diff options
36 files changed, 1177 insertions, 659 deletions
diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile index 6042981d030..1044fd787db 100644 --- a/arch/x86/kernel/cpu/Makefile +++ b/arch/x86/kernel/cpu/Makefile @@ -28,6 +28,11 @@ obj-$(CONFIG_CPU_SUP_UMC_32) += umc.o obj-$(CONFIG_PERF_EVENTS) += perf_event.o +ifdef CONFIG_PERF_EVENTS +obj-$(CONFIG_CPU_SUP_AMD) += perf_event_amd.o +obj-$(CONFIG_CPU_SUP_INTEL) += perf_event_p6.o perf_event_p4.o perf_event_intel_lbr.o perf_event_intel_ds.o perf_event_intel.o +endif + obj-$(CONFIG_X86_MCE) += mcheck/ obj-$(CONFIG_MTRR) += mtrr/ diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index cfa62ec090e..8ab89112f93 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -32,6 +32,8 @@ #include <asm/smp.h> #include <asm/alternative.h> +#include "perf_event.h" + #if 0 #undef wrmsrl #define wrmsrl(msr, val) \ @@ -43,283 +45,17 @@ do { \ } while (0) #endif -/* - * | NHM/WSM | SNB | - * register ------------------------------- - * | HT | no HT | HT | no HT | - *----------------------------------------- - * offcore | core | core | cpu | core | - * lbr_sel | core | core | cpu | core | - * ld_lat | cpu | core | cpu | core | - *----------------------------------------- - * - * Given that there is a small number of shared regs, - * we can pre-allocate their slot in the per-cpu - * per-core reg tables. - */ -enum extra_reg_type { - EXTRA_REG_NONE = -1, /* not used */ - - EXTRA_REG_RSP_0 = 0, /* offcore_response_0 */ - EXTRA_REG_RSP_1 = 1, /* offcore_response_1 */ - - EXTRA_REG_MAX /* number of entries needed */ -}; - -struct event_constraint { - union { - unsigned long idxmsk[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; - u64 idxmsk64; - }; - u64 code; - u64 cmask; - int weight; -}; - -struct amd_nb { - int nb_id; /* NorthBridge id */ - int refcnt; /* reference count */ - struct perf_event *owners[X86_PMC_IDX_MAX]; - struct event_constraint event_constraints[X86_PMC_IDX_MAX]; -}; - -struct intel_percore; - -#define MAX_LBR_ENTRIES 16 - -struct cpu_hw_events { - /* - * Generic x86 PMC bits - */ - struct perf_event *events[X86_PMC_IDX_MAX]; /* in counter order */ - unsigned long active_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; - unsigned long running[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; - int enabled; - - int n_events; - int n_added; - int n_txn; - int assign[X86_PMC_IDX_MAX]; /* event to counter assignment */ - u64 tags[X86_PMC_IDX_MAX]; - struct perf_event *event_list[X86_PMC_IDX_MAX]; /* in enabled order */ - - unsigned int group_flag; - - /* - * Intel DebugStore bits - */ - struct debug_store *ds; - u64 pebs_enabled; - - /* - * Intel LBR bits - */ - int lbr_users; - void *lbr_context; - struct perf_branch_stack lbr_stack; - struct perf_branch_entry lbr_entries[MAX_LBR_ENTRIES]; - - /* - * manage shared (per-core, per-cpu) registers - * used on Intel NHM/WSM/SNB - */ - struct intel_shared_regs *shared_regs; - - /* - * AMD specific bits - */ - struct amd_nb *amd_nb; -}; - -#define __EVENT_CONSTRAINT(c, n, m, w) {\ - { .idxmsk64 = (n) }, \ - .code = (c), \ - .cmask = (m), \ - .weight = (w), \ -} - -#define EVENT_CONSTRAINT(c, n, m) \ - __EVENT_CONSTRAINT(c, n, m, HWEIGHT(n)) - -/* - * Constraint on the Event code. - */ -#define INTEL_EVENT_CONSTRAINT(c, n) \ - EVENT_CONSTRAINT(c, n, ARCH_PERFMON_EVENTSEL_EVENT) - -/* - * Constraint on the Event code + UMask + fixed-mask - * - * filter mask to validate fixed counter events. - * the following filters disqualify for fixed counters: - * - inv - * - edge - * - cnt-mask - * The other filters are supported by fixed counters. - * The any-thread option is supported starting with v3. - */ -#define FIXED_EVENT_CONSTRAINT(c, n) \ - EVENT_CONSTRAINT(c, (1ULL << (32+n)), X86_RAW_EVENT_MASK) - -/* - * Constraint on the Event code + UMask - */ -#define INTEL_UEVENT_CONSTRAINT(c, n) \ - EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVENT_MASK) - -#define EVENT_CONSTRAINT_END \ - EVENT_CONSTRAINT(0, 0, 0) +struct x86_pmu x86_pmu __read_mostly; -#define for_each_event_constraint(e, c) \ - for ((e) = (c); (e)->weight; (e)++) - -/* - * Per register state. - */ -struct er_account { - raw_spinlock_t lock; /* per-core: protect structure */ - u64 config; /* extra MSR config */ - u64 reg; /* extra MSR number */ - atomic_t ref; /* reference count */ -}; - -/* - * Extra registers for specific events. - * - * Some events need large masks and require external MSRs. - * Those extra MSRs end up being shared for all events on - * a PMU and sometimes between PMU of sibling HT threads. - * In either case, the kernel needs to handle conflicting - * accesses to those extra, shared, regs. The data structure - * to manage those registers is stored in cpu_hw_event. - */ -struct extra_reg { - unsigned int event; - unsigned int msr; - u64 config_mask; - u64 valid_mask; - int idx; /* per_xxx->regs[] reg index */ -}; - -#define EVENT_EXTRA_REG(e, ms, m, vm, i) { \ - .event = (e), \ - .msr = (ms), \ - .config_mask = (m), \ - .valid_mask = (vm), \ - .idx = EXTRA_REG_##i \ - } - -#define INTEL_EVENT_EXTRA_REG(event, msr, vm, idx) \ - EVENT_EXTRA_REG(event, msr, ARCH_PERFMON_EVENTSEL_EVENT, vm, idx) - -#define EVENT_EXTRA_END EVENT_EXTRA_REG(0, 0, 0, 0, RSP_0) - -union perf_capabilities { - struct { - u64 lbr_format : 6; - u64 pebs_trap : 1; - u64 pebs_arch_reg : 1; - u64 pebs_format : 4; - u64 smm_freeze : 1; - }; - u64 capabilities; -}; - -/* - * struct x86_pmu - generic x86 pmu - */ -struct x86_pmu { - /* - * Generic x86 PMC bits - */ - const char *name; - int version; - int (*handle_irq)(struct pt_regs *); - void (*disable_all)(void); - void (*enable_all)(int added); - void (*enable)(struct perf_event *); - void (*disable)(struct perf_event *); - int (*hw_config)(struct perf_event *event); - int (*schedule_events)(struct cpu_hw_events *cpuc, int n, int *assign); - unsigned eventsel; - unsigned perfctr; - u64 (*event_map)(int); - int max_events; - int num_counters; - int num_counters_fixed; - int cntval_bits; - u64 cntval_mask; - int apic; - u64 max_period; - struct event_constraint * - (*get_event_constraints)(struct cpu_hw_events *cpuc, - struct perf_event *event); - - void (*put_event_constraints)(struct cpu_hw_events *cpuc, - 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); - void (*cpu_dying)(int cpu); - void (*cpu_dead)(int cpu); - - /* - * Intel Arch Perfmon v2+ - */ - u64 intel_ctrl; - union perf_capabilities intel_cap; - - /* - * Intel DebugStore bits - */ - int bts, pebs; - int bts_active, pebs_active; - int pebs_record_size; - void (*drain_pebs)(struct pt_regs *regs); - struct event_constraint *pebs_constraints; - - /* - * Intel LBR - */ - unsigned long lbr_tos, lbr_from, lbr_to; /* MSR base regs */ - int lbr_nr; /* hardware stack size */ - - /* - * Extra registers for events - */ - struct extra_reg *extra_regs; - unsigned int er_flags; -}; - -#define ERF_NO_HT_SHARING 1 -#define ERF_HAS_RSP_1 2 - -static struct x86_pmu x86_pmu __read_mostly; - -static DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = { +DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = { .enabled = 1, }; -static int x86_perf_event_set_period(struct perf_event *event); - -/* - * Generalized hw caching related hw_event table, filled - * in on a per model basis. A value of 0 means - * 'not supported', -1 means 'hw_event makes no sense on - * this CPU', any other value means the raw hw_event - * ID. - */ - -#define C(x) PERF_COUNT_HW_CACHE_##x - -static u64 __read_mostly hw_cache_event_ids +u64 __read_mostly hw_cache_event_ids [PERF_COUNT_HW_CACHE_MAX] [PERF_COUNT_HW_CACHE_OP_MAX] [PERF_COUNT_HW_CACHE_RESULT_MAX]; -static u64 __read_mostly hw_cache_extra_regs +u64 __read_mostly hw_cache_extra_regs [PERF_COUNT_HW_CACHE_MAX] [PERF_COUNT_HW_CACHE_OP_MAX] [PERF_COUNT_HW_CACHE_RESULT_MAX]; @@ -329,8 +65,7 @@ static u64 __read_mostly hw_cache_extra_regs * Can only be executed on the CPU where the event is active. * Returns the delta events processed. */ -static u64 -x86_perf_event_update(struct perf_event *event) +u64 x86_perf_event_update(struct perf_event *event) { struct hw_perf_event *hwc = &event->hw; int shift = 64 - x86_pmu.cntval_bits; @@ -373,30 +108,6 @@ again: return new_raw_count; } -static inline int x86_pmu_addr_offset(int index) -{ - int offset; - - /* offset = X86_FEATURE_PERFCTR_CORE ? index << 1 : index */ - alternative_io(ASM_NOP2, - "shll $1, %%eax", - X86_FEATURE_PERFCTR_CORE, - "=a" (offset), - "a" (index)); - - return offset; -} - -static inline unsigned int x86_pmu_config_addr(int index) -{ - return x86_pmu.eventsel + x86_pmu_addr_offset(index); -} - -static inline unsigned int x86_pmu_event_addr(int index) -{ - return x86_pmu.perfctr + x86_pmu_addr_offset(index); -} - /* * Find and validate any extra registers to set up. */ @@ -532,9 +243,6 @@ msr_fail: return false; } -static void reserve_ds_buffers(void); -static void release_ds_buffers(void); - static void hw_perf_event_destroy(struct perf_event *event) { if (atomic_dec_and_mutex_lock(&active_events, &pmc_reserve_mutex)) { @@ -583,7 +291,7 @@ set_ext_hw_attr(struct hw_perf_event *hwc, struct perf_event *event) return x86_pmu_extra_regs(val, event); } -static int x86_setup_perfctr(struct perf_event *event) +int x86_setup_perfctr(struct perf_event *event) { struct perf_event_attr *attr = &event->attr; struct hw_perf_event *hwc = &event->hw; @@ -647,7 +355,7 @@ static int x86_setup_perfctr(struct perf_event *event) return 0; } -static int x86_pmu_hw_config(struct perf_event *event) +int x86_pmu_hw_config(struct perf_event *event) { if (event->attr.precise_ip) { int precise = 0; @@ -723,7 +431,7 @@ static int __x86_pmu_event_init(struct perf_event *event) return x86_pmu.hw_config(event); } -static void x86_pmu_disable_all(void) +void x86_pmu_disable_all(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); int idx; @@ -758,15 +466,7 @@ static void x86_pmu_disable(struct pmu *pmu) x86_pmu.disable_all(); } -static inline void __x86_pmu_enable_event(struct hw_perf_event *hwc, - u64 enable_mask) -{ - if (hwc->extra_reg.reg) - wrmsrl(hwc->extra_reg.reg, hwc->extra_reg.config); - wrmsrl(hwc->config_base, hwc->config | enable_mask); -} - -static void x86_pmu_enable_all(int added) +void x86_pmu_enable_all(int added) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); int idx; @@ -788,7 +488,7 @@ static inline int is_x86_event(struct perf_event *event) return event->pmu == &pmu; } -static int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign) +int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign) { struct event_constraint *c, *constraints[X86_PMC_IDX_MAX]; unsigned long used_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; @@ -959,7 +659,6 @@ static inline int match_prev_assignment(struct hw_perf_event *hwc, } 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_enable(struct pmu *pmu) { @@ -1031,21 +730,13 @@ static void x86_pmu_enable(struct pmu *pmu) x86_pmu.enable_all(added); } -static inline void x86_pmu_disable_event(struct perf_event *event) -{ - struct hw_perf_event *hwc = &event->hw; - - wrmsrl(hwc->config_base, hwc->config); -} - static DEFINE_PER_CPU(u64 [X86_PMC_IDX_MAX], pmc_prev_left); /* * Set the next IRQ period, based on the hwc->period_left value. * To be called with the event disabled in hw: */ -static int -x86_perf_event_set_period(struct perf_event *event) +int x86_perf_event_set_period(struct perf_event *event) { struct hw_perf_event *hwc = &event->hw; s64 left = local64_read(&hwc->period_left); @@ -1105,7 +796,7 @@ x86_perf_event_set_period(struct perf_event *event) return ret; } -static void x86_pmu_enable_event(struct perf_event *event) +void x86_pmu_enable_event(struct perf_event *event) { if (__this_cpu_read(cpu_hw_events.enabled)) __x86_pmu_enable_event(&event->hw, @@ -1244,7 +935,7 @@ void perf_event_print_debug(void) local_irq_restore(flags); } -static void x86_pmu_stop(struct perf_event *event, int flags) +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; @@ -1297,7 +988,7 @@ static void x86_pmu_del(struct perf_event *event, int flags) perf_event_update_userpage(event); } -static int x86_pmu_handle_irq(struct pt_regs *regs) +int x86_pmu_handle_irq(struct pt_regs *regs) { struct perf_sample_data data; struct cpu_hw_events *cpuc; @@ -1437,39 +1128,19 @@ static __read_mostly struct notifier_block perf_event_nmi_notifier = { .priority = NMI_LOCAL_LOW_PRIOR, }; -static struct event_constraint unconstrained; -static struct event_constraint emptyconstraint; - -static struct event_constraint * -x86_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event) -{ - struct event_constraint *c; - - if (x86_pmu.event_constraints) { - for_each_event_constraint(c, x86_pmu.event_constraints) { - if ((event->hw.config & c->cmask) == c->code) - return c; - } - } - - return &unconstrained; -} - -#include "perf_event_amd.c" -#include "perf_event_p6.c" -#include "perf_event_p4.c" -#include "perf_event_intel_lbr.c" -#include "perf_event_intel_ds.c" -#include "perf_event_intel.c" +struct event_constraint emptyconstraint; +struct event_constraint unconstrained; static int __cpuinit x86_pmu_notifier(struct notifier_block *self, unsigned long action, void *hcpu) { unsigned int cpu = (long)hcpu; + struct cpu_hw_events *cpuc = &per_cpu(cpu_hw_events, cpu); int ret = NOTIFY_OK; switch (action & ~CPU_TASKS_FROZEN) { case CPU_UP_PREPARE: + cpuc->kfree_on_online = NULL; if (x86_pmu.cpu_prepare) ret = x86_pmu.cpu_prepare(cpu); break; @@ -1479,6 +1150,10 @@ x86_pmu_notifier(struct notifier_block *self, unsigned long action, void *hcpu) x86_pmu.cpu_starting(cpu); break; + case CPU_ONLINE: + kfree(cpuc->kfree_on_online); + break; + case CPU_DYING: if (x86_pmu.cpu_dying) x86_pmu.cpu_dying(cpu); diff --git a/arch/x86/kernel/cpu/perf_event.h b/arch/x86/kernel/cpu/perf_event.h new file mode 100644 index 00000000000..fb330b0a816 --- /dev/null +++ b/arch/x86/kernel/cpu/perf_event.h @@ -0,0 +1,493 @@ +/* + * Performance events x86 architecture header + * + * Copyright (C) 2008 Thomas Gleixner <tglx@linutronix.de> + * Copyright (C) 2008-2009 Red Hat, Inc., Ingo Molnar + * Copyright (C) 2009 Jaswinder Singh Rajput + * Copyright (C) 2009 Advanced Micro Devices, Inc., Robert Richter + * Copyright (C) 2008-2009 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com> + * Copyright (C) 2009 Intel Corporation, <markus.t.metzger@intel.com> + * Copyright (C) 2009 Google, Inc., Stephane Eranian + * + * For licencing details see kernel-base/COPYING + */ + +#include <linux/perf_event.h> + +/* + * | NHM/WSM | SNB | + * register ------------------------------- + * | HT | no HT | HT | no HT | + *----------------------------------------- + * offcore | core | core | cpu | core | + * lbr_sel | core | core | cpu | core | + * ld_lat | cpu | core | cpu | core | + *----------------------------------------- + * + * Given that there is a small number of shared regs, + * we can pre-allocate their slot in the per-cpu + * per-core reg tables. + */ +enum extra_reg_type { + EXTRA_REG_NONE = -1, /* not used */ + + EXTRA_REG_RSP_0 = 0, /* offcore_response_0 */ + EXTRA_REG_RSP_1 = 1, /* offcore_response_1 */ + + EXTRA_REG_MAX /* number of entries needed */ +}; + +struct event_constraint { + union { + unsigned long idxmsk[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; + u64 idxmsk64; + }; + u64 code; + u64 cmask; + int weight; +}; + +struct amd_nb { + int nb_id; /* NorthBridge id */ + int refcnt; /* reference count */ + struct perf_event *owners[X86_PMC_IDX_MAX]; + struct event_constraint event_constraints[X86_PMC_IDX_MAX]; +}; + +/* The maximal number of PEBS events: */ +#define MAX_PEBS_EVENTS 4 + +/* + * A debug store configuration. + * + * We only support architectures that use 64bit fields. + */ +struct debug_store { + u64 bts_buffer_base; + u64 bts_index; + u64 bts_absolute_maximum; + u64 bts_interrupt_threshold; + u64 pebs_buffer_base; + u64 pebs_index; + u64 pebs_absolute_maximum; + u64 pebs_interrupt_threshold; + u64 pebs_event_reset[MAX_PEBS_EVENTS]; +}; + +/* + * Per register state. + */ +struct er_account { + raw_spinlock_t lock; /* per-core: protect structure */ + u64 config; /* extra MSR config */ + u64 reg; /* extra MSR number */ + atomic_t ref; /* reference count */ +}; + +/* + * Per core/cpu state + * + * Used to coordinate shared registers between HT threads or + * among events on a single PMU. + */ +struct intel_shared_regs { + struct er_account regs[EXTRA_REG_MAX]; + int refcnt; /* per-core: #HT threads */ + unsigned core_id; /* per-core: core id */ +}; + +#define MAX_LBR_ENTRIES 16 + +struct cpu_hw_events { + /* + * Generic x86 PMC bits + */ + struct perf_event *events[X86_PMC_IDX_MAX]; /* in counter order */ + unsigned long active_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; + unsigned long running[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; + int enabled; + + int n_events; + int n_added; + int n_txn; + int assign[X86_PMC_IDX_MAX]; /* event to counter assignment */ + u64 tags[X86_PMC_IDX_MAX]; + struct perf_event *event_list[X86_PMC_IDX_MAX]; /* in enabled order */ + + unsigned int group_flag; + + /* + * Intel DebugStore bits + */ + struct debug_store *ds; + u64 pebs_enabled; + + /* + * Intel LBR bits + */ + int lbr_users; + void *lbr_context; + struct perf_branch_stack lbr_stack; + struct perf_branch_entry lbr_entries[MAX_LBR_ENTRIES]; + + /* + * manage shared (per-core, per-cpu) registers + * used on Intel NHM/WSM/SNB + */ + struct intel_shared_regs *shared_regs; + + /* + * AMD specific bits + */ + struct amd_nb *amd_nb; + + void *kfree_on_online; +}; + +#define __EVENT_CONSTRAINT(c, n, m, w) {\ + { .idxmsk64 = (n) }, \ + .code = (c), \ + .cmask = (m), \ + .weight = (w), \ +} + +#define EVENT_CONSTRAINT(c, n, m) \ + __EVENT_CONSTRAINT(c, n, m, HWEIGHT(n)) + +/* + * Constraint on the Event code. + */ +#define INTEL_EVENT_CONSTRAINT(c, n) \ + EVENT_CONSTRAINT(c, n, ARCH_PERFMON_EVENTSEL_EVENT) + +/* + * Constraint on the Event code + UMask + fixed-mask + * + * filter mask to validate fixed counter events. + * the following filters disqualify for fixed counters: + * - inv + * - edge + * - cnt-mask + * The other filters are supported by fixed counters. + * The any-thread option is supported starting with v3. + */ +#define FIXED_EVENT_CONSTRAINT(c, n) \ + EVENT_CONSTRAINT(c, (1ULL << (32+n)), X86_RAW_EVENT_MASK) + +/* + * Constraint on the Event code + UMask + */ +#define INTEL_UEVENT_CONSTRAINT(c, n) \ + EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVENT_MASK) + +#define EVENT_CONSTRAINT_END \ + EVENT_CONSTRAINT(0, 0, 0) + +#define for_each_event_constraint(e, c) \ + for ((e) = (c); (e)->weight; (e)++) + +/* + * Extra registers for specific events. + * + * Some events need large masks and require external MSRs. + * Those extra MSRs end up being shared for all events on + * a PMU and sometimes between PMU of sibling HT threads. + * In either case, the kernel needs to handle conflicting + * accesses to those extra, shared, regs. The data structure + * to manage those registers is stored in cpu_hw_event. + */ +struct extra_reg { + unsigned int event; + unsigned int msr; + u64 config_mask; + u64 valid_mask; + int idx; /* per_xxx->regs[] reg index */ +}; + +#define EVENT_EXTRA_REG(e, ms, m, vm, i) { \ + .event = (e), \ + .msr = (ms), \ + .config_mask = (m), \ + .valid_mask = (vm), \ + .idx = EXTRA_REG_##i \ + } + +#define INTEL_EVENT_EXTRA_REG(event, msr, vm, idx) \ + EVENT_EXTRA_REG(event, msr, ARCH_PERFMON_EVENTSEL_EVENT, vm, idx) + +#define EVENT_EXTRA_END EVENT_EXTRA_REG(0, 0, 0, 0, RSP_0) + +union perf_capabilities { + struct { + u64 lbr_format:6; + u64 pebs_trap:1; + u64 pebs_arch_reg:1; + u64 pebs_format:4; + u64 smm_freeze:1; + }; + u64 capabilities; +}; + +/* + * struct x86_pmu - generic x86 pmu + */ +struct x86_pmu { + /* + * Generic x86 PMC bits + */ + const char *name; + int version; + int (*handle_irq)(struct pt_regs *); + void (*disable_all)(void); + void (*enable_all)(int added); + void (*enable)(struct perf_event *); + void (*disable)(struct perf_event *); + int (*hw_config)(struct perf_event *event); + int (*schedule_events)(struct cpu_hw_events *cpuc, int n, int *assign); + unsigned eventsel; + unsigned perfctr; + u64 (*event_map)(int); + int max_events; + int num_counters; + int num_counters_fixed; + int cntval_bits; + u64 cntval_mask; + int apic; + u64 max_period; + struct event_constraint * + (*get_event_constraints)(struct cpu_hw_events *cpuc, + struct perf_event *event); + + void (*put_event_constraints)(struct cpu_hw_events *cpuc, + 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); + void (*cpu_dying)(int cpu); + void (*cpu_dead)(int cpu); + + /* + * Intel Arch Perfmon v2+ + */ + u64 intel_ctrl; + union perf_capabilities intel_cap; + + /* + * Intel DebugStore bits + */ + int bts, pebs; + int bts_active, pebs_active; + int pebs_record_size; + void (*drain_pebs)(struct pt_regs *regs); + struct event_constraint *pebs_constraints; + + /* + * Intel LBR + */ + unsigned long lbr_tos, lbr_from, lbr_to; /* MSR base regs */ + int lbr_nr; /* hardware stack size */ + + /* + * Extra registers for events + */ + struct extra_reg *extra_regs; + unsigned int er_flags; +}; + +#define ERF_NO_HT_SHARING 1 +#define ERF_HAS_RSP_1 2 + +extern struct x86_pmu x86_pmu __read_mostly; + +DECLARE_PER_CPU(struct cpu_hw_events, cpu_hw_events); + +int x86_perf_event_set_period(struct perf_event *event); + +/* + * Generalized hw caching related hw_event table, filled + * in on a per model basis. A value of 0 means + * 'not supported', -1 means 'hw_event makes no sense on + * this CPU', any other value means the raw hw_event + * ID. + */ + +#define C(x) PERF_COUNT_HW_CACHE_##x + +extern u64 __read_mostly hw_cache_event_ids + [PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX]; +extern u64 __read_mostly hw_cache_extra_regs + [PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX]; + +u64 x86_perf_event_update(struct perf_event *event); + +static inline int x86_pmu_addr_offset(int index) +{ + int offset; + + /* offset = X86_FEATURE_PERFCTR_CORE ? index << 1 : index */ + alternative_io(ASM_NOP2, + "shll $1, %%eax", + X86_FEATURE_PERFCTR_CORE, + "=a" (offset), + "a" (index)); + + return offset; +} + +static inline unsigned int x86_pmu_config_addr(int index) +{ + return x86_pmu.eventsel + x86_pmu_addr_offset(index); +} + +static inline unsigned int x86_pmu_event_addr(int index) +{ + return x86_pmu.perfctr + x86_pmu_addr_offset(index); +} + +int x86_setup_perfctr(struct perf_event *event); + +int x86_pmu_hw_config(struct perf_event *event); + +void x86_pmu_disable_all(void); + +static inline void __x86_pmu_enable_event(struct hw_perf_event *hwc, + u64 enable_mask) +{ + if (hwc->extra_reg.reg) + wrmsrl(hwc->extra_reg.reg, hwc->extra_reg.config); + wrmsrl(hwc->config_base, hwc->config | enable_mask); +} + +void x86_pmu_enable_all(int added); + +int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign); + +void x86_pmu_stop(struct perf_event *event, int flags); + +static inline void x86_pmu_disable_event(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + + wrmsrl(hwc->config_base, hwc->config); +} + +void x86_pmu_enable_event(struct perf_event *event); + +int x86_pmu_handle_irq(struct pt_regs *regs); + +extern struct event_constraint emptyconstraint; + +extern struct event_constraint unconstrained; + +#ifdef CONFIG_CPU_SUP_AMD + +int amd_pmu_init(void); + +#else /* CONFIG_CPU_SUP_AMD */ + +static inline int amd_pmu_init(void) +{ + return 0; +} + +#endif /* CONFIG_CPU_SUP_AMD */ + +#ifdef CONFIG_CPU_SUP_INTEL + +int intel_pmu_save_and_restart(struct perf_event *event); + +struct event_constraint * +x86_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event); + +struct intel_shared_regs *allocate_shared_regs(int cpu); + +int intel_pmu_init(void); + +void init_debug_store_on_cpu(int cpu); + +void fini_debug_store_on_cpu(int cpu); + +void release_ds_buffers(void); + +void reserve_ds_buffers(void); + +extern struct event_constraint bts_constraint; + +void intel_pmu_enable_bts(u64 config); + +void intel_pmu_disable_bts(void); + +int intel_pmu_drain_bts_buffer(void); + +extern struct event_constraint intel_core2_pebs_event_constraints[]; + +extern struct event_constraint intel_atom_pebs_event_constraints[]; + +extern struct event_constraint intel_nehalem_pebs_event_constraints[]; + +extern struct event_constraint intel_westmere_pebs_event_constraints[]; + +extern struct event_constraint intel_snb_pebs_event_constraints[]; + +struct event_constraint *intel_pebs_constraints(struct perf_event *event); + +void intel_pmu_pebs_enable(struct perf_event *event); + +void intel_pmu_pebs_disable(struct perf_event *event); + +void intel_pmu_pebs_enable_all(void); + +void intel_pmu_pebs_disable_all(void); + +void intel_ds_init(void); + +void intel_pmu_lbr_reset(void); + +void intel_pmu_lbr_enable(struct perf_event *event); + +void intel_pmu_lbr_disable(struct perf_event *event); + +void intel_pmu_lbr_enable_all(void); + +void intel_pmu_lbr_disable_all(void); + +void intel_pmu_lbr_read(void); + +void intel_pmu_lbr_init_core(void); + +void intel_pmu_lbr_init_nhm(void); + +void intel_pmu_lbr_init_atom(void); + +int p4_pmu_init(void); + +int p6_pmu_init(void); + +#else /* CONFIG_CPU_SUP_INTEL */ + +static inline void reserve_ds_buffers(void) +{ +} + +static inline void release_ds_buffers(void) +{ +} + +static inline int intel_pmu_init(void) +{ + return 0; +} + +static inline struct intel_shared_regs *allocate_shared_regs(int cpu) +{ + return NULL; +} + +#endif /* CONFIG_CPU_SUP_INTEL */ diff --git a/arch/x86/kernel/cpu/perf_event_amd.c b/arch/x86/kernel/cpu/perf_event_amd.c index 941caa2e449..384450d6712 100644 --- a/arch/x86/kernel/cpu/perf_event_amd.c +++ b/arch/x86/kernel/cpu/perf_event_amd.c @@ -1,4 +1,10 @@ -#ifdef CONFIG_CPU_SUP_AMD +#include <linux/perf_event.h> +#include <linux/types.h> +#include <linux/init.h> +#include <linux/slab.h> +#include <asm/apicdef.h> + +#include "perf_event.h" static __initconst const u64 amd_hw_cache_event_ids [PERF_COUNT_HW_CACHE_MAX] @@ -350,7 +356,7 @@ static void amd_pmu_cpu_starting(int cpu) continue; if (nb->nb_id == nb_id) { - kfree(cpuc->amd_nb); + cpuc->kfree_on_online = cpuc->amd_nb; cpuc->amd_nb = nb; break; } @@ -573,7 +579,7 @@ static __initconst const struct x86_pmu amd_pmu_f15h = { #endif }; -static __init int amd_pmu_init(void) +__init int amd_pmu_init(void) { /* Performance-monitoring supported from K7 and later: */ if (boot_cpu_data.x86 < 6) @@ -602,12 +608,3 @@ static __init int amd_pmu_init(void) return 0; } - -#else /* CONFIG_CPU_SUP_AMD */ - -static int amd_pmu_init(void) -{ - return 0; -} - -#endif diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index f88af2c2a56..61fa35750b9 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -1,16 +1,19 @@ -#ifdef CONFIG_CPU_SUP_INTEL - /* * Per core/cpu state * * Used to coordinate shared registers between HT threads or * among events on a single PMU. */ -struct intel_shared_regs { - struct er_account regs[EXTRA_REG_MAX]; - int refcnt; /* per-core: #HT threads */ - unsigned core_id; /* per-core: core id */ -}; + +#include <linux/stddef.h> +#include <linux/types.h> +#include <linux/init.h> +#include <linux/slab.h> + +#include <asm/hardirq.h> +#include <asm/apic.h> + +#include "perf_event.h" /* * Intel PerfMon, used on Core and later. @@ -945,7 +948,7 @@ static void intel_pmu_enable_event(struct perf_event *event) * Save and restart an expired event. Called by NMI contexts, * so it has to be careful about preempting normal event ops: */ -static int intel_pmu_save_and_restart(struct perf_event *event) +int intel_pmu_save_and_restart(struct perf_event *event) { x86_perf_event_update(event); return x86_perf_event_set_period(event); @@ -1197,6 +1200,21 @@ intel_shared_regs_constraints(struct cpu_hw_events *cpuc, return c; } +struct event_constraint * +x86_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event) +{ + struct event_constraint *c; + + if (x86_pmu.event_constraints) { + for_each_event_constraint(c, x86_pmu.event_constraints) { + if ((event->hw.config & c->cmask) == c->code) + return c; + } + } + + return &unconstrained; +} + static struct event_constraint * intel_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event) { @@ -1309,7 +1327,7 @@ static __initconst const struct x86_pmu core_pmu = { .event_constraints = intel_core_event_constraints, }; -static struct intel_shared_regs *allocate_shared_regs(int cpu) +struct intel_shared_regs *allocate_shared_regs(int cpu) { struct intel_shared_regs *regs; int i; @@ -1362,7 +1380,7 @@ static void intel_pmu_cpu_starting(int cpu) pc = per_cpu(cpu_hw_events, i).shared_regs; if (pc && pc->core_id == core_id) { - kfree(cpuc->shared_regs); + cpuc->kfree_on_online = cpuc->shared_regs; cpuc->shared_regs = pc; break; } @@ -1441,7 +1459,7 @@ static void intel_clovertown_quirks(void) x86_pmu.pebs_constraints = NULL; } -static __init int intel_pmu_init(void) +__init int intel_pmu_init(void) { union cpuid10_edx edx; union cpuid10_eax eax; @@ -1597,7 +1615,7 @@ static __init int intel_pmu_init(void) intel_pmu_lbr_init_nhm(); x86_pmu.event_constraints = intel_snb_event_constraints; - x86_pmu.pebs_constraints = intel_snb_pebs_events; + x86_pmu.pebs_constraints = intel_snb_pebs_event_constraints; x86_pmu.extra_regs = intel_snb_extra_regs; /* all extra regs are per-cpu when HT is on */ x86_pmu.er_flags |= ERF_HAS_RSP_1; @@ -1628,16 +1646,3 @@ static __init int intel_pmu_init(void) } return 0; } - -#else /* CONFIG_CPU_SUP_INTEL */ - -static int intel_pmu_init(void) -{ - return 0; -} - -static struct intel_shared_regs *allocate_shared_regs(int cpu) -{ - return NULL; -} -#endif /* CONFIG_CPU_SUP_INTEL */ diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c index 1b1ef3addcf..c0d238f49db 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_ds.c +++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c @@ -1,7 +1,10 @@ -#ifdef CONFIG_CPU_SUP_INTEL +#include <linux/bitops.h> +#include <linux/types.h> +#include <linux/slab.h> -/* The maximal number of PEBS events: */ -#define MAX_PEBS_EVENTS 4 +#include <asm/perf_event.h> + +#include "perf_event.h" /* The size of a BTS record in bytes: */ #define BTS_RECORD_SIZE 24 @@ -37,24 +40,7 @@ struct pebs_record_nhm { u64 status, dla, dse, lat; }; -/* - * A debug store configuration. - * - * We only support architectures that use 64bit fields. - */ -struct debug_store { - u64 bts_buffer_base; - u64 bts_index; - u64 bts_absolute_maximum; - u64 bts_interrupt_threshold; - u64 pebs_buffer_base; - u64 pebs_index; - u64 pebs_absolute_maximum; - u64 pebs_interrupt_threshold; - u64 pebs_event_reset[MAX_PEBS_EVENTS]; -}; - -static void init_debug_store_on_cpu(int cpu) +void init_debug_store_on_cpu(int cpu) { struct debug_store *ds = per_cpu(cpu_hw_events, cpu).ds; @@ -66,7 +52,7 @@ static void init_debug_store_on_cpu(int cpu) (u32)((u64)(unsigned long)ds >> 32)); } -static void fini_debug_store_on_cpu(int cpu) +void fini_debug_store_on_cpu(int cpu) { if (!per_cpu(cpu_hw_events, cpu).ds) return; @@ -175,7 +161,7 @@ static void release_ds_buffer(int cpu) kfree(ds); } -static void release_ds_buffers(void) +void release_ds_buffers(void) { int cpu; @@ -194,7 +180,7 @@ static void release_ds_buffers(void) put_online_cpus(); } -static void reserve_ds_buffers(void) +void reserve_ds_buffers(void) { int bts_err = 0, pebs_err = 0; int cpu; @@ -260,10 +246,10 @@ static void reserve_ds_buffers(void) * BTS */ -static struct event_constraint bts_constraint = +struct event_constraint bts_constraint = EVENT_CONSTRAINT(0, 1ULL << X86_PMC_IDX_FIXED_BTS, 0); -static void intel_pmu_enable_bts(u64 config) +void intel_pmu_enable_bts(u64 config) { unsigned long debugctlmsr; @@ -282,7 +268,7 @@ static void intel_pmu_enable_bts(u64 config) update_debugctlmsr(debugctlmsr); } -static void intel_pmu_disable_bts(void) +void intel_pmu_disable_bts(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); unsigned long debugctlmsr; @@ -299,7 +285,7 @@ static void intel_pmu_disable_bts(void) update_debugctlmsr(debugctlmsr); } -static int intel_pmu_drain_bts_buffer(void) +int intel_pmu_drain_bts_buffer(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct debug_store *ds = cpuc->ds; @@ -361,7 +347,7 @@ static int intel_pmu_drain_bts_buffer(void) /* * PEBS */ -static struct event_constraint intel_core2_pebs_event_constraints[] = { +struct event_constraint intel_core2_pebs_event_constraints[] = { INTEL_UEVENT_CONSTRAINT(0x00c0, 0x1), /* INST_RETIRED.ANY */ INTEL_UEVENT_CONSTRAINT(0xfec1, 0x1), /* X87_OPS_RETIRED.ANY */ INTEL_UEVENT_CONSTRAINT(0x00c5, 0x1), /* BR_INST_RETIRED.MISPRED */ @@ -370,14 +356,14 @@ static struct event_constraint intel_core2_pebs_event_constraints[] = { EVENT_CONSTRAINT_END }; -static struct event_constraint intel_atom_pebs_event_constraints[] = { +struct event_constraint intel_atom_pebs_event_constraints[] = { INTEL_UEVENT_CONSTRAINT(0x00c0, 0x1), /* INST_RETIRED.ANY */ INTEL_UEVENT_CONSTRAINT(0x00c5, 0x1), /* MISPREDICTED_BRANCH_RETIRED */ INTEL_EVENT_CONSTRAINT(0xcb, 0x1), /* MEM_LOAD_RETIRED.* */ EVENT_CONSTRAINT_END }; -static struct event_constraint intel_nehalem_pebs_event_constraints[] = { +struct event_constraint intel_nehalem_pebs_event_constraints[] = { INTEL_EVENT_CONSTRAINT(0x0b, 0xf), /* MEM_INST_RETIRED.* */ INTEL_EVENT_CONSTRAINT(0x0f, 0xf), /* MEM_UNCORE_RETIRED.* */ INTEL_UEVENT_CONSTRAINT(0x010c, 0xf), /* MEM_STORE_RETIRED.DTLB_MISS */ @@ -392,7 +378,7 @@ static struct event_constraint intel_nehalem_pebs_event_constraints[] = { EVENT_CONSTRAINT_END }; -static struct event_constraint intel_westmere_pebs_event_constraints[] = { +struct event_constraint intel_westmere_pebs_event_constraints[] = { INTEL_EVENT_CONSTRAINT(0x0b, 0xf), /* MEM_INST_RETIRED.* */ INTEL_EVENT_CONSTRAINT(0x0f, 0xf), /* MEM_UNCORE_RETIRED.* */ INTEL_UEVENT_CONSTRAINT(0x010c, 0xf), /* MEM_STORE_RETIRED.DTLB_MISS */ @@ -407,7 +393,7 @@ static struct event_constraint intel_westmere_pebs_event_constraints[] = { EVENT_CONSTRAINT_END }; -static struct event_constraint intel_snb_pebs_events[] = { +struct event_constraint intel_snb_pebs_event_constraints[] = { INTEL_UEVENT_CONSTRAINT(0x01c0, 0x2), /* INST_RETIRED.PRECDIST */ INTEL_UEVENT_CONSTRAINT(0x01c2, 0xf), /* UOPS_RETIRED.ALL */ INTEL_UEVENT_CONSTRAINT(0x02c2, 0xf), /* UOPS_RETIRED.RETIRE_SLOTS */ @@ -428,8 +414,7 @@ static struct event_constraint intel_snb_pebs_events[] = { EVENT_CONSTRAINT_END }; -static struct event_constraint * -intel_pebs_constraints(struct perf_event *event) +struct event_constraint *intel_pebs_constraints(struct perf_event *event) { struct event_constraint *c; @@ -446,7 +431,7 @@ intel_pebs_constraints(struct perf_event *event) return &emptyconstraint; } -static void intel_pmu_pebs_enable(struct perf_event *event) +void intel_pmu_pebs_enable(struct perf_event *event) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct hw_perf_event *hwc = &event->hw; @@ -460,7 +445,7 @@ static void intel_pmu_pebs_enable(struct perf_event *event) intel_pmu_lbr_enable(event); } -static void intel_pmu_pebs_disable(struct perf_event *event) +void intel_pmu_pebs_disable(struct perf_event *event) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct hw_perf_event *hwc = &event->hw; @@ -475,7 +460,7 @@ static void intel_pmu_pebs_disable(struct perf_event *event) intel_pmu_lbr_disable(event); } -static void intel_pmu_pebs_enable_all(void) +void intel_pmu_pebs_enable_all(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -483,7 +468,7 @@ static void intel_pmu_pebs_enable_all(void) wrmsrl(MSR_IA32_PEBS_ENABLE, cpuc->pebs_enabled); } -static void intel_pmu_pebs_disable_all(void) +void intel_pmu_pebs_disable_all(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -576,8 +561,6 @@ static int intel_pmu_pebs_fixup_ip(struct pt_regs *regs) return 0; } -static int intel_pmu_save_and_restart(struct perf_event *event); - static void __intel_pmu_pebs_event(struct perf_event *event, struct pt_regs *iregs, void *__pebs) { @@ -716,7 +699,7 @@ static void intel_pmu_drain_pebs_nhm(struct pt_regs *iregs) * BTS, PEBS probe and setup */ -static void intel_ds_init(void) +void intel_ds_init(void) { /* * No support for 32bit formats @@ -749,15 +732,3 @@ static void intel_ds_init(void) } } } - -#else /* CONFIG_CPU_SUP_INTEL */ - -static void reserve_ds_buffers(void) -{ -} - -static void release_ds_buffers(void) -{ -} - -#endif /* CONFIG_CPU_SUP_INTEL */ diff --git a/arch/x86/kernel/cpu/perf_event_intel_lbr.c b/arch/x86/kernel/cpu/perf_event_intel_lbr.c index d202c1bece1..3fab3de3ce9 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_lbr.c +++ b/arch/x86/kernel/cpu/perf_event_intel_lbr.c @@ -1,4 +1,10 @@ -#ifdef CONFIG_CPU_SUP_INTEL +#include <linux/perf_event.h> +#include <linux/types.h> + +#include <asm/perf_event.h> +#include <asm/msr.h> + +#include "perf_event.h" enum { LBR_FORMAT_32 = 0x00, @@ -48,7 +54,7 @@ static void intel_pmu_lbr_reset_64(void) } } -static void intel_pmu_lbr_reset(void) +void intel_pmu_lbr_reset(void) { if (!x86_pmu.lbr_nr) return; @@ -59,7 +65,7 @@ static void intel_pmu_lbr_reset(void) intel_pmu_lbr_reset_64(); } -static void intel_pmu_lbr_enable(struct perf_event *event) +void intel_pmu_lbr_enable(struct perf_event *event) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -81,7 +87,7 @@ static void intel_pmu_lbr_enable(struct perf_event *event) cpuc->lbr_users++; } -static void intel_pmu_lbr_disable(struct perf_event *event) +void intel_pmu_lbr_disable(struct perf_event *event) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -95,7 +101,7 @@ static void intel_pmu_lbr_disable(struct perf_event *event) __intel_pmu_lbr_disable(); } -static void intel_pmu_lbr_enable_all(void) +void intel_pmu_lbr_enable_all(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -103,7 +109,7 @@ static void intel_pmu_lbr_enable_all(void) __intel_pmu_lbr_enable(); } -static void intel_pmu_lbr_disable_all(void) +void intel_pmu_lbr_disable_all(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -178,7 +184,7 @@ static void intel_pmu_lbr_read_64(struct cpu_hw_events *cpuc) cpuc->lbr_stack.nr = i; } -static void intel_pmu_lbr_read(void) +void intel_pmu_lbr_read(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -191,7 +197,7 @@ static void intel_pmu_lbr_read(void) intel_pmu_lbr_read_64(cpuc); } -static void intel_pmu_lbr_init_core(void) +void intel_pmu_lbr_init_core(void) { x86_pmu.lbr_nr = 4; x86_pmu.lbr_tos = 0x01c9; @@ -199,7 +205,7 @@ static void intel_pmu_lbr_init_core(void) x86_pmu.lbr_to = 0x60; } -static void intel_pmu_lbr_init_nhm(void) +void intel_pmu_lbr_init_nhm(void) { x86_pmu.lbr_nr = 16; x86_pmu.lbr_tos = 0x01c9; @@ -207,12 +213,10 @@ static void intel_pmu_lbr_init_nhm(void) x86_pmu.lbr_to = 0x6c0; } -static void intel_pmu_lbr_init_atom(void) +void intel_pmu_lbr_init_atom(void) { x86_pmu.lbr_nr = 8; x86_pmu.lbr_tos = 0x01c9; x86_pmu.lbr_from = 0x40; x86_pmu.lbr_to = 0x60; } - -#endif /* CONFIG_CPU_SUP_INTEL */ diff --git a/arch/x86/kernel/cpu/perf_event_p4.c b/arch/x86/kernel/cpu/perf_event_p4.c index 7809d2bcb20..492bf1358a7 100644 --- a/arch/x86/kernel/cpu/perf_event_p4.c +++ b/arch/x86/kernel/cpu/perf_event_p4.c @@ -7,9 +7,13 @@ * For licencing details see kernel-base/COPYING */ -#ifdef CONFIG_CPU_SUP_INTEL +#include <linux/perf_event.h> #include <asm/perf_event_p4.h> +#include <asm/hardirq.h> +#include <asm/apic.h> + +#include "perf_event.h" #define P4_CNTR_LIMIT 3 /* @@ -1303,7 +1307,7 @@ static __initconst const struct x86_pmu p4_pmu = { .perfctr_second_write = 1, }; -static __init int p4_pmu_init(void) +__init int p4_pmu_init(void) { unsigned int low, high; @@ -1326,5 +1330,3 @@ static __init int p4_pmu_init(void) return 0; } - -#endif /* CONFIG_CPU_SUP_INTEL */ diff --git a/arch/x86/kernel/cpu/perf_event_p6.c b/arch/x86/kernel/cpu/perf_event_p6.c index 20c097e3386..c7181befecd 100644 --- a/arch/x86/kernel/cpu/perf_event_p6.c +++ b/arch/x86/kernel/cpu/perf_event_p6.c @@ -1,4 +1,7 @@ -#ifdef CONFIG_CPU_SUP_INTEL +#include <linux/perf_event.h> +#include <linux/types.h> + +#include "perf_event.h" /* * Not sure about some of these @@ -114,7 +117,7 @@ static __initconst const struct x86_pmu p6_pmu = { .event_constraints = p6_event_constraints, }; -static __init int p6_pmu_init(void) +__init int p6_pmu_init(void) { switch (boot_cpu_data.x86_model) { case 1: @@ -138,5 +141,3 @@ static __init int p6_pmu_init(void) return 0; } - -#endif /* CONFIG_CPU_SUP_INTEL */ diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c index 94b745045e4..d90528ea541 100644 --- a/arch/x86/oprofile/op_model_ppro.c +++ b/arch/x86/oprofile/op_model_ppro.c @@ -28,7 +28,7 @@ static int counter_width = 32; #define MSR_PPRO_EVENTSEL_RESERVED ((0xFFFFFFFFULL<<32)|(1ULL<<21)) -static u64 *reset_value; +static u64 reset_value[OP_MAX_COUNTER]; static void ppro_shutdown(struct op_msrs const * const msrs) { @@ -40,10 +40,6 @@ static void ppro_shutdown(struct op_msrs const * const msrs) release_perfctr_nmi(MSR_P6_PERFCTR0 + i); release_evntsel_nmi(MSR_P6_EVNTSEL0 + i); } - if (reset_value) { - kfree(reset_value); - reset_value = NULL; - } } static int ppro_fill_in_addresses(struct op_msrs * const msrs) @@ -79,13 +75,6 @@ static void ppro_setup_ctrs(struct op_x86_model_spec const *model, u64 val; int i; - if (!reset_value) { - reset_value = kzalloc(sizeof(reset_value[0]) * num_counters, - GFP_ATOMIC); - if (!reset_value) - return; - } - if (cpu_has_arch_perfmon) { union cpuid10_eax eax; eax.full = cpuid_eax(0xa); @@ -141,13 +130,6 @@ static int ppro_check_ctrs(struct pt_regs * const regs, u64 val; int i; - /* - * This can happen if perf counters are in use when - * we steal the die notifier NMI. - */ - if (unlikely(!reset_value)) - goto out; - for (i = 0; i < num_counters; ++i) { if (!reset_value[i]) continue; @@ -158,7 +140,6 @@ static int ppro_check_ctrs(struct pt_regs * const regs, wrmsrl(msrs->counters[i].addr, -reset_value[i]); } -out: /* Only P6 based Pentium M need to re-unmask the apic vector but it * doesn't hurt other P6 variant */ apic_write(APIC_LVTPC, apic_read(APIC_LVTPC) & ~APIC_LVT_MASKED); @@ -179,8 +160,6 @@ static void ppro_start(struct op_msrs const * const msrs) u64 val; int i; - if (!reset_value) - return; for (i = 0; i < num_counters; ++i) { if (reset_value[i]) { rdmsrl(msrs->controls[i].addr, val); @@ -196,8 +175,6 @@ static void ppro_stop(struct op_msrs const * const msrs) u64 val; int i; - if (!reset_value) - return; for (i = 0; i < num_counters; ++i) { if (!reset_value[i]) continue; @@ -242,7 +219,7 @@ static void arch_perfmon_setup_counters(void) eax.split.bit_width = 40; } - num_counters = eax.split.num_counters; + num_counters = min((int)eax.split.num_counters, OP_MAX_COUNTER); op_arch_perfmon_spec.num_counters = num_counters; op_arch_perfmon_spec.num_controls = num_counters; diff --git a/include/trace/ftrace.h b/include/trace/ftrace.h index 533c49f4804..769724944fc 100644 --- a/include/trace/ftrace.h +++ b/include/trace/ftrace.h @@ -711,6 +711,9 @@ __attribute__((section("_ftrace_events"))) *__event_##call = &event_##call #undef __perf_count #define __perf_count(c) __count = (c) +#undef TP_perf_assign +#define TP_perf_assign(args...) args + #undef DECLARE_EVENT_CLASS #define DECLARE_EVENT_CLASS(call, proto, args, tstruct, assign, print) \ static notrace void \ diff --git a/kernel/events/core.c b/kernel/events/core.c index 0f857782d06..d1a1bee3522 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -29,6 +29,7 @@ #include <linux/hardirq.h> #include <linux/rculist.h> #include <linux/uaccess.h> +#include <linux/suspend.h> #include <linux/syscalls.h> #include <linux/anon_inodes.h> #include <linux/kernel_stat.h> @@ -5758,6 +5759,7 @@ struct pmu *perf_init_event(struct perf_event *event) pmu = idr_find(&pmu_idr, event->attr.type); rcu_read_unlock(); if (pmu) { + event->pmu = pmu; ret = pmu->event_init(event); if (ret) pmu = ERR_PTR(ret); @@ -5765,6 +5767,7 @@ struct pmu *perf_init_event(struct perf_event *event) } list_for_each_entry_rcu(pmu, &pmus, entry) { + event->pmu = pmu; ret = pmu->event_init(event); if (!ret) goto unlock; @@ -5891,8 +5894,6 @@ done: return ERR_PTR(err); } - event->pmu = pmu; - if (!event->parent) { if (event->attach_state & PERF_ATTACH_TASK) jump_label_inc(&perf_sched_events); @@ -6852,7 +6853,7 @@ static void __cpuinit perf_event_init_cpu(int cpu) struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu); mutex_lock(&swhash->hlist_mutex); - if (swhash->hlist_refcount > 0) { + if (swhash->hlist_refcount > 0 && !swhash->swevent_hlist) { struct swevent_hlist *hlist; hlist = kzalloc_node(sizeof(*hlist), GFP_KERNEL, cpu_to_node(cpu)); @@ -6941,7 +6942,14 @@ perf_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { unsigned int cpu = (long)hcpu; - switch (action & ~CPU_TASKS_FROZEN) { + /* + * Ignore suspend/resume action, the perf_pm_notifier will + * take care of that. + */ + if (action & CPU_TASKS_FROZEN) + return NOTIFY_OK; + + switch (action) { case CPU_UP_PREPARE: case CPU_DOWN_FAILED: @@ -6960,6 +6968,90 @@ perf_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) return NOTIFY_OK; } +static void perf_pm_resume_cpu(void *unused) +{ + struct perf_cpu_context *cpuctx; + struct perf_event_context *ctx; + struct pmu *pmu; + int idx; + + idx = srcu_read_lock(&pmus_srcu); + list_for_each_entry_rcu(pmu, &pmus, entry) { + cpuctx = this_cpu_ptr(pmu->pmu_cpu_context); + ctx = cpuctx->task_ctx; + + perf_ctx_lock(cpuctx, ctx); + perf_pmu_disable(cpuctx->ctx.pmu); + + cpu_ctx_sched_out(cpuctx, EVENT_ALL); + if (ctx) + ctx_sched_out(ctx, cpuctx, EVENT_ALL); + + perf_pmu_enable(cpuctx->ctx.pmu); + perf_ctx_unlock(cpuctx, ctx); + } + srcu_read_unlock(&pmus_srcu, idx); +} + +static void perf_pm_suspend_cpu(void *unused) +{ + struct perf_cpu_context *cpuctx; + struct perf_event_context *ctx; + struct pmu *pmu; + int idx; + + idx = srcu_read_lock(&pmus_srcu); + list_for_each_entry_rcu(pmu, &pmus, entry) { + cpuctx = this_cpu_ptr(pmu->pmu_cpu_context); + ctx = cpuctx->task_ctx; + + perf_ctx_lock(cpuctx, ctx); + perf_pmu_disable(cpuctx->ctx.pmu); + + perf_event_sched_in(cpuctx, ctx, current); + + perf_pmu_enable(cpuctx->ctx.pmu); + perf_ctx_unlock(cpuctx, ctx); + } + srcu_read_unlock(&pmus_srcu, idx); +} + +static int perf_resume(void) +{ + get_online_cpus(); + smp_call_function(perf_pm_resume_cpu, NULL, 1); + put_online_cpus(); + + return NOTIFY_OK; +} + +static int perf_suspend(void) +{ + get_online_cpus(); + smp_call_function(perf_pm_suspend_cpu, NULL, 1); + put_online_cpus(); + + return NOTIFY_OK; +} + +static int perf_pm(struct notifier_block *self, unsigned long action, void *ptr) +{ + switch (action) { + case PM_POST_HIBERNATION: + case PM_POST_SUSPEND: + return perf_resume(); + case PM_HIBERNATION_PREPARE: + case PM_SUSPEND_PREPARE: + return perf_suspend(); + default: + return NOTIFY_DONE; + } +} + +static struct notifier_block perf_pm_notifier = { + .notifier_call = perf_pm, +}; + void __init perf_event_init(void) { int ret; @@ -6974,6 +7066,7 @@ void __init perf_event_init(void) perf_tp_register(); perf_cpu_notifier(perf_cpu_notify); register_reboot_notifier(&perf_reboot_notifier); + register_pm_notifier(&perf_pm_notifier); ret = init_hw_breakpoint(); WARN(ret, "hw_breakpoint initialization failed with: %d", ret); diff --git a/kernel/watchdog.c b/kernel/watchdog.c index 36491cd5b7d..d680381b0e9 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -321,7 +321,7 @@ static enum hrtimer_restart watchdog_timer_fn(struct hrtimer *hrtimer) */ static int watchdog(void *unused) { - static struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 }; + struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 }; struct hrtimer *hrtimer = &__raw_get_cpu_var(watchdog_hrtimer); sched_setscheduler(current, SCHED_FIFO, ¶m); @@ -350,7 +350,8 @@ static int watchdog(void *unused) set_current_state(TASK_INTERRUPTIBLE); } __set_current_state(TASK_RUNNING); - + param.sched_priority = 0; + sched_setscheduler(current, SCHED_NORMAL, ¶m); return 0; } @@ -438,7 +439,7 @@ static int watchdog_enable(int cpu) /* create the watchdog thread */ if (!p) { - p = kthread_create(watchdog, (void *)(unsigned long)cpu, "watchdog/%d", cpu); + p = kthread_create_on_node(watchdog, NULL, cpu_to_node(cpu), "watchdog/%d", cpu); if (IS_ERR(p)) { printk(KERN_ERR "softlockup watchdog for %i failed\n", cpu); if (!err) { diff --git a/tools/perf/Documentation/perf-annotate.txt b/tools/perf/Documentation/perf-annotate.txt index 85c5f026930..0102d83600d 100644 --- a/tools/perf/Documentation/perf-annotate.txt +++ b/tools/perf/Documentation/perf-annotate.txt @@ -72,6 +72,20 @@ OPTIONS CPUs are specified with -: 0-2. Default is to report samples on all CPUs. +--asm-raw:: + Show raw instruction encoding of assembly instructions. They + are displayed by default, disable with --no-asm-raw. + +--source:: + Interleave source code with assembly code. Enabled by default, + disable with --no-source. + +--symfs=<directory>:: + Look for files with symbols relative to this directory. + +-M:: +--disassembler-style=:: Set disassembler style for objdump. + SEE ALSO -------- linkperf:perf-record[1], linkperf:perf-report[1] diff --git a/tools/perf/Documentation/perf-buildid-list.txt b/tools/perf/Documentation/perf-buildid-list.txt index 5eaac6f26d5..cc22325ffd1 100644 --- a/tools/perf/Documentation/perf-buildid-list.txt +++ b/tools/perf/Documentation/perf-buildid-list.txt @@ -16,6 +16,9 @@ This command displays the buildids found in a perf.data file, so that other tools can be used to fetch packages with matching symbol tables for use by perf report. +It can also be used to show the build id of the running kernel or in an ELF +file using -i/--input. + OPTIONS ------- -H:: @@ -27,6 +30,9 @@ OPTIONS -f:: --force:: Don't do ownership validation. +-k:: +--kernel:: + Show running kernel build id. -v:: --verbose:: Be more verbose. diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt index 04253c07d19..6349b6c0e3e 100644 --- a/tools/perf/Documentation/perf-report.txt +++ b/tools/perf/Documentation/perf-report.txt @@ -134,6 +134,9 @@ OPTIONS CPUs are specified with -: 0-2. Default is to report samples on all CPUs. +-M:: +--disassembler-style=:: Set disassembler style for objdump. + SEE ALSO -------- linkperf:perf-stat[1] diff --git a/tools/perf/Documentation/perf-sched.txt b/tools/perf/Documentation/perf-sched.txt index 46822d5fde1..5b212b57f70 100644 --- a/tools/perf/Documentation/perf-sched.txt +++ b/tools/perf/Documentation/perf-sched.txt @@ -8,7 +8,7 @@ perf-sched - Tool to trace/measure scheduler properties (latencies) SYNOPSIS -------- [verse] -'perf sched' {record|latency|map|replay|trace} +'perf sched' {record|latency|map|replay|script} DESCRIPTION ----------- @@ -20,8 +20,8 @@ There are five variants of perf sched: 'perf sched latency' to report the per task scheduling latencies and other scheduling properties of the workload. - 'perf sched trace' to see a detailed trace of the workload that - was recorded. + 'perf sched script' to see a detailed trace of the workload that + was recorded (aliased to 'perf script' for now). 'perf sched replay' to simulate the workload that was recorded via perf sched record. (this is done by starting up mockup threads diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt index 918cc38ee6d..8966b9ab201 100644 --- a/tools/perf/Documentation/perf-stat.txt +++ b/tools/perf/Documentation/perf-stat.txt @@ -94,6 +94,22 @@ an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must ha corresponding events, i.e., they always refer to events defined earlier on the command line. +-o file:: +--output file:: +Print the output into the designated file. + +--append:: +Append to the output file designated with the -o option. Ignored if -o is not specified. + +--log-fd:: + +Log output to fd, instead of stderr. Complementary to --output, and mutually exclusive +with it. --append may be used here. Examples: + 3>results perf stat --log-fd 3 -- $cmd + 3>>results perf stat --log-fd 3 --append -- $cmd + + + EXAMPLES -------- diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c index 555aefd7fe0..cf68819f745 100644 --- a/tools/perf/builtin-annotate.c +++ b/tools/perf/builtin-annotate.c @@ -267,6 +267,14 @@ static const struct option options[] = { OPT_BOOLEAN('P', "full-paths", &full_paths, "Don't shorten the displayed pathnames"), OPT_STRING('c', "cpu", &cpu_list, "cpu", "list of cpus to profile"), + OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory", + "Look for files with symbols relative to this directory"), + OPT_BOOLEAN('0', "source", &symbol_conf.annotate_src, + "Interleave source code with assembly code (default)"), + OPT_BOOLEAN('0', "asm-raw", &symbol_conf.annotate_asm_raw, + "Display raw encoding of assembly instructions (default)"), + OPT_STRING('M', "disassembler-style", &disassembler_style, "disassembler style", + "Specify disassembler style (e.g. -M intel for intel syntax)"), OPT_END() }; diff --git a/tools/perf/builtin-buildid-list.c b/tools/perf/builtin-buildid-list.c index 5af32ae9031..cb690a65bf0 100644 --- a/tools/perf/builtin-buildid-list.c +++ b/tools/perf/builtin-buildid-list.c @@ -1,7 +1,8 @@ /* * builtin-buildid-list.c * - * Builtin buildid-list command: list buildids in perf.data + * Builtin buildid-list command: list buildids in perf.data, in the running + * kernel and in ELF files. * * Copyright (C) 2009, Red Hat Inc. * Copyright (C) 2009, Arnaldo Carvalho de Melo <acme@redhat.com> @@ -15,8 +16,11 @@ #include "util/session.h" #include "util/symbol.h" +#include <libelf.h> + static char const *input_name = "perf.data"; static bool force; +static bool show_kernel; static bool with_hits; static const char * const buildid_list_usage[] = { @@ -29,12 +33,13 @@ static const struct option options[] = { OPT_STRING('i', "input", &input_name, "file", "input file name"), OPT_BOOLEAN('f', "force", &force, "don't complain, do it"), + OPT_BOOLEAN('k', "kernel", &show_kernel, "Show current kernel build id"), OPT_INCR('v', "verbose", &verbose, "be more verbose"), OPT_END() }; -static int __cmd_buildid_list(void) +static int perf_session__list_build_ids(void) { struct perf_session *session; @@ -52,6 +57,49 @@ static int __cmd_buildid_list(void) return 0; } +static int sysfs__fprintf_build_id(FILE *fp) +{ + u8 kallsyms_build_id[BUILD_ID_SIZE]; + char sbuild_id[BUILD_ID_SIZE * 2 + 1]; + + if (sysfs__read_build_id("/sys/kernel/notes", kallsyms_build_id, + sizeof(kallsyms_build_id)) != 0) + return -1; + + build_id__sprintf(kallsyms_build_id, sizeof(kallsyms_build_id), + sbuild_id); + fprintf(fp, "%s\n", sbuild_id); + return 0; +} + +static int filename__fprintf_build_id(const char *name, FILE *fp) +{ + u8 build_id[BUILD_ID_SIZE]; + char sbuild_id[BUILD_ID_SIZE * 2 + 1]; + + if (filename__read_build_id(name, build_id, + sizeof(build_id)) != sizeof(build_id)) + return 0; + + build_id__sprintf(build_id, sizeof(build_id), sbuild_id); + return fprintf(fp, "%s\n", sbuild_id); +} + +static int __cmd_buildid_list(void) +{ + if (show_kernel) + return sysfs__fprintf_build_id(stdout); + + elf_version(EV_CURRENT); + /* + * See if this is an ELF file first: + */ + if (filename__fprintf_build_id(input_name, stdout)) + return 0; + + return perf_session__list_build_ids(); +} + int cmd_buildid_list(int argc, const char **argv, const char *prefix __used) { argc = parse_options(argc, argv, options, buildid_list_usage, 0); diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index f4c3fbee4ba..dd6467872f6 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -73,6 +73,7 @@ static off_t post_processing_offset; static struct perf_session *session; static const char *cpu_list; +static const char *progname; static void advance_output(size_t size) { @@ -137,17 +138,29 @@ static void mmap_read(struct perf_mmap *md) static volatile int done = 0; static volatile int signr = -1; +static volatile int child_finished = 0; static void sig_handler(int sig) { + if (sig == SIGCHLD) + child_finished = 1; + done = 1; signr = sig; } static void sig_atexit(void) { - if (child_pid > 0) - kill(child_pid, SIGTERM); + int status; + + if (child_pid > 0) { + if (!child_finished) + kill(child_pid, SIGTERM); + + wait(&status); + if (WIFSIGNALED(status)) + psignal(WTERMSIG(status), progname); + } if (signr == -1 || signr == SIGUSR1) return; @@ -446,6 +459,8 @@ static int __cmd_record(int argc, const char **argv) char buf; struct machine *machine; + progname = argv[0]; + page_size = sysconf(_SC_PAGE_SIZE); atexit(sig_atexit); diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c index d7ff277bdb7..3d58334909a 100644 --- a/tools/perf/builtin-report.c +++ b/tools/perf/builtin-report.c @@ -229,10 +229,7 @@ static int perf_evlist__tty_browse_hists(struct perf_evlist *evlist, list_for_each_entry(pos, &evlist->entries, node) { struct hists *hists = &pos->hists; - const char *evname = NULL; - - if (rb_first(&hists->entries) != rb_last(&hists->entries)) - evname = event_name(pos); + const char *evname = event_name(pos); hists__fprintf_nr_sample_events(hists, evname, stdout); hists__fprintf(hists, NULL, false, stdout); @@ -487,6 +484,8 @@ static const struct option options[] = { OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory", "Look for files with symbols relative to this directory"), OPT_STRING('c', "cpu", &cpu_list, "cpu", "list of cpus to profile"), + OPT_STRING('M', "disassembler-style", &disassembler_style, "disassembler style", + "Specify disassembler style (e.g. -M intel for intel syntax)"), OPT_END() }; diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index 5deb17d9e79..7ce65f52415 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -194,6 +194,9 @@ static const char *cpu_list; static const char *csv_sep = NULL; static bool csv_output = false; static bool group = false; +static const char *output_name = NULL; +static FILE *output = NULL; +static int output_fd; static volatile int done = 0; @@ -251,8 +254,13 @@ static double avg_stats(struct stats *stats) */ static double stddev_stats(struct stats *stats) { - double variance = stats->M2 / (stats->n - 1); - double variance_mean = variance / stats->n; + double variance, variance_mean; + + if (!stats->n) + return 0.0; + + variance = stats->M2 / (stats->n - 1); + variance_mean = variance / stats->n; return sqrt(variance_mean); } @@ -352,7 +360,7 @@ static int read_counter_aggr(struct perf_evsel *counter) update_stats(&ps->res_stats[i], count[i]); if (verbose) { - fprintf(stderr, "%s: %" PRIu64 " %" PRIu64 " %" PRIu64 "\n", + fprintf(output, "%s: %" PRIu64 " %" PRIu64 " %" PRIu64 "\n", event_name(counter), count[0], count[1], count[2]); } @@ -487,6 +495,8 @@ static int run_perf_stat(int argc __used, const char **argv) if (forks) { close(go_pipe[1]); wait(&status); + if (WIFSIGNALED(status)) + psignal(WTERMSIG(status), argv[0]); } else { while(!done) sleep(1); } @@ -519,9 +529,9 @@ static void print_noise_pct(double total, double avg) pct = 100.0*total/avg; if (csv_output) - fprintf(stderr, "%s%.2f%%", csv_sep, pct); - else - fprintf(stderr, " ( +-%6.2f%% )", pct); + fprintf(output, "%s%.2f%%", csv_sep, pct); + else if (pct) + fprintf(output, " ( +-%6.2f%% )", pct); } static void print_noise(struct perf_evsel *evsel, double avg) @@ -546,16 +556,17 @@ static void nsec_printout(int cpu, struct perf_evsel *evsel, double avg) csv_output ? 0 : -4, evsel_list->cpus->map[cpu], csv_sep); - fprintf(stderr, fmt, cpustr, msecs, csv_sep, event_name(evsel)); + fprintf(output, fmt, cpustr, msecs, csv_sep, event_name(evsel)); if (evsel->cgrp) - fprintf(stderr, "%s%s", csv_sep, evsel->cgrp->name); + fprintf(output, "%s%s", csv_sep, evsel->cgrp->name); if (csv_output) return; if (perf_evsel__match(evsel, SOFTWARE, SW_TASK_CLOCK)) - fprintf(stderr, " # %8.3f CPUs utilized ", avg / avg_stats(&walltime_nsecs_stats)); + fprintf(output, " # %8.3f CPUs utilized ", + avg / avg_stats(&walltime_nsecs_stats)); } static void print_stalled_cycles_frontend(int cpu, struct perf_evsel *evsel __used, double avg) @@ -576,9 +587,9 @@ static void print_stalled_cycles_frontend(int cpu, struct perf_evsel *evsel __us else if (ratio > 10.0) color = PERF_COLOR_YELLOW; - fprintf(stderr, " # "); - color_fprintf(stderr, color, "%6.2f%%", ratio); - fprintf(stderr, " frontend cycles idle "); + fprintf(output, " # "); + color_fprintf(output, color, "%6.2f%%", ratio); + fprintf(output, " frontend cycles idle "); } static void print_stalled_cycles_backend(int cpu, struct perf_evsel *evsel __used, double avg) @@ -599,9 +610,9 @@ static void print_stalled_cycles_backend(int cpu, struct perf_evsel *evsel __use else if (ratio > 20.0) color = PERF_COLOR_YELLOW; - fprintf(stderr, " # "); - color_fprintf(stderr, color, "%6.2f%%", ratio); - fprintf(stderr, " backend cycles idle "); + fprintf(output, " # "); + color_fprintf(output, color, "%6.2f%%", ratio); + fprintf(output, " backend cycles idle "); } static void print_branch_misses(int cpu, struct perf_evsel *evsel __used, double avg) @@ -622,9 +633,9 @@ static void print_branch_misses(int cpu, struct perf_evsel *evsel __used, double else if (ratio > 5.0) color = PERF_COLOR_YELLOW; - fprintf(stderr, " # "); - color_fprintf(stderr, color, "%6.2f%%", ratio); - fprintf(stderr, " of all branches "); + fprintf(output, " # "); + color_fprintf(output, color, "%6.2f%%", ratio); + fprintf(output, " of all branches "); } static void print_l1_dcache_misses(int cpu, struct perf_evsel *evsel __used, double avg) @@ -645,9 +656,9 @@ static void print_l1_dcache_misses(int cpu, struct perf_evsel *evsel __used, dou else if (ratio > 5.0) color = PERF_COLOR_YELLOW; - fprintf(stderr, " # "); - color_fprintf(stderr, color, "%6.2f%%", ratio); - fprintf(stderr, " of all L1-dcache hits "); + fprintf(output, " # "); + color_fprintf(output, color, "%6.2f%%", ratio); + fprintf(output, " of all L1-dcache hits "); } static void print_l1_icache_misses(int cpu, struct perf_evsel *evsel __used, double avg) @@ -668,9 +679,9 @@ static void print_l1_icache_misses(int cpu, struct perf_evsel *evsel __used, dou else if (ratio > 5.0) color = PERF_COLOR_YELLOW; - fprintf(stderr, " # "); - color_fprintf(stderr, color, "%6.2f%%", ratio); - fprintf(stderr, " of all L1-icache hits "); + fprintf(output, " # "); + color_fprintf(output, color, "%6.2f%%", ratio); + fprintf(output, " of all L1-icache hits "); } static void print_dtlb_cache_misses(int cpu, struct perf_evsel *evsel __used, double avg) @@ -691,9 +702,9 @@ static void print_dtlb_cache_misses(int cpu, struct perf_evsel *evsel __used, do else if (ratio > 5.0) color = PERF_COLOR_YELLOW; - fprintf(stderr, " # "); - color_fprintf(stderr, color, "%6.2f%%", ratio); - fprintf(stderr, " of all dTLB cache hits "); + fprintf(output, " # "); + color_fprintf(output, color, "%6.2f%%", ratio); + fprintf(output, " of all dTLB cache hits "); } static void print_itlb_cache_misses(int cpu, struct perf_evsel *evsel __used, double avg) @@ -714,9 +725,9 @@ static void print_itlb_cache_misses(int cpu, struct perf_evsel *evsel __used, do else if (ratio > 5.0) color = PERF_COLOR_YELLOW; - fprintf(stderr, " # "); - color_fprintf(stderr, color, "%6.2f%%", ratio); - fprintf(stderr, " of all iTLB cache hits "); + fprintf(output, " # "); + color_fprintf(output, color, "%6.2f%%", ratio); + fprintf(output, " of all iTLB cache hits "); } static void print_ll_cache_misses(int cpu, struct perf_evsel *evsel __used, double avg) @@ -737,9 +748,9 @@ static void print_ll_cache_misses(int cpu, struct perf_evsel *evsel __used, doub else if (ratio > 5.0) color = PERF_COLOR_YELLOW; - fprintf(stderr, " # "); - color_fprintf(stderr, color, "%6.2f%%", ratio); - fprintf(stderr, " of all LL-cache hits "); + fprintf(output, " # "); + color_fprintf(output, color, "%6.2f%%", ratio); + fprintf(output, " of all LL-cache hits "); } static void abs_printout(int cpu, struct perf_evsel *evsel, double avg) @@ -762,10 +773,10 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg) else cpu = 0; - fprintf(stderr, fmt, cpustr, avg, csv_sep, event_name(evsel)); + fprintf(output, fmt, cpustr, avg, csv_sep, event_name(evsel)); if (evsel->cgrp) - fprintf(stderr, "%s%s", csv_sep, evsel->cgrp->name); + fprintf(output, "%s%s", csv_sep, evsel->cgrp->name); if (csv_output) return; @@ -776,14 +787,14 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg) if (total) ratio = avg / total; - fprintf(stderr, " # %5.2f insns per cycle ", ratio); + fprintf(output, " # %5.2f insns per cycle ", ratio); total = avg_stats(&runtime_stalled_cycles_front_stats[cpu]); total = max(total, avg_stats(&runtime_stalled_cycles_back_stats[cpu])); if (total && avg) { ratio = total / avg; - fprintf(stderr, "\n # %5.2f stalled cycles per insn", ratio); + fprintf(output, "\n # %5.2f stalled cycles per insn", ratio); } } else if (perf_evsel__match(evsel, HARDWARE, HW_BRANCH_MISSES) && @@ -831,7 +842,7 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg) if (total) ratio = avg * 100 / total; - fprintf(stderr, " # %8.3f %% of all cache refs ", ratio); + fprintf(output, " # %8.3f %% of all cache refs ", ratio); } else if (perf_evsel__match(evsel, HARDWARE, HW_STALLED_CYCLES_FRONTEND)) { print_stalled_cycles_frontend(cpu, evsel, avg); @@ -843,16 +854,16 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg) if (total) ratio = 1.0 * avg / total; - fprintf(stderr, " # %8.3f GHz ", ratio); + fprintf(output, " # %8.3f GHz ", ratio); } else if (runtime_nsecs_stats[cpu].n != 0) { total = avg_stats(&runtime_nsecs_stats[cpu]); if (total) ratio = 1000.0 * avg / total; - fprintf(stderr, " # %8.3f M/sec ", ratio); + fprintf(output, " # %8.3f M/sec ", ratio); } else { - fprintf(stderr, " "); + fprintf(output, " "); } } @@ -867,7 +878,7 @@ static void print_counter_aggr(struct perf_evsel *counter) int scaled = counter->counts->scaled; if (scaled == -1) { - fprintf(stderr, "%*s%s%*s", + fprintf(output, "%*s%s%*s", csv_output ? 0 : 18, counter->supported ? CNTR_NOT_COUNTED : CNTR_NOT_SUPPORTED, csv_sep, @@ -875,9 +886,9 @@ static void print_counter_aggr(struct perf_evsel *counter) event_name(counter)); if (counter->cgrp) - fprintf(stderr, "%s%s", csv_sep, counter->cgrp->name); + fprintf(output, "%s%s", csv_sep, counter->cgrp->name); - fputc('\n', stderr); + fputc('\n', output); return; } @@ -889,7 +900,7 @@ static void print_counter_aggr(struct perf_evsel *counter) print_noise(counter, avg); if (csv_output) { - fputc('\n', stderr); + fputc('\n', output); return; } @@ -899,9 +910,9 @@ static void print_counter_aggr(struct perf_evsel *counter) avg_enabled = avg_stats(&ps->res_stats[1]); avg_running = avg_stats(&ps->res_stats[2]); - fprintf(stderr, " [%5.2f%%]", 100 * avg_running / avg_enabled); + fprintf(output, " [%5.2f%%]", 100 * avg_running / avg_enabled); } - fprintf(stderr, "\n"); + fprintf(output, "\n"); } /* @@ -918,7 +929,7 @@ static void print_counter(struct perf_evsel *counter) ena = counter->counts->cpu[cpu].ena; run = counter->counts->cpu[cpu].run; if (run == 0 || ena == 0) { - fprintf(stderr, "CPU%*d%s%*s%s%*s", + fprintf(output, "CPU%*d%s%*s%s%*s", csv_output ? 0 : -4, evsel_list->cpus->map[cpu], csv_sep, csv_output ? 0 : 18, @@ -928,9 +939,10 @@ static void print_counter(struct perf_evsel *counter) event_name(counter)); if (counter->cgrp) - fprintf(stderr, "%s%s", csv_sep, counter->cgrp->name); + fprintf(output, "%s%s", + csv_sep, counter->cgrp->name); - fputc('\n', stderr); + fputc('\n', output); continue; } @@ -943,9 +955,10 @@ static void print_counter(struct perf_evsel *counter) print_noise(counter, 1.0); if (run != ena) - fprintf(stderr, " (%.2f%%)", 100.0 * run / ena); + fprintf(output, " (%.2f%%)", + 100.0 * run / ena); } - fputc('\n', stderr); + fputc('\n', output); } } @@ -957,21 +970,21 @@ static void print_stat(int argc, const char **argv) fflush(stdout); if (!csv_output) { - fprintf(stderr, "\n"); - fprintf(stderr, " Performance counter stats for "); + fprintf(output, "\n"); + fprintf(output, " Performance counter stats for "); if(target_pid == -1 && target_tid == -1) { - fprintf(stderr, "\'%s", argv[0]); + fprintf(output, "\'%s", argv[0]); for (i = 1; i < argc; i++) - fprintf(stderr, " %s", argv[i]); + fprintf(output, " %s", argv[i]); } else if (target_pid != -1) - fprintf(stderr, "process id \'%d", target_pid); + fprintf(output, "process id \'%d", target_pid); else - fprintf(stderr, "thread id \'%d", target_tid); + fprintf(output, "thread id \'%d", target_tid); - fprintf(stderr, "\'"); + fprintf(output, "\'"); if (run_count > 1) - fprintf(stderr, " (%d runs)", run_count); - fprintf(stderr, ":\n\n"); + fprintf(output, " (%d runs)", run_count); + fprintf(output, ":\n\n"); } if (no_aggr) { @@ -984,15 +997,15 @@ static void print_stat(int argc, const char **argv) if (!csv_output) { if (!null_run) - fprintf(stderr, "\n"); - fprintf(stderr, " %17.9f seconds time elapsed", + fprintf(output, "\n"); + fprintf(output, " %17.9f seconds time elapsed", avg_stats(&walltime_nsecs_stats)/1e9); if (run_count > 1) { - fprintf(stderr, " "); + fprintf(output, " "); print_noise_pct(stddev_stats(&walltime_nsecs_stats), avg_stats(&walltime_nsecs_stats)); } - fprintf(stderr, "\n\n"); + fprintf(output, "\n\n"); } } @@ -1030,6 +1043,8 @@ static int stat__set_big_num(const struct option *opt __used, return 0; } +static bool append_file; + static const struct option options[] = { OPT_CALLBACK('e', "event", &evsel_list, "event", "event selector. use 'perf list' to list available events", @@ -1070,6 +1085,11 @@ static const struct option options[] = { OPT_CALLBACK('G', "cgroup", &evsel_list, "name", "monitor event in cgroup name only", parse_cgroups), + OPT_STRING('o', "output", &output_name, "file", + "output file name"), + OPT_BOOLEAN(0, "append", &append_file, "append to the output file"), + OPT_INTEGER(0, "log-fd", &output_fd, + "log output to fd, instead of stderr"), OPT_END() }; @@ -1141,6 +1161,7 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used) { struct perf_evsel *pos; int status = -ENOMEM; + const char *mode; setlocale(LC_ALL, ""); @@ -1151,16 +1172,46 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used) argc = parse_options(argc, argv, options, stat_usage, PARSE_OPT_STOP_AT_NON_OPTION); - if (csv_sep) + output = stderr; + if (output_name && strcmp(output_name, "-")) + output = NULL; + + if (output_name && output_fd) { + fprintf(stderr, "cannot use both --output and --log-fd\n"); + usage_with_options(stat_usage, options); + } + if (!output) { + struct timespec tm; + mode = append_file ? "a" : "w"; + + output = fopen(output_name, mode); + if (!output) { + perror("failed to create output file"); + exit(-1); + } + clock_gettime(CLOCK_REALTIME, &tm); + fprintf(output, "# started on %s\n", ctime(&tm.tv_sec)); + } else if (output_fd != 2) { + mode = append_file ? "a" : "w"; + output = fdopen(output_fd, mode); + if (!output) { + perror("Failed opening logfd"); + return -errno; + } + } + + if (csv_sep) { csv_output = true; - else + if (!strcmp(csv_sep, "\\t")) + csv_sep = "\t"; + } else csv_sep = DEFAULT_SEPARATOR; /* * let the spreadsheet do the pretty-printing */ if (csv_output) { - /* User explicitely passed -B? */ + /* User explicitly passed -B? */ if (big_num_opt == 1) { fprintf(stderr, "-B option not supported with -x\n"); usage_with_options(stat_usage, options); @@ -1226,7 +1277,8 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used) status = 0; for (run_idx = 0; run_idx < run_count; run_idx++) { if (run_count != 1 && verbose) - fprintf(stderr, "[ perf stat: executing run #%d ... ]\n", run_idx + 1); + fprintf(output, "[ perf stat: executing run #%d ... ]\n", + run_idx + 1); if (sync_run) sync(); diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c index d28013b7d61..5ede7d7c923 100644 --- a/tools/perf/builtin-top.c +++ b/tools/perf/builtin-top.c @@ -251,7 +251,7 @@ static void __list_insert_active_sym(struct sym_entry *syme) list_add(&syme->node, &top.active_symbols); } -static void print_sym_table(struct perf_session *session) +static void print_sym_table(void) { char bf[160]; int printed = 0; @@ -271,10 +271,11 @@ static void print_sym_table(struct perf_session *session) printf("%-*.*s\n", win_width, win_width, graph_dotted_line); - if (session->hists.stats.total_lost != 0) { + if (top.total_lost_warned != top.session->hists.stats.total_lost) { + top.total_lost_warned = top.session->hists.stats.total_lost; color_fprintf(stdout, PERF_COLOR_RED, "WARNING:"); printf(" LOST %" PRIu64 " events, Check IO/CPU overload\n", - session->hists.stats.total_lost); + top.total_lost_warned); } if (top.sym_filter_entry) { @@ -475,7 +476,7 @@ static int key_mapped(int c) return 0; } -static void handle_keypress(struct perf_session *session, int c) +static void handle_keypress(int c) { if (!key_mapped(c)) { struct pollfd stdin_poll = { .fd = 0, .events = POLLIN }; @@ -551,7 +552,7 @@ static void handle_keypress(struct perf_session *session, int c) case 'Q': printf("exiting.\n"); if (dump_symtab) - perf_session__fprintf_dsos(session, stderr); + perf_session__fprintf_dsos(top.session, stderr); exit(0); case 's': prompt_symbol(&top.sym_filter_entry, "Enter details symbol"); @@ -603,7 +604,6 @@ static void *display_thread(void *arg __used) struct pollfd stdin_poll = { .fd = 0, .events = POLLIN }; struct termios tc, save; int delay_msecs, c; - struct perf_session *session = (struct perf_session *) arg; tcgetattr(0, &save); tc = save; @@ -618,13 +618,13 @@ repeat: getc(stdin); do { - print_sym_table(session); + print_sym_table(); } while (!poll(&stdin_poll, 1, delay_msecs) == 1); c = getc(stdin); tcsetattr(0, TCSAFLUSH, &save); - handle_keypress(session, c); + handle_keypress(c); goto repeat; return NULL; @@ -936,27 +936,27 @@ static int __cmd_top(void) * FIXME: perf_session__new should allow passing a O_MMAP, so that all this * mmap reading, etc is encapsulated in it. Use O_WRONLY for now. */ - struct perf_session *session = perf_session__new(NULL, O_WRONLY, false, false, NULL); - if (session == NULL) + top.session = perf_session__new(NULL, O_WRONLY, false, false, NULL); + if (top.session == NULL) return -ENOMEM; if (top.target_tid != -1) perf_event__synthesize_thread_map(top.evlist->threads, - perf_event__process, session); + perf_event__process, top.session); else - perf_event__synthesize_threads(perf_event__process, session); + perf_event__synthesize_threads(perf_event__process, top.session); start_counters(top.evlist); - session->evlist = top.evlist; - perf_session__update_sample_type(session); + top.session->evlist = top.evlist; + perf_session__update_sample_type(top.session); /* Wait for a minimal set of events before starting the snapshot */ poll(top.evlist->pollfd, top.evlist->nr_fds, 100); - perf_session__mmap_read(session); + perf_session__mmap_read(top.session); if (pthread_create(&thread, NULL, (use_browser > 0 ? display_thread_tui : - display_thread), session)) { + display_thread), NULL)) { printf("Could not create display thread.\n"); exit(-1); } @@ -974,7 +974,7 @@ static int __cmd_top(void) while (1) { u64 hits = top.samples; - perf_session__mmap_read(session); + perf_session__mmap_read(top.session); if (hits == top.samples) ret = poll(top.evlist->pollfd, top.evlist->nr_fds, 100); diff --git a/tools/perf/scripts/python/bin/net_dropmonitor-record b/tools/perf/scripts/python/bin/net_dropmonitor-record new file mode 100755 index 00000000000..423fb81dada --- /dev/null +++ b/tools/perf/scripts/python/bin/net_dropmonitor-record @@ -0,0 +1,2 @@ +#!/bin/bash +perf record -e skb:kfree_skb $@ diff --git a/tools/perf/scripts/python/bin/net_dropmonitor-report b/tools/perf/scripts/python/bin/net_dropmonitor-report new file mode 100755 index 00000000000..8d698f5a06a --- /dev/null +++ b/tools/perf/scripts/python/bin/net_dropmonitor-report @@ -0,0 +1,4 @@ +#!/bin/bash +# description: display a table of dropped frames + +perf script -s "$PERF_EXEC_PATH"/scripts/python/net_dropmonitor.py $@ diff --git a/tools/perf/scripts/python/net_dropmonitor.py b/tools/perf/scripts/python/net_dropmonitor.py new file mode 100755 index 00000000000..a4ffc950002 --- /dev/null +++ b/tools/perf/scripts/python/net_dropmonitor.py @@ -0,0 +1,72 @@ +# Monitor the system for dropped packets and proudce a report of drop locations and counts + +import os +import sys + +sys.path.append(os.environ['PERF_EXEC_PATH'] + \ + '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') + +from perf_trace_context import * +from Core import * +from Util import * + +drop_log = {} +kallsyms = [] + +def get_kallsyms_table(): + global kallsyms + try: + f = open("/proc/kallsyms", "r") + linecount = 0 + for line in f: + linecount = linecount+1 + f.seek(0) + except: + return + + + j = 0 + for line in f: + loc = int(line.split()[0], 16) + name = line.split()[2] + j = j +1 + if ((j % 100) == 0): + print "\r" + str(j) + "/" + str(linecount), + kallsyms.append({ 'loc': loc, 'name' : name}) + + print "\r" + str(j) + "/" + str(linecount) + kallsyms.sort() + return + +def get_sym(sloc): + loc = int(sloc) + for i in kallsyms: + if (i['loc'] >= loc): + return (i['name'], i['loc']-loc) + return (None, 0) + +def print_drop_table(): + print "%25s %25s %25s" % ("LOCATION", "OFFSET", "COUNT") + for i in drop_log.keys(): + (sym, off) = get_sym(i) + if sym == None: + sym = i + print "%25s %25s %25s" % (sym, off, drop_log[i]) + + +def trace_begin(): + print "Starting trace (Ctrl-C to dump results)" + +def trace_end(): + print "Gathering kallsyms data" + get_kallsyms_table() + print_drop_table() + +# called from perf, when it finds a correspoinding event +def skb__kfree_skb(name, context, cpu, sec, nsec, pid, comm, + skbaddr, protocol, location): + slocation = str(location) + try: + drop_log[slocation] = drop_log[slocation] + 1 + except: + drop_log[slocation] = 1 diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c index e01af2b1a46..bc8f4773d4d 100644 --- a/tools/perf/util/annotate.c +++ b/tools/perf/util/annotate.c @@ -16,6 +16,8 @@ #include "annotate.h" #include <pthread.h> +const char *disassembler_style; + int symbol__annotate_init(struct map *map __used, struct symbol *sym) { struct annotation *notes = symbol__annotation(sym); @@ -323,10 +325,15 @@ fallback: dso, dso->long_name, sym, sym->name); snprintf(command, sizeof(command), - "objdump --start-address=0x%016" PRIx64 - " --stop-address=0x%016" PRIx64 " -dS -C %s|grep -v %s|expand", + "objdump %s%s --start-address=0x%016" PRIx64 + " --stop-address=0x%016" PRIx64 + " -d %s %s -C %s|grep -v %s|expand", + disassembler_style ? "-M " : "", + disassembler_style ? disassembler_style : "", map__rip_2objdump(map, sym->start), map__rip_2objdump(map, sym->end), + symbol_conf.annotate_asm_raw ? "" : "--no-show-raw", + symbol_conf.annotate_src ? "-S" : "", symfs_filename, filename); pr_debug("Executing: %s\n", command); diff --git a/tools/perf/util/annotate.h b/tools/perf/util/annotate.h index c2c28689680..6ede1286ee7 100644 --- a/tools/perf/util/annotate.h +++ b/tools/perf/util/annotate.h @@ -100,4 +100,6 @@ int symbol__tui_annotate(struct symbol *sym, struct map *map, int evidx, int refresh); #endif +extern const char *disassembler_style; + #endif /* __PERF_ANNOTATE_H */ diff --git a/tools/perf/util/color.c b/tools/perf/util/color.c index e191eb9a667..521c38a7919 100644 --- a/tools/perf/util/color.c +++ b/tools/perf/util/color.c @@ -200,7 +200,7 @@ static int __color_vfprintf(FILE *fp, const char *color, const char *fmt, * Auto-detect: */ if (perf_use_color_default < 0) { - if (isatty(1) || pager_in_use()) + if (isatty(fileno(fp)) || pager_in_use()) perf_use_color_default = 1; else perf_use_color_default = 0; diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c index a16ecab5229..9cf0d4393c8 100644 --- a/tools/perf/util/map.c +++ b/tools/perf/util/map.c @@ -18,6 +18,13 @@ static inline int is_anon_memory(const char *filename) return strcmp(filename, "//anon") == 0; } +static inline int is_no_dso_memory(const char *filename) +{ + return !strcmp(filename, "[stack]") || + !strcmp(filename, "[vdso]") || + !strcmp(filename, "[heap]"); +} + void map__init(struct map *self, enum map_type type, u64 start, u64 end, u64 pgoff, struct dso *dso) { @@ -42,9 +49,10 @@ struct map *map__new(struct list_head *dsos__list, u64 start, u64 len, if (self != NULL) { char newfilename[PATH_MAX]; struct dso *dso; - int anon; + int anon, no_dso; anon = is_anon_memory(filename); + no_dso = is_no_dso_memory(filename); if (anon) { snprintf(newfilename, sizeof(newfilename), "/tmp/perf-%d.map", pid); @@ -57,12 +65,16 @@ struct map *map__new(struct list_head *dsos__list, u64 start, u64 len, map__init(self, type, start, start + len, pgoff, dso); - if (anon) { -set_identity: + if (anon || no_dso) { self->map_ip = self->unmap_ip = identity__map_ip; - } else if (strcmp(filename, "[vdso]") == 0) { - dso__set_loaded(dso, self->type); - goto set_identity; + + /* + * Set memory without DSO as loaded. All map__find_* + * functions still return NULL, and we avoid the + * unnecessary map__load warning. + */ + if (no_dso) + dso__set_loaded(dso, self->type); } } return self; @@ -220,55 +232,55 @@ u64 map__objdump_2ip(struct map *map, u64 addr) return ip; } -void map_groups__init(struct map_groups *self) +void map_groups__init(struct map_groups *mg) { int i; for (i = 0; i < MAP__NR_TYPES; ++i) { - self->maps[i] = RB_ROOT; - INIT_LIST_HEAD(&self->removed_maps[i]); + mg->maps[i] = RB_ROOT; + INIT_LIST_HEAD(&mg->removed_maps[i]); } - self->machine = NULL; + mg->machine = NULL; } -static void maps__delete(struct rb_root *self) +static void maps__delete(struct rb_root *maps) { - struct rb_node *next = rb_first(self); + struct rb_node *next = rb_first(maps); while (next) { struct map *pos = rb_entry(next, struct map, rb_node); next = rb_next(&pos->rb_node); - rb_erase(&pos->rb_node, self); + rb_erase(&pos->rb_node, maps); map__delete(pos); } } -static void maps__delete_removed(struct list_head *self) +static void maps__delete_removed(struct list_head *maps) { struct map *pos, *n; - list_for_each_entry_safe(pos, n, self, node) { + list_for_each_entry_safe(pos, n, maps, node) { list_del(&pos->node); map__delete(pos); } } -void map_groups__exit(struct map_groups *self) +void map_groups__exit(struct map_groups *mg) { int i; for (i = 0; i < MAP__NR_TYPES; ++i) { - maps__delete(&self->maps[i]); - maps__delete_removed(&self->removed_maps[i]); + maps__delete(&mg->maps[i]); + maps__delete_removed(&mg->removed_maps[i]); } } -void map_groups__flush(struct map_groups *self) +void map_groups__flush(struct map_groups *mg) { int type; for (type = 0; type < MAP__NR_TYPES; type++) { - struct rb_root *root = &self->maps[type]; + struct rb_root *root = &mg->maps[type]; struct rb_node *next = rb_first(root); while (next) { @@ -280,17 +292,17 @@ void map_groups__flush(struct map_groups *self) * instance in some hist_entry instances, so * just move them to a separate list. */ - list_add_tail(&pos->node, &self->removed_maps[pos->type]); + list_add_tail(&pos->node, &mg->removed_maps[pos->type]); } } } -struct symbol *map_groups__find_symbol(struct map_groups *self, +struct symbol *map_groups__find_symbol(struct map_groups *mg, enum map_type type, u64 addr, struct map **mapp, symbol_filter_t filter) { - struct map *map = map_groups__find(self, type, addr); + struct map *map = map_groups__find(mg, type, addr); if (map != NULL) { if (mapp != NULL) @@ -301,7 +313,7 @@ struct symbol *map_groups__find_symbol(struct map_groups *self, return NULL; } -struct symbol *map_groups__find_symbol_by_name(struct map_groups *self, +struct symbol *map_groups__find_symbol_by_name(struct map_groups *mg, enum map_type type, const char *name, struct map **mapp, @@ -309,7 +321,7 @@ struct symbol *map_groups__find_symbol_by_name(struct map_groups *self, { struct rb_node *nd; - for (nd = rb_first(&self->maps[type]); nd; nd = rb_next(nd)) { + for (nd = rb_first(&mg->maps[type]); nd; nd = rb_next(nd)) { struct map *pos = rb_entry(nd, struct map, rb_node); struct symbol *sym = map__find_symbol_by_name(pos, name, filter); @@ -323,13 +335,13 @@ struct symbol *map_groups__find_symbol_by_name(struct map_groups *self, return NULL; } -size_t __map_groups__fprintf_maps(struct map_groups *self, +size_t __map_groups__fprintf_maps(struct map_groups *mg, enum map_type type, int verbose, FILE *fp) { size_t printed = fprintf(fp, "%s:\n", map_type__name[type]); struct rb_node *nd; - for (nd = rb_first(&self->maps[type]); nd; nd = rb_next(nd)) { + for (nd = rb_first(&mg->maps[type]); nd; nd = rb_next(nd)) { struct map *pos = rb_entry(nd, struct map, rb_node); printed += fprintf(fp, "Map:"); printed += map__fprintf(pos, fp); @@ -342,22 +354,22 @@ size_t __map_groups__fprintf_maps(struct map_groups *self, return printed; } -size_t map_groups__fprintf_maps(struct map_groups *self, int verbose, FILE *fp) +size_t map_groups__fprintf_maps(struct map_groups *mg, int verbose, FILE *fp) { size_t printed = 0, i; for (i = 0; i < MAP__NR_TYPES; ++i) - printed += __map_groups__fprintf_maps(self, i, verbose, fp); + printed += __map_groups__fprintf_maps(mg, i, verbose, fp); return printed; } -static size_t __map_groups__fprintf_removed_maps(struct map_groups *self, +static size_t __map_groups__fprintf_removed_maps(struct map_groups *mg, enum map_type type, int verbose, FILE *fp) { struct map *pos; size_t printed = 0; - list_for_each_entry(pos, &self->removed_maps[type], node) { + list_for_each_entry(pos, &mg->removed_maps[type], node) { printed += fprintf(fp, "Map:"); printed += map__fprintf(pos, fp); if (verbose > 1) { @@ -368,26 +380,26 @@ static size_t __map_groups__fprintf_removed_maps(struct map_groups *self, return printed; } -static size_t map_groups__fprintf_removed_maps(struct map_groups *self, +static size_t map_groups__fprintf_removed_maps(struct map_groups *mg, int verbose, FILE *fp) { size_t printed = 0, i; for (i = 0; i < MAP__NR_TYPES; ++i) - printed += __map_groups__fprintf_removed_maps(self, i, verbose, fp); + printed += __map_groups__fprintf_removed_maps(mg, i, verbose, fp); return printed; } -size_t map_groups__fprintf(struct map_groups *self, int verbose, FILE *fp) +size_t map_groups__fprintf(struct map_groups *mg, int verbose, FILE *fp) { - size_t printed = map_groups__fprintf_maps(self, verbose, fp); + size_t printed = map_groups__fprintf_maps(mg, verbose, fp); printed += fprintf(fp, "Removed maps:\n"); - return printed + map_groups__fprintf_removed_maps(self, verbose, fp); + return printed + map_groups__fprintf_removed_maps(mg, verbose, fp); } -int map_groups__fixup_overlappings(struct map_groups *self, struct map *map, +int map_groups__fixup_overlappings(struct map_groups *mg, struct map *map, int verbose, FILE *fp) { - struct rb_root *root = &self->maps[map->type]; + struct rb_root *root = &mg->maps[map->type]; struct rb_node *next = rb_first(root); int err = 0; @@ -418,7 +430,7 @@ int map_groups__fixup_overlappings(struct map_groups *self, struct map *map, } before->end = map->start - 1; - map_groups__insert(self, before); + map_groups__insert(mg, before); if (verbose >= 2) map__fprintf(before, fp); } @@ -432,7 +444,7 @@ int map_groups__fixup_overlappings(struct map_groups *self, struct map *map, } after->start = map->end + 1; - map_groups__insert(self, after); + map_groups__insert(mg, after); if (verbose >= 2) map__fprintf(after, fp); } @@ -441,7 +453,7 @@ move_map: * If we have references, just move them to a separate list. */ if (pos->referenced) - list_add_tail(&pos->node, &self->removed_maps[map->type]); + list_add_tail(&pos->node, &mg->removed_maps[map->type]); else map__delete(pos); @@ -455,7 +467,7 @@ move_map: /* * XXX This should not really _copy_ te maps, but refcount them. */ -int map_groups__clone(struct map_groups *self, +int map_groups__clone(struct map_groups *mg, struct map_groups *parent, enum map_type type) { struct rb_node *nd; @@ -464,7 +476,7 @@ int map_groups__clone(struct map_groups *self, struct map *new = map__clone(map); if (new == NULL) return -ENOMEM; - map_groups__insert(self, new); + map_groups__insert(mg, new); } return 0; } diff --git a/tools/perf/util/map.h b/tools/perf/util/map.h index b397c038372..890d85545d0 100644 --- a/tools/perf/util/map.h +++ b/tools/perf/util/map.h @@ -123,17 +123,17 @@ void map__fixup_end(struct map *self); void map__reloc_vmlinux(struct map *self); -size_t __map_groups__fprintf_maps(struct map_groups *self, +size_t __map_groups__fprintf_maps(struct map_groups *mg, enum map_type type, int verbose, FILE *fp); void maps__insert(struct rb_root *maps, struct map *map); -void maps__remove(struct rb_root *self, struct map *map); +void maps__remove(struct rb_root *maps, struct map *map); struct map *maps__find(struct rb_root *maps, u64 addr); -void map_groups__init(struct map_groups *self); -void map_groups__exit(struct map_groups *self); -int map_groups__clone(struct map_groups *self, +void map_groups__init(struct map_groups *mg); +void map_groups__exit(struct map_groups *mg); +int map_groups__clone(struct map_groups *mg, struct map_groups *parent, enum map_type type); -size_t map_groups__fprintf(struct map_groups *self, int verbose, FILE *fp); -size_t map_groups__fprintf_maps(struct map_groups *self, int verbose, FILE *fp); +size_t map_groups__fprintf(struct map_groups *mg, int verbose, FILE *fp); +size_t map_groups__fprintf_maps(struct map_groups *mg, int verbose, FILE *fp); typedef void (*machine__process_t)(struct machine *self, void *data); @@ -162,29 +162,29 @@ static inline bool machine__is_host(struct machine *self) return self ? self->pid == HOST_KERNEL_ID : false; } -static inline void map_groups__insert(struct map_groups *self, struct map *map) +static inline void map_groups__insert(struct map_groups *mg, struct map *map) { - maps__insert(&self->maps[map->type], map); - map->groups = self; + maps__insert(&mg->maps[map->type], map); + map->groups = mg; } -static inline void map_groups__remove(struct map_groups *self, struct map *map) +static inline void map_groups__remove(struct map_groups *mg, struct map *map) { - maps__remove(&self->maps[map->type], map); + maps__remove(&mg->maps[map->type], map); } -static inline struct map *map_groups__find(struct map_groups *self, +static inline struct map *map_groups__find(struct map_groups *mg, enum map_type type, u64 addr) { - return maps__find(&self->maps[type], addr); + return maps__find(&mg->maps[type], addr); } -struct symbol *map_groups__find_symbol(struct map_groups *self, +struct symbol *map_groups__find_symbol(struct map_groups *mg, enum map_type type, u64 addr, struct map **mapp, symbol_filter_t filter); -struct symbol *map_groups__find_symbol_by_name(struct map_groups *self, +struct symbol *map_groups__find_symbol_by_name(struct map_groups *mg, enum map_type type, const char *name, struct map **mapp, @@ -208,11 +208,11 @@ struct symbol *machine__find_kernel_function(struct machine *self, u64 addr, } static inline -struct symbol *map_groups__find_function_by_name(struct map_groups *self, +struct symbol *map_groups__find_function_by_name(struct map_groups *mg, const char *name, struct map **mapp, symbol_filter_t filter) { - return map_groups__find_symbol_by_name(self, MAP__FUNCTION, name, mapp, filter); + return map_groups__find_symbol_by_name(mg, MAP__FUNCTION, name, mapp, filter); } static inline @@ -225,13 +225,13 @@ struct symbol *machine__find_kernel_function_by_name(struct machine *self, filter); } -int map_groups__fixup_overlappings(struct map_groups *self, struct map *map, +int map_groups__fixup_overlappings(struct map_groups *mg, struct map *map, int verbose, FILE *fp); -struct map *map_groups__find_by_name(struct map_groups *self, +struct map *map_groups__find_by_name(struct map_groups *mg, enum map_type type, const char *name); struct map *machine__new_module(struct machine *self, u64 start, const char *filename); -void map_groups__flush(struct map_groups *self); +void map_groups__flush(struct map_groups *mg); #endif /* __PERF_MAP_H */ diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c index 40eeaf07725..077df15ee70 100644 --- a/tools/perf/util/symbol.c +++ b/tools/perf/util/symbol.c @@ -46,6 +46,8 @@ struct symbol_conf symbol_conf = { .exclude_other = true, .use_modules = true, .try_vmlinux_path = true, + .annotate_asm_raw = true, + .annotate_src = true, .symfs = "", }; diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h index 4f377d92e75..7733f0b3cd4 100644 --- a/tools/perf/util/symbol.h +++ b/tools/perf/util/symbol.h @@ -76,7 +76,9 @@ struct symbol_conf { exclude_other, show_cpu_utilization, initialized, - kptr_restrict; + kptr_restrict, + annotate_asm_raw, + annotate_src; const char *vmlinux_name, *kallsyms_name, *source_prefix, diff --git a/tools/perf/util/top.h b/tools/perf/util/top.h index bfbf95bcc60..b07b0410463 100644 --- a/tools/perf/util/top.h +++ b/tools/perf/util/top.h @@ -10,6 +10,7 @@ struct perf_evlist; struct perf_evsel; +struct perf_session; struct sym_entry { struct rb_node rb_node; @@ -38,6 +39,7 @@ struct perf_top { u64 kernel_samples, us_samples; u64 exact_samples; u64 guest_us_samples, guest_kernel_samples; + u64 total_lost_warned; int print_entries, count_filter, delay_secs; int display_weighted, freq, rb_entries; pid_t target_pid, target_tid; @@ -45,6 +47,7 @@ struct perf_top { const char *cpu_list; struct sym_entry *sym_filter_entry; struct perf_evsel *sym_evsel; + struct perf_session *session; }; size_t perf_top__header_snprintf(struct perf_top *top, char *bf, size_t size); diff --git a/tools/perf/util/ui/browsers/top.c b/tools/perf/util/ui/browsers/top.c index 88403cf8396..9b6b43b32ac 100644 --- a/tools/perf/util/ui/browsers/top.c +++ b/tools/perf/util/ui/browsers/top.c @@ -11,10 +11,12 @@ #include "../helpline.h" #include "../libslang.h" #include "../util.h" +#include "../ui.h" #include "../../evlist.h" #include "../../hist.h" #include "../../sort.h" #include "../../symbol.h" +#include "../../session.h" #include "../../top.h" struct perf_top_browser { @@ -43,10 +45,10 @@ static void perf_top_browser__write(struct ui_browser *browser, void *entry, int if (top->evlist->nr_entries == 1 || !top->display_weighted) { slsmg_printf("%20.2f ", syme->weight); - width -= 24; + width -= 21; } else { slsmg_printf("%9.1f %10ld ", syme->weight, syme->snap_count); - width -= 23; + width -= 20; } slsmg_printf("%4.1f%%", pcnt); @@ -143,6 +145,25 @@ do_annotation: symbol__tui_annotate(sym, syme->map, 0, top->delay_secs * 1000); } +static void perf_top_browser__warn_lost(struct perf_top_browser *browser) +{ + struct perf_top *top = browser->b.priv; + char msg[128]; + int len; + + top->total_lost_warned = top->session->hists.stats.total_lost; + pthread_mutex_lock(&ui__lock); + ui_browser__set_color(&browser->b, HE_COLORSET_TOP); + len = snprintf(msg, sizeof(msg), + " WARNING: LOST %" PRIu64 " events, Check IO/CPU overload", + top->total_lost_warned); + if (len > browser->b.width) + len = browser->b.width; + SLsmg_gotorc(0, browser->b.width - len); + slsmg_write_nstring(msg, len); + pthread_mutex_unlock(&ui__lock); +} + static int perf_top_browser__run(struct perf_top_browser *browser) { int key; @@ -174,6 +195,9 @@ static int perf_top_browser__run(struct perf_top_browser *browser) ui_browser__set_color(&browser->b, NEWT_COLORSET_ROOT); SLsmg_gotorc(0, 0); slsmg_write_nstring(title, browser->b.width); + + if (top->total_lost_warned != top->session->hists.stats.total_lost) + perf_top_browser__warn_lost(browser); break; case 'a': case NEWT_KEY_RIGHT: |