xfs: remove the i_new_size field in struct xfs_inode

Now that we use the VFS i_size field throughout XFS there is no need for the
i_new_size field any more given that the VFS i_size field gets updated
in ->write_end before unlocking the page, and thus is always uptodate when
writeback could see a page.  Removing i_new_size also has the advantage that
we will never have to trim back di_size during a failed buffered write,
given that it never gets updated past i_size.

Note that currently the generic direct I/O code only updates i_size after
calling our end_io handler, which requires a small workaround to make
sure di_size actually makes it to disk.  I hope to fix this properly in
the generic code.

A downside is that we lose the support for parallel non-overlapping O_DIRECT
appending writes that recently was added.  I don't think keeping the complex
and fragile i_new_size infrastructure for this is a good tradeoff - if we
really care about parallel appending writers we should investigate turning
the iolock into a range lock, which would also allow for parallel
non-overlapping buffered writers.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Ben Myers <bpm@sgi.com>

1 files changed, 16 insertions, 13 deletions

diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c
index 4d27ea117e0e..74b9baf36ac3 100644
--- a/fs/xfs/xfs_aops.c
+++ b/fs/xfs/xfs_aops.c
@@ -111,8 +111,7 @@ xfs_ioend_new_eof(
 	xfs_fsize_t		bsize;
 
 	bsize = ioend->io_offset + ioend->io_size;
-	isize = MAX(i_size_read(VFS_I(ip)), ip->i_new_size);
-	isize = MIN(isize, bsize);
+	isize = MIN(i_size_read(VFS_I(ip)), bsize);
 	return isize > ip->i_d.di_size ? isize : 0;
 }
 
@@ -126,11 +125,7 @@ static inline bool xfs_ioend_is_append(struct xfs_ioend *ioend)
 }
 
 /*
- * Update on-disk file size now that data has been written to disk.  The
- * current in-memory file size is i_size.  If a write is beyond eof i_new_size
- * will be the intended file size until i_size is updated.  If this write does
- * not extend all the way to the valid file size then restrict this update to
- * the end of the write.
+ * Update on-disk file size now that data has been written to disk.
  *
  * This function does not block as blocking on the inode lock in IO completion
  * can lead to IO completion order dependency deadlocks.. If it can't get the
@@ -1279,6 +1274,15 @@ xfs_end_io_direct_write(
 	struct xfs_ioend	*ioend = iocb->private;
 
 	/*
+	 * While the generic direct I/O code updates the inode size, it does
+	 * so only after the end_io handler is called, which means our
+	 * end_io handler thinks the on-disk size is outside the in-core
+	 * size.  To prevent this just update it a little bit earlier here.
+	 */
+	if (offset + size > i_size_read(ioend->io_inode))
+		i_size_write(ioend->io_inode, offset + size);
+
+	/*
 	 * blockdev_direct_IO can return an error even after the I/O
 	 * completion handler was called.  Thus we need to protect
 	 * against double-freeing.
@@ -1340,12 +1344,11 @@ xfs_vm_write_failed(
 
 	if (to > inode->i_size) {
 		/*
-		 * punch out the delalloc blocks we have already allocated. We
-		 * don't call xfs_setattr() to do this as we may be in the
-		 * middle of a multi-iovec write and so the vfs inode->i_size
-		 * will not match the xfs ip->i_size and so it will zero too
-		 * much. Hence we jus truncate the page cache to zero what is
-		 * necessary and punch the delalloc blocks directly.
+		 * Punch out the delalloc blocks we have already allocated.
+		 *
+		 * Don't bother with xfs_setattr given that nothing can have
+		 * made it to disk yet as the page is still locked at this
+		 * point.
 		 */
 		struct xfs_inode	*ip = XFS_I(inode);
 		xfs_fileoff_t		start_fsb;