lkml.org 
[lkml]   [2012]   [May]   [15]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: [GIT PULL 0/8] Annotation weekly ponies delivery
Date
Hi,

On Tue, 15 May 2012 00:06:25 +0900, Namhyung Kim wrote:
> 2012-05-14 (월), 13:55 +0200, Ingo Molnar:
>> Another detail seems to be that the bug takes per-task-inherited
>> profiling. Doing:
>>
>> perf record -a -e cycles:pp make -j64 bzImage
>>
>> produces the expected number of events. Without the '-a' the bug
>> Linus found triggers.
>>
>> Similarly, 'perf top' does not trigger the bug - because it's
>> using per CPU, not per task (inherited) profiling.
>>
>
> perf_evlist__config_attrs() has this:
>
> if (evlist->cpus->map[0] < 0)
> opts->no_inherit = true;
>
> meaning that per task profiling won't enable event inheritance. I don't
> know why it's needed though.

I found this on commit 5d2cd90922c7 ("perf evsel: Fix use of inherit"):

- /*
- * Don't allow mmap() of inherited per-task counters. This
- * would create a performance issue due to all children writing
- * to the same buffer.
- *
- * FIXME:
- * Proper fix is not to pass 'inherit' to perf_evsel__open*,
- * but a 'flags' parameter, with 'group' folded there as well,
- * then introduce a PERF_O_{MMAP,GROUP,INHERIT} enum, and if
- * O_MMAP is set, emit a warning if cpu < 0 and O_INHERIT is
- * set. Lets go for the minimal fix first tho.
- */
- evsel->attr.inherit = (cpus->map[cpu] >= 0) && inherit;
Hmm.. still don't know what to do then.

Thanks,
Namhyung
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2012-05-15 08:41    [from the cache]
©2003-2011 Jasper Spaans