block: fix req_bio_endio append error handling

Shinichiro Kawasaki reports that there is a bug in a recent
req_bio_endio() patch causing problems with zonefs. As Shinichiro
suggested, inverse the condition in zone append path to resemble how it
was before: fail when it's not fully completed.

Fixes: 478eb72 ("block: optimise req_bio_endio()")
Reported-by: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/344ea4e334aace9148b41af5f2426da38c8aa65a.1634914228.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block/blk-mq.c

@@ -640,7 +640,7 @@ static void req_bio_endio(struct request *rq, struct bio *bio,
 		 * Partial zone append completions cannot be supported as the
 		 * BIO fragments may end up not being written sequentially.
 		 */
-		if (bio->bi_iter.bi_size == nbytes)
+		if (bio->bi_iter.bi_size != nbytes)
 			bio->bi_status = BLK_STS_IOERR;
 		else
 			bio->bi_iter.bi_sector = rq->__sector;