perf: Reorganize scripting support
Subsystems (4)
Ian Rogers
reviewerPATCH v9 11/23 perf evlist: Add reference count checking
In reply to Ian RogersNow the evlist is reference counted, add reference count checking so that gets and puts are paired and easy to debug. Reference count checking is documented here:
https://perfwiki.github.io/main/reference-count-checking/
This large patch is adding accessors to evlist functions and switching to their use. There was some minor renaming as evlist__mmap is now an accessor to the mmap variable, and the original evlist__mmap is renamed to evlist__do_mmap.
---
v2:
1. Fixed Memory Leak in evlist__new : Added free(evlist) in the else branch if ADD_RC_CHK fails, preventing a leak of the allocated raw structure.
2. Fixed Potential NULL Dereference: Added a NULL check after from_list_start(_evlist) in perf_evlist__mmap_cb_get() .
3. Fixed Use-After-Free Risk: In evlist__add() , I changed entry->evlist = evlist; to entry->evlist = evlist__get(evlist); . This ensures that the evsel holds a valid reference (wrapper) to the evlist , preventing it from becoming a dangling pointer if the original wrapper is freed.
4. Fixed Test Masking Bug: In test__perf_time__parse_for_ranges() , I replaced TEST_ASSERT_VAL with a manual check and return false; to avoid boolean evaluation of -1 inadvertently passing the test.
5. Fix reference count checker memory leaks from missed puts and due to cyclic evsel to evlist references. A leak still exists in __perf_evlist__propagate_maps due to empty CPU maps and not deleting the removed evsel.
v8:
1. Handle evsel__new failure in sample-parsing.c test and extra ref count checking integrity in evlist.c
2. Use standard per_sample initialization in cs-etm.c.
--- tools/perf/arch/arm/util/cs-etm.c | 10 +- tools/perf/arch/arm64/util/arm-spe.c | 8 +- tools/perf/arch/arm64/util/hisi-ptt.c | 2 +- tools/perf/arch/x86/tests/hybrid.c | 20 +- tools/perf/arch/x86/util/auxtrace.c | 2 +- tools/perf/arch/x86/util/intel-bts.c | 6 +- tools/perf/arch/x86/util/intel-pt.c | 9 +- tools/perf/arch/x86/util/iostat.c | 6 +- tools/perf/bench/evlist-open-close.c | 11 +- tools/perf/builtin-annotate.c | 2 +- tools/perf/builtin-ftrace.c | 6 +- tools/perf/builtin-inject.c | 4 +- tools/perf/builtin-kvm.c | 10 +- tools/perf/builtin-kwork.c | 8 +- tools/perf/builtin-record.c | 91 ++--- tools/perf/builtin-report.c | 6 +- tools/perf/builtin-sched.c | 20 +- tools/perf/builtin-script.c | 13 +- tools/perf/builtin-stat.c | 71 ++-- tools/perf/builtin-top.c | 52 +-- tools/perf/builtin-trace.c | 22 +- tools/perf/tests/backward-ring-buffer.c | 8 +- tools/perf/tests/code-reading.c | 10 +- tools/perf/tests/event-times.c | 2 +- tools/perf/tests/event_update.c | 2 +- tools/perf/tests/expand-cgroup.c | 4 +- tools/perf/tests/hwmon_pmu.c | 5 +- tools/perf/tests/keep-tracking.c | 8 +- tools/perf/tests/mmap-basic.c | 6 +- tools/perf/tests/openat-syscall-tp-fields.c | 8 +- tools/perf/tests/parse-events.c | 135 +++---- tools/perf/tests/parse-metric.c | 4 +- tools/perf/tests/perf-record.c | 20 +- tools/perf/tests/perf-time-to-tsc.c | 10 +- tools/perf/tests/pfm.c | 8 +- tools/perf/tests/pmu-events.c | 5 +- tools/perf/tests/sample-parsing.c | 42 ++- tools/perf/tests/sw-clock.c | 6 +- tools/perf/tests/switch-tracking.c | 8 +- tools/perf/tests/task-exit.c | 6 +- tools/perf/tests/time-utils-test.c | 14 +- tools/perf/tests/tool_pmu.c | 5 +- tools/perf/tests/topology.c | 2 +- tools/perf/ui/browsers/annotate.c | 2 +- tools/perf/ui/browsers/hists.c | 22 +- tools/perf/util/amd-sample-raw.c | 2 +- tools/perf/util/annotate-data.c | 2 +- tools/perf/util/annotate.c | 10 +- tools/perf/util/auxtrace.c | 14 +- tools/perf/util/block-info.c | 4 +- tools/perf/util/bpf_counter.c | 2 +- tools/perf/util/bpf_counter_cgroup.c | 8 +- tools/perf/util/bpf_ftrace.c | 9 +- tools/perf/util/bpf_lock_contention.c | 12 +- tools/perf/util/bpf_off_cpu.c | 14 +- tools/perf/util/cgroup.c | 20 +- tools/perf/util/cs-etm.c | 5 +- tools/perf/util/evlist.c | 391 ++++++++++++-------- tools/perf/util/evlist.h | 251 ++++++++++++- tools/perf/util/evsel.c | 6 +- tools/perf/util/evsel.h | 4 +- tools/perf/util/header.c | 39 +- tools/perf/util/header.h | 2 +- tools/perf/util/intel-tpebs.c | 7 +- tools/perf/util/metricgroup.c | 6 +- tools/perf/util/parse-events.c | 6 +- tools/perf/util/pfm.c | 2 +- tools/perf/util/python.c | 30 +- tools/perf/util/record.c | 9 +- tools/perf/util/sample-raw.c | 4 +- tools/perf/util/session.c | 54 +-- tools/perf/util/sideband_evlist.c | 24 +- tools/perf/util/sort.c | 2 +- tools/perf/util/stat-display.c | 6 +- tools/perf/util/stat-shadow.c | 4 +- tools/perf/util/stat.c | 4 +- tools/perf/util/stream.c | 4 +- tools/perf/util/synthetic-events.c | 11 +- tools/perf/util/time-utils.c | 12 +- tools/perf/util/top.c | 4 +- 80 files changed, 1015 insertions(+), 690 deletions(-)
tools/perf/arch/arm/util/cs-etm.c +5 -5
@@ -201,7 +201,7 @@ static int cs_etm_validate_config(struct perf_pmu *cs_etm_pmu, {unsigned int idx;int err =0 ; -struct perf_cpu_map *event_cpus = evsel->evlist->core.user_requested_cpus; +struct perf_cpu_map *event_cpus =evlist__core (evsel->evlist)->user_requested_cpus;struct perf_cpu_map *intersect_cpus;struct perf_cpu cpu;
@@ -325,7 +325,7 @@ static int cs_etm_recording_options(struct auxtrace_record *itr, container_of(itr, struct cs_etm_recording, itr);struct perf_pmu *cs_etm_pmu = ptr->cs_etm_pmu;struct evsel *evsel, *cs_etm_evsel =NULL ; -struct perf_cpu_map *cpus = evlist->core.user_requested_cpus; +struct perf_cpu_map *cpus =evlist__core (evlist)->user_requested_cpus;bool privileged =perf_event_paranoid_check (-1 );int err =0 ;
@@ -551,7 +551,7 @@ cs_etm_info_priv_size(struct auxtrace_record *itr, {unsigned int idx;int etmv3 =0 , etmv4 =0 , ete =0 ; -struct perf_cpu_map *event_cpus = evlist->core.user_requested_cpus; +struct perf_cpu_map *event_cpus =evlist__core (evlist)->user_requested_cpus;struct perf_cpu_map *intersect_cpus;struct perf_cpu cpu;struct perf_pmu *cs_etm_pmu =cs_etm_get_pmu (itr);
@@ -790,7 +790,7 @@ static int cs_etm_info_fill(struct auxtrace_record *itr,u32 offset;u64 nr_cpu, type;struct perf_cpu_map *cpu_map; -struct perf_cpu_map *event_cpus = session->evlist->core.user_requested_cpus; +struct perf_cpu_map *event_cpus =evlist__core (session->evlist)->user_requested_cpus;struct perf_cpu_map *online_cpus =perf_cpu_map__new_online_cpus ();struct cs_etm_recording *ptr = container_of(itr, struct cs_etm_recording, itr);
@@ -800,7 +800,7 @@ static int cs_etm_info_fill(struct auxtrace_record *itr,if (priv_size !=cs_etm_info_priv_size (itr, session->evlist))return -EINVAL ; -if (!session->evlist->core.nr_mmaps) +if (!evlist__core (session->evlist)->nr_mmaps)return -EINVAL ;/* If the cpu_map has the "any" CPU all online CPUs are involved */
tools/perf/arch/arm64/util/arm-spe.c +4 -4
@@ -60,7 +60,7 @@ static bool arm_spe_is_set_freq(struct evsel *evsel)*/ static struct perf_cpu_map *arm_spe_find_cpus (struct evlist *evlist) { -struct perf_cpu_map *event_cpus = evlist->core.user_requested_cpus; +struct perf_cpu_map *event_cpus =evlist__core (evlist)->user_requested_cpus;struct perf_cpu_map *online_cpus =perf_cpu_map__new_online_cpus ();struct perf_cpu_map *intersect_cpus;
@@ -157,7 +157,7 @@ static int arm_spe_info_fill(struct auxtrace_record *itr,if (priv_size !=arm_spe_info_priv_size (itr, session->evlist))return -EINVAL ; -if (!session->evlist->core.nr_mmaps) +if (!evlist__core (session->evlist)->nr_mmaps)return -EINVAL ; cpu_map =arm_spe_find_cpus (session->evlist);
@@ -363,7 +363,7 @@ static int arm_spe_setup_tracking_event(struct evlist *evlist, {int err;struct evsel *tracking_evsel; -struct perf_cpu_map *cpus = evlist->core.user_requested_cpus; +struct perf_cpu_map *cpus =evlist__core (evlist)->user_requested_cpus;/* Add dummy event to keep tracking */ err =parse_event (evlist,"dummy:u" );
@@ -396,7 +396,7 @@ static int arm_spe_recording_options(struct auxtrace_record *itr,struct arm_spe_recording *sper = container_of(itr, struct arm_spe_recording, itr);struct evsel *evsel, *tmp; -struct perf_cpu_map *cpus = evlist->core.user_requested_cpus; +struct perf_cpu_map *cpus =evlist__core (evlist)->user_requested_cpus;bool discard = false;int err;u64 discard_bit;
tools/perf/arch/arm64/util/hisi-ptt.c +1 -1
@@ -53,7 +53,7 @@ static int hisi_ptt_info_fill(struct auxtrace_record *itr,if (priv_size !=HISI_PTT_AUXTRACE_PRIV_SIZE )return -EINVAL ; -if (!session->evlist->core.nr_mmaps) +if (!evlist__core (session->evlist)->nr_mmaps)return -EINVAL ; auxtrace_info->type =PERF_AUXTRACE_HISI_PTT ;
tools/perf/arch/x86/tests/hybrid.c +10 -10
@@ -26,7 +26,7 @@ static int test__hybrid_hw_event_with_pmu(struct evlist *evlist) {struct evsel *evsel =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,1 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,1 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_HARDWARE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong hybrid type" ,test_hybrid_type (evsel,PERF_TYPE_RAW ));TEST_ASSERT_VAL ("wrong config" ,test_config (evsel,PERF_COUNT_HW_CPU_CYCLES ));
@@ -38,7 +38,7 @@ static int test__hybrid_hw_group_event(struct evlist *evlist)struct evsel *evsel, *leader; evsel = leader =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,2 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,2 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_HARDWARE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong hybrid type" ,test_hybrid_type (evsel,PERF_TYPE_RAW ));TEST_ASSERT_VAL ("wrong config" ,test_config (evsel,PERF_COUNT_HW_CPU_CYCLES ));
@@ -57,7 +57,7 @@ static int test__hybrid_sw_hw_group_event(struct evlist *evlist)struct evsel *evsel, *leader; evsel = leader =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,2 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,2 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_SOFTWARE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong leader" ,evsel__has_leader (evsel, leader));
@@ -74,7 +74,7 @@ static int test__hybrid_hw_sw_group_event(struct evlist *evlist)struct evsel *evsel, *leader; evsel = leader =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,2 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,2 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_HARDWARE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong hybrid type" ,test_hybrid_type (evsel,PERF_TYPE_RAW ));TEST_ASSERT_VAL ("wrong config" ,test_config (evsel,PERF_COUNT_HW_CPU_CYCLES ));
@@ -91,7 +91,7 @@ static int test__hybrid_group_modifier1(struct evlist *evlist)struct evsel *evsel, *leader; evsel = leader =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,2 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,2 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_HARDWARE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong hybrid type" ,test_hybrid_type (evsel,PERF_TYPE_RAW ));TEST_ASSERT_VAL ("wrong config" ,test_config (evsel,PERF_COUNT_HW_CPU_CYCLES ));
@@ -113,7 +113,7 @@ static int test__hybrid_raw1(struct evlist *evlist) {struct perf_evsel *evsel; -perf_evlist__for_each_evsel (&evlist->core, evsel) { +perf_evlist__for_each_evsel (evlist__core (evlist), evsel) {struct perf_pmu *pmu =perf_pmus__find_by_type (evsel->attr.type);TEST_ASSERT_VAL ("missing pmu" , pmu);
@@ -127,7 +127,7 @@ static int test__hybrid_raw2(struct evlist *evlist) {struct evsel *evsel =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,1 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,1 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_RAW == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong config" ,test_config (evsel,0x1a ));return TEST_OK ;
@@ -137,7 +137,7 @@ static int test__hybrid_cache_event(struct evlist *evlist) {struct evsel *evsel =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,1 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,1 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_HW_CACHE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong config" ,0x2 == (evsel->core.attr.config &0xffffffff ));return TEST_OK ;
@@ -148,7 +148,7 @@ static int test__checkevent_pmu(struct evlist *evlist)struct evsel *evsel =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,1 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,1 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_RAW == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong config" ,10 == evsel->core.attr.config);TEST_ASSERT_VAL ("wrong config1" ,1 == evsel->core.attr.config1);
@@ -168,7 +168,7 @@ static int test__hybrid_hw_group_event_2(struct evlist *evlist)struct evsel *evsel, *leader; evsel = leader =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,2 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,2 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_HARDWARE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong hybrid type" ,test_hybrid_type (evsel,PERF_TYPE_RAW ));TEST_ASSERT_VAL ("wrong config" ,test_config (evsel,PERF_COUNT_HW_CPU_CYCLES ));
tools/perf/arch/x86/util/auxtrace.c +1 -1
@@ -55,7 +55,7 @@ struct auxtrace_record *auxtrace_record__init(struct evlist *evlist, int *err) {char buffer[64 ]; -struct perf_cpu cpu =perf_cpu_map__min (evlist->core.all_cpus); +struct perf_cpu cpu =perf_cpu_map__min (evlist__core (evlist)->all_cpus);int ret; *err =0 ;
tools/perf/arch/x86/util/intel-bts.c +3 -3
@@ -79,10 +79,10 @@ static int intel_bts_info_fill(struct auxtrace_record *itr,if (priv_size !=INTEL_BTS_AUXTRACE_PRIV_SIZE )return -EINVAL ; -if (!session->evlist->core.nr_mmaps) +if (!evlist__core (session->evlist)->nr_mmaps)return -EINVAL ; - pc = session->evlist->mmap[0 ].core.base; + pc =evlist__mmap (session->evlist)[0 ].core.base;if (pc) { err =perf_read_tsc_conversion (pc, &tc);if (err) {
@@ -114,7 +114,7 @@ static int intel_bts_recording_options(struct auxtrace_record *itr, container_of(itr, struct intel_bts_recording, itr);struct perf_pmu *intel_bts_pmu = btsr->intel_bts_pmu;struct evsel *evsel, *intel_bts_evsel =NULL ; -const struct perf_cpu_map *cpus = evlist->core.user_requested_cpus; +const struct perf_cpu_map *cpus =evlist__core (evlist)->user_requested_cpus;bool privileged =perf_event_paranoid_check (-1 );if (opts->auxtrace_sample_mode) {
tools/perf/arch/x86/util/intel-pt.c +5 -4
@@ -360,10 +360,10 @@ static int intel_pt_info_fill(struct auxtrace_record *itr, filter =intel_pt_find_filter (session->evlist, ptr->intel_pt_pmu); filter_str_len = filter ?strlen (filter) :0 ; -if (!session->evlist->core.nr_mmaps) +if (!evlist__core (session->evlist)->nr_mmaps)return -EINVAL ; - pc = session->evlist->mmap[0 ].core.base; + pc =evlist__mmap (session->evlist)[0 ].core.base;if (pc) { err =perf_read_tsc_conversion (pc, &tc);if (err) {
@@ -376,7 +376,8 @@ static int intel_pt_info_fill(struct auxtrace_record *itr,ui__warning ("Intel Processor Trace: TSC not available\n" ); } - per_cpu_mmaps = !perf_cpu_map__is_any_cpu_or_is_empty (session->evlist->core.user_requested_cpus); + per_cpu_mmaps = !perf_cpu_map__is_any_cpu_or_is_empty ( +evlist__core (session->evlist)->user_requested_cpus); auxtrace_info->type =PERF_AUXTRACE_INTEL_PT ; auxtrace_info->priv[INTEL_PT_PMU_TYPE ] = intel_pt_pmu->type;
@@ -621,7 +622,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr,struct perf_pmu *intel_pt_pmu = ptr->intel_pt_pmu;bool have_timing_info, need_immediate = false;struct evsel *evsel, *intel_pt_evsel =NULL ; -const struct perf_cpu_map *cpus = evlist->core.user_requested_cpus; +const struct perf_cpu_map *cpus =evlist__core (evlist)->user_requested_cpus;bool privileged =perf_event_paranoid_check (-1 );u64 tsc_bit;int err;
tools/perf/arch/x86/util/iostat.c +3 -3
@@ -334,7 +334,7 @@ static int iostat_event_group(struct evlist *evl,int iostat_prepare (struct evlist *evlist,struct perf_stat_config *config) { -if (evlist->core.nr_entries >0 ) { +if (evlist__nr_entries (evlist) >0 ) {pr_warning ("The -e and -M options are not supported." "All chosen events/metrics will be dropped\n" );evlist__put (evlist);
@@ -400,7 +400,7 @@ void iostat_prefix(struct evlist *evlist,struct perf_stat_config *config, char *prefix, struct timespec *ts) { - struct iio_root_port *rp = evlist->selected->priv; + struct iio_root_port *rp = evlist__selected(evlist)->priv;if (rp) {/*
@@ -463,7 +463,7 @@ void iostat_print_counters(struct evlist *evlist,iostat_prefix (evlist, config, prefix, ts);fprintf (config->output,"%s" , prefix);evlist__for_each_entry (evlist, counter) { - perf_device = evlist->selected->priv; + perf_device =evlist__selected (evlist)->priv;if (perf_device && perf_device != counter->priv) {evlist__set_selected (evlist, counter);iostat_prefix (evlist, config, prefix, ts);
tools/perf/bench/evlist-open-close.c +6 -5
@@ -116,7 +116,7 @@ static int bench__do_evlist_open_close(struct evlist *evlist)return err; } - err =evlist__mmap (evlist, opts.mmap_pages); + err =evlist__do_mmap (evlist, opts.mmap_pages);if (err <0 ) {pr_err ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));return err;
@@ -124,7 +124,7 @@ static int bench__do_evlist_open_close(struct evlist *evlist)evlist__enable (evlist);evlist__disable (evlist); -evlist__munmap (evlist); +evlist__do_munmap (evlist);evlist__close (evlist);return 0 ;
@@ -145,10 +145,11 @@ static int bench_evlist_open_close__run(char *evstr, const char *uid_str)init_stats (&time_stats); -printf (" Number of cpus:\t%d\n" ,perf_cpu_map__nr (evlist->core.user_requested_cpus)); -printf (" Number of threads:\t%d\n" , evlist->core.threads->nr); +printf (" Number of cpus:\t%d\n" , +perf_cpu_map__nr (evlist__core (evlist)->user_requested_cpus)); +printf (" Number of threads:\t%d\n" ,evlist__core (evlist)->threads->nr);printf (" Number of events:\t%d (%d fds)\n" , - evlist->core.nr_entries,evlist__count_evsel_fds (evlist)); +evlist__nr_entries (evlist),evlist__count_evsel_fds (evlist));printf (" Number of iterations:\t%d\n" , iterations);evlist__put (evlist);
tools/perf/builtin-annotate.c +1 -1
@@ -924,7 +924,7 @@ int cmd_annotate(int argc, const char **argv)*/ if ((use_browser ==1 || annotate.use_stdio2) && annotate.has_br_stack) { sort__mode =SORT_MODE__BRANCH ; -if (annotate.session->evlist->nr_br_cntr >0 ) +if (evlist__nr_br_cntr (annotate.session->evlist) >0 ) annotate_opts.show_br_cntr = true; }
tools/perf/builtin-ftrace.c +3 -3
@@ -377,9 +377,9 @@ static int set_tracing_pid(struct perf_ftrace *ftrace)if (target__has_cpu (&ftrace->target))return 0 ; -for (i =0 ; i <perf_thread_map__nr (ftrace->evlist->core.threads); i++) { +for (i =0 ; i <perf_thread_map__nr (evlist__core (ftrace->evlist)->threads); i++) {scnprintf (buf,sizeof (buf),"%d" , -perf_thread_map__pid (ftrace->evlist->core.threads, i)); +perf_thread_map__pid (evlist__core (ftrace->evlist)->threads, i));if (append_tracing_file ("set_ftrace_pid" , buf) <0 )return -1 ; }
@@ -413,7 +413,7 @@ static int set_tracing_cpumask(struct perf_cpu_map *cpumap)static int set_tracing_cpu (struct perf_ftrace *ftrace) { -struct perf_cpu_map *cpumap = ftrace->evlist->core.user_requested_cpus; +struct perf_cpu_map *cpumap =evlist__core (ftrace->evlist)->user_requested_cpus;if (!target__has_cpu (&ftrace->target))return 0 ;
tools/perf/builtin-inject.c +2 -2
@@ -1487,7 +1487,7 @@ static int synthesize_id_index(struct perf_inject *inject, size_t new_cnt)struct perf_session *session = inject->session;struct evlist *evlist = session->evlist;struct machine *machine = &session->machines.host; -size_t from = evlist->core.nr_entries - new_cnt; +size_t from =evlist__nr_entries (evlist) - new_cnt;return __perf_event__synthesize_id_index (&inject->tool, perf_event__repipe, evlist, machine, from);
@@ -2022,7 +2022,7 @@ static int host__finished_init(const struct perf_tool *tool, struct perf_sessionif (ret)return ret; - ret =synthesize_id_index (inject, gs->session->evlist->core.nr_entries); + ret =synthesize_id_index (inject,evlist__nr_entries (gs->session->evlist));if (ret) {pr_err ("Failed to synthesize id_index\n" );return ret;
tools/perf/builtin-kvm.c +5 -5
@@ -1218,7 +1218,7 @@ static s64 perf_kvm__mmap_read_idx(struct perf_kvm_stat *kvm, int idx,int err; *mmap_time =ULLONG_MAX ; - md = &evlist->mmap[idx]; + md = &evlist__mmap (evlist)[idx]; err =perf_mmap__read_init (&md->core);if (err <0 )return (err == -EAGAIN ) ?0 :-1 ;
@@ -1263,7 +1263,7 @@ static int perf_kvm__mmap_read(struct perf_kvm_stat *kvm)s64 n, ntotal =0 ;u64 flush_time =ULLONG_MAX , mmap_time; -for (i =0 ; i < kvm->evlist->core.nr_mmaps; i++) { +for (i =0 ; i <evlist__core (kvm->evlist)->nr_mmaps; i++) { n =perf_kvm__mmap_read_idx (kvm, i, &mmap_time);if (n <0 )return -1 ;
@@ -1446,7 +1446,7 @@ static int kvm_events_live_report(struct perf_kvm_stat *kvm)evlist__enable (kvm->evlist);while (!done) { -struct fdarray *fda = &kvm->evlist->core.pollfd; +struct fdarray *fda = &evlist__core (kvm->evlist)->pollfd;int rc; rc =perf_kvm__mmap_read (kvm);
@@ -1528,7 +1528,7 @@ static int kvm_live_open_events(struct perf_kvm_stat *kvm) goto out; } -if (evlist__mmap (evlist, kvm->opts.mmap_pages) <0 ) { +if (evlist__do_mmap (evlist, kvm->opts.mmap_pages) <0 ) {ui__error ("Failed to mmap the events: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));evlist__close (evlist);
@@ -1928,7 +1928,7 @@ static int kvm_events_live(struct perf_kvm_stat *kvm,perf_session__set_id_hdr_size (kvm->session);ordered_events__set_copy_on_queue (&kvm->session->ordered_events, true);machine__synthesize_threads (&kvm->session->machines.host, &kvm->opts.target, - kvm->evlist->core.threads, true, false,1 ); +evlist__core (kvm->evlist)->threads, true, false,1 ); err =kvm_live_open_events (kvm);if (err) goto out;
tools/perf/builtin-kwork.c +4 -4
@@ -1749,7 +1749,7 @@ static int perf_kwork__check_config(struct perf_kwork *kwork, } } -list_for_each_entry (evsel, &session->evlist->core.entries, core.node) { +list_for_each_entry (evsel, &evlist__core (session->evlist)->entries, core.node) {if (kwork->show_callchain && !evsel__has_callchain (evsel)) {pr_debug ("Samples do not have callchains\n" ); kwork->show_callchain =0 ;
@@ -1799,9 +1799,9 @@ static int perf_kwork__read_events(struct perf_kwork *kwork) goto out_delete; } - kwork->nr_events = session->evlist->stats.nr_events[0 ]; - kwork->nr_lost_events = session->evlist->stats.total_lost; - kwork->nr_lost_chunks = session->evlist->stats.nr_events[PERF_RECORD_LOST ]; + kwork->nr_events =evlist__stats (session->evlist)->nr_events[0 ]; + kwork->nr_lost_events =evlist__stats (session->evlist)->total_lost; + kwork->nr_lost_chunks =evlist__stats (session->evlist)->nr_events[PERF_RECORD_LOST ]; out_delete:perf_session__delete (session);
tools/perf/builtin-record.c +46 -45
268 lines not on this page · raw message
tools/perf/builtin-report.c +3 -3
@@ -557,7 +557,7 @@ static int evlist__tty_browse_hists(struct evlist *evlist, struct report *rep, cif (!quiet) {fprintf (stdout,"#\n# Total Lost Samples: %" PRIu64"\n#\n" , - evlist->stats.total_lost_samples); +evlist__stats (evlist)->total_lost_samples); }evlist__for_each_entry (evlist, pos) {
@@ -1152,7 +1152,7 @@ static int __cmd_report(struct report *rep) PERF_HPP_REPORT__BLOCK_AVG_CYCLES, }; -if (session->evlist->nr_br_cntr >0 ) +if (evlist__nr_br_cntr (session->evlist) >0 ) block_hpps[nr_hpps++] =PERF_HPP_REPORT__BLOCK_BRANCH_COUNTER ; block_hpps[nr_hpps++] =PERF_HPP_REPORT__BLOCK_RANGE ;
@@ -1287,7 +1287,7 @@ static int process_attr(const struct perf_tool *tool __maybe_unused,* on events sample_type. */ sample_type =evlist__combined_sample_type (*pevlist); - session = (*pevlist)->session; + session =evlist__session (*pevlist);callchain_param_setup (sample_type,perf_session__e_machine (session,/*e_flags=*/ NULL ));return 0 ; }
tools/perf/builtin-sched.c +10 -10
@@ -1950,9 +1950,9 @@ static int perf_sched__read_events(struct perf_sched *sched) goto out_delete; } - sched->nr_events = session->evlist->stats.nr_events[0 ]; - sched->nr_lost_events = session->evlist->stats.total_lost; - sched->nr_lost_chunks = session->evlist->stats.nr_events[PERF_RECORD_LOST ]; + sched->nr_events =evlist__stats (session->evlist)->nr_events[0 ]; + sched->nr_lost_events =evlist__stats (session->evlist)->total_lost; + sched->nr_lost_chunks =evlist__stats (session->evlist)->nr_events[PERF_RECORD_LOST ]; } rc =0 ;
@@ -3200,7 +3200,7 @@ static int timehist_check_attr(struct perf_sched *sched,struct evsel *evsel;struct evsel_runtime *er; -list_for_each_entry (evsel, &evlist->core.entries, core.node) { +list_for_each_entry (evsel, &evlist__core (evlist)->entries, core.node) { er =evsel__get_runtime (evsel);if (er ==NULL ) {pr_err ("Failed to allocate memory for evsel runtime data\n" );
@@ -3372,9 +3372,9 @@ static int perf_sched__timehist(struct perf_sched *sched) goto out; } - sched->nr_events = evlist->stats.nr_events[0 ]; - sched->nr_lost_events = evlist->stats.total_lost; - sched->nr_lost_chunks = evlist->stats.nr_events[PERF_RECORD_LOST ]; + sched->nr_events =evlist__stats (evlist)->nr_events[0 ]; + sched->nr_lost_events =evlist__stats (evlist)->total_lost; + sched->nr_lost_chunks =evlist__stats (evlist)->nr_events[PERF_RECORD_LOST ];if (sched->summary)timehist_print_summary (sched, session);
@@ -3881,7 +3881,7 @@ static int perf_sched__schedstat_record(struct perf_sched *sched,if (err <0 ) goto out; - user_requested_cpus = evlist->core.user_requested_cpus; + user_requested_cpus =evlist__core (evlist)->user_requested_cpus; err = perf_event__synthesize_schedstat(&(sched->tool), process_synthesized_schedstat_event,
@@ -4598,7 +4598,7 @@ static int perf_sched__schedstat_report(struct perf_sched *sched)if (err <0 ) goto out; - user_requested_cpus = session->evlist->core.user_requested_cpus; + user_requested_cpus =evlist__core (session->evlist)->user_requested_cpus; err =perf_session__process_events (session);
@@ -4774,7 +4774,7 @@ static int perf_sched__schedstat_live(struct perf_sched *sched,if (err <0 ) goto out; - user_requested_cpus = evlist->core.user_requested_cpus; + user_requested_cpus =evlist__core (evlist)->user_requested_cpus; err = perf_event__synthesize_schedstat(&(sched->tool), process_synthesized_event_live,
tools/perf/builtin-script.c +7 -6
@@ -2225,9 +2225,10 @@ static int script_find_metrics(const struct pmu_metric *pm,evlist__for_each_entry (metric_evlist, metric_evsel) {struct evsel *script_evsel =map_metric_evsel_to_script_evsel (script_evlist, metric_evsel); -struct metric_event *metric_me =metricgroup__lookup (&metric_evlist->metric_events, - metric_evsel, -/*create=*/ false); +struct metric_event *metric_me = +metricgroup__lookup (evlist__metric_events (metric_evlist), + metric_evsel, +/*create=*/ false);if (script_evsel->metric_id ==NULL ) { script_evsel->metric_id = metric_evsel->metric_id;
@@ -2247,7 +2248,7 @@ static int script_find_metrics(const struct pmu_metric *pm,if (metric_me) {struct metric_expr *expr;struct metric_event *script_me = -metricgroup__lookup (&script_evlist->metric_events, +metricgroup__lookup (evlist__metric_events (script_evlist), script_evsel,/*create=*/ true);
@@ -2317,7 +2318,7 @@ static void perf_sample__fprint_metric(struct thread *thread,assert (stat_config.aggr_mode ==AGGR_GLOBAL ); stat_config.aggr_get_id = script_aggr_cpu_id_get; stat_config.aggr_map = -cpu_aggr_map__new (evsel->evlist->core.user_requested_cpus, +cpu_aggr_map__new (evlist__core (evsel->evlist)->user_requested_cpus, aggr_cpu_id__global,/*data=*/ NULL ,/*needs_sort=*/ false); }
@@ -3903,7 +3904,7 @@ static int set_maps(struct perf_script *script)if (WARN_ONCE (script->allocated,"stats double allocation\n" ))return -EINVAL ; -perf_evlist__set_maps (&evlist->core, script->cpus, script->threads); +perf_evlist__set_maps (evlist__core (evlist), script->cpus, script->threads);if (evlist__alloc_stats (&stat_config, evlist,/*alloc_raw=*/ true))return -ENOMEM ;
tools/perf/builtin-stat.c +38 -33
198 lines not on this page · raw message
tools/perf/builtin-top.c +26 -26
175 lines not on this page · raw message
tools/perf/builtin-trace.c +11 -11
@@ -2026,7 +2026,7 @@ static int trace__symbols_init(struct trace *trace, int argc, const char **argv, goto out; err = __machine__synthesize_threads(trace->host, &trace->tool, &trace->opts.target, - evlist->core.threads, trace__tool_process, + evlist__core(evlist)->threads, trace__tool_process, /*needs_mmap=*/callchain_param.enabled && !trace->summary_only, /*mmap_data=*/false,
@@ -4199,7 +4199,7 @@ static int trace__set_filter_pids(struct trace *trace) err =augmented_syscalls__set_filter_pids (trace->filter_pids.nr, trace->filter_pids.entries); } - }else if (perf_thread_map__pid (trace->evlist->core.threads,0 ) ==-1 ) { + }else if (perf_thread_map__pid (evlist__core (trace->evlist)->threads,0 ) ==-1 ) { err =trace__set_filter_loop_pids (trace); }
@@ -4513,7 +4513,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv)fprintf (trace->output,"Couldn't run the workload!\n" ); goto out_put_evlist; } - workload_pid = evlist->workload.pid; + workload_pid =evlist__workload_pid (evlist); } err =evlist__open (evlist);
@@ -4565,7 +4565,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv) goto out_error_apply_filters;if (!trace->summary_only || !trace->summary_bpf) { - err =evlist__mmap (evlist, trace->opts.mmap_pages); + err =evlist__do_mmap (evlist, trace->opts.mmap_pages);if (err <0 ) goto out_error_mmap; }
@@ -4584,8 +4584,8 @@ static int trace__run(struct trace *trace, int argc, const char **argv)if (trace->summary_bpf)trace_start_bpf_summary (); - trace->multiple_threads =perf_thread_map__pid (evlist->core.threads,0 ) ==-1 || -perf_thread_map__nr (evlist->core.threads) >1 || + trace->multiple_threads =perf_thread_map__pid (evlist__core (evlist)->threads,0 ) ==-1 || +perf_thread_map__nr (evlist__core (evlist)->threads) >1 ||evlist__first (evlist)->core.attr.inherit;/*
@@ -4602,11 +4602,11 @@ static int trace__run(struct trace *trace, int argc, const char **argv) again: before = trace->nr_events; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) {union perf_event *event;struct mmap *md; - md = &evlist->mmap[i]; + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;
@@ -5306,7 +5306,7 @@ static int trace__parse_cgroups(const struct option *opt, const char *str, int u {struct trace *trace = opt->value; -if (!list_empty (&trace->evlist->core.entries)) { +if (!list_empty (&evlist__core (trace->evlist)->entries)) {struct option o = { .value = &trace->evlist, };
@@ -5580,7 +5580,7 @@ int cmd_trace(int argc, const char **argv)* .perfconfig trace.add_events, and filter those out. */ if (!trace.trace_syscalls && !trace.trace_pgfaults && - trace.evlist->core.nr_entries ==0 /* Was --events used? */ ) { +evlist__nr_entries (trace.evlist) ==0 /* Was --events used? */ ) { trace.trace_syscalls = true; }/*
@@ -5666,7 +5666,7 @@ int cmd_trace(int argc, const char **argv) symbol_conf.use_callchain = true; } -if (trace.evlist->core.nr_entries >0 ) { +if (evlist__nr_entries (trace.evlist) >0 ) {bool use_btf = false;evlist__set_default_evsel_handler (trace.evlist, trace__event_handler);
tools/perf/tests/backward-ring-buffer.c +4 -4
@@ -34,8 +34,8 @@ static int count_samples(struct evlist *evlist, int *sample_count, {int i; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { -struct mmap *map = &evlist->overwrite_mmap[i]; +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) { +struct mmap *map = &evlist__overwrite_mmap (evlist)[i];union perf_event *event;perf_mmap__read_init (&map->core);
@@ -65,7 +65,7 @@ static int do_test(struct evlist *evlist, int mmap_pages,int err;char sbuf[STRERR_BUFSIZE ]; - err =evlist__mmap (evlist, mmap_pages); + err =evlist__do_mmap (evlist, mmap_pages);if (err <0 ) {pr_debug ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));
@@ -77,7 +77,7 @@ static int do_test(struct evlist *evlist, int mmap_pages,evlist__disable (evlist); err =count_samples (evlist, sample_count, comm_count); -evlist__munmap (evlist); +evlist__do_munmap (evlist);return err; }
tools/perf/tests/code-reading.c +5 -5
@@ -589,8 +589,8 @@ static int process_events(struct machine *machine, struct evlist *evlist,struct mmap *md;int i, ret; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { - md = &evlist->mmap[i]; +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) { + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;
@@ -778,7 +778,7 @@ static int do_test_code_reading(bool try_kcore) goto out_put; } -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads); str = events[evidx];pr_debug ("Parsing event '%s'\n" , str);
@@ -806,7 +806,7 @@ static int do_test_code_reading(bool try_kcore)pr_debug ("perf_evlist__open() failed!\n%s\n" , errbuf); } -perf_evlist__set_maps (&evlist->core,NULL ,NULL ); +perf_evlist__set_maps (evlist__core (evlist),NULL ,NULL );evlist__put (evlist); evlist =NULL ;continue ;
@@ -817,7 +817,7 @@ static int do_test_code_reading(bool try_kcore)if (events[evidx] ==NULL ) goto out_put; - ret =evlist__mmap (evlist,UINT_MAX ); + ret =evlist__do_mmap (evlist,UINT_MAX );if (ret <0 ) {pr_debug ("evlist__mmap failed\n" ); goto out_put;
tools/perf/tests/event-times.c +1 -1
@@ -50,7 +50,7 @@ static int attach__enable_on_exec(struct evlist *evlist)static int detach__enable_on_exec (struct evlist *evlist) { -waitpid (evlist->workload.pid,NULL ,0 ); +waitpid (evlist__workload_pid (evlist),NULL ,0 );return 0 ; }
tools/perf/tests/event_update.c +1 -1
@@ -92,7 +92,7 @@ static int test__event_update(struct test_suite *test __maybe_unused, int subtesTEST_ASSERT_VAL ("failed to allocate ids" , !perf_evsel__alloc_id (&evsel->core,1 ,1 )); -perf_evlist__id_add (&evlist->core, &evsel->core,0 ,0 ,123 ); +perf_evlist__id_add (evlist__core (evlist), &evsel->core,0 ,0 ,123 );free ((char *)evsel->unit); evsel->unit =strdup ("KRAVA" );
tools/perf/tests/expand-cgroup.c +2 -2
@@ -28,7 +28,7 @@ static int test_expand_events(struct evlist *evlist)TEST_ASSERT_VAL ("evlist is empty" , !evlist__empty (evlist)); - nr_events = evlist->core.nr_entries; + nr_events =evlist__nr_entries (evlist); ev_name =calloc (nr_events,sizeof (*ev_name));if (ev_name ==NULL ) {pr_debug ("memory allocation failure\n" );
@@ -54,7 +54,7 @@ static int test_expand_events(struct evlist *evlist) } ret =TEST_FAIL ; -if (evlist->core.nr_entries != nr_events * nr_cgrps) { +if (evlist__nr_entries (evlist) != nr_events * nr_cgrps) {pr_debug ("event count doesn't match\n" ); goto out; }
tools/perf/tests/hwmon_pmu.c +3 -2
@@ -183,9 +183,10 @@ static int do_test(size_t i, bool with_pmu, bool with_alias) } ret =TEST_OK ; -if (with_pmu ? (evlist->core.nr_entries !=1 ) : (evlist->core.nr_entries <1 )) { +if (with_pmu ? (evlist__nr_entries (evlist) !=1 ) + : (evlist__nr_entries (evlist) <1 )) {pr_debug ("FAILED %s:%d Unexpected number of events for '%s' of %d\n" , - __FILE__, __LINE__, str, evlist->core.nr_entries); + __FILE__, __LINE__, str,evlist__nr_entries (evlist)); ret =TEST_FAIL ; goto out; }
tools/perf/tests/keep-tracking.c +4 -4
@@ -37,8 +37,8 @@ static int find_comm(struct evlist *evlist, const char *comm)int i, found; found =0 ; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { - md = &evlist->mmap[i]; +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) { + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;while ((event =perf_mmap__read_event (&md->core)) !=NULL ) {
@@ -87,7 +87,7 @@ static int test__keep_tracking(struct test_suite *test __maybe_unused, int subte evlist =evlist__new ();CHECK_NOT_NULL__ (evlist); -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads);CHECK__ (parse_event (evlist,"dummy:u" ));CHECK__ (parse_event (evlist,"cpu-cycles:u" ));
@@ -106,7 +106,7 @@ static int test__keep_tracking(struct test_suite *test __maybe_unused, int subte goto out_err; } -CHECK__ (evlist__mmap (evlist,UINT_MAX )); +CHECK__ (evlist__do_mmap (evlist,UINT_MAX ));/* * First, test that a 'comm' event can be found when the event is
tools/perf/tests/mmap-basic.c +3 -3
@@ -81,7 +81,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtest goto out_free_cpus; } -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads);for (i =0 ; i < nsyscalls; ++i) {char name[64 ];
@@ -113,7 +113,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtest expected_nr_events[i] =1 +rand () %127 ; } -if (evlist__mmap (evlist,128 ) <0 ) { +if (evlist__do_mmap (evlist,128 ) <0 ) {pr_debug ("failed to mmap events: %d (%s)\n" , errno,str_error_r (errno, sbuf,sizeof (sbuf))); goto out_put_evlist;
@@ -124,7 +124,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtest syscalls[i](); } - md = &evlist->mmap[0 ]; + md = &evlist__mmap (evlist)[0 ];if (perf_mmap__read_init (&md->core) <0 ) goto out_init;
tools/perf/tests/openat-syscall-tp-fields.c +4 -4
@@ -64,7 +64,7 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unusedevsel__config (evsel, &opts,NULL ); -perf_thread_map__set_pid (evlist->core.threads,0 ,getpid ()); +perf_thread_map__set_pid (evlist__core (evlist)->threads,0 ,getpid ()); err =evlist__open (evlist);if (err <0 ) {
@@ -73,7 +73,7 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unused goto out_put_evlist; } - err =evlist__mmap (evlist,UINT_MAX ); + err =evlist__do_mmap (evlist,UINT_MAX );if (err <0 ) {pr_debug ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));
@@ -90,11 +90,11 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unusedwhile (1 ) {int before = nr_events; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) {union perf_event *event;struct mmap *md; - md = &evlist->mmap[i]; + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;
tools/perf/tests/parse-events.c +70 -65
519 lines not on this page · raw message
tools/perf/tests/parse-metric.c +2 -2
@@ -53,7 +53,7 @@ static double compute_single(struct evlist *evlist, const char *name)struct evsel *evsel;evlist__for_each_entry (evlist, evsel) { - me =metricgroup__lookup (&evlist->metric_events, evsel, false); + me =metricgroup__lookup (evlist__metric_events (evlist), evsel, false);if (me !=NULL ) { list_for_each_entry (mexp, &me->head, nd) {if (strcmp (mexp->metric_name, name))
@@ -88,7 +88,7 @@ static int __compute_metric(const char *name, struct value *vals,return -ENOMEM ; } -perf_evlist__set_maps (&evlist->core, cpus,NULL ); +perf_evlist__set_maps (evlist__core (evlist), cpus,NULL );/* Parse the metric into metric_events list. */ pme_test =find_core_metrics_table ("testarch" ,"testcpu" );
tools/perf/tests/perf-record.c +10 -10
@@ -129,7 +129,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtestevsel__set_sample_bit (evsel,TIME );evlist__config (evlist, &opts,NULL ); - err =sched__get_first_possible_cpu (evlist->workload.pid, cpu_mask); + err =sched__get_first_possible_cpu (evlist__workload_pid (evlist), cpu_mask);if (err <0 ) {pr_debug ("sched__get_first_possible_cpu: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));
@@ -142,7 +142,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtest/* * So that we can check perf_sample.cpu on all the samples. */ -if (sched_setaffinity (evlist->workload.pid, cpu_mask_size, cpu_mask) <0 ) { +if (sched_setaffinity (evlist__workload_pid (evlist), cpu_mask_size, cpu_mask) <0 ) {pr_debug ("sched_setaffinity: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));evlist__cancel_workload (evlist);
@@ -166,7 +166,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtest* fds in the same CPU to be injected in the same mmap ring buffer * (using ioctl(PERF_EVENT_IOC_SET_OUTPUT)). */ - err =evlist__mmap (evlist, opts.mmap_pages); + err =evlist__do_mmap (evlist, opts.mmap_pages);if (err <0 ) {pr_debug ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));
@@ -188,11 +188,11 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtestwhile (1 ) {int before = total_events; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) {union perf_event *event;struct mmap *md; - md = &evlist->mmap[i]; + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;
@@ -231,15 +231,15 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtest ++errs; } -if ((pid_t )sample.pid != evlist->workload.pid) { +if ((pid_t )sample.pid !=evlist__workload_pid (evlist)) {pr_debug ("%s with unexpected pid, expected %d, got %d\n" , - name, evlist->workload.pid, sample.pid); + name,evlist__workload_pid (evlist), sample.pid); ++errs; } -if ((pid_t )sample.tid != evlist->workload.pid) { +if ((pid_t )sample.tid !=evlist__workload_pid (evlist)) {pr_debug ("%s with unexpected tid, expected %d, got %d\n" , - name, evlist->workload.pid, sample.tid); + name,evlist__workload_pid (evlist), sample.tid); ++errs; }
@@ -248,7 +248,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtest type == PERF_RECORD_MMAP2 || type == PERF_RECORD_FORK || type == PERF_RECORD_EXIT) && - (pid_t)event->comm.pid != evlist->workload.pid) { + (pid_t)event->comm.pid != evlist__workload_pid(evlist)) {pr_debug ("%s with unexpected pid/tid\n" , name); ++errs; }
tools/perf/tests/perf-time-to-tsc.c +5 -5
@@ -99,7 +99,7 @@ static int test__perf_time_to_tsc(struct test_suite *test __maybe_unused, int su evlist =evlist__new ();CHECK_NOT_NULL__ (evlist); -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads);CHECK__ (parse_event (evlist,"cpu-cycles:u" ));
@@ -121,9 +121,9 @@ static int test__perf_time_to_tsc(struct test_suite *test __maybe_unused, int su goto out_err; } -CHECK__ (evlist__mmap (evlist,UINT_MAX )); +CHECK__ (evlist__do_mmap (evlist,UINT_MAX )); - pc = evlist->mmap[0 ].core.base; + pc =evlist__mmap (evlist)[0 ].core.base; ret =perf_read_tsc_conversion (pc, &tc);if (ret) {if (ret == -EOPNOTSUPP ) {
@@ -145,8 +145,8 @@ static int test__perf_time_to_tsc(struct test_suite *test __maybe_unused, int suevlist__disable (evlist); -for (i =0 ; i < evlist->core.nr_mmaps; i++) { - md = &evlist->mmap[i]; +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) { + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;
tools/perf/tests/pfm.c +4 -4
@@ -69,12 +69,12 @@ static int test__pfm_events(struct test_suite *test __maybe_unused,if (evlist ==NULL )return -ENOMEM ; - opt.value = evlist; + opt.value = &evlist;parse_libpfm_events_option (&opt, table[i].events,0 );TEST_ASSERT_EQUAL (table[i].events, -count_pfm_events (&evlist->core), +count_pfm_events (evlist__core (evlist)), table[i].nr_events); TEST_ASSERT_EQUAL(table[i].events, evlist__nr_groups(evlist),
@@ -154,12 +154,12 @@ static int test__pfm_group(struct test_suite *test __maybe_unused,if (evlist ==NULL )return -ENOMEM ; - opt.value = evlist; + opt.value = &evlist;parse_libpfm_events_option (&opt, table[i].events,0 );TEST_ASSERT_EQUAL (table[i].events, -count_pfm_events (&evlist->core), +count_pfm_events (evlist__core (evlist)), table[i].nr_events); TEST_ASSERT_EQUAL(table[i].events, evlist__nr_groups(evlist),
tools/perf/tests/pmu-events.c +3 -2
@@ -869,7 +869,7 @@ static int test__parsing_callback(const struct pmu_metric *pm,return -ENOMEM ; } -perf_evlist__set_maps (&evlist->core, cpus,NULL ); +perf_evlist__set_maps (evlist__core (evlist), cpus,NULL ); err =metricgroup__parse_groups_test (evlist, table, pm->metric_name);if (err) {
@@ -895,7 +895,8 @@ static int test__parsing_callback(const struct pmu_metric *pm, k++; }evlist__for_each_entry (evlist, evsel) { -struct metric_event *me =metricgroup__lookup (&evlist->metric_events, evsel, false); +struct metric_event *me =metricgroup__lookup (evlist__metric_events (evlist), + evsel, false);if (me !=NULL ) {struct metric_expr *mexp;
tools/perf/tests/sample-parsing.c +22 -20
105 lines not on this page · raw message
tools/perf/tests/sw-clock.c +3 -3
@@ -71,7 +71,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id) goto out_put_evlist; } -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads);if (evlist__open (evlist)) {const char *knob ="/proc/sys/kernel/perf_event_max_sample_rate" ;
@@ -83,7 +83,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id) goto out_put_evlist; } - err =evlist__mmap (evlist,128 ); + err =evlist__do_mmap (evlist,128 );if (err <0 ) {pr_debug ("failed to mmap event: %d (%s)\n" , errno,str_error_r (errno, sbuf,sizeof (sbuf)));
@@ -98,7 +98,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id)evlist__disable (evlist); - md = &evlist->mmap[0 ]; + md = &evlist__mmap (evlist)[0 ];if (perf_mmap__read_init (&md->core) <0 ) goto out_init;
tools/perf/tests/switch-tracking.c +4 -4
@@ -282,8 +282,8 @@ static int process_events(struct evlist *evlist,struct mmap *md;int i, ret; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { - md = &evlist->mmap[i]; +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) { + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;
@@ -374,7 +374,7 @@ static int test__switch_tracking(struct test_suite *test __maybe_unused, int sub goto out_err; } -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads);/* First event */ err =parse_event (evlist,"cpu-clock:u" );
@@ -471,7 +471,7 @@ static int test__switch_tracking(struct test_suite *test __maybe_unused, int sub goto out; } - err =evlist__mmap (evlist,UINT_MAX ); + err =evlist__do_mmap (evlist,UINT_MAX );if (err) {pr_debug ("evlist__mmap failed!\n" ); goto out_err;
tools/perf/tests/task-exit.c +3 -3
@@ -77,7 +77,7 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _ goto out_put_evlist; } -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads); err =evlist__prepare_workload (evlist, &target, argv, false, workload_exec_failed_signal);if (err <0 ) {
@@ -104,7 +104,7 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _ goto out_put_evlist; } -if (evlist__mmap (evlist,128 ) <0 ) { +if (evlist__do_mmap (evlist,128 ) <0 ) {pr_debug ("failed to mmap events: %d (%s)\n" , errno,str_error_r (errno, sbuf,sizeof (sbuf))); err =-1 ;
@@ -114,7 +114,7 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _evlist__start_workload (evlist); retry: - md = &evlist->mmap[0 ]; + md = &evlist__mmap (evlist)[0 ];if (perf_mmap__read_init (&md->core) <0 ) goto out_init;
tools/perf/tests/time-utils-test.c +9 -5
@@ -69,16 +69,19 @@ struct test_data {static bool test__perf_time__parse_for_ranges (struct test_data *d) { -struct evlist evlist = { - .first_sample_time = d->first, - .last_sample_time = d->last, - }; -struct perf_session session = { .evlist = &evlist }; +struct evlist *evlist =evlist__new (); +struct perf_session session = { .evlist = evlist };struct perf_time_interval *ptime =NULL ;int range_size, range_num;bool pass = false;int i, err; +if (!evlist) { +pr_debug ("Missing evlist\n" ); +return false; + } +evlist__set_first_sample_time (evlist, d->first); +evlist__set_last_sample_time (evlist, d->last);pr_debug ("\nperf_time__parse_for_ranges(\"%s\")\n" , d->str);if (strchr (d->str,'%' ))
@@ -127,6 +130,7 @@ static bool test__perf_time__parse_for_ranges(struct test_data *d) pass = true; out: +evlist__put (evlist);free (ptime);return pass; }
tools/perf/tests/tool_pmu.c +3 -2
@@ -40,9 +40,10 @@ static int do_test(enum tool_pmu_event ev, bool with_pmu) } ret =TEST_OK ; -if (with_pmu ? (evlist->core.nr_entries !=1 ) : (evlist->core.nr_entries <1 )) { +if (with_pmu ? (evlist__nr_entries (evlist) !=1 ) + : (evlist__nr_entries (evlist) <1 )) {pr_debug ("FAILED %s:%d Unexpected number of events for '%s' of %d\n" , - __FILE__, __LINE__, str, evlist->core.nr_entries); + __FILE__, __LINE__, str,evlist__nr_entries (evlist)); ret =TEST_FAIL ; goto out; }
tools/perf/tests/topology.c +1 -1
@@ -45,7 +45,7 @@ static int session_write_header(char *path) session->evlist =evlist__new_default (&target,/*sample_callchains=*/ false);TEST_ASSERT_VAL ("can't get evlist" , session->evlist); - session->evlist->session = session; +evlist__set_session (session->evlist, session);perf_header__set_feat (&session->header,HEADER_CPU_TOPOLOGY );perf_header__set_feat (&session->header,HEADER_NRCPUS );
tools/perf/ui/browsers/annotate.c +1 -1
@@ -594,7 +594,7 @@ static bool annotate_browser__callq(struct annotate_browser *browser, notes =symbol__annotation (dl->ops.target.sym);annotation__lock (notes); -if (!symbol__hists (dl->ops.target.sym, evsel->evlist->core.nr_entries)) { +if (!symbol__hists (dl->ops.target.sym,evlist__nr_entries (evsel->evlist))) {annotation__unlock (notes);ui__warning ("Not enough memory for annotating '%s' symbol!\n" , dl->ops.target.sym->name);
tools/perf/ui/browsers/hists.c +11 -11
@@ -688,10 +688,10 @@ static int hist_browser__handle_hotkey(struct hist_browser *browser, bool warn_lui_browser__update_nr_entries (&browser->b, nr_entries);if (warn_lost_event && - (evsel->evlist->stats.nr_lost_warned != - evsel->evlist->stats.nr_events[PERF_RECORD_LOST ])) { - evsel->evlist->stats.nr_lost_warned = - evsel->evlist->stats.nr_events[PERF_RECORD_LOST ]; + (evlist__stats (evsel->evlist)->nr_lost_warned != +evlist__stats (evsel->evlist)->nr_events[PERF_RECORD_LOST ])) { +evlist__stats (evsel->evlist)->nr_lost_warned = +evlist__stats (evsel->evlist)->nr_events[PERF_RECORD_LOST ];ui_browser__warn_lost_events (&browser->b); }
@@ -3321,7 +3321,7 @@ static int evsel__hists_browse(struct evsel *evsel, int nr_events, const char *h* No need to refresh, resort/decay histogram * entries if we are not collecting samples: */ -if (top->evlist->enabled) { +if (evlist__enabled (top->evlist)) { helpline ="Press 'f' to disable the events or 'h' to see other hotkeys" ; hbt->refresh = delay_secs; }else {
@@ -3493,7 +3493,7 @@ static void perf_evsel_menu__write(struct ui_browser *browser, unit, unit == ' ' ? "" : " ", ev_name);ui_browser__printf (browser,"%s" , bf); - nr_events = evsel->evlist->stats.nr_events[PERF_RECORD_LOST ]; + nr_events =evlist__stats (evsel->evlist)->nr_events[PERF_RECORD_LOST ];if (nr_events !=0 ) { menu->lost_events = true;if (!current_entry)
@@ -3559,13 +3559,13 @@ static int perf_evsel_menu__run(struct evsel_menu *menu,ui_browser__show_title (&menu->b, title);switch (key) {case K_TAB : -if (pos->core.node.next == &evlist->core.entries) +if (pos->core.node.next == &evlist__core (evlist)->entries) pos =evlist__first (evlist);else pos =evsel__next (pos); goto browse_hists;case K_UNTAB : -if (pos->core.node.prev == &evlist->core.entries) +if (pos->core.node.prev == &evlist__core (evlist)->entries) pos =evlist__last (evlist);else pos =evsel__prev (pos);
@@ -3618,7 +3618,7 @@ static int __evlist__tui_browse_hists(struct evlist *evlist, int nr_entries, construct evsel *pos;struct evsel_menu menu = { .b = { - .entries = &evlist->core.entries, + .entries = &evlist__core (evlist)->entries, .refresh = ui_browser__list_head_refresh, .seek = ui_browser__list_head_seek, .write = perf_evsel_menu__write,
@@ -3646,7 +3646,7 @@ static int __evlist__tui_browse_hists(struct evlist *evlist, int nr_entries, constatic bool evlist__single_entry (struct evlist *evlist) { -int nr_entries = evlist->core.nr_entries; +int nr_entries =evlist__nr_entries (evlist);if (nr_entries ==1 )return true;
@@ -3664,7 +3664,7 @@ static bool evlist__single_entry(struct evlist *evlist)int evlist__tui_browse_hists (struct evlist *evlist,const char *help,struct hist_browser_timer *hbt,float min_pcnt,struct perf_env *env,bool warn_lost_event) { -int nr_entries = evlist->core.nr_entries; +int nr_entries =evlist__nr_entries (evlist);if (evlist__single_entry (evlist)) { single_entry: {
tools/perf/util/amd-sample-raw.c +1 -1
@@ -354,7 +354,7 @@ static void parse_cpuid(struct perf_env *env)*/ bool evlist__has_amd_ibs (struct evlist *evlist) { -struct perf_env *env =perf_session__env (evlist->session); +struct perf_env *env =perf_session__env (evlist__session (evlist));int ret, nr_pmu_mappings =perf_env__nr_pmu_mappings (env);const char *pmu_mapping =perf_env__pmu_mappings (env);char name[sizeof ("ibs_fetch" )];
tools/perf/util/annotate-data.c +1 -1
@@ -1829,7 +1829,7 @@ int annotated_data_type__update_samples(struct annotated_data_type *adt,return 0 ;if (adt->histograms ==NULL ) { -int nr = evsel->evlist->core.nr_entries; +int nr =evlist__nr_entries (evsel->evlist);if (alloc_data_type_histograms (adt, nr) <0 )return -1 ;
tools/perf/util/annotate.c +5 -5
@@ -327,7 +327,7 @@ static int symbol__inc_addr_samples(struct map_symbol *ms,if (sym ==NULL )return 0 ; - src =symbol__hists (sym, sample->evsel->evlist->core.nr_entries); + src =symbol__hists (sym,evlist__nr_entries (sample->evsel->evlist));return src ?__symbol__inc_addr_samples (ms, src, addr, sample) :0 ; }
@@ -338,7 +338,7 @@ static int symbol__account_br_cntr(struct annotated_branch *branch, {unsigned int br_cntr_nr =evsel__leader (evsel)->br_cntr_nr;unsigned int base =evsel__leader (evsel)->br_cntr_idx; -unsigned int off = offset * evsel->evlist->nr_br_cntr; +unsigned int off = offset *evlist__nr_br_cntr (evsel->evlist);u64 *branch_br_cntr = branch->br_cntr;unsigned int i, mask, width;
@@ -368,7 +368,7 @@ static int symbol__account_cycles(u64 addr, u64 start, struct symbol *sym,if (sym ==NULL )return 0 ; - branch =symbol__find_branch_hist (sym, evsel->evlist->nr_br_cntr); + branch =symbol__find_branch_hist (sym,evlist__nr_br_cntr (evsel->evlist));if (!branch)return -ENOMEM ;if (addr < sym->start || addr >= sym->end)
@@ -510,7 +510,7 @@ static void annotation__count_and_fill(struct annotation *notes, u64 start, u64static int annotation__compute_ipc (struct annotation *notes,size_t size,struct evsel *evsel) { -unsigned int br_cntr_nr = evsel->evlist->nr_br_cntr; +unsigned int br_cntr_nr =evlist__nr_br_cntr (evsel->evlist);int err =0 ;s64 offset;
@@ -1812,7 +1812,7 @@ int annotation_br_cntr_abbr_list(char **str, struct evsel *evsel, bool header)struct evsel *pos;struct strbuf sb; -if (evsel->evlist->nr_br_cntr <=0 ) +if (evlist__nr_br_cntr (evsel->evlist) <=0 )return -ENOTSUP ;strbuf_init (&sb,/*hint=*/ 0 );
tools/perf/util/auxtrace.c +7 -7
@@ -191,7 +191,7 @@ void auxtrace_mmap_params__set_idx(struct auxtrace_mmap_params *mp,struct evlist *evlist, struct evsel *evsel, int idx) { - bool per_cpu = !perf_cpu_map__has_any_cpu(evlist->core.user_requested_cpus); + bool per_cpu = !perf_cpu_map__has_any_cpu(evlist__core(evlist)->user_requested_cpus); mp->mmap_needed = evsel->needs_auxtrace_mmap;
@@ -201,11 +201,11 @@ void auxtrace_mmap_params__set_idx(struct auxtrace_mmap_params *mp, mp->idx = idx;if (per_cpu) { - mp->cpu =perf_cpu_map__cpu (evlist->core.all_cpus, idx); - mp->tid =perf_thread_map__pid (evlist->core.threads,0 ); + mp->cpu =perf_cpu_map__cpu (evlist__core (evlist)->all_cpus, idx); + mp->tid =perf_thread_map__pid (evlist__core (evlist)->threads,0 ); }else { mp->cpu.cpu =-1 ; - mp->tid =perf_thread_map__pid (evlist->core.threads, idx); + mp->tid =perf_thread_map__pid (evlist__core (evlist)->threads, idx); } }
@@ -667,10 +667,10 @@ int auxtrace_parse_snapshot_options(struct auxtrace_record *itr,static int evlist__enable_event_idx (struct evlist *evlist,struct evsel *evsel,int idx) { -bool per_cpu_mmaps = !perf_cpu_map__has_any_cpu (evlist->core.user_requested_cpus); +bool per_cpu_mmaps = !perf_cpu_map__has_any_cpu (evlist__core (evlist)->user_requested_cpus);if (per_cpu_mmaps) { -struct perf_cpu evlist_cpu =perf_cpu_map__cpu (evlist->core.all_cpus, idx); +struct perf_cpu evlist_cpu =perf_cpu_map__cpu (evlist__core (evlist)->all_cpus, idx);int cpu_map_idx =perf_cpu_map__idx (evsel->core.cpus, evlist_cpu);if (cpu_map_idx ==-1 )
@@ -1806,7 +1806,7 @@ void perf_session__auxtrace_error_inc(struct perf_session *session,struct perf_record_auxtrace_error *e = &event->auxtrace_error;if (e->type <PERF_AUXTRACE_ERROR_MAX ) - session->evlist->stats.nr_auxtrace_errors[e->type] +=1 ; +evlist__stats (session->evlist)->nr_auxtrace_errors[e->type] +=1 ; }void events_stats__auxtrace_error_warn (const struct events_stats *stats)
tools/perf/util/block-info.c +2 -2
@@ -472,7 +472,7 @@ struct block_report *block_info__create_report(struct evlist *evlist, int *nr_reps) {struct block_report *block_reports; -int nr_hists = evlist->core.nr_entries, i =0 ; +int nr_hists =evlist__nr_entries (evlist), i =0 ;struct evsel *pos; block_reports =calloc (nr_hists,sizeof (struct block_report ));
@@ -483,7 +483,7 @@ struct block_report *block_info__create_report(struct evlist *evlist,struct hists *hists =evsel__hists (pos);process_block_report (hists, &block_reports[i], total_cycles, - block_hpps, nr_hpps, evlist->nr_br_cntr); + block_hpps, nr_hpps,evlist__nr_br_cntr (evlist)); i++; }
tools/perf/util/bpf_counter.c +1 -1
@@ -443,7 +443,7 @@ static int bperf_check_target(struct evsel *evsel, }else if (target->tid) { *filter_type =BPERF_FILTER_PID ; *filter_entry_cnt =perf_thread_map__nr (evsel->core.threads); - }else if (target->pid || evsel->evlist->workload.pid !=-1 ) { + }else if (target->pid ||evlist__workload_pid (evsel->evlist) !=-1 ) { *filter_type =BPERF_FILTER_TGID ; *filter_entry_cnt =perf_thread_map__nr (evsel->core.threads); }else {
tools/perf/util/bpf_counter_cgroup.c +4 -4
@@ -111,7 +111,7 @@ static int bperf_load_program(struct evlist *evlist)pr_err ("Failed to open cgroup skeleton\n" );return -1 ; } -setup_rodata (skel, evlist->core.nr_entries); +setup_rodata (skel,evlist__nr_entries (evlist)); err =bperf_cgroup_bpf__load (skel);if (err) {
@@ -122,12 +122,12 @@ static int bperf_load_program(struct evlist *evlist) err =-1 ; cgrp_switch =evsel__new (&cgrp_switch_attr); -if (evsel__open_per_cpu (cgrp_switch, evlist->core.all_cpus,-1 ) <0 ) { +if (evsel__open_per_cpu (cgrp_switch,evlist__core (evlist)->all_cpus,-1 ) <0 ) {pr_err ("Failed to open cgroup switches event\n" ); goto out; } -perf_cpu_map__for_each_cpu (cpu, i, evlist->core.all_cpus) { +perf_cpu_map__for_each_cpu (cpu, i,evlist__core (evlist)->all_cpus) { link =bpf_program__attach_perf_event (skel->progs.on_cgrp_switch,FD (cgrp_switch, i));if (IS_ERR (link)) {
@@ -238,7 +238,7 @@ static int bperf_cgrp__sync_counters(struct evlist *evlist)unsigned int idx;int prog_fd =bpf_program__fd (skel->progs.trigger_read); -perf_cpu_map__for_each_cpu (cpu, idx, evlist->core.all_cpus) +perf_cpu_map__for_each_cpu (cpu, idx,evlist__core (evlist)->all_cpus)bperf_trigger_reading (prog_fd, cpu.cpu);return 0 ;
tools/perf/util/bpf_ftrace.c +5 -4
@@ -59,13 +59,13 @@ int perf_ftrace__latency_prepare_bpf(struct perf_ftrace *ftrace)/* don't need to set cpu filter for system-wide mode */ if (ftrace->target.cpu_list) { - ncpus =perf_cpu_map__nr (ftrace->evlist->core.user_requested_cpus); + ncpus =perf_cpu_map__nr (evlist__core (ftrace->evlist)->user_requested_cpus);bpf_map__set_max_entries (skel->maps.cpu_filter, ncpus); skel->rodata->has_cpu =1 ; }if (target__has_task (&ftrace->target) ||target__none (&ftrace->target)) { - ntasks =perf_thread_map__nr (ftrace->evlist->core.threads); + ntasks =perf_thread_map__nr (evlist__core (ftrace->evlist)->threads);bpf_map__set_max_entries (skel->maps.task_filter, ntasks); skel->rodata->has_task =1 ; }
@@ -87,7 +87,8 @@ int perf_ftrace__latency_prepare_bpf(struct perf_ftrace *ftrace) fd =bpf_map__fd (skel->maps.cpu_filter);for (i =0 ; i < ncpus; i++) { - cpu =perf_cpu_map__cpu (ftrace->evlist->core.user_requested_cpus, i).cpu; + cpu =perf_cpu_map__cpu ( +evlist__core (ftrace->evlist)->user_requested_cpus, i).cpu;bpf_map_update_elem (fd, &cpu, &val,BPF_ANY ); } }
@@ -99,7 +100,7 @@ int perf_ftrace__latency_prepare_bpf(struct perf_ftrace *ftrace) fd =bpf_map__fd (skel->maps.task_filter);for (i =0 ; i < ntasks; i++) { - pid =perf_thread_map__pid (ftrace->evlist->core.threads, i); + pid =perf_thread_map__pid (evlist__core (ftrace->evlist)->threads, i);bpf_map_update_elem (fd, &pid, &val,BPF_ANY ); } }
tools/perf/util/bpf_lock_contention.c +6 -6
@@ -222,11 +222,11 @@ int lock_contention_prepare(struct lock_contention *con)if (target__has_cpu (target)) { skel->rodata->has_cpu =1 ; - ncpus =perf_cpu_map__nr (evlist->core.user_requested_cpus); + ncpus =perf_cpu_map__nr (evlist__core (evlist)->user_requested_cpus); }if (target__has_task (target)) { skel->rodata->has_task =1 ; - ntasks =perf_thread_map__nr (evlist->core.threads); + ntasks =perf_thread_map__nr (evlist__core (evlist)->threads); }if (con->filters->nr_types) { skel->rodata->has_type =1 ;
@@ -327,7 +327,7 @@ int lock_contention_prepare(struct lock_contention *con) fd =bpf_map__fd (skel->maps.cpu_filter);for (i =0 ; i < ncpus; i++) { - cpu =perf_cpu_map__cpu (evlist->core.user_requested_cpus, i).cpu; + cpu =perf_cpu_map__cpu (evlist__core (evlist)->user_requested_cpus, i).cpu;bpf_map_update_elem (fd, &cpu, &val,BPF_ANY ); } }
@@ -339,13 +339,13 @@ int lock_contention_prepare(struct lock_contention *con) fd =bpf_map__fd (skel->maps.task_filter);for (i =0 ; i < ntasks; i++) { - pid =perf_thread_map__pid (evlist->core.threads, i); + pid =perf_thread_map__pid (evlist__core (evlist)->threads, i);bpf_map_update_elem (fd, &pid, &val,BPF_ANY ); } } -if (target__none (target) && evlist->workload.pid >0 ) { -u32 pid = evlist->workload.pid; +if (target__none (target) &&evlist__workload_pid (evlist) >0 ) { +u32 pid =evlist__workload_pid (evlist);u8 val =1 ; fd =bpf_map__fd (skel->maps.task_filter);
tools/perf/util/bpf_off_cpu.c +7 -7
@@ -73,13 +73,13 @@ static void off_cpu_start(void *arg)/* update task filter for the given workload */ if (skel->rodata->has_task && skel->rodata->uses_tgid && -perf_thread_map__pid (evlist->core.threads,0 ) !=-1 ) { +perf_thread_map__pid (evlist__core (evlist)->threads,0 ) !=-1 ) {int fd;u32 pid;u8 val =1 ; fd =bpf_map__fd (skel->maps.task_filter); - pid =perf_thread_map__pid (evlist->core.threads,0 ); + pid =perf_thread_map__pid (evlist__core (evlist)->threads,0 );bpf_map_update_elem (fd, &pid, &val,BPF_ANY ); }
@@ -168,7 +168,7 @@ int off_cpu_prepare(struct evlist *evlist, struct target *target,/* don't need to set cpu filter for system-wide mode */ if (target->cpu_list) { - ncpus =perf_cpu_map__nr (evlist->core.user_requested_cpus); + ncpus =perf_cpu_map__nr (evlist__core (evlist)->user_requested_cpus);bpf_map__set_max_entries (skel->maps.cpu_filter, ncpus); skel->rodata->has_cpu =1 ; }
@@ -199,7 +199,7 @@ int off_cpu_prepare(struct evlist *evlist, struct target *target, skel->rodata->has_task =1 ; skel->rodata->uses_tgid =1 ; }else if (target__has_task (target)) { - ntasks =perf_thread_map__nr (evlist->core.threads); + ntasks =perf_thread_map__nr (evlist__core (evlist)->threads);bpf_map__set_max_entries (skel->maps.task_filter, ntasks); skel->rodata->has_task =1 ; }else if (target__none (target)) {
@@ -209,7 +209,7 @@ int off_cpu_prepare(struct evlist *evlist, struct target *target, }if (evlist__first (evlist)->cgrp) { - ncgrps = evlist->core.nr_entries -1 ;/* excluding a dummy */ + ncgrps =evlist__nr_entries (evlist) -1 ;/* excluding a dummy */ bpf_map__set_max_entries (skel->maps.cgroup_filter, ncgrps);if (!cgroup_is_v2 ("perf_event" ))
@@ -240,7 +240,7 @@ int off_cpu_prepare(struct evlist *evlist, struct target *target, fd =bpf_map__fd (skel->maps.cpu_filter);for (i =0 ; i < ncpus; i++) { - cpu =perf_cpu_map__cpu (evlist->core.user_requested_cpus, i).cpu; + cpu =perf_cpu_map__cpu (evlist__core (evlist)->user_requested_cpus, i).cpu;bpf_map_update_elem (fd, &cpu, &val,BPF_ANY ); } }
@@ -269,7 +269,7 @@ int off_cpu_prepare(struct evlist *evlist, struct target *target, fd =bpf_map__fd (skel->maps.task_filter);for (i =0 ; i < ntasks; i++) { - pid =perf_thread_map__pid (evlist->core.threads, i); + pid =perf_thread_map__pid (evlist__core (evlist)->threads, i);bpf_map_update_elem (fd, &pid, &val,BPF_ANY ); } }
tools/perf/util/cgroup.c +10 -10
@@ -367,7 +367,7 @@ int parse_cgroups(const struct option *opt, const char *str,char *s;int ret, i; -if (list_empty (&evlist->core.entries)) { +if (list_empty (&evlist__core (evlist)->entries)) {fprintf (stderr,"must define events before cgroups\n" );return -1 ; }
@@ -423,7 +423,7 @@ int evlist__expand_cgroup(struct evlist *evlist, const char *str, bool open_cgroint ret =-1 ;int prefix_len; -if (evlist->core.nr_entries ==0 ) { +if (evlist__nr_entries (evlist) ==0 ) {fprintf (stderr,"must define events before cgroups\n" );return -EINVAL ; }
@@ -436,11 +436,11 @@ int evlist__expand_cgroup(struct evlist *evlist, const char *str, bool open_cgro }/* save original events and init evlist */ -evlist__splice_list_tail (orig_list, &evlist->core.entries); - evlist->core.nr_entries =0 ; +evlist__splice_list_tail (orig_list, &evlist__core (evlist)->entries); +evlist__core (evlist)->nr_entries =0 ; - orig_metric_events = evlist->metric_events; -metricgroup__rblist_init (&evlist->metric_events); + orig_metric_events = *evlist__metric_events (evlist); +metricgroup__rblist_init (evlist__metric_events (evlist));if (has_pattern_string (str)) prefix_len =match_cgroups (str);
@@ -503,15 +503,15 @@ int evlist__expand_cgroup(struct evlist *evlist, const char *str, bool open_cgro nr_cgroups++;if (metricgroup__copy_metric_events (tmp_list, cgrp, - &evlist->metric_events, +evlist__metric_events (evlist), &orig_metric_events) <0 ) goto out_err; -evlist__splice_list_tail (evlist, &tmp_list->core.entries); - tmp_list->core.nr_entries =0 ; +evlist__splice_list_tail (evlist, &evlist__core (tmp_list)->entries); +evlist__core (tmp_list)->nr_entries =0 ; } -if (list_empty (&evlist->core.entries)) { +if (list_empty (&evlist__core (evlist)->entries)) {fprintf (stderr,"no cgroup matched: %s\n" , str); goto out_err; }
tools/perf/util/cs-etm.c +4 -1
@@ -1636,8 +1636,10 @@ static int cs_etm__synth_branch_sample(struct cs_etm_queue *etmq, {int ret =0 ;struct cs_etm_auxtrace *etm = etmq->etm; -struct perf_sample sample = {.ip =0 ,}; +struct perf_sample sample;union perf_event *event = tidq->event_buf; + +perf_sample__init (&sample,/*all=*/ true);struct dummy_branch_stack {u64 nr;u64 hw_idx;
@@ -1698,6 +1700,7 @@ static int cs_etm__synth_branch_sample(struct cs_etm_queue *etmq,"CS ETM Trace: failed to deliver instruction event, error %d\n" , ret); +perf_sample__exit (&sample);return ret; }
tools/perf/util/evlist.c +229 -162
282 lines not on this page · raw message
tools/perf/util/evlist.h +235 -16
251 lines not on this page · raw message
tools/perf/util/session.c +30 -24
54 lines not on this page · raw message
The rest of this diff is not shown. View the raw message.
Signature · 1 line
2.54.0.794.g4f17f83d09-goog
Part of the diff is not shown here. View the raw message.
[email protected]
AI reviewRe: [PATCH v9 11/23] perf evlist: Add reference count checking
Thank you for your contribution! Sashiko AI review found 5 potential issue(s) to consider:
Thank you for your contribution! Sashiko AI review found 5 potential issue(s) to consider:
New issues:
- [High] Asymmetric reference counting of sample->evsel causes memory leaks and premature frees when processing synthesized events or deferred samples.
- [High] Concurrent calls to evlist__put can result in a double-free or Use-After-Free of the evlist object due to a TOCTOU race.
- [Low] Missing cleanup of dynamically allocated evsel on malloc failure in do_test().
- [Low] Mixed declaration and statement violates C89 standard in cs_etm__synth_branch_sample.
Pre-existing issues:
- [High] Reassigning the passed evlist pointer by value in iostat_prepare() causes a Use-After-Free in the caller. --
commit f7fbc61f05122c32f095347ca87933eb1cf01ec1
Author: Ian Rogers <[email protected]> Date: Fri May 22 22:04:22 2026 +0000
perf evlist: Add reference count checking
Now the evlist is reference counted, add reference count checking so that gets and puts are paired and easy to debug. This patch is adding accessors to evlist functions and switching to their use.
Quoted 13 lines
tools/perf/arch/x86/util/iostat.c +1
-1@@ -334,7 +334,7 @@ static int iostat_event_group(struct evlist *evl,int iostat_prepare (struct evlist *evlist,struct perf_stat_config *config) { -if (evlist->core.nr_entries >0 ) { +if (evlist__nr_entries (evlist) >0 ) {pr_warning ("The -e and -M options are not supported." "All chosen events/metrics will be dropped\n" );evlist__put (evlist);
Quoted 1 line
evlist =
evlist__new();
[Severity: High]
This is a pre-existing issue, but doesn't reassigning the evlist parameter by value here cause a use-after-free in the caller? Since the caller in cmd_stat retains its original pointer after iostat_prepare returns, it seems it would point to freed memory.
Quoted 4 lines
tools/perf/tests/sample-parsing.c +0
-0
[ ... ]
tools/perf/tests/sample-parsing.c
Quoted 12 lines
tools/perf/tests/sample-parsing.c +1
-1@@ -311,7 +312,7 @@ static int do_test(u64 sample_type, u64 sample_regs, u64 read_format) } sz =perf_event__sample_event_size (&sample, sample_type, read_format, - evsel.core.attr.branch_sample_type); + evsel->core.attr.branch_sample_type); bufsz = sz +4096 ;/* Add a bit for overrun checking */ event =malloc (bufsz);if (!event) {tools/perf/tests/sample-parsing.c +0
-0pr_debug ("malloc failed\n" );return -1 ; }
[Severity: Low]
Does this leak the dynamically allocated evsel when malloc fails? Because the function returns -1 directly here, it bypasses the out_free label and evsel__put, losing the reference to the newly created evsel.
Quoted 6 lines
tools/perf/util/cs-etm.c +0
-0@@ -1636,8 +1636,10 @@ static int cs_etm__synth_branch_sample(struct cs_etm_queue *etmq, {
tools/perf/util/cs-etm.c static int cs_etm__synth_branch_sample(struct cs_etm_queue *etmq, · line 1645
Quoted 10 lines
tools/perf/util/cs-etm.c +3
-1int ret =0 ;struct cs_etm_auxtrace *etm = etmq->etm; -struct perf_sample sample = {.ip =0 ,}; +struct perf_sample sample;union perf_event *event = tidq->event_buf; + +perf_sample__init (&sample,/*all=*/ true);struct dummy_branch_stack {u64 nr;u64 hw_idx;
[Severity: Low]
Could this cause a build failure on compilers enforcing C89 standards? The function call to perf_sample__init is placed before the declaration of the dummy_branch_stack struct, creating a mixed declaration and statement.
Quoted 4 lines
tools/perf/util/evlist.c +0
-0
[ ... ]
Quoted 24 lines
tools/perf/util/evlist.c +15
-2@@ -208,18 +215,35 @@ static void evlist__exit(struct evlist *evlist)void evlist__put (struct evlist *evlist) { +struct evsel *evsel; +unsigned int count, ref_cnt; +if (evlist ==NULL )return ; -if (!refcount_dec_and_test (&evlist->refcnt)) -return ; +if (refcount_dec_and_test (evlist__refcnt (evlist))) + goto out_delete; + +retry: + count =refcount_read (evlist__refcnt (evlist)); + ref_cnt = count; +evlist__for_each_entry (evlist, evsel) { +if (RC_CHK_EQUAL (evsel->evlist, evlist) && count) + count--; + } +if (refcount_read (evlist__refcnt (evlist)) != ref_cnt) + goto retry;
tools/perf/util/evlist.c static void evlist__exit(struct evlist *evlist) · line 245
Quoted 10 lines
tools/perf/util/evlist.c +9
-0+if (count !=0 ) { +/* +* Not the last reference except for back references from +* evsels. +*/ +RC_CHK_PUT (evlist); +return ; + } +out_delete:
[Severity: High]
Is this custom cyclic reference checking algorithm thread-safe? If two threads concurrently drop an external reference via evlist__put, could one thread reach 0 and proceed to out_delete to free the object, while the other thread evaluates count == 0 after subtracting cyclic references and jumps to out_delete as well? This TOCTOU race seems like it could result in a double-free or use-after-free.
Quoted 4 lines
tools/perf/util/session.c +0
-0
[ ... ]
Quoted 9 lines
tools/perf/util/session.c +2
-0@@ -1400,6 +1402,8 @@ static int evlist__deliver_deferred_callchain(struct evlist *evlist, orig_sample.deferred_callchain = false; orig_sample.evsel =evlist__id2evsel (evlist, orig_sample.id); +if (orig_sample.evsel) + orig_sample.evsel =evsel__get (orig_sample.evsel); ret =evlist__deliver_sample (evlist, tool, de->event, &orig_sample, machine);
[ ... ]
Quoted 9 lines
tools/perf/util/session.c +2
-0@@ -1438,6 +1442,8 @@ static int session__flush_deferred_samples(struct perf_session *session, } sample.evsel =evlist__id2evsel (evlist, sample.id); +if (sample.evsel) + sample.evsel =evsel__get (sample.evsel); ret =evlist__deliver_sample (evlist, tool, de->event, &sample, machine);
[ ... ]
tools/perf/util/session.c static int machines__deliver_event(struct machines *machines, · line 1476
Quoted 11 lines
tools/perf/util/session.c +4
-2@@ -1465,21 +1469,23 @@ static int machines__deliver_event(struct machines *machines,dump_event (evlist, event, file_offset, sample, file_path); -if (!sample->evsel) +if (!sample->evsel) { sample->evsel =evlist__id2evsel (evlist, sample->id); -else +if (sample->evsel) + sample->evsel =evsel__get (sample->evsel); + }else {
[Severity: High]
Can the asymmetric reference counting of sample->evsel cause a memory leak or premature free? It appears that evsel__get is called in these new code blocks when resolving the evsel by ID, but perf_sample__exit does not call evsel__put. This looks like it would result in a permanent memory leak.
Additionally, if the evsel is populated via evlist__parse_sample it does not hold a reference, creating an ownership inconsistency that could lead to a use-after-free if an explicit put is called elsewhere expecting the get to have occurred.
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=11
Ian Rogers
reviewerPATCH v9 12/23 perf python: Use evsel in sample in pyrf_event
In reply to Ian RogersAvoid a duplicated evsel by using the one in sample. Add evsel__get/put to the evsel in perf_sample.
--- tools/perf/util/evsel.c | 2 +- tools/perf/util/python.c | 10 +++------- tools/perf/util/sample.c | 17 ++++++++++++----- 3 files changed, 16 insertions(+), 13 deletions(-)
tools/perf/util/evsel.c +1 -1
@@ -3244,7 +3244,7 @@ int evsel__parse_sample(struct evsel *evsel, union perf_event *event,union u64_swap u;perf_sample__init (data,/*all=*/ true); - data->evsel = evsel; + data->evsel =evsel__get (evsel); data->cpu = data->pid = data->tid =-1 ; data->stream_id = data->id = data->time =-1ULL ; data->period = evsel->core.attr.sample_period;
tools/perf/util/python.c +3 -7
@@ -45,7 +45,6 @@ PyMODINIT_FUNC PyInit_perf(void);struct pyrf_event {PyObject_HEAD - struct evsel *evsel; struct perf_sample sample;union perf_event event; };
@@ -276,7 +275,6 @@ static PyMemberDef pyrf_sample_event__members[] = {static void pyrf_sample_event__delete (struct pyrf_event *pevent) { -evsel__put (pevent->evsel);perf_sample__exit (&pevent->sample);Py_TYPE (pevent)->tp_free ((PyObject *)pevent); }
@@ -298,7 +296,7 @@ static PyObject *pyrf_sample_event__repr(const struct pyrf_event *pevent)#ifdef HAVE_LIBTRACEEVENT static bool is_tracepoint (const struct pyrf_event *pevent) { -return pevent->evsel->core.attr.type ==PERF_TYPE_TRACEPOINT ; +return pevent->sample.evsel->core.attr.type ==PERF_TYPE_TRACEPOINT ; } static PyObject*
@@ -345,7 +343,7 @@ tracepoint_field(const struct pyrf_event *pe, struct tep_format_field *field)static PyObject *get_tracepoint_field (struct pyrf_event *pevent,PyObject *attr_name) { -struct evsel *evsel = pevent->evsel; +struct evsel *evsel = pevent->sample.evsel;struct tep_event *tp_format =evsel__tp_format (evsel);struct tep_format_field *field;
@@ -511,7 +509,7 @@ static PyObject *pyrf_event__new(const union perf_event *event) pevent = PyObject_New(struct pyrf_event, ptype);if (pevent !=NULL ) {memcpy (&pevent->event, event, event->header.size); - pevent->evsel =NULL ; +perf_sample__init (&pevent->sample,/*all=*/ false); }return (PyObject *)pevent; }
@@ -1802,8 +1800,6 @@ static PyObject *pyrf_evlist__read_on_cpu(struct pyrf_evlist *pevlist,return Py_None; } - pevent->evsel =evsel__get (evsel); -perf_mmap__consume (&md->core); err =evsel__parse_sample (evsel, &pevent->event, &pevent->sample);
tools/perf/util/sample.c +12 -5
@@ -1,18 +1,23 @@/* SPDX-License-Identifier: GPL-2.0 */ #include "sample.h" + +#include <stdlib.h> +#include <string.h> + +#include <elf.h> +#include <linux/zalloc.h> + +#include "../../arch/x86/include/asm/insn.h" #include "debug.h" +#include "evsel.h" #include "thread.h" -#include <elf.h> +#ifndef EM_CSKY #define EM_CSKY 252#endif #ifndef EM_LOONGARCH #define EM_LOONGARCH 258#endif -#include <linux/zalloc.h> -#include <stdlib.h> -#include <string.h> -#include "../../arch/x86/include/asm/insn.h" void perf_sample__init (struct perf_sample *sample,bool all) {
@@ -29,6 +34,8 @@ void perf_sample__init(struct perf_sample *sample, bool all)void perf_sample__exit (struct perf_sample *sample) { +evsel__put (sample->evsel); + sample->evsel =NULL ;zfree (&sample->user_regs);zfree (&sample->intr_regs);if (sample->merged_callchain) {
Signature · 1 line
2.54.0.794.g4f17f83d09-goog
Ian Rogers
reviewerPATCH v9 13/23 perf python: Add wrapper for perf_data file abstraction
In reply to Ian RogersThe perf_data struct is needed for session supported.
---
v2:
1. Fixed Memory & FD Leaks in pyrf_data__init : Added cleanup of old state (closing file and freeing path) if __init__ is called multiple times on the same object.
2. Fixed Invalid Free in pyrf_data__delete : Ensured pdata->data.path is always dynamically allocated via strdup() , even for the default "perf.data" . This avoids passing a static string literal to free().
3. Fixed NULL Pointer Dereference in pyrf_data__str : Added a check for NULL path to prevent segfaults if str() or repr() is called on an uninitialized object.
4. Guarded fd Argument Usage: Added a check to ensure that if an fd is provided, it corresponds to a pipe, failing gracefully with a ValueError otherwise.
v7:
- Added pyrf_data__new to zero-initialize pdata->data to fix crash on re-initialization.
--- tools/perf/util/python.c | 103 ++++++++++++++++++++++++++++++++++++++- 1 file changed, 102 insertions(+), 1 deletion(-)
tools/perf/util/python.c +102 -1
@@ -12,6 +12,7 @@#include "callchain.h" #include "counts.h" +#include "data.h" #include "event.h" #include "evlist.h" #include "evsel.h"
@@ -2320,6 +2321,102 @@ static PyObject *pyrf__metrics(PyObject *self, PyObject *args)return list; } +struct pyrf_data { +PyObject_HEAD + + struct perf_data data; +}; + +static int pyrf_data__init (struct pyrf_data *pdata,PyObject *args,PyObject *kwargs) +{ +static char *kwlist[] = {"path" ,"fd" ,NULL }; +char *path =NULL ; +int fd =-1 ; + +if (!PyArg_ParseTupleAndKeywords (args, kwargs,"|si" , kwlist, &path, &fd)) +return -1 ; + +if (pdata->data.open) +perf_data__close (&pdata->data); +free ((char *)pdata->data.path); + +if (!path) + path ="perf.data" ; + + pdata->data.path =strdup (path); +if (!pdata->data.path) { +PyErr_NoMemory (); +return -1 ; + } + +if (fd !=-1 ) { +struct stat st; + +if (fstat (fd, &st) <0 || !S_ISFIFO (st.st_mode)) { +PyErr_SetString (PyExc_ValueError, +"fd argument is only supported for pipes" ); +free ((char *)pdata->data.path); + pdata->data.path =NULL ; +return -1 ; + } + } + + pdata->data.mode =PERF_DATA_MODE_READ ; + pdata->data.file.fd = fd; +if (perf_data__open (&pdata->data) <0 ) { +PyErr_Format (PyExc_IOError,"Failed to open perf data: %s" , + pdata->data.path ? pdata->data.path :"perf.data" ); +return -1 ; + } +return 0 ; +} + +static void pyrf_data__delete (struct pyrf_data *pdata) +{ +perf_data__close (&pdata->data); +free ((char *)pdata->data.path); +Py_TYPE (pdata)->tp_free ((PyObject *)pdata); +} + +static PyObject *pyrf_data__str (PyObject *self) +{ +const struct pyrf_data *pdata = (const struct pyrf_data *)self; + +if (!pdata->data.path) +return PyUnicode_FromString ("[uninitialized]" ); +return PyUnicode_FromString (pdata->data.path); +} + +static PyObject *pyrf_data__new (PyTypeObject *type,PyObject *args,PyObject *kwargs) +{ +struct pyrf_data *pdata; + + pdata = (struct pyrf_data *)PyType_GenericNew (type, args, kwargs); +if (pdata) +memset (&pdata->data,0 ,sizeof (pdata->data)); +return (PyObject *)pdata; +} + +static const char pyrf_data__doc[] =PyDoc_STR ("perf data file object." ); + +static PyTypeObject pyrf_data__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.data" , + .tp_basicsize =sizeof (struct pyrf_data ), + .tp_dealloc = (destructor )pyrf_data__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_data__doc, + .tp_init = (initproc )pyrf_data__init, + .tp_repr = pyrf_data__str, + .tp_str = pyrf_data__str, +}; + +static int pyrf_data__setup_types (void ) +{ + pyrf_data__type.tp_new = pyrf_data__new; +return PyType_Ready (&pyrf_data__type); +} +static PyMethodDef perf__methods[] = { { .ml_name ="metrics" ,
@@ -2382,7 +2479,8 @@ PyMODINIT_FUNC PyInit_perf(void)pyrf_cpu_map__setup_types () <0 ||pyrf_pmu_iterator__setup_types () <0 ||pyrf_pmu__setup_types () <0 || - pyrf_counts_values__setup_types() < 0) +pyrf_counts_values__setup_types () <0 || + pyrf_data__setup_types() < 0) return module;/* The page_size is placed in util object. */
@@ -2430,6 +2528,9 @@ PyMODINIT_FUNC PyInit_perf(void)Py_INCREF (&pyrf_counts_values__type);PyModule_AddObject (module,"counts_values" , (PyObject *)&pyrf_counts_values__type); +Py_INCREF (&pyrf_data__type); +PyModule_AddObject (module,"data" , (PyObject *)&pyrf_data__type); + dict =PyModule_GetDict (module);if (dict ==NULL ) goto error;
Signature · 1 line
2.54.0.794.g4f17f83d09-goog
[email protected]
AI reviewRe: [PATCH v9 13/23] perf python: Add wrapper for perf_data file abstraction
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider: - [High] User-provided file descriptor is silently ignored, overwritten, and leaked unless `path` is…
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [High] User-provided file descriptor is silently ignored, overwritten, and leaked unless path is explicitly set to "-".
- [High] Python wrapper inadvertently closes standard streams or user-managed file descriptors upon garbage collection. --
commit 0c13b6fbaf09ddda5964c9a789eb34aaeae8f382
Author: Ian Rogers <[email protected]>
perf python: Add wrapper for perf_data file abstraction
This commit adds a python wrapper for the perf_data file abstraction required to support session handling.
Quoted 4 lines
tools/perf/util/python.c +0
-0
[ ... ]
Quoted 37 lines
tools/perf/util/python.c +33
-0@@ -2320,6 +2321,102 @@ static PyObject *pyrf__metrics(PyObject *self, PyObject *args)return list; } +struct pyrf_data { +PyObject_HEAD + + struct perf_data data; +}; + +static int pyrf_data__init (struct pyrf_data *pdata,PyObject *args,PyObject *kwargs) +{ +static char *kwlist[] = {"path" ,"fd" ,NULL }; +char *path =NULL ; +int fd =-1 ; + +if (!PyArg_ParseTupleAndKeywords (args, kwargs,"|si" , kwlist, &path, &fd)) +return -1 ; + +if (pdata->data.open) +perf_data__close (&pdata->data); +free ((char *)pdata->data.path); + +if (!path) + path ="perf.data" ; + + pdata->data.path =strdup (path); +if (!pdata->data.path) { +PyErr_NoMemory (); +return -1 ; + } + +if (fd !=-1 ) { +struct stat st; + +if (fstat (fd, &st) <0 || !S_ISFIFO (st.st_mode)) { + PyErr_SetString(PyExc_ValueError,
tools/perf/util/python.c static PyObject *pyrf__metrics(PyObject *self, PyObject *args) · line 2366
Quoted 10 lines
tools/perf/util/python.c +10
-0+ "fd argument is only supported for pipes"); +free ((char *)pdata->data.path); + pdata->data.path =NULL ; +return -1 ; + } + } + + pdata->data.mode =PERF_DATA_MODE_READ ; + pdata->data.file.fd = fd; +if (perf_data__open (&pdata->data) <0 ) {
[Severity: High]
Does this code silently ignore the user-provided fd unless the path is explicitly set to "-"?
If fd is provided but path is not, path defaults to "perf.data". When perf_data__open() is called, it checks if it's a pipe using check_pipe(), which only recognizes a pipe if data->path is "-".
Because it does not recognize the pipe, it will open the local "perf.data" file instead and unconditionally overwrite data->file.fd, dropping the provided fd.
tools/perf/util/python.c static PyObject *pyrf__metrics(PyObject *self, PyObject *args) · line 2376
Quoted 10 lines
tools/perf/util/python.c +10
-0+PyErr_Format (PyExc_IOError,"Failed to open perf data: %s" , + pdata->data.path ? pdata->data.path :"perf.data" ); +return -1 ; + } +return 0 ; +} + +static void pyrf_data__delete (struct pyrf_data *pdata) +{ +perf_data__close (&pdata->data);
[Severity: High]
Does this inadvertently close standard streams or user-managed file descriptors when the Python object is garbage collected?
Since perf_data__close() unconditionally closes data->file.fd, if the user initialized the object to read from standard input (e.g. path="-"), this will call close(0), abruptly closing standard input for the entire Python host process. There doesn't appear to be a way to prevent this behavior.
Quoted 3 lines
tools/perf/util/python.c +3
-0+free ((char *)pdata->data.path); +Py_TYPE (pdata)->tp_free ((PyObject *)pdata); +}
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=13
Ian Rogers
reviewerPATCH v9 14/23 perf python: Add python session abstraction wrapping perf's session
In reply to Ian RogersSessions are necessary to be able to use perf.data files within a tool. Add a wrapper python type that incorporates the tool. Allow a sample callback to be passed when creating the session. When process_events is run this callback will be called, if supplied, for sample events.
An example use looks like:
```
$ perf record -e cycles,instructions -a sleep 3
$ PYTHONPATH=..../perf/python python3
Python 3.13.7 (main, Aug 20 2025, 22:17:40) [GCC 14.2.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import perf
>>> count=0
... def handle_sample(x):
... global count
... if count < 3:
... print(dir(x))
... count = count + 1
... perf.session(perf.data("perf.data"),sample=handle_sample).process_events()
...
['__class__', '__delattr__', '__dir__', '__doc__', '__eq__', '__format__', '__ge__', '__getattribute__', '__getstate__', '__gt__', '__hash__', '__init__', '__init_subclass__', '__le__', '__lt__', '__ne__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', 'sample_addr', 'sample_cpu', 'sample_id', 'sample_ip', 'sample_period', 'sample_pid', 'sample_stream_id', 'sample_tid', 'sample_time', 'type']
['__class__', '__delattr__', '__dir__', '__doc__', '__eq__', '__format__', '__ge__', '__getattribute__', '__getstate__', '__gt__', '__hash__', '__init__', '__init_subclass__', '__le__', '__lt__', '__ne__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', 'sample_addr', 'sample_cpu', 'sample_id', 'sample_ip', 'sample_period', 'sample_pid', 'sample_stream_id', 'sample_tid', 'sample_time', 'type']
['__class__', '__delattr__', '__dir__', '__doc__', '__eq__', '__format__', '__ge__', '__getattribute__', '__getstate__', '__gt__', '__hash__', '__init__', '__init_subclass__', '__le__', '__lt__', '__ne__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', 'sample_addr', 'sample_cpu', 'sample_id', 'sample_ip', 'sample_period', 'sample_pid', 'sample_stream_id', 'sample_tid', 'sample_time', 'type']
```Also, add the ability to get the thread associated with a session. For threads, allow the comm string to be retrieved. This can be useful for filtering threads. Connect up some of the standard event handling in psession->tool to better support queries of the machine. Also connect up the symbols.
---
v2:
1. Fixed Potential Crash in pyrf_thread__comm : Used thread__comm_str() to safely retrieve the command name, avoiding a crash if thread__comm() returns NULL.
2. Fixed Double Free Risk: Zeroed out user_regs , intr_regs , and callchain in the shallow copy of perf_sample to prevent Python from attempting to free pointers it doesn't own.
3. Fixed Memory Leak & Exception Handling in Callback: Handled the return value of PyObject_CallFunction() to avoid leaks, and checked for failure to abort the loop and propagate Python exceptions cleanly.
4. Enforced Type Safety: Used O! with &pyrf_data__type in PyArg_ParseTupleAndKeywords to prevent bad casts from passing arbitrary objects as perf.data.
5. Added Missing Build ID Handler: Registered perf_event__process_build_id to allow correct symbol resolution.
6. Fixed Double Free Crash on Init Failure: Set session and pdata to NULL on failure to prevent tp_dealloc from double-freeing them.
7. Preserved C-level Errors: Made pyrf_session__process_events return the error code integer rather than always returning None .
v7:
- Fixed NULL comm handling.
- Avoided swallowing exceptions in module init.
- Fixed checkpatch warning for missing blank line.
v8:
- Switch from pyrf_session__init to pyrf_session__new to avoid dealing with a potentially NULL session variable.
- Added pid, tid, ppid, and cpu attributes to perf.thread.
--- tools/perf/util/python.c | 310 ++++++++++++++++++++++++++++++++++++++- 1 file changed, 304 insertions(+), 6 deletions(-)
tools/perf/util/python.c +304 -6
@@ -11,8 +11,10 @@#include <structmember.h> #include "callchain.h" +#include "comm.h" #include "counts.h" #include "data.h" +#include "debug.h" #include "event.h" #include "evlist.h" #include "evsel.h"
@@ -22,8 +24,12 @@#include "pmus.h" #include "print_binary.h" #include "record.h" +#include "session.h" #include "strbuf.h" +#include "symbol.h" +#include "thread.h" #include "thread_map.h" +#include "tool.h" #include "tp_pmu.h" #include "trace-event.h" #include "util/sample.h"
@@ -2417,6 +2423,287 @@ static int pyrf_data__setup_types(void)return PyType_Ready (&pyrf_data__type); } +struct pyrf_thread { +PyObject_HEAD + + struct thread *thread; +}; + +static void pyrf_thread__delete (struct pyrf_thread *pthread) +{ +thread__put (pthread->thread); +Py_TYPE (pthread)->tp_free ((PyObject *)pthread); +} + +static PyObject *pyrf_thread__comm (PyObject *obj) +{ +struct pyrf_thread *pthread = (void *)obj; +const char *str =thread__comm_str (pthread->thread); + +if (!str) + Py_RETURN_NONE; + +return PyUnicode_FromString (str); +} + +static PyMethodDef pyrf_thread__methods[] = { + { + .ml_name ="comm" , + .ml_meth = (PyCFunction )pyrf_thread__comm, + .ml_flags =METH_NOARGS , + .ml_doc =PyDoc_STR ("Comm(and) associated with this thread." ) + }, + { .ml_name =NULL , } +}; + +static PyObject *pyrf_thread__get_pid (struct pyrf_thread *pthread,void *closure__maybe_unused ) +{ +return PyLong_FromLong (thread__pid (pthread->thread)); +} + +static PyObject *pyrf_thread__get_tid (struct pyrf_thread *pthread,void *closure__maybe_unused ) +{ +return PyLong_FromLong (thread__tid (pthread->thread)); +} + +static PyObject *pyrf_thread__get_ppid (struct pyrf_thread *pthread,void *closure__maybe_unused ) +{ +return PyLong_FromLong (thread__ppid (pthread->thread)); +} + +static PyObject *pyrf_thread__get_cpu (struct pyrf_thread *pthread,void *closure__maybe_unused ) +{ +return PyLong_FromLong (thread__cpu (pthread->thread)); +} + +static PyGetSetDef pyrf_thread__getset[] = { + { .name ="pid" , .get = (getter )pyrf_thread__get_pid, .doc ="process ID" }, + { .name ="tid" , .get = (getter )pyrf_thread__get_tid, .doc ="thread ID" }, + { .name ="ppid" , .get = (getter )pyrf_thread__get_ppid, .doc ="parent process ID" }, + { .name ="cpu" , .get = (getter )pyrf_thread__get_cpu, .doc ="cpu number" }, + { .name =NULL } +}; + +static const char pyrf_thread__doc[] =PyDoc_STR ("perf thread object." ); + +static PyTypeObject pyrf_thread__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.thread" , + .tp_basicsize =sizeof (struct pyrf_thread ), + .tp_dealloc = (destructor )pyrf_thread__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_methods = pyrf_thread__methods, + .tp_getset = pyrf_thread__getset, + .tp_doc = pyrf_thread__doc, +}; + +static int pyrf_thread__setup_types (void ) +{ +return PyType_Ready (&pyrf_thread__type); +} + +static PyObject *pyrf_thread__from_thread (struct thread *thread) +{ + struct pyrf_thread *pthread = PyObject_New(struct pyrf_thread, &pyrf_thread__type); + +if (!pthread) +return NULL ; + + pthread->thread =thread__get (thread); +return (PyObject *)pthread; +} + +struct pyrf_session { +PyObject_HEAD + + struct perf_session *session; +struct perf_tool tool; +struct pyrf_data *pdata; +PyObject *sample; +}; + +static int pyrf_session_tool__sample (const struct perf_tool *tool, +union perf_event *event, +struct perf_sample *sample, +struct machine *machine__maybe_unused ) +{ + struct pyrf_session *psession = container_of(tool, struct pyrf_session, tool); +PyObject *pyevent =pyrf_event__new (event); +struct pyrf_event *pevent = (struct pyrf_event *)pyevent; +PyObject *ret; + +if (pyevent ==NULL ) +return -ENOMEM ; + +memcpy (&pevent->event, event, event->header.size); +if (evsel__parse_sample (sample->evsel, &pevent->event, &pevent->sample) <0 ) { +Py_DECREF (pyevent); +return -1 ; + } + + ret =PyObject_CallFunction (psession->sample,"O" , pyevent); +if (!ret) { +Py_DECREF (pyevent); +return -1 ; + } +Py_DECREF (ret); +Py_DECREF (pyevent); +return 0 ; +} + +static PyObject *pyrf_session__find_thread (struct pyrf_session *psession,PyObject *args) +{ +struct machine *machine; +struct thread *thread =NULL ; +PyObject *result; +int pid; + +if (!PyArg_ParseTuple (args,"i" , &pid)) +return NULL ; + + machine = &psession->session->machines.host; + thread =machine__find_thread (machine, pid, pid); + +if (!thread) { + machine =perf_session__find_machine (psession->session, pid); +if (machine) + thread =machine__find_thread (machine, pid, pid); + } + +if (!thread) { +PyErr_Format (PyExc_TypeError,"Failed to find thread %d" , pid); +return NULL ; + } + result =pyrf_thread__from_thread (thread); +thread__put (thread); +return result; +} + +static PyObject *pyrf_session__new (PyTypeObject *type,PyObject *args,PyObject *kwargs) +{ +struct pyrf_data *pdata; +PyObject *sample =NULL ; +static char *kwlist[] = {"data" ,"sample" ,NULL }; +struct pyrf_session *psession; +struct perf_session *session; + +if (!PyArg_ParseTupleAndKeywords (args, kwargs,"O!|O" , kwlist, &pyrf_data__type, &pdata, + &sample)) +return NULL ; + + psession = PyObject_New(struct pyrf_session, type); +if (!psession) +return NULL ; + + psession->session =NULL ; + psession->sample =NULL ; + psession->pdata =NULL ; + +Py_INCREF (pdata); + psession->pdata = pdata; + +perf_tool__init (&psession->tool,/*ordered_events=*/ true); + psession->tool.ordering_requires_timestamps = true; + +#define ADD_TOOL (name) \ + do { \ + if (name) { \ + if (!PyCallable_Check(name)) { \ + PyErr_SetString(PyExc_TypeError, #name " must be callable"); \ + goto err_out; \ + } \ + psession->tool.name = pyrf_session_tool__##name; \ + Py_INCREF(name); \ + psession->name = name; \ + } \ + } while (0) + +ADD_TOOL (sample); +#undef ADD_TOOL + + psession->tool.comm = perf_event__process_comm; + psession->tool.mmap = perf_event__process_mmap; + psession->tool.mmap2 = perf_event__process_mmap2; + psession->tool.namespaces = perf_event__process_namespaces; + psession->tool.cgroup = perf_event__process_cgroup; + psession->tool.exit = perf_event__process_exit; + psession->tool.fork = perf_event__process_fork; + psession->tool.ksymbol = perf_event__process_ksymbol; + psession->tool.text_poke = perf_event__process_text_poke; + psession->tool.build_id = perf_event__process_build_id; + session =perf_session__new (&pdata->data, &psession->tool); +if (IS_ERR (session)) { +PyErr_Format (PyExc_IOError,"failed to create session: %ld" ,PTR_ERR (session)); + goto err_out; + } + psession->session = session; + +if (symbol__init (perf_session__env (session)) <0 ) { +PyErr_SetString (PyExc_OSError,"perf: symbol__init failed" ); + goto err_out; + } + +if (perf_session__create_kernel_maps (session) <0 ) +pr_warning ("Cannot read kernel map\n" ); + +return (PyObject *)psession; +err_out: +Py_DECREF (psession); +return NULL ; +} + +static void pyrf_session__delete (struct pyrf_session *psession) +{ +perf_session__delete (psession->session); +Py_XDECREF (psession->pdata); +Py_XDECREF (psession->sample); +Py_TYPE (psession)->tp_free ((PyObject *)psession); +} + +static PyObject *pyrf_session__find_thread_events (struct pyrf_session *psession) +{ +int err =perf_session__process_events (psession->session); + +if (PyErr_Occurred ()) +return NULL ; + +return PyLong_FromLong (err); +} + +static PyMethodDef pyrf_session__methods[] = { + { + .ml_name ="process_events" , + .ml_meth = (PyCFunction )pyrf_session__find_thread_events, + .ml_flags =METH_NOARGS , + .ml_doc =PyDoc_STR ("Iterate and process events." ) + }, + { + .ml_name ="find_thread" , + .ml_meth = (PyCFunction )pyrf_session__find_thread, + .ml_flags =METH_VARARGS , + .ml_doc =PyDoc_STR ("Returns the thread associated with a pid." ) + }, + { .ml_name =NULL , } +}; + +static const char pyrf_session__doc[] =PyDoc_STR ("perf session object." ); + +static PyTypeObject pyrf_session__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.session" , + .tp_basicsize =sizeof (struct pyrf_session ), + .tp_dealloc = (destructor )pyrf_session__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_methods = pyrf_session__methods, + .tp_doc = pyrf_session__doc, + .tp_new = pyrf_session__new, +}; + +static int pyrf_session__setup_types (void ) +{ +return PyType_Ready (&pyrf_session__type); +} +static PyMethodDef perf__methods[] = { { .ml_name ="metrics" ,
@@ -2471,8 +2758,10 @@ PyMODINIT_FUNC PyInit_perf(void) };PyObject *module =PyModule_Create (&moduledef); - if (module == NULL || - pyrf_event__setup_types() < 0 || +if (module ==NULL ) +return NULL ; + + if (pyrf_event__setup_types() < 0 || pyrf_evlist__setup_types() < 0 || pyrf_evsel__setup_types() < 0 || pyrf_thread_map__setup_types() < 0 ||
@@ -2480,8 +2769,12 @@ PyMODINIT_FUNC PyInit_perf(void) pyrf_pmu_iterator__setup_types() < 0 || pyrf_pmu__setup_types() < 0 || pyrf_counts_values__setup_types() < 0 || - pyrf_data__setup_types() < 0) - return module; + pyrf_data__setup_types() < 0 || + pyrf_session__setup_types() < 0 || + pyrf_thread__setup_types() < 0) { +Py_DECREF (module); +return NULL ; + }/* The page_size is placed in util object. */ page_size =sysconf (_SC_PAGE_SIZE);
@@ -2531,6 +2824,9 @@ PyMODINIT_FUNC PyInit_perf(void)Py_INCREF (&pyrf_data__type);PyModule_AddObject (module,"data" , (PyObject *)&pyrf_data__type); +Py_INCREF (&pyrf_session__type); +PyModule_AddObject (module,"session" , (PyObject *)&pyrf_session__type); + dict =PyModule_GetDict (module);if (dict ==NULL ) goto error;
@@ -2544,7 +2840,9 @@ PyMODINIT_FUNC PyInit_perf(void) } error: -if (PyErr_Occurred ()) -PyErr_SetString (PyExc_ImportError,"perf: Init failed!" ); +if (PyErr_Occurred ()) { +Py_XDECREF (module); +return NULL ; + }return module; }
Signature · 1 line
2.54.0.794.g4f17f83d09-goog
Ian Rogers
reviewerPATCH v9 15/23 perf python: Refactor and add accessors to sample event
In reply to Ian RogersAdd common evsel field for events and move sample specific fields to only be present in sample events. Add accessors for sample events. Ensure offsets are within the bounds of the event. Allocate just enough memory for the copied event, don't make the maximum event size each time.
---
v5:
1. Fix Uninitialized Memory: Restore zero-initialization of pevent->sample in pyrf_event__new() to prevent wild free crashes on error paths.
v6:
- Refactored pyrf_event__new to take evsel and session, and use dynamic allocation based on event size. Updated callers.
v8:
- Ensure events are properly deallocated.
--- tools/perf/util/python.c | 618 +++++++++++++++++++++++++++++++++------ 1 file changed, 534 insertions(+), 84 deletions(-)
tools/perf/util/python.c +534 -84
@@ -10,21 +10,28 @@#include <perf/mmap.h> #include <structmember.h> +#include "addr_location.h" +#include "build-id.h" #include "callchain.h" #include "comm.h" #include "counts.h" #include "data.h" #include "debug.h" +#include "dso.h" #include "event.h" #include "evlist.h" #include "evsel.h" #include "expr.h" +#include "map.h" #include "metricgroup.h" #include "mmap.h" #include "pmus.h" #include "print_binary.h" #include "record.h" +#include "sample.h" #include "session.h" +#include "srccode.h" +#include "srcline.h" #include "strbuf.h" #include "symbol.h" #include "thread.h"
@@ -32,7 +39,6 @@#include "tool.h" #include "tp_pmu.h" #include "trace-event.h" -#include "util/sample.h" #ifdef HAVE_LIBTRACEEVENT #include <event-parse.h>
@@ -40,6 +46,8 @@PyMODINIT_FUNC PyInit_perf (void ); +static PyObject *pyrf_evsel__from_evsel (struct evsel *evsel); +#define member_def (type, member, ptype, help) \ { #member, ptype, \ offsetof(struct pyrf_event, event) + offsetof(struct type, member), \
@@ -52,21 +60,53 @@ PyMODINIT_FUNC PyInit_perf(void);struct pyrf_event {PyObject_HEAD +/** @sample: The parsed sample from the event. */ struct perf_sample sample; -union perf_event event; +/** @al: The address location from machine__resolve, lazily computed. */ +struct addr_location al; +/** @al_resolved: True when machine__resolve been called. */ +bool al_resolved; +/** @event: The underlying perf_event that may be in a file or ring buffer. */ +union perf_event event; };#define sample_members \ - sample_member_def(sample_ip, ip, T_ULONGLONG, "event ip"), \ sample_member_def(sample_pid, pid, T_INT, "event pid"), \ sample_member_def(sample_tid, tid, T_INT, "event tid"), \ sample_member_def(sample_time, time, T_ULONGLONG, "event timestamp"), \ - sample_member_def(sample_addr, addr, T_ULONGLONG, "event addr"), \ sample_member_def(sample_id, id, T_ULONGLONG, "event id"), \ sample_member_def(sample_stream_id, stream_id, T_ULONGLONG, "event stream id"), \ sample_member_def(sample_period, period, T_ULONGLONG, "event period"), \ sample_member_def(sample_cpu, cpu, T_UINT, "event cpu"), +static PyObject *pyrf_event__get_evsel (PyObject *self,void *closure__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; + +if (!pevent->sample.evsel) + Py_RETURN_NONE; + +return pyrf_evsel__from_evsel (pevent->sample.evsel); +} + +static PyGetSetDef pyrf_event__getset[] = { + { + .name ="evsel" , + .get = pyrf_event__get_evsel, + .set =NULL , + .doc ="tracking event." , + }, + { .name =NULL , }, +}; + +static void pyrf_event__delete (struct pyrf_event *pevent) +{ +if (pevent->al_resolved) +addr_location__exit (&pevent->al); +perf_sample__exit (&pevent->sample); +Py_TYPE (pevent)->tp_free ((PyObject *)pevent); +} +static const char pyrf_mmap_event__doc[] =PyDoc_STR ("perf mmap event object." );static PyMemberDef pyrf_mmap_event__members[] = {
@@ -105,9 +145,11 @@ static PyTypeObject pyrf_mmap_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.mmap_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_mmap_event__doc, .tp_members = pyrf_mmap_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_mmap_event__repr, };
@@ -140,9 +182,11 @@ static PyTypeObject pyrf_task_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.task_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_task_event__doc, .tp_members = pyrf_task_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_task_event__repr, };
@@ -169,9 +213,11 @@ static PyTypeObject pyrf_comm_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.comm_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_comm_event__doc, .tp_members = pyrf_comm_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_comm_event__repr, };
@@ -201,9 +247,11 @@ static PyTypeObject pyrf_throttle_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.throttle_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_throttle_event__doc, .tp_members = pyrf_throttle_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_throttle_event__repr, };
@@ -236,9 +284,11 @@ static PyTypeObject pyrf_lost_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.lost_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_lost_event__doc, .tp_members = pyrf_lost_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_lost_event__repr, };
@@ -266,9 +316,11 @@ static PyTypeObject pyrf_read_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.read_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_read_event__doc, .tp_members = pyrf_read_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_read_event__repr, };
@@ -276,16 +328,17 @@ static const char pyrf_sample_event__doc[] = PyDoc_STR("perf sample event objectstatic PyMemberDef pyrf_sample_event__members[] = { sample_members +sample_member_def (sample_ip, ip,T_ULONGLONG ,"event ip" ), +sample_member_def (sample_addr, addr,T_ULONGLONG ,"event addr" ), +sample_member_def (sample_phys_addr, phys_addr,T_ULONGLONG ,"event physical addr" ), +sample_member_def (sample_weight, weight,T_ULONGLONG ,"event weight" ), +sample_member_def (sample_data_src, data_src,T_ULONGLONG ,"event data source" ), +sample_member_def (sample_insn_count, insn_cnt,T_ULONGLONG ,"event instruction count" ), +sample_member_def (sample_cyc_count, cyc_cnt,T_ULONGLONG ,"event cycle count" ),member_def (perf_event_header, type,T_UINT ,"event type" ), { .name =NULL , }, }; -static void pyrf_sample_event__delete (struct pyrf_event *pevent) -{ -perf_sample__exit (&pevent->sample); -Py_TYPE (pevent)->tp_free ((PyObject *)pevent); -} -static PyObject *pyrf_sample_event__repr (const struct pyrf_event *pevent) {PyObject *ret;
@@ -303,6 +356,8 @@ static PyObject *pyrf_sample_event__repr(const struct pyrf_event *pevent)#ifdef HAVE_LIBTRACEEVENT static bool is_tracepoint (const struct pyrf_event *pevent) { +if (!pevent->sample.evsel) +return false;return pevent->sample.evsel->core.attr.type ==PERF_TYPE_TRACEPOINT ; }
@@ -373,6 +428,199 @@ get_tracepoint_field(struct pyrf_event *pevent, PyObject *attr_name) }#endif /* HAVE_LIBTRACEEVENT */ +static int pyrf_sample_event__resolve_al (struct pyrf_event *pevent) +{ +struct evsel *evsel = pevent->sample.evsel; +struct evlist *evlist = evsel ? evsel->evlist :NULL ; +struct perf_session *session = evlist ?evlist__session (evlist) :NULL ; + +if (pevent->al_resolved) +return 0 ; + +if (!session) +return -1 ; + +addr_location__init (&pevent->al); +if (machine__resolve (&session->machines.host, &pevent->al, &pevent->sample) <0 ) { +addr_location__exit (&pevent->al); +return -1 ; + } + + pevent->al_resolved = true; +return 0 ; +} + +static PyObject *pyrf_sample_event__get_dso (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.map) + Py_RETURN_NONE; + +return PyUnicode_FromString (dso__name (map__dso (pevent->al.map))); +} + +static PyObject *pyrf_sample_event__get_dso_long_name (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.map) + Py_RETURN_NONE; + +return PyUnicode_FromString (dso__long_name (map__dso (pevent->al.map))); +} + +static PyObject *pyrf_sample_event__get_dso_bid (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +char sbuild_id[SBUILD_ID_SIZE ]; + +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.map) + Py_RETURN_NONE; + +build_id__snprintf (dso__bid (map__dso (pevent->al.map)), sbuild_id,sizeof (sbuild_id)); +return PyUnicode_FromString (sbuild_id); +} + +static PyObject *pyrf_sample_event__get_map_start (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.map) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLong (map__start (pevent->al.map)); +} + +static PyObject *pyrf_sample_event__get_map_end (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.map) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLong (map__end (pevent->al.map)); +} + +static PyObject *pyrf_sample_event__get_map_pgoff (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.map) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLongLong (map__pgoff (pevent->al.map)); +} + +static PyObject *pyrf_sample_event__get_symbol (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.sym) + Py_RETURN_NONE; + +return PyUnicode_FromString (pevent->al.sym->name); +} + +static PyObject *pyrf_sample_event__get_sym_start (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.sym) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLongLong (pevent->al.sym->start); +} + +static PyObject *pyrf_sample_event__get_sym_end (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.sym) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLongLong (pevent->al.sym->end); +} + +static PyObject *pyrf_sample_event__get_raw_buf (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pevent->event.header.type !=PERF_RECORD_SAMPLE ) + Py_RETURN_NONE; + +return PyBytes_FromStringAndSize ((const char *)pevent->sample.raw_data, + pevent->sample.raw_size); +} + +static PyObject *pyrf_sample_event__srccode (PyObject *self,PyObject *args) +{ +struct pyrf_event *pevent = (void *)self; +u64 addr = pevent->sample.ip; +char *srcfile =NULL ; +char *srccode =NULL ; +unsigned int line =0 ; +int len =0 ; +PyObject *result; +struct addr_location al; + +if (!PyArg_ParseTuple (args,"|K" , &addr)) +return NULL ; + +if (pyrf_sample_event__resolve_al (pevent) <0 ) + Py_RETURN_NONE; + +if (addr != pevent->sample.ip) { +addr_location__init (&al); +thread__find_symbol_fb (pevent->al.thread, pevent->sample.cpumode, addr, &al); + }else { +addr_location__init (&al); + al.thread =thread__get (pevent->al.thread); + al.map =map__get (pevent->al.map); + al.sym = pevent->al.sym; + al.addr = pevent->al.addr; + } + +if (al.map) { +struct dso *dso =map__dso (al.map); + +if (dso) { + srcfile =get_srcline_split (dso,map__rip_2objdump (al.map, addr), + &line); + } + } +addr_location__exit (&al); + +if (srcfile) { + srccode =find_sourceline (srcfile, line, &len); + result =Py_BuildValue ("(sIs#)" , srcfile, line, srccode, (Py_ssize_t )len); +free (srcfile); + }else { + result =Py_BuildValue ("(sIs#)" ,NULL ,0 ,NULL , (Py_ssize_t )0 ); + } + +return result; +} + +static PyObject *pyrf_sample_event__insn (PyObject *self,PyObject *args__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; +struct thread *thread; +struct machine *machine; + +if (pyrf_sample_event__resolve_al (pevent) <0 ) + Py_RETURN_NONE; + + thread = pevent->al.thread; + +if (!thread || !thread__maps (thread)) + Py_RETURN_NONE; + + machine =maps__machine (thread__maps (thread)); +if (!machine) + Py_RETURN_NONE; + +if (pevent->sample.ip && !pevent->sample.insn_len) +perf_sample__fetch_insn (&pevent->sample, thread, machine); + +if (!pevent->sample.insn_len) + Py_RETURN_NONE; + +return PyBytes_FromStringAndSize ((const char *)pevent->sample.insn, + pevent->sample.insn_len); +} +static PyObject *pyrf_sample_event__getattro (struct pyrf_event *pevent,PyObject *attr_name) {
@@ -386,13 +634,102 @@ pyrf_sample_event__getattro(struct pyrf_event *pevent, PyObject *attr_name)return obj ?:PyObject_GenericGetAttr ((PyObject *) pevent, attr_name); } +static PyGetSetDef pyrf_sample_event__getset[] = { + { + .name ="raw_buf" , + .get = (getter )pyrf_sample_event__get_raw_buf, + .set =NULL , + .doc ="event raw buffer." , + }, + { + .name ="evsel" , + .get = pyrf_event__get_evsel, + .set =NULL , + .doc ="tracking event." , + }, + { + .name ="dso" , + .get = (getter )pyrf_sample_event__get_dso, + .set =NULL , + .doc ="event dso short name." , + }, + { + .name ="dso_long_name" , + .get = (getter )pyrf_sample_event__get_dso_long_name, + .set =NULL , + .doc ="event dso long name." , + }, + { + .name ="dso_bid" , + .get = (getter )pyrf_sample_event__get_dso_bid, + .set =NULL , + .doc ="event dso build id." , + }, + { + .name ="map_start" , + .get = (getter )pyrf_sample_event__get_map_start, + .set =NULL , + .doc ="event map start address." , + }, + { + .name ="map_end" , + .get = (getter )pyrf_sample_event__get_map_end, + .set =NULL , + .doc ="event map end address." , + }, + { + .name ="map_pgoff" , + .get = (getter )pyrf_sample_event__get_map_pgoff, + .set =NULL , + .doc ="event map page offset." , + }, + { + .name ="symbol" , + .get = (getter )pyrf_sample_event__get_symbol, + .set =NULL , + .doc ="event symbol name." , + }, + { + .name ="sym_start" , + .get = (getter )pyrf_sample_event__get_sym_start, + .set =NULL , + .doc ="event symbol start address." , + }, + { + .name ="sym_end" , + .get = (getter )pyrf_sample_event__get_sym_end, + .set =NULL , + .doc ="event symbol end address." , + }, + { .name =NULL , }, +}; + +static PyMethodDef pyrf_sample_event__methods[] = { + { + .ml_name ="srccode" , + .ml_meth = (PyCFunction )pyrf_sample_event__srccode, + .ml_flags =METH_VARARGS , + .ml_doc =PyDoc_STR ("Get source code for an address." ) + }, + { + .ml_name ="insn" , + .ml_meth = (PyCFunction )pyrf_sample_event__insn, + .ml_flags =METH_NOARGS , + .ml_doc =PyDoc_STR ("Get instruction bytes for a sample." ) + }, + { .ml_name =NULL , } +}; +static PyTypeObject pyrf_sample_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name ="perf.sample_event" , .tp_basicsize =sizeof (struct pyrf_event ), + .tp_dealloc = (destructor )pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_sample_event__doc, .tp_members = pyrf_sample_event__members, + .tp_getset = pyrf_sample_event__getset, + .tp_methods = pyrf_sample_event__methods, .tp_repr = (reprfunc )pyrf_sample_event__repr, .tp_getattro = (getattrofunc ) pyrf_sample_event__getattro, };
@@ -428,25 +765,17 @@ static PyTypeObject pyrf_context_switch_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.context_switch_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_context_switch_event__doc, .tp_members = pyrf_context_switch_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_context_switch_event__repr, };static int pyrf_event__setup_types (void ) {int err; - pyrf_mmap_event__type.tp_new = - pyrf_task_event__type.tp_new = - pyrf_comm_event__type.tp_new = - pyrf_lost_event__type.tp_new = - pyrf_read_event__type.tp_new = - pyrf_sample_event__type.tp_new = - pyrf_context_switch_event__type.tp_new = - pyrf_throttle_event__type.tp_new = PyType_GenericNew; - - pyrf_sample_event__type.tp_dealloc = (destructor )pyrf_sample_event__delete, err =PyType_Ready (&pyrf_mmap_event__type);if (err <0 )
@@ -490,33 +819,175 @@ static PyTypeObject *pyrf_event__type[] = { [PERF_RECORD_SWITCH_CPU_WIDE] = &pyrf_context_switch_event__type, }; -static PyObject *pyrf_event__new (const union perf_event *event) +static PyObject *pyrf_event__new (const union perf_event *event,struct evsel *evsel, +struct perf_session *session__maybe_unused ) {struct pyrf_event *pevent; -PyTypeObject *ptype; +size_t size; +int err; +size_t min_size =sizeof (struct perf_event_header ); -if ((event->header.type <PERF_RECORD_MMAP || - event->header.type >PERF_RECORD_SAMPLE ) && - !(event->header.type ==PERF_RECORD_SWITCH || - event->header.type ==PERF_RECORD_SWITCH_CPU_WIDE )) { -PyErr_Format (PyExc_TypeError,"Unexpected header type %u" , +if (event->header.type >=ARRAY_SIZE (pyrf_event__type) || + pyrf_event__type[event->header.type] ==NULL ) { +return PyErr_Format (PyExc_TypeError,"Unexpected header type %u" , event->header.type); -return NULL ; } -// FIXME this better be dynamic or we need to parse everything -// before calling perf_mmap__consume(), including tracepoint fields. -if (sizeof (pevent->event) < event->header.size) { -PyErr_Format (PyExc_TypeError,"Unexpected event size: %zd < %u" , -sizeof (pevent->event), event->header.size); -return NULL ; +switch (event->header.type) { +case PERF_RECORD_MMAP : + min_size = offsetof(struct perf_record_mmap , filename) +1 ; +break ; +case PERF_RECORD_MMAP2 : + min_size = offsetof(struct perf_record_mmap2 , filename) +1 ; +break ; +case PERF_RECORD_COMM : + min_size =sizeof (struct perf_record_comm ); +break ; +case PERF_RECORD_FORK : +case PERF_RECORD_EXIT : + min_size =sizeof (struct perf_record_fork ); +break ; +case PERF_RECORD_THROTTLE : +case PERF_RECORD_UNTHROTTLE : + min_size =sizeof (struct perf_record_throttle ); +break ; +case PERF_RECORD_LOST : + min_size =sizeof (struct perf_record_lost ); +break ; +case PERF_RECORD_READ : + min_size =sizeof (struct perf_record_read ); +break ; +case PERF_RECORD_SWITCH : +case PERF_RECORD_SWITCH_CPU_WIDE : + min_size =sizeof (struct perf_record_switch ); +break ; +case PERF_RECORD_AUX : + min_size =sizeof (struct perf_record_aux ); +break ; +case PERF_RECORD_ITRACE_START : + min_size =sizeof (struct perf_record_itrace_start ); +break ; +case PERF_RECORD_LOST_SAMPLES : + min_size =sizeof (struct perf_record_lost_samples ); +break ; +case PERF_RECORD_NAMESPACES : + min_size = offsetof(struct perf_record_namespaces , link_info); +break ; +case PERF_RECORD_KSYMBOL : + min_size =sizeof (struct perf_record_ksymbol ); +break ; +case PERF_RECORD_BPF_EVENT : + min_size =sizeof (struct perf_record_bpf_event ); +break ; +case PERF_RECORD_CGROUP : + min_size =sizeof (struct perf_record_cgroup ); +break ; +case PERF_RECORD_TEXT_POKE : + min_size = offsetof(struct perf_record_text_poke_event , bytes); +break ; +case PERF_RECORD_AUX_OUTPUT_HW_ID : + min_size =sizeof (struct perf_record_aux_output_hw_id ); +break ; +case PERF_RECORD_CALLCHAIN_DEFERRED : + min_size =sizeof (struct perf_record_callchain_deferred ); +break ; +case PERF_RECORD_HEADER_ATTR : + min_size =sizeof (struct perf_record_header_attr ); +break ; +case PERF_RECORD_HEADER_TRACING_DATA : + min_size =sizeof (struct perf_record_header_tracing_data ); +break ; +case PERF_RECORD_HEADER_BUILD_ID : + min_size = offsetof(struct perf_record_header_build_id , filename) +1 ; +break ; +case PERF_RECORD_ID_INDEX : + min_size = offsetof(struct perf_record_id_index , entries); +break ; +case PERF_RECORD_AUXTRACE_INFO : + min_size = offsetof(struct perf_record_auxtrace_info , priv); +break ; +case PERF_RECORD_AUXTRACE : + min_size =sizeof (struct perf_record_auxtrace ); +break ; +case PERF_RECORD_AUXTRACE_ERROR : + min_size =sizeof (struct perf_record_auxtrace_error ); +break ; +case PERF_RECORD_THREAD_MAP : + min_size = offsetof(struct perf_record_thread_map , entries); +break ; +case PERF_RECORD_CPU_MAP : + min_size =sizeof (struct perf_record_cpu_map ); +break ; +case PERF_RECORD_STAT_CONFIG : + min_size = offsetof(struct perf_record_stat_config , data); +break ; +case PERF_RECORD_STAT : + min_size =sizeof (struct perf_record_stat ); +break ; +case PERF_RECORD_STAT_ROUND : + min_size =sizeof (struct perf_record_stat_round ); +break ; +case PERF_RECORD_EVENT_UPDATE : + min_size =sizeof (struct perf_record_event_update ); +break ; +case PERF_RECORD_TIME_CONV : + min_size =sizeof (struct perf_record_time_conv ); +break ; +case PERF_RECORD_HEADER_FEATURE : + min_size = offsetof(struct perf_record_header_feature , data); +break ; +case PERF_RECORD_COMPRESSED : + min_size = offsetof(struct perf_record_compressed , data); +break ; +case PERF_RECORD_COMPRESSED2 : + min_size = offsetof(struct perf_record_compressed2 , data); +break ; +case PERF_RECORD_BPF_METADATA : + min_size = offsetof(struct perf_record_bpf_metadata , entries); +break ; +case PERF_RECORD_SCHEDSTAT_CPU : + min_size =sizeof (struct perf_record_schedstat_cpu ); +break ; +case PERF_RECORD_SCHEDSTAT_DOMAIN : + min_size =sizeof (struct perf_record_schedstat_domain ); +break ; +default : +break ; } +if (event->header.size < min_size) +return PyErr_Format (PyExc_ValueError,"Event size %u too small for type %u" , + event->header.size, event->header.type); + +/* Allocate just enough memory for the size of event. */ + size = offsetof(struct pyrf_event , event) + event->header.size; + pevent = (struct pyrf_event *)PyObject_Malloc (size); +if (pevent ==NULL ) +return PyErr_NoMemory (); + +/* Copy the event for memory safety and initilaize variables. */ +PyObject_Init ((PyObject *)pevent, pyrf_event__type[event->header.type]); +memcpy (&pevent->event, event, event->header.size); + +if (event->header.type ==PERF_RECORD_MMAP ) { +/* Ensure '\0' string termination. */ +size_t max_len = pevent->event.header.size - offsetof(struct perf_record_mmap , filename); - ptype = pyrf_event__type[event->header.type]; - pevent = PyObject_New(struct pyrf_event, ptype); -if (pevent !=NULL ) { -memcpy (&pevent->event, event, event->header.size); -perf_sample__init (&pevent->sample,/*all=*/ false); + pevent->event.mmap.filename[max_len -1 ] ='\0' ; + } + +perf_sample__init (&pevent->sample,/*all=*/ true); + pevent->al_resolved = false; +addr_location__init (&pevent->al); + +if (!evsel) +return (PyObject *)pevent; + +/* Parse the sample again so that pointers are within the copied event. */ + err =evsel__parse_sample (evsel, &pevent->event, &pevent->sample); +if (err <0 ) { +Py_DECREF (pevent); +return PyErr_Format (PyExc_OSError, +"perf: can't parse sample, err=%d" , err); }return (PyObject *)pevent; }
@@ -1209,7 +1680,7 @@ static PyObject *pyrf_evsel__str(PyObject *self)struct pyrf_evsel *pevsel = (void *)self;struct evsel *evsel = pevsel->evsel; -return PyUnicode_FromFormat ("evsel(%s/%s/)" ,evsel__pmu_name (evsel),evsel__name (evsel)); +return PyUnicode_FromFormat ("evsel(%s)" ,evsel__name (evsel)); }static PyMethodDef pyrf_evsel__methods[] = {
@@ -1773,9 +2244,11 @@ static PyObject *pyrf_evlist__read_on_cpu(struct pyrf_evlist *pevlist, { struct evlist *evlist = pevlist->evlist; union perf_event *event; + struct evsel *evsel; int sample_id_all = 1, cpu; static char *kwlist[] = { "cpu", "sample_id_all", NULL }; struct mmap *md; + PyObject *pyevent; int err; if (!PyArg_ParseTupleAndKeywords(args, kwargs, "i|i", kwlist,
@@ -1783,44 +2256,31 @@ static PyObject *pyrf_evlist__read_on_cpu(struct pyrf_evlist *pevlist,return NULL ; md =get_md (evlist, cpu); -if (!md) { -PyErr_Format (PyExc_TypeError,"Unknown CPU '%d'" , cpu); -return NULL ; - } +if (!md) +return PyErr_Format (PyExc_TypeError,"Unknown CPU '%d'" , cpu); -if (perf_mmap__read_init (&md->core) <0 ) - goto end; + err =perf_mmap__read_init (&md->core); +if (err <0 ) { +return PyErr_Format (PyExc_OSError, +"perf: error mmap read init, err=%d" , err); + } event =perf_mmap__read_event (&md->core); -if (event !=NULL ) { -PyObject *pyevent =pyrf_event__new (event); -struct pyrf_event *pevent = (struct pyrf_event *)pyevent; -struct evsel *evsel; - -if (pyevent ==NULL ) -return PyErr_NoMemory (); - - evsel =evlist__event2evsel (evlist, event); -if (!evsel) { -Py_DECREF (pyevent); -Py_INCREF (Py_None); -return Py_None; - } +if (event ==NULL ) + Py_RETURN_NONE; + evsel =evlist__event2evsel (evlist, event); +if (!evsel) { +/* Unknown evsel. */ perf_mmap__consume (&md->core); - - err =evsel__parse_sample (evsel, &pevent->event, &pevent->sample); -if (err) { -Py_DECREF (pyevent); -return PyErr_Format (PyExc_OSError, -"perf: can't parse sample, err=%d" , err); - } - -return pyevent; + Py_RETURN_NONE; } -end: -Py_INCREF (Py_None); -return Py_None; + pyevent =pyrf_event__new (event, evsel,evlist__session (evlist)); +perf_mmap__consume (&md->core); +if (pyevent ==NULL ) +return PyErr_Occurred () ?NULL :PyErr_NoMemory (); + +return pyevent; } static PyObject *pyrf_evlist__open(struct pyrf_evlist *pevlist,
@@ -2015,10 +2475,7 @@ static PyObject *pyrf_evlist__str(PyObject *self)evlist__for_each_entry (pevlist->evlist, pos) {if (!first)strbuf_addch (&sb,',' ); -if (!pos->pmu) -strbuf_addstr (&sb,evsel__name (pos)); -else -strbuf_addf (&sb,"%s/%s/" , pos->pmu->name,evsel__name (pos)); +strbuf_addstr (&sb,evsel__name (pos)); first = false; }strbuf_addstr (&sb,"])" );
@@ -2528,19 +2985,12 @@ static int pyrf_session_tool__sample(const struct perf_tool *tool, struct machine *machine __maybe_unused) { struct pyrf_session *psession = container_of(tool, struct pyrf_session, tool); -PyObject *pyevent =pyrf_event__new (event); -struct pyrf_event *pevent = (struct pyrf_event *)pyevent; +PyObject *pyevent =pyrf_event__new (event, sample->evsel, psession->session);PyObject *ret;if (pyevent ==NULL )return -ENOMEM ; -memcpy (&pevent->event, event, event->header.size); -if (evsel__parse_sample (sample->evsel, &pevent->event, &pevent->sample) <0 ) { -Py_DECREF (pyevent); -return -1 ; - } - ret =PyObject_CallFunction (psession->sample,"O" , pyevent);if (!ret) {Py_DECREF (pyevent);
Signature · 1 line
2.54.0.794.g4f17f83d09-goog
Ian Rogers
reviewerPATCH v9 16/23 perf python: Add mmap2 event
In reply to Ian RogersIf mmap is handled so should mmap2 events. Add support as a distinct python event type.
--- tools/perf/util/python.c | 60 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 60 insertions(+)
tools/perf/util/python.c +60 -0
@@ -153,6 +153,54 @@ static PyTypeObject pyrf_mmap_event__type = { .tp_repr = (reprfunc)pyrf_mmap_event__repr, }; +static const char pyrf_mmap2_event__doc[] =PyDoc_STR ("perf mmap2 event object." ); + +static PyMemberDef pyrf_mmap2_event__members[] = { + sample_members +member_def (perf_event_header, type,T_UINT ,"event type" ), +member_def (perf_event_header, misc,T_UINT ,"event misc" ), +member_def (perf_record_mmap2, pid,T_UINT ,"event pid" ), +member_def (perf_record_mmap2, tid,T_UINT ,"event tid" ), +member_def (perf_record_mmap2, start,T_ULONGLONG ,"start of the map" ), +member_def (perf_record_mmap2, len,T_ULONGLONG ,"map length" ), +member_def (perf_record_mmap2, pgoff,T_ULONGLONG ,"page offset" ), +member_def (perf_record_mmap2, prot,T_UINT ,"protection" ), +member_def (perf_record_mmap2, flags,T_UINT ,"flags" ), +member_def (perf_record_mmap2, filename,T_STRING_INPLACE ,"backing store" ), + { .name =NULL , }, +}; + +static PyObject *pyrf_mmap2_event__repr (const struct pyrf_event *pevent) +{ +PyObject *ret; +char *s; + +if (asprintf (&s,"{ type: mmap2, pid: %u, tid: %u, start: %#" PRI_lx64", " +"length: %#" PRI_lx64", offset: %#" PRI_lx64", " +"flags: %#x, prot: %#x, filename: %s }" , + pevent->event.mmap2.pid, pevent->event.mmap2.tid, + pevent->event.mmap2.start, pevent->event.mmap2.len, + pevent->event.mmap2.pgoff, pevent->event.mmap2.flags, + pevent->event.mmap2.prot, pevent->event.mmap2.filename) <0 ) +return PyErr_NoMemory (); + + ret =PyUnicode_FromString (s); +free (s); +return ret; +} + +static PyTypeObject pyrf_mmap2_event__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.mmap2_event" , + .tp_basicsize =sizeof (struct pyrf_event ), + .tp_dealloc = (destructor )pyrf_event__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_mmap2_event__doc, + .tp_members = pyrf_mmap2_event__members, + .tp_getset = pyrf_event__getset, + .tp_repr = (reprfunc )pyrf_mmap2_event__repr, +}; +static const char pyrf_task_event__doc[] =PyDoc_STR ("perf task (fork/exit) event object." );static PyMemberDef pyrf_task_event__members[] = {
@@ -778,6 +826,9 @@ static int pyrf_event__setup_types(void)int err; err =PyType_Ready (&pyrf_mmap_event__type); +if (err <0 ) + goto out; + err =PyType_Ready (&pyrf_mmap2_event__type);if (err <0 ) goto out; err =PyType_Ready (&pyrf_lost_event__type);
@@ -807,6 +858,7 @@ static int pyrf_event__setup_types(void)static PyTypeObject *pyrf_event__type[] = { [PERF_RECORD_MMAP ] = &pyrf_mmap_event__type, + [PERF_RECORD_MMAP2 ] = &pyrf_mmap2_event__type, [PERF_RECORD_LOST ] = &pyrf_lost_event__type, [PERF_RECORD_COMM ] = &pyrf_comm_event__type, [PERF_RECORD_EXIT ] = &pyrf_task_event__type,
@@ -973,6 +1025,11 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evsize_t max_len = pevent->event.header.size - offsetof(struct perf_record_mmap , filename); pevent->event.mmap.filename[max_len -1 ] ='\0' ; + }else if (event->header.type ==PERF_RECORD_MMAP2 ) { +/* Ensure '\0' string termination. */ +size_t max_len = pevent->event.header.size - offsetof(struct perf_record_mmap2 , filename); + + pevent->event.mmap2.filename[max_len -1 ] ='\0' ; }perf_sample__init (&pevent->sample,/*all=*/ true);
@@ -3238,6 +3295,9 @@ PyMODINIT_FUNC PyInit_perf(void)Py_INCREF (&pyrf_mmap_event__type);PyModule_AddObject (module,"mmap_event" , (PyObject *)&pyrf_mmap_event__type); +Py_INCREF (&pyrf_mmap2_event__type); +PyModule_AddObject (module,"mmap2_event" , (PyObject *)&pyrf_mmap2_event__type); +Py_INCREF (&pyrf_lost_event__type);PyModule_AddObject (module,"lost_event" , (PyObject *)&pyrf_lost_event__type);
Signature · 1 line
2.54.0.794.g4f17f83d09-goog
[email protected]
AI reviewRe: [PATCH v9 16/23] perf python: Add mmap2 event
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider: - [Medium] The newly added `mmap2_event` Python type omits the core fields that distinguish it from the…
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [Medium] The newly added mmap2_event Python type omits the core fields that distinguish it from the mmap event (maj, min, ino, ino_generation, or build_id).
- [Medium] Incorrect type identifier 'T_UINT' used for the 16-bit 'misc' field, causing Python to read corrupted data for the event.
--
commit 32ac3903a38f752d22cb2845337c77142842e9ac
Author: Ian Rogers <[email protected]>
perf python: Add mmap2 event
This commit adds support for the PERF_RECORD_MMAP2 event type to the perf python API. It registers a new mmap2_event Python object and handlers to process these events.
Quoted 5 lines
tools/perf/util/python.c +0
-0
tools/perf/util/python.c static PyTypeObject pyrf_mmap_event__type = { · line 161
Quoted 9 lines
tools/perf/util/python.c +6
-0@@ -153,6 +153,54 @@ static PyTypeObject pyrf_mmap_event__type = { .tp_repr = (reprfunc)pyrf_mmap_event__repr, }; +static const char pyrf_mmap2_event__doc[] =PyDoc_STR ("perf mmap2 event object." ); + +static PyMemberDef pyrf_mmap2_event__members[] = { + sample_members +member_def (perf_event_header, type,T_UINT ,"event type" ), +member_def (perf_event_header, misc,T_UINT ,"event misc" ),
[Severity: Medium]
Is T_UINT the correct type identifier for the misc field?
Looking at struct perf_event_header, misc is a 16-bit unsigned integer (__u16). Using T_UINT instructs the Python C API to read a 32-bit integer, which might inadvertently read the adjacent 16-bit size field as well and return a corrupted value.
Would T_USHORT be more appropriate here?
tools/perf/util/python.c static PyTypeObject pyrf_mmap_event__type = { · line 169
Quoted 8 lines
tools/perf/util/python.c +8
-0+member_def (perf_record_mmap2, pid,T_UINT ,"event pid" ), +member_def (perf_record_mmap2, tid,T_UINT ,"event tid" ), +member_def (perf_record_mmap2, start,T_ULONGLONG ,"start of the map" ), +member_def (perf_record_mmap2, len,T_ULONGLONG ,"map length" ), +member_def (perf_record_mmap2, pgoff,T_ULONGLONG ,"page offset" ), +member_def (perf_record_mmap2, prot,T_UINT ,"protection" ), +member_def (perf_record_mmap2, flags,T_UINT ,"flags" ), + member_def(perf_record_mmap2, filename, T_STRING_INPLACE, "backing store"),
[Severity: Medium]
Does this structure omit the core fields that distinguish mmap2 from mmap?
The PERF_RECORD_MMAP2 event is designed to include additional identifying fields, such as maj, min, ino, ino_generation, and build_id.
Since these fields are missing from pyrf_mmap2_event__members, the Python API might not expose the essential data that differentiates an mmap2 event from a regular mmap event. Should these fields be added so scripts can access them?
Quoted 2 lines
tools/perf/util/python.c +2
-0+ { .name = NULL, }, +};
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=16
Ian Rogers
reviewerPATCH v9 17/23 perf python: Add callchain support
In reply to Ian RogersImplement pyrf_callchain_node and pyrf_callchain types for lazy iteration over callchain frames. Add callchain property to sample_event.
---
v2:
1. Eager Callchain Resolution: Moved the callchain resolution from deferred iteration to eager processing in pyrf_session_tool__sample() . This avoids risks of reading from unmapped memory or following dangling pointers to closed sessions.
2. Cached Callchain: Added a callchain field to struct pyrf_event to store the resolved object.
3. Simplified Access: pyrf_sample_event__get_callchain() now just returns the cached object if available.
4. Avoided Double Free: Handled lazy cleanups properly.
v6:
- Moved callchain resolution from session_tool__sample to pyrf_event__new.
--- tools/perf/util/python.c | 241 ++++++++++++++++++++++++++++++++++++++- 1 file changed, 240 insertions(+), 1 deletion(-)
tools/perf/util/python.c +240 -1
@@ -66,6 +66,8 @@ struct pyrf_event {struct addr_location al;/** @al_resolved: True when machine__resolve been called. */ bool al_resolved; +/** @callchain: Resolved callchain, eagerly computed if requested. */ +PyObject *callchain;/** @event: The underlying perf_event that may be in a file or ring buffer. */ union perf_event event; };
@@ -103,6 +105,7 @@ static void pyrf_event__delete(struct pyrf_event *pevent) {if (pevent->al_resolved)addr_location__exit (&pevent->al); +Py_XDECREF (pevent->callchain);perf_sample__exit (&pevent->sample);Py_TYPE (pevent)->tp_free ((PyObject *)pevent); }
@@ -669,6 +672,181 @@ static PyObject *pyrf_sample_event__insn(PyObject *self, PyObject *args __maybe_ pevent->sample.insn_len); } +struct pyrf_callchain_node { +PyObject_HEAD + u64 ip; +struct map *map; +struct symbol *sym; +}; + +static void pyrf_callchain_node__delete (struct pyrf_callchain_node *pnode) +{ +map__put (pnode->map); +Py_TYPE (pnode)->tp_free ((PyObject *)pnode); +} + +static PyObject *pyrf_callchain_node__get_ip (struct pyrf_callchain_node *pnode, +void *closure__maybe_unused ) +{ +return PyLong_FromUnsignedLongLong (pnode->ip); +} + +static PyObject *pyrf_callchain_node__get_symbol (struct pyrf_callchain_node *pnode, +void *closure__maybe_unused ) +{ +if (pnode->sym) +return PyUnicode_FromString (pnode->sym->name); +return PyUnicode_FromString ("[unknown]" ); +} + +static PyObject *pyrf_callchain_node__get_dso (struct pyrf_callchain_node *pnode, +void *closure__maybe_unused ) +{ +const char *dsoname ="[unknown]" ; + +if (pnode->map) { +struct dso *dso =map__dso (pnode->map); +if (dso) { +if (symbol_conf.show_kernel_path &&dso__long_name (dso)) + dsoname =dso__long_name (dso); +else + dsoname =dso__name (dso); + } + } +return PyUnicode_FromString (dsoname); +} + +static PyGetSetDef pyrf_callchain_node__getset[] = { + { .name ="ip" , .get = (getter )pyrf_callchain_node__get_ip, }, + { .name ="symbol" , .get = (getter )pyrf_callchain_node__get_symbol, }, + { .name ="dso" , .get = (getter )pyrf_callchain_node__get_dso, }, + { .name =NULL , }, +}; + +static PyTypeObject pyrf_callchain_node__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.callchain_node" , + .tp_basicsize =sizeof (struct pyrf_callchain_node ), + .tp_dealloc = (destructor )pyrf_callchain_node__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc ="perf callchain node object." , + .tp_getset = pyrf_callchain_node__getset, +}; + +struct pyrf_callchain_frame { +u64 ip; +struct map *map; +struct symbol *sym; +}; + +struct pyrf_callchain { +PyObject_HEAD + struct pyrf_event *pevent; +struct pyrf_callchain_frame *frames; +u64 nr_frames; +u64 pos; +bool resolved; +}; + +static void pyrf_callchain__delete (struct pyrf_callchain *pchain) +{ +Py_XDECREF (pchain->pevent); +if (pchain->frames) { +for (u64 i =0 ; i < pchain->nr_frames; i++) +map__put (pchain->frames[i].map); +free (pchain->frames); + } +Py_TYPE (pchain)->tp_free ((PyObject *)pchain); +} + +static PyObject *pyrf_callchain__next (struct pyrf_callchain *pchain) +{ +struct pyrf_callchain_node *pnode; + +if (!pchain->resolved) { +struct evsel *evsel = pchain->pevent->sample.evsel; +struct evlist *evlist = evsel->evlist; +struct perf_session *session = evlist ?evlist__session (evlist) :NULL ; +struct addr_location al; +struct callchain_cursor *cursor; +struct callchain_cursor_node *node; +u64 i; + +if (!session || !pchain->pevent->sample.callchain) +return NULL ; + +addr_location__init (&al); +if (machine__resolve (&session->machines.host, &al, &pchain->pevent->sample) <0 ) { +addr_location__exit (&al); +return NULL ; + } + + cursor =get_tls_callchain_cursor (); +if (thread__resolve_callchain (al.thread, cursor, + &pchain->pevent->sample,NULL ,NULL , +PERF_MAX_STACK_DEPTH ) !=0 ) { +addr_location__exit (&al); +return NULL ; + } +callchain_cursor_commit (cursor); + + pchain->nr_frames = cursor->nr; +if (pchain->nr_frames >0 ) { + pchain->frames =calloc (pchain->nr_frames,sizeof (*pchain->frames)); +if (!pchain->frames) { +addr_location__exit (&al); +return PyErr_NoMemory (); + } + +for (i =0 ; i < pchain->nr_frames; i++) { + node =callchain_cursor_current (cursor); + pchain->frames[i].ip = node->ip; + pchain->frames[i].map =map__get (node->ms.map); + pchain->frames[i].sym = node->ms.sym; +callchain_cursor_advance (cursor); + } + } + pchain->resolved = true; +addr_location__exit (&al); + } + +if (pchain->pos >= pchain->nr_frames) +return NULL ; + + pnode = PyObject_New(struct pyrf_callchain_node, &pyrf_callchain_node__type); +if (!pnode) +return NULL ; + + pnode->ip = pchain->frames[pchain->pos].ip; + pnode->map =map__get (pchain->frames[pchain->pos].map); + pnode->sym = pchain->frames[pchain->pos].sym; + + pchain->pos++; +return (PyObject *)pnode; +} + +static PyTypeObject pyrf_callchain__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.callchain" , + .tp_basicsize =sizeof (struct pyrf_callchain ), + .tp_dealloc = (destructor )pyrf_callchain__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc ="perf callchain object." , + .tp_iter = PyObject_SelfIter, + .tp_iternext = (iternextfunc )pyrf_callchain__next, +}; + +static PyObject *pyrf_sample_event__get_callchain (PyObject *self,void *closure__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; + +if (!pevent->callchain) + Py_RETURN_NONE; + +Py_INCREF (pevent->callchain); +return pevent->callchain; +} +static PyObject *pyrf_sample_event__getattro (struct pyrf_event *pevent,PyObject *attr_name) {
@@ -683,6 +861,12 @@ pyrf_sample_event__getattro(struct pyrf_event *pevent, PyObject *attr_name) }static PyGetSetDef pyrf_sample_event__getset[] = { + { + .name ="callchain" , + .get = pyrf_sample_event__get_callchain, + .set =NULL , + .doc ="event callchain." , + }, { .name ="raw_buf" , .get = (getter )pyrf_sample_event__get_raw_buf,
@@ -852,6 +1036,12 @@ static int pyrf_event__setup_types(void) err =PyType_Ready (&pyrf_context_switch_event__type);if (err <0 ) goto out; + err =PyType_Ready (&pyrf_callchain_node__type); +if (err <0 ) + goto out; + err =PyType_Ready (&pyrf_callchain__type); +if (err <0 ) + goto out; out:return err; }
@@ -872,9 +1062,11 @@ static PyTypeObject *pyrf_event__type[] = { };static PyObject *pyrf_event__new (const union perf_event *event,struct evsel *evsel, -struct perf_session *session__maybe_unused ) +struct perf_session *session) {struct pyrf_event *pevent; +struct perf_sample *sample; +struct machine *machine = session ? &session->machines.host :NULL ;size_t size;int err;size_t min_size =sizeof (struct perf_event_header );
@@ -1033,6 +1225,7 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *ev }perf_sample__init (&pevent->sample,/*all=*/ true); + pevent->callchain =NULL ; pevent->al_resolved = false;addr_location__init (&pevent->al);
@@ -1046,6 +1239,49 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evreturn PyErr_Format (PyExc_OSError,"perf: can't parse sample, err=%d" , err); } + sample = &pevent->sample; +if (machine && sample->callchain) { +struct addr_location al; +struct callchain_cursor *cursor; +u64 i; +struct pyrf_callchain *pchain; + +addr_location__init (&al); +if (machine__resolve (machine, &al, sample) >=0 ) { + cursor =get_tls_callchain_cursor (); +if (thread__resolve_callchain (al.thread, cursor, sample, +NULL ,NULL ,PERF_MAX_STACK_DEPTH ) ==0 ) { +callchain_cursor_commit (cursor); + + pchain = PyObject_New(struct pyrf_callchain, &pyrf_callchain__type); +if (pchain) { + pchain->pevent = pevent; +Py_INCREF (pevent); + pchain->nr_frames = cursor->nr; + pchain->pos =0 ; + pchain->resolved = true; + pchain->frames =calloc (pchain->nr_frames, +sizeof (*pchain->frames)); +if (pchain->frames) { +struct callchain_cursor_node *node; + +for (i =0 ; i < pchain->nr_frames; i++) { + node =callchain_cursor_current (cursor); + pchain->frames[i].ip = node->ip; + pchain->frames[i].map = +map__get (node->ms.map); + pchain->frames[i].sym = node->ms.sym; +callchain_cursor_advance (cursor); + } + pevent->callchain = (PyObject *)pchain; + }else { +Py_DECREF (pchain); + } + } + } +addr_location__exit (&al); + } + }return (PyObject *)pevent; }
@@ -3145,6 +3381,9 @@ static PyObject *pyrf_session__new(PyTypeObject *type, PyObject *args, PyObject } psession->session = session; + symbol_conf.use_callchain = true; + symbol_conf.show_kernel_path = true; + symbol_conf.inline_name = false;if (symbol__init (perf_session__env (session)) <0 ) {PyErr_SetString (PyExc_OSError,"perf: symbol__init failed" ); goto err_out;
Signature · 1 line
2.54.0.794.g4f17f83d09-goog
[email protected]
AI reviewRe: [PATCH v9 17/23] perf python: Add callchain support
Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider: - [High] A circular reference between `pyrf_event` and `pyrf_callchain` creates a permanent memory leak for…
Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
- [High] A circular reference between pyrf_event and pyrf_callchain creates a permanent memory leak for every processed event containing a callchain.
- [Medium] Eager callchain resolution contradicts the 'lazy iteration' design, causing severe parsing overhead and leaving behind dead code.
- [Medium] The .callchain property returns an exhaustible iterator instead of a repeatable iterable, causing silent failures on subsequent accesses. --
commit 2ae22d7a748138d318f9939992a43e386f407964
Author: Ian Rogers <[email protected]>
perf python: Add callchain support
This commit adds support for callchains to the perf Python extension, introducing new types for callchains and nodes, and appending a callchain property to sample events.
Quoted 4 lines
tools/perf/util/python.c +0
-0
[ ... ]
Quoted 1 line
tools/perf/util/python.c +0
-0@@ -669,6 +672,181 @@ static PyObject *pyrf_sample_event__insn(PyObject *self, PyObject *args __maybe_
[ ... ]
tools/perf/util/python.c static PyObject *pyrf_sample_event__insn(PyObject *self, PyObject *args __maybe_ · line 768
Quoted 7 lines
tools/perf/util/python.c +7
-0+static PyObject *pyrf_callchain__next (struct pyrf_callchain *pchain) +{ +struct pyrf_callchain_node *pnode; + +if (!pchain->resolved) { +struct evsel *evsel = pchain->pevent->sample.evsel; +struct evlist *evlist = evsel->evlist;
[Severity: Medium]
The commit message states this implements "lazy iteration over callchain frames", but is this block ever executed?
Since thread__resolve_callchain() is called eagerly during pyrf_event__new(), pchain->resolved is always true here. This seems to defeat the lazy resolution and leaves this block as unused dead code.
Does this also force heavy symbol resolution on all events immediately upon creation, even if the Python script doesn't access the callchain property?
Quoted 3 lines
tools/perf/util/python.c +3
-0+struct perf_session *session = evlist ?evlist__session (evlist) :NULL ; +struct addr_location al; +struct callchain_cursor *cursor;
[ ... ]
Quoted 6 lines
tools/perf/util/python.c +6
-0+ pchain->resolved = true; +addr_location__exit (&al); + } + +if (pchain->pos >= pchain->nr_frames) +return NULL ;
tools/perf/util/python.c static PyObject *pyrf_sample_event__insn(PyObject *self, PyObject *args __maybe_ · line 824
Quoted 10 lines
tools/perf/util/python.c +10
-0+ + pnode = PyObject_New(struct pyrf_callchain_node, &pyrf_callchain_node__type); +if (!pnode) +return NULL ; + + pnode->ip = pchain->frames[pchain->pos].ip; + pnode->map =map__get (pchain->frames[pchain->pos].map); + pnode->sym = pchain->frames[pchain->pos].sym; + + pchain->pos++;
[Severity: Medium]
Does returning an exhaustible iterator from the property cause issues on subsequent accesses?
Because the .callchain property caches this single iterator instance and pyrf_callchain__next() increments pchain->pos without any reset mechanism, the iterator will be permanently exhausted on the first pass.
Will subsequent attempts to iterate over event.callchain silently yield zero elements instead of repeating the iteration?
Quoted 2 lines
tools/perf/util/python.c +2
-0+return (PyObject *)pnode; +}
[ ... ]
Quoted 12 lines
tools/perf/util/python.c +8
-0@@ -1046,6 +1239,49 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evreturn PyErr_Format (PyExc_OSError,"perf: can't parse sample, err=%d" , err); } + sample = &pevent->sample; +if (machine && sample->callchain) { +struct addr_location al; +struct callchain_cursor *cursor; +u64 i; +struct pyrf_callchain *pchain; + +addr_location__init (&al);
tools/perf/util/python.c static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *ev · line 1259
Quoted 10 lines
tools/perf/util/python.c +10
-0+if (machine__resolve (machine, &al, sample) >=0 ) { + cursor =get_tls_callchain_cursor (); +if (thread__resolve_callchain (al.thread, cursor, sample, +NULL ,NULL ,PERF_MAX_STACK_DEPTH ) ==0 ) { +callchain_cursor_commit (cursor); + + pchain = PyObject_New(struct pyrf_callchain, &pyrf_callchain__type); +if (pchain) { + pchain->pevent = pevent; +Py_INCREF (pevent);
[Severity: High]
Does this create a circular reference that leaks memory?
The pchain object holds a strong reference to pevent via Py_INCREF, and then pevent takes ownership of pchain by storing it in pevent->callchain below.
Since neither pyrf_event__type nor pyrf_callchain__type implement Python's garbage collector traversal functions (tp_traverse, tp_clear), this cycle is invisible to the GC. When processing large files, will this permanently leak the event object, the callchain object, and all associated kernel map references for every event with a callchain?
Quoted 27 lines
tools/perf/util/python.c +25
-0+ pchain->nr_frames = cursor->nr; + pchain->pos =0 ; + pchain->resolved = true; + pchain->frames =calloc (pchain->nr_frames, +sizeof (*pchain->frames)); +if (pchain->frames) { +struct callchain_cursor_node *node; + +for (i =0 ; i < pchain->nr_frames; i++) { + node =callchain_cursor_current (cursor); + pchain->frames[i].ip = node->ip; + pchain->frames[i].map = +map__get (node->ms.map); + pchain->frames[i].sym = node->ms.sym; +callchain_cursor_advance (cursor); + } + pevent->callchain = (PyObject *)pchain; + }else { +Py_DECREF (pchain); + } + } + } +addr_location__exit (&al); + } + }return (PyObject *)pevent; }
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=17
Ian Rogers
reviewerPATCH v9 18/23 perf python: Extend API for stat events in python.c
In reply to Ian RogersAdd stat information to the session. Add call backs for stat events.
---
v5:
1. Fix Memory Corruption: Corrected the memory offset for stat_round_type in pyrf_stat_round_event__members by adding the base offset of struct pyrf_event.
2. Fix Memory Leak: Added Py_XDECREF() to free the unused return value of the Python callback in pyrf_session_tool__stat_round().
---
v7:
- Added comprehensive size checks in pyrf_event__new for all event types.
- Fixed line length warning.
--- tools/perf/util/python.c | 180 +++++++++++++++++++++++++++++++++++++-- 1 file changed, 173 insertions(+), 7 deletions(-)
tools/perf/util/python.c +173 -7
@@ -343,6 +343,77 @@ static PyTypeObject pyrf_lost_event__type = { .tp_repr = (reprfunc)pyrf_lost_event__repr, }; +static const char pyrf_stat_event__doc[] =PyDoc_STR ("perf stat event object." ); + +static PyMemberDef pyrf_stat_event__members[] = { + sample_members +member_def (perf_event_header, type,T_UINT ,"event type" ), +member_def (perf_record_stat, id,T_ULONGLONG ,"event id" ), +member_def (perf_record_stat, cpu,T_UINT ,"event cpu" ), +member_def (perf_record_stat, thread,T_UINT ,"event thread" ), +member_def (perf_record_stat, val,T_ULONGLONG ,"counter value" ), +member_def (perf_record_stat, ena,T_ULONGLONG ,"enabled time" ), +member_def (perf_record_stat, run,T_ULONGLONG ,"running time" ), + { .name =NULL , }, +}; + +static PyObject *pyrf_stat_event__repr (const struct pyrf_event *pevent) +{ +return PyUnicode_FromFormat ( +"{ type: stat, id: %llu, cpu: %u, thread: %u, val: %llu, ena: %llu, run: %llu }" , + pevent->event.stat.id, + pevent->event.stat.cpu, + pevent->event.stat.thread, + pevent->event.stat.val, + pevent->event.stat.ena, + pevent->event.stat.run); +} + +static PyTypeObject pyrf_stat_event__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.stat_event" , + .tp_basicsize =sizeof (struct pyrf_event ), + .tp_new = PyType_GenericNew, + .tp_dealloc = (destructor )pyrf_event__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_stat_event__doc, + .tp_members = pyrf_stat_event__members, + .tp_getset = pyrf_event__getset, + .tp_repr = (reprfunc )pyrf_stat_event__repr, +}; + +static const char pyrf_stat_round_event__doc[] =PyDoc_STR ("perf stat round event object." ); + +static PyMemberDef pyrf_stat_round_event__members[] = { + sample_members +member_def (perf_event_header, type,T_UINT ,"event type" ), + { .name ="stat_round_type" , .type =T_ULONGLONG , + .offset = offsetof(struct pyrf_event , event) + offsetof(struct perf_record_stat_round , type), + .doc ="round type" }, +member_def (perf_record_stat_round, time,T_ULONGLONG ,"round time" ), + { .name =NULL , }, +}; + +static PyObject *pyrf_stat_round_event__repr (const struct pyrf_event *pevent) +{ +return PyUnicode_FromFormat ("{ type: stat_round, type: %llu, time: %llu }" , + pevent->event.stat_round.type, + pevent->event.stat_round.time); +} + +static PyTypeObject pyrf_stat_round_event__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.stat_round_event" , + .tp_basicsize =sizeof (struct pyrf_event ), + .tp_new = PyType_GenericNew, + .tp_dealloc = (destructor )pyrf_event__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_stat_round_event__doc, + .tp_members = pyrf_stat_round_event__members, + .tp_getset = pyrf_event__getset, + .tp_repr = (reprfunc )pyrf_stat_round_event__repr, +}; +static const char pyrf_read_event__doc[] =PyDoc_STR ("perf read event object." );static PyMemberDef pyrf_read_event__members[] = {
@@ -1034,6 +1105,12 @@ static int pyrf_event__setup_types(void)if (err <0 ) goto out; err =PyType_Ready (&pyrf_context_switch_event__type); +if (err <0 ) + goto out; + err =PyType_Ready (&pyrf_stat_event__type); +if (err <0 ) + goto out; + err =PyType_Ready (&pyrf_stat_round_event__type);if (err <0 ) goto out; err =PyType_Ready (&pyrf_callchain_node__type);
@@ -1059,6 +1136,8 @@ static PyTypeObject *pyrf_event__type[] = { [PERF_RECORD_SAMPLE] = &pyrf_sample_event__type, [PERF_RECORD_SWITCH] = &pyrf_context_switch_event__type, [PERF_RECORD_SWITCH_CPU_WIDE] = &pyrf_context_switch_event__type, + [PERF_RECORD_STAT] = &pyrf_stat_event__type, + [PERF_RECORD_STAT_ROUND] = &pyrf_stat_round_event__type, }; static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evsel,
@@ -1091,13 +1170,13 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evcase PERF_RECORD_EXIT : min_size =sizeof (struct perf_record_fork );break ; +case PERF_RECORD_LOST : + min_size =sizeof (struct perf_record_lost ); +break ;case PERF_RECORD_THROTTLE :case PERF_RECORD_UNTHROTTLE : min_size =sizeof (struct perf_record_throttle );break ; -case PERF_RECORD_LOST : - min_size =sizeof (struct perf_record_lost ); -break ;case PERF_RECORD_READ : min_size =sizeof (struct perf_record_read );break ;
@@ -2119,7 +2198,40 @@ static PyObject *pyrf_evsel__get_attr_wakeup_events(PyObject *self, void */*closreturn PyLong_FromUnsignedLong (pevsel->evsel->core.attr.wakeup_events); } +static PyObject *pyrf_evsel__get_ids (struct pyrf_evsel *pevsel,void *closure__maybe_unused ) +{ +struct evsel *evsel = pevsel->evsel; +PyObject *list =PyList_New (0 ); + +if (!list) +return NULL ; + +for (u32 i =0 ; i < evsel->core.ids; i++) { +PyObject *id =PyLong_FromUnsignedLongLong (evsel->core.id[i]); +int ret; + +if (!id) { +Py_DECREF (list); +return NULL ; + } + ret =PyList_Append (list, id); +Py_DECREF (id); +if (ret <0 ) { +Py_DECREF (list); +return NULL ; + } + } + +return list; +} +static PyGetSetDef pyrf_evsel__getset[] = { + { + .name ="ids" , + .get = (getter )pyrf_evsel__get_ids, + .set =NULL , + .doc ="event IDs." , + }, { .name ="tracking" , .get = pyrf_evsel__get_tracking,
@@ -2894,6 +3006,8 @@ static const struct perf_constant perf__constants[] = {PERF_CONST (RECORD_LOST_SAMPLES ),PERF_CONST (RECORD_SWITCH ),PERF_CONST (RECORD_SWITCH_CPU_WIDE ), +PERF_CONST (RECORD_STAT ), +PERF_CONST (RECORD_STAT_ROUND ),PERF_CONST (RECORD_MISC_SWITCH_OUT ), { .name = NULL, },
@@ -3270,6 +3384,7 @@ struct pyrf_session {struct perf_tool tool;struct pyrf_data *pdata;PyObject *sample; +PyObject *stat; }; static int pyrf_session_tool__sample(const struct perf_tool *tool,
@@ -3294,6 +3409,50 @@ static int pyrf_session_tool__sample(const struct perf_tool *tool,return 0 ; } +static int pyrf_session_tool__stat (const struct perf_tool *tool, +struct perf_session *session, +union perf_event *event) +{ + struct pyrf_session *psession = container_of(tool, struct pyrf_session, tool); +struct evsel *evsel =evlist__id2evsel (session->evlist, event->stat.id); +PyObject *pyevent =pyrf_event__new (event, evsel, psession->session); +const char *name = evsel ?evsel__name (evsel) :"unknown" ; +PyObject *ret; + +if (pyevent ==NULL ) +return -ENOMEM ; + + ret =PyObject_CallFunction (psession->stat,"Os" , pyevent, name); +if (!ret) { +Py_DECREF (pyevent); +return -1 ; + } +Py_DECREF (ret); +Py_DECREF (pyevent); +return 0 ; +} + +static int pyrf_session_tool__stat_round (const struct perf_tool *tool, +struct perf_session *session__maybe_unused , +union perf_event *event) +{ + struct pyrf_session *psession = container_of(tool, struct pyrf_session, tool); +PyObject *pyevent =pyrf_event__new (event,/*evsel=*/ NULL , psession->session); +PyObject *ret; + +if (pyevent ==NULL ) +return -ENOMEM ; + + ret =PyObject_CallFunction (psession->stat,"O" , pyevent); +if (!ret) { +Py_DECREF (pyevent); +return -1 ; + } +Py_DECREF (ret); +Py_DECREF (pyevent); +return 0 ; +} +static PyObject *pyrf_session__find_thread (struct pyrf_session *psession,PyObject *args) {struct machine *machine;
@@ -3325,13 +3484,13 @@ static PyObject *pyrf_session__find_thread(struct pyrf_session *psession, PyObjestatic PyObject *pyrf_session__new (PyTypeObject *type,PyObject *args,PyObject *kwargs) {struct pyrf_data *pdata; -PyObject *sample =NULL ; -static char *kwlist[] = {"data" ,"sample" ,NULL }; +PyObject *sample =NULL , *stat =NULL ; +static char *kwlist[] = {"data" ,"sample" ,"stat" ,NULL };struct pyrf_session *psession;struct perf_session *session; -if (!PyArg_ParseTupleAndKeywords (args, kwargs,"O!|O" , kwlist, &pyrf_data__type, &pdata, - &sample)) +if (!PyArg_ParseTupleAndKeywords (args, kwargs,"O!|OO" , kwlist, &pyrf_data__type, &pdata, + &sample, &stat))return NULL ; psession = PyObject_New(struct pyrf_session, type);
@@ -3340,6 +3499,7 @@ static PyObject *pyrf_session__new(PyTypeObject *type, PyObject *args, PyObject psession->session =NULL ; psession->sample =NULL ; + psession->stat =NULL ; psession->pdata =NULL ;Py_INCREF (pdata);
@@ -3362,8 +3522,13 @@ static PyObject *pyrf_session__new(PyTypeObject *type, PyObject *args, PyObject }while (0 )ADD_TOOL (sample); +ADD_TOOL (stat);#undef ADD_TOOL +if (stat) + psession->tool.stat_round = pyrf_session_tool__stat_round; + + psession->tool.comm = perf_event__process_comm; psession->tool.mmap = perf_event__process_mmap; psession->tool.mmap2 = perf_event__process_mmap2;
@@ -3403,6 +3568,7 @@ static void pyrf_session__delete(struct pyrf_session *psession)perf_session__delete (psession->session);Py_XDECREF (psession->pdata);Py_XDECREF (psession->sample); +Py_XDECREF (psession->stat);Py_TYPE (psession)->tp_free ((PyObject *)psession); }
Signature · 1 line
2.54.0.794.g4f17f83d09-goog
[email protected]
AI reviewRe: [PATCH v9 18/23] perf python: Extend API for stat events in python.c
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider: - [Medium] The same Python callback is used for both 'stat' and 'stat_round' events, but is invoked with…
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [Medium] The same Python callback is used for both 'stat' and 'stat_round' events, but is invoked with different numbers of arguments.
- [Medium] The new 'perf.stat_event' and 'perf.stat_round_event' types are not added to the Python module namespace. --
commit 98e6cac4d71c905dd640dddfae6ccd27bae799e7
Author: Ian Rogers <[email protected]> Date: Fri May 22 22:04:29 2026 +0000
perf python: Extend API for stat events in python.c
This commit adds stat information to the perf python session API. It introduces definitions for stat and stat_round event types and links them into the perf tool callbacks.
Quoted 4 lines
tools/perf/util/python.c +0
-0
[ ... ]
tools/perf/util/python.c static int pyrf_event__setup_types(void) · line 1113
Quoted 10 lines
tools/perf/util/python.c +6
-0@@ -1034,6 +1105,12 @@ static int pyrf_event__setup_types(void)if (err <0 ) goto out; err =PyType_Ready (&pyrf_context_switch_event__type); +if (err <0 ) + goto out; + err =PyType_Ready (&pyrf_stat_event__type); +if (err <0 ) + goto out; + err =PyType_Ready (&pyrf_stat_round_event__type);
[Severity: Medium]
Will these new event types be missing from the Python module namespace?
It looks like the types are prepared here, but they aren't exported using PyModule_AddObject() in the module initialization function like the other event types. This would prevent users from accessing them directly (for instance, to check isinstance(event, perf.stat_event)).
Quoted 3 lines
tools/perf/util/python.c +0
-0if (err <0 ) goto out; err =PyType_Ready (&pyrf_callchain_node__type);
[ ... ]
Quoted 8 lines
tools/perf/util/python.c +4
-0@@ -3294,6 +3409,50 @@ static int pyrf_session_tool__sample(const struct perf_tool *tool,return 0 ; } +static int pyrf_session_tool__stat (const struct perf_tool *tool, +struct perf_session *session, +union perf_event *event) +{
tools/perf/util/python.c static int pyrf_session_tool__sample(const struct perf_tool *tool, · line 3425
Quoted 10 lines
tools/perf/util/python.c +10
-0+ struct pyrf_session *psession = container_of(tool, struct pyrf_session, tool); +struct evsel *evsel =evlist__id2evsel (session->evlist, event->stat.id); +PyObject *pyevent =pyrf_event__new (event, evsel, psession->session); +const char *name = evsel ?evsel__name (evsel) :"unknown" ; +PyObject *ret; + +if (pyevent ==NULL ) +return -ENOMEM ; + + ret =PyObject_CallFunction (psession->stat,"Os" , pyevent, name);
[Severity: Medium]
Since both PERF_RECORD_STAT and PERF_RECORD_STAT_ROUND share the same psession->stat callback, will calling it with two arguments here ("Os") and one argument below ("O") cause issues?
If a Python function is defined to take two arguments for stat events, it will raise a TypeError when called with one argument for stat_round events, which would abort the perf session event processing unless the user implements a workaround like *args.
Quoted 30 lines
tools/perf/util/python.c +30
-0+if (!ret) { +Py_DECREF (pyevent); +return -1 ; + } +Py_DECREF (ret); +Py_DECREF (pyevent); +return 0 ; +} + +static int pyrf_session_tool__stat_round (const struct perf_tool *tool, +struct perf_session *session__maybe_unused , +union perf_event *event) +{ + struct pyrf_session *psession = container_of(tool, struct pyrf_session, tool); +PyObject *pyevent =pyrf_event__new (event,/*evsel=*/ NULL , psession->session); +PyObject *ret; + +if (pyevent ==NULL ) +return -ENOMEM ; + + ret =PyObject_CallFunction (psession->stat,"O" , pyevent); +if (!ret) { +Py_DECREF (pyevent); +return -1 ; + } +Py_DECREF (ret); +Py_DECREF (pyevent); +return 0 ; +} +
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=18
Ian Rogers
reviewerPATCH v9 19/23 perf python: Expose brstack in sample event
In reply to Ian RogersImplement pyrf_branch_entry and pyrf_branch_stack for lazy iteration over branch stack entries.
---
v2:
1. Avoided Keyword Collision: Renamed the properties "from" and "to" to "from_ip" and "to_ip" in pyrf_branch_entry__getset[] to prevent syntax errors in Python.
2. Eager Branch Stack Resolution: Updated pyrf_session_tool__sample() to allocate and copy the branch stack entries eagerly when the event is processed, instead of deferring it to iteration time. This avoids reading from potentially overwritten or unmapped mmap buffers.
3. Updated Iterators: Updated pyrf_branch_stack and pyrf_branch_stack__next() to use the copied entries rather than pointing directly to the sample's buffer.
4. Avoided Reference Leak on Init Failure: Added proper error checking for PyModule_AddObject() in the module initialization function, decrementing references on failure.
v6:
- Moved branch stack resolution from session_tool__sample to pyrf_event__new.
--- tools/perf/util/python.c | 188 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 188 insertions(+)
tools/perf/util/python.c +188 -0
@@ -19,6 +19,7 @@#include "debug.h" #include "dso.h" #include "event.h" +#include "branch.h" #include "evlist.h" #include "evsel.h" #include "expr.h"
@@ -68,6 +69,8 @@ struct pyrf_event {bool al_resolved;/** @callchain: Resolved callchain, eagerly computed if requested. */ PyObject *callchain; +/** @brstack: Resolved branch stack, eagerly computed if requested. */ +PyObject *brstack;/** @event: The underlying perf_event that may be in a file or ring buffer. */ union perf_event event; };
@@ -106,6 +109,7 @@ static void pyrf_event__delete(struct pyrf_event *pevent)if (pevent->al_resolved)addr_location__exit (&pevent->al);Py_XDECREF (pevent->callchain); +Py_XDECREF (pevent->brstack);perf_sample__exit (&pevent->sample);Py_TYPE (pevent)->tp_free ((PyObject *)pevent); }
@@ -918,6 +922,144 @@ static PyObject *pyrf_sample_event__get_callchain(PyObject *self, void *closurereturn pevent->callchain; } +struct pyrf_branch_entry { +PyObject_HEAD + u64 from; +u64 to; +struct branch_flags flags; +}; + +static void pyrf_branch_entry__delete (struct pyrf_branch_entry *pentry) +{ +Py_TYPE (pentry)->tp_free ((PyObject *)pentry); +} + +static PyObject *pyrf_branch_entry__get_from (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyLong_FromUnsignedLongLong (pentry->from); +} + +static PyObject *pyrf_branch_entry__get_to (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyLong_FromUnsignedLongLong (pentry->to); +} + +static PyObject *pyrf_branch_entry__get_mispred (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyBool_FromLong (pentry->flags.mispred); +} + +static PyObject *pyrf_branch_entry__get_predicted (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyBool_FromLong (pentry->flags.predicted); +} + +static PyObject *pyrf_branch_entry__get_in_tx (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyBool_FromLong (pentry->flags.in_tx); +} + +static PyObject *pyrf_branch_entry__get_abort (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyBool_FromLong (pentry->flags.abort); +} + +static PyObject *pyrf_branch_entry__get_cycles (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyLong_FromUnsignedLongLong (pentry->flags.cycles); +} + +static PyObject *pyrf_branch_entry__get_type (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyLong_FromLong (pentry->flags.type); +} + +static PyGetSetDef pyrf_branch_entry__getset[] = { + { .name ="from_ip" , .get = (getter )pyrf_branch_entry__get_from, }, + { .name ="to_ip" , .get = (getter )pyrf_branch_entry__get_to, }, + { .name ="mispred" , .get = (getter )pyrf_branch_entry__get_mispred, }, + { .name ="predicted" , .get = (getter )pyrf_branch_entry__get_predicted, }, + { .name ="in_tx" , .get = (getter )pyrf_branch_entry__get_in_tx, }, + { .name ="abort" , .get = (getter )pyrf_branch_entry__get_abort, }, + { .name ="cycles" , .get = (getter )pyrf_branch_entry__get_cycles, }, + { .name ="type" , .get = (getter )pyrf_branch_entry__get_type, }, + { .name =NULL , }, +}; + +static PyTypeObject pyrf_branch_entry__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.branch_entry" , + .tp_basicsize =sizeof (struct pyrf_branch_entry ), + .tp_dealloc = (destructor )pyrf_branch_entry__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc ="perf branch entry object." , + .tp_getset = pyrf_branch_entry__getset, +}; + +struct pyrf_branch_stack { +PyObject_HEAD + struct pyrf_event *pevent; +struct branch_entry *entries; +u64 nr; +u64 pos; +}; + +static void pyrf_branch_stack__delete (struct pyrf_branch_stack *pstack) +{ +Py_XDECREF (pstack->pevent); +free (pstack->entries); +Py_TYPE (pstack)->tp_free ((PyObject *)pstack); +} + +static PyObject *pyrf_branch_stack__next (struct pyrf_branch_stack *pstack) +{ +struct pyrf_branch_entry *pentry; + +if (pstack->pos >= pstack->nr) +return NULL ; + + pentry = PyObject_New(struct pyrf_branch_entry, &pyrf_branch_entry__type); +if (!pentry) +return NULL ; + + pentry->from = pstack->entries[pstack->pos].from; + pentry->to = pstack->entries[pstack->pos].to; + pentry->flags = pstack->entries[pstack->pos].flags; + + pstack->pos++; +return (PyObject *)pentry; +} + +static PyTypeObject pyrf_branch_stack__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.branch_stack" , + .tp_basicsize =sizeof (struct pyrf_branch_stack ), + .tp_dealloc = (destructor )pyrf_branch_stack__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc ="perf branch stack object." , + .tp_iter = PyObject_SelfIter, + .tp_iternext = (iternextfunc )pyrf_branch_stack__next, +}; + +static PyObject *pyrf_sample_event__get_brstack (PyObject *self,void *closure__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; + +if (!pevent->brstack) + Py_RETURN_NONE; + +Py_INCREF (pevent->brstack); +return pevent->brstack; +} +static PyObject *pyrf_sample_event__getattro (struct pyrf_event *pevent,PyObject *attr_name) {
@@ -938,6 +1080,12 @@ static PyGetSetDef pyrf_sample_event__getset[] = { .set = NULL, .doc = "event callchain.", }, + { + .name = "brstack", + .get = pyrf_sample_event__get_brstack, + .set = NULL, + .doc = "event branch stack.", + }, { .name = "raw_buf", .get = (getter)pyrf_sample_event__get_raw_buf,
@@ -1119,6 +1267,12 @@ static int pyrf_event__setup_types(void) err =PyType_Ready (&pyrf_callchain__type);if (err <0 ) goto out; + err =PyType_Ready (&pyrf_branch_entry__type); +if (err <0 ) + goto out; + err =PyType_Ready (&pyrf_branch_stack__type); +if (err <0 ) + goto out; out:return err; }
@@ -1305,6 +1459,7 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evperf_sample__init (&pevent->sample,/*all=*/ true); pevent->callchain =NULL ; + pevent->brstack =NULL ; pevent->al_resolved = false;addr_location__init (&pevent->al);
@@ -1361,6 +1516,27 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evaddr_location__exit (&al); } } +if (sample->branch_stack) { +struct branch_stack *bs = sample->branch_stack; +struct branch_entry *entries =perf_sample__branch_entries (sample); +struct pyrf_branch_stack *pstack; + + pstack = PyObject_New(struct pyrf_branch_stack, &pyrf_branch_stack__type); +if (pstack) { +Py_INCREF (pevent); + pstack->pevent = pevent; + pstack->pos =0 ; + pstack->nr = bs->nr; + pstack->entries =calloc (bs->nr,sizeof (struct branch_entry )); +if (pstack->entries) { +memcpy (pstack->entries, entries, + bs->nr *sizeof (struct branch_entry )); + pevent->brstack = (PyObject *)pstack; + }else { +Py_DECREF (pstack); + } + } + }return (PyObject *)pevent; }
@@ -3742,6 +3918,18 @@ PyMODINIT_FUNC PyInit_perf(void)Py_INCREF (&pyrf_session__type);PyModule_AddObject (module,"session" , (PyObject *)&pyrf_session__type); +Py_INCREF (&pyrf_branch_entry__type); +if (PyModule_AddObject (module,"branch_entry" , (PyObject *)&pyrf_branch_entry__type) <0 ) { +Py_DECREF (&pyrf_branch_entry__type); + goto error; + } + +Py_INCREF (&pyrf_branch_stack__type); +if (PyModule_AddObject (module,"branch_stack" , (PyObject *)&pyrf_branch_stack__type) <0 ) { +Py_DECREF (&pyrf_branch_stack__type); + goto error; + } + dict =PyModule_GetDict (module);if (dict ==NULL ) goto error;
Signature · 1 line
2.54.0.794.g4f17f83d09-goog
[email protected]
AI reviewRe: [PATCH v9 19/23] perf python: Expose brstack in sample event
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider: - [High] An unbreakable circular reference between `pyrf_event` and `pyrf_branch_stack` causes a memory…
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [High] An unbreakable circular reference between pyrf_event and pyrf_branch_stack causes a memory leak for every event containing a branch stack.
- [Medium] Returning a valid object while a Python exception is set triggers a fatal SystemError in the interpreter.
--
commit c44b624d7ee17549d2a93cdf268bc3540d223c20
Author: Ian Rogers <[email protected]>
perf python: Expose brstack in sample event
This commit implements pyrf_branch_entry and pyrf_branch_stack objects in the perf python extension. This allows lazy iteration over branch stack entries in a sample event.
Quoted 4 lines
tools/perf/util/python.c +0
-0
[ ... ]
tools/perf/util/python.c static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *ev · line 1524
Quoted 10 lines
tools/perf/util/python.c +6
-0@@ -1361,6 +1516,27 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evaddr_location__exit (&al); } } +if (sample->branch_stack) { +struct branch_stack *bs = sample->branch_stack; +struct branch_entry *entries =perf_sample__branch_entries (sample); +struct pyrf_branch_stack *pstack; + + pstack = PyObject_New(struct pyrf_branch_stack, &pyrf_branch_stack__type);
[Severity: Medium]
If PyObject_New() fails to allocate memory for pstack, it returns NULL and sets a Python MemoryError exception. Because the code falls through and returns pevent below, does returning a valid object while an exception is set trigger a fatal SystemError in the Python interpreter?
To handle OOM without crashing the interpreter, it seems the code needs to either propagate the error by returning NULL, or clear the exception using PyErr_Clear() before returning pevent.
tools/perf/util/python.c static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *ev · line 1534
Quoted 10 lines
tools/perf/util/python.c +10
-0+if (pstack) { +Py_INCREF (pevent); + pstack->pevent = pevent; + pstack->pos =0 ; + pstack->nr = bs->nr; + pstack->entries =calloc (bs->nr,sizeof (struct branch_entry )); +if (pstack->entries) { +memcpy (pstack->entries, entries, + bs->nr *sizeof (struct branch_entry )); + pevent->brstack = (PyObject *)pstack;
[Severity: High]
Does this create an unbreakable circular reference between pevent and pstack?
pstack holds a reference to pevent via Py_INCREF(pevent), and pevent holds a reference to pstack via pevent->brstack = (PyObject *)pstack. Since neither type uses the Python cyclic garbage collector (via Py_TPFLAGS_HAVE_GC and tp_traverse), it appears this cycle cannot be broken by Python.
This could cause an indefinite memory leak of the event object, the branch stack object, and the dynamically allocated pstack->entries array for every event containing a branch stack.
Quoted 7 lines
tools/perf/util/python.c +5
-0+ }else { +Py_DECREF (pstack); + } + } + }return (PyObject *)pevent; }
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=19
Ian Rogers
reviewerPATCH v9 20/23 perf python: Add syscall name/id to convert syscall number and name
In reply to Ian RogersUse perf's syscalltbl support to convert syscall number to name assuming the number is for the host machine. This avoids python libaudit support as tools/perf/scripts/python/syscall-counts.py requires.
---
v2:
1. Guarded with HAVE_LIBTRACEEVENT : Wrapped the syscall functions and their entries in perf__methods with #ifdef HAVE_LIBTRACEEVENT to avoid potential linker errors if CONFIG_TRACE is disabled.
2. Changed Exception Type: Updated pyrf__syscall_id to raise a ValueError instead of a TypeError when a syscall is not found.
3. Fixed Typo: Corrected "name number" to "name" in the docstring for syscall_id.
v6:
- Added optional keyword-only elf_machine argument to syscall_name and syscall_id.
v7:
- Made syscalltbl.o unconditional in Build to fix undefined symbol when building without libtraceevent.
--- tools/perf/util/Build | 1 - tools/perf/util/python.c | 48 ++++++++++++++++++++++++++++++++++++++++ 2 files changed, 48 insertions(+), 1 deletion(-)
tools/perf/util/Build +0 -1
@@ -75,7 +75,6 @@ perf-util-y += sample.o perf-util-y += sample-raw.o perf-util-y += s390-sample-raw.o perf-util-y += amd-sample-raw.o - perf-util-y += ordered-events.o perf-util-y += namespaces.o perf-util-y += comm.o
tools/perf/util/python.c +48 -0
72 lines not on this page · raw message
Signature · 1 line
2.54.0.794.g4f17f83d09-goog
[email protected]
AI reviewRe: [PATCH v9 20/23] perf python: Add syscall name/id to convert syscall number and name
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider: - [Medium] Incomplete implementation: Python scripts are not updated to use the new syscall API. - [Low]…
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [Medium] Incomplete implementation: Python scripts are not updated to use the new syscall API.
- [Low] Incompatible pointer type warning causing build failure under -Werror. --
commit e9d75f77bbf3346ce89ce6e85ac1c93ce060af99
Author: Ian Rogers <[email protected]>
perf python: Add syscall name/id to convert syscall number and name
This commit adds Python bindings for syscall name and ID conversion using perf's syscalltbl support. It aims to avoid the need for python libaudit support in syscall-counts.py.
Quoted 6 lines
perf python: Add syscall name/id to convert syscall number and name
Use perf's syscalltbl support to convert syscall number to name assuming the number is for the host machine. This avoids python libaudit support as
tools/perf/scripts/python/syscall-counts.pyrequires.
[Severity: Medium]
The commit message states this patch avoids python libaudit support for syscall-counts.py, but the diff doesn't appear to update syscall-counts.py or its helper scripts to use this new API.
Are the script updates missing from this patch?
Quoted 4 lines
tools/perf/util/python.c +0
-0
[ ... ]
tools/perf/util/python.c static int pyrf_session__setup_types(void) · line 3802
Quoted 10 lines
tools/perf/util/python.c +6
-0@@ -3792,6 +3794,40 @@ static int pyrf_session__setup_types(void)return PyType_Ready (&pyrf_session__type); } +static PyObject *pyrf__syscall_name (PyObject *self,PyObject *args,PyObject *kwargs) +{ +const char *name; +int id; +int elf_machine =EM_HOST ; +static char *const kwlist[] = {"id" ,"elf_machine" ,NULL };
[Severity: Low]
Will this cause an incompatible pointer type warning?
PyArg_ParseTupleAndKeywords() expects either a char ** or a const char * const * depending on the python version.
Passing a char * const * discards the const qualifier and might trigger a build failure since perf builds with -Werror.
Quoted 3 lines
tools/perf/util/python.c +3
-0+ +if (!PyArg_ParseTupleAndKeywords (args, kwargs,"i|$i" , kwlist, &id, &elf_machine)) +return NULL ;
[ ... ]
tools/perf/util/python.c static int pyrf_session__setup_types(void) · line 3818
Quoted 6 lines
tools/perf/util/python.c +6
-0+static PyObject *pyrf__syscall_id (PyObject *self,PyObject *args,PyObject *kwargs) +{ +const char *name; +int id; +int elf_machine =EM_HOST ; +static char *const kwlist[] = {"name" ,"elf_machine" ,NULL };
[Severity: Low]
Does this have the same discarded qualifier issue with the kwlist type as seen in pyrf__syscall_name() above?
Quoted 3 lines
tools/perf/util/python.c +3
-0+ +if (!PyArg_ParseTupleAndKeywords (args, kwargs,"s|$i" , kwlist, &name, &elf_machine)) +return NULL ;
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=20
Ian Rogers
reviewerPATCH v9 21/23 perf python: Add config file access
In reply to Ian RogersAdd perf.config_get(name) to expose the perf configuration system.
--- tools/perf/util/python.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+)
tools/perf/util/python.c +20 -0
@@ -14,6 +14,7 @@#include "build-id.h" #include "callchain.h" #include "comm.h" +#include "config.h" #include "counts.h" #include "data.h" #include "debug.h"
@@ -3828,7 +3829,26 @@ static PyObject *pyrf__syscall_id(PyObject *self, PyObject *args, PyObject *kwarreturn PyLong_FromLong (id); } +static PyObject *pyrf__config_get (PyObject *self,PyObject *args) +{ +const char *config_name, *val; + +if (!PyArg_ParseTuple (args,"s" , &config_name)) +return NULL ; + + val =perf_config_get (config_name); +if (!val) + Py_RETURN_NONE; +return PyUnicode_FromString (val); +} +static PyMethodDef perf__methods[] = { + { + .ml_name ="config_get" , + .ml_meth = (PyCFunction ) pyrf__config_get, + .ml_flags =METH_VARARGS , + .ml_doc =PyDoc_STR ("Get a perf config value." ) + }, { .ml_name ="metrics" , .ml_meth = (PyCFunction ) pyrf__metrics,
Signature · 1 line
2.54.0.794.g4f17f83d09-goog
Ian Rogers
reviewerPATCH v9 22/23 perf python: Add perf.pyi stubs file
In reply to Ian RogersAdd Python type stubs for the perf module to improve IDE support and static analysis. Includes docstrings for classes, methods, and constants derived from C source and JSON definitions.
---
v2:
1. Added Missing Module Functions: Added parse_metrics and pmus. Renamed metrics to parse_metrics to match python.c .
2. Added Constructors: Added __init__ methods for data , evsel , and evlist with their appropriate arguments.
3. Removed sample_comm : Removed it from sample_event since it is not exported in python.c .
4. Keyword Handling in branch_entry : I used from_ip and to_ip in the stubs to match the rename I did in python.c (in turn 145) to avoid the Python from keyword conflict.
5. Added Missing Event Classes: Added mmap_event , lost_event , comm_event , task_event , throttle_event , read_event , and switch_event .
6. Added Missing evlist Methods: Added get_pollfd and add .
7. Updated Return Types: Changed process_events to return int .
v6:
- Updated perf.pyi to use find_thread and elf_machine.
v8:
- Added mmap2_event class and new evsel attributes to perf.pyi.
- Added pid, tid, ppid, cpu attributes to class thread in perf.pyi.
--- tools/perf/python/perf.pyi | 605 +++++++++++++++++++++++++++++++++++++ 1 file changed, 605 insertions(+) create mode 100644 tools/perf/python/perf.pyi
tools/perf/python/perf.pyi new file +605 -0
@@ -0,0 +1,605 @@ +"""Type stubs for the perf Python module.""" +from typing import Callable, Dict, List, Optional, Any, Iterator + +def config_get(name: str) -> Optional[str]: + """Get a configuration value from perf config. + + Args: + name: The configuration variable name (e.g., 'colors.top'). + + Returns: + The configuration value as a string, or None if not set. + """ + ... + +def metrics() -> List[Dict[str, str]]: + """Get a list of available metrics. + + Returns: + A list of dictionaries, each describing a metric. + """ + ... + +def syscall_name(sc_id: int, *, elf_machine: Optional[int] = None) -> str: + """Convert a syscall number to its name. + + Args: + sc_id: The syscall number. + elf_machine: Optional ELF machine type. + + Returns: + The name of the syscall. + """ + ... + +def syscall_id(name: str, *, elf_machine: Optional[int] = None) -> int: + """Convert a syscall name to its number. + + Args: + name: The syscall name. + elf_machine: Optional ELF machine type. + + Returns: + The number of the syscall. + """ + ... + +def parse_events( + event_string: str, + cpus: Optional[cpu_map] = None, + threads: Optional[Any] = None +) -> 'evlist': + """Parse an event string and return an evlist. + + Args: + event_string: The event string (e.g., 'cycles,instructions'). + cpus: Optional CPU map to bind events to. + threads: Optional thread map to bind events to. + + Returns: + An evlist containing the parsed events. + """ + ... + +def parse_metrics(metrics_string: str) -> 'evlist': + """Parse a string of metrics or metric groups and return an evlist.""" + ... + +def pmus() -> Iterator[Any]: + """Returns a sequence of pmus.""" + ... + +class data: + """Represents a perf data file.""" + def __init__(self, path: str = ..., fd: int = ...) -> None: ... + +class thread: + """Represents a thread in the system.""" + def comm(self) -> str: + """Get the command name of the thread.""" + ... + pid: int + tid: int + ppid: int + cpu: int + +class counts_values: + """Raw counter values.""" + val: int + ena: int + run: int + +class thread_map: + """Map of threads being monitored.""" + def __init__(self, pid: int = -1, tid: int = -1) -> None: + """Initialize a thread map. + + Args: + pid: Process ID to monitor (-1 for all). + tid: Thread ID to monitor (-1 for all). + """ + ... + def __len__(self) -> int: ... + def __getitem__(self, index: int) -> int: ... + def __iter__(self) -> Iterator[int]: ... + +class evsel: + """Event selector, represents a single event being monitored.""" + def __init__( + self, + type: int = ..., + config: int = ..., + sample_freq: int = ..., + sample_period: int = ..., + sample_type: int = ..., + read_format: int = ..., + disabled: bool = ..., + inherit: bool = ..., + pinned: bool = ..., + exclusive: bool = ..., + exclude_user: bool = ..., + exclude_kernel: bool = ..., + exclude_hv: bool = ..., + exclude_idle: bool = ..., + mmap: bool = ..., + context_switch: bool = ..., + comm: bool = ..., + freq: bool = ..., + idx: int = ..., + ) -> None: ... + def __str__(self) -> str: + """Return string representation of the event.""" + ... + def open(self) -> None: + """Open the event selector file descriptor table.""" + ... + def read(self, cpu: int, thread: int) -> counts_values: + """Read counter values for a specific CPU and thread.""" + ... + ids: List[int] + def cpus(self) -> cpu_map: + """Get CPU map for this event.""" + ... + def threads(self) -> thread_map: + """Get thread map for this event.""" + ... + tracking: bool + config: int + read_format: int + sample_period: int + sample_type: int + size: int + type: int + wakeup_events: int + + +class sample_event: + """Represents a sample event from perf.""" + evsel: evsel + sample_cpu: int + sample_time: int + sample_pid: int + type: int + brstack: Optional['branch_stack'] + callchain: Optional['callchain'] + def __getattr__(self, name: str) -> Any: ... + +class mmap_event: + """Represents a mmap event from perf.""" + type: int + pid: int + tid: int + addr: int + len: int + pgoff: int + filename: str + +class mmap2_event: + """Represents a mmap2 event from perf.""" + type: int + pid: int + tid: int + addr: int + len: int + pgoff: int + prot: int + flags: int + filename: str + +class lost_event: + """Represents a lost events record.""" + type: int + id: int + lost: int + +class comm_event: + """Represents a COMM record.""" + type: int + pid: int + tid: int + comm: str + +class task_event: + """Represents an EXIT or FORK record.""" + type: int + pid: int + ppid: int + tid: int + ptid: int + time: int + +class throttle_event: + """Represents a THROTTLE or UNTHROTTLE record.""" + type: int + time: int + id: int + stream_id: int + +class read_event: + """Represents a READ record.""" + type: int + pid: int + tid: int + value: int + +class switch_event: + """Represents a SWITCH or SWITCH_CPU_WIDE record.""" + type: int + +class branch_entry: + """Represents a branch entry in the branch stack. + + Attributes: + from_ip: Source address of the branch (corresponds to 'from' keyword in C). + to_ip: Destination address of the branch. + mispred: True if the branch was mispredicted. + predicted: True if the branch was predicted. + in_tx: True if the branch was in a transaction. + abort: True if the branch was an abort. + cycles: Number of cycles since the last branch. + type: Type of branch. + """ + from_ip: int + to_ip: int + mispred: bool + predicted: bool + in_tx: bool + abort: bool + cycles: int + type: int + +class branch_stack: + """Iterator over branch entries in the branch stack.""" + def __iter__(self) -> Iterator[branch_entry]: ... + def __next__(self) -> branch_entry: ... + +class callchain_node: + """Represents a frame in the callchain.""" + ip: int + sym: Optional[Any] + map: Optional[Any] + +class callchain: + """Iterator over callchain frames.""" + def __iter__(self) -> Iterator[callchain_node]: ... + def __next__(self) -> callchain_node: ... + +class stat_event: + """Represents a stat event from perf.""" + type: int + id: int + cpu: int + thread: int + val: int + ena: int + run: int + +class stat_round_event: + """Represents a stat round event from perf.""" + type: int + time: int + +class cpu_map: + """Map of CPUs being monitored.""" + def __init__(self, cpustr: Optional[str] = None) -> None: ... + def __len__(self) -> int: ... + def __getitem__(self, index: int) -> int: ... + def __iter__(self) -> Iterator[int]: ... + + +class evlist: + def __init__(self, cpus: cpu_map, threads: thread_map) -> None: ... + def open(self) -> None: + """Open the events in the list.""" + ... + def close(self) -> None: + """Close the events in the list.""" + ... + def mmap(self) -> None: + """Memory map the event buffers.""" + ... + def poll(self, timeout: int) -> int: + """Poll for events. + + Args: + timeout: Timeout in milliseconds. + + Returns: + Number of events ready. + """ + ... + def read_on_cpu(self, cpu: int) -> Optional[sample_event]: + """Read a sample event from a specific CPU. + + Args: + cpu: The CPU number. + + Returns: + A sample_event if available, or None. + """ + ... + def all_cpus(self) -> cpu_map: + """Get a cpu_map of all CPUs in the system.""" + ... + def metrics(self) -> List[str]: + """Get a list of metric names within the evlist.""" + ... + def compute_metric(self, metric: str, cpu: int, thread: int) -> float: + """Compute metric for given name, cpu and thread. + + Args: + metric: The metric name. + cpu: The CPU number. + thread: The thread ID. + + Returns: + The computed metric value. + """ + ... + def config(self) -> None: + """Configure the events in the list.""" + ... + def disable(self) -> None: + """Disable all events in the list.""" + ... + def enable(self) -> None: + """Enable all events in the list.""" + ... + def get_pollfd(self) -> List[int]: + """Get a list of file descriptors for polling.""" + ... + def add(self, evsel: evsel) -> int: + """Add an event to the list.""" + ... + def __iter__(self) -> Iterator[evsel]: + """Iterate over the events (evsel) in the list.""" + ... + + +class session: + def __init__( + self, + data: data, + sample: Optional[Callable[[sample_event], None]] = None, + stat: Optional[Callable[[Any, Optional[str]], None]] = None + ) -> None: + """Initialize a perf session. + + Args: + data: The perf data file to read. + sample: Callback for sample events. + stat: Callback for stat events. + """ + ... + def process_events(self) -> int: + """Process all events in the session.""" + ... + def find_thread(self, pid: int) -> thread: + """Returns the thread associated with a pid.""" + ... + +# Event Types +TYPE_HARDWARE: int +"""Hardware event.""" + +TYPE_SOFTWARE: int +"""Software event.""" + +TYPE_TRACEPOINT: int +"""Tracepoint event.""" + +TYPE_HW_CACHE: int +"""Hardware cache event.""" + +TYPE_RAW: int +"""Raw hardware event.""" + +TYPE_BREAKPOINT: int +"""Breakpoint event.""" + + +# Hardware Counters +COUNT_HW_CPU_CYCLES: int +"""Total cycles. Be wary of what happens during CPU frequency scaling.""" + +COUNT_HW_INSTRUCTIONS: int +"""Retired instructions. Be careful, these can be affected by various issues, +most notably hardware interrupt counts.""" + +COUNT_HW_CACHE_REFERENCES: int +"""Cache accesses. Usually this indicates Last Level Cache accesses but this +may vary depending on your CPU.""" + +COUNT_HW_CACHE_MISSES: int +"""Cache misses. Usually this indicates Last Level Cache misses.""" + +COUNT_HW_BRANCH_INSTRUCTIONS: int +"""Retired branch instructions.""" + +COUNT_HW_BRANCH_MISSES: int +"""Mispredicted branch instructions.""" + +COUNT_HW_BUS_CYCLES: int +"""Bus cycles, which can be different from total cycles.""" + +COUNT_HW_STALLED_CYCLES_FRONTEND: int +"""Stalled cycles during issue [This event is an alias of idle-cycles-frontend].""" + +COUNT_HW_STALLED_CYCLES_BACKEND: int +"""Stalled cycles during retirement [This event is an alias of idle-cycles-backend].""" + +COUNT_HW_REF_CPU_CYCLES: int +"""Total cycles; not affected by CPU frequency scaling.""" + + +# Cache Counters +COUNT_HW_CACHE_L1D: int +"""Level 1 data cache.""" + +COUNT_HW_CACHE_L1I: int +"""Level 1 instruction cache.""" + +COUNT_HW_CACHE_LL: int +"""Last Level Cache.""" + +COUNT_HW_CACHE_DTLB: int +"""Data TLB.""" + +COUNT_HW_CACHE_ITLB: int +"""Instruction TLB.""" + +COUNT_HW_CACHE_BPU: int +"""Branch Processing Unit.""" + +COUNT_HW_CACHE_OP_READ: int +"""Read accesses.""" + +COUNT_HW_CACHE_OP_WRITE: int +"""Write accesses.""" + +COUNT_HW_CACHE_OP_PREFETCH: int +"""Prefetch accesses.""" + +COUNT_HW_CACHE_RESULT_ACCESS: int +"""Accesses.""" + +COUNT_HW_CACHE_RESULT_MISS: int +"""Misses.""" + + +# Software Counters +COUNT_SW_CPU_CLOCK: int +"""CPU clock event.""" + +COUNT_SW_TASK_CLOCK: int +"""Task clock event.""" + +COUNT_SW_PAGE_FAULTS: int +"""Page faults.""" + +COUNT_SW_CONTEXT_SWITCHES: int +"""Context switches.""" + +COUNT_SW_CPU_MIGRATIONS: int +"""CPU migrations.""" + +COUNT_SW_PAGE_FAULTS_MIN: int +"""Minor page faults.""" + +COUNT_SW_PAGE_FAULTS_MAJ: int +"""Major page faults.""" + +COUNT_SW_ALIGNMENT_FAULTS: int +"""Alignment faults.""" + +COUNT_SW_EMULATION_FAULTS: int +"""Emulation faults.""" + +COUNT_SW_DUMMY: int +"""Dummy event.""" + + +# Sample Fields +SAMPLE_IP: int +"""Instruction pointer.""" + +SAMPLE_TID: int +"""Process and thread ID.""" + +SAMPLE_TIME: int +"""Timestamp.""" + +SAMPLE_ADDR: int +"""Sampled address.""" + +SAMPLE_READ: int +"""Read barcode.""" + +SAMPLE_CALLCHAIN: int +"""Call chain.""" + +SAMPLE_ID: int +"""Unique ID.""" + +SAMPLE_CPU: int +"""CPU number.""" + +SAMPLE_PERIOD: int +"""Sample period.""" + +SAMPLE_STREAM_ID: int +"""Stream ID.""" + +SAMPLE_RAW: int +"""Raw sample.""" + + +# Format Fields +FORMAT_TOTAL_TIME_ENABLED: int +"""Total time enabled.""" + +FORMAT_TOTAL_TIME_RUNNING: int +"""Total time running.""" + +FORMAT_ID: int +"""Event ID.""" + +FORMAT_GROUP: int +"""Event group.""" + + +# Record Types +RECORD_MMAP: int +"""MMAP record. Contains header, pid, tid, addr, len, pgoff, filename, and sample_id.""" + +RECORD_LOST: int +"""Lost events record. Contains header, id, lost count, and sample_id.""" + +RECORD_COMM: int +"""COMM record. Contains header, pid, tid, comm, and sample_id.""" + +RECORD_EXIT: int +"""EXIT record. Contains header, pid, ppid, tid, ptid, time, and sample_id.""" + +RECORD_THROTTLE: int +"""THROTTLE record. Contains header, time, id, stream_id, and sample_id.""" + +RECORD_UNTHROTTLE: int +"""UNTHROTTLE record. Contains header, time, id, stream_id, and sample_id.""" + +RECORD_FORK: int +"""FORK record. Contains header, pid, ppid, tid, ptid, time, and sample_id.""" + +RECORD_READ: int +"""READ record. Contains header, and read values.""" + +RECORD_SAMPLE: int +"""SAMPLE record. Contains header, and sample data requested by sample_type.""" + +RECORD_MMAP2: int +"""MMAP2 record. Contains header, pid, tid, addr, len, pgoff, maj, min, ino, +ino_generation, prot, flags, filename, and sample_id.""" + +RECORD_AUX: int +"""AUX record. Contains header, aux_offset, aux_size, flags, and sample_id.""" + +RECORD_ITRACE_START: int +"""ITRACE_START record. Contains header, pid, tid, and sample_id.""" + +RECORD_LOST_SAMPLES: int +"""LOST_SAMPLES record. Contains header, lost count, and sample_id.""" + +RECORD_SWITCH: int +"""SWITCH record. Contains header, and sample_id.""" + +RECORD_SWITCH_CPU_WIDE: int +"""SWITCH_CPU_WIDE record. Contains header, and sample_id.""" + +RECORD_STAT: int +"""STAT record.""" + +RECORD_STAT_ROUND: int +"""STAT_ROUND record.""" + +RECORD_MISC_SWITCH_OUT: int +"""MISC_SWITCH_OUT record."""
Signature · 1 line
2.54.0.794.g4f17f83d09-goog
[email protected]
AI reviewRe: [PATCH v9 22/23] perf python: Add perf.pyi stubs file
Thank you for your contribution! Sashiko AI review found 4 potential issue(s) to consider: - [High] Several type stubs omit optional arguments, include incorrect arguments, or provide incorrect…
Thank you for your contribution! Sashiko AI review found 4 potential issue(s) to consider:
- [High] Several type stubs omit optional arguments, include incorrect arguments, or provide incorrect callback signatures compared to the underlying C implementation.
- [High] Event classes in perf.pyi omit numerous attributes exposed by the C extension and declare attributes that do not exist.
- [Medium] The perf.pyi type stub file is not included in the Python extension distribution.
- [Medium] The module-level tracepoint method is missing from the type stubs.
--
commit 55dca7ad7944e4ffd824e2f8978f664aa6e28e2b
Author: Ian Rogers <[email protected]>
perf python: Add perf.pyi stubs file
This commit adds Python type stubs for the perf module to improve IDE support and static analysis. It provides classes, methods, and constants derived from the C source definitions.
tools/perf/python/perf.pyi line 2
Quoted 8 lines
tools/perf/python/perf.pyi new file +2
-0@@ -0,0 +1,605 @@ +"""Type stubs for the perf Python module.""" +from typing import Callable, Dict, List, Optional, Any, Iterator
[Severity: Medium]
Will this stub file actually be installed into the user site-packages directory?
Looking at tools/perf/util/setup.py and tools/perf/Makefile.perf, there doesn't appear to be any logic to copy perf.pyi into the build output directory alongside the compiled perf extension.
Without deployment, this regression might prevent IDEs and static analysis tools from accessing these stubs.
tools/perf/python/perf.pyi line 66
Quoted 4 lines
[ ... ]
tools/perf/python/perf.pyi new file +3
-0+def parse_metrics(metrics_string: str) -> 'evlist': + """Parse a string of metrics or metric groups and return an evlist.""" + ...
[Severity: High]
Does this match the C implementation of parse_metrics?
The C extension appears to accept additional parameters like pmu, cpus, and threads. Missing these arguments in the stub is a regression that might cause type checkers to incorrectly flag valid calls.
tools/perf/python/perf.pyi line 70
Quoted 3 lines
tools/perf/python/perf.pyi new file +3
-0+def pmus() -> Iterator[Any]: + """Returns a sequence of pmus.""" + ...
[Severity: Medium]
Is the module-level tracepoint method missing from these stubs?
In tools/perf/util/python.c, the perf__methods array exports a tracepoint method mapped to pyrf__tracepoint. Omitting it here is a regression that might cause IDEs to report calls to perf.tracepoint as unresolved references.
tools/perf/python/perf.pyi line 90
Quoted 6 lines
[ ... ]
tools/perf/python/perf.pyi new file +5
-0+class counts_values: + """Raw counter values.""" + val: int + ena: int + run: int
[Severity: High]
Do these attributes exist on the counts_values class at runtime?
The C code exposes a single values attribute returning a list of the three integers using pyrf_counts_values_getset. Developers relying on val, ena, and run in the stubs might encounter AttributeError regressions.
tools/perf/python/perf.pyi line 129
Quoted 11 lines
[ ... ]
tools/perf/python/perf.pyi new file +9
-0+class evsel: + """Event selector, represents a single event being monitored.""" + def __init__( + self, + type: int = ..., + config: int = ..., + freq: bool = ..., + idx: int = ..., + ) -> None: ...
[Severity: High]
Does this constructor signature match what the C extension expects?
The C constructor only has 29 strings in kwlist and parses 29 format characters. It appears to miss over 10 configuration parameters like inherit_stat or enable_on_exec, and incorrectly includes an idx parameter.
Passing idx based on this stub will raise a runtime TypeError regression.
tools/perf/python/perf.pyi line 172
Quoted 7 lines
[ ... ]
tools/perf/python/perf.pyi new file +6
-0+class mmap_event: + """Represents a mmap event from perf.""" + type: int + pid: int + tid: int + addr: int
[Severity: High]
Should addr be named start instead?
In the C code, pyrf_mmap_event__members exports this field as start.
There are also attributes injected by the sample_members macro like sample_pid, sample_tid, and sample_time that appear to be missing from the event classes here.
tools/perf/python/perf.pyi line 223
Quoted 7 lines
[ ... ]
tools/perf/python/perf.pyi new file +6
-0+class read_event: + """Represents a READ record.""" + type: int + pid: int + tid: int + value: int
[Severity: High]
Do the type and value fields exist in the C members array for read_event?
It looks like pyrf_read_event__members and pyrf_lost_event__members do not contain these fields.
tools/perf/python/perf.pyi line 227
Quoted 4 lines
[ ... ]
tools/perf/python/perf.pyi new file +3
-0+class switch_event: + """Represents a SWITCH or SWITCH_CPU_WIDE record.""" + type: int
[Severity: High]
Are next_prev_pid and next_prev_tid missing from this class?
These fields are present in the C structure but omitted here.
tools/perf/python/perf.pyi line 300
Quoted 6 lines
[ ... ]
tools/perf/python/perf.pyi new file +4
-0+class evlist: + def mmap(self) -> None: + """Memory map the event buffers.""" + ...
[Severity: High]
Are the pages and overwrite arguments missing from this stub?
tools/perf/python/perf.pyi line 310
Quoted 5 lines
tools/perf/python/perf.pyi new file +4
-0+ def poll(self, timeout: int) -> int: + """Poll for events. + """ + ...
[Severity: High]
Is timeout an optional parameter in the C implementation?
If it doesn't have a default value in the stub, type checkers will require it, even if the C extension allows calling poll without arguments.
tools/perf/python/perf.pyi line 320
Quoted 5 lines
tools/perf/python/perf.pyi new file +4
-0+ def read_on_cpu(self, cpu: int) -> Optional[sample_event]: + """Read a sample event from a specific CPU. + """ + ...
[Severity: High]
Is the sample_id_all parameter missing from this method signature?
tools/perf/python/perf.pyi line 365
Quoted 8 lines
[ ... ]
tools/perf/python/perf.pyi new file +7
-0+class session: + def __init__( + self, + data: data, + sample: Optional[Callable[[sample_event], None]] = None, + stat: Optional[Callable[[Any, Optional[str]], None]] = None + ) -> None:
[Severity: High]
Does the stat callback signature match how the C runtime invokes it?
The C code in pyrf_session_tool__stat_round calls it with exactly one argument for RECORD_STAT_ROUND events. If users follow this stub and expect two arguments, it might result in a TypeError regression.
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=22
Ian Rogers
reviewerPATCH v9 23/23 perf python: Add LiveSession helper
In reply to Ian RogersAdd LiveSession class in tools/perf/python/perf_live.py to support live event collection using perf.evlist and perf.parse_events, avoiding the need to fork a separate perf record process.
---
v2:
1. Fixed File Descriptor Leak: I moved self.evlist.mmap() inside the try block so that if it raises an exception, the finally block will still be executed and call self.evlist.close() , preventing file descriptor leaks.
2. Handled InterruptedError in poll() : I wrapped the poll() call in a try-except block to catch InterruptedError and continue the loop. This prevents the live session from crashing on non-fatal signals like SIGWINCH .
3. Added evlist.config() : I added a call to self.evlist.config() in the constructor after parse_events() . This applies the default record options to the events, enabling sampling and setting up PERF_SAMPLE_* fields so that the kernel will actually generate PERF_RECORD_SAMPLE events.
4. Enable the evlist and be robust to exceptions from reading unsupported events like mmap2.
v8:
- Drain all events from a CPU before moving to the next.
--- tools/perf/python/perf_live.py | 53 ++++++++++++++++++++++++++++++++++ 1 file changed, 53 insertions(+) create mode 100755 tools/perf/python/perf_live.py
tools/perf/python/perf_live.py new file +53 -0
@@ -0,0 +1,53 @@ +# SPDX-License-Identifier: GPL-2.0 +""" +Live event session helper using perf.evlist. + +This module provides a LiveSession class that allows running a callback +for each event collected live from the system, similar to perf.session +but without requiring a perf.data file. +""" + +import perf + + +class LiveSession: + """Represents a live event collection session.""" + + def __init__(self, event_string: str, sample_callback): + self.event_string = event_string + self.sample_callback = sample_callback + # Create a cpu map for all online CPUs + self.cpus = perf.cpu_map() + # Parse events and set maps + self.evlist = perf.parse_events(self.event_string, self.cpus) + self.evlist.config() + + def run(self): + """Run the live session.""" + self.evlist.open() + try: + self.evlist.mmap() + self.evlist.enable() + + while True: + # Poll for events with 100ms timeout + try: + self.evlist.poll(100) + except InterruptedError: + continue + for cpu in self.cpus: + while True: + try: + event = self.evlist.read_on_cpu(cpu) + if event is None: + break + if event.type == perf.RECORD_SAMPLE: + self.sample_callback(event) + except Exception as e: + import sys + print(f"Error processing event on CPU {cpu}: {e}", file=sys.stderr) + break + except KeyboardInterrupt: + pass + finally: + self.evlist.close()
Signature · 1 line
2.54.0.794.g4f17f83d09-goog
[email protected]
AI reviewRe: [PATCH v9 23/23] perf python: Add LiveSession helper
Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider: - [Medium] LiveSession.run() will enter an infinite busy-loop pegging the CPU to 100% and flooding stderr…
Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
- [Medium] LiveSession.run() will enter an infinite busy-loop pegging the CPU to 100% and flooding stderr if a file descriptor encounters an unrecoverable error.
- [Medium] Draining all events from one CPU before moving to the next causes starvation and event loss for other CPUs during high event rates.
- [Low] Potential file descriptor leak if an asynchronous exception occurs between evlist.open() and the try block.
--
commit d20b9507bd8c4d01a239c0c3dc0e2b42402bcfe1
Author: Ian Rogers <[email protected]>
perf python: Add LiveSession helper
Add LiveSession class in tools/perf/python/perf_live.py to support live event collection using perf.evlist and perf.parse_events, avoiding the need to fork a separate perf record process.
Quoted 5 lines
tools/perf/python/perf_live.py new file +0
-0
[ ... ]
tools/perf/python/perf_live.py line 28
Quoted 4 lines
tools/perf/python/perf_live.py new file +4
-0+ def run(self): + """Run the live session.""" + self.evlist.open() + try:
[Severity: Low]
Can this leak file descriptors if an asynchronous exception, like a KeyboardInterrupt, is raised right after open() but before entering the try block?
If the exception occurs here, the finally block will be bypassed and close() won't be called. Should open() be moved inside the try block to guarantee cleanup?
tools/perf/python/perf_live.py line 35
Quoted 7 lines
tools/perf/python/perf_live.py new file +7
-0+ self.evlist.mmap() + self.evlist.enable() + + while True: + # Poll for events with 100ms timeout + try: + self.evlist.poll(100)
[Severity: Medium]
What happens if a polled file descriptor encounters an unrecoverable error like POLLHUP or POLLERR?
It seems poll() will return immediately without blocking, and the subsequent read_on_cpu() will raise an exception. Since the offending file descriptor is never removed from the evlist or disabled, poll() will be called again and return immediately, leading to an infinite busy-loop that pegs the CPU at 100% and floods stderr.
tools/perf/python/perf_live.py line 45
Quoted 10 lines
tools/perf/python/perf_live.py new file +10
-0+ except InterruptedError: + continue + for cpu in self.cpus: + while True: + try: + event = self.evlist.read_on_cpu(cpu) + if event is None: + break + if event.type == perf.RECORD_SAMPLE: + self.sample_callback(event)
[Severity: Medium]
Will this continuous inner loop cause starvation and event loss for other CPUs?
If a particular CPU generates events at a rate comparable to or faster than the script can process them, this inner loop might never exit. This would stall the outer loop, preventing other CPUs from being processed. Their ring buffers would fill up and the kernel would drop their events.
Should the processing be interleaved or bounded per CPU to ensure fairness?
Quoted 4 lines
tools/perf/python/perf_live.py new file +4
-0+ except Exception as e: + import sys + print(f"Error processing event on CPU {cpu}: {e}", file=sys.stderr) + break
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=23
Arnaldo Carvalho de Melo
maintainerRe: [PATCH v9 00/23] perf python: Modernize and extend Python API (Phase 1)
In reply to Ian RogersQuoted 28 lines · Ian Rogers
The perf script command has long supported running Python and Perl scripts by embedding libpython and libperl. This approach has several drawbacks:
- overhead by creating Python dictionaries for every event (whether used or not),
- complex build dependencies on specific Python/Perl versions,
- complications with threading due to perf being the interpreter,
- no clear way to run standalone scripts likeilist.py.This series takes a different approach with some initial implementation posted as an RFC last October:
https://lore.kernel.org/linux-perf-users/[email protected]/
with the motivation coming up on the mailing list earlier:
https://lore.kernel.org/lkml/CAP-5=fWDqE8SYfOLZkg_0=4Ayx6E7O+h7uUp4NDeCFkiN4b7-w@mail.gmail.com/The ultimate goal is to remove the embedded libpython and libperl support from perf entirely, expanding the existing perf Python module to provide full access to perf data files and events, allowing scripts to be run as standalone Python applications.
To make the review process more manageable, the original 58-patch series has been split. This v9 series represents "Phase 1: API & Infrastructure" (23 patches). It contains:
1. Missed explicit dependency cleanups and header sorting.
2. Crucial core safety infrastructure (reference counting for evlist/evsel) to support safe lifecycle management in garbage-collected Python.
3. The core Python API extensions (session wrappers,perf_datawrappers, sample accessors, stubs, and LiveSession helper).
Quoted 3 lines
The subsequent "Phase 2" series will contain the actual porting of all existing Python/Perl scripts to the new API (which yields up to 35x speedups as demonstrated previously) and the final removal of embedded interpreters.
I tried applying it now and got this:
Cover: ./v9_20260522_irogers_perf_python_modernize_and_extend_python_api_phase_1.cover
Base: not specified
git am ./v9_20260522_irogers_perf_python_modernize_and_extend_python_api_phase_1.mbx
⬢ [acme@toolbx perf-tools-next2]$ git am ./v9_20260522_irogers_perf_python_modernize_and_extend_python_api_phase_1.mbx
Applying: perf arch arm: Sort includes and add missed explicit dependencies
Applying: perf arch x86: Sort includes and add missed explicit dependencies
Applying: perf tests: Sort includes and add missed explicit dependencies
Applying: perf script: Sort includes and add missed explicit dependencies
Applying: perf util: Sort includes and add missed explicit dependencies
error: patch failed: tools/perf/util/evsel.c:11
error: tools/perf/util/evsel.c: patch does not apply
Patch failed at 0005 perf util: Sort includes and add missed explicit dependencies
hint: Use 'git am --show-current-patch=diff' to see the failed patch
hint: When you have resolved this problem, run "git am --continue".
hint: If you prefer to skip this patch, run "git am --skip" instead.
hint: To restore the original branch and stop patching, run "git am --abort".
hint: Disable this message with "git config set advice.mergeConflict false" ⬢ [acme@toolbx perf-tools-next2]$ git am --abort
⬢ [acme@toolbx perf-tools-next2]$ git diff
⬢ [acme@toolbx perf-tools-next2]$
I agree with the intent and plan on reviewing and testing it.
- Arnaldo
Arnaldo Carvalho de Melo
maintainerRe: [PATCH v9 00/23] perf python: Modernize and extend Python API (Phase 1)
Quoted 36 lines · Arnaldo Carvalho de Melo
Quoted 10 lines · Ian Rogers
This series takes a different approach with some initial implementation posted as an RFC last October:
https://lore.kernel.org/linux-perf-users/[email protected]/
with the motivation coming up on the mailing list earlier:
https://lore.kernel.org/lkml/CAP-5=fWDqE8SYfOLZkg_0=4Ayx6E7O+h7uUp4NDeCFkiN4b7-w@mail.gmail.com/The ultimate goal is to remove the embedded libpython and libperl support from perf entirely, expanding the existing perf Python module to provide full access to perf data files and events, allowing scripts to be run as standalone Python applications.
I tried applying it now and got this:
Cover: ./v9_20260522_irogers_perf_python_modernize_and_extend_python_api_phase_1.cover
Base: not specified
git am ./v9_20260522_irogers_perf_python_modernize_and_extend_python_api_phase_1.mbx⬢ [acme@toolbx perf-tools-next2]$ git am ./v9_20260522_irogers_perf_python_modernize_and_extend_python_api_phase_1.mbx
Applying: perf arch arm: Sort includes and add missed explicit dependencies
Applying: perf arch x86: Sort includes and add missed explicit dependencies
Applying: perf tests: Sort includes and add missed explicit dependencies
Applying: perf script: Sort includes and add missed explicit dependencies
Applying: perf util: Sort includes and add missed explicit dependencies
error: patch failed:tools/perf/util/evsel.c:11
error:tools/perf/util/evsel.c: patch does not apply
Patch failed at 0005 perf util: Sort includes and add missed explicit dependencies
hint: Use 'git am --show-current-patch=diff' to see the failed patch
hint: When you have resolved this problem, run "git am --continue".
hint: If you prefer to skip this patch, run "git am --skip" instead.
hint: To restore the original branch and stop patching, run "git am --abort".
hint: Disable this message with "git config set advice.mergeConflict false" ⬢ [acme@toolbx perf-tools-next2]$ git am --abort
⬢ [acme@toolbx perf-tools-next2]$ git diff
⬢ [acme@toolbx perf-tools-next2]$
Quoted 1 line
I agree with the intent and plan on reviewing and testing it.
What I have now is at tmp.perf-tools-next if you wish to rebase from there,
Regards,
- Arnaldo
Ian Rogers
reviewerRe: [PATCH v9 00/23] perf python: Modernize and extend Python API (Phase 1)
Quoted 38 lines · Arnaldo Carvalho de Melo
Quoted 35 lines · Arnaldo Carvalho de Melo
Quoted 10 lines · Ian Rogers
This series takes a different approach with some initial implementation posted as an RFC last October:
https://lore.kernel.org/linux-perf-users/[email protected]/
with the motivation coming up on the mailing list earlier:
https://lore.kernel.org/lkml/CAP-5=fWDqE8SYfOLZkg_0=4Ayx6E7O+h7uUp4NDeCFkiN4b7-w@mail.gmail.com/The ultimate goal is to remove the embedded libpython and libperl support from perf entirely, expanding the existing perf Python module to provide full access to perf data files and events, allowing scripts to be run as standalone Python applications.
I tried applying it now and got this:
Cover: ./v9_20260522_irogers_perf_python_modernize_and_extend_python_api_phase_1.cover
Base: not specified
git am ./v9_20260522_irogers_perf_python_modernize_and_extend_python_api_phase_1.mbx⬢ [acme@toolbx perf-tools-next2]$ git am ./v9_20260522_irogers_perf_python_modernize_and_extend_python_api_phase_1.mbx
Applying: perf arch arm: Sort includes and add missed explicit dependencies
Applying: perf arch x86: Sort includes and add missed explicit dependencies
Applying: perf tests: Sort includes and add missed explicit dependencies
Applying: perf script: Sort includes and add missed explicit dependencies
Applying: perf util: Sort includes and add missed explicit dependencies
error: patch failed:tools/perf/util/evsel.c:11
error:tools/perf/util/evsel.c: patch does not apply
Patch failed at 0005 perf util: Sort includes and add missed explicit dependencies
hint: Use 'git am --show-current-patch=diff' to see the failed patch
hint: When you have resolved this problem, run "git am --continue".
hint: If you prefer to skip this patch, run "git am --skip" instead.
hint: To restore the original branch and stop patching, run "git am --abort".
hint: Disable this message with "git config set advice.mergeConflict false" ⬢ [acme@toolbx perf-tools-next2]$ git am --abort
⬢ [acme@toolbx perf-tools-next2]$ git diff
⬢ [acme@toolbx perf-tools-next2]$
Quoted 4 lines
Quoted 1 line
I agree with the intent and plan on reviewing and testing it.
What I have now is at tmp.perf-tools-next if you wish to rebase from there,
Thanks Arnaldo, I'll do that.
Ian
Quoted 3 lines
Regards,
- Arnaldo
Ian Rogers
reviewerv10 cover perf python: Modernize and extend Python API (Phase 1)
In reply to Ian RogersThe perf script command has long supported running Python and Perl scripts by embedding libpython and libperl. This approach has several drawbacks:
- overhead by creating Python dictionaries for every event (whether used or not),
- complex build dependencies on specific Python/Perl versions,
- complications with threading due to perf being the interpreter,
- no clear way to run standalone scripts like ilist.py.
This series takes a different approach with some initial implementation posted as an RFC last October:
https://lore.kernel.org/linux-perf-users/[email protected]/
with the motivation coming up on the mailing list earlier:
https://lore.kernel.org/lkml/CAP-5=fWDqE8SYfOLZkg_0=4Ayx6E7O+h7uUp4NDeCFkiN4b7-w@mail.gmail.com/
The ultimate goal is to remove the embedded libpython and libperl support from perf entirely, expanding the existing perf Python module to provide full access to perf data files and events, allowing scripts to be run as standalone Python applications.
To make the review process more manageable, the original 58-patch series has been split. This v10 series represents "Phase 1: API & Infrastructure" (19 patches). The first 4 patches of Phase 1 (cleanups and arch-specific header sorting) have already been merged upstream.
This remaining set contains:
1. Missed explicit dependency cleanups and header sorting for util/ and python.
2. Crucial core safety infrastructure (reference counting for evlist/evsel) to support safe lifecycle management in garbage-collected Python.
3. The core Python API extensions (session wrappers, perf_data wrappers, sample accessors, stubs, and LiveSession helper).
The subsequent "Phase 2" series will contain the actual porting of all existing Python/Perl scripts to the new API (which yields up to 35x speedups as demonstrated previously) and the final removal of embedded interpreters.
Note: The preliminary clang-format patch has been separated from this series to be sent independently.
---
v10 Changes
-----------
- The preliminary clang-format patch has been separated from this series and updated independently based on review feedback (using Priority -1 for Python.h and eliminating the unused Priority 1 gap).
- Fixed Type Confusion on Python Object instantiation:
- Added strict type verification O!O! to pyrf_evlist__init for CPU and Thread maps to prevent interpreter crashes from bad inputs.
- Added .tp_getattro and .tp_setattro handlers to evlist, evsel, and session Python types to safely raise ValueError on uninitialized access (such as bypassing __init__ and invoking __new__ directly) instead of dereferencing NULL.
- Handled NULL structures safely inside string/representation callbacks for evlist and evsel types.
- Fixed Circular Reference memory leaks:
- Eliminated the pevent pointer and unused lazy resolution code from pyrf_callchain, breaking the cycle that prevented GC collection.
- Eliminated pevent pointer from pyrf_branch_stack, breaking the cycle.
- This prevents permanent leaks of all map, symbol, and event structures for every sample containing a callchain or branch stack.
- Converted callchain and branch_stack to true Python Sequence types:
- Replaced custom exhaustible iterators with .tp_as_sequence sequence methods (sq_length and sq_item).
- This allows standard Sequence operations like len(event.callchain), list index access (e.g. event.callchain[0]), and infinite re-iterations.
- Fixed potential Memory Leak on test failure:
- Fixed a leak of the newly allocated evsel structure in do_test inside sample-parsing.c on malloc failure by routing through the cleanup block.
- Fixed compiler warning for mixed declarations:
- Moved the C-statement perf_sample__init after struct and variable declarations in cs_etm__synth_branch_sample to ensure C89/compiler compatibility.
- Fixed compiler warning for discarded const qualifiers:
- Reverted custom static char * const kwlist[] to the standard static char *kwlist[] for pyrf__syscall_name and pyrf__syscall_id in python.c.
- Fixed unrecoverable file descriptor leak and busy loop in LiveSession:
- Moved self.evlist.open() inside the try block of LiveSession.run() to guarantee finally block cleanup and prevent descriptor leaks on early interrupts.
- Handled unrecoverable OSError (like mmap read init errors) by propagating/raising them to safely terminate the session instead of getting stuck in an infinite CPU-pegging poll busy loop.
- Bounded read_on_cpu() reads to at most 1000 events per CPU per poll iteration to prevent high-volume starvation of other CPU channels.
- Exported stat_event and stat_round_event to the perf module namespace to allow type checks like isinstance(event, perf.stat_event).
- Fixed Python stat callback signatures:
- Changed Python stat callbacks in C to consistently pass 2 arguments (using "Oz" and passing None for stat_round events), preventing TypeError failures in scripts defining standard 2-parameter signatures.
- Fixed data-corruption in misc fields:
- Changed misc member definition from T_UINT to T_USHORT in both mmap and mmap2 events to avoid reading adjacent size bytes from 16-bit struct headers.
- Exposed maj, min, ino, and ino_generation members to Python mmap2 events.
- Fixed mmap2 union overlay field access:
- Added custom getters to mmap2 events for maj, min, ino, ino_generation, and build_id based on the PERF_RECORD_MISC_MMAP_BUILD_ID misc header flag.
- This correctly exposes build_id (as bytes) when present, or maj/min/ino when build ID is absent, returning None for the inactive union fields.
- Enabled deployment of stubs:
- Updated setup.py to install perf.pyi alongside the extension in site-packages.
- Updated Makefile.perf to copy perf.pyi to the build directory for in-tree usage.
- Added tracepoint and missing optional parse_metrics arguments to stubs.
- Fixed a pre-existing Use-After-Free bug in iostat_prepare:
- Modified iostat_prepare to take struct evlist **evlist_ptr, allowing it to properly reassign the caller's evlist pointer, avoiding use-after-free in cmd_stat.
- Fixed in-tree python test runner stability:
- Updated setup_python.sh to automatically export PYTHONPATH pointing to the in-tree built perf.so directory (handling O= build folders), preventing Python from accidentally loading the system-wide outdated /usr/lib/perf.so module and failing with AttributeError during perf test.
v9 Changes
----------
- This series is now split, containing only the first 23 patches of the previous 58-patch series. This "Phase 1: API & Infrastructure" set focuses on modernizing and extending the Python API and adding crucial safety infrastructure (reference counting). The script porting and legacy interpreter removal will be sent in a subsequent Phase 2.
- Fixed Type Confusion in pyrf_evlist__init: Added strict type validation to CPU and Thread map arguments (using O!O!) to prevent crashes from unsafe casts.
- Fixed Infinite Loop in LiveSession.run: Added a break statement in the exception block of the event reading loop to prevent 100% CPU spinning on persistent OS errors (like mmap read init failures).
- Fixed Inconsistent Exception Handling in Session Callbacks:
- Removed the swallowing PyErr_Print() call from pyrf_session_tool__stat to preserve exceptions.
- Updated pyrf_session_tool__stat_round to check the callback return value and return -1 on failure, aborting the event loop and propagating the exception cleanly.
- Fixed Uninitialized State in pyrf_session__new: Added explicit psession->pdata = NULL initialization immediately after allocation to prevent potential crashes in tp_dealloc on early failures.
v8 Changes
----------
- Make schedstat and itrace=L fixes separate patches:
https://lore.kernel.org/lkml/[email protected]/
https://lore.kernel.org/lkml/[email protected]/
- Fixed Heap Out-Of-Bounds / Uninitialized Memory in pyrf_event__new:
Use /*all=*/true in perf_sample__init to prevent garbage memory in sample structures.
- Fixed Type Confusion in pyrf_evlist__add: Added strict O! type validation to avoid unsafe casts when adding evsels to an evlist.
- Exposed Thread Identifiers: Added pid, tid, ppid, and cpu attributes to the Python perf.thread type to allow thread identification.
- Fixed Process Resolution: Wrapped thread resolution in compaction-times.py, check-perf-trace.py, and task-analyzer.py in try-except blocks to safely handle untracked PIDs instead of raising uncaught TypeError crashes.
- Fixed Potential Data Loss in futex-contention.py: Updated process resolution in handle_start to fall back to 'unknown' on lookup errors, ensuring events are always tracked.
- Synchronized Type Stubs File: Added the mmap2_event class and new evsel and thread attributes to perf.pyi.
v7 Changes
----------
- Fixed heap out-of-bounds in pyrf_event__new by adding comprehensive size checks for all event types.
- Fixed undefined symbol syscalltbl__id when building without libtraceevent by making syscalltbl.o unconditional in Build.
- Fixed several issues in python.c:
- Handled NULL return from thread__comm_str in pyrf_thread__comm.
- Avoided swallowing exceptions in module initialization.
- Added custom tp_new methods for evlist, evsel, and data types to zero-initialize pointers and avoid crashes on re-initialization.
- Fixed lower priority review comments:
- Avoided permanent iterator exhaustion on brstack in perf_brstack_max.py by converting it to a list.
- Removed dead code (unused self.unhandled dictionary) in failed-syscalls-by-pid.py.
v6 Changes
----------
- Refactored pyrf_event__new to take evsel and session arguments, and use dynamic allocation based on the actual event size to improve memory safety and efficiency.
- Moved callchain and branch stack resolution logic from pyrf_session_tool__sample into pyrf_event__new, centralizing initialization.
- Added an optional keyword-only elf_machine argument to syscall_name and syscall_id functions to allow specifying non-host architectures, defaulting to EM_HOST.
- Renamed process method to find_thread in the Python API and C implementation for better intention-revealing naming.
- Fixed a terminal injection vulnerability in flamegraph.py by not printing unverified downloaded content in the prompt.
- Fixed CWD exposure and symlink attack risks in gecko.py by using a secure temporary directory for the HTTP server.
- Fixed a severe performance issue in event_analyzing_sample.py by removing SQLite autocommit mode and batching commits.
- Fixed AttributeError crashes in rw-by-file.py and rw-by-pid.py by correctly extracting event names.
- Fixed man page formatting issues in perf-script-python.txt by using indented code blocks.
- Updated perf.pyi stubs file to reflect all API changes.
- Verified all commit messages with checkpatch.pl and ensured lines are wrapped appropriately.
- Fixed segmentation faults in perf sched stats in diff mode.
v5 Changes
----------
Resending due to partial send of v4 due to a quota limit.
v4 Changes
----------
1. Git Fixup Cleanups
- Squashed the lingering fixup! commit remaining from the previous session back into perf check-perf-trace: Port check-perf-trace to use python module.
v3 Changes
----------
1. Memory Safety & Reference Counting Fixes
- Stored transient mmap event data inside the Python object's permanent pevent->event and invoked evsel__parse_sample() to safely point attributes into it, resolving Use-After-Free vulnerabilities.
- Nullified sample->evsel after calling evsel__put() in perf_sample__exit() to protect against potential refcount double-put crashes in error paths.
- Reordered operations inside evlist__remove() to invoke perf_evlist__remove() before reference release.
- Patched an evsel reference leak inside evlist__deliver_deferred_callchain().
2. Sashiko AI Review Cleanups
- Corrected the broken event name equality check in gecko.py to search for a substring match within the parsed event string.
- Fixed a latent AttributeError crash in task-analyzer.py by properly assigning the session instance.
- Safeguarded thread reporting in check-perf-trace.py by utilizing sample_tid instead of sample_pid, and wrapping the session thread resolution in a try-except block.
3. Omitted Minor Issues
- The minor review comments (such as permanent iterator exhaustion on brstack, or dead-code in failed-syscalls-by-pid.py) have been omitted because they do not affect correctness, lead to crashes, or require significant architectural rework.
v2 Changes
----------
1. String Match and Event Name Accuracy
- Replaced loose substring event matching across the script suite with exact matches or specific prefix constraints (syscalls:sys_exit_, evsel(skb:kfree_skb), etc.).
- Added getattr() safety checks to prevent script failures caused by unresolved attributes from older kernel traces.
2. OOM and Memory Protections
- Refactored netdev-times.py to compute and process network statistics chronologically on-the-fly, eliminating an unbounded in-memory list that caused Out-Of-Memory crashes on large files.
- Implemented threshold limits on intel-pt-events.py to cap memory allocation during event interleaving.
- Optimized export-to-sqlite.py to periodically commit database transactions (every 10,000 samples) to reduce temporary SQLite journal sizes.
3. Portability & Environment Independence
- Re-keyed internal tracking dictionaries in scripts like powerpc-hcalls.py to use thread PIDs instead of CPUs, ensuring correctness when threads migrate.
- Switched net_dropmonitor.py from host-specific /proc/kallsyms parsing to perf's built-in symbol resolution API.
- Added the --iomem parameter to mem-phys-addr.py to support offline analysis of data collected on different architectures.
4. Standalone Scripting Improvements
- Patched builtin-script.c to ensure --input parameters are successfully passed down to standalone execution pipelines via execvp().
- Guarded against string buffer overflows during .py extension path resolving.
5. Code Cleanups
- Removed stale perl subdirectories from being detected by the TUI script browser.
- Ran the entire script suite through mypy and pylint to achieve strict static type checking and resolve unreferenced variables.
Ian Rogers (19):
perf util: Sort includes and add missed explicit dependencies perf python: Add missed explicit dependencies
perf evsel/evlist: Avoid unnecessary #includes
perf data: Add open flag
perf evlist: Add reference count
perf evsel: Add reference count
perf evlist: Add reference count checking
perf python: Use evsel in sample in pyrf_event
perf python: Add wrapper for perf_data file abstraction perf python: Add python session abstraction wrapping perf's session perf python: Refactor and add accessors to sample event perf python: Add mmap2 event
perf python: Add callchain support
perf python: Extend API for stat events in python.c
perf python: Expose brstack in sample event
perf python: Add syscall name/id to convert syscall number and name perf python: Add config file access
perf python: Add perf.pyi stubs file
perf python: Add LiveSession helper
tools/perf/Makefile.perf | 3 +- tools/perf/arch/arm/util/cs-etm.c | 10 +- tools/perf/arch/arm64/util/arm-spe.c | 8 +- tools/perf/arch/arm64/util/hisi-ptt.c | 2 +- tools/perf/arch/x86/tests/hybrid.c | 22 +- tools/perf/arch/x86/tests/topdown.c | 4 +- tools/perf/arch/x86/util/auxtrace.c | 2 +- tools/perf/arch/x86/util/intel-bts.c | 6 +- tools/perf/arch/x86/util/intel-pt.c | 9 +- tools/perf/arch/x86/util/iostat.c | 14 +- tools/perf/bench/evlist-open-close.c | 29 +- tools/perf/builtin-annotate.c | 5 +- tools/perf/builtin-ftrace.c | 14 +- tools/perf/builtin-inject.c | 4 +- tools/perf/builtin-kvm.c | 14 +- tools/perf/builtin-kwork.c | 8 +- tools/perf/builtin-lock.c | 2 +- tools/perf/builtin-record.c | 95 +- tools/perf/builtin-report.c | 6 +- tools/perf/builtin-sched.c | 26 +- tools/perf/builtin-script.c | 15 +- tools/perf/builtin-stat.c | 83 +- tools/perf/builtin-top.c | 104 +- tools/perf/builtin-trace.c | 60 +- tools/perf/python/perf.pyi | 629 +++++ tools/perf/python/perf_live.py | 56 + tools/perf/tests/backward-ring-buffer.c | 26 +- tools/perf/tests/code-reading.c | 14 +- tools/perf/tests/event-times.c | 6 +- tools/perf/tests/event_update.c | 4 +- tools/perf/tests/evsel-roundtrip-name.c | 8 +- tools/perf/tests/evsel-tp-sched.c | 4 +- tools/perf/tests/expand-cgroup.c | 12 +- tools/perf/tests/hists_cumulate.c | 2 +- tools/perf/tests/hists_filter.c | 2 +- tools/perf/tests/hists_link.c | 2 +- tools/perf/tests/hists_output.c | 2 +- tools/perf/tests/hwmon_pmu.c | 7 +- tools/perf/tests/keep-tracking.c | 10 +- tools/perf/tests/mmap-basic.c | 24 +- tools/perf/tests/openat-syscall-all-cpus.c | 6 +- tools/perf/tests/openat-syscall-tp-fields.c | 26 +- tools/perf/tests/openat-syscall.c | 6 +- tools/perf/tests/parse-events.c | 139 +- tools/perf/tests/parse-metric.c | 8 +- tools/perf/tests/parse-no-sample-id-all.c | 2 +- tools/perf/tests/perf-record.c | 38 +- tools/perf/tests/perf-time-to-tsc.c | 12 +- tools/perf/tests/pfm.c | 12 +- tools/perf/tests/pmu-events.c | 11 +- tools/perf/tests/pmu.c | 4 +- tools/perf/tests/sample-parsing.c | 44 +- tools/perf/tests/shell/lib/setup_python.sh | 12 + tools/perf/tests/sw-clock.c | 20 +- tools/perf/tests/switch-tracking.c | 10 +- tools/perf/tests/task-exit.c | 20 +- tools/perf/tests/time-utils-test.c | 14 +- tools/perf/tests/tool_pmu.c | 7 +- tools/perf/tests/topology.c | 4 +- tools/perf/tests/uncore-event-sorting.c | 2 +- tools/perf/ui/browsers/annotate.c | 2 +- tools/perf/ui/browsers/hists.c | 22 +- tools/perf/util/Build | 1 - tools/perf/util/amd-sample-raw.c | 2 +- tools/perf/util/annotate-data.c | 2 +- tools/perf/util/annotate.c | 10 +- tools/perf/util/auxtrace.c | 14 +- tools/perf/util/block-info.c | 4 +- tools/perf/util/bpf_counter.c | 2 +- tools/perf/util/bpf_counter_cgroup.c | 10 +- tools/perf/util/bpf_ftrace.c | 9 +- tools/perf/util/bpf_lock_contention.c | 12 +- tools/perf/util/bpf_off_cpu.c | 44 +- tools/perf/util/bpf_trace_augment.c | 8 +- tools/perf/util/cgroup.c | 26 +- tools/perf/util/cs-etm.c | 5 +- tools/perf/util/data-convert-bt.c | 2 +- tools/perf/util/data.c | 27 +- tools/perf/util/data.h | 4 +- tools/perf/util/evlist.c | 492 ++-- tools/perf/util/evlist.h | 273 +- tools/perf/util/evsel.c | 39 +- tools/perf/util/evsel.h | 40 +- tools/perf/util/expr.c | 2 +- tools/perf/util/header.c | 61 +- tools/perf/util/header.h | 2 +- tools/perf/util/intel-tpebs.c | 7 +- tools/perf/util/iostat.c | 2 +- tools/perf/util/iostat.h | 2 +- tools/perf/util/map.h | 9 +- tools/perf/util/metricgroup.c | 12 +- tools/perf/util/parse-events.c | 10 +- tools/perf/util/parse-events.y | 2 +- tools/perf/util/perf_api_probe.c | 20 +- tools/perf/util/pfm.c | 4 +- tools/perf/util/print-events.c | 2 +- tools/perf/util/python.c | 2703 ++++++++++++++++--- tools/perf/util/record.c | 11 +- tools/perf/util/s390-sample-raw.c | 19 +- tools/perf/util/sample-raw.c | 4 +- tools/perf/util/sample.c | 17 +- tools/perf/util/session.c | 59 +- tools/perf/util/setup.py | 1 + tools/perf/util/sideband_evlist.c | 40 +- tools/perf/util/sort.c | 2 +- tools/perf/util/stat-display.c | 6 +- tools/perf/util/stat-shadow.c | 24 +- tools/perf/util/stat.c | 20 +- tools/perf/util/stream.c | 4 +- tools/perf/util/synthetic-events.c | 11 +- tools/perf/util/time-utils.c | 12 +- tools/perf/util/top.c | 4 +- 112 files changed, 4456 insertions(+), 1416 deletions(-) create mode 100644 tools/perf/python/perf.pyi create mode 100755 tools/perf/python/perf_live.py
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
Ian Rogers
reviewerPATCH v10 1/19 perf util: Sort includes and add missed explicit dependencies
Fix missing includes found while cleaning the evsel/evlist header files. Sort the remaining header files for consistency with the rest of the code.
--- tools/perf/util/bpf_off_cpu.c | 30 +++++----- tools/perf/util/bpf_trace_augment.c | 8 +-- tools/perf/util/evlist.c | 91 +++++++++++++++-------------- tools/perf/util/map.h | 9 ++- tools/perf/util/perf_api_probe.c | 18 +++--- tools/perf/util/s390-sample-raw.c | 19 +++--- tools/perf/util/stat-shadow.c | 20 ++++--- tools/perf/util/stat.c | 16 +++-- 8 files changed, 113 insertions(+), 98 deletions(-)
tools/perf/util/bpf_off_cpu.c +16 -14
@@ -1,23 +1,25 @@// SPDX-License-Identifier: GPL-2.0 -#include "util/bpf_counter.h" -#include "util/debug.h" -#include "util/evsel.h" -#include "util/evlist.h" -#include "util/off_cpu.h" -#include "util/perf-hooks.h" -#include "util/record.h" -#include "util/session.h" -#include "util/target.h" -#include "util/cpumap.h" -#include "util/thread_map.h" -#include "util/cgroup.h" -#include "util/strlist.h" +#include <linux/time64.h> +#include <bpf/bpf.h> #include <bpf/btf.h> #include <internal/xyarray.h> -#include <linux/time64.h> +#include "bpf_counter.h" #include "bpf_skel/off_cpu.skel.h" +#include "cgroup.h" +#include "cpumap.h" +#include "debug.h" +#include "evlist.h" +#include "evsel.h" +#include "off_cpu.h" +#include "parse-events.h" +#include "perf-hooks.h" +#include "record.h" +#include "session.h" +#include "strlist.h" +#include "target.h" +#include "thread_map.h" #define MAX_STACKS 32#define MAX_PROC 4096
tools/perf/util/bpf_trace_augment.c +4 -4
@@ -1,11 +1,11 @@#include <bpf/libbpf.h> #include <internal/xyarray.h> -#include "util/debug.h" -#include "util/evlist.h" -#include "util/trace_augment.h" -#include "bpf_skel/augmented_raw_syscalls.skel.h" +#include "debug.h" +#include "evlist.h" +#include "parse-events.h" +#include "trace_augment.h" static struct augmented_raw_syscalls_bpf *skel;static struct evsel *bpf_output;
tools/perf/util/evlist.c +46 -45
113 lines not on this page · raw message
tools/perf/util/map.h +4 -5
@@ -2,14 +2,13 @@#ifndef __PERF_MAP_H#define __PERF_MAP_H -#include <linux/refcount.h> -#include <linux/compiler.h> -#include <linux/list.h> -#include <linux/rbtree.h> +#include <stdbool.h> #include <stdio.h> #include <string.h> -#include <stdbool.h> + +#include <linux/refcount.h> #include <linux/types.h> +#include <internal/rc_check.h> struct dso ;
tools/perf/util/perf_api_probe.c +11 -7
@@ -1,14 +1,18 @@/* SPDX-License-Identifier: GPL-2.0 */ +#include "perf_api_probe.h" -#include "perf-sys.h" -#include "util/cloexec.h" -#include "util/evlist.h" -#include "util/evsel.h" -#include "util/parse-events.h" -#include "util/perf_api_probe.h" -#include <perf/cpumap.h> #include <errno.h> +#include <perf/cpumap.h> + +#include "cloexec.h" +#include "evlist.h" +#include "evsel.h" +#include "parse-events.h" +#include "perf-sys.h" +#include "pmu.h" +#include "pmus.h" +typedef void (*setup_probe_fn_t )(struct evsel *evsel);static int perf_do_probe_api (setup_probe_fn_t fn,struct perf_cpu cpu,const char *str)
tools/perf/util/s390-sample-raw.c +10 -9
@@ -12,25 +12,26 @@* sample was taken from. */ -#include <unistd.h> +#include <inttypes.h> #include <stdio.h> #include <string.h> -#include <inttypes.h> -#include <sys/stat.h> +#include <asm/byteorder.h> #include <linux/compiler.h> #include <linux/err.h> -#include <asm/byteorder.h> +#include <sys/stat.h> +#include <unistd.h> +#include "color.h" #include "debug.h" -#include "session.h" #include "evlist.h" -#include "color.h" #include "hashmap.h" -#include "sample-raw.h" +#include "pmu.h" +#include "pmus.h" #include "s390-cpumcf-kernel.h" -#include "util/pmu.h" -#include "util/sample.h" +#include "sample-raw.h" +#include "sample.h" +#include "session.h" static size_t ctrset_size (struct cf_ctrset_entry *set) {
tools/perf/util/stat-shadow.c +12 -8
@@ -2,20 +2,24 @@#include <errno.h> #include <math.h> #include <stdio.h> -#include "evsel.h" -#include "stat.h" + +#include <linux/zalloc.h> + +#include "cgroup.h" #include "color.h" #include "debug.h" -#include "pmu.h" -#include "rblist.h" #include "evlist.h" +#include "evsel.h" #include "expr.h" -#include "metricgroup.h" -#include "cgroup.h" -#include "units.h" +#include "hashmap.h" #include "iostat.h" -#include "util/hashmap.h" +#include "metricgroup.h" +#include "pmu.h" +#include "pmus.h" +#include "rblist.h" +#include "stat.h" #include "tool_pmu.h" +#include "units.h" static bool tool_pmu__is_time_event (const struct perf_stat_config *config,const struct evsel *evsel,int *tool_aggr_idx)
tools/perf/util/stat.c +10 -6
@@ -1,21 +1,25 @@// SPDX-License-Identifier: GPL-2.0 +#include "stat.h" +#include <errno.h> -#include <linux/err.h> #include <inttypes.h> #include <math.h> #include <string.h> + +#include <linux/err.h> +#include <linux/zalloc.h> +#include "counts.h" #include "cpumap.h" #include "debug.h" +#include "evlist.h" +#include "evsel.h" +#include "hashmap.h" #include "header.h" -#include "stat.h" +#include "pmu.h" #include "session.h" #include "target.h" -#include "evlist.h" -#include "evsel.h" #include "thread_map.h" -#include "util/hashmap.h" -#include <linux/zalloc.h> void update_stats (struct stats *stats,u64 val) {
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
Ian Rogers
reviewerPATCH v10 2/19 perf python: Add missed explicit dependencies
In reply to Ian RogersFix missing #include of pmus.h found while cleaning the evsel/evlist header files.
Also define PY_SSIZE_T_CLEAN before including Python.h to comply with modern Python 3 C-API requirements, and introduce CHECK_INITIALIZED and CHECK_INITIALIZED_INT safety macros to be used by subsequent patches.
--- tools/perf/util/python.c | 41 +++++++++++++++++++++++++++++++--------- 1 file changed, 32 insertions(+), 9 deletions(-)
tools/perf/util/python.c +32 -9
@@ -1,30 +1,37 @@// SPDX-License-Identifier: GPL-2.0 +#define PY_SSIZE_T_CLEAN #include <Python.h> -#include <structmember.h> +#include <inttypes.h> -#include <poll.h> +#include <linux/err.h> +#include <poll.h> + +#include <internal/lib.h> #include <perf/cpumap.h> -#ifdef HAVE_LIBTRACEEVENT -#include <event-parse.h> -#endif #include <perf/mmap.h> +#include <structmember.h> +#include "callchain.h" #include "counts.h" +#include "event.h" #include "evlist.h" #include "evsel.h" -#include "event.h" #include "expr.h" +#include "metricgroup.h" +#include "mmap.h" +#include "pmus.h" #include "print_binary.h" #include "record.h" #include "strbuf.h" #include "thread_map.h" #include "tp_pmu.h" #include "trace-event.h" -#include "metricgroup.h" -#include "mmap.h" #include "util/sample.h" -#include <internal/lib.h> + +#ifdef HAVE_LIBTRACEEVENT +#include <event-parse.h> +#endif PyMODINIT_FUNC PyInit_perf (void );
@@ -38,6 +45,22 @@ PyMODINIT_FUNC PyInit_perf(void); offsetof(struct pyrf_event , sample) + offsetof(struct perf_sample , member), \ 0, help } +#define CHECK_INITIALIZED (ptr, msg) \ + do { \ + if (!(ptr)) { \ + PyErr_SetString(PyExc_ValueError, msg " not initialized"); \ + return NULL; \ + } \ + } while (0) + +#define CHECK_INITIALIZED_INT (ptr, msg) \ + do { \ + if (!(ptr)) { \ + PyErr_SetString(PyExc_ValueError, msg " not initialized"); \ + return -1; \ + } \ + } while (0) +struct pyrf_event {PyObject_HEAD struct evsel *evsel;
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
Ian Rogers
reviewerPATCH v10 3/19 perf evsel/evlist: Avoid unnecessary #includes
In reply to Ian RogersUse forward declarations and remove unnecessary #includes in evsel.h. Sort the forward declarations in evsel.h and evlist.h. Move some PMU code into evsel.c.
--- tools/perf/util/evlist.h | 15 +++++++++------ tools/perf/util/evsel.c | 5 +++++ tools/perf/util/evsel.h | 25 ++++++++++++------------- 3 files changed, 26 insertions(+), 19 deletions(-)
tools/perf/util/evlist.h +9 -6
@@ -2,29 +2,32 @@#ifndef __PERF_EVLIST_H#define __PERF_EVLIST_H 1 +#include <signal.h> +#include <linux/compiler.h> #include <linux/kernel.h> -#include <linux/refcount.h> #include <linux/list.h> +#include <linux/refcount.h> +#include <pthread.h> +#include <unistd.h> +#include <api/fd/array.h> #include <internal/evlist.h> #include <internal/evsel.h> #include <perf/evlist.h> +#include "affinity.h" #include "events_stats.h" #include "evsel.h" #include "rblist.h" -#include <pthread.h> -#include <signal.h> -#include <unistd.h> -struct pollfd ; -struct thread_map ;struct perf_cpu_map ;struct perf_stat_config ; +struct pollfd ;struct record_opts ;struct strbuf ;struct target ; +struct thread_map ;/* * State machine of bkw_mmap_state:
tools/perf/util/evsel.c +5 -0
@@ -370,6 +370,11 @@ void evsel__set_sample_id(struct evsel *evsel, evsel->core.attr.read_format |=PERF_FORMAT_ID ; } +bool evsel__is_non_perf_event_open_pmu (const struct evsel *evsel) +{ +return evsel->pmu && evsel->pmu->type >PERF_PMU_TYPE_PE_END ; +} +/** * evsel__is_function_event - Return whether given evsel is a function * trace event
tools/perf/util/evsel.h +12 -13
@@ -2,28 +2,30 @@#ifndef __PERF_EVSEL_H#define __PERF_EVSEL_H 1 -#include <linux/list.h> #include <stdbool.h> -#include <sys/types.h> + +#include <linux/list.h> #include <linux/perf_event.h> #include <linux/types.h> +#include <sys/types.h> +#include <internal/evsel.h> #include <perf/evsel.h> +#include "symbol_conf.h" -#include "pmus.h" -#include "pmu.h" +struct bperf_follower_bpf ; +struct bperf_leader_bpf ; +struct bpf_counter_ops ;struct bpf_object ;struct cgroup ; +struct hashmap ;struct perf_counts ; +struct perf_pmu ;struct perf_stat_config ;struct perf_stat_evsel ; -union perf_event ; -struct bpf_counter_ops ;struct target ; -struct hashmap ; -struct bperf_leader_bpf ; -struct bperf_follower_bpf ; +union perf_event ;typedef int (evsel__sb_cb_t )(union perf_event *event,void *data);
@@ -356,10 +358,6 @@ void arch_evsel__apply_ratio_to_prev(struct evsel *evsel, struct perf_event_attrint evsel__set_filter (struct evsel *evsel,const char *filter);int evsel__append_tp_filter (struct evsel *evsel,const char *filter);int evsel__append_addr_filter (struct evsel *evsel,const char *filter); -static inline bool evsel__is_non_perf_event_open_pmu (const struct evsel *evsel) -{ -return evsel->pmu && evsel->pmu->type >PERF_PMU_TYPE_PE_END ; -}int evsel__enable_cpu (struct evsel *evsel,int cpu_map_idx);int evsel__enable (struct evsel *evsel);
@@ -478,6 +476,7 @@ static inline bool evsel__is_group_event(struct evsel *evsel)return evsel__is_group_leader (evsel) && evsel->core.nr_members >1 ; } +bool evsel__is_non_perf_event_open_pmu (const struct evsel *evsel);bool evsel__is_function_event (struct evsel *evsel);static inline bool evsel__is_bpf_output (struct evsel *evsel)
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
Ian Rogers
reviewerPATCH v10 4/19 perf data: Add open flag
In reply to Ian RogersAvoid double opens and ensure only open files are closed. This addresses some issues with python integration where the data file wants to be opened before being given to a session.
---
Changes in v2:
1. Fixed File Rotation: In perf_data__switch() , I added data->open = false; after the file is closed. This ensures that the subsequent perf_data__open() call will not exit early and will successfully open the new file.
2. Fixed Memory Leak: In open_dir() , I added a call to zfree(&data->file.path) if mkdir() fails, preventing the leak of the path string.
--- tools/perf/util/data.c | 27 ++++++++++++++++++++++----- tools/perf/util/data.h | 4 +++- 2 files changed, 25 insertions(+), 6 deletions(-)
tools/perf/util/data.c +22 -5
@@ -345,9 +345,10 @@ static int open_dir(struct perf_data *data)if (asprintf (&data->file.path,"%s/data" , data->path) <0 )return -1 ; -if (perf_data__is_write (data) && -mkdir (data->path,S_IRWXU ) <0 ) +if (perf_data__is_write (data) &&mkdir (data->path,0700 ) <0 ) { +zfree (&data->file.path);return -1 ; + } ret =open_file (data);
@@ -360,9 +361,16 @@ static int open_dir(struct perf_data *data)int perf_data__open (struct perf_data *data) { -if (check_pipe (data)) +int ret; + +if (data->open)return 0 ; +if (check_pipe (data)) { + data->open = true; +return 0 ; + } +/* currently it allows stdio for pipe only */ data->file.use_stdio = false;
@@ -375,16 +383,24 @@ int perf_data__open(struct perf_data *data)if (perf_data__is_read (data)) data->is_dir =is_dir (data); -return perf_data__is_dir (data) ? -open_dir (data) :open_file_dup (data); + ret =perf_data__is_dir (data) ?open_dir (data) :open_file_dup (data); + +if (!ret) + data->open = true; + +return ret; }void perf_data__close (struct perf_data *data) { +if (!data->open) +return ; +if (perf_data__is_dir (data))perf_data__close_dir (data);perf_data_file__close (&data->file); + data->open = false; }static ssize_t perf_data_file__read (struct perf_data_file *file,void *buf,size_t size)
@@ -457,6 +473,7 @@ int perf_data__switch(struct perf_data *data,if (!at_exit) {perf_data_file__close (&data->file); + data->open = false; ret =perf_data__open (data);if (ret <0 ) goto out;
tools/perf/util/data.h +3 -1
@@ -50,6 +50,8 @@ struct perf_data {const char *path;/** @file: Underlying file to be used. */ struct perf_data_file file; +/** @open: Has the file or directory been opened. */ +bool open;/** @is_pipe: Underlying file is a pipe. */ bool is_pipe;/** @is_dir: Underlying file is a directory. */
@@ -59,7 +61,7 @@ struct perf_data {/** @in_place_update: A file opened for reading but will be written to. */ bool in_place_update;/** @mode: Read or write mode. */ -enum perf_data_mode mode; + enum perf_data_mode mode:8;struct {/** @version: perf_dir_version. */
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
[email protected]
AI reviewRe: [PATCH v10 04/19] perf data: Add open flag
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
Pre-existing issues:
- [High] A time-of-check to time-of-use (TOCTOU) vulnerability exists when opening the perf data file for writing, which can be exploited to truncate arbitrary files. --
commit 6ea2670fcd0afaed23b29eb27c1bda1c709e9986
Author: Ian Rogers <[email protected]>
perf data: Add open flag
Avoid double opens and ensure only open files are closed. This addresses some issues with python integration where the data file wants to be opened before being given to a session.
Quoted 4 lines
tools/perf/util/data.c +0
-0
[ ... ]
tools/perf/util/data.c int perf_data__open(struct perf_data *data) · line 386
Quoted 7 lines
tools/perf/util/data.c +1
-2@@ -375,16 +383,24 @@ int perf_data__open(struct perf_data *data)if (perf_data__is_read (data)) data->is_dir =is_dir (data); -return perf_data__is_dir (data) ? -open_dir (data) :open_file_dup (data); + ret =perf_data__is_dir (data) ?open_dir (data) :open_file_dup (data);
[Severity: High]
This isn't a bug introduced by this patch, but is there a time-of-check to time-of-use (TOCTOU) vulnerability when opening the perf data file for writing?
In perf_data__open(), check_backup() is called to verify if the target file already exists, renaming it to .old if so. However, there is a race condition between this check and the subsequent file creation in open_file_write() called via open_file_dup():
tools/perf/util/data.c:check_backup() {
if (!stat(data->path, &st) && st.st_size) {
snprintf(oldname, sizeof(oldname), "%s.old", data->path);
...
}An attacker with write access to the target directory could create a malicious symlink at the target path in the brief window after check_backup() but before open_file_write():
tools/perf/util/data.c:open_file_write() {
int fd = open(data->file.path, O_CREAT|O_RDWR|O_TRUNC|O_CLOEXEC, S_IRUSR|S_IWUSR);
...
}Because open_file_write() opens the file without O_EXCL or O_NOFOLLOW, the open() call will follow the symlink and truncate the target file.
Could this be exploited to truncate arbitrary files if perf is executed by a privileged user in a shared directory?
Quoted 6 lines
tools/perf/util/data.c +5
-0+ +if (!ret) + data->open = true; + +return ret; }
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=4
Ian Rogers
reviewerPATCH v10 5/19 perf evlist: Add reference count
In reply to Ian RogersThis a no-op for most of the perf tool. The reference count is set to 1 at allocation, the put will see the 1, decrement it and perform the delete. The purpose for adding the reference count is for the python code. Prior to this change the python code would clone evlists, but this has issues if events are opened, etc. This change adds a reference count for the evlists and a later change will add it to evsels. The combination is needed for the python code to operate correctly (not hit asserts in the evsel clone), but the changes are broken apart for the sake of smaller patches.
---
v2: Added evlist__put to pyrf_evlist__init in case init is called more
than once.
I double-checked trace__replay() and confirmed that trace->evlist
is not assigned to session->evlist in that function.
trace__replay creates a new session and uses its own evlist for
processing file events, leaving trace->evlist pointing to the
empty list created at startup. Therefore, the
evlist__put(trace->evlist) call in trace__exit() is safe and
correct to avoid leaking that empty list.v7:
- Added pyrf_evlist__new to zero-initialize pevlist->evlist to fix crash on re-initialization.
--- tools/perf/arch/x86/tests/hybrid.c | 2 +- tools/perf/arch/x86/tests/topdown.c | 4 +- tools/perf/arch/x86/util/iostat.c | 2 +- tools/perf/bench/evlist-open-close.c | 18 +- tools/perf/builtin-ftrace.c | 8 +- tools/perf/builtin-kvm.c | 4 +- tools/perf/builtin-lock.c | 2 +- tools/perf/builtin-record.c | 4 +- tools/perf/builtin-sched.c | 6 +- tools/perf/builtin-script.c | 2 +- tools/perf/builtin-stat.c | 10 +- tools/perf/builtin-top.c | 52 ++--- tools/perf/builtin-trace.c | 26 +-- tools/perf/tests/backward-ring-buffer.c | 18 +- tools/perf/tests/code-reading.c | 4 +- tools/perf/tests/event-times.c | 4 +- tools/perf/tests/event_update.c | 2 +- tools/perf/tests/evsel-roundtrip-name.c | 8 +- tools/perf/tests/expand-cgroup.c | 8 +- tools/perf/tests/hists_cumulate.c | 2 +- tools/perf/tests/hists_filter.c | 2 +- tools/perf/tests/hists_link.c | 2 +- tools/perf/tests/hists_output.c | 2 +- tools/perf/tests/hwmon_pmu.c | 2 +- tools/perf/tests/keep-tracking.c | 2 +- tools/perf/tests/mmap-basic.c | 18 +- tools/perf/tests/openat-syscall-tp-fields.c | 18 +- tools/perf/tests/parse-events.c | 4 +- tools/perf/tests/parse-metric.c | 4 +- tools/perf/tests/parse-no-sample-id-all.c | 2 +- tools/perf/tests/perf-record.c | 18 +- tools/perf/tests/perf-time-to-tsc.c | 2 +- tools/perf/tests/pfm.c | 4 +- tools/perf/tests/pmu-events.c | 6 +- tools/perf/tests/pmu.c | 4 +- tools/perf/tests/sw-clock.c | 14 +- tools/perf/tests/switch-tracking.c | 2 +- tools/perf/tests/task-exit.c | 14 +- tools/perf/tests/tool_pmu.c | 2 +- tools/perf/tests/topology.c | 2 +- tools/perf/tests/uncore-event-sorting.c | 2 +- tools/perf/util/cgroup.c | 4 +- tools/perf/util/data-convert-bt.c | 2 +- tools/perf/util/evlist.c | 20 +- tools/perf/util/evlist.h | 7 +- tools/perf/util/expr.c | 2 +- tools/perf/util/header.c | 22 +- tools/perf/util/metricgroup.c | 6 +- tools/perf/util/parse-events.c | 4 +- tools/perf/util/perf_api_probe.c | 2 +- tools/perf/util/python.c | 242 +++++++++----------- tools/perf/util/record.c | 2 +- tools/perf/util/session.c | 2 +- tools/perf/util/sideband_evlist.c | 16 +- 54 files changed, 315 insertions(+), 328 deletions(-)
tools/perf/arch/x86/tests/hybrid.c +1 -1
@@ -268,7 +268,7 @@ static int test_event(const struct evlist_test *e) ret = e->check (evlist); }parse_events_error__exit (&err); -evlist__delete (evlist); +evlist__put (evlist);return ret; }
tools/perf/arch/x86/tests/topdown.c +2 -2
@@ -56,7 +56,7 @@ static int event_cb(void *state, struct pmu_event_info *info) *ret =TEST_FAIL ; } } -evlist__delete (evlist); +evlist__put (evlist);return 0 ; }
@@ -174,7 +174,7 @@ static int test_sort(const char *str, int expected_slots_group_size,CHECK_COND (slots_seen,"slots seen" ); ret =TEST_OK ; out_err: -evlist__delete (evlist); +evlist__put (evlist);parse_events_error__exit (&err);return ret; }
tools/perf/arch/x86/util/iostat.c +1 -1
@@ -337,7 +337,7 @@ int iostat_prepare(struct evlist *evlist, struct perf_stat_config *config)if (evlist->core.nr_entries >0 ) {pr_warning ("The -e and -M options are not supported." "All chosen events/metrics will be dropped\n" ); -evlist__delete (evlist); +evlist__put (evlist); evlist =evlist__new ();if (!evlist)return -ENOMEM ;
tools/perf/bench/evlist-open-close.c +9 -9
@@ -76,7 +76,7 @@ static struct evlist *bench__create_evlist(char *evstr, const char *uid_str)parse_events_error__exit (&err);pr_err ("Run 'perf list' for a list of valid events\n" ); ret =1 ; - goto out_delete_evlist; + goto out_put_evlist; }parse_events_error__exit (&err);if (uid_str) {
@@ -85,24 +85,24 @@ static struct evlist *bench__create_evlist(char *evstr, const char *uid_str)if (uid ==UINT_MAX ) {pr_err ("Invalid User: %s" , uid_str); ret = -EINVAL ; - goto out_delete_evlist; + goto out_put_evlist; } ret =parse_uid_filter (evlist, uid);if (ret) - goto out_delete_evlist; + goto out_put_evlist; } ret =evlist__create_maps (evlist, &opts.target);if (ret <0 ) {pr_err ("Not enough memory to create thread/cpu maps\n" ); - goto out_delete_evlist; + goto out_put_evlist; }evlist__config (evlist, &opts,NULL );return evlist; -out_delete_evlist: -evlist__delete (evlist); +out_put_evlist: +evlist__put (evlist);return NULL ; }
@@ -151,7 +151,7 @@ static int bench_evlist_open_close__run(char *evstr, const char *uid_str) evlist->core.nr_entries, evlist__count_evsel_fds(evlist));printf (" Number of iterations:\t%d\n" , iterations); -evlist__delete (evlist); +evlist__put (evlist);for (i =0 ; i < iterations; i++) {pr_debug ("Started iteration %d\n" , i);
@@ -162,7 +162,7 @@ static int bench_evlist_open_close__run(char *evstr, const char *uid_str)gettimeofday (&start,NULL ); err =bench__do_evlist_open_close (evlist);if (err) { -evlist__delete (evlist); +evlist__put (evlist);return err; }
@@ -171,7 +171,7 @@ static int bench_evlist_open_close__run(char *evstr, const char *uid_str) runtime_us =timeval2usec (&diff);update_stats (&time_stats, runtime_us); -evlist__delete (evlist); +evlist__put (evlist);pr_debug ("Iteration %d took:\t%" PRIu64"us\n" , i, runtime_us); }
tools/perf/builtin-ftrace.c +4 -4
@@ -1999,20 +1999,20 @@ int cmd_ftrace(int argc, const char **argv) ret =evlist__create_maps (ftrace.evlist, &ftrace.target);if (ret <0 ) - goto out_delete_evlist; + goto out_put_evlist;if (argc) { ret =evlist__prepare_workload (ftrace.evlist, &ftrace.target, argv, false, ftrace__workload_exec_failed_signal);if (ret <0 ) - goto out_delete_evlist; + goto out_put_evlist; } ret =cmd_func (&ftrace); -out_delete_evlist: -evlist__delete (ftrace.evlist); +out_put_evlist: +evlist__put (ftrace.evlist); out_delete_filters:delete_filter_func (&ftrace.filters);
tools/perf/builtin-kvm.c +2 -2
@@ -1810,7 +1810,7 @@ static struct evlist *kvm_live_event_list(void) out:if (err) { -evlist__delete (evlist); +evlist__put (evlist); evlist =NULL ; }
@@ -1941,7 +1941,7 @@ static int kvm_events_live(struct perf_kvm_stat *kvm, out:perf_session__delete (kvm->session); kvm->session =NULL ; -evlist__delete (kvm->evlist); +evlist__put (kvm->evlist);return err; }
tools/perf/builtin-lock.c +1 -1
@@ -2149,7 +2149,7 @@ static int __cmd_contention(int argc, const char **argv) out_delete:lock_filter_finish (); -evlist__delete (con.evlist); +evlist__put (con.evlist);lock_contention_finish (&con);perf_session__delete (session);perf_env__exit (&host_env);
tools/perf/builtin-record.c +2 -2
@@ -4290,7 +4290,7 @@ int cmd_record(int argc, const char **argv) goto out;evlist__splice_list_tail (rec->evlist, &def_evlist->core.entries); -evlist__delete (def_evlist); +evlist__put (def_evlist); }if (rec->opts.target.tid && !rec->opts.no_inherit_set)
@@ -4400,7 +4400,7 @@ int cmd_record(int argc, const char **argv)auxtrace_record__free (rec->itr); out_opts:evlist__close_control (rec->opts.ctl_fd, rec->opts.ctl_fd_ack, &rec->opts.ctl_fd_close); -evlist__delete (rec->evlist); +evlist__put (rec->evlist);return err; }
tools/perf/builtin-sched.c +3 -3
@@ -3848,7 +3848,7 @@ static int perf_sched__schedstat_record(struct perf_sched *sched, session =perf_session__new (&data, &sched->tool);if (IS_ERR (session)) {pr_err ("Perf session creation failed.\n" ); -evlist__delete (evlist); +evlist__put (evlist);return PTR_ERR (session); }
@@ -3947,7 +3947,7 @@ static int perf_sched__schedstat_record(struct perf_sched *sched,else fprintf (stderr,"[ perf sched stats: Failed !! ]\n" ); -evlist__delete (evlist); +evlist__put (evlist);close (fd);return err; }
@@ -4851,7 +4851,7 @@ static int perf_sched__schedstat_live(struct perf_sched *sched,free_cpu_domain_info (cd_map, sv, nr); out:free_schedstat (&cpu_head); -evlist__delete (evlist); +evlist__put (evlist);return err; }
tools/perf/builtin-script.c +1 -1
@@ -2274,7 +2274,7 @@ static int script_find_metrics(const struct pmu_metric *pm, }pr_debug ("Found metric '%s' whose evsels match those of in the perf data\n" , pm->metric_name); -evlist__delete (metric_evlist); +evlist__put (metric_evlist); out:return 0 ; }
tools/perf/builtin-stat.c +5 -5
@@ -2113,7 +2113,7 @@ static int add_default_events(void) stat_config.user_requested_cpu_list, stat_config.system_wide, stat_config.hardware_aware_grouping) < 0) { -evlist__delete (metric_evlist); +evlist__put (metric_evlist); ret =-1 ;break ; }
@@ -2125,7 +2125,7 @@ static int add_default_events(void)metricgroup__copy_metric_events (evlist,/*cgrp=*/ NULL , &evlist->metric_events, &metric_evlist->metric_events); -evlist__delete (metric_evlist); +evlist__put (metric_evlist); }list_sort (/*priv=*/ NULL , &evlist->core.entries, default_evlist_evsel_cmp);
@@ -2146,7 +2146,7 @@ static int add_default_events(void)metricgroup__copy_metric_events (evsel_list,/*cgrp=*/ NULL , &evsel_list->metric_events, &evlist->metric_events); -evlist__delete (evlist); +evlist__put (evlist);return ret; }
@@ -2381,7 +2381,7 @@ static int __cmd_report(int argc, const char **argv) perf_stat.session = session; stat_config.output = stderr; -evlist__delete (evsel_list); +evlist__put (evsel_list); evsel_list = session->evlist; ret =perf_session__process_events (session);
@@ -3060,7 +3060,7 @@ int cmd_stat(int argc, const char **argv)if (smi_cost && smi_reset)sysfs__write_int (FREEZE_ON_SMI_PATH ,0 ); -evlist__delete (evsel_list); +evlist__put (evsel_list);evlist__close_control (stat_config.ctl_fd, stat_config.ctl_fd_ack, &stat_config.ctl_fd_close);
tools/perf/builtin-top.c +26 -26
184 lines not on this page · raw message
tools/perf/builtin-trace.c +13 -13
@@ -4418,7 +4418,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv)if (trace->summary_bpf) {if (trace_prepare_bpf_summary (trace->summary_mode) <0 ) - goto out_delete_evlist; + goto out_put_evlist;if (trace->summary_only) goto create_maps;
@@ -4486,19 +4486,19 @@ static int trace__run(struct trace *trace, int argc, const char **argv) err =evlist__create_maps (evlist, &trace->opts.target);if (err <0 ) {fprintf (trace->output,"Problems parsing the target to trace, check your options!\n" ); - goto out_delete_evlist; + goto out_put_evlist; } err =trace__symbols_init (trace, argc, argv, evlist);if (err <0 ) {fprintf (trace->output,"Problems initializing symbol libraries!\n" ); - goto out_delete_evlist; + goto out_put_evlist; }if (trace->summary_mode ==SUMMARY__BY_TOTAL && !trace->summary_bpf) { trace->syscall_stats =alloc_syscall_stats ();if (!trace->syscall_stats) - goto out_delete_evlist; + goto out_put_evlist; }evlist__config (evlist, &trace->opts, &callchain_param);
@@ -4507,7 +4507,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv) err =evlist__prepare_workload (evlist, &trace->opts.target, argv, false,NULL );if (err <0 ) {fprintf (trace->output,"Couldn't run the workload!\n" ); - goto out_delete_evlist; + goto out_put_evlist; } workload_pid = evlist->workload.pid; }
@@ -4555,7 +4555,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv) err =trace__expand_filters (trace, &evsel);if (err) - goto out_delete_evlist; + goto out_put_evlist; err =evlist__apply_filters (evlist, &evsel, &trace->opts.target);if (err <0 ) goto out_error_apply_filters;
@@ -4672,12 +4672,12 @@ static int trace__run(struct trace *trace, int argc, const char **argv) } } -out_delete_evlist: +out_put_evlist:trace_cleanup_bpf_summary ();delete_syscall_stats (trace->syscall_stats);trace__symbols__exit (trace);evlist__free_syscall_tp_fields (evlist); -evlist__delete (evlist); +evlist__put (evlist);cgroup__put (trace->cgroup); trace->evlist =NULL ; trace->live = false;
@@ -4702,21 +4702,21 @@ static int trace__run(struct trace *trace, int argc, const char **argv) out_error:fprintf (trace->output,"%s\n" , errbuf); - goto out_delete_evlist; + goto out_put_evlist; out_error_apply_filters:fprintf (trace->output,"Failed to set filter \"%s\" on event %s: %m\n" , evsel->filter,evsel__name (evsel)); - goto out_delete_evlist; + goto out_put_evlist; } out_error_mem:fprintf (trace->output,"Not enough memory to run!\n" ); - goto out_delete_evlist; + goto out_put_evlist; out_errno:fprintf (trace->output,"%m\n" ); - goto out_delete_evlist; + goto out_put_evlist; }static int trace__replay (struct trace *trace)
@@ -5396,7 +5396,7 @@ static void trace__exit(struct trace *trace)zfree (&trace->syscalls.table); }zfree (&trace->perfconfig_events); -evlist__delete (trace->evlist); +evlist__put (trace->evlist); trace->evlist =NULL ;ordered_events__free (&trace->oe.data);#ifdef HAVE_LIBBPF_SUPPORT
tools/perf/tests/backward-ring-buffer.c +9 -9
@@ -111,7 +111,7 @@ static int test__backward_ring_buffer(struct test_suite *test __maybe_unused, in err =evlist__create_maps (evlist, &opts.target);if (err <0 ) {pr_debug ("Not enough memory to create thread/cpu maps\n" ); - goto out_delete_evlist; + goto out_put_evlist; }parse_events_error__init (&parse_error);
@@ -124,7 +124,7 @@ static int test__backward_ring_buffer(struct test_suite *test __maybe_unused, inif (err) {pr_debug ("Failed to parse tracepoint event, try use root\n" ); ret =TEST_SKIP ; - goto out_delete_evlist; + goto out_put_evlist; }evlist__config (evlist, &opts,NULL );
@@ -133,19 +133,19 @@ static int test__backward_ring_buffer(struct test_suite *test __maybe_unused, inif (err <0 ) {pr_debug ("perf_evlist__open: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf))); - goto out_delete_evlist; + goto out_put_evlist; } ret =TEST_FAIL ; err =do_test (evlist, opts.mmap_pages, &sample_count, &comm_count);if (err !=TEST_OK ) - goto out_delete_evlist; + goto out_put_evlist;if ((sample_count !=NR_ITERS ) || (comm_count !=NR_ITERS )) {pr_err ("Unexpected counter: sample_count=%d, comm_count=%d\n" , sample_count, comm_count); - goto out_delete_evlist; + goto out_put_evlist; }evlist__close (evlist);
@@ -154,16 +154,16 @@ static int test__backward_ring_buffer(struct test_suite *test __maybe_unused, inif (err <0 ) {pr_debug ("perf_evlist__open: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf))); - goto out_delete_evlist; + goto out_put_evlist; } err =do_test (evlist,1 , &sample_count, &comm_count);if (err !=TEST_OK ) - goto out_delete_evlist; + goto out_put_evlist; ret =TEST_OK ; -out_delete_evlist: -evlist__delete (evlist); +out_put_evlist: +evlist__put (evlist);return ret; }
tools/perf/tests/code-reading.c +2 -2
@@ -807,7 +807,7 @@ static int do_test_code_reading(bool try_kcore) }perf_evlist__set_maps (&evlist->core,NULL ,NULL ); -evlist__delete (evlist); +evlist__put (evlist); evlist =NULL ;continue ; }
@@ -844,7 +844,7 @@ static int do_test_code_reading(bool try_kcore) out_put:thread__put (thread); out_err: -evlist__delete (evlist); +evlist__put (evlist);perf_cpu_map__put (cpus);perf_thread_map__put (threads);machine__delete (machine);
tools/perf/tests/event-times.c +2 -2
@@ -186,7 +186,7 @@ static int test_times(int (attach)(struct evlist *), err =attach (evlist);if (err ==TEST_SKIP ) {pr_debug (" SKIP : not enough rights\n" ); -evlist__delete (evlist); +evlist__put (evlist);return err; }
@@ -205,7 +205,7 @@ static int test_times(int (attach)(struct evlist *), count.ena, count.run); out_err: -evlist__delete (evlist); +evlist__put (evlist);return !err ?TEST_OK :TEST_FAIL ; }
tools/perf/tests/event_update.c +1 -1
@@ -117,7 +117,7 @@ static int test__event_update(struct test_suite *test __maybe_unused, int subtesTEST_ASSERT_VAL ("failed to synthesize attr update cpus" , !perf_event__synthesize_event_update_cpus (&tmp.tool, evsel, process_event_cpus)); -evlist__delete (evlist); +evlist__put (evlist);return 0 ; }
tools/perf/tests/evsel-roundtrip-name.c +4 -4
@@ -33,7 +33,7 @@ static int perf_evsel__roundtrip_cache_name_test(void)if (err) {pr_debug ("Failure to parse cache event '%s' possibly as PMUs don't support it" , name); -evlist__delete (evlist); +evlist__put (evlist);continue ; }evlist__for_each_entry (evlist, evsel) {
@@ -42,7 +42,7 @@ static int perf_evsel__roundtrip_cache_name_test(void) ret =TEST_FAIL ; } } -evlist__delete (evlist); +evlist__put (evlist); } } }
@@ -66,7 +66,7 @@ static int perf_evsel__name_array_test(const char *const names[], int nr_names)if (err) {pr_debug ("failed to parse event '%s', err %d\n" , names[i], err); -evlist__delete (evlist); +evlist__put (evlist); ret =TEST_FAIL ;continue ; }
@@ -76,7 +76,7 @@ static int perf_evsel__name_array_test(const char *const names[], int nr_names) ret =TEST_FAIL ; } } -evlist__delete (evlist); +evlist__put (evlist); }return ret; }
tools/perf/tests/expand-cgroup.c +4 -4
@@ -106,7 +106,7 @@ static int expand_default_events(void)TEST_ASSERT_VAL ("failed to get evlist" , evlist); ret =test_expand_events (evlist); -evlist__delete (evlist); +evlist__put (evlist);return ret; }
@@ -133,7 +133,7 @@ static int expand_group_events(void) ret =test_expand_events (evlist); out:parse_events_error__exit (&err); -evlist__delete (evlist); +evlist__put (evlist);return ret; }
@@ -164,7 +164,7 @@ static int expand_libpfm_events(void) ret =test_expand_events (evlist); out: -evlist__delete (evlist); +evlist__put (evlist);return ret; }
@@ -188,7 +188,7 @@ static int expand_metric_events(void) ret =test_expand_events (evlist); out: -evlist__delete (evlist); +evlist__put (evlist);return ret; }
tools/perf/tests/hists_cumulate.c +1 -1
@@ -743,7 +743,7 @@ static int test__hists_cumulate(struct test_suite *test __maybe_unused, int subt out:/* tear down everything */ -evlist__delete (evlist); +evlist__put (evlist);machines__exit (&machines);put_fake_samples ();
tools/perf/tests/hists_filter.c +1 -1
@@ -331,7 +331,7 @@ static int test__hists_filter(struct test_suite *test __maybe_unused, int subtes out:/* tear down everything */ -evlist__delete (evlist); +evlist__put (evlist);reset_output_field ();machines__exit (&machines);put_fake_samples ();
tools/perf/tests/hists_link.c +1 -1
@@ -352,7 +352,7 @@ static int test__hists_link(struct test_suite *test __maybe_unused, int subtest out:/* tear down everything */ -evlist__delete (evlist); +evlist__put (evlist);reset_output_field ();machines__exit (&machines);put_fake_samples ();
tools/perf/tests/hists_output.c +1 -1
@@ -630,7 +630,7 @@ static int test__hists_output(struct test_suite *test __maybe_unused, int subtes out:/* tear down everything */ -evlist__delete (evlist); +evlist__put (evlist);machines__exit (&machines);put_fake_samples ();
tools/perf/tests/hwmon_pmu.c +1 -1
@@ -215,7 +215,7 @@ static int do_test(size_t i, bool with_pmu, bool with_alias) out:parse_events_error__exit (&err); -evlist__delete (evlist); +evlist__put (evlist);return ret; }
tools/perf/tests/keep-tracking.c +1 -1
@@ -153,7 +153,7 @@ static int test__keep_tracking(struct test_suite *test __maybe_unused, int subte out_err:if (evlist) {evlist__disable (evlist); -evlist__delete (evlist); +evlist__put (evlist); }perf_cpu_map__put (cpus);perf_thread_map__put (threads);
tools/perf/tests/mmap-basic.c +9 -9
@@ -94,7 +94,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtest/* Permissions failure, flag the failure as a skip. */ err =TEST_SKIP ; } - goto out_delete_evlist; + goto out_put_evlist; } evsels[i]->core.attr.wakeup_events =1 ;
@@ -106,7 +106,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtestpr_debug ("failed to open counter: %s, " "tweak /proc/sys/kernel/perf_event_paranoid?\n" ,str_error_r (errno, sbuf,sizeof (sbuf))); - goto out_delete_evlist; + goto out_put_evlist; } nr_events[i] =0 ;
@@ -116,7 +116,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtestif (evlist__mmap (evlist,128 ) <0 ) {pr_debug ("failed to mmap events: %d (%s)\n" , errno,str_error_r (errno, sbuf,sizeof (sbuf))); - goto out_delete_evlist; + goto out_put_evlist; }for (i =0 ; i < nsyscalls; ++i)
@@ -134,7 +134,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtestif (event->header.type !=PERF_RECORD_SAMPLE ) {pr_debug ("unexpected %s event\n" ,perf_event__name (event->header.type)); - goto out_delete_evlist; + goto out_put_evlist; }perf_sample__init (&sample,/*all=*/ false);
@@ -142,7 +142,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtestif (err) {pr_err ("Can't parse sample, err = %d\n" , err);perf_sample__exit (&sample); - goto out_delete_evlist; + goto out_put_evlist; } err =-1 ;
@@ -153,7 +153,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtestif (evsel ==NULL ) {pr_debug ("event with id %" PRIu64" doesn't map to an evsel\n" , sample.id); - goto out_delete_evlist; + goto out_put_evlist; } nr_events[evsel->core.idx]++;perf_mmap__consume (&md->core);
@@ -168,12 +168,12 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtest expected_nr_events[evsel->core.idx], evsel__name(evsel), nr_events[evsel->core.idx]); err =-1 ; - goto out_delete_evlist; + goto out_put_evlist; } } -out_delete_evlist: -evlist__delete (evlist); +out_put_evlist: +evlist__put (evlist); out_free_cpus:perf_cpu_map__put (cpus); out_free_threads:
tools/perf/tests/openat-syscall-tp-fields.c +9 -9
@@ -51,7 +51,7 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unusedif (IS_ERR (evsel)) {pr_debug ("%s: evsel__newtp\n" , __func__); ret =PTR_ERR (evsel) == -EACCES ?TEST_SKIP :TEST_FAIL ; - goto out_delete_evlist; + goto out_put_evlist; }evlist__add (evlist, evsel);
@@ -59,7 +59,7 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unused err =evlist__create_maps (evlist, &opts.target);if (err <0 ) {pr_debug ("%s: evlist__create_maps\n" , __func__); - goto out_delete_evlist; + goto out_put_evlist; }evsel__config (evsel, &opts,NULL );
@@ -70,14 +70,14 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unusedif (err <0 ) {pr_debug ("perf_evlist__open: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf))); - goto out_delete_evlist; + goto out_put_evlist; } err =evlist__mmap (evlist,UINT_MAX );if (err <0 ) {pr_debug ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf))); - goto out_delete_evlist; + goto out_put_evlist; }evlist__enable (evlist);
@@ -115,7 +115,7 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unusedif (err) {pr_debug ("Can't parse sample, err = %d\n" , err);perf_sample__exit (&sample); - goto out_delete_evlist; + goto out_put_evlist; } tp_flags =perf_sample__intval (&sample,"flags" );
@@ -126,7 +126,7 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unused (tp_flags & flags) != flags) {pr_debug ("%s: Expected flags=%#x, got %#x\n" , __func__, flags, tp_flags); - goto out_delete_evlist; + goto out_put_evlist; } goto out_ok;
@@ -139,13 +139,13 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unusedif (++nr_polls >5 ) {pr_debug ("%s: no events!\n" , __func__); - goto out_delete_evlist; + goto out_put_evlist; } } out_ok: ret =TEST_OK ; -out_delete_evlist: -evlist__delete (evlist); +out_put_evlist: +evlist__put (evlist); out:return ret; }
tools/perf/tests/parse-events.c +2 -2
@@ -2568,7 +2568,7 @@ static int test_event(const struct evlist_test *e) ret = e->check (evlist); }parse_events_error__exit (&err); -evlist__delete (evlist); +evlist__put (evlist);return ret; }
@@ -2594,7 +2594,7 @@ static int test_event_fake_pmu(const char *str) }parse_events_error__exit (&err); -evlist__delete (evlist); +evlist__put (evlist);return ret; }
tools/perf/tests/parse-metric.c +2 -2
@@ -84,7 +84,7 @@ static int __compute_metric(const char *name, struct value *vals, cpus =perf_cpu_map__new ("0" );if (!cpus) { -evlist__delete (evlist); +evlist__put (evlist);return -ENOMEM ; }
@@ -113,7 +113,7 @@ static int __compute_metric(const char *name, struct value *vals,/* ... cleanup. */ evlist__free_stats (evlist);perf_cpu_map__put (cpus); -evlist__delete (evlist); +evlist__put (evlist);return err; }
tools/perf/tests/parse-no-sample-id-all.c +1 -1
@@ -49,7 +49,7 @@ static int process_events(union perf_event **events, size_t count)for (i =0 ; i < count && !err; i++) err =process_event (&evlist, events[i]); -evlist__delete (evlist); +evlist__put (evlist);return err; }
tools/perf/tests/perf-record.c +9 -9
@@ -105,7 +105,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtest err =evlist__create_maps (evlist, &opts.target);if (err <0 ) {pr_debug ("Not enough memory to create thread/cpu maps\n" ); - goto out_delete_evlist; + goto out_put_evlist; }/*
@@ -117,7 +117,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtest err =evlist__prepare_workload (evlist, &opts.target, argv, false,NULL );if (err <0 ) {pr_debug ("Couldn't run the workload!\n" ); - goto out_delete_evlist; + goto out_put_evlist; }/*
@@ -134,7 +134,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtestpr_debug ("sched__get_first_possible_cpu: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));evlist__cancel_workload (evlist); - goto out_delete_evlist; + goto out_put_evlist; } cpu = err;
@@ -146,7 +146,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtestpr_debug ("sched_setaffinity: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));evlist__cancel_workload (evlist); - goto out_delete_evlist; + goto out_put_evlist; }/*
@@ -158,7 +158,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtestpr_debug ("perf_evlist__open: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));evlist__cancel_workload (evlist); - goto out_delete_evlist; + goto out_put_evlist; }/*
@@ -171,7 +171,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtestpr_debug ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));evlist__cancel_workload (evlist); - goto out_delete_evlist; + goto out_put_evlist; }/*
@@ -209,7 +209,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtestif (verbose >0 )perf_event__fprintf (event,NULL , stderr);pr_debug ("Couldn't parse sample\n" ); - goto out_delete_evlist; + goto out_put_evlist; }if (verbose >0 ) {
@@ -350,9 +350,9 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtestpr_debug ("PERF_RECORD_MMAP for %s missing!\n" ,"[vdso]" ); ++errs; } -out_delete_evlist: +out_put_evlist:CPU_FREE (cpu_mask); -evlist__delete (evlist); +evlist__put (evlist); out:perf_sample__exit (&sample);if (err == -EACCES )
tools/perf/tests/perf-time-to-tsc.c +1 -1
@@ -201,7 +201,7 @@ static int test__perf_time_to_tsc(struct test_suite *test __maybe_unused, int su err =TEST_OK ; out_err: -evlist__delete (evlist); +evlist__put (evlist);perf_cpu_map__put (cpus);perf_thread_map__put (threads);return err;
tools/perf/tests/pfm.c +2 -2
@@ -80,7 +80,7 @@ static int test__pfm_events(struct test_suite *test __maybe_unused,evlist__nr_groups (evlist), 0); -evlist__delete (evlist); +evlist__put (evlist); }return 0 ; }
@@ -165,7 +165,7 @@ static int test__pfm_group(struct test_suite *test __maybe_unused,evlist__nr_groups (evlist), table[i].nr_groups); -evlist__delete (evlist); +evlist__put (evlist); }return 0 ; }
tools/perf/tests/pmu-events.c +3 -3
@@ -798,7 +798,7 @@ static int check_parse_id(const char *id, struct parse_events_error *error) /*warn_if_reordered=*/true, /*fake_tp=*/false);free (dup); -evlist__delete (evlist); +evlist__put (evlist);return ret; }
@@ -865,7 +865,7 @@ static int test__parsing_callback(const struct pmu_metric *pm, cpus =perf_cpu_map__new ("0" );if (!cpus) { -evlist__delete (evlist); +evlist__put (evlist);return -ENOMEM ; }
@@ -919,7 +919,7 @@ static int test__parsing_callback(const struct pmu_metric *pm,/* ... cleanup. */ evlist__free_stats (evlist);perf_cpu_map__put (cpus); -evlist__delete (evlist); +evlist__put (evlist);return err; }
tools/perf/tests/pmu.c +2 -2
@@ -294,7 +294,7 @@ static int test__pmu_config_helpers(struct test_suite *test __maybe_unused, ret =TEST_OK ; err_out:parse_events_terms__exit (&terms); -evlist__delete (evlist); +evlist__put (evlist);test_pmu_put (dir, pmu);return ret; }
@@ -346,7 +346,7 @@ static int test__pmu_events(struct test_suite *test __maybe_unused, int subtest ret =TEST_OK ; err_out:parse_events_error__exit (&err); -evlist__delete (evlist); +evlist__put (evlist);test_pmu_put (dir, pmu);return ret; }
tools/perf/tests/sw-clock.c +7 -7
@@ -59,7 +59,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id) evsel =evsel__new (&attr);if (evsel ==NULL ) {pr_debug ("evsel__new\n" ); - goto out_delete_evlist; + goto out_put_evlist; }evlist__add (evlist, evsel);
@@ -68,7 +68,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id)if (!cpus || !threads) { err = -ENOMEM ;pr_debug ("Not enough memory to create thread/cpu maps\n" ); - goto out_delete_evlist; + goto out_put_evlist; }perf_evlist__set_maps (&evlist->core, cpus, threads);
@@ -80,14 +80,14 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id)pr_debug ("Couldn't open evlist: %s\nHint: check %s, using %" PRIu64" in this test.\n" ,str_error_r (errno, sbuf,sizeof (sbuf)), knob, (u64 )attr.sample_freq); - goto out_delete_evlist; + goto out_put_evlist; } err =evlist__mmap (evlist,128 );if (err <0 ) {pr_debug ("failed to mmap event: %d (%s)\n" , errno,str_error_r (errno, sbuf,sizeof (sbuf))); - goto out_delete_evlist; + goto out_put_evlist; }evlist__enable (evlist);
@@ -113,7 +113,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id)if (err <0 ) {pr_debug ("Error during parse sample\n" );perf_sample__exit (&sample); - goto out_delete_evlist; + goto out_put_evlist; } total_periods += sample.period;
@@ -131,10 +131,10 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id) err =-1 ; } -out_delete_evlist: +out_put_evlist:perf_cpu_map__put (cpus);perf_thread_map__put (threads); -evlist__delete (evlist); +evlist__put (evlist);return err; }
tools/perf/tests/switch-tracking.c +1 -1
@@ -582,7 +582,7 @@ static int test__switch_tracking(struct test_suite *test __maybe_unused, int sub out:if (evlist) {evlist__disable (evlist); -evlist__delete (evlist); +evlist__put (evlist); }perf_cpu_map__put (cpus);perf_thread_map__put (threads);
tools/perf/tests/task-exit.c +7 -7
@@ -74,7 +74,7 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _if (!cpus || !threads) { err = -ENOMEM ;pr_debug ("Not enough memory to create thread/cpu maps\n" ); - goto out_delete_evlist; + goto out_put_evlist; }perf_evlist__set_maps (&evlist->core, cpus, threads);
@@ -82,7 +82,7 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _ err =evlist__prepare_workload (evlist, &target, argv, false, workload_exec_failed_signal);if (err <0 ) {pr_debug ("Couldn't run the workload!\n" ); - goto out_delete_evlist; + goto out_put_evlist; } evsel =evlist__first (evlist);
@@ -101,14 +101,14 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _if (err <0 ) {pr_debug ("Couldn't open the evlist: %s\n" ,str_error_r (-err, sbuf,sizeof (sbuf))); - goto out_delete_evlist; + goto out_put_evlist; }if (evlist__mmap (evlist,128 ) <0 ) {pr_debug ("failed to mmap events: %d (%s)\n" , errno,str_error_r (errno, sbuf,sizeof (sbuf))); err =-1 ; - goto out_delete_evlist; + goto out_put_evlist; }evlist__start_workload (evlist);
@@ -133,7 +133,7 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _if (retry_count++ >1000 ) {pr_debug ("Failed after retrying 1000 times\n" ); err =-1 ; - goto out_delete_evlist; + goto out_put_evlist; } goto retry;
@@ -144,10 +144,10 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _ err =-1 ; } -out_delete_evlist: +out_put_evlist:perf_cpu_map__put (cpus);perf_thread_map__put (threads); -evlist__delete (evlist); +evlist__put (evlist);return err; }
tools/perf/tests/tool_pmu.c +1 -1
@@ -67,7 +67,7 @@ static int do_test(enum tool_pmu_event ev, bool with_pmu) out:parse_events_error__exit (&err); -evlist__delete (evlist); +evlist__put (evlist);return ret; }
tools/perf/tests/topology.c +1 -1
@@ -58,7 +58,7 @@ static int session_write_header(char *path) !perf_session__write_header (session, session->evlist, perf_data__fd(&data), true)); -evlist__delete (session->evlist); +evlist__put (session->evlist);perf_session__delete (session);return 0 ;
tools/perf/tests/uncore-event-sorting.c +1 -1
@@ -165,7 +165,7 @@ static int test__uncore_event_sorting(struct test_suite *test __maybe_unused, ret =TEST_OK ; out_err: -evlist__delete (evlist); +evlist__put (evlist);parse_events_error__exit (&err);zfree (&pmu_prefix);zfree (&m.event1);
tools/perf/util/cgroup.c +2 -2
@@ -520,8 +520,8 @@ int evlist__expand_cgroup(struct evlist *evlist, const char *str, bool open_cgro cgrp_event_expanded = true; out_err: -evlist__delete (orig_list); -evlist__delete (tmp_list); +evlist__put (orig_list); +evlist__put (tmp_list);metricgroup__rblist_exit (&orig_metric_events);release_cgroup_list ();
tools/perf/util/data-convert-bt.c +1 -1
@@ -1362,7 +1362,7 @@ static void cleanup_events(struct perf_session *session)zfree (&evsel->priv); } -evlist__delete (evlist); +evlist__put (evlist); session->evlist =NULL ; }
tools/perf/util/evlist.c +15 -5
@@ -75,7 +75,7 @@ int sigqueue(pid_t pid, int sig, const union sigval value);#define FD (e, x, y) (*(int *)xyarray__entry(e->core.fd, x, y))#define SID (e, x, y) xyarray__entry(e->core.sample_id, x, y) -void evlist__init (struct evlist *evlist,struct perf_cpu_map *cpus, +static void evlist__init (struct evlist *evlist,struct perf_cpu_map *cpus,struct perf_thread_map *threads) {perf_evlist__init (&evlist->core);
@@ -88,6 +88,7 @@ void evlist__init(struct evlist *evlist, struct perf_cpu_map *cpus, evlist->nr_br_cntr =-1 ;metricgroup__rblist_init (&evlist->metric_events);INIT_LIST_HEAD (&evlist->deferred_samples); +refcount_set (&evlist->refcnt,1 ); }struct evlist *evlist__new (void )
@@ -139,7 +140,7 @@ struct evlist *evlist__new_default(const struct target *target, bool sample_callreturn evlist; out_err: -evlist__delete (evlist); +evlist__put (evlist);return NULL ; }
@@ -148,13 +149,19 @@ struct evlist *evlist__new_dummy(void)struct evlist *evlist =evlist__new ();if (evlist &&evlist__add_dummy (evlist)) { -evlist__delete (evlist); +evlist__put (evlist); evlist =NULL ; }return evlist; } +struct evlist *evlist__get (struct evlist *evlist) +{ +refcount_inc (&evlist->refcnt); +return evlist; +} +/** * evlist__set_id_pos - set the positions of event ids. * @evlist: selected event list
@@ -193,7 +200,7 @@ static void evlist__purge(struct evlist *evlist) evlist->core.nr_entries =0 ; } -void evlist__exit (struct evlist *evlist) +static void evlist__exit (struct evlist *evlist) {metricgroup__rblist_exit (&evlist->metric_events);event_enable_timer__exit (&evlist->eet);
@@ -202,11 +209,14 @@ void evlist__exit(struct evlist *evlist)perf_evlist__exit (&evlist->core); } -void evlist__delete (struct evlist *evlist) +void evlist__put (struct evlist *evlist) {if (evlist ==NULL )return ; +if (!refcount_dec_and_test (&evlist->refcnt)) +return ; +evlist__free_stats (evlist);evlist__munmap (evlist);evlist__close (evlist);
tools/perf/util/evlist.h +3 -4
@@ -61,6 +61,7 @@ struct event_enable_timer;struct evlist {struct perf_evlist core; +refcount_t refcnt;bool enabled;bool no_affinity;int id_pos;
@@ -109,10 +110,8 @@ struct evsel_str_handler {struct evlist *evlist__new (void );struct evlist *evlist__new_default (const struct target *target,bool sample_callchains);struct evlist *evlist__new_dummy (void ); -void evlist__init (struct evlist *evlist,struct perf_cpu_map *cpus, -struct perf_thread_map *threads); -void evlist__exit (struct evlist *evlist); -void evlist__delete (struct evlist *evlist); +struct evlist *evlist__get (struct evlist *evlist); +void evlist__put (struct evlist *evlist);void evlist__add (struct evlist *evlist,struct evsel *entry);void evlist__remove (struct evlist *evlist,struct evsel *evsel);
tools/perf/util/expr.c +1 -1
@@ -468,7 +468,7 @@ double expr__has_event(const struct expr_parse_ctx *ctx, bool compute_ids, const ret =parse_event (tmp, id) ?0 :1 ; } out: -evlist__delete (tmp); +evlist__put (tmp);return ret; }
tools/perf/util/header.c +11 -11
@@ -5187,7 +5187,7 @@ int perf_session__read_header(struct perf_session *session)pr_err ("Invalid ids section size %" PRIu64" for attr %d, not aligned to u64\n" , f_attr.ids.size, i); err = -EINVAL ; - goto out_delete_evlist; + goto out_put_evlist; }/*
@@ -5200,7 +5200,7 @@ int perf_session__read_header(struct perf_session *session)pr_err ("Invalid ids section size %" PRIu64" for attr %d, too many IDs\n" , f_attr.ids.size, i); err = -EINVAL ; - goto out_delete_evlist; + goto out_put_evlist; }/*
@@ -5213,19 +5213,19 @@ int perf_session__read_header(struct perf_session *session)pr_err ("Invalid ids section for attr %d: offset=%" PRIu64" size=%" PRIu64" exceeds file size %" PRIu64"\n" , i, f_attr.ids.offset, f_attr.ids.size, (u64 )input_stat.st_size); err = -EINVAL ; - goto out_delete_evlist; + goto out_put_evlist; } tmp =lseek (fd,0 ,SEEK_CUR ); evsel =evsel__new (&f_attr.attr);if (evsel ==NULL ) - goto out_delete_evlist; + goto out_put_evlist; evsel->needs_swap = header->needs_swap;/* * Do it before so that if perf_evsel__alloc_id fails, this -* entry gets purged too at evlist__delete(). +* entry gets purged too at evlist__put(). */ evlist__add (session->evlist, evsel);
@@ -5236,7 +5236,7 @@ int perf_session__read_header(struct perf_session *session)* hattr->ids threads. */ if (perf_evsel__alloc_id (&evsel->core,1 , nr_ids)) - goto out_delete_evlist; + goto out_put_evlist;lseek (fd, f_attr.ids.offset,SEEK_SET );
@@ -5266,18 +5266,18 @@ int perf_session__read_header(struct perf_session *session) err =perf_header__process_sections (header, fd, &session->tevent, perf_file_section__process);if (err <0 ) - goto out_delete_evlist; + goto out_put_evlist;if (evlist__prepare_tracepoint_events (session->evlist, session->tevent.pevent)) { err = -ENOMEM ; - goto out_delete_evlist; + goto out_put_evlist; }#else err =perf_header__process_sections (header, fd,NULL , perf_file_section__process);if (err <0 ) - goto out_delete_evlist; + goto out_put_evlist;#endif }
@@ -5303,8 +5303,8 @@ int perf_session__read_header(struct perf_session *session) out_errno:return -errno; -out_delete_evlist: -evlist__delete (session->evlist); +out_put_evlist: +evlist__put (session->evlist); session->evlist =NULL ;return err; }
tools/perf/util/metricgroup.c +3 -3
@@ -214,7 +214,7 @@ static void metric__free(struct metric *m)zfree (&m->metric_refs);expr__ctx_free (m->pctx);zfree (&m->modifier); -evlist__delete (m->evlist); +evlist__put (m->evlist);free (m); }
@@ -1331,7 +1331,7 @@ static int parse_ids(bool metric_no_merge, bool fake_pmu, parsed_evlist =NULL ; err_out:parse_events_error__exit (&parse_error); -evlist__delete (parsed_evlist); +evlist__put (parsed_evlist);strbuf_release (&events);return ret; }
@@ -1542,7 +1542,7 @@ static int parse_groups(struct evlist *perf_evlist,if (combined_evlist) {evlist__splice_list_tail (perf_evlist, &combined_evlist->core.entries); -evlist__delete (combined_evlist); +evlist__put (combined_evlist); }list_for_each_entry (m, &metric_list, nd) {
tools/perf/util/parse-events.c +2 -2
@@ -2343,7 +2343,7 @@ int __parse_events(struct evlist *evlist, const char *str, const char *pmu_filte/* * There are 2 users - builtin-record and builtin-test objects. -* Both call evlist__delete in case of error, so we dont +* Both call evlist__put in case of error, so we dont * need to bother. */ return ret;
@@ -2546,7 +2546,7 @@ int parse_events_option_new_evlist(const struct option *opt, const char *str, in } ret =parse_events_option (opt, str, unset);if (ret) { -evlist__delete (*args->evlistp); +evlist__put (*args->evlistp); *args->evlistp =NULL ; }
tools/perf/util/perf_api_probe.c +1 -1
@@ -57,7 +57,7 @@ static int perf_do_probe_api(setup_probe_fn_t fn, struct perf_cpu cpu, const cha err =0 ; out_delete: -evlist__delete (evlist); +evlist__put (evlist);return err; }
tools/perf/util/python.c +110 -132
451 lines not on this page · raw message
tools/perf/util/record.c +1 -1
@@ -264,7 +264,7 @@ bool evlist__can_select_event(struct evlist *evlist, const char *str) ret = true; out_delete: -evlist__delete (temp_evlist); +evlist__put (temp_evlist);return ret; }
tools/perf/util/session.c +1 -1
@@ -264,7 +264,7 @@ void perf_session__delete(struct perf_session *session)machines__exit (&session->machines);if (session->data) {if (perf_data__is_read (session->data)) -evlist__delete (session->evlist); +evlist__put (session->evlist);perf_data__close (session->data); }#ifdef HAVE_LIBTRACEEVENT
tools/perf/util/sideband_evlist.c +8 -8
@@ -102,7 +102,7 @@ int evlist__start_sb_thread(struct evlist *evlist, struct target *target)return 0 ;if (evlist__create_maps (evlist, target)) - goto out_delete_evlist; + goto out_put_evlist;if (evlist->core.nr_entries >1 ) {bool can_sample_identifier =perf_can_sample_identifier ();
@@ -116,25 +116,25 @@ int evlist__start_sb_thread(struct evlist *evlist, struct target *target)evlist__for_each_entry (evlist, counter) {if (evsel__open (counter, evlist->core.user_requested_cpus, evlist->core.threads) <0 ) - goto out_delete_evlist; + goto out_put_evlist; }if (evlist__mmap (evlist,UINT_MAX )) - goto out_delete_evlist; + goto out_put_evlist;evlist__for_each_entry (evlist, counter) {if (evsel__enable (counter)) - goto out_delete_evlist; + goto out_put_evlist; } evlist->thread.done =0 ;if (pthread_create (&evlist->thread.th,NULL , perf_evlist__poll_thread, evlist)) - goto out_delete_evlist; + goto out_put_evlist;return 0 ; -out_delete_evlist: -evlist__delete (evlist); +out_put_evlist: +evlist__put (evlist); evlist =NULL ;return -1 ; }
@@ -145,5 +145,5 @@ void evlist__stop_sb_thread(struct evlist *evlist)return ; evlist->thread.done =1 ;pthread_join (evlist->thread.th,NULL ); -evlist__delete (evlist); +evlist__put (evlist); }
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
[email protected]
AI reviewRe: [PATCH v10 05/19] perf evlist: Add reference count
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
Pre-existing issues:
- [High] Missing type checks when parsing arguments in pyrf__parse_events and pyrf__parse_metrics lead to memory corruption if arbitrary objects are passed. --
commit cb6079407cc6561ad1c951d274d9edb326c3f418
Author: Ian Rogers <[email protected]>
perf evlist: Add reference count
This commit adds a reference count to the evlist structure. The reference count is initialized to 1 at allocation and helps manage the lifecycle of evlists, particularly avoiding asserts and issues when cloning evlists in Python code.
Quoted 3 lines
tools/perf/util/python.c +0
-0
[ ... ]
Quoted 12 lines
tools/perf/util/python.c +1
-2@@ -2088,14 +2088,17 @@ static PyObject *pyrf__parse_events(PyObject *self, PyObject *args) {const char *input; -struct evlist evlist = {}; +struct evlist *evlist =evlist__new ();struct parse_events_error err;PyObject *result;PyObject *pcpus =NULL , *pthreads =NULL ;struct perf_cpu_map *cpus;struct perf_thread_map *threads; -if (!PyArg_ParseTuple (args,"s|OO" , &input, &pcpus, &pthreads))
tools/perf/util/python.c static PyObject *pyrf__parse_events(PyObject *self, PyObject *args) · line 2070
Quoted 10 lines
tools/perf/util/python.c +6
-0+if (!evlist) +return PyErr_NoMemory (); + +if (!PyArg_ParseTuple (args,"s|OO" , &input, &pcpus, &pthreads)) { +evlist__put (evlist);return NULL ; + } threads = pthreads ? ((struct pyrf_thread_map *)pthreads)->threads :NULL ; cpus = pcpus ? ((struct pyrf_cpu_map *)pcpus)->cpus :NULL ;
[Severity: High]
This is a pre-existing issue, but could this cause memory corruption if arbitrary types are passed from Python? The untyped Python objects (parsed with 'O') are blindly cast to pyrf_thread_map and pyrf_cpu_map structs without validation. Should this use 'O!' with the corresponding type objects to ensure the inputs are actually the expected map types?
[ ... ]
Quoted 12 lines
tools/perf/util/python.c +1
-2@@ -2114,14 +2114,18 @@ static PyObject *pyrf__parse_metrics(PyObject *self, PyObject *args) {const char *input, *pmu =NULL ; -struct evlist evlist = {}; +struct evlist *evlist =evlist__new ();PyObject *result;PyObject *pcpus =NULL , *pthreads =NULL ;struct perf_cpu_map *cpus;struct perf_thread_map *threads;int ret; -if (!PyArg_ParseTuple (args,"s|sOO" , &input, &pmu, &pcpus, &pthreads))
tools/perf/util/python.c static PyObject *pyrf__parse_metrics(PyObject *self, PyObject *args) · line 2104
Quoted 10 lines
tools/perf/util/python.c +6
-0+if (!evlist) +return PyErr_NoMemory (); + +if (!PyArg_ParseTuple (args,"s|sOO" , &input, &pmu, &pcpus, &pthreads)) { +evlist__put (evlist);return NULL ; + } threads = pthreads ? ((struct pyrf_thread_map *)pthreads)->threads :NULL ; cpus = pcpus ? ((struct pyrf_cpu_map *)pcpus)->cpus :NULL ;
[Severity: High]
This is a pre-existing issue, but similarly to pyrf__parse_events() above, does this code blindly cast the untyped Python objects to pyrf_thread_map and pyrf_cpu_map structs? Using 'O!' in PyArg_ParseTuple would allow the Python C API to validate the type before casting, avoiding potential crashes.
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=5
Ian Rogers
reviewerPATCH v10 6/19 perf evsel: Add reference count
In reply to Ian RogersAs with evlist this a no-op for most of the perf tool. The reference count is set to 1 at allocation, the put will see the 1, decrement it and perform the delete. The purpose for adding the reference count is for the python code. Prior to this change the python code would clone evsels, but this has issues if events are opened, etc. leading to assertion failures. With a reference count the same evsel can be used and the reference count incremented for the python usage. To not change the python evsel API getset functions are added for the evsel members, no set function is provided for size as it doesn't make sense to alter this.
---
v2:
1. Fixed Potential Crash in pyrf_event__new : Initialized pevent->evsel = NULL; to avoid garbage pointer dereference if evlist__event2evsel() fails in read_on_cpu .
2. Fixed Memory Leak in pyrf_evsel__init : Added evsel__put(pevsel->evsel) before overwriting it to handle repeated __init__ calls.
3. Fixed Exception Contract: Added PyErr_NoMemory() when evsel__new() fails in pyrf_evsel__init .
4. Fixed NULL Pointer Dereference on Property Access: Added a custom tp_getattro ( pyrf_evsel__getattro ) to pyrf_evsel__type to check if pevsel->evsel is NULL and raise a ValueError if so, covering all property accesses.
5. Fixed Reference Count in pyrf_evlist__add : Added evsel__get(evsel) when adding to the evlist .
6. Fixed Reference Count in pyrf_evlist__read_on_cpu : Added evsel__get(evsel) when assigning to pevent->evsel .
v7:
- Added pyrf_evsel__new to zero-initialize pevsel->evsel to fix crash on re-initialization.
v8:
- Added O! type validation to pyrf_evlist__add to prevent type confusion.
v10:
- Added strict CHECK_INITIALIZED/CHECK_INITIALIZED_INT checks to all pyrf_evsel methods and property getters/setters to prevent NULL pointer dereferences if the object is instantiated without calling __init__.
--- tools/perf/builtin-trace.c | 12 +- tools/perf/tests/evsel-tp-sched.c | 4 +- tools/perf/tests/openat-syscall-all-cpus.c | 6 +- tools/perf/tests/openat-syscall.c | 6 +- tools/perf/util/bpf_counter_cgroup.c | 2 +- tools/perf/util/cgroup.c | 2 +- tools/perf/util/evlist.c | 2 +- tools/perf/util/evsel.c | 26 +- tools/perf/util/evsel.h | 11 +- tools/perf/util/parse-events.y | 2 +- tools/perf/util/pfm.c | 2 +- tools/perf/util/print-events.c | 2 +- tools/perf/util/python.c | 311 ++++++++++++++++++--- tools/perf/util/session.c | 3 + 14 files changed, 310 insertions(+), 81 deletions(-)
tools/perf/builtin-trace.c +6 -6
@@ -460,10 +460,10 @@ static int evsel__init_tp_ptr_field(struct evsel *evsel, struct tp_field *field, ({ struct syscall_tp *sc = __evsel__syscall_tp(evsel);\ evsel__init_tp_ptr_field(evsel, &sc->name, #name); }) -static void evsel__delete_priv(struct evsel *evsel) +static void evsel__put_and_free_priv(struct evsel *evsel) { zfree(&evsel->priv); - evsel__delete(evsel); + evsel__put(evsel); } static int evsel__init_syscall_tp(struct evsel *evsel)
@@ -543,7 +543,7 @@ static struct evsel *perf_evsel__raw_syscall_newtp(const char *direction, void *return evsel; out_delete: -evsel__delete_priv (evsel); +evsel__put_and_free_priv (evsel);return NULL ; }
@@ -3612,7 +3612,7 @@ static bool evlist__add_vfs_getname(struct evlist *evlist)list_del_init (&evsel->core.node); evsel->evlist =NULL ; -evsel__delete (evsel); +evsel__put (evsel); }return found;
@@ -3728,9 +3728,9 @@ static int trace__add_syscall_newtp(struct trace *trace)return ret; out_delete_sys_exit: -evsel__delete_priv (sys_exit); +evsel__put_and_free_priv (sys_exit); out_delete_sys_enter: -evsel__delete_priv (sys_enter); +evsel__put_and_free_priv (sys_enter); goto out; }
tools/perf/tests/evsel-tp-sched.c +2 -2
@@ -64,7 +64,7 @@ static int test__perf_evsel__tp_sched_test(struct test_suite *test __maybe_unuseif (evsel__test_field (evsel,"next_prio" ,4 , true)) ret =TEST_FAIL ; -evsel__delete (evsel); +evsel__put (evsel); evsel =evsel__newtp ("sched" ,"sched_wakeup" );
@@ -85,7 +85,7 @@ static int test__perf_evsel__tp_sched_test(struct test_suite *test __maybe_unuseif (evsel__test_field (evsel,"target_cpu" ,4 , true)) ret =TEST_FAIL ; -evsel__delete (evsel); +evsel__put (evsel);return ret; }
tools/perf/tests/openat-syscall-all-cpus.c +3 -3
@@ -59,7 +59,7 @@ static int test__openat_syscall_event_on_all_cpus(struct test_suite *test __mayb"tweak /proc/sys/kernel/perf_event_paranoid?\n" , str_error_r(errno, sbuf, sizeof(sbuf))); err =TEST_SKIP ; - goto out_evsel_delete; + goto out_evsel_put; }perf_cpu_map__for_each_cpu (cpu, idx, cpus) {
@@ -116,8 +116,8 @@ static int test__openat_syscall_event_on_all_cpus(struct test_suite *test __maybevsel__free_counts (evsel); out_close_fd:perf_evsel__close_fd (&evsel->core); -out_evsel_delete: -evsel__delete (evsel); +out_evsel_put: +evsel__put (evsel); out_cpu_map_delete:perf_cpu_map__put (cpus); out_thread_map_delete:
tools/perf/tests/openat-syscall.c +3 -3
@@ -42,7 +42,7 @@ static int test__openat_syscall_event(struct test_suite *test __maybe_unused,"tweak /proc/sys/kernel/perf_event_paranoid?\n" , str_error_r(errno, sbuf, sizeof(sbuf))); err =TEST_SKIP ; - goto out_evsel_delete; + goto out_evsel_put; }for (i =0 ; i < nr_openat_calls; ++i) {
@@ -64,8 +64,8 @@ static int test__openat_syscall_event(struct test_suite *test __maybe_unused, err =TEST_OK ; out_close_fd:perf_evsel__close_fd (&evsel->core); -out_evsel_delete: -evsel__delete (evsel); +out_evsel_put: +evsel__put (evsel); out_thread_map_delete:perf_thread_map__put (threads);return err;
tools/perf/util/bpf_counter_cgroup.c +1 -1
@@ -336,7 +336,7 @@ static int bperf_cgrp__destroy(struct evsel *evsel)return 0 ;bperf_cgroup_bpf__destroy (skel); -evsel__delete (cgrp_switch);// it'll destroy on_switch progs too +evsel__put (cgrp_switch);// it'll destroy on_switch progs too return 0 ; }
tools/perf/util/cgroup.c +1 -1
@@ -469,7 +469,7 @@ int evlist__expand_cgroup(struct evlist *evlist, const char *str, bool open_cgro/* copy the list and set to the new cgroup. */ evlist__for_each_entry (orig_list, pos) { -struct evsel *evsel =evsel__clone (/*dest=*/ NULL , pos); +struct evsel *evsel =evsel__clone (pos);if (evsel ==NULL ) goto out_err;
tools/perf/util/evlist.c +1 -1
@@ -194,7 +194,7 @@ static void evlist__purge(struct evlist *evlist)evlist__for_each_entry_safe (evlist, n, pos) {list_del_init (&pos->core.node); pos->evlist =NULL ; -evsel__delete (pos); +evsel__put (pos); } evlist->core.nr_entries =0 ;
tools/perf/util/evsel.c +16 -10
@@ -393,10 +393,11 @@ bool evsel__is_function_event(struct evsel *evsel)#undef FUNCTION_EVENT } -void evsel__init (struct evsel *evsel, +static void evsel__init (struct evsel *evsel,struct perf_event_attr *attr,int idx) {perf_evsel__init (&evsel->core, attr, idx); +refcount_set (&evsel->refcnt,1 ); evsel->tracking = !idx; evsel->unit =strdup ("" ); evsel->scale =1.0 ;
@@ -477,7 +478,7 @@ static int evsel__copy_config_terms(struct evsel *dst, struct evsel *src)* The assumption is that @orig is not configured nor opened yet. * So we only care about the attributes that can be set while it's parsed. */ -struct evsel *evsel__clone (struct evsel *dest,struct evsel *orig) +struct evsel *evsel__clone (struct evsel *orig) {struct evsel *evsel;
@@ -490,11 +491,7 @@ struct evsel *evsel__clone(struct evsel *dest, struct evsel *orig)if (orig->bpf_obj)return NULL ; -if (dest) - evsel = dest; -else - evsel =evsel__new (&orig->core.attr); - + evsel =evsel__new (&orig->core.attr);if (evsel ==NULL )return NULL ;
@@ -579,7 +576,7 @@ struct evsel *evsel__clone(struct evsel *dest, struct evsel *orig)return evsel; out_err: -evsel__delete (evsel); +evsel__put (evsel);return NULL ; }
@@ -638,6 +635,12 @@ struct evsel *evsel__newtp_idx(const char *sys, const char *name, int idx, boolreturn ERR_PTR (err); } +struct evsel *evsel__get (struct evsel *evsel) +{ +refcount_inc (&evsel->refcnt); +return evsel; +} +#ifdef HAVE_LIBTRACEEVENT struct tep_event *evsel__tp_format (struct evsel *evsel) {
@@ -1976,7 +1979,7 @@ void evsel__set_priv_destructor(void (*destructor)(void *priv)) evsel__priv_destructor = destructor; } -void evsel__exit (struct evsel *evsel) +static void evsel__exit (struct evsel *evsel) {assert (list_empty (&evsel->core.node));assert (evsel->evlist ==NULL );
@@ -2013,11 +2016,14 @@ void evsel__exit(struct evsel *evsel) } } -void evsel__delete (struct evsel *evsel) +void evsel__put (struct evsel *evsel) {if (!evsel)return ; +if (!refcount_dec_and_test (&evsel->refcnt)) +return ; +evsel__exit (evsel);free (evsel); }
tools/perf/util/evsel.h +6 -5
@@ -6,6 +6,7 @@#include <linux/list.h> #include <linux/perf_event.h> +#include <linux/refcount.h> #include <linux/types.h> #include <sys/types.h>
@@ -47,6 +48,7 @@ typedef int (evsel__sb_cb_t)(union perf_event *event, void *data);struct evsel {struct perf_evsel core;struct evlist *evlist; +refcount_t refcnt;off_t id_offset;int id_pos;int is_pos;
@@ -268,7 +270,7 @@ static inline struct evsel *evsel__new(struct perf_event_attr *attr)return evsel__new_idx (attr,0 ); } -struct evsel *evsel__clone (struct evsel *dest,struct evsel *orig); +struct evsel *evsel__clone (struct evsel *orig);int copy_config_terms (struct list_head *dst,struct list_head *src);void free_config_terms (struct list_head *config_terms);
@@ -283,14 +285,13 @@ static inline struct evsel *evsel__newtp(const char *sys, const char *name)return evsel__newtp_idx (sys, name,0 , true); } +struct evsel *evsel__get (struct evsel *evsel); +void evsel__put (struct evsel *evsel); +#ifdef HAVE_LIBTRACEEVENT struct tep_event *evsel__tp_format (struct evsel *evsel);#endif -void evsel__init (struct evsel *evsel,struct perf_event_attr *attr,int idx); -void evsel__exit (struct evsel *evsel); -void evsel__delete (struct evsel *evsel); -void evsel__set_priv_destructor (void (*destructor)(void *priv));struct callchain_param ;
tools/perf/util/parse-events.y +1 -1
@@ -47,7 +47,7 @@ static void free_list_evsel(struct list_head* list_evsel) list_for_each_entry_safe(evsel, tmp, list_evsel, core.node) { list_del_init(&evsel->core.node); - evsel__delete(evsel); + evsel__put(evsel); } free(list_evsel); }
tools/perf/util/pfm.c +1 -1
@@ -159,7 +159,7 @@ static bool is_libpfm_event_supported(const char *name, struct perf_cpu_map *cpu result = false;evsel__close (evsel); -evsel__delete (evsel); +evsel__put (evsel);return result; }
tools/perf/util/print-events.c +1 -1
@@ -174,7 +174,7 @@ bool is_event_supported(u8 type, u64 config) }evsel__close (evsel); -evsel__delete (evsel); +evsel__put (evsel); }perf_thread_map__put (tmap);
tools/perf/util/python.c +265 -46
444 lines not on this page · raw message
tools/perf/util/session.c +3 -0
@@ -1844,7 +1844,10 @@ static int evlist__deliver_deferred_callchain(struct evlist *evlist,struct evsel *saved_evsel = sample->evsel; sample->evsel =evlist__id2evsel (evlist, sample->id); +if (sample->evsel) + sample->evsel =evsel__get (sample->evsel); ret = tool->callchain_deferred (tool, event, sample, machine); +evsel__put (sample->evsel); sample->evsel = saved_evsel;return ret; }
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
Ian Rogers
reviewerPATCH v10 7/19 perf evlist: Add reference count checking
In reply to Ian RogersNow the evlist is reference counted, add reference count checking so that gets and puts are paired and easy to debug. Reference count checking is documented here:
https://perfwiki.github.io/main/reference-count-checking/
This large patch is adding accessors to evlist functions and switching to their use. There was some minor renaming as evlist__mmap is now an accessor to the mmap variable, and the original evlist__mmap is renamed to evlist__do_mmap.
---
v2:
1. Fixed Memory Leak in evlist__new : Added free(evlist) in the else branch if ADD_RC_CHK fails, preventing a leak of the allocated raw structure.
2. Fixed Potential NULL Dereference: Added a NULL check after from_list_start(_evlist) in perf_evlist__mmap_cb_get() .
3. Fixed Use-After-Free Risk: In evlist__add() , I changed entry->evlist = evlist; to entry->evlist = evlist__get(evlist); . This ensures that the evsel holds a valid reference (wrapper) to the evlist , preventing it from becoming a dangling pointer if the original wrapper is freed.
4. Fixed Test Masking Bug: In test__perf_time__parse_for_ranges() , I replaced TEST_ASSERT_VAL with a manual check and return false; to avoid boolean evaluation of -1 inadvertently passing the test.
5. Fix reference count checker memory leaks from missed puts and due to cyclic evsel to evlist references. A leak still exists in __perf_evlist__propagate_maps due to empty CPU maps and not deleting the removed evsel.
v8:
1. Handle evsel__new failure in sample-parsing.c test and extra ref count checking integrity in evlist.c
2. Use standard per_sample initialization in cs-etm.c.
v10:
- Added CHECK_INITIALIZED checks to pyrf_evlist methods to prevent NULL pointer dereferences on uninitialized objects.
--- tools/perf/arch/arm/util/cs-etm.c | 10 +- tools/perf/arch/arm64/util/arm-spe.c | 8 +- tools/perf/arch/arm64/util/hisi-ptt.c | 2 +- tools/perf/arch/x86/tests/hybrid.c | 20 +- tools/perf/arch/x86/util/auxtrace.c | 2 +- tools/perf/arch/x86/util/intel-bts.c | 6 +- tools/perf/arch/x86/util/intel-pt.c | 9 +- tools/perf/arch/x86/util/iostat.c | 12 +- tools/perf/bench/evlist-open-close.c | 11 +- tools/perf/builtin-annotate.c | 5 +- tools/perf/builtin-ftrace.c | 6 +- tools/perf/builtin-inject.c | 4 +- tools/perf/builtin-kvm.c | 10 +- tools/perf/builtin-kwork.c | 8 +- tools/perf/builtin-record.c | 91 ++--- tools/perf/builtin-report.c | 6 +- tools/perf/builtin-sched.c | 20 +- tools/perf/builtin-script.c | 13 +- tools/perf/builtin-stat.c | 73 ++-- tools/perf/builtin-top.c | 52 +-- tools/perf/builtin-trace.c | 22 +- tools/perf/tests/backward-ring-buffer.c | 8 +- tools/perf/tests/code-reading.c | 10 +- tools/perf/tests/event-times.c | 2 +- tools/perf/tests/event_update.c | 2 +- tools/perf/tests/expand-cgroup.c | 4 +- tools/perf/tests/hwmon_pmu.c | 5 +- tools/perf/tests/keep-tracking.c | 8 +- tools/perf/tests/mmap-basic.c | 6 +- tools/perf/tests/openat-syscall-tp-fields.c | 8 +- tools/perf/tests/parse-events.c | 135 +++---- tools/perf/tests/parse-metric.c | 4 +- tools/perf/tests/perf-record.c | 20 +- tools/perf/tests/perf-time-to-tsc.c | 10 +- tools/perf/tests/pfm.c | 8 +- tools/perf/tests/pmu-events.c | 5 +- tools/perf/tests/sample-parsing.c | 44 +-- tools/perf/tests/sw-clock.c | 6 +- tools/perf/tests/switch-tracking.c | 8 +- tools/perf/tests/task-exit.c | 6 +- tools/perf/tests/time-utils-test.c | 14 +- tools/perf/tests/tool_pmu.c | 5 +- tools/perf/tests/topology.c | 2 +- tools/perf/ui/browsers/annotate.c | 2 +- tools/perf/ui/browsers/hists.c | 22 +- tools/perf/util/amd-sample-raw.c | 2 +- tools/perf/util/annotate-data.c | 2 +- tools/perf/util/annotate.c | 10 +- tools/perf/util/auxtrace.c | 14 +- tools/perf/util/block-info.c | 4 +- tools/perf/util/bpf_counter.c | 2 +- tools/perf/util/bpf_counter_cgroup.c | 8 +- tools/perf/util/bpf_ftrace.c | 9 +- tools/perf/util/bpf_lock_contention.c | 12 +- tools/perf/util/bpf_off_cpu.c | 14 +- tools/perf/util/cgroup.c | 20 +- tools/perf/util/cs-etm.c | 5 +- tools/perf/util/evlist.c | 391 ++++++++++++-------- tools/perf/util/evlist.h | 251 ++++++++++++- tools/perf/util/evsel.c | 6 +- tools/perf/util/evsel.h | 4 +- tools/perf/util/header.c | 39 +- tools/perf/util/header.h | 2 +- tools/perf/util/intel-tpebs.c | 7 +- tools/perf/util/iostat.c | 2 +- tools/perf/util/iostat.h | 2 +- tools/perf/util/metricgroup.c | 6 +- tools/perf/util/parse-events.c | 6 +- tools/perf/util/pfm.c | 2 +- tools/perf/util/python.c | 92 +++-- tools/perf/util/record.c | 9 +- tools/perf/util/sample-raw.c | 4 +- tools/perf/util/session.c | 54 +-- tools/perf/util/sideband_evlist.c | 24 +- tools/perf/util/sort.c | 2 +- tools/perf/util/stat-display.c | 6 +- tools/perf/util/stat-shadow.c | 4 +- tools/perf/util/stat.c | 4 +- tools/perf/util/stream.c | 4 +- tools/perf/util/synthetic-events.c | 11 +- tools/perf/util/time-utils.c | 12 +- tools/perf/util/top.c | 4 +- 82 files changed, 1076 insertions(+), 708 deletions(-)
tools/perf/arch/arm/util/cs-etm.c +5 -5
@@ -201,7 +201,7 @@ static int cs_etm_validate_config(struct perf_pmu *cs_etm_pmu, {unsigned int idx;int err =0 ; -struct perf_cpu_map *event_cpus = evsel->evlist->core.user_requested_cpus; +struct perf_cpu_map *event_cpus =evlist__core (evsel->evlist)->user_requested_cpus;struct perf_cpu_map *intersect_cpus;struct perf_cpu cpu;
@@ -325,7 +325,7 @@ static int cs_etm_recording_options(struct auxtrace_record *itr, container_of(itr, struct cs_etm_recording, itr);struct perf_pmu *cs_etm_pmu = ptr->cs_etm_pmu;struct evsel *evsel, *cs_etm_evsel =NULL ; -struct perf_cpu_map *cpus = evlist->core.user_requested_cpus; +struct perf_cpu_map *cpus =evlist__core (evlist)->user_requested_cpus;bool privileged =perf_event_paranoid_check (-1 );int err =0 ;
@@ -551,7 +551,7 @@ cs_etm_info_priv_size(struct auxtrace_record *itr, {unsigned int idx;int etmv3 =0 , etmv4 =0 , ete =0 ; -struct perf_cpu_map *event_cpus = evlist->core.user_requested_cpus; +struct perf_cpu_map *event_cpus =evlist__core (evlist)->user_requested_cpus;struct perf_cpu_map *intersect_cpus;struct perf_cpu cpu;struct perf_pmu *cs_etm_pmu =cs_etm_get_pmu (itr);
@@ -790,7 +790,7 @@ static int cs_etm_info_fill(struct auxtrace_record *itr,u32 offset;u64 nr_cpu, type;struct perf_cpu_map *cpu_map; -struct perf_cpu_map *event_cpus = session->evlist->core.user_requested_cpus; +struct perf_cpu_map *event_cpus =evlist__core (session->evlist)->user_requested_cpus;struct perf_cpu_map *online_cpus =perf_cpu_map__new_online_cpus ();struct cs_etm_recording *ptr = container_of(itr, struct cs_etm_recording, itr);
@@ -800,7 +800,7 @@ static int cs_etm_info_fill(struct auxtrace_record *itr,if (priv_size !=cs_etm_info_priv_size (itr, session->evlist))return -EINVAL ; -if (!session->evlist->core.nr_mmaps) +if (!evlist__core (session->evlist)->nr_mmaps)return -EINVAL ;/* If the cpu_map has the "any" CPU all online CPUs are involved */
tools/perf/arch/arm64/util/arm-spe.c +4 -4
@@ -60,7 +60,7 @@ static bool arm_spe_is_set_freq(struct evsel *evsel)*/ static struct perf_cpu_map *arm_spe_find_cpus (struct evlist *evlist) { -struct perf_cpu_map *event_cpus = evlist->core.user_requested_cpus; +struct perf_cpu_map *event_cpus =evlist__core (evlist)->user_requested_cpus;struct perf_cpu_map *online_cpus =perf_cpu_map__new_online_cpus ();struct perf_cpu_map *intersect_cpus;
@@ -157,7 +157,7 @@ static int arm_spe_info_fill(struct auxtrace_record *itr,if (priv_size !=arm_spe_info_priv_size (itr, session->evlist))return -EINVAL ; -if (!session->evlist->core.nr_mmaps) +if (!evlist__core (session->evlist)->nr_mmaps)return -EINVAL ; cpu_map =arm_spe_find_cpus (session->evlist);
@@ -363,7 +363,7 @@ static int arm_spe_setup_tracking_event(struct evlist *evlist, {int err;struct evsel *tracking_evsel; -struct perf_cpu_map *cpus = evlist->core.user_requested_cpus; +struct perf_cpu_map *cpus =evlist__core (evlist)->user_requested_cpus;/* Add dummy event to keep tracking */ err =parse_event (evlist,"dummy:u" );
@@ -396,7 +396,7 @@ static int arm_spe_recording_options(struct auxtrace_record *itr,struct arm_spe_recording *sper = container_of(itr, struct arm_spe_recording, itr);struct evsel *evsel, *tmp; -struct perf_cpu_map *cpus = evlist->core.user_requested_cpus; +struct perf_cpu_map *cpus =evlist__core (evlist)->user_requested_cpus;bool discard = false;int err;u64 discard_bit;
tools/perf/arch/arm64/util/hisi-ptt.c +1 -1
@@ -53,7 +53,7 @@ static int hisi_ptt_info_fill(struct auxtrace_record *itr,if (priv_size !=HISI_PTT_AUXTRACE_PRIV_SIZE )return -EINVAL ; -if (!session->evlist->core.nr_mmaps) +if (!evlist__core (session->evlist)->nr_mmaps)return -EINVAL ; auxtrace_info->type =PERF_AUXTRACE_HISI_PTT ;
tools/perf/arch/x86/tests/hybrid.c +10 -10
@@ -26,7 +26,7 @@ static int test__hybrid_hw_event_with_pmu(struct evlist *evlist) {struct evsel *evsel =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,1 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,1 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_HARDWARE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong hybrid type" ,test_hybrid_type (evsel,PERF_TYPE_RAW ));TEST_ASSERT_VAL ("wrong config" ,test_config (evsel,PERF_COUNT_HW_CPU_CYCLES ));
@@ -38,7 +38,7 @@ static int test__hybrid_hw_group_event(struct evlist *evlist)struct evsel *evsel, *leader; evsel = leader =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,2 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,2 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_HARDWARE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong hybrid type" ,test_hybrid_type (evsel,PERF_TYPE_RAW ));TEST_ASSERT_VAL ("wrong config" ,test_config (evsel,PERF_COUNT_HW_CPU_CYCLES ));
@@ -57,7 +57,7 @@ static int test__hybrid_sw_hw_group_event(struct evlist *evlist)struct evsel *evsel, *leader; evsel = leader =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,2 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,2 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_SOFTWARE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong leader" ,evsel__has_leader (evsel, leader));
@@ -74,7 +74,7 @@ static int test__hybrid_hw_sw_group_event(struct evlist *evlist)struct evsel *evsel, *leader; evsel = leader =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,2 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,2 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_HARDWARE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong hybrid type" ,test_hybrid_type (evsel,PERF_TYPE_RAW ));TEST_ASSERT_VAL ("wrong config" ,test_config (evsel,PERF_COUNT_HW_CPU_CYCLES ));
@@ -91,7 +91,7 @@ static int test__hybrid_group_modifier1(struct evlist *evlist)struct evsel *evsel, *leader; evsel = leader =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,2 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,2 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_HARDWARE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong hybrid type" ,test_hybrid_type (evsel,PERF_TYPE_RAW ));TEST_ASSERT_VAL ("wrong config" ,test_config (evsel,PERF_COUNT_HW_CPU_CYCLES ));
@@ -113,7 +113,7 @@ static int test__hybrid_raw1(struct evlist *evlist) {struct perf_evsel *evsel; -perf_evlist__for_each_evsel (&evlist->core, evsel) { +perf_evlist__for_each_evsel (evlist__core (evlist), evsel) {struct perf_pmu *pmu =perf_pmus__find_by_type (evsel->attr.type);TEST_ASSERT_VAL ("missing pmu" , pmu);
@@ -127,7 +127,7 @@ static int test__hybrid_raw2(struct evlist *evlist) {struct evsel *evsel =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,1 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,1 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_RAW == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong config" ,test_config (evsel,0x1a ));return TEST_OK ;
@@ -137,7 +137,7 @@ static int test__hybrid_cache_event(struct evlist *evlist) {struct evsel *evsel =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,1 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,1 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_HW_CACHE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong config" ,0x2 == (evsel->core.attr.config &0xffffffff ));return TEST_OK ;
@@ -148,7 +148,7 @@ static int test__checkevent_pmu(struct evlist *evlist)struct evsel *evsel =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,1 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,1 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_RAW == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong config" ,10 == evsel->core.attr.config);TEST_ASSERT_VAL ("wrong config1" ,1 == evsel->core.attr.config1);
@@ -168,7 +168,7 @@ static int test__hybrid_hw_group_event_2(struct evlist *evlist)struct evsel *evsel, *leader; evsel = leader =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,2 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,2 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_HARDWARE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong hybrid type" ,test_hybrid_type (evsel,PERF_TYPE_RAW ));TEST_ASSERT_VAL ("wrong config" ,test_config (evsel,PERF_COUNT_HW_CPU_CYCLES ));
tools/perf/arch/x86/util/auxtrace.c +1 -1
@@ -55,7 +55,7 @@ struct auxtrace_record *auxtrace_record__init(struct evlist *evlist, int *err) {char buffer[64 ]; -struct perf_cpu cpu =perf_cpu_map__min (evlist->core.all_cpus); +struct perf_cpu cpu =perf_cpu_map__min (evlist__core (evlist)->all_cpus);int ret; *err =0 ;
tools/perf/arch/x86/util/intel-bts.c +3 -3
@@ -79,10 +79,10 @@ static int intel_bts_info_fill(struct auxtrace_record *itr,if (priv_size !=INTEL_BTS_AUXTRACE_PRIV_SIZE )return -EINVAL ; -if (!session->evlist->core.nr_mmaps) +if (!evlist__core (session->evlist)->nr_mmaps)return -EINVAL ; - pc = session->evlist->mmap[0 ].core.base; + pc =evlist__mmap (session->evlist)[0 ].core.base;if (pc) { err =perf_read_tsc_conversion (pc, &tc);if (err) {
@@ -114,7 +114,7 @@ static int intel_bts_recording_options(struct auxtrace_record *itr, container_of(itr, struct intel_bts_recording, itr);struct perf_pmu *intel_bts_pmu = btsr->intel_bts_pmu;struct evsel *evsel, *intel_bts_evsel =NULL ; -const struct perf_cpu_map *cpus = evlist->core.user_requested_cpus; +const struct perf_cpu_map *cpus =evlist__core (evlist)->user_requested_cpus;bool privileged =perf_event_paranoid_check (-1 );if (opts->auxtrace_sample_mode) {
tools/perf/arch/x86/util/intel-pt.c +5 -4
@@ -360,10 +360,10 @@ static int intel_pt_info_fill(struct auxtrace_record *itr, filter =intel_pt_find_filter (session->evlist, ptr->intel_pt_pmu); filter_str_len = filter ?strlen (filter) :0 ; -if (!session->evlist->core.nr_mmaps) +if (!evlist__core (session->evlist)->nr_mmaps)return -EINVAL ; - pc = session->evlist->mmap[0 ].core.base; + pc =evlist__mmap (session->evlist)[0 ].core.base;if (pc) { err =perf_read_tsc_conversion (pc, &tc);if (err) {
@@ -376,7 +376,8 @@ static int intel_pt_info_fill(struct auxtrace_record *itr,ui__warning ("Intel Processor Trace: TSC not available\n" ); } - per_cpu_mmaps = !perf_cpu_map__is_any_cpu_or_is_empty (session->evlist->core.user_requested_cpus); + per_cpu_mmaps = !perf_cpu_map__is_any_cpu_or_is_empty ( +evlist__core (session->evlist)->user_requested_cpus); auxtrace_info->type =PERF_AUXTRACE_INTEL_PT ; auxtrace_info->priv[INTEL_PT_PMU_TYPE ] = intel_pt_pmu->type;
@@ -621,7 +622,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr,struct perf_pmu *intel_pt_pmu = ptr->intel_pt_pmu;bool have_timing_info, need_immediate = false;struct evsel *evsel, *intel_pt_evsel =NULL ; -const struct perf_cpu_map *cpus = evlist->core.user_requested_cpus; +const struct perf_cpu_map *cpus =evlist__core (evlist)->user_requested_cpus;bool privileged =perf_event_paranoid_check (-1 );u64 tsc_bit;int err;
tools/perf/arch/x86/util/iostat.c +7 -5
@@ -332,13 +332,15 @@ static int iostat_event_group(struct evlist *evl,return ret; } -int iostat_prepare (struct evlist *evlist,struct perf_stat_config *config) +int iostat_prepare (struct evlist **evlist_ptr,struct perf_stat_config *config) { -if (evlist->core.nr_entries >0 ) { +struct evlist *evlist = *evlist_ptr; + +if (evlist__nr_entries (evlist) >0 ) {pr_warning ("The -e and -M options are not supported." "All chosen events/metrics will be dropped\n" );evlist__put (evlist); - evlist =evlist__new (); + *evlist_ptr = evlist =evlist__new ();if (!evlist)return -ENOMEM ; }
@@ -400,7 +402,7 @@ void iostat_prefix(struct evlist *evlist,struct perf_stat_config *config, char *prefix, struct timespec *ts) { - struct iio_root_port *rp = evlist->selected->priv; + struct iio_root_port *rp = evlist__selected(evlist)->priv;if (rp) {/*
@@ -463,7 +465,7 @@ void iostat_print_counters(struct evlist *evlist,iostat_prefix (evlist, config, prefix, ts);fprintf (config->output,"%s" , prefix);evlist__for_each_entry (evlist, counter) { - perf_device = evlist->selected->priv; + perf_device =evlist__selected (evlist)->priv;if (perf_device && perf_device != counter->priv) {evlist__set_selected (evlist, counter);iostat_prefix (evlist, config, prefix, ts);
tools/perf/bench/evlist-open-close.c +6 -5
@@ -116,7 +116,7 @@ static int bench__do_evlist_open_close(struct evlist *evlist)return err; } - err =evlist__mmap (evlist, opts.mmap_pages); + err =evlist__do_mmap (evlist, opts.mmap_pages);if (err <0 ) {pr_err ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));return err;
@@ -124,7 +124,7 @@ static int bench__do_evlist_open_close(struct evlist *evlist)evlist__enable (evlist);evlist__disable (evlist); -evlist__munmap (evlist); +evlist__do_munmap (evlist);evlist__close (evlist);return 0 ;
@@ -145,10 +145,11 @@ static int bench_evlist_open_close__run(char *evstr, const char *uid_str)init_stats (&time_stats); -printf (" Number of cpus:\t%d\n" ,perf_cpu_map__nr (evlist->core.user_requested_cpus)); -printf (" Number of threads:\t%d\n" , evlist->core.threads->nr); +printf (" Number of cpus:\t%d\n" , +perf_cpu_map__nr (evlist__core (evlist)->user_requested_cpus)); +printf (" Number of threads:\t%d\n" ,evlist__core (evlist)->threads->nr);printf (" Number of events:\t%d (%d fds)\n" , - evlist->core.nr_entries,evlist__count_evsel_fds (evlist)); +evlist__nr_entries (evlist),evlist__count_evsel_fds (evlist));printf (" Number of iterations:\t%d\n" , iterations);evlist__put (evlist);
tools/perf/builtin-annotate.c +4 -1
@@ -928,8 +928,11 @@ int cmd_annotate(int argc, const char **argv)* branch counters, if the corresponding branch info is available * in the perf data in the TUI mode. */ -if ((use_browser ==1 || annotate.use_stdio2) && annotate.has_br_stack) +if ((use_browser ==1 || annotate.use_stdio2) && annotate.has_br_stack) { sort__mode =SORT_MODE__BRANCH ; +if (evlist__nr_br_cntr (annotate.session->evlist) >0 ) + annotate_opts.show_br_cntr = true; + }if (setup_sorting (/*evlist=*/ NULL ,perf_session__env (annotate.session)) <0 )usage_with_options (annotate_usage, options);
tools/perf/builtin-ftrace.c +3 -3
@@ -377,9 +377,9 @@ static int set_tracing_pid(struct perf_ftrace *ftrace)if (target__has_cpu (&ftrace->target))return 0 ; -for (i =0 ; i <perf_thread_map__nr (ftrace->evlist->core.threads); i++) { +for (i =0 ; i <perf_thread_map__nr (evlist__core (ftrace->evlist)->threads); i++) {scnprintf (buf,sizeof (buf),"%d" , -perf_thread_map__pid (ftrace->evlist->core.threads, i)); +perf_thread_map__pid (evlist__core (ftrace->evlist)->threads, i));if (append_tracing_file ("set_ftrace_pid" , buf) <0 )return -1 ; }
@@ -413,7 +413,7 @@ static int set_tracing_cpumask(struct perf_cpu_map *cpumap)static int set_tracing_cpu (struct perf_ftrace *ftrace) { -struct perf_cpu_map *cpumap = ftrace->evlist->core.user_requested_cpus; +struct perf_cpu_map *cpumap =evlist__core (ftrace->evlist)->user_requested_cpus;if (!target__has_cpu (&ftrace->target))return 0 ;
tools/perf/builtin-inject.c +2 -2
@@ -1496,7 +1496,7 @@ static int synthesize_id_index(struct perf_inject *inject, size_t new_cnt)struct perf_session *session = inject->session;struct evlist *evlist = session->evlist;struct machine *machine = &session->machines.host; -size_t from = evlist->core.nr_entries - new_cnt; +size_t from =evlist__nr_entries (evlist) - new_cnt;return __perf_event__synthesize_id_index (&inject->tool, perf_event__repipe, evlist, machine, from);
@@ -2031,7 +2031,7 @@ static int host__finished_init(const struct perf_tool *tool, struct perf_sessionif (ret)return ret; - ret =synthesize_id_index (inject, gs->session->evlist->core.nr_entries); + ret =synthesize_id_index (inject,evlist__nr_entries (gs->session->evlist));if (ret) {pr_err ("Failed to synthesize id_index\n" );return ret;
tools/perf/builtin-kvm.c +5 -5
@@ -1221,7 +1221,7 @@ static s64 perf_kvm__mmap_read_idx(struct perf_kvm_stat *kvm, int idx,int err; *mmap_time =ULLONG_MAX ; - md = &evlist->mmap[idx]; + md = &evlist__mmap (evlist)[idx]; err =perf_mmap__read_init (&md->core);if (err <0 )return (err == -EAGAIN ) ?0 :-1 ;
@@ -1266,7 +1266,7 @@ static int perf_kvm__mmap_read(struct perf_kvm_stat *kvm)s64 n, ntotal =0 ;u64 flush_time =ULLONG_MAX , mmap_time; -for (i =0 ; i < kvm->evlist->core.nr_mmaps; i++) { +for (i =0 ; i <evlist__core (kvm->evlist)->nr_mmaps; i++) { n =perf_kvm__mmap_read_idx (kvm, i, &mmap_time);if (n <0 )return -1 ;
@@ -1449,7 +1449,7 @@ static int kvm_events_live_report(struct perf_kvm_stat *kvm)evlist__enable (kvm->evlist);while (!done) { -struct fdarray *fda = &kvm->evlist->core.pollfd; +struct fdarray *fda = &evlist__core (kvm->evlist)->pollfd;int rc; rc =perf_kvm__mmap_read (kvm);
@@ -1531,7 +1531,7 @@ static int kvm_live_open_events(struct perf_kvm_stat *kvm) goto out; } -if (evlist__mmap (evlist, kvm->opts.mmap_pages) <0 ) { +if (evlist__do_mmap (evlist, kvm->opts.mmap_pages) <0 ) {ui__error ("Failed to mmap the events: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));evlist__close (evlist);
@@ -1931,7 +1931,7 @@ static int kvm_events_live(struct perf_kvm_stat *kvm,perf_session__set_id_hdr_size (kvm->session);ordered_events__set_copy_on_queue (&kvm->session->ordered_events, true);machine__synthesize_threads (&kvm->session->machines.host, &kvm->opts.target, - kvm->evlist->core.threads, true, false,1 ); +evlist__core (kvm->evlist)->threads, true, false,1 ); err =kvm_live_open_events (kvm);if (err) goto out;
tools/perf/builtin-kwork.c +4 -4
@@ -1814,7 +1814,7 @@ static int perf_kwork__check_config(struct perf_kwork *kwork, } } -list_for_each_entry (evsel, &session->evlist->core.entries, core.node) { +list_for_each_entry (evsel, &evlist__core (session->evlist)->entries, core.node) {if (kwork->show_callchain && !evsel__has_callchain (evsel)) {pr_debug ("Samples do not have callchains\n" ); kwork->show_callchain =0 ;
@@ -1864,9 +1864,9 @@ static int perf_kwork__read_events(struct perf_kwork *kwork) goto out_delete; } - kwork->nr_events = session->evlist->stats.nr_events[0 ]; - kwork->nr_lost_events = session->evlist->stats.total_lost; - kwork->nr_lost_chunks = session->evlist->stats.nr_events[PERF_RECORD_LOST ]; + kwork->nr_events =evlist__stats (session->evlist)->nr_events[0 ]; + kwork->nr_lost_events =evlist__stats (session->evlist)->total_lost; + kwork->nr_lost_chunks =evlist__stats (session->evlist)->nr_events[PERF_RECORD_LOST ]; out_delete:perf_session__delete (session);
tools/perf/builtin-record.c +46 -45
268 lines not on this page · raw message
tools/perf/builtin-report.c +3 -3
@@ -561,7 +561,7 @@ static int evlist__tty_browse_hists(struct evlist *evlist, struct report *rep, cif (!quiet) {fprintf (stdout,"#\n# Total Lost Samples: %" PRIu64"\n#\n" , - evlist->stats.total_lost_samples); +evlist__stats (evlist)->total_lost_samples); }evlist__for_each_entry (evlist, pos) {
@@ -1156,7 +1156,7 @@ static int __cmd_report(struct report *rep) PERF_HPP_REPORT__BLOCK_AVG_CYCLES, }; -if (session->evlist->nr_br_cntr >0 ) +if (evlist__nr_br_cntr (session->evlist) >0 ) block_hpps[nr_hpps++] =PERF_HPP_REPORT__BLOCK_BRANCH_COUNTER ; block_hpps[nr_hpps++] =PERF_HPP_REPORT__BLOCK_RANGE ;
@@ -1291,7 +1291,7 @@ static int process_attr(const struct perf_tool *tool __maybe_unused,* on events sample_type. */ sample_type =evlist__combined_sample_type (*pevlist); - session = (*pevlist)->session; + session =evlist__session (*pevlist);callchain_param_setup (sample_type,perf_session__e_machine (session,/*e_flags=*/ NULL ));return 0 ; }
tools/perf/builtin-sched.c +10 -10
@@ -1972,9 +1972,9 @@ static int perf_sched__read_events(struct perf_sched *sched) goto out_delete; } - sched->nr_events = session->evlist->stats.nr_events[0 ]; - sched->nr_lost_events = session->evlist->stats.total_lost; - sched->nr_lost_chunks = session->evlist->stats.nr_events[PERF_RECORD_LOST ]; + sched->nr_events =evlist__stats (session->evlist)->nr_events[0 ]; + sched->nr_lost_events =evlist__stats (session->evlist)->total_lost; + sched->nr_lost_chunks =evlist__stats (session->evlist)->nr_events[PERF_RECORD_LOST ]; } rc =0 ;
@@ -3227,7 +3227,7 @@ static int timehist_check_attr(struct perf_sched *sched,struct evsel *evsel;struct evsel_runtime *er; -list_for_each_entry (evsel, &evlist->core.entries, core.node) { +list_for_each_entry (evsel, &evlist__core (evlist)->entries, core.node) { er =evsel__get_runtime (evsel);if (er ==NULL ) {pr_err ("Failed to allocate memory for evsel runtime data\n" );
@@ -3399,9 +3399,9 @@ static int perf_sched__timehist(struct perf_sched *sched) goto out; } - sched->nr_events = evlist->stats.nr_events[0 ]; - sched->nr_lost_events = evlist->stats.total_lost; - sched->nr_lost_chunks = evlist->stats.nr_events[PERF_RECORD_LOST ]; + sched->nr_events =evlist__stats (evlist)->nr_events[0 ]; + sched->nr_lost_events =evlist__stats (evlist)->total_lost; + sched->nr_lost_chunks =evlist__stats (evlist)->nr_events[PERF_RECORD_LOST ];if (sched->summary)timehist_print_summary (sched, session);
@@ -3906,7 +3906,7 @@ static int perf_sched__schedstat_record(struct perf_sched *sched,if (err <0 ) goto out; - user_requested_cpus = evlist->core.user_requested_cpus; + user_requested_cpus =evlist__core (evlist)->user_requested_cpus; err = perf_event__synthesize_schedstat(&(sched->tool), process_synthesized_schedstat_event,
@@ -4623,7 +4623,7 @@ static int perf_sched__schedstat_report(struct perf_sched *sched)if (err <0 ) goto out; - user_requested_cpus = session->evlist->core.user_requested_cpus; + user_requested_cpus =evlist__core (session->evlist)->user_requested_cpus; err =perf_session__process_events (session);
@@ -4799,7 +4799,7 @@ static int perf_sched__schedstat_live(struct perf_sched *sched,if (err <0 ) goto out; - user_requested_cpus = evlist->core.user_requested_cpus; + user_requested_cpus =evlist__core (evlist)->user_requested_cpus; err = perf_event__synthesize_schedstat(&(sched->tool), process_synthesized_event_live,
tools/perf/builtin-script.c +7 -6
@@ -2229,9 +2229,10 @@ static int script_find_metrics(const struct pmu_metric *pm,evlist__for_each_entry (metric_evlist, metric_evsel) {struct evsel *script_evsel =map_metric_evsel_to_script_evsel (script_evlist, metric_evsel); -struct metric_event *metric_me =metricgroup__lookup (&metric_evlist->metric_events, - metric_evsel, -/*create=*/ false); +struct metric_event *metric_me = +metricgroup__lookup (evlist__metric_events (metric_evlist), + metric_evsel, +/*create=*/ false);if (script_evsel->metric_id ==NULL ) { script_evsel->metric_id = metric_evsel->metric_id;
@@ -2251,7 +2252,7 @@ static int script_find_metrics(const struct pmu_metric *pm,if (metric_me) {struct metric_expr *expr;struct metric_event *script_me = -metricgroup__lookup (&script_evlist->metric_events, +metricgroup__lookup (evlist__metric_events (script_evlist), script_evsel,/*create=*/ true);
@@ -2321,7 +2322,7 @@ static void perf_sample__fprint_metric(struct thread *thread,assert (stat_config.aggr_mode ==AGGR_GLOBAL ); stat_config.aggr_get_id = script_aggr_cpu_id_get; stat_config.aggr_map = -cpu_aggr_map__new (evsel->evlist->core.user_requested_cpus, +cpu_aggr_map__new (evlist__core (evsel->evlist)->user_requested_cpus, aggr_cpu_id__global,/*data=*/ NULL ,/*needs_sort=*/ false); }
@@ -3909,7 +3910,7 @@ static int set_maps(struct perf_script *script)if (WARN_ONCE (script->allocated,"stats double allocation\n" ))return -EINVAL ; -perf_evlist__set_maps (&evlist->core, script->cpus, script->threads); +perf_evlist__set_maps (evlist__core (evlist), script->cpus, script->threads);if (evlist__alloc_stats (&stat_config, evlist,/*alloc_raw=*/ true))return -ENOMEM ;
tools/perf/builtin-stat.c +39 -34
206 lines not on this page · raw message
tools/perf/builtin-top.c +26 -26
175 lines not on this page · raw message
tools/perf/builtin-trace.c +11 -11
@@ -2023,7 +2023,7 @@ static int trace__symbols_init(struct trace *trace, int argc, const char **argv, goto out; err = __machine__synthesize_threads(trace->host, &trace->tool, &trace->opts.target, - evlist->core.threads, trace__tool_process, + evlist__core(evlist)->threads, trace__tool_process, /*needs_mmap=*/callchain_param.enabled && !trace->summary_only, /*mmap_data=*/false,
@@ -4195,7 +4195,7 @@ static int trace__set_filter_pids(struct trace *trace) err =augmented_syscalls__set_filter_pids (trace->filter_pids.nr, trace->filter_pids.entries); } - }else if (perf_thread_map__pid (trace->evlist->core.threads,0 ) ==-1 ) { + }else if (perf_thread_map__pid (evlist__core (trace->evlist)->threads,0 ) ==-1 ) { err =trace__set_filter_loop_pids (trace); }
@@ -4509,7 +4509,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv)fprintf (trace->output,"Couldn't run the workload!\n" ); goto out_put_evlist; } - workload_pid = evlist->workload.pid; + workload_pid =evlist__workload_pid (evlist); } err =evlist__open (evlist);
@@ -4561,7 +4561,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv) goto out_error_apply_filters;if (!trace->summary_only || !trace->summary_bpf) { - err =evlist__mmap (evlist, trace->opts.mmap_pages); + err =evlist__do_mmap (evlist, trace->opts.mmap_pages);if (err <0 ) goto out_error_mmap; }
@@ -4580,8 +4580,8 @@ static int trace__run(struct trace *trace, int argc, const char **argv)if (trace->summary_bpf)trace_start_bpf_summary (); - trace->multiple_threads =perf_thread_map__pid (evlist->core.threads,0 ) ==-1 || -perf_thread_map__nr (evlist->core.threads) >1 || + trace->multiple_threads =perf_thread_map__pid (evlist__core (evlist)->threads,0 ) ==-1 || +perf_thread_map__nr (evlist__core (evlist)->threads) >1 ||evlist__first (evlist)->core.attr.inherit;/*
@@ -4598,11 +4598,11 @@ static int trace__run(struct trace *trace, int argc, const char **argv) again: before = trace->nr_events; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) {union perf_event *event;struct mmap *md; - md = &evlist->mmap[i]; + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;
@@ -5304,7 +5304,7 @@ static int trace__parse_cgroups(const struct option *opt, const char *str, int u {struct trace *trace = opt->value; -if (!list_empty (&trace->evlist->core.entries)) { +if (!list_empty (&evlist__core (trace->evlist)->entries)) {struct option o = { .value = &trace->evlist, };
@@ -5578,7 +5578,7 @@ int cmd_trace(int argc, const char **argv)* .perfconfig trace.add_events, and filter those out. */ if (!trace.trace_syscalls && !trace.trace_pgfaults && - trace.evlist->core.nr_entries ==0 /* Was --events used? */ ) { +evlist__nr_entries (trace.evlist) ==0 /* Was --events used? */ ) { trace.trace_syscalls = true; }/*
@@ -5664,7 +5664,7 @@ int cmd_trace(int argc, const char **argv) symbol_conf.use_callchain = true; } -if (trace.evlist->core.nr_entries >0 ) { +if (evlist__nr_entries (trace.evlist) >0 ) {bool use_btf = false;evlist__set_default_evsel_handler (trace.evlist, trace__event_handler);
tools/perf/tests/backward-ring-buffer.c +4 -4
@@ -34,8 +34,8 @@ static int count_samples(struct evlist *evlist, int *sample_count, {int i; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { -struct mmap *map = &evlist->overwrite_mmap[i]; +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) { +struct mmap *map = &evlist__overwrite_mmap (evlist)[i];union perf_event *event;perf_mmap__read_init (&map->core);
@@ -65,7 +65,7 @@ static int do_test(struct evlist *evlist, int mmap_pages,int err;char sbuf[STRERR_BUFSIZE ]; - err =evlist__mmap (evlist, mmap_pages); + err =evlist__do_mmap (evlist, mmap_pages);if (err <0 ) {pr_debug ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));
@@ -77,7 +77,7 @@ static int do_test(struct evlist *evlist, int mmap_pages,evlist__disable (evlist); err =count_samples (evlist, sample_count, comm_count); -evlist__munmap (evlist); +evlist__do_munmap (evlist);return err; }
tools/perf/tests/code-reading.c +5 -5
@@ -589,8 +589,8 @@ static int process_events(struct machine *machine, struct evlist *evlist,struct mmap *md;int i, ret; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { - md = &evlist->mmap[i]; +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) { + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;
@@ -778,7 +778,7 @@ static int do_test_code_reading(bool try_kcore) goto out_put; } -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads); str = events[evidx];pr_debug ("Parsing event '%s'\n" , str);
@@ -806,7 +806,7 @@ static int do_test_code_reading(bool try_kcore)pr_debug ("perf_evlist__open() failed!\n%s\n" , errbuf); } -perf_evlist__set_maps (&evlist->core,NULL ,NULL ); +perf_evlist__set_maps (evlist__core (evlist),NULL ,NULL );evlist__put (evlist); evlist =NULL ;continue ;
@@ -817,7 +817,7 @@ static int do_test_code_reading(bool try_kcore)if (events[evidx] ==NULL ) goto out_put; - ret =evlist__mmap (evlist,UINT_MAX ); + ret =evlist__do_mmap (evlist,UINT_MAX );if (ret <0 ) {pr_debug ("evlist__mmap failed\n" ); goto out_put;
tools/perf/tests/event-times.c +1 -1
@@ -50,7 +50,7 @@ static int attach__enable_on_exec(struct evlist *evlist)static int detach__enable_on_exec (struct evlist *evlist) { -waitpid (evlist->workload.pid,NULL ,0 ); +waitpid (evlist__workload_pid (evlist),NULL ,0 );return 0 ; }
tools/perf/tests/event_update.c +1 -1
@@ -92,7 +92,7 @@ static int test__event_update(struct test_suite *test __maybe_unused, int subtesTEST_ASSERT_VAL ("failed to allocate ids" , !perf_evsel__alloc_id (&evsel->core,1 ,1 )); -perf_evlist__id_add (&evlist->core, &evsel->core,0 ,0 ,123 ); +perf_evlist__id_add (evlist__core (evlist), &evsel->core,0 ,0 ,123 );free ((char *)evsel->unit); evsel->unit =strdup ("KRAVA" );
tools/perf/tests/expand-cgroup.c +2 -2
@@ -28,7 +28,7 @@ static int test_expand_events(struct evlist *evlist)TEST_ASSERT_VAL ("evlist is empty" , !evlist__empty (evlist)); - nr_events = evlist->core.nr_entries; + nr_events =evlist__nr_entries (evlist); ev_name =calloc (nr_events,sizeof (*ev_name));if (ev_name ==NULL ) {pr_debug ("memory allocation failure\n" );
@@ -54,7 +54,7 @@ static int test_expand_events(struct evlist *evlist) } ret =TEST_FAIL ; -if (evlist->core.nr_entries != nr_events * nr_cgrps) { +if (evlist__nr_entries (evlist) != nr_events * nr_cgrps) {pr_debug ("event count doesn't match\n" ); goto out; }
tools/perf/tests/hwmon_pmu.c +3 -2
@@ -184,9 +184,10 @@ static int do_test(size_t i, bool with_pmu, bool with_alias) } ret =TEST_OK ; -if (with_pmu ? (evlist->core.nr_entries !=1 ) : (evlist->core.nr_entries <1 )) { +if (with_pmu ? (evlist__nr_entries (evlist) !=1 ) + : (evlist__nr_entries (evlist) <1 )) {pr_debug ("FAILED %s:%d Unexpected number of events for '%s' of %d\n" , - __FILE__, __LINE__, str, evlist->core.nr_entries); + __FILE__, __LINE__, str,evlist__nr_entries (evlist)); ret =TEST_FAIL ; goto out; }
tools/perf/tests/keep-tracking.c +4 -4
@@ -37,8 +37,8 @@ static int find_comm(struct evlist *evlist, const char *comm)int i, found; found =0 ; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { - md = &evlist->mmap[i]; +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) { + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;while ((event =perf_mmap__read_event (&md->core)) !=NULL ) {
@@ -87,7 +87,7 @@ static int test__keep_tracking(struct test_suite *test __maybe_unused, int subte evlist =evlist__new ();CHECK_NOT_NULL__ (evlist); -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads);CHECK__ (parse_event (evlist,"dummy:u" ));CHECK__ (parse_event (evlist,"cpu-cycles:u" ));
@@ -106,7 +106,7 @@ static int test__keep_tracking(struct test_suite *test __maybe_unused, int subte goto out_err; } -CHECK__ (evlist__mmap (evlist,UINT_MAX )); +CHECK__ (evlist__do_mmap (evlist,UINT_MAX ));/* * First, test that a 'comm' event can be found when the event is
tools/perf/tests/mmap-basic.c +3 -3
@@ -81,7 +81,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtest goto out_free_cpus; } -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads);for (i =0 ; i < nsyscalls; ++i) {char name[64 ];
@@ -113,7 +113,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtest expected_nr_events[i] =1 +rand () %127 ; } -if (evlist__mmap (evlist,128 ) <0 ) { +if (evlist__do_mmap (evlist,128 ) <0 ) {pr_debug ("failed to mmap events: %d (%s)\n" , errno,str_error_r (errno, sbuf,sizeof (sbuf))); goto out_put_evlist;
@@ -124,7 +124,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtest syscalls[i](); } - md = &evlist->mmap[0 ]; + md = &evlist__mmap (evlist)[0 ];if (perf_mmap__read_init (&md->core) <0 ) goto out_init;
tools/perf/tests/openat-syscall-tp-fields.c +4 -4
@@ -64,7 +64,7 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unusedevsel__config (evsel, &opts,NULL ); -perf_thread_map__set_pid (evlist->core.threads,0 ,getpid ()); +perf_thread_map__set_pid (evlist__core (evlist)->threads,0 ,getpid ()); err =evlist__open (evlist);if (err <0 ) {
@@ -73,7 +73,7 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unused goto out_put_evlist; } - err =evlist__mmap (evlist,UINT_MAX ); + err =evlist__do_mmap (evlist,UINT_MAX );if (err <0 ) {pr_debug ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));
@@ -90,11 +90,11 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unusedwhile (1 ) {int before = nr_events; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) {union perf_event *event;struct mmap *md; - md = &evlist->mmap[i]; + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;
tools/perf/tests/parse-events.c +70 -65
519 lines not on this page · raw message
tools/perf/tests/parse-metric.c +2 -2
@@ -53,7 +53,7 @@ static double compute_single(struct evlist *evlist, const char *name)struct evsel *evsel;evlist__for_each_entry (evlist, evsel) { - me =metricgroup__lookup (&evlist->metric_events, evsel, false); + me =metricgroup__lookup (evlist__metric_events (evlist), evsel, false);if (me !=NULL ) { list_for_each_entry (mexp, &me->head, nd) {if (strcmp (mexp->metric_name, name))
@@ -88,7 +88,7 @@ static int __compute_metric(const char *name, struct value *vals,return -ENOMEM ; } -perf_evlist__set_maps (&evlist->core, cpus,NULL ); +perf_evlist__set_maps (evlist__core (evlist), cpus,NULL );/* Parse the metric into metric_events list. */ pme_test =find_core_metrics_table ("testarch" ,"testcpu" );
tools/perf/tests/perf-record.c +10 -10
@@ -129,7 +129,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtestevsel__set_sample_bit (evsel,TIME );evlist__config (evlist, &opts,NULL ); - err =sched__get_first_possible_cpu (evlist->workload.pid, cpu_mask); + err =sched__get_first_possible_cpu (evlist__workload_pid (evlist), cpu_mask);if (err <0 ) {pr_debug ("sched__get_first_possible_cpu: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));
@@ -142,7 +142,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtest/* * So that we can check perf_sample.cpu on all the samples. */ -if (sched_setaffinity (evlist->workload.pid, cpu_mask_size, cpu_mask) <0 ) { +if (sched_setaffinity (evlist__workload_pid (evlist), cpu_mask_size, cpu_mask) <0 ) {pr_debug ("sched_setaffinity: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));evlist__cancel_workload (evlist);
@@ -166,7 +166,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtest* fds in the same CPU to be injected in the same mmap ring buffer * (using ioctl(PERF_EVENT_IOC_SET_OUTPUT)). */ - err =evlist__mmap (evlist, opts.mmap_pages); + err =evlist__do_mmap (evlist, opts.mmap_pages);if (err <0 ) {pr_debug ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));
@@ -188,11 +188,11 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtestwhile (1 ) {int before = total_events; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) {union perf_event *event;struct mmap *md; - md = &evlist->mmap[i]; + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;
@@ -231,15 +231,15 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtest ++errs; } -if ((pid_t )sample.pid != evlist->workload.pid) { +if ((pid_t )sample.pid !=evlist__workload_pid (evlist)) {pr_debug ("%s with unexpected pid, expected %d, got %d\n" , - name, evlist->workload.pid, sample.pid); + name,evlist__workload_pid (evlist), sample.pid); ++errs; } -if ((pid_t )sample.tid != evlist->workload.pid) { +if ((pid_t )sample.tid !=evlist__workload_pid (evlist)) {pr_debug ("%s with unexpected tid, expected %d, got %d\n" , - name, evlist->workload.pid, sample.tid); + name,evlist__workload_pid (evlist), sample.tid); ++errs; }
@@ -248,7 +248,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtest type == PERF_RECORD_MMAP2 || type == PERF_RECORD_FORK || type == PERF_RECORD_EXIT) && - (pid_t)event->comm.pid != evlist->workload.pid) { + (pid_t)event->comm.pid != evlist__workload_pid(evlist)) {pr_debug ("%s with unexpected pid/tid\n" , name); ++errs; }
tools/perf/tests/perf-time-to-tsc.c +5 -5
@@ -99,7 +99,7 @@ static int test__perf_time_to_tsc(struct test_suite *test __maybe_unused, int su evlist =evlist__new ();CHECK_NOT_NULL__ (evlist); -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads);CHECK__ (parse_event (evlist,"cpu-cycles:u" ));
@@ -121,9 +121,9 @@ static int test__perf_time_to_tsc(struct test_suite *test __maybe_unused, int su goto out_err; } -CHECK__ (evlist__mmap (evlist,UINT_MAX )); +CHECK__ (evlist__do_mmap (evlist,UINT_MAX )); - pc = evlist->mmap[0 ].core.base; + pc =evlist__mmap (evlist)[0 ].core.base; ret =perf_read_tsc_conversion (pc, &tc);if (ret) {if (ret == -EOPNOTSUPP ) {
@@ -145,8 +145,8 @@ static int test__perf_time_to_tsc(struct test_suite *test __maybe_unused, int suevlist__disable (evlist); -for (i =0 ; i < evlist->core.nr_mmaps; i++) { - md = &evlist->mmap[i]; +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) { + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;
tools/perf/tests/pfm.c +4 -4
@@ -69,12 +69,12 @@ static int test__pfm_events(struct test_suite *test __maybe_unused,if (evlist ==NULL )return -ENOMEM ; - opt.value = evlist; + opt.value = &evlist;parse_libpfm_events_option (&opt, table[i].events,0 );TEST_ASSERT_EQUAL (table[i].events, -count_pfm_events (&evlist->core), +count_pfm_events (evlist__core (evlist)), table[i].nr_events); TEST_ASSERT_EQUAL(table[i].events, evlist__nr_groups(evlist),
@@ -154,12 +154,12 @@ static int test__pfm_group(struct test_suite *test __maybe_unused,if (evlist ==NULL )return -ENOMEM ; - opt.value = evlist; + opt.value = &evlist;parse_libpfm_events_option (&opt, table[i].events,0 );TEST_ASSERT_EQUAL (table[i].events, -count_pfm_events (&evlist->core), +count_pfm_events (evlist__core (evlist)), table[i].nr_events); TEST_ASSERT_EQUAL(table[i].events, evlist__nr_groups(evlist),
tools/perf/tests/pmu-events.c +3 -2
@@ -869,7 +869,7 @@ static int test__parsing_callback(const struct pmu_metric *pm,return -ENOMEM ; } -perf_evlist__set_maps (&evlist->core, cpus,NULL ); +perf_evlist__set_maps (evlist__core (evlist), cpus,NULL ); err =metricgroup__parse_groups_test (evlist, table, pm->metric_name);if (err) {
@@ -895,7 +895,8 @@ static int test__parsing_callback(const struct pmu_metric *pm, k++; }evlist__for_each_entry (evlist, evsel) { -struct metric_event *me =metricgroup__lookup (&evlist->metric_events, evsel, false); +struct metric_event *me =metricgroup__lookup (evlist__metric_events (evlist), + evsel, false);if (me !=NULL ) {struct metric_expr *mexp;
tools/perf/tests/sample-parsing.c +23 -21
111 lines not on this page · raw message
tools/perf/tests/sw-clock.c +3 -3
@@ -71,7 +71,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id) goto out_put_evlist; } -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads);if (evlist__open (evlist)) {const char *knob ="/proc/sys/kernel/perf_event_max_sample_rate" ;
@@ -83,7 +83,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id) goto out_put_evlist; } - err =evlist__mmap (evlist,128 ); + err =evlist__do_mmap (evlist,128 );if (err <0 ) {pr_debug ("failed to mmap event: %d (%s)\n" , errno,str_error_r (errno, sbuf,sizeof (sbuf)));
@@ -98,7 +98,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id)evlist__disable (evlist); - md = &evlist->mmap[0 ]; + md = &evlist__mmap (evlist)[0 ];if (perf_mmap__read_init (&md->core) <0 ) goto out_init;
tools/perf/tests/switch-tracking.c +4 -4
@@ -282,8 +282,8 @@ static int process_events(struct evlist *evlist,struct mmap *md;int i, ret; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { - md = &evlist->mmap[i]; +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) { + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;
@@ -374,7 +374,7 @@ static int test__switch_tracking(struct test_suite *test __maybe_unused, int sub goto out_err; } -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads);/* First event */ err =parse_event (evlist,"cpu-clock:u" );
@@ -471,7 +471,7 @@ static int test__switch_tracking(struct test_suite *test __maybe_unused, int sub goto out; } - err =evlist__mmap (evlist,UINT_MAX ); + err =evlist__do_mmap (evlist,UINT_MAX );if (err) {pr_debug ("evlist__mmap failed!\n" ); goto out_err;
tools/perf/tests/task-exit.c +3 -3
@@ -77,7 +77,7 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _ goto out_put_evlist; } -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads); err =evlist__prepare_workload (evlist, &target, argv, false, workload_exec_failed_signal);if (err <0 ) {
@@ -104,7 +104,7 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _ goto out_put_evlist; } -if (evlist__mmap (evlist,128 ) <0 ) { +if (evlist__do_mmap (evlist,128 ) <0 ) {pr_debug ("failed to mmap events: %d (%s)\n" , errno,str_error_r (errno, sbuf,sizeof (sbuf))); err =-1 ;
@@ -114,7 +114,7 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _evlist__start_workload (evlist); retry: - md = &evlist->mmap[0 ]; + md = &evlist__mmap (evlist)[0 ];if (perf_mmap__read_init (&md->core) <0 ) goto out_init;
tools/perf/tests/time-utils-test.c +9 -5
@@ -69,16 +69,19 @@ struct test_data {static bool test__perf_time__parse_for_ranges (struct test_data *d) { -struct evlist evlist = { - .first_sample_time = d->first, - .last_sample_time = d->last, - }; -struct perf_session session = { .evlist = &evlist }; +struct evlist *evlist =evlist__new (); +struct perf_session session = { .evlist = evlist };struct perf_time_interval *ptime =NULL ;int range_size, range_num;bool pass = false;int i, err; +if (!evlist) { +pr_debug ("Missing evlist\n" ); +return false; + } +evlist__set_first_sample_time (evlist, d->first); +evlist__set_last_sample_time (evlist, d->last);pr_debug ("\nperf_time__parse_for_ranges(\"%s\")\n" , d->str);if (strchr (d->str,'%' ))
@@ -127,6 +130,7 @@ static bool test__perf_time__parse_for_ranges(struct test_data *d) pass = true; out: +evlist__put (evlist);free (ptime);return pass; }
tools/perf/tests/tool_pmu.c +3 -2
@@ -40,9 +40,10 @@ static int do_test(enum tool_pmu_event ev, bool with_pmu) } ret =TEST_OK ; -if (with_pmu ? (evlist->core.nr_entries !=1 ) : (evlist->core.nr_entries <1 )) { +if (with_pmu ? (evlist__nr_entries (evlist) !=1 ) + : (evlist__nr_entries (evlist) <1 )) {pr_debug ("FAILED %s:%d Unexpected number of events for '%s' of %d\n" , - __FILE__, __LINE__, str, evlist->core.nr_entries); + __FILE__, __LINE__, str,evlist__nr_entries (evlist)); ret =TEST_FAIL ; goto out; }
tools/perf/tests/topology.c +1 -1
@@ -46,7 +46,7 @@ static int session_write_header(char *path) session->evlist =evlist__new_default (&target,/*sample_callchains=*/ false);TEST_ASSERT_VAL ("can't get evlist" , session->evlist); - session->evlist->session = session; +evlist__set_session (session->evlist, session);perf_header__set_feat (&session->header,HEADER_CPU_TOPOLOGY );perf_header__set_feat (&session->header,HEADER_NRCPUS );
tools/perf/ui/browsers/annotate.c +1 -1
@@ -597,7 +597,7 @@ static bool annotate_browser__callq(struct annotate_browser *browser, notes =symbol__annotation (dl->ops.target.sym);annotation__lock (notes); -if (!symbol__hists (dl->ops.target.sym, evsel->evlist->core.nr_entries)) { +if (!symbol__hists (dl->ops.target.sym,evlist__nr_entries (evsel->evlist))) {annotation__unlock (notes);ui__warning ("Not enough memory for annotating '%s' symbol!\n" , dl->ops.target.sym->name);
tools/perf/ui/browsers/hists.c +11 -11
@@ -688,10 +688,10 @@ static int hist_browser__handle_hotkey(struct hist_browser *browser, bool warn_lui_browser__update_nr_entries (&browser->b, nr_entries);if (warn_lost_event && - (evsel->evlist->stats.nr_lost_warned != - evsel->evlist->stats.nr_events[PERF_RECORD_LOST ])) { - evsel->evlist->stats.nr_lost_warned = - evsel->evlist->stats.nr_events[PERF_RECORD_LOST ]; + (evlist__stats (evsel->evlist)->nr_lost_warned != +evlist__stats (evsel->evlist)->nr_events[PERF_RECORD_LOST ])) { +evlist__stats (evsel->evlist)->nr_lost_warned = +evlist__stats (evsel->evlist)->nr_events[PERF_RECORD_LOST ];ui_browser__warn_lost_events (&browser->b); }
@@ -3321,7 +3321,7 @@ static int evsel__hists_browse(struct evsel *evsel, int nr_events, const char *h* No need to refresh, resort/decay histogram * entries if we are not collecting samples: */ -if (top->evlist->enabled) { +if (evlist__enabled (top->evlist)) { helpline ="Press 'f' to disable the events or 'h' to see other hotkeys" ; hbt->refresh = delay_secs; }else {
@@ -3493,7 +3493,7 @@ static void perf_evsel_menu__write(struct ui_browser *browser, unit, unit == ' ' ? "" : " ", ev_name);ui_browser__printf (browser,"%s" , bf); - nr_events = evsel->evlist->stats.nr_events[PERF_RECORD_LOST ]; + nr_events =evlist__stats (evsel->evlist)->nr_events[PERF_RECORD_LOST ];if (nr_events !=0 ) { menu->lost_events = true;if (!current_entry)
@@ -3559,13 +3559,13 @@ static int perf_evsel_menu__run(struct evsel_menu *menu,ui_browser__show_title (&menu->b, title);switch (key) {case K_TAB : -if (pos->core.node.next == &evlist->core.entries) +if (pos->core.node.next == &evlist__core (evlist)->entries) pos =evlist__first (evlist);else pos =evsel__next (pos); goto browse_hists;case K_UNTAB : -if (pos->core.node.prev == &evlist->core.entries) +if (pos->core.node.prev == &evlist__core (evlist)->entries) pos =evlist__last (evlist);else pos =evsel__prev (pos);
@@ -3618,7 +3618,7 @@ static int __evlist__tui_browse_hists(struct evlist *evlist, int nr_entries, construct evsel *pos;struct evsel_menu menu = { .b = { - .entries = &evlist->core.entries, + .entries = &evlist__core (evlist)->entries, .refresh = ui_browser__list_head_refresh, .seek = ui_browser__list_head_seek, .write = perf_evsel_menu__write,
@@ -3646,7 +3646,7 @@ static int __evlist__tui_browse_hists(struct evlist *evlist, int nr_entries, constatic bool evlist__single_entry (struct evlist *evlist) { -int nr_entries = evlist->core.nr_entries; +int nr_entries =evlist__nr_entries (evlist);if (nr_entries ==1 )return true;
@@ -3664,7 +3664,7 @@ static bool evlist__single_entry(struct evlist *evlist)int evlist__tui_browse_hists (struct evlist *evlist,const char *help,struct hist_browser_timer *hbt,float min_pcnt,struct perf_env *env,bool warn_lost_event) { -int nr_entries = evlist->core.nr_entries; +int nr_entries =evlist__nr_entries (evlist);if (evlist__single_entry (evlist)) { single_entry: {
tools/perf/util/amd-sample-raw.c +1 -1
@@ -421,7 +421,7 @@ static void parse_cpuid(struct perf_env *env)*/ bool evlist__has_amd_ibs (struct evlist *evlist) { -struct perf_env *env =perf_session__env (evlist->session); +struct perf_env *env =perf_session__env (evlist__session (evlist));int ret, nr_pmu_mappings =perf_env__nr_pmu_mappings (env);const char *pmu_mapping =perf_env__pmu_mappings (env);char name[sizeof ("ibs_fetch" )];
tools/perf/util/annotate-data.c +1 -1
@@ -1829,7 +1829,7 @@ int annotated_data_type__update_samples(struct annotated_data_type *adt,return 0 ;if (adt->histograms ==NULL ) { -int nr = evsel->evlist->core.nr_entries; +int nr =evlist__nr_entries (evsel->evlist);if (alloc_data_type_histograms (adt, nr) <0 )return -1 ;
tools/perf/util/annotate.c +5 -5
@@ -328,7 +328,7 @@ static int symbol__inc_addr_samples(struct map_symbol *ms,if (sym ==NULL )return 0 ; - src =symbol__hists (sym, sample->evsel->evlist->core.nr_entries); + src =symbol__hists (sym,evlist__nr_entries (sample->evsel->evlist));return src ?__symbol__inc_addr_samples (ms, src, addr, sample) :0 ; }
@@ -339,7 +339,7 @@ static int symbol__account_br_cntr(struct annotated_branch *branch, {unsigned int br_cntr_nr =evsel__leader (evsel)->br_cntr_nr;unsigned int base =evsel__leader (evsel)->br_cntr_idx; -unsigned int off = offset * evsel->evlist->nr_br_cntr; +unsigned int off = offset *evlist__nr_br_cntr (evsel->evlist);u64 *branch_br_cntr = branch->br_cntr;unsigned int i, mask, width;
@@ -369,7 +369,7 @@ static int symbol__account_cycles(u64 addr, u64 start, struct symbol *sym,if (sym ==NULL )return 0 ; - branch =symbol__find_branch_hist (sym, evsel->evlist->nr_br_cntr); + branch =symbol__find_branch_hist (sym,evlist__nr_br_cntr (evsel->evlist));if (!branch)return -ENOMEM ;if (addr < sym->start || addr >= sym->end)
@@ -511,7 +511,7 @@ static void annotation__count_and_fill(struct annotation *notes, u64 start, u64static int annotation__compute_ipc (struct annotation *notes,size_t size,struct evsel *evsel) { -unsigned int br_cntr_nr = evsel->evlist->nr_br_cntr; +unsigned int br_cntr_nr =evlist__nr_br_cntr (evsel->evlist);int err =0 ;s64 offset;
@@ -1813,7 +1813,7 @@ int annotation_br_cntr_abbr_list(char **str, struct evsel *evsel, bool header)struct evsel *pos;struct strbuf sb; -if (evsel->evlist->nr_br_cntr <=0 ) +if (evlist__nr_br_cntr (evsel->evlist) <=0 )return -ENOTSUP ;strbuf_init (&sb,/*hint=*/ 0 );
tools/perf/util/auxtrace.c +7 -7
@@ -191,7 +191,7 @@ void auxtrace_mmap_params__set_idx(struct auxtrace_mmap_params *mp,struct evlist *evlist, struct evsel *evsel, int idx) { - bool per_cpu = !perf_cpu_map__has_any_cpu(evlist->core.user_requested_cpus); + bool per_cpu = !perf_cpu_map__has_any_cpu(evlist__core(evlist)->user_requested_cpus); mp->mmap_needed = evsel->needs_auxtrace_mmap;
@@ -201,11 +201,11 @@ void auxtrace_mmap_params__set_idx(struct auxtrace_mmap_params *mp, mp->idx = idx;if (per_cpu) { - mp->cpu =perf_cpu_map__cpu (evlist->core.all_cpus, idx); - mp->tid =perf_thread_map__pid (evlist->core.threads,0 ); + mp->cpu =perf_cpu_map__cpu (evlist__core (evlist)->all_cpus, idx); + mp->tid =perf_thread_map__pid (evlist__core (evlist)->threads,0 ); }else { mp->cpu.cpu =-1 ; - mp->tid =perf_thread_map__pid (evlist->core.threads, idx); + mp->tid =perf_thread_map__pid (evlist__core (evlist)->threads, idx); } }
@@ -667,10 +667,10 @@ int auxtrace_parse_snapshot_options(struct auxtrace_record *itr,static int evlist__enable_event_idx (struct evlist *evlist,struct evsel *evsel,int idx) { -bool per_cpu_mmaps = !perf_cpu_map__has_any_cpu (evlist->core.user_requested_cpus); +bool per_cpu_mmaps = !perf_cpu_map__has_any_cpu (evlist__core (evlist)->user_requested_cpus);if (per_cpu_mmaps) { -struct perf_cpu evlist_cpu =perf_cpu_map__cpu (evlist->core.all_cpus, idx); +struct perf_cpu evlist_cpu =perf_cpu_map__cpu (evlist__core (evlist)->all_cpus, idx);int cpu_map_idx =perf_cpu_map__idx (evsel->core.cpus, evlist_cpu);if (cpu_map_idx ==-1 )
@@ -1837,7 +1837,7 @@ void perf_session__auxtrace_error_inc(struct perf_session *session,struct perf_record_auxtrace_error *e = &event->auxtrace_error;if (e->type <PERF_AUXTRACE_ERROR_MAX ) - session->evlist->stats.nr_auxtrace_errors[e->type] +=1 ; +evlist__stats (session->evlist)->nr_auxtrace_errors[e->type] +=1 ; }void events_stats__auxtrace_error_warn (const struct events_stats *stats)
tools/perf/util/block-info.c +2 -2
@@ -472,7 +472,7 @@ struct block_report *block_info__create_report(struct evlist *evlist, int *nr_reps) {struct block_report *block_reports; -int nr_hists = evlist->core.nr_entries, i =0 ; +int nr_hists =evlist__nr_entries (evlist), i =0 ;struct evsel *pos; block_reports =calloc (nr_hists,sizeof (struct block_report ));
@@ -483,7 +483,7 @@ struct block_report *block_info__create_report(struct evlist *evlist,struct hists *hists =evsel__hists (pos);process_block_report (hists, &block_reports[i], total_cycles, - block_hpps, nr_hpps, evlist->nr_br_cntr); + block_hpps, nr_hpps,evlist__nr_br_cntr (evlist)); i++; }
tools/perf/util/bpf_counter.c +1 -1
@@ -443,7 +443,7 @@ static int bperf_check_target(struct evsel *evsel, }else if (target->tid) { *filter_type =BPERF_FILTER_PID ; *filter_entry_cnt =perf_thread_map__nr (evsel->core.threads); - }else if (target->pid || evsel->evlist->workload.pid !=-1 ) { + }else if (target->pid ||evlist__workload_pid (evsel->evlist) !=-1 ) { *filter_type =BPERF_FILTER_TGID ; *filter_entry_cnt =perf_thread_map__nr (evsel->core.threads); }else {
tools/perf/util/bpf_counter_cgroup.c +4 -4
@@ -116,7 +116,7 @@ static int bperf_load_program(struct evlist *evlist)pr_err ("Failed to open cgroup skeleton\n" );return -1 ; } -setup_rodata (skel, evlist->core.nr_entries); +setup_rodata (skel,evlist__nr_entries (evlist)); err =bperf_cgroup_bpf__load (skel);if (err) {
@@ -127,12 +127,12 @@ static int bperf_load_program(struct evlist *evlist) err =-1 ; cgrp_switch =evsel__new (&cgrp_switch_attr); -if (evsel__open_per_cpu (cgrp_switch, evlist->core.all_cpus,-1 ) <0 ) { +if (evsel__open_per_cpu (cgrp_switch,evlist__core (evlist)->all_cpus,-1 ) <0 ) {pr_err ("Failed to open cgroup switches event\n" ); goto out; } -perf_cpu_map__for_each_cpu (cpu, i, evlist->core.all_cpus) { +perf_cpu_map__for_each_cpu (cpu, i,evlist__core (evlist)->all_cpus) { link =bpf_program__attach_perf_event (skel->progs.on_cgrp_switch,FD (cgrp_switch, i));if (IS_ERR (link)) {
@@ -258,7 +258,7 @@ static int bperf_cgrp__sync_counters(struct evlist *evlist)unsigned int idx;int prog_fd =bpf_program__fd (skel->progs.trigger_read); -perf_cpu_map__for_each_cpu (cpu, idx, evlist->core.all_cpus) +perf_cpu_map__for_each_cpu (cpu, idx,evlist__core (evlist)->all_cpus)bperf_trigger_reading (prog_fd, cpu.cpu);return 0 ;
tools/perf/util/bpf_ftrace.c +5 -4
@@ -59,13 +59,13 @@ int perf_ftrace__latency_prepare_bpf(struct perf_ftrace *ftrace)/* don't need to set cpu filter for system-wide mode */ if (ftrace->target.cpu_list) { - ncpus =perf_cpu_map__nr (ftrace->evlist->core.user_requested_cpus); + ncpus =perf_cpu_map__nr (evlist__core (ftrace->evlist)->user_requested_cpus);bpf_map__set_max_entries (skel->maps.cpu_filter, ncpus); skel->rodata->has_cpu =1 ; }if (target__has_task (&ftrace->target) ||target__none (&ftrace->target)) { - ntasks =perf_thread_map__nr (ftrace->evlist->core.threads); + ntasks =perf_thread_map__nr (evlist__core (ftrace->evlist)->threads);bpf_map__set_max_entries (skel->maps.task_filter, ntasks); skel->rodata->has_task =1 ; }
@@ -87,7 +87,8 @@ int perf_ftrace__latency_prepare_bpf(struct perf_ftrace *ftrace) fd =bpf_map__fd (skel->maps.cpu_filter);for (i =0 ; i < ncpus; i++) { - cpu =perf_cpu_map__cpu (ftrace->evlist->core.user_requested_cpus, i).cpu; + cpu =perf_cpu_map__cpu ( +evlist__core (ftrace->evlist)->user_requested_cpus, i).cpu;bpf_map_update_elem (fd, &cpu, &val,BPF_ANY ); } }
@@ -99,7 +100,7 @@ int perf_ftrace__latency_prepare_bpf(struct perf_ftrace *ftrace) fd =bpf_map__fd (skel->maps.task_filter);for (i =0 ; i < ntasks; i++) { - pid =perf_thread_map__pid (ftrace->evlist->core.threads, i); + pid =perf_thread_map__pid (evlist__core (ftrace->evlist)->threads, i);bpf_map_update_elem (fd, &pid, &val,BPF_ANY ); } }
tools/perf/util/bpf_lock_contention.c +6 -6
@@ -223,11 +223,11 @@ int lock_contention_prepare(struct lock_contention *con)if (target__has_cpu (target)) { skel->rodata->has_cpu =1 ; - ncpus =perf_cpu_map__nr (evlist->core.user_requested_cpus); + ncpus =perf_cpu_map__nr (evlist__core (evlist)->user_requested_cpus); }if (target__has_task (target)) { skel->rodata->has_task =1 ; - ntasks =perf_thread_map__nr (evlist->core.threads); + ntasks =perf_thread_map__nr (evlist__core (evlist)->threads); }if (con->filters->nr_types) { skel->rodata->has_type =1 ;
@@ -334,7 +334,7 @@ int lock_contention_prepare(struct lock_contention *con) fd =bpf_map__fd (skel->maps.cpu_filter);for (i =0 ; i < ncpus; i++) { - cpu =perf_cpu_map__cpu (evlist->core.user_requested_cpus, i).cpu; + cpu =perf_cpu_map__cpu (evlist__core (evlist)->user_requested_cpus, i).cpu;bpf_map_update_elem (fd, &cpu, &val,BPF_ANY ); } }
@@ -346,13 +346,13 @@ int lock_contention_prepare(struct lock_contention *con) fd =bpf_map__fd (skel->maps.task_filter);for (i =0 ; i < ntasks; i++) { - pid =perf_thread_map__pid (evlist->core.threads, i); + pid =perf_thread_map__pid (evlist__core (evlist)->threads, i);bpf_map_update_elem (fd, &pid, &val,BPF_ANY ); } } -if (target__none (target) && evlist->workload.pid >0 ) { -u32 pid = evlist->workload.pid; +if (target__none (target) &&evlist__workload_pid (evlist) >0 ) { +u32 pid =evlist__workload_pid (evlist);u8 val =1 ; fd =bpf_map__fd (skel->maps.task_filter);
tools/perf/util/bpf_off_cpu.c +7 -7
@@ -73,13 +73,13 @@ static void off_cpu_start(void *arg)/* update task filter for the given workload */ if (skel->rodata->has_task && skel->rodata->uses_tgid && -perf_thread_map__pid (evlist->core.threads,0 ) !=-1 ) { +perf_thread_map__pid (evlist__core (evlist)->threads,0 ) !=-1 ) {int fd;u32 pid;u8 val =1 ; fd =bpf_map__fd (skel->maps.task_filter); - pid =perf_thread_map__pid (evlist->core.threads,0 ); + pid =perf_thread_map__pid (evlist__core (evlist)->threads,0 );bpf_map_update_elem (fd, &pid, &val,BPF_ANY ); }
@@ -168,7 +168,7 @@ int off_cpu_prepare(struct evlist *evlist, struct target *target,/* don't need to set cpu filter for system-wide mode */ if (target->cpu_list) { - ncpus =perf_cpu_map__nr (evlist->core.user_requested_cpus); + ncpus =perf_cpu_map__nr (evlist__core (evlist)->user_requested_cpus);bpf_map__set_max_entries (skel->maps.cpu_filter, ncpus); skel->rodata->has_cpu =1 ; }
@@ -199,7 +199,7 @@ int off_cpu_prepare(struct evlist *evlist, struct target *target, skel->rodata->has_task =1 ; skel->rodata->uses_tgid =1 ; }else if (target__has_task (target)) { - ntasks =perf_thread_map__nr (evlist->core.threads); + ntasks =perf_thread_map__nr (evlist__core (evlist)->threads);bpf_map__set_max_entries (skel->maps.task_filter, ntasks); skel->rodata->has_task =1 ; }else if (target__none (target)) {
@@ -209,7 +209,7 @@ int off_cpu_prepare(struct evlist *evlist, struct target *target, }if (evlist__first (evlist)->cgrp) { - ncgrps = evlist->core.nr_entries -1 ;/* excluding a dummy */ + ncgrps =evlist__nr_entries (evlist) -1 ;/* excluding a dummy */ bpf_map__set_max_entries (skel->maps.cgroup_filter, ncgrps);if (!cgroup_is_v2 ("perf_event" ))
@@ -240,7 +240,7 @@ int off_cpu_prepare(struct evlist *evlist, struct target *target, fd =bpf_map__fd (skel->maps.cpu_filter);for (i =0 ; i < ncpus; i++) { - cpu =perf_cpu_map__cpu (evlist->core.user_requested_cpus, i).cpu; + cpu =perf_cpu_map__cpu (evlist__core (evlist)->user_requested_cpus, i).cpu;bpf_map_update_elem (fd, &cpu, &val,BPF_ANY ); } }
@@ -269,7 +269,7 @@ int off_cpu_prepare(struct evlist *evlist, struct target *target, fd =bpf_map__fd (skel->maps.task_filter);for (i =0 ; i < ntasks; i++) { - pid =perf_thread_map__pid (evlist->core.threads, i); + pid =perf_thread_map__pid (evlist__core (evlist)->threads, i);bpf_map_update_elem (fd, &pid, &val,BPF_ANY ); } }
tools/perf/util/cgroup.c +10 -10
@@ -367,7 +367,7 @@ int parse_cgroups(const struct option *opt, const char *str,char *s;int ret, i; -if (list_empty (&evlist->core.entries)) { +if (list_empty (&evlist__core (evlist)->entries)) {fprintf (stderr,"must define events before cgroups\n" );return -1 ; }
@@ -423,7 +423,7 @@ int evlist__expand_cgroup(struct evlist *evlist, const char *str, bool open_cgroint ret =-1 ;int prefix_len; -if (evlist->core.nr_entries ==0 ) { +if (evlist__nr_entries (evlist) ==0 ) {fprintf (stderr,"must define events before cgroups\n" );return -EINVAL ; }
@@ -436,11 +436,11 @@ int evlist__expand_cgroup(struct evlist *evlist, const char *str, bool open_cgro }/* save original events and init evlist */ -evlist__splice_list_tail (orig_list, &evlist->core.entries); - evlist->core.nr_entries =0 ; +evlist__splice_list_tail (orig_list, &evlist__core (evlist)->entries); +evlist__core (evlist)->nr_entries =0 ; - orig_metric_events = evlist->metric_events; -metricgroup__rblist_init (&evlist->metric_events); + orig_metric_events = *evlist__metric_events (evlist); +metricgroup__rblist_init (evlist__metric_events (evlist));if (has_pattern_string (str)) prefix_len =match_cgroups (str);
@@ -503,15 +503,15 @@ int evlist__expand_cgroup(struct evlist *evlist, const char *str, bool open_cgro nr_cgroups++;if (metricgroup__copy_metric_events (tmp_list, cgrp, - &evlist->metric_events, +evlist__metric_events (evlist), &orig_metric_events) <0 ) goto out_err; -evlist__splice_list_tail (evlist, &tmp_list->core.entries); - tmp_list->core.nr_entries =0 ; +evlist__splice_list_tail (evlist, &evlist__core (tmp_list)->entries); +evlist__core (tmp_list)->nr_entries =0 ; } -if (list_empty (&evlist->core.entries)) { +if (list_empty (&evlist__core (evlist)->entries)) {fprintf (stderr,"no cgroup matched: %s\n" , str); goto out_err; }
tools/perf/util/cs-etm.c +4 -1
@@ -1636,8 +1636,9 @@ static int cs_etm__synth_branch_sample(struct cs_etm_queue *etmq, {int ret =0 ;struct cs_etm_auxtrace *etm = etmq->etm; -struct perf_sample sample = {.ip =0 ,}; +struct perf_sample sample;union perf_event *event = tidq->event_buf; +struct dummy_branch_stack {u64 nr;u64 hw_idx;
@@ -1645,6 +1646,7 @@ static int cs_etm__synth_branch_sample(struct cs_etm_queue *etmq, } dummy_bs;u64 ip; +perf_sample__init (&sample,/*all=*/ true); ip =cs_etm__last_executed_instr (tidq->prev_packet); event->sample.header.type =PERF_RECORD_SAMPLE ;
@@ -1698,6 +1700,7 @@ static int cs_etm__synth_branch_sample(struct cs_etm_queue *etmq,"CS ETM Trace: failed to deliver instruction event, error %d\n" , ret); +perf_sample__exit (&sample);return ret; }
tools/perf/util/evlist.c +229 -162
248 lines not on this page · raw message
tools/perf/util/evlist.h +235 -16
251 lines not on this page · raw message
tools/perf/util/python.c +65 -27
92 lines not on this page · raw message
tools/perf/util/session.c +30 -24
54 lines not on this page · raw message
The rest of this diff is not shown. View the raw message.
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
Part of the diff is not shown here. View the raw message.
[email protected]
AI reviewRe: [PATCH v10 07/19] perf evlist: Add reference count checking
Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider: - [High] The cycle-breaking logic in `evlist__put()` suffers from a Time-of-Check to Time-of-Use (TOCTOU)…
Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
- [High] The cycle-breaking logic in evlist__put() suffers from a Time-of-Check to Time-of-Use (TOCTOU) race condition that can result in a double free if multiple threads drop their references concurrently.
- [High] Added evsel__get() calls in util/session.c for deferred sample delivery unconditionally leak evsel references for every processed sample where !sample->evsel is initially true or during deferred callchain delivery.
- [Medium] Raw pointer assignments of the newly reference-counted evlist handle are used instead of acquiring explicit references via evlist__get().
--
commit 58a195e1af66c9cff3bb966d4b7b789d6cee0f42
Author: Ian Rogers <[email protected]>
perf evlist: Add reference count checking
This commit introduces reference count checking for the perf evlist subsystem. It replaces direct struct accesses with new accessor functions and adds lifecycle macros to pair gets and puts.
Quoted 4 lines
tools/perf/arch/arm/util/cs-etm.c +0
-0@@ -319,7 +319,7 @@ static int cs_etm_recording_options(struct auxtrace_record *itr,
[ ... ]
tools/perf/arch/arm/util/cs-etm.c
Quoted 5 lines
tools/perf/arch/arm/util/cs-etm.c +0
-0/* no need to continue if at least one event of interest was found */ if (!cs_etm_evsel)return 0 ; ptr->evlist = evlist;
[Severity: Medium]
Should this raw pointer assignment use evlist__get() instead?
With the transition to reference-counted handles, bypassing the explicit lifecycle helpers leaves the raw pointer vulnerable to becoming a dangling handle if the original reference is dropped.
Quoted 3 lines
tools/perf/arch/arm/util/cs-etm.c +0
-0ptr->snapshot_mode = opts->auxtrace_snapshot_mode; if (!record_opts__no_switch_events(opts) &&
[ ... ]
Quoted 4 lines
tools/perf/builtin-sched.c +0
-0@@ -3825,7 +3825,7 @@ static int perf_sched__schedstat_record(struct perf_sched *sched,
[ ... ]
tools/perf/builtin-sched.c
Quoted 7 lines
tools/perf/builtin-sched.c +0
-0if (IS_ERR (session)) {pr_err ("Perf session creation failed.\n" );evlist__put (evlist);return PTR_ERR (session); } session->evlist = evlist;
[Severity: Medium]
Does this assignment bypass the reference counting API?
Assigning the handle directly without evlist__get() might result in a use-after-free if the main reference drops.
Quoted 3 lines
tools/perf/builtin-sched.c +0
-0sched->session = session; sched->data = &data;
[ ... ]
Quoted 4 lines
tools/perf/builtin-top.c +0
-0@@ -1296,7 +1296,7 @@ static int __cmd_top(struct perf_top *top)
[ ... ]
tools/perf/builtin-top.c
Quoted 6 lines
tools/perf/builtin-top.c +0
-0evlist__uniquify_evsel_names (top->evlist, &stat_config); ret =perf_top__start_counters (top);if (ret)return ret; top->session->evlist = top->evlist;
[Severity: Medium]
Could this be updated to top->session->evlist = evlist__get(top->evlist) to ensure the reference is correctly tracked?
Quoted 3 lines
tools/perf/builtin-top.c +0
-0perf_session__set_id_hdr_size (top->session);/*
[ ... ]
Quoted 4 lines
tools/perf/util/evlist.c +0
-0@@ -200,52 +200,81 @@ static void evlist__purge(struct evlist *evlist)
[ ... ]
Quoted 22 lines
void
evlist__put(struct evlist *evlist)
{tools/perf/util/evlist.c +15
-2+struct evsel *evsel; +unsigned int count, ref_cnt; +if (evlist ==NULL )return ; -if (!refcount_dec_and_test (&evlist->refcnt)) -return ; +if (refcount_dec_and_test (evlist__refcnt (evlist))) + goto out_delete; + +retry: + count =refcount_read (evlist__refcnt (evlist)); + ref_cnt = count; +evlist__for_each_entry (evlist, evsel) { +if (RC_CHK_EQUAL (evsel->evlist, evlist) && count) + count--; + } +if (refcount_read (evlist__refcnt (evlist)) != ref_cnt) + goto retry;
tools/perf/util/evlist.c static void evlist__purge(struct evlist *evlist) · line 253
Quoted 10 lines
tools/perf/util/evlist.c +9
-0+if (count !=0 ) { +/* +* Not the last reference except for back references from +* evsels. +*/ +RC_CHK_PUT (evlist); +return ; + } +out_delete:
[Severity: High]
Can this result in a double free if multiple threads drop their references concurrently?
If two threads execute this path simultaneously after failing refcount_dec_and_test(), they could both read the same count, apply the same decrements, evaluate count == 0 as true, and fall through to out_delete concurrently.
Would it be safer to protect this back-reference resolution with a lock?
Quoted 1 line
evlist__free_stats(evlist);
[ ... ]
Quoted 4 lines
tools/perf/util/session.c +0
-0@@ -1852,7 +1852,7 @@ static int evlist__deliver_deferred_callchain(struct evlist *evlist,
[ ... ]
tools/perf/util/session.c static int evlist__deliver_deferred_callchain(struct evlist *evlist, · line 1881
Quoted 8 lines
tools/perf/util/session.c +2
-0orig_sample.deferred_callchain = false; orig_sample.evsel =evlist__id2evsel (evlist, orig_sample.id); +if (orig_sample.evsel) + orig_sample.evsel =evsel__get (orig_sample.evsel); ret =evlist__deliver_sample (evlist, tool, de->event, &orig_sample, machine);
[Severity: High]
Does this sequence leak the evsel reference during deferred callchain delivery?
We acquire a new reference via evsel__get(), but the lifecycle of the sample is managed by perf_sample__init() and perf_sample__exit().
The perf_sample__exit() function does not release the sample->evsel reference, as it has historically treated the pointer as borrowed.
Without a corresponding evsel__put(), won't every such processed event or deferred sample permanently leak its evsel reference?
Quoted 1 line
tools/perf/util/session.c +0
-0@@ -1901,7 +1903,7 @@ static int session__flush_deferred_samples(struct perf_session *session,
[ ... ]
tools/perf/util/session.c static int session__flush_deferred_samples(struct perf_session *session, · line 1922
Quoted 7 lines
tools/perf/util/session.c +2
-0sample.file_offset = de->file_offset; sample.evsel =evlist__id2evsel (evlist, sample.id); +if (sample.evsel) + sample.evsel =evsel__get (sample.evsel); ret =evlist__deliver_sample (evlist, tool, de->event, &sample, machine);
[Severity: High]
Can this also leak the reference for the same reasons as above?
When sample.evsel is populated using evsel__get(), it seems we are missing a cleanup step after the sample is delivered.
Quoted 1 line
tools/perf/util/session.c +0
-0@@ -1959,21 +1963,23 @@ static int machines__deliver_event(struct machines *machines,
[ ... ]
tools/perf/util/session.c static int machines__deliver_event(struct machines *machines, · line 1972
Quoted 12 lines
tools/perf/util/session.c +5
-3dump_event (evlist, event, file_offset, sample, file_path); -if (!sample->evsel) +if (!sample->evsel) { sample->evsel =evlist__id2evsel (evlist, sample.id); -else +if (sample->evsel) + sample->evsel =evsel__get (sample->evsel); + }else {assert (sample->evsel ==evlist__id2evsel (evlist, sample.id)); - + }
[Severity: High]
Does this similarly leak an evsel reference for every processed sample where sample->evsel is initially missing?
The evsel__get() call holds a reference, but it does not appear to be tracked or released when the sample processing concludes.
Quoted 1 line
machine =
machines__find_for_cpumode(machines, event, sample);
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=7
Ian Rogers
reviewerPATCH v10 8/19 perf python: Use evsel in sample in pyrf_event
In reply to Ian RogersAvoid a duplicated evsel by using the one in sample. Add evsel__get/put to the evsel in perf_sample.
--- tools/perf/util/evsel.c | 2 +- tools/perf/util/python.c | 10 +++------- tools/perf/util/sample.c | 17 ++++++++++++----- 3 files changed, 16 insertions(+), 13 deletions(-)
tools/perf/util/evsel.c +1 -1
@@ -3365,7 +3365,7 @@ int evsel__parse_sample(struct evsel *evsel, union perf_event *event,union u64_swap u;perf_sample__init (data,/*all=*/ true); - data->evsel = evsel; + data->evsel =evsel__get (evsel); data->cpu = data->pid = data->tid =-1 ; data->stream_id = data->id = data->time =-1ULL ; data->period = evsel->core.attr.sample_period;
tools/perf/util/python.c +3 -7
@@ -63,7 +63,6 @@ PyMODINIT_FUNC PyInit_perf(void);struct pyrf_event {PyObject_HEAD - struct evsel *evsel; struct perf_sample sample;union perf_event event; };
@@ -294,7 +293,6 @@ static PyMemberDef pyrf_sample_event__members[] = {static void pyrf_sample_event__delete (struct pyrf_event *pevent) { -evsel__put (pevent->evsel);perf_sample__exit (&pevent->sample);Py_TYPE (pevent)->tp_free ((PyObject *)pevent); }
@@ -316,7 +314,7 @@ static PyObject *pyrf_sample_event__repr(const struct pyrf_event *pevent)#ifdef HAVE_LIBTRACEEVENT static bool is_tracepoint (const struct pyrf_event *pevent) { -return pevent->evsel->core.attr.type ==PERF_TYPE_TRACEPOINT ; +return pevent->sample.evsel->core.attr.type ==PERF_TYPE_TRACEPOINT ; } static PyObject*
@@ -363,7 +361,7 @@ tracepoint_field(const struct pyrf_event *pe, struct tep_format_field *field)static PyObject *get_tracepoint_field (struct pyrf_event *pevent,PyObject *attr_name) { -struct evsel *evsel = pevent->evsel; +struct evsel *evsel = pevent->sample.evsel;struct tep_event *tp_format =evsel__tp_format (evsel);struct tep_format_field *field;
@@ -529,7 +527,7 @@ static PyObject *pyrf_event__new(const union perf_event *event) pevent = PyObject_New(struct pyrf_event, ptype);if (pevent !=NULL ) {memcpy (&pevent->event, event, event->header.size); - pevent->evsel =NULL ; +perf_sample__init (&pevent->sample,/*all=*/ false); }return (PyObject *)pevent; }
@@ -1899,8 +1897,6 @@ static PyObject *pyrf_evlist__read_on_cpu(struct pyrf_evlist *pevlist,return Py_None; } - pevent->evsel =evsel__get (evsel); -perf_mmap__consume (&md->core); err =evsel__parse_sample (evsel, &pevent->event, &pevent->sample);
tools/perf/util/sample.c +12 -5
@@ -1,18 +1,23 @@/* SPDX-License-Identifier: GPL-2.0 */ #include "sample.h" + +#include <stdlib.h> +#include <string.h> + +#include <elf.h> +#include <linux/zalloc.h> + +#include "../../arch/x86/include/asm/insn.h" #include "debug.h" +#include "evsel.h" #include "thread.h" -#include <elf.h> +#ifndef EM_CSKY #define EM_CSKY 252#endif #ifndef EM_LOONGARCH #define EM_LOONGARCH 258#endif -#include <linux/zalloc.h> -#include <stdlib.h> -#include <string.h> -#include "../../arch/x86/include/asm/insn.h" void perf_sample__init (struct perf_sample *sample,bool all) {
@@ -29,6 +34,8 @@ void perf_sample__init(struct perf_sample *sample, bool all)void perf_sample__exit (struct perf_sample *sample) { +evsel__put (sample->evsel); + sample->evsel =NULL ;zfree (&sample->user_regs);zfree (&sample->intr_regs);if (sample->merged_callchain) {
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
[email protected]
AI reviewRe: [PATCH v10 08/19] perf python: Use evsel in sample in pyrf_event
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [Low] The patch introduces a memory leak of the `evsel` reference on error paths in the…
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] The patch introduces a memory leak of the evsel reference on error paths in the perf-time-to-tsc test executable. --
commit 7d5a63d3f3c57b9056699588a949237eae0cad32
Author: Ian Rogers <[email protected]>
perf python: Use evsel in sample in pyrf_event
Avoid a duplicated evsel by using the one in sample. Add evsel__get/put to the evsel in perf_sample.
tools/perf/util/evsel.c int evsel__parse_sample(struct evsel *evsel, union perf_event *event, · line 3368
Quoted 10 lines
tools/perf/util/evsel.c +1
-1@@ -3365,7 +3365,7 @@ int evsel__parse_sample(struct evsel *evsel, union perf_event *event,union u64_swap u;perf_sample__init (data,/*all=*/ true); - data->evsel = evsel; + data->evsel =evsel__get (evsel);
[Severity: Low]
If evsel__parse_sample() encounters an error later in the function (such as returning -EFAULT) after acquiring this reference, does the caller properly clean it up?
For example, looking at test__perf_time_to_tsc() in tools/perf/tests/perf-time-to-tsc.c, the CHECK__ macro evaluates evsel__parse_sample() and jumps to out_err on failure:
tools/perf/tests/perf-time-to-tsc.c:test__perf_time_to_tsc() {
...
CHECK__(evsel__parse_sample(evsel, event, &sample));
...
}Because CHECK__ handles failures by jumping to out_err, it bypasses the perf_sample__exit() cleanup call.
Does this leak the evsel reference?
Quoted 3 lines
tools/perf/util/evsel.c +0
-0data->cpu = data->pid = data->tid =-1 ; data->stream_id = data->id = data->time =-1ULL ; data->period = evsel->core.attr.sample_period;
[ ... ]
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=8