[PATCH 3.16.y-ckt 023/142] ext4: fix crashes in dioread_nolock mode

Luis Henriques luis.henriques at canonical.com
Tue Mar 22 10:39:12 UTC 2016


3.16.7-ckt26 -stable review patch.  If anyone has any objections, please let me know.

---8<------------------------------------------------------------

From: Jan Kara <jack at suse.cz>

commit 74dae4278546b897eb81784fdfcce872ddd8b2b8 upstream.

Competing overwrite DIO in dioread_nolock mode will just overwrite
pointer to io_end in the inode. This may result in data corruption or
extent conversion happening from IO completion interrupt because we
don't properly set buffer_defer_completion() when unlocked DIO races
with locked DIO to unwritten extent.

Since unlocked DIO doesn't need io_end for anything, just avoid
allocating it and corrupting pointer from inode for locked DIO.
A cleaner fix would be to avoid these games with io_end pointer from the
inode but that requires more intrusive changes so we leave that for
later.

Signed-off-by: Jan Kara <jack at suse.cz>
Signed-off-by: Theodore Ts'o <tytso at mit.edu>
Signed-off-by: Luis Henriques <luis.henriques at canonical.com>
---
 fs/ext4/inode.c | 40 ++++++++++++++++++++--------------------
 1 file changed, 20 insertions(+), 20 deletions(-)

diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 2d7e9a181be0..9af7ee7d0193 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -3201,29 +3201,29 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb,
 	 * case, we allocate an io_end structure to hook to the iocb.
 	 */
 	iocb->private = NULL;
-	ext4_inode_aio_set(inode, NULL);
-	if (!is_sync_kiocb(iocb)) {
-		io_end = ext4_init_io_end(inode, GFP_NOFS);
-		if (!io_end) {
-			ret = -ENOMEM;
-			goto retake_lock;
-		}
-		/*
-		 * Grab reference for DIO. Will be dropped in ext4_end_io_dio()
-		 */
-		iocb->private = ext4_get_io_end(io_end);
-		/*
-		 * we save the io structure for current async direct
-		 * IO, so that later ext4_map_blocks() could flag the
-		 * io structure whether there is a unwritten extents
-		 * needs to be converted when IO is completed.
-		 */
-		ext4_inode_aio_set(inode, io_end);
-	}
-
 	if (overwrite) {
 		get_block_func = ext4_get_block_write_nolock;
 	} else {
+		ext4_inode_aio_set(inode, NULL);
+		if (!is_sync_kiocb(iocb)) {
+			io_end = ext4_init_io_end(inode, GFP_NOFS);
+			if (!io_end) {
+				ret = -ENOMEM;
+				goto retake_lock;
+			}
+			/*
+			 * Grab reference for DIO. Will be dropped in
+			 * ext4_end_io_dio()
+			 */
+			iocb->private = ext4_get_io_end(io_end);
+			/*
+			 * we save the io structure for current async direct
+			 * IO, so that later ext4_map_blocks() could flag the
+			 * io structure whether there is a unwritten extents
+			 * needs to be converted when IO is completed.
+			 */
+			ext4_inode_aio_set(inode, io_end);
+		}
 		get_block_func = ext4_get_block_write;
 		dio_flags = DIO_LOCKING;
 	}




More information about the kernel-team mailing list