From mboxrd@z Thu Jan 1 00:00:00 1970 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 0CFE0C10F16 for ; Mon, 6 May 2024 14:02:33 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 75F316B0083; Mon, 6 May 2024 10:02:33 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 70F576B0087; Mon, 6 May 2024 10:02:33 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 5D7286B0088; Mon, 6 May 2024 10:02:33 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0011.hostedemail.com [216.40.44.11]) by kanga.kvack.org (Postfix) with ESMTP id 40F596B0083 for ; Mon, 6 May 2024 10:02:33 -0400 (EDT) Received: from smtpin30.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay02.hostedemail.com (Postfix) with ESMTP id B45FA120AD0 for ; Mon, 6 May 2024 14:02:32 +0000 (UTC) X-FDA: 82088136144.30.ECB7650 Received: from dggsgout11.his.huawei.com (dggsgout11.his.huawei.com [45.249.212.51]) by imf10.hostedemail.com (Postfix) with ESMTP id 62F1DC002D for ; Mon, 6 May 2024 14:02:26 +0000 (UTC) Authentication-Results: imf10.hostedemail.com; dkim=none; dmarc=none; spf=pass (imf10.hostedemail.com: domain of yi.zhang@huaweicloud.com designates 45.249.212.51 as permitted sender) smtp.mailfrom=yi.zhang@huaweicloud.com ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1715004150; 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=tzirr46AhQV2sydmBjvhNQR+wDT1i0w3GQuUVDio51E=; b=OhMmcWZSEGLrFRp9SZ0V7V9dbpgUduA46JTYCp4xczdk91o/b85xaYT02sRCEhF9kf6X0Z wwK+TmuFL2Vxnh2KKYCxGd/gCfzR8V7/B/hIJsQy9xAbBRF1iViMljUswnnsbuq9rwhaXY Gq3y0HbQMraIIiZLUrpsOQ3/owQQJV0= ARC-Authentication-Results: i=1; imf10.hostedemail.com; dkim=none; dmarc=none; spf=pass (imf10.hostedemail.com: domain of yi.zhang@huaweicloud.com designates 45.249.212.51 as permitted sender) smtp.mailfrom=yi.zhang@huaweicloud.com ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1715004150; a=rsa-sha256; cv=none; b=f3EcdFpT2KKNSEUm0nCKXH1Uzjmxm05TM5jieR1SkU3zOuKsA7kUBSJYqnDTs+lQ7+Zj42 a4j2Mk0gGu5kd5Tik6CmohsgMOppGagwpGfj47HfVxSLEWGHlG1szPwxaJZl21LX1BZ03b t5TVM/ImbocnVb2Qaysyn1E4NYUGZMk= Received: from mail.maildlp.com (unknown [172.19.163.235]) by dggsgout11.his.huawei.com (SkyGuard) with ESMTP id 4VY34g0dyXz4f3lfT for ; Mon, 6 May 2024 22:02:11 +0800 (CST) Received: from mail02.huawei.com (unknown [10.116.40.75]) by mail.maildlp.com (Postfix) with ESMTP id D23871A0572 for ; Mon, 6 May 2024 22:02:20 +0800 (CST) Received: from [10.174.179.80] (unknown [10.174.179.80]) by APP2 (Coremail) with SMTP id Syh0CgDnCg7q4jhmTvnUMA--.48428S3; Mon, 06 May 2024 22:02:20 +0800 (CST) Subject: Re: [RFC PATCH v4 33/34] ext4: don't mark IOMAP_F_DIRTY for buffer write To: Dave Chinner Cc: linux-ext4@vger.kernel.org, linux-fsdevel@vger.kernel.org, linux-mm@kvack.org, linux-kernel@vger.kernel.org, tytso@mit.edu, adilger.kernel@dilger.ca, jack@suse.cz, ritesh.list@gmail.com, hch@infradead.org, djwong@kernel.org, willy@infradead.org, zokeefe@google.com, yi.zhang@huawei.com, chengzhihao1@huawei.com, yukuai3@huawei.com, wangkefeng.wang@huawei.com References: <20240410142948.2817554-1-yi.zhang@huaweicloud.com> <20240410150313.2820364-5-yi.zhang@huaweicloud.com> From: Zhang Yi Message-ID: <349d0d44-2ffc-6177-cf00-db7a3f48b60d@huaweicloud.com> Date: Mon, 6 May 2024 22:02:18 +0800 User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:78.0) Gecko/20100101 Thunderbird/78.12.0 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 7bit X-CM-TRANSID:Syh0CgDnCg7q4jhmTvnUMA--.48428S3 X-Coremail-Antispam: 1UD129KBjvJXoWxur1DAF1DKw4rZr1kAF47Jwb_yoW5ArWDpr Z7Ka1rX3WkXry7ur4IqF47ZrZ7tayUKr4DGrW8tw12y398Gw1xtF1FgFyrAF98Gr1fWw42 q3Wjqr97Ww17AaDanT9S1TB71UUUUUUqnTZGkaVYY2UrUUUUjbIjqfuFe4nvWSU5nxnvy2 9KBjDU0xBIdaVrnRJUUUvab4IE77IF4wAFF20E14v26ryj6rWUM7CY07I20VC2zVCF04k2 6cxKx2IYs7xG6rWj6s0DM7CIcVAFz4kK6r1j6r18M28lY4IEw2IIxxk0rwA2F7IY1VAKz4 vEj48ve4kI8wA2z4x0Y4vE2Ix0cI8IcVAFwI0_Xr0_Ar1l84ACjcxK6xIIjxv20xvEc7Cj xVAFwI0_Cr0_Gr1UM28EF7xvwVC2z280aVAFwI0_GcCE3s1l84ACjcxK6I8E87Iv6xkF7I 0E14v26rxl6s0DM2AIxVAIcxkEcVAq07x20xvEncxIr21l5I8CrVACY4xI64kE6c02F40E x7xfMcIj6xIIjxv20xvE14v26r1j6r18McIj6I8E87Iv67AKxVWUJVW8JwAm72CE4IkC6x 0Yz7v_Jr0_Gr1lF7xvr2IY64vIr41lFIxGxcIEc7CjxVA2Y2ka0xkIwI1lc7I2V7IY0VAS 07AlzVAYIcxG8wCF04k20xvY0x0EwIxGrwCFx2IqxVCFs4IE7xkEbVWUJVW8JwC20s026c 02F40E14v26r1j6r18MI8I3I0E7480Y4vE14v26r106r1rMI8E67AF67kF1VAFwI0_GFv_ WrylIxkGc2Ij64vIr41lIxAIcVC0I7IYx2IY67AKxVWUJVWUCwCI42IY6xIIjxv20xvEc7 CjxVAFwI0_Gr0_Cr1lIxAIcVCF04k26cxKx2IYs7xG6rWUJVWrZr1UMIIF0xvEx4A2jsIE 14v26r1j6r4UMIIF0xvEx4A2jsIEc7CjxVAFwI0_Gr1j6F4UJbIYCTnIWIevJa73UjIFyT uYvjxUFDGOUUUUU X-CM-SenderInfo: d1lo6xhdqjqx5xdzvxpfor3voofrz/ X-Rspamd-Server: rspam01 X-Stat-Signature: no1kteccxnuw391eu85u9yjpbsqysqb4 X-Rspam-User: X-Rspamd-Queue-Id: 62F1DC002D X-HE-Tag: 1715004146-300763 X-HE-Meta: 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 3AyspOr7 BMu7X0INQWv6jXscCk2rXC3YCQfVKSQ4Jw4dVUeBgc5yU9nZulqOYVdts225Zj/ADoPm3jtu7Roz4N+bvfbuGy1JoFEP80s5JcQPOhUTJstbz4FEy83k5KhEmUEOuOCf/V+CWNMiJDA1QZt0JWHGLZMxY2P+FZcdNkttvYU+ABV7/AFXokp6I3FK9M8Eg51wV3CBVJUeR+Hje3UsqbTa5Bfy5Pa1kC7o9ClhOxGIrZaKDRc6JsaSRh3qPEWW1lbI/2OQoaZ0hiuJeErw= 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: On 2024/5/1 17:27, Dave Chinner wrote: > On Wed, Apr 10, 2024 at 11:03:12PM +0800, Zhang Yi wrote: >> From: Zhang Yi >> >> The data sync dirty check in ext4_inode_datasync_dirty() is expansive >> since jbd2_transaction_committed() holds journal->j_state lock when >> journal is enabled, it costs a lot in high-concurrency iomap buffered >> read/write paths, but we never check IOMAP_F_DIRTY in these cases, so >> let's check it only in swap file, dax and direct IO cases. Tested by >> Unixbench on 100GB ramdisk: >> >> ./Run -c 128 -i 10 fstime fsbuffer fsdisk >> >> == without this patch == >> 128 CPUs in system; running 128 parallel copies of tests >> >> File Copy 1024 bufsize 2000 maxblocks 6332521.0 KBps >> File Copy 256 bufsize 500 maxblocks 1639726.0 KBps >> File Copy 4096 bufsize 8000 maxblocks 24018572.0 KBps >> >> == with this patch == >> 128 CPUs in system; running 128 parallel copies of tests >> >> File Copy 1024 bufsize 2000 maxblocks 49229257.0 KBps >> File Copy 256 bufsize 500 maxblocks 24057510.0 KBps >> File Copy 4096 bufsize 8000 maxblocks 75704437.0 KBps >> >> Signed-off-by: Zhang Yi >> --- >> fs/ext4/inode.c | 10 +++++++--- >> 1 file changed, 7 insertions(+), 3 deletions(-) >> >> diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c >> index 1cb219d347af..269503749ef5 100644 >> --- a/fs/ext4/inode.c >> +++ b/fs/ext4/inode.c >> @@ -3281,9 +3281,13 @@ static void ext4_set_iomap(struct inode *inode, struct iomap *iomap, >> * there is no other metadata changes being made or are pending. >> */ >> iomap->flags = 0; >> - if (ext4_inode_datasync_dirty(inode) || >> - offset + length > i_size_read(inode)) >> - iomap->flags |= IOMAP_F_DIRTY; >> + if ((flags & (IOMAP_DAX | IOMAP_REPORT)) || >> + ((flags & (IOMAP_WRITE | IOMAP_DIRECT)) == >> + (IOMAP_WRITE | IOMAP_DIRECT))) { >> + if (offset + length > i_size_read(inode) || >> + ext4_inode_datasync_dirty(inode)) >> + iomap->flags |= IOMAP_F_DIRTY; >> + } > > NACK. This just adds a nasty landmine that anyone working on the > iomap infrastructure can step on. i.e. any time we add a new check > for IOMAP_F_DIRTY in the generic infrastructure, ext4 is going to > break because it won't set the IOMAP_F_DIRTY flag correctly. > > If checking an inode is dirty is expensive on ext4, then make it > less expensive and everyone will benefit. > > /me goes and looks at jbd2_transaction_committed() > > Oh, it it's just a sequence number comparison, and it needs a lock > because it has to dereference the running/committed transactions > structures to get the current sequence numbers. Why not just store > the commiting/running transaction tids in the journal_t, and then > you can sample them without needing any locking and the whole > ext4_inode_datasync_dirty() scalability problem goes away... > Indeed, it could be useful, and it seems could also simplify many other jbd2 processes. Thanks, Yi.