rbd: Fix error recovery in rbd_obj_read_sync()

commit a8d4205 upstream.

When we fail to allocate page vector in rbd_obj_read_sync() we just
basically ignore the problem and continue which will result in an oops
later. Fix the problem by returning proper error.

CC: Yehuda Sadeh <yehuda@inktank.com>
CC: Sage Weil <sage@inktank.com>
CC: ceph-devel@vger.kernel.org
Coverity-id: 1226882
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Ilya Dryomov <idryomov@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/block/rbd.c

@@ -3227,7 +3227,7 @@ static int rbd_obj_read_sync(struct rbd_device *rbd_dev,
 	page_count = (u32) calc_pages_for(offset, length);
 	pages = ceph_alloc_page_vector(page_count, GFP_KERNEL);
 	if (IS_ERR(pages))
-		ret = PTR_ERR(pages);
+		return PTR_ERR(pages);
 
 	ret = -ENOMEM;
 	obj_request = rbd_obj_request_create(object_name, offset, length,