set max_pfn_mapped to 512 MiB if we enter via head_64.S head_64.S maps up to 512 MiB, but that is not necessarity true for other entry paths, such as Xen. Thus, co-locate the setting of max_pfn_mapped with the code to actually set up the page tables in head_64.S. The 32-bit code is already so co-located. (The Xen code already sets max_pfn_mapped correctly for its own use case.) -v2: Yinghai fixed the following bug in this patch: | | max_pfn_mapped is in .bss section, so we need to set that | after bss get cleared. Without that we crash on bootup. | | That is safe because Xen does not call x86_64_start_kernel(). | Signed-off-by: Jeremy Fitzhardinge Fixed-by: Yinghai Lu Signed-off-by: H. Peter Anvin LKML-Reference: <4CB6AB24.9020504@kernel.org> Signed-off-by: Ingo Molnar --- arch/x86/kernel/head64.c | 2 ++ arch/x86/kernel/setup.c | 1 - 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index 97adf9828b9..2d2673c28af 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -80,6 +80,8 @@ void __init x86_64_start_kernel(char * real_mode_data) /* Cleanup the over mapped high alias */ cleanup_highmap(); + max_pfn_mapped = KERNEL_IMAGE_SIZE >> PAGE_SHIFT; + for (i = 0; i < NUM_EXCEPTION_VECTORS; i++) { #ifdef CONFIG_EARLY_PRINTK set_intr_gate(i, &early_idt_handlers[i]); diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index b11a238b2e3..c3cebfe7bfc 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -932,7 +932,6 @@ void __init setup_arch(char **cmdline_p) max_low_pfn = max_pfn; high_memory = (void *)__va(max_pfn * PAGE_SIZE - 1) + 1; - max_pfn_mapped = KERNEL_IMAGE_SIZE >> PAGE_SHIFT; #endif /* -- cgit v1.2.3-70-g09d2 From 0eead9ab41da33644ae2c97c57ad03da636a0422 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 14 Oct 2010 10:57:40 -0700 Subject: Don't dump task struct in a.out core-dumps akiphie points out that a.out core-dumps have that odd task struct dumping that was never used and was never really a good idea (it goes back into the mists of history, probably the original core-dumping code). Just remove it. Also do the access_ok() check on dump_write(). It probably doesn't matter (since normal filesystems all seem to do it anyway), but he points out that it's normally done by the VFS layer, so ... [ I suspect that we should possibly do "vfs_write()" instead of calling ->write directly. That also does the whole fsnotify and write statistics thing, which may or may not be a good idea. ] And just to be anal, do this all for the x86-64 32-bit a.out emulation code too, even though it's not enabled (and won't currently even compile) Reported-by: akiphie Signed-off-by: Linus Torvalds --- arch/x86/ia32/ia32_aout.c | 22 +++++----------------- fs/binfmt_aout.c | 4 ---- include/linux/coredump.h | 2 +- 3 files changed, 6 insertions(+), 22 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/ia32/ia32_aout.c b/arch/x86/ia32/ia32_aout.c index 0350311906a..2d93bdbc9ac 100644 --- a/arch/x86/ia32/ia32_aout.c +++ b/arch/x86/ia32/ia32_aout.c @@ -34,7 +34,7 @@ #include #undef WARN_OLD -#undef CORE_DUMP /* probably broken */ +#undef CORE_DUMP /* definitely broken */ static int load_aout_binary(struct linux_binprm *, struct pt_regs *regs); static int load_aout_library(struct file *); @@ -131,21 +131,15 @@ static void set_brk(unsigned long start, unsigned long end) * macros to write out all the necessary info. */ -static int dump_write(struct file *file, const void *addr, int nr) -{ - return file->f_op->write(file, addr, nr, &file->f_pos) == nr; -} +#include #define DUMP_WRITE(addr, nr) \ if (!dump_write(file, (void *)(addr), (nr))) \ goto end_coredump; -#define DUMP_SEEK(offset) \ - if (file->f_op->llseek) { \ - if (file->f_op->llseek(file, (offset), 0) != (offset)) \ - goto end_coredump; \ - } else \ - file->f_pos = (offset) +#define DUMP_SEEK(offset) \ + if (!dump_seek(file, offset)) \ + goto end_coredump; #define START_DATA() (u.u_tsize << PAGE_SHIFT) #define START_STACK(u) (u.start_stack) @@ -217,12 +211,6 @@ static int aout_core_dump(long signr, struct pt_regs *regs, struct file *file, dump_size = dump.u_ssize << PAGE_SHIFT; DUMP_WRITE(dump_start, dump_size); } - /* - * Finally dump the task struct. Not be used by gdb, but - * could be useful - */ - set_fs(KERNEL_DS); - DUMP_WRITE(current, sizeof(*current)); end_coredump: set_fs(fs); return has_dumped; diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c index f96eff04e11..a6395bdb26a 100644 --- a/fs/binfmt_aout.c +++ b/fs/binfmt_aout.c @@ -134,10 +134,6 @@ static int aout_core_dump(struct coredump_params *cprm) if (!dump_write(file, dump_start, dump_size)) goto end_coredump; } -/* Finally dump the task struct. Not be used by gdb, but could be useful */ - set_fs(KERNEL_DS); - if (!dump_write(file, current, sizeof(*current))) - goto end_coredump; end_coredump: set_fs(fs); return has_dumped; diff --git a/include/linux/coredump.h b/include/linux/coredump.h index 8ba66a9d902..59579cfee6a 100644 --- a/include/linux/coredump.h +++ b/include/linux/coredump.h @@ -11,7 +11,7 @@ */ static inline int dump_write(struct file *file, const void *addr, int nr) { - return file->f_op->write(file, addr, nr, &file->f_pos) == nr; + return access_ok(VERIFY_READ, addr, nr) && file->f_op->write(file, addr, nr, &file->f_pos) == nr; } static inline int dump_seek(struct file *file, loff_t off) -- cgit v1.2.3-70-g09d2 From ebc8827f75954fe315492883eee5cb3f355d547d Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Mon, 27 Sep 2010 18:50:51 +0200 Subject: x86: Barf when vmalloc and kmemcheck faults happen in NMI In x86, faults exit by executing the iret instruction, which then reenables NMIs if we faulted in NMI context. Then if a fault happens in NMI, another NMI can nest after the fault exits. But we don't yet support nested NMIs because we have only one NMI stack. To prevent from that, check that vmalloc and kmemcheck faults don't happen in this context. Most of the other kernel faults in NMIs can be more easily spotted by finding explicit copy_from,to_user() calls on review. Signed-off-by: Frederic Weisbecker Cc: Ingo Molnar Cc: Thomas Gleixner Cc: H. Peter Anvin Cc: Mathieu Desnoyers Cc: Peter Zijlstra --- arch/x86/mm/fault.c | 4 ++++ arch/x86/mm/kmemcheck/kmemcheck.c | 2 ++ 2 files changed, 6 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 4c4508e8a20..a24c6cfdccc 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -251,6 +251,8 @@ static noinline __kprobes int vmalloc_fault(unsigned long address) if (!(address >= VMALLOC_START && address < VMALLOC_END)) return -1; + WARN_ON_ONCE(in_nmi()); + /* * Synchronize this task's top level page-table * with the 'reference' page table. @@ -369,6 +371,8 @@ static noinline __kprobes int vmalloc_fault(unsigned long address) if (!(address >= VMALLOC_START && address < VMALLOC_END)) return -1; + WARN_ON_ONCE(in_nmi()); + /* * Copy kernel mappings over when needed. This can also * happen within a race in page table update. In the later diff --git a/arch/x86/mm/kmemcheck/kmemcheck.c b/arch/x86/mm/kmemcheck/kmemcheck.c index b3b531a4f8e..d87dd6d042d 100644 --- a/arch/x86/mm/kmemcheck/kmemcheck.c +++ b/arch/x86/mm/kmemcheck/kmemcheck.c @@ -631,6 +631,8 @@ bool kmemcheck_fault(struct pt_regs *regs, unsigned long address, if (!pte) return false; + WARN_ON_ONCE(in_nmi()); + if (error_code & 2) kmemcheck_access(regs, address, KMEMCHECK_WRITE); else -- cgit v1.2.3-70-g09d2 From 72441cb1fd77d092f09ddfac748955703884c9a7 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Wed, 13 Oct 2010 17:12:30 -0400 Subject: ftrace/x86: Add support for C version of recordmcount This patch adds the support for the C version of recordmcount and compile times show ~ 12% improvement. After verifying this works, other archs can add: HAVE_C_MCOUNT_RECORD in its Kconfig and it will use the C version of recordmcount instead of the perl version. Cc: Cc: Michal Marek Cc: linux-kbuild@vger.kernel.org Cc: John Reiser Signed-off-by: Steven Rostedt --- Makefile | 6 ++++++ arch/x86/Kconfig | 1 + kernel/trace/Kconfig | 5 +++++ scripts/Makefile | 1 + scripts/Makefile.build | 4 ++++ 5 files changed, 17 insertions(+) (limited to 'arch/x86') diff --git a/Makefile b/Makefile index 534c09c255d..0dd3a8d9313 100644 --- a/Makefile +++ b/Makefile @@ -568,6 +568,12 @@ endif ifdef CONFIG_FUNCTION_TRACER KBUILD_CFLAGS += -pg +ifdef CONFIG_DYNAMIC_FTRACE + ifdef CONFIG_HAVE_C_MCOUNT_RECORD + BUILD_C_RECORDMCOUNT := y + export BUILD_C_RECORDMCOUNT + endif +endif endif # We trigger additional mismatches with less inlining diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index c14d8b4d2f7..788b50ef5fc 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -33,6 +33,7 @@ config X86 select HAVE_KRETPROBES select HAVE_OPTPROBES select HAVE_FTRACE_MCOUNT_RECORD + select HAVE_C_MCOUNT_RECORD select HAVE_DYNAMIC_FTRACE select HAVE_FUNCTION_TRACER select HAVE_FUNCTION_GRAPH_TRACER diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index 538501c6ea5..df00fbbaf60 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -49,6 +49,11 @@ config HAVE_SYSCALL_TRACEPOINTS help See Documentation/trace/ftrace-design.txt +config HAVE_C_MCOUNT_RECORD + bool + help + C version of recordmcount available? + config TRACER_MAX_TRACE bool diff --git a/scripts/Makefile b/scripts/Makefile index 842dbc2d5ae..2e088109fbd 100644 --- a/scripts/Makefile +++ b/scripts/Makefile @@ -11,6 +11,7 @@ hostprogs-$(CONFIG_KALLSYMS) += kallsyms hostprogs-$(CONFIG_LOGO) += pnmtologo hostprogs-$(CONFIG_VT) += conmakehash hostprogs-$(CONFIG_IKCONFIG) += bin2c +hostprogs-$(BUILD_C_RECORDMCOUNT) += recordmcount always := $(hostprogs-y) $(hostprogs-m) diff --git a/scripts/Makefile.build b/scripts/Makefile.build index a1a5cf95a68..4d03a7efc68 100644 --- a/scripts/Makefile.build +++ b/scripts/Makefile.build @@ -209,12 +209,16 @@ cmd_modversions = \ endif ifdef CONFIG_FTRACE_MCOUNT_RECORD +ifdef BUILD_C_RECORDMCOUNT +cmd_record_mcount = $(srctree)/scripts/recordmcount "$(@)"; +else cmd_record_mcount = set -e ; perl $(srctree)/scripts/recordmcount.pl "$(ARCH)" \ "$(if $(CONFIG_CPU_BIG_ENDIAN),big,little)" \ "$(if $(CONFIG_64BIT),64,32)" \ "$(OBJDUMP)" "$(OBJCOPY)" "$(CC)" "$(LD)" "$(NM)" "$(RM)" "$(MV)" \ "$(if $(part-of-module),1,0)" "$(@)"; endif +endif define rule_cc_o_c $(call echo-cmd,checksrc) $(cmd_checksrc) \ -- cgit v1.2.3-70-g09d2 From cf4db2597ae93b60efc0a7a4ec08690b75d629b1 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Thu, 14 Oct 2010 23:32:44 -0400 Subject: ftrace: Rename config option HAVE_C_MCOUNT_RECORD to HAVE_C_RECORDMCOUNT The config option used by archs to let the build system know that the C version of the recordmcount works for said arch is currently called HAVE_C_MCOUNT_RECORD which enables BUILD_C_RECORDMCOUNT. To be more consistent with the name that all archs may use, it has been renamed to HAVE_C_RECORDMCOUNT. This will be less confusing since we are building a C recordmcount and not a mcount_record. Suggested-by: Ingo Molnar Cc: Cc: Michal Marek Cc: linux-kbuild@vger.kernel.org Cc: John Reiser Signed-off-by: Steven Rostedt --- Makefile | 2 +- arch/x86/Kconfig | 2 +- kernel/trace/Kconfig | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/Makefile b/Makefile index 0dd3a8d9313..c2362c27300 100644 --- a/Makefile +++ b/Makefile @@ -569,7 +569,7 @@ endif ifdef CONFIG_FUNCTION_TRACER KBUILD_CFLAGS += -pg ifdef CONFIG_DYNAMIC_FTRACE - ifdef CONFIG_HAVE_C_MCOUNT_RECORD + ifdef CONFIG_HAVE_C_RECORDMCOUNT BUILD_C_RECORDMCOUNT := y export BUILD_C_RECORDMCOUNT endif diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 788b50ef5fc..9815221976a 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -33,7 +33,7 @@ config X86 select HAVE_KRETPROBES select HAVE_OPTPROBES select HAVE_FTRACE_MCOUNT_RECORD - select HAVE_C_MCOUNT_RECORD + select HAVE_C_RECORDMCOUNT select HAVE_DYNAMIC_FTRACE select HAVE_FUNCTION_TRACER select HAVE_FUNCTION_GRAPH_TRACER diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index df00fbbaf60..e550d2eda1d 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -49,7 +49,7 @@ config HAVE_SYSCALL_TRACEPOINTS help See Documentation/trace/ftrace-design.txt -config HAVE_C_MCOUNT_RECORD +config HAVE_C_RECORDMCOUNT bool help C version of recordmcount available? -- cgit v1.2.3-70-g09d2 From 9e9006e9090dc1f88d5127cb69f416013e7ecd60 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 14 Oct 2010 10:13:13 -0700 Subject: x86, olpc: XO-1 uses/depends on PCI olpc-xo1 uses pci_*() interfaces so it should depend on PCI. Otherwise we get build failure like: arch/x86/kernel/olpc-xo1.c:65: error: implicit declaration of function 'pci_enable_device_io' arch/x86/kernel/olpc-xo1.c:71: error: implicit declaration of function 'pci_request_region' arch/x86/kernel/olpc-xo1.c:80: error: implicit declaration of function 'pci_release_region' Signed-off-by: Randy Dunlap Acked-by: Daniel Drake Cc: Stephen Rothwell LKML-Reference: <20101014101313.adf7eb2a.randy.dunlap@oracle.com> Signed-off-by: Ingo Molnar --- arch/x86/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 0bcb8765b1c..c010b8d8271 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -2068,7 +2068,7 @@ config OLPC config OLPC_XO1 tristate "OLPC XO-1 support" - depends on OLPC + depends on OLPC && PCI ---help--- Add support for non-essential features of the OLPC XO-1 laptop. -- cgit v1.2.3-70-g09d2 From 6038f373a3dc1f1c26496e60b6c40b164716f07e Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Sun, 15 Aug 2010 18:52:59 +0200 Subject: llseek: automatically add .llseek fop All file_operations should get a .llseek operation so we can make nonseekable_open the default for future file operations without a .llseek pointer. The three cases that we can automatically detect are no_llseek, seq_lseek and default_llseek. For cases where we can we can automatically prove that the file offset is always ignored, we use noop_llseek, which maintains the current behavior of not returning an error from a seek. New drivers should normally not use noop_llseek but instead use no_llseek and call nonseekable_open at open time. Existing drivers can be converted to do the same when the maintainer knows for certain that no user code relies on calling seek on the device file. The generated code is often incorrectly indented and right now contains comments that clarify for each added line why a specific variant was chosen. In the version that gets submitted upstream, the comments will be gone and I will manually fix the indentation, because there does not seem to be a way to do that using coccinelle. Some amount of new code is currently sitting in linux-next that should get the same modifications, which I will do at the end of the merge window. Many thanks to Julia Lawall for helping me learn to write a semantic patch that does all this. ===== begin semantic patch ===== // This adds an llseek= method to all file operations, // as a preparation for making no_llseek the default. // // The rules are // - use no_llseek explicitly if we do nonseekable_open // - use seq_lseek for sequential files // - use default_llseek if we know we access f_pos // - use noop_llseek if we know we don't access f_pos, // but we still want to allow users to call lseek // @ open1 exists @ identifier nested_open; @@ nested_open(...) { <+... nonseekable_open(...) ...+> } @ open exists@ identifier open_f; identifier i, f; identifier open1.nested_open; @@ int open_f(struct inode *i, struct file *f) { <+... ( nonseekable_open(...) | nested_open(...) ) ...+> } @ read disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ read_no_fpos disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { ... when != off } @ write @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ write_no_fpos @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { ... when != off } @ fops0 @ identifier fops; @@ struct file_operations fops = { ... }; @ has_llseek depends on fops0 @ identifier fops0.fops; identifier llseek_f; @@ struct file_operations fops = { ... .llseek = llseek_f, ... }; @ has_read depends on fops0 @ identifier fops0.fops; identifier read_f; @@ struct file_operations fops = { ... .read = read_f, ... }; @ has_write depends on fops0 @ identifier fops0.fops; identifier write_f; @@ struct file_operations fops = { ... .write = write_f, ... }; @ has_open depends on fops0 @ identifier fops0.fops; identifier open_f; @@ struct file_operations fops = { ... .open = open_f, ... }; // use no_llseek if we call nonseekable_open //////////////////////////////////////////// @ nonseekable1 depends on !has_llseek && has_open @ identifier fops0.fops; identifier nso ~= "nonseekable_open"; @@ struct file_operations fops = { ... .open = nso, ... +.llseek = no_llseek, /* nonseekable */ }; @ nonseekable2 depends on !has_llseek @ identifier fops0.fops; identifier open.open_f; @@ struct file_operations fops = { ... .open = open_f, ... +.llseek = no_llseek, /* open uses nonseekable */ }; // use seq_lseek for sequential files ///////////////////////////////////// @ seq depends on !has_llseek @ identifier fops0.fops; identifier sr ~= "seq_read"; @@ struct file_operations fops = { ... .read = sr, ... +.llseek = seq_lseek, /* we have seq_read */ }; // use default_llseek if there is a readdir /////////////////////////////////////////// @ fops1 depends on !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier readdir_e; @@ // any other fop is used that changes pos struct file_operations fops = { ... .readdir = readdir_e, ... +.llseek = default_llseek, /* readdir is present */ }; // use default_llseek if at least one of read/write touches f_pos ///////////////////////////////////////////////////////////////// @ fops2 depends on !fops1 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read.read_f; @@ // read fops use offset struct file_operations fops = { ... .read = read_f, ... +.llseek = default_llseek, /* read accesses f_pos */ }; @ fops3 depends on !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, ... + .llseek = default_llseek, /* write accesses f_pos */ }; // Use noop_llseek if neither read nor write accesses f_pos /////////////////////////////////////////////////////////// @ fops4 depends on !fops1 && !fops2 && !fops3 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; identifier write_no_fpos.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, .read = read_f, ... +.llseek = noop_llseek, /* read and write both use no f_pos */ }; @ depends on has_write && !has_read && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write_no_fpos.write_f; @@ struct file_operations fops = { ... .write = write_f, ... +.llseek = noop_llseek, /* write uses no f_pos */ }; @ depends on has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; @@ struct file_operations fops = { ... .read = read_f, ... +.llseek = noop_llseek, /* read uses no f_pos */ }; @ depends on !has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; @@ struct file_operations fops = { ... +.llseek = noop_llseek, /* no read or write fn */ }; ===== End semantic patch ===== Signed-off-by: Arnd Bergmann Cc: Julia Lawall Cc: Christoph Hellwig --- arch/arm/kernel/etm.c | 1 + arch/arm/mach-msm/last_radio_log.c | 3 ++- arch/arm/mach-msm/smd_debug.c | 1 + arch/arm/plat-mxc/audmux-v2.c | 1 + arch/avr32/boards/mimc200/fram.c | 1 + arch/blackfin/kernel/kgdb_test.c | 1 + arch/blackfin/mach-bf561/coreb.c | 1 + arch/cris/arch-v10/drivers/ds1302.c | 1 + arch/cris/arch-v10/drivers/gpio.c | 1 + arch/cris/arch-v10/drivers/i2c.c | 1 + arch/cris/arch-v10/drivers/pcf8563.c | 1 + arch/cris/arch-v10/drivers/sync_serial.c | 3 ++- arch/cris/arch-v32/drivers/cryptocop.c | 3 ++- arch/cris/arch-v32/drivers/i2c.c | 1 + arch/cris/arch-v32/drivers/mach-a3/gpio.c | 1 + arch/cris/arch-v32/drivers/mach-fs/gpio.c | 1 + arch/cris/arch-v32/drivers/pcf8563.c | 1 + arch/cris/arch-v32/drivers/sync_serial.c | 3 ++- arch/cris/kernel/profile.c | 1 + arch/ia64/kernel/salinfo.c | 2 ++ arch/ia64/sn/kernel/sn2/sn_hwperf.c | 1 + arch/m68k/bvme6000/rtc.c | 1 + arch/m68k/mvme16x/rtc.c | 1 + arch/mips/kernel/rtlx.c | 3 ++- arch/mips/kernel/vpe.c | 3 ++- arch/mips/sibyte/common/sb_tbprof.c | 1 + arch/powerpc/kernel/lparcfg.c | 1 + arch/powerpc/kernel/rtas_flash.c | 3 +++ arch/powerpc/kernel/rtasd.c | 1 + arch/powerpc/platforms/iseries/mf.c | 1 + arch/powerpc/platforms/pseries/reconfig.c | 3 ++- arch/powerpc/platforms/pseries/scanlog.c | 1 + arch/s390/crypto/prng.c | 1 + arch/s390/hypfs/hypfs_diag.c | 1 + arch/s390/hypfs/hypfs_vm.c | 1 + arch/s390/hypfs/inode.c | 1 + arch/s390/kernel/debug.c | 1 + arch/sh/boards/mach-landisk/gio.c | 1 + arch/sparc/kernel/apc.c | 1 + arch/sparc/kernel/mdesc.c | 1 + arch/tile/kernel/hardwall.c | 1 + arch/um/drivers/harddog_kern.c | 1 + arch/um/drivers/mconsole_kern.c | 1 + arch/um/drivers/mmapper_kern.c | 1 + arch/um/drivers/random.c | 1 + arch/x86/kernel/apm_32.c | 1 + arch/x86/kernel/cpu/mcheck/mce-severity.c | 1 + arch/x86/kernel/cpu/mcheck/mce.c | 1 + arch/x86/kernel/kdebugfs.c | 1 + arch/x86/kernel/microcode_core.c | 1 + arch/x86/kernel/tlb_uv.c | 1 + arch/x86/xen/debugfs.c | 1 + block/bsg.c | 1 + drivers/acpi/apei/erst-dbg.c | 1 + drivers/acpi/debugfs.c | 1 + drivers/acpi/ec_sys.c | 1 + drivers/acpi/event.c | 1 + drivers/block/DAC960.c | 3 ++- drivers/block/aoe/aoechr.c | 1 + drivers/block/paride/pg.c | 1 + drivers/block/paride/pt.c | 1 + drivers/block/pktcdvd.c | 1 + drivers/bluetooth/btmrvl_debugfs.c | 10 +++++++ drivers/bluetooth/hci_vhci.c | 1 + drivers/char/apm-emulation.c | 1 + drivers/char/bfin-otp.c | 1 + drivers/char/briq_panel.c | 1 + drivers/char/bsr.c | 1 + drivers/char/cs5535_gpio.c | 3 ++- drivers/char/ds1302.c | 1 + drivers/char/ds1620.c | 1 + drivers/char/dsp56k.c | 1 + drivers/char/dtlk.c | 1 + drivers/char/genrtc.c | 1 + drivers/char/hw_random/core.c | 1 + drivers/char/ip2/ip2main.c | 1 + drivers/char/ipmi/ipmi_devintf.c | 1 + drivers/char/ipmi/ipmi_watchdog.c | 1 + drivers/char/istallion.c | 1 + drivers/char/lp.c | 1 + drivers/char/mem.c | 3 +++ drivers/char/misc.c | 1 + drivers/char/mmtimer.c | 1 + drivers/char/mspec.c | 9 ++++--- drivers/char/mwave/mwavedd.c | 3 ++- drivers/char/nwbutton.c | 1 + drivers/char/pc8736x_gpio.c | 1 + drivers/char/pcmcia/cm4000_cs.c | 1 + drivers/char/pcmcia/cm4040_cs.c | 1 + drivers/char/random.c | 2 ++ drivers/char/rio/rio_linux.c | 1 + drivers/char/scx200_gpio.c | 1 + drivers/char/snsc.c | 1 + drivers/char/stallion.c | 1 + drivers/char/sx.c | 1 + drivers/char/sysrq.c | 1 + drivers/char/tb0219.c | 1 + drivers/char/tlclk.c | 1 + drivers/char/toshiba.c | 1 + drivers/char/uv_mmtimer.c | 1 + drivers/char/xilinx_hwicap/xilinx_hwicap.c | 1 + drivers/dma/coh901318.c | 1 + drivers/firewire/nosy.c | 1 + drivers/gpu/drm/drm_drv.c | 3 ++- drivers/gpu/drm/i810/i810_dma.c | 1 + drivers/gpu/drm/i830/i830_dma.c | 1 + drivers/gpu/drm/i915/i915_debugfs.c | 1 + drivers/gpu/vga/vgaarb.c | 1 + drivers/hid/hid-debug.c | 1 + drivers/hid/hid-roccat.c | 1 + drivers/hid/hidraw.c | 1 + drivers/hid/usbhid/hiddev.c | 1 + drivers/hwmon/asus_atk0110.c | 1 + drivers/ide/ide-tape.c | 1 + drivers/idle/i7300_idle.c | 1 + drivers/infiniband/hw/ipath/ipath_diag.c | 4 ++- drivers/infiniband/hw/ipath/ipath_file_ops.c | 3 ++- drivers/infiniband/hw/ipath/ipath_fs.c | 3 +++ drivers/infiniband/hw/qib/qib_diag.c | 4 ++- drivers/infiniband/hw/qib/qib_file_ops.c | 3 ++- drivers/infiniband/hw/qib/qib_fs.c | 1 + drivers/input/evdev.c | 3 ++- drivers/input/input.c | 1 + drivers/input/joydev.c | 1 + drivers/input/misc/uinput.c | 1 + drivers/input/mousedev.c | 1 + drivers/input/serio/serio_raw.c | 1 + drivers/isdn/mISDN/timerdev.c | 1 + drivers/lguest/lguest_user.c | 1 + drivers/macintosh/ans-lcd.c | 1 + drivers/macintosh/via-pmu.c | 1 + drivers/md/dm-ioctl.c | 1 + drivers/media/IR/imon.c | 6 +++-- drivers/media/IR/lirc_dev.c | 1 + drivers/media/dvb/bt8xx/dst_ca.c | 3 ++- drivers/media/dvb/dvb-core/dmxdev.c | 2 ++ drivers/media/dvb/dvb-core/dvb_ca_en50221.c | 1 + drivers/media/dvb/dvb-core/dvb_frontend.c | 3 ++- drivers/media/dvb/dvb-core/dvb_net.c | 1 + drivers/media/dvb/dvb-core/dvbdev.c | 1 + drivers/media/dvb/firewire/firedtv-ci.c | 1 + drivers/media/dvb/ttpci/av7110.c | 1 + drivers/media/dvb/ttpci/av7110_av.c | 2 ++ drivers/media/dvb/ttpci/av7110_ca.c | 1 + drivers/media/dvb/ttpci/av7110_ir.c | 1 + drivers/mfd/ab3100-core.c | 1 + drivers/misc/hpilo.c | 1 + drivers/misc/phantom.c | 1 + drivers/misc/sgi-gru/grufile.c | 1 + drivers/mmc/core/debugfs.c | 1 + drivers/mtd/ubi/cdev.c | 1 + drivers/net/caif/caif_spi.c | 6 +++-- drivers/net/cxgb4/cxgb4_main.c | 1 + drivers/net/ppp_generic.c | 3 ++- drivers/net/wimax/i2400m/debugfs.c | 2 ++ drivers/net/wireless/airo.c | 24 +++++++++++------ drivers/net/wireless/ath/ath5k/debug.c | 7 +++++ drivers/net/wireless/ath/ath9k/debug.c | 33 ++++++++++++++++-------- drivers/net/wireless/ath/ath9k/htc_drv_main.c | 9 ++++--- drivers/net/wireless/iwlwifi/iwl-3945-rs.c | 1 + drivers/net/wireless/iwlwifi/iwl-agn-rs.c | 3 +++ drivers/net/wireless/iwmc3200wifi/debugfs.c | 4 +++ drivers/net/wireless/iwmc3200wifi/sdio.c | 1 + drivers/net/wireless/libertas/debugfs.c | 1 + drivers/net/wireless/ray_cs.c | 2 ++ drivers/net/wireless/rt2x00/rt2x00debug.c | 4 +++ drivers/net/wireless/wl12xx/wl1251_debugfs.c | 2 ++ drivers/net/wireless/wl12xx/wl1271_debugfs.c | 4 ++- drivers/oprofile/oprofile_files.c | 8 +++++- drivers/oprofile/oprofilefs.c | 3 +++ drivers/pci/pcie/aer/aer_inject.c | 1 + drivers/platform/x86/sony-laptop.c | 1 + drivers/rtc/rtc-m41t80.c | 1 + drivers/s390/block/dasd_eer.c | 1 + drivers/s390/char/fs3270.c | 1 + drivers/s390/char/monreader.c | 1 + drivers/s390/char/monwriter.c | 1 + drivers/s390/char/tape_char.c | 1 + drivers/s390/char/vmcp.c | 1 + drivers/s390/char/vmlogrdr.c | 1 + drivers/s390/char/vmwatchdog.c | 1 + drivers/s390/char/zcore.c | 2 ++ drivers/s390/cio/chsc_sch.c | 1 + drivers/s390/cio/css.c | 1 + drivers/s390/crypto/zcrypt_api.c | 3 ++- drivers/s390/scsi/zfcp_cfdc.c | 3 ++- drivers/sbus/char/display7seg.c | 1 + drivers/sbus/char/envctrl.c | 1 + drivers/scsi/3w-9xxx.c | 3 ++- drivers/scsi/3w-sas.c | 3 ++- drivers/scsi/3w-xxxx.c | 3 ++- drivers/scsi/aacraid/linit.c | 1 + drivers/scsi/ch.c | 1 + drivers/scsi/dpt_i2o.c | 1 + drivers/scsi/gdth.c | 1 + drivers/scsi/megaraid.c | 1 + drivers/scsi/megaraid/megaraid_mm.c | 1 + drivers/scsi/megaraid/megaraid_sas.c | 1 + drivers/scsi/mpt2sas/mpt2sas_ctl.c | 1 + drivers/scsi/osd/osd_uld.c | 1 + drivers/scsi/pmcraid.c | 1 + drivers/scsi/qla2xxx/qla_os.c | 1 + drivers/scsi/scsi_tgt_if.c | 1 + drivers/scsi/sg.c | 1 + drivers/serial/mfd.c | 2 ++ drivers/spi/dw_spi.c | 1 + drivers/spi/spidev.c | 1 + drivers/staging/comedi/comedi_fops.c | 1 + drivers/staging/crystalhd/crystalhd_lnx.c | 1 + drivers/staging/dream/camera/msm_camera.c | 3 +++ drivers/staging/dream/pmem.c | 2 ++ drivers/staging/dream/qdsp5/adsp_driver.c | 1 + drivers/staging/dream/qdsp5/audio_aac.c | 1 + drivers/staging/dream/qdsp5/audio_amrnb.c | 1 + drivers/staging/dream/qdsp5/audio_evrc.c | 1 + drivers/staging/dream/qdsp5/audio_in.c | 2 ++ drivers/staging/dream/qdsp5/audio_mp3.c | 1 + drivers/staging/dream/qdsp5/audio_out.c | 2 ++ drivers/staging/dream/qdsp5/audio_qcelp.c | 1 + drivers/staging/dream/qdsp5/evlog.h | 1 + drivers/staging/dream/qdsp5/snd.c | 1 + drivers/staging/frontier/alphatrack.c | 1 + drivers/staging/frontier/tranzport.c | 1 + drivers/staging/iio/industrialio-core.c | 1 + drivers/staging/iio/industrialio-ring.c | 1 + drivers/staging/lirc/lirc_imon.c | 3 ++- drivers/staging/lirc/lirc_it87.c | 1 + drivers/staging/lirc/lirc_sasem.c | 1 + drivers/staging/memrar/memrar_handler.c | 1 + drivers/staging/panel/panel.c | 1 + drivers/staging/tidspbridge/rmgr/drv_interface.c | 1 + drivers/telephony/ixj.c | 3 ++- drivers/telephony/phonedev.c | 1 + drivers/uio/uio.c | 1 + drivers/usb/class/cdc-wdm.c | 3 ++- drivers/usb/class/usblp.c | 1 + drivers/usb/class/usbtmc.c | 1 + drivers/usb/core/file.c | 1 + drivers/usb/gadget/f_hid.c | 1 + drivers/usb/gadget/printer.c | 3 ++- drivers/usb/host/ehci-dbg.c | 4 +++ drivers/usb/host/ohci-dbg.c | 3 +++ drivers/usb/image/mdc800.c | 1 + drivers/usb/misc/adutux.c | 1 + drivers/usb/misc/idmouse.c | 1 + drivers/usb/misc/iowarrior.c | 1 + drivers/usb/misc/ldusb.c | 1 + drivers/usb/misc/rio500.c | 1 + drivers/usb/misc/usblcd.c | 1 + drivers/usb/usb-skeleton.c | 1 + drivers/vhost/net.c | 1 + drivers/video/fbmem.c | 1 + drivers/video/mbx/mbxdebugfs.c | 6 +++++ drivers/watchdog/ar7_wdt.c | 1 + drivers/watchdog/cpwd.c | 1 + drivers/watchdog/ep93xx_wdt.c | 1 + drivers/watchdog/omap_wdt.c | 1 + drivers/xen/evtchn.c | 1 + drivers/xen/xenfs/super.c | 1 + drivers/xen/xenfs/xenbus.c | 1 + fs/afs/mntpt.c | 1 + fs/autofs4/dev-ioctl.c | 1 + fs/binfmt_misc.c | 3 +++ fs/btrfs/super.c | 1 + fs/cachefiles/daemon.c | 1 + fs/char_dev.c | 1 + fs/coda/pioctl.c | 1 + fs/coda/psdev.c | 1 + fs/debugfs/file.c | 3 +++ fs/dlm/debug_fs.c | 3 ++- fs/dlm/plock.c | 3 ++- fs/dlm/user.c | 3 +++ fs/ecryptfs/file.c | 1 + fs/ecryptfs/miscdev.c | 1 + fs/eventfd.c | 1 + fs/eventpoll.c | 3 ++- fs/fifo.c | 1 + fs/fuse/control.c | 4 +++ fs/fuse/cuse.c | 1 + fs/gfs2/file.c | 2 ++ fs/hppfs/hppfs.c | 1 + fs/hugetlbfs/inode.c | 1 + fs/logfs/dir.c | 1 + fs/nfsd/nfsctl.c | 1 + fs/no-block.c | 1 + fs/notify/fanotify/fanotify_user.c | 1 + fs/notify/inotify/inotify_user.c | 1 + fs/ocfs2/dlmfs/dlmfs.c | 1 + fs/ocfs2/stack_user.c | 1 + fs/proc/base.c | 8 ++++++ fs/proc/proc_sysctl.c | 1 + fs/proc/root.c | 1 + fs/proc/task_mmu.c | 1 + fs/romfs/super.c | 1 + fs/signalfd.c | 1 + fs/squashfs/dir.c | 3 ++- fs/timerfd.c | 1 + fs/ubifs/debug.c | 1 + ipc/mqueue.c | 1 + ipc/shm.c | 2 ++ kernel/configs.c | 1 + kernel/gcov/fs.c | 1 + kernel/kprobes.c | 1 + kernel/pm_qos_params.c | 1 + kernel/profile.c | 1 + kernel/trace/blktrace.c | 2 ++ kernel/trace/ftrace.c | 2 ++ kernel/trace/ring_buffer.c | 1 + kernel/trace/trace_events.c | 6 +++++ kernel/trace/trace_stack.c | 1 + lib/dma-debug.c | 1 + net/atm/proc.c | 1 + net/dccp/probe.c | 1 + net/ipv4/tcp_probe.c | 1 + net/mac80211/debugfs.c | 19 +++++++++----- net/mac80211/rate.c | 1 + net/mac80211/rc80211_minstrel_debugfs.c | 1 + net/mac80211/rc80211_pid_debugfs.c | 1 + net/netfilter/xt_recent.c | 1 + net/nonet.c | 1 + net/rfkill/core.c | 1 + net/sctp/probe.c | 1 + net/socket.c | 1 + net/sunrpc/cache.c | 2 ++ net/wireless/debugfs.c | 1 + samples/kfifo/bytestream-example.c | 1 + samples/kfifo/inttype-example.c | 1 + samples/kfifo/record-example.c | 1 + samples/tracepoints/tracepoint-sample.c | 1 + security/apparmor/apparmorfs.c | 9 ++++--- security/inode.c | 1 + security/smack/smackfs.c | 5 ++++ sound/core/seq/oss/seq_oss.c | 1 + sound/core/sound.c | 3 ++- sound/oss/msnd_pinnacle.c | 1 + sound/soc/soc-core.c | 1 + sound/soc/soc-dapm.c | 1 + sound/sound_core.c | 1 + virt/kvm/kvm_main.c | 3 +++ 339 files changed, 538 insertions(+), 73 deletions(-) (limited to 'arch/x86') diff --git a/arch/arm/kernel/etm.c b/arch/arm/kernel/etm.c index 33c7077174d..30b8878f120 100644 --- a/arch/arm/kernel/etm.c +++ b/arch/arm/kernel/etm.c @@ -314,6 +314,7 @@ static const struct file_operations etb_fops = { .read = etb_read, .open = etb_open, .release = etb_release, + .llseek = no_llseek, }; static struct miscdevice etb_miscdev = { diff --git a/arch/arm/mach-msm/last_radio_log.c b/arch/arm/mach-msm/last_radio_log.c index b64ba5a9868..1e243f46a96 100644 --- a/arch/arm/mach-msm/last_radio_log.c +++ b/arch/arm/mach-msm/last_radio_log.c @@ -48,7 +48,8 @@ static ssize_t last_radio_log_read(struct file *file, char __user *buf, } static struct file_operations last_radio_log_fops = { - .read = last_radio_log_read + .read = last_radio_log_read, + .llseek = default_llseek, }; void msm_init_last_radio_log(struct module *owner) diff --git a/arch/arm/mach-msm/smd_debug.c b/arch/arm/mach-msm/smd_debug.c index 3b2dd717b78..f91c3b7bc65 100644 --- a/arch/arm/mach-msm/smd_debug.c +++ b/arch/arm/mach-msm/smd_debug.c @@ -212,6 +212,7 @@ static int debug_open(struct inode *inode, struct file *file) static const struct file_operations debug_ops = { .read = debug_read, .open = debug_open, + .llseek = default_llseek, }; static void debug_create(const char *name, mode_t mode, diff --git a/arch/arm/plat-mxc/audmux-v2.c b/arch/arm/plat-mxc/audmux-v2.c index f9e7cdbd000..25ad95ba92a 100644 --- a/arch/arm/plat-mxc/audmux-v2.c +++ b/arch/arm/plat-mxc/audmux-v2.c @@ -137,6 +137,7 @@ static ssize_t audmux_read_file(struct file *file, char __user *user_buf, static const struct file_operations audmux_debugfs_fops = { .open = audmux_open_file, .read = audmux_read_file, + .llseek = default_llseek, }; static void audmux_debugfs_init(void) diff --git a/arch/avr32/boards/mimc200/fram.c b/arch/avr32/boards/mimc200/fram.c index 54fbd95cee9..9764a1a1073 100644 --- a/arch/avr32/boards/mimc200/fram.c +++ b/arch/avr32/boards/mimc200/fram.c @@ -41,6 +41,7 @@ static int fram_mmap(struct file *filp, struct vm_area_struct *vma) static const struct file_operations fram_fops = { .owner = THIS_MODULE, .mmap = fram_mmap, + .llseek = noop_llseek, }; #define FRAM_MINOR 0 diff --git a/arch/blackfin/kernel/kgdb_test.c b/arch/blackfin/kernel/kgdb_test.c index 9a4b0759438..08c0236acf3 100644 --- a/arch/blackfin/kernel/kgdb_test.c +++ b/arch/blackfin/kernel/kgdb_test.c @@ -88,6 +88,7 @@ static const struct file_operations kgdb_test_proc_fops = { .owner = THIS_MODULE, .read = kgdb_test_proc_read, .write = kgdb_test_proc_write, + .llseek = noop_llseek, }; static int __init kgdbtest_init(void) diff --git a/arch/blackfin/mach-bf561/coreb.c b/arch/blackfin/mach-bf561/coreb.c index deb2271d09a..c6a4c8f2d37 100644 --- a/arch/blackfin/mach-bf561/coreb.c +++ b/arch/blackfin/mach-bf561/coreb.c @@ -51,6 +51,7 @@ coreb_ioctl(struct file *file, unsigned int cmd, unsigned long arg) static const struct file_operations coreb_fops = { .owner = THIS_MODULE, .unlocked_ioctl = coreb_ioctl, + .llseek = noop_llseek, }; static struct miscdevice coreb_dev = { diff --git a/arch/cris/arch-v10/drivers/ds1302.c b/arch/cris/arch-v10/drivers/ds1302.c index 884275629ef..95aadb4a8cf 100644 --- a/arch/cris/arch-v10/drivers/ds1302.c +++ b/arch/cris/arch-v10/drivers/ds1302.c @@ -387,6 +387,7 @@ print_rtc_status(void) static const struct file_operations rtc_fops = { .owner = THIS_MODULE, .unlocked_ioctl = rtc_unlocked_ioctl, + .llseek = noop_llseek, }; /* Probe for the chip by writing something to its RAM and try reading it back. */ diff --git a/arch/cris/arch-v10/drivers/gpio.c b/arch/cris/arch-v10/drivers/gpio.c index a07b6d25b0c..a276f081173 100644 --- a/arch/cris/arch-v10/drivers/gpio.c +++ b/arch/cris/arch-v10/drivers/gpio.c @@ -745,6 +745,7 @@ static const struct file_operations gpio_fops = { .write = gpio_write, .open = gpio_open, .release = gpio_release, + .llseek = noop_llseek, }; static void ioif_watcher(const unsigned int gpio_in_available, diff --git a/arch/cris/arch-v10/drivers/i2c.c b/arch/cris/arch-v10/drivers/i2c.c index 77a94181381..c413539d420 100644 --- a/arch/cris/arch-v10/drivers/i2c.c +++ b/arch/cris/arch-v10/drivers/i2c.c @@ -617,6 +617,7 @@ static const struct file_operations i2c_fops = { .unlocked_ioctl = i2c_ioctl, .open = i2c_open, .release = i2c_release, + .llseek = noop_llseek, }; int __init diff --git a/arch/cris/arch-v10/drivers/pcf8563.c b/arch/cris/arch-v10/drivers/pcf8563.c index 7dcb1f85f42..7aa6ff00a11 100644 --- a/arch/cris/arch-v10/drivers/pcf8563.c +++ b/arch/cris/arch-v10/drivers/pcf8563.c @@ -64,6 +64,7 @@ static int voltage_low; static const struct file_operations pcf8563_fops = { .owner = THIS_MODULE, .unlocked_ioctl = pcf8563_unlocked_ioctl, + .llseek = noop_llseek, }; unsigned char diff --git a/arch/cris/arch-v10/drivers/sync_serial.c b/arch/cris/arch-v10/drivers/sync_serial.c index ee2dd4323da..e501632fa8c 100644 --- a/arch/cris/arch-v10/drivers/sync_serial.c +++ b/arch/cris/arch-v10/drivers/sync_serial.c @@ -250,7 +250,8 @@ static const struct file_operations sync_serial_fops = { .poll = sync_serial_poll, .unlocked_ioctl = sync_serial_ioctl, .open = sync_serial_open, - .release = sync_serial_release + .release = sync_serial_release, + .llseek = noop_llseek, }; static int __init etrax_sync_serial_init(void) diff --git a/arch/cris/arch-v32/drivers/cryptocop.c b/arch/cris/arch-v32/drivers/cryptocop.c index b07646a3050..dcb43fddfb9 100644 --- a/arch/cris/arch-v32/drivers/cryptocop.c +++ b/arch/cris/arch-v32/drivers/cryptocop.c @@ -281,7 +281,8 @@ const struct file_operations cryptocop_fops = { .owner = THIS_MODULE, .open = cryptocop_open, .release = cryptocop_release, - .unlocked_ioctl = cryptocop_ioctl + .unlocked_ioctl = cryptocop_ioctl, + .llseek = noop_llseek, }; diff --git a/arch/cris/arch-v32/drivers/i2c.c b/arch/cris/arch-v32/drivers/i2c.c index 5a3e900c9a7..ddb23996f11 100644 --- a/arch/cris/arch-v32/drivers/i2c.c +++ b/arch/cris/arch-v32/drivers/i2c.c @@ -698,6 +698,7 @@ static const struct file_operations i2c_fops = { .unlocked_ioctl = i2c_ioctl, .open = i2c_open, .release = i2c_release, + .llseek = noop_llseek, }; static int __init i2c_init(void) diff --git a/arch/cris/arch-v32/drivers/mach-a3/gpio.c b/arch/cris/arch-v32/drivers/mach-a3/gpio.c index 2dcd27adbad..06b24ebda41 100644 --- a/arch/cris/arch-v32/drivers/mach-a3/gpio.c +++ b/arch/cris/arch-v32/drivers/mach-a3/gpio.c @@ -893,6 +893,7 @@ static const struct file_operations gpio_fops = { .write = gpio_write, .open = gpio_open, .release = gpio_release, + .llseek = noop_llseek, }; #ifdef CONFIG_ETRAX_VIRTUAL_GPIO diff --git a/arch/cris/arch-v32/drivers/mach-fs/gpio.c b/arch/cris/arch-v32/drivers/mach-fs/gpio.c index 5ec8a7d4e7d..0649c8bea67 100644 --- a/arch/cris/arch-v32/drivers/mach-fs/gpio.c +++ b/arch/cris/arch-v32/drivers/mach-fs/gpio.c @@ -870,6 +870,7 @@ static const struct file_operations gpio_fops = { .write = gpio_write, .open = gpio_open, .release = gpio_release, + .llseek = noop_llseek, }; #ifdef CONFIG_ETRAX_VIRTUAL_GPIO diff --git a/arch/cris/arch-v32/drivers/pcf8563.c b/arch/cris/arch-v32/drivers/pcf8563.c index bef6eb53b15..0f7b101ee5e 100644 --- a/arch/cris/arch-v32/drivers/pcf8563.c +++ b/arch/cris/arch-v32/drivers/pcf8563.c @@ -60,6 +60,7 @@ static int voltage_low; static const struct file_operations pcf8563_fops = { .owner = THIS_MODULE, .unlocked_ioctl = pcf8563_unlocked_ioctl, + .llseek = noop_llseek, }; unsigned char diff --git a/arch/cris/arch-v32/drivers/sync_serial.c b/arch/cris/arch-v32/drivers/sync_serial.c index ca248f3adb8..a2e8a8c3985 100644 --- a/arch/cris/arch-v32/drivers/sync_serial.c +++ b/arch/cris/arch-v32/drivers/sync_serial.c @@ -247,7 +247,8 @@ static const struct file_operations sync_serial_fops = { .poll = sync_serial_poll, .unlocked_ioctl = sync_serial_ioctl, .open = sync_serial_open, - .release = sync_serial_release + .release = sync_serial_release, + .llseek = noop_llseek, }; static int __init etrax_sync_serial_init(void) diff --git a/arch/cris/kernel/profile.c b/arch/cris/kernel/profile.c index 195ec5fa0dd..b82e08615d1 100644 --- a/arch/cris/kernel/profile.c +++ b/arch/cris/kernel/profile.c @@ -59,6 +59,7 @@ write_cris_profile(struct file *file, const char __user *buf, static const struct file_operations cris_proc_profile_operations = { .read = read_cris_profile, .write = write_cris_profile, + .llseek = default_llseek, }; static int __init init_cris_profile(void) diff --git a/arch/ia64/kernel/salinfo.c b/arch/ia64/kernel/salinfo.c index aa8b5fa1a8d..6f22c404163 100644 --- a/arch/ia64/kernel/salinfo.c +++ b/arch/ia64/kernel/salinfo.c @@ -354,6 +354,7 @@ retry: static const struct file_operations salinfo_event_fops = { .open = salinfo_event_open, .read = salinfo_event_read, + .llseek = noop_llseek, }; static int @@ -571,6 +572,7 @@ static const struct file_operations salinfo_data_fops = { .release = salinfo_log_release, .read = salinfo_log_read, .write = salinfo_log_write, + .llseek = default_llseek, }; static int __cpuinit diff --git a/arch/ia64/sn/kernel/sn2/sn_hwperf.c b/arch/ia64/sn/kernel/sn2/sn_hwperf.c index fa1eceed0d2..30862c0358c 100644 --- a/arch/ia64/sn/kernel/sn2/sn_hwperf.c +++ b/arch/ia64/sn/kernel/sn2/sn_hwperf.c @@ -860,6 +860,7 @@ error: static const struct file_operations sn_hwperf_fops = { .unlocked_ioctl = sn_hwperf_ioctl, + .llseek = noop_llseek, }; static struct miscdevice sn_hwperf_dev = { diff --git a/arch/m68k/bvme6000/rtc.c b/arch/m68k/bvme6000/rtc.c index cb8617bb194..1c4d4c7bf4d 100644 --- a/arch/m68k/bvme6000/rtc.c +++ b/arch/m68k/bvme6000/rtc.c @@ -155,6 +155,7 @@ static const struct file_operations rtc_fops = { .unlocked_ioctl = rtc_ioctl, .open = rtc_open, .release = rtc_release, + .llseek = noop_llseek, }; static struct miscdevice rtc_dev = { diff --git a/arch/m68k/mvme16x/rtc.c b/arch/m68k/mvme16x/rtc.c index 11ac6f63967..39c79ebcd18 100644 --- a/arch/m68k/mvme16x/rtc.c +++ b/arch/m68k/mvme16x/rtc.c @@ -144,6 +144,7 @@ static const struct file_operations rtc_fops = { .unlocked_ioctl = rtc_ioctl, .open = rtc_open, .release = rtc_release, + .llseek = noop_llseek, }; static struct miscdevice rtc_dev= diff --git a/arch/mips/kernel/rtlx.c b/arch/mips/kernel/rtlx.c index 26f9b9ab19c..557ef72472e 100644 --- a/arch/mips/kernel/rtlx.c +++ b/arch/mips/kernel/rtlx.c @@ -468,7 +468,8 @@ static const struct file_operations rtlx_fops = { .release = file_release, .write = file_write, .read = file_read, - .poll = file_poll + .poll = file_poll, + .llseek = noop_llseek, }; static struct irqaction rtlx_irq = { diff --git a/arch/mips/kernel/vpe.c b/arch/mips/kernel/vpe.c index 2bd2151c586..3eb3cde2f66 100644 --- a/arch/mips/kernel/vpe.c +++ b/arch/mips/kernel/vpe.c @@ -1192,7 +1192,8 @@ static const struct file_operations vpe_fops = { .owner = THIS_MODULE, .open = vpe_open, .release = vpe_release, - .write = vpe_write + .write = vpe_write, + .llseek = noop_llseek, }; /* module wrapper entry points */ diff --git a/arch/mips/sibyte/common/sb_tbprof.c b/arch/mips/sibyte/common/sb_tbprof.c index d4ed7a9156f..ca35b730d18 100644 --- a/arch/mips/sibyte/common/sb_tbprof.c +++ b/arch/mips/sibyte/common/sb_tbprof.c @@ -545,6 +545,7 @@ static const struct file_operations sbprof_tb_fops = { .unlocked_ioctl = sbprof_tb_ioctl, .compat_ioctl = sbprof_tb_ioctl, .mmap = NULL, + .llseek = default_llseek, }; static struct class *tb_class; diff --git a/arch/powerpc/kernel/lparcfg.c b/arch/powerpc/kernel/lparcfg.c index 50362b6ef6e..b1dd962e247 100644 --- a/arch/powerpc/kernel/lparcfg.c +++ b/arch/powerpc/kernel/lparcfg.c @@ -780,6 +780,7 @@ static const struct file_operations lparcfg_fops = { .write = lparcfg_write, .open = lparcfg_open, .release = single_release, + .llseek = seq_lseek, }; static int __init lparcfg_init(void) diff --git a/arch/powerpc/kernel/rtas_flash.c b/arch/powerpc/kernel/rtas_flash.c index 67a84d8f118..2b442e6c21e 100644 --- a/arch/powerpc/kernel/rtas_flash.c +++ b/arch/powerpc/kernel/rtas_flash.c @@ -716,6 +716,7 @@ static const struct file_operations rtas_flash_operations = { .write = rtas_flash_write, .open = rtas_excl_open, .release = rtas_flash_release, + .llseek = default_llseek, }; static const struct file_operations manage_flash_operations = { @@ -724,6 +725,7 @@ static const struct file_operations manage_flash_operations = { .write = manage_flash_write, .open = rtas_excl_open, .release = rtas_excl_release, + .llseek = default_llseek, }; static const struct file_operations validate_flash_operations = { @@ -732,6 +734,7 @@ static const struct file_operations validate_flash_operations = { .write = validate_flash_write, .open = rtas_excl_open, .release = validate_flash_release, + .llseek = default_llseek, }; static int __init rtas_flash_init(void) diff --git a/arch/powerpc/kernel/rtasd.c b/arch/powerpc/kernel/rtasd.c index 638883e23e3..0438f819fe6 100644 --- a/arch/powerpc/kernel/rtasd.c +++ b/arch/powerpc/kernel/rtasd.c @@ -354,6 +354,7 @@ static const struct file_operations proc_rtas_log_operations = { .poll = rtas_log_poll, .open = rtas_log_open, .release = rtas_log_release, + .llseek = noop_llseek, }; static int enable_surveillance(int timeout) diff --git a/arch/powerpc/platforms/iseries/mf.c b/arch/powerpc/platforms/iseries/mf.c index 33e5fc7334f..42d0a886de0 100644 --- a/arch/powerpc/platforms/iseries/mf.c +++ b/arch/powerpc/platforms/iseries/mf.c @@ -1249,6 +1249,7 @@ out: static const struct file_operations proc_vmlinux_operations = { .write = proc_mf_change_vmlinux, + .llseek = default_llseek, }; static int __init mf_proc_init(void) diff --git a/arch/powerpc/platforms/pseries/reconfig.c b/arch/powerpc/platforms/pseries/reconfig.c index 57ddbb43b33..1de2cbb9230 100644 --- a/arch/powerpc/platforms/pseries/reconfig.c +++ b/arch/powerpc/platforms/pseries/reconfig.c @@ -539,7 +539,8 @@ out: } static const struct file_operations ofdt_fops = { - .write = ofdt_write + .write = ofdt_write, + .llseek = noop_llseek, }; /* create /proc/powerpc/ofdt write-only by root */ diff --git a/arch/powerpc/platforms/pseries/scanlog.c b/arch/powerpc/platforms/pseries/scanlog.c index 80e9e7652a4..554457294a2 100644 --- a/arch/powerpc/platforms/pseries/scanlog.c +++ b/arch/powerpc/platforms/pseries/scanlog.c @@ -170,6 +170,7 @@ const struct file_operations scanlog_fops = { .write = scanlog_write, .open = scanlog_open, .release = scanlog_release, + .llseek = noop_llseek, }; static int __init scanlog_init(void) diff --git a/arch/s390/crypto/prng.c b/arch/s390/crypto/prng.c index aa819dac236..975e3ab13cb 100644 --- a/arch/s390/crypto/prng.c +++ b/arch/s390/crypto/prng.c @@ -152,6 +152,7 @@ static const struct file_operations prng_fops = { .open = &prng_open, .release = NULL, .read = &prng_read, + .llseek = noop_llseek, }; static struct miscdevice prng_dev = { diff --git a/arch/s390/hypfs/hypfs_diag.c b/arch/s390/hypfs/hypfs_diag.c index 1211bb1d2f2..020e51c063d 100644 --- a/arch/s390/hypfs/hypfs_diag.c +++ b/arch/s390/hypfs/hypfs_diag.c @@ -618,6 +618,7 @@ static const struct file_operations dbfs_d204_ops = { .open = dbfs_d204_open, .read = dbfs_d204_read, .release = dbfs_d204_release, + .llseek = no_llseek, }; static int hypfs_dbfs_init(void) diff --git a/arch/s390/hypfs/hypfs_vm.c b/arch/s390/hypfs/hypfs_vm.c index ee5ab1a578e..26cf177f6a3 100644 --- a/arch/s390/hypfs/hypfs_vm.c +++ b/arch/s390/hypfs/hypfs_vm.c @@ -275,6 +275,7 @@ static const struct file_operations dbfs_d2fc_ops = { .open = dbfs_d2fc_open, .read = dbfs_d2fc_read, .release = dbfs_d2fc_release, + .llseek = no_llseek, }; int hypfs_vm_init(void) diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c index 98a4a4c267a..74d98670be2 100644 --- a/arch/s390/hypfs/inode.c +++ b/arch/s390/hypfs/inode.c @@ -449,6 +449,7 @@ static const struct file_operations hypfs_file_ops = { .write = do_sync_write, .aio_read = hypfs_aio_read, .aio_write = hypfs_aio_write, + .llseek = no_llseek, }; static struct file_system_type hypfs_type = { diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c index 98192261491..5ad6bc078bf 100644 --- a/arch/s390/kernel/debug.c +++ b/arch/s390/kernel/debug.c @@ -174,6 +174,7 @@ static const struct file_operations debug_file_ops = { .write = debug_input, .open = debug_open, .release = debug_close, + .llseek = no_llseek, }; static struct dentry *debug_debugfs_root_entry; diff --git a/arch/sh/boards/mach-landisk/gio.c b/arch/sh/boards/mach-landisk/gio.c index 01e6abb769b..8132dff078f 100644 --- a/arch/sh/boards/mach-landisk/gio.c +++ b/arch/sh/boards/mach-landisk/gio.c @@ -128,6 +128,7 @@ static const struct file_operations gio_fops = { .open = gio_open, /* open */ .release = gio_close, /* release */ .unlocked_ioctl = gio_ioctl, + .llseek = noop_llseek, }; static int __init gio_init(void) diff --git a/arch/sparc/kernel/apc.c b/arch/sparc/kernel/apc.c index 2c0046ecc71..52de4a9424e 100644 --- a/arch/sparc/kernel/apc.c +++ b/arch/sparc/kernel/apc.c @@ -132,6 +132,7 @@ static const struct file_operations apc_fops = { .unlocked_ioctl = apc_ioctl, .open = apc_open, .release = apc_release, + .llseek = noop_llseek, }; static struct miscdevice apc_miscdev = { APC_MINOR, APC_DEVNAME, &apc_fops }; diff --git a/arch/sparc/kernel/mdesc.c b/arch/sparc/kernel/mdesc.c index 83e85c2e802..6addb914fcc 100644 --- a/arch/sparc/kernel/mdesc.c +++ b/arch/sparc/kernel/mdesc.c @@ -890,6 +890,7 @@ static ssize_t mdesc_read(struct file *file, char __user *buf, static const struct file_operations mdesc_fops = { .read = mdesc_read, .owner = THIS_MODULE, + .llseek = noop_llseek, }; static struct miscdevice mdesc_misc = { diff --git a/arch/tile/kernel/hardwall.c b/arch/tile/kernel/hardwall.c index 584b965dc82..1e54a784341 100644 --- a/arch/tile/kernel/hardwall.c +++ b/arch/tile/kernel/hardwall.c @@ -774,6 +774,7 @@ static const struct file_operations dev_hardwall_fops = { #endif .flush = hardwall_flush, .release = hardwall_release, + .llseek = noop_llseek, }; static struct cdev hardwall_dev; diff --git a/arch/um/drivers/harddog_kern.c b/arch/um/drivers/harddog_kern.c index cfcac1ff4cf..dd1e6f871fe 100644 --- a/arch/um/drivers/harddog_kern.c +++ b/arch/um/drivers/harddog_kern.c @@ -166,6 +166,7 @@ static const struct file_operations harddog_fops = { .unlocked_ioctl = harddog_ioctl, .open = harddog_open, .release = harddog_release, + .llseek = no_llseek, }; static struct miscdevice harddog_miscdev = { diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c index ebc680717e5..975613b23dc 100644 --- a/arch/um/drivers/mconsole_kern.c +++ b/arch/um/drivers/mconsole_kern.c @@ -843,6 +843,7 @@ static ssize_t mconsole_proc_write(struct file *file, static const struct file_operations mconsole_proc_fops = { .owner = THIS_MODULE, .write = mconsole_proc_write, + .llseek = noop_llseek, }; static int create_proc_mconsole(void) diff --git a/arch/um/drivers/mmapper_kern.c b/arch/um/drivers/mmapper_kern.c index 7158393b679..8501e7d0015 100644 --- a/arch/um/drivers/mmapper_kern.c +++ b/arch/um/drivers/mmapper_kern.c @@ -93,6 +93,7 @@ static const struct file_operations mmapper_fops = { .mmap = mmapper_mmap, .open = mmapper_open, .release = mmapper_release, + .llseek = default_llseek, }; /* diff --git a/arch/um/drivers/random.c b/arch/um/drivers/random.c index 4949044773b..981085a93f3 100644 --- a/arch/um/drivers/random.c +++ b/arch/um/drivers/random.c @@ -100,6 +100,7 @@ static const struct file_operations rng_chrdev_ops = { .owner = THIS_MODULE, .open = rng_dev_open, .read = rng_dev_read, + .llseek = noop_llseek, }; /* rng_init shouldn't be called more than once at boot time */ diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c index 4c9c67bf09b..fbbc4dadecc 100644 --- a/arch/x86/kernel/apm_32.c +++ b/arch/x86/kernel/apm_32.c @@ -1926,6 +1926,7 @@ static const struct file_operations apm_bios_fops = { .unlocked_ioctl = do_ioctl, .open = do_open, .release = do_release, + .llseek = noop_llseek, }; static struct miscdevice apm_device = { diff --git a/arch/x86/kernel/cpu/mcheck/mce-severity.c b/arch/x86/kernel/cpu/mcheck/mce-severity.c index 8a85dd1b1aa..1e8d66c1336 100644 --- a/arch/x86/kernel/cpu/mcheck/mce-severity.c +++ b/arch/x86/kernel/cpu/mcheck/mce-severity.c @@ -192,6 +192,7 @@ static const struct file_operations severities_coverage_fops = { .release = seq_release, .read = seq_read, .write = severities_coverage_write, + .llseek = seq_lseek, }; static int __init severities_debugfs_init(void) diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index ed41562909f..7a35b72d7c0 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -1665,6 +1665,7 @@ struct file_operations mce_chrdev_ops = { .read = mce_read, .poll = mce_poll, .unlocked_ioctl = mce_ioctl, + .llseek = no_llseek, }; EXPORT_SYMBOL_GPL(mce_chrdev_ops); diff --git a/arch/x86/kernel/kdebugfs.c b/arch/x86/kernel/kdebugfs.c index 8afd9f321f1..90fcf62854b 100644 --- a/arch/x86/kernel/kdebugfs.c +++ b/arch/x86/kernel/kdebugfs.c @@ -78,6 +78,7 @@ static int setup_data_open(struct inode *inode, struct file *file) static const struct file_operations fops_setup_data = { .read = setup_data_read, .open = setup_data_open, + .llseek = default_llseek, }; static int __init diff --git a/arch/x86/kernel/microcode_core.c b/arch/x86/kernel/microcode_core.c index fa6551d36c1..0b3d37e8360 100644 --- a/arch/x86/kernel/microcode_core.c +++ b/arch/x86/kernel/microcode_core.c @@ -232,6 +232,7 @@ static const struct file_operations microcode_fops = { .owner = THIS_MODULE, .write = microcode_write, .open = microcode_open, + .llseek = no_llseek, }; static struct miscdevice microcode_dev = { diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c index 312ef029281..50ac949c7f1 100644 --- a/arch/x86/kernel/tlb_uv.c +++ b/arch/x86/kernel/tlb_uv.c @@ -1285,6 +1285,7 @@ static const struct file_operations tunables_fops = { .open = tunables_open, .read = tunables_read, .write = tunables_write, + .llseek = default_llseek, }; static int __init uv_ptc_init(void) diff --git a/arch/x86/xen/debugfs.c b/arch/x86/xen/debugfs.c index 1304bcec8ee..7c0fedd98ea 100644 --- a/arch/x86/xen/debugfs.c +++ b/arch/x86/xen/debugfs.c @@ -106,6 +106,7 @@ static const struct file_operations u32_array_fops = { .open = u32_array_open, .release= xen_array_release, .read = u32_array_read, + .llseek = no_llseek, }; struct dentry *xen_debugfs_create_u32_array(const char *name, mode_t mode, diff --git a/block/bsg.c b/block/bsg.c index 82d58829ba5..83e381a26b5 100644 --- a/block/bsg.c +++ b/block/bsg.c @@ -968,6 +968,7 @@ static const struct file_operations bsg_fops = { .release = bsg_release, .unlocked_ioctl = bsg_ioctl, .owner = THIS_MODULE, + .llseek = default_llseek, }; void bsg_unregister_queue(struct request_queue *q) diff --git a/drivers/acpi/apei/erst-dbg.c b/drivers/acpi/apei/erst-dbg.c index 5281ddda277..cbab9b07bf2 100644 --- a/drivers/acpi/apei/erst-dbg.c +++ b/drivers/acpi/apei/erst-dbg.c @@ -180,6 +180,7 @@ static const struct file_operations erst_dbg_ops = { .read = erst_dbg_read, .write = erst_dbg_write, .unlocked_ioctl = erst_dbg_ioctl, + .llseek = no_llseek, }; static struct miscdevice erst_dbg_dev = { diff --git a/drivers/acpi/debugfs.c b/drivers/acpi/debugfs.c index 7de27d49c4b..6355b575ee5 100644 --- a/drivers/acpi/debugfs.c +++ b/drivers/acpi/debugfs.c @@ -69,6 +69,7 @@ static ssize_t cm_write(struct file *file, const char __user * user_buf, static const struct file_operations cm_fops = { .write = cm_write, + .llseek = default_llseek, }; int __init acpi_debugfs_init(void) diff --git a/drivers/acpi/ec_sys.c b/drivers/acpi/ec_sys.c index 0e869b3f81c..411620ef84c 100644 --- a/drivers/acpi/ec_sys.c +++ b/drivers/acpi/ec_sys.c @@ -101,6 +101,7 @@ static struct file_operations acpi_ec_io_ops = { .open = acpi_ec_open_io, .read = acpi_ec_read_io, .write = acpi_ec_write_io, + .llseek = default_llseek, }; int acpi_ec_add_debugfs(struct acpi_ec *ec, unsigned int ec_device_count) diff --git a/drivers/acpi/event.c b/drivers/acpi/event.c index d439314a75d..85d90899380 100644 --- a/drivers/acpi/event.c +++ b/drivers/acpi/event.c @@ -110,6 +110,7 @@ static const struct file_operations acpi_system_event_ops = { .read = acpi_system_read_event, .release = acpi_system_close_event, .poll = acpi_system_poll_event, + .llseek = default_llseek, }; #endif /* CONFIG_ACPI_PROC_EVENT */ diff --git a/drivers/block/DAC960.c b/drivers/block/DAC960.c index 4e2c367fec1..dfcb33e8d40 100644 --- a/drivers/block/DAC960.c +++ b/drivers/block/DAC960.c @@ -7062,7 +7062,8 @@ static long DAC960_gam_ioctl(struct file *file, unsigned int Request, static const struct file_operations DAC960_gam_fops = { .owner = THIS_MODULE, - .unlocked_ioctl = DAC960_gam_ioctl + .unlocked_ioctl = DAC960_gam_ioctl, + .llseek = noop_llseek, }; static struct miscdevice DAC960_gam_dev = { diff --git a/drivers/block/aoe/aoechr.c b/drivers/block/aoe/aoechr.c index 4a1b9e7464a..32b484ba21b 100644 --- a/drivers/block/aoe/aoechr.c +++ b/drivers/block/aoe/aoechr.c @@ -265,6 +265,7 @@ static const struct file_operations aoe_fops = { .open = aoechr_open, .release = aoechr_rel, .owner = THIS_MODULE, + .llseek = noop_llseek, }; static char *aoe_devnode(struct device *dev, mode_t *mode) diff --git a/drivers/block/paride/pg.c b/drivers/block/paride/pg.c index c397b3ddba9..aa27cd84f63 100644 --- a/drivers/block/paride/pg.c +++ b/drivers/block/paride/pg.c @@ -234,6 +234,7 @@ static const struct file_operations pg_fops = { .write = pg_write, .open = pg_open, .release = pg_release, + .llseek = noop_llseek, }; static void pg_init_units(void) diff --git a/drivers/block/paride/pt.c b/drivers/block/paride/pt.c index bc5825fdeaa..c372c32e0db 100644 --- a/drivers/block/paride/pt.c +++ b/drivers/block/paride/pt.c @@ -239,6 +239,7 @@ static const struct file_operations pt_fops = { .unlocked_ioctl = pt_ioctl, .open = pt_open, .release = pt_release, + .llseek = noop_llseek, }; /* sysfs class support */ diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c index b1cbeb59bb7..6a4642dd828 100644 --- a/drivers/block/pktcdvd.c +++ b/drivers/block/pktcdvd.c @@ -3046,6 +3046,7 @@ static const struct file_operations pkt_ctl_fops = { .compat_ioctl = pkt_ctl_compat_ioctl, #endif .owner = THIS_MODULE, + .llseek = no_llseek, }; static struct miscdevice pkt_misc = { diff --git a/drivers/bluetooth/btmrvl_debugfs.c b/drivers/bluetooth/btmrvl_debugfs.c index 54739b08c30..fd6305bf953 100644 --- a/drivers/bluetooth/btmrvl_debugfs.c +++ b/drivers/bluetooth/btmrvl_debugfs.c @@ -92,6 +92,7 @@ static const struct file_operations btmrvl_hscfgcmd_fops = { .read = btmrvl_hscfgcmd_read, .write = btmrvl_hscfgcmd_write, .open = btmrvl_open_generic, + .llseek = default_llseek, }; static ssize_t btmrvl_psmode_write(struct file *file, const char __user *ubuf, @@ -130,6 +131,7 @@ static const struct file_operations btmrvl_psmode_fops = { .read = btmrvl_psmode_read, .write = btmrvl_psmode_write, .open = btmrvl_open_generic, + .llseek = default_llseek, }; static ssize_t btmrvl_pscmd_write(struct file *file, const char __user *ubuf, @@ -173,6 +175,7 @@ static const struct file_operations btmrvl_pscmd_fops = { .read = btmrvl_pscmd_read, .write = btmrvl_pscmd_write, .open = btmrvl_open_generic, + .llseek = default_llseek, }; static ssize_t btmrvl_gpiogap_write(struct file *file, const char __user *ubuf, @@ -211,6 +214,7 @@ static const struct file_operations btmrvl_gpiogap_fops = { .read = btmrvl_gpiogap_read, .write = btmrvl_gpiogap_write, .open = btmrvl_open_generic, + .llseek = default_llseek, }; static ssize_t btmrvl_hscmd_write(struct file *file, const char __user *ubuf, @@ -252,6 +256,7 @@ static const struct file_operations btmrvl_hscmd_fops = { .read = btmrvl_hscmd_read, .write = btmrvl_hscmd_write, .open = btmrvl_open_generic, + .llseek = default_llseek, }; static ssize_t btmrvl_hsmode_write(struct file *file, const char __user *ubuf, @@ -289,6 +294,7 @@ static const struct file_operations btmrvl_hsmode_fops = { .read = btmrvl_hsmode_read, .write = btmrvl_hsmode_write, .open = btmrvl_open_generic, + .llseek = default_llseek, }; static ssize_t btmrvl_curpsmode_read(struct file *file, char __user *userbuf, @@ -306,6 +312,7 @@ static ssize_t btmrvl_curpsmode_read(struct file *file, char __user *userbuf, static const struct file_operations btmrvl_curpsmode_fops = { .read = btmrvl_curpsmode_read, .open = btmrvl_open_generic, + .llseek = default_llseek, }; static ssize_t btmrvl_psstate_read(struct file *file, char __user * userbuf, @@ -323,6 +330,7 @@ static ssize_t btmrvl_psstate_read(struct file *file, char __user * userbuf, static const struct file_operations btmrvl_psstate_fops = { .read = btmrvl_psstate_read, .open = btmrvl_open_generic, + .llseek = default_llseek, }; static ssize_t btmrvl_hsstate_read(struct file *file, char __user *userbuf, @@ -340,6 +348,7 @@ static ssize_t btmrvl_hsstate_read(struct file *file, char __user *userbuf, static const struct file_operations btmrvl_hsstate_fops = { .read = btmrvl_hsstate_read, .open = btmrvl_open_generic, + .llseek = default_llseek, }; static ssize_t btmrvl_txdnldready_read(struct file *file, char __user *userbuf, @@ -358,6 +367,7 @@ static ssize_t btmrvl_txdnldready_read(struct file *file, char __user *userbuf, static const struct file_operations btmrvl_txdnldready_fops = { .read = btmrvl_txdnldready_read, .open = btmrvl_open_generic, + .llseek = default_llseek, }; void btmrvl_debugfs_init(struct hci_dev *hdev) diff --git a/drivers/bluetooth/hci_vhci.c b/drivers/bluetooth/hci_vhci.c index 3aa7b2a54b6..67c180c2c1e 100644 --- a/drivers/bluetooth/hci_vhci.c +++ b/drivers/bluetooth/hci_vhci.c @@ -282,6 +282,7 @@ static const struct file_operations vhci_fops = { .poll = vhci_poll, .open = vhci_open, .release = vhci_release, + .llseek = no_llseek, }; static struct miscdevice vhci_miscdev= { diff --git a/drivers/char/apm-emulation.c b/drivers/char/apm-emulation.c index 033e1505fca..5ffa6904ea6 100644 --- a/drivers/char/apm-emulation.c +++ b/drivers/char/apm-emulation.c @@ -402,6 +402,7 @@ static const struct file_operations apm_bios_fops = { .unlocked_ioctl = apm_ioctl, .open = apm_open, .release = apm_release, + .llseek = noop_llseek, }; static struct miscdevice apm_device = { diff --git a/drivers/char/bfin-otp.c b/drivers/char/bfin-otp.c index 836d4f0a876..44660f1c484 100644 --- a/drivers/char/bfin-otp.c +++ b/drivers/char/bfin-otp.c @@ -222,6 +222,7 @@ static const struct file_operations bfin_otp_fops = { .unlocked_ioctl = bfin_otp_ioctl, .read = bfin_otp_read, .write = bfin_otp_write, + .llseek = default_llseek, }; static struct miscdevice bfin_otp_misc_device = { diff --git a/drivers/char/briq_panel.c b/drivers/char/briq_panel.c index d5fa113afe3..f6718f05dad 100644 --- a/drivers/char/briq_panel.c +++ b/drivers/char/briq_panel.c @@ -186,6 +186,7 @@ static const struct file_operations briq_panel_fops = { .write = briq_panel_write, .open = briq_panel_open, .release = briq_panel_release, + .llseek = noop_llseek, }; static struct miscdevice briq_panel_miscdev = { diff --git a/drivers/char/bsr.c b/drivers/char/bsr.c index 91917133ae0..a4a6c2f044b 100644 --- a/drivers/char/bsr.c +++ b/drivers/char/bsr.c @@ -155,6 +155,7 @@ static const struct file_operations bsr_fops = { .owner = THIS_MODULE, .mmap = bsr_mmap, .open = bsr_open, + .llseek = noop_llseek, }; static void bsr_cleanup_devs(void) diff --git a/drivers/char/cs5535_gpio.c b/drivers/char/cs5535_gpio.c index 4d830dc482e..0cf1e5fad9a 100644 --- a/drivers/char/cs5535_gpio.c +++ b/drivers/char/cs5535_gpio.c @@ -169,7 +169,8 @@ static const struct file_operations cs5535_gpio_fops = { .owner = THIS_MODULE, .write = cs5535_gpio_write, .read = cs5535_gpio_read, - .open = cs5535_gpio_open + .open = cs5535_gpio_open, + .llseek = no_llseek, }; static int __init cs5535_gpio_init(void) diff --git a/drivers/char/ds1302.c b/drivers/char/ds1302.c index 170693c93c7..4f7aa364167 100644 --- a/drivers/char/ds1302.c +++ b/drivers/char/ds1302.c @@ -288,6 +288,7 @@ get_rtc_status(char *buf) static const struct file_operations rtc_fops = { .owner = THIS_MODULE, .unlocked_ioctl = rtc_ioctl, + .llseek = noop_llseek, }; /* Probe for the chip by writing something to its RAM and try reading it back. */ diff --git a/drivers/char/ds1620.c b/drivers/char/ds1620.c index dbee8688f75..50462b63e51 100644 --- a/drivers/char/ds1620.c +++ b/drivers/char/ds1620.c @@ -357,6 +357,7 @@ static const struct file_operations ds1620_fops = { .open = ds1620_open, .read = ds1620_read, .unlocked_ioctl = ds1620_unlocked_ioctl, + .llseek = no_llseek, }; static struct miscdevice ds1620_miscdev = { diff --git a/drivers/char/dsp56k.c b/drivers/char/dsp56k.c index 8a1b28a10ef..353be4707d9 100644 --- a/drivers/char/dsp56k.c +++ b/drivers/char/dsp56k.c @@ -482,6 +482,7 @@ static const struct file_operations dsp56k_fops = { .unlocked_ioctl = dsp56k_ioctl, .open = dsp56k_open, .release = dsp56k_release, + .llseek = noop_llseek, }; diff --git a/drivers/char/dtlk.c b/drivers/char/dtlk.c index e3859d4eaea..007eca3ed77 100644 --- a/drivers/char/dtlk.c +++ b/drivers/char/dtlk.c @@ -105,6 +105,7 @@ static const struct file_operations dtlk_fops = .unlocked_ioctl = dtlk_ioctl, .open = dtlk_open, .release = dtlk_release, + .llseek = no_llseek, }; /* local prototypes */ diff --git a/drivers/char/genrtc.c b/drivers/char/genrtc.c index b6c2cc167c1..eaa0e4264e1 100644 --- a/drivers/char/genrtc.c +++ b/drivers/char/genrtc.c @@ -497,6 +497,7 @@ static const struct file_operations gen_rtc_fops = { .unlocked_ioctl = gen_rtc_unlocked_ioctl, .open = gen_rtc_open, .release = gen_rtc_release, + .llseek = noop_llseek, }; static struct miscdevice rtc_gen_dev = diff --git a/drivers/char/hw_random/core.c b/drivers/char/hw_random/core.c index 3d9c61e5acb..788da05190c 100644 --- a/drivers/char/hw_random/core.c +++ b/drivers/char/hw_random/core.c @@ -170,6 +170,7 @@ static const struct file_operations rng_chrdev_ops = { .owner = THIS_MODULE, .open = rng_dev_open, .read = rng_dev_read, + .llseek = noop_llseek, }; static struct miscdevice rng_miscdev = { diff --git a/drivers/char/ip2/ip2main.c b/drivers/char/ip2/ip2main.c index d4b71e8d0d2..2e49e515d8e 100644 --- a/drivers/char/ip2/ip2main.c +++ b/drivers/char/ip2/ip2main.c @@ -236,6 +236,7 @@ static const struct file_operations ip2_ipl = { .write = ip2_ipl_write, .unlocked_ioctl = ip2_ipl_ioctl, .open = ip2_ipl_open, + .llseek = noop_llseek, }; static unsigned long irq_counter; diff --git a/drivers/char/ipmi/ipmi_devintf.c b/drivers/char/ipmi/ipmi_devintf.c index d8ec92a3898..c6709d6b0fd 100644 --- a/drivers/char/ipmi/ipmi_devintf.c +++ b/drivers/char/ipmi/ipmi_devintf.c @@ -850,6 +850,7 @@ static const struct file_operations ipmi_fops = { .release = ipmi_release, .fasync = ipmi_fasync, .poll = ipmi_poll, + .llseek = noop_llseek, }; #define DEVICE_NAME "ipmidev" diff --git a/drivers/char/ipmi/ipmi_watchdog.c b/drivers/char/ipmi/ipmi_watchdog.c index 654d566ca57..da041f88f64 100644 --- a/drivers/char/ipmi/ipmi_watchdog.c +++ b/drivers/char/ipmi/ipmi_watchdog.c @@ -909,6 +909,7 @@ static const struct file_operations ipmi_wdog_fops = { .open = ipmi_open, .release = ipmi_close, .fasync = ipmi_fasync, + .llseek = no_llseek, }; static struct miscdevice ipmi_wdog_miscdev = { diff --git a/drivers/char/istallion.c b/drivers/char/istallion.c index be28391adb7..667abd23ad6 100644 --- a/drivers/char/istallion.c +++ b/drivers/char/istallion.c @@ -704,6 +704,7 @@ static const struct file_operations stli_fsiomem = { .read = stli_memread, .write = stli_memwrite, .unlocked_ioctl = stli_memioctl, + .llseek = default_llseek, }; /*****************************************************************************/ diff --git a/drivers/char/lp.c b/drivers/char/lp.c index 938a3a27388..d2344117eaf 100644 --- a/drivers/char/lp.c +++ b/drivers/char/lp.c @@ -748,6 +748,7 @@ static const struct file_operations lp_fops = { #ifdef CONFIG_PARPORT_1284 .read = lp_read, #endif + .llseek = noop_llseek, }; /* --- support for console on the line printer ----------------- */ diff --git a/drivers/char/mem.c b/drivers/char/mem.c index a398ecdbd75..8ebd232132f 100644 --- a/drivers/char/mem.c +++ b/drivers/char/mem.c @@ -804,6 +804,7 @@ static const struct file_operations full_fops = { static const struct file_operations oldmem_fops = { .read = read_oldmem, .open = open_oldmem, + .llseek = default_llseek, }; #endif @@ -830,6 +831,7 @@ static ssize_t kmsg_write(struct file *file, const char __user *buf, static const struct file_operations kmsg_fops = { .write = kmsg_write, + .llseek = noop_llseek, }; static const struct memdev { @@ -881,6 +883,7 @@ static int memory_open(struct inode *inode, struct file *filp) static const struct file_operations memory_fops = { .open = memory_open, + .llseek = noop_llseek, }; static char *mem_devnode(struct device *dev, mode_t *mode) diff --git a/drivers/char/misc.c b/drivers/char/misc.c index abdafd48898..778273c9324 100644 --- a/drivers/char/misc.c +++ b/drivers/char/misc.c @@ -162,6 +162,7 @@ static struct class *misc_class; static const struct file_operations misc_fops = { .owner = THIS_MODULE, .open = misc_open, + .llseek = noop_llseek, }; /** diff --git a/drivers/char/mmtimer.c b/drivers/char/mmtimer.c index ea7c99fa978..1c4070ced06 100644 --- a/drivers/char/mmtimer.c +++ b/drivers/char/mmtimer.c @@ -72,6 +72,7 @@ static const struct file_operations mmtimer_fops = { .owner = THIS_MODULE, .mmap = mmtimer_mmap, .unlocked_ioctl = mmtimer_ioctl, + .llseek = noop_llseek, }; /* diff --git a/drivers/char/mspec.c b/drivers/char/mspec.c index ecb89d798e3..966a95bc974 100644 --- a/drivers/char/mspec.c +++ b/drivers/char/mspec.c @@ -316,7 +316,8 @@ uncached_mmap(struct file *file, struct vm_area_struct *vma) static const struct file_operations fetchop_fops = { .owner = THIS_MODULE, - .mmap = fetchop_mmap + .mmap = fetchop_mmap, + .llseek = noop_llseek, }; static struct miscdevice fetchop_miscdev = { @@ -327,7 +328,8 @@ static struct miscdevice fetchop_miscdev = { static const struct file_operations cached_fops = { .owner = THIS_MODULE, - .mmap = cached_mmap + .mmap = cached_mmap, + .llseek = noop_llseek, }; static struct miscdevice cached_miscdev = { @@ -338,7 +340,8 @@ static struct miscdevice cached_miscdev = { static const struct file_operations uncached_fops = { .owner = THIS_MODULE, - .mmap = uncached_mmap + .mmap = uncached_mmap, + .llseek = noop_llseek, }; static struct miscdevice uncached_miscdev = { diff --git a/drivers/char/mwave/mwavedd.c b/drivers/char/mwave/mwavedd.c index a4ec50c9507..0464822eac5 100644 --- a/drivers/char/mwave/mwavedd.c +++ b/drivers/char/mwave/mwavedd.c @@ -479,7 +479,8 @@ static const struct file_operations mwave_fops = { .write = mwave_write, .unlocked_ioctl = mwave_ioctl, .open = mwave_open, - .release = mwave_close + .release = mwave_close, + .llseek = default_llseek, }; diff --git a/drivers/char/nwbutton.c b/drivers/char/nwbutton.c index 2604246501e..8994ce32e6c 100644 --- a/drivers/char/nwbutton.c +++ b/drivers/char/nwbutton.c @@ -182,6 +182,7 @@ static int button_read (struct file *filp, char __user *buffer, static const struct file_operations button_fops = { .owner = THIS_MODULE, .read = button_read, + .llseek = noop_llseek, }; /* diff --git a/drivers/char/pc8736x_gpio.c b/drivers/char/pc8736x_gpio.c index 8ecbcc174c1..b304ec05250 100644 --- a/drivers/char/pc8736x_gpio.c +++ b/drivers/char/pc8736x_gpio.c @@ -234,6 +234,7 @@ static const struct file_operations pc8736x_gpio_fileops = { .open = pc8736x_gpio_open, .write = nsc_gpio_write, .read = nsc_gpio_read, + .llseek = no_llseek, }; static void __init pc8736x_init_shadow(void) diff --git a/drivers/char/pcmcia/cm4000_cs.c b/drivers/char/pcmcia/cm4000_cs.c index ec73d9f6d9e..c99f6997e5e 100644 --- a/drivers/char/pcmcia/cm4000_cs.c +++ b/drivers/char/pcmcia/cm4000_cs.c @@ -1880,6 +1880,7 @@ static const struct file_operations cm4000_fops = { .unlocked_ioctl = cmm_ioctl, .open = cmm_open, .release= cmm_close, + .llseek = no_llseek, }; static struct pcmcia_device_id cm4000_ids[] = { diff --git a/drivers/char/pcmcia/cm4040_cs.c b/drivers/char/pcmcia/cm4040_cs.c index 815cde1d057..9ecc58baa8f 100644 --- a/drivers/char/pcmcia/cm4040_cs.c +++ b/drivers/char/pcmcia/cm4040_cs.c @@ -650,6 +650,7 @@ static const struct file_operations reader_fops = { .open = cm4040_open, .release = cm4040_close, .poll = cm4040_poll, + .llseek = no_llseek, }; static struct pcmcia_device_id cm4040_ids[] = { diff --git a/drivers/char/random.c b/drivers/char/random.c index caef35a4689..5a1aa64f4e7 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -1165,6 +1165,7 @@ const struct file_operations random_fops = { .poll = random_poll, .unlocked_ioctl = random_ioctl, .fasync = random_fasync, + .llseek = noop_llseek, }; const struct file_operations urandom_fops = { @@ -1172,6 +1173,7 @@ const struct file_operations urandom_fops = { .write = random_write, .unlocked_ioctl = random_ioctl, .fasync = random_fasync, + .llseek = noop_llseek, }; /*************************************************************** diff --git a/drivers/char/rio/rio_linux.c b/drivers/char/rio/rio_linux.c index d58c2eb07f0..1e87a93164b 100644 --- a/drivers/char/rio/rio_linux.c +++ b/drivers/char/rio/rio_linux.c @@ -241,6 +241,7 @@ static struct real_driver rio_real_driver = { static const struct file_operations rio_fw_fops = { .owner = THIS_MODULE, .unlocked_ioctl = rio_fw_ioctl, + .llseek = noop_llseek, }; static struct miscdevice rio_fw_device = { diff --git a/drivers/char/scx200_gpio.c b/drivers/char/scx200_gpio.c index 99e5272e3c5..0bc135b9b16 100644 --- a/drivers/char/scx200_gpio.c +++ b/drivers/char/scx200_gpio.c @@ -67,6 +67,7 @@ static const struct file_operations scx200_gpio_fileops = { .read = nsc_gpio_read, .open = scx200_gpio_open, .release = scx200_gpio_release, + .llseek = no_llseek, }; static struct cdev scx200_gpio_cdev; /* use 1 cdev for all pins */ diff --git a/drivers/char/snsc.c b/drivers/char/snsc.c index 32b74de18f5..444ce17ac72 100644 --- a/drivers/char/snsc.c +++ b/drivers/char/snsc.c @@ -357,6 +357,7 @@ static const struct file_operations scdrv_fops = { .poll = scdrv_poll, .open = scdrv_open, .release = scdrv_release, + .llseek = noop_llseek, }; static struct class *snsc_class; diff --git a/drivers/char/stallion.c b/drivers/char/stallion.c index f2167f8e5aa..8ef16490810 100644 --- a/drivers/char/stallion.c +++ b/drivers/char/stallion.c @@ -608,6 +608,7 @@ static unsigned int sc26198_baudtable[] = { static const struct file_operations stl_fsiomem = { .owner = THIS_MODULE, .unlocked_ioctl = stl_memioctl, + .llseek = noop_llseek, }; static struct class *stallion_class; diff --git a/drivers/char/sx.c b/drivers/char/sx.c index 5b24db4ff7f..e53f1686539 100644 --- a/drivers/char/sx.c +++ b/drivers/char/sx.c @@ -397,6 +397,7 @@ static struct real_driver sx_real_driver = { static const struct file_operations sx_fw_fops = { .owner = THIS_MODULE, .unlocked_ioctl = sx_fw_ioctl, + .llseek = noop_llseek, }; static struct miscdevice sx_fw_device = { diff --git a/drivers/char/sysrq.c b/drivers/char/sysrq.c index ef31bb81e84..f3019f53e87 100644 --- a/drivers/char/sysrq.c +++ b/drivers/char/sysrq.c @@ -772,6 +772,7 @@ static ssize_t write_sysrq_trigger(struct file *file, const char __user *buf, static const struct file_operations proc_sysrq_trigger_operations = { .write = write_sysrq_trigger, + .llseek = noop_llseek, }; static void sysrq_init_procfs(void) diff --git a/drivers/char/tb0219.c b/drivers/char/tb0219.c index cad4eb65f13..ad264185eb1 100644 --- a/drivers/char/tb0219.c +++ b/drivers/char/tb0219.c @@ -261,6 +261,7 @@ static const struct file_operations tb0219_fops = { .write = tanbac_tb0219_write, .open = tanbac_tb0219_open, .release = tanbac_tb0219_release, + .llseek = no_llseek, }; static void tb0219_restart(char *command) diff --git a/drivers/char/tlclk.c b/drivers/char/tlclk.c index 80ea6bcfffd..d087532b29d 100644 --- a/drivers/char/tlclk.c +++ b/drivers/char/tlclk.c @@ -267,6 +267,7 @@ static const struct file_operations tlclk_fops = { .read = tlclk_read, .open = tlclk_open, .release = tlclk_release, + .llseek = noop_llseek, }; diff --git a/drivers/char/toshiba.c b/drivers/char/toshiba.c index f8bc79f6de3..3d6e9617245 100644 --- a/drivers/char/toshiba.c +++ b/drivers/char/toshiba.c @@ -95,6 +95,7 @@ static long tosh_ioctl(struct file *, unsigned int, static const struct file_operations tosh_fops = { .owner = THIS_MODULE, .unlocked_ioctl = tosh_ioctl, + .llseek = noop_llseek, }; static struct miscdevice tosh_device = { diff --git a/drivers/char/uv_mmtimer.c b/drivers/char/uv_mmtimer.c index c7072ba14f4..493b47a0d51 100644 --- a/drivers/char/uv_mmtimer.c +++ b/drivers/char/uv_mmtimer.c @@ -52,6 +52,7 @@ static const struct file_operations uv_mmtimer_fops = { .owner = THIS_MODULE, .mmap = uv_mmtimer_mmap, .unlocked_ioctl = uv_mmtimer_ioctl, + .llseek = noop_llseek, }; /** diff --git a/drivers/char/xilinx_hwicap/xilinx_hwicap.c b/drivers/char/xilinx_hwicap/xilinx_hwicap.c index b663d573aad..be6d6fb47cc 100644 --- a/drivers/char/xilinx_hwicap/xilinx_hwicap.c +++ b/drivers/char/xilinx_hwicap/xilinx_hwicap.c @@ -567,6 +567,7 @@ static const struct file_operations hwicap_fops = { .read = hwicap_read, .open = hwicap_open, .release = hwicap_release, + .llseek = noop_llseek, }; static int __devinit hwicap_setup(struct device *dev, int id, diff --git a/drivers/dma/coh901318.c b/drivers/dma/coh901318.c index 557e2272e5b..ae2b8714d19 100644 --- a/drivers/dma/coh901318.c +++ b/drivers/dma/coh901318.c @@ -157,6 +157,7 @@ static const struct file_operations coh901318_debugfs_status_operations = { .owner = THIS_MODULE, .open = coh901318_debugfs_open, .read = coh901318_debugfs_read, + .llseek = default_llseek, }; diff --git a/drivers/firewire/nosy.c b/drivers/firewire/nosy.c index 8528b10763e..bf184fb59a5 100644 --- a/drivers/firewire/nosy.c +++ b/drivers/firewire/nosy.c @@ -405,6 +405,7 @@ static const struct file_operations nosy_ops = { .poll = nosy_poll, .open = nosy_open, .release = nosy_release, + .llseek = noop_llseek, }; #define PHY_PACKET_SIZE 12 /* 1 payload, 1 inverse, 1 ack = 3 quadlets */ diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c index 84da748555b..ff6690f4fc8 100644 --- a/drivers/gpu/drm/drm_drv.c +++ b/drivers/gpu/drm/drm_drv.c @@ -284,7 +284,8 @@ EXPORT_SYMBOL(drm_exit); /** File operations structure */ static const struct file_operations drm_stub_fops = { .owner = THIS_MODULE, - .open = drm_stub_open + .open = drm_stub_open, + .llseek = noop_llseek, }; static int __init drm_core_init(void) diff --git a/drivers/gpu/drm/i810/i810_dma.c b/drivers/gpu/drm/i810/i810_dma.c index 61b4caf220f..00f1bdaa65c 100644 --- a/drivers/gpu/drm/i810/i810_dma.c +++ b/drivers/gpu/drm/i810/i810_dma.c @@ -119,6 +119,7 @@ static const struct file_operations i810_buffer_fops = { .unlocked_ioctl = drm_ioctl, .mmap = i810_mmap_buffers, .fasync = drm_fasync, + .llseek = noop_llseek, }; static int i810_map_buffer(struct drm_buf *buf, struct drm_file *file_priv) diff --git a/drivers/gpu/drm/i830/i830_dma.c b/drivers/gpu/drm/i830/i830_dma.c index 671aa18415a..5c6eb65f4e5 100644 --- a/drivers/gpu/drm/i830/i830_dma.c +++ b/drivers/gpu/drm/i830/i830_dma.c @@ -121,6 +121,7 @@ static const struct file_operations i830_buffer_fops = { .unlocked_ioctl = drm_ioctl, .mmap = i830_mmap_buffers, .fasync = drm_fasync, + .llseek = noop_llseek, }; static int i830_map_buffer(struct drm_buf *buf, struct drm_file *file_priv) diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index 5e43d707678..048149748fd 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -782,6 +782,7 @@ static const struct file_operations i915_wedged_fops = { .open = i915_wedged_open, .read = i915_wedged_read, .write = i915_wedged_write, + .llseek = default_llseek, }; /* As the drm_debugfs_init() routines are called before dev->dev_private is diff --git a/drivers/gpu/vga/vgaarb.c b/drivers/gpu/vga/vgaarb.c index b87569e96b1..3b7e0acf816 100644 --- a/drivers/gpu/vga/vgaarb.c +++ b/drivers/gpu/vga/vgaarb.c @@ -1211,6 +1211,7 @@ static const struct file_operations vga_arb_device_fops = { .poll = vga_arb_fpoll, .open = vga_arb_open, .release = vga_arb_release, + .llseek = noop_llseek, }; static struct miscdevice vga_arb_device = { diff --git a/drivers/hid/hid-debug.c b/drivers/hid/hid-debug.c index 850d02a7a92..61a3e572224 100644 --- a/drivers/hid/hid-debug.c +++ b/drivers/hid/hid-debug.c @@ -1051,6 +1051,7 @@ static const struct file_operations hid_debug_events_fops = { .read = hid_debug_events_read, .poll = hid_debug_events_poll, .release = hid_debug_events_release, + .llseek = noop_llseek, }; diff --git a/drivers/hid/hid-roccat.c b/drivers/hid/hid-roccat.c index f6e80c7ca61..5a6879e235a 100644 --- a/drivers/hid/hid-roccat.c +++ b/drivers/hid/hid-roccat.c @@ -384,6 +384,7 @@ static const struct file_operations roccat_ops = { .poll = roccat_poll, .open = roccat_open, .release = roccat_release, + .llseek = noop_llseek, }; static int __init roccat_init(void) diff --git a/drivers/hid/hidraw.c b/drivers/hid/hidraw.c index 47d70c523d9..bb98fa87aa8 100644 --- a/drivers/hid/hidraw.c +++ b/drivers/hid/hidraw.c @@ -329,6 +329,7 @@ static const struct file_operations hidraw_ops = { .open = hidraw_open, .release = hidraw_release, .unlocked_ioctl = hidraw_ioctl, + .llseek = noop_llseek, }; void hidraw_report_event(struct hid_device *hid, u8 *data, int len) diff --git a/drivers/hid/usbhid/hiddev.c b/drivers/hid/usbhid/hiddev.c index 0a29c51114a..e8f45d7ef3b 100644 --- a/drivers/hid/usbhid/hiddev.c +++ b/drivers/hid/usbhid/hiddev.c @@ -847,6 +847,7 @@ static const struct file_operations hiddev_fops = { #ifdef CONFIG_COMPAT .compat_ioctl = hiddev_compat_ioctl, #endif + .llseek = noop_llseek, }; static char *hiddev_devnode(struct device *dev, mode_t *mode) diff --git a/drivers/hwmon/asus_atk0110.c b/drivers/hwmon/asus_atk0110.c index 653db1bda93..23b8555215d 100644 --- a/drivers/hwmon/asus_atk0110.c +++ b/drivers/hwmon/asus_atk0110.c @@ -762,6 +762,7 @@ static const struct file_operations atk_debugfs_ggrp_fops = { .read = atk_debugfs_ggrp_read, .open = atk_debugfs_ggrp_open, .release = atk_debugfs_ggrp_release, + .llseek = no_llseek, }; static void atk_debugfs_init(struct atk_data *data) diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c index 6d622cb5ac8..23d1d1c5587 100644 --- a/drivers/ide/ide-tape.c +++ b/drivers/ide/ide-tape.c @@ -1903,6 +1903,7 @@ static const struct file_operations idetape_fops = { .unlocked_ioctl = idetape_chrdev_ioctl, .open = idetape_chrdev_open, .release = idetape_chrdev_release, + .llseek = noop_llseek, }; static int idetape_open(struct block_device *bdev, fmode_t mode) diff --git a/drivers/idle/i7300_idle.c b/drivers/idle/i7300_idle.c index 15341fc1c68..c976285d313 100644 --- a/drivers/idle/i7300_idle.c +++ b/drivers/idle/i7300_idle.c @@ -536,6 +536,7 @@ static ssize_t stats_read_ul(struct file *fp, char __user *ubuf, size_t count, static const struct file_operations idle_fops = { .open = stats_open_generic, .read = stats_read_ul, + .llseek = default_llseek, }; struct debugfs_file_info { diff --git a/drivers/infiniband/hw/ipath/ipath_diag.c b/drivers/infiniband/hw/ipath/ipath_diag.c index d4ce8b63e19..daef61d5e5b 100644 --- a/drivers/infiniband/hw/ipath/ipath_diag.c +++ b/drivers/infiniband/hw/ipath/ipath_diag.c @@ -65,7 +65,8 @@ static const struct file_operations diag_file_ops = { .write = ipath_diag_write, .read = ipath_diag_read, .open = ipath_diag_open, - .release = ipath_diag_release + .release = ipath_diag_release, + .llseek = default_llseek, }; static ssize_t ipath_diagpkt_write(struct file *fp, @@ -75,6 +76,7 @@ static ssize_t ipath_diagpkt_write(struct file *fp, static const struct file_operations diagpkt_file_ops = { .owner = THIS_MODULE, .write = ipath_diagpkt_write, + .llseek = noop_llseek, }; static atomic_t diagpkt_count = ATOMIC_INIT(0); diff --git a/drivers/infiniband/hw/ipath/ipath_file_ops.c b/drivers/infiniband/hw/ipath/ipath_file_ops.c index 65eb8929db2..6078992da3f 100644 --- a/drivers/infiniband/hw/ipath/ipath_file_ops.c +++ b/drivers/infiniband/hw/ipath/ipath_file_ops.c @@ -63,7 +63,8 @@ static const struct file_operations ipath_file_ops = { .open = ipath_open, .release = ipath_close, .poll = ipath_poll, - .mmap = ipath_mmap + .mmap = ipath_mmap, + .llseek = noop_llseek, }; /* diff --git a/drivers/infiniband/hw/ipath/ipath_fs.c b/drivers/infiniband/hw/ipath/ipath_fs.c index 2fca70836da..d13e72685dc 100644 --- a/drivers/infiniband/hw/ipath/ipath_fs.c +++ b/drivers/infiniband/hw/ipath/ipath_fs.c @@ -103,6 +103,7 @@ static ssize_t atomic_stats_read(struct file *file, char __user *buf, static const struct file_operations atomic_stats_ops = { .read = atomic_stats_read, + .llseek = default_llseek, }; static ssize_t atomic_counters_read(struct file *file, char __user *buf, @@ -120,6 +121,7 @@ static ssize_t atomic_counters_read(struct file *file, char __user *buf, static const struct file_operations atomic_counters_ops = { .read = atomic_counters_read, + .llseek = default_llseek, }; static ssize_t flash_read(struct file *file, char __user *buf, @@ -224,6 +226,7 @@ bail: static const struct file_operations flash_ops = { .read = flash_read, .write = flash_write, + .llseek = default_llseek, }; static int create_device_files(struct super_block *sb, diff --git a/drivers/infiniband/hw/qib/qib_diag.c b/drivers/infiniband/hw/qib/qib_diag.c index 05dcf0d9a7d..204c4dd9dce 100644 --- a/drivers/infiniband/hw/qib/qib_diag.c +++ b/drivers/infiniband/hw/qib/qib_diag.c @@ -136,7 +136,8 @@ static const struct file_operations diag_file_ops = { .write = qib_diag_write, .read = qib_diag_read, .open = qib_diag_open, - .release = qib_diag_release + .release = qib_diag_release, + .llseek = default_llseek, }; static atomic_t diagpkt_count = ATOMIC_INIT(0); @@ -149,6 +150,7 @@ static ssize_t qib_diagpkt_write(struct file *fp, const char __user *data, static const struct file_operations diagpkt_file_ops = { .owner = THIS_MODULE, .write = qib_diagpkt_write, + .llseek = noop_llseek, }; int qib_diag_add(struct qib_devdata *dd) diff --git a/drivers/infiniband/hw/qib/qib_file_ops.c b/drivers/infiniband/hw/qib/qib_file_ops.c index 6b11645edf3..aa2be214270 100644 --- a/drivers/infiniband/hw/qib/qib_file_ops.c +++ b/drivers/infiniband/hw/qib/qib_file_ops.c @@ -63,7 +63,8 @@ static const struct file_operations qib_file_ops = { .open = qib_open, .release = qib_close, .poll = qib_poll, - .mmap = qib_mmapf + .mmap = qib_mmapf, + .llseek = noop_llseek, }; /* diff --git a/drivers/infiniband/hw/qib/qib_fs.c b/drivers/infiniband/hw/qib/qib_fs.c index 9f989c0ba9d..a0e6613e8be 100644 --- a/drivers/infiniband/hw/qib/qib_fs.c +++ b/drivers/infiniband/hw/qib/qib_fs.c @@ -367,6 +367,7 @@ bail: static const struct file_operations flash_ops = { .read = flash_read, .write = flash_write, + .llseek = default_llseek, }; static int add_cntr_files(struct super_block *sb, struct qib_devdata *dd) diff --git a/drivers/input/evdev.c b/drivers/input/evdev.c index c908c5f8364..51330363c0e 100644 --- a/drivers/input/evdev.c +++ b/drivers/input/evdev.c @@ -761,7 +761,8 @@ static const struct file_operations evdev_fops = { .compat_ioctl = evdev_ioctl_compat, #endif .fasync = evdev_fasync, - .flush = evdev_flush + .flush = evdev_flush, + .llseek = no_llseek, }; static int evdev_install_chrdev(struct evdev *evdev) diff --git a/drivers/input/input.c b/drivers/input/input.c index ab698205651..7919c253722 100644 --- a/drivers/input/input.c +++ b/drivers/input/input.c @@ -2047,6 +2047,7 @@ out: static const struct file_operations input_fops = { .owner = THIS_MODULE, .open = input_open_file, + .llseek = noop_llseek, }; static int __init input_init(void) diff --git a/drivers/input/joydev.c b/drivers/input/joydev.c index d85bd8a7967..502b2f73b43 100644 --- a/drivers/input/joydev.c +++ b/drivers/input/joydev.c @@ -736,6 +736,7 @@ static const struct file_operations joydev_fops = { .compat_ioctl = joydev_compat_ioctl, #endif .fasync = joydev_fasync, + .llseek = no_llseek, }; static int joydev_install_chrdev(struct joydev *joydev) diff --git a/drivers/input/misc/uinput.c b/drivers/input/misc/uinput.c index 0d4266a533a..2771ea778d3 100644 --- a/drivers/input/misc/uinput.c +++ b/drivers/input/misc/uinput.c @@ -804,6 +804,7 @@ static const struct file_operations uinput_fops = { #ifdef CONFIG_COMPAT .compat_ioctl = uinput_compat_ioctl, #endif + .llseek = no_llseek, }; static struct miscdevice uinput_misc = { diff --git a/drivers/input/mousedev.c b/drivers/input/mousedev.c index d528a2dba06..31ec7265aac 100644 --- a/drivers/input/mousedev.c +++ b/drivers/input/mousedev.c @@ -792,6 +792,7 @@ static const struct file_operations mousedev_fops = { .open = mousedev_open, .release = mousedev_release, .fasync = mousedev_fasync, + .llseek = noop_llseek, }; static int mousedev_install_chrdev(struct mousedev *mousedev) diff --git a/drivers/input/serio/serio_raw.c b/drivers/input/serio/serio_raw.c index 99866485444..cd82bb12591 100644 --- a/drivers/input/serio/serio_raw.c +++ b/drivers/input/serio/serio_raw.c @@ -243,6 +243,7 @@ static const struct file_operations serio_raw_fops = { .write = serio_raw_write, .poll = serio_raw_poll, .fasync = serio_raw_fasync, + .llseek = noop_llseek, }; diff --git a/drivers/isdn/mISDN/timerdev.c b/drivers/isdn/mISDN/timerdev.c index de43c8c70ad..859c81e9483 100644 --- a/drivers/isdn/mISDN/timerdev.c +++ b/drivers/isdn/mISDN/timerdev.c @@ -267,6 +267,7 @@ static const struct file_operations mISDN_fops = { .unlocked_ioctl = mISDN_ioctl, .open = mISDN_open, .release = mISDN_close, + .llseek = no_llseek, }; static struct miscdevice mISDNtimer = { diff --git a/drivers/lguest/lguest_user.c b/drivers/lguest/lguest_user.c index 85b714df8ea..3c781cdddda 100644 --- a/drivers/lguest/lguest_user.c +++ b/drivers/lguest/lguest_user.c @@ -514,6 +514,7 @@ static const struct file_operations lguest_fops = { .release = close, .write = write, .read = read, + .llseek = default_llseek, }; /* diff --git a/drivers/macintosh/ans-lcd.c b/drivers/macintosh/ans-lcd.c index a3d25da2f27..1a57e88a38f 100644 --- a/drivers/macintosh/ans-lcd.c +++ b/drivers/macintosh/ans-lcd.c @@ -137,6 +137,7 @@ const struct file_operations anslcd_fops = { .write = anslcd_write, .unlocked_ioctl = anslcd_ioctl, .open = anslcd_open, + .llseek = default_llseek, }; static struct miscdevice anslcd_dev = { diff --git a/drivers/macintosh/via-pmu.c b/drivers/macintosh/via-pmu.c index 2d17e76066b..44d171cc225 100644 --- a/drivers/macintosh/via-pmu.c +++ b/drivers/macintosh/via-pmu.c @@ -2398,6 +2398,7 @@ static const struct file_operations pmu_device_fops = { #endif .open = pmu_open, .release = pmu_release, + .llseek = noop_llseek, }; static struct miscdevice pmu_device = { diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index 3e39193e503..4b54618b415 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -1596,6 +1596,7 @@ static const struct file_operations _ctl_fops = { .unlocked_ioctl = dm_ctl_ioctl, .compat_ioctl = dm_compat_ctl_ioctl, .owner = THIS_MODULE, + .llseek = noop_llseek, }; static struct miscdevice _dm_misc = { diff --git a/drivers/media/IR/imon.c b/drivers/media/IR/imon.c index c185422ef28..faed5a332c7 100644 --- a/drivers/media/IR/imon.c +++ b/drivers/media/IR/imon.c @@ -151,7 +151,8 @@ static const struct file_operations vfd_fops = { .owner = THIS_MODULE, .open = &display_open, .write = &vfd_write, - .release = &display_close + .release = &display_close, + .llseek = noop_llseek, }; /* lcd character device file operations */ @@ -159,7 +160,8 @@ static const struct file_operations lcd_fops = { .owner = THIS_MODULE, .open = &display_open, .write = &lcd_write, - .release = &display_close + .release = &display_close, + .llseek = noop_llseek, }; enum { diff --git a/drivers/media/IR/lirc_dev.c b/drivers/media/IR/lirc_dev.c index 5b145e8672f..0acf6396e06 100644 --- a/drivers/media/IR/lirc_dev.c +++ b/drivers/media/IR/lirc_dev.c @@ -163,6 +163,7 @@ static struct file_operations fops = { .unlocked_ioctl = lirc_dev_fop_ioctl, .open = lirc_dev_fop_open, .release = lirc_dev_fop_close, + .llseek = noop_llseek, }; static int lirc_cdev_add(struct irctl *ir) diff --git a/drivers/media/dvb/bt8xx/dst_ca.c b/drivers/media/dvb/bt8xx/dst_ca.c index cf870516284..7ed74dfc6b9 100644 --- a/drivers/media/dvb/bt8xx/dst_ca.c +++ b/drivers/media/dvb/bt8xx/dst_ca.c @@ -694,7 +694,8 @@ static const struct file_operations dst_ca_fops = { .open = dst_ca_open, .release = dst_ca_release, .read = dst_ca_read, - .write = dst_ca_write + .write = dst_ca_write, + .llseek = noop_llseek, }; static struct dvb_device dvbdev_ca = { diff --git a/drivers/media/dvb/dvb-core/dmxdev.c b/drivers/media/dvb/dvb-core/dmxdev.c index 0042306ea11..75c20ac82c0 100644 --- a/drivers/media/dvb/dvb-core/dmxdev.c +++ b/drivers/media/dvb/dvb-core/dmxdev.c @@ -1150,6 +1150,7 @@ static const struct file_operations dvb_demux_fops = { .open = dvb_demux_open, .release = dvb_demux_release, .poll = dvb_demux_poll, + .llseek = default_llseek, }; static struct dvb_device dvbdev_demux = { @@ -1225,6 +1226,7 @@ static const struct file_operations dvb_dvr_fops = { .open = dvb_dvr_open, .release = dvb_dvr_release, .poll = dvb_dvr_poll, + .llseek = default_llseek, }; static struct dvb_device dvbdev_dvr = { diff --git a/drivers/media/dvb/dvb-core/dvb_ca_en50221.c b/drivers/media/dvb/dvb-core/dvb_ca_en50221.c index cb97e6b8543..ff8921dd737 100644 --- a/drivers/media/dvb/dvb-core/dvb_ca_en50221.c +++ b/drivers/media/dvb/dvb-core/dvb_ca_en50221.c @@ -1628,6 +1628,7 @@ static const struct file_operations dvb_ca_fops = { .open = dvb_ca_en50221_io_open, .release = dvb_ca_en50221_io_release, .poll = dvb_ca_en50221_io_poll, + .llseek = noop_llseek, }; static struct dvb_device dvbdev_ca = { diff --git a/drivers/media/dvb/dvb-core/dvb_frontend.c b/drivers/media/dvb/dvb-core/dvb_frontend.c index 4d45b7d6b3f..970c9b8882d 100644 --- a/drivers/media/dvb/dvb-core/dvb_frontend.c +++ b/drivers/media/dvb/dvb-core/dvb_frontend.c @@ -2034,7 +2034,8 @@ static const struct file_operations dvb_frontend_fops = { .unlocked_ioctl = dvb_generic_ioctl, .poll = dvb_frontend_poll, .open = dvb_frontend_open, - .release = dvb_frontend_release + .release = dvb_frontend_release, + .llseek = noop_llseek, }; int dvb_register_frontend(struct dvb_adapter* dvb, diff --git a/drivers/media/dvb/dvb-core/dvb_net.c b/drivers/media/dvb/dvb-core/dvb_net.c index 6c3a8a06cca..82636f517b9 100644 --- a/drivers/media/dvb/dvb-core/dvb_net.c +++ b/drivers/media/dvb/dvb-core/dvb_net.c @@ -1475,6 +1475,7 @@ static const struct file_operations dvb_net_fops = { .unlocked_ioctl = dvb_net_ioctl, .open = dvb_generic_open, .release = dvb_net_close, + .llseek = noop_llseek, }; static struct dvb_device dvbdev_net = { diff --git a/drivers/media/dvb/dvb-core/dvbdev.c b/drivers/media/dvb/dvb-core/dvbdev.c index b915c39d782..774b40e4f58 100644 --- a/drivers/media/dvb/dvb-core/dvbdev.c +++ b/drivers/media/dvb/dvb-core/dvbdev.c @@ -105,6 +105,7 @@ static const struct file_operations dvb_device_fops = { .owner = THIS_MODULE, .open = dvb_device_open, + .llseek = noop_llseek, }; static struct cdev dvb_device_cdev; diff --git a/drivers/media/dvb/firewire/firedtv-ci.c b/drivers/media/dvb/firewire/firedtv-ci.c index d3c2cf60de7..8ffb565f070 100644 --- a/drivers/media/dvb/firewire/firedtv-ci.c +++ b/drivers/media/dvb/firewire/firedtv-ci.c @@ -220,6 +220,7 @@ static const struct file_operations fdtv_ca_fops = { .open = dvb_generic_open, .release = dvb_generic_release, .poll = fdtv_ca_io_poll, + .llseek = noop_llseek, }; static struct dvb_device fdtv_ca = { diff --git a/drivers/media/dvb/ttpci/av7110.c b/drivers/media/dvb/ttpci/av7110.c index a6be529eec5..893fbc57c72 100644 --- a/drivers/media/dvb/ttpci/av7110.c +++ b/drivers/media/dvb/ttpci/av7110.c @@ -730,6 +730,7 @@ static const struct file_operations dvb_osd_fops = { .unlocked_ioctl = dvb_generic_ioctl, .open = dvb_generic_open, .release = dvb_generic_release, + .llseek = noop_llseek, }; static struct dvb_device dvbdev_osd = { diff --git a/drivers/media/dvb/ttpci/av7110_av.c b/drivers/media/dvb/ttpci/av7110_av.c index 13efba942da..6ef3996565a 100644 --- a/drivers/media/dvb/ttpci/av7110_av.c +++ b/drivers/media/dvb/ttpci/av7110_av.c @@ -1521,6 +1521,7 @@ static const struct file_operations dvb_video_fops = { .open = dvb_video_open, .release = dvb_video_release, .poll = dvb_video_poll, + .llseek = noop_llseek, }; static struct dvb_device dvbdev_video = { @@ -1539,6 +1540,7 @@ static const struct file_operations dvb_audio_fops = { .open = dvb_audio_open, .release = dvb_audio_release, .poll = dvb_audio_poll, + .llseek = noop_llseek, }; static struct dvb_device dvbdev_audio = { diff --git a/drivers/media/dvb/ttpci/av7110_ca.c b/drivers/media/dvb/ttpci/av7110_ca.c index 4eba35a018e..43f61f2eca9 100644 --- a/drivers/media/dvb/ttpci/av7110_ca.c +++ b/drivers/media/dvb/ttpci/av7110_ca.c @@ -353,6 +353,7 @@ static const struct file_operations dvb_ca_fops = { .open = dvb_ca_open, .release = dvb_generic_release, .poll = dvb_ca_poll, + .llseek = default_llseek, }; static struct dvb_device dvbdev_ca = { diff --git a/drivers/media/dvb/ttpci/av7110_ir.c b/drivers/media/dvb/ttpci/av7110_ir.c index b070e88d8c6..908f272fe26 100644 --- a/drivers/media/dvb/ttpci/av7110_ir.c +++ b/drivers/media/dvb/ttpci/av7110_ir.c @@ -312,6 +312,7 @@ static ssize_t av7110_ir_proc_write(struct file *file, const char __user *buffer static const struct file_operations av7110_ir_proc_fops = { .owner = THIS_MODULE, .write = av7110_ir_proc_write, + .llseek = noop_llseek, }; /* interrupt handler */ diff --git a/drivers/mfd/ab3100-core.c b/drivers/mfd/ab3100-core.c index 66379b41390..b048ecc56db 100644 --- a/drivers/mfd/ab3100-core.c +++ b/drivers/mfd/ab3100-core.c @@ -583,6 +583,7 @@ static ssize_t ab3100_get_set_reg(struct file *file, static const struct file_operations ab3100_get_set_reg_fops = { .open = ab3100_get_set_reg_open_file, .write = ab3100_get_set_reg, + .llseek = noop_llseek, }; static struct dentry *ab3100_dir; diff --git a/drivers/misc/hpilo.c b/drivers/misc/hpilo.c index 557a8c2a733..69c1f2fca14 100644 --- a/drivers/misc/hpilo.c +++ b/drivers/misc/hpilo.c @@ -640,6 +640,7 @@ static const struct file_operations ilo_fops = { .poll = ilo_poll, .open = ilo_open, .release = ilo_close, + .llseek = noop_llseek, }; static irqreturn_t ilo_isr(int irq, void *data) diff --git a/drivers/misc/phantom.c b/drivers/misc/phantom.c index 75ee0d3f6f4..6b38b596429 100644 --- a/drivers/misc/phantom.c +++ b/drivers/misc/phantom.c @@ -279,6 +279,7 @@ static const struct file_operations phantom_file_ops = { .unlocked_ioctl = phantom_ioctl, .compat_ioctl = phantom_compat_ioctl, .poll = phantom_poll, + .llseek = no_llseek, }; static irqreturn_t phantom_isr(int irq, void *data) diff --git a/drivers/misc/sgi-gru/grufile.c b/drivers/misc/sgi-gru/grufile.c index cb3b4d22847..28852dfa310 100644 --- a/drivers/misc/sgi-gru/grufile.c +++ b/drivers/misc/sgi-gru/grufile.c @@ -587,6 +587,7 @@ static const struct file_operations gru_fops = { .owner = THIS_MODULE, .unlocked_ioctl = gru_file_unlocked_ioctl, .mmap = gru_file_mmap, + .llseek = noop_llseek, }; static struct miscdevice gru_miscdev = { diff --git a/drivers/mmc/core/debugfs.c b/drivers/mmc/core/debugfs.c index 53cb380c098..46bc6d7551a 100644 --- a/drivers/mmc/core/debugfs.c +++ b/drivers/mmc/core/debugfs.c @@ -245,6 +245,7 @@ static const struct file_operations mmc_dbg_ext_csd_fops = { .open = mmc_ext_csd_open, .read = mmc_ext_csd_read, .release = mmc_ext_csd_release, + .llseek = default_llseek, }; void mmc_add_card_debugfs(struct mmc_card *card) diff --git a/drivers/mtd/ubi/cdev.c b/drivers/mtd/ubi/cdev.c index 3d2d1a69e9a..af9fb0ff821 100644 --- a/drivers/mtd/ubi/cdev.c +++ b/drivers/mtd/ubi/cdev.c @@ -1100,4 +1100,5 @@ const struct file_operations ubi_ctrl_cdev_operations = { .owner = THIS_MODULE, .unlocked_ioctl = ctrl_cdev_ioctl, .compat_ioctl = ctrl_cdev_compat_ioctl, + .llseek = noop_llseek, }; diff --git a/drivers/net/caif/caif_spi.c b/drivers/net/caif/caif_spi.c index f5058ff2b21..8427533fe31 100644 --- a/drivers/net/caif/caif_spi.c +++ b/drivers/net/caif/caif_spi.c @@ -240,13 +240,15 @@ static ssize_t dbgfs_frame(struct file *file, char __user *user_buf, static const struct file_operations dbgfs_state_fops = { .open = dbgfs_open, .read = dbgfs_state, - .owner = THIS_MODULE + .owner = THIS_MODULE, + .llseek = default_llseek, }; static const struct file_operations dbgfs_frame_fops = { .open = dbgfs_open, .read = dbgfs_frame, - .owner = THIS_MODULE + .owner = THIS_MODULE, + .llseek = default_llseek, }; static inline void dev_debugfs_add(struct cfspi *cfspi) diff --git a/drivers/net/cxgb4/cxgb4_main.c b/drivers/net/cxgb4/cxgb4_main.c index c327527fbbc..e2bf10d90ad 100644 --- a/drivers/net/cxgb4/cxgb4_main.c +++ b/drivers/net/cxgb4/cxgb4_main.c @@ -2026,6 +2026,7 @@ static const struct file_operations mem_debugfs_fops = { .owner = THIS_MODULE, .open = mem_open, .read = mem_read, + .llseek = default_llseek, }; static void __devinit add_debugfs_mem(struct adapter *adap, const char *name, diff --git a/drivers/net/ppp_generic.c b/drivers/net/ppp_generic.c index 6695a51e09e..323e81e9e80 100644 --- a/drivers/net/ppp_generic.c +++ b/drivers/net/ppp_generic.c @@ -856,7 +856,8 @@ static const struct file_operations ppp_device_fops = { .poll = ppp_poll, .unlocked_ioctl = ppp_ioctl, .open = ppp_open, - .release = ppp_release + .release = ppp_release, + .llseek = noop_llseek, }; static __net_init int ppp_init_net(struct net *net) diff --git a/drivers/net/wimax/i2400m/debugfs.c b/drivers/net/wimax/i2400m/debugfs.c index b1aec3e1892..9c70b5fa3f5 100644 --- a/drivers/net/wimax/i2400m/debugfs.c +++ b/drivers/net/wimax/i2400m/debugfs.c @@ -119,6 +119,7 @@ const struct file_operations i2400m_rx_stats_fops = { .open = i2400m_stats_open, .read = i2400m_rx_stats_read, .write = i2400m_rx_stats_write, + .llseek = default_llseek, }; @@ -171,6 +172,7 @@ const struct file_operations i2400m_tx_stats_fops = { .open = i2400m_stats_open, .read = i2400m_tx_stats_read, .write = i2400m_tx_stats_write, + .llseek = default_llseek, }; diff --git a/drivers/net/wireless/airo.c b/drivers/net/wireless/airo.c index 1d05445d4ba..ce77575e88b 100644 --- a/drivers/net/wireless/airo.c +++ b/drivers/net/wireless/airo.c @@ -4430,21 +4430,24 @@ static const struct file_operations proc_statsdelta_ops = { .owner = THIS_MODULE, .read = proc_read, .open = proc_statsdelta_open, - .release = proc_close + .release = proc_close, + .llseek = default_llseek, }; static const struct file_operations proc_stats_ops = { .owner = THIS_MODULE, .read = proc_read, .open = proc_stats_open, - .release = proc_close + .release = proc_close, + .llseek = default_llseek, }; static const struct file_operations proc_status_ops = { .owner = THIS_MODULE, .read = proc_read, .open = proc_status_open, - .release = proc_close + .release = proc_close, + .llseek = default_llseek, }; static const struct file_operations proc_SSID_ops = { @@ -4452,7 +4455,8 @@ static const struct file_operations proc_SSID_ops = { .read = proc_read, .write = proc_write, .open = proc_SSID_open, - .release = proc_close + .release = proc_close, + .llseek = default_llseek, }; static const struct file_operations proc_BSSList_ops = { @@ -4460,7 +4464,8 @@ static const struct file_operations proc_BSSList_ops = { .read = proc_read, .write = proc_write, .open = proc_BSSList_open, - .release = proc_close + .release = proc_close, + .llseek = default_llseek, }; static const struct file_operations proc_APList_ops = { @@ -4468,7 +4473,8 @@ static const struct file_operations proc_APList_ops = { .read = proc_read, .write = proc_write, .open = proc_APList_open, - .release = proc_close + .release = proc_close, + .llseek = default_llseek, }; static const struct file_operations proc_config_ops = { @@ -4476,7 +4482,8 @@ static const struct file_operations proc_config_ops = { .read = proc_read, .write = proc_write, .open = proc_config_open, - .release = proc_close + .release = proc_close, + .llseek = default_llseek, }; static const struct file_operations proc_wepkey_ops = { @@ -4484,7 +4491,8 @@ static const struct file_operations proc_wepkey_ops = { .read = proc_read, .write = proc_write, .open = proc_wepkey_open, - .release = proc_close + .release = proc_close, + .llseek = default_llseek, }; static struct proc_dir_entry *airo_entry; diff --git a/drivers/net/wireless/ath/ath5k/debug.c b/drivers/net/wireless/ath/ath5k/debug.c index 4cccc29964f..fb339c3852e 100644 --- a/drivers/net/wireless/ath/ath5k/debug.c +++ b/drivers/net/wireless/ath/ath5k/debug.c @@ -271,6 +271,7 @@ static const struct file_operations fops_beacon = { .write = write_file_beacon, .open = ath5k_debugfs_open, .owner = THIS_MODULE, + .llseek = default_llseek, }; @@ -290,6 +291,7 @@ static const struct file_operations fops_reset = { .write = write_file_reset, .open = ath5k_debugfs_open, .owner = THIS_MODULE, + .llseek = noop_llseek, }; @@ -369,6 +371,7 @@ static const struct file_operations fops_debug = { .write = write_file_debug, .open = ath5k_debugfs_open, .owner = THIS_MODULE, + .llseek = default_llseek, }; @@ -480,6 +483,7 @@ static const struct file_operations fops_antenna = { .write = write_file_antenna, .open = ath5k_debugfs_open, .owner = THIS_MODULE, + .llseek = default_llseek, }; @@ -591,6 +595,7 @@ static const struct file_operations fops_frameerrors = { .write = write_file_frameerrors, .open = ath5k_debugfs_open, .owner = THIS_MODULE, + .llseek = default_llseek, }; @@ -748,6 +753,7 @@ static const struct file_operations fops_ani = { .write = write_file_ani, .open = ath5k_debugfs_open, .owner = THIS_MODULE, + .llseek = default_llseek, }; @@ -811,6 +817,7 @@ static const struct file_operations fops_queue = { .write = write_file_queue, .open = ath5k_debugfs_open, .owner = THIS_MODULE, + .llseek = default_llseek, }; diff --git a/drivers/net/wireless/ath/ath9k/debug.c b/drivers/net/wireless/ath/ath9k/debug.c index 54aae931424..cf500bf25ad 100644 --- a/drivers/net/wireless/ath/ath9k/debug.c +++ b/drivers/net/wireless/ath/ath9k/debug.c @@ -71,7 +71,8 @@ static const struct file_operations fops_debug = { .read = read_file_debug, .write = write_file_debug, .open = ath9k_debugfs_open, - .owner = THIS_MODULE + .owner = THIS_MODULE, + .llseek = default_llseek, }; #endif @@ -116,7 +117,8 @@ static const struct file_operations fops_tx_chainmask = { .read = read_file_tx_chainmask, .write = write_file_tx_chainmask, .open = ath9k_debugfs_open, - .owner = THIS_MODULE + .owner = THIS_MODULE, + .llseek = default_llseek, }; @@ -158,7 +160,8 @@ static const struct file_operations fops_rx_chainmask = { .read = read_file_rx_chainmask, .write = write_file_rx_chainmask, .open = ath9k_debugfs_open, - .owner = THIS_MODULE + .owner = THIS_MODULE, + .llseek = default_llseek, }; @@ -259,7 +262,8 @@ static ssize_t read_file_dma(struct file *file, char __user *user_buf, static const struct file_operations fops_dma = { .read = read_file_dma, .open = ath9k_debugfs_open, - .owner = THIS_MODULE + .owner = THIS_MODULE, + .llseek = default_llseek, }; @@ -375,7 +379,8 @@ static ssize_t read_file_interrupt(struct file *file, char __user *user_buf, static const struct file_operations fops_interrupt = { .read = read_file_interrupt, .open = ath9k_debugfs_open, - .owner = THIS_MODULE + .owner = THIS_MODULE, + .llseek = default_llseek, }; void ath_debug_stat_rc(struct ath_softc *sc, int final_rate) @@ -464,7 +469,8 @@ static ssize_t read_file_rcstat(struct file *file, char __user *user_buf, static const struct file_operations fops_rcstat = { .read = read_file_rcstat, .open = ath9k_debugfs_open, - .owner = THIS_MODULE + .owner = THIS_MODULE, + .llseek = default_llseek, }; static const char * ath_wiphy_state_str(enum ath_wiphy_state state) @@ -623,7 +629,8 @@ static const struct file_operations fops_wiphy = { .read = read_file_wiphy, .write = write_file_wiphy, .open = ath9k_debugfs_open, - .owner = THIS_MODULE + .owner = THIS_MODULE, + .llseek = default_llseek, }; #define PR(str, elem) \ @@ -702,7 +709,8 @@ void ath_debug_stat_tx(struct ath_softc *sc, struct ath_txq *txq, static const struct file_operations fops_xmit = { .read = read_file_xmit, .open = ath9k_debugfs_open, - .owner = THIS_MODULE + .owner = THIS_MODULE, + .llseek = default_llseek, }; static ssize_t read_file_recv(struct file *file, char __user *user_buf, @@ -814,7 +822,8 @@ void ath_debug_stat_rx(struct ath_softc *sc, struct ath_rx_status *rs) static const struct file_operations fops_recv = { .read = read_file_recv, .open = ath9k_debugfs_open, - .owner = THIS_MODULE + .owner = THIS_MODULE, + .llseek = default_llseek, }; static ssize_t read_file_regidx(struct file *file, char __user *user_buf, @@ -852,7 +861,8 @@ static const struct file_operations fops_regidx = { .read = read_file_regidx, .write = write_file_regidx, .open = ath9k_debugfs_open, - .owner = THIS_MODULE + .owner = THIS_MODULE, + .llseek = default_llseek, }; static ssize_t read_file_regval(struct file *file, char __user *user_buf, @@ -894,7 +904,8 @@ static const struct file_operations fops_regval = { .read = read_file_regval, .write = write_file_regval, .open = ath9k_debugfs_open, - .owner = THIS_MODULE + .owner = THIS_MODULE, + .llseek = default_llseek, }; int ath9k_init_debug(struct ath_hw *ah) diff --git a/drivers/net/wireless/ath/ath9k/htc_drv_main.c b/drivers/net/wireless/ath/ath9k/htc_drv_main.c index 7d09b4b17bb..bc2ca7d898e 100644 --- a/drivers/net/wireless/ath/ath9k/htc_drv_main.c +++ b/drivers/net/wireless/ath/ath9k/htc_drv_main.c @@ -536,7 +536,8 @@ static ssize_t read_file_tgt_stats(struct file *file, char __user *user_buf, static const struct file_operations fops_tgt_stats = { .read = read_file_tgt_stats, .open = ath9k_debugfs_open, - .owner = THIS_MODULE + .owner = THIS_MODULE, + .llseek = default_llseek, }; static ssize_t read_file_xmit(struct file *file, char __user *user_buf, @@ -584,7 +585,8 @@ static ssize_t read_file_xmit(struct file *file, char __user *user_buf, static const struct file_operations fops_xmit = { .read = read_file_xmit, .open = ath9k_debugfs_open, - .owner = THIS_MODULE + .owner = THIS_MODULE, + .llseek = default_llseek, }; static ssize_t read_file_recv(struct file *file, char __user *user_buf, @@ -613,7 +615,8 @@ static ssize_t read_file_recv(struct file *file, char __user *user_buf, static const struct file_operations fops_recv = { .read = read_file_recv, .open = ath9k_debugfs_open, - .owner = THIS_MODULE + .owner = THIS_MODULE, + .llseek = default_llseek, }; int ath9k_htc_init_debug(struct ath_hw *ah) diff --git a/drivers/net/wireless/iwlwifi/iwl-3945-rs.c b/drivers/net/wireless/iwlwifi/iwl-3945-rs.c index 8e84a08ff95..293e1dbc166 100644 --- a/drivers/net/wireless/iwlwifi/iwl-3945-rs.c +++ b/drivers/net/wireless/iwlwifi/iwl-3945-rs.c @@ -873,6 +873,7 @@ static ssize_t iwl3945_sta_dbgfs_stats_table_read(struct file *file, static const struct file_operations rs_sta_dbgfs_stats_table_ops = { .read = iwl3945_sta_dbgfs_stats_table_read, .open = iwl3945_open_file_generic, + .llseek = default_llseek, }; static void iwl3945_add_debugfs(void *priv, void *priv_sta, diff --git a/drivers/net/wireless/iwlwifi/iwl-agn-rs.c b/drivers/net/wireless/iwlwifi/iwl-agn-rs.c index 23e5c42e7d7..a4378ba31ef 100644 --- a/drivers/net/wireless/iwlwifi/iwl-agn-rs.c +++ b/drivers/net/wireless/iwlwifi/iwl-agn-rs.c @@ -2873,6 +2873,7 @@ static const struct file_operations rs_sta_dbgfs_scale_table_ops = { .write = rs_sta_dbgfs_scale_table_write, .read = rs_sta_dbgfs_scale_table_read, .open = open_file_generic, + .llseek = default_llseek, }; static ssize_t rs_sta_dbgfs_stats_table_read(struct file *file, char __user *user_buf, size_t count, loff_t *ppos) @@ -2915,6 +2916,7 @@ static ssize_t rs_sta_dbgfs_stats_table_read(struct file *file, static const struct file_operations rs_sta_dbgfs_stats_table_ops = { .read = rs_sta_dbgfs_stats_table_read, .open = open_file_generic, + .llseek = default_llseek, }; static ssize_t rs_sta_dbgfs_rate_scale_data_read(struct file *file, @@ -2946,6 +2948,7 @@ static ssize_t rs_sta_dbgfs_rate_scale_data_read(struct file *file, static const struct file_operations rs_sta_dbgfs_rate_scale_data_ops = { .read = rs_sta_dbgfs_rate_scale_data_read, .open = open_file_generic, + .llseek = default_llseek, }; static void rs_add_debugfs(void *priv, void *priv_sta, diff --git a/drivers/net/wireless/iwmc3200wifi/debugfs.c b/drivers/net/wireless/iwmc3200wifi/debugfs.c index 53b0b7711f0..0a0cc9667cd 100644 --- a/drivers/net/wireless/iwmc3200wifi/debugfs.c +++ b/drivers/net/wireless/iwmc3200wifi/debugfs.c @@ -402,24 +402,28 @@ static const struct file_operations iwm_debugfs_txq_fops = { .owner = THIS_MODULE, .open = iwm_generic_open, .read = iwm_debugfs_txq_read, + .llseek = default_llseek, }; static const struct file_operations iwm_debugfs_tx_credit_fops = { .owner = THIS_MODULE, .open = iwm_generic_open, .read = iwm_debugfs_tx_credit_read, + .llseek = default_llseek, }; static const struct file_operations iwm_debugfs_rx_ticket_fops = { .owner = THIS_MODULE, .open = iwm_generic_open, .read = iwm_debugfs_rx_ticket_read, + .llseek = default_llseek, }; static const struct file_operations iwm_debugfs_fw_err_fops = { .owner = THIS_MODULE, .open = iwm_generic_open, .read = iwm_debugfs_fw_err_read, + .llseek = default_llseek, }; void iwm_debugfs_init(struct iwm_priv *iwm) diff --git a/drivers/net/wireless/iwmc3200wifi/sdio.c b/drivers/net/wireless/iwmc3200wifi/sdio.c index edcb52330cf..56383e7be83 100644 --- a/drivers/net/wireless/iwmc3200wifi/sdio.c +++ b/drivers/net/wireless/iwmc3200wifi/sdio.c @@ -364,6 +364,7 @@ static const struct file_operations iwm_debugfs_sdio_fops = { .owner = THIS_MODULE, .open = iwm_debugfs_sdio_open, .read = iwm_debugfs_sdio_read, + .llseek = default_llseek, }; static void if_sdio_debugfs_init(struct iwm_priv *iwm, struct dentry *parent_dir) diff --git a/drivers/net/wireless/libertas/debugfs.c b/drivers/net/wireless/libertas/debugfs.c index 74e94cc10e0..fbf3b0332bb 100644 --- a/drivers/net/wireless/libertas/debugfs.c +++ b/drivers/net/wireless/libertas/debugfs.c @@ -962,6 +962,7 @@ static const struct file_operations lbs_debug_fops = { .open = open_file_generic, .write = lbs_debugfs_write, .read = lbs_debugfs_read, + .llseek = default_llseek, }; /** diff --git a/drivers/net/wireless/ray_cs.c b/drivers/net/wireless/ray_cs.c index 88560d0ae50..dab30a8c747 100644 --- a/drivers/net/wireless/ray_cs.c +++ b/drivers/net/wireless/ray_cs.c @@ -2802,6 +2802,7 @@ static ssize_t ray_cs_essid_proc_write(struct file *file, static const struct file_operations ray_cs_essid_proc_fops = { .owner = THIS_MODULE, .write = ray_cs_essid_proc_write, + .llseek = noop_llseek, }; static ssize_t int_proc_write(struct file *file, const char __user *buffer, @@ -2835,6 +2836,7 @@ static ssize_t int_proc_write(struct file *file, const char __user *buffer, static const struct file_operations int_proc_fops = { .owner = THIS_MODULE, .write = int_proc_write, + .llseek = noop_llseek, }; #endif diff --git a/drivers/net/wireless/rt2x00/rt2x00debug.c b/drivers/net/wireless/rt2x00/rt2x00debug.c index 7d6f19a2805..cea81e4c5c8 100644 --- a/drivers/net/wireless/rt2x00/rt2x00debug.c +++ b/drivers/net/wireless/rt2x00/rt2x00debug.c @@ -315,6 +315,7 @@ static const struct file_operations rt2x00debug_fop_queue_dump = { .poll = rt2x00debug_poll_queue_dump, .open = rt2x00debug_open_queue_dump, .release = rt2x00debug_release_queue_dump, + .llseek = default_llseek, }; static ssize_t rt2x00debug_read_queue_stats(struct file *file, @@ -371,6 +372,7 @@ static const struct file_operations rt2x00debug_fop_queue_stats = { .read = rt2x00debug_read_queue_stats, .open = rt2x00debug_file_open, .release = rt2x00debug_file_release, + .llseek = default_llseek, }; #ifdef CONFIG_RT2X00_LIB_CRYPTO @@ -423,6 +425,7 @@ static const struct file_operations rt2x00debug_fop_crypto_stats = { .read = rt2x00debug_read_crypto_stats, .open = rt2x00debug_file_open, .release = rt2x00debug_file_release, + .llseek = default_llseek, }; #endif @@ -543,6 +546,7 @@ static const struct file_operations rt2x00debug_fop_dev_flags = { .read = rt2x00debug_read_dev_flags, .open = rt2x00debug_file_open, .release = rt2x00debug_file_release, + .llseek = default_llseek, }; static struct dentry *rt2x00debug_create_file_driver(const char *name, diff --git a/drivers/net/wireless/wl12xx/wl1251_debugfs.c b/drivers/net/wireless/wl12xx/wl1251_debugfs.c index a4ae7c4d94b..fa620a5e530 100644 --- a/drivers/net/wireless/wl12xx/wl1251_debugfs.c +++ b/drivers/net/wireless/wl12xx/wl1251_debugfs.c @@ -238,6 +238,7 @@ static ssize_t tx_queue_len_read(struct file *file, char __user *userbuf, static const struct file_operations tx_queue_len_ops = { .read = tx_queue_len_read, .open = wl1251_open_file_generic, + .llseek = generic_file_llseek, }; static ssize_t tx_queue_status_read(struct file *file, char __user *userbuf, @@ -259,6 +260,7 @@ static ssize_t tx_queue_status_read(struct file *file, char __user *userbuf, static const struct file_operations tx_queue_status_ops = { .read = tx_queue_status_read, .open = wl1251_open_file_generic, + .llseek = generic_file_llseek, }; static void wl1251_debugfs_delete_files(struct wl1251 *wl) diff --git a/drivers/net/wireless/wl12xx/wl1271_debugfs.c b/drivers/net/wireless/wl12xx/wl1271_debugfs.c index 6e25303a8e7..66c2b90ddfd 100644 --- a/drivers/net/wireless/wl12xx/wl1271_debugfs.c +++ b/drivers/net/wireless/wl12xx/wl1271_debugfs.c @@ -239,6 +239,7 @@ static ssize_t tx_queue_len_read(struct file *file, char __user *userbuf, static const struct file_operations tx_queue_len_ops = { .read = tx_queue_len_read, .open = wl1271_open_file_generic, + .llseek = default_llseek, }; static ssize_t gpio_power_read(struct file *file, char __user *user_buf, @@ -293,7 +294,8 @@ out: static const struct file_operations gpio_power_ops = { .read = gpio_power_read, .write = gpio_power_write, - .open = wl1271_open_file_generic + .open = wl1271_open_file_generic, + .llseek = default_llseek, }; static void wl1271_debugfs_delete_files(struct wl1271 *wl) diff --git a/drivers/oprofile/oprofile_files.c b/drivers/oprofile/oprofile_files.c index bbd7516e086..9bb4d4bdc23 100644 --- a/drivers/oprofile/oprofile_files.c +++ b/drivers/oprofile/oprofile_files.c @@ -59,6 +59,7 @@ static ssize_t timeout_write(struct file *file, char const __user *buf, static const struct file_operations timeout_fops = { .read = timeout_read, .write = timeout_write, + .llseek = default_llseek, }; #endif @@ -93,7 +94,8 @@ static ssize_t depth_write(struct file *file, char const __user *buf, size_t cou static const struct file_operations depth_fops = { .read = depth_read, - .write = depth_write + .write = depth_write, + .llseek = default_llseek, }; @@ -105,6 +107,7 @@ static ssize_t pointer_size_read(struct file *file, char __user *buf, size_t cou static const struct file_operations pointer_size_fops = { .read = pointer_size_read, + .llseek = default_llseek, }; @@ -116,6 +119,7 @@ static ssize_t cpu_type_read(struct file *file, char __user *buf, size_t count, static const struct file_operations cpu_type_fops = { .read = cpu_type_read, + .llseek = default_llseek, }; @@ -151,6 +155,7 @@ static ssize_t enable_write(struct file *file, char const __user *buf, size_t co static const struct file_operations enable_fops = { .read = enable_read, .write = enable_write, + .llseek = default_llseek, }; @@ -163,6 +168,7 @@ static ssize_t dump_write(struct file *file, char const __user *buf, size_t coun static const struct file_operations dump_fops = { .write = dump_write, + .llseek = noop_llseek, }; void oprofile_create_files(struct super_block *sb, struct dentry *root) diff --git a/drivers/oprofile/oprofilefs.c b/drivers/oprofile/oprofilefs.c index 2766a6d3c2e..6b6a1f71957 100644 --- a/drivers/oprofile/oprofilefs.c +++ b/drivers/oprofile/oprofilefs.c @@ -117,12 +117,14 @@ static const struct file_operations ulong_fops = { .read = ulong_read_file, .write = ulong_write_file, .open = default_open, + .llseek = default_llseek, }; static const struct file_operations ulong_ro_fops = { .read = ulong_read_file, .open = default_open, + .llseek = default_llseek, }; @@ -183,6 +185,7 @@ static ssize_t atomic_read_file(struct file *file, char __user *buf, size_t coun static const struct file_operations atomic_ro_fops = { .read = atomic_read_file, .open = default_open, + .llseek = default_llseek, }; diff --git a/drivers/pci/pcie/aer/aer_inject.c b/drivers/pci/pcie/aer/aer_inject.c index 909924692b8..b3cf6223f63 100644 --- a/drivers/pci/pcie/aer/aer_inject.c +++ b/drivers/pci/pcie/aer/aer_inject.c @@ -472,6 +472,7 @@ static ssize_t aer_inject_write(struct file *filp, const char __user *ubuf, static const struct file_operations aer_inject_fops = { .write = aer_inject_write, .owner = THIS_MODULE, + .llseek = noop_llseek, }; static struct miscdevice aer_inject_device = { diff --git a/drivers/platform/x86/sony-laptop.c b/drivers/platform/x86/sony-laptop.c index e3154ff7a39..f200677851b 100644 --- a/drivers/platform/x86/sony-laptop.c +++ b/drivers/platform/x86/sony-laptop.c @@ -2360,6 +2360,7 @@ static const struct file_operations sonypi_misc_fops = { .release = sonypi_misc_release, .fasync = sonypi_misc_fasync, .unlocked_ioctl = sonypi_misc_ioctl, + .llseek = noop_llseek, }; static struct miscdevice sonypi_misc_device = { diff --git a/drivers/rtc/rtc-m41t80.c b/drivers/rtc/rtc-m41t80.c index d60557cae8e..ef151ce0443 100644 --- a/drivers/rtc/rtc-m41t80.c +++ b/drivers/rtc/rtc-m41t80.c @@ -748,6 +748,7 @@ static const struct file_operations wdt_fops = { .write = wdt_write, .open = wdt_open, .release = wdt_release, + .llseek = no_llseek, }; static struct miscdevice wdt_dev = { diff --git a/drivers/s390/block/dasd_eer.c b/drivers/s390/block/dasd_eer.c index 7158f9528ec..c71d89dba30 100644 --- a/drivers/s390/block/dasd_eer.c +++ b/drivers/s390/block/dasd_eer.c @@ -670,6 +670,7 @@ static const struct file_operations dasd_eer_fops = { .read = &dasd_eer_read, .poll = &dasd_eer_poll, .owner = THIS_MODULE, + .llseek = noop_llseek, }; static struct miscdevice *dasd_eer_dev = NULL; diff --git a/drivers/s390/char/fs3270.c b/drivers/s390/char/fs3270.c index 857dfcb7b35..eb28fb01a38 100644 --- a/drivers/s390/char/fs3270.c +++ b/drivers/s390/char/fs3270.c @@ -520,6 +520,7 @@ static const struct file_operations fs3270_fops = { .compat_ioctl = fs3270_ioctl, /* ioctl */ .open = fs3270_open, /* open */ .release = fs3270_close, /* release */ + .llseek = no_llseek, }; /* diff --git a/drivers/s390/char/monreader.c b/drivers/s390/char/monreader.c index e021ec663ef..5b8b8592d31 100644 --- a/drivers/s390/char/monreader.c +++ b/drivers/s390/char/monreader.c @@ -447,6 +447,7 @@ static const struct file_operations mon_fops = { .release = &mon_close, .read = &mon_read, .poll = &mon_poll, + .llseek = noop_llseek, }; static struct miscdevice mon_dev = { diff --git a/drivers/s390/char/monwriter.c b/drivers/s390/char/monwriter.c index 572a1e7fd09..e0702d3ea33 100644 --- a/drivers/s390/char/monwriter.c +++ b/drivers/s390/char/monwriter.c @@ -274,6 +274,7 @@ static const struct file_operations monwrite_fops = { .open = &monwrite_open, .release = &monwrite_close, .write = &monwrite_write, + .llseek = noop_llseek, }; static struct miscdevice mon_dev = { diff --git a/drivers/s390/char/tape_char.c b/drivers/s390/char/tape_char.c index 539045acaad..883e2db02bd 100644 --- a/drivers/s390/char/tape_char.c +++ b/drivers/s390/char/tape_char.c @@ -53,6 +53,7 @@ static const struct file_operations tape_fops = #endif .open = tapechar_open, .release = tapechar_release, + .llseek = no_llseek, }; static int tapechar_major = TAPECHAR_MAJOR; diff --git a/drivers/s390/char/vmcp.c b/drivers/s390/char/vmcp.c index 04e532eec03..0e7cb1a8415 100644 --- a/drivers/s390/char/vmcp.c +++ b/drivers/s390/char/vmcp.c @@ -177,6 +177,7 @@ static const struct file_operations vmcp_fops = { .write = vmcp_write, .unlocked_ioctl = vmcp_ioctl, .compat_ioctl = vmcp_ioctl, + .llseek = no_llseek, }; static struct miscdevice vmcp_dev = { diff --git a/drivers/s390/char/vmlogrdr.c b/drivers/s390/char/vmlogrdr.c index e40a1b89286..0d6dc4b92cc 100644 --- a/drivers/s390/char/vmlogrdr.c +++ b/drivers/s390/char/vmlogrdr.c @@ -97,6 +97,7 @@ static const struct file_operations vmlogrdr_fops = { .open = vmlogrdr_open, .release = vmlogrdr_release, .read = vmlogrdr_read, + .llseek = no_llseek, }; diff --git a/drivers/s390/char/vmwatchdog.c b/drivers/s390/char/vmwatchdog.c index e13508c98b1..12ef9121d4f 100644 --- a/drivers/s390/char/vmwatchdog.c +++ b/drivers/s390/char/vmwatchdog.c @@ -297,6 +297,7 @@ static const struct file_operations vmwdt_fops = { .unlocked_ioctl = &vmwdt_ioctl, .write = &vmwdt_write, .owner = THIS_MODULE, + .llseek = noop_llseek, }; static struct miscdevice vmwdt_dev = { diff --git a/drivers/s390/char/zcore.c b/drivers/s390/char/zcore.c index f5ea3384a4b..3b94044027c 100644 --- a/drivers/s390/char/zcore.c +++ b/drivers/s390/char/zcore.c @@ -459,6 +459,7 @@ static const struct file_operations zcore_memmap_fops = { .read = zcore_memmap_read, .open = zcore_memmap_open, .release = zcore_memmap_release, + .llseek = no_llseek, }; static ssize_t zcore_reipl_write(struct file *filp, const char __user *buf, @@ -486,6 +487,7 @@ static const struct file_operations zcore_reipl_fops = { .write = zcore_reipl_write, .open = zcore_reipl_open, .release = zcore_reipl_release, + .llseek = no_llseek, }; #ifdef CONFIG_32BIT diff --git a/drivers/s390/cio/chsc_sch.c b/drivers/s390/cio/chsc_sch.c index a83877c664a..f2b77e7bfc6 100644 --- a/drivers/s390/cio/chsc_sch.c +++ b/drivers/s390/cio/chsc_sch.c @@ -806,6 +806,7 @@ static const struct file_operations chsc_fops = { .open = nonseekable_open, .unlocked_ioctl = chsc_ioctl, .compat_ioctl = chsc_ioctl, + .llseek = no_llseek, }; static struct miscdevice chsc_misc_device = { diff --git a/drivers/s390/cio/css.c b/drivers/s390/cio/css.c index ac94ac75145..ca8e1c240c3 100644 --- a/drivers/s390/cio/css.c +++ b/drivers/s390/cio/css.c @@ -1067,6 +1067,7 @@ static ssize_t cio_settle_write(struct file *file, const char __user *buf, static const struct file_operations cio_settle_proc_fops = { .open = nonseekable_open, .write = cio_settle_write, + .llseek = no_llseek, }; static int __init cio_settle_init(void) diff --git a/drivers/s390/crypto/zcrypt_api.c b/drivers/s390/crypto/zcrypt_api.c index 41e0aaefafd..f5221749d18 100644 --- a/drivers/s390/crypto/zcrypt_api.c +++ b/drivers/s390/crypto/zcrypt_api.c @@ -897,7 +897,8 @@ static const struct file_operations zcrypt_fops = { .compat_ioctl = zcrypt_compat_ioctl, #endif .open = zcrypt_open, - .release = zcrypt_release + .release = zcrypt_release, + .llseek = no_llseek, }; /* diff --git a/drivers/s390/scsi/zfcp_cfdc.c b/drivers/s390/scsi/zfcp_cfdc.c index fcbd2b756da..1838cda68ba 100644 --- a/drivers/s390/scsi/zfcp_cfdc.c +++ b/drivers/s390/scsi/zfcp_cfdc.c @@ -251,8 +251,9 @@ static const struct file_operations zfcp_cfdc_fops = { .open = nonseekable_open, .unlocked_ioctl = zfcp_cfdc_dev_ioctl, #ifdef CONFIG_COMPAT - .compat_ioctl = zfcp_cfdc_dev_ioctl + .compat_ioctl = zfcp_cfdc_dev_ioctl, #endif + .llseek = no_llseek, }; struct miscdevice zfcp_cfdc_misc = { diff --git a/drivers/sbus/char/display7seg.c b/drivers/sbus/char/display7seg.c index 1690e53fb84..55f71ea9c41 100644 --- a/drivers/sbus/char/display7seg.c +++ b/drivers/sbus/char/display7seg.c @@ -162,6 +162,7 @@ static const struct file_operations d7s_fops = { .compat_ioctl = d7s_ioctl, .open = d7s_open, .release = d7s_release, + .llseek = noop_llseek, }; static struct miscdevice d7s_miscdev = { diff --git a/drivers/sbus/char/envctrl.c b/drivers/sbus/char/envctrl.c index 078e5f4520e..8ce414e3948 100644 --- a/drivers/sbus/char/envctrl.c +++ b/drivers/sbus/char/envctrl.c @@ -720,6 +720,7 @@ static const struct file_operations envctrl_fops = { #endif .open = envctrl_open, .release = envctrl_release, + .llseek = noop_llseek, }; static struct miscdevice envctrl_dev = { diff --git a/drivers/scsi/3w-9xxx.c b/drivers/scsi/3w-9xxx.c index e20b7bdd4c7..67aad69cfbc 100644 --- a/drivers/scsi/3w-9xxx.c +++ b/drivers/scsi/3w-9xxx.c @@ -222,7 +222,8 @@ static const struct file_operations twa_fops = { .owner = THIS_MODULE, .unlocked_ioctl = twa_chrdev_ioctl, .open = twa_chrdev_open, - .release = NULL + .release = NULL, + .llseek = noop_llseek, }; /* This function will complete an aen request from the isr */ diff --git a/drivers/scsi/3w-sas.c b/drivers/scsi/3w-sas.c index f481e734aad..7afac93d889 100644 --- a/drivers/scsi/3w-sas.c +++ b/drivers/scsi/3w-sas.c @@ -890,7 +890,8 @@ static const struct file_operations twl_fops = { .owner = THIS_MODULE, .unlocked_ioctl = twl_chrdev_ioctl, .open = twl_chrdev_open, - .release = NULL + .release = NULL, + .llseek = noop_llseek, }; /* This function passes sense data from firmware to scsi layer */ diff --git a/drivers/scsi/3w-xxxx.c b/drivers/scsi/3w-xxxx.c index 30d735ad35b..5a233730603 100644 --- a/drivers/scsi/3w-xxxx.c +++ b/drivers/scsi/3w-xxxx.c @@ -1059,7 +1059,8 @@ static const struct file_operations tw_fops = { .owner = THIS_MODULE, .unlocked_ioctl = tw_chrdev_ioctl, .open = tw_chrdev_open, - .release = NULL + .release = NULL, + .llseek = noop_llseek, }; /* This function will free up device extension resources */ diff --git a/drivers/scsi/aacraid/linit.c b/drivers/scsi/aacraid/linit.c index cad6f9abaeb..13af86eec96 100644 --- a/drivers/scsi/aacraid/linit.c +++ b/drivers/scsi/aacraid/linit.c @@ -1039,6 +1039,7 @@ static const struct file_operations aac_cfg_fops = { .compat_ioctl = aac_compat_cfg_ioctl, #endif .open = aac_cfg_open, + .llseek = noop_llseek, }; static struct scsi_host_template aac_driver_template = { diff --git a/drivers/scsi/ch.c b/drivers/scsi/ch.c index d6532187f61..e40c9f7a002 100644 --- a/drivers/scsi/ch.c +++ b/drivers/scsi/ch.c @@ -981,6 +981,7 @@ static const struct file_operations changer_fops = { #ifdef CONFIG_COMPAT .compat_ioctl = ch_ioctl_compat, #endif + .llseek = noop_llseek, }; static int __init init_ch_module(void) diff --git a/drivers/scsi/dpt_i2o.c b/drivers/scsi/dpt_i2o.c index ffc1edf5e80..7a61206ed1c 100644 --- a/drivers/scsi/dpt_i2o.c +++ b/drivers/scsi/dpt_i2o.c @@ -126,6 +126,7 @@ static const struct file_operations adpt_fops = { #ifdef CONFIG_COMPAT .compat_ioctl = compat_adpt_ioctl, #endif + .llseek = noop_llseek, }; /* Structures and definitions for synchronous message posting. diff --git a/drivers/scsi/gdth.c b/drivers/scsi/gdth.c index b860d650a56..9886586df01 100644 --- a/drivers/scsi/gdth.c +++ b/drivers/scsi/gdth.c @@ -372,6 +372,7 @@ static const struct file_operations gdth_fops = { .unlocked_ioctl = gdth_unlocked_ioctl, .open = gdth_open, .release = gdth_close, + .llseek = noop_llseek, }; #include "gdth_proc.h" diff --git a/drivers/scsi/megaraid.c b/drivers/scsi/megaraid.c index 0b6e3228610..aac80f7bb99 100644 --- a/drivers/scsi/megaraid.c +++ b/drivers/scsi/megaraid.c @@ -101,6 +101,7 @@ static const struct file_operations megadev_fops = { .owner = THIS_MODULE, .unlocked_ioctl = megadev_unlocked_ioctl, .open = megadev_open, + .llseek = noop_llseek, }; /* diff --git a/drivers/scsi/megaraid/megaraid_mm.c b/drivers/scsi/megaraid/megaraid_mm.c index 41f82f76d88..ab801232d77 100644 --- a/drivers/scsi/megaraid/megaraid_mm.c +++ b/drivers/scsi/megaraid/megaraid_mm.c @@ -75,6 +75,7 @@ static const struct file_operations lsi_fops = { .compat_ioctl = mraid_mm_compat_ioctl, #endif .owner = THIS_MODULE, + .llseek = noop_llseek, }; static struct miscdevice megaraid_mm_dev = { diff --git a/drivers/scsi/megaraid/megaraid_sas.c b/drivers/scsi/megaraid/megaraid_sas.c index 99e4478c3f3..209cc87b9a3 100644 --- a/drivers/scsi/megaraid/megaraid_sas.c +++ b/drivers/scsi/megaraid/megaraid_sas.c @@ -3957,6 +3957,7 @@ static const struct file_operations megasas_mgmt_fops = { #ifdef CONFIG_COMPAT .compat_ioctl = megasas_mgmt_compat_ioctl, #endif + .llseek = noop_llseek, }; /* diff --git a/drivers/scsi/mpt2sas/mpt2sas_ctl.c b/drivers/scsi/mpt2sas/mpt2sas_ctl.c index b774973f076..9a4e584ae3c 100644 --- a/drivers/scsi/mpt2sas/mpt2sas_ctl.c +++ b/drivers/scsi/mpt2sas/mpt2sas_ctl.c @@ -2952,6 +2952,7 @@ static const struct file_operations ctl_fops = { #ifdef CONFIG_COMPAT .compat_ioctl = _ctl_ioctl_compat, #endif + .llseek = noop_llseek, }; static struct miscdevice ctl_dev = { diff --git a/drivers/scsi/osd/osd_uld.c b/drivers/scsi/osd/osd_uld.c index ffdd9fdb999..b31a8e3841d 100644 --- a/drivers/scsi/osd/osd_uld.c +++ b/drivers/scsi/osd/osd_uld.c @@ -182,6 +182,7 @@ static const struct file_operations osd_fops = { .open = osd_uld_open, .release = osd_uld_release, .unlocked_ioctl = osd_uld_ioctl, + .llseek = noop_llseek, }; struct osd_dev *osduld_path_lookup(const char *name) diff --git a/drivers/scsi/pmcraid.c b/drivers/scsi/pmcraid.c index ecc45c8b4e6..4b8765785ae 100644 --- a/drivers/scsi/pmcraid.c +++ b/drivers/scsi/pmcraid.c @@ -4165,6 +4165,7 @@ static const struct file_operations pmcraid_fops = { #ifdef CONFIG_COMPAT .compat_ioctl = pmcraid_chr_ioctl, #endif + .llseek = noop_llseek, }; diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index 1e4bff69525..9946fac5425 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -3948,6 +3948,7 @@ static struct pci_driver qla2xxx_pci_driver = { static struct file_operations apidev_fops = { .owner = THIS_MODULE, + .llseek = noop_llseek, }; /** diff --git a/drivers/scsi/scsi_tgt_if.c b/drivers/scsi/scsi_tgt_if.c index a87e21c35ef..bbb02f6e917 100644 --- a/drivers/scsi/scsi_tgt_if.c +++ b/drivers/scsi/scsi_tgt_if.c @@ -333,6 +333,7 @@ static const struct file_operations tgt_fops = { .poll = tgt_poll, .write = tgt_write, .mmap = tgt_mmap, + .llseek = noop_llseek, }; static struct miscdevice tgt_miscdev = { diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c index 78d616315d8..00baea1e2db 100644 --- a/drivers/scsi/sg.c +++ b/drivers/scsi/sg.c @@ -1351,6 +1351,7 @@ static const struct file_operations sg_fops = { .mmap = sg_mmap, .release = sg_release, .fasync = sg_fasync, + .llseek = no_llseek, }; static struct class *sg_sysfs_class; diff --git a/drivers/serial/mfd.c b/drivers/serial/mfd.c index bc9af503907..6703f3e802a 100644 --- a/drivers/serial/mfd.c +++ b/drivers/serial/mfd.c @@ -227,12 +227,14 @@ static const struct file_operations port_regs_ops = { .owner = THIS_MODULE, .open = hsu_show_regs_open, .read = port_show_regs, + .llseek = default_llseek, }; static const struct file_operations dma_regs_ops = { .owner = THIS_MODULE, .open = hsu_show_regs_open, .read = dma_show_regs, + .llseek = default_llseek, }; static int hsu_debugfs_init(struct hsu_port *hsu) diff --git a/drivers/spi/dw_spi.c b/drivers/spi/dw_spi.c index d256cb00604..a32aa66d66b 100644 --- a/drivers/spi/dw_spi.c +++ b/drivers/spi/dw_spi.c @@ -131,6 +131,7 @@ static const struct file_operations mrst_spi_regs_ops = { .owner = THIS_MODULE, .open = spi_show_regs_open, .read = spi_show_regs, + .llseek = default_llseek, }; static int mrst_spi_debugfs_init(struct dw_spi *dws) diff --git a/drivers/spi/spidev.c b/drivers/spi/spidev.c index ea1bec3c9a1..4e6245e6799 100644 --- a/drivers/spi/spidev.c +++ b/drivers/spi/spidev.c @@ -545,6 +545,7 @@ static const struct file_operations spidev_fops = { .unlocked_ioctl = spidev_ioctl, .open = spidev_open, .release = spidev_release, + .llseek = no_llseek, }; /*-------------------------------------------------------------------------*/ diff --git a/drivers/staging/comedi/comedi_fops.c b/drivers/staging/comedi/comedi_fops.c index 14091313ceb..fecb89e8c66 100644 --- a/drivers/staging/comedi/comedi_fops.c +++ b/drivers/staging/comedi/comedi_fops.c @@ -1922,6 +1922,7 @@ const struct file_operations comedi_fops = { .mmap = comedi_mmap, .poll = comedi_poll, .fasync = comedi_fasync, + .llseek = noop_llseek, }; struct class *comedi_class; diff --git a/drivers/staging/crystalhd/crystalhd_lnx.c b/drivers/staging/crystalhd/crystalhd_lnx.c index fbb80f09a3d..af258991fe7 100644 --- a/drivers/staging/crystalhd/crystalhd_lnx.c +++ b/drivers/staging/crystalhd/crystalhd_lnx.c @@ -351,6 +351,7 @@ static const struct file_operations chd_dec_fops = { .unlocked_ioctl = chd_dec_ioctl, .open = chd_dec_open, .release = chd_dec_close, + .llseek = noop_llseek, }; static int __devinit chd_dec_init_chdev(struct crystalhd_adp *adp) diff --git a/drivers/staging/dream/camera/msm_camera.c b/drivers/staging/dream/camera/msm_camera.c index 81bd71fd816..de4ab61efd4 100644 --- a/drivers/staging/dream/camera/msm_camera.c +++ b/drivers/staging/dream/camera/msm_camera.c @@ -1941,6 +1941,7 @@ static const struct file_operations msm_fops_config = { .open = msm_open, .unlocked_ioctl = msm_ioctl_config, .release = msm_release_config, + .llseek = no_llseek, }; static const struct file_operations msm_fops_control = { @@ -1948,6 +1949,7 @@ static const struct file_operations msm_fops_control = { .open = msm_open_control, .unlocked_ioctl = msm_ioctl_control, .release = msm_release_control, + .llseek = no_llseek, }; static const struct file_operations msm_fops_frame = { @@ -1956,6 +1958,7 @@ static const struct file_operations msm_fops_frame = { .unlocked_ioctl = msm_ioctl_frame, .release = msm_release_frame, .poll = msm_poll_frame, + .llseek = no_llseek, }; static int msm_setup_cdev(struct msm_device *msm, diff --git a/drivers/staging/dream/pmem.c b/drivers/staging/dream/pmem.c index 7d6bbadd7fc..3640d1f2376 100644 --- a/drivers/staging/dream/pmem.c +++ b/drivers/staging/dream/pmem.c @@ -180,6 +180,7 @@ const struct file_operations pmem_fops = { .mmap = pmem_mmap, .open = pmem_open, .unlocked_ioctl = pmem_ioctl, + .llseek = noop_llseek, }; static int get_id(struct file *file) @@ -1204,6 +1205,7 @@ static ssize_t debug_read(struct file *file, char __user *buf, size_t count, static struct file_operations debug_fops = { .read = debug_read, .open = debug_open, + .llseek = default_llseek, }; #endif diff --git a/drivers/staging/dream/qdsp5/adsp_driver.c b/drivers/staging/dream/qdsp5/adsp_driver.c index 8197765aae1..28a6f8da947 100644 --- a/drivers/staging/dream/qdsp5/adsp_driver.c +++ b/drivers/staging/dream/qdsp5/adsp_driver.c @@ -582,6 +582,7 @@ static struct file_operations adsp_fops = { .open = adsp_open, .unlocked_ioctl = adsp_ioctl, .release = adsp_release, + .llseek = no_llseek, }; static void adsp_create(struct adsp_device *adev, const char *name, diff --git a/drivers/staging/dream/qdsp5/audio_aac.c b/drivers/staging/dream/qdsp5/audio_aac.c index a373f352238..45f4c78ab6e 100644 --- a/drivers/staging/dream/qdsp5/audio_aac.c +++ b/drivers/staging/dream/qdsp5/audio_aac.c @@ -1030,6 +1030,7 @@ static struct file_operations audio_aac_fops = { .read = audio_read, .write = audio_write, .unlocked_ioctl = audio_ioctl, + .llseek = noop_llseek, }; struct miscdevice audio_aac_misc = { diff --git a/drivers/staging/dream/qdsp5/audio_amrnb.c b/drivers/staging/dream/qdsp5/audio_amrnb.c index 07b79d5836e..402bbc13281 100644 --- a/drivers/staging/dream/qdsp5/audio_amrnb.c +++ b/drivers/staging/dream/qdsp5/audio_amrnb.c @@ -841,6 +841,7 @@ static struct file_operations audio_amrnb_fops = { .read = audamrnb_read, .write = audamrnb_write, .unlocked_ioctl = audamrnb_ioctl, + .llseek = noop_llseek, }; struct miscdevice audio_amrnb_misc = { diff --git a/drivers/staging/dream/qdsp5/audio_evrc.c b/drivers/staging/dream/qdsp5/audio_evrc.c index ad989ee8769..24a89264737 100644 --- a/drivers/staging/dream/qdsp5/audio_evrc.c +++ b/drivers/staging/dream/qdsp5/audio_evrc.c @@ -813,6 +813,7 @@ static struct file_operations audio_evrc_fops = { .read = audevrc_read, .write = audevrc_write, .unlocked_ioctl = audevrc_ioctl, + .llseek = noop_llseek, }; struct miscdevice audio_evrc_misc = { diff --git a/drivers/staging/dream/qdsp5/audio_in.c b/drivers/staging/dream/qdsp5/audio_in.c index 6ae48e72d14..b51fa096074 100644 --- a/drivers/staging/dream/qdsp5/audio_in.c +++ b/drivers/staging/dream/qdsp5/audio_in.c @@ -921,12 +921,14 @@ static struct file_operations audio_fops = { .read = audio_in_read, .write = audio_in_write, .unlocked_ioctl = audio_in_ioctl, + .llseek = noop_llseek, }; static struct file_operations audpre_fops = { .owner = THIS_MODULE, .open = audpre_open, .unlocked_ioctl = audpre_ioctl, + .llseek = noop_llseek, }; struct miscdevice audio_in_misc = { diff --git a/drivers/staging/dream/qdsp5/audio_mp3.c b/drivers/staging/dream/qdsp5/audio_mp3.c index 530e1f35eed..409a19ce603 100644 --- a/drivers/staging/dream/qdsp5/audio_mp3.c +++ b/drivers/staging/dream/qdsp5/audio_mp3.c @@ -948,6 +948,7 @@ static struct file_operations audio_mp3_fops = { .read = audio_read, .write = audio_write, .unlocked_ioctl = audio_ioctl, + .llseek = noop_llseek, }; struct miscdevice audio_mp3_misc = { diff --git a/drivers/staging/dream/qdsp5/audio_out.c b/drivers/staging/dream/qdsp5/audio_out.c index 76d7fa5667d..d20e8954156 100644 --- a/drivers/staging/dream/qdsp5/audio_out.c +++ b/drivers/staging/dream/qdsp5/audio_out.c @@ -807,12 +807,14 @@ static struct file_operations audio_fops = { .read = audio_read, .write = audio_write, .unlocked_ioctl = audio_ioctl, + .llseek = noop_llseek, }; static struct file_operations audpp_fops = { .owner = THIS_MODULE, .open = audpp_open, .unlocked_ioctl = audpp_ioctl, + .llseek = noop_llseek, }; struct miscdevice audio_misc = { diff --git a/drivers/staging/dream/qdsp5/audio_qcelp.c b/drivers/staging/dream/qdsp5/audio_qcelp.c index effa96f34fd..911bab416b8 100644 --- a/drivers/staging/dream/qdsp5/audio_qcelp.c +++ b/drivers/staging/dream/qdsp5/audio_qcelp.c @@ -824,6 +824,7 @@ static struct file_operations audio_qcelp_fops = { .read = audqcelp_read, .write = audqcelp_write, .unlocked_ioctl = audqcelp_ioctl, + .llseek = noop_llseek, }; struct miscdevice audio_qcelp_misc = { diff --git a/drivers/staging/dream/qdsp5/evlog.h b/drivers/staging/dream/qdsp5/evlog.h index 922ce670a32..e5ab86b9dd7 100644 --- a/drivers/staging/dream/qdsp5/evlog.h +++ b/drivers/staging/dream/qdsp5/evlog.h @@ -123,6 +123,7 @@ static int ev_log_open(struct inode *inode, struct file *file) static const struct file_operations ev_log_ops = { .read = ev_log_read, .open = ev_log_open, + .llseek = default_llseek, }; static int ev_log_init(struct ev_log *log) diff --git a/drivers/staging/dream/qdsp5/snd.c b/drivers/staging/dream/qdsp5/snd.c index 037d7ffb7e6..e0f2f7bca29 100644 --- a/drivers/staging/dream/qdsp5/snd.c +++ b/drivers/staging/dream/qdsp5/snd.c @@ -247,6 +247,7 @@ static struct file_operations snd_fops = { .open = snd_open, .release = snd_release, .unlocked_ioctl = snd_ioctl, + .llseek = noop_llseek, }; struct miscdevice snd_misc = { diff --git a/drivers/staging/frontier/alphatrack.c b/drivers/staging/frontier/alphatrack.c index 4e52105e607..689099b57fd 100644 --- a/drivers/staging/frontier/alphatrack.c +++ b/drivers/staging/frontier/alphatrack.c @@ -641,6 +641,7 @@ static const struct file_operations usb_alphatrack_fops = { .open = usb_alphatrack_open, .release = usb_alphatrack_release, .poll = usb_alphatrack_poll, + .llseek = no_llseek, }; /* diff --git a/drivers/staging/frontier/tranzport.c b/drivers/staging/frontier/tranzport.c index eed74f0fe0b..3d12c1737ed 100644 --- a/drivers/staging/frontier/tranzport.c +++ b/drivers/staging/frontier/tranzport.c @@ -767,6 +767,7 @@ static const struct file_operations usb_tranzport_fops = { .open = usb_tranzport_open, .release = usb_tranzport_release, .poll = usb_tranzport_poll, + .llseek = no_llseek, }; /* diff --git a/drivers/staging/iio/industrialio-core.c b/drivers/staging/iio/industrialio-core.c index dd4d87a8bca..92a212f064b 100644 --- a/drivers/staging/iio/industrialio-core.c +++ b/drivers/staging/iio/industrialio-core.c @@ -349,6 +349,7 @@ static const struct file_operations iio_event_chrdev_fileops = { .release = iio_event_chrdev_release, .open = iio_event_chrdev_open, .owner = THIS_MODULE, + .llseek = noop_llseek, }; static void iio_event_dev_release(struct device *dev) diff --git a/drivers/staging/iio/industrialio-ring.c b/drivers/staging/iio/industrialio-ring.c index 6ab578e4f5f..1c5f67253b8 100644 --- a/drivers/staging/iio/industrialio-ring.c +++ b/drivers/staging/iio/industrialio-ring.c @@ -133,6 +133,7 @@ static const struct file_operations iio_ring_fileops = { .release = iio_ring_release, .open = iio_ring_open, .owner = THIS_MODULE, + .llseek = noop_llseek, }; /** diff --git a/drivers/staging/lirc/lirc_imon.c b/drivers/staging/lirc/lirc_imon.c index 66493253042..ed5c5fe022c 100644 --- a/drivers/staging/lirc/lirc_imon.c +++ b/drivers/staging/lirc/lirc_imon.c @@ -115,7 +115,8 @@ static const struct file_operations display_fops = { .owner = THIS_MODULE, .open = &display_open, .write = &vfd_write, - .release = &display_close + .release = &display_close, + .llseek = noop_llseek, }; /* diff --git a/drivers/staging/lirc/lirc_it87.c b/drivers/staging/lirc/lirc_it87.c index ec11c0e949a..543c5c3bf90 100644 --- a/drivers/staging/lirc/lirc_it87.c +++ b/drivers/staging/lirc/lirc_it87.c @@ -342,6 +342,7 @@ static const struct file_operations lirc_fops = { .unlocked_ioctl = lirc_ioctl, .open = lirc_open, .release = lirc_close, + .llseek = noop_llseek, }; static int set_use_inc(void *data) diff --git a/drivers/staging/lirc/lirc_sasem.c b/drivers/staging/lirc/lirc_sasem.c index 73166c3f581..8f72a84f34e 100644 --- a/drivers/staging/lirc/lirc_sasem.c +++ b/drivers/staging/lirc/lirc_sasem.c @@ -125,6 +125,7 @@ static const struct file_operations vfd_fops = { .write = &vfd_write, .unlocked_ioctl = &vfd_ioctl, .release = &vfd_close, + .llseek = noop_llseek, }; /* USB Device ID for Sasem USB Control Board */ diff --git a/drivers/staging/memrar/memrar_handler.c b/drivers/staging/memrar/memrar_handler.c index a98b3f1f11e..cfcaa8e5b8e 100644 --- a/drivers/staging/memrar/memrar_handler.c +++ b/drivers/staging/memrar/memrar_handler.c @@ -890,6 +890,7 @@ static const struct file_operations memrar_fops = { .mmap = memrar_mmap, .open = memrar_open, .release = memrar_release, + .llseek = no_llseek, }; static struct miscdevice memrar_miscdev = { diff --git a/drivers/staging/panel/panel.c b/drivers/staging/panel/panel.c index 3221814a856..6885f9a4660 100644 --- a/drivers/staging/panel/panel.c +++ b/drivers/staging/panel/panel.c @@ -1631,6 +1631,7 @@ static const struct file_operations keypad_fops = { .read = keypad_read, /* read */ .open = keypad_open, /* open */ .release = keypad_release, /* close */ + .llseek = default_llseek, }; static struct miscdevice keypad_dev = { diff --git a/drivers/staging/tidspbridge/rmgr/drv_interface.c b/drivers/staging/tidspbridge/rmgr/drv_interface.c index 7ee89492a75..7b3a7d04a10 100644 --- a/drivers/staging/tidspbridge/rmgr/drv_interface.c +++ b/drivers/staging/tidspbridge/rmgr/drv_interface.c @@ -144,6 +144,7 @@ static const struct file_operations bridge_fops = { .release = bridge_release, .unlocked_ioctl = bridge_ioctl, .mmap = bridge_mmap, + .llseek = noop_llseek, }; #ifdef CONFIG_PM diff --git a/drivers/telephony/ixj.c b/drivers/telephony/ixj.c index b53deee25d7..5c6239e5aa2 100644 --- a/drivers/telephony/ixj.c +++ b/drivers/telephony/ixj.c @@ -6676,7 +6676,8 @@ static const struct file_operations ixj_fops = .poll = ixj_poll, .unlocked_ioctl = ixj_ioctl, .release = ixj_release, - .fasync = ixj_fasync + .fasync = ixj_fasync, + .llseek = default_llseek, }; static int ixj_linetest(IXJ *j) diff --git a/drivers/telephony/phonedev.c b/drivers/telephony/phonedev.c index f3873f650bb..1915af20117 100644 --- a/drivers/telephony/phonedev.c +++ b/drivers/telephony/phonedev.c @@ -130,6 +130,7 @@ static const struct file_operations phone_fops = { .owner = THIS_MODULE, .open = phone_open, + .llseek = noop_llseek, }; /* diff --git a/drivers/uio/uio.c b/drivers/uio/uio.c index bff1afbde5a..4d3a6fd1a15 100644 --- a/drivers/uio/uio.c +++ b/drivers/uio/uio.c @@ -740,6 +740,7 @@ static const struct file_operations uio_fops = { .mmap = uio_mmap, .poll = uio_poll, .fasync = uio_fasync, + .llseek = noop_llseek, }; static int uio_major_init(void) diff --git a/drivers/usb/class/cdc-wdm.c b/drivers/usb/class/cdc-wdm.c index 094c76b5de1..6ee4451bfe2 100644 --- a/drivers/usb/class/cdc-wdm.c +++ b/drivers/usb/class/cdc-wdm.c @@ -584,7 +584,8 @@ static const struct file_operations wdm_fops = { .open = wdm_open, .flush = wdm_flush, .release = wdm_release, - .poll = wdm_poll + .poll = wdm_poll, + .llseek = noop_llseek, }; static struct usb_class_driver wdm_class = { diff --git a/drivers/usb/class/usblp.c b/drivers/usb/class/usblp.c index e325162859b..9eca4053312 100644 --- a/drivers/usb/class/usblp.c +++ b/drivers/usb/class/usblp.c @@ -1043,6 +1043,7 @@ static const struct file_operations usblp_fops = { .compat_ioctl = usblp_ioctl, .open = usblp_open, .release = usblp_release, + .llseek = noop_llseek, }; static char *usblp_devnode(struct device *dev, mode_t *mode) diff --git a/drivers/usb/class/usbtmc.c b/drivers/usb/class/usbtmc.c index 3e7c1b800eb..6a54634ab82 100644 --- a/drivers/usb/class/usbtmc.c +++ b/drivers/usb/class/usbtmc.c @@ -987,6 +987,7 @@ static const struct file_operations fops = { .open = usbtmc_open, .release = usbtmc_release, .unlocked_ioctl = usbtmc_ioctl, + .llseek = default_llseek, }; static struct usb_class_driver usbtmc_class = { diff --git a/drivers/usb/core/file.c b/drivers/usb/core/file.c index f06f5dbc8cd..580bcd39683 100644 --- a/drivers/usb/core/file.c +++ b/drivers/usb/core/file.c @@ -59,6 +59,7 @@ static int usb_open(struct inode * inode, struct file * file) static const struct file_operations usb_fops = { .owner = THIS_MODULE, .open = usb_open, + .llseek = noop_llseek, }; static struct usb_class { diff --git a/drivers/usb/gadget/f_hid.c b/drivers/usb/gadget/f_hid.c index 53e120208e9..2b98bd26364 100644 --- a/drivers/usb/gadget/f_hid.c +++ b/drivers/usb/gadget/f_hid.c @@ -451,6 +451,7 @@ const struct file_operations f_hidg_fops = { .write = f_hidg_write, .read = f_hidg_read, .poll = f_hidg_poll, + .llseek = noop_llseek, }; static int __init hidg_bind(struct usb_configuration *c, struct usb_function *f) diff --git a/drivers/usb/gadget/printer.c b/drivers/usb/gadget/printer.c index cf241c371a7..327a92a137b 100644 --- a/drivers/usb/gadget/printer.c +++ b/drivers/usb/gadget/printer.c @@ -884,7 +884,8 @@ static const struct file_operations printer_io_operations = { .fsync = printer_fsync, .poll = printer_poll, .unlocked_ioctl = printer_ioctl, - .release = printer_close + .release = printer_close, + .llseek = noop_llseek, }; /*-------------------------------------------------------------------------*/ diff --git a/drivers/usb/host/ehci-dbg.c b/drivers/usb/host/ehci-dbg.c index 76b7fd2d838..86afdc73322 100644 --- a/drivers/usb/host/ehci-dbg.c +++ b/drivers/usb/host/ehci-dbg.c @@ -369,18 +369,21 @@ static const struct file_operations debug_async_fops = { .open = debug_async_open, .read = debug_output, .release = debug_close, + .llseek = default_llseek, }; static const struct file_operations debug_periodic_fops = { .owner = THIS_MODULE, .open = debug_periodic_open, .read = debug_output, .release = debug_close, + .llseek = default_llseek, }; static const struct file_operations debug_registers_fops = { .owner = THIS_MODULE, .open = debug_registers_open, .read = debug_output, .release = debug_close, + .llseek = default_llseek, }; static const struct file_operations debug_lpm_fops = { .owner = THIS_MODULE, @@ -388,6 +391,7 @@ static const struct file_operations debug_lpm_fops = { .read = debug_lpm_read, .write = debug_lpm_write, .release = debug_lpm_close, + .llseek = noop_llseek, }; static struct dentry *ehci_debug_root; diff --git a/drivers/usb/host/ohci-dbg.c b/drivers/usb/host/ohci-dbg.c index 36abd2baa3e..d7d34492934 100644 --- a/drivers/usb/host/ohci-dbg.c +++ b/drivers/usb/host/ohci-dbg.c @@ -413,18 +413,21 @@ static const struct file_operations debug_async_fops = { .open = debug_async_open, .read = debug_output, .release = debug_close, + .llseek = default_llseek, }; static const struct file_operations debug_periodic_fops = { .owner = THIS_MODULE, .open = debug_periodic_open, .read = debug_output, .release = debug_close, + .llseek = default_llseek, }; static const struct file_operations debug_registers_fops = { .owner = THIS_MODULE, .open = debug_registers_open, .read = debug_output, .release = debug_close, + .llseek = default_llseek, }; static struct dentry *ohci_debug_root; diff --git a/drivers/usb/image/mdc800.c b/drivers/usb/image/mdc800.c index e192e8f7c56..575b56c79e9 100644 --- a/drivers/usb/image/mdc800.c +++ b/drivers/usb/image/mdc800.c @@ -963,6 +963,7 @@ static const struct file_operations mdc800_device_ops = .write = mdc800_device_write, .open = mdc800_device_open, .release = mdc800_device_release, + .llseek = noop_llseek, }; diff --git a/drivers/usb/misc/adutux.c b/drivers/usb/misc/adutux.c index 801324af947..44f8b922505 100644 --- a/drivers/usb/misc/adutux.c +++ b/drivers/usb/misc/adutux.c @@ -679,6 +679,7 @@ static const struct file_operations adu_fops = { .write = adu_write, .open = adu_open, .release = adu_release, + .llseek = noop_llseek, }; /* diff --git a/drivers/usb/misc/idmouse.c b/drivers/usb/misc/idmouse.c index a54c3cb804c..c6184b4d169 100644 --- a/drivers/usb/misc/idmouse.c +++ b/drivers/usb/misc/idmouse.c @@ -105,6 +105,7 @@ static const struct file_operations idmouse_fops = { .read = idmouse_read, .open = idmouse_open, .release = idmouse_release, + .llseek = default_llseek, }; /* class driver information */ diff --git a/drivers/usb/misc/iowarrior.c b/drivers/usb/misc/iowarrior.c index bc88c79875a..9b50db25701 100644 --- a/drivers/usb/misc/iowarrior.c +++ b/drivers/usb/misc/iowarrior.c @@ -730,6 +730,7 @@ static const struct file_operations iowarrior_fops = { .open = iowarrior_open, .release = iowarrior_release, .poll = iowarrior_poll, + .llseek = noop_llseek, }; static char *iowarrior_devnode(struct device *dev, mode_t *mode) diff --git a/drivers/usb/misc/ldusb.c b/drivers/usb/misc/ldusb.c index dd41d871004..edffef64233 100644 --- a/drivers/usb/misc/ldusb.c +++ b/drivers/usb/misc/ldusb.c @@ -613,6 +613,7 @@ static const struct file_operations ld_usb_fops = { .open = ld_usb_open, .release = ld_usb_release, .poll = ld_usb_poll, + .llseek = no_llseek, }; /* diff --git a/drivers/usb/misc/rio500.c b/drivers/usb/misc/rio500.c index cc13ae61712..4e23d3841b4 100644 --- a/drivers/usb/misc/rio500.c +++ b/drivers/usb/misc/rio500.c @@ -439,6 +439,7 @@ static const struct file_operations usb_rio_fops = { .unlocked_ioctl = ioctl_rio, .open = open_rio, .release = close_rio, + .llseek = noop_llseek, }; static struct usb_class_driver usb_rio_class = { diff --git a/drivers/usb/misc/usblcd.c b/drivers/usb/misc/usblcd.c index d00dde19194..51648154bb4 100644 --- a/drivers/usb/misc/usblcd.c +++ b/drivers/usb/misc/usblcd.c @@ -282,6 +282,7 @@ static const struct file_operations lcd_fops = { .open = lcd_open, .unlocked_ioctl = lcd_ioctl, .release = lcd_release, + .llseek = noop_llseek, }; /* diff --git a/drivers/usb/usb-skeleton.c b/drivers/usb/usb-skeleton.c index 552679b8dbd..e24ce312307 100644 --- a/drivers/usb/usb-skeleton.c +++ b/drivers/usb/usb-skeleton.c @@ -507,6 +507,7 @@ static const struct file_operations skel_fops = { .open = skel_open, .release = skel_release, .flush = skel_flush, + .llseek = noop_llseek, }; /* diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c index 29e850a7a2f..c8523ce2e4c 100644 --- a/drivers/vhost/net.c +++ b/drivers/vhost/net.c @@ -869,6 +869,7 @@ static const struct file_operations vhost_net_fops = { .compat_ioctl = vhost_net_compat_ioctl, #endif .open = vhost_net_open, + .llseek = noop_llseek, }; static struct miscdevice vhost_net_misc = { diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c index b06647517c0..42e303ff862 100644 --- a/drivers/video/fbmem.c +++ b/drivers/video/fbmem.c @@ -1439,6 +1439,7 @@ static const struct file_operations fb_fops = { #ifdef CONFIG_FB_DEFERRED_IO .fsync = fb_deferred_io_fsync, #endif + .llseek = default_llseek, }; struct class *fb_class; diff --git a/drivers/video/mbx/mbxdebugfs.c b/drivers/video/mbx/mbxdebugfs.c index ecad9652457..12dec7634c5 100644 --- a/drivers/video/mbx/mbxdebugfs.c +++ b/drivers/video/mbx/mbxdebugfs.c @@ -175,36 +175,42 @@ static const struct file_operations sysconf_fops = { .read = sysconf_read_file, .write = write_file_dummy, .open = open_file_generic, + .llseek = default_llseek, }; static const struct file_operations clock_fops = { .read = clock_read_file, .write = write_file_dummy, .open = open_file_generic, + .llseek = default_llseek, }; static const struct file_operations display_fops = { .read = display_read_file, .write = write_file_dummy, .open = open_file_generic, + .llseek = default_llseek, }; static const struct file_operations gsctl_fops = { .read = gsctl_read_file, .write = write_file_dummy, .open = open_file_generic, + .llseek = default_llseek, }; static const struct file_operations sdram_fops = { .read = sdram_read_file, .write = write_file_dummy, .open = open_file_generic, + .llseek = default_llseek, }; static const struct file_operations misc_fops = { .read = misc_read_file, .write = write_file_dummy, .open = open_file_generic, + .llseek = default_llseek, }; static void __devinit mbxfb_debugfs_init(struct fb_info *fbi) diff --git a/drivers/watchdog/ar7_wdt.c b/drivers/watchdog/ar7_wdt.c index c764c52412e..b2922178359 100644 --- a/drivers/watchdog/ar7_wdt.c +++ b/drivers/watchdog/ar7_wdt.c @@ -267,6 +267,7 @@ static const struct file_operations ar7_wdt_fops = { .unlocked_ioctl = ar7_wdt_ioctl, .open = ar7_wdt_open, .release = ar7_wdt_release, + .llseek = no_llseek, }; static struct miscdevice ar7_wdt_miscdev = { diff --git a/drivers/watchdog/cpwd.c b/drivers/watchdog/cpwd.c index 566343b3c13..51ae4e84f4f 100644 --- a/drivers/watchdog/cpwd.c +++ b/drivers/watchdog/cpwd.c @@ -524,6 +524,7 @@ static const struct file_operations cpwd_fops = { .write = cpwd_write, .read = cpwd_read, .release = cpwd_release, + .llseek = no_llseek, }; static int __devinit cpwd_probe(struct platform_device *op, diff --git a/drivers/watchdog/ep93xx_wdt.c b/drivers/watchdog/ep93xx_wdt.c index 59359c9a5e0..726b7df61fd 100644 --- a/drivers/watchdog/ep93xx_wdt.c +++ b/drivers/watchdog/ep93xx_wdt.c @@ -188,6 +188,7 @@ static const struct file_operations ep93xx_wdt_fops = { .unlocked_ioctl = ep93xx_wdt_ioctl, .open = ep93xx_wdt_open, .release = ep93xx_wdt_release, + .llseek = no_llseek, }; static struct miscdevice ep93xx_wdt_miscdev = { diff --git a/drivers/watchdog/omap_wdt.c b/drivers/watchdog/omap_wdt.c index 76b58abf445..81e3d610089 100644 --- a/drivers/watchdog/omap_wdt.c +++ b/drivers/watchdog/omap_wdt.c @@ -258,6 +258,7 @@ static const struct file_operations omap_wdt_fops = { .unlocked_ioctl = omap_wdt_ioctl, .open = omap_wdt_open, .release = omap_wdt_release, + .llseek = no_llseek, }; static int __devinit omap_wdt_probe(struct platform_device *pdev) diff --git a/drivers/xen/evtchn.c b/drivers/xen/evtchn.c index 66e185cfe92..fec6ba3c08a 100644 --- a/drivers/xen/evtchn.c +++ b/drivers/xen/evtchn.c @@ -467,6 +467,7 @@ static const struct file_operations evtchn_fops = { .fasync = evtchn_fasync, .open = evtchn_open, .release = evtchn_release, + .llseek = noop_llseek, }; static struct miscdevice evtchn_miscdev = { diff --git a/drivers/xen/xenfs/super.c b/drivers/xen/xenfs/super.c index 78bfab0700b..bd96340063c 100644 --- a/drivers/xen/xenfs/super.c +++ b/drivers/xen/xenfs/super.c @@ -35,6 +35,7 @@ static ssize_t capabilities_read(struct file *file, char __user *buf, static const struct file_operations capabilities_file_ops = { .read = capabilities_read, + .llseek = default_llseek, }; static int xenfs_fill_super(struct super_block *sb, void *data, int silent) diff --git a/drivers/xen/xenfs/xenbus.c b/drivers/xen/xenfs/xenbus.c index 3b39c3752e2..1c1236087f7 100644 --- a/drivers/xen/xenfs/xenbus.c +++ b/drivers/xen/xenfs/xenbus.c @@ -594,4 +594,5 @@ const struct file_operations xenbus_file_ops = { .open = xenbus_file_open, .release = xenbus_file_release, .poll = xenbus_file_poll, + .llseek = no_llseek, }; diff --git a/fs/afs/mntpt.c b/fs/afs/mntpt.c index 6d552686c49..6153417caf5 100644 --- a/fs/afs/mntpt.c +++ b/fs/afs/mntpt.c @@ -29,6 +29,7 @@ static void afs_mntpt_expiry_timed_out(struct work_struct *work); const struct file_operations afs_mntpt_file_operations = { .open = afs_mntpt_open, + .llseek = noop_llseek, }; const struct inode_operations afs_mntpt_inode_operations = { diff --git a/fs/autofs4/dev-ioctl.c b/fs/autofs4/dev-ioctl.c index ba4a38b9c22..eff9a419469 100644 --- a/fs/autofs4/dev-ioctl.c +++ b/fs/autofs4/dev-ioctl.c @@ -724,6 +724,7 @@ static const struct file_operations _dev_ioctl_fops = { .unlocked_ioctl = autofs_dev_ioctl, .compat_ioctl = autofs_dev_ioctl_compat, .owner = THIS_MODULE, + .llseek = noop_llseek, }; static struct miscdevice _autofs_dev_ioctl_misc = { diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index fd0cc0bf9a4..139fc8083f5 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -576,6 +576,7 @@ static ssize_t bm_entry_write(struct file *file, const char __user *buffer, static const struct file_operations bm_entry_operations = { .read = bm_entry_read, .write = bm_entry_write, + .llseek = default_llseek, }; /* /register */ @@ -643,6 +644,7 @@ out: static const struct file_operations bm_register_operations = { .write = bm_register_write, + .llseek = noop_llseek, }; /* /status */ @@ -680,6 +682,7 @@ static ssize_t bm_status_write(struct file * file, const char __user * buffer, static const struct file_operations bm_status_operations = { .read = bm_status_read, .write = bm_status_write, + .llseek = default_llseek, }; /* Superblock handling */ diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 1776dbd8dc9..144f8a5730f 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -815,6 +815,7 @@ static const struct file_operations btrfs_ctl_fops = { .unlocked_ioctl = btrfs_control_ioctl, .compat_ioctl = btrfs_control_ioctl, .owner = THIS_MODULE, + .llseek = noop_llseek, }; static struct miscdevice btrfs_misc = { diff --git a/fs/cachefiles/daemon.c b/fs/cachefiles/daemon.c index 727caedcdd9..0a1467b1551 100644 --- a/fs/cachefiles/daemon.c +++ b/fs/cachefiles/daemon.c @@ -55,6 +55,7 @@ const struct file_operations cachefiles_daemon_fops = { .read = cachefiles_daemon_read, .write = cachefiles_daemon_write, .poll = cachefiles_daemon_poll, + .llseek = noop_llseek, }; struct cachefiles_daemon_cmd { diff --git a/fs/char_dev.c b/fs/char_dev.c index f80a4f25123..e4a3318b812 100644 --- a/fs/char_dev.c +++ b/fs/char_dev.c @@ -454,6 +454,7 @@ static void cdev_purge(struct cdev *cdev) */ const struct file_operations def_chr_fops = { .open = chrdev_open, + .llseek = noop_llseek, }; static struct kobject *exact_match(dev_t dev, int *part, void *data) diff --git a/fs/coda/pioctl.c b/fs/coda/pioctl.c index ca25d96d45c..028a9a0f588 100644 --- a/fs/coda/pioctl.c +++ b/fs/coda/pioctl.c @@ -39,6 +39,7 @@ const struct inode_operations coda_ioctl_inode_operations = { const struct file_operations coda_ioctl_operations = { .owner = THIS_MODULE, .unlocked_ioctl = coda_pioctl, + .llseek = noop_llseek, }; /* the coda pioctl inode ops */ diff --git a/fs/coda/psdev.c b/fs/coda/psdev.c index de89645777c..9fa280bfcff 100644 --- a/fs/coda/psdev.c +++ b/fs/coda/psdev.c @@ -346,6 +346,7 @@ static const struct file_operations coda_psdev_fops = { .unlocked_ioctl = coda_psdev_ioctl, .open = coda_psdev_open, .release = coda_psdev_release, + .llseek = noop_llseek, }; static int init_coda_psdev(void) diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c index 0210898458b..89d394d8fe2 100644 --- a/fs/debugfs/file.c +++ b/fs/debugfs/file.c @@ -43,6 +43,7 @@ const struct file_operations debugfs_file_operations = { .read = default_read_file, .write = default_write_file, .open = default_open, + .llseek = noop_llseek, }; static void *debugfs_follow_link(struct dentry *dentry, struct nameidata *nd) @@ -454,6 +455,7 @@ static const struct file_operations fops_bool = { .read = read_file_bool, .write = write_file_bool, .open = default_open, + .llseek = default_llseek, }; /** @@ -498,6 +500,7 @@ static ssize_t read_file_blob(struct file *file, char __user *user_buf, static const struct file_operations fops_blob = { .read = read_file_blob, .open = default_open, + .llseek = default_llseek, }; /** diff --git a/fs/dlm/debug_fs.c b/fs/dlm/debug_fs.c index c6cf2515874..6b42ba807df 100644 --- a/fs/dlm/debug_fs.c +++ b/fs/dlm/debug_fs.c @@ -643,7 +643,8 @@ static ssize_t waiters_read(struct file *file, char __user *userbuf, static const struct file_operations waiters_fops = { .owner = THIS_MODULE, .open = waiters_open, - .read = waiters_read + .read = waiters_read, + .llseek = default_llseek, }; void dlm_delete_debug_file(struct dlm_ls *ls) diff --git a/fs/dlm/plock.c b/fs/dlm/plock.c index d45c02db694..30d8b85febb 100644 --- a/fs/dlm/plock.c +++ b/fs/dlm/plock.c @@ -412,7 +412,8 @@ static const struct file_operations dev_fops = { .read = dev_read, .write = dev_write, .poll = dev_poll, - .owner = THIS_MODULE + .owner = THIS_MODULE, + .llseek = noop_llseek, }; static struct miscdevice plock_dev_misc = { diff --git a/fs/dlm/user.c b/fs/dlm/user.c index b6272853130..66d6c16bf44 100644 --- a/fs/dlm/user.c +++ b/fs/dlm/user.c @@ -1009,6 +1009,7 @@ static const struct file_operations device_fops = { .write = device_write, .poll = device_poll, .owner = THIS_MODULE, + .llseek = noop_llseek, }; static const struct file_operations ctl_device_fops = { @@ -1017,6 +1018,7 @@ static const struct file_operations ctl_device_fops = { .read = device_read, .write = device_write, .owner = THIS_MODULE, + .llseek = noop_llseek, }; static struct miscdevice ctl_device = { @@ -1029,6 +1031,7 @@ static const struct file_operations monitor_device_fops = { .open = monitor_device_open, .release = monitor_device_close, .owner = THIS_MODULE, + .llseek = noop_llseek, }; static struct miscdevice monitor_device = { diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index 622c9514080..86e9da1c787 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -332,6 +332,7 @@ const struct file_operations ecryptfs_dir_fops = { .fsync = ecryptfs_fsync, .fasync = ecryptfs_fasync, .splice_read = generic_file_splice_read, + .llseek = default_llseek, }; const struct file_operations ecryptfs_main_fops = { diff --git a/fs/ecryptfs/miscdev.c b/fs/ecryptfs/miscdev.c index 00208c3d7e9..940a82e63dc 100644 --- a/fs/ecryptfs/miscdev.c +++ b/fs/ecryptfs/miscdev.c @@ -482,6 +482,7 @@ static const struct file_operations ecryptfs_miscdev_fops = { .read = ecryptfs_miscdev_read, .write = ecryptfs_miscdev_write, .release = ecryptfs_miscdev_release, + .llseek = noop_llseek, }; static struct miscdevice ecryptfs_miscdev = { diff --git a/fs/eventfd.c b/fs/eventfd.c index 6bd3f76fdf8..e0194b3e14d 100644 --- a/fs/eventfd.c +++ b/fs/eventfd.c @@ -293,6 +293,7 @@ static const struct file_operations eventfd_fops = { .poll = eventfd_poll, .read = eventfd_read, .write = eventfd_write, + .llseek = noop_llseek, }; /** diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 3817149919c..256bb7bb102 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -674,7 +674,8 @@ static unsigned int ep_eventpoll_poll(struct file *file, poll_table *wait) /* File callbacks that implement the eventpoll file behaviour */ static const struct file_operations eventpoll_fops = { .release = ep_eventpoll_release, - .poll = ep_eventpoll_poll + .poll = ep_eventpoll_poll, + .llseek = noop_llseek, }; /* Fast test to see if the file is an evenpoll file */ diff --git a/fs/fifo.c b/fs/fifo.c index 5d6606ffc2d..4e303c22d5e 100644 --- a/fs/fifo.c +++ b/fs/fifo.c @@ -151,4 +151,5 @@ err_nocleanup: */ const struct file_operations def_fifo_fops = { .open = fifo_open, /* will set read_ or write_pipefifo_fops */ + .llseek = noop_llseek, }; diff --git a/fs/fuse/control.c b/fs/fuse/control.c index 3773fd63d2f..7367e177186 100644 --- a/fs/fuse/control.c +++ b/fs/fuse/control.c @@ -179,23 +179,27 @@ static ssize_t fuse_conn_congestion_threshold_write(struct file *file, static const struct file_operations fuse_ctl_abort_ops = { .open = nonseekable_open, .write = fuse_conn_abort_write, + .llseek = no_llseek, }; static const struct file_operations fuse_ctl_waiting_ops = { .open = nonseekable_open, .read = fuse_conn_waiting_read, + .llseek = no_llseek, }; static const struct file_operations fuse_conn_max_background_ops = { .open = nonseekable_open, .read = fuse_conn_max_background_read, .write = fuse_conn_max_background_write, + .llseek = no_llseek, }; static const struct file_operations fuse_conn_congestion_threshold_ops = { .open = nonseekable_open, .read = fuse_conn_congestion_threshold_read, .write = fuse_conn_congestion_threshold_write, + .llseek = no_llseek, }; static struct dentry *fuse_ctl_add_dentry(struct dentry *parent, diff --git a/fs/fuse/cuse.c b/fs/fuse/cuse.c index e1f8171278b..3e87cce5837 100644 --- a/fs/fuse/cuse.c +++ b/fs/fuse/cuse.c @@ -182,6 +182,7 @@ static const struct file_operations cuse_frontend_fops = { .unlocked_ioctl = cuse_file_ioctl, .compat_ioctl = cuse_file_compat_ioctl, .poll = fuse_file_poll, + .llseek = noop_llseek, }; diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index 4edd662c823..55d25a68b49 100644 --- a/fs/gfs2/file.c +++ b/fs/gfs2/file.c @@ -771,6 +771,7 @@ const struct file_operations gfs2_dir_fops = { .fsync = gfs2_fsync, .lock = gfs2_lock, .flock = gfs2_flock, + .llseek = default_llseek, }; #endif /* CONFIG_GFS2_FS_LOCKING_DLM */ @@ -797,5 +798,6 @@ const struct file_operations gfs2_dir_fops_nolock = { .open = gfs2_open, .release = gfs2_close, .fsync = gfs2_fsync, + .llseek = default_llseek, }; diff --git a/fs/hppfs/hppfs.c b/fs/hppfs/hppfs.c index 7b027720d82..4e2a45ea614 100644 --- a/fs/hppfs/hppfs.c +++ b/fs/hppfs/hppfs.c @@ -598,6 +598,7 @@ static const struct file_operations hppfs_dir_fops = { .readdir = hppfs_readdir, .open = hppfs_dir_open, .fsync = hppfs_fsync, + .llseek = default_llseek, }; static int hppfs_statfs(struct dentry *dentry, struct kstatfs *sf) diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 6e5bd42f386..113eba3d3c3 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -674,6 +674,7 @@ const struct file_operations hugetlbfs_file_operations = { .mmap = hugetlbfs_file_mmap, .fsync = noop_fsync, .get_unmapped_area = hugetlb_get_unmapped_area, + .llseek = default_llseek, }; static const struct inode_operations hugetlbfs_dir_inode_operations = { diff --git a/fs/logfs/dir.c b/fs/logfs/dir.c index 9777eb5b552..1eb4e89e045 100644 --- a/fs/logfs/dir.c +++ b/fs/logfs/dir.c @@ -827,4 +827,5 @@ const struct file_operations logfs_dir_fops = { .unlocked_ioctl = logfs_ioctl, .readdir = logfs_readdir, .read = generic_read_dir, + .llseek = default_llseek, }; diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index b53b1d042f1..06fa87e52e8 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -137,6 +137,7 @@ static const struct file_operations transaction_ops = { .write = nfsctl_transaction_write, .read = nfsctl_transaction_read, .release = simple_transaction_release, + .llseek = default_llseek, }; static int exports_open(struct inode *inode, struct file *file) diff --git a/fs/no-block.c b/fs/no-block.c index d269a93d346..6e40e42a43d 100644 --- a/fs/no-block.c +++ b/fs/no-block.c @@ -19,4 +19,5 @@ static int no_blkdev_open(struct inode * inode, struct file * filp) const struct file_operations def_blk_fops = { .open = no_blkdev_open, + .llseek = noop_llseek, }; diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c index 5ed8e58d7bf..bbcb98e7fcc 100644 --- a/fs/notify/fanotify/fanotify_user.c +++ b/fs/notify/fanotify/fanotify_user.c @@ -433,6 +433,7 @@ static const struct file_operations fanotify_fops = { .release = fanotify_release, .unlocked_ioctl = fanotify_ioctl, .compat_ioctl = fanotify_ioctl, + .llseek = noop_llseek, }; static void fanotify_free_mark(struct fsnotify_mark *fsn_mark) diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c index bf7f6d776c3..24edc1185d5 100644 --- a/fs/notify/inotify/inotify_user.c +++ b/fs/notify/inotify/inotify_user.c @@ -344,6 +344,7 @@ static const struct file_operations inotify_fops = { .release = inotify_release, .unlocked_ioctl = inotify_ioctl, .compat_ioctl = inotify_ioctl, + .llseek = noop_llseek, }; diff --git a/fs/ocfs2/dlmfs/dlmfs.c b/fs/ocfs2/dlmfs/dlmfs.c index c2903b84bb7..a7ebd9d42dc 100644 --- a/fs/ocfs2/dlmfs/dlmfs.c +++ b/fs/ocfs2/dlmfs/dlmfs.c @@ -612,6 +612,7 @@ static const struct file_operations dlmfs_file_operations = { .poll = dlmfs_file_poll, .read = dlmfs_file_read, .write = dlmfs_file_write, + .llseek = default_llseek, }; static const struct inode_operations dlmfs_dir_inode_operations = { diff --git a/fs/ocfs2/stack_user.c b/fs/ocfs2/stack_user.c index 2dc57bca068..0dbc6dae1de 100644 --- a/fs/ocfs2/stack_user.c +++ b/fs/ocfs2/stack_user.c @@ -628,6 +628,7 @@ static const struct file_operations ocfs2_control_fops = { .read = ocfs2_control_read, .write = ocfs2_control_write, .owner = THIS_MODULE, + .llseek = default_llseek, }; static struct miscdevice ocfs2_control_device = { diff --git a/fs/proc/base.c b/fs/proc/base.c index a1c43e7c8a7..bc307d7a5b7 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1151,6 +1151,7 @@ static ssize_t oom_score_adj_write(struct file *file, const char __user *buf, static const struct file_operations proc_oom_score_adj_operations = { .read = oom_score_adj_read, .write = oom_score_adj_write, + .llseek = default_llseek, }; #ifdef CONFIG_AUDITSYSCALL @@ -2039,11 +2040,13 @@ static ssize_t proc_fdinfo_read(struct file *file, char __user *buf, static const struct file_operations proc_fdinfo_file_operations = { .open = nonseekable_open, .read = proc_fdinfo_read, + .llseek = no_llseek, }; static const struct file_operations proc_fd_operations = { .read = generic_read_dir, .readdir = proc_readfd, + .llseek = default_llseek, }; /* @@ -2112,6 +2115,7 @@ static int proc_readfdinfo(struct file *filp, void *dirent, filldir_t filldir) static const struct file_operations proc_fdinfo_operations = { .read = generic_read_dir, .readdir = proc_readfdinfo, + .llseek = default_llseek, }; /* @@ -2343,6 +2347,7 @@ static int proc_attr_dir_readdir(struct file * filp, static const struct file_operations proc_attr_dir_operations = { .read = generic_read_dir, .readdir = proc_attr_dir_readdir, + .llseek = default_llseek, }; static struct dentry *proc_attr_dir_lookup(struct inode *dir, @@ -2751,6 +2756,7 @@ static int proc_tgid_base_readdir(struct file * filp, static const struct file_operations proc_tgid_base_operations = { .read = generic_read_dir, .readdir = proc_tgid_base_readdir, + .llseek = default_llseek, }; static struct dentry *proc_tgid_base_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd){ @@ -3088,6 +3094,7 @@ static struct dentry *proc_tid_base_lookup(struct inode *dir, struct dentry *den static const struct file_operations proc_tid_base_operations = { .read = generic_read_dir, .readdir = proc_tid_base_readdir, + .llseek = default_llseek, }; static const struct inode_operations proc_tid_base_inode_operations = { @@ -3324,4 +3331,5 @@ static const struct inode_operations proc_task_inode_operations = { static const struct file_operations proc_task_operations = { .read = generic_read_dir, .readdir = proc_task_readdir, + .llseek = default_llseek, }; diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index 5be436ea088..2fc52552271 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -364,6 +364,7 @@ static int proc_sys_getattr(struct vfsmount *mnt, struct dentry *dentry, struct static const struct file_operations proc_sys_file_operations = { .read = proc_sys_read, .write = proc_sys_write, + .llseek = default_llseek, }; static const struct file_operations proc_sys_dir_file_operations = { diff --git a/fs/proc/root.c b/fs/proc/root.c index 4258384ed22..93d99b31632 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -179,6 +179,7 @@ static int proc_root_readdir(struct file * filp, static const struct file_operations proc_root_operations = { .read = generic_read_dir, .readdir = proc_root_readdir, + .llseek = default_llseek, }; /* diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 271afc48b9a..aa56920df33 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -539,6 +539,7 @@ static ssize_t clear_refs_write(struct file *file, const char __user *buf, const struct file_operations proc_clear_refs_operations = { .write = clear_refs_write, + .llseek = noop_llseek, }; struct pagemapread { diff --git a/fs/romfs/super.c b/fs/romfs/super.c index 42d21354689..268580535c9 100644 --- a/fs/romfs/super.c +++ b/fs/romfs/super.c @@ -282,6 +282,7 @@ error: static const struct file_operations romfs_dir_operations = { .read = generic_read_dir, .readdir = romfs_readdir, + .llseek = default_llseek, }; static const struct inode_operations romfs_dir_inode_operations = { diff --git a/fs/signalfd.c b/fs/signalfd.c index 1c5a6add779..74047304b01 100644 --- a/fs/signalfd.c +++ b/fs/signalfd.c @@ -206,6 +206,7 @@ static const struct file_operations signalfd_fops = { .release = signalfd_release, .poll = signalfd_poll, .read = signalfd_read, + .llseek = noop_llseek, }; SYSCALL_DEFINE4(signalfd4, int, ufd, sigset_t __user *, user_mask, diff --git a/fs/squashfs/dir.c b/fs/squashfs/dir.c index 12b933ac658..0dc340aa2be 100644 --- a/fs/squashfs/dir.c +++ b/fs/squashfs/dir.c @@ -230,5 +230,6 @@ failed_read: const struct file_operations squashfs_dir_ops = { .read = generic_read_dir, - .readdir = squashfs_readdir + .readdir = squashfs_readdir, + .llseek = default_llseek, }; diff --git a/fs/timerfd.c b/fs/timerfd.c index b86ab8eff79..8c4fc1425b3 100644 --- a/fs/timerfd.c +++ b/fs/timerfd.c @@ -144,6 +144,7 @@ static const struct file_operations timerfd_fops = { .release = timerfd_release, .poll = timerfd_poll, .read = timerfd_read, + .llseek = noop_llseek, }; static struct file *timerfd_fget(int fd) diff --git a/fs/ubifs/debug.c b/fs/ubifs/debug.c index c2a68baa782..c6c553fd0b3 100644 --- a/fs/ubifs/debug.c +++ b/fs/ubifs/debug.c @@ -2625,6 +2625,7 @@ static const struct file_operations dfs_fops = { .open = open_debugfs_file, .write = write_debugfs_file, .owner = THIS_MODULE, + .llseek = default_llseek, }; /** diff --git a/ipc/mqueue.c b/ipc/mqueue.c index c60e519e291..e1e7b9635f5 100644 --- a/ipc/mqueue.c +++ b/ipc/mqueue.c @@ -1219,6 +1219,7 @@ static const struct file_operations mqueue_file_operations = { .flush = mqueue_flush_file, .poll = mqueue_poll_file, .read = mqueue_read_file, + .llseek = default_llseek, }; static const struct super_operations mqueue_super_ops = { diff --git a/ipc/shm.c b/ipc/shm.c index 52ed77eb971..7bc46a9fe1f 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -298,6 +298,7 @@ static const struct file_operations shm_file_operations = { #ifndef CONFIG_MMU .get_unmapped_area = shm_get_unmapped_area, #endif + .llseek = noop_llseek, }; static const struct file_operations shm_file_operations_huge = { @@ -305,6 +306,7 @@ static const struct file_operations shm_file_operations_huge = { .fsync = shm_fsync, .release = shm_release, .get_unmapped_area = shm_get_unmapped_area, + .llseek = noop_llseek, }; int is_file_shm_hugepages(struct file *file) diff --git a/kernel/configs.c b/kernel/configs.c index abaee684ecb..b4066b44a99 100644 --- a/kernel/configs.c +++ b/kernel/configs.c @@ -66,6 +66,7 @@ ikconfig_read_current(struct file *file, char __user *buf, static const struct file_operations ikconfig_file_ops = { .owner = THIS_MODULE, .read = ikconfig_read_current, + .llseek = default_llseek, }; static int __init ikconfig_init(void) diff --git a/kernel/gcov/fs.c b/kernel/gcov/fs.c index f83972b1656..9bd0934f6c3 100644 --- a/kernel/gcov/fs.c +++ b/kernel/gcov/fs.c @@ -561,6 +561,7 @@ static ssize_t reset_read(struct file *file, char __user *addr, size_t len, static const struct file_operations gcov_reset_fops = { .write = reset_write, .read = reset_read, + .llseek = noop_llseek, }; /* diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 282035f3ae9..8b5ff2655ae 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -1992,6 +1992,7 @@ static ssize_t write_enabled_file_bool(struct file *file, static const struct file_operations fops_kp = { .read = read_enabled_file_bool, .write = write_enabled_file_bool, + .llseek = default_llseek, }; static int __kprobes debugfs_kprobe_init(void) diff --git a/kernel/pm_qos_params.c b/kernel/pm_qos_params.c index 645e541a45f..a96b850ba08 100644 --- a/kernel/pm_qos_params.c +++ b/kernel/pm_qos_params.c @@ -110,6 +110,7 @@ static const struct file_operations pm_qos_power_fops = { .write = pm_qos_power_write, .open = pm_qos_power_open, .release = pm_qos_power_release, + .llseek = noop_llseek, }; /* unlocked internal variant */ diff --git a/kernel/profile.c b/kernel/profile.c index b22a899934c..66f841b7fbd 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -555,6 +555,7 @@ static ssize_t write_profile(struct file *file, const char __user *buf, static const struct file_operations proc_profile_operations = { .read = read_profile, .write = write_profile, + .llseek = default_llseek, }; #ifdef CONFIG_SMP diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 959f8d6c8cc..2d5f3a75731 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -326,6 +326,7 @@ static const struct file_operations blk_dropped_fops = { .owner = THIS_MODULE, .open = blk_dropped_open, .read = blk_dropped_read, + .llseek = default_llseek, }; static int blk_msg_open(struct inode *inode, struct file *filp) @@ -365,6 +366,7 @@ static const struct file_operations blk_msg_fops = { .owner = THIS_MODULE, .open = blk_msg_open, .write = blk_msg_write, + .llseek = noop_llseek, }; /* diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index fa7ece649fe..5e1ad476309 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -800,6 +800,7 @@ static const struct file_operations ftrace_profile_fops = { .open = tracing_open_generic, .read = ftrace_profile_read, .write = ftrace_profile_write, + .llseek = default_llseek, }; /* used to initialize the real stat files */ @@ -2632,6 +2633,7 @@ static const struct file_operations ftrace_graph_fops = { .read = seq_read, .write = ftrace_graph_write, .release = ftrace_graph_release, + .llseek = seq_lseek, }; #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 492197e2f86..3aea966d16d 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -3965,6 +3965,7 @@ static const struct file_operations rb_simple_fops = { .open = tracing_open_generic, .read = rb_simple_read, .write = rb_simple_write, + .llseek = default_llseek, }; diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index 4c758f14632..0369c5e0998 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -951,6 +951,7 @@ static const struct file_operations ftrace_enable_fops = { .open = tracing_open_generic, .read = event_enable_read, .write = event_enable_write, + .llseek = default_llseek, }; static const struct file_operations ftrace_event_format_fops = { @@ -963,29 +964,34 @@ static const struct file_operations ftrace_event_format_fops = { static const struct file_operations ftrace_event_id_fops = { .open = tracing_open_generic, .read = event_id_read, + .llseek = default_llseek, }; static const struct file_operations ftrace_event_filter_fops = { .open = tracing_open_generic, .read = event_filter_read, .write = event_filter_write, + .llseek = default_llseek, }; static const struct file_operations ftrace_subsystem_filter_fops = { .open = tracing_open_generic, .read = subsystem_filter_read, .write = subsystem_filter_write, + .llseek = default_llseek, }; static const struct file_operations ftrace_system_enable_fops = { .open = tracing_open_generic, .read = system_enable_read, .write = system_enable_write, + .llseek = default_llseek, }; static const struct file_operations ftrace_show_header_fops = { .open = tracing_open_generic, .read = show_header, + .llseek = default_llseek, }; static struct dentry *event_trace_events_dir(void) diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c index a6b7e0e0f3e..4c5dead0c23 100644 --- a/kernel/trace/trace_stack.c +++ b/kernel/trace/trace_stack.c @@ -195,6 +195,7 @@ static const struct file_operations stack_max_size_fops = { .open = tracing_open_generic, .read = stack_max_size_read, .write = stack_max_size_write, + .llseek = default_llseek, }; static void * diff --git a/lib/dma-debug.c b/lib/dma-debug.c index 01e64270e24..4bfb0471f10 100644 --- a/lib/dma-debug.c +++ b/lib/dma-debug.c @@ -590,6 +590,7 @@ out_unlock: static const struct file_operations filter_fops = { .read = filter_read, .write = filter_write, + .llseek = default_llseek, }; static int dma_debug_fs_init(void) diff --git a/net/atm/proc.c b/net/atm/proc.c index 6262aeae398..f85da0779e5 100644 --- a/net/atm/proc.c +++ b/net/atm/proc.c @@ -38,6 +38,7 @@ static ssize_t proc_dev_atm_read(struct file *file, char __user *buf, static const struct file_operations proc_atm_dev_ops = { .owner = THIS_MODULE, .read = proc_dev_atm_read, + .llseek = noop_llseek, }; static void add_stats(struct seq_file *seq, const char *aal, diff --git a/net/dccp/probe.c b/net/dccp/probe.c index 078e48d442f..33d0e6297c2 100644 --- a/net/dccp/probe.c +++ b/net/dccp/probe.c @@ -149,6 +149,7 @@ static const struct file_operations dccpprobe_fops = { .owner = THIS_MODULE, .open = dccpprobe_open, .read = dccpprobe_read, + .llseek = noop_llseek, }; static __init int dccpprobe_init(void) diff --git a/net/ipv4/tcp_probe.c b/net/ipv4/tcp_probe.c index f8efada580e..6211e211417 100644 --- a/net/ipv4/tcp_probe.c +++ b/net/ipv4/tcp_probe.c @@ -214,6 +214,7 @@ static const struct file_operations tcpprobe_fops = { .owner = THIS_MODULE, .open = tcpprobe_open, .read = tcpprobe_read, + .llseek = noop_llseek, }; static __init int tcpprobe_init(void) diff --git a/net/mac80211/debugfs.c b/net/mac80211/debugfs.c index 4a4d35c750c..b8b0ae79a74 100644 --- a/net/mac80211/debugfs.c +++ b/net/mac80211/debugfs.c @@ -102,7 +102,8 @@ static ssize_t tsf_write(struct file *file, static const struct file_operations tsf_ops = { .read = tsf_read, .write = tsf_write, - .open = mac80211_open_file_generic + .open = mac80211_open_file_generic, + .llseek = default_llseek, }; static ssize_t reset_write(struct file *file, const char __user *user_buf, @@ -121,6 +122,7 @@ static ssize_t reset_write(struct file *file, const char __user *user_buf, static const struct file_operations reset_ops = { .write = reset_write, .open = mac80211_open_file_generic, + .llseek = noop_llseek, }; static ssize_t noack_read(struct file *file, char __user *user_buf, @@ -156,7 +158,8 @@ static ssize_t noack_write(struct file *file, static const struct file_operations noack_ops = { .read = noack_read, .write = noack_write, - .open = mac80211_open_file_generic + .open = mac80211_open_file_generic, + .llseek = default_llseek, }; static ssize_t uapsd_queues_read(struct file *file, char __user *user_buf, @@ -202,7 +205,8 @@ static ssize_t uapsd_queues_write(struct file *file, static const struct file_operations uapsd_queues_ops = { .read = uapsd_queues_read, .write = uapsd_queues_write, - .open = mac80211_open_file_generic + .open = mac80211_open_file_generic, + .llseek = default_llseek, }; static ssize_t uapsd_max_sp_len_read(struct file *file, char __user *user_buf, @@ -248,7 +252,8 @@ static ssize_t uapsd_max_sp_len_write(struct file *file, static const struct file_operations uapsd_max_sp_len_ops = { .read = uapsd_max_sp_len_read, .write = uapsd_max_sp_len_write, - .open = mac80211_open_file_generic + .open = mac80211_open_file_generic, + .llseek = default_llseek, }; static ssize_t channel_type_read(struct file *file, char __user *user_buf, @@ -280,7 +285,8 @@ static ssize_t channel_type_read(struct file *file, char __user *user_buf, static const struct file_operations channel_type_ops = { .read = channel_type_read, - .open = mac80211_open_file_generic + .open = mac80211_open_file_generic, + .llseek = default_llseek, }; static ssize_t queues_read(struct file *file, char __user *user_buf, @@ -303,7 +309,8 @@ static ssize_t queues_read(struct file *file, char __user *user_buf, static const struct file_operations queues_ops = { .read = queues_read, - .open = mac80211_open_file_generic + .open = mac80211_open_file_generic, + .llseek = default_llseek, }; /* statistics stuff */ diff --git a/net/mac80211/rate.c b/net/mac80211/rate.c index be04d46110f..334cbd3d2aa 100644 --- a/net/mac80211/rate.c +++ b/net/mac80211/rate.c @@ -145,6 +145,7 @@ static ssize_t rcname_read(struct file *file, char __user *userbuf, static const struct file_operations rcname_ops = { .read = rcname_read, .open = mac80211_open_file_generic, + .llseek = default_llseek, }; #endif diff --git a/net/mac80211/rc80211_minstrel_debugfs.c b/net/mac80211/rc80211_minstrel_debugfs.c index 241e76f3fdf..a290ad231d7 100644 --- a/net/mac80211/rc80211_minstrel_debugfs.c +++ b/net/mac80211/rc80211_minstrel_debugfs.c @@ -122,6 +122,7 @@ static const struct file_operations minstrel_stat_fops = { .open = minstrel_stats_open, .read = minstrel_stats_read, .release = minstrel_stats_release, + .llseek = default_llseek, }; void diff --git a/net/mac80211/rc80211_pid_debugfs.c b/net/mac80211/rc80211_pid_debugfs.c index 47438b4a9af..7905f79cc2e 100644 --- a/net/mac80211/rc80211_pid_debugfs.c +++ b/net/mac80211/rc80211_pid_debugfs.c @@ -206,6 +206,7 @@ static const struct file_operations rc_pid_fop_events = { .poll = rate_control_pid_events_poll, .open = rate_control_pid_events_open, .release = rate_control_pid_events_release, + .llseek = noop_llseek, }; void rate_control_pid_add_sta_debugfs(void *priv, void *priv_sta, diff --git a/net/netfilter/xt_recent.c b/net/netfilter/xt_recent.c index 76aec6a4476..d2ff15a2412 100644 --- a/net/netfilter/xt_recent.c +++ b/net/netfilter/xt_recent.c @@ -567,6 +567,7 @@ static const struct file_operations recent_mt_fops = { .write = recent_mt_proc_write, .release = seq_release_private, .owner = THIS_MODULE, + .llseek = seq_lseek, }; static int __net_init recent_proc_net_init(struct net *net) diff --git a/net/nonet.c b/net/nonet.c index 92e76640c7c..b1a73fda9c1 100644 --- a/net/nonet.c +++ b/net/nonet.c @@ -22,4 +22,5 @@ static int sock_no_open(struct inode *irrelevant, struct file *dontcare) const struct file_operations bad_sock_fops = { .owner = THIS_MODULE, .open = sock_no_open, + .llseek = noop_llseek, }; diff --git a/net/rfkill/core.c b/net/rfkill/core.c index 51875a0c5d4..04f599089e6 100644 --- a/net/rfkill/core.c +++ b/net/rfkill/core.c @@ -1241,6 +1241,7 @@ static const struct file_operations rfkill_fops = { .unlocked_ioctl = rfkill_fop_ioctl, .compat_ioctl = rfkill_fop_ioctl, #endif + .llseek = no_llseek, }; static struct miscdevice rfkill_miscdev = { diff --git a/net/sctp/probe.c b/net/sctp/probe.c index db3a42b8b34..289b1ba62ca 100644 --- a/net/sctp/probe.c +++ b/net/sctp/probe.c @@ -117,6 +117,7 @@ static const struct file_operations sctpprobe_fops = { .owner = THIS_MODULE, .open = sctpprobe_open, .read = sctpprobe_read, + .llseek = noop_llseek, }; sctp_disposition_t jsctp_sf_eat_sack(const struct sctp_endpoint *ep, diff --git a/net/socket.c b/net/socket.c index 2270b941bcc..9eac5c39413 100644 --- a/net/socket.c +++ b/net/socket.c @@ -502,6 +502,7 @@ static int sock_no_open(struct inode *irrelevant, struct file *dontcare) const struct file_operations bad_sock_fops = { .owner = THIS_MODULE, .open = sock_no_open, + .llseek = noop_llseek, }; /** diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c index 2b06410e584..6bc692582de 100644 --- a/net/sunrpc/cache.c +++ b/net/sunrpc/cache.c @@ -1441,6 +1441,7 @@ static const struct file_operations cache_flush_operations_procfs = { .read = read_flush_procfs, .write = write_flush_procfs, .release = release_flush_procfs, + .llseek = no_llseek, }; static void remove_cache_proc_entries(struct cache_detail *cd) @@ -1646,6 +1647,7 @@ const struct file_operations cache_flush_operations_pipefs = { .read = read_flush_pipefs, .write = write_flush_pipefs, .release = release_flush_pipefs, + .llseek = no_llseek, }; int sunrpc_cache_register_pipefs(struct dentry *parent, diff --git a/net/wireless/debugfs.c b/net/wireless/debugfs.c index 3f9a57e9650..39765bcfb47 100644 --- a/net/wireless/debugfs.c +++ b/net/wireless/debugfs.c @@ -103,6 +103,7 @@ static ssize_t ht40allow_map_read(struct file *file, static const struct file_operations ht40allow_map_ops = { .read = ht40allow_map_read, .open = cfg80211_open_file_generic, + .llseek = default_llseek, }; #define DEBUGFS_ADD(name) \ diff --git a/samples/kfifo/bytestream-example.c b/samples/kfifo/bytestream-example.c index 178061e87ff..cfe40addda7 100644 --- a/samples/kfifo/bytestream-example.c +++ b/samples/kfifo/bytestream-example.c @@ -148,6 +148,7 @@ static const struct file_operations fifo_fops = { .owner = THIS_MODULE, .read = fifo_read, .write = fifo_write, + .llseek = noop_llseek, }; static int __init example_init(void) diff --git a/samples/kfifo/inttype-example.c b/samples/kfifo/inttype-example.c index 71b2aabca96..6f8e79e76c9 100644 --- a/samples/kfifo/inttype-example.c +++ b/samples/kfifo/inttype-example.c @@ -141,6 +141,7 @@ static const struct file_operations fifo_fops = { .owner = THIS_MODULE, .read = fifo_read, .write = fifo_write, + .llseek = noop_llseek, }; static int __init example_init(void) diff --git a/samples/kfifo/record-example.c b/samples/kfifo/record-example.c index e68bd16a5da..2d7529eeb29 100644 --- a/samples/kfifo/record-example.c +++ b/samples/kfifo/record-example.c @@ -155,6 +155,7 @@ static const struct file_operations fifo_fops = { .owner = THIS_MODULE, .read = fifo_read, .write = fifo_write, + .llseek = noop_llseek, }; static int __init example_init(void) diff --git a/samples/tracepoints/tracepoint-sample.c b/samples/tracepoints/tracepoint-sample.c index 26fab33ffa8..f4d89e008c3 100644 --- a/samples/tracepoints/tracepoint-sample.c +++ b/samples/tracepoints/tracepoint-sample.c @@ -30,6 +30,7 @@ static int my_open(struct inode *inode, struct file *file) static const struct file_operations mark_ops = { .open = my_open, + .llseek = noop_llseek, }; static int __init sample_init(void) diff --git a/security/apparmor/apparmorfs.c b/security/apparmor/apparmorfs.c index 7320331b44a..a27086d16f0 100644 --- a/security/apparmor/apparmorfs.c +++ b/security/apparmor/apparmorfs.c @@ -86,7 +86,8 @@ static ssize_t profile_load(struct file *f, const char __user *buf, size_t size, } static const struct file_operations aa_fs_profile_load = { - .write = profile_load + .write = profile_load, + .llseek = default_llseek, }; /* .replace file hook fn to load and/or replace policy */ @@ -107,7 +108,8 @@ static ssize_t profile_replace(struct file *f, const char __user *buf, } static const struct file_operations aa_fs_profile_replace = { - .write = profile_replace + .write = profile_replace, + .llseek = default_llseek, }; /* .remove file hook fn to remove loaded policy */ @@ -134,7 +136,8 @@ static ssize_t profile_remove(struct file *f, const char __user *buf, } static const struct file_operations aa_fs_profile_remove = { - .write = profile_remove + .write = profile_remove, + .llseek = default_llseek, }; /** Base file system setup **/ diff --git a/security/inode.c b/security/inode.c index 8c777f022ad..88839866cbc 100644 --- a/security/inode.c +++ b/security/inode.c @@ -53,6 +53,7 @@ static const struct file_operations default_file_ops = { .read = default_read_file, .write = default_write_file, .open = default_open, + .llseek = noop_llseek, }; static struct inode *get_inode(struct super_block *sb, int mode, dev_t dev) diff --git a/security/smack/smackfs.c b/security/smack/smackfs.c index a2b72d77f92..7512502d016 100644 --- a/security/smack/smackfs.c +++ b/security/smack/smackfs.c @@ -968,6 +968,7 @@ static ssize_t smk_write_doi(struct file *file, const char __user *buf, static const struct file_operations smk_doi_ops = { .read = smk_read_doi, .write = smk_write_doi, + .llseek = default_llseek, }; /** @@ -1031,6 +1032,7 @@ static ssize_t smk_write_direct(struct file *file, const char __user *buf, static const struct file_operations smk_direct_ops = { .read = smk_read_direct, .write = smk_write_direct, + .llseek = default_llseek, }; /** @@ -1112,6 +1114,7 @@ static ssize_t smk_write_ambient(struct file *file, const char __user *buf, static const struct file_operations smk_ambient_ops = { .read = smk_read_ambient, .write = smk_write_ambient, + .llseek = default_llseek, }; /** @@ -1191,6 +1194,7 @@ static ssize_t smk_write_onlycap(struct file *file, const char __user *buf, static const struct file_operations smk_onlycap_ops = { .read = smk_read_onlycap, .write = smk_write_onlycap, + .llseek = default_llseek, }; /** @@ -1255,6 +1259,7 @@ static ssize_t smk_write_logging(struct file *file, const char __user *buf, static const struct file_operations smk_logging_ops = { .read = smk_read_logging, .write = smk_write_logging, + .llseek = default_llseek, }; /** * smk_fill_super - fill the /smackfs superblock diff --git a/sound/core/seq/oss/seq_oss.c b/sound/core/seq/oss/seq_oss.c index f25e3cc7ddf..a1f1a2f00cc 100644 --- a/sound/core/seq/oss/seq_oss.c +++ b/sound/core/seq/oss/seq_oss.c @@ -220,6 +220,7 @@ static const struct file_operations seq_oss_f_ops = .poll = odev_poll, .unlocked_ioctl = odev_ioctl, .compat_ioctl = odev_ioctl_compat, + .llseek = noop_llseek, }; static int __init diff --git a/sound/core/sound.c b/sound/core/sound.c index ac42af42b78..62a093efb45 100644 --- a/sound/core/sound.c +++ b/sound/core/sound.c @@ -184,7 +184,8 @@ static int snd_open(struct inode *inode, struct file *file) static const struct file_operations snd_fops = { .owner = THIS_MODULE, - .open = snd_open + .open = snd_open, + .llseek = noop_llseek, }; #ifdef CONFIG_SND_DYNAMIC_MINORS diff --git a/sound/oss/msnd_pinnacle.c b/sound/oss/msnd_pinnacle.c index 2e48b17667d..ca942f7cd23 100644 --- a/sound/oss/msnd_pinnacle.c +++ b/sound/oss/msnd_pinnacle.c @@ -1117,6 +1117,7 @@ static const struct file_operations dev_fileops = { .unlocked_ioctl = dev_ioctl, .open = dev_open, .release = dev_release, + .llseek = noop_llseek, }; static int reset_dsp(void) diff --git a/sound/soc/soc-core.c b/sound/soc/soc-core.c index acc91daa1c5..4057d35343b 100644 --- a/sound/soc/soc-core.c +++ b/sound/soc/soc-core.c @@ -223,6 +223,7 @@ static const struct file_operations codec_reg_fops = { .open = codec_reg_open_file, .read = codec_reg_read_file, .write = codec_reg_write_file, + .llseek = default_llseek, }; static void soc_init_codec_debugfs(struct snd_soc_codec *codec) diff --git a/sound/soc/soc-dapm.c b/sound/soc/soc-dapm.c index 03cb7c05ebe..72a53d0a41e 100644 --- a/sound/soc/soc-dapm.c +++ b/sound/soc/soc-dapm.c @@ -1089,6 +1089,7 @@ static ssize_t dapm_widget_power_read_file(struct file *file, static const struct file_operations dapm_widget_power_fops = { .open = dapm_widget_power_open_file, .read = dapm_widget_power_read_file, + .llseek = default_llseek, }; void snd_soc_dapm_debugfs_init(struct snd_soc_codec *codec) diff --git a/sound/sound_core.c b/sound/sound_core.c index cb61317df50..c03bbaefdbc 100644 --- a/sound/sound_core.c +++ b/sound/sound_core.c @@ -165,6 +165,7 @@ static const struct file_operations soundcore_fops = /* We must have an owner or the module locking fails */ .owner = THIS_MODULE, .open = soundcore_open, + .llseek = noop_llseek, }; /* diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index d4853a54771..e039f641d66 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -1305,6 +1305,7 @@ static struct file_operations kvm_vcpu_fops = { .unlocked_ioctl = kvm_vcpu_ioctl, .compat_ioctl = kvm_vcpu_ioctl, .mmap = kvm_vcpu_mmap, + .llseek = noop_llseek, }; /* @@ -1774,6 +1775,7 @@ static struct file_operations kvm_vm_fops = { .compat_ioctl = kvm_vm_compat_ioctl, #endif .mmap = kvm_vm_mmap, + .llseek = noop_llseek, }; static int kvm_dev_ioctl_create_vm(void) @@ -1867,6 +1869,7 @@ out: static struct file_operations kvm_chardev_ops = { .unlocked_ioctl = kvm_dev_ioctl, .compat_ioctl = kvm_dev_ioctl, + .llseek = noop_llseek, }; static struct miscdevice kvm_dev = { -- cgit v1.2.3-70-g09d2 From 940b3c7b193ec54141976a8642c00582f423690c Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 15 Oct 2010 19:36:26 +0200 Subject: x86: sfi: Make local functions static Signed-off-by: Thomas Gleixner Cc: Len Brown --- arch/x86/kernel/sfi.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/sfi.c b/arch/x86/kernel/sfi.c index cb22acf3ed0..dd4c281ffe5 100644 --- a/arch/x86/kernel/sfi.c +++ b/arch/x86/kernel/sfi.c @@ -34,7 +34,7 @@ #ifdef CONFIG_X86_LOCAL_APIC static unsigned long sfi_lapic_addr __initdata = APIC_DEFAULT_PHYS_BASE; -void __init mp_sfi_register_lapic_address(unsigned long address) +static void __init mp_sfi_register_lapic_address(unsigned long address) { mp_lapic_addr = address; @@ -46,7 +46,7 @@ void __init mp_sfi_register_lapic_address(unsigned long address) } /* All CPUs enumerated by SFI must be present and enabled */ -void __cpuinit mp_sfi_register_lapic(u8 id) +static void __cpuinit mp_sfi_register_lapic(u8 id) { if (MAX_APICS - id <= 0) { pr_warning("Processor #%d invalid (max %d)\n", -- cgit v1.2.3-70-g09d2 From 40ffa93791985ab300fd488072e9f37ccf72e88c Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 15 Oct 2010 21:08:14 +0200 Subject: x86: Remove stale pmtimer_64.c This file is unused since the apic unification in 2.6.29, but nobody noticed. Signed-off-by: Thomas Gleixner --- arch/x86/kernel/Makefile | 1 - arch/x86/kernel/pmtimer_64.c | 69 -------------------------------------------- include/linux/acpi_pmtmr.h | 2 -- 3 files changed, 72 deletions(-) delete mode 100644 arch/x86/kernel/pmtimer_64.c (limited to 'arch/x86') diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 0925676266b..dd9a2e459c7 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -120,7 +120,6 @@ obj-$(CONFIG_SWIOTLB) += pci-swiotlb.o # 64 bit specific files ifeq ($(CONFIG_X86_64),y) obj-$(CONFIG_X86_UV) += tlb_uv.o bios_uv.o uv_irq.o uv_sysfs.o uv_time.o - obj-$(CONFIG_X86_PM_TIMER) += pmtimer_64.o obj-$(CONFIG_AUDIT) += audit_64.o obj-$(CONFIG_GART_IOMMU) += pci-gart_64.o aperture_64.o diff --git a/arch/x86/kernel/pmtimer_64.c b/arch/x86/kernel/pmtimer_64.c deleted file mode 100644 index b112406f199..00000000000 --- a/arch/x86/kernel/pmtimer_64.c +++ /dev/null @@ -1,69 +0,0 @@ -/* Ported over from i386 by AK, original copyright was: - * - * (C) Dominik Brodowski 2003 - * - * Driver to use the Power Management Timer (PMTMR) available in some - * southbridges as primary timing source for the Linux kernel. - * - * Based on parts of linux/drivers/acpi/hardware/hwtimer.c, timer_pit.c, - * timer_hpet.c, and on Arjan van de Ven's implementation for 2.4. - * - * This file is licensed under the GPL v2. - * - * Dropped all the hardware bug workarounds for now. Hopefully they - * are not needed on 64bit chipsets. - */ - -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include - -static inline u32 cyc2us(u32 cycles) -{ - /* The Power Management Timer ticks at 3.579545 ticks per microsecond. - * 1 / PM_TIMER_FREQUENCY == 0.27936511 =~ 286/1024 [error: 0.024%] - * - * Even with HZ = 100, delta is at maximum 35796 ticks, so it can - * easily be multiplied with 286 (=0x11E) without having to fear - * u32 overflows. - */ - cycles *= 286; - return (cycles >> 10); -} - -static unsigned pmtimer_wait_tick(void) -{ - u32 a, b; - for (a = b = inl(pmtmr_ioport) & ACPI_PM_MASK; - a == b; - b = inl(pmtmr_ioport) & ACPI_PM_MASK) - cpu_relax(); - return b; -} - -/* note: wait time is rounded up to one tick */ -void pmtimer_wait(unsigned us) -{ - u32 a, b; - a = pmtimer_wait_tick(); - do { - b = inl(pmtmr_ioport); - cpu_relax(); - } while (cyc2us(b - a) < us); -} - -static int __init nopmtimer_setup(char *s) -{ - pmtmr_ioport = 0; - return 1; -} - -__setup("nopmtimer", nopmtimer_setup); diff --git a/include/linux/acpi_pmtmr.h b/include/linux/acpi_pmtmr.h index 7e3d2859be5..1d0ef1ae803 100644 --- a/include/linux/acpi_pmtmr.h +++ b/include/linux/acpi_pmtmr.h @@ -25,8 +25,6 @@ static inline u32 acpi_pm_read_early(void) return acpi_pm_read_verified() & ACPI_PM_MASK; } -extern void pmtimer_wait(unsigned); - #else static inline u32 acpi_pm_read_early(void) -- cgit v1.2.3-70-g09d2 From ba0cef3d149ce4db293c572bf36ed352b11ce7b9 Mon Sep 17 00:00:00 2001 From: Stephane Eranian Date: Fri, 15 Oct 2010 15:15:01 +0200 Subject: perf_events: Fix bogus AMD64 generic TLB events PERF_COUNT_HW_CACHE_DTLB:READ:MISS had a bogus umask value of 0 which counts nothing. Needed to be 0x7 (to count all possibilities). PERF_COUNT_HW_CACHE_ITLB:READ:MISS had a bogus umask value of 0 which counts nothing. Needed to be 0x3 (to count all possibilities). Signed-off-by: Stephane Eranian Signed-off-by: Peter Zijlstra Cc: Robert Richter Cc: # as far back as it applies LKML-Reference: <4cb85478.41e9d80a.44e2.3f00@mx.google.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event_amd.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/perf_event_amd.c b/arch/x86/kernel/cpu/perf_event_amd.c index c2897b7b4a3..46d58448c3a 100644 --- a/arch/x86/kernel/cpu/perf_event_amd.c +++ b/arch/x86/kernel/cpu/perf_event_amd.c @@ -52,7 +52,7 @@ static __initconst const u64 amd_hw_cache_event_ids [ C(DTLB) ] = { [ C(OP_READ) ] = { [ C(RESULT_ACCESS) ] = 0x0040, /* Data Cache Accesses */ - [ C(RESULT_MISS) ] = 0x0046, /* L1 DTLB and L2 DLTB Miss */ + [ C(RESULT_MISS) ] = 0x0746, /* L1_DTLB_AND_L2_DLTB_MISS.ALL */ }, [ C(OP_WRITE) ] = { [ C(RESULT_ACCESS) ] = 0, @@ -66,7 +66,7 @@ static __initconst const u64 amd_hw_cache_event_ids [ C(ITLB) ] = { [ C(OP_READ) ] = { [ C(RESULT_ACCESS) ] = 0x0080, /* Instruction fecthes */ - [ C(RESULT_MISS) ] = 0x0085, /* Instr. fetch ITLB misses */ + [ C(RESULT_MISS) ] = 0x0385, /* L1_ITLB_AND_L2_ITLB_MISS.ALL */ }, [ C(OP_WRITE) ] = { [ C(RESULT_ACCESS) ] = -1, -- cgit v1.2.3-70-g09d2 From e360adbe29241a0194e10e20595360dd7b98a2b3 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 14 Oct 2010 14:01:34 +0800 Subject: irq_work: Add generic hardirq context callbacks Provide a mechanism that allows running code in IRQ context. It is most useful for NMI code that needs to interact with the rest of the system -- like wakeup a task to drain buffers. Perf currently has such a mechanism, so extract that and provide it as a generic feature, independent of perf so that others may also benefit. The IRQ context callback is generated through self-IPIs where possible, or on architectures like powerpc the decrementer (the built-in timer facility) is set to generate an interrupt immediately. Architectures that don't have anything like this get to do with a callback from the timer tick. These architectures can call irq_work_run() at the tail of any IRQ handlers that might enqueue such work (like the perf IRQ handler) to avoid undue latencies in processing the work. Signed-off-by: Peter Zijlstra Acked-by: Kyle McMartin Acked-by: Martin Schwidefsky [ various fixes ] Signed-off-by: Huang Ying LKML-Reference: <1287036094.7768.291.camel@yhuang-dev> Signed-off-by: Ingo Molnar --- arch/alpha/Kconfig | 1 + arch/alpha/include/asm/perf_event.h | 5 -- arch/alpha/kernel/time.c | 30 +++---- arch/arm/Kconfig | 1 + arch/arm/include/asm/perf_event.h | 12 --- arch/arm/kernel/perf_event.c | 8 +- arch/frv/Kconfig | 1 + arch/frv/lib/Makefile | 2 +- arch/frv/lib/perf_event.c | 19 ---- arch/parisc/Kconfig | 1 + arch/parisc/include/asm/perf_event.h | 3 +- arch/powerpc/Kconfig | 1 + arch/powerpc/include/asm/paca.h | 2 +- arch/powerpc/kernel/time.c | 42 ++++----- arch/s390/Kconfig | 1 + arch/s390/include/asm/perf_event.h | 3 +- arch/sh/Kconfig | 1 + arch/sh/include/asm/perf_event.h | 7 -- arch/sparc/Kconfig | 2 + arch/sparc/include/asm/perf_event.h | 4 - arch/sparc/kernel/pcr.c | 8 +- arch/x86/Kconfig | 1 + arch/x86/include/asm/entry_arch.h | 4 +- arch/x86/include/asm/hardirq.h | 2 +- arch/x86/include/asm/hw_irq.h | 2 +- arch/x86/include/asm/irq_vectors.h | 4 +- arch/x86/kernel/Makefile | 1 + arch/x86/kernel/cpu/perf_event.c | 19 ---- arch/x86/kernel/entry_64.S | 6 +- arch/x86/kernel/irq.c | 8 +- arch/x86/kernel/irq_work.c | 30 +++++++ arch/x86/kernel/irqinit.c | 6 +- include/linux/irq_work.h | 20 +++++ include/linux/perf_event.h | 11 +-- init/Kconfig | 8 ++ kernel/Makefile | 2 + kernel/irq_work.c | 164 +++++++++++++++++++++++++++++++++++ kernel/perf_event.c | 104 ++-------------------- kernel/timer.c | 7 +- 39 files changed, 311 insertions(+), 242 deletions(-) delete mode 100644 arch/frv/lib/perf_event.c create mode 100644 arch/x86/kernel/irq_work.c create mode 100644 include/linux/irq_work.h create mode 100644 kernel/irq_work.c (limited to 'arch/x86') diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig index b9647bb66d1..d04ccd73af4 100644 --- a/arch/alpha/Kconfig +++ b/arch/alpha/Kconfig @@ -9,6 +9,7 @@ config ALPHA select HAVE_IDE select HAVE_OPROFILE select HAVE_SYSCALL_WRAPPERS + select HAVE_IRQ_WORK select HAVE_PERF_EVENTS select HAVE_DMA_ATTRS help diff --git a/arch/alpha/include/asm/perf_event.h b/arch/alpha/include/asm/perf_event.h index 4157cd3c44a..fe792ca818f 100644 --- a/arch/alpha/include/asm/perf_event.h +++ b/arch/alpha/include/asm/perf_event.h @@ -1,11 +1,6 @@ #ifndef __ASM_ALPHA_PERF_EVENT_H #define __ASM_ALPHA_PERF_EVENT_H -/* Alpha only supports software events through this interface. */ -extern void set_perf_event_pending(void); - -#define PERF_EVENT_INDEX_OFFSET 0 - #ifdef CONFIG_PERF_EVENTS extern void init_hw_perf_events(void); #else diff --git a/arch/alpha/kernel/time.c b/arch/alpha/kernel/time.c index 396af1799ea..0f1d8493cfc 100644 --- a/arch/alpha/kernel/time.c +++ b/arch/alpha/kernel/time.c @@ -41,7 +41,7 @@ #include #include #include -#include +#include #include #include @@ -83,25 +83,25 @@ static struct { unsigned long est_cycle_freq; -#ifdef CONFIG_PERF_EVENTS +#ifdef CONFIG_IRQ_WORK -DEFINE_PER_CPU(u8, perf_event_pending); +DEFINE_PER_CPU(u8, irq_work_pending); -#define set_perf_event_pending_flag() __get_cpu_var(perf_event_pending) = 1 -#define test_perf_event_pending() __get_cpu_var(perf_event_pending) -#define clear_perf_event_pending() __get_cpu_var(perf_event_pending) = 0 +#define set_irq_work_pending_flag() __get_cpu_var(irq_work_pending) = 1 +#define test_irq_work_pending() __get_cpu_var(irq_work_pending) +#define clear_irq_work_pending() __get_cpu_var(irq_work_pending) = 0 -void set_perf_event_pending(void) +void set_irq_work_pending(void) { - set_perf_event_pending_flag(); + set_irq_work_pending_flag(); } -#else /* CONFIG_PERF_EVENTS */ +#else /* CONFIG_IRQ_WORK */ -#define test_perf_event_pending() 0 -#define clear_perf_event_pending() +#define test_irq_work_pending() 0 +#define clear_irq_work_pending() -#endif /* CONFIG_PERF_EVENTS */ +#endif /* CONFIG_IRQ_WORK */ static inline __u32 rpcc(void) @@ -191,9 +191,9 @@ irqreturn_t timer_interrupt(int irq, void *dev) write_sequnlock(&xtime_lock); - if (test_perf_event_pending()) { - clear_perf_event_pending(); - perf_event_do_pending(); + if (test_irq_work_pending()) { + clear_irq_work_pending(); + irq_work_run(); } #ifndef CONFIG_SMP diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 88c97bc7a6f..7c0dfccd05b 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -23,6 +23,7 @@ config ARM select HAVE_KERNEL_GZIP select HAVE_KERNEL_LZO select HAVE_KERNEL_LZMA + select HAVE_IRQ_WORK select HAVE_PERF_EVENTS select PERF_USE_VMALLOC select HAVE_REGS_AND_STACK_ACCESS_API diff --git a/arch/arm/include/asm/perf_event.h b/arch/arm/include/asm/perf_event.h index b5799a3b711..c4aa4e8c6af 100644 --- a/arch/arm/include/asm/perf_event.h +++ b/arch/arm/include/asm/perf_event.h @@ -12,18 +12,6 @@ #ifndef __ARM_PERF_EVENT_H__ #define __ARM_PERF_EVENT_H__ -/* - * NOP: on *most* (read: all supported) ARM platforms, the performance - * counter interrupts are regular interrupts and not an NMI. This - * means that when we receive the interrupt we can call - * perf_event_do_pending() that handles all of the work with - * interrupts disabled. - */ -static inline void -set_perf_event_pending(void) -{ -} - /* ARM performance counters start from 1 (in the cp15 accesses) so use the * same indexes here for consistency. */ #define PERF_EVENT_INDEX_OFFSET 1 diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index 6cc6521881a..49643b1467e 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -1092,7 +1092,7 @@ armv6pmu_handle_irq(int irq_num, * platforms that can have the PMU interrupts raised as an NMI, this * will not work. */ - perf_event_do_pending(); + irq_work_run(); return IRQ_HANDLED; } @@ -2068,7 +2068,7 @@ static irqreturn_t armv7pmu_handle_irq(int irq_num, void *dev) * platforms that can have the PMU interrupts raised as an NMI, this * will not work. */ - perf_event_do_pending(); + irq_work_run(); return IRQ_HANDLED; } @@ -2436,7 +2436,7 @@ xscale1pmu_handle_irq(int irq_num, void *dev) armpmu->disable(hwc, idx); } - perf_event_do_pending(); + irq_work_run(); /* * Re-enable the PMU. @@ -2763,7 +2763,7 @@ xscale2pmu_handle_irq(int irq_num, void *dev) armpmu->disable(hwc, idx); } - perf_event_do_pending(); + irq_work_run(); /* * Re-enable the PMU. diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig index 16399bd2499..0f2417df632 100644 --- a/arch/frv/Kconfig +++ b/arch/frv/Kconfig @@ -7,6 +7,7 @@ config FRV default y select HAVE_IDE select HAVE_ARCH_TRACEHOOK + select HAVE_IRQ_WORK select HAVE_PERF_EVENTS config ZONE_DMA diff --git a/arch/frv/lib/Makefile b/arch/frv/lib/Makefile index f4709756d0d..4ff2fb1e6b1 100644 --- a/arch/frv/lib/Makefile +++ b/arch/frv/lib/Makefile @@ -5,4 +5,4 @@ lib-y := \ __ashldi3.o __lshrdi3.o __muldi3.o __ashrdi3.o __negdi2.o __ucmpdi2.o \ checksum.o memcpy.o memset.o atomic-ops.o atomic64-ops.o \ - outsl_ns.o outsl_sw.o insl_ns.o insl_sw.o cache.o perf_event.o + outsl_ns.o outsl_sw.o insl_ns.o insl_sw.o cache.o diff --git a/arch/frv/lib/perf_event.c b/arch/frv/lib/perf_event.c deleted file mode 100644 index 9ac5acfd2e9..00000000000 --- a/arch/frv/lib/perf_event.c +++ /dev/null @@ -1,19 +0,0 @@ -/* Performance event handling - * - * Copyright (C) 2009 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public Licence - * as published by the Free Software Foundation; either version - * 2 of the Licence, or (at your option) any later version. - */ - -#include - -/* - * mark the performance event as pending - */ -void set_perf_event_pending(void) -{ -} diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig index 907417d187e..79a04a9394d 100644 --- a/arch/parisc/Kconfig +++ b/arch/parisc/Kconfig @@ -16,6 +16,7 @@ config PARISC select RTC_DRV_GENERIC select INIT_ALL_POSSIBLE select BUG + select HAVE_IRQ_WORK select HAVE_PERF_EVENTS select GENERIC_ATOMIC64 if !64BIT help diff --git a/arch/parisc/include/asm/perf_event.h b/arch/parisc/include/asm/perf_event.h index cc146427d8f..1e0fd8ba6c0 100644 --- a/arch/parisc/include/asm/perf_event.h +++ b/arch/parisc/include/asm/perf_event.h @@ -1,7 +1,6 @@ #ifndef __ASM_PARISC_PERF_EVENT_H #define __ASM_PARISC_PERF_EVENT_H -/* parisc only supports software events through this interface. */ -static inline void set_perf_event_pending(void) { } +/* Empty, just to avoid compiling error */ #endif /* __ASM_PARISC_PERF_EVENT_H */ diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 631e5a0fb6a..4b1e521d966 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -138,6 +138,7 @@ config PPC select HAVE_OPROFILE select HAVE_SYSCALL_WRAPPERS if PPC64 select GENERIC_ATOMIC64 if PPC32 + select HAVE_IRQ_WORK select HAVE_PERF_EVENTS select HAVE_REGS_AND_STACK_ACCESS_API select HAVE_HW_BREAKPOINT if PERF_EVENTS && PPC_BOOK3S_64 diff --git a/arch/powerpc/include/asm/paca.h b/arch/powerpc/include/asm/paca.h index 1ff6662f7fa..9b287fdd8ea 100644 --- a/arch/powerpc/include/asm/paca.h +++ b/arch/powerpc/include/asm/paca.h @@ -129,7 +129,7 @@ struct paca_struct { u8 soft_enabled; /* irq soft-enable flag */ u8 hard_enabled; /* set if irqs are enabled in MSR */ u8 io_sync; /* writel() needs spin_unlock sync */ - u8 perf_event_pending; /* PM interrupt while soft-disabled */ + u8 irq_work_pending; /* IRQ_WORK interrupt while soft-disable */ /* Stuff for accurate time accounting */ u64 user_time; /* accumulated usermode TB ticks */ diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index 8533b3b83f5..54888eb10c3 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -53,7 +53,7 @@ #include #include #include -#include +#include #include #include @@ -493,60 +493,60 @@ void __init iSeries_time_init_early(void) } #endif /* CONFIG_PPC_ISERIES */ -#ifdef CONFIG_PERF_EVENTS +#ifdef CONFIG_IRQ_WORK /* * 64-bit uses a byte in the PACA, 32-bit uses a per-cpu variable... */ #ifdef CONFIG_PPC64 -static inline unsigned long test_perf_event_pending(void) +static inline unsigned long test_irq_work_pending(void) { unsigned long x; asm volatile("lbz %0,%1(13)" : "=r" (x) - : "i" (offsetof(struct paca_struct, perf_event_pending))); + : "i" (offsetof(struct paca_struct, irq_work_pending))); return x; } -static inline void set_perf_event_pending_flag(void) +static inline void set_irq_work_pending_flag(void) { asm volatile("stb %0,%1(13)" : : "r" (1), - "i" (offsetof(struct paca_struct, perf_event_pending))); + "i" (offsetof(struct paca_struct, irq_work_pending))); } -static inline void clear_perf_event_pending(void) +static inline void clear_irq_work_pending(void) { asm volatile("stb %0,%1(13)" : : "r" (0), - "i" (offsetof(struct paca_struct, perf_event_pending))); + "i" (offsetof(struct paca_struct, irq_work_pending))); } #else /* 32-bit */ -DEFINE_PER_CPU(u8, perf_event_pending); +DEFINE_PER_CPU(u8, irq_work_pending); -#define set_perf_event_pending_flag() __get_cpu_var(perf_event_pending) = 1 -#define test_perf_event_pending() __get_cpu_var(perf_event_pending) -#define clear_perf_event_pending() __get_cpu_var(perf_event_pending) = 0 +#define set_irq_work_pending_flag() __get_cpu_var(irq_work_pending) = 1 +#define test_irq_work_pending() __get_cpu_var(irq_work_pending) +#define clear_irq_work_pending() __get_cpu_var(irq_work_pending) = 0 #endif /* 32 vs 64 bit */ -void set_perf_event_pending(void) +void set_irq_work_pending(void) { preempt_disable(); - set_perf_event_pending_flag(); + set_irq_work_pending_flag(); set_dec(1); preempt_enable(); } -#else /* CONFIG_PERF_EVENTS */ +#else /* CONFIG_IRQ_WORK */ -#define test_perf_event_pending() 0 -#define clear_perf_event_pending() +#define test_irq_work_pending() 0 +#define clear_irq_work_pending() -#endif /* CONFIG_PERF_EVENTS */ +#endif /* CONFIG_IRQ_WORK */ /* * For iSeries shared processors, we have to let the hypervisor @@ -587,9 +587,9 @@ void timer_interrupt(struct pt_regs * regs) calculate_steal_time(); - if (test_perf_event_pending()) { - clear_perf_event_pending(); - perf_event_do_pending(); + if (test_irq_work_pending()) { + clear_irq_work_pending(); + irq_work_run(); } #ifdef CONFIG_PPC_ISERIES diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index f0777a47e3a..958f0dadead 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -95,6 +95,7 @@ config S390 select HAVE_KVM if 64BIT select HAVE_ARCH_TRACEHOOK select INIT_ALL_POSSIBLE + select HAVE_IRQ_WORK select HAVE_PERF_EVENTS select HAVE_KERNEL_GZIP select HAVE_KERNEL_BZIP2 diff --git a/arch/s390/include/asm/perf_event.h b/arch/s390/include/asm/perf_event.h index 3840cbe7763..a75f168d271 100644 --- a/arch/s390/include/asm/perf_event.h +++ b/arch/s390/include/asm/perf_event.h @@ -4,7 +4,6 @@ * Copyright 2009 Martin Schwidefsky, IBM Corporation. */ -static inline void set_perf_event_pending(void) {} -static inline void clear_perf_event_pending(void) {} +/* Empty, just to avoid compiling error */ #define PERF_EVENT_INDEX_OFFSET 0 diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig index 35b6c3f8517..35b6879628a 100644 --- a/arch/sh/Kconfig +++ b/arch/sh/Kconfig @@ -16,6 +16,7 @@ config SUPERH select HAVE_ARCH_TRACEHOOK select HAVE_DMA_API_DEBUG select HAVE_DMA_ATTRS + select HAVE_IRQ_WORK select HAVE_PERF_EVENTS select PERF_USE_VMALLOC select HAVE_KERNEL_GZIP diff --git a/arch/sh/include/asm/perf_event.h b/arch/sh/include/asm/perf_event.h index 3d0c9f36d15..14308bed7ea 100644 --- a/arch/sh/include/asm/perf_event.h +++ b/arch/sh/include/asm/perf_event.h @@ -26,11 +26,4 @@ extern int register_sh_pmu(struct sh_pmu *); extern int reserve_pmc_hardware(void); extern void release_pmc_hardware(void); -static inline void set_perf_event_pending(void) -{ - /* Nothing to see here, move along. */ -} - -#define PERF_EVENT_INDEX_OFFSET 0 - #endif /* __ASM_SH_PERF_EVENT_H */ diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig index 9212cd42a83..3e9d31401fb 100644 --- a/arch/sparc/Kconfig +++ b/arch/sparc/Kconfig @@ -26,6 +26,7 @@ config SPARC select ARCH_WANT_OPTIONAL_GPIOLIB select RTC_CLASS select RTC_DRV_M48T59 + select HAVE_IRQ_WORK select HAVE_PERF_EVENTS select PERF_USE_VMALLOC select HAVE_DMA_ATTRS @@ -54,6 +55,7 @@ config SPARC64 select RTC_DRV_BQ4802 select RTC_DRV_SUN4V select RTC_DRV_STARFIRE + select HAVE_IRQ_WORK select HAVE_PERF_EVENTS select PERF_USE_VMALLOC diff --git a/arch/sparc/include/asm/perf_event.h b/arch/sparc/include/asm/perf_event.h index 727af70646c..6e8bfa1786d 100644 --- a/arch/sparc/include/asm/perf_event.h +++ b/arch/sparc/include/asm/perf_event.h @@ -1,10 +1,6 @@ #ifndef __ASM_SPARC_PERF_EVENT_H #define __ASM_SPARC_PERF_EVENT_H -extern void set_perf_event_pending(void); - -#define PERF_EVENT_INDEX_OFFSET 0 - #ifdef CONFIG_PERF_EVENTS #include diff --git a/arch/sparc/kernel/pcr.c b/arch/sparc/kernel/pcr.c index c4a6a50b484..b87873c0e8e 100644 --- a/arch/sparc/kernel/pcr.c +++ b/arch/sparc/kernel/pcr.c @@ -7,7 +7,7 @@ #include #include -#include +#include #include #include @@ -43,14 +43,14 @@ void __irq_entry deferred_pcr_work_irq(int irq, struct pt_regs *regs) old_regs = set_irq_regs(regs); irq_enter(); -#ifdef CONFIG_PERF_EVENTS - perf_event_do_pending(); +#ifdef CONFIG_IRQ_WORK + irq_work_run(); #endif irq_exit(); set_irq_regs(old_regs); } -void set_perf_event_pending(void) +void arch_irq_work_raise(void) { set_softint(1 << PIL_DEFERRED_PCR_WORK); } diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 9815221976a..fd227d6b8d9 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -25,6 +25,7 @@ config X86 select HAVE_IDE select HAVE_OPROFILE select HAVE_PERF_EVENTS if (!M386 && !M486) + select HAVE_IRQ_WORK select HAVE_IOREMAP_PROT select HAVE_KPROBES select ARCH_WANT_OPTIONAL_GPIOLIB diff --git a/arch/x86/include/asm/entry_arch.h b/arch/x86/include/asm/entry_arch.h index 8e8ec663a98..b8e96a18676 100644 --- a/arch/x86/include/asm/entry_arch.h +++ b/arch/x86/include/asm/entry_arch.h @@ -49,8 +49,8 @@ BUILD_INTERRUPT(apic_timer_interrupt,LOCAL_TIMER_VECTOR) BUILD_INTERRUPT(error_interrupt,ERROR_APIC_VECTOR) BUILD_INTERRUPT(spurious_interrupt,SPURIOUS_APIC_VECTOR) -#ifdef CONFIG_PERF_EVENTS -BUILD_INTERRUPT(perf_pending_interrupt, LOCAL_PENDING_VECTOR) +#ifdef CONFIG_IRQ_WORK +BUILD_INTERRUPT(irq_work_interrupt, IRQ_WORK_VECTOR) #endif #ifdef CONFIG_X86_THERMAL_VECTOR diff --git a/arch/x86/include/asm/hardirq.h b/arch/x86/include/asm/hardirq.h index aeab29aee61..55e4de613f0 100644 --- a/arch/x86/include/asm/hardirq.h +++ b/arch/x86/include/asm/hardirq.h @@ -14,7 +14,7 @@ typedef struct { #endif unsigned int x86_platform_ipis; /* arch dependent */ unsigned int apic_perf_irqs; - unsigned int apic_pending_irqs; + unsigned int apic_irq_work_irqs; #ifdef CONFIG_SMP unsigned int irq_resched_count; unsigned int irq_call_count; diff --git a/arch/x86/include/asm/hw_irq.h b/arch/x86/include/asm/hw_irq.h index 46c0fe05f23..3a54a1ca1a0 100644 --- a/arch/x86/include/asm/hw_irq.h +++ b/arch/x86/include/asm/hw_irq.h @@ -29,7 +29,7 @@ extern void apic_timer_interrupt(void); extern void x86_platform_ipi(void); extern void error_interrupt(void); -extern void perf_pending_interrupt(void); +extern void irq_work_interrupt(void); extern void spurious_interrupt(void); extern void thermal_interrupt(void); diff --git a/arch/x86/include/asm/irq_vectors.h b/arch/x86/include/asm/irq_vectors.h index e2ca3009255..6af0894dafb 100644 --- a/arch/x86/include/asm/irq_vectors.h +++ b/arch/x86/include/asm/irq_vectors.h @@ -114,9 +114,9 @@ #define X86_PLATFORM_IPI_VECTOR 0xed /* - * Performance monitoring pending work vector: + * IRQ work vector: */ -#define LOCAL_PENDING_VECTOR 0xec +#define IRQ_WORK_VECTOR 0xec #define UV_BAU_MESSAGE 0xea diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 9d3f485e5dd..7490bf8d145 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -35,6 +35,7 @@ obj-y := process_$(BITS).o signal.o entry_$(BITS).o obj-y += traps.o irq.o irq_$(BITS).o dumpstack_$(BITS).o obj-y += time.o ioport.o ldt.o dumpstack.o obj-y += setup.o x86_init.o i8259.o irqinit.o jump_label.o +obj-$(CONFIG_IRQ_WORK) += irq_work.o obj-$(CONFIG_X86_VISWS) += visws_quirks.o obj-$(CONFIG_X86_32) += probe_roms_32.o obj-$(CONFIG_X86_32) += sys_i386_32.o i386_ksyms_32.o diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index e2513f26ba8..fe73c1844a9 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -1196,25 +1196,6 @@ static int x86_pmu_handle_irq(struct pt_regs *regs) return handled; } -void smp_perf_pending_interrupt(struct pt_regs *regs) -{ - irq_enter(); - ack_APIC_irq(); - inc_irq_stat(apic_pending_irqs); - perf_event_do_pending(); - irq_exit(); -} - -void set_perf_event_pending(void) -{ -#ifdef CONFIG_X86_LOCAL_APIC - if (!x86_pmu.apic || !x86_pmu_initialized()) - return; - - apic->send_IPI_self(LOCAL_PENDING_VECTOR); -#endif -} - void perf_events_lapic_init(void) { if (!x86_pmu.apic || !x86_pmu_initialized()) diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index 17be5ec7cbb..c375c79065f 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -1023,9 +1023,9 @@ apicinterrupt ERROR_APIC_VECTOR \ apicinterrupt SPURIOUS_APIC_VECTOR \ spurious_interrupt smp_spurious_interrupt -#ifdef CONFIG_PERF_EVENTS -apicinterrupt LOCAL_PENDING_VECTOR \ - perf_pending_interrupt smp_perf_pending_interrupt +#ifdef CONFIG_IRQ_WORK +apicinterrupt IRQ_WORK_VECTOR \ + irq_work_interrupt smp_irq_work_interrupt #endif /* diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c index 91fd0c70a18..44edb03fc9e 100644 --- a/arch/x86/kernel/irq.c +++ b/arch/x86/kernel/irq.c @@ -67,10 +67,10 @@ static int show_other_interrupts(struct seq_file *p, int prec) for_each_online_cpu(j) seq_printf(p, "%10u ", irq_stats(j)->apic_perf_irqs); seq_printf(p, " Performance monitoring interrupts\n"); - seq_printf(p, "%*s: ", prec, "PND"); + seq_printf(p, "%*s: ", prec, "IWI"); for_each_online_cpu(j) - seq_printf(p, "%10u ", irq_stats(j)->apic_pending_irqs); - seq_printf(p, " Performance pending work\n"); + seq_printf(p, "%10u ", irq_stats(j)->apic_irq_work_irqs); + seq_printf(p, " IRQ work interrupts\n"); #endif if (x86_platform_ipi_callback) { seq_printf(p, "%*s: ", prec, "PLT"); @@ -185,7 +185,7 @@ u64 arch_irq_stat_cpu(unsigned int cpu) sum += irq_stats(cpu)->apic_timer_irqs; sum += irq_stats(cpu)->irq_spurious_count; sum += irq_stats(cpu)->apic_perf_irqs; - sum += irq_stats(cpu)->apic_pending_irqs; + sum += irq_stats(cpu)->apic_irq_work_irqs; #endif if (x86_platform_ipi_callback) sum += irq_stats(cpu)->x86_platform_ipis; diff --git a/arch/x86/kernel/irq_work.c b/arch/x86/kernel/irq_work.c new file mode 100644 index 00000000000..ca8f703a1e7 --- /dev/null +++ b/arch/x86/kernel/irq_work.c @@ -0,0 +1,30 @@ +/* + * x86 specific code for irq_work + * + * Copyright (C) 2010 Red Hat, Inc., Peter Zijlstra + */ + +#include +#include +#include +#include + +void smp_irq_work_interrupt(struct pt_regs *regs) +{ + irq_enter(); + ack_APIC_irq(); + inc_irq_stat(apic_irq_work_irqs); + irq_work_run(); + irq_exit(); +} + +void arch_irq_work_raise(void) +{ +#ifdef CONFIG_X86_LOCAL_APIC + if (!cpu_has_apic) + return; + + apic->send_IPI_self(IRQ_WORK_VECTOR); + apic_wait_icr_idle(); +#endif +} diff --git a/arch/x86/kernel/irqinit.c b/arch/x86/kernel/irqinit.c index 990ae7cfc57..713969b9266 100644 --- a/arch/x86/kernel/irqinit.c +++ b/arch/x86/kernel/irqinit.c @@ -224,9 +224,9 @@ static void __init apic_intr_init(void) alloc_intr_gate(SPURIOUS_APIC_VECTOR, spurious_interrupt); alloc_intr_gate(ERROR_APIC_VECTOR, error_interrupt); - /* Performance monitoring interrupts: */ -# ifdef CONFIG_PERF_EVENTS - alloc_intr_gate(LOCAL_PENDING_VECTOR, perf_pending_interrupt); + /* IRQ work interrupts: */ +# ifdef CONFIG_IRQ_WORK + alloc_intr_gate(IRQ_WORK_VECTOR, irq_work_interrupt); # endif #endif diff --git a/include/linux/irq_work.h b/include/linux/irq_work.h new file mode 100644 index 00000000000..4fa09d4d0b7 --- /dev/null +++ b/include/linux/irq_work.h @@ -0,0 +1,20 @@ +#ifndef _LINUX_IRQ_WORK_H +#define _LINUX_IRQ_WORK_H + +struct irq_work { + struct irq_work *next; + void (*func)(struct irq_work *); +}; + +static inline +void init_irq_work(struct irq_work *entry, void (*func)(struct irq_work *)) +{ + entry->next = NULL; + entry->func = func; +} + +bool irq_work_queue(struct irq_work *entry); +void irq_work_run(void); +void irq_work_sync(struct irq_work *entry); + +#endif /* _LINUX_IRQ_WORK_H */ diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index a9227e98520..2ebfc9ae475 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -486,6 +486,7 @@ struct perf_guest_info_callbacks { #include #include #include +#include #include #include @@ -672,11 +673,6 @@ struct perf_buffer { void *data_pages[0]; }; -struct perf_pending_entry { - struct perf_pending_entry *next; - void (*func)(struct perf_pending_entry *); -}; - struct perf_sample_data; typedef void (*perf_overflow_handler_t)(struct perf_event *, int, @@ -784,7 +780,7 @@ struct perf_event { int pending_wakeup; int pending_kill; int pending_disable; - struct perf_pending_entry pending; + struct irq_work pending; atomic_t event_limit; @@ -898,8 +894,6 @@ extern int perf_event_init_task(struct task_struct *child); extern void perf_event_exit_task(struct task_struct *child); extern void perf_event_free_task(struct task_struct *task); extern void perf_event_delayed_put(struct task_struct *task); -extern void set_perf_event_pending(void); -extern void perf_event_do_pending(void); extern void perf_event_print_debug(void); extern void perf_pmu_disable(struct pmu *pmu); extern void perf_pmu_enable(struct pmu *pmu); @@ -1078,7 +1072,6 @@ static inline int perf_event_init_task(struct task_struct *child) { return 0; } static inline void perf_event_exit_task(struct task_struct *child) { } static inline void perf_event_free_task(struct task_struct *task) { } static inline void perf_event_delayed_put(struct task_struct *task) { } -static inline void perf_event_do_pending(void) { } static inline void perf_event_print_debug(void) { } static inline int perf_event_task_disable(void) { return -EINVAL; } static inline int perf_event_task_enable(void) { return -EINVAL; } diff --git a/init/Kconfig b/init/Kconfig index 2de5b1cbadd..1ef0b439908 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -21,6 +21,13 @@ config CONSTRUCTORS depends on !UML default y +config HAVE_IRQ_WORK + bool + +config IRQ_WORK + bool + depends on HAVE_IRQ_WORK + menu "General setup" config EXPERIMENTAL @@ -987,6 +994,7 @@ config PERF_EVENTS default y if (PROFILING || PERF_COUNTERS) depends on HAVE_PERF_EVENTS select ANON_INODES + select IRQ_WORK help Enable kernel support for various performance events provided by software and hardware. diff --git a/kernel/Makefile b/kernel/Makefile index d52b473c99a..4d9bf5f8531 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -23,6 +23,7 @@ CFLAGS_REMOVE_rtmutex-debug.o = -pg CFLAGS_REMOVE_cgroup-debug.o = -pg CFLAGS_REMOVE_sched_clock.o = -pg CFLAGS_REMOVE_perf_event.o = -pg +CFLAGS_REMOVE_irq_work.o = -pg endif obj-$(CONFIG_FREEZER) += freezer.o @@ -100,6 +101,7 @@ obj-$(CONFIG_TRACING) += trace/ obj-$(CONFIG_X86_DS) += trace/ obj-$(CONFIG_RING_BUFFER) += trace/ obj-$(CONFIG_SMP) += sched_cpupri.o +obj-$(CONFIG_IRQ_WORK) += irq_work.o obj-$(CONFIG_PERF_EVENTS) += perf_event.o obj-$(CONFIG_HAVE_HW_BREAKPOINT) += hw_breakpoint.o obj-$(CONFIG_USER_RETURN_NOTIFIER) += user-return-notifier.o diff --git a/kernel/irq_work.c b/kernel/irq_work.c new file mode 100644 index 00000000000..f16763ff848 --- /dev/null +++ b/kernel/irq_work.c @@ -0,0 +1,164 @@ +/* + * Copyright (C) 2010 Red Hat, Inc., Peter Zijlstra + * + * Provides a framework for enqueueing and running callbacks from hardirq + * context. The enqueueing is NMI-safe. + */ + +#include +#include +#include +#include + +/* + * An entry can be in one of four states: + * + * free NULL, 0 -> {claimed} : free to be used + * claimed NULL, 3 -> {pending} : claimed to be enqueued + * pending next, 3 -> {busy} : queued, pending callback + * busy NULL, 2 -> {free, claimed} : callback in progress, can be claimed + * + * We use the lower two bits of the next pointer to keep PENDING and BUSY + * flags. + */ + +#define IRQ_WORK_PENDING 1UL +#define IRQ_WORK_BUSY 2UL +#define IRQ_WORK_FLAGS 3UL + +static inline bool irq_work_is_set(struct irq_work *entry, int flags) +{ + return (unsigned long)entry->next & flags; +} + +static inline struct irq_work *irq_work_next(struct irq_work *entry) +{ + unsigned long next = (unsigned long)entry->next; + next &= ~IRQ_WORK_FLAGS; + return (struct irq_work *)next; +} + +static inline struct irq_work *next_flags(struct irq_work *entry, int flags) +{ + unsigned long next = (unsigned long)entry; + next |= flags; + return (struct irq_work *)next; +} + +static DEFINE_PER_CPU(struct irq_work *, irq_work_list); + +/* + * Claim the entry so that no one else will poke at it. + */ +static bool irq_work_claim(struct irq_work *entry) +{ + struct irq_work *next, *nflags; + + do { + next = entry->next; + if ((unsigned long)next & IRQ_WORK_PENDING) + return false; + nflags = next_flags(next, IRQ_WORK_FLAGS); + } while (cmpxchg(&entry->next, next, nflags) != next); + + return true; +} + + +void __weak arch_irq_work_raise(void) +{ + /* + * Lame architectures will get the timer tick callback + */ +} + +/* + * Queue the entry and raise the IPI if needed. + */ +static void __irq_work_queue(struct irq_work *entry) +{ + struct irq_work **head, *next; + + head = &get_cpu_var(irq_work_list); + + do { + next = *head; + /* Can assign non-atomic because we keep the flags set. */ + entry->next = next_flags(next, IRQ_WORK_FLAGS); + } while (cmpxchg(head, next, entry) != next); + + /* The list was empty, raise self-interrupt to start processing. */ + if (!irq_work_next(entry)) + arch_irq_work_raise(); + + put_cpu_var(irq_work_list); +} + +/* + * Enqueue the irq_work @entry, returns true on success, failure when the + * @entry was already enqueued by someone else. + * + * Can be re-enqueued while the callback is still in progress. + */ +bool irq_work_queue(struct irq_work *entry) +{ + if (!irq_work_claim(entry)) { + /* + * Already enqueued, can't do! + */ + return false; + } + + __irq_work_queue(entry); + return true; +} +EXPORT_SYMBOL_GPL(irq_work_queue); + +/* + * Run the irq_work entries on this cpu. Requires to be ran from hardirq + * context with local IRQs disabled. + */ +void irq_work_run(void) +{ + struct irq_work *list, **head; + + head = &__get_cpu_var(irq_work_list); + if (*head == NULL) + return; + + BUG_ON(!in_irq()); + BUG_ON(!irqs_disabled()); + + list = xchg(head, NULL); + while (list != NULL) { + struct irq_work *entry = list; + + list = irq_work_next(list); + + /* + * Clear the PENDING bit, after this point the @entry + * can be re-used. + */ + entry->next = next_flags(NULL, IRQ_WORK_BUSY); + entry->func(entry); + /* + * Clear the BUSY bit and return to the free state if + * no-one else claimed it meanwhile. + */ + cmpxchg(&entry->next, next_flags(NULL, IRQ_WORK_BUSY), NULL); + } +} +EXPORT_SYMBOL_GPL(irq_work_run); + +/* + * Synchronize against the irq_work @entry, ensures the entry is not + * currently in use. + */ +void irq_work_sync(struct irq_work *entry) +{ + WARN_ON_ONCE(irqs_disabled()); + + while (irq_work_is_set(entry, IRQ_WORK_BUSY)) + cpu_relax(); +} +EXPORT_SYMBOL_GPL(irq_work_sync); diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 634f86a4b2f..99b9700e74d 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -2206,12 +2206,11 @@ static void free_event_rcu(struct rcu_head *head) kfree(event); } -static void perf_pending_sync(struct perf_event *event); static void perf_buffer_put(struct perf_buffer *buffer); static void free_event(struct perf_event *event) { - perf_pending_sync(event); + irq_work_sync(&event->pending); if (!event->parent) { atomic_dec(&nr_events); @@ -3162,16 +3161,7 @@ void perf_event_wakeup(struct perf_event *event) } } -/* - * Pending wakeups - * - * Handle the case where we need to wakeup up from NMI (or rq->lock) context. - * - * The NMI bit means we cannot possibly take locks. Therefore, maintain a - * single linked list and use cmpxchg() to add entries lockless. - */ - -static void perf_pending_event(struct perf_pending_entry *entry) +static void perf_pending_event(struct irq_work *entry) { struct perf_event *event = container_of(entry, struct perf_event, pending); @@ -3187,89 +3177,6 @@ static void perf_pending_event(struct perf_pending_entry *entry) } } -#define PENDING_TAIL ((struct perf_pending_entry *)-1UL) - -static DEFINE_PER_CPU(struct perf_pending_entry *, perf_pending_head) = { - PENDING_TAIL, -}; - -static void perf_pending_queue(struct perf_pending_entry *entry, - void (*func)(struct perf_pending_entry *)) -{ - struct perf_pending_entry **head; - - if (cmpxchg(&entry->next, NULL, PENDING_TAIL) != NULL) - return; - - entry->func = func; - - head = &get_cpu_var(perf_pending_head); - - do { - entry->next = *head; - } while (cmpxchg(head, entry->next, entry) != entry->next); - - set_perf_event_pending(); - - put_cpu_var(perf_pending_head); -} - -static int __perf_pending_run(void) -{ - struct perf_pending_entry *list; - int nr = 0; - - list = xchg(&__get_cpu_var(perf_pending_head), PENDING_TAIL); - while (list != PENDING_TAIL) { - void (*func)(struct perf_pending_entry *); - struct perf_pending_entry *entry = list; - - list = list->next; - - func = entry->func; - entry->next = NULL; - /* - * Ensure we observe the unqueue before we issue the wakeup, - * so that we won't be waiting forever. - * -- see perf_not_pending(). - */ - smp_wmb(); - - func(entry); - nr++; - } - - return nr; -} - -static inline int perf_not_pending(struct perf_event *event) -{ - /* - * If we flush on whatever cpu we run, there is a chance we don't - * need to wait. - */ - get_cpu(); - __perf_pending_run(); - put_cpu(); - - /* - * Ensure we see the proper queue state before going to sleep - * so that we do not miss the wakeup. -- see perf_pending_handle() - */ - smp_rmb(); - return event->pending.next == NULL; -} - -static void perf_pending_sync(struct perf_event *event) -{ - wait_event(event->waitq, perf_not_pending(event)); -} - -void perf_event_do_pending(void) -{ - __perf_pending_run(); -} - /* * We assume there is only KVM supporting the callbacks. * Later on, we might change it to a list if there is @@ -3319,8 +3226,7 @@ static void perf_output_wakeup(struct perf_output_handle *handle) if (handle->nmi) { handle->event->pending_wakeup = 1; - perf_pending_queue(&handle->event->pending, - perf_pending_event); + irq_work_queue(&handle->event->pending); } else perf_event_wakeup(handle->event); } @@ -4356,8 +4262,7 @@ static int __perf_event_overflow(struct perf_event *event, int nmi, event->pending_kill = POLL_HUP; if (nmi) { event->pending_disable = 1; - perf_pending_queue(&event->pending, - perf_pending_event); + irq_work_queue(&event->pending); } else perf_event_disable(event); } @@ -5374,6 +5279,7 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu, INIT_LIST_HEAD(&event->event_entry); INIT_LIST_HEAD(&event->sibling_list); init_waitqueue_head(&event->waitq); + init_irq_work(&event->pending, perf_pending_event); mutex_init(&event->mmap_mutex); diff --git a/kernel/timer.c b/kernel/timer.c index 97bf05baade..68a9ae7679b 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -37,7 +37,7 @@ #include #include #include -#include +#include #include #include @@ -1279,7 +1279,10 @@ void update_process_times(int user_tick) run_local_timers(); rcu_check_callbacks(cpu, user_tick); printk_tick(); - perf_event_do_pending(); +#ifdef CONFIG_IRQ_WORK + if (in_irq()) + irq_work_run(); +#endif scheduler_tick(); run_posix_cpu_timers(p); } -- cgit v1.2.3-70-g09d2 From e82b8e4ea4f3dffe6e7939f90e78da675fcc450e Mon Sep 17 00:00:00 2001 From: Venkatesh Pallipadi Date: Mon, 4 Oct 2010 17:03:20 -0700 Subject: x86: Add IRQ_TIME_ACCOUNTING This patch adds IRQ_TIME_ACCOUNTING option on x86 and runtime enables it when TSC is enabled. This change just enables fine grained irq time accounting, isn't used yet. Following patches use it for different purposes. Signed-off-by: Venkatesh Pallipadi Signed-off-by: Peter Zijlstra LKML-Reference: <1286237003-12406-6-git-send-email-venki@google.com> Signed-off-by: Ingo Molnar --- Documentation/kernel-parameters.txt | 4 ++++ arch/x86/Kconfig | 11 +++++++++++ arch/x86/kernel/tsc.c | 8 ++++++++ 3 files changed, 23 insertions(+) (limited to 'arch/x86') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 8dd7248508a..ed05a4a0d24 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -2435,6 +2435,10 @@ and is between 256 and 4096 characters. It is defined in the file disables clocksource verification at runtime. Used to enable high-resolution timer mode on older hardware, and in virtualized environment. + [x86] noirqtime: Do not use TSC to do irq accounting. + Used to run time disable IRQ_TIME_ACCOUNTING on any + platforms where RDTSC is slow and this accounting + can add overhead. turbografx.map[2|3]= [HW,JOY] TurboGraFX parallel port interface diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index cea0cd9a316..f4c70c246ff 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -795,6 +795,17 @@ config SCHED_MC making when dealing with multi-core CPU chips at a cost of slightly increased overhead in some places. If unsure say N here. +config IRQ_TIME_ACCOUNTING + bool "Fine granularity task level IRQ time accounting" + default n + ---help--- + Select this option to enable fine granularity task irq time + accounting. This is done by reading a timestamp on each + transitions between softirq and hardirq state, so there can be a + small performance impact. + + If in doubt, say N here. + source "kernel/Kconfig.preempt" config X86_UP_APIC diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index 26a863a9c2a..a1c2cd76853 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -104,10 +104,14 @@ int __init notsc_setup(char *str) __setup("notsc", notsc_setup); +static int no_sched_irq_time; + static int __init tsc_setup(char *str) { if (!strcmp(str, "reliable")) tsc_clocksource_reliable = 1; + if (!strncmp(str, "noirqtime", 9)) + no_sched_irq_time = 1; return 1; } @@ -801,6 +805,7 @@ void mark_tsc_unstable(char *reason) if (!tsc_unstable) { tsc_unstable = 1; sched_clock_stable = 0; + disable_sched_clock_irqtime(); printk(KERN_INFO "Marking TSC unstable due to %s\n", reason); /* Change only the rating, when not registered */ if (clocksource_tsc.mult) @@ -987,6 +992,9 @@ void __init tsc_init(void) /* now allow native_sched_clock() to use rdtsc */ tsc_disabled = 0; + if (!no_sched_irq_time) + enable_sched_clock_irqtime(); + lpj = ((u64)tsc_khz * 1000); do_div(lpj, HZ); lpj_fine = lpj; -- cgit v1.2.3-70-g09d2 From 9717967c4b704ce344c954afb5bb160aa9c01c34 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Mon, 18 Oct 2010 13:47:48 -0700 Subject: x86: ioapic: Call free_irte only if interrupt remapping enabled On a system that support intr-rempping when booting with "intremap=off" [ 177.895501] BUG: unable to handle kernel NULL pointer dereference at 00000000000000f8 [ 177.913316] IP: [] free_irte+0x47/0xc0 ... [ 178.173326] Call Trace: [ 178.173574] [] destroy_irq+0x3a/0x75 [ 178.192934] [] arch_teardown_msi_irq+0xe/0x10 [ 178.193418] [] arch_teardown_msi_irqs+0x56/0x7f [ 178.213021] [] free_msi_irqs+0x8d/0xeb Call free_irte only when interrupt remapping is enabled. Signed-off-by: Yinghai Lu LKML-Reference: <4CBCB274.7010108@kernel.org> Signed-off-by: Thomas Gleixner --- arch/x86/kernel/apic/io_apic.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 20e47e04539..8ae808d110f 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -3109,7 +3109,8 @@ void destroy_irq(unsigned int irq) irq_set_status_flags(irq, IRQ_NOREQUEST|IRQ_NOPROBE); - free_irte(irq); + if (intr_remapping_enabled) + free_irte(irq); raw_spin_lock_irqsave(&vector_lock, flags); __clear_irq_vector(irq, cfg); raw_spin_unlock_irqrestore(&vector_lock, flags); -- cgit v1.2.3-70-g09d2 From 9581d442b9058d3699b4be568b6e5eae38a41493 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Tue, 19 Oct 2010 16:46:55 +0200 Subject: KVM: Fix fs/gs reload oops with invalid ldt kvm reloads the host's fs and gs blindly, however the underlying segment descriptors may be invalid due to the user modifying the ldt after loading them. Fix by using the safe accessors (loadsegment() and load_gs_index()) instead of home grown unsafe versions. This is CVE-2010-3698. KVM-Stable-Tag. Signed-off-by: Avi Kivity Signed-off-by: Marcelo Tosatti --- arch/x86/include/asm/kvm_host.h | 24 ------------------------ arch/x86/kvm/svm.c | 15 ++++++++++----- arch/x86/kvm/vmx.c | 24 +++++++++--------------- 3 files changed, 19 insertions(+), 44 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 502e53f999c..c52e2eb40a1 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -652,20 +652,6 @@ static inline struct kvm_mmu_page *page_header(hpa_t shadow_page) return (struct kvm_mmu_page *)page_private(page); } -static inline u16 kvm_read_fs(void) -{ - u16 seg; - asm("mov %%fs, %0" : "=g"(seg)); - return seg; -} - -static inline u16 kvm_read_gs(void) -{ - u16 seg; - asm("mov %%gs, %0" : "=g"(seg)); - return seg; -} - static inline u16 kvm_read_ldt(void) { u16 ldt; @@ -673,16 +659,6 @@ static inline u16 kvm_read_ldt(void) return ldt; } -static inline void kvm_load_fs(u16 sel) -{ - asm("mov %0, %%fs" : : "rm"(sel)); -} - -static inline void kvm_load_gs(u16 sel) -{ - asm("mov %0, %%gs" : : "rm"(sel)); -} - static inline void kvm_load_ldt(u16 sel) { asm("lldt %0" : : "rm"(sel)); diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 81ed28cb36e..8a3f9f64f86 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -3163,8 +3163,8 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) sync_lapic_to_cr8(vcpu); save_host_msrs(vcpu); - fs_selector = kvm_read_fs(); - gs_selector = kvm_read_gs(); + savesegment(fs, fs_selector); + savesegment(gs, gs_selector); ldt_selector = kvm_read_ldt(); svm->vmcb->save.cr2 = vcpu->arch.cr2; /* required for live migration with NPT */ @@ -3251,10 +3251,15 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) vcpu->arch.regs[VCPU_REGS_RSP] = svm->vmcb->save.rsp; vcpu->arch.regs[VCPU_REGS_RIP] = svm->vmcb->save.rip; - kvm_load_fs(fs_selector); - kvm_load_gs(gs_selector); - kvm_load_ldt(ldt_selector); load_host_msrs(vcpu); + loadsegment(fs, fs_selector); +#ifdef CONFIG_X86_64 + load_gs_index(gs_selector); + wrmsrl(MSR_KERNEL_GS_BASE, current->thread.gs); +#else + loadsegment(gs, gs_selector); +#endif + kvm_load_ldt(ldt_selector); reload_tss(vcpu); diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 49b25eee25a..7bddfab1201 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -803,7 +803,7 @@ static void vmx_save_host_state(struct kvm_vcpu *vcpu) */ vmx->host_state.ldt_sel = kvm_read_ldt(); vmx->host_state.gs_ldt_reload_needed = vmx->host_state.ldt_sel; - vmx->host_state.fs_sel = kvm_read_fs(); + savesegment(fs, vmx->host_state.fs_sel); if (!(vmx->host_state.fs_sel & 7)) { vmcs_write16(HOST_FS_SELECTOR, vmx->host_state.fs_sel); vmx->host_state.fs_reload_needed = 0; @@ -811,7 +811,7 @@ static void vmx_save_host_state(struct kvm_vcpu *vcpu) vmcs_write16(HOST_FS_SELECTOR, 0); vmx->host_state.fs_reload_needed = 1; } - vmx->host_state.gs_sel = kvm_read_gs(); + savesegment(gs, vmx->host_state.gs_sel); if (!(vmx->host_state.gs_sel & 7)) vmcs_write16(HOST_GS_SELECTOR, vmx->host_state.gs_sel); else { @@ -841,27 +841,21 @@ static void vmx_save_host_state(struct kvm_vcpu *vcpu) static void __vmx_load_host_state(struct vcpu_vmx *vmx) { - unsigned long flags; - if (!vmx->host_state.loaded) return; ++vmx->vcpu.stat.host_state_reload; vmx->host_state.loaded = 0; if (vmx->host_state.fs_reload_needed) - kvm_load_fs(vmx->host_state.fs_sel); + loadsegment(fs, vmx->host_state.fs_sel); if (vmx->host_state.gs_ldt_reload_needed) { kvm_load_ldt(vmx->host_state.ldt_sel); - /* - * If we have to reload gs, we must take care to - * preserve our gs base. - */ - local_irq_save(flags); - kvm_load_gs(vmx->host_state.gs_sel); #ifdef CONFIG_X86_64 - wrmsrl(MSR_GS_BASE, vmcs_readl(HOST_GS_BASE)); + load_gs_index(vmx->host_state.gs_sel); + wrmsrl(MSR_KERNEL_GS_BASE, current->thread.gs); +#else + loadsegment(gs, vmx->host_state.gs_sel); #endif - local_irq_restore(flags); } reload_tss(); #ifdef CONFIG_X86_64 @@ -2589,8 +2583,8 @@ static int vmx_vcpu_setup(struct vcpu_vmx *vmx) vmcs_write16(HOST_CS_SELECTOR, __KERNEL_CS); /* 22.2.4 */ vmcs_write16(HOST_DS_SELECTOR, __KERNEL_DS); /* 22.2.4 */ vmcs_write16(HOST_ES_SELECTOR, __KERNEL_DS); /* 22.2.4 */ - vmcs_write16(HOST_FS_SELECTOR, kvm_read_fs()); /* 22.2.4 */ - vmcs_write16(HOST_GS_SELECTOR, kvm_read_gs()); /* 22.2.4 */ + vmcs_write16(HOST_FS_SELECTOR, 0); /* 22.2.4 */ + vmcs_write16(HOST_GS_SELECTOR, 0); /* 22.2.4 */ vmcs_write16(HOST_SS_SELECTOR, __KERNEL_DS); /* 22.2.4 */ #ifdef CONFIG_X86_64 rdmsrl(MSR_FS_BASE, a); -- cgit v1.2.3-70-g09d2 From 44235dcde416104b8e1db7606c283f4c0149c760 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 14 Oct 2010 17:04:59 -0700 Subject: x86, mm: Fix bogus whitespace in sync_global_pgds() Whitespace cleanup only. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: H. Peter Anvin --- arch/x86/mm/init_64.c | 44 ++++++++++++++++++++++---------------------- 1 file changed, 22 insertions(+), 22 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 74f0f358b07..1ad7c0ff5d2 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -103,28 +103,28 @@ __setup("noexec32=", nonx32_setup); */ void sync_global_pgds(unsigned long start, unsigned long end) { - unsigned long address; - - for (address = start; address <= end; address += PGDIR_SIZE) { - const pgd_t *pgd_ref = pgd_offset_k(address); - unsigned long flags; - struct page *page; - - if (pgd_none(*pgd_ref)) - continue; - - spin_lock_irqsave(&pgd_lock, flags); - list_for_each_entry(page, &pgd_list, lru) { - pgd_t *pgd; - pgd = (pgd_t *)page_address(page) + pgd_index(address); - if (pgd_none(*pgd)) - set_pgd(pgd, *pgd_ref); - else - BUG_ON(pgd_page_vaddr(*pgd) - != pgd_page_vaddr(*pgd_ref)); - } - spin_unlock_irqrestore(&pgd_lock, flags); - } + unsigned long address; + + for (address = start; address <= end; address += PGDIR_SIZE) { + const pgd_t *pgd_ref = pgd_offset_k(address); + unsigned long flags; + struct page *page; + + if (pgd_none(*pgd_ref)) + continue; + + spin_lock_irqsave(&pgd_lock, flags); + list_for_each_entry(page, &pgd_list, lru) { + pgd_t *pgd; + pgd = (pgd_t *)page_address(page) + pgd_index(address); + if (pgd_none(*pgd)) + set_pgd(pgd, *pgd_ref); + else + BUG_ON(pgd_page_vaddr(*pgd) + != pgd_page_vaddr(*pgd_ref)); + } + spin_unlock_irqrestore(&pgd_lock, flags); + } } /* -- cgit v1.2.3-70-g09d2 From 617d34d9e5d8326ec8f188c616aa06ac59d083fe Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Tue, 21 Sep 2010 12:01:51 -0700 Subject: x86, mm: Hold mm->page_table_lock while doing vmalloc_sync Take mm->page_table_lock while syncing the vmalloc region. This prevents a race with the Xen pagetable pin/unpin code, which expects that the page_table_lock is already held. If this race occurs, then Xen can see an inconsistent page type (a page can either be read/write or a pagetable page, and pin/unpin converts it between them), which will cause either the pin or the set_p[gm]d to fail; either will crash the kernel. vmalloc_sync_all() should be called rarely, so this extra use of page_table_lock should not interfere with its normal users. The mm pointer is stashed in the pgd page's index field, as that won't be otherwise used for pgds. Reported-by: Ian Campbell Originally-by: Jan Beulich LKML-Reference: <4CB88A4C.1080305@goop.org> Signed-off-by: Jeremy Fitzhardinge Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/pgtable.h | 2 ++ arch/x86/mm/fault.c | 11 ++++++++++- arch/x86/mm/init_64.c | 7 +++++++ arch/x86/mm/pgtable.c | 20 +++++++++++++++++--- 4 files changed, 36 insertions(+), 4 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h index 2d0a33bd297..ada823a13c7 100644 --- a/arch/x86/include/asm/pgtable.h +++ b/arch/x86/include/asm/pgtable.h @@ -28,6 +28,8 @@ extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)]; extern spinlock_t pgd_lock; extern struct list_head pgd_list; +extern struct mm_struct *pgd_page_get_mm(struct page *page); + #ifdef CONFIG_PARAVIRT #include #else /* !CONFIG_PARAVIRT */ diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index caec22906d7..6c27c39f8a3 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -229,7 +229,16 @@ void vmalloc_sync_all(void) spin_lock_irqsave(&pgd_lock, flags); list_for_each_entry(page, &pgd_list, lru) { - if (!vmalloc_sync_one(page_address(page), address)) + spinlock_t *pgt_lock; + int ret; + + pgt_lock = &pgd_page_get_mm(page)->page_table_lock; + + spin_lock(pgt_lock); + ret = vmalloc_sync_one(page_address(page), address); + spin_unlock(pgt_lock); + + if (!ret) break; } spin_unlock_irqrestore(&pgd_lock, flags); diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 1ad7c0ff5d2..4d323fb770c 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -116,12 +116,19 @@ void sync_global_pgds(unsigned long start, unsigned long end) spin_lock_irqsave(&pgd_lock, flags); list_for_each_entry(page, &pgd_list, lru) { pgd_t *pgd; + spinlock_t *pgt_lock; + pgd = (pgd_t *)page_address(page) + pgd_index(address); + pgt_lock = &pgd_page_get_mm(page)->page_table_lock; + spin_lock(pgt_lock); + if (pgd_none(*pgd)) set_pgd(pgd, *pgd_ref); else BUG_ON(pgd_page_vaddr(*pgd) != pgd_page_vaddr(*pgd_ref)); + + spin_unlock(pgt_lock); } spin_unlock_irqrestore(&pgd_lock, flags); } diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c index 5c4ee422590..c70e57dbb49 100644 --- a/arch/x86/mm/pgtable.c +++ b/arch/x86/mm/pgtable.c @@ -87,7 +87,19 @@ static inline void pgd_list_del(pgd_t *pgd) #define UNSHARED_PTRS_PER_PGD \ (SHARED_KERNEL_PMD ? KERNEL_PGD_BOUNDARY : PTRS_PER_PGD) -static void pgd_ctor(pgd_t *pgd) + +static void pgd_set_mm(pgd_t *pgd, struct mm_struct *mm) +{ + BUILD_BUG_ON(sizeof(virt_to_page(pgd)->index) < sizeof(mm)); + virt_to_page(pgd)->index = (pgoff_t)mm; +} + +struct mm_struct *pgd_page_get_mm(struct page *page) +{ + return (struct mm_struct *)page->index; +} + +static void pgd_ctor(struct mm_struct *mm, pgd_t *pgd) { /* If the pgd points to a shared pagetable level (either the ptes in non-PAE, or shared PMD in PAE), then just copy the @@ -105,8 +117,10 @@ static void pgd_ctor(pgd_t *pgd) } /* list required to sync kernel mapping updates */ - if (!SHARED_KERNEL_PMD) + if (!SHARED_KERNEL_PMD) { + pgd_set_mm(pgd, mm); pgd_list_add(pgd); + } } static void pgd_dtor(pgd_t *pgd) @@ -272,7 +286,7 @@ pgd_t *pgd_alloc(struct mm_struct *mm) */ spin_lock_irqsave(&pgd_lock, flags); - pgd_ctor(pgd); + pgd_ctor(mm, pgd); pgd_prepopulate_pmd(mm, pgd, pmds); spin_unlock_irqrestore(&pgd_lock, flags); -- cgit v1.2.3-70-g09d2 From 3234282f33b29d349bcada40204fc7c8fda7fe72 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Tue, 19 Oct 2010 14:52:26 +0100 Subject: x86, asm: Fix CFI macro invocations to deal with shortcomings in gas gas prior to (perhaps) 2.16.90 has problems with passing non- parenthesized expressions containing spaces to macros. Spaces, however, get inserted by cpp between any macro expanding to a number and a subsequent + or -. For the +, current x86 gas then removes the space again (future gas may not do so), but for the - the space gets retained and is then considered a separator between macro arguments. Fix the respective definitions for both the - and + cases, so that they neither contain spaces nor make cpp insert any (the latter by adding seemingly redundant parentheses). Signed-off-by: Jan Beulich LKML-Reference: <4CBDBEBA020000780001E05A@vpn.id2.novell.com> Cc: Alexander van Heukelum Signed-off-by: H. Peter Anvin --- Kbuild | 2 +- arch/x86/include/asm/calling.h | 52 ++++++++++++++++++++------------------- arch/x86/include/asm/entry_arch.h | 19 +++----------- arch/x86/include/asm/segment.h | 32 ++++++++++++------------ arch/x86/kernel/asm-offsets_32.c | 4 +-- arch/x86/kernel/entry_32.S | 6 ++--- arch/x86/kernel/entry_64.S | 20 +++------------ 7 files changed, 56 insertions(+), 79 deletions(-) (limited to 'arch/x86') diff --git a/Kbuild b/Kbuild index e3737ad72b5..399593942a9 100644 --- a/Kbuild +++ b/Kbuild @@ -53,7 +53,7 @@ targets += arch/$(SRCARCH)/kernel/asm-offsets.s # Default sed regexp - multiline due to syntax constraints define sed-y "/^->/{s:->#\(.*\):/* \1 */:; \ - s:^->\([^ ]*\) [\$$#]*\([^ ]*\) \(.*\):#define \1 \2 /* \3 */:; \ + s:^->\([^ ]*\) [\$$#]*\([^ ]*\) \(.*\):#define \1 (\2) /* \3 */:; \ s:->::; p;}" endef diff --git a/arch/x86/include/asm/calling.h b/arch/x86/include/asm/calling.h index 0e63c9a2a8d..30af5a83216 100644 --- a/arch/x86/include/asm/calling.h +++ b/arch/x86/include/asm/calling.h @@ -48,36 +48,38 @@ For 32-bit we have the following conventions - kernel is built with /* - * 64-bit system call stack frame layout defines and helpers, - * for assembly code: + * 64-bit system call stack frame layout defines and helpers, for + * assembly code (note that the seemingly unnecessary parentheses + * are to prevent cpp from inserting spaces in expressions that get + * passed to macros): */ -#define R15 0 -#define R14 8 -#define R13 16 -#define R12 24 -#define RBP 32 -#define RBX 40 +#define R15 (0) +#define R14 (8) +#define R13 (16) +#define R12 (24) +#define RBP (32) +#define RBX (40) /* arguments: interrupts/non tracing syscalls only save up to here: */ -#define R11 48 -#define R10 56 -#define R9 64 -#define R8 72 -#define RAX 80 -#define RCX 88 -#define RDX 96 -#define RSI 104 -#define RDI 112 -#define ORIG_RAX 120 /* + error_code */ +#define R11 (48) +#define R10 (56) +#define R9 (64) +#define R8 (72) +#define RAX (80) +#define RCX (88) +#define RDX (96) +#define RSI (104) +#define RDI (112) +#define ORIG_RAX (120) /* + error_code */ /* end of arguments */ /* cpu exception frame or undefined in case of fast syscall: */ -#define RIP 128 -#define CS 136 -#define EFLAGS 144 -#define RSP 152 -#define SS 160 +#define RIP (128) +#define CS (136) +#define EFLAGS (144) +#define RSP (152) +#define SS (160) #define ARGOFFSET R11 #define SWFRAME ORIG_RAX @@ -111,7 +113,7 @@ For 32-bit we have the following conventions - kernel is built with .endif .endm -#define ARG_SKIP 9*8 +#define ARG_SKIP (9*8) .macro RESTORE_ARGS skiprax=0, addskip=0, skiprcx=0, skipr11=0, \ skipr8910=0, skiprdx=0 @@ -169,7 +171,7 @@ For 32-bit we have the following conventions - kernel is built with .endif .endm -#define REST_SKIP 6*8 +#define REST_SKIP (6*8) .macro SAVE_REST subq $REST_SKIP, %rsp diff --git a/arch/x86/include/asm/entry_arch.h b/arch/x86/include/asm/entry_arch.h index 8e8ec663a98..4d2966e7e85 100644 --- a/arch/x86/include/asm/entry_arch.h +++ b/arch/x86/include/asm/entry_arch.h @@ -16,22 +16,11 @@ BUILD_INTERRUPT(call_function_single_interrupt,CALL_FUNCTION_SINGLE_VECTOR) BUILD_INTERRUPT(irq_move_cleanup_interrupt,IRQ_MOVE_CLEANUP_VECTOR) BUILD_INTERRUPT(reboot_interrupt,REBOOT_VECTOR) -BUILD_INTERRUPT3(invalidate_interrupt0,INVALIDATE_TLB_VECTOR_START+0, - smp_invalidate_interrupt) -BUILD_INTERRUPT3(invalidate_interrupt1,INVALIDATE_TLB_VECTOR_START+1, - smp_invalidate_interrupt) -BUILD_INTERRUPT3(invalidate_interrupt2,INVALIDATE_TLB_VECTOR_START+2, - smp_invalidate_interrupt) -BUILD_INTERRUPT3(invalidate_interrupt3,INVALIDATE_TLB_VECTOR_START+3, - smp_invalidate_interrupt) -BUILD_INTERRUPT3(invalidate_interrupt4,INVALIDATE_TLB_VECTOR_START+4, - smp_invalidate_interrupt) -BUILD_INTERRUPT3(invalidate_interrupt5,INVALIDATE_TLB_VECTOR_START+5, - smp_invalidate_interrupt) -BUILD_INTERRUPT3(invalidate_interrupt6,INVALIDATE_TLB_VECTOR_START+6, - smp_invalidate_interrupt) -BUILD_INTERRUPT3(invalidate_interrupt7,INVALIDATE_TLB_VECTOR_START+7, +.irpc idx, "01234567" +BUILD_INTERRUPT3(invalidate_interrupt\idx, + (INVALIDATE_TLB_VECTOR_START)+\idx, smp_invalidate_interrupt) +.endr #endif BUILD_INTERRUPT(x86_platform_ipi, X86_PLATFORM_IPI_VECTOR) diff --git a/arch/x86/include/asm/segment.h b/arch/x86/include/asm/segment.h index 14e0ed86a6f..231f1c1d660 100644 --- a/arch/x86/include/asm/segment.h +++ b/arch/x86/include/asm/segment.h @@ -73,31 +73,31 @@ #define GDT_ENTRY_DEFAULT_USER_DS 15 -#define GDT_ENTRY_KERNEL_BASE 12 +#define GDT_ENTRY_KERNEL_BASE (12) -#define GDT_ENTRY_KERNEL_CS (GDT_ENTRY_KERNEL_BASE + 0) +#define GDT_ENTRY_KERNEL_CS (GDT_ENTRY_KERNEL_BASE+0) -#define GDT_ENTRY_KERNEL_DS (GDT_ENTRY_KERNEL_BASE + 1) +#define GDT_ENTRY_KERNEL_DS (GDT_ENTRY_KERNEL_BASE+1) -#define GDT_ENTRY_TSS (GDT_ENTRY_KERNEL_BASE + 4) -#define GDT_ENTRY_LDT (GDT_ENTRY_KERNEL_BASE + 5) +#define GDT_ENTRY_TSS (GDT_ENTRY_KERNEL_BASE+4) +#define GDT_ENTRY_LDT (GDT_ENTRY_KERNEL_BASE+5) -#define GDT_ENTRY_PNPBIOS_BASE (GDT_ENTRY_KERNEL_BASE + 6) -#define GDT_ENTRY_APMBIOS_BASE (GDT_ENTRY_KERNEL_BASE + 11) +#define GDT_ENTRY_PNPBIOS_BASE (GDT_ENTRY_KERNEL_BASE+6) +#define GDT_ENTRY_APMBIOS_BASE (GDT_ENTRY_KERNEL_BASE+11) -#define GDT_ENTRY_ESPFIX_SS (GDT_ENTRY_KERNEL_BASE + 14) -#define __ESPFIX_SS (GDT_ENTRY_ESPFIX_SS * 8) +#define GDT_ENTRY_ESPFIX_SS (GDT_ENTRY_KERNEL_BASE+14) +#define __ESPFIX_SS (GDT_ENTRY_ESPFIX_SS*8) -#define GDT_ENTRY_PERCPU (GDT_ENTRY_KERNEL_BASE + 15) +#define GDT_ENTRY_PERCPU (GDT_ENTRY_KERNEL_BASE+15) #ifdef CONFIG_SMP #define __KERNEL_PERCPU (GDT_ENTRY_PERCPU * 8) #else #define __KERNEL_PERCPU 0 #endif -#define GDT_ENTRY_STACK_CANARY (GDT_ENTRY_KERNEL_BASE + 16) +#define GDT_ENTRY_STACK_CANARY (GDT_ENTRY_KERNEL_BASE+16) #ifdef CONFIG_CC_STACKPROTECTOR -#define __KERNEL_STACK_CANARY (GDT_ENTRY_STACK_CANARY * 8) +#define __KERNEL_STACK_CANARY (GDT_ENTRY_STACK_CANARY*8) #else #define __KERNEL_STACK_CANARY 0 #endif @@ -182,10 +182,10 @@ #endif -#define __KERNEL_CS (GDT_ENTRY_KERNEL_CS * 8) -#define __KERNEL_DS (GDT_ENTRY_KERNEL_DS * 8) -#define __USER_DS (GDT_ENTRY_DEFAULT_USER_DS* 8 + 3) -#define __USER_CS (GDT_ENTRY_DEFAULT_USER_CS* 8 + 3) +#define __KERNEL_CS (GDT_ENTRY_KERNEL_CS*8) +#define __KERNEL_DS (GDT_ENTRY_KERNEL_DS*8) +#define __USER_DS (GDT_ENTRY_DEFAULT_USER_DS*8+3) +#define __USER_CS (GDT_ENTRY_DEFAULT_USER_CS*8+3) #ifndef CONFIG_PARAVIRT #define get_kernel_rpl() 0 #endif diff --git a/arch/x86/kernel/asm-offsets_32.c b/arch/x86/kernel/asm-offsets_32.c index dfdbf640389..1a4088dda37 100644 --- a/arch/x86/kernel/asm-offsets_32.c +++ b/arch/x86/kernel/asm-offsets_32.c @@ -99,9 +99,7 @@ void foo(void) DEFINE(PAGE_SIZE_asm, PAGE_SIZE); DEFINE(PAGE_SHIFT_asm, PAGE_SHIFT); - DEFINE(PTRS_PER_PTE, PTRS_PER_PTE); - DEFINE(PTRS_PER_PMD, PTRS_PER_PMD); - DEFINE(PTRS_PER_PGD, PTRS_PER_PGD); + DEFINE(THREAD_SIZE_asm, THREAD_SIZE); OFFSET(crypto_tfm_ctx_offset, crypto_tfm, __crt_ctx); diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S index 9fb188d7bc7..f73a4b881aa 100644 --- a/arch/x86/kernel/entry_32.S +++ b/arch/x86/kernel/entry_32.S @@ -382,20 +382,20 @@ sysenter_past_esp: * enough kernel state to call TRACE_IRQS_OFF can be called - but * we immediately enable interrupts at that point anyway. */ - pushl_cfi $(__USER_DS) + pushl_cfi $__USER_DS /*CFI_REL_OFFSET ss, 0*/ pushl_cfi %ebp CFI_REL_OFFSET esp, 0 pushfl_cfi orl $X86_EFLAGS_IF, (%esp) - pushl_cfi $(__USER_CS) + pushl_cfi $__USER_CS /*CFI_REL_OFFSET cs, 0*/ /* * Push current_thread_info()->sysenter_return to the stack. * A tiny bit of offset fixup is necessary - 4*4 means the 4 words * pushed above; +8 corresponds to copy_thread's esp0 setting. */ - pushl_cfi (TI_sysenter_return-THREAD_SIZE+8+4*4)(%esp) + pushl_cfi TI_sysenter_return-THREAD_SIZE_asm+8+4*4(%esp) CFI_REL_OFFSET eip, 0 pushl_cfi %eax diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index 8851a2bb8c0..9cc9a71f8fb 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -963,22 +963,10 @@ apicinterrupt X86_PLATFORM_IPI_VECTOR \ x86_platform_ipi smp_x86_platform_ipi #ifdef CONFIG_SMP -apicinterrupt INVALIDATE_TLB_VECTOR_START+0 \ - invalidate_interrupt0 smp_invalidate_interrupt -apicinterrupt INVALIDATE_TLB_VECTOR_START+1 \ - invalidate_interrupt1 smp_invalidate_interrupt -apicinterrupt INVALIDATE_TLB_VECTOR_START+2 \ - invalidate_interrupt2 smp_invalidate_interrupt -apicinterrupt INVALIDATE_TLB_VECTOR_START+3 \ - invalidate_interrupt3 smp_invalidate_interrupt -apicinterrupt INVALIDATE_TLB_VECTOR_START+4 \ - invalidate_interrupt4 smp_invalidate_interrupt -apicinterrupt INVALIDATE_TLB_VECTOR_START+5 \ - invalidate_interrupt5 smp_invalidate_interrupt -apicinterrupt INVALIDATE_TLB_VECTOR_START+6 \ - invalidate_interrupt6 smp_invalidate_interrupt -apicinterrupt INVALIDATE_TLB_VECTOR_START+7 \ - invalidate_interrupt7 smp_invalidate_interrupt +.irpc idx, "01234567" +apicinterrupt (INVALIDATE_TLB_VECTOR_START)+\idx \ + invalidate_interrupt\idx smp_invalidate_interrupt +.endr #endif apicinterrupt THRESHOLD_APIC_VECTOR \ -- cgit v1.2.3-70-g09d2 From a68c439b1966c91f0ef474e2bf275d6792312726 Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Wed, 6 Oct 2010 12:27:53 +0200 Subject: apic, x86: Check if EILVT APIC registers are available (AMD only) This patch implements checks for the availability of LVT entries (APIC500-530) and reserves it if used. The check becomes necessary since we want to let the BIOS provide the LVT offsets. The offsets should be determined by the subsystems using it like those for MCE threshold or IBS. On K8 only offset 0 (APIC500) and MCE interrupts are supported. Beginning with family 10h at least 4 offsets are available. Since offsets must be consistent for all cores, we keep track of the LVT offsets in software and reserve the offset for the same vector also to be used on other cores. An offset is freed by setting the entry to APIC_EILVT_MASKED. If the BIOS is right, there should be no conflicts. Otherwise a "[Firmware Bug]: ..." error message is generated. However, if software does not properly determines the offsets, it is not necessarily a BIOS bug. Signed-off-by: Robert Richter LKML-Reference: <1286360874-1471-2-git-send-email-robert.richter@amd.com> Signed-off-by: Ingo Molnar --- arch/x86/include/asm/apicdef.h | 1 + arch/x86/kernel/apic/apic.c | 83 +++++++++++++++++++++++++++++++++++++----- 2 files changed, 75 insertions(+), 9 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/apicdef.h b/arch/x86/include/asm/apicdef.h index 7fe3b3060f0..a859ca461fb 100644 --- a/arch/x86/include/asm/apicdef.h +++ b/arch/x86/include/asm/apicdef.h @@ -131,6 +131,7 @@ #define APIC_EILVTn(n) (0x500 + 0x10 * n) #define APIC_EILVT_NR_AMD_K8 1 /* # of extended interrupts */ #define APIC_EILVT_NR_AMD_10H 4 +#define APIC_EILVT_NR_MAX APIC_EILVT_NR_AMD_10H #define APIC_EILVT_LVTOFF(x) (((x) >> 4) & 0xF) #define APIC_EILVT_MSG_FIX 0x0 #define APIC_EILVT_MSG_SMI 0x2 diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index 8cf86fb3b4e..2bfeafd24f5 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -52,6 +52,7 @@ #include #include #include +#include unsigned int num_processors; @@ -370,24 +371,88 @@ static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen) } /* - * Setup extended LVT, AMD specific (K8, family 10h) + * Setup extended LVT, AMD specific * - * Vector mappings are hard coded. On K8 only offset 0 (APIC500) and - * MCE interrupts are supported. Thus MCE offset must be set to 0. + * Software should use the LVT offsets the BIOS provides. The offsets + * are determined by the subsystems using it like those for MCE + * threshold or IBS. On K8 only offset 0 (APIC500) and MCE interrupts + * are supported. Beginning with family 10h at least 4 offsets are + * available. * - * If mask=1, the LVT entry does not generate interrupts while mask=0 - * enables the vector. See also the BKDGs. + * Since the offsets must be consistent for all cores, we keep track + * of the LVT offsets in software and reserve the offset for the same + * vector also to be used on other cores. An offset is freed by + * setting the entry to APIC_EILVT_MASKED. + * + * If the BIOS is right, there should be no conflicts. Otherwise a + * "[Firmware Bug]: ..." error message is generated. However, if + * software does not properly determines the offsets, it is not + * necessarily a BIOS bug. */ #define APIC_EILVT_LVTOFF_MCE 0 #define APIC_EILVT_LVTOFF_IBS 1 -static void setup_APIC_eilvt(u8 lvt_off, u8 vector, u8 msg_type, u8 mask) +static atomic_t eilvt_offsets[APIC_EILVT_NR_MAX]; + +static inline int eilvt_entry_is_changeable(unsigned int old, unsigned int new) +{ + return (old & APIC_EILVT_MASKED) + || (new == APIC_EILVT_MASKED) + || ((new & ~APIC_EILVT_MASKED) == old); +} + +static unsigned int reserve_eilvt_offset(int offset, unsigned int new) +{ + unsigned int rsvd; /* 0: uninitialized */ + + if (offset >= APIC_EILVT_NR_MAX) + return ~0; + + rsvd = atomic_read(&eilvt_offsets[offset]) & ~APIC_EILVT_MASKED; + do { + if (rsvd && + !eilvt_entry_is_changeable(rsvd, new)) + /* may not change if vectors are different */ + return rsvd; + rsvd = atomic_cmpxchg(&eilvt_offsets[offset], rsvd, new); + } while (rsvd != new); + + return new; +} + +/* + * If mask=1, the LVT entry does not generate interrupts while mask=0 + * enables the vector. See also the BKDGs. + */ + +static int setup_APIC_eilvt(u8 offset, u8 vector, u8 msg_type, u8 mask) { - unsigned long reg = (lvt_off << 4) + APIC_EILVTn(0); - unsigned int v = (mask << 16) | (msg_type << 8) | vector; + unsigned long reg = APIC_EILVTn(offset); + unsigned int new, old, reserved; + + new = (mask << 16) | (msg_type << 8) | vector; + old = apic_read(reg); + reserved = reserve_eilvt_offset(offset, new); - apic_write(reg, v); + if (reserved != new) { + pr_err(FW_BUG "cpu %d, try to setup vector 0x%x, but " + "vector 0x%x was already reserved by another core, " + "APIC%lX=0x%x\n", + smp_processor_id(), new, reserved, reg, old); + return -EINVAL; + } + + if (!eilvt_entry_is_changeable(old, new)) { + pr_err(FW_BUG "cpu %d, try to setup vector 0x%x but " + "register already in use, APIC%lX=0x%x\n", + smp_processor_id(), new, reg, old); + return -EBUSY; + } + + apic_write(reg, new); + + return 0; } u8 setup_APIC_eilvt_mce(u8 vector, u8 msg_type, u8 mask) -- cgit v1.2.3-70-g09d2 From 27afdf2008da0b8878a73e32e4eb12381b84e224 Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Wed, 6 Oct 2010 12:27:54 +0200 Subject: apic, x86: Use BIOS settings for IBS and MCE threshold interrupt LVT offsets We want the BIOS to setup the EILVT APIC registers. The offsets were hardcoded and BIOS settings were overwritten by the OS. Now, the subsystems for MCE threshold and IBS determine the LVT offset from the registers the BIOS has setup. If the BIOS setup is buggy on a family 10h system, a workaround enables IBS. If the OS determines an invalid register setup, a "[Firmware Bug]: " error message is reported. We need this change also for upcomming cpu families. Signed-off-by: Robert Richter LKML-Reference: <1286360874-1471-3-git-send-email-robert.richter@amd.com> Signed-off-by: Ingo Molnar --- arch/x86/include/asm/apic.h | 4 +- arch/x86/kernel/apic/apic.c | 19 +---- arch/x86/kernel/cpu/mcheck/mce_amd.c | 27 ++++++- arch/x86/oprofile/op_model_amd.c | 145 ++++++++++++++++++++++++++++++----- 4 files changed, 154 insertions(+), 41 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h index 1fa03e04ae4..286de34b0ed 100644 --- a/arch/x86/include/asm/apic.h +++ b/arch/x86/include/asm/apic.h @@ -252,9 +252,7 @@ static inline int apic_is_clustered_box(void) } #endif -extern u8 setup_APIC_eilvt_mce(u8 vector, u8 msg_type, u8 mask); -extern u8 setup_APIC_eilvt_ibs(u8 vector, u8 msg_type, u8 mask); - +extern int setup_APIC_eilvt(u8 lvt_off, u8 vector, u8 msg_type, u8 mask); #else /* !CONFIG_X86_LOCAL_APIC */ static inline void lapic_shutdown(void) { } diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index 2bfeafd24f5..850657d1b0e 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -390,9 +390,6 @@ static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen) * necessarily a BIOS bug. */ -#define APIC_EILVT_LVTOFF_MCE 0 -#define APIC_EILVT_LVTOFF_IBS 1 - static atomic_t eilvt_offsets[APIC_EILVT_NR_MAX]; static inline int eilvt_entry_is_changeable(unsigned int old, unsigned int new) @@ -426,7 +423,7 @@ static unsigned int reserve_eilvt_offset(int offset, unsigned int new) * enables the vector. See also the BKDGs. */ -static int setup_APIC_eilvt(u8 offset, u8 vector, u8 msg_type, u8 mask) +int setup_APIC_eilvt(u8 offset, u8 vector, u8 msg_type, u8 mask) { unsigned long reg = APIC_EILVTn(offset); unsigned int new, old, reserved; @@ -454,19 +451,7 @@ static int setup_APIC_eilvt(u8 offset, u8 vector, u8 msg_type, u8 mask) return 0; } - -u8 setup_APIC_eilvt_mce(u8 vector, u8 msg_type, u8 mask) -{ - setup_APIC_eilvt(APIC_EILVT_LVTOFF_MCE, vector, msg_type, mask); - return APIC_EILVT_LVTOFF_MCE; -} - -u8 setup_APIC_eilvt_ibs(u8 vector, u8 msg_type, u8 mask) -{ - setup_APIC_eilvt(APIC_EILVT_LVTOFF_IBS, vector, msg_type, mask); - return APIC_EILVT_LVTOFF_IBS; -} -EXPORT_SYMBOL_GPL(setup_APIC_eilvt_ibs); +EXPORT_SYMBOL_GPL(setup_APIC_eilvt); /* * Program the next event, relative to now diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd.c b/arch/x86/kernel/cpu/mcheck/mce_amd.c index 39aaee5c1ab..80c482382d5 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_amd.c +++ b/arch/x86/kernel/cpu/mcheck/mce_amd.c @@ -131,7 +131,8 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c) u32 low = 0, high = 0, address = 0; unsigned int bank, block; struct thresh_restart tr; - u8 lvt_off; + int lvt_off = -1; + u8 offset; for (bank = 0; bank < NR_BANKS; ++bank) { for (block = 0; block < NR_BLOCKS; ++block) { @@ -162,8 +163,28 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c) if (shared_bank[bank] && c->cpu_core_id) break; #endif - lvt_off = setup_APIC_eilvt_mce(THRESHOLD_APIC_VECTOR, - APIC_EILVT_MSG_FIX, 0); + offset = (high & MASK_LVTOFF_HI) >> 20; + if (lvt_off < 0) { + if (setup_APIC_eilvt(offset, + THRESHOLD_APIC_VECTOR, + APIC_EILVT_MSG_FIX, 0)) { + pr_err(FW_BUG "cpu %d, failed to " + "setup threshold interrupt " + "for bank %d, block %d " + "(MSR%08X=0x%x%08x)", + smp_processor_id(), bank, block, + address, high, low); + continue; + } + lvt_off = offset; + } else if (lvt_off != offset) { + pr_err(FW_BUG "cpu %d, invalid threshold " + "interrupt offset %d for bank %d," + "block %d (MSR%08X=0x%x%08x)", + smp_processor_id(), lvt_off, bank, + block, address, high, low); + continue; + } high &= ~MASK_LVTOFF_HI; high |= lvt_off << 20; diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c index b67a6b5aa8d..42fb46f8388 100644 --- a/arch/x86/oprofile/op_model_amd.c +++ b/arch/x86/oprofile/op_model_amd.c @@ -64,15 +64,22 @@ static u64 ibs_op_ctl; * IBS cpuid feature detection */ -#define IBS_CPUID_FEATURES 0x8000001b +#define IBS_CPUID_FEATURES 0x8000001b /* * Same bit mask as for IBS cpuid feature flags (Fn8000_001B_EAX), but * bit 0 is used to indicate the existence of IBS. */ -#define IBS_CAPS_AVAIL (1LL<<0) -#define IBS_CAPS_RDWROPCNT (1LL<<3) -#define IBS_CAPS_OPCNT (1LL<<4) +#define IBS_CAPS_AVAIL (1U<<0) +#define IBS_CAPS_RDWROPCNT (1U<<3) +#define IBS_CAPS_OPCNT (1U<<4) + +/* + * IBS APIC setup + */ +#define IBSCTL 0x1cc +#define IBSCTL_LVT_OFFSET_VALID (1ULL<<8) +#define IBSCTL_LVT_OFFSET_MASK 0x0F /* * IBS randomization macros @@ -266,6 +273,74 @@ static void op_amd_stop_ibs(void) wrmsrl(MSR_AMD64_IBSOPCTL, 0); } +static inline int eilvt_is_available(int offset) +{ + /* check if we may assign a vector */ + return !setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_NMI, 1); +} + +static inline int ibs_eilvt_valid(void) +{ + u64 val; + int offset; + + rdmsrl(MSR_AMD64_IBSCTL, val); + if (!(val & IBSCTL_LVT_OFFSET_VALID)) { + pr_err(FW_BUG "cpu %d, invalid IBS " + "interrupt offset %d (MSR%08X=0x%016llx)", + smp_processor_id(), offset, + MSR_AMD64_IBSCTL, val); + return 0; + } + + offset = val & IBSCTL_LVT_OFFSET_MASK; + + if (eilvt_is_available(offset)) + return !0; + + pr_err(FW_BUG "cpu %d, IBS interrupt offset %d " + "not available (MSR%08X=0x%016llx)", + smp_processor_id(), offset, + MSR_AMD64_IBSCTL, val); + + return 0; +} + +static inline int get_ibs_offset(void) +{ + u64 val; + + rdmsrl(MSR_AMD64_IBSCTL, val); + if (!(val & IBSCTL_LVT_OFFSET_VALID)) + return -EINVAL; + + return val & IBSCTL_LVT_OFFSET_MASK; +} + +static void setup_APIC_ibs(void) +{ + int offset; + + offset = get_ibs_offset(); + if (offset < 0) + goto failed; + + if (!setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_NMI, 0)) + return; +failed: + pr_warn("oprofile: IBS APIC setup failed on cpu #%d\n", + smp_processor_id()); +} + +static void clear_APIC_ibs(void) +{ + int offset; + + offset = get_ibs_offset(); + if (offset >= 0) + setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_FIX, 1); +} + #ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX static void op_mux_switch_ctrl(struct op_x86_model_spec const *model, @@ -376,13 +451,13 @@ static void op_amd_setup_ctrs(struct op_x86_model_spec const *model, } if (ibs_caps) - setup_APIC_eilvt_ibs(0, APIC_EILVT_MSG_NMI, 0); + setup_APIC_ibs(); } static void op_amd_cpu_shutdown(void) { if (ibs_caps) - setup_APIC_eilvt_ibs(0, APIC_EILVT_MSG_FIX, 1); + clear_APIC_ibs(); } static int op_amd_check_ctrs(struct pt_regs * const regs, @@ -445,16 +520,11 @@ static void op_amd_stop(struct op_msrs const * const msrs) op_amd_stop_ibs(); } -static int __init_ibs_nmi(void) +static int setup_ibs_ctl(int ibs_eilvt_off) { -#define IBSCTL_LVTOFFSETVAL (1 << 8) -#define IBSCTL 0x1cc struct pci_dev *cpu_cfg; int nodes; u32 value = 0; - u8 ibs_eilvt_off; - - ibs_eilvt_off = setup_APIC_eilvt_ibs(0, APIC_EILVT_MSG_FIX, 1); nodes = 0; cpu_cfg = NULL; @@ -466,21 +536,60 @@ static int __init_ibs_nmi(void) break; ++nodes; pci_write_config_dword(cpu_cfg, IBSCTL, ibs_eilvt_off - | IBSCTL_LVTOFFSETVAL); + | IBSCTL_LVT_OFFSET_VALID); pci_read_config_dword(cpu_cfg, IBSCTL, &value); - if (value != (ibs_eilvt_off | IBSCTL_LVTOFFSETVAL)) { + if (value != (ibs_eilvt_off | IBSCTL_LVT_OFFSET_VALID)) { pci_dev_put(cpu_cfg); printk(KERN_DEBUG "Failed to setup IBS LVT offset, " - "IBSCTL = 0x%08x", value); - return 1; + "IBSCTL = 0x%08x\n", value); + return -EINVAL; } } while (1); if (!nodes) { - printk(KERN_DEBUG "No CPU node configured for IBS"); - return 1; + printk(KERN_DEBUG "No CPU node configured for IBS\n"); + return -ENODEV; + } + + return 0; +} + +static int force_ibs_eilvt_setup(void) +{ + int i; + int ret; + + /* find the next free available EILVT entry */ + for (i = 1; i < 4; i++) { + if (!eilvt_is_available(i)) + continue; + ret = setup_ibs_ctl(i); + if (ret) + return ret; + return 0; } + printk(KERN_DEBUG "No EILVT entry available\n"); + + return -EBUSY; +} + +static int __init_ibs_nmi(void) +{ + int ret; + + if (ibs_eilvt_valid()) + return 0; + + ret = force_ibs_eilvt_setup(); + if (ret) + return ret; + + if (!ibs_eilvt_valid()) + return -EFAULT; + + pr_err(FW_BUG "workaround enabled for IBS LVT offset\n"); + return 0; } -- cgit v1.2.3-70-g09d2 From f01f7c56a1425b9749a99af821e1de334fb64d7e Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Tue, 19 Oct 2010 22:17:37 +0000 Subject: x86, mm: Fix incorrect data type in vmalloc_sync_all() arch/x86/mm/fault.c: In function 'vmalloc_sync_all': arch/x86/mm/fault.c:238: warning: assignment makes integer from pointer without a cast introduced by 617d34d9e5d8326ec8f188c616aa06ac59d083fe. Signed-off-by: Borislav Petkov LKML-Reference: <20101020103642.GA3135@kryptos.osrc.amd.com> Signed-off-by: H. Peter Anvin --- arch/x86/mm/fault.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 6c27c39f8a3..0cdb8d493f6 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -230,7 +230,7 @@ void vmalloc_sync_all(void) spin_lock_irqsave(&pgd_lock, flags); list_for_each_entry(page, &pgd_list, lru) { spinlock_t *pgt_lock; - int ret; + pmd_t *ret; pgt_lock = &pgd_page_get_mm(page)->page_table_lock; -- cgit v1.2.3-70-g09d2 From b40827fa7268fda8a62490728a61c2856f33830b Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Sat, 28 Aug 2010 15:58:33 +0200 Subject: x86-32, mm: Add an initial page table for core bootstrapping This patch adds an initial page table with low mappings used exclusively for booting APs/resuming after ACPI suspend/machine restart. After this, there's no need to add low mappings to swapper_pg_dir and zap them later or create own swsusp PGD page solely for ACPI sleep needs - we have initial_page_table for that. Signed-off-by: Borislav Petkov LKML-Reference: <20101020070526.GA9588@liondog.tnic> Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/pgtable_32.h | 2 +- arch/x86/include/asm/tlbflush.h | 2 -- arch/x86/include/asm/trampoline.h | 3 --- arch/x86/kernel/acpi/sleep.c | 7 ++++- arch/x86/kernel/head32.c | 1 + arch/x86/kernel/head_32.S | 55 ++++++++++++++++++--------------------- arch/x86/kernel/reboot.c | 10 ++----- arch/x86/kernel/setup.c | 18 ++++++++++++- arch/x86/kernel/smpboot.c | 16 +++--------- arch/x86/kernel/trampoline.c | 16 ------------ arch/x86/mm/init_32.c | 45 -------------------------------- 11 files changed, 56 insertions(+), 119 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/pgtable_32.h b/arch/x86/include/asm/pgtable_32.h index f686f49e8b7..8abde9ec90b 100644 --- a/arch/x86/include/asm/pgtable_32.h +++ b/arch/x86/include/asm/pgtable_32.h @@ -26,7 +26,7 @@ struct mm_struct; struct vm_area_struct; extern pgd_t swapper_pg_dir[1024]; -extern pgd_t trampoline_pg_dir[1024]; +extern pgd_t initial_page_table[1024]; static inline void pgtable_cache_init(void) { } static inline void check_pgt_cache(void) { } diff --git a/arch/x86/include/asm/tlbflush.h b/arch/x86/include/asm/tlbflush.h index 7f3eba08e7d..169be8938b9 100644 --- a/arch/x86/include/asm/tlbflush.h +++ b/arch/x86/include/asm/tlbflush.h @@ -172,6 +172,4 @@ static inline void flush_tlb_kernel_range(unsigned long start, flush_tlb_all(); } -extern void zap_low_mappings(bool early); - #endif /* _ASM_X86_TLBFLUSH_H */ diff --git a/arch/x86/include/asm/trampoline.h b/arch/x86/include/asm/trampoline.h index 4dde797c057..f4500fb3b48 100644 --- a/arch/x86/include/asm/trampoline.h +++ b/arch/x86/include/asm/trampoline.h @@ -13,16 +13,13 @@ extern unsigned char *trampoline_base; extern unsigned long init_rsp; extern unsigned long initial_code; -extern unsigned long initial_page_table; extern unsigned long initial_gs; #define TRAMPOLINE_SIZE roundup(trampoline_end - trampoline_data, PAGE_SIZE) extern unsigned long setup_trampoline(void); -extern void __init setup_trampoline_page_table(void); extern void __init reserve_trampoline_memory(void); #else -static inline void setup_trampoline_page_table(void) {} static inline void reserve_trampoline_memory(void) {} #endif /* CONFIG_X86_TRAMPOLINE */ diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c index 33cec152070..b35e1ab8ba0 100644 --- a/arch/x86/kernel/acpi/sleep.c +++ b/arch/x86/kernel/acpi/sleep.c @@ -12,6 +12,11 @@ #include #include +#ifdef CONFIG_X86_32 +#include +#include +#endif + #include "realmode/wakeup.h" #include "sleep.h" @@ -90,7 +95,7 @@ int acpi_save_state_mem(void) #ifndef CONFIG_64BIT header->pmode_entry = (u32)&wakeup_pmode_return; - header->pmode_cr3 = (u32)(swsusp_pg_dir - __PAGE_OFFSET); + header->pmode_cr3 = (u32)__pa(&initial_page_table); saved_magic = 0x12345678; #else /* CONFIG_64BIT */ header->trampoline_segment = setup_trampoline() >> 4; diff --git a/arch/x86/kernel/head32.c b/arch/x86/kernel/head32.c index 784360c0625..8b9c2019fc0 100644 --- a/arch/x86/kernel/head32.c +++ b/arch/x86/kernel/head32.c @@ -17,6 +17,7 @@ #include #include #include +#include static void __init i386_default_early_setup(void) { diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S index fa8c1b8e09f..bcece91dd31 100644 --- a/arch/x86/kernel/head_32.S +++ b/arch/x86/kernel/head_32.S @@ -183,13 +183,12 @@ default_entry: #ifdef CONFIG_X86_PAE /* - * In PAE mode swapper_pg_dir is statically defined to contain enough - * entries to cover the VMSPLIT option (that is the top 1, 2 or 3 - * entries). The identity mapping is handled by pointing two PGD - * entries to the first kernel PMD. + * In PAE mode initial_page_table is statically defined to contain + * enough entries to cover the VMSPLIT option (that is the top 1, 2 or 3 + * entries). The identity mapping is handled by pointing two PGD entries + * to the first kernel PMD. * - * Note the upper half of each PMD or PTE are always zero at - * this stage. + * Note the upper half of each PMD or PTE are always zero at this stage. */ #define KPMDS (((-__PAGE_OFFSET) >> 30) & 3) /* Number of kernel PMDs */ @@ -197,7 +196,7 @@ default_entry: xorl %ebx,%ebx /* %ebx is kept at zero */ movl $pa(__brk_base), %edi - movl $pa(swapper_pg_pmd), %edx + movl $pa(initial_pg_pmd), %edx movl $PTE_IDENT_ATTR, %eax 10: leal PDE_IDENT_ATTR(%edi),%ecx /* Create PMD entry */ @@ -226,14 +225,14 @@ default_entry: movl %eax, pa(max_pfn_mapped) /* Do early initialization of the fixmap area */ - movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR,%eax - movl %eax,pa(swapper_pg_pmd+0x1000*KPMDS-8) + movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,%eax + movl %eax,pa(initial_pg_pmd+0x1000*KPMDS-8) #else /* Not PAE */ page_pde_offset = (__PAGE_OFFSET >> 20); movl $pa(__brk_base), %edi - movl $pa(swapper_pg_dir), %edx + movl $pa(initial_page_table), %edx movl $PTE_IDENT_ATTR, %eax 10: leal PDE_IDENT_ATTR(%edi),%ecx /* Create PDE entry */ @@ -257,8 +256,8 @@ page_pde_offset = (__PAGE_OFFSET >> 20); movl %eax, pa(max_pfn_mapped) /* Do early initialization of the fixmap area */ - movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR,%eax - movl %eax,pa(swapper_pg_dir+0xffc) + movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,%eax + movl %eax,pa(initial_page_table+0xffc) #endif jmp 3f /* @@ -334,7 +333,7 @@ ENTRY(startup_32_smp) /* * Enable paging */ - movl pa(initial_page_table), %eax + movl $pa(initial_page_table), %eax movl %eax,%cr3 /* set the page table pointer.. */ movl %cr0,%eax orl $X86_CR0_PG,%eax @@ -614,8 +613,6 @@ ignore_int: .align 4 ENTRY(initial_code) .long i386_start_kernel -ENTRY(initial_page_table) - .long pa(swapper_pg_dir) /* * BSS section @@ -623,20 +620,18 @@ ENTRY(initial_page_table) __PAGE_ALIGNED_BSS .align PAGE_SIZE_asm #ifdef CONFIG_X86_PAE -swapper_pg_pmd: +initial_pg_pmd: .fill 1024*KPMDS,4,0 #else -ENTRY(swapper_pg_dir) +ENTRY(initial_page_table) .fill 1024,4,0 #endif -swapper_pg_fixmap: +initial_pg_fixmap: .fill 1024,4,0 -#ifdef CONFIG_X86_TRAMPOLINE -ENTRY(trampoline_pg_dir) - .fill 1024,4,0 -#endif ENTRY(empty_zero_page) .fill 4096,1,0 +ENTRY(swapper_pg_dir) + .fill 1024,4,0 /* * This starts the data section. @@ -645,20 +640,20 @@ ENTRY(empty_zero_page) __PAGE_ALIGNED_DATA /* Page-aligned for the benefit of paravirt? */ .align PAGE_SIZE_asm -ENTRY(swapper_pg_dir) - .long pa(swapper_pg_pmd+PGD_IDENT_ATTR),0 /* low identity map */ +ENTRY(initial_page_table) + .long pa(initial_pg_pmd+PGD_IDENT_ATTR),0 /* low identity map */ # if KPMDS == 3 - .long pa(swapper_pg_pmd+PGD_IDENT_ATTR),0 - .long pa(swapper_pg_pmd+PGD_IDENT_ATTR+0x1000),0 - .long pa(swapper_pg_pmd+PGD_IDENT_ATTR+0x2000),0 + .long pa(initial_pg_pmd+PGD_IDENT_ATTR),0 + .long pa(initial_pg_pmd+PGD_IDENT_ATTR+0x1000),0 + .long pa(initial_pg_pmd+PGD_IDENT_ATTR+0x2000),0 # elif KPMDS == 2 .long 0,0 - .long pa(swapper_pg_pmd+PGD_IDENT_ATTR),0 - .long pa(swapper_pg_pmd+PGD_IDENT_ATTR+0x1000),0 + .long pa(initial_pg_pmd+PGD_IDENT_ATTR),0 + .long pa(initial_pg_pmd+PGD_IDENT_ATTR+0x1000),0 # elif KPMDS == 1 .long 0,0 .long 0,0 - .long pa(swapper_pg_pmd+PGD_IDENT_ATTR),0 + .long pa(initial_pg_pmd+PGD_IDENT_ATTR),0 # else # error "Kernel PMDs should be 1, 2 or 3" # endif diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 7a4cf14223b..f7f53dcd3e0 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -371,16 +371,10 @@ void machine_real_restart(const unsigned char *code, int length) CMOS_WRITE(0x00, 0x8f); spin_unlock(&rtc_lock); - /* Remap the kernel at virtual address zero, as well as offset zero - from the kernel segment. This assumes the kernel segment starts at - virtual address PAGE_OFFSET. */ - memcpy(swapper_pg_dir, swapper_pg_dir + KERNEL_PGD_BOUNDARY, - sizeof(swapper_pg_dir [0]) * KERNEL_PGD_PTRS); - /* - * Use `swapper_pg_dir' as our page directory. + * Switch back to the initial page table. */ - load_cr3(swapper_pg_dir); + load_cr3(initial_page_table); /* Write 0x1234 to absolute memory location 0x472. The BIOS reads this on booting to tell it to "Bypass memory test (also warm diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 322b24fbeaf..af6cf2bfcee 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -728,6 +728,17 @@ void __init setup_arch(char **cmdline_p) #ifdef CONFIG_X86_32 memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data)); visws_early_detect(); + + /* + * copy kernel address range established so far and switch + * to the proper swapper page table + */ + clone_pgd_range(swapper_pg_dir + KERNEL_PGD_BOUNDARY, + initial_page_table + KERNEL_PGD_BOUNDARY, + KERNEL_PGD_PTRS); + + load_cr3(swapper_pg_dir); + __flush_tlb_all(); #else printk(KERN_INFO "Command line: %s\n", boot_command_line); #endif @@ -1009,7 +1020,12 @@ void __init setup_arch(char **cmdline_p) paging_init(); x86_init.paging.pagetable_setup_done(swapper_pg_dir); - setup_trampoline_page_table(); +#ifdef CONFIG_X86_32 + /* sync back kernel address range */ + clone_pgd_range(initial_page_table + KERNEL_PGD_BOUNDARY, + swapper_pg_dir + KERNEL_PGD_BOUNDARY, + KERNEL_PGD_PTRS); +#endif tboot_probe(); diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 63a1a5596ac..e63bb518585 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -298,22 +298,16 @@ notrace static void __cpuinit start_secondary(void *unused) * fragile that we want to limit the things done here to the * most necessary things. */ + cpu_init(); + preempt_disable(); + smp_callin(); #ifdef CONFIG_X86_32 - /* - * Switch away from the trampoline page-table - * - * Do this before cpu_init() because it needs to access per-cpu - * data which may not be mapped in the trampoline page-table. - */ + /* switch away from the initial page table */ load_cr3(swapper_pg_dir); __flush_tlb_all(); #endif - cpu_init(); - preempt_disable(); - smp_callin(); - /* otherwise gcc will move up smp_processor_id before the cpu_init */ barrier(); /* @@ -772,7 +766,6 @@ do_rest: #ifdef CONFIG_X86_32 /* Stack for startup_32 can be just as for start_secondary onwards */ irq_ctx_init(cpu); - initial_page_table = __pa(&trampoline_pg_dir); #else clear_tsk_thread_flag(c_idle.idle, TIF_FORK); initial_gs = per_cpu_offset(cpu); @@ -921,7 +914,6 @@ int __cpuinit native_cpu_up(unsigned int cpu) per_cpu(cpu_state, cpu) = CPU_UP_PREPARE; err = do_boot_cpu(apicid, cpu); - if (err) { pr_debug("do_boot_cpu failed %d\n", err); return -EIO; diff --git a/arch/x86/kernel/trampoline.c b/arch/x86/kernel/trampoline.c index e2a59525739..f1488a344cd 100644 --- a/arch/x86/kernel/trampoline.c +++ b/arch/x86/kernel/trampoline.c @@ -38,19 +38,3 @@ unsigned long __trampinit setup_trampoline(void) memcpy(trampoline_base, trampoline_data, TRAMPOLINE_SIZE); return virt_to_phys(trampoline_base); } - -void __init setup_trampoline_page_table(void) -{ -#ifdef CONFIG_X86_32 - /* Copy kernel address range */ - clone_pgd_range(trampoline_pg_dir + KERNEL_PGD_BOUNDARY, - swapper_pg_dir + KERNEL_PGD_BOUNDARY, - KERNEL_PGD_PTRS); - - /* Initialize low mappings */ - clone_pgd_range(trampoline_pg_dir, - swapper_pg_dir + KERNEL_PGD_BOUNDARY, - min_t(unsigned long, KERNEL_PGD_PTRS, - KERNEL_PGD_BOUNDARY)); -#endif -} diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index 558f2d33207..1aeac2d9df8 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -548,48 +548,6 @@ static void __init pagetable_init(void) permanent_kmaps_init(pgd_base); } -#ifdef CONFIG_ACPI_SLEEP -/* - * ACPI suspend needs this for resume, because things like the intel-agp - * driver might have split up a kernel 4MB mapping. - */ -char swsusp_pg_dir[PAGE_SIZE] - __attribute__ ((aligned(PAGE_SIZE))); - -static inline void save_pg_dir(void) -{ - copy_page(swsusp_pg_dir, swapper_pg_dir); -} -#else /* !CONFIG_ACPI_SLEEP */ -static inline void save_pg_dir(void) -{ -} -#endif /* !CONFIG_ACPI_SLEEP */ - -void zap_low_mappings(bool early) -{ - int i; - - /* - * Zap initial low-memory mappings. - * - * Note that "pgd_clear()" doesn't do it for - * us, because pgd_clear() is a no-op on i386. - */ - for (i = 0; i < KERNEL_PGD_BOUNDARY; i++) { -#ifdef CONFIG_X86_PAE - set_pgd(swapper_pg_dir+i, __pgd(1 + __pa(empty_zero_page))); -#else - set_pgd(swapper_pg_dir+i, __pgd(0)); -#endif - } - - if (early) - __flush_tlb(); - else - flush_tlb_all(); -} - pteval_t __supported_pte_mask __read_mostly = ~(_PAGE_NX | _PAGE_GLOBAL | _PAGE_IOMAP); EXPORT_SYMBOL_GPL(__supported_pte_mask); @@ -958,9 +916,6 @@ void __init mem_init(void) if (boot_cpu_data.wp_works_ok < 0) test_wp_bit(); - - save_pg_dir(); - zap_low_mappings(true); } #ifdef CONFIG_MEMORY_HOTPLUG -- cgit v1.2.3-70-g09d2 From 932967202182743c01a2eee4bdfa2c42697bc586 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Wed, 20 Oct 2010 11:07:03 +0800 Subject: x86: Spread tlb flush vector between nodes Currently flush tlb vector allocation is based on below equation: sender = smp_processor_id() % 8 This isn't optimal, CPUs from different node can have the same vector, this causes a lot of lock contention. Instead, we can assign the same vectors to CPUs from the same node, while different node has different vectors. This has below advantages: a. if there is lock contention, the lock contention is between CPUs from one node. This should be much cheaper than the contention between nodes. b. completely avoid lock contention between nodes. This especially benefits kswapd, which is the biggest user of tlb flush, since kswapd sets its affinity to specific node. In my test, this could reduce > 20% CPU overhead in extreme case.The test machine has 4 nodes and each node has 16 CPUs. I then bind each node's kswapd to the first CPU of the node. I run a workload with 4 sequential mmap file read thread. The files are empty sparse file. This workload will trigger a lot of page reclaim and tlbflush. The kswapd bind is to easy trigger the extreme tlb flush lock contention because otherwise kswapd keeps migrating between CPUs of a node and I can't get stable result. Sure in real workload, we can't always see so big tlb flush lock contention, but it's possible. [ hpa: folded in fix from Eric Dumazet to use this_cpu_read() ] Signed-off-by: Shaohua Li LKML-Reference: <1287544023.4571.8.camel@sli10-conroe.sh.intel.com> Cc: Eric Dumazet Signed-off-by: H. Peter Anvin --- arch/x86/mm/tlb.c | 48 +++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 47 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c index c03f14ab666..49358481c73 100644 --- a/arch/x86/mm/tlb.c +++ b/arch/x86/mm/tlb.c @@ -5,6 +5,7 @@ #include #include #include +#include #include #include @@ -52,6 +53,8 @@ union smp_flush_state { want false sharing in the per cpu data segment. */ static union smp_flush_state flush_state[NUM_INVALIDATE_TLB_VECTORS]; +static DEFINE_PER_CPU_READ_MOSTLY(int, tlb_vector_offset); + /* * We cannot call mmdrop() because we are in interrupt context, * instead update mm->cpu_vm_mask. @@ -173,7 +176,7 @@ static void flush_tlb_others_ipi(const struct cpumask *cpumask, union smp_flush_state *f; /* Caller has disabled preemption */ - sender = smp_processor_id() % NUM_INVALIDATE_TLB_VECTORS; + sender = this_cpu_read(tlb_vector_offset); f = &flush_state[sender]; /* @@ -218,6 +221,47 @@ void native_flush_tlb_others(const struct cpumask *cpumask, flush_tlb_others_ipi(cpumask, mm, va); } +static void __cpuinit calculate_tlb_offset(void) +{ + int cpu, node, nr_node_vecs; + /* + * we are changing tlb_vector_offset for each CPU in runtime, but this + * will not cause inconsistency, as the write is atomic under X86. we + * might see more lock contentions in a short time, but after all CPU's + * tlb_vector_offset are changed, everything should go normal + * + * Note: if NUM_INVALIDATE_TLB_VECTORS % nr_online_nodes !=0, we might + * waste some vectors. + **/ + if (nr_online_nodes > NUM_INVALIDATE_TLB_VECTORS) + nr_node_vecs = 1; + else + nr_node_vecs = NUM_INVALIDATE_TLB_VECTORS/nr_online_nodes; + + for_each_online_node(node) { + int node_offset = (node % NUM_INVALIDATE_TLB_VECTORS) * + nr_node_vecs; + int cpu_offset = 0; + for_each_cpu(cpu, cpumask_of_node(node)) { + per_cpu(tlb_vector_offset, cpu) = node_offset + + cpu_offset; + cpu_offset++; + cpu_offset = cpu_offset % nr_node_vecs; + } + } +} + +static int tlb_cpuhp_notify(struct notifier_block *n, + unsigned long action, void *hcpu) +{ + switch (action & 0xf) { + case CPU_ONLINE: + case CPU_DEAD: + calculate_tlb_offset(); + } + return NOTIFY_OK; +} + static int __cpuinit init_smp_flush(void) { int i; @@ -225,6 +269,8 @@ static int __cpuinit init_smp_flush(void) for (i = 0; i < ARRAY_SIZE(flush_state); i++) raw_spin_lock_init(&flush_state[i].tlbstate_lock); + calculate_tlb_offset(); + hotcpu_notifier(tlb_cpuhp_notify, 0); return 0; } core_initcall(init_smp_flush); -- cgit v1.2.3-70-g09d2 From 66f2b061546974b96b7b238a92ce89a87ecf0754 Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Wed, 20 Oct 2010 15:55:35 -0700 Subject: x86, mm: Enable ARCH_DMA_ADDR_T_64BIT with X86_64 || HIGHMEM64G Set CONFIG_ARCH_DMA_ADDR_T_64BIT when we set dma_addr_t to 64 bits in ; this allows Kconfig decisions based on this property. Signed-off-by: FUJITA Tomonori LKML-Reference: <201010202255.o9KMtZXu009370@imap1.linux-foundation.org> Acked-by: "H. Peter Anvin" Signed-off-by: Andrew Morton Signed-off-by: H. Peter Anvin --- arch/x86/Kconfig | 3 +++ 1 file changed, 3 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index cea0cd9a316..2924f4e7779 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -1148,6 +1148,9 @@ config X86_PAE config ARCH_PHYS_ADDR_T_64BIT def_bool X86_64 || X86_PAE +config ARCH_DMA_ADDR_T_64BIT + def_bool X86_64 || HIGHMEM64G + config DIRECT_GBPAGES bool "Enable 1GB pages for kernel pagetables" if EMBEDDED default y -- cgit v1.2.3-70-g09d2 From 07bd8516a2f967aa67904c68ab97bb896a448b09 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Fri, 22 Oct 2010 08:22:35 +0100 Subject: x86, asm: Restore parentheses around one pushl_cfi argument These were (intentionally) stripped by "fix CFI macro invocations to deal with shortcomings in gas" to expose problems with unexpected splitting of arguments by older gas also on newer versions, but as it turns out there is at least one distro (Ubuntu 6.06) where even not having *any* spaces in a macro argument doesn't reliably prevent splitting into multiple arguments. Signed-off-by: Jan Beulich Acked-by: Alexander van Heukelum LKML-Reference: <4CC157DB020000780001E8A2@vpn.id2.novell.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/entry_32.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S index f73a4b881aa..59e175e8959 100644 --- a/arch/x86/kernel/entry_32.S +++ b/arch/x86/kernel/entry_32.S @@ -395,7 +395,7 @@ sysenter_past_esp: * A tiny bit of offset fixup is necessary - 4*4 means the 4 words * pushed above; +8 corresponds to copy_thread's esp0 setting. */ - pushl_cfi TI_sysenter_return-THREAD_SIZE_asm+8+4*4(%esp) + pushl_cfi (TI_sysenter_return-THREAD_SIZE_asm+8+4*4)(%esp) CFI_REL_OFFSET eip, 0 pushl_cfi %eax -- cgit v1.2.3-70-g09d2 From fad99fac2627e2cc0ebfe07fcb5046c0b4e103f9 Mon Sep 17 00:00:00 2001 From: Jason Wessel Date: Wed, 20 Oct 2010 08:20:00 -0500 Subject: x86,kgdb: fix debugger hw breakpoint test regression in 2.6.35 HW breakpoints events stopped working correctly with kgdb as a result of commit: 018cbffe6819f6f8db20a0a3acd9bab9bfd667e4 (Merge commit 'v2.6.33' into perf/core), later commit: ba773f7c510c0b252145933926c636c439889207 (x86,kgdb: Fix hw breakpoint regression) allowed breakpoints to propagate to the debugger core but did not completely address the original regression in functionality found in 2.6.35. When the DR_STEP flag is set in dr6 along with any of the DR_TRAP bits, the kgdb exception handler will enter once from the hw_breakpoint API call back and again from the die notifier for do_debug(), which causes the debugger to stop twice and also for the kgdb regression tests to fail running under kvm with: echo V2I1 > /sys/module/kgdbts/parameters/kgdbts To address the problem, the kgdb overflow handler needs to implement the same logic as the ptrace overflow handler call back with respect to updating the virtual copy of dr6. This will allow the kgdb do_debug() die notifier to properly handle the exception and the attached debugger, or kgdb test suite, will only receive a single notification. Signed-off-by: Jason Wessel CC: Frederic Weisbecker CC: x86@kernel.org --- arch/x86/kernel/kgdb.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c index 852b81967a3..497f9738641 100644 --- a/arch/x86/kernel/kgdb.c +++ b/arch/x86/kernel/kgdb.c @@ -621,7 +621,12 @@ int kgdb_arch_init(void) static void kgdb_hw_overflow_handler(struct perf_event *event, int nmi, struct perf_sample_data *data, struct pt_regs *regs) { - kgdb_ll_trap(DIE_DEBUG, "debug", regs, 0, 0, SIGTRAP); + struct task_struct *tsk = current; + int i; + + for (i = 0; i < 4; i++) + if (breakinfo[i].enabled) + tsk->thread.debugreg6 |= (DR_TRAP0 << i); } void kgdb_arch_late(void) -- cgit v1.2.3-70-g09d2 From 91b152aa85bbcf076e269565394c31964f940371 Mon Sep 17 00:00:00 2001 From: Jason Wessel Date: Mon, 23 Aug 2010 09:20:14 -0500 Subject: kdb,kgdb: fix sparse fixups Fix the following sparse warnings: kdb_main.c:328:5: warning: symbol 'kdbgetu64arg' was not declared. Should it be static? kgdboc.c:246:12: warning: symbol 'kgdboc_early_init' was not declared. Should it be static? kgdb.c:652:26: warning: incorrect type in argument 1 (different address spaces) kgdb.c:652:26: expected void const *ptr kgdb.c:652:26: got struct perf_event *[noderef] *pev The one in kgdb.c required the (void * __force) because of the return code from register_wide_hw_breakpoint looking like: return (void __percpu __force *)ERR_PTR(err); Signed-off-by: Jason Wessel --- arch/x86/kernel/kgdb.c | 2 +- drivers/serial/kgdboc.c | 2 +- include/linux/kdb.h | 8 ++++++++ kernel/debug/kdb/kdb_private.h | 9 +-------- 4 files changed, 11 insertions(+), 10 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c index 497f9738641..101bf22cf16 100644 --- a/arch/x86/kernel/kgdb.c +++ b/arch/x86/kernel/kgdb.c @@ -649,7 +649,7 @@ void kgdb_arch_late(void) if (breakinfo[i].pev) continue; breakinfo[i].pev = register_wide_hw_breakpoint(&attr, NULL); - if (IS_ERR(breakinfo[i].pev)) { + if (IS_ERR((void * __force)breakinfo[i].pev)) { printk(KERN_ERR "kgdb: Could not allocate hw" "breakpoints\nDisabling the kernel debugger\n"); breakinfo[i].pev = NULL; diff --git a/drivers/serial/kgdboc.c b/drivers/serial/kgdboc.c index 39f9a1adaa7..d4b711c9a41 100644 --- a/drivers/serial/kgdboc.c +++ b/drivers/serial/kgdboc.c @@ -243,7 +243,7 @@ static struct kgdb_io kgdboc_io_ops = { #ifdef CONFIG_KGDB_SERIAL_CONSOLE /* This is only available if kgdboc is a built in for early debugging */ -int __init kgdboc_early_init(char *opt) +static int __init kgdboc_early_init(char *opt) { /* save the first character of the config string because the * init routine can destroy it. diff --git a/include/linux/kdb.h b/include/linux/kdb.h index deda197ced6..aadff7cc2b8 100644 --- a/include/linux/kdb.h +++ b/include/linux/kdb.h @@ -139,6 +139,14 @@ int kdb_process_cpu(const struct task_struct *p) /* kdb access to register set for stack dumping */ extern struct pt_regs *kdb_current_regs; +#ifdef CONFIG_KALLSYMS +extern const char *kdb_walk_kallsyms(loff_t *pos); +#else /* ! CONFIG_KALLSYMS */ +static inline const char *kdb_walk_kallsyms(loff_t *pos) +{ + return NULL; +} +#endif /* ! CONFIG_KALLSYMS */ /* Dynamic kdb shell command registration */ extern int kdb_register(char *, kdb_func_t, char *, char *, short); diff --git a/kernel/debug/kdb/kdb_private.h b/kernel/debug/kdb/kdb_private.h index 1921e6e4c0b..35d69ed1dfb 100644 --- a/kernel/debug/kdb/kdb_private.h +++ b/kernel/debug/kdb/kdb_private.h @@ -105,6 +105,7 @@ extern int kdb_getword(unsigned long *, unsigned long, size_t); extern int kdb_putword(unsigned long, unsigned long, size_t); extern int kdbgetularg(const char *, unsigned long *); +extern int kdbgetu64arg(const char *, u64 *); extern char *kdbgetenv(const char *); extern int kdbgetaddrarg(int, const char **, int*, unsigned long *, long *, char **); @@ -216,14 +217,6 @@ extern void kdb_ps1(const struct task_struct *p); extern void kdb_print_nameval(const char *name, unsigned long val); extern void kdb_send_sig_info(struct task_struct *p, struct siginfo *info); extern void kdb_meminfo_proc_show(void); -#ifdef CONFIG_KALLSYMS -extern const char *kdb_walk_kallsyms(loff_t *pos); -#else /* ! CONFIG_KALLSYMS */ -static inline const char *kdb_walk_kallsyms(loff_t *pos) -{ - return NULL; -} -#endif /* ! CONFIG_KALLSYMS */ extern char *kdb_getstr(char *, size_t, char *); /* Defines for kdb_symbol_print */ -- cgit v1.2.3-70-g09d2 From 39a0715f5ace92268190c89e246fd1cf741dbaea Mon Sep 17 00:00:00 2001 From: Dongdong Deng Date: Mon, 13 Sep 2010 06:58:00 -0500 Subject: x86,kgdb: remove unnecessary call to kgdb_correct_hw_break() The kernel debug_core invokes hw breakpoint install and removal via call backs. The architecture specific kgdb stubs only need to implement the call backs and not actually call the functions. Signed-off-by: Dongdong Deng Signed-off-by: Jason Wessel CC: x86@kernel.org CC: Thomas Gleixner CC: Ingo Molnar CC: H. Peter Anvin --- arch/x86/kernel/kgdb.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c index 101bf22cf16..d81cfebb848 100644 --- a/arch/x86/kernel/kgdb.c +++ b/arch/x86/kernel/kgdb.c @@ -477,8 +477,6 @@ int kgdb_arch_handle_exception(int e_vector, int signo, int err_code, raw_smp_processor_id()); } - kgdb_correct_hw_break(); - return 0; } -- cgit v1.2.3-70-g09d2