From: Chao Yu <chao2.yu@samsung.com>
To: Jaegeuk Kim <jaegeuk@kernel.org>
Cc: linux-f2fs-devel@lists.sourceforge.net, linux-kernel@vger.kernel.org
Subject: [PATCH 5/8] f2fs: support data flush in checkpoint
Date: Tue, 15 Dec 2015 13:33:18 +0800
Message-id: <00f801d136fa$324a5070$96def150$@samsung.com>
MIME-version: 1.0
Content-type: text/plain; charset=us-ascii
Content-transfer-encoding: 7bit
Thread-index: AdE29i7bGhFTI9+eSEeo0oJ8hbTkBg==
Content-language: zh-cn
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 2561
Lines: 80

Previously, when finishing a checkpoint, we only keep consistency of all fs
meta info including meta inode, node inode, dentry page of directory inode,
so, after a sudden power cut, f2fs can recover from last checkpoint with
full directory structure.

But during checkpoint, we didn't flush dirty pages of regular and symlink
inode, so such dirty datas still in memory will be lost in that moment of
power off.

In order to reduce the chance of lost data, this patch tries to flush user
data before starting a checkpoint. So user's data written between two
checkpoint which may not be fsynced could be saved.

Signed-off-by: Chao Yu <chao2.yu@samsung.com>
---
 fs/f2fs/checkpoint.c | 11 +++++++++++
 fs/f2fs/f2fs.h       |  5 +++++
 2 files changed, 16 insertions(+)

diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
index f33c4d7..217571f 100644
--- a/fs/f2fs/checkpoint.c
+++ b/fs/f2fs/checkpoint.c
@@ -849,6 +849,17 @@ static int block_operations(struct f2fs_sb_info *sbi)
 
 	blk_start_plug(&plug);
 
+retry_flush_datas:
+	/* write all the dirty data pages */
+	if (get_pages(sbi, F2FS_DIRTY_DATAS)) {
+		sync_dirty_inodes(sbi, FILE_INODE);
+		if (unlikely(f2fs_cp_error(sbi))) {
+			err = -EIO;
+			goto out;
+		}
+		goto retry_flush_datas;
+	}
+
 retry_flush_dents:
 	f2fs_lock_all(sbi);
 	/* write all the dirty dentry pages */
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index d8bef3c..2477b2f5 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -648,6 +648,7 @@ struct f2fs_sm_info {
 enum count_type {
 	F2FS_WRITEBACK,
 	F2FS_DIRTY_DENTS,
+	F2FS_DIRTY_DATAS,
 	F2FS_DIRTY_NODES,
 	F2FS_DIRTY_META,
 	F2FS_INMEM_PAGES,
@@ -1068,6 +1069,8 @@ static inline void inode_inc_dirty_pages(struct inode *inode)
 	atomic_inc(&F2FS_I(inode)->dirty_pages);
 	if (S_ISDIR(inode->i_mode))
 		inc_page_count(F2FS_I_SB(inode), F2FS_DIRTY_DENTS);
+	else
+		inc_page_count(F2FS_I_SB(inode), F2FS_DIRTY_DATAS);
 }
 
 static inline void dec_page_count(struct f2fs_sb_info *sbi, int count_type)
@@ -1085,6 +1088,8 @@ static inline void inode_dec_dirty_pages(struct inode *inode)
 
 	if (S_ISDIR(inode->i_mode))
 		dec_page_count(F2FS_I_SB(inode), F2FS_DIRTY_DENTS);
+	else
+		dec_page_count(F2FS_I_SB(inode), F2FS_DIRTY_DATAS);
 }
 
 static inline int get_pages(struct f2fs_sb_info *sbi, int count_type)
-- 
2.6.3


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/