* Re: [ck] 2.6.22-rc3-ck1
[not found] <200705270800.06612.kernel@kolivas.org>
@ 2007-05-29 19:53 ` Antonino Ingargiola
2007-05-29 19:59 ` Antonino Ingargiola
0 siblings, 1 reply; 3+ messages in thread
From: Antonino Ingargiola @ 2007-05-29 19:53 UTC (permalink / raw)
To: Con Kolivas; +Cc: ck list, linux kernel mailing list
Hi,
2007/5/27, Con Kolivas <kernel@kolivas.org>:
> New test release:
>
> http://ck.kolivas.org/patches/pre-releases/2.6.22-rc3/2.6.22-rc3-ck1/
>
> This is a resync with newer -rc and contains the updated swap prefetch code.
With this patch there is a stunning difference with the no swap
prefetch case on my system. With previous SP the difference was
around 2 seconds, now is more than 7 seconds.
Swap Prefetch OFF
# ./sp_tester
Ram 776388000 Swap 514040000
Total ram to be malloced: 1033408000 bytes
Starting first malloc of 516704000 bytes
Starting 1st read of first malloc
Touching this much ram takes 1642 milliseconds
Starting second malloc of 516704000 bytes
Completed second malloc and free
Sleeping for 60 seconds
Important part - starting reread of first malloc
Completed read of first malloc
Timed portion 9089 milliseconds
Swap Prefetch OFF
# ./sp_tester
Ram 776388000 Swap 514040000
Total ram to be malloced: 1033408000 bytes
Starting first malloc of 516704000 bytes
Starting 1st read of first malloc
Touching this much ram takes 1635 milliseconds
Starting second malloc of 516704000 bytes
Completed second malloc and free
Sleeping for 60 seconds
Important part - starting reread of first malloc
Completed read of first malloc
Timed portion 1783 milliseconds
Regards,
~ Antonio
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [ck] 2.6.22-rc3-ck1
2007-05-29 19:53 ` [ck] 2.6.22-rc3-ck1 Antonino Ingargiola
@ 2007-05-29 19:59 ` Antonino Ingargiola
2007-05-31 2:03 ` Re:swap prefetch improvements Con Kolivas
0 siblings, 1 reply; 3+ messages in thread
From: Antonino Ingargiola @ 2007-05-29 19:59 UTC (permalink / raw)
To: Con Kolivas; +Cc: ck list, linux kernel mailing list
2007/5/29, Antonino Ingargiola <tritemio@gmail.com>:
[cut]
> Swap Prefetch OFF
> # ./sp_tester
> Ram 776388000 Swap 514040000
> Total ram to be malloced: 1033408000 bytes
> Starting first malloc of 516704000 bytes
> Starting 1st read of first malloc
> Touching this much ram takes 1642 milliseconds
> Starting second malloc of 516704000 bytes
> Completed second malloc and free
> Sleeping for 60 seconds
> Important part - starting reread of first malloc
> Completed read of first malloc
> Timed portion 9089 milliseconds
>
>
> Swap Prefetch OFF
> # ./sp_tester
> Ram 776388000 Swap 514040000
> Total ram to be malloced: 1033408000 bytes
> Starting first malloc of 516704000 bytes
> Starting 1st read of first malloc
> Touching this much ram takes 1635 milliseconds
> Starting second malloc of 516704000 bytes
> Completed second malloc and free
> Sleeping for 60 seconds
> Important part - starting reread of first malloc
> Completed read of first malloc
> Timed portion 1783 milliseconds
The second case is clearly with swap prefetch *ON*, sorry.
~ Antonio
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re:swap prefetch improvements
2007-05-29 19:59 ` Antonino Ingargiola
@ 2007-05-31 2:03 ` Con Kolivas
0 siblings, 0 replies; 3+ messages in thread
From: Con Kolivas @ 2007-05-31 2:03 UTC (permalink / raw)
To: Antonino Ingargiola; +Cc: ck list, linux kernel mailing list
On Wednesday 30 May 2007 05:59, Antonino Ingargiola wrote:
> 2007/5/29, Antonino Ingargiola <tritemio@gmail.com>:
> [cut]
>
> > Swap Prefetch OFF
> > # ./sp_tester
> > Ram 776388000 Swap 514040000
> > Total ram to be malloced: 1033408000 bytes
> > Starting first malloc of 516704000 bytes
> > Starting 1st read of first malloc
> > Touching this much ram takes 1642 milliseconds
> > Starting second malloc of 516704000 bytes
> > Completed second malloc and free
> > Sleeping for 60 seconds
> > Important part - starting reread of first malloc
> > Completed read of first malloc
> > Timed portion 9089 milliseconds
> >
> >
> > Swap Prefetch OFF
> > # ./sp_tester
> > Ram 776388000 Swap 514040000
> > Total ram to be malloced: 1033408000 bytes
> > Starting first malloc of 516704000 bytes
> > Starting 1st read of first malloc
> > Touching this much ram takes 1635 milliseconds
> > Starting second malloc of 516704000 bytes
> > Completed second malloc and free
> > Sleeping for 60 seconds
> > Important part - starting reread of first malloc
> > Completed read of first malloc
> > Timed portion 1783 milliseconds
>
> The second case is clearly with swap prefetch *ON*, sorry.
Thanks very much for testing! The patch has been taken up by Andrew for the
next -mm.
どうも
--
-ck
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2007-05-30 1:04 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <200705270800.06612.kernel@kolivas.org>
2007-05-29 19:53 ` [ck] 2.6.22-rc3-ck1 Antonino Ingargiola
2007-05-29 19:59 ` Antonino Ingargiola
2007-05-31 2:03 ` Re:swap prefetch improvements Con Kolivas
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox