Received: by 2002:a05:6512:3d0e:0:0:0:0 with SMTP id d14csp48824lfv; Tue, 12 Apr 2022 16:48:34 -0700 (PDT) X-Google-Smtp-Source: ABdhPJxvbGIFVKjvAzyqx0qD3fQnwfBqCi8HYQwEKZZCNTXu1qfLMbw6Kmqq5hZ8EkLyorPJSNTT X-Received: by 2002:a17:902:c951:b0:154:38b8:aa30 with SMTP id i17-20020a170902c95100b0015438b8aa30mr39857090pla.145.1649807192239; Tue, 12 Apr 2022 16:46:32 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1649807192; cv=none; d=google.com; s=arc-20160816; b=OKfFzLw12Zck3XEEl9tiMbEgc5XZ36+ilsDYSLL07fCEz7OgwqI3jXmGJLIcXmfTjN HhJMNSMDnBAz7gyjotr5wM1ST7ALzjjDJnJLQW9b2e7kSmBQ7rBdpw07ueFpg/+DIA5K ETUBxQRRRS1243Hom8fkDJnXi0CIco0GEkmrQYWHcX791u/dz4UeSq1aNl7TS00bwHzL s8tjDzmZxdItQ4AgAZTYAU3IbZeMkMRk4rDiUAN6WMnrT5C7ZGUO9WKd62HPf7Mlb/E/ U/CU6ZXfDDSqLMJk8rSo1obvd5vM2e+WX5AhMKLaV/XVGh9fz0qM+VNVbb67J1TRtgtC Ql/Q== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:in-reply-to:content-disposition:mime-version :references:message-id:subject:cc:to:from:date:dkim-signature; bh=xdu0d36HDdyHscsaxtx/1Tx7/27I7rb1js0FQcC8pvs=; b=II6OzkNClReXaIVOVr1q+HFfWDR01OlBpG1eoKgrtld+Lmb758JVZXCAd6JZjF9dVt 4Itj9qFOVOv8oKNRuqrz+GQ3XCB0a3ywMIvypXA2U+MLkxTgsL3t667qWoY3Y4YUQOoQ 0YSvy0btvXB6ChdDEElI2WlDsXtWsT30XzaJKZ+kbqwy4jDzyDbmJh0G8OkAnP03tvwt ftADV1qD+6vbFIYPfuzmMypIhkla0RBae8zOZ7FTIB2lJXIvDyP4nEKHQHnVoKe2ZGIe s7UHNS3yN4a4XxEQXeOsI2H+6eNTn2CpSqpXZcG4B2IZfQ/CyKSQhoAbcRJwTLt7Vu3k I9XA== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@kernel.org header.s=k20201202 header.b="nyNi4/EO"; spf=softfail (google.com: domain of transitioning linux-kernel-owner@vger.kernel.org does not designate 23.128.96.19 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=kernel.org Return-Path: Received: from lindbergh.monkeyblade.net (lindbergh.monkeyblade.net. [23.128.96.19]) by mx.google.com with ESMTPS id h68-20020a636c47000000b0039961d076f8si4215300pgc.51.2022.04.12.16.46.31 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 12 Apr 2022 16:46:32 -0700 (PDT) Received-SPF: softfail (google.com: domain of transitioning linux-kernel-owner@vger.kernel.org does not designate 23.128.96.19 as permitted sender) client-ip=23.128.96.19; Authentication-Results: mx.google.com; dkim=pass header.i=@kernel.org header.s=k20201202 header.b="nyNi4/EO"; spf=softfail (google.com: domain of transitioning linux-kernel-owner@vger.kernel.org does not designate 23.128.96.19 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=kernel.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by lindbergh.monkeyblade.net (Postfix) with ESMTP id ADF301D4C2D; Tue, 12 Apr 2022 14:38:52 -0700 (PDT) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1349829AbiDKUWv (ORCPT + 99 others); Mon, 11 Apr 2022 16:22:51 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:37244 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S233857AbiDKUWu (ORCPT ); Mon, 11 Apr 2022 16:22:50 -0400 Received: from ams.source.kernel.org (ams.source.kernel.org [145.40.68.75]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 593F625C52 for ; Mon, 11 Apr 2022 13:20:35 -0700 (PDT) Received: from smtp.kernel.org (relay.kernel.org [52.25.139.140]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ams.source.kernel.org (Postfix) with ESMTPS id D14C9B818AE for ; Mon, 11 Apr 2022 20:20:33 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id 573ABC385A6; Mon, 11 Apr 2022 20:20:32 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1649708432; bh=25sHb8HttN5LykPduZQYnoCC5sTx6+iuY5/D1DMWVeA=; h=Date:From:To:Cc:Subject:References:In-Reply-To:From; b=nyNi4/EOILdhFmaXIlM5ymPCb/oZ8Ra5ooJFcFev3elbI6CWDaq0Gi27KVfbv++pq ko2Mi42fR3qsb2n2T7t+jWZSieALid/YQ853MGCSZSTs6RCcaAMXcJS7ajDe3Hqdd5 KK37nIY2p0jjo/yHmc+a0KrmqK2M80aevacuPqIV8fYY99v2h9AJwPwseocl2Cqj76 BlOdBFUAEV1nFPMTyaEvD4TPmnp4YhFOmjOzcCX3UmSXPZLE/xYYdpmV09F8ZssmB0 Kf5/ATM5Yn0+v4r0da1lVQzCwrb72jsg3G0Eolr7Os30RN+ugs6dS3i/J6/MwuWAN5 KvNm2/u92YGzg== Date: Mon, 11 Apr 2022 13:20:30 -0700 From: Jaegeuk Kim To: Chao Yu Cc: linux-f2fs-devel@lists.sourceforge.net, linux-kernel@vger.kernel.org, Chao Yu Subject: Re: [PATCH v3] f2fs: give priority to select unpinned section for foreground GC Message-ID: References: <20220406152651.5142-1-chao@kernel.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20220406152651.5142-1-chao@kernel.org> X-Spam-Status: No, score=-2.3 required=5.0 tests=BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,DKIM_VALID_EF,MAILING_LIST_MULTI, RDNS_NONE,SPF_HELO_NONE,T_SCC_BODY_TEXT_LINE autolearn=unavailable autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on lindbergh.monkeyblade.net Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 04/06, Chao Yu wrote: > Previously, during foreground GC, if victims contain data of pinned file, > it will fail migration of the data, and meanwhile i_gc_failures of that > pinned file may increase, and when it exceeds threshold, GC will unpin > the file, result in breaking pinfile's semantics. > > In order to mitigate such condition, let's record and skip section which > has pinned file's data and give priority to select unpinned one. > > Signed-off-by: Chao Yu > --- > v3: > - check pin status before pinning section in pin_section(). > fs/f2fs/gc.c | 56 ++++++++++++++++++++++++++++++++++++++++++++--- > fs/f2fs/segment.c | 7 ++++++ > fs/f2fs/segment.h | 2 ++ > 3 files changed, 62 insertions(+), 3 deletions(-) > > diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c > index 6a7e4148ff9d..df23824ae3c2 100644 > --- a/fs/f2fs/gc.c > +++ b/fs/f2fs/gc.c > @@ -646,6 +646,37 @@ static void release_victim_entry(struct f2fs_sb_info *sbi) > f2fs_bug_on(sbi, !list_empty(&am->victim_list)); > } > > +static void pin_section(struct f2fs_sb_info *sbi, unsigned int segno) Need f2fs_...? > +{ > + struct dirty_seglist_info *dirty_i = DIRTY_I(sbi); > + unsigned int secno = GET_SEC_FROM_SEG(sbi, segno); > + > + if (test_bit(secno, dirty_i->pinned_secmap)) > + return; > + set_bit(secno, dirty_i->pinned_secmap); > + dirty_i->pinned_secmap_cnt++; > +} > + > +static bool pinned_section_exists(struct dirty_seglist_info *dirty_i) > +{ > + return dirty_i->pinned_secmap_cnt; > +} > + > +static bool section_is_pinned(struct dirty_seglist_info *dirty_i, > + unsigned int secno) > +{ > + return pinned_section_exists(dirty_i) && > + test_bit(secno, dirty_i->pinned_secmap); > +} > + > +static void unpin_all_sections(struct f2fs_sb_info *sbi) > +{ > + unsigned int bitmap_size = f2fs_bitmap_size(MAIN_SECS(sbi)); > + > + memset(DIRTY_I(sbi)->pinned_secmap, 0, bitmap_size); > + DIRTY_I(sbi)->pinned_secmap_cnt = 0; > +} > + > /* > * This function is called from two paths. > * One is garbage collection and the other is SSR segment selection. > @@ -787,6 +818,9 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi, > if (gc_type == BG_GC && test_bit(secno, dirty_i->victim_secmap)) > goto next; > > + if (gc_type == FG_GC && section_is_pinned(dirty_i, secno)) > + goto next; > + > if (is_atgc) { > add_victim_entry(sbi, &p, segno); > goto next; > @@ -1202,8 +1236,10 @@ static int move_data_block(struct inode *inode, block_t bidx, > } > > if (f2fs_is_pinned_file(inode)) { > - if (gc_type == FG_GC) > + if (gc_type == FG_GC) { > f2fs_pin_file_control(inode, true); > + pin_section(F2FS_I_SB(inode), segno); Do we need to check unpinning the inode? if (!f2fs_pin_file_control()) f2fs_set_pin_section(); > + } > err = -EAGAIN; > goto out; > } > @@ -1352,8 +1388,10 @@ static int move_data_page(struct inode *inode, block_t bidx, int gc_type, > goto out; > } > if (f2fs_is_pinned_file(inode)) { > - if (gc_type == FG_GC) > + if (gc_type == FG_GC) { > f2fs_pin_file_control(inode, true); > + pin_section(F2FS_I_SB(inode), segno); > + } > err = -EAGAIN; > goto out; > } > @@ -1485,6 +1523,7 @@ static int gc_data_segment(struct f2fs_sb_info *sbi, struct f2fs_summary *sum, > gc_type == FG_GC) { > f2fs_pin_file_control(inode, true); > iput(inode); > + pin_section(sbi, segno); We don't have this code. > return submitted; > } > > @@ -1766,9 +1805,17 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync, > ret = -EINVAL; > goto stop; > } > +retry: > ret = __get_victim(sbi, &segno, gc_type); > - if (ret) > + if (ret) { > + /* allow to search victim from sections has pinned data */ > + if (ret == -ENODATA && gc_type == FG_GC && > + pinned_section_exists(DIRTY_I(sbi))) { > + unpin_all_sections(sbi); > + goto retry; > + } > goto stop; > + } > > seg_freed = do_garbage_collect(sbi, segno, &gc_list, gc_type, force); > if (gc_type == FG_GC && > @@ -1811,6 +1858,9 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync, > SIT_I(sbi)->last_victim[ALLOC_NEXT] = 0; > SIT_I(sbi)->last_victim[FLUSH_DEVICE] = init_segno; > > + if (gc_type == FG_GC && pinned_section_exists(DIRTY_I(sbi))) > + unpin_all_sections(sbi); > + > trace_f2fs_gc_end(sbi->sb, ret, total_freed, sec_freed, > get_pages(sbi, F2FS_DIRTY_NODES), > get_pages(sbi, F2FS_DIRTY_DENTS), > diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c > index 012524db7437..1c20d7c9eca3 100644 > --- a/fs/f2fs/segment.c > +++ b/fs/f2fs/segment.c > @@ -4736,6 +4736,12 @@ static int init_victim_secmap(struct f2fs_sb_info *sbi) > dirty_i->victim_secmap = f2fs_kvzalloc(sbi, bitmap_size, GFP_KERNEL); > if (!dirty_i->victim_secmap) > return -ENOMEM; > + > + dirty_i->pinned_secmap = f2fs_kvzalloc(sbi, bitmap_size, GFP_KERNEL); > + if (!dirty_i->pinned_secmap) > + return -ENOMEM; > + > + dirty_i->pinned_secmap_cnt = 0; > return 0; > } > > @@ -5324,6 +5330,7 @@ static void destroy_victim_secmap(struct f2fs_sb_info *sbi) > { > struct dirty_seglist_info *dirty_i = DIRTY_I(sbi); > > + kvfree(dirty_i->pinned_secmap); > kvfree(dirty_i->victim_secmap); > } > > diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h > index 5c94caf0c0a1..fd6f246e649c 100644 > --- a/fs/f2fs/segment.h > +++ b/fs/f2fs/segment.h > @@ -294,6 +294,8 @@ struct dirty_seglist_info { > struct mutex seglist_lock; /* lock for segment bitmaps */ > int nr_dirty[NR_DIRTY_TYPE]; /* # of dirty segments */ > unsigned long *victim_secmap; /* background GC victims */ > + unsigned long *pinned_secmap; /* pinned victims from foreground GC */ > + unsigned int pinned_secmap_cnt; /* count of victims which has pinned data */ > }; > > /* victim selection function for cleaning and SSR */ > -- > 2.32.0