lkml.org 
[lkml]   [2013]   [Feb]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 1/1] perf/tools: Fix "perf record -C... workload" behaviour
On Wed, Feb 06, 2013 at 08:42:58PM +0100, Oleg Nesterov wrote:
> Test-case:
>
> # perf record -e probe_libc:syscall -C0 true
> true: Terminated
>
> The child process is killed by perf_record__sig_exit(). This is because
> perf_evlist__prepare_workload() doesn't initialize threads->map[] and
> then perf_event__synthesize_mmap_events() (silently!) fails. This patch

hi,
sorry for late reply..

I think the reason it fails is that it actually got inside
perf_event__synthesize_mmap_events while there's cpu list
specified.

I sent possible fix:
http://marc.info/?l=linux-kernel&m=136178604718144&w=2


> seems to fix the problem, but perhaps we should also move
> 'pr_err("Couldn't run the workload!\n")' down to 'out_delete_session:',

I'll check ;-)

thanks,
jirka


\
 
 \ /
  Last update: 2013-02-25 11:43    [W:0.052 / U:1.540 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site