gfs: Don't use GFP_NOFS in gfs2_unstuff_dinode

Revert the rest of commit 220cca2 ("GFS2: Change truncate page
allocation to be GFP_NOFS"):

In gfs2_unstuff_dinode(), there is no need to carry out the page cache
allocation under GFP_NOFS because inodes on the "regular" filesystem are
never un-inlined under memory pressure, so switch back from
find_or_create_page() to grab_cache_page() here as well.

Inodes on the "metadata" filesystem can theoretically be un-inlined
under memory pressure, but any page cache allocations in that context
would happen in GFP_NOFS context because those inodes have
inode->i_mapping->gfp_mask set to GFP_NOFS (see the previous patch).

Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>

fs/gfs2/bmap.c

@@ -161,7 +161,7 @@ int gfs2_unstuff_dinode(struct gfs2_inode *ip)
 	int error;
 
 	down_write(&ip->i_rw_mutex);
-	page = find_or_create_page(inode->i_mapping, 0, GFP_NOFS);
+	page = grab_cache_page(inode->i_mapping, 0);
 	error = -ENOMEM;
 	if (!page)
 		goto out;