fs: always maintain i_dio_count

Maintain i_dio_count for all filesystems, not just those using DIO_LOCKING.
This these filesystems to also protect truncate against direct I/O requests
by using common code.  Right now the only non-DIO_LOCKING filesystem that
appears to do so is XFS, which uses an opencoded variant of the i_dio_count
scheme.

Behaviour doesn't change for filesystems never calling inode_dio_wait.
For ext4 behaviour changes when using the dioread_nonlock option, which
previously was missing any protection between truncate and direct I/O reads.
For ocfs2 that handcrafted i_dio_count manipulations are replaced with
the common code now enable.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

1 files changed, 13 insertions, 12 deletions

diff --git a/fs/direct-io.c b/fs/direct-io.c
index 354cbdbc14bd..0a073c7125a6 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -297,8 +297,7 @@ static ssize_t dio_complete(struct dio *dio, loff_t offset, ssize_t ret, bool is
                 aio_complete(dio->iocb, ret, 0);
         }
 
-        if (dio->flags & DIO_LOCKING)
-                inode_dio_done(dio->inode);
+        inode_dio_done(dio->inode);
         return ret;
 }
 
@@ -1185,14 +1184,16 @@ direct_io_worker(int rw, struct kiocb *iocb, struct inode *inode,
  *    For writes this function is called under i_mutex and returns with
  *    i_mutex held, for reads, i_mutex is not held on entry, but it is
  *    taken and dropped again before returning.
- *    The i_dio_count counter keeps track of the number of outstanding
- *    direct I/O requests, and truncate waits for it to reach zero.
- *    New references to i_dio_count must only be grabbed with i_mutex
- *    held.
- *
  *  - if the flags value does NOT contain DIO_LOCKING we don't use any
  *    internal locking but rather rely on the filesystem to synchronize
  *    direct I/O reads/writes versus each other and truncate.
+ *
+ * To help with locking against truncate we incremented the i_dio_count
+ * counter before starting direct I/O, and decrement it once we are done.
+ * Truncate can wait for it to reach zero to provide exclusion.  It is
+ * expected that filesystem provide exclusion between new direct I/O
+ * and truncates.  For DIO_LOCKING filesystems this is done by i_mutex,
+ * but other filesystems need to take care of this on their own.
  */
 ssize_t
 __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
@@ -1270,14 +1271,14 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
                                 goto out;
                         }
                 }
-
-                /*
-                 * Will be decremented at I/O completion time.
-                 */
-                atomic_inc(&inode->i_dio_count);
         }
 
         /*
+         * Will be decremented at I/O completion time.
+         */
+        atomic_inc(&inode->i_dio_count);
+
+        /*
          * 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