From patchwork Mon Jul 24 09:43:52 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Qi Zheng X-Patchwork-Id: 13323996 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from lists.xenproject.org (lists.xenproject.org [192.237.175.120]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id BA372C04FDF for ; Mon, 24 Jul 2023 09:54:20 +0000 (UTC) Received: from list by lists.xenproject.org with outflank-mailman.568618.888273 (Exim 4.92) (envelope-from ) id 1qNsGY-0001gV-Li; Mon, 24 Jul 2023 09:54:10 +0000 X-Outflank-Mailman: Message body and most headers restored to incoming version Received: by outflank-mailman (output) from mailman id 568618.888273; Mon, 24 Jul 2023 09:54:10 +0000 Received: from localhost ([127.0.0.1] helo=lists.xenproject.org) by lists.xenproject.org with esmtp (Exim 4.92) (envelope-from ) id 1qNsGY-0001gL-I0; Mon, 24 Jul 2023 09:54:10 +0000 Received: by outflank-mailman (input) for mailman id 568618; Mon, 24 Jul 2023 09:54:08 +0000 Received: from se1-gles-flk1-in.inumbo.com ([94.247.172.50] helo=se1-gles-flk1.inumbo.com) by lists.xenproject.org with esmtp (Exim 4.92) (envelope-from ) id 1qNsGW-0008Df-Gl for xen-devel@lists.xenproject.org; Mon, 24 Jul 2023 09:54:08 +0000 Received: from mail-pl1-x62d.google.com (mail-pl1-x62d.google.com [2607:f8b0:4864:20::62d]) by se1-gles-flk1.inumbo.com (Halon) with ESMTPS id 06c8b78f-2a08-11ee-8612-37d641c3527e; Mon, 24 Jul 2023 11:54:06 +0200 (CEST) Received: by mail-pl1-x62d.google.com with SMTP id d9443c01a7336-1b7dfb95761so5887145ad.1 for ; Mon, 24 Jul 2023 02:54:06 -0700 (PDT) Received: from C02DW0BEMD6R.bytedance.net ([203.208.167.147]) by smtp.gmail.com with ESMTPSA id d5-20020a170902c18500b001bb20380bf2sm8467233pld.13.2023.07.24.02.53.53 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 24 Jul 2023 02:54:04 -0700 (PDT) X-BeenThere: xen-devel@lists.xenproject.org List-Id: Xen developer discussion List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Errors-To: xen-devel-bounces@lists.xenproject.org Precedence: list Sender: "Xen-devel" X-Inumbo-ID: 06c8b78f-2a08-11ee-8612-37d641c3527e DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=bytedance.com; s=google; t=1690192445; x=1690797245; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=FywtQwu9mwqdLnpgsxjfsrD11oDXqt6ENbvCNRWNCVA=; b=DikUT4CKvV9RWn+fl0EOFm95TR3cSja3PlgjCDFqYMWPqMDDzd0vmehY70wgwFHQZD xId6TApEDoayIU5bQp8ozmrUPwrV+fZdWpug1R+OQH5DrriDxnV5YGJ3iOuChC8/ataP +R3gMMBTSe4nwS0xpXqJAoWzm6vYwBlpoaI2H2xyHy5N1Vcg7pPEK5Dz8BrqbjXdo46a BIBOWCEzFcgC3VjgRrMmpjjBZE50INCoJYWq8LivZP/kkdwHaluoJFKZZtpIIdRFhRls Kq/htL25Wjh18OwWW51R+/aE+ahnELBQ5oiFmyJ9PiIhZjCyUAuo6LkxYtq+ZqQ6xAXz Fmyg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1690192445; x=1690797245; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=FywtQwu9mwqdLnpgsxjfsrD11oDXqt6ENbvCNRWNCVA=; b=M+gBc4r4ld+4SusH9m8TMuvR7lyA7qzFfaZkL4HKezMMX0sFO6XDy4Mn9KflITijeP 84+M1a+ixc5sb04rg3yVmkqr6y0V3Fx9C/l6BLLO6wVWvG7j6pyq57Lg/m3pJtifVH8x nD83uKh+TTmB4kFPJCoeihDta+ZhsAtm8sBTx+Alp7+CvoOn/UbDHeM0y6+b/gxQDbc0 UWwYNcV0CtU2pEepjZapfBP9+7GstehVMb1KQVAowOaBHiULy58CHhKrH+M2VdtW6JmY 5PyXpfCrK/DmxRbFkcXe9lL3ZEMjg9T+zjqe+bgSvchrqLl5ui/n7k+9UlHQNbbSl0rk di5g== X-Gm-Message-State: ABy/qLbYoRSQ7jWYfGA0JyiS2CUaILS6rntFc3kcdVrwIUgrorFUGdqE wuPM8u6dDCANNL62uHdUiGmlIA== X-Google-Smtp-Source: APBJJlGle8sGTFCIz3U+SEQARZPGtcsatZf97BgxEHL69WWMCLjqGjoS5fsqRbwtJvxWPn7Q2rRfzw== X-Received: by 2002:a17:902:cecd:b0:1b8:70d2:eb3a with SMTP id d13-20020a170902cecd00b001b870d2eb3amr12272091plg.0.1690192445165; Mon, 24 Jul 2023 02:54:05 -0700 (PDT) From: Qi Zheng To: akpm@linux-foundation.org, david@fromorbit.com, tkhai@ya.ru, vbabka@suse.cz, roman.gushchin@linux.dev, djwong@kernel.org, brauner@kernel.org, paulmck@kernel.org, tytso@mit.edu, steven.price@arm.com, cel@kernel.org, senozhatsky@chromium.org, yujie.liu@intel.com, gregkh@linuxfoundation.org, muchun.song@linux.dev Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org, x86@kernel.org, kvm@vger.kernel.org, xen-devel@lists.xenproject.org, linux-erofs@lists.ozlabs.org, linux-f2fs-devel@lists.sourceforge.net, cluster-devel@redhat.com, linux-nfs@vger.kernel.org, linux-mtd@lists.infradead.org, rcu@vger.kernel.org, netdev@vger.kernel.org, dri-devel@lists.freedesktop.org, linux-arm-msm@vger.kernel.org, dm-devel@redhat.com, linux-raid@vger.kernel.org, linux-bcache@vger.kernel.org, virtualization@lists.linux-foundation.org, linux-fsdevel@vger.kernel.org, linux-ext4@vger.kernel.org, linux-xfs@vger.kernel.org, linux-btrfs@vger.kernel.org, Qi Zheng Subject: [PATCH v2 45/47] mm: shrinker: make memcg slab shrink lockless Date: Mon, 24 Jul 2023 17:43:52 +0800 Message-Id: <20230724094354.90817-46-zhengqi.arch@bytedance.com> X-Mailer: git-send-email 2.24.3 (Apple Git-128) In-Reply-To: <20230724094354.90817-1-zhengqi.arch@bytedance.com> References: <20230724094354.90817-1-zhengqi.arch@bytedance.com> MIME-Version: 1.0 Like global slab shrink, this commit also uses refcount+RCU method to make memcg slab shrink lockless. Use the following script to do slab shrink stress test: ``` DIR="/root/shrinker/memcg/mnt" do_create() { mkdir -p /sys/fs/cgroup/memory/test echo 4G > /sys/fs/cgroup/memory/test/memory.limit_in_bytes for i in `seq 0 $1`; do mkdir -p /sys/fs/cgroup/memory/test/$i; echo $$ > /sys/fs/cgroup/memory/test/$i/cgroup.procs; mkdir -p $DIR/$i; done } do_mount() { for i in `seq $1 $2`; do mount -t tmpfs $i $DIR/$i; done } do_touch() { for i in `seq $1 $2`; do echo $$ > /sys/fs/cgroup/memory/test/$i/cgroup.procs; dd if=/dev/zero of=$DIR/$i/file$i bs=1M count=1 & done } case "$1" in touch) do_touch $2 $3 ;; test) do_create 4000 do_mount 0 4000 do_touch 0 3000 ;; *) exit 1 ;; esac ``` Save the above script, then run test and touch commands. Then we can use the following perf command to view hotspots: perf top -U -F 999 1) Before applying this patchset: 40.44% [kernel] [k] down_read_trylock 17.59% [kernel] [k] up_read 13.64% [kernel] [k] pv_native_safe_halt 11.90% [kernel] [k] shrink_slab 8.21% [kernel] [k] idr_find 2.71% [kernel] [k] _find_next_bit 1.36% [kernel] [k] shrink_node 0.81% [kernel] [k] shrink_lruvec 0.80% [kernel] [k] __radix_tree_lookup 0.50% [kernel] [k] do_shrink_slab 0.21% [kernel] [k] list_lru_count_one 0.16% [kernel] [k] mem_cgroup_iter 2) After applying this patchset: 60.17% [kernel] [k] shrink_slab 20.42% [kernel] [k] pv_native_safe_halt 3.03% [kernel] [k] do_shrink_slab 2.73% [kernel] [k] shrink_node 2.27% [kernel] [k] shrink_lruvec 2.00% [kernel] [k] __rcu_read_unlock 1.92% [kernel] [k] mem_cgroup_iter 0.98% [kernel] [k] __rcu_read_lock 0.91% [kernel] [k] osq_lock 0.63% [kernel] [k] mem_cgroup_calculate_protection 0.55% [kernel] [k] shrinker_put 0.46% [kernel] [k] list_lru_count_one We can see that the first perf hotspot becomes shrink_slab, which is what we expect. Signed-off-by: Qi Zheng --- mm/shrinker.c | 121 +++++++++++++++++++++++++++----------------- mm/shrinker_debug.c | 4 +- 2 files changed, 76 insertions(+), 49 deletions(-) diff --git a/mm/shrinker.c b/mm/shrinker.c index 8e3334749552..744361afd520 100644 --- a/mm/shrinker.c +++ b/mm/shrinker.c @@ -107,6 +107,12 @@ static struct shrinker_info *shrinker_info_protected(struct mem_cgroup *memcg, lockdep_is_held(&shrinker_rwsem)); } +static struct shrinker_info *shrinker_info_rcu(struct mem_cgroup *memcg, + int nid) +{ + return rcu_dereference(memcg->nodeinfo[nid]->shrinker_info); +} + static int expand_one_shrinker_info(struct mem_cgroup *memcg, int new_size, int old_size, int new_nr_max) { @@ -152,11 +158,11 @@ static int expand_shrinker_info(int new_id) int new_size, old_size = 0; struct mem_cgroup *memcg; + down_write(&shrinker_rwsem); + if (!root_mem_cgroup) goto out; - lockdep_assert_held(&shrinker_rwsem); - new_size = shrinker_unit_size(new_nr_max); old_size = shrinker_unit_size(shrinker_nr_max); @@ -173,6 +179,8 @@ static int expand_shrinker_info(int new_id) if (!ret) shrinker_nr_max = new_nr_max; + up_write(&shrinker_rwsem); + return ret; } @@ -198,7 +206,7 @@ void set_shrinker_bit(struct mem_cgroup *memcg, int nid, int shrinker_id) struct shrinker_info_unit *unit; rcu_read_lock(); - info = rcu_dereference(memcg->nodeinfo[nid]->shrinker_info); + info = shrinker_info_rcu(memcg, nid); unit = info->unit[shriner_id_to_index(shrinker_id)]; if (!WARN_ON_ONCE(shrinker_id >= info->map_nr_max)) { /* Pairs with smp mb in shrink_slab() */ @@ -211,39 +219,40 @@ void set_shrinker_bit(struct mem_cgroup *memcg, int nid, int shrinker_id) static DEFINE_IDR(shrinker_idr); -static int prealloc_memcg_shrinker(struct shrinker *shrinker) +static int shrinker_memcg_alloc(struct shrinker *shrinker) { - int id, ret = -ENOMEM; + int id; if (mem_cgroup_disabled()) return -ENOSYS; - down_write(&shrinker_rwsem); - /* This may call shrinker, so it must use down_read_trylock() */ - id = idr_alloc(&shrinker_idr, shrinker, 0, 0, GFP_KERNEL); + spin_lock(&shrinker_lock); + id = idr_alloc(&shrinker_idr, shrinker, 0, 0, GFP_ATOMIC); + spin_unlock(&shrinker_lock); + if (id < 0) - goto unlock; + return -ENOMEM; if (id >= shrinker_nr_max) { if (expand_shrinker_info(id)) { + spin_lock(&shrinker_lock); idr_remove(&shrinker_idr, id); - goto unlock; + spin_unlock(&shrinker_lock); + return -ENOMEM; } } shrinker->id = id; - ret = 0; -unlock: - up_write(&shrinker_rwsem); - return ret; + + return 0; } -static void unregister_memcg_shrinker(struct shrinker *shrinker) +static void shrinker_memcg_remove(struct shrinker *shrinker) { int id = shrinker->id; BUG_ON(id < 0); - lockdep_assert_held(&shrinker_rwsem); + lockdep_assert_held(&shrinker_lock); idr_remove(&shrinker_idr, id); } @@ -253,10 +262,15 @@ static long xchg_nr_deferred_memcg(int nid, struct shrinker *shrinker, { struct shrinker_info *info; struct shrinker_info_unit *unit; + long nr_deferred; - info = shrinker_info_protected(memcg, nid); + rcu_read_lock(); + info = shrinker_info_rcu(memcg, nid); unit = info->unit[shriner_id_to_index(shrinker->id)]; - return atomic_long_xchg(&unit->nr_deferred[shriner_id_to_offset(shrinker->id)], 0); + nr_deferred = atomic_long_xchg(&unit->nr_deferred[shriner_id_to_offset(shrinker->id)], 0); + rcu_read_unlock(); + + return nr_deferred; } static long add_nr_deferred_memcg(long nr, int nid, struct shrinker *shrinker, @@ -264,10 +278,16 @@ static long add_nr_deferred_memcg(long nr, int nid, struct shrinker *shrinker, { struct shrinker_info *info; struct shrinker_info_unit *unit; + long nr_deferred; - info = shrinker_info_protected(memcg, nid); + rcu_read_lock(); + info = shrinker_info_rcu(memcg, nid); unit = info->unit[shriner_id_to_index(shrinker->id)]; - return atomic_long_add_return(nr, &unit->nr_deferred[shriner_id_to_offset(shrinker->id)]); + nr_deferred = + atomic_long_add_return(nr, &unit->nr_deferred[shriner_id_to_offset(shrinker->id)]); + rcu_read_unlock(); + + return nr_deferred; } void reparent_shrinker_deferred(struct mem_cgroup *memcg) @@ -299,12 +319,12 @@ void reparent_shrinker_deferred(struct mem_cgroup *memcg) up_read(&shrinker_rwsem); } #else -static int prealloc_memcg_shrinker(struct shrinker *shrinker) +static int shrinker_memcg_alloc(struct shrinker *shrinker) { return -ENOSYS; } -static void unregister_memcg_shrinker(struct shrinker *shrinker) +static void shrinker_memcg_remove(struct shrinker *shrinker) { } @@ -458,6 +478,8 @@ void shrinker_put(struct shrinker *shrinker) if (refcount_dec_and_test(&shrinker->refcount)) { spin_lock(&shrinker_lock); list_del_rcu(&shrinker->list); + if (shrinker->flags & SHRINKER_MEMCG_AWARE) + shrinker_memcg_remove(shrinker); spin_unlock(&shrinker_lock); kfree(shrinker->nr_deferred); @@ -476,18 +498,23 @@ static unsigned long shrink_slab_memcg(gfp_t gfp_mask, int nid, if (!mem_cgroup_online(memcg)) return 0; - if (!down_read_trylock(&shrinker_rwsem)) - return 0; - - info = shrinker_info_protected(memcg, nid); +again: + rcu_read_lock(); + info = shrinker_info_rcu(memcg, nid); if (unlikely(!info)) goto unlock; - for (; index < shriner_id_to_index(info->map_nr_max); index++) { + if (index < shriner_id_to_index(info->map_nr_max)) { struct shrinker_info_unit *unit; unit = info->unit[index]; + /* + * The shrinker_info_unit will not be freed, so we can + * safely release the RCU lock here. + */ + rcu_read_unlock(); + for_each_set_bit(offset, unit->map, SHRINKER_UNIT_BITS) { struct shrink_control sc = { .gfp_mask = gfp_mask, @@ -497,13 +524,14 @@ static unsigned long shrink_slab_memcg(gfp_t gfp_mask, int nid, struct shrinker *shrinker; int shrinker_id = calc_shrinker_id(index, offset); + rcu_read_lock(); shrinker = idr_find(&shrinker_idr, shrinker_id); - if (unlikely(!shrinker || - !READ_ONCE(shrinker->registered))) { - if (!shrinker) - clear_bit(offset, unit->map); + if (unlikely(!shrinker || !shrinker_try_get(shrinker))) { + clear_bit(offset, unit->map); + rcu_read_unlock(); continue; } + rcu_read_unlock(); /* Call non-slab shrinkers even though kmem is disabled */ if (!memcg_kmem_online() && @@ -536,15 +564,20 @@ static unsigned long shrink_slab_memcg(gfp_t gfp_mask, int nid, set_shrinker_bit(memcg, nid, shrinker_id); } freed += ret; - - if (rwsem_is_contended(&shrinker_rwsem)) { - freed = freed ? : 1; - goto unlock; - } + shrinker_put(shrinker); } + + /* + * We have already exited the read-side of rcu critical section + * before calling do_shrink_slab(), the shrinker_info may be + * released in expand_one_shrinker_info(), so reacquire the + * shrinker_info. + */ + index++; + goto again; } unlock: - up_read(&shrinker_rwsem); + rcu_read_unlock(); return freed; } #else /* !CONFIG_MEMCG */ @@ -652,7 +685,7 @@ struct shrinker *shrinker_alloc(unsigned int flags, const char *fmt, ...) shrinker->flags = flags; if (flags & SHRINKER_MEMCG_AWARE) { - err = prealloc_memcg_shrinker(shrinker); + err = shrinker_memcg_alloc(shrinker); if (err == -ENOSYS) shrinker->flags &= ~SHRINKER_MEMCG_AWARE; else if (err == 0) @@ -697,9 +730,9 @@ void shrinker_free_non_registered(struct shrinker *shrinker) shrinker->name = NULL; #endif if (shrinker->flags & SHRINKER_MEMCG_AWARE) { - down_write(&shrinker_rwsem); - unregister_memcg_shrinker(shrinker); - up_write(&shrinker_rwsem); + spin_lock(&shrinker_lock); + shrinker_memcg_remove(shrinker); + spin_unlock(&shrinker_lock); } kfree(shrinker->nr_deferred); @@ -731,13 +764,7 @@ void shrinker_unregister(struct shrinker *shrinker) return; WRITE_ONCE(shrinker->registered, false); - - down_write(&shrinker_rwsem); - if (shrinker->flags & SHRINKER_MEMCG_AWARE) - unregister_memcg_shrinker(shrinker); debugfs_entry = shrinker_debugfs_detach(shrinker, &debugfs_id); - up_write(&shrinker_rwsem); - shrinker_debugfs_remove(debugfs_entry, debugfs_id); shrinker_put(shrinker); diff --git a/mm/shrinker_debug.c b/mm/shrinker_debug.c index c5573066adbf..badda35464c3 100644 --- a/mm/shrinker_debug.c +++ b/mm/shrinker_debug.c @@ -246,8 +246,7 @@ struct dentry *shrinker_debugfs_detach(struct shrinker *shrinker, { struct dentry *entry = shrinker->debugfs_entry; - lockdep_assert_held(&shrinker_rwsem); - + down_write(&shrinker_rwsem); kfree_const(shrinker->name); shrinker->name = NULL; @@ -258,6 +257,7 @@ struct dentry *shrinker_debugfs_detach(struct shrinker *shrinker, */ smp_wmb(); shrinker->debugfs_entry = NULL; + up_write(&shrinker_rwsem); return entry; }