From: Ian Rogers <irogers@google.com>
Date: Sun, 25 Jun 2023 22:30:48 -0700
Subject: perf pmu: Correct auto_merge_stats test
Git-commit: 710dffc969023315c3c32717dc08b543012e60d8
Patch-mainline: v6.5-rc1
References: perf-v6.7 (jsc#PED-6012 jsc#PED-6121)
The original logic was to check is_pmu_hybrid() like in the below.
It just checks the name of PMU specifically for Intel hybrid systems
which means uncore PMU events should return false.
https://lore.kernel.org/all/20230527072210.2900565-35-irogers@google.com/
The is_pmu_hybrid() was replaced by arch-agnostic way but with the
incorrect condition which was fixed for core PMUs but not uncore.
This change fixes both.
Fixes: e23421426e13 ("perf pmu: Correct perf_pmu__auto_merge_stats() affecting hybrid")
Signed-off-by: Ian Rogers <irogers@google.com>
Tested-by: Namhyung Kim <namhyung@kernel.org>
Cc: James Clark <james.clark@arm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Link: https://lore.kernel.org/all/CAP-5=fXOi=xQ4=j5xAq+jWLR9n7uvfsWK+PzXkY1MZ3Fz-xccw@mail.gmail.com/
Link: https://lore.kernel.org/r/20230626053048.257959-1-irogers@google.com
[ rephrase the commit log a bit ]
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Tony Jones <tonyj@suse.de>
---
tools/perf/util/pmu.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c
index 64fa568a5426..8d5ecd4ff1a9 100644
--- a/tools/perf/util/pmu.c
+++ b/tools/perf/util/pmu.c
@@ -1427,7 +1427,7 @@ bool perf_pmu__supports_legacy_cache(const struct perf_pmu *pmu)
bool perf_pmu__auto_merge_stats(const struct perf_pmu *pmu)
{
- return pmu->is_core && perf_pmus__num_core_pmus() == 1;
+ return !pmu->is_core || perf_pmus__num_core_pmus() == 1;
}
bool perf_pmu__have_event(const struct perf_pmu *pmu, const char *name)