From: David Hildenbrand <david@redhat.com> To: Michal Hocko <mhocko@kernel.org> Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org, virtio-dev@lists.oasis-open.org, virtualization@lists.linux-foundation.org, kvm@vger.kernel.org, Andrew Morton <akpm@linux-foundation.org>, "Michael S . Tsirkin" <mst@redhat.com> Subject: Re: [PATCH RFC v4 12/13] mm/vmscan: Export drop_slab() and drop_slab_node() Date: Tue, 25 Feb 2020 16:09:29 +0100 [thread overview] Message-ID: <afdf8600-2339-6d74-5e3d-fa1a23384318@redhat.com> (raw) In-Reply-To: <20200225145829.GW22443@dhcp22.suse.cz> On 25.02.20 15:58, Michal Hocko wrote: > On Thu 12-12-19 18:11:36, David Hildenbrand wrote: >> We already have a way to trigger reclaiming of all reclaimable slab objects >> from user space (echo 2 > /proc/sys/vm/drop_caches). Let's allow drivers >> to also trigger this when they really want to make progress and know what >> they are doing. > > I cannot say I would be fan of this. This is a global action with user > visible performance impact. I am worried that we will find out that all > sorts of drivers have a very good idea that dropping slab caches is > going to help their problem whatever it is. We have seen the same patter > in the userspace already and that is the reason we are logging the usage > to the log and count invocations in the counter. Yeah, I decided to hold back patch 11-13 for the v1 (which I am planning to post in March after more testing). What we really want is to make memory offlining an alloc_contig_range() work better with reclaimable objects. > >> virtio-mem wants to use these functions when it failed to unplug memory >> for quite some time (e.g., after 30 minutes). It will then try to >> free up reclaimable objects by dropping the slab caches every now and >> then (e.g., every 30 minutes) as long as necessary. There will be a way to >> disable this feature and info messages will be logged. >> >> In the future, we want to have a drop_slab_range() functionality >> instead. Memory offlining code has similar demands and also other >> alloc_contig_range() users (e.g., gigantic pages) could make good use of >> this feature. Adding it, however, requires more work/thought. > > We already do have a memory_notify(MEM_GOING_OFFLINE) for that purpose > and slab allocator implements a callback (slab_mem_going_offline_callback). > The callback is quite dumb and it doesn't really try to free objects > from the given memory range or even try to drop active objects which > might turn out to be hard but this sounds like a more robust way to > achieve what you want. Two things: 1. memory_notify(MEM_GOING_OFFLINE) is called after trying to isolate the page range and checking if we only have movable pages. Won't help much I guess. 2. alloc_contig_range() won't benefit from that. Something like drop_slab_range() would be better, and calling it from the right spots in the core (e.g., set_migratetype_isolate() see below). Especially, have a look at mm/page_isolation.c:set_migratetype_isolate() "FIXME: Now, memory hotplug doesn't call shrink_slab() by itself. We just check MOVABLE pages." -- Thanks, David / dhildenb
WARNING: multiple messages have this Message-ID (diff)
From: David Hildenbrand <david@redhat.com> To: Michal Hocko <mhocko@kernel.org> Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org, virtio-dev@lists.oasis-open.org, virtualization@lists.linux-foundation.org, kvm@vger.kernel.org, Andrew Morton <akpm@linux-foundation.org>, "Michael S . Tsirkin" <mst@redhat.com> Subject: [virtio-dev] Re: [PATCH RFC v4 12/13] mm/vmscan: Export drop_slab() and drop_slab_node() Date: Tue, 25 Feb 2020 16:09:29 +0100 [thread overview] Message-ID: <afdf8600-2339-6d74-5e3d-fa1a23384318@redhat.com> (raw) In-Reply-To: <20200225145829.GW22443@dhcp22.suse.cz> On 25.02.20 15:58, Michal Hocko wrote: > On Thu 12-12-19 18:11:36, David Hildenbrand wrote: >> We already have a way to trigger reclaiming of all reclaimable slab objects >> from user space (echo 2 > /proc/sys/vm/drop_caches). Let's allow drivers >> to also trigger this when they really want to make progress and know what >> they are doing. > > I cannot say I would be fan of this. This is a global action with user > visible performance impact. I am worried that we will find out that all > sorts of drivers have a very good idea that dropping slab caches is > going to help their problem whatever it is. We have seen the same patter > in the userspace already and that is the reason we are logging the usage > to the log and count invocations in the counter. Yeah, I decided to hold back patch 11-13 for the v1 (which I am planning to post in March after more testing). What we really want is to make memory offlining an alloc_contig_range() work better with reclaimable objects. > >> virtio-mem wants to use these functions when it failed to unplug memory >> for quite some time (e.g., after 30 minutes). It will then try to >> free up reclaimable objects by dropping the slab caches every now and >> then (e.g., every 30 minutes) as long as necessary. There will be a way to >> disable this feature and info messages will be logged. >> >> In the future, we want to have a drop_slab_range() functionality >> instead. Memory offlining code has similar demands and also other >> alloc_contig_range() users (e.g., gigantic pages) could make good use of >> this feature. Adding it, however, requires more work/thought. > > We already do have a memory_notify(MEM_GOING_OFFLINE) for that purpose > and slab allocator implements a callback (slab_mem_going_offline_callback). > The callback is quite dumb and it doesn't really try to free objects > from the given memory range or even try to drop active objects which > might turn out to be hard but this sounds like a more robust way to > achieve what you want. Two things: 1. memory_notify(MEM_GOING_OFFLINE) is called after trying to isolate the page range and checking if we only have movable pages. Won't help much I guess. 2. alloc_contig_range() won't benefit from that. Something like drop_slab_range() would be better, and calling it from the right spots in the core (e.g., set_migratetype_isolate() see below). Especially, have a look at mm/page_isolation.c:set_migratetype_isolate() "FIXME: Now, memory hotplug doesn't call shrink_slab() by itself. We just check MOVABLE pages." -- Thanks, David / dhildenb --------------------------------------------------------------------- To unsubscribe, e-mail: virtio-dev-unsubscribe@lists.oasis-open.org For additional commands, e-mail: virtio-dev-help@lists.oasis-open.org
next prev parent reply other threads:[~2020-02-25 15:09 UTC|newest] Thread overview: 116+ messages / expand[flat|nested] mbox.gz Atom feed top 2019-12-12 17:11 [PATCH RFC v4 00/13] virtio-mem: paravirtualized memory David Hildenbrand 2019-12-12 17:11 ` [virtio-dev] " David Hildenbrand 2019-12-12 17:11 ` [PATCH RFC v4 01/13] ACPI: NUMA: export pxm_to_node David Hildenbrand 2019-12-12 17:11 ` [virtio-dev] " David Hildenbrand 2019-12-12 21:43 ` Rafael J. Wysocki 2019-12-13 9:41 ` David Hildenbrand 2019-12-13 9:41 ` David Hildenbrand 2019-12-13 9:41 ` [virtio-dev] " David Hildenbrand 2019-12-13 9:47 ` Rafael J. Wysocki 2019-12-13 9:47 ` Rafael J. Wysocki 2019-12-13 9:47 ` Rafael J. Wysocki 2019-12-12 17:11 ` David Hildenbrand 2019-12-12 17:11 ` [PATCH RFC v4 02/13] virtio-mem: Paravirtualized memory hotplug David Hildenbrand 2019-12-12 17:11 ` [virtio-dev] " David Hildenbrand 2019-12-12 17:11 ` David Hildenbrand 2019-12-12 17:11 ` [PATCH RFC v4 03/13] virtio-mem: Paravirtualized memory hotunplug part 1 David Hildenbrand 2019-12-12 17:11 ` [virtio-dev] " David Hildenbrand 2019-12-12 17:11 ` David Hildenbrand 2019-12-12 17:11 ` [PATCH RFC v4 04/13] mm: Export alloc_contig_range() / free_contig_range() David Hildenbrand 2019-12-12 17:11 ` [virtio-dev] " David Hildenbrand 2019-12-12 17:11 ` David Hildenbrand 2019-12-12 17:11 ` [PATCH RFC v4 05/13] virtio-mem: Paravirtualized memory hotunplug part 2 David Hildenbrand 2019-12-12 17:11 ` [virtio-dev] " David Hildenbrand 2019-12-12 17:11 ` David Hildenbrand 2019-12-12 17:11 ` [PATCH RFC v4 06/13] mm: Allow to offline unmovable PageOffline() pages via MEM_GOING_OFFLINE David Hildenbrand 2019-12-12 17:11 ` David Hildenbrand 2019-12-12 17:11 ` [virtio-dev] " David Hildenbrand 2020-02-25 18:26 ` Alexander Duyck 2020-02-25 18:26 ` [virtio-dev] " Alexander Duyck 2020-02-25 18:26 ` Alexander Duyck 2020-02-25 18:49 ` David Hildenbrand 2020-02-25 18:49 ` [virtio-dev] " David Hildenbrand 2020-02-25 18:49 ` David Hildenbrand 2020-02-25 21:46 ` Alexander Duyck 2020-02-25 21:46 ` [virtio-dev] " Alexander Duyck 2020-02-25 21:46 ` Alexander Duyck 2020-02-25 22:19 ` David Hildenbrand 2020-02-25 22:19 ` [virtio-dev] " David Hildenbrand 2020-02-25 22:19 ` David Hildenbrand 2020-02-26 16:27 ` Alexander Duyck 2020-02-26 16:27 ` [virtio-dev] " Alexander Duyck 2020-02-26 16:27 ` Alexander Duyck 2019-12-12 17:11 ` [PATCH RFC v4 07/13] virtio-mem: Allow to offline partially unplugged memory blocks David Hildenbrand 2019-12-12 17:11 ` David Hildenbrand 2019-12-12 17:11 ` [virtio-dev] " David Hildenbrand 2019-12-12 17:11 ` [PATCH RFC v4 08/13] mm/memory_hotplug: Introduce offline_and_remove_memory() David Hildenbrand 2019-12-12 17:11 ` [virtio-dev] " David Hildenbrand 2020-02-25 14:11 ` Michal Hocko 2020-02-25 14:27 ` David Hildenbrand 2020-02-25 14:27 ` [virtio-dev] " David Hildenbrand 2020-03-02 12:48 ` Michal Hocko 2020-03-02 12:53 ` David Hildenbrand 2020-03-02 12:53 ` [virtio-dev] " David Hildenbrand 2019-12-12 17:11 ` David Hildenbrand 2019-12-12 17:11 ` [PATCH RFC v4 09/13] virtio-mem: Offline and remove completely unplugged memory blocks David Hildenbrand 2019-12-12 17:11 ` [virtio-dev] " David Hildenbrand 2019-12-12 17:11 ` David Hildenbrand 2019-12-12 17:11 ` [PATCH RFC v4 10/13] virtio-mem: Better retry handling David Hildenbrand 2019-12-12 17:11 ` David Hildenbrand 2019-12-12 17:11 ` [virtio-dev] " David Hildenbrand 2019-12-12 17:11 ` [PATCH RFC v4 11/13] mm/vmscan: Move count_vm_event(DROP_SLAB) into drop_slab() David Hildenbrand 2019-12-12 17:11 ` David Hildenbrand 2019-12-12 17:11 ` [virtio-dev] " David Hildenbrand 2020-02-25 14:13 ` Michal Hocko 2019-12-12 17:11 ` [PATCH RFC v4 12/13] mm/vmscan: Export drop_slab() and drop_slab_node() David Hildenbrand 2019-12-12 17:11 ` [virtio-dev] " David Hildenbrand 2020-02-25 14:58 ` Michal Hocko 2020-02-25 15:09 ` David Hildenbrand [this message] 2020-02-25 15:09 ` [virtio-dev] " David Hildenbrand 2020-02-25 17:06 ` Michal Hocko 2020-02-25 17:23 ` David Hildenbrand 2020-02-25 17:23 ` [virtio-dev] " David Hildenbrand 2019-12-12 17:11 ` David Hildenbrand 2019-12-12 17:11 ` [PATCH RFC v4 13/13] virtio-mem: Drop slab objects when unplug continues to fail David Hildenbrand 2019-12-12 17:11 ` [virtio-dev] " David Hildenbrand 2019-12-12 17:11 ` David Hildenbrand 2019-12-13 20:15 ` [PATCH RFC v4 00/13] virtio-mem: paravirtualized memory Konrad Rzeszutek Wilk 2019-12-13 20:15 ` Konrad Rzeszutek Wilk 2019-12-16 11:03 ` David Hildenbrand 2019-12-16 11:03 ` David Hildenbrand 2019-12-16 11:03 ` [virtio-dev] " David Hildenbrand 2019-12-24 6:58 ` teawater 2019-12-24 9:28 ` David Hildenbrand 2019-12-24 9:28 ` [virtio-dev] " David Hildenbrand 2019-12-24 9:28 ` David Hildenbrand 2020-01-09 13:48 ` David Hildenbrand 2020-01-09 13:48 ` [virtio-dev] " David Hildenbrand 2020-01-29 9:41 ` David Hildenbrand 2020-01-29 9:41 ` [virtio-dev] " David Hildenbrand 2020-01-29 9:41 ` David Hildenbrand 2020-02-25 9:58 ` David Hildenbrand 2020-02-25 9:58 ` [virtio-dev] " David Hildenbrand 2020-02-25 9:58 ` David Hildenbrand 2020-06-05 8:55 ` Alex Shi 2020-06-05 8:55 ` Alex Shi 2020-06-05 9:08 ` David Hildenbrand 2020-06-05 9:08 ` [virtio-dev] " David Hildenbrand 2020-06-05 9:08 ` David Hildenbrand 2020-06-05 9:36 ` David Hildenbrand 2020-06-05 9:36 ` [virtio-dev] " David Hildenbrand 2020-06-05 9:36 ` David Hildenbrand 2020-06-05 10:05 ` David Hildenbrand 2020-06-05 10:05 ` [virtio-dev] " David Hildenbrand 2020-06-05 10:05 ` David Hildenbrand 2020-06-05 10:46 ` Alex Shi 2020-06-05 10:46 ` Alex Shi 2020-06-05 12:18 ` David Hildenbrand 2020-06-05 12:18 ` [virtio-dev] " David Hildenbrand 2020-06-05 12:18 ` David Hildenbrand 2020-06-09 3:05 ` Alex Shi 2020-06-09 3:05 ` Alex Shi 2020-06-05 10:08 ` Alex Shi 2020-06-05 10:08 ` Alex Shi 2020-06-05 10:06 ` Alex Shi 2020-06-05 10:06 ` Alex Shi 2020-01-09 13:48 ` David Hildenbrand
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=afdf8600-2339-6d74-5e3d-fa1a23384318@redhat.com \ --to=david@redhat.com \ --cc=akpm@linux-foundation.org \ --cc=kvm@vger.kernel.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-mm@kvack.org \ --cc=mhocko@kernel.org \ --cc=mst@redhat.com \ --cc=virtio-dev@lists.oasis-open.org \ --cc=virtualization@lists.linux-foundation.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.