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=-2.8 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, MAILING_LIST_MULTI,SPF_PASS,URIBL_BLOCKED,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 2105FECDFB8 for ; Wed, 18 Jul 2018 02:02:45 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id D1CCF2075A for ; Wed, 18 Jul 2018 02:02:44 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org D1CCF2075A Authentication-Results: mail.kernel.org; dmarc=none (p=none dis=none) header.from=huawei.com 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 S1731537AbeGRCiI (ORCPT ); Tue, 17 Jul 2018 22:38:08 -0400 Received: from szxga06-in.huawei.com ([45.249.212.32]:53057 "EHLO huawei.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1731194AbeGRCiI (ORCPT ); Tue, 17 Jul 2018 22:38:08 -0400 Received: from DGGEMS411-HUB.china.huawei.com (unknown [172.30.72.60]) by Forcepoint Email with ESMTP id 9CDCDDFCCEA60; Wed, 18 Jul 2018 10:02:38 +0800 (CST) Received: from huawei.com (10.113.189.234) by DGGEMS411-HUB.china.huawei.com (10.3.19.211) with Microsoft SMTP Server id 14.3.382.0; Wed, 18 Jul 2018 10:02:32 +0800 From: Yunlong Song To: , , , , CC: , , , , , Subject: [PATCH v2] f2fs: issue discard align to section in LFS mode Date: Wed, 18 Jul 2018 10:02:27 +0800 Message-ID: <1531879347-102427-1-git-send-email-yunlong.song@huawei.com> X-Mailer: git-send-email 1.8.5.2 In-Reply-To: <1531836275-94933-1-git-send-email-yunlong.song@huawei.com> References: <1531836275-94933-1-git-send-email-yunlong.song@huawei.com> MIME-Version: 1.0 Content-Type: text/plain X-Originating-IP: [10.113.189.234] X-CFilter-Loop: Reflected Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org For the case when sbi->segs_per_sec > 1 with lfs mode, take section:segment = 5 for example, if the section prefree_map is ...previous section | current section (1 1 0 1 1) | next section..., then the start = x, end = x + 1, after start = start_segno + sbi->segs_per_sec, start = x + 5, then it will skip x + 3 and x + 4, but their bitmap is still set, which will cause duplicated f2fs_issue_discard of this same section in the next write_checkpoint: round 1: section bitmap : 1 1 1 1 1, all valid, prefree_map: 0 0 0 0 0 then rm data block NO.2, block NO.2 becomes invalid, prefree_map: 0 0 1 0 0 write_checkpoint: section bitmap: 1 1 0 1 1, prefree_map: 0 0 0 0 0, prefree of NO.2 is cleared, and no discard issued round 2: rm data block NO.0, NO.1, NO.3, NO.4 all invalid, but prefree bit of NO.2 is set and cleared in round 1, then prefree_map: 1 1 0 1 1 write_checkpoint: section bitmap: 0 0 0 0 0, prefree_map: 0 0 0 1 1, no valid blocks of this section, so discard issued, but this time prefree bit of NO.3 and NO.4 is skipped due to start = start_segno + sbi->segs_per_sec; round 3: write_checkpoint: section bitmap: 0 0 0 0 0, prefree_map: 0 0 0 1 1 -> 0 0 0 0 0, no valid blocks of this section, so discard issued, this time prefree bit of NO.3 and NO.4 is cleared, but the discard of this section is sent again... To fix this problem, we can align the start and end value to section boundary for fstrim and real-time discard operation, and decide to issue discard only when the whole section is invalid, which can issue discard aligned to section size as much as possible and avoid redundant discard. Signed-off-by: Yunlong Song Signed-off-by: Chao Yu --- fs/f2fs/segment.c | 21 ++++++++++++++++----- 1 file changed, 16 insertions(+), 5 deletions(-) diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index cfff7cf..c25e2eb 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -1710,6 +1710,7 @@ void f2fs_clear_prefree_segments(struct f2fs_sb_info *sbi, unsigned int start = 0, end = -1; unsigned int secno, start_segno; bool force = (cpc->reason & CP_DISCARD); + bool need_align = test_opt(sbi, LFS) && sbi->segs_per_sec > 1; mutex_lock(&dirty_i->seglist_lock); @@ -1721,14 +1722,19 @@ void f2fs_clear_prefree_segments(struct f2fs_sb_info *sbi, end = find_next_zero_bit(prefree_map, MAIN_SEGS(sbi), start + 1); - for (i = start; i < end; i++) - clear_bit(i, prefree_map); - - dirty_i->nr_dirty[PRE] -= end - start; - if (!test_opt(sbi, DISCARD)) continue; + if (need_align) { + start = rounddown(start, sbi->segs_per_sec); + end = roundup(end, sbi->segs_per_sec); + } + + for (i = start; i < end; i++) { + if (test_and_clear_bit(i, prefree_map)) + dirty_i->nr_dirty[PRE]--; + } + if (force && start >= cpc->trim_start && (end - 1) <= cpc->trim_end) continue; @@ -2511,6 +2517,7 @@ int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range) struct discard_policy dpolicy; unsigned long long trimmed = 0; int err = 0; + bool need_align = test_opt(sbi, LFS) && sbi->segs_per_sec > 1; if (start >= MAX_BLKADDR(sbi) || range->len < sbi->blocksize) return -EINVAL; @@ -2528,6 +2535,10 @@ int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range) start_segno = (start <= MAIN_BLKADDR(sbi)) ? 0 : GET_SEGNO(sbi, start); end_segno = (end >= MAX_BLKADDR(sbi)) ? MAIN_SEGS(sbi) - 1 : GET_SEGNO(sbi, end); + if (need_align) { + start_segno = rounddown(start_segno, sbi->segs_per_sec); + end_segno = roundup(end_segno + 1, sbi->segs_per_sec) - 1; + } cpc.reason = CP_DISCARD; cpc.trim_minlen = max_t(__u64, 1, F2FS_BYTES_TO_BLK(range->minlen)); -- 1.8.5.2