From patchwork Fri Feb 16 08:55:05 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Chengming Zhou X-Patchwork-Id: 13559703 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 kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by smtp.lore.kernel.org (Postfix) with ESMTP id 47046C48BEC for ; Fri, 16 Feb 2024 08:55:41 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id C79898D0014; Fri, 16 Feb 2024 03:55:40 -0500 (EST) Received: by kanga.kvack.org (Postfix, from userid 40) id C03C58D0001; Fri, 16 Feb 2024 03:55:40 -0500 (EST) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id AA1D98D0014; Fri, 16 Feb 2024 03:55:40 -0500 (EST) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0013.hostedemail.com [216.40.44.13]) by kanga.kvack.org (Postfix) with ESMTP id 8FA218D0001 for ; Fri, 16 Feb 2024 03:55:40 -0500 (EST) Received: from smtpin06.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay07.hostedemail.com (Postfix) with ESMTP id 5D02A1603B0 for ; Fri, 16 Feb 2024 08:55:40 +0000 (UTC) X-FDA: 81797058840.06.012CBDD Received: from out-184.mta0.migadu.com (out-184.mta0.migadu.com [91.218.175.184]) by imf10.hostedemail.com (Postfix) with ESMTP id 90BA2C0009 for ; Fri, 16 Feb 2024 08:55:38 +0000 (UTC) Authentication-Results: imf10.hostedemail.com; dkim=none; dmarc=fail reason="SPF not aligned (relaxed), No valid DKIM" header.from=bytedance.com (policy=quarantine); spf=pass (imf10.hostedemail.com: domain of chengming.zhou@linux.dev designates 91.218.175.184 as permitted sender) smtp.mailfrom=chengming.zhou@linux.dev ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1708073738; a=rsa-sha256; cv=none; b=HgffLkdU/RuGZph3PvoJsJAwXhInkIIEbWaO9AgrTCZKk0LWmFSK6lzfsCEOj81SpDMPl+ qEDyi577tJGed1vZBOIQEiTG7DSqO7MoeCXktIUPYYkWpCLO9SV9hfDwjEhQ+zUCrOZl+2 1QmV+liqgtwQj23C7zYMPhA7i0LY7uc= ARC-Authentication-Results: i=1; imf10.hostedemail.com; dkim=none; dmarc=fail reason="SPF not aligned (relaxed), No valid DKIM" header.from=bytedance.com (policy=quarantine); spf=pass (imf10.hostedemail.com: domain of chengming.zhou@linux.dev designates 91.218.175.184 as permitted sender) smtp.mailfrom=chengming.zhou@linux.dev ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1708073738; h=from:from:sender:reply-to:subject:subject:date:date: message-id:message-id:to:to:cc:cc:mime-version:mime-version: content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=l7jMbQSDWATokAiFpRFYHIQIVCg9/U265+bQ86tP450=; b=FG9oRQ5X3CpwJX+tRBGsV8Qo+OK2QOKb++3EsYLNM23ZipEL2JAxt+BFHomk3T1okKS/cY YcF5pQ9pzeQ2pX0YgnAl9fVKbgpXW/UKXJSeSumaa+mOPKMUpgL1NjT/iqjhmUhSAdShV+ To7obQja1udDRy791zPj5plEjjJOFVA= X-Report-Abuse: Please report any abuse attempt to abuse@migadu.com and include these headers. From: Chengming Zhou Date: Fri, 16 Feb 2024 08:55:05 +0000 Subject: [PATCH v3 2/2] mm/zswap: change zswap_pool kref to percpu_ref MIME-Version: 1.0 Message-Id: <20240210-zswap-global-lru-v3-2-200495333595@bytedance.com> References: <20240210-zswap-global-lru-v3-0-200495333595@bytedance.com> In-Reply-To: <20240210-zswap-global-lru-v3-0-200495333595@bytedance.com> To: Johannes Weiner , Yosry Ahmed , Nhat Pham , Andrew Morton Cc: linux-mm@kvack.org, Yosry Ahmed , linux-kernel@vger.kernel.org, Chengming Zhou X-Migadu-Flow: FLOW_OUT X-Rspamd-Server: rspam08 X-Rspamd-Queue-Id: 90BA2C0009 X-Stat-Signature: ertki8yyzq9e4xemnyky8os5r1unsru4 X-Rspamd-Pre-Result: action=add header; module=dmarc; Action set by DMARC X-Rspam-User: X-Rspam: Yes X-HE-Tag: 1708073738-280415 X-HE-Meta: U2FsdGVkX19SKE/xnDUZNPp8i+WWb70HGg8haCy5m9NyvInoXuaHq6mGhYjzi9DbYeFjlDUUDbVsg9toC3w3ZAvkISULxg9RvVwL0f2GRwhYkhthZEvP5F9jPocrZ+YagutRGmpzTgs+jOO2UX0/rM+yURz501bdjw0ltm7mYRmYLuRt6pZpXldL3NmIxbdlES6hpd5i+C2HqPTB7rL7CmmIeJ2y4zCsTgYttZLGGsgWWqoQUPmu4fe6ga3L0CLeG28i8xIVl6FMHGGpyWMheIqIC69ttxppXK+mDJas7ijq4j4RHyKoAAezRQA2oA9IJfSzxt0tmqzC/YlqI3YieiMi3QSN3DVvykuFnwLBEElU6Zyb68hKppocT7oSxhQ3jzsAtO6ZG7QSiwXiP/AtOe9LQgqe6JC2FtconMkM2g2Ga3RUEqfn7RtYpnSt3MQN+hcA9yGOvQaiA1zuMqCsAT5AQ1CuE3Lk6uMbRebciNu/0vGT3TJb89g2Vet3T5pMazIthBLWXgCUNyH5Y6hRAgjX6U3YGeTJ4SLkqlHjWnGm5JQclFLth55GvxacFAsAGskZW13ayTSWR6xUZxDSClYBeZPQbaKBGy17R8/4XVQAJxpxkruJIfbwCGhqQ0SPDKKherAQzAFhj5x22SD/e/1DCrOUAsJelb/2uNodIeq/E5EmcXQ5/YcOGGjvuascxsyv6V2tfzM1Bx1BGmiYvsIMOczGLyZUbeEZ+byGs/W+GGPwBjwHhqsw4MMKeU+/oeWDRvI3pTTeiBrXjtgnXZJ77EuGjys8KnwNORgtfIucx17m+uJ+R3cQxWr2HVWvNJewyvu6H+IahkLi62FQhMDOpG6kwsJzvYgGWmLLw8DiCGIzpDa+PbVihUcSTIzh6b4mN/L29pSclOb5JoIl+GgIBA8lrfF/OaWlEwlsR+S4Ev7w9kw1dVAHFUW0RX9Jrno6C7no/oZbHnqvk9i 0gQV/Gss SBYPVnXOsgpGFOKC+NS28k8WdII9gj4uFikzL+N/1jSv80VbcTecM095+RRGsWH4bU7LNEQRh3LC3MIHy+qeecKMpw56nJUWLGNN8DMjsH+eZ8lgn6zwm2bAF/Qe156L8vO1/E1dMUSnJrSjIs0CV+FjEyirDFWsCWDKpZs/Hc3vUoQ885m8VDC4YCsVAlNGrPcJUTrMRlAhxya0hhODtPmvNW2MnjP8zRig2KTvByE1cS584i8YPbDwA0RM8D+EB1osdS4fr7ayMjAe//BOod5oo2J1XG6zoETp+C1wMtsbBNIdiS7EFQe7Ooumqu4IzM485vLzDBc23Zc2PebBjNAUOId8jxJkilnL+2hU8bTsQPkXjG1+RT2KjWw== 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: List-Subscribe: List-Unsubscribe: All zswap entries will take a reference of zswap_pool when zswap_store(), and drop it when free. Change it to use the percpu_ref is better for scalability performance. Although percpu_ref use a bit more memory which should be ok for our use case, since we almost have only one zswap_pool to be using. The performance gain is for zswap_store/load hotpath. Testing kernel build (32 threads) in tmpfs with memory.max=2GB. (zswap shrinker and writeback enabled with one 50GB swapfile, on a 128 CPUs x86-64 machine, below is the average of 5 runs) mm-unstable zswap-global-lru real 63.20 63.12 user 1061.75 1062.95 sys 268.74 264.44 Signed-off-by: Chengming Zhou Reviewed-by: Nhat Pham --- mm/zswap.c | 36 +++++++++++++++++++++++++++--------- 1 file changed, 27 insertions(+), 9 deletions(-) diff --git a/mm/zswap.c b/mm/zswap.c index d275eb523fc4..961349162997 100644 --- a/mm/zswap.c +++ b/mm/zswap.c @@ -173,7 +173,7 @@ struct crypto_acomp_ctx { struct zswap_pool { struct zpool *zpools[ZSWAP_NR_ZPOOLS]; struct crypto_acomp_ctx __percpu *acomp_ctx; - struct kref kref; + struct percpu_ref ref; struct list_head list; struct work_struct release_work; struct hlist_node node; @@ -305,6 +305,7 @@ static void zswap_update_total_size(void) /********************************* * pool functions **********************************/ +static void __zswap_pool_empty(struct percpu_ref *ref); static struct zswap_pool *zswap_pool_create(char *type, char *compressor) { @@ -358,13 +359,18 @@ static struct zswap_pool *zswap_pool_create(char *type, char *compressor) /* being the current pool takes 1 ref; this func expects the * caller to always add the new pool as the current pool */ - kref_init(&pool->kref); + ret = percpu_ref_init(&pool->ref, __zswap_pool_empty, + PERCPU_REF_ALLOW_REINIT, GFP_KERNEL); + if (ret) + goto ref_fail; INIT_LIST_HEAD(&pool->list); zswap_pool_debug("created", pool); return pool; +ref_fail: + cpuhp_state_remove_instance(CPUHP_MM_ZSWP_POOL_PREPARE, &pool->node); error: if (pool->acomp_ctx) free_percpu(pool->acomp_ctx); @@ -437,8 +443,9 @@ static void __zswap_pool_release(struct work_struct *work) synchronize_rcu(); - /* nobody should have been able to get a kref... */ - WARN_ON(kref_get_unless_zero(&pool->kref)); + /* nobody should have been able to get a ref... */ + WARN_ON(!percpu_ref_is_zero(&pool->ref)); + percpu_ref_exit(&pool->ref); /* pool is now off zswap_pools list and has no references. */ zswap_pool_destroy(pool); @@ -446,11 +453,11 @@ static void __zswap_pool_release(struct work_struct *work) static struct zswap_pool *zswap_pool_current(void); -static void __zswap_pool_empty(struct kref *kref) +static void __zswap_pool_empty(struct percpu_ref *ref) { struct zswap_pool *pool; - pool = container_of(kref, typeof(*pool), kref); + pool = container_of(ref, typeof(*pool), ref); spin_lock(&zswap_pools_lock); @@ -469,12 +476,12 @@ static int __must_check zswap_pool_get(struct zswap_pool *pool) if (!pool) return 0; - return kref_get_unless_zero(&pool->kref); + return percpu_ref_tryget(&pool->ref); } static void zswap_pool_put(struct zswap_pool *pool) { - kref_put(&pool->kref, __zswap_pool_empty); + percpu_ref_put(&pool->ref); } static struct zswap_pool *__zswap_pool_current(void) @@ -604,6 +611,17 @@ static int __zswap_param_set(const char *val, const struct kernel_param *kp, if (!pool) pool = zswap_pool_create(type, compressor); + else { + /* + * Restore the initial ref dropped by percpu_ref_kill() + * when the pool was decommissioned and switch it again + * to percpu mode. + */ + percpu_ref_resurrect(&pool->ref); + + /* Drop the ref from zswap_pool_find_get(). */ + zswap_pool_put(pool); + } if (pool) ret = param_set_charp(s, kp); @@ -642,7 +660,7 @@ static int __zswap_param_set(const char *val, const struct kernel_param *kp, * or the new pool we failed to add */ if (put_pool) - zswap_pool_put(put_pool); + percpu_ref_kill(&put_pool->ref); return ret; }