diff options
64 files changed, 746 insertions, 371 deletions
diff --git a/Documentation/trace/kprobetrace.txt b/Documentation/trace/kprobetrace.txt index ea52ec1f8484..e4991fb1eedc 100644 --- a/Documentation/trace/kprobetrace.txt +++ b/Documentation/trace/kprobetrace.txt @@ -44,8 +44,8 @@ Synopsis of kprobe_events +|-offs(FETCHARG) : Fetch memory at FETCHARG +|- offs address.(**) NAME=FETCHARG : Set NAME as the argument name of FETCHARG. FETCHARG:TYPE : Set TYPE as the type of FETCHARG. Currently, basic types - (u8/u16/u32/u64/s8/s16/s32/s64), "string" and bitfield - are supported. + (u8/u16/u32/u64/s8/s16/s32/s64), hexadecimal types + (x8/x16/x32/x64), "string" and bitfield are supported. (*) only for return probe. (**) this is useful for fetching a field of data structures. @@ -54,7 +54,10 @@ Types ----- Several types are supported for fetch-args. Kprobe tracer will access memory by given type. Prefix 's' and 'u' means those types are signed and unsigned -respectively. Traced arguments are shown in decimal (signed) or hex (unsigned). +respectively. 'x' prefix implies it is unsigned. Traced arguments are shown +in decimal ('s' and 'u') or hexadecimal ('x'). Without type casting, 'x32' +or 'x64' is used depends on the architecture (e.g. x86-32 uses x32, and +x86-64 uses x64). String type is a special type, which fetches a "null-terminated" string from kernel space. This means it will fail and store NULL if the string container has been paged out. diff --git a/Documentation/trace/uprobetracer.txt b/Documentation/trace/uprobetracer.txt index 72d1cd4f7bf3..94b6b4581763 100644 --- a/Documentation/trace/uprobetracer.txt +++ b/Documentation/trace/uprobetracer.txt @@ -40,8 +40,8 @@ Synopsis of uprobe_tracer +|-offs(FETCHARG) : Fetch memory at FETCHARG +|- offs address.(**) NAME=FETCHARG : Set NAME as the argument name of FETCHARG. FETCHARG:TYPE : Set TYPE as the type of FETCHARG. Currently, basic types - (u8/u16/u32/u64/s8/s16/s32/s64), "string" and bitfield - are supported. + (u8/u16/u32/u64/s8/s16/s32/s64), hexadecimal types + (x8/x16/x32/x64), "string" and bitfield are supported. (*) only for return probe. (**) this is useful for fetching a field of data structures. @@ -50,7 +50,10 @@ Types ----- Several types are supported for fetch-args. Uprobe tracer will access memory by given type. Prefix 's' and 'u' means those types are signed and unsigned -respectively. Traced arguments are shown in decimal (signed) or hex (unsigned). +respectively. 'x' prefix implies it is unsigned. Traced arguments are shown +in decimal ('s' and 'u') or hexadecimal ('x'). Without type casting, 'x32' +or 'x64' is used depends on the architecture (e.g. x86-32 uses x32, and +x86-64 uses x64). String type is a special type, which fetches a "null-terminated" string from user space. Bitfield is another special type, which takes 3 parameters, bit-width, bit- diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c index d0efb5cb1b00..18a1acf86c90 100644 --- a/arch/x86/events/core.c +++ b/arch/x86/events/core.c @@ -1201,6 +1201,9 @@ static int x86_pmu_add(struct perf_event *event, int flags) * If group events scheduling transaction was started, * skip the schedulability test here, it will be performed * at commit time (->commit_txn) as a whole. + * + * If commit fails, we'll call ->del() on all events + * for which ->add() was called. */ if (cpuc->txn_flags & PERF_PMU_TXN_ADD) goto done_collect; @@ -1223,6 +1226,14 @@ done_collect: cpuc->n_added += n - n0; cpuc->n_txn += n - n0; + if (x86_pmu.add) { + /* + * This is before x86_pmu_enable() will call x86_pmu_start(), + * so we enable LBRs before an event needs them etc.. + */ + x86_pmu.add(event); + } + ret = 0; out: return ret; @@ -1346,7 +1357,7 @@ static void x86_pmu_del(struct perf_event *event, int flags) event->hw.flags &= ~PERF_X86_EVENT_COMMITTED; /* - * If we're called during a txn, we don't need to do anything. + * If we're called during a txn, we only need to undo x86_pmu.add. * The events never got scheduled and ->cancel_txn will truncate * the event_list. * @@ -1354,7 +1365,7 @@ static void x86_pmu_del(struct perf_event *event, int flags) * an event added during that same TXN. */ if (cpuc->txn_flags & PERF_PMU_TXN_ADD) - return; + goto do_del; /* * Not a TXN, therefore cleanup properly. @@ -1384,6 +1395,15 @@ static void x86_pmu_del(struct perf_event *event, int flags) --cpuc->n_events; perf_event_update_userpage(event); + +do_del: + if (x86_pmu.del) { + /* + * This is after x86_pmu_stop(); so we disable LBRs after any + * event can need them etc.. + */ + x86_pmu.del(event); + } } int x86_pmu_handle_irq(struct pt_regs *regs) diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c index 2cbde2f449aa..88792f846d12 100644 --- a/arch/x86/events/intel/core.c +++ b/arch/x86/events/intel/core.c @@ -1907,13 +1907,6 @@ static void intel_pmu_disable_event(struct perf_event *event) cpuc->intel_ctrl_host_mask &= ~(1ull << hwc->idx); cpuc->intel_cp_status &= ~(1ull << hwc->idx); - /* - * must disable before any actual event - * because any event may be combined with LBR - */ - if (needs_branch_stack(event)) - intel_pmu_lbr_disable(event); - if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) { intel_pmu_disable_fixed(hwc); return; @@ -1925,6 +1918,14 @@ static void intel_pmu_disable_event(struct perf_event *event) intel_pmu_pebs_disable(event); } +static void intel_pmu_del_event(struct perf_event *event) +{ + if (needs_branch_stack(event)) + intel_pmu_lbr_del(event); + if (event->attr.precise_ip) + intel_pmu_pebs_del(event); +} + static void intel_pmu_enable_fixed(struct hw_perf_event *hwc) { int idx = hwc->idx - INTEL_PMC_IDX_FIXED; @@ -1968,12 +1969,6 @@ static void intel_pmu_enable_event(struct perf_event *event) intel_pmu_enable_bts(hwc->config); return; } - /* - * must enabled before any actual event - * because any event may be combined with LBR - */ - if (needs_branch_stack(event)) - intel_pmu_lbr_enable(event); if (event->attr.exclude_host) cpuc->intel_ctrl_guest_mask |= (1ull << hwc->idx); @@ -1994,6 +1989,14 @@ static void intel_pmu_enable_event(struct perf_event *event) __x86_pmu_enable_event(hwc, ARCH_PERFMON_EVENTSEL_ENABLE); } +static void intel_pmu_add_event(struct perf_event *event) +{ + if (event->attr.precise_ip) + intel_pmu_pebs_add(event); + if (needs_branch_stack(event)) + intel_pmu_lbr_add(event); +} + /* * Save and restart an expired event. Called by NMI contexts, * so it has to be careful about preempting normal event ops: @@ -3290,6 +3293,8 @@ static __initconst const struct x86_pmu intel_pmu = { .enable_all = intel_pmu_enable_all, .enable = intel_pmu_enable_event, .disable = intel_pmu_disable_event, + .add = intel_pmu_add_event, + .del = intel_pmu_del_event, .hw_config = intel_pmu_hw_config, .schedule_events = x86_schedule_events, .eventsel = MSR_ARCH_PERFMON_EVENTSEL0, diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c index 7ce9f3f669e6..e0288d555367 100644 --- a/arch/x86/events/intel/ds.c +++ b/arch/x86/events/intel/ds.c @@ -806,9 +806,65 @@ struct event_constraint *intel_pebs_constraints(struct perf_event *event) return &emptyconstraint; } -static inline bool pebs_is_enabled(struct cpu_hw_events *cpuc) +/* + * We need the sched_task callback even for per-cpu events when we use + * the large interrupt threshold, such that we can provide PID and TID + * to PEBS samples. + */ +static inline bool pebs_needs_sched_cb(struct cpu_hw_events *cpuc) +{ + return cpuc->n_pebs && (cpuc->n_pebs == cpuc->n_large_pebs); +} + +static inline void pebs_update_threshold(struct cpu_hw_events *cpuc) +{ + struct debug_store *ds = cpuc->ds; + u64 threshold; + + if (cpuc->n_pebs == cpuc->n_large_pebs) { + threshold = ds->pebs_absolute_maximum - + x86_pmu.max_pebs_events * x86_pmu.pebs_record_size; + } else { + threshold = ds->pebs_buffer_base + x86_pmu.pebs_record_size; + } + + ds->pebs_interrupt_threshold = threshold; +} + +static void +pebs_update_state(bool needed_cb, struct cpu_hw_events *cpuc, struct pmu *pmu) +{ + /* + * Make sure we get updated with the first PEBS + * event. It will trigger also during removal, but + * that does not hurt: + */ + bool update = cpuc->n_pebs == 1; + + if (needed_cb != pebs_needs_sched_cb(cpuc)) { + if (!needed_cb) + perf_sched_cb_inc(pmu); + else + perf_sched_cb_dec(pmu); + + update = true; + } + + if (update) + pebs_update_threshold(cpuc); +} + +void intel_pmu_pebs_add(struct perf_event *event) { - return (cpuc->pebs_enabled & ((1ULL << MAX_PEBS_EVENTS) - 1)); + struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); + struct hw_perf_event *hwc = &event->hw; + bool needed_cb = pebs_needs_sched_cb(cpuc); + + cpuc->n_pebs++; + if (hwc->flags & PERF_X86_EVENT_FREERUNNING) + cpuc->n_large_pebs++; + + pebs_update_state(needed_cb, cpuc, event->ctx->pmu); } void intel_pmu_pebs_enable(struct perf_event *event) @@ -816,12 +872,9 @@ void intel_pmu_pebs_enable(struct perf_event *event) struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); struct hw_perf_event *hwc = &event->hw; struct debug_store *ds = cpuc->ds; - bool first_pebs; - u64 threshold; hwc->config &= ~ARCH_PERFMON_EVENTSEL_INT; - first_pebs = !pebs_is_enabled(cpuc); cpuc->pebs_enabled |= 1ULL << hwc->idx; if (event->hw.flags & PERF_X86_EVENT_PEBS_LDLAT) @@ -830,46 +883,34 @@ void intel_pmu_pebs_enable(struct perf_event *event) cpuc->pebs_enabled |= 1ULL << 63; /* - * When the event is constrained enough we can use a larger - * threshold and run the event with less frequent PMI. + * Use auto-reload if possible to save a MSR write in the PMI. + * This must be done in pmu::start(), because PERF_EVENT_IOC_PERIOD. */ - if (hwc->flags & PERF_X86_EVENT_FREERUNNING) { - threshold = ds->pebs_absolute_maximum - - x86_pmu.max_pebs_events * x86_pmu.pebs_record_size; - - if (first_pebs) - perf_sched_cb_inc(event->ctx->pmu); - } else { - threshold = ds->pebs_buffer_base + x86_pmu.pebs_record_size; - - /* - * If not all events can use larger buffer, - * roll back to threshold = 1 - */ - if (!first_pebs && - (ds->pebs_interrupt_threshold > threshold)) - perf_sched_cb_dec(event->ctx->pmu); - } - - /* Use auto-reload if possible to save a MSR write in the PMI */ if (hwc->flags & PERF_X86_EVENT_AUTO_RELOAD) { ds->pebs_event_reset[hwc->idx] = (u64)(-hwc->sample_period) & x86_pmu.cntval_mask; } +} + +void intel_pmu_pebs_del(struct perf_event *event) +{ + struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); + struct hw_perf_event *hwc = &event->hw; + bool needed_cb = pebs_needs_sched_cb(cpuc); - if (first_pebs || ds->pebs_interrupt_threshold > threshold) - ds->pebs_interrupt_threshold = threshold; + cpuc->n_pebs--; + if (hwc->flags & PERF_X86_EVENT_FREERUNNING) + cpuc->n_large_pebs--; + + pebs_update_state(needed_cb, cpuc, event->ctx->pmu); } void intel_pmu_pebs_disable(struct perf_event *event) { struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); struct hw_perf_event *hwc = &event->hw; - struct debug_store *ds = cpuc->ds; - bool large_pebs = ds->pebs_interrupt_threshold > - ds->pebs_buffer_base + x86_pmu.pebs_record_size; - if (large_pebs) + if (cpuc->n_pebs == cpuc->n_large_pebs) intel_pmu_drain_pebs_buffer(); cpuc->pebs_enabled &= ~(1ULL << hwc->idx); @@ -879,9 +920,6 @@ void intel_pmu_pebs_disable(struct perf_event *event) else if (event->hw.flags & PERF_X86_EVENT_PEBS_ST) cpuc->pebs_enabled &= ~(1ULL << 63); - if (large_pebs && !pebs_is_enabled(cpuc)) - perf_sched_cb_dec(event->ctx->pmu); - if (cpuc->enabled) wrmsrl(MSR_IA32_PEBS_ENABLE, cpuc->pebs_enabled); diff --git a/arch/x86/events/intel/lbr.c b/arch/x86/events/intel/lbr.c index 707d358e0dff..fc6cf21c535e 100644 --- a/arch/x86/events/intel/lbr.c +++ b/arch/x86/events/intel/lbr.c @@ -380,7 +380,6 @@ static void __intel_pmu_lbr_save(struct x86_perf_task_context *task_ctx) void intel_pmu_lbr_sched_task(struct perf_event_context *ctx, bool sched_in) { - struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); struct x86_perf_task_context *task_ctx; /* @@ -390,31 +389,21 @@ void intel_pmu_lbr_sched_task(struct perf_event_context *ctx, bool sched_in) */ task_ctx = ctx ? ctx->task_ctx_data : NULL; if (task_ctx) { - if (sched_in) { + if (sched_in) __intel_pmu_lbr_restore(task_ctx); - cpuc->lbr_context = ctx; - } else { + else __intel_pmu_lbr_save(task_ctx); - } return; } /* - * When sampling the branck stack in system-wide, it may be - * necessary to flush the stack on context switch. This happens - * when the branch stack does not tag its entries with the pid - * of the current task. Otherwise it becomes impossible to - * associate a branch entry with a task. This ambiguity is more - * likely to appear when the branch stack supports priv level - * filtering and the user sets it to monitor only at the user - * level (which could be a useful measurement in system-wide - * mode). In that case, the risk is high of having a branch - * stack with branch from multiple tasks. - */ - if (sched_in) { + * Since a context switch can flip the address space and LBR entries + * are not tagged with an identifier, we need to wipe the LBR, even for + * per-cpu events. You simply cannot resolve the branches from the old + * address space. + */ + if (sched_in) intel_pmu_lbr_reset(); - cpuc->lbr_context = ctx; - } } static inline bool branch_user_callstack(unsigned br_sel) @@ -422,7 +411,7 @@ static inline bool branch_user_callstack(unsigned br_sel) return (br_sel & X86_BR_USER) && (br_sel & X86_BR_CALL_STACK); } -void intel_pmu_lbr_enable(struct perf_event *event) +void intel_pmu_lbr_add(struct perf_event *event) { struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); struct x86_perf_task_context *task_ctx; @@ -430,27 +419,38 @@ void intel_pmu_lbr_enable(struct perf_event *event) if (!x86_pmu.lbr_nr) return; - /* - * Reset the LBR stack if we changed task context to - * avoid data leaks. - */ - if (event->ctx->task && cpuc->lbr_context != event->ctx) { - intel_pmu_lbr_reset(); - cpuc->lbr_context = event->ctx; - } cpuc->br_sel = event->hw.branch_reg.reg; - if (branch_user_callstack(cpuc->br_sel) && event->ctx && - event->ctx->task_ctx_data) { + if (branch_user_callstack(cpuc->br_sel) && event->ctx->task_ctx_data) { task_ctx = event->ctx->task_ctx_data; task_ctx->lbr_callstack_users++; } - cpuc->lbr_users++; + /* + * Request pmu::sched_task() callback, which will fire inside the + * regular perf event scheduling, so that call will: + * + * - restore or wipe; when LBR-callstack, + * - wipe; otherwise, + * + * when this is from __perf_event_task_sched_in(). + * + * However, if this is from perf_install_in_context(), no such callback + * will follow and we'll need to reset the LBR here if this is the + * first LBR event. + * + * The problem is, we cannot tell these cases apart... but we can + * exclude the biggest chunk of cases by looking at + * event->total_time_running. An event that has accrued runtime cannot + * be 'new'. Conversely, a new event can get installed through the + * context switch path for the first time. + */ perf_sched_cb_inc(event->ctx->pmu); + if (!cpuc->lbr_users++ && !event->total_time_running) + intel_pmu_lbr_reset(); } -void intel_pmu_lbr_disable(struct perf_event *event) +void intel_pmu_lbr_del(struct perf_event *event) { struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); struct x86_perf_task_context *task_ctx; @@ -467,12 +467,6 @@ void intel_pmu_lbr_disable(struct perf_event *event) cpuc->lbr_users--; WARN_ON_ONCE(cpuc->lbr_users < 0); perf_sched_cb_dec(event->ctx->pmu); - - if (cpuc->enabled && !cpuc->lbr_users) { - __intel_pmu_lbr_disable(); - /* avoid stale pointer */ - cpuc->lbr_context = NULL; - } } void intel_pmu_lbr_enable_all(bool pmi) diff --git a/arch/x86/events/intel/rapl.c b/arch/x86/events/intel/rapl.c index 28865938aadf..62bebcca64cc 100644 --- a/arch/x86/events/intel/rapl.c +++ b/arch/x86/events/intel/rapl.c @@ -357,6 +357,8 @@ static int rapl_pmu_event_init(struct perf_event *event) if (event->cpu < 0) return -EINVAL; + event->event_caps |= PERF_EV_CAP_READ_ACTIVE_PKG; + /* * check event is known (determines counter) */ diff --git a/arch/x86/events/intel/uncore.c b/arch/x86/events/intel/uncore.c index 463dc7a5a6c3..7b3cc8be3282 100644 --- a/arch/x86/events/intel/uncore.c +++ b/arch/x86/events/intel/uncore.c @@ -664,6 +664,8 @@ static int uncore_pmu_event_init(struct perf_event *event) event->cpu = box->cpu; event->pmu_private = box; + event->event_caps |= PERF_EV_CAP_READ_ACTIVE_PKG; + event->hw.idx = -1; event->hw.last_tag = ~0ULL; event->hw.extra_reg.idx = EXTRA_REG_NONE; diff --git a/arch/x86/events/intel/uncore_snb.c b/arch/x86/events/intel/uncore_snb.c index 9d35ec0cb8fc..5f845eef9a4d 100644 --- a/arch/x86/events/intel/uncore_snb.c +++ b/arch/x86/events/intel/uncore_snb.c @@ -388,6 +388,8 @@ static int snb_uncore_imc_event_init(struct perf_event *event) event->cpu = box->cpu; event->pmu_private = box; + event->event_caps |= PERF_EV_CAP_READ_ACTIVE_PKG; + event->hw.idx = -1; event->hw.last_tag = ~0ULL; event->hw.extra_reg.idx = EXTRA_REG_NONE; diff --git a/arch/x86/events/perf_event.h b/arch/x86/events/perf_event.h index 8c4a47706296..5874d8de1f8d 100644 --- a/arch/x86/events/perf_event.h +++ b/arch/x86/events/perf_event.h @@ -194,12 +194,13 @@ struct cpu_hw_events { */ struct debug_store *ds; u64 pebs_enabled; + int n_pebs; + int n_large_pebs; /* * Intel LBR bits */ int lbr_users; - void *lbr_context; struct perf_branch_stack lbr_stack; struct perf_branch_entry lbr_entries[MAX_LBR_ENTRIES]; struct er_account *lbr_sel; @@ -508,6 +509,8 @@ struct x86_pmu { void (*enable_all)(int added); void (*enable)(struct perf_event *); void (*disable)(struct perf_event *); + void (*add)(struct perf_event *); + void (*del)(struct perf_event *); int (*hw_config)(struct perf_event *event); int (*schedule_events)(struct cpu_hw_events *cpuc, int n, int *assign); unsigned eventsel; @@ -888,6 +891,10 @@ extern struct event_constraint intel_skl_pebs_event_constraints[]; struct event_constraint *intel_pebs_constraints(struct perf_event *event); +void intel_pmu_pebs_add(struct perf_event *event); + +void intel_pmu_pebs_del(struct perf_event *event); + void intel_pmu_pebs_enable(struct perf_event *event); void intel_pmu_pebs_disable(struct perf_event *event); @@ -906,9 +913,9 @@ u64 lbr_from_signext_quirk_wr(u64 val); void intel_pmu_lbr_reset(void); -void intel_pmu_lbr_enable(struct perf_event *event); +void intel_pmu_lbr_add(struct perf_event *event); -void intel_pmu_lbr_disable(struct perf_event *event); +void intel_pmu_lbr_del(struct perf_event *event); void intel_pmu_lbr_enable_all(bool pmi); diff --git a/include/linux/bitmap.h b/include/linux/bitmap.h index 598bc999f4c2..3b77588a9360 100644 --- a/include/linux/bitmap.h +++ b/include/linux/bitmap.h @@ -339,6 +339,24 @@ static inline int bitmap_parse(const char *buf, unsigned int buflen, return __bitmap_parse(buf, buflen, 0, maskp, nmaskbits); } +/* + * bitmap_from_u64 - Check and swap words within u64. + * @mask: source bitmap + * @dst: destination bitmap + * + * In 32-bit Big Endian kernel, when using (u32 *)(&val)[*] + * to read u64 mask, we will get the wrong word. + * That is "(u32 *)(&val)[0]" gets the upper 32 bits, + * but we expect the lower 32-bits of u64. + */ +static inline void bitmap_from_u64(unsigned long *dst, u64 mask) +{ + dst[0] = mask & ULONG_MAX; + + if (sizeof(mask) > sizeof(unsigned long)) + dst[1] = mask >> 32; +} + #endif /* __ASSEMBLY__ */ #endif /* __LINUX_BITMAP_H */ diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 2b6b43cc0dd5..5c5362584aba 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -510,9 +510,15 @@ typedef void (*perf_overflow_handler_t)(struct perf_event *, struct perf_sample_data *, struct pt_regs *regs); -enum perf_group_flag { - PERF_GROUP_SOFTWARE = 0x1, -}; +/* + * Event capabilities. For event_caps and groups caps. + * + * PERF_EV_CAP_SOFTWARE: Is a software event. + * PERF_EV_CAP_READ_ACTIVE_PKG: A CPU event (or cgroup event) that can be read + * from any CPU in the package where it is active. + */ +#define PERF_EV_CAP_SOFTWARE BIT(0) +#define PERF_EV_CAP_READ_ACTIVE_PKG BIT(1) #define SWEVENT_HLIST_BITS 8 #define SWEVENT_HLIST_SIZE (1 << SWEVENT_HLIST_BITS) @@ -568,7 +574,12 @@ struct perf_event { struct hlist_node hlist_entry; struct list_head active_entry; int nr_siblings; - int group_flags; + + /* Not serialized. Only written during event initialization. */ + int event_caps; + /* The cumulative AND of all event_caps for events in this group. */ + int group_caps; + struct perf_event *group_leader; struct pmu *pmu; void *pmu_private; @@ -774,6 +785,9 @@ struct perf_cpu_context { #ifdef CONFIG_CGROUP_PERF struct perf_cgroup *cgrp; #endif + + struct list_head sched_cb_entry; + int sched_cb_usage; }; struct perf_output_handle { @@ -985,7 +999,7 @@ static inline bool is_sampling_event(struct perf_event *event) */ static inline int is_software_event(struct perf_event *event) { - return event->pmu->task_ctx_nr == perf_sw_context; + return event->event_caps & PERF_EV_CAP_SOFTWARE; } extern struct static_key perf_swevent_enabled[PERF_COUNT_SW_MAX]; diff --git a/kernel/events/core.c b/kernel/events/core.c index 07ac8596a728..dff00c787867 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -1475,8 +1475,7 @@ list_add_event(struct perf_event *event, struct perf_event_context *ctx) if (event->group_leader == event) { struct list_head *list; - if (is_software_event(event)) - event->group_flags |= PERF_GROUP_SOFTWARE; + event->group_caps = event->event_caps; list = ctx_group_list(event, ctx); list_add_tail(&event->group_entry, list); @@ -1630,9 +1629,7 @@ static void perf_group_attach(struct perf_event *event) WARN_ON_ONCE(group_leader->ctx != event->ctx); - if (group_leader->group_flags & PERF_GROUP_SOFTWARE && - !is_software_event(event)) - group_leader->group_flags &= ~PERF_GROUP_SOFTWARE; + group_leader->group_caps &= event->event_caps; list_add_tail(&event->group_entry, &group_leader->sibling_list); group_leader->nr_siblings++; @@ -1723,7 +1720,7 @@ static void perf_group_detach(struct perf_event *event) sibling->group_leader = sibling; /* Inherit group flags from the previous leader */ - sibling->group_flags = event->group_flags; + sibling->group_caps = event->group_caps; WARN_ON_ONCE(sibling->ctx != event->ctx); } @@ -1832,6 +1829,8 @@ group_sched_out(struct perf_event *group_event, struct perf_event *event; int state = group_event->state; + perf_pmu_disable(ctx->pmu); + event_sched_out(group_event, cpuctx, ctx); /* @@ -1840,6 +1839,8 @@ group_sched_out(struct perf_event *group_event, list_for_each_entry(event, &group_event->sibling_list, group_entry) event_sched_out(event, cpuctx, ctx); + perf_pmu_enable(ctx->pmu); + if (state == PERF_EVENT_STATE_ACTIVE && group_event->attr.exclusive) cpuctx->exclusive = 0; } @@ -2145,7 +2146,7 @@ static int group_can_go_on(struct perf_event *event, /* * Groups consisting entirely of software events can always go on. */ - if (event->group_flags & PERF_GROUP_SOFTWARE) + if (event->group_caps & PERF_EV_CAP_SOFTWARE) return 1; /* * If an exclusive group is already on, no other hardware @@ -2837,19 +2838,36 @@ unlock: } } +static DEFINE_PER_CPU(struct list_head, sched_cb_list); + void perf_sched_cb_dec(struct pmu *pmu) { + struct perf_cpu_context *cpuctx = this_cpu_ptr(pmu->pmu_cpu_context); + this_cpu_dec(perf_sched_cb_usages); + + if (!--cpuctx->sched_cb_usage) + list_del(&cpuctx->sched_cb_entry); } + void perf_sched_cb_inc(struct pmu *pmu) { + struct perf_cpu_context *cpuctx = this_cpu_ptr(pmu->pmu_cpu_context); + + if (!cpuctx->sched_cb_usage++) + list_add(&cpuctx->sched_cb_entry, this_cpu_ptr(&sched_cb_list)); + this_cpu_inc(perf_sched_cb_usages); } /* * This function provides the context switch callback to the lower code * layer. It is invoked ONLY when the context switch callback is enabled. + * + * This callback is relevant even to per-cpu events; for example multi event + * PEBS requires this to provide PID/TID information. This requires we flush + * all queued PEBS records before we context switch to a new task. */ static void perf_pmu_sched_task(struct task_struct *prev, struct task_struct *next, @@ -2857,34 +2875,24 @@ static void perf_pmu_sched_task(struct task_struct *prev, { struct perf_cpu_context *cpuctx; struct pmu *pmu; - unsigned long flags; if (prev == next) return; - local_irq_save(flags); - - rcu_read_lock(); - - list_for_each_entry_rcu(pmu, &pmus, entry) { - if (pmu->sched_task) { - cpuctx = this_cpu_ptr(pmu->pmu_cpu_context); - - perf_ctx_lock(cpuctx, cpuctx->task_ctx); + list_for_each_entry(cpuctx, this_cpu_ptr(&sched_cb_list), sched_cb_entry) { + pmu = cpuctx->unique_pmu; /* software PMUs will not have sched_task */ - perf_pmu_disable(pmu); + if (WARN_ON_ONCE(!pmu->sched_task)) + continue; - pmu->sched_task(cpuctx->task_ctx, sched_in); + perf_ctx_lock(cpuctx, cpuctx->task_ctx); + perf_pmu_disable(pmu); - perf_pmu_enable(pmu); + pmu->sched_task(cpuctx->task_ctx, sched_in); - perf_ctx_unlock(cpuctx, cpuctx->task_ctx); - } + perf_pmu_enable(pmu); + perf_ctx_unlock(cpuctx, cpuctx->task_ctx); } - - rcu_read_unlock(); - - local_irq_restore(flags); } static void perf_event_switch(struct task_struct *task, @@ -3416,6 +3424,22 @@ struct perf_read_data { int ret; }; +static int find_cpu_to_read(struct perf_event *event, int local_cpu) +{ + int event_cpu = event->oncpu; + u16 local_pkg, event_pkg; + + if (event->group_caps & PERF_EV_CAP_READ_ACTIVE_PKG) { + event_pkg = topology_physical_package_id(event_cpu); + local_pkg = topology_physical_package_id(local_cpu); + + if (event_pkg == local_pkg) + return local_cpu; + } + + return event_cpu; +} + /* * Cross CPU call to read the hardware event */ @@ -3537,7 +3561,7 @@ u64 perf_event_read_local(struct perf_event *event) static int perf_event_read(struct perf_event *event, bool group) { - int ret = 0; + int ret = 0, cpu_to_read, local_cpu; /* * If event is enabled and currently active on a CPU, update the @@ -3549,6 +3573,11 @@ static int perf_event_read(struct perf_event *event, bool group) .group = group, .ret = 0, }; + + local_cpu = get_cpu(); + cpu_to_read = find_cpu_to_read(event, local_cpu); + put_cpu(); + /* * Purposely ignore the smp_call_function_single() return * value. @@ -3559,7 +3588,7 @@ static int perf_event_read(struct perf_event *event, bool group) * Therefore, either way, we'll have an up-to-date event count * after this. */ - (void)smp_call_function_single(event->oncpu, __perf_event_read, &data, 1); + (void)smp_call_function_single(cpu_to_read, __perf_event_read, &data, 1); ret = data.ret; } else if (event->state == PERF_EVENT_STATE_INACTIVE) { struct perf_event_context *ctx = event->ctx; @@ -5337,9 +5366,10 @@ perf_output_sample_regs(struct perf_output_handle *handle, struct pt_regs *regs, u64 mask) { int bit; + DECLARE_BITMAP(_mask, 64); - for_each_set_bit(bit, (const unsigned long *) &mask, - sizeof(mask) * BITS_PER_BYTE) { + bitmap_from_u64(_mask, mask); + for_each_set_bit(bit, _mask, sizeof(mask) * BITS_PER_BYTE) { u64 val; val = perf_reg_value(regs, bit); @@ -9486,6 +9516,9 @@ SYSCALL_DEFINE5(perf_event_open, goto err_alloc; } + if (pmu->task_ctx_nr == perf_sw_context) + event->event_caps |= PERF_EV_CAP_SOFTWARE; + if (group_leader && (is_software_event(event) != is_software_event(group_leader))) { if (is_software_event(event)) { @@ -9499,7 +9532,7 @@ SYSCALL_DEFINE5(perf_event_open, */ pmu = group_leader->pmu; } else if (is_software_event(group_leader) && - (group_leader->group_flags & PERF_GROUP_SOFTWARE)) { + (group_leader->group_caps & PERF_EV_CAP_SOFTWARE)) { /* * In case the group is a pure software group, and we * try to add a hardware event, move the whole group to @@ -10434,6 +10467,8 @@ static void __init perf_event_init_all_cpus(void) INIT_LIST_HEAD(&per_cpu(pmu_sb_events.list, cpu)); raw_spin_lock_init(&per_cpu(pmu_sb_events.lock, cpu)); + + INIT_LIST_HEAD(&per_cpu(sched_cb_list, cpu)); } } diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 8c50276b60d1..d4129bb05e5d 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -150,7 +150,7 @@ static loff_t vaddr_to_offset(struct vm_area_struct *vma, unsigned long vaddr) * Returns 0 on success, -EFAULT on failure. */ static int __replace_page(struct vm_area_struct *vma, unsigned long addr, - struct page *page, struct page *kpage) + struct page *old_page, struct page *new_page) { struct mm_struct *mm = vma->vm_mm; spinlock_t *ptl; @@ -161,49 +161,49 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr, const unsigned long mmun_end = addr + PAGE_SIZE; struct mem_cgroup *memcg; - err = mem_cgroup_try_charge(kpage, vma->vm_mm, GFP_KERNEL, &memcg, + err = mem_cgroup_try_charge(new_page, vma->vm_mm, GFP_KERNEL, &memcg, false); if (err) return err; /* For try_to_free_swap() and munlock_vma_page() below */ - lock_page(page); + lock_page(old_page); mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end); err = -EAGAIN; - ptep = page_check_address(page, mm, addr, &ptl, 0); + ptep = page_check_address(old_page, mm, addr, &ptl, 0); if (!ptep) { - mem_cgroup_cancel_charge(kpage, memcg, false); + mem_cgroup_cancel_charge(new_page, memcg, false); goto unlock; } - get_page(kpage); - page_add_new_anon_rmap(kpage, vma, addr, false); - mem_cgroup_commit_charge(kpage, memcg, false, false); - lru_cache_add_active_or_unevictable(kpage, vma); + get_page(new_page); + page_add_new_anon_rmap(new_page, vma, addr, false); + mem_cgroup_commit_charge(new_page, memcg, false, false); + lru_cache_add_active_or_unevictable(new_page, vma); - if (!PageAnon(page)) { - dec_mm_counter(mm, mm_counter_file(page)); + if (!PageAnon(old_page)) { + dec_mm_counter(mm, mm_counter_file(old_page)); inc_mm_counter(mm, MM_ANONPAGES); } flush_cache_page(vma, addr, pte_pfn(*ptep)); ptep_clear_flush_notify(vma, addr, ptep); - set_pte_at_notify(mm, addr, ptep, mk_pte(kpage, vma->vm_page_prot)); + set_pte_at_notify(mm, addr, ptep, mk_pte(new_page, vma->vm_page_prot)); - page_remove_rmap(page, false); - if (!page_mapped(page)) - try_to_free_swap(page); + page_remove_rmap(old_page, false); + if (!page_mapped(old_page)) + try_to_free_swap(old_page); pte_unmap_unlock(ptep, ptl); if (vma->vm_flags & VM_LOCKED) - munlock_vma_page(page); - put_page(page); + munlock_vma_page(old_page); + put_page(old_page); err = 0; unlock: mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end); - unlock_page(page); + unlock_page(old_page); return err; } diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index dade4c9559cc..1e2ce3b52e51 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -4123,6 +4123,30 @@ static const char readme_msg[] = "\t\t\t traces\n" #endif #endif /* CONFIG_STACK_TRACER */ +#ifdef CONFIG_KPROBE_EVENT + " kprobe_events\t\t- Add/remove/show the kernel dynamic events\n" + "\t\t\t Write into this file to define/undefine new trace events.\n" +#endif +#ifdef CONFIG_UPROBE_EVENT + " uprobe_events\t\t- Add/remove/show the userspace dynamic events\n" + "\t\t\t Write into this file to define/undefine new trace events.\n" +#endif +#if defined(CONFIG_KPROBE_EVENT) || defined(CONFIG_UPROBE_EVENT) + "\t accepts: event-definitions (one definition per line)\n" + "\t Format: p|r[:[<group>/]<event>] <place> [<args>]\n" + "\t -:[<group>/]<event>\n" +#ifdef CONFIG_KPROBE_EVENT + "\t place: [<module>:]<symbol>[+<offset>]|<memaddr>\n" +#endif +#ifdef CONFIG_UPROBE_EVENT + "\t place: <path>:<offset>\n" +#endif + "\t args: <name>=fetcharg[:type]\n" + "\t fetcharg: %<register>, @<address>, @<symbol>[+|-<offset>],\n" + "\t $stack<index>, $stack, $retval, $comm\n" + "\t type: s8/16/32/64, u8/16/32/64, x8/16/32/64, string,\n" + "\t b<bit-width>@<bit-offset>/<container-size>\n" +#endif " events/\t\t- Directory containing all trace event subsystems:\n" " enable\t\t- Write 0/1 to enable/disable tracing of all events\n" " events/<system>/\t- Directory containing all trace events for <system>:\n" diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 9aedb0b06683..eb6c9f1d3a93 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -253,6 +253,10 @@ static const struct fetch_type kprobes_fetch_type_table[] = { ASSIGN_FETCH_TYPE(s16, u16, 1), ASSIGN_FETCH_TYPE(s32, u32, 1), ASSIGN_FETCH_TYPE(s64, u64, 1), + ASSIGN_FETCH_TYPE_ALIAS(x8, u8, u8, 0), + ASSIGN_FETCH_TYPE_ALIAS(x16, u16, u16, 0), + ASSIGN_FETCH_TYPE_ALIAS(x32, u32, u32, 0), + ASSIGN_FETCH_TYPE_ALIAS(x64, u64, u64, 0), ASSIGN_FETCH_TYPE_END }; diff --git a/kernel/trace/trace_probe.c b/kernel/trace/trace_probe.c index 74e80a582c28..8c0553d9afd3 100644 --- a/kernel/trace/trace_probe.c +++ b/kernel/trace/trace_probe.c @@ -36,24 +36,28 @@ const char *reserved_field_names[] = { }; /* Printing in basic type function template */ -#define DEFINE_BASIC_PRINT_TYPE_FUNC(type, fmt) \ -int PRINT_TYPE_FUNC_NAME(type)(struct trace_seq *s, const char *name, \ +#define DEFINE_BASIC_PRINT_TYPE_FUNC(tname, type, fmt) \ +int PRINT_TYPE_FUNC_NAME(tname)(struct trace_seq *s, const char *name, \ void *data, void *ent) \ { \ trace_seq_printf(s, " %s=" fmt, name, *(type *)data); \ return !trace_seq_has_overflowed(s); \ } \ -const char PRINT_TYPE_FMT_NAME(type)[] = fmt; \ -NOKPROBE_SYMBOL(PRINT_TYPE_FUNC_NAME(type)); - -DEFINE_BASIC_PRINT_TYPE_FUNC(u8 , "0x%x") -DEFINE_BASIC_PRINT_TYPE_FUNC(u16, "0x%x") -DEFINE_BASIC_PRINT_TYPE_FUNC(u32, "0x%x") -DEFINE_BASIC_PRINT_TYPE_FUNC(u64, "0x%Lx") -DEFINE_BASIC_PRINT_TYPE_FUNC(s8, "%d") -DEFINE_BASIC_PRINT_TYPE_FUNC(s16, "%d") -DEFINE_BASIC_PRINT_TYPE_FUNC(s32, "%d") -DEFINE_BASIC_PRINT_TYPE_FUNC(s64, "%Ld") +const char PRINT_TYPE_FMT_NAME(tname)[] = fmt; \ +NOKPROBE_SYMBOL(PRINT_TYPE_FUNC_NAME(tname)); + +DEFINE_BASIC_PRINT_TYPE_FUNC(u8, u8, "%u") +DEFINE_BASIC_PRINT_TYPE_FUNC(u16, u16, "%u") +DEFINE_BASIC_PRINT_TYPE_FUNC(u32, u32, "%u") +DEFINE_BASIC_PRINT_TYPE_FUNC(u64, u64, "%Lu") +DEFINE_BASIC_PRINT_TYPE_FUNC(s8, s8, "%d") +DEFINE_BASIC_PRINT_TYPE_FUNC(s16, s16, "%d") +DEFINE_BASIC_PRINT_TYPE_FUNC(s32, s32, "%d") +DEFINE_BASIC_PRINT_TYPE_FUNC(s64, s64, "%Ld") +DEFINE_BASIC_PRINT_TYPE_FUNC(x8, u8, "0x%x") +DEFINE_BASIC_PRINT_TYPE_FUNC(x16, u16, "0x%x") +DEFINE_BASIC_PRINT_TYPE_FUNC(x32, u32, "0x%x") +DEFINE_BASIC_PRINT_TYPE_FUNC(x64, u64, "0x%Lx") /* Print type function for string type */ int PRINT_TYPE_FUNC_NAME(string)(struct trace_seq *s, const char *name, diff --git a/kernel/trace/trace_probe.h b/kernel/trace/trace_probe.h index 45400ca5ded1..0c0ae54d44c6 100644 --- a/kernel/trace/trace_probe.h +++ b/kernel/trace/trace_probe.h @@ -149,6 +149,11 @@ DECLARE_BASIC_PRINT_TYPE_FUNC(s8); DECLARE_BASIC_PRINT_TYPE_FUNC(s16); DECLARE_BASIC_PRINT_TYPE_FUNC(s32); DECLARE_BASIC_PRINT_TYPE_FUNC(s64); +DECLARE_BASIC_PRINT_TYPE_FUNC(x8); +DECLARE_BASIC_PRINT_TYPE_FUNC(x16); +DECLARE_BASIC_PRINT_TYPE_FUNC(x32); +DECLARE_BASIC_PRINT_TYPE_FUNC(x64); + DECLARE_BASIC_PRINT_TYPE_FUNC(string); #define FETCH_FUNC_NAME(method, type) fetch_##method##_##type @@ -203,7 +208,7 @@ DEFINE_FETCH_##method(u32) \ DEFINE_FETCH_##method(u64) /* Default (unsigned long) fetch type */ -#define __DEFAULT_FETCH_TYPE(t) u##t +#define __DEFAULT_FETCH_TYPE(t) x##t #define _DEFAULT_FETCH_TYPE(t) __DEFAULT_FETCH_TYPE(t) #define DEFAULT_FETCH_TYPE _DEFAULT_FETCH_TYPE(BITS_PER_LONG) #define DEFAULT_FETCH_TYPE_STR __stringify(DEFAULT_FETCH_TYPE) @@ -234,6 +239,10 @@ ASSIGN_FETCH_FUNC(file_offset, ftype), \ #define ASSIGN_FETCH_TYPE(ptype, ftype, sign) \ __ASSIGN_FETCH_TYPE(#ptype, ptype, ftype, sizeof(ftype), sign, #ptype) +/* If ptype is an alias of atype, use this macro (show atype in format) */ +#define ASSIGN_FETCH_TYPE_ALIAS(ptype, atype, ftype, sign) \ + __ASSIGN_FETCH_TYPE(#ptype, ptype, ftype, sizeof(ftype), sign, #atype) + #define ASSIGN_FETCH_TYPE_END {} #define FETCH_TYPE_STRING 0 diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c index c53485441c88..7a687320f867 100644 --- a/kernel/trace/trace_uprobe.c +++ b/kernel/trace/trace_uprobe.c @@ -211,6 +211,10 @@ static const struct fetch_type uprobes_fetch_type_table[] = { ASSIGN_FETCH_TYPE(s16, u16, 1), ASSIGN_FETCH_TYPE(s32, u32, 1), ASSIGN_FETCH_TYPE(s64, u64, 1), + ASSIGN_FETCH_TYPE_ALIAS(x8, u8, u8, 0), + ASSIGN_FETCH_TYPE_ALIAS(x16, u16, u16, 0), + ASSIGN_FETCH_TYPE_ALIAS(x32, u32, u32, 0), + ASSIGN_FETCH_TYPE_ALIAS(x64, u64, u64, 0), ASSIGN_FETCH_TYPE_END }; diff --git a/tools/include/linux/coresight-pmu.h b/tools/include/linux/coresight-pmu.h new file mode 100644 index 000000000000..7d410260661b --- /dev/null +++ b/tools/include/linux/coresight-pmu.h @@ -0,0 +1,39 @@ +/* + * Copyright(C) 2015 Linaro Limited. All rights reserved. + * Author: Mathieu Poirier <[email protected]> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published by + * the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef _LINUX_CORESIGHT_PMU_H +#define _LINUX_CORESIGHT_PMU_H + +#define CORESIGHT_ETM_PMU_NAME "cs_etm" +#define CORESIGHT_ETM_PMU_SEED 0x10 + +/* ETMv3.5/PTM's ETMCR config bit */ +#define ETM_OPT_CYCACC 12 +#define ETM_OPT_TS 28 + +static inline int coresight_get_trace_id(int cpu) +{ + /* + * A trace ID of value 0 is invalid, so let's start at some + * random value that fits in 7 bits and go from there. Since + * the common convention is to have data trace IDs be I(N) + 1, + * set instruction trace IDs as a function of the CPU number. + */ + return (CORESIGHT_ETM_PMU_SEED + (cpu * 2)); +} + +#endif diff --git a/tools/include/linux/time64.h b/tools/include/linux/time64.h new file mode 100644 index 000000000000..df9265483d65 --- /dev/null +++ b/tools/include/linux/time64.h @@ -0,0 +1,12 @@ +#ifndef _TOOLS_LINUX_TIME64_H +#define _TOOLS_LINUX_TIME64_H + +#define MSEC_PER_SEC 1000L +#define USEC_PER_MSEC 1000L +#define NSEC_PER_USEC 1000L +#define NSEC_PER_MSEC 1000000L +#define USEC_PER_SEC 1000000L +#define NSEC_PER_SEC 1000000000L +#define FSEC_PER_SEC 1000000000000000LL + +#endif /* _LINUX_TIME64_H */ diff --git a/tools/perf/Documentation/perf-config.txt b/tools/perf/Documentation/perf-config.txt index 15949e2a7805..68c89190e79c 100644 --- a/tools/perf/Documentation/perf-config.txt +++ b/tools/perf/Documentation/perf-config.txt @@ -382,6 +382,10 @@ call-graph.*:: histogram entry. Default is 0 which means no limitation. report.*:: + report.sort_order:: + Allows changing the default sort order from "comm,dso,symbol" to + some other default, for instance "sym,dso" may be more fitting for + kernel developers. report.percent-limit:: This one is mostly the same as call-graph.threshold but works for histogram entries. Entries having an overhead lower than this diff --git a/tools/perf/Documentation/perf-probe.txt b/tools/perf/Documentation/perf-probe.txt index b303bcdd8ed1..f37d123d5dac 100644 --- a/tools/perf/Documentation/perf-probe.txt +++ b/tools/perf/Documentation/perf-probe.txt @@ -176,13 +176,12 @@ Each probe argument follows below syntax. 'NAME' specifies the name of this argument (optional). You can use the name of local variable, local data structure member (e.g. var->field, var.field2), local array with fixed index (e.g. array[1], var->array[0], var->pointer[2]), or kprobe-tracer argument format (e.g. $retval, %ax, etc). Note that the name of this argument will be set as the last member name if you specify a local data structure member (e.g. field2 for 'var->field1.field2'.) '$vars' and '$params' special arguments are also available for NAME, '$vars' is expanded to the local variables (including function parameters) which can access at given probe point. '$params' is expanded to only the function parameters. -'TYPE' casts the type of this argument (optional). If omitted, perf probe automatically set the type based on debuginfo. Currently, basic types (u8/u16/u32/u64/s8/s16/s32/s64), signedness casting (u/s), "string" and bitfield are supported. (see TYPES for detail) - +'TYPE' casts the type of this argument (optional). If omitted, perf probe automatically set the type based on debuginfo (*). Currently, basic types (u8/u16/u32/u64/s8/s16/s32/s64), hexadecimal integers (x/x8/x16/x32/x64), signedness casting (u/s), "string" and bitfield are supported. (see TYPES for detail) On x86 systems %REG is always the short form of the register: for example %AX. %RAX or %EAX is not valid. TYPES ----- -Basic types (u8/u16/u32/u64/s8/s16/s32/s64) are integer types. Prefix 's' and 'u' means those types are signed and unsigned respectively. Traced arguments are shown in decimal (signed) or hex (unsigned). You can also use 's' or 'u' to specify only signedness and leave its size auto-detected by perf probe. +Basic types (u8/u16/u32/u64/s8/s16/s32/s64) and hexadecimal integers (x8/x16/x32/x64) are integer types. Prefix 's' and 'u' means those types are signed and unsigned respectively, and 'x' means that is shown in hexadecimal format. Traced arguments are shown in decimal (sNN/uNN) or hex (xNN). You can also use 's' or 'u' to specify only signedness and leave its size auto-detected by perf probe. Moreover, you can use 'x' to explicitly specify to be shown in hexadecimal (the size is also auto-detected). String type is a special type, which fetches a "null-terminated" string from kernel space. This means it will fail and store NULL if the string container has been paged out. You can specify 'string' type only for the local variable or structure member which is an array of or a pointer to 'char' or 'unsigned char' type. Bitfield is another special type, which takes 3 parameters, bit-width, bit-offset, and container-size (usually 32). The syntax is; diff --git a/tools/perf/MANIFEST b/tools/perf/MANIFEST index ad2534df4ba6..ff200c6cb790 100644 --- a/tools/perf/MANIFEST +++ b/tools/perf/MANIFEST @@ -60,6 +60,7 @@ tools/include/asm-generic/bitops.h tools/include/linux/atomic.h tools/include/linux/bitops.h tools/include/linux/compiler.h +tools/include/linux/coresight-pmu.h tools/include/linux/filter.h tools/include/linux/hash.h tools/include/linux/kernel.h @@ -77,4 +78,5 @@ tools/include/linux/stringify.h tools/include/linux/types.h tools/include/linux/err.h tools/include/linux/bitmap.h +tools/include/linux/time64.h tools/arch/*/include/uapi/asm/perf_regs.h diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf index 2d9087501633..828cfd766e31 100644 --- a/tools/perf/Makefile.perf +++ b/tools/perf/Makefile.perf @@ -165,7 +165,7 @@ SUBCMD_DIR = $(srctree)/tools/lib/subcmd/ # non-config cases config := 1 -NON_CONFIG_TARGETS := clean TAGS tags cscope help +NON_CONFIG_TARGETS := clean TAGS tags cscope help install-doc ifdef MAKECMDGOALS ifeq ($(filter-out $(NON_CONFIG_TARGETS),$(MAKECMDGOALS)),) @@ -429,6 +429,9 @@ $(PERF_IN): prepare FORCE @(test -f ../../include/asm-generic/bitops/fls64.h && ( \ (diff -B ../include/asm-generic/bitops/fls64.h ../../include/asm-generic/bitops/fls64.h >/dev/null) \ || echo "Warning: tools/include/asm-generic/bitops/fls64.h differs from kernel" >&2 )) || true + @(test -f ../../include/linux/coresight-pmu.h && ( \ + (diff -B ../include/linux/coresight-pmu.h ../../include/linux/coresight-pmu.h >/dev/null) \ + || echo "Warning: tools/include/linux/coresight-pmu.h differs from kernel" >&2 )) || true $(Q)$(MAKE) $(build)=perf $(OUTPUT)perf: $(PERFLIBS) $(PERF_IN) $(LIBTRACEEVENT_DYNAMIC_LIST) diff --git a/tools/perf/bench/futex-requeue.c b/tools/perf/bench/futex-requeue.c index f96e22ed9f87..2b9705a8734c 100644 --- a/tools/perf/bench/futex-requeue.c +++ b/tools/perf/bench/futex-requeue.c @@ -16,6 +16,7 @@ #include <subcmd/parse-options.h> #include <linux/compiler.h> #include <linux/kernel.h> +#include <linux/time64.h> #include <errno.h> #include "bench.h" #include "futex.h" @@ -62,7 +63,7 @@ static void print_summary(void) printf("Requeued %d of %d threads in %.4f ms (+-%.2f%%)\n", requeued_avg, nthreads, - requeuetime_avg/1e3, + requeuetime_avg / USEC_PER_MSEC, rel_stddev_stats(requeuetime_stddev, requeuetime_avg)); } @@ -184,7 +185,7 @@ int bench_futex_requeue(int argc, const char **argv, if (!silent) { printf("[Run %d]: Requeued %d of %d threads in %.4f ms\n", - j + 1, nrequeued, nthreads, runtime.tv_usec/1e3); + j + 1, nrequeued, nthreads, runtime.tv_usec / (double)USEC_PER_MSEC); } /* everybody should be blocked on futex2, wake'em up */ diff --git a/tools/perf/bench/futex-wake-parallel.c b/tools/perf/bench/futex-wake-parallel.c index 4a2ecd7438ca..2c8fa67ad537 100644 --- a/tools/perf/bench/futex-wake-parallel.c +++ b/tools/perf/bench/futex-wake-parallel.c @@ -15,6 +15,7 @@ #include <subcmd/parse-options.h> #include <linux/compiler.h> #include <linux/kernel.h> +#include <linux/time64.h> #include <errno.h> #include "bench.h" #include "futex.h" @@ -156,7 +157,7 @@ static void print_run(struct thread_data *waking_worker, unsigned int run_num) printf("[Run %d]: Avg per-thread latency (waking %d/%d threads) " "in %.4f ms (+-%.2f%%)\n", run_num + 1, wakeup_avg, - nblocked_threads, waketime_avg/1e3, + nblocked_threads, waketime_avg / USEC_PER_MSEC, rel_stddev_stats(waketime_stddev, waketime_avg)); } @@ -172,7 +173,7 @@ static void print_summary(void) printf("Avg per-thread latency (waking %d/%d threads) in %.4f ms (+-%.2f%%)\n", wakeup_avg, nblocked_threads, - waketime_avg/1e3, + waketime_avg / USEC_PER_MSEC, rel_stddev_stats(waketime_stddev, waketime_avg)); } diff --git a/tools/perf/bench/futex-wake.c b/tools/perf/bench/futex-wake.c index 87d8f4f292d9..e246b1b8388a 100644 --- a/tools/perf/bench/futex-wake.c +++ b/tools/perf/bench/futex-wake.c @@ -16,6 +16,7 @@ #include <subcmd/parse-options.h> #include <linux/compiler.h> #include <linux/kernel.h> +#include <linux/time64.h> #include <errno.h> #include "bench.h" #include "futex.h" @@ -81,7 +82,7 @@ static void print_summary(void) printf("Wokeup %d of %d threads in %.4f ms (+-%.2f%%)\n", wakeup_avg, nthreads, - waketime_avg/1e3, + waketime_avg / USEC_PER_MSEC, rel_stddev_stats(waketime_stddev, waketime_avg)); } @@ -182,7 +183,7 @@ int bench_futex_wake(int argc, const char **argv, if (!silent) { printf("[Run %d]: Wokeup %d of %d threads in %.4f ms\n", - j + 1, nwoken, nthreads, runtime.tv_usec/1e3); + j + 1, nwoken, nthreads, runtime.tv_usec / (double)USEC_PER_MSEC); } for (i = 0; i < nthreads; i++) { diff --git a/tools/perf/bench/mem-functions.c b/tools/perf/bench/mem-functions.c index 2b54d0f2672a..c684910e5a48 100644 --- a/tools/perf/bench/mem-functions.c +++ b/tools/perf/bench/mem-functions.c @@ -21,6 +21,7 @@ #include <string.h> #include <sys/time.h> #include <errno.h> +#include <linux/time64.h> #define K 1024 @@ -89,7 +90,7 @@ static u64 get_cycles(void) static double timeval2double(struct timeval *ts) { - return (double)ts->tv_sec + (double)ts->tv_usec / (double)1000000; + return (double)ts->tv_sec + (double)ts->tv_usec / (double)USEC_PER_SEC; } #define print_bps(x) do { \ diff --git a/tools/perf/bench/numa.c b/tools/perf/bench/numa.c index f7f530081aa9..8efe904e486b 100644 --- a/tools/perf/bench/numa.c +++ b/tools/perf/bench/numa.c @@ -30,6 +30,7 @@ #include <sys/wait.h> #include <sys/prctl.h> #include <sys/types.h> +#include <linux/time64.h> #include <numa.h> #include <numaif.h> @@ -1004,7 +1005,7 @@ static void calc_convergence(double runtime_ns_max, double *convergence) if (strong && process_groups == g->p.nr_proc) { if (!*convergence) { *convergence = runtime_ns_max; - tprintf(" (%6.1fs converged)\n", *convergence/1e9); + tprintf(" (%6.1fs converged)\n", *convergence / NSEC_PER_SEC); if (g->p.measure_convergence) { g->all_converged = true; g->stop_work = true; @@ -1012,7 +1013,7 @@ static void calc_convergence(double runtime_ns_max, double *convergence) } } else { if (*convergence) { - tprintf(" (%6.1fs de-converged)", runtime_ns_max/1e9); + tprintf(" (%6.1fs de-converged)", runtime_ns_max / NSEC_PER_SEC); *convergence = 0; } tprintf("\n"); @@ -1022,7 +1023,7 @@ static void calc_convergence(double runtime_ns_max, double *convergence) static void show_summary(double runtime_ns_max, int l, double *convergence) { tprintf("\r # %5.1f%% [%.1f mins]", - (double)(l+1)/g->p.nr_loops*100.0, runtime_ns_max/1e9 / 60.0); + (double)(l+1)/g->p.nr_loops*100.0, runtime_ns_max / NSEC_PER_SEC / 60.0); calc_convergence(runtime_ns_max, convergence); @@ -1179,8 +1180,8 @@ static void *worker_thread(void *__tdata) if (details >= 3) { timersub(&stop, &start, &diff); - runtime_ns_max = diff.tv_sec * 1000000000; - runtime_ns_max += diff.tv_usec * 1000; + runtime_ns_max = diff.tv_sec * NSEC_PER_SEC; + runtime_ns_max += diff.tv_usec * NSEC_PER_USEC; if (details >= 0) { printf(" #%2d / %2d: %14.2lf nsecs/op [val: %016"PRIx64"]\n", @@ -1192,23 +1193,23 @@ static void *worker_thread(void *__tdata) continue; timersub(&stop, &start0, &diff); - runtime_ns_max = diff.tv_sec * 1000000000ULL; - runtime_ns_max += diff.tv_usec * 1000ULL; + runtime_ns_max = diff.tv_sec * NSEC_PER_SEC; + runtime_ns_max += diff.tv_usec * NSEC_PER_USEC; show_summary(runtime_ns_max, l, &convergence); } gettimeofday(&stop, NULL); timersub(&stop, &start0, &diff); - td->runtime_ns = diff.tv_sec * 1000000000ULL; - td->runtime_ns += diff.tv_usec * 1000ULL; - td->speed_gbs = bytes_done / (td->runtime_ns / 1e9) / 1e9; + td->runtime_ns = diff.tv_sec * NSEC_PER_SEC; + td->runtime_ns += diff.tv_usec * NSEC_PER_USEC; + td->speed_gbs = bytes_done / (td->runtime_ns / NSEC_PER_SEC) / 1e9; getrusage(RUSAGE_THREAD, &rusage); - td->system_time_ns = rusage.ru_stime.tv_sec * 1000000000ULL; - td->system_time_ns += rusage.ru_stime.tv_usec * 1000ULL; - td->user_time_ns = rusage.ru_utime.tv_sec * 1000000000ULL; - td->user_time_ns += rusage.ru_utime.tv_usec * 1000ULL; + td->system_time_ns = rusage.ru_stime.tv_sec * NSEC_PER_SEC; + td->system_time_ns += rusage.ru_stime.tv_usec * NSEC_PER_USEC; + td->user_time_ns = rusage.ru_utime.tv_sec * NSEC_PER_SEC; + td->user_time_ns += rusage.ru_utime.tv_usec * NSEC_PER_USEC; free_data(thread_data, g->p.bytes_thread); @@ -1469,7 +1470,7 @@ static int __bench_numa(const char *name) } /* Wait for all the threads to start up: */ while (g->nr_tasks_started != g->p.nr_tasks) - usleep(1000); + usleep(USEC_PER_MSEC); BUG_ON(g->nr_tasks_started != g->p.nr_tasks); @@ -1488,9 +1489,9 @@ static int __bench_numa(const char *name) timersub(&stop, &start, &diff); - startup_sec = diff.tv_sec * 1000000000.0; - startup_sec += diff.tv_usec * 1000.0; - startup_sec /= 1e9; + startup_sec = diff.tv_sec * NSEC_PER_SEC; + startup_sec += diff.tv_usec * NSEC_PER_USEC; + startup_sec /= NSEC_PER_SEC; tprintf(" threads initialized in %.6f seconds.\n", startup_sec); tprintf(" #\n"); @@ -1529,14 +1530,14 @@ static int __bench_numa(const char *name) tprintf("\n ###\n"); tprintf("\n"); - runtime_sec_max = diff.tv_sec * 1000000000.0; - runtime_sec_max += diff.tv_usec * 1000.0; - runtime_sec_max /= 1e9; + runtime_sec_max = diff.tv_sec * NSEC_PER_SEC; + runtime_sec_max += diff.tv_usec * NSEC_PER_USEC; + runtime_sec_max /= NSEC_PER_SEC; - runtime_sec_min = runtime_ns_min/1e9; + runtime_sec_min = runtime_ns_min / NSEC_PER_SEC; bytes = g->bytes_done; - runtime_avg = (double)runtime_ns_sum / g->p.nr_tasks / 1e9; + runtime_avg = (double)runtime_ns_sum / g->p.nr_tasks / NSEC_PER_SEC; if (g->p.measure_convergence) { print_res(name, runtime_sec_max, @@ -1562,7 +1563,7 @@ static int __bench_numa(const char *name) print_res(name, bytes / 1e9, "GB,", "data-total", "GB data processed, total"); - print_res(name, runtime_sec_max * 1e9 / (bytes / g->p.nr_tasks), + print_res(name, runtime_sec_max * NSEC_PER_SEC / (bytes / g->p.nr_tasks), "nsecs,", "runtime/byte/thread","nsecs/byte/thread runtime"); print_res(name, bytes / g->p.nr_tasks / 1e9 / runtime_sec_max, @@ -1581,9 +1582,9 @@ static int __bench_numa(const char *name) snprintf(tname, 32, "process%d:thread%d", p, t); print_res(tname, td->speed_gbs, "GB/sec", "thread-speed", "GB/sec/thread speed"); - print_res(tname, td->system_time_ns / 1e9, + print_res(tname, td->system_time_ns / NSEC_PER_SEC, "secs", "thread-system-time", "system CPU time/thread"); - print_res(tname, td->user_time_ns / 1e9, + print_res(tname, td->user_time_ns / NSEC_PER_SEC, "secs", "thread-user-time", "user CPU time/thread"); } } diff --git a/tools/perf/bench/sched-messaging.c b/tools/perf/bench/sched-messaging.c index bfaf9503de8e..6a111e775210 100644 --- a/tools/perf/bench/sched-messaging.c +++ b/tools/perf/bench/sched-messaging.c @@ -29,6 +29,7 @@ #include <poll.h> #include <limits.h> #include <err.h> +#include <linux/time64.h> #define DATASIZE 100 @@ -312,11 +313,11 @@ int bench_sched_messaging(int argc, const char **argv, thread_mode ? "threads" : "processes"); printf(" %14s: %lu.%03lu [sec]\n", "Total time", diff.tv_sec, - (unsigned long) (diff.tv_usec/1000)); + (unsigned long) (diff.tv_usec / USEC_PER_MSEC)); break; case BENCH_FORMAT_SIMPLE: printf("%lu.%03lu\n", diff.tv_sec, - (unsigned long) (diff.tv_usec/1000)); + (unsigned long) (diff.tv_usec / USEC_PER_MSEC)); break; default: /* reaching here is something disaster */ diff --git a/tools/perf/bench/sched-pipe.c b/tools/perf/bench/sched-pipe.c index 1dc2d13cc272..2243f0150d76 100644 --- a/tools/perf/bench/sched-pipe.c +++ b/tools/perf/bench/sched-pipe.c @@ -25,6 +25,7 @@ #include <sys/time.h> #include <sys/types.h> #include <sys/syscall.h> +#include <linux/time64.h> #include <pthread.h> @@ -153,24 +154,24 @@ int bench_sched_pipe(int argc, const char **argv, const char *prefix __maybe_unu printf("# Executed %d pipe operations between two %s\n\n", loops, threaded ? "threads" : "processes"); - result_usec = diff.tv_sec * 1000000; + result_usec = diff.tv_sec * USEC_PER_SEC; result_usec += diff.tv_usec; printf(" %14s: %lu.%03lu [sec]\n\n", "Total time", diff.tv_sec, - (unsigned long) (diff.tv_usec/1000)); + (unsigned long) (diff.tv_usec / USEC_PER_MSEC)); printf(" %14lf usecs/op\n", (double)result_usec / (double)loops); printf(" %14d ops/sec\n", (int)((double)loops / - ((double)result_usec / (double)1000000))); + ((double)result_usec / (double)USEC_PER_SEC))); break; case BENCH_FORMAT_SIMPLE: printf("%lu.%03lu\n", diff.tv_sec, - (unsigned long) (diff.tv_usec / 1000)); + (unsigned long) (diff.tv_usec / USEC_PER_MSEC)); break; default: diff --git a/tools/perf/builtin-diff.c b/tools/perf/builtin-diff.c index 21ee753211ad..9ff0db4e2d0c 100644 --- a/tools/perf/builtin-diff.c +++ b/tools/perf/builtin-diff.c @@ -1033,7 +1033,9 @@ static int hpp__entry_global(struct perf_hpp_fmt *_fmt, struct perf_hpp *hpp, } static int hpp__header(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp, - struct hists *hists __maybe_unused) + struct hists *hists __maybe_unused, + int line __maybe_unused, + int *span __maybe_unused) { struct diff_hpp_fmt *dfmt = container_of(fmt, struct diff_hpp_fmt, fmt); diff --git a/tools/perf/builtin-kvm.c b/tools/perf/builtin-kvm.c index 5e2127e04f83..08fa88f62a24 100644 --- a/tools/perf/builtin-kvm.c +++ b/tools/perf/builtin-kvm.c @@ -24,6 +24,7 @@ #include <sys/timerfd.h> #endif +#include <linux/time64.h> #include <termios.h> #include <semaphore.h> #include <pthread.h> @@ -362,7 +363,7 @@ static bool handle_end_event(struct perf_kvm_stat *kvm, if (!skip_event(decode)) { pr_info("%" PRIu64 " VM %d, vcpu %d: %s event took %" PRIu64 "usec\n", sample->time, sample->pid, vcpu_record->vcpu_id, - decode, time_diff/1000); + decode, time_diff / NSEC_PER_USEC); } } @@ -608,15 +609,15 @@ static void print_result(struct perf_kvm_stat *kvm) pr_info("%10llu ", (unsigned long long)ecount); pr_info("%8.2f%% ", (double)ecount / kvm->total_count * 100); pr_info("%8.2f%% ", (double)etime / kvm->total_time * 100); - pr_info("%9.2fus ", (double)min / 1e3); - pr_info("%9.2fus ", (double)max / 1e3); - pr_info("%9.2fus ( +-%7.2f%% )", (double)etime / ecount/1e3, + pr_info("%9.2fus ", (double)min / NSEC_PER_USEC); + pr_info("%9.2fus ", (double)max / NSEC_PER_USEC); + pr_info("%9.2fus ( +-%7.2f%% )", (double)etime / ecount / NSEC_PER_USEC, kvm_event_rel_stddev(vcpu, event)); pr_info("\n"); } pr_info("\nTotal Samples:%" PRIu64 ", Total events handled time:%.2fus.\n\n", - kvm->total_count, kvm->total_time / 1e3); + kvm->total_count, kvm->total_time / (double)NSEC_PER_USEC); if (kvm->lost_events) pr_info("\nLost events: %" PRIu64 "\n\n", kvm->lost_events); diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index 6355902fbfc8..03251c7f14ec 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -42,7 +42,7 @@ #include <sched.h> #include <sys/mman.h> #include <asm/bug.h> - +#include <linux/time64.h> struct record { struct perf_tool tool; @@ -96,7 +96,7 @@ backward_rb_find_range(void *buf, int mask, u64 head, u64 *start, u64 *end) *start = head; while (true) { if (evt_head - head >= (unsigned int)size) { - pr_debug("Finshed reading backward ring buffer: rewind\n"); + pr_debug("Finished reading backward ring buffer: rewind\n"); if (evt_head - head > (unsigned int)size) evt_head -= pheader->size; *end = evt_head; @@ -106,7 +106,7 @@ backward_rb_find_range(void *buf, int mask, u64 head, u64 *start, u64 *end) pheader = (struct perf_event_header *)(buf + (evt_head & mask)); if (pheader->size == 0) { - pr_debug("Finshed reading backward ring buffer: get start\n"); + pr_debug("Finished reading backward ring buffer: get start\n"); *end = evt_head; return 0; } @@ -954,7 +954,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv) } if (opts->initial_delay) { - usleep(opts->initial_delay * 1000); + usleep(opts->initial_delay * USEC_PER_MSEC); perf_evlist__enable(rec->evlist); } diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c index 949e5a15c960..b9e046baa5fc 100644 --- a/tools/perf/builtin-report.c +++ b/tools/perf/builtin-report.c @@ -89,6 +89,10 @@ static int report__config(const char *var, const char *value, void *cb) rep->queue_size = perf_config_u64(var, value); return 0; } + if (!strcmp(var, "report.sort_order")) { + default_sort_order = strdup(value); + return 0; + } return 0; } diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c index 0dfe8df2ab9b..f5503ca22e1c 100644 --- a/tools/perf/builtin-sched.c +++ b/tools/perf/builtin-sched.c @@ -26,6 +26,7 @@ #include <pthread.h> #include <math.h> #include <api/fs/fs.h> +#include <linux/time64.h> #define PR_SET_NAME 15 /* Set process name */ #define MAX_CPUS 4096 @@ -199,7 +200,7 @@ static u64 get_nsecs(void) clock_gettime(CLOCK_MONOTONIC, &ts); - return ts.tv_sec * 1000000000ULL + ts.tv_nsec; + return ts.tv_sec * NSEC_PER_SEC + ts.tv_nsec; } static void burn_nsecs(struct perf_sched *sched, u64 nsecs) @@ -223,7 +224,7 @@ static void sleep_nsecs(u64 nsecs) static void calibrate_run_measurement_overhead(struct perf_sched *sched) { - u64 T0, T1, delta, min_delta = 1000000000ULL; + u64 T0, T1, delta, min_delta = NSEC_PER_SEC; int i; for (i = 0; i < 10; i++) { @@ -240,7 +241,7 @@ static void calibrate_run_measurement_overhead(struct perf_sched *sched) static void calibrate_sleep_measurement_overhead(struct perf_sched *sched) { - u64 T0, T1, delta, min_delta = 1000000000ULL; + u64 T0, T1, delta, min_delta = NSEC_PER_SEC; int i; for (i = 0; i < 10; i++) { @@ -452,8 +453,8 @@ static u64 get_cpu_usage_nsec_parent(void) err = getrusage(RUSAGE_SELF, &ru); BUG_ON(err); - sum = ru.ru_utime.tv_sec*1e9 + ru.ru_utime.tv_usec*1e3; - sum += ru.ru_stime.tv_sec*1e9 + ru.ru_stime.tv_usec*1e3; + sum = ru.ru_utime.tv_sec * NSEC_PER_SEC + ru.ru_utime.tv_usec * NSEC_PER_USEC; + sum += ru.ru_stime.tv_sec * NSEC_PER_SEC + ru.ru_stime.tv_usec * NSEC_PER_USEC; return sum; } @@ -667,12 +668,12 @@ static void run_one_test(struct perf_sched *sched) sched->run_avg = delta; sched->run_avg = (sched->run_avg * (sched->replay_repeat - 1) + delta) / sched->replay_repeat; - printf("#%-3ld: %0.3f, ", sched->nr_runs, (double)delta / 1000000.0); + printf("#%-3ld: %0.3f, ", sched->nr_runs, (double)delta / NSEC_PER_MSEC); - printf("ravg: %0.2f, ", (double)sched->run_avg / 1e6); + printf("ravg: %0.2f, ", (double)sched->run_avg / NSEC_PER_MSEC); printf("cpu: %0.2f / %0.2f", - (double)sched->cpu_usage / 1e6, (double)sched->runavg_cpu_usage / 1e6); + (double)sched->cpu_usage / NSEC_PER_MSEC, (double)sched->runavg_cpu_usage / NSEC_PER_MSEC); #if 0 /* @@ -680,8 +681,8 @@ static void run_one_test(struct perf_sched *sched) * accurate than the sched->sum_exec_runtime based statistics: */ printf(" [%0.2f / %0.2f]", - (double)sched->parent_cpu_usage/1e6, - (double)sched->runavg_parent_cpu_usage/1e6); + (double)sched->parent_cpu_usage / NSEC_PER_MSEC, + (double)sched->runavg_parent_cpu_usage / NSEC_PER_MSEC); #endif printf("\n"); @@ -696,13 +697,13 @@ static void test_calibrations(struct perf_sched *sched) u64 T0, T1; T0 = get_nsecs(); - burn_nsecs(sched, 1e6); + burn_nsecs(sched, NSEC_PER_MSEC); T1 = get_nsecs(); printf("the run test took %" PRIu64 " nsecs\n", T1 - T0); T0 = get_nsecs(); - sleep_nsecs(1e6); + sleep_nsecs(NSEC_PER_MSEC); T1 = get_nsecs(); printf("the sleep test took %" PRIu64 " nsecs\n", T1 - T0); @@ -1213,10 +1214,10 @@ static void output_lat_thread(struct perf_sched *sched, struct work_atoms *work_ avg = work_list->total_lat / work_list->nb_atoms; printf("|%11.3f ms |%9" PRIu64 " | avg:%9.3f ms | max:%9.3f ms | max at: %13.6f s\n", - (double)work_list->total_runtime / 1e6, - work_list->nb_atoms, (double)avg / 1e6, - (double)work_list->max_lat / 1e6, - (double)work_list->max_lat_at / 1e9); + (double)work_list->total_runtime / NSEC_PER_MSEC, + work_list->nb_atoms, (double)avg / NSEC_PER_MSEC, + (double)work_list->max_lat / NSEC_PER_MSEC, + (double)work_list->max_lat_at / NSEC_PER_SEC); } static int pid_cmp(struct work_atoms *l, struct work_atoms *r) @@ -1491,7 +1492,7 @@ static int map_switch_event(struct perf_sched *sched, struct perf_evsel *evsel, if (sched->map.cpus && !cpu_map__has(sched->map.cpus, this_cpu)) goto out; - color_fprintf(stdout, color, " %12.6f secs ", (double)timestamp/1e9); + color_fprintf(stdout, color, " %12.6f secs ", (double)timestamp / NSEC_PER_SEC); if (new_shortname) { const char *pid_color = color; @@ -1753,7 +1754,7 @@ static int perf_sched__lat(struct perf_sched *sched) printf(" -----------------------------------------------------------------------------------------------------------------\n"); printf(" TOTAL: |%11.3f ms |%9" PRIu64 " |\n", - (double)sched->all_runtime / 1e6, sched->all_count); + (double)sched->all_runtime / NSEC_PER_MSEC, sched->all_count); printf(" ---------------------------------------------------\n"); diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c index c859e59dfe3e..6b3c8b0d3276 100644 --- a/tools/perf/builtin-script.c +++ b/tools/perf/builtin-script.c @@ -24,6 +24,7 @@ #include "util/thread-stack.h" #include <linux/bitmap.h> #include <linux/stringify.h> +#include <linux/time64.h> #include "asm/bug.h" #include "util/mem-events.h" @@ -464,9 +465,9 @@ static void print_sample_start(struct perf_sample *sample, if (PRINT_FIELD(TIME)) { nsecs = sample->time; - secs = nsecs / NSECS_PER_SEC; - nsecs -= secs * NSECS_PER_SEC; - usecs = nsecs / NSECS_PER_USEC; + secs = nsecs / NSEC_PER_SEC; + nsecs -= secs * NSEC_PER_SEC; + usecs = nsecs / NSEC_PER_USEC; if (nanosecs) printf("%5lu.%09llu: ", secs, nsecs); else diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index 3c7452b39f57..90882b1d6a91 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -65,6 +65,7 @@ #include "util/group.h" #include "asm/bug.h" +#include <linux/time64.h> #include <api/fs/fs.h> #include <stdlib.h> #include <sys/prctl.h> @@ -172,7 +173,7 @@ static inline void diff_timespec(struct timespec *r, struct timespec *a, { r->tv_sec = a->tv_sec - b->tv_sec; if (a->tv_nsec < b->tv_nsec) { - r->tv_nsec = a->tv_nsec + 1000000000L - b->tv_nsec; + r->tv_nsec = a->tv_nsec + NSEC_PER_SEC - b->tv_nsec; r->tv_sec--; } else { r->tv_nsec = a->tv_nsec - b->tv_nsec ; @@ -354,7 +355,7 @@ static void process_interval(void) diff_timespec(&rs, &ts, &ref_time); if (STAT_RECORD) { - if (WRITE_STAT_ROUND_EVENT(rs.tv_sec * NSECS_PER_SEC + rs.tv_nsec, INTERVAL)) + if (WRITE_STAT_ROUND_EVENT(rs.tv_sec * NSEC_PER_SEC + rs.tv_nsec, INTERVAL)) pr_err("failed to write stat round event\n"); } @@ -364,7 +365,7 @@ static void process_interval(void) static void enable_counters(void) { if (initial_delay) - usleep(initial_delay * 1000); + usleep(initial_delay * USEC_PER_MSEC); /* * We need to enable counters only if: @@ -541,8 +542,8 @@ static int __run_perf_stat(int argc, const char **argv) bool is_pipe = STAT_RECORD ? perf_stat.file.is_pipe : false; if (interval) { - ts.tv_sec = interval / 1000; - ts.tv_nsec = (interval % 1000) * 1000000; + ts.tv_sec = interval / USEC_PER_MSEC; + ts.tv_nsec = (interval % USEC_PER_MSEC) * NSEC_PER_MSEC; } else { ts.tv_sec = 1; ts.tv_nsec = 0; @@ -971,7 +972,7 @@ static void print_metric_header(void *ctx, const char *color __maybe_unused, static void nsec_printout(int id, int nr, struct perf_evsel *evsel, double avg) { FILE *output = stat_config.output; - double msecs = avg / 1e6; + double msecs = avg / NSEC_PER_MSEC; const char *fmt_v, *fmt_n; char name[25]; @@ -1460,7 +1461,7 @@ static void print_footer(void) if (!null_run) fprintf(output, "\n"); fprintf(output, " %17.9f seconds time elapsed", - avg_stats(&walltime_nsecs_stats)/1e9); + avg_stats(&walltime_nsecs_stats) / NSEC_PER_SEC); if (run_count > 1) { fprintf(output, " "); print_noise_pct(stddev_stats(&walltime_nsecs_stats), @@ -2175,8 +2176,8 @@ static int process_stat_round_event(struct perf_tool *tool __maybe_unused, update_stats(&walltime_nsecs_stats, stat_round->time); if (stat_config.interval && stat_round->time) { - tsh.tv_sec = stat_round->time / NSECS_PER_SEC; - tsh.tv_nsec = stat_round->time % NSECS_PER_SEC; + tsh.tv_sec = stat_round->time / NSEC_PER_SEC; + tsh.tv_nsec = stat_round->time % NSEC_PER_SEC; ts = &tsh; } diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c index 733a55422d03..e7eaa298d34a 100644 --- a/tools/perf/builtin-timechart.c +++ b/tools/perf/builtin-timechart.c @@ -24,6 +24,7 @@ #include "util/evlist.h" #include "util/evsel.h" #include <linux/rbtree.h> +#include <linux/time64.h> #include "util/symbol.h" #include "util/callchain.h" #include "util/strlist.h" @@ -1288,9 +1289,9 @@ static void draw_process_bars(struct timechart *tchart) if (c->comm) { char comm[256]; if (c->total_time > 5000000000) /* 5 seconds */ - sprintf(comm, "%s:%i (%2.2fs)", c->comm, p->pid, c->total_time / 1000000000.0); + sprintf(comm, "%s:%i (%2.2fs)", c->comm, p->pid, c->total_time / (double)NSEC_PER_SEC); else - sprintf(comm, "%s:%i (%3.1fms)", c->comm, p->pid, c->total_time / 1000000.0); + sprintf(comm, "%s:%i (%3.1fms)", c->comm, p->pid, c->total_time / (double)NSEC_PER_MSEC); svg_text(Y, c->start_time, comm); } @@ -1637,7 +1638,7 @@ static int __cmd_timechart(struct timechart *tchart, const char *output_name) write_svg_file(tchart, output_name); pr_info("Written %2.1f seconds of trace to %s.\n", - (tchart->last_time - tchart->first_time) / 1000000000.0, output_name); + (tchart->last_time - tchart->first_time) / (double)NSEC_PER_SEC, output_name); out_delete: perf_session__delete(session); return ret; @@ -1901,10 +1902,10 @@ parse_time(const struct option *opt, const char *arg, int __maybe_unused unset) if (sscanf(arg, "%" PRIu64 "%cs", value, &unit) > 0) { switch (unit) { case 'm': - *value *= 1000000; + *value *= NSEC_PER_MSEC; break; case 'u': - *value *= 1000; + *value *= NSEC_PER_USEC; break; case 'n': break; @@ -1928,7 +1929,7 @@ int cmd_timechart(int argc, const char **argv, .ordered_events = true, }, .proc_num = 15, - .min_time = 1000000, + .min_time = NSEC_PER_MSEC, .merge_dist = 1000, }; const char *output_name = "output.svg"; diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c index 418ed94756d3..a3223aa22213 100644 --- a/tools/perf/builtin-top.c +++ b/tools/perf/builtin-top.c @@ -68,6 +68,7 @@ #include <sys/mman.h> #include <linux/stringify.h> +#include <linux/time64.h> #include <linux/types.h> static volatile int done; @@ -624,7 +625,7 @@ static void *display_thread(void *arg) display_setup_sig(); pthread__unblock_sigwinch(); repeat: - delay_msecs = top->delay_secs * 1000; + delay_msecs = top->delay_secs * MSEC_PER_SEC; set_term_quiet_input(&save); /* trash return*/ getc(stdin); diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c index b8c6766301db..b4fc1ab3d2a7 100644 --- a/tools/perf/builtin-trace.c +++ b/tools/perf/builtin-trace.c @@ -45,6 +45,7 @@ #include <linux/audit.h> #include <linux/random.h> #include <linux/stringify.h> +#include <linux/time64.h> #ifndef O_CLOEXEC # define O_CLOEXEC 02000000 diff --git a/tools/perf/perf.h b/tools/perf/perf.h index cb0f1356ff81..9a0236a4cf95 100644 --- a/tools/perf/perf.h +++ b/tools/perf/perf.h @@ -14,13 +14,6 @@ void test_attr__open(struct perf_event_attr *attr, pid_t pid, int cpu, #define HAVE_ATTR_TEST #include "perf-sys.h" -#ifndef NSEC_PER_SEC -# define NSEC_PER_SEC 1000000000ULL -#endif -#ifndef NSEC_PER_USEC -# define NSEC_PER_USEC 1000ULL -#endif - static inline unsigned long long rdclock(void) { struct timespec ts; diff --git a/tools/perf/tests/backward-ring-buffer.c b/tools/perf/tests/backward-ring-buffer.c index 615780cbfe1d..e6d1816e431a 100644 --- a/tools/perf/tests/backward-ring-buffer.c +++ b/tools/perf/tests/backward-ring-buffer.c @@ -97,7 +97,7 @@ int test__backward_ring_buffer(int subtest __maybe_unused) evlist = perf_evlist__new(); if (!evlist) { - pr_debug("No ehough memory to create evlist\n"); + pr_debug("No enough memory to create evlist\n"); return TEST_FAIL; } diff --git a/tools/perf/tests/bpf.c b/tools/perf/tests/bpf.c index fc54064b9186..2673e86ed50f 100644 --- a/tools/perf/tests/bpf.c +++ b/tools/perf/tests/bpf.c @@ -125,7 +125,7 @@ static int do_test(struct bpf_object *obj, int (*func)(void), /* Instead of perf_evlist__new_default, don't add default events */ evlist = perf_evlist__new(); if (!evlist) { - pr_debug("No ehough memory to create evlist\n"); + pr_debug("No enough memory to create evlist\n"); return TEST_FAIL; } diff --git a/tools/perf/ui/browsers/hists.c b/tools/perf/ui/browsers/hists.c index 13d414384739..f0611c937d4b 100644 --- a/tools/perf/ui/browsers/hists.c +++ b/tools/perf/ui/browsers/hists.c @@ -69,8 +69,11 @@ static u32 hist_browser__nr_entries(struct hist_browser *hb) static void hist_browser__update_rows(struct hist_browser *hb) { struct ui_browser *browser = &hb->b; - u16 header_offset = hb->show_headers ? 1 : 0, index_row; + struct hists *hists = hb->hists; + struct perf_hpp_list *hpp_list = hists->hpp_list; + u16 header_offset, index_row; + header_offset = hb->show_headers ? hpp_list->nr_header_lines : 0; browser->rows = browser->height - header_offset; /* * Verify if we were at the last line and that line isn't @@ -99,8 +102,11 @@ static void hist_browser__refresh_dimensions(struct ui_browser *browser) static void hist_browser__gotorc(struct hist_browser *browser, int row, int column) { - u16 header_offset = browser->show_headers ? 1 : 0; + struct hists *hists = browser->hists; + struct perf_hpp_list *hpp_list = hists->hpp_list; + u16 header_offset; + header_offset = browser->show_headers ? hpp_list->nr_header_lines : 0; ui_browser__gotorc(&browser->b, row + header_offset, column); } @@ -1496,7 +1502,9 @@ static int advance_hpp_check(struct perf_hpp *hpp, int inc) return hpp->size <= 0; } -static int hists_browser__scnprintf_headers(struct hist_browser *browser, char *buf, size_t size) +static int +hists_browser__scnprintf_headers(struct hist_browser *browser, char *buf, + size_t size, int line) { struct hists *hists = browser->hists; struct perf_hpp dummy_hpp = { @@ -1506,6 +1514,7 @@ static int hists_browser__scnprintf_headers(struct hist_browser *browser, char * struct perf_hpp_fmt *fmt; size_t ret = 0; int column = 0; + int span = 0; if (symbol_conf.use_callchain) { ret = scnprintf(buf, size, " "); @@ -1517,10 +1526,13 @@ static int hists_browser__scnprintf_headers(struct hist_browser *browser, char * if (perf_hpp__should_skip(fmt, hists) || column++ < browser->b.horiz_scroll) continue; - ret = fmt->header(fmt, &dummy_hpp, hists); + ret = fmt->header(fmt, &dummy_hpp, hists, line, &span); if (advance_hpp_check(&dummy_hpp, ret)) break; + if (span) + continue; + ret = scnprintf(dummy_hpp.buf, dummy_hpp.size, " "); if (advance_hpp_check(&dummy_hpp, ret)) break; @@ -1554,7 +1566,7 @@ static int hists_browser__scnprintf_hierarchy_headers(struct hist_browser *brows if (column++ < browser->b.horiz_scroll) continue; - ret = fmt->header(fmt, &dummy_hpp, hists); + ret = fmt->header(fmt, &dummy_hpp, hists, 0, NULL); if (advance_hpp_check(&dummy_hpp, ret)) break; @@ -1591,7 +1603,7 @@ static int hists_browser__scnprintf_hierarchy_headers(struct hist_browser *brows } first_col = false; - ret = fmt->header(fmt, &dummy_hpp, hists); + ret = fmt->header(fmt, &dummy_hpp, hists, 0, NULL); dummy_hpp.buf[ret] = '\0'; start = trim(dummy_hpp.buf); @@ -1622,14 +1634,21 @@ static void hists_browser__hierarchy_headers(struct hist_browser *browser) static void hists_browser__headers(struct hist_browser *browser) { - char headers[1024]; + struct hists *hists = browser->hists; + struct perf_hpp_list *hpp_list = hists->hpp_list; - hists_browser__scnprintf_headers(browser, headers, - sizeof(headers)); + int line; - ui_browser__gotorc(&browser->b, 0, 0); - ui_browser__set_color(&browser->b, HE_COLORSET_ROOT); - ui_browser__write_nstring(&browser->b, headers, browser->b.width + 1); + for (line = 0; line < hpp_list->nr_header_lines; line++) { + char headers[1024]; + + hists_browser__scnprintf_headers(browser, headers, + sizeof(headers), line); + + ui_browser__gotorc(&browser->b, line, 0); + ui_browser__set_color(&browser->b, HE_COLORSET_ROOT); + ui_browser__write_nstring(&browser->b, headers, browser->b.width + 1); + } } static void hist_browser__show_headers(struct hist_browser *browser) @@ -1656,10 +1675,13 @@ static unsigned int hist_browser__refresh(struct ui_browser *browser) u16 header_offset = 0; struct rb_node *nd; struct hist_browser *hb = container_of(browser, struct hist_browser, b); + struct hists *hists = hb->hists; if (hb->show_headers) { + struct perf_hpp_list *hpp_list = hists->hpp_list; + hist_browser__show_headers(hb); - header_offset = 1; + header_offset = hpp_list->nr_header_lines; } ui_browser__hists_init_top(browser); @@ -2418,8 +2440,6 @@ do_zoom_dso(struct hist_browser *browser, struct popup_action *act) browser->hists->dso_filter = NULL; ui_helpline__pop(); } else { - if (map == NULL) - return 0; ui_helpline__fpush("To zoom out press ESC or ENTER + \"Zoom out of %s DSO\"", __map__is_kernel(map) ? "the Kernel" : map->dso->short_name); browser->hists->dso_filter = map->dso; diff --git a/tools/perf/ui/gtk/hists.c b/tools/perf/ui/gtk/hists.c index c5f3677f6679..a4f02de7c1b5 100644 --- a/tools/perf/ui/gtk/hists.c +++ b/tools/perf/ui/gtk/hists.c @@ -549,7 +549,7 @@ static void perf_gtk__show_hierarchy(GtkWidget *window, struct hists *hists, strcat(buf, "+"); first_col = false; - fmt->header(fmt, &hpp, hists); + fmt->header(fmt, &hpp, hists, 0, NULL); strcat(buf, ltrim(rtrim(hpp.buf))); } } diff --git a/tools/perf/ui/hist.c b/tools/perf/ui/hist.c index 4274969ddc89..b47fafc8ee2a 100644 --- a/tools/perf/ui/hist.c +++ b/tools/perf/ui/hist.c @@ -230,7 +230,8 @@ static int hpp__width_fn(struct perf_hpp_fmt *fmt, } static int hpp__header_fn(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp, - struct hists *hists) + struct hists *hists, int line __maybe_unused, + int *span __maybe_unused) { int len = hpp__width_fn(fmt, hpp, hists); return scnprintf(hpp->buf, hpp->size, "%*s", len, fmt->name); @@ -441,6 +442,7 @@ struct perf_hpp_fmt perf_hpp__format[] = { struct perf_hpp_list perf_hpp_list = { .fields = LIST_HEAD_INIT(perf_hpp_list.fields), .sorts = LIST_HEAD_INIT(perf_hpp_list.sorts), + .nr_header_lines = 1, }; #undef HPP__COLOR_PRINT_FNS diff --git a/tools/perf/ui/stdio/hist.c b/tools/perf/ui/stdio/hist.c index f04a63112079..9b65f4a6b35a 100644 --- a/tools/perf/ui/stdio/hist.c +++ b/tools/perf/ui/stdio/hist.c @@ -549,7 +549,7 @@ static int print_hierarchy_header(struct hists *hists, struct perf_hpp *hpp, struct perf_hpp_list_node, list); perf_hpp_list__for_each_format(&fmt_node->hpp, fmt) { - fmt->header(fmt, hpp, hists); + fmt->header(fmt, hpp, hists, 0, NULL); fprintf(fp, "%s%s", hpp->buf, sep ?: " "); } @@ -569,7 +569,7 @@ static int print_hierarchy_header(struct hists *hists, struct perf_hpp *hpp, header_width += fprintf(fp, "+"); first_col = false; - fmt->header(fmt, hpp, hists); + fmt->header(fmt, hpp, hists, 0, NULL); header_width += fprintf(fp, "%s", trim(hpp->buf)); } @@ -639,33 +639,52 @@ hists__fprintf_hierarchy_headers(struct hists *hists, return print_hierarchy_header(hists, hpp, symbol_conf.field_sep, fp); } -static int -hists__fprintf_standard_headers(struct hists *hists, - struct perf_hpp *hpp, - FILE *fp) +static void fprintf_line(struct hists *hists, struct perf_hpp *hpp, + int line, FILE *fp) { struct perf_hpp_fmt *fmt; - unsigned int width; const char *sep = symbol_conf.field_sep; bool first = true; + int span = 0; hists__for_each_format(hists, fmt) { if (perf_hpp__should_skip(fmt, hists)) continue; - if (!first) + if (!first && !span) fprintf(fp, "%s", sep ?: " "); else first = false; - fmt->header(fmt, hpp, hists); - fprintf(fp, "%s", hpp->buf); + fmt->header(fmt, hpp, hists, line, &span); + + if (!span) + fprintf(fp, "%s", hpp->buf); } +} - fprintf(fp, "\n"); +static int +hists__fprintf_standard_headers(struct hists *hists, + struct perf_hpp *hpp, + FILE *fp) +{ + struct perf_hpp_list *hpp_list = hists->hpp_list; + struct perf_hpp_fmt *fmt; + unsigned int width; + const char *sep = symbol_conf.field_sep; + bool first = true; + int line; + + for (line = 0; line < hpp_list->nr_header_lines; line++) { + /* first # is displayed one level up */ + if (line) + fprintf(fp, "# "); + fprintf_line(hists, hpp, line, fp); + fprintf(fp, "\n"); + } if (sep) - return 1; + return hpp_list->nr_header_lines; first = true; @@ -689,7 +708,7 @@ hists__fprintf_standard_headers(struct hists *hists, fprintf(fp, "\n"); fprintf(fp, "#\n"); - return 3; + return hpp_list->nr_header_lines + 2; } static int hists__fprintf_headers(struct hists *hists, FILE *fp) diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c index 4024d309bb00..25a9259a6a6e 100644 --- a/tools/perf/util/annotate.c +++ b/tools/perf/util/annotate.c @@ -1162,53 +1162,60 @@ int symbol__strerror_disassemble(struct symbol *sym __maybe_unused, struct map * return 0; } -int symbol__disassemble(struct symbol *sym, struct map *map, size_t privsize) +static int dso__disassemble_filename(struct dso *dso, char *filename, size_t filename_size) { - struct dso *dso = map->dso; - char *filename = dso__build_id_filename(dso, NULL, 0); - bool free_filename = true; - char command[PATH_MAX * 2]; - FILE *file; - int err = 0; - char symfs_filename[PATH_MAX]; - struct kcore_extract kce; - bool delete_extract = false; - int stdout_fd[2]; - int lineno = 0; - int nline; - pid_t pid; + char linkname[PATH_MAX]; + char *build_id_filename; - if (filename) - symbol__join_symfs(symfs_filename, filename); + if (dso->symtab_type == DSO_BINARY_TYPE__KALLSYMS && + !dso__is_kcore(dso)) + return SYMBOL_ANNOTATE_ERRNO__NO_VMLINUX; - if (filename == NULL) { + build_id_filename = dso__build_id_filename(dso, NULL, 0); + if (build_id_filename) { + __symbol__join_symfs(filename, filename_size, build_id_filename); + free(build_id_filename); + } else { if (dso->has_build_id) return ENOMEM; goto fallback; - } else if (dso__is_kcore(dso) || - readlink(symfs_filename, command, sizeof(command)) < 0 || - strstr(command, DSO__NAME_KALLSYMS) || - access(symfs_filename, R_OK)) { - free(filename); + } + + if (dso__is_kcore(dso) || + readlink(filename, linkname, sizeof(linkname)) < 0 || + strstr(linkname, DSO__NAME_KALLSYMS) || + access(filename, R_OK)) { fallback: /* * If we don't have build-ids or the build-id file isn't in the * cache, or is just a kallsyms file, well, lets hope that this * DSO is the same as when 'perf record' ran. */ - filename = (char *)dso->long_name; - symbol__join_symfs(symfs_filename, filename); - free_filename = false; + __symbol__join_symfs(filename, filename_size, dso->long_name); } - if (dso->symtab_type == DSO_BINARY_TYPE__KALLSYMS && - !dso__is_kcore(dso)) { - err = SYMBOL_ANNOTATE_ERRNO__NO_VMLINUX; - goto out_free_filename; - } + return 0; +} + +int symbol__disassemble(struct symbol *sym, struct map *map, size_t privsize) +{ + struct dso *dso = map->dso; + char command[PATH_MAX * 2]; + FILE *file; + char symfs_filename[PATH_MAX]; + struct kcore_extract kce; + bool delete_extract = false; + int stdout_fd[2]; + int lineno = 0; + int nline; + pid_t pid; + int err = dso__disassemble_filename(dso, symfs_filename, sizeof(symfs_filename)); + + if (err) + return err; pr_debug("%s: filename=%s, sym=%s, start=%#" PRIx64 ", end=%#" PRIx64 "\n", __func__, - filename, sym->name, map->unmap_ip(map, sym->start), + symfs_filename, sym->name, map->unmap_ip(map, sym->start), map->unmap_ip(map, sym->end)); pr_debug("annotating [%p] %30s : [%p] %30s\n", @@ -1223,11 +1230,6 @@ fallback: delete_extract = true; strlcpy(symfs_filename, kce.extract_filename, sizeof(symfs_filename)); - if (free_filename) { - free(filename); - free_filename = false; - } - filename = symfs_filename; } } else if (dso__needs_decompress(dso)) { char tmp[PATH_MAX]; @@ -1236,14 +1238,14 @@ fallback: bool ret; if (kmod_path__parse_ext(&m, symfs_filename)) - goto out_free_filename; + goto out; snprintf(tmp, PATH_MAX, "/tmp/perf-kmod-XXXXXX"); fd = mkstemp(tmp); if (fd < 0) { free(m.ext); - goto out_free_filename; + goto out; } ret = decompress_to_file(m.ext, symfs_filename, fd); @@ -1255,7 +1257,7 @@ fallback: close(fd); if (!ret) - goto out_free_filename; + goto out; strcpy(symfs_filename, tmp); } @@ -1271,7 +1273,7 @@ fallback: map__rip_2objdump(map, sym->end), symbol_conf.annotate_asm_raw ? "" : "--no-show-raw", symbol_conf.annotate_src ? "-S" : "", - symfs_filename, filename); + symfs_filename, symfs_filename); pr_debug("Executing: %s\n", command); @@ -1333,11 +1335,10 @@ out_remove_tmp: if (dso__needs_decompress(dso)) unlink(symfs_filename); -out_free_filename: + if (delete_extract) kcore_extract__delete(&kce); - if (free_filename) - free(filename); +out: return err; out_close_stdout: diff --git a/tools/perf/util/bpf-loader.c b/tools/perf/util/bpf-loader.c index 1f12e4e40006..2b2c9b82f5ab 100644 --- a/tools/perf/util/bpf-loader.c +++ b/tools/perf/util/bpf-loader.c @@ -531,7 +531,7 @@ static int map_prologue(struct perf_probe_event *pev, int *mapping, ptevs = malloc(array_sz); if (!ptevs) { - pr_debug("No ehough memory: alloc ptevs failed\n"); + pr_debug("No enough memory: alloc ptevs failed\n"); return -ENOMEM; } diff --git a/tools/perf/util/debug.c b/tools/perf/util/debug.c index 8c4212abd19b..c1838b643108 100644 --- a/tools/perf/util/debug.c +++ b/tools/perf/util/debug.c @@ -6,6 +6,7 @@ #include <stdarg.h> #include <stdio.h> #include <api/debug.h> +#include <linux/time64.h> #include "cache.h" #include "color.h" @@ -14,9 +15,6 @@ #include "util.h" #include "target.h" -#define NSECS_PER_SEC 1000000000ULL -#define NSECS_PER_USEC 1000ULL - int verbose; bool dump_trace = false, quiet = false; int debug_ordered_events; @@ -54,9 +52,9 @@ static int veprintf_time(u64 t, const char *fmt, va_list args) int ret = 0; u64 secs, usecs, nsecs = t; - secs = nsecs / NSECS_PER_SEC; - nsecs -= secs * NSECS_PER_SEC; - usecs = nsecs / NSECS_PER_USEC; + secs = nsecs / NSEC_PER_SEC; + nsecs -= secs * NSEC_PER_SEC; + usecs = nsecs / NSEC_PER_USEC; ret = fprintf(stderr, "[%13" PRIu64 ".%06" PRIu64 "] ", secs, usecs); diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c index 8f0db4007282..85dd0db0a127 100644 --- a/tools/perf/util/header.c +++ b/tools/perf/util/header.c @@ -828,8 +828,7 @@ static int write_group_desc(int fd, struct perf_header *h __maybe_unused, * default get_cpuid(): nothing gets recorded * actual implementation must be in arch/$(ARCH)/util/header.c */ -int __attribute__ ((weak)) get_cpuid(char *buffer __maybe_unused, - size_t sz __maybe_unused) +int __weak get_cpuid(char *buffer __maybe_unused, size_t sz __maybe_unused) { return -1; } diff --git a/tools/perf/util/hist.h b/tools/perf/util/hist.h index 0a1edf1ab450..a002c93fe422 100644 --- a/tools/perf/util/hist.h +++ b/tools/perf/util/hist.h @@ -230,7 +230,7 @@ struct perf_hpp { struct perf_hpp_fmt { const char *name; int (*header)(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp, - struct hists *hists); + struct hists *hists, int line, int *span); int (*width)(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp, struct hists *hists); int (*color)(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp, @@ -259,6 +259,7 @@ struct perf_hpp_list { struct list_head fields; struct list_head sorts; + int nr_header_lines; int need_collapse; int parent; int sym; diff --git a/tools/perf/util/probe-file.c b/tools/perf/util/probe-file.c index 9c3b9ed5b3c3..697ef66bff91 100644 --- a/tools/perf/util/probe-file.c +++ b/tools/perf/util/probe-file.c @@ -877,3 +877,60 @@ int probe_cache__show_all_caches(struct strfilter *filter) return 0; } + +static struct { + const char *pattern; + bool avail; + bool checked; +} probe_type_table[] = { +#define DEFINE_TYPE(idx, pat, def_avail) \ + [idx] = {.pattern = pat, .avail = (def_avail)} + DEFINE_TYPE(PROBE_TYPE_U, "* u8/16/32/64,*", true), + DEFINE_TYPE(PROBE_TYPE_S, "* s8/16/32/64,*", true), + DEFINE_TYPE(PROBE_TYPE_X, "* x8/16/32/64,*", false), + DEFINE_TYPE(PROBE_TYPE_STRING, "* string,*", true), + DEFINE_TYPE(PROBE_TYPE_BITFIELD, + "* b<bit-width>@<bit-offset>/<container-size>", true), +}; + +bool probe_type_is_available(enum probe_type type) +{ + FILE *fp; + char *buf = NULL; + size_t len = 0; + bool target_line = false; + bool ret = probe_type_table[type].avail; + + if (type >= PROBE_TYPE_END) + return false; + /* We don't have to check the type which supported by default */ + if (ret || probe_type_table[type].checked) + return ret; + + if (asprintf(&buf, "%s/README", tracing_path) < 0) + return ret; + + fp = fopen(buf, "r"); + if (!fp) + goto end; + + zfree(&buf); + while (getline(&buf, &len, fp) > 0 && !ret) { + if (!target_line) { + target_line = !!strstr(buf, " type: "); + if (!target_line) + continue; + } else if (strstr(buf, "\t ") != buf) + break; + ret = strglobmatch(buf, probe_type_table[type].pattern); + } + /* Cache the result */ + probe_type_table[type].checked = true; + probe_type_table[type].avail = ret; + + fclose(fp); +end: + free(buf); + + return ret; +} diff --git a/tools/perf/util/probe-file.h b/tools/perf/util/probe-file.h index 9577b5c0b487..eba44c3e9dca 100644 --- a/tools/perf/util/probe-file.h +++ b/tools/perf/util/probe-file.h @@ -19,6 +19,15 @@ struct probe_cache { struct list_head entries; }; +enum probe_type { + PROBE_TYPE_U = 0, + PROBE_TYPE_S, + PROBE_TYPE_X, + PROBE_TYPE_STRING, + PROBE_TYPE_BITFIELD, + PROBE_TYPE_END, +}; + #define PF_FL_UPROBE 1 #define PF_FL_RW 2 #define for_each_probe_cache_entry(entry, pcache) \ @@ -54,6 +63,7 @@ struct probe_cache_entry *probe_cache__find(struct probe_cache *pcache, struct probe_cache_entry *probe_cache__find_by_name(struct probe_cache *pcache, const char *group, const char *event); int probe_cache__show_all_caches(struct strfilter *filter); +bool probe_type_is_available(enum probe_type type); #else /* ! HAVE_LIBELF_SUPPORT */ static inline struct probe_cache *probe_cache__new(const char *tgt __maybe_unused) { diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c index 5c290c682afe..ac4740f8ee3a 100644 --- a/tools/perf/util/probe-finder.c +++ b/tools/perf/util/probe-finder.c @@ -39,6 +39,7 @@ #include "util.h" #include "symbol.h" #include "probe-finder.h" +#include "probe-file.h" /* Kprobe tracer basic type is up to u64 */ #define MAX_BASIC_TYPE_BITS 64 @@ -297,13 +298,13 @@ static int convert_variable_type(Dwarf_Die *vr_die, char sbuf[STRERR_BUFSIZE]; int bsize, boffs, total; int ret; - char sign; + char prefix; /* TODO: check all types */ - if (cast && strcmp(cast, "string") != 0 && + if (cast && strcmp(cast, "string") != 0 && strcmp(cast, "x") != 0 && strcmp(cast, "s") != 0 && strcmp(cast, "u") != 0) { /* Non string type is OK */ - /* and respect signedness cast */ + /* and respect signedness/hexadecimal cast */ tvar->type = strdup(cast); return (tvar->type == NULL) ? -ENOMEM : 0; } @@ -365,11 +366,15 @@ static int convert_variable_type(Dwarf_Die *vr_die, } if (cast && (strcmp(cast, "u") == 0)) - sign = 'u'; + prefix = 'u'; else if (cast && (strcmp(cast, "s") == 0)) - sign = 's'; + prefix = 's'; + else if (cast && (strcmp(cast, "x") == 0) && + probe_type_is_available(PROBE_TYPE_X)) + prefix = 'x'; else - sign = die_is_signed_type(&type) ? 's' : 'u'; + prefix = die_is_signed_type(&type) ? 's' : + probe_type_is_available(PROBE_TYPE_X) ? 'x' : 'u'; ret = dwarf_bytesize(&type); if (ret <= 0) @@ -383,7 +388,7 @@ static int convert_variable_type(Dwarf_Die *vr_die, dwarf_diename(&type), MAX_BASIC_TYPE_BITS); ret = MAX_BASIC_TYPE_BITS; } - ret = snprintf(buf, 16, "%c%d", sign, ret); + ret = snprintf(buf, 16, "%c%d", prefix, ret); formatted: if (ret < 0 || ret >= 16) { diff --git a/tools/perf/util/scripting-engines/trace-event-perl.c b/tools/perf/util/scripting-engines/trace-event-perl.c index 5d1eb1ccd96c..e55a132f69b7 100644 --- a/tools/perf/util/scripting-engines/trace-event-perl.c +++ b/tools/perf/util/scripting-engines/trace-event-perl.c @@ -25,6 +25,7 @@ #include <ctype.h> #include <errno.h> #include <linux/bitmap.h> +#include <linux/time64.h> #include "../util.h" #include <EXTERN.h> @@ -359,8 +360,8 @@ static void perl_process_tracepoint(struct perf_sample *sample, if (!test_and_set_bit(event->id, events_defined)) define_event_symbols(event, handler, event->print_fmt.args); - s = nsecs / NSECS_PER_SEC; - ns = nsecs - s * NSECS_PER_SEC; + s = nsecs / NSEC_PER_SEC; + ns = nsecs - s * NSEC_PER_SEC; scripting_context->event_data = data; scripting_context->pevent = evsel->tp_format->pevent; diff --git a/tools/perf/util/scripting-engines/trace-event-python.c b/tools/perf/util/scripting-engines/trace-event-python.c index e0203b979474..089438da1f7f 100644 --- a/tools/perf/util/scripting-engines/trace-event-python.c +++ b/tools/perf/util/scripting-engines/trace-event-python.c @@ -27,6 +27,7 @@ #include <stdbool.h> #include <errno.h> #include <linux/bitmap.h> +#include <linux/time64.h> #include "../../perf.h" #include "../debug.h" @@ -426,8 +427,8 @@ static void python_process_tracepoint(struct perf_sample *sample, if (!dict) Py_FatalError("couldn't create Python dict"); } - s = nsecs / NSECS_PER_SEC; - ns = nsecs - s * NSECS_PER_SEC; + s = nsecs / NSEC_PER_SEC; + ns = nsecs - s * NSEC_PER_SEC; scripting_context->event_data = data; scripting_context->pevent = evsel->tp_format->pevent; diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c index 3d3cb8392c86..1884d7f9b9d2 100644 --- a/tools/perf/util/sort.c +++ b/tools/perf/util/sort.c @@ -11,7 +11,7 @@ regex_t parent_regex; const char default_parent_pattern[] = "^sys_|^do_page_fault"; const char *parent_pattern = default_parent_pattern; -const char default_sort_order[] = "comm,dso,symbol"; +const char *default_sort_order = "comm,dso,symbol"; const char default_branch_sort_order[] = "comm,dso_from,symbol_from,symbol_to,cycles"; const char default_mem_sort_order[] = "local_weight,mem,sym,dso,symbol_daddr,dso_daddr,snoop,tlb,locked"; const char default_top_sort_order[] = "dso,symbol"; @@ -1492,7 +1492,8 @@ void perf_hpp__reset_sort_width(struct perf_hpp_fmt *fmt, struct hists *hists) } static int __sort__hpp_header(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp, - struct hists *hists) + struct hists *hists, int line __maybe_unused, + int *span __maybe_unused) { struct hpp_sort_entry *hse; size_t len = fmt->user_len; @@ -1797,7 +1798,9 @@ static void update_dynamic_len(struct hpp_dynamic_entry *hde, } static int __sort__hde_header(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp, - struct hists *hists __maybe_unused) + struct hists *hists __maybe_unused, + int line __maybe_unused, + int *span __maybe_unused) { struct hpp_dynamic_entry *hde; size_t len = fmt->user_len; diff --git a/tools/perf/util/sort.h b/tools/perf/util/sort.h index 7ca37ea17395..28c0524c8702 100644 --- a/tools/perf/util/sort.h +++ b/tools/perf/util/sort.h @@ -28,7 +28,7 @@ extern const char *sort_order; extern const char *field_order; extern const char default_parent_pattern[]; extern const char *parent_pattern; -extern const char default_sort_order[]; +extern const char *default_sort_order; extern regex_t ignore_callees_regex; extern int have_ignore_callees; extern enum sort_mode sort__mode; diff --git a/tools/perf/util/svghelper.c b/tools/perf/util/svghelper.c index eec6c1149f44..1cbada2dc6be 100644 --- a/tools/perf/util/svghelper.c +++ b/tools/perf/util/svghelper.c @@ -18,6 +18,7 @@ #include <unistd.h> #include <string.h> #include <linux/bitmap.h> +#include <linux/time64.h> #include "perf.h" #include "svghelper.h" @@ -274,14 +275,14 @@ static char *time_to_string(u64 duration) text[0] = 0; - if (duration < 1000) /* less than 1 usec */ + if (duration < NSEC_PER_USEC) /* less than 1 usec */ return text; - if (duration < 1000 * 1000) { /* less than 1 msec */ - sprintf(text, "%.1f us", duration / 1000.0); + if (duration < NSEC_PER_MSEC) { /* less than 1 msec */ + sprintf(text, "%.1f us", duration / (double)NSEC_PER_USEC); return text; } - sprintf(text, "%.1f ms", duration / 1000.0 / 1000); + sprintf(text, "%.1f ms", duration / (double)NSEC_PER_MSEC); return text; } @@ -297,7 +298,7 @@ void svg_waiting(int Yslot, int cpu, u64 start, u64 end, const char *backtrace) style = "waiting"; - if (end-start > 10 * 1000000) /* 10 msec */ + if (end-start > 10 * NSEC_PER_MSEC) /* 10 msec */ style = "WAITING"; text = time_to_string(end-start); diff --git a/tools/perf/util/util.c b/tools/perf/util/util.c index cee559d8c9e8..85c56800f17a 100644 --- a/tools/perf/util/util.c +++ b/tools/perf/util/util.c @@ -15,6 +15,7 @@ #include <byteswap.h> #include <linux/kernel.h> #include <linux/log2.h> +#include <linux/time64.h> #include <unistd.h> #include "callchain.h" #include "strlist.h" diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h index e5f55477491d..43899e0d6fa1 100644 --- a/tools/perf/util/util.h +++ b/tools/perf/util/util.h @@ -179,10 +179,6 @@ static inline void *zalloc(size_t size) #undef tolower #undef toupper -#ifndef NSEC_PER_MSEC -#define NSEC_PER_MSEC 1000000L -#endif - int parse_nsec_time(const char *str, u64 *ptime); extern unsigned char sane_ctype[256]; |