perf stat: Introduce config stat.bpf-counter-events
authorSong Liu <song@kernel.org>
Sun, 25 Apr 2021 21:43:31 +0000 (14:43 -0700)
committerArnaldo Carvalho de Melo <acme@redhat.com>
Thu, 29 Apr 2021 13:30:58 +0000 (10:30 -0300)
commit112cb56164bc2108a55aee785d841a35aab0616a
treefec53329a1865ca4f3a26f09e4f74f2b816b1e5b
parentfe3dd8263b9f3912a0f3a2f66c0fdb3987d69a1a
perf stat: Introduce config stat.bpf-counter-events

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>
tools/perf/Documentation/perf-stat.txt
tools/perf/builtin-stat.c
tools/perf/util/bpf_counter.c
tools/perf/util/config.c
tools/perf/util/evsel.c
tools/perf/util/evsel.h
tools/perf/util/target.h