ext3: Add replace-on-truncate hueristics for data=writeback mode

In data=writeback mode, start an asynchronous flush when closing a
file which had been previously truncated down to zero.  This lowers
the probability of data loss in the case of applications that attempt
to replace a file using truncate.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>

fs/ext3/file.c

@@ -33,6 +33,10 @@
  */
 static int ext3_release_file (struct inode * inode, struct file * filp)
 {
+	if (EXT3_I(inode)->i_state & EXT3_STATE_FLUSH_ON_CLOSE) {
+		filemap_flush(inode->i_mapping);
+		EXT3_I(inode)->i_state &= ~EXT3_STATE_FLUSH_ON_CLOSE;
+	}
 	/* if we are the last writer on the inode, drop the block reservation */
 	if ((filp->f_mode & FMODE_WRITE) &&
 			(atomic_read(&inode->i_writecount) == 1))

fs/ext3/inode.c

@@ -2346,6 +2346,9 @@ void ext3_truncate(struct inode *inode)
 	if (!ext3_can_truncate(inode))
 		return;
 
+	if (inode->i_size == 0 && ext3_should_writeback_data(inode))
+		ei->i_state |= EXT3_STATE_FLUSH_ON_CLOSE;
+
 	/*
 	 * We have to lock the EOF page here, because lock_page() nests
 	 * outside journal_start().

include/linux/ext3_fs.h

@@ -208,6 +208,7 @@ static inline __u32 ext3_mask_flags(umode_t mode, __u32 flags)
 #define EXT3_STATE_JDATA		0x00000001 /* journaled data exists */
 #define EXT3_STATE_NEW			0x00000002 /* inode is newly created */
 #define EXT3_STATE_XATTR		0x00000004 /* has in-inode xattrs */
+#define EXT3_STATE_FLUSH_ON_CLOSE	0x00000008
 
 /* Used to pass group descriptor data when online resize is done */
 struct ext3_new_group_input {