lkml.org 
[lkml]   [2011]   [Mar]   [8]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 1/2] x86: separating entry text section

* Jiri Olsa <jolsa@redhat.com> wrote:

> On Mon, Mar 07, 2011 at 04:29:21PM +0100, Ingo Molnar wrote:
> >
> > * Jiri Olsa <jolsa@redhat.com> wrote:
> >
> > > hi,
> > > any feedback?
> >
> > Did you get my reply below?
> >
> > Thanks,
> >
> > Ingo
> >
>
> hi,
>
> strange, I believe I did reply on this.. got missed
> somehow or I'm missing simething :) reattached..

hm, the direct mail to me got lost, but i found your reply in the lkml folder.
That's how i missed your reply.

> I made another test with "reseting" the system state as suggested and only for
> cache-misses together with instructions and cycles events.
>
> I can see even bigger drop of icache load misses than before
> from 19359739 to 16448709 (about 15%).
>
> The instruction/cycles count is slightly bigger in the patched
> kernel run though..

Ok, elapsed time is more stable, it went from:

> 0.205402048 seconds time elapsed ( +- 0.299% )
> 0.206267711 seconds time elapsed ( +- 0.138% )

to:

> 0.206072845 seconds time elapsed ( +- 0.276% )
> 0.206747566 seconds time elapsed ( +- 0.137% )

which is 0.1% slower, within the noise of the measurement.

Will apply the patches, thanks Jiri!

Thanks,

Ingo


\
 
 \ /
  Last update: 2011-03-08 17:17    [W:0.075 / U:2.992 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site