[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20140602115550.GA1215@krava.brq.redhat.com>
Date: Mon, 2 Jun 2014 13:56:40 +0200
From: Jiri Olsa <jolsa@...hat.com>
To: Stephane Eranian <eranian@...gle.com>,
Yan Zheng <zheng.z.yan@...el.com>,
Andi Kleen <andi@...stfloor.org>
Cc: linux-kernel@...r.kernel.org,
Arnaldo Carvalho de Melo <acme@...nel.org>,
Corey Ashford <cjashfor@...ux.vnet.ibm.com>,
Frederic Weisbecker <fweisbec@...il.com>,
Ingo Molnar <mingo@...nel.org>,
Paul Mackerras <paulus@...ba.org>,
Peter Zijlstra <a.p.zijlstra@...llo.nl>,
Stephane Eranian <eranian@...gle.com>,
Joe Mario <jmario@...hat.com>,
Richard Fowles <rfowles@...hat.com>,
Don Zickus <dzickus@...hat.com>
Subject: [RFC/BUG] not working LLC-load-misses on SandyBridge (42) and Haswell
hi,
Joe asked about issues with LLC-load-misses on Haswell. I tried
and found it non functional on SandyBridge (42) as well:
[jolsa@...va perf]$ ./perf stat -e LLC-load-misses ls
...
Performance counter stats for 'ls':
<not supported> LLC-load-misses
0.007195547 seconds time elapsed
both SandyBridge (42 and 45) share following config arrays:
snb_hw_cache_event_ids and snb_hw_cache_extra_regs
but have different 'x86_pmu.extra_regs' which makes LLC-load-misses
event fail (in valid_mask check) in x86_pmu_extra_regs function
Haswell is sharing the same extra regs/ids config as SandyBridge (42),
so it suffers in the same way.
Not sure if thats known issue, but it looks like we need special
set of snb_hw_cache_event_ids for SandyBridge (42) ?
thanks for info,
jirka
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists