Received-SPF: pass (google.com: domain of linux-kernel+bounces-97503-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
Precedence: bulk
MIME-Version: 1.0
References: <ZcWOh9u3uqZjNFMa@chrisdown.name> <20240229235134.2447718-1-axelrasmussen@google.com>
 <ZeEhvV15IWllPKvM@chrisdown.name>
In-Reply-To: <ZeEhvV15IWllPKvM@chrisdown.name>
From: Axel Rasmussen <axelrasmussen@google.com>
Date: Fri, 8 Mar 2024 11:18:28 -0800
Message-ID: <CAJHvVch2qVUDTJjNeSMqLBx0yoEm4zzb=ZXmABbd_5dWGQTpNg@mail.gmail.com>
Subject: Re: MGLRU premature memcg OOM on slow writes
To: Chris Down <chris@chrisdown.name>
Cc: cgroups@vger.kernel.org, hannes@cmpxchg.org, kernel-team@fb.com, 
	linux-kernel@vger.kernel.org, linux-mm@kvack.org, yuzhao@google.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Thu, Feb 29, 2024 at 4:30=E2=80=AFPM Chris Down <chris@chrisdown.name> w=
rote:
>
> Axel Rasmussen writes:
> >A couple of dumb questions. In your test, do you have any of the followi=
ng
> >configured / enabled?
> >
> >/proc/sys/vm/laptop_mode
> >memory.low
> >memory.min
>
> None of these are enabled. The issue is trivially reproducible by writing=
 to
> any slow device with memory.max enabled, but from the code it looks like =
MGLRU
> is also susceptible to this on global reclaim (although it's less likely =
due to
> page diversity).
>
> >Besides that, it looks like the place non-MGLRU reclaim wakes up the
> >flushers is in shrink_inactive_list() (which calls wakeup_flusher_thread=
s()).
> >Since MGLRU calls shrink_folio_list() directly (from evict_folios()), I =
agree it
> >looks like it simply will not do this.
> >
> >Yosry pointed out [1], where MGLRU used to call this but stopped doing t=
hat. It
> >makes sense to me at least that doing writeback every time we age is too
> >aggressive, but doing it in evict_folios() makes some sense to me, basic=
ally to
> >copy the behavior the non-MGLRU path (shrink_inactive_list()) has.
>
> Thanks! We may also need reclaim_throttle(), depending on how you impleme=
nt it.
> Current non-MGLRU behaviour on slow storage is also highly suspect in ter=
ms of
> (lack of) throttling after moving away from VMSCAN_THROTTLE_WRITEBACK, bu=
t one
> thing at a time :-)


Hmm, so I have a patch which I think will help with this situation,
but I'm having some trouble reproducing the problem on 6.8-rc7 (so
then I can verify the patch fixes it).

If I understand the issue right, all we should need to do is get a
slow filesystem, and then generate a bunch of dirty file pages on it,
while running in a tightly constrained memcg. To that end, I tried the
following script. But, in reality I seem to get little or no
accumulation of dirty file pages.

I thought maybe fio does something different than rsync which you said
you originally tried, so I also tried rsync (copying /usr/bin into
this loop mount) and didn't run into an OOM situation either.

Maybe some dirty ratio settings need tweaking or something to get the
behavior you see? Or maybe my test has a dumb mistake in it. :)


#!/usr/bin/env bash

echo 0 > /proc/sys/vm/laptop_mode || exit 1
echo y > /sys/kernel/mm/lru_gen/enabled || exit 1

echo "Allocate disk image"
IMAGE_SIZE_MIB=3D1024
IMAGE_PATH=3D/tmp/slow.img
dd if=3D/dev/zero of=3D$IMAGE_PATH bs=3D1024k count=3D$IMAGE_SIZE_MIB || ex=
it 1

echo "Setup loop device"
LOOP_DEV=3D$(losetup --show --find $IMAGE_PATH) || exit 1
LOOP_BLOCKS=3D$(blockdev --getsize $LOOP_DEV) || exit 1

echo "Create dm-slow"
DM_NAME=3Ddm-slow
DM_DEV=3D/dev/mapper/$DM_NAME
echo "0 $LOOP_BLOCKS delay $LOOP_DEV 0 100" | dmsetup create $DM_NAME || ex=
it 1

echo "Create fs"
mkfs.ext4 "$DM_DEV" || exit 1

echo "Mount fs"
MOUNT_PATH=3D"/tmp/$DM_NAME"
mkdir -p "$MOUNT_PATH" || exit 1
mount -t ext4 "$DM_DEV" "$MOUNT_PATH" || exit 1

echo "Generate dirty file pages"
systemd-run --wait --pipe --collect -p MemoryMax=3D32M \
        fio -name=3Dwrites -directory=3D$MOUNT_PATH -readwrite=3Drandwrite =
\
        -numjobs=3D10 -nrfiles=3D90 -filesize=3D1048576 \
        -fallocate=3Dposix \
        -blocksize=3D4k -ioengine=3Dmmap \
        -direct=3D0 -buffered=3D1 -fsync=3D0 -fdatasync=3D0 -sync=3D0 \
        -runtime=3D300 -time_based