lkml.org 
[lkml]   [2010]   [Nov]   [18]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [RFC][PATCH v2 5/7] taskstats: Improve cumulative CPU time accounting
Sorry for delay.

On 11/16, Michael Holzheu wrote:
>
> On Sat, 2010-11-13 at 19:38 +0100, Oleg Nesterov wrote:
> > First of all, let me repeat, I am not going to discuss whether we need
> > these changes or not, I do not know even if I understand your motivation.
>
> Sorry, if I was not clear enough with my descriptions. Let me try to
> describe my motivation again:

Yes, more or less I see what this patch does (I hope ;).

> 2. Because of reparent to init, there are situations where it is
> not clear to which tasks the CPU time of dead tasks between
> two snapshots has been accounted. This is a problem for example 2.

Yes, I see.

But I must admit, _personally_ I am not sure this problem worth the
trouble. And, I already mentioned daemonize() case, IOW I am not sure
it is _always_ right to choose exiting_parent->parent for accounting.
To me, this can be equally confusing. A user sees the running deamon
with ppid = 1, then this daemon exits and top reports the "unrelated"
process as cpu consumer.


But once again. I am _not_ against this patch. I never know when it
comes to new features. Certainly you know better if this suits top.

What I actually meant is: dear CC list, please look at this change
and comment ;)

Oleg.



\
 
 \ /
  Last update: 2010-11-18 18:19    [W:0.180 / U:0.632 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site