lkml.org 
[lkml]   [2013]   [Sep]   [8]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [sched next] overflowed cpu time for kernel threads in /proc/PID/stat
On Sat, Aug 31, 2013 at 10:01:17PM +0300, Sergey Senozhatsky wrote:
> On (08/31/13 01:04), Frederic Weisbecker wrote:
> > But stime should always be below rtime due to the calculation done by scale_stime()
> > which roughly sums up to:
> >
> > stime = (stime / stime + utime) * rtime
> >
> > So this shouldn't happen.
> >
> > I couldn't manage to reproduce it though. Can you still trigger it with latest -next?
> >
> > Thanks.
>
> Hello,
> I can't pull current -next (away from good network connection for a
> couple of days). What I have is 3.11.0-rc7-next-20130829:
>
> ps aux | grep rcu
> root 8 0.0 0.0 0 0 ? S Aug30 0:00 [rcuc/0]
> root 9 0.0 0.0 0 0 ? S Aug30 0:00 [rcub/0]
> root 10 21376376 0.0 0 0 ? S Aug30 21114581:36 [rcu_preempt]
> root 11 21376376 0.0 0 0 ? S Aug30 21114581:35 [rcuop/0]
> root 12 21376376 0.0 0 0 ? S Aug30 21114581:35 [rcuop/1]
>
> cat /proc/10/stat
> 10 (rcu_preempt) S 2 0 0 0 -1 2129984 0 0 0 0 1844674407370 715 0 0 20 0 1 0 10 0 0 18446744073709551615 0 0 0 0 0 0 0 2147483647 0 18446744073709551615 0 0 17 0 0 0 0 0 0 0 0 0 0 0 0 0 0
>
> -ss

Sorry, I did not realize this message was posted before you tested the patch.


\
 
 \ /
  Last update: 2013-09-08 13:21    [W:0.133 / U:0.056 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site