From patchwork Mon Dec 14 22:37:18 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Yang Shi X-Patchwork-Id: 11973295 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-13.5 required=3.0 tests=BAYES_00, DKIM_ADSP_CUSTOM_MED,DKIM_INVALID,DKIM_SIGNED,FREEMAIL_FORGED_FROMDOMAIN, FREEMAIL_FROM,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER, INCLUDES_PATCH,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,USER_AGENT_GIT autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id C11D7C4361B for ; Mon, 14 Dec 2020 22:37:54 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id 6B30322509 for ; Mon, 14 Dec 2020 22:37:54 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 6B30322509 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=gmail.com Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=owner-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix) id 0758A6B0070; Mon, 14 Dec 2020 17:37:54 -0500 (EST) Received: by kanga.kvack.org (Postfix, from userid 40) id 0001B6B0071; Mon, 14 Dec 2020 17:37:53 -0500 (EST) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id DE23F6B0072; Mon, 14 Dec 2020 17:37:53 -0500 (EST) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0186.hostedemail.com [216.40.44.186]) by kanga.kvack.org (Postfix) with ESMTP id BF1316B0070 for ; Mon, 14 Dec 2020 17:37:53 -0500 (EST) Received: from smtpin14.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay04.hostedemail.com (Postfix) with ESMTP id 75F391DE1 for ; Mon, 14 Dec 2020 22:37:53 +0000 (UTC) X-FDA: 77593351626.14.beef89_0a0fb372741e Received: from filter.hostedemail.com (10.5.16.251.rfc1918.com [10.5.16.251]) by smtpin14.hostedemail.com (Postfix) with ESMTP id 58EF618229818 for ; Mon, 14 Dec 2020 22:37:53 +0000 (UTC) X-HE-Tag: beef89_0a0fb372741e X-Filterd-Recvd-Size: 12075 Received: from mail-pl1-f193.google.com (mail-pl1-f193.google.com [209.85.214.193]) by imf03.hostedemail.com (Postfix) with ESMTP for ; Mon, 14 Dec 2020 22:37:52 +0000 (UTC) Received: by mail-pl1-f193.google.com with SMTP id 4so9747984plk.5 for ; Mon, 14 Dec 2020 14:37:52 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=from:to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; bh=ssXnZiLmnDnhCBaKkF0ViwDaE3pTbiD3bgrO3REdu1o=; b=Kdyag+afxDemndXxDu2neVdeMrMnJnpsM79GpbcCiOrzOzOecSQNwuUwNkXBaAfsUm amDvyRcr96/3GBLplUfOrS53VyimB/bD7uc8cL7ubwF/h2bPq8ljNIV1s7se7hquiP9c 3RvhvrRw/x7f2Khxqt8p3gz0hMoWA+vH1wdzGMUQd4f/e2P0RPJwesMT33kdRgRR7ZQa tTYXem2DTnlGvmP2C9FZi3W0VC37WChq1xtvTnSlMHjlpmKdvYYrKxL/nXqDKvyRHxDf CTV27ONp3jyp+fxK52/nnbpluX0KBiHPx8zrahhBuZE/0VU2+hJ6L+reyM7zj7X9D8qb /LOw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references:mime-version:content-transfer-encoding; bh=ssXnZiLmnDnhCBaKkF0ViwDaE3pTbiD3bgrO3REdu1o=; b=eBAt6RnZ7CsAnPwzqK2oqRnC6yaqnyOwIH4TzGZWGm55t3cBRe0mgSA7XffJGWwfva Tf3m3KEnDtPFz/uq3Vi/TA6rEZsH0aQhHziPY6fnSm7C+GnoXbXx45kfM0kW17NBoco0 Srab9AIs29RWF8ujyEUUaR/+iU1AiZI8m8eW9bfZ8EvoJjs2tkhhNwKphUkNR5g1mgae PV8oUWQ9ytMWtT1x8TtzLsO4klzfeRtuB4kJo5TfkzqSZbtZOVk79a6yyOskDrA8awMC KePJJHMMNROSe29lWgI3R4GjLoF7pfqjWt7+2AUD+5iSBalm5uX547Gd9wEqtDaItkVw jklA== X-Gm-Message-State: AOAM5320ApiXeIS+yXhxiSHG/iECdGh5qydxiguHCpPjZEei/0S9adrI rQ2hVUW6npsMUxkFAnTMfbU= X-Google-Smtp-Source: ABdhPJyWkKVRTRNzs5DbHGN2QcNv2wKI3ximz4j2UUbDkjomVlfb5vfSXO6RIa099VMG93AU7IgDsw== X-Received: by 2002:a17:90b:4b07:: with SMTP id lx7mr28007598pjb.230.1607985471993; Mon, 14 Dec 2020 14:37:51 -0800 (PST) Received: from localhost.localdomain (c-73-93-239-127.hsd1.ca.comcast.net. [73.93.239.127]) by smtp.gmail.com with ESMTPSA id d4sm20610758pfo.127.2020.12.14.14.37.49 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 14 Dec 2020 14:37:51 -0800 (PST) From: Yang Shi To: guro@fb.com, ktkhai@virtuozzo.com, shakeelb@google.com, david@fromorbit.com, hannes@cmpxchg.org, mhocko@suse.com, akpm@linux-foundation.org Cc: shy828301@gmail.com, linux-mm@kvack.org, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [v2 PATCH 5/9] mm: memcontrol: add per memcg shrinker nr_deferred Date: Mon, 14 Dec 2020 14:37:18 -0800 Message-Id: <20201214223722.232537-6-shy828301@gmail.com> X-Mailer: git-send-email 2.26.2 In-Reply-To: <20201214223722.232537-1-shy828301@gmail.com> References: <20201214223722.232537-1-shy828301@gmail.com> MIME-Version: 1.0 X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: Currently the number of deferred objects are per shrinker, but some slabs, for example, vfs inode/dentry cache are per memcg, this would result in poor isolation among memcgs. The deferred objects typically are generated by __GFP_NOFS allocations, one memcg with excessive __GFP_NOFS allocations may blow up deferred objects, then other innocent memcgs may suffer from over shrink, excessive reclaim latency, etc. For example, two workloads run in memcgA and memcgB respectively, workload in B is vfs heavy workload. Workload in A generates excessive deferred objects, then B's vfs cache might be hit heavily (drop half of caches) by B's limit reclaim or global reclaim. We observed this hit in our production environment which was running vfs heavy workload shown as the below tracing log: <...>-409454 [016] .... 28286961.747146: mm_shrink_slab_start: super_cache_scan+0x0/0x1a0 ffff9a83046f3458: nid: 1 objects to shrink 3641681686040 gfp_flags GFP_HIGHUSER_MOVABLE|__GFP_ZERO pgs_scanned 1 lru_pgs 15721 cache items 246404277 delta 31345 total_scan 123202138 <...>-409454 [022] .... 28287105.928018: mm_shrink_slab_end: super_cache_scan+0x0/0x1a0 ffff9a83046f3458: nid: 1 unused scan count 3641681686040 new scan count 3641798379189 total_scan 602 last shrinker return val 123186855 The vfs cache and page cache ration was 10:1 on this machine, and half of caches were dropped. This also resulted in significant amount of page caches were dropped due to inodes eviction. Make nr_deferred per memcg for memcg aware shrinkers would solve the unfairness and bring better isolation. When memcg is not enabled (!CONFIG_MEMCG or memcg disabled), the shrinker's nr_deferred would be used. And non memcg aware shrinkers use shrinker's nr_deferred all the time. Signed-off-by: Yang Shi --- include/linux/memcontrol.h | 9 +++ mm/memcontrol.c | 110 ++++++++++++++++++++++++++++++++++++- mm/vmscan.c | 4 ++ 3 files changed, 120 insertions(+), 3 deletions(-) diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 922a7f600465..1b343b268359 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -92,6 +92,13 @@ struct lruvec_stat { long count[NR_VM_NODE_STAT_ITEMS]; }; + +/* Shrinker::id indexed nr_deferred of memcg-aware shrinkers. */ +struct memcg_shrinker_deferred { + struct rcu_head rcu; + atomic_long_t nr_deferred[]; +}; + /* * Bitmap of shrinker::id corresponding to memcg-aware shrinkers, * which have elements charged to this memcg. @@ -119,6 +126,7 @@ struct mem_cgroup_per_node { struct mem_cgroup_reclaim_iter iter; struct memcg_shrinker_map __rcu *shrinker_map; + struct memcg_shrinker_deferred __rcu *shrinker_deferred; struct rb_node tree_node; /* RB tree node */ unsigned long usage_in_excess;/* Set to the value by which */ @@ -1489,6 +1497,7 @@ static inline bool mem_cgroup_under_socket_pressure(struct mem_cgroup *memcg) } extern int memcg_expand_shrinker_maps(int new_id); +extern int memcg_expand_shrinker_deferred(int new_id); extern void memcg_set_shrinker_bit(struct mem_cgroup *memcg, int nid, int shrinker_id); diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 3d4ddbb84a01..321d1818ce3d 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -394,14 +394,20 @@ DEFINE_STATIC_KEY_FALSE(memcg_kmem_enabled_key); EXPORT_SYMBOL(memcg_kmem_enabled_key); #endif -/* It is only can be changed with holding shrinker_rwsem exclusively */ +/* They are only can be changed with holding shrinker_rwsem exclusively */ static int memcg_shrinker_map_size; +static int memcg_shrinker_deferred_size; static void memcg_free_shrinker_map_rcu(struct rcu_head *head) { kvfree(container_of(head, struct memcg_shrinker_map, rcu)); } +static void memcg_free_shrinker_deferred_rcu(struct rcu_head *head) +{ + kvfree(container_of(head, struct memcg_shrinker_deferred, rcu)); +} + static int memcg_expand_one_shrinker_map(struct mem_cgroup *memcg, int size, int old_size) { @@ -430,6 +436,34 @@ static int memcg_expand_one_shrinker_map(struct mem_cgroup *memcg, return 0; } +static int memcg_expand_one_shrinker_deferred(struct mem_cgroup *memcg, + int size, int old_size) +{ + struct memcg_shrinker_deferred *new, *old; + int nid; + + for_each_node(nid) { + old = rcu_dereference_protected( + mem_cgroup_nodeinfo(memcg, nid)->shrinker_deferred, true); + /* Not yet online memcg */ + if (!old) + return 0; + + new = kvmalloc_node(sizeof(*new) + size, GFP_KERNEL, nid); + if (!new) + return -ENOMEM; + + /* Copy all old values, and clear all new ones */ + memcpy((void *)new->nr_deferred, (void *)old->nr_deferred, old_size); + memset((void *)new->nr_deferred + old_size, 0, size - old_size); + + rcu_assign_pointer(memcg->nodeinfo[nid]->shrinker_deferred, new); + call_rcu(&old->rcu, memcg_free_shrinker_deferred_rcu); + } + + return 0; +} + static void memcg_free_shrinker_maps(struct mem_cgroup *memcg) { struct mem_cgroup_per_node *pn; @@ -448,6 +482,21 @@ static void memcg_free_shrinker_maps(struct mem_cgroup *memcg) } } +static void memcg_free_shrinker_deferred(struct mem_cgroup *memcg) +{ + struct mem_cgroup_per_node *pn; + struct memcg_shrinker_deferred *deferred; + int nid; + + for_each_node(nid) { + pn = mem_cgroup_nodeinfo(memcg, nid); + deferred = rcu_dereference_protected(pn->shrinker_deferred, true); + if (deferred) + kvfree(deferred); + rcu_assign_pointer(pn->shrinker_deferred, NULL); + } +} + static int memcg_alloc_shrinker_maps(struct mem_cgroup *memcg) { struct memcg_shrinker_map *map; @@ -472,6 +521,27 @@ static int memcg_alloc_shrinker_maps(struct mem_cgroup *memcg) return ret; } +static int memcg_alloc_shrinker_deferred(struct mem_cgroup *memcg) +{ + struct memcg_shrinker_deferred *deferred; + int nid, size, ret = 0; + + down_read(&shrinker_rwsem); + size = memcg_shrinker_deferred_size; + for_each_node(nid) { + deferred = kvzalloc_node(sizeof(*deferred) + size, GFP_KERNEL, nid); + if (!deferred) { + memcg_free_shrinker_deferred(memcg); + ret = -ENOMEM; + break; + } + rcu_assign_pointer(memcg->nodeinfo[nid]->shrinker_deferred, deferred); + } + up_read(&shrinker_rwsem); + + return ret; +} + int memcg_expand_shrinker_maps(int new_id) { int size, old_size, ret = 0; @@ -501,6 +571,33 @@ int memcg_expand_shrinker_maps(int new_id) return ret; } +int memcg_expand_shrinker_deferred(int new_id) +{ + int size, old_size, ret = 0; + struct mem_cgroup *memcg; + + size = (new_id + 1) * sizeof(atomic_long_t); + old_size = memcg_shrinker_deferred_size; + if (size <= old_size) + return 0; + + if (!root_mem_cgroup) + goto out; + + for_each_mem_cgroup(memcg) { + ret = memcg_expand_one_shrinker_deferred(memcg, size, old_size); + if (ret) { + mem_cgroup_iter_break(NULL, memcg); + goto out; + } + } +out: + if (!ret) + memcg_shrinker_deferred_size = size; + + return ret; +} + void memcg_set_shrinker_bit(struct mem_cgroup *memcg, int nid, int shrinker_id) { if (shrinker_id >= 0 && memcg && !mem_cgroup_is_root(memcg)) { @@ -5397,8 +5494,8 @@ static int mem_cgroup_css_online(struct cgroup_subsys_state *css) struct mem_cgroup *memcg = mem_cgroup_from_css(css); /* - * A memcg must be visible for memcg_expand_shrinker_maps() - * by the time the maps are allocated. So, we allocate maps + * A memcg must be visible for memcg_expand_shrinker_{maps|deferred}() + * by the time the maps are allocated. So, we allocate maps and deferred * here, when for_each_mem_cgroup() can't skip it. */ if (memcg_alloc_shrinker_maps(memcg)) { @@ -5406,6 +5503,12 @@ static int mem_cgroup_css_online(struct cgroup_subsys_state *css) return -ENOMEM; } + if (memcg_alloc_shrinker_deferred(memcg)) { + memcg_free_shrinker_maps(memcg); + mem_cgroup_id_remove(memcg); + return -ENOMEM; + } + /* * Barrier for CSS_ONLINE, so that shrink_slab_memcg() sees shirnker_maps * and shrinker_deferred before CSS_ONLINE. It pairs with the read barrier @@ -5473,6 +5576,7 @@ static void mem_cgroup_css_free(struct cgroup_subsys_state *css) cancel_work_sync(&memcg->high_work); mem_cgroup_remove_from_trees(memcg); memcg_free_shrinker_maps(memcg); + memcg_free_shrinker_deferred(memcg); memcg_free_kmem(memcg); mem_cgroup_free(memcg); } diff --git a/mm/vmscan.c b/mm/vmscan.c index 16c9d2aeeb26..bf34167dd67e 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -219,6 +219,10 @@ static int prealloc_memcg_shrinker(struct shrinker *shrinker) goto unlock; } + if (memcg_expand_shrinker_deferred(id)) { + idr_remove(&shrinker_idr, id); + goto unlock; + } shrinker_nr_max = id + 1; } shrinker->id = id;