From: Namhyung Kim <namhyung@kernel.org>
To: Arnaldo Carvalho de Melo <acme@kernel.org>,
Ian Rogers <irogers@google.com>,
Kan Liang <kan.liang@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>,
Adrian Hunter <adrian.hunter@intel.com>,
Peter Zijlstra <peterz@infradead.org>,
Ingo Molnar <mingo@kernel.org>,
LKML <linux-kernel@vger.kernel.org>,
linux-perf-users@vger.kernel.org,
Ravi Bangoria <ravi.bangoria@amd.com>,
Mark Rutland <mark.rutland@arm.com>,
James Clark <james.clark@arm.com>,
Kajol Jain <kjain@linux.ibm.com>,
Thomas Richter <tmricht@linux.ibm.com>,
Atish Patra <atishp@atishpatra.org>,
Palmer Dabbelt <palmer@rivosinc.com>,
Mingwei Zhang <mizhang@google.com>,
James Clark <james.clark@linaro.org>
Subject: [PATCH v5 1/9] perf tools: Add fallback for exclude_guest
Date: Tue, 15 Oct 2024 23:23:51 -0700 [thread overview]
Message-ID: <20241016062359.264929-2-namhyung@kernel.org> (raw)
In-Reply-To: <20241016062359.264929-1-namhyung@kernel.org>
Commit 7b100989b4f6bce70 ("perf evlist: Remove __evlist__add_default")
changed to parse "cycles:P" event instead of creating a new cycles
event for perf record. But it also changed the way how modifiers are
handled so it doesn't set the exclude_guest bit by default.
It seems Apple M1 PMU requires exclude_guest set and returns EOPNOTSUPP
if not. Let's add a fallback so that it can work with default events.
Also update perf stat hybrid tests to handle possible u or H modifiers.
Fixes: 7b100989b4f6bce70 ("perf evlist: Remove __evlist__add_default")
Reviewed-by: Ian Rogers <irogers@google.com>
Acked-by: Kan Liang <kan.liang@linux.intel.com>
Reviewed-by: James Clark <james.clark@linaro.org>
Reviewed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
---
tools/perf/builtin-stat.c | 18 +++++++++++++++---
tools/perf/tests/shell/stat.sh | 2 +-
tools/perf/util/evsel.c | 21 +++++++++++++++++++++
3 files changed, 37 insertions(+), 4 deletions(-)
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index b5e18853982cdb0f..0de3db9810db16f8 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -640,8 +640,7 @@ static enum counter_recovery stat_handle_error(struct evsel *counter)
* (behavior changed with commit b0a873e).
*/
if (errno == EINVAL || errno == ENOSYS ||
- errno == ENOENT || errno == EOPNOTSUPP ||
- errno == ENXIO) {
+ errno == ENOENT || errno == ENXIO) {
if (verbose > 0)
ui__warning("%s event is not supported by the kernel.\n",
evsel__name(counter));
@@ -659,7 +658,7 @@ static enum counter_recovery stat_handle_error(struct evsel *counter)
if (verbose > 0)
ui__warning("%s\n", msg);
return COUNTER_RETRY;
- } else if (target__has_per_thread(&target) &&
+ } else if (target__has_per_thread(&target) && errno != EOPNOTSUPP &&
evsel_list->core.threads &&
evsel_list->core.threads->err_thread != -1) {
/*
@@ -680,6 +679,19 @@ static enum counter_recovery stat_handle_error(struct evsel *counter)
return COUNTER_SKIP;
}
+ if (errno == EOPNOTSUPP) {
+ if (verbose > 0) {
+ ui__warning("%s event is not supported by the kernel.\n",
+ evsel__name(counter));
+ }
+ counter->supported = false;
+ counter->errored = true;
+
+ if ((evsel__leader(counter) != counter) ||
+ !(counter->core.leader->nr_members > 1))
+ return COUNTER_SKIP;
+ }
+
evsel__open_strerror(counter, &target, errno, msg, sizeof(msg));
ui__error("%s\n", msg);
diff --git a/tools/perf/tests/shell/stat.sh b/tools/perf/tests/shell/stat.sh
index f25ef468b8fca992..7a8adf81e4b39338 100755
--- a/tools/perf/tests/shell/stat.sh
+++ b/tools/perf/tests/shell/stat.sh
@@ -185,7 +185,7 @@ test_hybrid() {
fi
# Run default Perf stat
- cycles_events=$(perf stat -- true 2>&1 | grep -E "/cycles/| cycles " | wc -l)
+ cycles_events=$(perf stat -- true 2>&1 | grep -E "/cycles/[uH]*| cycles[:uH]* " -c)
if [ "$pmus" -ne "$cycles_events" ]
then
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index e9e1f576d34ddf64..c965ec5676f8d54a 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -3037,6 +3037,27 @@ bool evsel__fallback(struct evsel *evsel, struct target *target, int err,
evsel->core.attr.exclude_kernel = 1;
evsel->core.attr.exclude_hv = 1;
+ return true;
+ } else if (err == EOPNOTSUPP && !evsel->core.attr.exclude_guest &&
+ !evsel->exclude_GH) {
+ const char *name = evsel__name(evsel);
+ char *new_name;
+ const char *sep = ":";
+
+ /* Is there already the separator in the name. */
+ if (strchr(name, '/') ||
+ (strchr(name, ':') && !evsel->is_libpfm_event))
+ sep = "";
+
+ if (asprintf(&new_name, "%s%sH", name, sep) < 0)
+ return false;
+
+ free(evsel->name);
+ evsel->name = new_name;
+ /* Apple M1 requires exclude_guest */
+ scnprintf(msg, msgsize, "trying to fall back to excluding guest samples");
+ evsel->core.attr.exclude_guest = 1;
+
return true;
}
--
2.47.0.rc1.288.g06298d1525-goog
next prev parent reply other threads:[~2024-10-16 6:24 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-10-16 6:23 [PATCHSET 0/9] perf tools: Do not set attr.exclude_guest by default (v5) Namhyung Kim
2024-10-16 6:23 ` Namhyung Kim [this message]
2024-10-16 6:23 ` [PATCH v5 2/9] perf tools: Don't set attr.exclude_guest by default Namhyung Kim
2024-10-16 6:23 ` [PATCH v5 3/9] perf tools: Simplify evsel__add_modifier() Namhyung Kim
2024-10-16 6:23 ` [PATCH v5 4/9] perf tools: Do not set exclude_guest for precise_ip Namhyung Kim
2024-10-16 6:23 ` [PATCH v5 5/9] perf tools: Detect missing kernel features properly Namhyung Kim
2024-10-16 6:23 ` [PATCH v5 6/9] perf tools: Move x86__is_amd_cpu() to util/env.c Namhyung Kim
2024-10-16 6:23 ` [PATCH v5 7/9] perf tools: Check fallback error and order Namhyung Kim
2024-10-16 6:23 ` [PATCH v5 8/9] perf record: Just use "cycles:P" as the default event Namhyung Kim
2024-10-16 6:23 ` [PATCH v5 9/9] perf test: Add precise_max subtest to the perf record shell test Namhyung Kim
2024-10-23 18:01 ` [PATCHSET 0/9] perf tools: Do not set attr.exclude_guest by default (v5) Namhyung Kim
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20241016062359.264929-2-namhyung@kernel.org \
--to=namhyung@kernel.org \
--cc=acme@kernel.org \
--cc=adrian.hunter@intel.com \
--cc=atishp@atishpatra.org \
--cc=irogers@google.com \
--cc=james.clark@arm.com \
--cc=james.clark@linaro.org \
--cc=jolsa@kernel.org \
--cc=kan.liang@linux.intel.com \
--cc=kjain@linux.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=mark.rutland@arm.com \
--cc=mingo@kernel.org \
--cc=mizhang@google.com \
--cc=palmer@rivosinc.com \
--cc=peterz@infradead.org \
--cc=ravi.bangoria@amd.com \
--cc=tmricht@linux.ibm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).