Messages in this thread Patch in this message | | | From | Namhyung Kim <> | Subject | [RFC] perf arm-spe: Track task context switch for cpu-mode events | Date | Wed, 15 Sep 2021 17:17:48 -0700 |
| |
When perf report synthesize events from ARM SPE data, it refers to current cpu, pid and tid in the machine. But there's no place to set them in the ARM SPE decoder. I'm seeing all pid/tid is set to -1 and user symbols are not resolved in the output.
# perf record -a -e arm_spe_0/ts_enable=1/ sleep 1
# perf report -q | head 8.77% 8.77% :-1 [kernel.kallsyms] [k] format_decode 7.02% 7.02% :-1 [kernel.kallsyms] [k] seq_printf 7.02% 7.02% :-1 [unknown] [.] 0x0000ffff9f687c34 5.26% 5.26% :-1 [kernel.kallsyms] [k] vsnprintf 3.51% 3.51% :-1 [kernel.kallsyms] [k] string 3.51% 3.51% :-1 [unknown] [.] 0x0000ffff9f66ae20 3.51% 3.51% :-1 [unknown] [.] 0x0000ffff9f670b3c 3.51% 3.51% :-1 [unknown] [.] 0x0000ffff9f67c040 1.75% 1.75% :-1 [kernel.kallsyms] [k] ___cache_free 1.75% 1.75% :-1 [kernel.kallsyms] [k] __count_memcg_events
Like Intel PT, add context switch records to track task info. As ARM SPE support was added later than PERF_RECORD_SWITCH_CPU_WIDE, I think we can safely set the attr.context_switch bit and use it.
Cc: Leo Yan <leo.yan@linaro.org> Cc: James Clark <james.clark@arm.com> Cc: Tan Xiaojun <tanxiaojun@huawei.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org> --- tools/perf/arch/arm64/util/arm-spe.c | 6 +++++- tools/perf/util/arm-spe.c | 24 ++++++++++++++++++++++++ 2 files changed, 29 insertions(+), 1 deletion(-)
diff --git a/tools/perf/arch/arm64/util/arm-spe.c b/tools/perf/arch/arm64/util/arm-spe.c index a4420d4df503..58ba8d15c573 100644 --- a/tools/perf/arch/arm64/util/arm-spe.c +++ b/tools/perf/arch/arm64/util/arm-spe.c @@ -166,8 +166,12 @@ static int arm_spe_recording_options(struct auxtrace_record *itr, tracking_evsel->core.attr.sample_period = 1; /* In per-cpu case, always need the time of mmap events etc */ - if (!perf_cpu_map__empty(cpus)) + if (!perf_cpu_map__empty(cpus)) { evsel__set_sample_bit(tracking_evsel, TIME); + evsel__set_sample_bit(tracking_evsel, CPU); + /* also track task context switch */ + tracking_evsel->core.attr.context_switch = 1; + } return 0; } diff --git a/tools/perf/util/arm-spe.c b/tools/perf/util/arm-spe.c index 58b7069c5a5f..0acac0431c48 100644 --- a/tools/perf/util/arm-spe.c +++ b/tools/perf/util/arm-spe.c @@ -681,6 +681,25 @@ static int arm_spe_process_timeless_queues(struct arm_spe *spe, pid_t tid, return 0; } +static int arm_spe_context_switch(struct arm_spe *spe, union perf_event *event, + struct perf_sample *sample) +{ + pid_t pid, tid; + int cpu; + + if (!(event->header.misc & PERF_RECORD_MISC_SWITCH_OUT)) + return 0; + + pid = event->context_switch.next_prev_pid; + tid = event->context_switch.next_prev_tid; + cpu = sample->cpu; + + if (tid == -1) + pr_warn("context_switch event has no tid\n"); + + return machine__set_current_tid(spe->machine, cpu, pid, tid); +} + static int arm_spe_process_event(struct perf_session *session, union perf_event *event, struct perf_sample *sample, @@ -718,6 +737,11 @@ static int arm_spe_process_event(struct perf_session *session, } } else if (timestamp) { err = arm_spe_process_queues(spe, timestamp); + if (err) + return err; + + if (event->header.type == PERF_RECORD_SWITCH_CPU_WIDE) + err = arm_spe_context_switch(spe, event, sample); } return err; -- 2.33.0.464.g1972c5931b-goog
| |