[lkml]   [2009]   [Mar]   [9]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [patch] Re: scheduler oddity [bug?]
On Mon, 2009-03-09 at 16:57 +0100, Balazs Scheidler wrote:
> Hi,
> Just an interesting sidenote:
> I've ported the quoted patch and
> 38736f475071b80b66be28af7b44c854073699cc (the one I've found via
> bisect) to 2.6.27 but these didn't resolve my scheduling problem, both
> my test program and my application still uses only one CPU. So probably
> the rest of the scheduling patches between 2.6.27..2.6.28 have some
> effect too.

It's probably the changes to wake_affine(). 27 doesn't have the disable
sync if avg_overlap is too large tests.


 \ /
  Last update: 2009-03-10 04:19    [W:0.059 / U:1.116 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site