* Re: VM patches (please review)
@ 2004-02-06 23:00 marcel cotta
2004-02-06 23:05 ` Nick Piggin
0 siblings, 1 reply; 6+ messages in thread
From: marcel cotta @ 2004-02-06 23:00 UTC (permalink / raw)
To: linux-kernel; +Cc: Nick Piggin
hey nick
these patches are working great for me (k7-2400+,256mb ram)
whenever about 100M of swap were in use, the system was
nearly unusable when i switched to another app under X
with your patches i got interactivity back and it nearly feals like 2.4
thanks :)
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: VM patches (please review)
2004-02-06 23:00 VM patches (please review) marcel cotta
@ 2004-02-06 23:05 ` Nick Piggin
0 siblings, 0 replies; 6+ messages in thread
From: Nick Piggin @ 2004-02-06 23:05 UTC (permalink / raw)
To: marcel cotta; +Cc: linux-kernel
marcel cotta wrote:
> hey nick
>
> these patches are working great for me (k7-2400+,256mb ram)
>
> whenever about 100M of swap were in use, the system was
> nearly unusable when i switched to another app under X
>
> with your patches i got interactivity back and it nearly feals like 2.4
>
> thanks :)
>
That's great Marcel, thanks for the feedback.
The patches are in Andrew's -mm tree now so they should
work their way into 2.6 after some more review and testing.
Nick
^ permalink raw reply [flat|nested] 6+ messages in thread
* VM patches (please review)
@ 2004-02-04 3:24 Nick Piggin
2004-02-04 17:16 ` Bill Davidsen
0 siblings, 1 reply; 6+ messages in thread
From: Nick Piggin @ 2004-02-04 3:24 UTC (permalink / raw)
To: Andrew Morton, Andrea Arcangeli, linux-kernel, linux-mm
http://www.kerneltrap.org/~npiggin/vm/
(may need to reload)
Here are the patches to go with my earlier post.
kernel is 2.6.2-rc3-mm1.
I'm suire I've upset at least one uncommented^Wdivine
balance so if anyone has time to review and comment
it would be appreciated.
I can email the patches to the lists if anyone would
like?
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: VM patches (please review)
2004-02-04 3:24 Nick Piggin
@ 2004-02-04 17:16 ` Bill Davidsen
2004-02-04 17:40 ` Rik van Riel
0 siblings, 1 reply; 6+ messages in thread
From: Bill Davidsen @ 2004-02-04 17:16 UTC (permalink / raw)
To: Nick Piggin; +Cc: Andrew Morton, Andrea Arcangeli, linux-kernel, linux-mm
Nick Piggin wrote:
> http://www.kerneltrap.org/~npiggin/vm/
> (may need to reload)
>
> Here are the patches to go with my earlier post.
> kernel is 2.6.2-rc3-mm1.
>
> I'm suire I've upset at least one uncommented^Wdivine
> balance so if anyone has time to review and comment
> it would be appreciated.
>
> I can email the patches to the lists if anyone would
> like?
Since this is broken down nicely, a line or two about what each patch
does or doesn't address would be useful. In particular, having just
gotten a working RSS I'm suspicious of the patch named vm-no-rss-limit
being desirable ;-)
Nice work, but it would be nice to see what problem a patch addresses to
check for blowback under some other load.
--
bill davidsen <davidsen@tmr.com>
CTO TMR Associates, Inc
Doing interesting things with small computers since 1979
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: VM patches (please review)
2004-02-04 17:16 ` Bill Davidsen
@ 2004-02-04 17:40 ` Rik van Riel
2004-02-04 23:55 ` Bill Davidsen
0 siblings, 1 reply; 6+ messages in thread
From: Rik van Riel @ 2004-02-04 17:40 UTC (permalink / raw)
To: Bill Davidsen
Cc: Nick Piggin, Andrew Morton, Andrea Arcangeli, linux-kernel,
linux-mm
On Wed, 4 Feb 2004, Bill Davidsen wrote:
> Since this is broken down nicely, a line or two about what each patch
> does or doesn't address would be useful. In particular, having just
> gotten a working RSS I'm suspicious of the patch named vm-no-rss-limit
> being desirable ;-)
The bug with the RSS limit patch is that I forgot to
change the exec() code, so when init is exec()d it
gets an RSS limit of zero, which is inherited by all
its children --> always over the RSS limit, no page
aging, etc.
I need to find the cleanest way to add the inheriting
of RSS limit at exec time and send a patchlet for that
to akpm...
--
"Debugging is twice as hard as writing the code in the first place.
Therefore, if you write the code as cleverly as possible, you are,
by definition, not smart enough to debug it." - Brian W. Kernighan
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: VM patches (please review)
2004-02-04 17:40 ` Rik van Riel
@ 2004-02-04 23:55 ` Bill Davidsen
0 siblings, 0 replies; 6+ messages in thread
From: Bill Davidsen @ 2004-02-04 23:55 UTC (permalink / raw)
To: Rik van Riel
Cc: Nick Piggin, Andrew Morton, Andrea Arcangeli, linux-kernel,
linux-mm
Rik van Riel wrote:
> On Wed, 4 Feb 2004, Bill Davidsen wrote:
>
>
>>Since this is broken down nicely, a line or two about what each patch
>>does or doesn't address would be useful. In particular, having just
>>gotten a working RSS I'm suspicious of the patch named vm-no-rss-limit
>>being desirable ;-)
>
>
> The bug with the RSS limit patch is that I forgot to
> change the exec() code, so when init is exec()d it
> gets an RSS limit of zero, which is inherited by all
> its children --> always over the RSS limit, no page
> aging, etc.
>
> I need to find the cleanest way to add the inheriting
> of RSS limit at exec time and send a patchlet for that
> to akpm...
Thank you! I had assumed that the info was being carried through exec,
fork, and pthread_* and similar, and just not enforced.
I still think this will be great for those applications known to
occasionally get a tough data set and not converge or recurse beyond all
sense.
--
bill davidsen <davidsen@tmr.com>
CTO TMR Associates, Inc
Doing interesting things with small computers since 1979
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2004-02-06 23:08 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2004-02-06 23:00 VM patches (please review) marcel cotta
2004-02-06 23:05 ` Nick Piggin
-- strict thread matches above, loose matches on Subject: below --
2004-02-04 3:24 Nick Piggin
2004-02-04 17:16 ` Bill Davidsen
2004-02-04 17:40 ` Rik van Riel
2004-02-04 23:55 ` Bill Davidsen
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox