iv/linux
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

perf stat: Enable counting events for BPF programs

Browse Source 
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>
This commit is contained in:
Song Liu
2020-12-29 13:42:14 -08:00
committed by Arnaldo Carvalho de Melo
parent fbcdaa1908
commit fa853c4b83
14 changed files with 645 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
tools/perf/util/python.c
View File

@ -79,6 +79,27 @@ int metricgroup__copy_metric_events(struct evlist *evlist, struct cgroup *cgrp,
return 0;
}
/*
* XXX: All these evsel destructors need some better mechanism, like a linked
* list of destructors registered when the relevant code indeed is used instead
* of having more and more calls in perf_evsel__delete(). -- acme
*
* For now, add some more:
*
* Not to drag the BPF bandwagon...
*/
void bpf_counter__destroy(struct evsel *evsel);
int bpf_counter__install_pe(struct evsel *evsel, int cpu, int fd);
void bpf_counter__destroy(struct evsel *evsel __maybe_unused)
{
}
int bpf_counter__install_pe(struct evsel *evsel __maybe_unused, int cpu __maybe_unused, int fd __maybe_unused)
{
return 0;
}
/*
* Support debug printing even though util/debug.c is not linked. That means
* implementing 'verbose' and 'eprintf'.