diff options
| author | Leo Yan <[email protected]> | 2020-08-11 13:35:05 +0800 |
|---|---|---|
| committer | Will Deacon <[email protected]> | 2020-09-07 11:42:38 +0100 |
| commit | ffdbd3d83553beb0fda00081293d5640cba477a2 (patch) | |
| tree | d1235a343357b23a12e599796801dc686cd2dc01 | |
| parent | f75aef392f869018f78cfedf3c320a6b3fcfda6b (diff) | |
arm64: perf: Add general hardware LLC events for PMUv3
This patch is to add the general hardware last level cache (LLC) events
for PMUv3: one event is for LLC access and another is for LLC miss.
With this change, perf tool can support last level cache profiling,
below is an example to demonstrate the usage on Arm64:
$ perf stat -e LLC-load-misses -e LLC-loads -- \
perf bench mem memcpy -s 1024MB -l 100 -f default
[...]
Performance counter stats for 'perf bench mem memcpy -s 1024MB -l 100 -f default':
35,534,262 LLC-load-misses # 2.16% of all LL-cache hits
1,643,946,443 LLC-loads
[...]
Signed-off-by: Leo Yan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Will Deacon <[email protected]>
| -rw-r--r-- | arch/arm64/kernel/perf_event.c | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/arch/arm64/kernel/perf_event.c b/arch/arm64/kernel/perf_event.c index 462f9a9cc44b..86e2328b0c2e 100644 --- a/arch/arm64/kernel/perf_event.c +++ b/arch/arm64/kernel/perf_event.c @@ -69,6 +69,9 @@ static const unsigned armv8_pmuv3_perf_cache_map[PERF_COUNT_HW_CACHE_MAX] [C(ITLB)][C(OP_READ)][C(RESULT_MISS)] = ARMV8_PMUV3_PERFCTR_L1I_TLB_REFILL, [C(ITLB)][C(OP_READ)][C(RESULT_ACCESS)] = ARMV8_PMUV3_PERFCTR_L1I_TLB, + [C(LL)][C(OP_READ)][C(RESULT_MISS)] = ARMV8_PMUV3_PERFCTR_LL_CACHE_MISS_RD, + [C(LL)][C(OP_READ)][C(RESULT_ACCESS)] = ARMV8_PMUV3_PERFCTR_LL_CACHE_RD, + [C(BPU)][C(OP_READ)][C(RESULT_ACCESS)] = ARMV8_PMUV3_PERFCTR_BR_PRED, [C(BPU)][C(OP_READ)][C(RESULT_MISS)] = ARMV8_PMUV3_PERFCTR_BR_MIS_PRED, }; |