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 X-Spam-Level: * X-Spam-Status: No, score=1.7 required=3.0 tests=DKIM_SIGNED,DKIM_VALID, DKIM_VALID_AU,MAILING_LIST_MULTI,SPF_PASS,T_DKIMWL_WL_HIGH, UNWANTED_LANGUAGE_BODY 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 90E51C3279B for ; Fri, 6 Jul 2018 23:35:41 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 40FDA22B46 for ; Fri, 6 Jul 2018 23:35:41 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (1024-bit key) header.d=kernel.org header.i=@kernel.org header.b="oYbRIQVk" DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 40FDA22B46 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=kernel.org Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=linux-kernel-owner@vger.kernel.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933516AbeGFXfi (ORCPT ); Fri, 6 Jul 2018 19:35:38 -0400 Received: from mail.kernel.org ([198.145.29.99]:38154 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932774AbeGFXfg (ORCPT ); Fri, 6 Jul 2018 19:35:36 -0400 Received: from [192.168.0.101] (unknown [49.77.239.198]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPSA id 71D7322B40; Fri, 6 Jul 2018 23:35:34 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=default; t=1530920135; bh=W/YLi+rWc9tLRut1WfqxOyoEeWdT6fIBTBXTY5JGbBc=; h=Subject:To:Cc:References:From:Date:In-Reply-To:From; b=oYbRIQVk4BU+utMuf6+zcn00QnPws/RbAyFV1GUcCaGz0wjl8/wtfbRnF5aVOu/ZS Zq40eU2MzjH3db7+4dpf1nf0pDMEdGRVIi7dgLUQOVGXl1zCTnh2bI8YFgCBsCnYB3 fjuFVio7aUSAedE0rxfQCC9qP32ECmSXqhrMYfq4= Subject: Re: [PATCH] f2fs: split discard command in prior to block layer To: Jaegeuk Kim Cc: linux-f2fs-devel@lists.sourceforge.net, linux-kernel@vger.kernel.org, Chao Yu References: <20180704153746.2510-1-chao@kernel.org> <20180706224533.GC77984@jaegeuk-macbookpro.roam.corp.google.com> From: Chao Yu Message-ID: Date: Sat, 7 Jul 2018 07:35:31 +0800 User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:52.0) Gecko/20100101 Thunderbird/52.8.0 MIME-Version: 1.0 In-Reply-To: <20180706224533.GC77984@jaegeuk-macbookpro.roam.corp.google.com> Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Jaegeuk, On 2018/7/7 6:45, Jaegeuk Kim wrote: > On 07/04, Chao Yu wrote: >> From: Chao Yu >> >> Some devices has small max_{hw,}discard_sectors, so that in >> __blkdev_issue_discard(), one big size discard bio can be split >> into multiple small size discard bios, result in heavy load in IO >> scheduler and device, which can hang other sync IO for long time. >> >> Now, f2fs is trying to control discard commands more elaboratively, >> in order to make less conflict in between discard IO and user IO >> to enhance application's performance, so in this patch, we will >> split discard bio in f2fs in prior to in block layer to reduce >> issuing multiple discard bios in a short time. > > Hi Chao, > > In terms of # of candidates, can we control this when actually issuing > the discard commands? IIUC, you mean once we pick one discard entry in rbtree, if max_{hw,}discard_sectors is smaller than size of this discard, then we can split it into smaller ones by discard_sectors, and just issue one or partials of them? Thanks, > > Thanks, > >> >> Signed-off-by: Chao Yu >> --- >> fs/f2fs/f2fs.h | 13 ++++++------- >> fs/f2fs/segment.c | 25 ++++++++++++++++++++++--- >> 2 files changed, 28 insertions(+), 10 deletions(-) >> >> diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h >> index a9da5a089cb4..a09d2b2d9520 100644 >> --- a/fs/f2fs/f2fs.h >> +++ b/fs/f2fs/f2fs.h >> @@ -178,7 +178,6 @@ enum { >> >> #define MAX_DISCARD_BLOCKS(sbi) BLKS_PER_SEC(sbi) >> #define DEF_MAX_DISCARD_REQUEST 8 /* issue 8 discards per round */ >> -#define DEF_MAX_DISCARD_LEN 512 /* Max. 2MB per discard */ >> #define DEF_MIN_DISCARD_ISSUE_TIME 50 /* 50 ms, if exists */ >> #define DEF_MID_DISCARD_ISSUE_TIME 500 /* 500 ms, if device busy */ >> #define DEF_MAX_DISCARD_ISSUE_TIME 60000 /* 60 s, if no candidates */ >> @@ -701,22 +700,22 @@ static inline void set_extent_info(struct extent_info *ei, unsigned int fofs, >> } >> >> static inline bool __is_discard_mergeable(struct discard_info *back, >> - struct discard_info *front) >> + struct discard_info *front, unsigned int max_len) >> { >> return (back->lstart + back->len == front->lstart) && >> - (back->len + front->len < DEF_MAX_DISCARD_LEN); >> + (back->len + front->len <= max_len); >> } >> >> static inline bool __is_discard_back_mergeable(struct discard_info *cur, >> - struct discard_info *back) >> + struct discard_info *back, unsigned int max_len) >> { >> - return __is_discard_mergeable(back, cur); >> + return __is_discard_mergeable(back, cur, max_len); >> } >> >> static inline bool __is_discard_front_mergeable(struct discard_info *cur, >> - struct discard_info *front) >> + struct discard_info *front, unsigned int max_len) >> { >> - return __is_discard_mergeable(cur, front); >> + return __is_discard_mergeable(cur, front, max_len); >> } >> >> static inline bool __is_extent_mergeable(struct extent_info *back, >> diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c >> index 4648561e2bfd..8e417a12684d 100644 >> --- a/fs/f2fs/segment.c >> +++ b/fs/f2fs/segment.c >> @@ -1086,6 +1086,9 @@ static void __update_discard_tree_range(struct f2fs_sb_info *sbi, >> struct discard_cmd *dc; >> struct discard_info di = {0}; >> struct rb_node **insert_p = NULL, *insert_parent = NULL; >> + struct request_queue *q = bdev_get_queue(bdev); >> + unsigned int max_discard_blocks = >> + SECTOR_TO_BLOCK(q->limits.max_discard_sectors); >> block_t end = lstart + len; >> >> mutex_lock(&dcc->cmd_lock); >> @@ -1129,7 +1132,8 @@ static void __update_discard_tree_range(struct f2fs_sb_info *sbi, >> >> if (prev_dc && prev_dc->state == D_PREP && >> prev_dc->bdev == bdev && >> - __is_discard_back_mergeable(&di, &prev_dc->di)) { >> + __is_discard_back_mergeable(&di, &prev_dc->di, >> + max_discard_blocks)) { >> prev_dc->di.len += di.len; >> dcc->undiscard_blks += di.len; >> __relocate_discard_cmd(dcc, prev_dc); >> @@ -1140,7 +1144,8 @@ static void __update_discard_tree_range(struct f2fs_sb_info *sbi, >> >> if (next_dc && next_dc->state == D_PREP && >> next_dc->bdev == bdev && >> - __is_discard_front_mergeable(&di, &next_dc->di)) { >> + __is_discard_front_mergeable(&di, &next_dc->di, >> + max_discard_blocks)) { >> next_dc->di.lstart = di.lstart; >> next_dc->di.len += di.len; >> next_dc->di.start = di.start; >> @@ -1170,7 +1175,11 @@ static void __update_discard_tree_range(struct f2fs_sb_info *sbi, >> static int __queue_discard_cmd(struct f2fs_sb_info *sbi, >> struct block_device *bdev, block_t blkstart, block_t blklen) >> { >> + struct request_queue *q = bdev_get_queue(bdev); >> + unsigned int max_discard_blocks = >> + SECTOR_TO_BLOCK(q->limits.max_discard_sectors); >> block_t lblkstart = blkstart; >> + block_t total_len = blklen; >> >> trace_f2fs_queue_discard(bdev, blkstart, blklen); >> >> @@ -1179,7 +1188,17 @@ static int __queue_discard_cmd(struct f2fs_sb_info *sbi, >> >> blkstart -= FDEV(devi).start_blk; >> } >> - __update_discard_tree_range(sbi, bdev, lblkstart, blkstart, blklen); >> + >> + while (total_len) { >> + if (blklen > max_discard_blocks) >> + blklen = max_discard_blocks; >> + __update_discard_tree_range(sbi, bdev, lblkstart, >> + blkstart, blklen); >> + lblkstart += blklen; >> + blkstart += blklen; >> + total_len -= blklen; >> + blklen = total_len; >> + } >> return 0; >> } >> >> -- >> 2.16.2.17.g38e79b1fd