Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1470911
| From | Vineet Gupta <Vineet.Gupta1@synopsys.com> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | Re: [PATCH] arc: perf: Enable generic "cache-references" and "cache-misses" events |
| Date | 2016-08-26 19:50 +0200 |
| Message-ID | <satqp-1GW-25@gated-at.bofh.it> (permalink) |
| References | <sa1u9-tW-3@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
On 08/25/2016 04:49 AM, Alexey Brodkin wrote:
> ...
> [PERF_COUNT_ARC_EDTLB] = "edtlb", /* D-TLB Miss */
> [PERF_COUNT_ARC_EITLB] = "eitlb", /* I-TLB Miss */
> +
> + [PERF_COUNT_HW_CACHE_REFERENCES] = "imemrdc", /* Instr: mem read cached */
> + [PERF_COUNT_HW_CACHE_MISSES] = "dclm", /* D-cache Load Miss */
I think this is duplicating a mistake we already have. I vaguely remember when
doing some hackbench profiling last year with range based profiling confined to
memset routine and saw that L1-dcache-misses was counting zero. This is because it
only counts LD misses while memset only does ST.
Performance counter stats for '/sbin/hackbench':
0 L1-dcache-misses
0 L1-dcache-load-misses
1846082 L1-dcache-store-misses
@PeterZ do you concur that is wrong and we ought to setup 2 counters to do this
correctly ?
-Vineet
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
[PATCH] arc: perf: Enable generic "cache-references" and "cache-misses" events Alexey Brodkin <Alexey.Brodkin@synopsys.com> - 2016-08-25 14:00 +0200
Re: [PATCH] arc: perf: Enable generic "cache-references" and "cache-misses" events Vineet Gupta <Vineet.Gupta1@synopsys.com> - 2016-08-26 19:50 +0200
Re: [PATCH] arc: perf: Enable generic "cache-references" and "cache-misses" events Vineet Gupta <Vineet.Gupta1@synopsys.com> - 2016-08-31 21:10 +0200
Re: [PATCH] arc: perf: Enable generic "cache-references" and "cache-misses" events Peter Zijlstra <peterz@infradead.org> - 2016-09-01 10:40 +0200
csiph-web