lkml.org 
[lkml]   [2008]   [Oct]   [2]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 0/4] Reclaim page capture v4
On Wed,  1 Oct 2008 13:30:57 +0100
Andy Whitcroft <apw@shadowen.org> wrote:

> For sometime we have been looking at mechanisms for improving the availability
> of larger allocations under load. One of the options we have explored is
> the capturing of pages freed under direct reclaim in order to increase the
> chances of free pages coelescing before they are subject to reallocation
> by racing allocators.
>
> Following this email is a patch stack implementing page capture during
> direct reclaim. It consits of four patches. The first two simply pull
> out existing code into helpers for reuse. The third makes buddy's use
> of struct page explicit. The fourth contains the meat of the changes,
> and its leader contains a much fuller description of the feature.
>
> This update represents a rebase to -mm and incorporates feedback from
> KOSAKI Motohiro. It also incorporates an accounting fix which was
> preventing some captures.
>
> I have done a lot of comparitive testing with and without this patch
> set and in broad brush I am seeing improvements in hugepage allocations
> (worst case size) success on all of my test systems. These tests consist
> of placing a constant stream of high order allocations on the system,
> at varying rates. The results for these various runs are then averaged
> to give an overall improvement.
>
> Absolute Effective
> x86-64 2.48% 4.58%
> powerpc 5.55% 25.22%
>
> x86-64 has a relatively small huge page size and so is always much more
> effective at allocating huge pages. Even there we get a measurable
> improvement. On powerpc the huge pages are much larger and much harder
> to recover. Here we see a full 25% increase in page recovery.
>
> It should be noted that these are worst case testing, and very agressive
> taking every possible page in the system. It would be helpful to get
> wider testing in -mm.
>
> Against: 2.6.27-rc1-mm1
>
> Andrew, please consider for -mm.
>

Hmm, can't we use "MIGRATE_ISOLATE" pageblock type for this purpose ?
The page allocater skips pageblock marked as MIGRATE_ISOLATE at allocation.
(pageblock-size is equal to HUGEPAGE size in general.)

Of course, "where should be isolated" is a problem.

Thanks,
-Kame



\
 
 \ /
  Last update: 2008-10-02 08:41    [from the cache]
©2003-2011 Jasper Spaans