From: Theodore Ts'o Subject: [PATCH 2.6.33.y 40/40] ext4: Make fsync sync new parent directories in no-journal mode Date: Tue, 1 Jun 2010 08:03:27 -0400 Message-ID: <1275393807-14369-40-git-send-email-tytso@mit.edu> References: <1275393807-14369-1-git-send-email-tytso@mit.edu> Cc: Ext4 Developers List , Frank Mayhar , "Theodore Ts'o" To: stable@vger.kernel.org Return-path: Received: from THUNK.ORG ([69.25.196.29]:47307 "EHLO thunker.thunk.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756088Ab0FAMDf (ORCPT ); Tue, 1 Jun 2010 08:03:35 -0400 In-Reply-To: <1275393807-14369-1-git-send-email-tytso@mit.edu> Sender: linux-ext4-owner@vger.kernel.org List-ID: From: Frank Mayhar commit 14ece1028b3ed53ffec1b1213ffc6acaf79ad77c upstream (as of v2.6.34-git13) Add a new ext4 state to tell us when a file has been newly created; use that state in ext4_sync_file in no-journal mode to tell us when we need to sync the parent directory as well as the inode and data itself. This fixes a problem in which a panic or power failure may lose the entire file even when using fsync, since the parent directory entry is lost. Addresses-Google-Bug: #2480057 Signed-off-by: Frank Mayhar Signed-off-by: "Theodore Ts'o" --- fs/ext4/ext4.h | 1 + fs/ext4/fsync.c | 31 +++++++++++++++++++++++++++++-- fs/ext4/namei.c | 2 ++ 3 files changed, 32 insertions(+), 2 deletions(-) diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index e231c63..2ec215f 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -1148,6 +1148,7 @@ enum { EXT4_STATE_DA_ALLOC_CLOSE, /* Alloc DA blks on close */ EXT4_STATE_EXT_MIGRATE, /* Inode is migrating */ EXT4_STATE_DIO_UNWRITTEN, /* need convert on dio done*/ + EXT4_STATE_NEWENTRY, /* File just added to dir */ }; #define EXT4_INODE_BIT_FNS(name, field) \ diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c index eb52837..5240354 100644 --- a/fs/ext4/fsync.c +++ b/fs/ext4/fsync.c @@ -35,6 +35,29 @@ #include /* + * If we're not journaling and this is a just-created file, we have to + * sync our parent directory (if it was freshly created) since + * otherwise it will only be written by writeback, leaving a huge + * window during which a crash may lose the file. This may apply for + * the parent directory's parent as well, and so on recursively, if + * they are also freshly created. + */ +static void ext4_sync_parent(struct inode *inode) +{ + struct dentry *dentry = NULL; + + while (inode && ext4_test_inode_state(inode, EXT4_STATE_NEWENTRY)) { + ext4_clear_inode_state(inode, EXT4_STATE_NEWENTRY); + dentry = list_entry(inode->i_dentry.next, + struct dentry, d_alias); + if (!dentry || !dentry->d_parent || !dentry->d_parent->d_inode) + break; + inode = dentry->d_parent->d_inode; + sync_mapping_buffers(inode->i_mapping); + } +} + +/* * akpm: A new design for ext4_sync_file(). * * This is only called from sys_fsync(), sys_fdatasync() and sys_msync(). @@ -67,8 +90,12 @@ int ext4_sync_file(struct file *file, struct dentry *dentry, int datasync) if (ret < 0) return ret; - if (!journal) - return simple_fsync(file, dentry, datasync); + if (!journal) { + ret = simple_fsync(file, dentry, datasync); + if (!ret && !list_empty(&inode->i_dentry)) + ext4_sync_parent(inode); + return ret; + } /* * data=writeback,ordered: diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index e99c4f1..c3b6ad0 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -1525,6 +1525,8 @@ static int ext4_add_entry(handle_t *handle, struct dentry *dentry, de->rec_len = ext4_rec_len_to_disk(blocksize, blocksize); retval = add_dirent_to_buf(handle, dentry, inode, de, bh); brelse(bh); + if (retval == 0) + ext4_set_inode_state(inode, EXT4_STATE_NEWENTRY); return retval; } -- 1.6.6.1.1.g974db.dirty