Re: [LKP] [mm] ac5b2c1891: vm-scalability.throughput -61.3% regression

From: David Rientjes
Date: Fri Dec 21 2018 - 17:15:18 EST

On Fri, 14 Dec 2018, Mel Gorman wrote:

> > In other words, I think there is a lot of potential stranding that occurs
> > for both scanners that could otherwise result in completely free
> > pageblocks. If there a single movable page present near the end of the
> > zone in an otherwise fully free pageblock, surely we can do better than
> > the current implementation that would never consider this very easy to
> > compact memory.
> >
> While it's somewhat premature, I posted a series before I had a full set
> of results because it uses free lists to reduce searches and reduces
> inference between multiple scanners. Preliminary results indicated it
> boosted allocation success rates by 20%ish, reduced migration scanning
> by 99% and free scanning by 27%.

Always good to have code to look at, I'll take a closer look. I've
unfortunately been distracted with other kernel issues lately :/