[lkml]   [2013]   [Sep]   [26]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCHSET 0/8] perf tools: Fix scalability problem on callchain merging (v4)

On Thu, 26 Sep 2013 11:07:49 -0300, Arnaldo Carvalho de Melo wrote:
> Em Thu, Sep 26, 2013 at 07:46:36AM -0600, David Ahern escreveu:
>> On 9/26/13 2:58 AM, Namhyung Kim wrote:
>> >This is a new version of callchain improvement patchset. I found and
>> >fixed bugs in the previous version. I verified that it produced
>> >exactly same output before and after applying rbtree conversion patch
>> >(#1). However after Frederic's new comm infrastructure patches are
>> >applied it'd be little different.
>> >The patches are on 'perf/callchain-v4' branch in my tree
>> > git://
>> Given recent breakage, has the patchset been run through any tests
>> on older kernels that do not support sample_id_all? e.g., 2.6.34
>> (WRL4). What about tests with the other perf commands -- script to
>> dump events, trace on a file with with multiple processes -- to
>> verify no impact on comm output, especially multithreaded processes
>> with named threads. I can certainly do those tests in time, but
>> can't guarantee a timeframe and want to make sure it gets done
>> before merging.
> Right, and I don't think this is perf/urgent material at this point in
> time, when merged will be for 3.13.

Agreed. It definitely needs more testing. I'll try to do some testing
mentioned above.


 \ /
  Last update: 2013-09-27 04:21    [W:0.044 / U:2.332 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site