Messages in this thread | | | Subject | Re: [patch 2/2] sched: fix select_idle_sibling() logic in select_task_rq_fair() | From | Mike Galbraith <> | Date | Tue, 09 Mar 2010 07:05:49 +0100 |
| |
On Mon, 2010-03-08 at 14:24 -0800, Suresh Siddha wrote: > On Fri, 2010-03-05 at 12:25 -0800, Mike Galbraith wrote: > > On Fri, 2010-03-05 at 10:39 -0800, Suresh Siddha wrote: > > > plain text document attachment (fix_lat_ctx.patch) > > > Performance improvements with this patch: > > > "lat_ctx -s 0 2" ~22usec (before-this-patch) ~5usec (after-this-patch) > > > > Hm. On my Q6600 box, it's nowhere near that. > > My numbers are based on an atom netbook.
Wish I had an atom to play with, hard to even imagine 22 usec lat_ctx, that's some _serious_ pain.
> > Calling the waking cpu idle in that case is a mistake. Just because the > > sync hint was used does not mean there is no gain to be had. > > Ok. I dropped that part in v2 patches that I just posted.
Yeah, I see mails (ramble ramble) crossed in the night. I'll take them out for a spin, see if your box can get fixed up without busting mine :)
-Mike
| |