Skip to content

Commit

Permalink
direct-io: add flag to allow aio writes beyond i_size
Browse files Browse the repository at this point in the history
Some filesystems can handle direct I/O writes beyond i_size safely,
so allow them to opt into receiving them.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
  • Loading branch information
Christoph Hellwig authored and Dave Chinner committed Feb 9, 2014
1 parent 38dbfb5 commit 6039257
Show file tree
Hide file tree
Showing 2 changed files with 15 additions and 6 deletions.
18 changes: 12 additions & 6 deletions fs/direct-io.c
Original file line number Diff line number Diff line change
Expand Up @@ -1194,13 +1194,19 @@ do_blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
}

/*
* For file extending writes updating i_size before data
* writeouts complete can expose uninitialized blocks. So
* even for AIO, we need to wait for i/o to complete before
* returning in this case.
* For file extending writes updating i_size before data writeouts
* complete can expose uninitialized blocks in dumb filesystems.
* In that case we need to wait for I/O completion even if asked
* for an asynchronous write.
*/
dio->is_async = !is_sync_kiocb(iocb) && !((rw & WRITE) &&
(end > i_size_read(inode)));
if (is_sync_kiocb(iocb))
dio->is_async = false;
else if (!(dio->flags & DIO_ASYNC_EXTEND) &&
(rw & WRITE) && end > i_size_read(inode))
dio->is_async = false;
else
dio->is_async = true;

dio->inode = inode;
dio->rw = rw;

Expand Down
3 changes: 3 additions & 0 deletions include/linux/fs.h
Original file line number Diff line number Diff line change
Expand Up @@ -2527,6 +2527,9 @@ enum {

/* filesystem does not support filling holes */
DIO_SKIP_HOLES = 0x02,

/* filesystem can handle aio writes beyond i_size */
DIO_ASYNC_EXTEND = 0x04,
};

void dio_end_io(struct bio *bio, int error);
Expand Down

0 comments on commit 6039257

Please sign in to comment.