| Message ID | 20171115115419.21862-1-eguan@redhat.com |
|---|---|
| State | Accepted, archived |
| Headers | show |
| Series | ext4: fix fdatasync(2) after fallocate(2) operation | expand |
On Wed, Nov 15, 2017 at 07:54:19PM +0800, Eryu Guan wrote: > Currently, fallocate(2) with KEEP_SIZE followed by a fdatasync(2) > then crash, we'll see wrong allocated block number (stat -c %b), the > blocks allocated beyond EOF are all lost. fstests generic/468 > exposes this bug. > > Commit 67a7d5f561f4 ("ext4: fix fdatasync(2) after extent > manipulation operations") fixed all the other extent manipulation > operation paths such as hole punch, zero range, collapse range etc., > but forgot the fallocate case. > > So similarly, fix it by recording the correct journal tid in ext4 > inode in fallocate(2) path, so that ext4_sync_file() will wait for > the right tid to be committed on fdatasync(2). > > Signed-off-by: Eryu Guan <eguan@redhat.com> Thanks, applied. - Ted
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 07bca11749d4..c941251ac0c0 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -4722,6 +4722,7 @@ static int ext4_alloc_file_blocks(struct file *file, ext4_lblk_t offset, EXT4_INODE_EOFBLOCKS); } ext4_mark_inode_dirty(handle, inode); + ext4_update_inode_fsync_trans(handle, inode, 1); ret2 = ext4_journal_stop(handle); if (ret2) break;
Currently, fallocate(2) with KEEP_SIZE followed by a fdatasync(2) then crash, we'll see wrong allocated block number (stat -c %b), the blocks allocated beyond EOF are all lost. fstests generic/468 exposes this bug. Commit 67a7d5f561f4 ("ext4: fix fdatasync(2) after extent manipulation operations") fixed all the other extent manipulation operation paths such as hole punch, zero range, collapse range etc., but forgot the fallocate case. So similarly, fix it by recording the correct journal tid in ext4 inode in fallocate(2) path, so that ext4_sync_file() will wait for the right tid to be committed on fdatasync(2). Signed-off-by: Eryu Guan <eguan@redhat.com> --- Note that fstests case generic/468 has not been pushed to upstream yet, please find it at https://patchwork.kernel.org/patch/7284531/ So the final test seq number can be different, though it's unlikely to change. If changed I'll send an updated patch. I've tested this patch with the pending generic/468 test and a auto group run of fstests with default ext4 configuration (data=ordered and 4k block size), test results all look good. fs/ext4/extents.c | 1 + 1 file changed, 1 insertion(+)