lkml.org 
[lkml]   [2005]   [Oct]   [31]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH]: Clean up of __alloc_pages
From
Date
On Sat, 2005-10-29 at 17:16 -0700, Paul Jackson wrote:
> Seth wroteL
> > @@ -851,19 +853,11 @@
> > * Ignore cpuset if GFP_ATOMIC (!wait) rather than fail alloc.
> > * See also cpuset_zone_allowed() comment in kernel/cpuset.c.
> > */
> > - for (i = 0; (z = zones[i]) != NULL; i++) {
> > - if (!zone_watermark_ok(z, order, z->pages_min,
> > - classzone_idx, can_try_harder,
> > - gfp_mask & __GFP_HIGH))
> > - continue;
> > -
> > - if (wait && !cpuset_zone_allowed(z, gfp_mask))
> > - continue;
> > -
> > - page = buffered_rmqueue(z, order, gfp_mask);
> > - if (page)
> > - goto got_pg;
> > - }
> > + if (!wait)
> > + page = get_page_from_freelist(gfp_mask, order, zones,
> > + can_try_harder);
>
> Thanks for the clean-up work. Good stuff.
>
> I think you've changed the affect that the cpuset check has on the
> above pass.
>
> As you know, the above is the last chance we have for GFP_ATOMIC (can't
> wait) allocations before getting into the oom_kill code. The code had
> been written to ignore cpuset constraints for GFP_ATOMIC (that is,
> "!wait") allocations. The intent is to allow taking GFP_ATOMIC memory
> from any damn node we can find it on, rather than start killing.
>
> Your change will call into get_page_from_freelist() in such cases,
> where the cpuset check is still done.


Shooo....I will fix it.

-rohit

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-10-31 20:05    [W:0.135 / U:0.096 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site