Re: [PATCH] mm: madvise: fix uneven accounting of psi

[Johannes Weiner <hannes@cmpxchg.org>]

On Fri, Jun 09, 2023 at 04:13:14PM -0700, Suren Baghdasaryan wrote:
> On Fri, Jun 9, 2023 at 5:42 AM Charan Teja Kalla
> <quic_charante@quicinc.com> wrote:
> >
> > Thanks Suren & Johannes,
> >
> > On 6/7/2023 1:18 AM, Suren Baghdasaryan wrote:
> > > Hi Folks. Sorry for being late to the party.
> > > Yeah, userspace does not have a crystal ball to predict future user
> > > behavior, so there will always be pathological cases when usual
> > > assumptions and resulting madvise() would make things worse.
> > >
> > > I think this discussion can be split into several questions/issues:
> > > 1. Inconsistency in how madvise(MADV_PAGEOUT) would affect PSI
> > > calculation when the page is refaulted, based on the path it took
> > > before being evicted by madvise(). In your initial description case
> > > (a) is inconsistent with (b) and (c) and it's probably worth fixing.
> > > IMHO (a) should be made consistent with others, not the other way
> > > around. My reasoning is that page was expelled from the active list,
> > > so it was part of the active workingset.
> > >
> > That means we should be setting Workingset on the page while it is on
> > the active list and when it is being pageout through madvising. Right? I
> > see, this makes it consistent.
> 
> This was my opinion but others might think otherwise, like I found out
> in some recent conversations. So, it would be great to get some more
> feedback before making the change.

I also agree with the consistency fix: it should set Workingset when
madvise zaps pages from the active list.