similar to: Balloon pressuring page cache

> On Feb 3, 2020, at 2:50 PM, Nadav Amit <namit at vmware.com> wrote: > >> On Feb 3, 2020, at 8:34 AM, David Hildenbrand <david at redhat.com> wrote: >> >> On 03.02.20 17:18, Alexander Duyck wrote: >>> On Mon, 2020-02-03 at 08:11 -0500, Michael S. Tsirkin wrote: >>>> On Thu, Jan 30, 2020 at 11:59:46AM -0800, Tyler Sanderson wrote:

> On Feb 3, 2020, at 2:50 PM, Nadav Amit <namit at vmware.com> wrote: > >> On Feb 3, 2020, at 8:34 AM, David Hildenbrand <david at redhat.com> wrote: >> >> On 03.02.20 17:18, Alexander Duyck wrote: >>> On Mon, 2020-02-03 at 08:11 -0500, Michael S. Tsirkin wrote: >>>> On Thu, Jan 30, 2020 at 11:59:46AM -0800, Tyler Sanderson wrote:

On Wed, Feb 05, 2020 at 09:19:58AM +0100, David Hildenbrand wrote: > On 05.02.20 08:35, Nadav Amit wrote: > >> On Feb 3, 2020, at 2:50 PM, Nadav Amit <namit at vmware.com> wrote: > >> > >>> On Feb 3, 2020, at 8:34 AM, David Hildenbrand <david at redhat.com> wrote: > >>> > >>> On 03.02.20 17:18, Alexander Duyck wrote: >

There were apparently good reasons for moving away from OOM notifier callback: https://lkml.org/lkml/2018/7/12/314 https://lkml.org/lkml/2018/8/2/322 In particular the OOM notifier is worse than the shrinker because: 1. It is last-resort, which means the system has already gone through heroics to prevent OOM. Those heroic reclaim efforts are expensive and impact application performance.

There were apparently good reasons for moving away from OOM notifier callback: https://lkml.org/lkml/2018/7/12/314 https://lkml.org/lkml/2018/8/2/322 In particular the OOM notifier is worse than the shrinker because: 1. It is last-resort, which means the system has already gone through heroics to prevent OOM. Those heroic reclaim efforts are expensive and impact application performance.

Two fixes for issues I stumbled over while working on patch #3. Switch back to the good ol' OOM handler for VIRTIO_BALLOON_F_DEFLATE_ON_OOM as the switch to the shrinker introduce some undesired side effects. Keep the shrinker in place to handle VIRTIO_BALLOON_F_FREE_PAGE_HINT. Lengthy discussion under [1]. I tested with QEMU and "deflate-on-oom=on". Works as expected. Did not test

On Wed, Feb 5, 2020 at 8:34 AM David Hildenbrand <david at redhat.com> wrote: > Commit 71994620bb25 ("virtio_balloon: replace oom notifier with shrinker") > changed the behavior when deflation happens automatically. Instead of > deflating when called by the OOM handler, the shrinker is used. > > However, the balloon is not simply some slab cache that should be >

On Wed, Feb 5, 2020 at 8:34 AM David Hildenbrand <david at redhat.com> wrote: > Commit 71994620bb25 ("virtio_balloon: replace oom notifier with shrinker") > changed the behavior when deflation happens automatically. Instead of > deflating when called by the OOM handler, the shrinker is used. > > However, the balloon is not simply some slab cache that should be >

Commit 71994620bb25 ("virtio_balloon: replace oom notifier with shrinker") changed the behavior when deflation happens automatically. Instead of deflating when called by the OOM handler, the shrinker is used. However, the balloon is not simply some other slab cache that should be shrunk when under memory pressure. The shrinker does not have a concept of priorities yet, so this behavior

Commit 71994620bb25 ("virtio_balloon: replace oom notifier with shrinker") changed the behavior when deflation happens automatically. Instead of deflating when called by the OOM handler, the shrinker is used. However, the balloon is not simply some other slab cache that should be shrunk when under memory pressure. The shrinker does not have a concept of priorities yet, so this behavior

On 05.02.20 17:34, David Hildenbrand wrote: > Commit 71994620bb25 ("virtio_balloon: replace oom notifier with shrinker") > changed the behavior when deflation happens automatically. Instead of > deflating when called by the OOM handler, the shrinker is used. > > However, the balloon is not simply some slab cache that should be > shrunk when under memory pressure. The

On 05.02.20 17:34, David Hildenbrand wrote: > Commit 71994620bb25 ("virtio_balloon: replace oom notifier with shrinker") > changed the behavior when deflation happens automatically. Instead of > deflating when called by the OOM handler, the shrinker is used. > > However, the balloon is not simply some slab cache that should be > shrunk when under memory pressure. The

On 10.03.20 20:05, David Rientjes wrote: > On Tue, 10 Mar 2020, David Hildenbrand wrote: > >> Commit 71994620bb25 ("virtio_balloon: replace oom notifier with shrinker") >> changed the behavior when deflation happens automatically. Instead of >> deflating when called by the OOM handler, the shrinker is used. >> >> However, the balloon is not simply

On 10.03.20 20:05, David Rientjes wrote: > On Tue, 10 Mar 2020, David Hildenbrand wrote: > >> Commit 71994620bb25 ("virtio_balloon: replace oom notifier with shrinker") >> changed the behavior when deflation happens automatically. Instead of >> deflating when called by the OOM handler, the shrinker is used. >> >> However, the balloon is not simply

@Andrew, as this fix is based on free page reporting, can this go through your tree? Patch #1 contains a proper description. v3 -> v4: - Add Ack from David Rientjes - Minor tweaks to test details in the patch description v2 -> v3: - Use vb->vdev instead of vdev in all feature checks. We'll clean the other ones up later. - Add one empty line virtballoon_probe() to make it look

On Wed, Feb 05, 2020 at 05:34:02PM +0100, David Hildenbrand wrote: > Commit 71994620bb25 ("virtio_balloon: replace oom notifier with shrinker") > changed the behavior when deflation happens automatically. Instead of > deflating when called by the OOM handler, the shrinker is used. > > However, the balloon is not simply some slab cache that should be > shrunk when under

Commit 71994620bb25 ("virtio_balloon: replace oom notifier with shrinker") changed the behavior when deflation happens automatically. Instead of deflating when called by the OOM handler, the shrinker is used. However, the balloon is not simply some other slab cache that should be shrunk when under memory pressure. The shrinker does not have a concept of priorities yet, so this behavior

Commit 71994620bb25 ("virtio_balloon: replace oom notifier with shrinker") changed the behavior when deflation happens automatically. Instead of deflating when called by the OOM handler, the shrinker is used. However, the balloon is not simply some other slab cache that should be shrunk when under memory pressure. The shrinker does not have a concept of priorities yet, so this behavior

On Wed, Feb 05, 2020 at 05:34:02PM +0100, David Hildenbrand wrote: > Commit 71994620bb25 ("virtio_balloon: replace oom notifier with shrinker") > changed the behavior when deflation happens automatically. Instead of > deflating when called by the OOM handler, the shrinker is used. > > However, the balloon is not simply some slab cache that should be > shrunk when under

On Wed, Feb 05, 2020 at 05:34:02PM +0100, David Hildenbrand wrote: > Commit 71994620bb25 ("virtio_balloon: replace oom notifier with shrinker") > changed the behavior when deflation happens automatically. Instead of > deflating when called by the OOM handler, the shrinker is used. > > However, the balloon is not simply some slab cache that should be > shrunk when under