Received-SPF: pass (google.com: domain of linux-kernel+bounces-69689-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Precedence: bulk
MIME-Version: 1.0
References: <20240208061825.36640-1-byungchul@sk.com> <CAOUHufYUC-oWePfqbbmm15Ue9QLfPg1G2nhXn6iSX_A460O6Uw@mail.gmail.com>
 <20240216072434.GC32626@system.software.com>
In-Reply-To: <20240216072434.GC32626@system.software.com>
From: Yu Zhao <yuzhao@google.com>
Date: Sat, 17 Feb 2024 00:11:25 -0500
Message-ID: <CAOUHufbGQ_ZFLhVSPG78pbMtvcfZ5v-E3oRdfZDP2mtHtkrPVg@mail.gmail.com>
Subject: Re: [PATCH] mm, vmscan: Don't turn on cache_trim_mode at the highest
 scan priority
To: Byungchul Park <byungchul@sk.com>
Cc: akpm@linux-foundation.org, linux-kernel@vger.kernel.org, 
	linux-mm@kvack.org, kernel_team@skhynix.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Fri, Feb 16, 2024 at 2:24=E2=80=AFAM Byungchul Park <byungchul@sk.com> w=
rote:
>
> On Fri, Feb 16, 2024 at 12:55:17AM -0500, Yu Zhao wrote:
> > On Thu, Feb 8, 2024 at 1:18=E2=80=AFAM Byungchul Park <byungchul@sk.com=
> wrote:
> > >
> > > With cache_trim_mode on, reclaim logic doesn't bother reclaiming anon
> > > pages. However, it should be more careful to turn on the mode because
> > > it's going to prevent anon pages from reclaimed even if there are hug=
e
> > > ammount of anon pages that are very cold so should be reclaimed. Even
> > > worse, that can lead kswapd_failures to be MAX_RECLAIM_RETRIES and st=
op
> > > until direct reclaim eventually works to resume kswapd.
> >
> > Is a theory or something observed in the real world? If it's the
> > former, would this change risk breaking existing use cases? It's the
>
> I faced the latter case.
>
> > latter, where are the performance numbers to show what it looks like
> > before and after this patch?

Let me ask again: where are the performance numbers to show what it
looks like before and after this patch?

> Before:
>
> Whenever the system meets the condition to turn on cache_trim_mode but
> few cache pages to trim, kswapd fails without scanning anon pages that
> are plenty and cold for sure and it retries 8 times and looks *stopped
> for ever*.
>
> After:
>
> When the system meets the condition to turn on cache_trim_mode but few
> cache pages to trim, kswapd finally works at the highest scan priority.
> So kswap looks working well even in the same condition.

These are not performance numbers -- what test cases can prove what's
described here?

> > > Signed-off-by: Byungchul Park <byungchul@sk.com>
> > > ---
> > >  mm/vmscan.c | 3 ++-
> > >  1 file changed, 2 insertions(+), 1 deletion(-)
> > >
> > > diff --git a/mm/vmscan.c b/mm/vmscan.c
> > > index bba207f41b14..25b55fdc0d41 100644
> > > --- a/mm/vmscan.c
> > > +++ b/mm/vmscan.c
> > > @@ -2268,7 +2268,8 @@ static void prepare_scan_control(pg_data_t *pgd=
at, struct scan_control *sc)
> > >          * anonymous pages.
> > >          */
> > >         file =3D lruvec_page_state(target_lruvec, NR_INACTIVE_FILE);
> > > -       if (file >> sc->priority && !(sc->may_deactivate & DEACTIVATE=
_FILE))
> > > +       if (sc->priority !=3D 1 && file >> sc->priority &
> >
> > Why 1?
>
> It means the highest scan priority. The priority goes from DEF_PRIORITY
> to 1.

This is not true -- sc->priority can go all the way to zero.

>         Byungchul
>
> > > +           !(sc->may_deactivate & DEACTIVATE_FILE))
> > >                 sc->cache_trim_mode =3D 1;
> > >         else
> > >                 sc->cache_trim_mode =3D 0;