[PATCH 4/4] mm: numa: Slow PTE scan rate if migration failures occur

[Dave Chinner]

On Thu, Mar 12, 2015 at 06:49:26PM +0000, Mel Gorman wrote:
> On Thu, Mar 12, 2015 at 09:20:36AM -0700, Linus Torvalds wrote:
> > On Thu, Mar 12, 2015 at 6:10 AM, Mel Gorman <mgorman at suse.de> wrote:
> > >
> > > I believe you're correct and it matches what was observed. I'm still
> > > travelling and wireless is dirt but managed to queue a test using pmd_dirty
> > 
> > Ok, thanks.
> > 
> > I'm not entirely happy with that change, and I suspect the whole
> > heuristic should be looked at much more (maybe it should also look at
> > whether it's executable, for example), but it's a step in the right
> > direction.
> > 
> 
> I can follow up when I'm back in work properly. As you have already pulled
> this in directly, can you also consider pulling in "mm: thp: return the
> correct value for change_huge_pmd" please? The other two patches were very
> minor can be resent through the normal paths later.

TO close the loop here, now I'm back home and can run tests:

config                            3.19      4.0-rc1     4.0-rc4
defaults                         8m08s        9m34s       9m14s
-o ag_stride=-1                  4m04s        4m38s       4m11s
-o bhash=101073                  6m04s       17m43s       7m35s
-o ag_stride=-1,bhash=101073     4m54s        9m58s       7m50s

It's better but there are still significant regressions, especially
for the large memory footprint cases. I haven't had a chance to look
at any stats or profiles yet, so I don't know yet whether this is
still page fault related or some other problem....

Cheers,

Dave
-- 
Dave Chinner
david at fromorbit.com