perf already support ignore_missing_thread for -p, but not yet
applied to `perf stat -p <pid>`. This patch enables ignore_missing_thread
for `perf stat -p <pid>`.
Committer notes:
And here is a refresher about the 'ignore_missing_thread' knob, from a
previous patch using it:
ca8000684e ("perf evsel: Enable ignore_missing_thread for pid option")
---
While monitoring a multithread process with pid option, perf sometimes
may return sys_perf_event_open failure with 3(No such process) if any of
the process's threads die before we open the event. However, we want
perf continue monitoring the remaining threads and do not exit with
error.
---
Signed-off-by: Gang Li <ligang.bdlg@bytedance.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220622030037.15005-1-ligang.bdlg@bytedance.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Uncore events require a CPU i.e. it cannot be -1.
The evsel system_wide flag is intended for events that should be on every
CPU, which does not make sense for uncore events because uncore events do
not map one-to-one with CPUs.
These 2 requirements are not exactly the same, so introduce a new flag
'requires_cpu' for the uncore case.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Ian Rogers <irogers@google.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Leo Yan <leo.yan@linaro.org>
Link: https://lore.kernel.org/r/20220524075436.29144-13-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
If any member in a group has a different cpu mask than the other
members, the current perf stat disables group. when the perf metrics
topdown events are part of the group, the below <not supported> error
will be triggered.
$ perf stat -e "{slots,topdown-retiring,uncore_imc_free_running_0/dclk/}" -a sleep 1
WARNING: grouped events cpus do not match, disabling group:
anon group { slots, topdown-retiring, uncore_imc_free_running_0/dclk/ }
Performance counter stats for 'system wide':
141,465,174 slots
<not supported> topdown-retiring
1,605,330,334 uncore_imc_free_running_0/dclk/
The perf metrics topdown events must always be grouped with a slots
event as leader.
Factor out evsel__remove_from_group() to only remove the regular events
from the group.
Remove evsel__must_be_in_group(), since no one use it anymore.
With the patch, the topdown events aren't broken from the group for the
splitting.
$ perf stat -e "{slots,topdown-retiring,uncore_imc_free_running_0/dclk/}" -a sleep 1
WARNING: grouped events cpus do not match, disabling group:
anon group { slots, topdown-retiring, uncore_imc_free_running_0/dclk/ }
Performance counter stats for 'system wide':
346,110,588 slots
124,608,256 topdown-retiring
1,606,869,976 uncore_imc_free_running_0/dclk/
1.003877592 seconds time elapsed
Fixes: a9a1790247 ("perf stat: Ensure group is defined on top of the same cpu mask")
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Acked-by: Ian Rogers <irogers@google.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Xing Zhengjun <zhengjun.xing@linux.intel.com>
Link: https://lore.kernel.org/r/20220518143900.1493980-3-kan.liang@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Since for cpu_core or cpu_atom, they have different topdown events
groups.
For cpu_core, --topdown equals to:
"{slots,cpu_core/topdown-retiring/,cpu_core/topdown-bad-spec/,
cpu_core/topdown-fe-bound/,cpu_core/topdown-be-bound/,
cpu_core/topdown-heavy-ops/,cpu_core/topdown-br-mispredict/,
cpu_core/topdown-fetch-lat/,cpu_core/topdown-mem-bound/}"
For cpu_atom, --topdown equals to:
"{cpu_atom/topdown-retiring/,cpu_atom/topdown-bad-spec/,
cpu_atom/topdown-fe-bound/,cpu_atom/topdown-be-bound/}"
To simplify the implementation, on hybrid, --topdown is used
together with --cputype. If without --cputype, it uses cpu_core
topdown events by default.
# ./perf stat --topdown -a sleep 1
WARNING: default to use cpu_core topdown events
Performance counter stats for 'system wide':
retiring bad speculation frontend bound backend bound heavy operations light operations branch mispredict machine clears fetch latency fetch bandwidth memory bound Core bound
4.1% 0.0% 5.1% 90.8% 2.3% 1.8% 0.0% 0.0% 4.2% 0.9% 9.9% 81.0%
1.002624229 seconds time elapsed
# ./perf stat --topdown -a --cputype atom sleep 1
Performance counter stats for 'system wide':
retiring bad speculation frontend bound backend bound
13.5% 0.1% 31.2% 55.2%
1.002366987 seconds time elapsed
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Xing Zhengjun <zhengjun.xing@linux.intel.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220422065635.767648-3-zhengjun.xing@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
For hybrid events, by default stat aggregates and reports the event counts
per pmu.
# ./perf stat -e cycles -a sleep 1
Performance counter stats for 'system wide':
14,066,877,268 cpu_core/cycles/
6,814,443,147 cpu_atom/cycles/
1.002760625 seconds time elapsed
Sometimes, it's also useful to aggregate event counts from all PMUs.
Create a new option '--hybrid-merge' to enable that behavior and report
the counts without PMUs.
# ./perf stat -e cycles -a --hybrid-merge sleep 1
Performance counter stats for 'system wide':
20,732,982,512 cycles
1.002776793 seconds time elapsed
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Xing Zhengjun <zhengjun.xing@linux.intel.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220422065635.767648-2-zhengjun.xing@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
It bothered me that during benchmarking using 'perf stat' (to collect
for example CPU cache events) I could not simultaneously retrieve the
times spend in user or kernel mode in a machine readable format.
When running 'perf stat' the output for humans contains the times
reported by rusage and wait4.
$ perf stat -e cache-misses:u -- true
Performance counter stats for 'true':
4,206 cache-misses:u
0.001113619 seconds time elapsed
0.001175000 seconds user
0.000000000 seconds sys
But 'perf stat's machine-readable format does not provide this information.
$ perf stat -x, -e cache-misses:u -- true
4282,,cache-misses:u,492859,100.00,,
I found no way to retrieve this information using the available events
while using machine-readable output.
This patch adds two new tool internal events 'user_time' and
'system_time', similarly to the already present 'duration_time' event.
Both events use the already collected rusage information obtained by
wait4 and tracked in the global ru_stats.
Examples presenting cache-misses and rusage information in both human
and machine-readable form:
$ perf stat -e duration_time,user_time,system_time,cache-misses -- grep -q -r duration_time .
Performance counter stats for 'grep -q -r duration_time .':
67,422,542 ns duration_time:u
50,517,000 ns user_time:u
16,839,000 ns system_time:u
30,937 cache-misses:u
0.067422542 seconds time elapsed
0.050517000 seconds user
0.016839000 seconds sys
$ perf stat -x, -e duration_time,user_time,system_time,cache-misses -- grep -q -r duration_time .
72134524,ns,duration_time:u,72134524,100.00,,
65225000,ns,user_time:u,65225000,100.00,,
6865000,ns,system_time:u,6865000,100.00,,
38705,,cache-misses:u,71189328,100.00,,
Signed-off-by: Florian Fischer <florian.fischer@muhq.space>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Xing Zhengjun <zhengjun.xing@linux.intel.com>
Link: https://lore.kernel.org/r/20220420102354.468173-3-florian.fischer@muhq.space
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
This is preparation for exporting rusage values as tool events.
Add new global stats tracking the values obtained via rusage.
For now only ru_utime and ru_stime are part of the tracked stats.
Both are stored as nanoseconds to be consistent with 'duration_time',
although the finest resolution the struct timeval data in rusage
provides are microseconds.
Signed-off-by: Florian Fischer <florian.fischer@muhq.space>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Xing Zhengjun <zhengjun.xing@linux.intel.com>
Link: https://lore.kernel.org/r/20220420102354.468173-2-florian.fischer@muhq.space
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
evlist contains cpus and all_cpus. all_cpus is the union of the cpu maps
of all evsels.
For non-task targets, cpus is set to be cpus requested from the command
line, defaulting to all online cpus if no cpus are specified.
For an uncore event, all_cpus may be just CPU 0 or every online CPU.
This causes all_cpus to have fewer values than the cpus variable which
is confusing given the 'all' in the name.
To try to make the behavior clearer, rename cpus to user_requested_cpus
and add comments on the two struct variables.
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Antonov <alexander.antonov@linux.intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: German Gomez <german.gomez@arm.com>
Cc: James Clark <james.clark@arm.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Fastabend <john.fastabend@gmail.com>
Cc: John Garry <john.garry@huawei.com>
Cc: KP Singh <kpsingh@kernel.org>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Martin KaFai Lau <kafai@fb.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Mike Leach <mike.leach@linaro.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Riccardo Mancini <rickyman7@gmail.com>
Cc: Song Liu <songliubraving@fb.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Suzuki Poulouse <suzuki.poulose@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: Yonghong Song <yhs@fb.com>
Cc: bpf@vger.kernel.org
Cc: coresight@lists.linaro.org
Cc: linux-arm-kernel@lists.infradead.org
Cc: netdev@vger.kernel.org
Link: http://lore.kernel.org/lkml/20220328232648.2127340-3-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Passing NULL to perf_cpu_map__max doesn't make sense as there is no
valid max. Avoid this problem by null checking in
perf_stat_init_aggr_mode.
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Antonov <alexander.antonov@linux.intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: German Gomez <german.gomez@arm.com>
Cc: James Clark <james.clark@arm.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Fastabend <john.fastabend@gmail.com>
Cc: John Garry <john.garry@huawei.com>
Cc: KP Singh <kpsingh@kernel.org>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Martin KaFai Lau <kafai@fb.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Mike Leach <mike.leach@linaro.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Riccardo Mancini <rickyman7@gmail.com>
Cc: Song Liu <songliubraving@fb.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Suzuki Poulouse <suzuki.poulose@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: Yonghong Song <yhs@fb.com>
Cc: bpf@vger.kernel.org
Cc: coresight@lists.linaro.org
Cc: linux-arm-kernel@lists.infradead.org
Cc: netdev@vger.kernel.org
Link: http://lore.kernel.org/lkml/20220328062414.1893550-5-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
We support short command 'rec*' for 'record' and 'rep*' for 'report' in
lots of sub-commands, but the matching is not quite strict currnetly.
It may be puzzling sometime, like we mis-type a 'recport' to report but
it will perform 'record' in fact without any message.
To fix this, add a check to ensure that the short cmd is valid prefix
of the real command.
Committer testing:
[root@quaco ~]# perf c2c re sleep 1
Usage: perf c2c {record|report}
-v, --verbose be more verbose (show counter open errors, etc)
# perf c2c rec sleep 1
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.038 MB perf.data (16 samples) ]
# perf c2c recport sleep 1
Usage: perf c2c {record|report}
-v, --verbose be more verbose (show counter open errors, etc)
# perf c2c record sleep 1
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.038 MB perf.data (15 samples) ]
# perf c2c records sleep 1
Usage: perf c2c {record|report}
-v, --verbose be more verbose (show counter open errors, etc)
#
Signed-off-by: Wei Li <liwei391@huawei.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Hanjun Guo <guohanjun@huawei.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Rui Xiang <rui.xiang@huawei.com>
Link: http://lore.kernel.org/lkml/20220325092032.2956161-1-liwei391@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
I have run into the following issue:
# perf stat -a -e new_pmu/INSTRUCTION_7/ -- mytest -c1 7
Performance counter stats for 'system wide':
0 new_pmu/INSTRUCTION_7/
0.000366428 seconds time elapsed
#
The new PMU for s390 counts the execution of certain CPU instructions.
The root cause is the extremely small run time of the mytest program. It
just executes some assembly instructions and then exits.
In above invocation the instruction is executed exactly one time (-c1
option). The PMU is expected to report this one time execution by a
counter value of one, but fails to do so in some cases, not all.
Debugging reveals the invocation of the child process is done
*before* the counter events are installed and enabled.
Tracing reveals that sometimes the child process starts and exits before
the event is installed on all CPUs. The more CPUs the machine has, the
more often this miscount happens.
Fix this by reversing the start of the work load after the events have
been installed on the specified CPUs. Now the comment also matches the
code.
Output after:
# perf stat -a -e new_pmu/INSTRUCTION_7/ -- mytest -c1 7
Performance counter stats for 'system wide':
1 new_pmu/INSTRUCTION_7/
0.000366428 seconds time elapsed
#
Now the correct result is reported rock solid all the time regardless
how many CPUs are online.
Reviewers notes:
Jiri:
Right, without -a the event has enable_on_exec so the race does not
matter, but it's a problem for system wide with fork.
Namhyung:
Agreed. Also we may move the enable_counters() and the clock code out of
the if block to be shared with the else block.
Fixes: acf2892270 ("perf stat: Use perf_evlist__prepare/start_workload()")
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Cc: Heiko Carstens <hca@linux.ibm.com>
Cc: Sven Schnelle <svens@linux.ibm.com>
Cc: Vasily Gorbik <gor@linux.ibm.com>
Link: https://lore.kernel.org/r/20220317155346.577384-1-tmricht@linux.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Switch from directly accessing the perf_cpu_map to using the appropriate
libperf API when possible. Using the API simplifies the job of
refactoring use of perf_cpu_map.
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: André Almeida <andrealmeid@collabora.com>
Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Darren Hart <dvhart@infradead.org>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Dmitriy Vyukov <dvyukov@google.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: German Gomez <german.gomez@arm.com>
Cc: James Clark <james.clark@arm.com>
Cc: Jin Yao <yao.jin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Madhavan Srinivasan <maddy@linux.ibm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Miaoqian Lin <linmq006@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Riccardo Mancini <rickyman7@gmail.com>
Cc: Shunsuke Nakamura <nakamura.shun@fujitsu.com>
Cc: Song Liu <song@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Stephen Brennan <stephen.s.brennan@oracle.com>
Cc: Steven Rostedt (VMware) <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Thomas Richter <tmricht@linux.ibm.com>
Cc: Yury Norov <yury.norov@gmail.com>
Link: http://lore.kernel.org/lkml/20220122045811.3402706-3-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
I.e. the simple:
$ perf stat sleep 1
Uses a dummy CPU map and thus there is no need to setup/cleanup
affinities to avoid IPIs, etc.
With this we're down to a sched_getaffinity() call, in the libnuma
initialization, that probably can be removed in a followup patch.
Acked-by: Ian Rogers <irogers@google.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20220117160931.1191712-3-acme@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
A common problem is confusing CPU map indices with the CPU, by wrapping
the CPU with a struct then this is avoided. This approach is similar to
atomic_t.
Committer notes:
To make it build with BUILD_BPF_SKEL=1 these files needed the
conversions to 'struct perf_cpu' usage:
tools/perf/util/bpf_counter.c
tools/perf/util/bpf_counter_cgroup.c
tools/perf/util/bpf_ftrace.c
Also perf_env__get_cpu() was removed back in "perf cpumap: Switch
cpu_map__build_map to cpu function".
Additionally these needed to be fixed for the ARM builds to complete:
tools/perf/arch/arm/util/cs-etm.c
tools/perf/arch/arm64/util/pmu.c
Suggested-by: John Garry <john.garry@huawei.com>
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Clark <james.clark@arm.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Mike Leach <mike.leach@linaro.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Clarke <pc@us.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Riccardo Mancini <rickyman7@gmail.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Suzuki Poulouse <suzuki.poulose@arm.com>
Cc: Vineet Singh <vineet.singh@intel.com>
Cc: coresight@lists.linaro.org
Cc: linux-arm-kernel@lists.infradead.org
Cc: zhengjun.xing@intel.com
Link: https://lore.kernel.org/r/20220105061351.120843-49-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
evsel__write_stat_event() was incorrectly passing a cpu map index rather
than a CPU to perf_event__synthesize_stat().
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Clark <james.clark@arm.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Mike Leach <mike.leach@linaro.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Clarke <pc@us.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Riccardo Mancini <rickyman7@gmail.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Suzuki Poulouse <suzuki.poulose@arm.com>
Cc: Vineet Singh <vineet.singh@intel.com>
Cc: coresight@lists.linaro.org
Cc: linux-arm-kernel@lists.infradead.org
Cc: zhengjun.xing@intel.com
Link: https://lore.kernel.org/r/20220105061351.120843-36-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Previously evlist__for_each_cpu() needed to iterate over the evlist in
an inner loop and call "skip" routines. Refactor this so that the
iteratr is smarter and the next function can update both the current CPU
and evsel.
By using a cpu map index, fix apparent off-by-1 in __run_perf_stat's
call to perf_evsel__close_cpu().
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Clark <james.clark@arm.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Mike Leach <mike.leach@linaro.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Clarke <pc@us.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Riccardo Mancini <rickyman7@gmail.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Suzuki Poulouse <suzuki.poulose@arm.com>
Cc: Vineet Singh <vineet.singh@intel.com>
Cc: coresight@lists.linaro.org
Cc: linux-arm-kernel@lists.infradead.org
Cc: zhengjun.xing@intel.com
Link: https://lore.kernel.org/r/20220105061351.120843-35-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The functions don't use a cpu_map so reduce them to being like
constructors of aggr_cpu_id.
Reviewed-by: James Clark <james.clark@arm.com>
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Mike Leach <mike.leach@linaro.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Clarke <pc@us.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Riccardo Mancini <rickyman7@gmail.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Suzuki Poulouse <suzuki.poulose@arm.com>
Cc: Vineet Singh <vineet.singh@intel.com>
Cc: coresight@lists.linaro.org
Cc: linux-arm-kernel@lists.infradead.org
Cc: zhengjun.xing@intel.com
Link: https://lore.kernel.org/r/20220105061351.120843-20-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Remove cpu_map from name as a cpu_map isn't used. Pass a const pointer
rather than by value to avoid unnecessary copying.
Reviewed-by: James Clark <james.clark@arm.com>
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Mike Leach <mike.leach@linaro.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Clarke <pc@us.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Riccardo Mancini <rickyman7@gmail.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Suzuki Poulouse <suzuki.poulose@arm.com>
Cc: Vineet Singh <vineet.singh@intel.com>
Cc: coresight@lists.linaro.org
Cc: linux-arm-kernel@lists.infradead.org
Cc: zhengjun.xing@intel.com
Link: https://lore.kernel.org/r/20220105061351.120843-15-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Avoid possible bugs where the wrong index is passed with the cpu_map.
Reviewed-by: James Clark <james.clark@arm.com>
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Mike Leach <mike.leach@linaro.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Clarke <pc@us.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Riccardo Mancini <rickyman7@gmail.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Suzuki Poulouse <suzuki.poulose@arm.com>
Cc: Vineet Singh <vineet.singh@intel.com>
Cc: coresight@lists.linaro.org
Cc: linux-arm-kernel@lists.infradead.org
Cc: zhengjun.xing@intel.com
Link: https://lore.kernel.org/r/20220105061351.120843-6-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
In previous patch, we have supported the syntax which enables
the event on a specified pmu, such as:
cpu_core/<event>/
cpu_atom/<event>/
While this syntax is not very easy for applying on a set of
events or applying on a group. In following example, we have to
explicitly assign the pmu prefix.
# ./perf stat -e '{cpu_core/cycles/,cpu_core/instructions/}' -- sleep 1
Performance counter stats for 'sleep 1':
1,158,545 cpu_core/cycles/
1,003,113 cpu_core/instructions/
1.002428712 seconds time elapsed
A much easier way is:
# ./perf stat --cputype core -e '{cycles,instructions}' -- sleep 1
Performance counter stats for 'sleep 1':
1,101,071 cpu_core/cycles/
939,892 cpu_core/instructions/
1.002363142 seconds time elapsed
For this example, the '--cputype' enables the events from specified
pmu (cpu_core).
If '--cputype' conflicts with pmu prefix, '--cputype' is ignored.
# ./perf stat --cputype core -e cycles,cpu_atom/instructions/ -a -- sleep 1
Performance counter stats for 'system wide':
21,003,407 cpu_core/cycles/
367,886 cpu_atom/instructions/
1.002203520 seconds time elapsed
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jin Yao <yao.jin@intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20210909062215.10278-1-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
parse_events() may succeed but leave string memory allocations reachable
in the error.
Add an init/exit that must be called to initialize and clean up the
error. This fixes a leak in metricgroup parse_ids.
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lore.kernel.org/lkml/20211107090002.3784612-2-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Group error functions and name after the data type they manipulate.
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lore.kernel.org/lkml/20211107090002.3784612-1-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
An iostate use case like "perf iostat 0000:16,0000:97 -- ls" should be
implemented to work in system-wide mode to ensure that the output from
print_header() is consistent with the user documentation perf-iostat.txt,
rather than incorrectly assuming that the kernel does not support it:
Error:
The sys_perf_event_open() syscall returned with 22 (Invalid argument) \
for event (uncore_iio_0/event=0x83,umask=0x04,ch_mask=0xF,fc_mask=0x07/).
/bin/dmesg | grep -i perf may provide additional information.
This error is easily fixed by assigning system-wide mode by default
for IOSTAT_RUN only when the target cpu_list is unspecified.
Fixes: f07952b179 ("perf stat: Basic support for iostat in perf")
Signed-off-by: Like Xu <likexu@tencent.com>
Cc: Alexander Antonov <alexander.antonov@linux.intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lore.kernel.org/lkml/20210927081115.39568-1-likexu@tencent.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The cgroup mode should work with cpu events. Warn if --for-each-cgroup
option is used with a task target like existing -G option.
# perf stat --for-each-cgroup . sleep 1
both cgroup and no-aggregation modes only available in system-wide mode
Usage: perf stat [<options>] [<command>]
-G, --cgroup <name> monitor event in cgroup name only
-A, --no-aggr disable CPU count aggregation
-a, --all-cpus system-wide collection from all CPUs
--for-each-cgroup <name>
expand events for each cgroup
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20210830170200.55652-1-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The 'perf record' and 'perf stat' commands have supported the option
'-C/--cpus' to count or collect only on the list of CPUs provided. This
option needs to be supported for hybrid as well.
For hybrid support, it needs to check that the cpu list are available
on hybrid PMU. One example for AlderLake, cpu0-7 is 'cpu_core', cpu8-11
is 'cpu_atom'.
Before:
# perf stat -e cpu_core/cycles/ -C11 -- sleep 1
Performance counter stats for 'CPU(s) 11':
<not supported> cpu_core/cycles/
1.006179431 seconds time elapsed
The 'perf stat' command silently returned "<not supported>" without any
helpful information. It should error out pointing out that that cpu11
was not 'cpu_core'.
After:
# perf stat -e cpu_core/cycles/ -C11 -- sleep 1
WARNING: 11 isn't a 'cpu_core', please use a CPU list in the 'cpu_core' range (0-7)
failed to use cpu list 11
We also need to support the events without pmu prefix specified.
# perf stat -e cycles -C11 -- sleep 1
WARNING: 11 isn't a 'cpu_core', please use a CPU list in the 'cpu_core' range (0-7)
Performance counter stats for 'CPU(s) 11':
1,067,373 cpu_atom/cycles/
1.005544738 seconds time elapsed
The perf tool creates two cycles events automatically, cpu_core/cycles/ and
cpu_atom/cycles/. It checks that cpu11 is not 'cpu_core', then shows a warning
for cpu_core/cycles/ and only count the cpu_atom/cycles/.
If part of cpus are 'cpu_core' and part of cpus are 'cpu_atom', for example,
# perf stat -e cycles -C0,11 -- sleep 1
WARNING: use 0 in 'cpu_core' for 'cycles', skip other cpus in list.
WARNING: use 11 in 'cpu_atom' for 'cycles', skip other cpus in list.
Performance counter stats for 'CPU(s) 0,11':
1,914,704 cpu_core/cycles/
2,036,983 cpu_atom/cycles/
1.005815641 seconds time elapsed
It now automatically selects cpu0 for cpu_core/cycles/, selects cpu11 for
cpu_atom/cycles/, and output with some warnings.
Some more complex examples,
# perf stat -e cycles,instructions -C0,11 -- sleep 1
WARNING: use 0 in 'cpu_core' for 'cycles', skip other cpus in list.
WARNING: use 11 in 'cpu_atom' for 'cycles', skip other cpus in list.
WARNING: use 0 in 'cpu_core' for 'instructions', skip other cpus in list.
WARNING: use 11 in 'cpu_atom' for 'instructions', skip other cpus in list.
Performance counter stats for 'CPU(s) 0,11':
2,780,387 cpu_core/cycles/
1,583,432 cpu_atom/cycles/
3,957,277 cpu_core/instructions/
1,167,089 cpu_atom/instructions/
1.006005124 seconds time elapsed
# perf stat -e cycles,cpu_atom/instructions/ -C0,11 -- sleep 1
WARNING: use 0 in 'cpu_core' for 'cycles', skip other cpus in list.
WARNING: use 11 in 'cpu_atom' for 'cycles', skip other cpus in list.
WARNING: use 11 in 'cpu_atom' for 'cpu_atom/instructions/', skip other cpus in list.
Performance counter stats for 'CPU(s) 0,11':
3,290,301 cpu_core/cycles/
1,953,073 cpu_atom/cycles/
1,407,869 cpu_atom/instructions/
1.006260912 seconds time elapsed
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jin Yao <yao.jin@intel.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https //lore.kernel.org/r/20210723063433.7318-4-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The repipe argument is only used by perf inject and the all others
passes 'false'. Let's remove it from the function signature and add
__perf_session__new() to be called from perf inject directly.
This is a preparation of the change the pipe input/output.
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20210719223153.1618812-2-namhyung@kernel.org
[ Fixed up some trivial conflicts as this patchset fell thru the cracks ;-( ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
On a hybrid platform, by default 'perf stat' aggregates and reports the
event counts per PMU. For example,
# perf stat -e cycles -a true
Performance counter stats for 'system wide':
1,400,445 cpu_core/cycles/
680,881 cpu_atom/cycles/
0.001770773 seconds time elapsed
But for uncore events that's not a suitable method. Uncore has nothing
to do with hybrid. So for uncore events, we aggregate event counts from
all PMUs and report the counts without PMUs.
Before:
# perf stat -e arb/event=0x81,umask=0x1/,arb/event=0x84,umask=0x1/ -a true
Performance counter stats for 'system wide':
2,058 uncore_arb_0/event=0x81,umask=0x1/
2,028 uncore_arb_1/event=0x81,umask=0x1/
0 uncore_arb_0/event=0x84,umask=0x1/
0 uncore_arb_1/event=0x84,umask=0x1/
0.000614498 seconds time elapsed
After:
# perf stat -e arb/event=0x81,umask=0x1/,arb/event=0x84,umask=0x1/ -a true
Performance counter stats for 'system wide':
3,996 arb/event=0x81,umask=0x1/
0 arb/event=0x84,umask=0x1/
0.000630046 seconds time elapsed
Of course, we also keep the '--no-merge' working for uncore events.
# perf stat -e arb/event=0x81,umask=0x1/,arb/event=0x84,umask=0x1/ --no-merge true
Performance counter stats for 'system wide':
1,952 uncore_arb_0/event=0x81,umask=0x1/
1,921 uncore_arb_1/event=0x81,umask=0x1/
0 uncore_arb_0/event=0x84,umask=0x1/
0 uncore_arb_1/event=0x84,umask=0x1/
0.000575536 seconds time elapsed
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210707055652.962-1-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The Topdown Microarchitecture Analysis (TMA) Method is a structured
analysis methodology to identify critical performance bottlenecks in
out-of-order processors.
The Topdown metrics L1 event was added as default in 42641d6f4d
("perf stat: Add Topdown metrics events as default events")
From the Sapphire Rapids server and later platforms, the same dedicated
"metrics" register is extended to support both L1 and L2 events.
Add both L1 and L2 Topdown metrics events as default to enrich the
default measuring information if the new measurement register is
available.
On legacy systems there is no change to avoid extra multiplexing.
The topdown_level indicates the max metrics level for the top-down
statistics. Set it to 2 to display all L1 and L2 Topdown metrics events.
With the patch:
$ perf stat sleep 1
Performance counter stats for 'sleep 1':
0.59 msec task-clock # 0.001 CPUs utilized
1 context-switches # 1.687 K/sec
0 cpu-migrations # 0.000 /sec
76 page-faults # 128.198 K/sec
1,405,318 cycles # 2.371 GHz
1,471,136 instructions # 1.05 insn per cycle
310,132 branches # 523.136 M/sec
10,435 branch-misses # 3.36% of all branches
8,431,908 slots # 14.223 G/sec
1,554,116 topdown-retiring # 18.4% retiring
1,289,585 topdown-bad-spec # 15.2% bad speculation
2,810,636 topdown-fe-bound # 33.2% frontend bound
2,810,636 topdown-be-bound # 33.2% backend bound
231,464 topdown-heavy-ops # 2.7% heavy operations # 15.6% light operations
1,223,453 topdown-br-mispredict # 14.5% branch mispredict # 0.8% machine clears
1,884,779 topdown-fetch-lat # 22.3% fetch latency # 10.9% fetch bandwidth
1,454,917 topdown-mem-bound # 17.2% memory bound # 16.0% Core bound
1.001179699 seconds time elapsed
0.000000000 seconds user
0.001238000 seconds sys
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Jin Yao <yao.jin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lore.kernel.org/lkml/1625760169-18396-1-git-send-email-kan.liang@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Move evsel::leader to perf_evsel::leader, so we can move the group
interface to libperf.
Also add several evsel helpers to ease up the transition:
struct evsel *evsel__leader(struct evsel *evsel);
- get leader evsel
bool evsel__has_leader(struct evsel *evsel, struct evsel *leader);
- true if evsel has leader as leader
bool evsel__is_leader(struct evsel *evsel);
- true if evsel is itw own leader
void evsel__set_leader(struct evsel *evsel, struct evsel *leader);
- set leader for evsel
Committer notes:
Fix this when building with 'make BUILD_BPF_SKEL=1'
tools/perf/util/bpf_counter.c
- if (evsel->leader->core.nr_members > 1) {
+ if (evsel->core.leader->nr_members > 1) {
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Requested-by: Shunsuke Nakamura <nakamura.shun@fujitsu.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Michael Petlan <mpetlan@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20210706151704.73662-4-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When all events of a perf-stat session use BPF, it is not necessary to
call evlist__enable() and evlist__disable(). Skip them when
all_counters_use_bpf is true.
Signed-off-by: Song Liu <song@kernel.org>
Reported-by: Jiri Olsa <jolsa@redhat.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
If a group has events which are from different hybrid PMUs,
shows a warning:
"WARNING: events in group from different hybrid PMUs!"
This is to remind the user not to put the core event and atom
event into one group.
Next, just disable grouping.
# perf stat -e "{cpu_core/cycles/,cpu_atom/cycles/}" -a -- sleep 1
WARNING: events in group from different hybrid PMUs!
WARNING: grouped events cpus do not match, disabling group:
anon group { cpu_core/cycles/, cpu_atom/cycles/ }
Performance counter stats for 'system wide':
5,438,125 cpu_core/cycles/
3,914,586 cpu_atom/cycles/
1.004250966 seconds time elapsed
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Reviewed-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210427070139.25256-17-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Previously if '-e' is not specified in perf stat, some software events
and hardware events are added to evlist by default.
Before:
# perf stat -a -- sleep 1
Performance counter stats for 'system wide':
24,044.40 msec cpu-clock # 23.946 CPUs utilized
99 context-switches # 4.117 /sec
24 cpu-migrations # 0.998 /sec
3 page-faults # 0.125 /sec
7,000,244 cycles # 0.000 GHz
2,955,024 instructions # 0.42 insn per cycle
608,941 branches # 25.326 K/sec
31,991 branch-misses # 5.25% of all branches
1.004106859 seconds time elapsed
Among the events, cycles, instructions, branches and branch-misses
are hardware events.
One hybrid platform, two hardware events are created for one
hardware event.
cpu_core/cycles/,
cpu_atom/cycles/,
cpu_core/instructions/,
cpu_atom/instructions/,
cpu_core/branches/,
cpu_atom/branches/,
cpu_core/branch-misses/,
cpu_atom/branch-misses/
These events would be added to evlist on hybrid platform.
Since parse_events() has been supported to create two hardware events
for one event on hybrid platform, so we just use parse_events(evlist,
"cycles,instructions,branches,branch-misses") to create the default
events and add them to evlist.
After:
# perf stat -a -- sleep 1
Performance counter stats for 'system wide':
24,043.99 msec cpu-clock # 23.991 CPUs utilized
139 context-switches # 5.781 /sec
25 cpu-migrations # 1.040 /sec
6 page-faults # 0.250 /sec
10,381,751 cpu_core/cycles/ # 431.782 K/sec
1,264,216 cpu_atom/cycles/ # 52.579 K/sec
3,406,958 cpu_core/instructions/ # 141.697 K/sec
414,588 cpu_atom/instructions/ # 17.243 K/sec
705,149 cpu_core/branches/ # 29.327 K/sec
82,358 cpu_atom/branches/ # 3.425 K/sec
40,821 cpu_core/branch-misses/ # 1.698 K/sec
9,086 cpu_atom/branch-misses/ # 377.891 /sec
1.002228863 seconds time elapsed
We can see two events are created for one hardware event.
One TODO is, the shadow stats looks a bit different, now it's just
'M/sec'.
The perf_stat__update_shadow_stats and perf_stat__print_shadow_stats
need to be improved in future if we want to get the original shadow
stats.
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Reviewed-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210427070139.25256-15-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
It would be useful to let user know the pmu which the event belongs to.
perf-stat has supported '--no-merge' option and it can print the pmu
name after the event name, such as:
"cycles [cpu_core]"
Now this option is enabled by default for hybrid platform but change
the format to:
"cpu_core/cycles/"
If user configs the name, we still use the user specified name.
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Reviewed-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
ink: https://lore.kernel.org/r/20210427070139.25256-8-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Currently, to use BPF to aggregate perf event counters, the user uses
--bpf-counters option. Enable "use bpf by default" events with a config
option, stat.bpf-counter-events. Events with name in the option will use
BPF.
This also enables mixed BPF event and regular event in the same sesssion.
For example:
perf config stat.bpf-counter-events=instructions
perf stat -e instructions,cs
The second command will use BPF for "instructions" but not "cs".
Signed-off-by: Song Liu <song@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Song Liu <songliubraving@fb.com>
Link: https://lore.kernel.org/r/20210425214333.1090950-4-song@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add basic flow for a new iostat mode in perf. Mode is intended to
provide four I/O performance metrics per each PCIe root port: Inbound Read,
Inbound Write, Outbound Read, Outbound Write.
The actual code to compute the metrics and attribute it to
root port is in follow-on patches.
Signed-off-by: Alexander Antonov <alexander.antonov@linux.intel.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexey V Bayduraev <alexey.v.bayduraev@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210419094147.15909-2-alexander.antonov@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The 'perf stat' subcommand supports the request for a summary of the
interval counter readings. But the summary lines break the CSV output
so it's hard for scripts to parse the result.
Before:
# perf stat -x, -I1000 --interval-count 1 --summary
1.001323097,8013.48,msec,cpu-clock,8013483384,100.00,8.013,CPUs utilized
1.001323097,270,,context-switches,8013513297,100.00,0.034,K/sec
1.001323097,13,,cpu-migrations,8013530032,100.00,0.002,K/sec
1.001323097,184,,page-faults,8013546992,100.00,0.023,K/sec
1.001323097,20574191,,cycles,8013551506,100.00,0.003,GHz
1.001323097,10562267,,instructions,8013564958,100.00,0.51,insn per cycle
1.001323097,2019244,,branches,8013575673,100.00,0.252,M/sec
1.001323097,106152,,branch-misses,8013585776,100.00,5.26,of all branches
8013.48,msec,cpu-clock,8013483384,100.00,7.984,CPUs utilized
270,,context-switches,8013513297,100.00,0.034,K/sec
13,,cpu-migrations,8013530032,100.00,0.002,K/sec
184,,page-faults,8013546992,100.00,0.023,K/sec
20574191,,cycles,8013551506,100.00,0.003,GHz
10562267,,instructions,8013564958,100.00,0.51,insn per cycle
2019244,,branches,8013575673,100.00,0.252,M/sec
106152,,branch-misses,8013585776,100.00,5.26,of all branches
The summary line loses the timestamp column, which breaks the CSV
output.
We add a column at the original 'timestamp' position and it just says
'summary' for the summary line.
After:
# perf stat -x, -I1000 --interval-count 1 --summary
1.001196053,8012.72,msec,cpu-clock,8012722903,100.00,8.013,CPUs utilized
1.001196053,218,,context-switches,8012753271,100.00,0.027,K/sec
1.001196053,9,,cpu-migrations,8012769767,100.00,0.001,K/sec
1.001196053,0,,page-faults,8012786257,100.00,0.000,K/sec
1.001196053,15004518,,cycles,8012790637,100.00,0.002,GHz
1.001196053,7954691,,instructions,8012804027,100.00,0.53,insn per cycle
1.001196053,1590259,,branches,8012814766,100.00,0.198,M/sec
1.001196053,82601,,branch-misses,8012824365,100.00,5.19,of all branches
summary,8012.72,msec,cpu-clock,8012722903,100.00,7.986,CPUs utilized
summary,218,,context-switches,8012753271,100.00,0.027,K/sec
summary,9,,cpu-migrations,8012769767,100.00,0.001,K/sec
summary,0,,page-faults,8012786257,100.00,0.000,K/sec
summary,15004518,,cycles,8012790637,100.00,0.002,GHz
summary,7954691,,instructions,8012804027,100.00,0.53,insn per cycle
summary,1590259,,branches,8012814766,100.00,0.198,M/sec
summary,82601,,branch-misses,8012824365,100.00,5.19,of all branches
Now it's easy for script to analyse the summary lines.
Of course, we also consider not to break possible existing scripts which
can continue to use the broken CSV format by using a new '--no-csv-summary.'
option.
# perf stat -x, -I1000 --interval-count 1 --summary --no-csv-summary
1.001213261,8012.67,msec,cpu-clock,8012672327,100.00,8.013,CPUs utilized
1.001213261,197,,context-switches,8012703742,100.00,24.586,/sec
1.001213261,9,,cpu-migrations,8012720902,100.00,1.123,/sec
1.001213261,644,,page-faults,8012738266,100.00,80.373,/sec
1.001213261,18350698,,cycles,8012744109,100.00,0.002,GHz
1.001213261,12745021,,instructions,8012759001,100.00,0.69,insn per cycle
1.001213261,2458033,,branches,8012770864,100.00,306.768,K/sec
1.001213261,102107,,branch-misses,8012781751,100.00,4.15,of all branches
8012.67,msec,cpu-clock,8012672327,100.00,7.985,CPUs utilized
197,,context-switches,8012703742,100.00,24.586,/sec
9,,cpu-migrations,8012720902,100.00,1.123,/sec
644,,page-faults,8012738266,100.00,80.373,/sec
18350698,,cycles,8012744109,100.00,0.002,GHz
12745021,,instructions,8012759001,100.00,0.69,insn per cycle
2458033,,branches,8012770864,100.00,306.768,K/sec
102107,,branch-misses,8012781751,100.00,4.15,of all branches
This option can be enabled in perf config by setting the variable
'stat.no-csv-summary'.
# perf config stat.no-csv-summary=true
# perf config -l
stat.no-csv-summary=true
# perf stat -x, -I1000 --interval-count 1 --summary
1.001330198,8013.28,msec,cpu-clock,8013279201,100.00,8.013,CPUs utilized
1.001330198,205,,context-switches,8013308394,100.00,25.583,/sec
1.001330198,10,,cpu-migrations,8013324681,100.00,1.248,/sec
1.001330198,0,,page-faults,8013340926,100.00,0.000,/sec
1.001330198,8027742,,cycles,8013344503,100.00,0.001,GHz
1.001330198,2871717,,instructions,8013356501,100.00,0.36,insn per cycle
1.001330198,553564,,branches,8013366204,100.00,69.081,K/sec
1.001330198,54021,,branch-misses,8013375952,100.00,9.76,of all branches
8013.28,msec,cpu-clock,8013279201,100.00,7.985,CPUs utilized
205,,context-switches,8013308394,100.00,25.583,/sec
10,,cpu-migrations,8013324681,100.00,1.248,/sec
0,,page-faults,8013340926,100.00,0.000,/sec
8027742,,cycles,8013344503,100.00,0.001,GHz
2871717,,instructions,8013356501,100.00,0.36,insn per cycle
553564,,branches,8013366204,100.00,69.081,K/sec
54021,,branch-misses,8013375952,100.00,9.76,of all branches
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Acked-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jin Yao <yao.jin@intel.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20210319070156.20394-1-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Take measurements of 't0' and 'ref_time' after enable_counters(), so
that they only measure the time consumed when the counters are enabled.
Signed-off-by: Song Liu <songliubraving@fb.com>
Acked-by: Andi Kleen <andi@firstfloor.org>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: kernel-team@fb.com
Link: http://lore.kernel.org/lkml/20210316211837.910506-3-songliubraving@fb.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The perf tool uses performance monitoring counters (PMCs) to monitor
system performance. The PMCs are limited hardware resources. For
example, Intel CPUs have 3x fixed PMCs and 4x programmable PMCs per cpu.
Modern data center systems use these PMCs in many different ways: system
level monitoring, (maybe nested) container level monitoring, per process
monitoring, profiling (in sample mode), etc. In some cases, there are
more active perf_events than available hardware PMCs. To allow all
perf_events to have a chance to run, it is necessary to do expensive
time multiplexing of events.
On the other hand, many monitoring tools count the common metrics
(cycles, instructions). It is a waste to have multiple tools create
multiple perf_events of "cycles" and occupy multiple PMCs.
bperf tries to reduce such wastes by allowing multiple perf_events of
"cycles" or "instructions" (at different scopes) to share PMUs. Instead
of having each perf-stat session to read its own perf_events, bperf uses
BPF programs to read the perf_events and aggregate readings to BPF maps.
Then, the perf-stat session(s) reads the values from these BPF maps.
Please refer to the comment before the definition of bperf_ops for the
description of bperf architecture.
bperf is off by default. To enable it, pass --bpf-counters option to
perf-stat. bperf uses a BPF hashmap to share information about BPF
programs and maps used by bperf. This map is pinned to bpffs. The
default path is /sys/fs/bpf/perf_attr_map. The user could change the
path with option --bpf-attr-map.
Committer testing:
# dmesg|grep "Performance Events" -A5
[ 0.225277] Performance Events: Fam17h+ core perfctr, AMD PMU driver.
[ 0.225280] ... version: 0
[ 0.225280] ... bit width: 48
[ 0.225281] ... generic registers: 6
[ 0.225281] ... value mask: 0000ffffffffffff
[ 0.225281] ... max period: 00007fffffffffff
#
# for a in $(seq 6) ; do perf stat -a -e cycles,instructions sleep 100000 & done
[1] 2436231
[2] 2436232
[3] 2436233
[4] 2436234
[5] 2436235
[6] 2436236
# perf stat -a -e cycles,instructions sleep 0.1
Performance counter stats for 'system wide':
310,326,987 cycles (41.87%)
236,143,290 instructions # 0.76 insn per cycle (41.87%)
0.100800885 seconds time elapsed
#
We can see that the counters were enabled for this workload 41.87% of
the time.
Now with --bpf-counters:
# for a in $(seq 32) ; do perf stat --bpf-counters -a -e cycles,instructions sleep 100000 & done
[1] 2436514
[2] 2436515
[3] 2436516
[4] 2436517
[5] 2436518
[6] 2436519
[7] 2436520
[8] 2436521
[9] 2436522
[10] 2436523
[11] 2436524
[12] 2436525
[13] 2436526
[14] 2436527
[15] 2436528
[16] 2436529
[17] 2436530
[18] 2436531
[19] 2436532
[20] 2436533
[21] 2436534
[22] 2436535
[23] 2436536
[24] 2436537
[25] 2436538
[26] 2436539
[27] 2436540
[28] 2436541
[29] 2436542
[30] 2436543
[31] 2436544
[32] 2436545
#
# ls -la /sys/fs/bpf/perf_attr_map
-rw-------. 1 root root 0 Mar 23 14:53 /sys/fs/bpf/perf_attr_map
# bpftool map | grep bperf | wc -l
64
#
# bpftool map | tail
1265: percpu_array name accum_readings flags 0x0
key 4B value 24B max_entries 1 memlock 4096B
1266: hash name filter flags 0x0
key 4B value 4B max_entries 1 memlock 4096B
1267: array name bperf_fo.bss flags 0x400
key 4B value 8B max_entries 1 memlock 4096B
btf_id 996
pids perf(2436545)
1268: percpu_array name accum_readings flags 0x0
key 4B value 24B max_entries 1 memlock 4096B
1269: hash name filter flags 0x0
key 4B value 4B max_entries 1 memlock 4096B
1270: array name bperf_fo.bss flags 0x400
key 4B value 8B max_entries 1 memlock 4096B
btf_id 997
pids perf(2436541)
1285: array name pid_iter.rodata flags 0x480
key 4B value 4B max_entries 1 memlock 4096B
btf_id 1017 frozen
pids bpftool(2437504)
1286: array flags 0x0
key 4B value 32B max_entries 1 memlock 4096B
#
# bpftool map dump id 1268 | tail
value (CPU 21):
8f f3 bc ca 00 00 00 00 80 fd 2a d1 4d 00 00 00
80 fd 2a d1 4d 00 00 00
value (CPU 22):
7e d5 64 4d 00 00 00 00 a4 8a 2e ee 4d 00 00 00
a4 8a 2e ee 4d 00 00 00
value (CPU 23):
a7 78 3e 06 01 00 00 00 b2 34 94 f6 4d 00 00 00
b2 34 94 f6 4d 00 00 00
Found 1 element
# bpftool map dump id 1268 | tail
value (CPU 21):
c6 8b d9 ca 00 00 00 00 20 c6 fc 83 4e 00 00 00
20 c6 fc 83 4e 00 00 00
value (CPU 22):
9c b4 d2 4d 00 00 00 00 3e 0c df 89 4e 00 00 00
3e 0c df 89 4e 00 00 00
value (CPU 23):
18 43 66 06 01 00 00 00 5b 69 ed 83 4e 00 00 00
5b 69 ed 83 4e 00 00 00
Found 1 element
# bpftool map dump id 1268 | tail
value (CPU 21):
f2 6e db ca 00 00 00 00 92 67 4c ba 4e 00 00 00
92 67 4c ba 4e 00 00 00
value (CPU 22):
dc 8e e1 4d 00 00 00 00 d9 32 7a c5 4e 00 00 00
d9 32 7a c5 4e 00 00 00
value (CPU 23):
bd 2b 73 06 01 00 00 00 7c 73 87 bf 4e 00 00 00
7c 73 87 bf 4e 00 00 00
Found 1 element
#
# perf stat --bpf-counters -a -e cycles,instructions sleep 0.1
Performance counter stats for 'system wide':
119,410,122 cycles
152,105,479 instructions # 1.27 insn per cycle
0.101395093 seconds time elapsed
#
See? We had the counters enabled all the time.
Signed-off-by: Song Liu <songliubraving@fb.com>
Reviewed-by: Jiri Olsa <jolsa@kernel.org>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: kernel-team@fb.com
Link: http://lore.kernel.org/lkml/20210316211837.910506-2-songliubraving@fb.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The TMA method level 2 metrics is supported from the Intel Sapphire
Rapids server, which expose four L2 Topdown metrics events to user
space. There are eight L2 events in total. The other four L2 Topdown
metrics events are calculated from the corresponding L1 and the exposed
L2 events.
Now, the --topdown prints the complete top-down metrics that supported
by the CPU. For the Intel Sapphire Rapids server, there are 4 L1 events
and 8 L2 events displyed in one line.
Add a new option, --td-level, to display the top-down statistics that
equal to or lower than the input level.
The L2 event is marked only when both its L1 parent event and itself
crosse the threshold.
Here is an example:
$ perf stat --topdown --td-level=2 --no-metric-only sleep 1
Topdown accuracy may decrease when measuring long periods.
Please print the result regularly, e.g. -I1000
Performance counter stats for 'sleep 1':
16,734,390 slots
2,100,001 topdown-retiring # 12.6% retiring
2,034,376 topdown-bad-spec # 12.3% bad speculation
4,003,128 topdown-fe-bound # 24.1% frontend bound
328,125 topdown-heavy-ops # 2.0% heavy operations # 10.6% light operations
1,968,751 topdown-br-mispredict # 11.9% branch mispredict # 0.4% machine clears
2,953,127 topdown-fetch-lat # 17.8% fetch latency # 6.3% fetch bandwidth
5,906,255 topdown-mem-bound # 35.6% memory bound # 15.4% core bound
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jin Yao <yao.jin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Madhavan Srinivasan <maddy@linux.vnet.ibm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lore.kernel.org/lkml/1612296553-21962-9-git-send-email-kan.liang@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The Topdown Microarchitecture Analysis (TMA) Method is a structured
analysis methodology to identify critical performance bottlenecks in
out-of-order processors. From the Ice Lake and later platforms, the
Topdown information can be retrieved from the dedicated "metrics"
register, which isn't impacted by other events. Also, the Topdown
metrics support both per thread/process and per core measuring. Adding
Topdown metrics events as default events can enrich the default
measuring information, and would not cost any extra multiplexing.
Introduce arch_evlist__add_default_attrs() to allow architecture
specific default events. Add the Topdown metrics events in the X86
specific arch_evlist__add_default_attrs(). Other architectures can add
their own default events later separately.
With the patch:
$ perf stat sleep 1
Performance counter stats for 'sleep 1':
0.82 msec task-clock:u # 0.001 CPUs utilized
0 context-switches:u # 0.000 K/sec
0 cpu-migrations:u # 0.000 K/sec
61 page-faults:u # 0.074 M/sec
319,941 cycles:u # 0.388 GHz
242,802 instructions:u # 0.76 insn per cycle
54,380 branches:u # 66.028 M/sec
4,043 branch-misses:u # 7.43% of all branches
1,585,555 slots:u # 1925.189 M/sec
238,941 topdown-retiring:u # 15.0% retiring
410,378 topdown-bad-spec:u # 25.8% bad speculation
634,222 topdown-fe-bound:u # 39.9% frontend bound
304,675 topdown-be-bound:u # 19.2% backend bound
1.001791625 seconds time elapsed
0.000000000 seconds user
0.001572000 seconds sys
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jin Yao <yao.jin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lore.kernel.org/lkml/20210121133752.118327-1-kan.liang@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add a control 'ping' command to detect if perf is up and its control
interface is operational.
It will be used in following daemon patches to synchronize with record
session - when control interface is up and running, we know that perf
record is monitoring and ready to receive signals.
Example session:
terminal 1:
# mkfifo control ack
# perf record --control=fifo:control,ack
terminal 2:
# echo ping > control
# cat ack
ack
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexei Budankov <abudankov@huawei.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Michael Petlan <mpetlan@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lore.kernel.org/lkml/20201226232038.390883-5-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Adding control 'stop' command to stop perf record.
When it is received, perf will set the 'done' variable to 1 to stop its
mmap ring buffer reading loop.
Example session:
terminal 1:
# mkfifo control ack
# perf record --control=fifo:control,ack
terminal 2:
# echo stop > control
terminal 1:
[ perf record: Woken up 7 times to write data ]
[ perf record: Captured and wrote 3.214 MB perf.data (38280 samples) ]
#
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexei Budankov <abudankov@huawei.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Michael Petlan <mpetlan@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lore.kernel.org/lkml/20201226232038.390883-4-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add a new 'evlist' control command to display all the evlist events.
When it is received, perf will scan and print current evlist into perf
record terminal.
The interface string for control file is:
evlist [-v|-g|-F]
The syntax follows perf evlist command:
-F Show just the sample frequency used for each event.
-v Show all fields.
-g Show event group information.
Example session:
terminal 1:
# mkfifo control ack
# perf record --control=fifo:control,ack -e '{cycles,instructions}'
terminal 2:
# echo evlist > control
terminal 1:
cycles
instructions
dummy:HG
terminal 2:
# echo 'evlist -v' > control
terminal 1:
cycles: size: 120, { sample_period, sample_freq }: 4000, sample_type: \
IP|TID|TIME|ID|CPU|PERIOD, read_format: ID, disabled: 1, inherit: 1, freq: 1, \
sample_id_all: 1, exclude_guest: 1
instructions: size: 120, config: 0x1, { sample_period, sample_freq }: 4000, \
sample_type: IP|TID|TIME|ID|CPU|PERIOD, read_format: ID, inherit: 1, freq: 1, \
sample_id_all: 1, exclude_guest: 1
dummy:HG: type: 1, size: 120, config: 0x9, { sample_period, sample_freq }: 4000, \
sample_type: IP|TID|TIME|ID|CPU|PERIOD, read_format: ID, inherit: 1, mmap: 1, \
comm: 1, freq: 1, task: 1, sample_id_all: 1, mmap2: 1, comm_exec: 1, ksymbol: 1, \
bpf_event: 1
terminal 2:
# echo 'evlist -g' > control
terminal 1:
{cycles,instructions}
dummy:HG
terminal 2:
# echo 'evlist -F' > control
terminal 1:
cycles: sample_freq=4000
instructions: sample_freq=4000
dummy:HG: sample_freq=4000
This new evlist command is handy to get real event names when
wildcards are used.
Adding evsel_fprintf.c object to python/perf.so build, because
it's now evlist.c dependency.
Adding PYTHON_PERF define for python/perf.so compilation, so we
can use it to compile in only evsel__fprintf from evsel_fprintf.c
object.
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexei Budankov <abudankov@huawei.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Michael Petlan <mpetlan@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lore.kernel.org/lkml/20201226232038.390883-3-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Adding new control events to enable/disable specific event.
The interface string for control file are:
'enable <EVENT NAME>'
'disable <EVENT NAME>'
when received the command, perf will scan the current evlist
for <EVENT NAME> and if found it's enabled/disabled.
Example session:
terminal 1:
# mkfifo control ack perf.pipe
# perf record --control=fifo:control,ack -D -1 --no-buffering -e 'sched:*' -o - > perf.pipe
terminal 2:
# cat perf.pipe | perf --no-pager script -i -
terminal 1:
Events disabled
NOTE Above message will show only after read side of the pipe ('>')
is started on 'terminal 2'. The 'terminal 1's bash does not execute
perf before that, hence the delyaed perf record message.
terminal 3:
# echo 'enable sched:sched_process_fork' > control
terminal 1:
event sched:sched_process_fork enabled
terminal 2:
bash 33349 [034] 149587.674295: sched:sched_process_fork: comm=bash pid=33349 child_comm=bash child_pid=34056
bash 33349 [034] 149588.239521: sched:sched_process_fork: comm=bash pid=33349 child_comm=bash child_pid=34057
terminal 3:
# echo 'enable sched:sched_wakeup_new' > control
terminal 1:
event sched:sched_wakeup_new enabled
terminal 2:
bash 33349 [034] 149632.228023: sched:sched_process_fork: comm=bash pid=33349 child_comm=bash child_pid=34059
bash 33349 [034] 149632.228050: sched:sched_wakeup_new: bash:34059 [120] success=1 CPU:036
bash 33349 [034] 149633.950005: sched:sched_process_fork: comm=bash pid=33349 child_comm=bash child_pid=34060
bash 33349 [034] 149633.950030: sched:sched_wakeup_new: bash:34060 [120] success=1 CPU:036
Committer testing:
If I use 'sched:*' and then enable all events, I can't get 'perf record'
to react to further commands, so I tested it with:
[root@five ~]# perf record --control=fifo:control,ack -D -1 --no-buffering -e 'sched:sched_process_*' -o - > perf.pipe
Events disabled
Events enabled
Events disabled
And then it works as expected, so we need to fix this pre-existing
problem.
Another issue, we need to check if a event is already enabled or
disabled and change the message to be clearer, i.e.:
[root@five ~]# perf record --control=fifo:control,ack -D -1 --no-buffering -e 'sched:sched_process_*' -o - > perf.pipe
Events disabled
If we receive a 'disable' command, then it should say:
[root@five ~]# perf record --control=fifo:control,ack -D -1 --no-buffering -e 'sched:sched_process_*' -o - > perf.pipe
Events disabled
Events already disabled
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexei Budankov <abudankov@huawei.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Michael Petlan <mpetlan@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lore.kernel.org/lkml/20201226232038.390883-2-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Introduce 'perf stat -b' option, which counts events for BPF programs, like:
[root@localhost ~]# ~/perf stat -e ref-cycles,cycles -b 254 -I 1000
1.487903822 115,200 ref-cycles
1.487903822 86,012 cycles
2.489147029 80,560 ref-cycles
2.489147029 73,784 cycles
3.490341825 60,720 ref-cycles
3.490341825 37,797 cycles
4.491540887 37,120 ref-cycles
4.491540887 31,963 cycles
The example above counts 'cycles' and 'ref-cycles' of BPF program of id
254. This is similar to bpftool-prog-profile command, but more
flexible.
'perf stat -b' creates per-cpu perf_event and loads fentry/fexit BPF
programs (monitor-progs) to the target BPF program (target-prog). The
monitor-progs read perf_event before and after the target-prog, and
aggregate the difference in a BPF map. Then the user space reads data
from these maps.
A new 'struct bpf_counter' is introduced to provide a common interface
that uses BPF programs/maps to count perf events.
Committer notes:
Removed all but bpf_counter.h includes from evsel.h, not needed at all.
Also BPF map lookups for PERCPU_ARRAYs need to have as its value receive
buffer passed to the kernel libbpf_num_possible_cpus() entries, not
evsel__nr_cpus(evsel), as the former uses
/sys/devices/system/cpu/possible while the later uses
/sys/devices/system/cpu/online, which may be less than the 'possible'
number making the bpf map lookup overwrite memory and cause hard to
debug memory corruption.
We need to continue using evsel__nr_cpus(evsel) when accessing the
perf_counts array tho, not to overwrite another are of memory :-)
Signed-off-by: Song Liu <songliubraving@fb.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Link: https://lore.kernel.org/lkml/20210120163031.GU12699@kernel.org/
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: kernel-team@fb.com
Link: http://lore.kernel.org/lkml/20201229214214.3413833-4-songliubraving@fb.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add core as a separate member so that it doesn't have to be packed into
the int value.
Signed-off-by: James Clark <james.clark@arm.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Tested-by: John Garry <john.garry@huawei.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Richter <tmricht@linux.ibm.com>
Link: https://lore.kernel.org/r/20201126141328.6509-12-james.clark@arm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>