perf evsel: Do not rely on errno values for precise_ip fallback
authorJiri Olsa <jolsa@kernel.org>
Wed, 3 Jul 2019 08:09:49 +0000 (10:09 +0200)
committerArnaldo Carvalho de Melo <acme@redhat.com>
Sat, 6 Jul 2019 17:30:30 +0000 (14:30 -0300)
commitcd136189370cc8a5aec0ea4b4ec517e5ee38d8a0
treedaaa7c3e2b1bbc12ac87817d7e82de482a0217ad
parent4c00af0e94cd01b8c5a5e6b3323d34677b04e192
perf evsel: Do not rely on errno values for precise_ip fallback

Konstantin reported problem with default perf record command, which
fails on some AMD servers, because of the default maximum precise
config.

The current fallback mechanism counts on getting ENOTSUP errno for
precise_ip fails, but that's not the case on some AMD servers.

We can fix this by removing the errno check completely, because the
precise_ip fallback is separated. We can just try  (if requested by
evsel->precise_max) all possible precise_ip, and if one succeeds we win,
if not, we continue with standard fallback.

Reported-by: Konstantin Kharlamov <Hi-Angel@yandex.ru>
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Quentin Monnet <quentin.monnet@netronome.com>
Cc: Kim Phillips <kim.phillips@amd.com>
Link: http://lkml.kernel.org/r/20190703080949.10356-1-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
tools/perf/util/evsel.c