fs: don't call ->writepage from __mpage_writepage

All callers of mpage_writepage use block_write_full_page as their
->writepage implementation when called from mpage_writepages
(although for ntfs3 this is obsfucated a bit).

Just call block_write_full_page directly instead of going through
the ->writepage indirection.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>

fs/mpage.c

@@ -626,7 +626,7 @@ static int __mpage_writepage(struct page *page, struct writeback_control *wbc,
 	/*
 	 * The caller has a ref on the inode, so *mapping is stable
 	 */
-	ret = mapping->a_ops->writepage(page, wbc);
+	ret = block_write_full_page(page, mpd->get_block, wbc);
 	mapping_set_error(mapping, ret);
 out:
 	mpd->bio = bio;