From c03914b7aa319fb2b6701a6427c13752c7418b9b Mon Sep 17 00:00:00 2001 From: Roman Gushchin Date: Thu, 11 Jul 2019 20:56:02 -0700 Subject: mm: memcg/slab: postpone kmem_cache memcg pointer initialization to memcg_link_cache() Patch series "mm: reparent slab memory on cgroup removal", v7. # Why do we need this? We've noticed that the number of dying cgroups is steadily growing on most of our hosts in production. The following investigation revealed an issue in the userspace memory reclaim code [1], accounting of kernel stacks [2], and also the main reason: slab objects. The underlying problem is quite simple: any page charged to a cgroup holds a reference to it, so the cgroup can't be reclaimed unless all charged pages are gone. If a slab object is actively used by other cgroups, it won't be reclaimed, and will prevent the origin cgroup from being reclaimed. Slab objects, and first of all vfs cache, is shared between cgroups, which are using the same underlying fs, and what's even more important, it's shared between multiple generations of the same workload. So if something is running periodically every time in a new cgroup (like how systemd works), we do accumulate multiple dying cgroups. Strictly speaking pagecache isn't different here, but there is a key difference: we disable protection and apply some extra pressure on LRUs of dying cgroups, and these LRUs contain all charged pages. My experiments show that with the disabled kernel memory accounting the number of dying cgroups stabilizes at a relatively small number (~100, depends on memory pressure and cgroup creation rate), and with kernel memory accounting it grows pretty steadily up to several thousands. Memory cgroups are quite complex and big objects (mostly due to percpu stats), so it leads to noticeable memory losses. Memory occupied by dying cgroups is measured in hundreds of megabytes. I've even seen a host with more than 100Gb of memory wasted for dying cgroups. It leads to a degradation of performance with the uptime, and generally limits the usage of cgroups. My previous attempt [3] to fix the problem by applying extra pressure on slab shrinker lists caused a regressions with xfs and ext4, and has been reverted [4]. The following attempts to find the right balance [5, 6] were not successful. So instead of trying to find a maybe non-existing balance, let's do reparent accounted slab caches to the parent cgroup on cgroup removal. # Implementation approach There is however a significant problem with reparenting of slab memory: there is no list of charged pages. Some of them are in shrinker lists, but not all. Introducing of a new list is really not an option. But fortunately there is a way forward: every slab page has a stable pointer to the corresponding kmem_cache. So the idea is to reparent kmem_caches instead of slab pages. It's actually simpler and cheaper, but requires some underlying changes: 1) Make kmem_caches to hold a single reference to the memory cgroup, instead of a separate reference per every slab page. 2) Stop setting page->mem_cgroup pointer for memcg slab pages and use page->kmem_cache->memcg indirection instead. It's used only on slab page release, so performance overhead shouldn't be a big issue. 3) Introduce a refcounter for non-root slab caches. It's required to be able to destroy kmem_caches when they become empty and release the associated memory cgroup. There is a bonus: currently we release all memcg kmem_caches all together with the memory cgroup itself. This patchset allows individual kmem_caches to be released as soon as they become inactive and free. Some additional implementation details are provided in corresponding commit messages. # Results Below is the average number of dying cgroups on two groups of our production hosts. They do run some sort of web frontend workload, the memory pressure is moderate. As we can see, with the kernel memory reparenting the number stabilizes in 60s range; however with the original version it grows almost linearly and doesn't show any signs of plateauing. The difference in slab and percpu usage between patched and unpatched versions also grows linearly. In 7 days it exceeded 200Mb. day 0 1 2 3 4 5 6 7 original 56 362 628 752 1070 1250 1490 1560 patched 23 46 51 55 60 57 67 69 mem diff(Mb) 22 74 123 152 164 182 214 241 # Links [1]: commit 68600f623d69 ("mm: don't miss the last page because of round-off error") [2]: commit 9b6f7e163cd0 ("mm: rework memcg kernel stack accounting") [3]: commit 172b06c32b94 ("mm: slowly shrink slabs with a relatively small number of objects") [4]: commit a9a238e83fbb ("Revert "mm: slowly shrink slabs with a relatively small number of objects") [5]: https://lkml.org/lkml/2019/1/28/1865 [6]: https://marc.info/?l=linux-mm&m=155064763626437&w=2 This patch (of 10): Initialize kmem_cache->memcg_params.memcg pointer in memcg_link_cache() rather than in init_memcg_params(). Once kmem_cache will hold a reference to the memory cgroup, it will simplify the refcounting. For non-root kmem_caches memcg_link_cache() is always called before the kmem_cache becomes visible to a user, so it's safe. Link: http://lkml.kernel.org/r/20190611231813.3148843-2-guro@fb.com Signed-off-by: Roman Gushchin Reviewed-by: Shakeel Butt Acked-by: Vladimir Davydov Acked-by: Johannes Weiner Cc: Waiman Long Cc: Michal Hocko Cc: Christoph Lameter Cc: Pekka Enberg Cc: David Rientjes Cc: Joonsoo Kim Cc: Andrei Vagin Cc: Qian Cai Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/slub.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'mm/slub.c') diff --git a/mm/slub.c b/mm/slub.c index 5e217653286c..e1402ed19e74 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -4199,7 +4199,7 @@ static struct kmem_cache * __init bootstrap(struct kmem_cache *static_cache) } slab_init_memcg_params(s); list_add(&s->list, &slab_caches); - memcg_link_cache(s); + memcg_link_cache(s, NULL); return s; } -- cgit v1.2.3