Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1681323 > unrolled thread

[tip:perf/urgent] perf evsel: Set attr.exclude_kernel when probing max attr.precise_ip

Back to article view | Back to linux.kernel

Contents

  [tip:perf/urgent] perf evsel: Set attr.exclude_kernel when probing  max attr.precise_ip tip-bot for Arnaldo Carvalho de Melo <tipbot@zytor.com> - 2017-07-05 09:20 +0200

#1681323 — [tip:perf/urgent] perf evsel: Set attr.exclude_kernel when probing max attr.precise_ip

Commit-ID:  97365e81366f5ca16a9ce66cff4dd4c5b0d9f4db
Gitweb:     http://git.kernel.org/tip/97365e81366f5ca16a9ce66cff4dd4c5b0d9f4db
Author:     Arnaldo Carvalho de Melo <acme@redhat.com>
AuthorDate: Mon, 3 Jul 2017 17:01:41 -0300
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Tue, 4 Jul 2017 11:42:21 -0300

perf evsel: Set attr.exclude_kernel when probing max attr.precise_ip

We should set attr.exclude_kernel when probing for attr.precise_ip
level, otherwise !CAP_SYS_ADMIN users will not default to skidless
samples in capable hardware.

The increase in the paranoid level in commit 0161028b7c8a ("perf/core:
Change the default paranoia level to 2") broke this, fix it by excluding
kernel samples when probing.

Before:

  $ perf record usleep 1
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.018 MB perf.data (6 samples) ]
  $ perf evlist -v
  cycles:u: sample_freq: 4000, sample_type: IP|TID|TIME|PERIOD, exclude_kernel: 1

After:

  $ perf record usleep 1
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ]
  $ perf evlist -v
  cycles:ppp: sample_freq: 4000, sample_type: IP|TID|TIME|PERIOD, exclude_kernel: 1, precise_ip: 3
                                                                                     ^^^^^^^^^^^^^
                                                                                     ^^^^^^^^^^^^^
                                                                                     ^^^^^^^^^^^^^
  $

To further clarify: we always set .exclude_kernel when non !CAP_SYS_ADMIN
users profile, its just on the attr.precise_ip probing that we weren't doing
so, fix it.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Fixes: 7f8d1ade1b19 ("perf tools: By default use the most precise "cycles" hw counter available")
Link: http://lkml.kernel.org/n/tip-t2qttwhbnua62o5gt75cueml@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/evsel.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index cda44b0..9119241 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -269,6 +269,7 @@ struct perf_evsel *perf_evsel__new_cycles(void)
 	struct perf_event_attr attr = {
 		.type	= PERF_TYPE_HARDWARE,
 		.config	= PERF_COUNT_HW_CPU_CYCLES,
+		.exclude_kernel	= 1,
 	};
 	struct perf_evsel *evsel;
 

[toc] | [standalone]

Back to top | Article view | linux.kernel

csiph-web