perf bpf_counter: Fix opening of "any"(-1) CPU events
Ian Rogers
reviewerPATCH v3 1/1 perf bpf_counter: Fix opening of "any"(-1) CPU events
The bperf BPF counter code doesn't handle "any"(-1) CPU events, always wanting to aggregate a count against a CPU, which avoids the need for atomics so let's not change that. Force evsels used for BPF counters to require a CPU when not in system-wide mode so that the "any"(-1) value isn't used during map propagation and evsel's CPU map matches that of the PMU.
--- tools/perf/builtin-stat.c | 13 +++++++++++++ tools/perf/util/bpf_counter.c | 7 ++++++- 2 files changed, 19 insertions(+), 1 deletion(-)
tools/perf/builtin-stat.c +13 -0
@@ -2540,6 +2540,7 @@ int cmd_stat(int argc, const char **argv)unsigned int interval, timeout;const char *const stat_subcommands[] = {"record" ,"report" };char errbuf[BUFSIZ ]; +struct evsel *counter;setlocale (LC_ALL ,"" );
@@ -2797,6 +2798,18 @@ int cmd_stat(int argc, const char **argv)evlist__warn_user_requested_cpus (evsel_list, target.cpu_list); +evlist__for_each_entry (evsel_list, counter) { +/* +* Setup BPF counters to require CPUs as any(-1) isn't +* supported. evlist__create_maps below will propagate this +* information to the evsels. Note, evsel__is_bperf isn't yet +* set up, and this change must happen early, so directly use +* the bpf_counter variable and target information. +*/ +if ((counter->bpf_counter || target.use_bpf) && !target__has_cpu (&target)) + counter->core.requires_cpu = true; + } +if (evlist__create_maps (evsel_list, &target) <0 ) {if (target__has_task (&target)) {pr_err ("Problems finding threads of monitor\n" );
tools/perf/util/bpf_counter.c +6 -1
@@ -460,6 +460,7 @@ static int bperf_reload_leader_program(struct evsel *evsel, int attr_map_fd,struct bperf_leader_bpf *skel =bperf_leader_bpf__open ();int link_fd, diff_map_fd, err;struct bpf_link *link =NULL ; +struct perf_thread_map *threads;if (!skel) {pr_err ("Failed to open leader skeleton\n" );
@@ -495,7 +496,11 @@ static int bperf_reload_leader_program(struct evsel *evsel, int attr_map_fd,* following evsel__open_per_cpu call */ evsel->leader_skel = skel; -evsel__open (evsel, evsel->core.cpus, evsel->core.threads); +assert (!perf_cpu_map__has_any_cpu_or_is_empty (evsel->core.cpus)); +/* Always open system wide. */ + threads =thread_map__new_by_tid (-1 ); +evsel__open (evsel, evsel->core.cpus, threads); +perf_thread_map__put (threads); out:bperf_leader_bpf__destroy (skel);
Signature · 1 line
2.51.0.710.ga91ca5db03-goog
Quoted 60 lines · Ian Rogers
The bperf BPF counter code doesn't handle "any"(-1) CPU events, always wanting to aggregate a count against a CPU, which avoids the need for atomics so let's not change that. Force evsels used for BPF counters to require a CPU when not in system-wide mode so that the "any"(-1) value isn't used during map propagation and evsel's CPU map matches that of the PMU.
Fixes b91917c0c6fa ("perf bpf_counter: Fix handling of cpumap fixing hybrid")Signed-off-by Ian Rogers <[email protected]>--- tools/perf/builtin-stat.c | 13 +++++++++++++ tools/perf/util/bpf_counter.c | 7 ++++++- 2 files changed, 19 insertions(+), 1 deletion(-)tools/perf/builtin-stat.c +13
-0@@ -2540,6 +2540,7 @@ int cmd_stat(int argc, const char **argv)unsigned int interval, timeout;const char *const stat_subcommands[] = {"record" ,"report" };char errbuf[BUFSIZ ]; +struct evsel *counter;setlocale (LC_ALL ,"" );@@ -2797,6 +2798,18 @@ int cmd_stat(int argc, const char **argv)evlist__warn_user_requested_cpus (evsel_list, target.cpu_list); +evlist__for_each_entry (evsel_list, counter) { +/* +* Setup BPF counters to require CPUs as any(-1) isn't +* supported. evlist__create_maps below will propagate this +* information to the evsels. Note, evsel__is_bperf isn't yet +* set up, and this change must happen early, so directly use +* the bpf_counter variable and target information. +*/ +if ((counter->bpf_counter || target.use_bpf) && !target__has_cpu (&target)) + counter->core.requires_cpu = true; + } +if (evlist__create_maps (evsel_list, &target) <0 ) {if (target__has_task (&target)) {pr_err ("Problems finding threads of monitor\n" );tools/perf/util/bpf_counter.c +1
-0@@ -460,6 +460,7 @@ static int bperf_reload_leader_program(struct evsel *evsel, int attr_map_fd,struct bperf_leader_bpf *skel =bperf_leader_bpf__open ();int link_fd, diff_map_fd, err;struct bpf_link *link =NULL ; +struct perf_thread_map *threads;if (!skel) {pr_err ("Failed to open leader skeleton\n" );@@ -495,7 +496,11 @@ static int bperf_reload_leader_program(struct evsel *evsel, int attr_map_fd,* following evsel__open_per_cpu call */
tools/perf/util/bpf_counter.c static int bperf_reload_leader_program(struct evsel *evsel, int attr_map_fd, · line 506
Quoted 10 lines
tools/perf/util/bpf_counter.c +5
-1evsel->leader_skel = skel; -evsel__open (evsel, evsel->core.cpus, evsel->core.threads); +assert (!perf_cpu_map__has_any_cpu_or_is_empty (evsel->core.cpus)); +/* Always open system wide. */ + threads =thread_map__new_by_tid (-1 ); +evsel__open (evsel, evsel->core.cpus, threads); +perf_thread_map__put (threads); out:bperf_leader_bpf__destroy (skel);
LGTM.
Tengda
Quoted 9 lines · Ian Rogers
The bperf BPF counter code doesn't handle "any"(-1) CPU events, always wanting to aggregate a count against a CPU, which avoids the need for atomics so let's not change that. Force evsels used for BPF counters to require a CPU when not in system-wide mode so that the "any"(-1) value isn't used during map propagation and evsel's CPU map matches that of the PMU.
Fixes b91917c0c6fa ("perf bpf_counter: Fix handling of cpumap fixing hybrid")Signed-off-by Ian Rogers <[email protected]>
Applied to perf-tools-next, thanks!
Best regards,
Namhyung