diff options
Diffstat (limited to 'arch/x86/kernel')
| -rw-r--r-- | arch/x86/kernel/apic/Makefile | 7 | ||||
| -rw-r--r-- | arch/x86/kernel/apic/hw_nmi.c | 107 | ||||
| -rw-r--r-- | arch/x86/kernel/apic/nmi.c | 7 | ||||
| -rw-r--r-- | arch/x86/kernel/cpu/perf_event.c | 62 | ||||
| -rw-r--r-- | arch/x86/kernel/cpu/perf_event_p4.c | 156 | ||||
| -rw-r--r-- | arch/x86/kernel/dumpstack.c | 1 | ||||
| -rw-r--r-- | arch/x86/kernel/dumpstack.h | 56 | ||||
| -rw-r--r-- | arch/x86/kernel/dumpstack_32.c | 2 | ||||
| -rw-r--r-- | arch/x86/kernel/dumpstack_64.c | 1 | ||||
| -rw-r--r-- | arch/x86/kernel/hw_breakpoint.c | 51 | ||||
| -rw-r--r-- | arch/x86/kernel/kprobes.c | 33 | ||||
| -rw-r--r-- | arch/x86/kernel/process_32.c | 4 | ||||
| -rw-r--r-- | arch/x86/kernel/process_64.c | 5 | ||||
| -rw-r--r-- | arch/x86/kernel/stacktrace.c | 31 | ||||
| -rw-r--r-- | arch/x86/kernel/traps.c | 7 |
15 files changed, 343 insertions, 187 deletions
diff --git a/arch/x86/kernel/apic/Makefile b/arch/x86/kernel/apic/Makefile index 565c1bf..910f20b4 100644 --- a/arch/x86/kernel/apic/Makefile +++ b/arch/x86/kernel/apic/Makefile @@ -2,7 +2,12 @@ # Makefile for local APIC drivers and for the IO-APIC code # -obj-$(CONFIG_X86_LOCAL_APIC) += apic.o apic_noop.o probe_$(BITS).o ipi.o nmi.o +obj-$(CONFIG_X86_LOCAL_APIC) += apic.o apic_noop.o probe_$(BITS).o ipi.o +ifneq ($(CONFIG_HARDLOCKUP_DETECTOR),y) +obj-$(CONFIG_X86_LOCAL_APIC) += nmi.o +endif +obj-$(CONFIG_HARDLOCKUP_DETECTOR) += hw_nmi.o + obj-$(CONFIG_X86_IO_APIC) += io_apic.o obj-$(CONFIG_SMP) += ipi.o diff --git a/arch/x86/kernel/apic/hw_nmi.c b/arch/x86/kernel/apic/hw_nmi.c new file mode 100644 index 0000000..cefd694 --- /dev/null +++ b/arch/x86/kernel/apic/hw_nmi.c @@ -0,0 +1,107 @@ +/* + * HW NMI watchdog support + * + * started by Don Zickus, Copyright (C) 2010 Red Hat, Inc. + * + * Arch specific calls to support NMI watchdog + * + * Bits copied from original nmi.c file + * + */ +#include <asm/apic.h> + +#include <linux/cpumask.h> +#include <linux/kdebug.h> +#include <linux/notifier.h> +#include <linux/kprobes.h> +#include <linux/nmi.h> +#include <linux/module.h> + +/* For reliability, we're prepared to waste bits here. */ +static DECLARE_BITMAP(backtrace_mask, NR_CPUS) __read_mostly; + +u64 hw_nmi_get_sample_period(void) +{ + return (u64)(cpu_khz) * 1000 * 60; +} + +#ifdef ARCH_HAS_NMI_WATCHDOG +void arch_trigger_all_cpu_backtrace(void) +{ + int i; + + cpumask_copy(to_cpumask(backtrace_mask), cpu_online_mask); + + printk(KERN_INFO "sending NMI to all CPUs:\n"); + apic->send_IPI_all(NMI_VECTOR); + + /* Wait for up to 10 seconds for all CPUs to do the backtrace */ + for (i = 0; i < 10 * 1000; i++) { + if (cpumask_empty(to_cpumask(backtrace_mask))) + break; + mdelay(1); + } +} + +static int __kprobes +arch_trigger_all_cpu_backtrace_handler(struct notifier_block *self, + unsigned long cmd, void *__args) +{ + struct die_args *args = __args; + struct pt_regs *regs; + int cpu = smp_processor_id(); + + switch (cmd) { + case DIE_NMI: + case DIE_NMI_IPI: + break; + + default: + return NOTIFY_DONE; + } + + regs = args->regs; + + if (cpumask_test_cpu(cpu, to_cpumask(backtrace_mask))) { + static arch_spinlock_t lock = __ARCH_SPIN_LOCK_UNLOCKED; + + arch_spin_lock(&lock); + printk(KERN_WARNING "NMI backtrace for cpu %d\n", cpu); + show_regs(regs); + dump_stack(); + arch_spin_unlock(&lock); + cpumask_clear_cpu(cpu, to_cpumask(backtrace_mask)); + return NOTIFY_STOP; + } + + return NOTIFY_DONE; +} + +static __read_mostly struct notifier_block backtrace_notifier = { + .notifier_call = arch_trigger_all_cpu_backtrace_handler, + .next = NULL, + .priority = 1 +}; + +static int __init register_trigger_all_cpu_backtrace(void) +{ + register_die_notifier(&backtrace_notifier); + return 0; +} +early_initcall(register_trigger_all_cpu_backtrace); +#endif + +/* STUB calls to mimic old nmi_watchdog behaviour */ +#if defined(CONFIG_X86_LOCAL_APIC) +unsigned int nmi_watchdog = NMI_NONE; +EXPORT_SYMBOL(nmi_watchdog); +void acpi_nmi_enable(void) { return; } +void acpi_nmi_disable(void) { return; } +#endif +atomic_t nmi_active = ATOMIC_INIT(0); /* oprofile uses this */ +EXPORT_SYMBOL(nmi_active); +int unknown_nmi_panic; +void cpu_nmi_set_wd_enabled(void) { return; } +void stop_apic_nmi_watchdog(void *unused) { return; } +void setup_apic_nmi_watchdog(void *unused) { return; } +int __init check_nmi_watchdog(void) { return 0; } diff --git a/arch/x86/kernel/apic/nmi.c b/arch/x86/kernel/apic/nmi.c index 1edaf15..a43f71c 100644 --- a/arch/x86/kernel/apic/nmi.c +++ b/arch/x86/kernel/apic/nmi.c @@ -401,13 +401,6 @@ nmi_watchdog_tick(struct pt_regs *regs, unsigned reason) int cpu = smp_processor_id(); int rc = 0; - /* check for other users first */ - if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) - == NOTIFY_STOP) { - rc = 1; - touched = 1; - } - sum = get_timer_irqs(cpu); if (__get_cpu_var(nmi_touch)) { diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 5db5b7d..f2da20f 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -220,6 +220,7 @@ struct x86_pmu { struct perf_event *event); struct event_constraint *event_constraints; void (*quirks)(void); + int perfctr_second_write; int (*cpu_prepare)(int cpu); void (*cpu_starting)(int cpu); @@ -295,10 +296,10 @@ x86_perf_event_update(struct perf_event *event) * count to the generic event atomically: */ again: - prev_raw_count = atomic64_read(&hwc->prev_count); + prev_raw_count = local64_read(&hwc->prev_count); rdmsrl(hwc->event_base + idx, new_raw_count); - if (atomic64_cmpxchg(&hwc->prev_count, prev_raw_count, + if (local64_cmpxchg(&hwc->prev_count, prev_raw_count, new_raw_count) != prev_raw_count) goto again; @@ -313,8 +314,8 @@ again: delta = (new_raw_count << shift) - (prev_raw_count << shift); delta >>= shift; - atomic64_add(delta, &event->count); - atomic64_sub(delta, &hwc->period_left); + local64_add(delta, &event->count); + local64_sub(delta, &hwc->period_left); return new_raw_count; } @@ -438,7 +439,7 @@ static int x86_setup_perfctr(struct perf_event *event) if (!hwc->sample_period) { hwc->sample_period = x86_pmu.max_period; hwc->last_period = hwc->sample_period; - atomic64_set(&hwc->period_left, hwc->sample_period); + local64_set(&hwc->period_left, hwc->sample_period); } else { /* * If we have a PMU initialized but no APIC @@ -885,7 +886,7 @@ static int x86_perf_event_set_period(struct perf_event *event) { struct hw_perf_event *hwc = &event->hw; - s64 left = atomic64_read(&hwc->period_left); + s64 left = local64_read(&hwc->period_left); s64 period = hwc->sample_period; int ret = 0, idx = hwc->idx; @@ -897,14 +898,14 @@ x86_perf_event_set_period(struct perf_event *event) */ if (unlikely(left <= -period)) { left = period; - atomic64_set(&hwc->period_left, left); + local64_set(&hwc->period_left, left); hwc->last_period = period; ret = 1; } if (unlikely(left <= 0)) { left += period; - atomic64_set(&hwc->period_left, left); + local64_set(&hwc->period_left, left); hwc->last_period = period; ret = 1; } @@ -923,10 +924,19 @@ x86_perf_event_set_period(struct perf_event *event) * The hw event starts counting from this event offset, * mark it to be able to extra future deltas: */ - atomic64_set(&hwc->prev_count, (u64)-left); + local64_set(&hwc->prev_count, (u64)-left); - wrmsrl(hwc->event_base + idx, + wrmsrl(hwc->event_base + idx, (u64)(-left) & x86_pmu.cntval_mask); + + /* + * Due to erratum on certan cpu we need + * a second write to be sure the register + * is updated properly + */ + if (x86_pmu.perfctr_second_write) { + wrmsrl(hwc->event_base + idx, (u64)(-left) & x86_pmu.cntval_mask); + } perf_event_update_userpage(event); @@ -969,7 +979,7 @@ static int x86_pmu_enable(struct perf_event *event) * skip the schedulability test here, it will be peformed * at commit time(->commit_txn) as a whole */ - if (cpuc->group_flag & PERF_EVENT_TXN_STARTED) + if (cpuc->group_flag & PERF_EVENT_TXN) goto out; ret = x86_pmu.schedule_events(cpuc, n, assign); @@ -1096,7 +1106,7 @@ static void x86_pmu_disable(struct perf_event *event) * The events never got scheduled and ->cancel_txn will truncate * the event_list. */ - if (cpuc->group_flag & PERF_EVENT_TXN_STARTED) + if (cpuc->group_flag & PERF_EVENT_TXN) return; x86_pmu_stop(event); @@ -1388,7 +1398,7 @@ static void x86_pmu_start_txn(const struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - cpuc->group_flag |= PERF_EVENT_TXN_STARTED; + cpuc->group_flag |= PERF_EVENT_TXN; cpuc->n_txn = 0; } @@ -1401,7 +1411,7 @@ static void x86_pmu_cancel_txn(const struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - cpuc->group_flag &= ~PERF_EVENT_TXN_STARTED; + cpuc->group_flag &= ~PERF_EVENT_TXN; /* * Truncate the collected events. */ @@ -1435,11 +1445,7 @@ static int x86_pmu_commit_txn(const struct pmu *pmu) */ memcpy(cpuc->assign, assign, n*sizeof(int)); - /* - * Clear out the txn count so that ->cancel_txn() which gets - * run after ->commit_txn() doesn't undo things. - */ - cpuc->n_txn = 0; + cpuc->group_flag &= ~PERF_EVENT_TXN; return 0; } @@ -1607,8 +1613,6 @@ static const struct stacktrace_ops backtrace_ops = { .walk_stack = print_context_stack_bp, }; -#include "../dumpstack.h" - static void perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry) { @@ -1730,22 +1734,6 @@ struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) return entry; } -void perf_arch_fetch_caller_regs(struct pt_regs *regs, unsigned long ip, int skip) -{ - regs->ip = ip; - /* - * perf_arch_fetch_caller_regs adds another call, we need to increment - * the skip level - */ - regs->bp = rewind_frame_pointer(skip + 1); - regs->cs = __KERNEL_CS; - /* - * We abuse bit 3 to pass exact information, see perf_misc_flags - * and the comment with PERF_EFLAGS_EXACT. - */ - regs->flags = 0; -} - unsigned long perf_instruction_pointer(struct pt_regs *regs) { unsigned long ip; diff --git a/arch/x86/kernel/cpu/perf_event_p4.c b/arch/x86/kernel/cpu/perf_event_p4.c index ae85d69..107711b 100644 --- a/arch/x86/kernel/cpu/perf_event_p4.c +++ b/arch/x86/kernel/cpu/perf_event_p4.c @@ -21,22 +21,36 @@ struct p4_event_bind { char cntr[2][P4_CNTR_LIMIT]; /* counter index (offset), -1 on abscence */ }; -struct p4_cache_event_bind { +struct p4_pebs_bind { unsigned int metric_pebs; unsigned int metric_vert; }; -#define P4_GEN_CACHE_EVENT_BIND(name) \ - [P4_CACHE__##name] = { \ - .metric_pebs = P4_PEBS__##name, \ - .metric_vert = P4_VERT__##name, \ +/* it sets P4_PEBS_ENABLE_UOP_TAG as well */ +#define P4_GEN_PEBS_BIND(name, pebs, vert) \ + [P4_PEBS_METRIC__##name] = { \ + .metric_pebs = pebs | P4_PEBS_ENABLE_UOP_TAG, \ + .metric_vert = vert, \ } -static struct p4_cache_event_bind p4_cache_event_bind_map[] = { - P4_GEN_CACHE_EVENT_BIND(1stl_cache_load_miss_retired), - P4_GEN_CACHE_EVENT_BIND(2ndl_cache_load_miss_retired), - P4_GEN_CACHE_EVENT_BIND(dtlb_load_miss_retired), - P4_GEN_CACHE_EVENT_BIND(dtlb_store_miss_retired), +/* + * note we have P4_PEBS_ENABLE_UOP_TAG always set here + * + * it's needed for mapping P4_PEBS_CONFIG_METRIC_MASK bits of + * event configuration to find out which values are to be + * written into MSR_IA32_PEBS_ENABLE and MSR_P4_PEBS_MATRIX_VERT + * resgisters + */ +static struct p4_pebs_bind p4_pebs_bind_map[] = { + P4_GEN_PEBS_BIND(1stl_cache_load_miss_retired, 0x0000001, 0x0000001), + P4_GEN_PEBS_BIND(2ndl_cache_load_miss_retired, 0x0000002, 0x0000001), + P4_GEN_PEBS_BIND(dtlb_load_miss_retired, 0x0000004, 0x0000001), + P4_GEN_PEBS_BIND(dtlb_store_miss_retired, 0x0000004, 0x0000002), + P4_GEN_PEBS_BIND(dtlb_all_miss_retired, 0x0000004, 0x0000003), + P4_GEN_PEBS_BIND(tagged_mispred_branch, 0x0018000, 0x0000010), + P4_GEN_PEBS_BIND(mob_load_replay_retired, 0x0000200, 0x0000001), + P4_GEN_PEBS_BIND(split_load_retired, 0x0000400, 0x0000001), + P4_GEN_PEBS_BIND(split_store_retired, 0x0000400, 0x0000002), }; /* @@ -281,10 +295,10 @@ static struct p4_event_bind p4_event_bind_map[] = { }, }; -#define P4_GEN_CACHE_EVENT(event, bit, cache_event) \ +#define P4_GEN_CACHE_EVENT(event, bit, metric) \ p4_config_pack_escr(P4_ESCR_EVENT(event) | \ P4_ESCR_EMASK_BIT(event, bit)) | \ - p4_config_pack_cccr(cache_event | \ + p4_config_pack_cccr(metric | \ P4_CCCR_ESEL(P4_OPCODE_ESEL(P4_OPCODE(event)))) static __initconst const u64 p4_hw_cache_event_ids @@ -296,34 +310,34 @@ static __initconst const u64 p4_hw_cache_event_ids [ C(OP_READ) ] = { [ C(RESULT_ACCESS) ] = 0x0, [ C(RESULT_MISS) ] = P4_GEN_CACHE_EVENT(P4_EVENT_REPLAY_EVENT, NBOGUS, - P4_CACHE__1stl_cache_load_miss_retired), + P4_PEBS_METRIC__1stl_cache_load_miss_retired), }, }, [ C(LL ) ] = { [ C(OP_READ) ] = { [ C(RESULT_ACCESS) ] = 0x0, [ C(RESULT_MISS) ] = P4_GEN_CACHE_EVENT(P4_EVENT_REPLAY_EVENT, NBOGUS, - P4_CACHE__2ndl_cache_load_miss_retired), + P4_PEBS_METRIC__2ndl_cache_load_miss_retired), }, }, [ C(DTLB) ] = { [ C(OP_READ) ] = { [ C(RESULT_ACCESS) ] = 0x0, [ C(RESULT_MISS) ] = P4_GEN_CACHE_EVENT(P4_EVENT_REPLAY_EVENT, NBOGUS, - P4_CACHE__dtlb_load_miss_retired), + P4_PEBS_METRIC__dtlb_load_miss_retired), }, [ C(OP_WRITE) ] = { [ C(RESULT_ACCESS) ] = 0x0, [ C(RESULT_MISS) ] = P4_GEN_CACHE_EVENT(P4_EVENT_REPLAY_EVENT, NBOGUS, - P4_CACHE__dtlb_store_miss_retired), + P4_PEBS_METRIC__dtlb_store_miss_retired), }, }, [ C(ITLB) ] = { [ C(OP_READ) ] = { [ C(RESULT_ACCESS) ] = P4_GEN_CACHE_EVENT(P4_EVENT_ITLB_REFERENCE, HIT, - P4_CACHE__itlb_reference_hit), + P4_PEBS_METRIC__none), [ C(RESULT_MISS) ] = P4_GEN_CACHE_EVENT(P4_EVENT_ITLB_REFERENCE, MISS, - P4_CACHE__itlb_reference_miss), + P4_PEBS_METRIC__none), }, [ C(OP_WRITE) ] = { [ C(RESULT_ACCESS) ] = -1, @@ -414,11 +428,37 @@ static u64 p4_pmu_event_map(int hw_event) return config; } +static int p4_validate_raw_event(struct perf_event *event) +{ + unsigned int v; + + /* 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); + return -EINVAL; + } + + /* + * it may have some screwed PEBS bits + */ + if (p4_config_pebs_has(event->attr.config, P4_PEBS_CONFIG_ENABLE)) { + pr_warning("P4 PMU: PEBS are not supported yet\n"); + 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); + return -EINVAL; + } + + return 0; +} + static int p4_hw_config(struct perf_event *event) { int cpu = get_cpu(); int rc = 0; - unsigned int evnt; u32 escr, cccr; /* @@ -438,12 +478,9 @@ static int p4_hw_config(struct perf_event *event) if (event->attr.type == PERF_TYPE_RAW) { - /* user data may have out-of-bound event index */ - evnt = p4_config_unpack_event(event->attr.config); - if (evnt >= ARRAY_SIZE(p4_event_bind_map)) { - rc = -EINVAL; + rc = p4_validate_raw_event(event); + if (rc) goto out; - } /* * We don't control raw events so it's up to the caller @@ -451,12 +488,15 @@ static int p4_hw_config(struct perf_event *event) * 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_config_pack_cccr(P4_CCCR_MASK_HT | P4_CCCR_RESERVED)); } rc = x86_setup_perfctr(event); @@ -482,6 +522,29 @@ static inline int p4_pmu_clear_cccr_ovf(struct hw_perf_event *hwc) return overflow; } +static void p4_pmu_disable_pebs(void) +{ + /* + * FIXME + * + * It's still allowed that two threads setup same cache + * events so we can't simply clear metrics until we knew + * noone is depending on us, so we need kind of counter + * for "ReplayEvent" users. + * + * What is more complex -- RAW events, if user (for some + * reason) will pass some cache event metric with improper + * event opcode -- it's fine from hardware point of view + * but completely nonsence from "meaning" of such action. + * + * So at moment let leave metrics turned on forever -- it's + * ok for now but need to be revisited! + * + * (void)checking_wrmsrl(MSR_IA32_PEBS_ENABLE, (u64)0); + * (void)checking_wrmsrl(MSR_P4_PEBS_MATRIX_VERT, (u64)0); + */ +} + static inline void p4_pmu_disable_event(struct perf_event *event) { struct hw_perf_event *hwc = &event->hw; @@ -507,6 +570,26 @@ static void p4_pmu_disable_all(void) continue; p4_pmu_disable_event(event); } + + p4_pmu_disable_pebs(); +} + +/* configuration must be valid */ +static void p4_pmu_enable_pebs(u64 config) +{ + struct p4_pebs_bind *bind; + unsigned int idx; + + BUILD_BUG_ON(P4_PEBS_METRIC__max > P4_PEBS_CONFIG_METRIC_MASK); + + idx = p4_config_unpack_metric(config); + if (idx == P4_PEBS_METRIC__none) + return; + + bind = &p4_pebs_bind_map[idx]; + + (void)checking_wrmsrl(MSR_IA32_PEBS_ENABLE, (u64)bind->metric_pebs); + (void)checking_wrmsrl(MSR_P4_PEBS_MATRIX_VERT, (u64)bind->metric_vert); } static void p4_pmu_enable_event(struct perf_event *event) @@ -515,9 +598,7 @@ static void p4_pmu_enable_event(struct perf_event *event) int thread = p4_ht_config_thread(hwc->config); u64 escr_conf = p4_config_unpack_escr(p4_clear_ht_bit(hwc->config)); unsigned int idx = p4_config_unpack_event(hwc->config); - unsigned int idx_cache = p4_config_unpack_cache_event(hwc->config); struct p4_event_bind *bind; - struct p4_cache_event_bind *bind_cache; u64 escr_addr, cccr; bind = &p4_event_bind_map[idx]; @@ -537,16 +618,10 @@ static void p4_pmu_enable_event(struct perf_event *event) cccr = p4_config_unpack_cccr(hwc->config); /* - * it could be Cache event so that we need to - * set metrics into additional MSRs + * it could be Cache event so we need to write metrics + * into additional MSRs */ - BUILD_BUG_ON(P4_CACHE__MAX > P4_CCCR_CACHE_OPS_MASK); - if (idx_cache > P4_CACHE__NONE && - idx_cache < ARRAY_SIZE(p4_cache_event_bind_map)) { - bind_cache = &p4_cache_event_bind_map[idx_cache]; - (void)checking_wrmsrl(MSR_IA32_PEBS_ENABLE, (u64)bind_cache->metric_pebs); - (void)checking_wrmsrl(MSR_P4_PEBS_MATRIX_VERT, (u64)bind_cache->metric_vert); - } + p4_pmu_enable_pebs(hwc->config); (void)checking_wrmsrl(escr_addr, escr_conf); (void)checking_wrmsrl(hwc->config_base + hwc->idx, @@ -829,6 +904,15 @@ static __initconst const struct x86_pmu p4_pmu = { .max_period = (1ULL << 39) - 1, .hw_config = p4_hw_config, .schedule_events = p4_pmu_schedule_events, + /* + * This handles erratum N15 in intel doc 249199-029, + * the counter may not be updated correctly on write + * so we need a second write operation to do the trick + * (the official workaround didn't work) + * + * the former idea is taken from OProfile code + */ + .perfctr_second_write = 1, }; static __init int p4_pmu_init(void) diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c index c89a386..6e8752c 100644 --- a/arch/x86/kernel/dumpstack.c +++ b/arch/x86/kernel/dumpstack.c @@ -18,7 +18,6 @@ #include <asm/stacktrace.h> -#include "dumpstack.h" int panic_on_unrecovered_nmi; int panic_on_io_nmi; diff --git a/arch/x86/kernel/dumpstack.h b/arch/x86/kernel/dumpstack.h deleted file mode 100644 index e1a93be..0000000 --- a/arch/x86/kernel/dumpstack.h +++ /dev/null @@ -1,56 +0,0 @@ -/* - * Copyright (C) 1991, 1992 Linus Torvalds - * Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs - */ - -#ifndef DUMPSTACK_H -#define DUMPSTACK_H - -#ifdef CONFIG_X86_32 -#define STACKSLOTS_PER_LINE 8 -#define get_bp(bp) asm("movl %%ebp, %0" : "=r" (bp) :) -#else -#define STACKSLOTS_PER_LINE 4 -#define get_bp(bp) asm("movq %%rbp, %0" : "=r" (bp) :) -#endif - -#include <linux/uaccess.h> - -extern void -show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, - unsigned long *stack, unsigned long bp, char *log_lvl); - -extern void -show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, - unsigned long *sp, unsigned long bp, char *log_lvl); - -extern unsigned int code_bytes; - -/* The form of the top of the frame on the stack */ -struct stack_frame { - struct stack_frame *next_frame; - unsigned long return_address; -}; - -struct stack_frame_ia32 { - u32 next_frame; - u32 return_address; -}; - -static inline unsigned long rewind_frame_pointer(int n) -{ - struct stack_frame *frame; - - get_bp(frame); - -#ifdef CONFIG_FRAME_POINTER - while (n--) { - if (probe_kernel_address(&frame->next_frame, frame)) - break; - } -#endif - - return (unsigned long)frame; -} - -#endif /* DUMPSTACK_H */ diff --git a/arch/x86/kernel/dumpstack_32.c b/arch/x86/kernel/dumpstack_32.c index 11540a1..0f6376f 100644 --- a/arch/x86/kernel/dumpstack_32.c +++ b/arch/x86/kernel/dumpstack_32.c @@ -16,8 +16,6 @@ #include <asm/stacktrace.h> -#include "dumpstack.h" - void dump_trace(struct task_struct *task, struct pt_regs *regs, unsigned long *stack, unsigned long bp, diff --git a/arch/x86/kernel/dumpstack_64.c b/arch/x86/kernel/dumpstack_64.c index 272c9f1..57a21f1 100644 --- a/arch/x86/kernel/dumpstack_64.c +++ b/arch/x86/kernel/dumpstack_64.c @@ -16,7 +16,6 @@ #include <asm/stacktrace.h> -#include "dumpstack.h" #define N_EXCEPTION_STACKS_END \ (N_EXCEPTION_STACKS + DEBUG_STKSZ/EXCEPTION_STKSZ - 2) diff --git a/arch/x86/kernel/hw_breakpoint.c b/arch/x86/kernel/hw_breakpoint.c index a8f1b80..a474ec3 100644 --- a/arch/x86/kernel/hw_breakpoint.c +++ b/arch/x86/kernel/hw_breakpoint.c @@ -208,6 +208,9 @@ int arch_bp_generic_fields(int x86_len, int x86_type, { /* Len */ switch (x86_len) { + case X86_BREAKPOINT_LEN_X: + *gen_len = sizeof(long); + break; case X86_BREAKPOINT_LEN_1: *gen_len = HW_BREAKPOINT_LEN_1; break; @@ -251,6 +254,29 @@ static int arch_build_bp_info(struct perf_event *bp) info->address = bp->attr.bp_addr; + /* Type */ + switch (bp->attr.bp_type) { + case HW_BREAKPOINT_W: + info->type = X86_BREAKPOINT_WRITE; + break; + case HW_BREAKPOINT_W | HW_BREAKPOINT_R: + info->type = X86_BREAKPOINT_RW; + break; + case HW_BREAKPOINT_X: + info->type = X86_BREAKPOINT_EXECUTE; + /* + * x86 inst breakpoints need to have a specific undefined len. + * But we still need to check userspace is not trying to setup + * an unsupported length, to get a range breakpoint for example. + */ + if (bp->attr.bp_len == sizeof(long)) { + info->len = X86_BREAKPOINT_LEN_X; + return 0; + } + default: + return -EINVAL; + } + /* Len */ switch (bp->attr.bp_len) { case HW_BREAKPOINT_LEN_1: @@ -271,21 +297,6 @@ static int arch_build_bp_info(struct perf_event *bp) return -EINVAL; } - /* Type */ - switch (bp->attr.bp_type) { - case HW_BREAKPOINT_W: - info->type = X86_BREAKPOINT_WRITE; - break; - case HW_BREAKPOINT_W | HW_BREAKPOINT_R: - info->type = X86_BREAKPOINT_RW; - break; - case HW_BREAKPOINT_X: - info->type = X86_BREAKPOINT_EXECUTE; - break; - default: - return -EINVAL; - } - return 0; } /* @@ -305,6 +316,9 @@ int arch_validate_hwbkpt_settings(struct perf_event *bp) ret = -EINVAL; switch (info->len) { + case X86_BREAKPOINT_LEN_X: + align = sizeof(long) -1; + break; case X86_BREAKPOINT_LEN_1: align = 0; break; @@ -466,6 +480,13 @@ static int __kprobes hw_breakpoint_handler(struct die_args *args) perf_bp_event(bp, args->regs); + /* + * Set up resume flag to avoid breakpoint recursion when + * returning back to origin. + */ + if (bp->hw.info.type == X86_BREAKPOINT_EXECUTE) + args->regs->flags |= X86_EFLAGS_RF; + rcu_read_unlock(); } /* diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c index 675879b..1bfb6cf 100644 --- a/arch/x86/kernel/kprobes.c +++ b/arch/x86/kernel/kprobes.c @@ -126,16 +126,22 @@ static void __kprobes synthesize_reljump(void *from, void *to) } /* - * Check for the REX prefix which can only exist on X86_64 - * X86_32 always returns 0 + * Skip the prefixes of the instruction. */ -static int __kprobes is_REX_prefix(kprobe_opcode_t *insn) +static kprobe_opcode_t *__kprobes skip_prefixes(kprobe_opcode_t *insn) { + insn_attr_t attr; + + attr = inat_get_opcode_attribute((insn_byte_t)*insn); + while (inat_is_legacy_prefix(attr)) { + insn++; + attr = inat_get_opcode_attribute((insn_byte_t)*insn); + } #ifdef CONFIG_X86_64 - if ((*insn & 0xf0) == 0x40) - return 1; + if (inat_is_rex_prefix(attr)) + insn++; #endif - return 0; + return insn; } /* @@ -272,6 +278,9 @@ static int __kprobes can_probe(unsigned long paddr) */ static int __kprobes is_IF_modifier(kprobe_opcode_t *insn) { + /* Skip prefixes */ + insn = skip_prefixes(insn); + switch (*insn) { case 0xfa: /* cli */ case 0xfb: /* sti */ @@ -280,13 +289,6 @@ static int __kprobes is_IF_modifier(kprobe_opcode_t *insn) return 1; } - /* - * on X86_64, 0x40-0x4f are REX prefixes so we need to look - * at the next byte instead.. but of course not recurse infinitely - */ - if (is_REX_prefix(insn)) - return is_IF_modifier(++insn); - return 0; } @@ -803,9 +805,8 @@ static void __kprobes resume_execution(struct kprobe *p, unsigned long orig_ip = (unsigned long)p->addr; kprobe_opcode_t *insn = p->ainsn.insn; - /*skip the REX prefix*/ - if (is_REX_prefix(insn)) - insn++; + /* Skip prefixes */ + insn = skip_prefixes(insn); regs->flags &= ~X86_EFLAGS_TF; switch (*insn) { diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index 8d12878..96586c3 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -57,6 +57,8 @@ #include <asm/syscalls.h> #include <asm/debugreg.h> +#include <trace/events/power.h> + asmlinkage void ret_from_fork(void) __asm__("ret_from_fork"); /* @@ -111,6 +113,8 @@ void cpu_idle(void) stop_critical_timings(); pm_idle(); start_critical_timings(); + + trace_power_end(smp_processor_id()); } tick_nohz_restart_sched_tick(); preempt_enable_no_resched(); diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 3c2422a..3d9ea53 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -51,6 +51,8 @@ #include <asm/syscalls.h> #include <asm/debugreg.h> +#include <trace/events/power.h> + asmlinkage extern void ret_from_fork(void); DEFINE_PER_CPU(unsigned long, old_rsp); @@ -138,6 +140,9 @@ void cpu_idle(void) stop_critical_timings(); pm_idle(); start_critical_timings(); + + trace_power_end(smp_processor_id()); + /* In many cases the interrupt that ended idle has already called exit_idle. But some idle loops can be woken up without interrupt. */ diff --git a/arch/x86/kernel/stacktrace.c b/arch/x86/kernel/stacktrace.c index 922eefb..b53c525 100644 --- a/arch/x86/kernel/stacktrace.c +++ b/arch/x86/kernel/stacktrace.c @@ -23,11 +23,16 @@ static int save_stack_stack(void *data, char *name) return 0; } -static void save_stack_address(void *data, unsigned long addr, int reliable) +static void +__save_stack_address(void *data, unsigned long addr, bool reliable, bool nosched) { struct stack_trace *trace = data; +#ifdef CONFIG_FRAME_POINTER if (!reliable) return; +#endif + if (nosched && in_sched_functions(addr)) + return; if (trace->skip > 0) { trace->skip--; return; @@ -36,20 +41,15 @@ static void save_stack_address(void *data, unsigned long addr, int reliable) trace->entries[trace->nr_entries++] = addr; } +static void save_stack_address(void *data, unsigned long addr, int reliable) +{ + return __save_stack_address(data, addr, reliable, false); +} + static void save_stack_address_nosched(void *data, unsigned long addr, int reliable) { - struct stack_trace *trace = (struct stack_trace *)data; - if (!reliable) - return; - if (in_sched_functions(addr)) - return; - if (trace->skip > 0) { - trace->skip--; - return; - } - if (trace->nr_entries < trace->max_entries) - trace->entries[trace->nr_entries++] = addr; + return __save_stack_address(data, addr, reliable, true); } static const struct stacktrace_ops save_stack_ops = { @@ -96,12 +96,13 @@ EXPORT_SYMBOL_GPL(save_stack_trace_tsk); /* Userspace stacktrace - based on kernel/trace/trace_sysprof.c */ -struct stack_frame { +struct stack_frame_user { const void __user *next_fp; unsigned long ret_addr; }; -static int copy_stack_frame(const void __user *fp, struct stack_frame *frame) +static int +copy_stack_frame(const void __user *fp, struct stack_frame_user *frame) { int ret; @@ -126,7 +127,7 @@ static inline void __save_stack_trace_user(struct stack_trace *trace) trace->entries[trace->nr_entries++] = regs->ip; while (trace->nr_entries < trace->max_entries) { - struct stack_frame frame; + struct stack_frame_user frame; frame.next_fp = NULL; frame.ret_addr = 0; diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index 725ef4d..60788de 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -392,7 +392,13 @@ static notrace __kprobes void default_do_nmi(struct pt_regs *regs) if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 2, SIGINT) == NOTIFY_STOP) return; + #ifdef CONFIG_X86_LOCAL_APIC + if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) + == NOTIFY_STOP) + return; + +#ifndef CONFIG_LOCKUP_DETECTOR /* * Ok, so this is none of the documented NMI sources, * so it must be the NMI watchdog. @@ -400,6 +406,7 @@ static notrace __kprobes void default_do_nmi(struct pt_regs *regs) if (nmi_watchdog_tick(regs, reason)) return; if (!do_nmi_callback(regs, cpu)) +#endif /* !CONFIG_LOCKUP_DETECTOR */ unknown_nmi_error(reason, regs); #else unknown_nmi_error(reason, regs); |
