Re: [PATCH v2 1/2] free_pcppages_bulk: do not hold lock when picking pages to free

From: Aaron Lu
Date: Thu Feb 22 2018 - 20:36:25 EST


On Thu, Feb 15, 2018 at 12:06:08PM +0000, Mel Gorman wrote:
> On Thu, Jan 25, 2018 at 03:21:44PM +0800, Aaron Lu wrote:
> > When freeing a batch of pages from Per-CPU-Pages(PCP) back to buddy,
> > the zone->lock is held and then pages are chosen from PCP's migratetype
> > list. While there is actually no need to do this 'choose part' under
> > lock since it's PCP pages, the only CPU that can touch them is us and
> > irq is also disabled.
> >
> > Moving this part outside could reduce lock held time and improve
> > performance. Test with will-it-scale/page_fault1 full load:
> >
> > kernel Broadwell(2S) Skylake(2S) Broadwell(4S) Skylake(4S)
> > v4.15-rc4 9037332 8000124 13642741 15728686
> > this patch 9608786 +6.3% 8368915 +4.6% 14042169 +2.9% 17433559 +10.8%
> >
> > What the test does is: starts $nr_cpu processes and each will repeatedly
> > do the following for 5 minutes:
> > 1 mmap 128M anonymouse space;
> > 2 write access to that space;
> > 3 munmap.
> > The score is the aggregated iteration.
> >
> > https://github.com/antonblanchard/will-it-scale/blob/master/tests/page_fault1.c
> >
> > Acked-by: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx>
> > Signed-off-by: Aaron Lu <aaron.lu@xxxxxxxxx>
>
> It looks like this series may have gotten lost because it was embedded
> within an existing thread or else it was the proximity to the merge
> window. I suggest a rebase, retest and resubmit unless there was some
> major objection that I missed. Patch 1 is fine by me at least. I never
> explicitly acked patch 2 but I've no major objection to it, just am a tad
> uncomfortable with prefetch magic sauce in general.

Thanks for the suggestion.
I just got back from vacation and will send out once I collected all the
required date.