gfs2: set FMODE_CAN_ODIRECT instead of a dummy direct_IO method

Since commit a2ad63da ("VFS: add FMODE_CAN_ODIRECT file flag"), file
systems can just set the FMODE_CAN_ODIRECT flag at open time instead of
wiring up a dummy direct_IO method to indicate support for direct I/O.

Remove .direct_IO from gfs2_aops and set FMODE_CAN_ODIRECT in
gfs2_open_common for regular files that do not use data journalling.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>

fs/gfs2/aops.c

@@ -750,7 +750,6 @@ static const struct address_space_operations gfs2_aops = {
 	.release_folio = iomap_release_folio,
 	.invalidate_folio = iomap_invalidate_folio,
 	.bmap = gfs2_bmap,
-	.direct_IO = noop_direct_IO,
 	.migrate_folio = filemap_migrate_folio,
 	.is_partially_uptodate = iomap_is_partially_uptodate,
 	.error_remove_page = generic_error_remove_page,

fs/gfs2/file.c

@@ -630,6 +630,9 @@ int gfs2_open_common(struct inode *inode, struct file *file)
 		ret = generic_file_open(inode, file);
 		if (ret)
 			return ret;
+
+		if (!gfs2_is_jdata(GFS2_I(inode)))
+			file->f_mode |= FMODE_CAN_ODIRECT;
 	}
 
 	fp = kzalloc(sizeof(struct gfs2_file), GFP_NOFS);