From mboxrd@z Thu Jan 1 00:00:00 1970 From: Chao Yu Subject: [PATCH 2/2] f2fs: avoid unneeded loop in build_sit_entries Date: Fri, 19 Aug 2016 23:13:47 +0800 Message-ID: <1471619627-4174-2-git-send-email-chao@kernel.org> References: <1471619627-4174-1-git-send-email-chao@kernel.org> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Received: from sog-mx-1.v43.ch3.sourceforge.com ([172.29.43.191] helo=mx.sourceforge.net) by sfs-ml-3.v29.ch3.sourceforge.com with esmtp (Exim 4.76) (envelope-from ) id 1balWg-00025V-2v for linux-f2fs-devel@lists.sourceforge.net; Fri, 19 Aug 2016 15:16:06 +0000 Received: from mail.kernel.org ([198.145.29.136]) by sog-mx-1.v43.ch3.sourceforge.com with esmtps (TLSv1:AES256-SHA:256) (Exim 4.76) id 1balWf-0006Dy-6z for linux-f2fs-devel@lists.sourceforge.net; Fri, 19 Aug 2016 15:16:06 +0000 In-Reply-To: <1471619627-4174-1-git-send-email-chao@kernel.org> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: linux-f2fs-devel-bounces@lists.sourceforge.net To: jaegeuk@kernel.org Cc: linux-kernel@vger.kernel.org, linux-f2fs-devel@lists.sourceforge.net From: Chao Yu When building each sit entry in cache, firstly, we will load it from sit page, and then check all entries in sit journal, if there is one updated entry in journal, cover cached entry with the journaled one. Actually, most of check operation is unneeded since we only need to update cached entries with journaled entries in batch, so changing the flow as below for more efficient: 1. load all sit entries into cache from sit pages; 2. update sit entries with journal. Signed-off-by: Chao Yu --- fs/f2fs/segment.c | 48 ++++++++++++++++++++++++++++++++---------------- 1 file changed, 32 insertions(+), 16 deletions(-) diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 59f578b..a394012 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -2262,22 +2262,11 @@ static void build_sit_entries(struct f2fs_sb_info *sbi) struct f2fs_sit_entry sit; struct page *page; - down_read(&curseg->journal_rwsem); - for (i = 0; i < sits_in_cursum(journal); i++) { - if (le32_to_cpu(segno_in_journal(journal, i)) - == start) { - sit = sit_in_journal(journal, i); - up_read(&curseg->journal_rwsem); - goto got_it; - } - } - up_read(&curseg->journal_rwsem); - page = get_current_sit_page(sbi, start); sit_blk = (struct f2fs_sit_block *)page_address(page); sit = sit_blk->entries[SIT_ENTRY_OFFSET(sit_i, start)]; f2fs_put_page(page, 1); -got_it: + check_block_count(sbi, start, &sit); seg_info_from_raw_sit(se, &sit); @@ -2289,13 +2278,40 @@ got_it: se->valid_blocks; } - if (sbi->segs_per_sec > 1) { - struct sec_entry *e = get_sec_entry(sbi, start); - e->valid_blocks += se->valid_blocks; - } + if (sbi->segs_per_sec > 1) + get_sec_entry(sbi, start)->valid_blocks += + se->valid_blocks; } start_blk += readed; } while (start_blk < sit_blk_cnt); + + down_read(&curseg->journal_rwsem); + for (i = 0; i < sits_in_cursum(journal); i++) { + struct f2fs_sit_entry sit; + struct seg_entry *se; + unsigned int old_valid_blocks; + + start = le32_to_cpu(segno_in_journal(journal, i)); + se = &sit_i->sentries[start]; + sit = sit_in_journal(journal, i); + + old_valid_blocks = se->valid_blocks; + + check_block_count(sbi, start, &sit); + seg_info_from_raw_sit(se, &sit); + + if (f2fs_discard_en(sbi)) { + memcpy(se->discard_map, se->cur_valid_map, + SIT_VBLOCK_MAP_SIZE); + sbi->discard_blks += old_valid_blocks - + se->valid_blocks; + } + + if (sbi->segs_per_sec > 1) + get_sec_entry(sbi, start)->valid_blocks += + se->valid_blocks - old_valid_blocks; + } + up_read(&curseg->journal_rwsem); } static void init_free_segmap(struct f2fs_sb_info *sbi) -- 2.7.2 ------------------------------------------------------------------------------ From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755465AbcHSPQC (ORCPT ); Fri, 19 Aug 2016 11:16:02 -0400 Received: from mail.kernel.org ([198.145.29.136]:53870 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755417AbcHSPQA (ORCPT ); Fri, 19 Aug 2016 11:16:00 -0400 From: Chao Yu To: jaegeuk@kernel.org Cc: linux-f2fs-devel@lists.sourceforge.net, linux-kernel@vger.kernel.org, Chao Yu Subject: [PATCH 2/2] f2fs: avoid unneeded loop in build_sit_entries Date: Fri, 19 Aug 2016 23:13:47 +0800 Message-Id: <1471619627-4174-2-git-send-email-chao@kernel.org> X-Mailer: git-send-email 2.7.2 In-Reply-To: <1471619627-4174-1-git-send-email-chao@kernel.org> References: <1471619627-4174-1-git-send-email-chao@kernel.org> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Chao Yu When building each sit entry in cache, firstly, we will load it from sit page, and then check all entries in sit journal, if there is one updated entry in journal, cover cached entry with the journaled one. Actually, most of check operation is unneeded since we only need to update cached entries with journaled entries in batch, so changing the flow as below for more efficient: 1. load all sit entries into cache from sit pages; 2. update sit entries with journal. Signed-off-by: Chao Yu --- fs/f2fs/segment.c | 48 ++++++++++++++++++++++++++++++++---------------- 1 file changed, 32 insertions(+), 16 deletions(-) diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 59f578b..a394012 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -2262,22 +2262,11 @@ static void build_sit_entries(struct f2fs_sb_info *sbi) struct f2fs_sit_entry sit; struct page *page; - down_read(&curseg->journal_rwsem); - for (i = 0; i < sits_in_cursum(journal); i++) { - if (le32_to_cpu(segno_in_journal(journal, i)) - == start) { - sit = sit_in_journal(journal, i); - up_read(&curseg->journal_rwsem); - goto got_it; - } - } - up_read(&curseg->journal_rwsem); - page = get_current_sit_page(sbi, start); sit_blk = (struct f2fs_sit_block *)page_address(page); sit = sit_blk->entries[SIT_ENTRY_OFFSET(sit_i, start)]; f2fs_put_page(page, 1); -got_it: + check_block_count(sbi, start, &sit); seg_info_from_raw_sit(se, &sit); @@ -2289,13 +2278,40 @@ got_it: se->valid_blocks; } - if (sbi->segs_per_sec > 1) { - struct sec_entry *e = get_sec_entry(sbi, start); - e->valid_blocks += se->valid_blocks; - } + if (sbi->segs_per_sec > 1) + get_sec_entry(sbi, start)->valid_blocks += + se->valid_blocks; } start_blk += readed; } while (start_blk < sit_blk_cnt); + + down_read(&curseg->journal_rwsem); + for (i = 0; i < sits_in_cursum(journal); i++) { + struct f2fs_sit_entry sit; + struct seg_entry *se; + unsigned int old_valid_blocks; + + start = le32_to_cpu(segno_in_journal(journal, i)); + se = &sit_i->sentries[start]; + sit = sit_in_journal(journal, i); + + old_valid_blocks = se->valid_blocks; + + check_block_count(sbi, start, &sit); + seg_info_from_raw_sit(se, &sit); + + if (f2fs_discard_en(sbi)) { + memcpy(se->discard_map, se->cur_valid_map, + SIT_VBLOCK_MAP_SIZE); + sbi->discard_blks += old_valid_blocks - + se->valid_blocks; + } + + if (sbi->segs_per_sec > 1) + get_sec_entry(sbi, start)->valid_blocks += + se->valid_blocks - old_valid_blocks; + } + up_read(&curseg->journal_rwsem); } static void init_free_segmap(struct f2fs_sb_info *sbi) -- 2.7.2