[lkml]   [2013]   [Dec]   [12]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [RFC PATCH 0/4] Configurable fair allocation zone policy
On Thu, Dec 12, 2013 at 03:06:15PM +0000, Mel Gorman wrote:
> Commit 81c0a2bb ("mm: page_alloc: fair zone allocator policy") solved a
> bug whereby new pages could be reclaimed before old pages because of how
> the page allocator and kswapd interacted on the per-zone LRU lists.
> Unfortunately a side-effect missed during review was that it's now very
> easy to allocate remote memory on NUMA machines. The problem is that
> it is not a simple case of just restoring local allocation policies as
> there are genuine reasons why global page aging may be prefereable. It's
> still a major change to default behaviour so this patch makes the policy
> configurable and sets what I think is a sensible default.
> The patches are on top of some NUMA balancing patches currently in -mm.
> The first patch in the series is a patch posted by Johannes that must be
> taken into account before any of my patches on top. The last patch of the
> series is what alters default behaviour and makes the fair zone allocator
> policy configurable.
> Sniff test results based on following kernels
> vanilla 3.13-rc3 stock
> instrument-v5r1 NUMA balancing patches just to rule out any conflicts there
> lruslabonly-v1r2 Patch 1 only
> local-v1r2 Full series

These figures need to be redone. The instrument-v5r1 and later kernels
included a debugging patch that increases migration rates to trigger
another bug. The figures of local-v1r2 relative to instrument-v5r1 are
fine but not relative to 3.13.0-rc3-vanilla

Mel Gorman

 \ /
  Last update: 2013-12-12 17:21    [W:0.031 / U:0.768 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site