If there's contention on slab_mutex, queueing the per-cache destruction work item on the system_wq can unnecessarily create and tie up a lot of kworkers.
Rename memcg_kmem_cache_create_wq to memcg_kmem_cache_wq and make it global and use that workqueue for the destruction work items too. While at it, convert the workqueue from an unbound workqueue to a per-cpu one with concurrency limited to 1. It's generally preferable to use per-cpu workqueues and concurrency limit of 1 is safe enough. This is suggested by Joonsoo Kim. Signed-off-by: Tejun Heo <[email protected]> Reported-by: Jay Vana <[email protected]> Cc: Vladimir Davydov <[email protected]> Cc: Christoph Lameter <[email protected]> Cc: Pekka Enberg <[email protected]> Cc: David Rientjes <[email protected]> Cc: Joonsoo Kim <[email protected]> Cc: Andrew Morton <[email protected]> --- include/linux/memcontrol.h | 1 + mm/memcontrol.c | 16 ++++++++-------- mm/slab_common.c | 2 +- 3 files changed, 10 insertions(+), 9 deletions(-) diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 4de925c..67f3303 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -810,6 +810,7 @@ void memcg_kmem_uncharge(struct page *page, int order); #if defined(CONFIG_MEMCG) && !defined(CONFIG_SLOB) extern struct static_key_false memcg_kmem_enabled_key; +extern struct workqueue_struct *memcg_kmem_cache_wq; extern int memcg_nr_cache_ids; void memcg_get_cache_ids(void); diff --git a/mm/memcontrol.c b/mm/memcontrol.c index a2b20f7f..8757403 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -317,6 +317,8 @@ void memcg_put_cache_ids(void) DEFINE_STATIC_KEY_FALSE(memcg_kmem_enabled_key); EXPORT_SYMBOL(memcg_kmem_enabled_key); +struct workqueue_struct *memcg_kmem_cache_wq; + #endif /* !CONFIG_SLOB */ /** @@ -2145,8 +2147,6 @@ struct memcg_kmem_cache_create_work { struct work_struct work; }; -static struct workqueue_struct *memcg_kmem_cache_create_wq; - static void memcg_kmem_cache_create_func(struct work_struct *w) { struct memcg_kmem_cache_create_work *cw = @@ -2178,7 +2178,7 @@ static void __memcg_schedule_kmem_cache_create(struct mem_cgroup *memcg, cw->cachep = cachep; INIT_WORK(&cw->work, memcg_kmem_cache_create_func); - queue_work(memcg_kmem_cache_create_wq, &cw->work); + queue_work(memcg_kmem_cache_wq, &cw->work); } static void memcg_schedule_kmem_cache_create(struct mem_cgroup *memcg, @@ -5780,12 +5780,12 @@ static int __init mem_cgroup_init(void) #ifndef CONFIG_SLOB /* * Kmem cache creation is mostly done with the slab_mutex held, - * so use a special workqueue to avoid stalling all worker - * threads in case lots of cgroups are created simultaneously. + * so use a workqueue with limited concurrency to avoid stalling + * all worker threads in case lots of cgroups are created and + * destroyed simultaneously. */ - memcg_kmem_cache_create_wq = - alloc_ordered_workqueue("memcg_kmem_cache_create", 0); - BUG_ON(!memcg_kmem_cache_create_wq); + memcg_kmem_cache_wq = alloc_workqueue("memcg_kmem_cache", 0, 1); + BUG_ON(!memcg_kmem_cache_wq); #endif cpuhp_setup_state_nocalls(CPUHP_MM_MEMCQ_DEAD, "mm/memctrl:dead", NULL, diff --git a/mm/slab_common.c b/mm/slab_common.c index 32610d1..5e6a98c 100644 --- a/mm/slab_common.c +++ b/mm/slab_common.c @@ -656,7 +656,7 @@ static void kmemcg_deactivate_rcufn(struct rcu_head *head) * initialized eariler. */ INIT_WORK(&s->memcg_params.deact_work, kmemcg_deactivate_workfn); - schedule_work(&s->memcg_params.deact_work); + queue_work(memcg_kmem_cache_wq, &s->memcg_params.deact_work); } /** -- 2.9.3
