perf: Add branch stack extra
Subsystems (3)
PATCH v3 1/6 perf: Add branch stack extra
From: Kan Liang <[email protected]>
Currently, the additional information of a branch entry is stored in a u64 space. With more and more information added, the space is running out. For example, the information of occurrences of events will be added for each branch.
Add a new branch sample type, PERF_SAMPLE_BRANCH_EXTRA, to indicate whether to support an extra space.
Two places were suggested to append the extra space.
https://lore.kernel.org/lkml/[email protected]/
One place is right after the flags of each branch entry. It changes the existing struct perf_branch_entry. In the later Intel-specific implementation, two separate spaces have to be created in the struct cpu_hw_events to store different branch entry structures. That duplicates space.
The other place is right after the entire struct perf_branch_stack. Only adding the new extra space in the struct cpu_hw_event is necessary. The disadvantage is that the pointer of the extra space has to be recorded. The common interface perf_sample_save_brstack() has to be updated as well.
The latter requires less space and is much straight forward. It is implemented in the patch.
Also, add a new branch sample type, PERF_SAMPLE_BRANCH_EVT_CNTRS, to indicate whether include occurrences of events in branch info. The information will be stored in the extra space.
Add two helper functions for the new branch sample types.
---
Changes since V2:
- Drop the new bit in struct perf_branch_entry
- Introduce a new sample type PERF_SAMPLE_BRANCH_EXTRA
arch/powerpc/perf/core-book3s.c | 2 +- arch/x86/events/amd/core.c | 2 +- arch/x86/events/core.c | 2 +- arch/x86/events/intel/core.c | 2 +- arch/x86/events/intel/ds.c | 4 ++-- include/linux/perf_event.h | 22 +++++++++++++++++++++- include/uapi/linux/perf_event.h | 9 +++++++++ kernel/events/core.c | 8 ++++++++ 8 files changed, 44 insertions(+), 7 deletions(-)
arch/powerpc/perf/core-book3s.c +1 -1
@@ -2313,7 +2313,7 @@ static void record_and_restart(struct perf_event *event, unsigned long val,struct cpu_hw_events *cpuhw; cpuhw =this_cpu_ptr (&cpu_hw_events);power_pmu_bhrb_read (event, cpuhw); -perf_sample_save_brstack (&data, event, &cpuhw->bhrb_stack); +perf_sample_save_brstack (&data, event, &cpuhw->bhrb_stack,NULL ); } if (event->attr.sample_type & PERF_SAMPLE_DATA_SRC &&
arch/x86/events/amd/core.c +1 -1
@@ -930,7 +930,7 @@ static int amd_pmu_v2_handle_irq(struct pt_regs *regs)continue ;if (has_branch_stack (event)) -perf_sample_save_brstack (&data, event, &cpuc->lbr_stack); +perf_sample_save_brstack (&data, event, &cpuc->lbr_stack,NULL );if (perf_event_overflow (event, &data, regs))x86_pmu_stop (event,0 );
arch/x86/events/core.c +1 -1
@@ -1702,7 +1702,7 @@ int x86_pmu_handle_irq(struct pt_regs *regs)perf_sample_data_init (&data,0 , event->hw.last_period);if (has_branch_stack (event)) -perf_sample_save_brstack (&data, event, &cpuc->lbr_stack); +perf_sample_save_brstack (&data, event, &cpuc->lbr_stack,NULL );if (perf_event_overflow (event, &data, regs))x86_pmu_stop (event,0 );
arch/x86/events/intel/core.c +1 -1
@@ -3058,7 +3058,7 @@ static int handle_pmi_common(struct pt_regs *regs, u64 status)perf_sample_data_init (&data,0 , event->hw.last_period);if (has_branch_stack (event)) -perf_sample_save_brstack (&data, event, &cpuc->lbr_stack); +perf_sample_save_brstack (&data, event, &cpuc->lbr_stack,NULL );if (perf_event_overflow (event, &data, regs))x86_pmu_stop (event,0 );
arch/x86/events/intel/ds.c +2 -2
@@ -1755,7 +1755,7 @@ static void setup_pebs_fixed_sample_data(struct perf_event *event,setup_pebs_time (event, data, pebs->tsc);if (has_branch_stack (event)) -perf_sample_save_brstack (data, event, &cpuc->lbr_stack); +perf_sample_save_brstack (data, event, &cpuc->lbr_stack,NULL ); } static void adaptive_pebs_save_regs(struct pt_regs *regs,
@@ -1912,7 +1912,7 @@ static void setup_pebs_adaptive_sample_data(struct perf_event *event,if (has_branch_stack (event)) {intel_pmu_store_pebs_lbrs (lbr); -perf_sample_save_brstack (data, event, &cpuc->lbr_stack); +perf_sample_save_brstack (data, event, &cpuc->lbr_stack,NULL ); } }
include/linux/perf_event.h +21 -1
@@ -1137,6 +1137,15 @@ static inline bool branch_sample_priv(const struct perf_event *event)return event->attr.branch_sample_type &PERF_SAMPLE_BRANCH_PRIV_SAVE ; } +static inline bool branch_sample_extra (const struct perf_event *event) +{ +return event->attr.branch_sample_type &PERF_SAMPLE_BRANCH_EXTRA ; +} + +static inline bool branch_sample_evt_cntrs (const struct perf_event *event) +{ +return event->attr.branch_sample_type &PERF_SAMPLE_BRANCH_EVT_CNTRS ; +}struct perf_sample_data {/*
@@ -1171,6 +1180,7 @@ struct perf_sample_data {struct perf_callchain_entry *callchain;struct perf_raw_record *raw;struct perf_branch_stack *br_stack; +u64 *br_stack_ext;union perf_sample_weight weight;union perf_mem_data_src data_src;u64 txn;
@@ -1248,7 +1258,8 @@ static inline void perf_sample_save_raw_data(struct perf_sample_data *data,static inline void perf_sample_save_brstack (struct perf_sample_data *data,struct perf_event *event, -struct perf_branch_stack *brs) +struct perf_branch_stack *brs, +u64 *brs_ext) {int size =sizeof (u64);/* nr */
@@ -1256,7 +1267,16 @@ static inline void perf_sample_save_brstack(struct perf_sample_data *data, size +=sizeof (u64); size += brs->nr *sizeof (struct perf_branch_entry ); +/* +* The extension space is appended after the struct perf_branch_stack. +* It is used to store the extra data of each branch, e.g., +* the occurrences of events since the last branch entry for Intel LBR. +*/ +if (branch_sample_extra (event)) + size += brs->nr *sizeof (u64); + data->br_stack = brs; + data->br_stack_ext = brs_ext; data->dyn_size += size; data->sample_flags |=PERF_SAMPLE_BRANCH_STACK ; }
include/uapi/linux/perf_event.h +9 -0
@@ -204,6 +204,10 @@ enum perf_branch_sample_type_shift {PERF_SAMPLE_BRANCH_PRIV_SAVE_SHIFT =18 ,/* save privilege mode */ +PERF_SAMPLE_BRANCH_EXTRA_SHIFT =19 ,/* support extra space */ + +PERF_SAMPLE_BRANCH_EVT_CNTRS_SHIFT =20 ,/* save occurrences of events on a branch */ + PERF_SAMPLE_BRANCH_MAX_SHIFT /* non-ABI */ };
@@ -235,6 +239,10 @@ enum perf_branch_sample_type {PERF_SAMPLE_BRANCH_PRIV_SAVE =1U <<PERF_SAMPLE_BRANCH_PRIV_SAVE_SHIFT , +PERF_SAMPLE_BRANCH_EXTRA =1U <<PERF_SAMPLE_BRANCH_EXTRA_SHIFT , + +PERF_SAMPLE_BRANCH_EVT_CNTRS =1U <<PERF_SAMPLE_BRANCH_EVT_CNTRS_SHIFT , + PERF_SAMPLE_BRANCH_MAX = 1U << PERF_SAMPLE_BRANCH_MAX_SHIFT, };
@@ -982,6 +990,7 @@ enum perf_event_type {* { u64 nr; * { u64 hw_idx; } && PERF_SAMPLE_BRANCH_HW_INDEX * { u64 from, to, flags } lbr[nr]; +* { u64 extra; } ext[nr] && PERF_SAMPLE_BRANCH_EXTRA * } && PERF_SAMPLE_BRANCH_STACK * * { u64 abi; # enum perf_sample_regs_abi
kernel/events/core.c +8 -0
@@ -7324,6 +7324,14 @@ void perf_output_sample(struct perf_output_handle *handle,if (branch_sample_hw_index (event))perf_output_put (handle, data->br_stack->hw_idx);perf_output_copy (handle, data->br_stack->entries, size); +/* +* Add the extension space which is appended +* right after the struct perf_branch_stack. +*/ +if (branch_sample_extra (event) && data->br_stack_ext) { + size = data->br_stack->nr *sizeof (u64); +perf_output_copy (handle, data->br_stack_ext, size); + } }else {/* * we always store at least the value of nr
Signature · 1 line
2.35.1
PATCH v3 2/6 perf/x86: Add PERF_X86_EVENT_NEEDS_BRANCH_STACK flag
From: Kan Liang <[email protected]>
Currently, branch_sample_type !=0 is used to check whether a branch stack setup is required. But it doesn't check the sample type, unnecessary branch stack setup may be done for a counting event. E.g., perf record -e "{branch-instructions,branch-misses}:S" -j any Also, the event with the new PERF_SAMPLE_BRANCH_EVT_CNTRS branch sample type may not require a branch stack setup either.
Add a new flag NEEDS_BRANCH_STACK to indicate whether the event requires a branch stack setup. Replace the needs_branch_stack() by checking the new flag.
The counting event check is implemented here. The later patch will take the new PERF_SAMPLE_BRANCH_EVT_CNTRS into account.
---
New patch
arch/x86/events/intel/core.c | 14 +++++++++++--- arch/x86/events/perf_event_flags.h | 1 + 2 files changed, 12 insertions(+), 3 deletions(-)
arch/x86/events/intel/core.c +11 -3
@@ -2518,9 +2518,14 @@ static void intel_pmu_assign_event(struct perf_event *event, int idx)perf_report_aux_output_id (event, idx); } +static __always_inline bool intel_pmu_needs_branch_stack(struct perf_event *event) +{ +return event->hw.flags &PERF_X86_EVENT_NEEDS_BRANCH_STACK ; +} +static void intel_pmu_del_event (struct perf_event *event) { -if (needs_branch_stack (event)) +if (intel_pmu_needs_branch_stack (event))intel_pmu_lbr_del (event);if (event->attr.precise_ip)intel_pmu_pebs_del (event);
@@ -2831,7 +2836,7 @@ 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)) +if (intel_pmu_needs_branch_stack (event))intel_pmu_lbr_add (event); }
@@ -3908,7 +3913,10 @@ static int intel_pmu_hw_config(struct perf_event *event) x86_pmu.pebs_aliases (event); } -if (needs_branch_stack (event)) { +if (needs_branch_stack (event) &&is_sampling_event (event)) + event->hw.flags |=PERF_X86_EVENT_NEEDS_BRANCH_STACK ; + +if (intel_pmu_needs_branch_stack (event)) { ret =intel_pmu_setup_lbr_filter (event);if (ret)return ret;
arch/x86/events/perf_event_flags.h +1 -0
@@ -20,3 +20,4 @@ PERF_ARCH(TOPDOWN, 0x04000) /* Count Topdown slots/metrics events */ PERF_ARCH(PEBS_STLAT, 0x08000) /* st+stlat data address sampling */ PERF_ARCH(AMD_BRS, 0x10000) /* AMD Branch Sampling */ PERF_ARCH(PEBS_LAT_HYBRID, 0x20000) /* ld and st lat for hybrid */ +PERF_ARCH (NEEDS_BRANCH_STACK ,0x40000 )/* require branch stack setup */
Signature · 1 line
2.35.1
PATCH v3 3/6 perf: Add branch_sample_call_stack
In reply to [email protected]From: Kan Liang <[email protected]>
Add a helper function to check call stack sample type.
The later patch will invoke the function in several places.
---
New patch
arch/x86/events/core.c | 2 +- include/linux/perf_event.h | 5 +++++ 2 files changed, 6 insertions(+), 1 deletion(-)
arch/x86/events/core.c +1 -1
@@ -601,7 +601,7 @@ int x86_pmu_hw_config(struct perf_event *event) } } -if (event->attr.branch_sample_type &PERF_SAMPLE_BRANCH_CALL_STACK ) +if (branch_sample_call_stack (event)) event->attach_state |=PERF_ATTACH_TASK_DATA ;/*
include/linux/perf_event.h +5 -0
@@ -1147,6 +1147,11 @@ static inline bool branch_sample_evt_cntrs(const struct perf_event *event)return event->attr.branch_sample_type &PERF_SAMPLE_BRANCH_EVT_CNTRS ; } +static inline bool branch_sample_call_stack (const struct perf_event *event) +{ +return event->attr.branch_sample_type &PERF_SAMPLE_BRANCH_CALL_STACK ; +} +struct perf_sample_data {/* * Fields set by perf_sample_data_init() unconditionally,
Signature · 1 line
2.35.1
PATCH v3 4/6 perf/x86/intel: Support LBR event logging
In reply to [email protected]From: Kan Liang <[email protected]>
The LBR event logging introduces a per-counter indication of precise event occurrences in LBRs. It can provide a means to attribute exposed retirement latency to combinations of events across a block of instructions. It also provides a means of attributing Timed LBR latencies to events.
The feature is first introduced on SRF/GRR. It is an enhancement of the ARCH LBR. It adds new fields in the LBR_INFO MSRs to log the occurrences of events on the GP counters. The information is displayed by the order of counters.
The design proposed in this patch requires that the events which are logged must be in a group with the event that has LBR. If there are more than one LBR group, the event logging information only from the current group (overflowed) are stored for the perf tool, otherwise the perf tool cannot know which and when other groups are scheduled especially when multiplexing is triggered. The user can ensure it uses the maximum number of counters that support LBR info (4 by now) by making the group large enough.
The HW only logs events by the order of counters. The order may be different from the order of enabling which the perf tool can understand. When parsing the information of each branch entry, convert the counter order to the enabled order, and store the enabled order in the extension space.
Unconditionally reset LBRs for an LBR event group when it's deleted. The logged events' occurrences information is only valid for the current LBR group. If another LBR group is scheduled later, the information from the stale LBRs would be otherwise wrongly interpreted.
Add a sanity check in intel_pmu_hw_config(). Disable the feature if other counter filters (inv, cmask, edge, in_tx) are set or LBR call stack mode is enabled. (For the LBR call stack mode, we cannot simply flush the LBR, since it will break the call stack. Also, there is no obvious usage with the call stack mode for now.)
Only applying the PERF_SAMPLE_BRANCH_EVT_CNTRS or the PERF_SAMPLE_BRANCH_EXTRA doesn't require any branch stack setup.
---
Changes since V2:
- Drop the SW bit LBR_EVENT_LOG_BIT and flag PERF_X86_EVENT_LBR_EVENT. They can be replaced by branch_sample_evt_cntrs() and branch_sample_extra().
arch/x86/events/intel/core.c | 62 ++++++++++++++++++-- arch/x86/events/intel/ds.c | 2 +- arch/x86/events/intel/lbr.c | 94 ++++++++++++++++++++++++++++++- arch/x86/events/perf_event.h | 7 +++ arch/x86/include/asm/msr-index.h | 2 + arch/x86/include/asm/perf_event.h | 4 ++ 6 files changed, 164 insertions(+), 7 deletions(-)
arch/x86/events/intel/core.c +57 -5
@@ -2803,6 +2803,7 @@ static void intel_pmu_enable_fixed(struct perf_event *event)static void intel_pmu_enable_event (struct perf_event *event) { +u64 enable_mask =ARCH_PERFMON_EVENTSEL_ENABLE ;struct hw_perf_event *hwc = &event->hw;int idx = hwc->idx;
@@ -2811,8 +2812,10 @@ static void intel_pmu_enable_event(struct perf_event *event)switch (idx) { case 0 ... INTEL_PMC_IDX_FIXED - 1: +if (branch_sample_evt_cntrs (event)) + enable_mask |=ARCH_PERFMON_EVENTSEL_LBR_LOG ;intel_set_masks (event, idx); -__x86_pmu_enable_event (hwc,ARCH_PERFMON_EVENTSEL_ENABLE ); +__x86_pmu_enable_event (hwc, enable_mask);break ; case INTEL_PMC_IDX_FIXED ... INTEL_PMC_IDX_FIXED_BTS - 1: case INTEL_PMC_IDX_METRIC_BASE ... INTEL_PMC_IDX_METRIC_END:
@@ -3063,7 +3066,7 @@ static int handle_pmi_common(struct pt_regs *regs, u64 status)perf_sample_data_init (&data,0 , event->hw.last_period);if (has_branch_stack (event)) -perf_sample_save_brstack (&data, event, &cpuc->lbr_stack,NULL ); +intel_pmu_lbr_save_brstack (&data, cpuc, event);if (perf_event_overflow (event, &data, regs))x86_pmu_stop (event,0 );
@@ -3628,6 +3631,13 @@ intel_get_event_constraints(struct cpu_hw_events *cpuc, int idx,if (cpuc->excl_cntrs)return intel_get_excl_constraints (cpuc, event, idx, c2); +/* The LBR event logging may not be available for all counters. */ +if (branch_sample_evt_cntrs (event)) { + c2 =dyn_constraint (cpuc, c2, idx); + c2->idxmsk64 &= x86_pmu.lbr_events; + c2->weight =hweight64 (c2->idxmsk64); + } +return c2; }
@@ -3916,6 +3926,44 @@ static int intel_pmu_hw_config(struct perf_event *event)if (needs_branch_stack (event) &&is_sampling_event (event)) event->hw.flags |=PERF_X86_EVENT_NEEDS_BRANCH_STACK ; +if (branch_sample_evt_cntrs (event) ||branch_sample_extra (event)) { +struct perf_event *leader, *sibling; + +if (!(x86_pmu.flags &PMU_FL_LBR_EVENT ) || + (event->attr.config & ~INTEL_ARCH_EVENT_MASK )) +return -EINVAL ; + +/* +* The event logging is not supported in the call stack mode +* yet, since we cannot simply flush the LBR during e.g., +* multiplexing. Also, there is no obvious usage with the call +* stack mode. Simply forbids it for now. +* +* If any events in the group enable the LBR event logging +* feature, the group is treated as a LBR event logging group, +* which requires the extra space. +*/ + leader = event->group_leader; +if (branch_sample_call_stack (leader) || !branch_sample_extra (leader)) +return -EINVAL ; + +for_each_sibling_event (sibling, leader) { +if (branch_sample_call_stack (sibling) || !branch_sample_extra (leader)) +return -EINVAL ; + } + +/* +* Only applying the PERF_SAMPLE_BRANCH_EVT_CNTRS or the +* PERF_SAMPLE_BRANCH_EXTRA doesn't require any branch stack setup. +* Clear the bit to avoid unnecessary branch stack setup. +*/ +if (0 == (event->attr.branch_sample_type & + ~(PERF_SAMPLE_BRANCH_PLM_ALL | +PERF_SAMPLE_BRANCH_EVT_CNTRS | +PERF_SAMPLE_BRANCH_EXTRA ))) + event->hw.flags &= ~PERF_X86_EVENT_NEEDS_BRANCH_STACK ; + } +if (intel_pmu_needs_branch_stack (event)) { ret =intel_pmu_setup_lbr_filter (event);if (ret)
@@ -4387,8 +4435,12 @@ cmt_get_event_constraints(struct cpu_hw_events *cpuc, int idx,*/ if (event->attr.precise_ip ==3 ) {/* Force instruction:ppp on PMC0, 1 and Fixed counter 0 */ -if (constraint_match (&fixed0_constraint, event->hw.config)) -return &fixed0_counter0_1_constraint; +if (constraint_match (&fixed0_constraint, event->hw.config)) { +if (branch_sample_evt_cntrs (event)) +return &counter0_1_constraint; +else +return &fixed0_counter0_1_constraint; + }switch (c->idxmsk64 &0x3ull ) {case 0x1 :
@@ -4567,7 +4619,7 @@ int intel_cpuc_prepare(struct cpu_hw_events *cpuc, int cpu) goto err; } -if (x86_pmu.flags & (PMU_FL_EXCL_CNTRS |PMU_FL_TFA )) { +if (x86_pmu.flags & (PMU_FL_EXCL_CNTRS |PMU_FL_TFA |PMU_FL_LBR_EVENT )) {size_t sz =X86_PMC_IDX_MAX *sizeof (struct event_constraint ); cpuc->constraint_list =kzalloc_node (sz,GFP_KERNEL ,cpu_to_node (cpu));
arch/x86/events/intel/ds.c +1 -1
@@ -1912,7 +1912,7 @@ static void setup_pebs_adaptive_sample_data(struct perf_event *event,if (has_branch_stack (event)) {intel_pmu_store_pebs_lbrs (lbr); -perf_sample_save_brstack (data, event, &cpuc->lbr_stack,NULL ); +intel_pmu_lbr_save_brstack (data, cpuc, event); } }
arch/x86/events/intel/lbr.c +93 -1
132 lines not on this page · raw message
arch/x86/events/perf_event.h +7 -0
@@ -283,6 +283,7 @@ struct cpu_hw_events {int lbr_pebs_users;struct perf_branch_stack lbr_stack;struct perf_branch_entry lbr_entries[MAX_LBR_ENTRIES ]; +u64 lbr_events[MAX_LBR_ENTRIES ];/* branch stack extra */ union {struct er_account *lbr_sel;struct er_account *lbr_ctl;
@@ -881,6 +882,7 @@ struct x86_pmu { unsigned int lbr_mispred:1; unsigned int lbr_timed_lbr:1; unsigned int lbr_br_type:1; + unsigned int lbr_events:4;void (*lbr_reset)(void );void (*lbr_read)(struct cpu_hw_events *cpuc);
@@ -1005,6 +1007,7 @@ do { \#define PMU_FL_INSTR_LATENCY 0x80/* Support Instruction Latency in PEBS Memory Info Record */ #define PMU_FL_MEM_LOADS_AUX 0x100/* Require an auxiliary event for the complete memory info */ #define PMU_FL_RETIRE_LATENCY 0x200/* Support Retire Latency in PEBS */ +#define PMU_FL_LBR_EVENT 0x400/* Support LBR event logging */ #define EVENT_VAR (_id) event_attr_##_id#define EVENT_PTR (_id) &event_attr_##_id.attr.attr
@@ -1545,6 +1548,10 @@ void intel_pmu_store_pebs_lbrs(struct lbr_entry *lbr);void intel_ds_init (void ); +void intel_pmu_lbr_save_brstack (struct perf_sample_data *data, +struct cpu_hw_events *cpuc, +struct perf_event *event); +void intel_pmu_lbr_swap_task_ctx (struct perf_event_pmu_context *prev_epc,struct perf_event_pmu_context *next_epc);
arch/x86/include/asm/msr-index.h +2 -0
@@ -224,6 +224,8 @@#define LBR_INFO_CYCLES 0xffff#define LBR_INFO_BR_TYPE_OFFSET 56#define LBR_INFO_BR_TYPE (0xfull << LBR_INFO_BR_TYPE_OFFSET) +#define LBR_INFO_EVENTS_OFFSET 32 +#define LBR_INFO_EVENTS (0xffull << LBR_INFO_EVENTS_OFFSET)#define MSR_ARCH_LBR_CTL 0x000014ce#define ARCH_LBR_CTL_LBREN BIT(0)
arch/x86/include/asm/perf_event.h +4 -0
@@ -31,6 +31,7 @@#define ARCH_PERFMON_EVENTSEL_ENABLE (1ULL << 22)#define ARCH_PERFMON_EVENTSEL_INV (1ULL << 23)#define ARCH_PERFMON_EVENTSEL_CMASK 0xFF000000ULL +#define ARCH_PERFMON_EVENTSEL_LBR_LOG (1ULL << 35)#define INTEL_FIXED_BITS_MASK 0xFULL#define INTEL_FIXED_BITS_STRIDE 4
@@ -216,6 +217,9 @@ union cpuid28_ecx { unsigned int lbr_timed_lbr:1;/* Branch Type Field Supported */ unsigned int lbr_br_type:1; + unsigned int reserved:13; +/* Event Logging Supported */ + unsigned int lbr_events:4; } split;unsigned int full; };
Signature · 1 line
2.35.1
PATCH v3 5/6 tools headers UAPI: Sync include/uapi/linux/perf_event.h header with the kernel
In reply to [email protected]From: Kan Liang <[email protected]>
Sync the new sample types and extra space for the branch event feature.
---
Changes since V2:
- Drop the new bit in struct perf_branch_entry
- Introduce a new sample type PERF_SAMPLE_BRANCH_EXTRA
tools/include/uapi/linux/perf_event.h | 9 +++++++++ 1 file changed, 9 insertions(+)
tools/include/uapi/linux/perf_event.h +9 -0
@@ -204,6 +204,10 @@ enum perf_branch_sample_type_shift {PERF_SAMPLE_BRANCH_PRIV_SAVE_SHIFT =18 ,/* save privilege mode */ +PERF_SAMPLE_BRANCH_EXTRA_SHIFT =19 ,/* support extra space */ + +PERF_SAMPLE_BRANCH_EVT_CNTRS_SHIFT =20 ,/* save occurrences of events on a branch */ + PERF_SAMPLE_BRANCH_MAX_SHIFT /* non-ABI */ };
@@ -235,6 +239,10 @@ enum perf_branch_sample_type {PERF_SAMPLE_BRANCH_PRIV_SAVE =1U <<PERF_SAMPLE_BRANCH_PRIV_SAVE_SHIFT , +PERF_SAMPLE_BRANCH_EXTRA =1U <<PERF_SAMPLE_BRANCH_EXTRA_SHIFT , + +PERF_SAMPLE_BRANCH_EVT_CNTRS =1U <<PERF_SAMPLE_BRANCH_EVT_CNTRS_SHIFT , + PERF_SAMPLE_BRANCH_MAX = 1U << PERF_SAMPLE_BRANCH_MAX_SHIFT, };
@@ -982,6 +990,7 @@ enum perf_event_type {* { u64 nr; * { u64 hw_idx; } && PERF_SAMPLE_BRANCH_HW_INDEX * { u64 from, to, flags } lbr[nr]; +* { u64 extra; } ext[nr] && PERF_SAMPLE_BRANCH_EXTRA * } && PERF_SAMPLE_BRANCH_STACK * * { u64 abi; # enum perf_sample_regs_abi
Signature · 1 line
2.35.1
PATCH v3 6/6 perf tools: Add branch event knob
In reply to [email protected]From: Kan Liang <[email protected]>
Add a new branch filter, "event", for the branch event option. It is used to mark the events which should be logged in the branch. If it is applied with the -j option, all the events should be logged in the branch. If the legacy kernel doesn't support the new branch sample type, switching off the branch event filter.
The branch event requires extra space. If the "event" branch filter is detected, the PERF_SAMPLE_BRANCH_EXTRA is applied automatically for the entire group.
The new extra space of each branch is dumped right after the regular branch stack information via perf report -D.
Usage examples:
perf record -e "{branch-instructions,branch-misses}:S" -j any,event
Only the first event, branch-instructions, collect the LBR. Both branch-instructions and branch-misses are marked as logged events. The occurrences information of them can be found in the branch stack extension space of each branch.
perf record -e "{cpu/branch-instructions,branch_type=any/, cpu/branch-misses,branch_type=event/}"
Only the first event, branch-instructions, collect the LBR. Only the branch-misses event is marked as a logged event.
---
Changes since V2:
- Drop the new bit in struct perf_branch_entry
- Support PERF_SAMPLE_BRANCH_EXTRA
tools/perf/Documentation/perf-record.txt | 4 ++++ tools/perf/arch/x86/util/evsel.c | 17 ++++++++++++++- tools/perf/util/evsel.c | 26 ++++++++++++++++++++++- tools/perf/util/evsel.h | 12 +++++++++++ tools/perf/util/parse-branch-options.c | 1 + tools/perf/util/perf_event_attr_fprintf.c | 1 + tools/perf/util/sample.h | 1 + tools/perf/util/session.c | 8 +++++++ 8 files changed, 68 insertions(+), 2 deletions(-)
tools/perf/Documentation/perf-record.txt +4 -0
@@ -456,6 +456,10 @@ following filters are defined: 4th-Gen Xeon+ server), the save branch type is unconditionally enabled when the taken branch stack sampling is enabled. - priv: save privilege state during sampling in case binary is not available later + - event: save occurrences of the event since the last branch entry. Currently, the + feature is only supported by a newer CPU, e.g., Intel Sierra Forest and + later platforms. An error out is expected if it's used on the unsupported + kernel or CPUs. + The option requires at least one branch type among any, any_call, any_ret, ind_call, cond.
tools/perf/arch/x86/util/evsel.c +16 -1
@@ -9,6 +9,7 @@#include "evsel.h" #include "util/debug.h" #include "env.h" +#include "../../../util/evlist.h" #define IBS_FETCH_L3MISSONLY (1ULL << 59)#define IBS_OP_L3MISSONLY (1ULL << 16)
@@ -80,9 +81,23 @@ void arch__post_evsel_config(struct evsel *evsel, struct perf_event_attr *attr)struct perf_pmu *evsel_pmu, *ibs_fetch_pmu, *ibs_op_pmu;static int warned_once; -if (warned_once || !x86__is_amd_cpu ()) +if (warned_once)return ; +if (!x86__is_amd_cpu ()) { +if (evsel__has_branch_evt_cntrs (evsel)) { +struct evsel *cur, *leader =evsel__leader (evsel); + +/* The extra space is required for the LBR event group */ +evlist__for_each_entry (evsel->evlist, cur) { +if (leader ==evsel__leader (cur)) + cur->core.attr.branch_sample_type |=PERF_SAMPLE_BRANCH_EXTRA ; + } + } + +return ; + } + evsel_pmu =evsel__find_pmu (evsel);if (!evsel_pmu)return ;
tools/perf/util/evsel.c +25 -1
@@ -1828,6 +1828,10 @@ static int __evsel__prepare_open(struct evsel *evsel, struct perf_cpu_map *cpus,static void evsel__disable_missing_features (struct evsel *evsel) { +if (perf_missing_features.branch_extra) + evsel->core.attr.branch_sample_type &= ~PERF_SAMPLE_BRANCH_EXTRA ; +if (perf_missing_features.branch_event) + evsel->core.attr.branch_sample_type &= ~PERF_SAMPLE_BRANCH_EVT_CNTRS ;if (perf_missing_features.read_lost) evsel->core.attr.read_format &= ~PERF_FORMAT_LOST ;if (perf_missing_features.weight_struct) {
@@ -1881,7 +1885,17 @@ bool evsel__detect_missing_features(struct evsel *evsel)* Must probe features in the order they were added to the * perf_event_attr interface. */ -if (!perf_missing_features.read_lost && +if (!perf_missing_features.branch_extra && + (evsel->core.attr.branch_sample_type &PERF_SAMPLE_BRANCH_EXTRA )) { + perf_missing_features.branch_extra = true; +pr_debug2 ("switching off branch extra space support\n" ); +return true; + }else if (!perf_missing_features.branch_event && + (evsel->core.attr.branch_sample_type &PERF_SAMPLE_BRANCH_EVT_CNTRS )) { + perf_missing_features.branch_event = true; +pr_debug2 ("switching off branch event support\n" ); +return true; + }else if (!perf_missing_features.read_lost && (evsel->core.attr.read_format &PERF_FORMAT_LOST )) { perf_missing_features.read_lost = true;pr_debug2 ("switching off PERF_FORMAT_LOST support\n" );
@@ -2574,6 +2588,16 @@ int evsel__parse_sample(struct evsel *evsel, union perf_event *event,OVERFLOW_CHECK (array, sz, max_size); array = (void *)array + sz; + +if (evsel__has_branch_extra (evsel)) { +OVERFLOW_CHECK_u64 (array); + + data->branch_stack_ext = (u64 *)array; + sz = data->branch_stack->nr *sizeof (u64); + +OVERFLOW_CHECK (array, sz, max_size); + array = (void *)array + sz; + } }if (type &PERF_SAMPLE_REGS_USER ) {
tools/perf/util/evsel.h +12 -0
@@ -191,6 +191,8 @@ struct perf_missing_features {bool code_page_size;bool weight_struct;bool read_lost; +bool branch_event; +bool branch_extra; };extern struct perf_missing_features perf_missing_features;
@@ -499,6 +501,16 @@ static inline bool evsel__has_branch_hw_idx(const struct evsel *evsel)return evsel->core.attr.branch_sample_type &PERF_SAMPLE_BRANCH_HW_INDEX ; } +static inline bool evsel__has_branch_evt_cntrs (const struct evsel *evsel) +{ +return evsel->core.attr.branch_sample_type &PERF_SAMPLE_BRANCH_EVT_CNTRS ; +} + +static inline bool evsel__has_branch_extra (const struct evsel *evsel) +{ +return evsel->core.attr.branch_sample_type &PERF_SAMPLE_BRANCH_EXTRA ; +} +static inline bool evsel__has_callchain (const struct evsel *evsel) {/*
tools/perf/util/parse-branch-options.c +1 -0
@@ -36,6 +36,7 @@ static const struct branch_mode branch_modes[] = {BRANCH_OPT ("stack" ,PERF_SAMPLE_BRANCH_CALL_STACK ),BRANCH_OPT ("hw_index" ,PERF_SAMPLE_BRANCH_HW_INDEX ),BRANCH_OPT ("priv" ,PERF_SAMPLE_BRANCH_PRIV_SAVE ), +BRANCH_OPT ("event" ,PERF_SAMPLE_BRANCH_EVT_CNTRS ), BRANCH_END };
tools/perf/util/perf_event_attr_fprintf.c +1 -0
@@ -55,6 +55,7 @@ static void __p_branch_sample_type(char *buf, size_t size, u64 value)bit_name (COND ),bit_name (CALL_STACK ),bit_name (IND_JUMP ),bit_name (CALL ),bit_name (NO_FLAGS ),bit_name (NO_CYCLES ),bit_name (TYPE_SAVE ),bit_name (HW_INDEX ),bit_name (PRIV_SAVE ), +bit_name (EVT_CNTRS ),bit_name (EXTRA ), { .name = NULL, } };#undef bit_name
tools/perf/util/sample.h +1 -0
@@ -113,6 +113,7 @@ struct perf_sample {void *raw_data;struct ip_callchain *callchain;struct branch_stack *branch_stack; +u64 *branch_stack_ext;struct regs_dump user_regs;struct regs_dump intr_regs;struct stack_dump user_stack;
tools/perf/util/session.c +8 -0
@@ -1153,6 +1153,7 @@ static void callchain__printf(struct evsel *evsel,static void branch_stack__printf (struct perf_sample *sample,bool callstack) {struct branch_entry *entries =perf_sample__branch_entries (sample); +u64 *branch_stack_ext = sample->branch_stack_ext;uint64_t i;if (!callstack) {
@@ -1194,6 +1195,13 @@ static void branch_stack__printf(struct perf_sample *sample, bool callstack) } } } + +if (branch_stack_ext) { +printf ("... branch stack extra: nr:%" PRIu64"\n" , sample->branch_stack->nr); +for (i =0 ; i < sample->branch_stack->nr; i++) { +printf ("..... %2" PRIu64": %016" PRIx64"\n" , i, branch_stack_ext[i]); + } + } }static void regs_dump__printf (u64 mask,u64 *regs,const char *arch)
Signature · 1 line
2.35.1