Merge branch 'for-linus' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:
"A smaller collection of fixes for the block core that would be nice to
have in -rc2. This pull request contains:

- Fixes for races in the wait/wakeup logic used in blk-mq from
Alexander. No issues have been observed, but it is definitely a
bit flakey currently. Alternatively, we may drop the cyclic
wakeups going forward, but that needs more testing.

- Some cleanups from Christoph.

- Fix for an oops in null_blk if queue_mode=1 and softirq completions
are used. From me.

- A fix for a regression caused by the chunk size setting. It
inadvertently used max_hw_sectors instead of max_sectors, which is
incorrect, and causes hangs on btrfs multi-disk setups (where hw
sectors apparently isn't set). From me.

- Removal of WQ_POWER_EFFICIENT in the kblockd creation. This was a
recent addition as well, but it actually breaks blk-mq which relies
on strict scheduling. If the workqueue power_efficient mode is
turned on, this breaks blk-mq. From Matias.

- null_blk module parameter description fix from Mike"

* 'for-linus' of git://git.kernel.dk/linux-block:
blk-mq: bitmap tag: fix races in bt_get() function
blk-mq: bitmap tag: fix race on blk_mq_bitmap_tags::wake_cnt
blk-mq: bitmap tag: fix races on shared ::wake_index fields
block: blk_max_size_offset() should check ->max_sectors
null_blk: fix softirq completions for queue_mode == 1
blk-mq: merge blk_mq_drain_queue and __blk_mq_drain_queue
blk-mq: properly drain stopped queues
block: remove WQ_POWER_EFFICIENT from kblockd
null_blk: fix name and description of 'queue_mode' module parameter
block: remove elv_abort_queue and blk_abort_flushes

block/blk-core.c

···
 
 	/* used for unplugging and affects IO latency/throughput - HIGHPRI */
 	kblockd_workqueue = alloc_workqueue("kblockd",
-					    WQ_MEM_RECLAIM | WQ_HIGHPRI |
-					    WQ_POWER_EFFICIENT, 0);
+					    WQ_MEM_RECLAIM | WQ_HIGHPRI, 0);
 	if (!kblockd_workqueue)
 		panic("Failed to create kblockd\n");
 

block/blk-flush.c

···
 }
 
 /**
- * blk_abort_flushes - @q is being aborted, abort flush requests
- * @q: request_queue being aborted
- *
- * To be called from elv_abort_queue().  @q is being aborted.  Prepare all
- * FLUSH/FUA requests for abortion.
- *
- * CONTEXT:
- * spin_lock_irq(q->queue_lock)
- */
-void blk_abort_flushes(struct request_queue *q)
-{
-	struct request *rq, *n;
-	int i;
-
-	/*
-	 * Requests in flight for data are already owned by the dispatch
-	 * queue or the device driver.  Just restore for normal completion.
-	 */
-	list_for_each_entry_safe(rq, n, &q->flush_data_in_flight, flush.list) {
-		list_del_init(&rq->flush.list);
-		blk_flush_restore_request(rq);
-	}
-
-	/*
-	 * We need to give away requests on flush queues.  Restore for
-	 * normal completion and put them on the dispatch queue.
-	 */
-	for (i = 0; i < ARRAY_SIZE(q->flush_queue); i++) {
-		list_for_each_entry_safe(rq, n, &q->flush_queue[i],
-					 flush.list) {
-			list_del_init(&rq->flush.list);
-			blk_flush_restore_request(rq);
-			list_add_tail(&rq->queuelist, &q->queue_head);
-		}
-	}
-}
-
-/**
  * blkdev_issue_flush - queue a flush
  * @bdev:	blockdev to issue flush for
  * @gfp_mask:	memory allocation flags (for bio_alloc)

block/blk-mq-tag.c

···
 	return bt_has_free_tags(&tags->bitmap_tags);
 }
 
-static inline void bt_index_inc(unsigned int *index)
+static inline int bt_index_inc(int index)
 {
-	*index = (*index + 1) & (BT_WAIT_QUEUES - 1);
+	return (index + 1) & (BT_WAIT_QUEUES - 1);
+}
+
+static inline void bt_index_atomic_inc(atomic_t *index)
+{
+	int old = atomic_read(index);
+	int new = bt_index_inc(old);
+	atomic_cmpxchg(index, old, new);
 }
 
 /*
···
 	int i, wake_index;
 
 	bt = &tags->bitmap_tags;
-	wake_index = bt->wake_index;
+	wake_index = atomic_read(&bt->wake_index);
 	for (i = 0; i < BT_WAIT_QUEUES; i++) {
 		struct bt_wait_state *bs = &bt->bs[wake_index];
 
 		if (waitqueue_active(&bs->wait))
 			wake_up(&bs->wait);
 
-		bt_index_inc(&wake_index);
+		wake_index = bt_index_inc(wake_index);
 	}
 }
 
···
 					 struct blk_mq_hw_ctx *hctx)
 {
 	struct bt_wait_state *bs;
+	int wait_index;
 
 	if (!hctx)
 		return &bt->bs[0];
 
-	bs = &bt->bs[hctx->wait_index];
-	bt_index_inc(&hctx->wait_index);
+	wait_index = atomic_read(&hctx->wait_index);
+	bs = &bt->bs[wait_index];
+	bt_index_atomic_inc(&hctx->wait_index);
 	return bs;
 }
 
···
 
 	bs = bt_wait_ptr(bt, hctx);
 	do {
-		bool was_empty;
-
-		was_empty = list_empty(&wait.task_list);
 		prepare_to_wait(&bs->wait, &wait, TASK_UNINTERRUPTIBLE);
 
 		tag = __bt_get(hctx, bt, last_tag);
 		if (tag != -1)
 			break;
-
-		if (was_empty)
-			atomic_set(&bs->wait_cnt, bt->wake_cnt);
 
 		blk_mq_put_ctx(data->ctx);
 
···
 {
 	int i, wake_index;
 
-	wake_index = bt->wake_index;
+	wake_index = atomic_read(&bt->wake_index);
 	for (i = 0; i < BT_WAIT_QUEUES; i++) {
 		struct bt_wait_state *bs = &bt->bs[wake_index];
 
 		if (waitqueue_active(&bs->wait)) {
-			if (wake_index != bt->wake_index)
-				bt->wake_index = wake_index;
+			int o = atomic_read(&bt->wake_index);
+			if (wake_index != o)
+				atomic_cmpxchg(&bt->wake_index, o, wake_index);
 
 			return bs;
 		}
 
-		bt_index_inc(&wake_index);
+		wake_index = bt_index_inc(wake_index);
 	}
 
 	return NULL;
···
 {
 	const int index = TAG_TO_INDEX(bt, tag);
 	struct bt_wait_state *bs;
+	int wait_cnt;
 
 	/*
 	 * The unlock memory barrier need to order access to req in free
···
 	clear_bit_unlock(TAG_TO_BIT(bt, tag), &bt->map[index].word);
 
 	bs = bt_wake_ptr(bt);
-	if (bs && atomic_dec_and_test(&bs->wait_cnt)) {
-		atomic_set(&bs->wait_cnt, bt->wake_cnt);
-		bt_index_inc(&bt->wake_index);
+	if (!bs)
+		return;
+
+	wait_cnt = atomic_dec_return(&bs->wait_cnt);
+	if (wait_cnt == 0) {
+wake:
+		atomic_add(bt->wake_cnt, &bs->wait_cnt);
+		bt_index_atomic_inc(&bt->wake_index);
 		wake_up(&bs->wait);
+	} else if (wait_cnt < 0) {
+		wait_cnt = atomic_inc_return(&bs->wait_cnt);
+		if (!wait_cnt)
+			goto wake;
 	}
 }
 
···
 		return -ENOMEM;
 	}
 
-	for (i = 0; i < BT_WAIT_QUEUES; i++)
-		init_waitqueue_head(&bt->bs[i].wait);
-
 	bt_update_count(bt, depth);
+
+	for (i = 0; i < BT_WAIT_QUEUES; i++) {
+		init_waitqueue_head(&bt->bs[i].wait);
+		atomic_set(&bt->bs[i].wait_cnt, bt->wake_cnt);
+	}
+
 	return 0;
 }
 

block/blk-mq-tag.h

···
 	unsigned int map_nr;
 	struct blk_align_bitmap *map;
 
-	unsigned int wake_index;
+	atomic_t wake_index;
 	struct bt_wait_state *bs;
 };
 

block/blk-mq.c

···
 	__percpu_counter_add(&q->mq_usage_counter, -1, 1000000);
 }
 
-static void __blk_mq_drain_queue(struct request_queue *q)
+void blk_mq_drain_queue(struct request_queue *q)
 {
 	while (true) {
 		s64 count;
···
 
 		if (count == 0)
 			break;
-		blk_mq_run_queues(q, false);
+		blk_mq_start_hw_queues(q);
 		msleep(10);
 	}
 }
···
 	spin_unlock_irq(q->queue_lock);
 
 	if (drain)
-		__blk_mq_drain_queue(q);
-}
-
-void blk_mq_drain_queue(struct request_queue *q)
-{
-	__blk_mq_drain_queue(q);
+		blk_mq_drain_queue(q);
 }
 
 static void blk_mq_unfreeze_queue(struct request_queue *q)

block/blk.h

···
 #define ELV_ON_HASH(rq) ((rq)->cmd_flags & REQ_HASHED)
 
 void blk_insert_flush(struct request *rq);
-void blk_abort_flushes(struct request_queue *q);
 
 static inline struct request *__elv_next_request(struct request_queue *q)
 {

block/elevator.c

···
 	return ELV_MQUEUE_MAY;
 }
 
-void elv_abort_queue(struct request_queue *q)
-{
-	struct request *rq;
-
-	blk_abort_flushes(q);
-
-	while (!list_empty(&q->queue_head)) {
-		rq = list_entry_rq(q->queue_head.next);
-		rq->cmd_flags |= REQ_QUIET;
-		trace_block_rq_abort(q, rq);
-		/*
-		 * Mark this request as started so we don't trigger
-		 * any debug logic in the end I/O path.
-		 */
-		blk_start_request(rq);
-		__blk_end_request_all(rq, -EIO);
-	}
-}
-EXPORT_SYMBOL(elv_abort_queue);
-
 void elv_completed_request(struct request_queue *q, struct request *rq)
 {
 	struct elevator_queue *e = q->elevator;

drivers/block/null_blk.c

···
 
 static int queue_mode = NULL_Q_MQ;
 module_param(queue_mode, int, S_IRUGO);
-MODULE_PARM_DESC(use_mq, "Use blk-mq interface (0=bio,1=rq,2=multiqueue)");
+MODULE_PARM_DESC(queue_mode, "Block interface to use (0=bio,1=rq,2=multiqueue)");
 
 static int gb = 250;
 module_param(gb, int, S_IRUGO);
···
 
 static void null_softirq_done_fn(struct request *rq)
 {
-	end_cmd(blk_mq_rq_to_pdu(rq));
+	if (queue_mode == NULL_Q_MQ)
+		end_cmd(blk_mq_rq_to_pdu(rq));
+	else
+		end_cmd(rq->special);
 }
 
 static inline void null_handle_cmd(struct nullb_cmd *cmd)

include/linux/blk-mq.h

···
 	unsigned int		nr_ctx;
 	struct blk_mq_ctx	**ctxs;
 
-	unsigned int		wait_index;
+	atomic_t		wait_index;
 
 	struct blk_mq_tags	*tags;
 

include/linux/blkdev.h

···
 					       sector_t offset)
 {
 	if (!q->limits.chunk_sectors)
-		return q->limits.max_hw_sectors;
+		return q->limits.max_sectors;
 
 	return q->limits.chunk_sectors -
 			(offset & (q->limits.chunk_sectors - 1));

include/linux/elevator.h

···
 extern int elv_register_queue(struct request_queue *q);
 extern void elv_unregister_queue(struct request_queue *q);
 extern int elv_may_queue(struct request_queue *, int);
-extern void elv_abort_queue(struct request_queue *);
 extern void elv_completed_request(struct request_queue *, struct request *);
 extern int elv_set_request(struct request_queue *q, struct request *rq,
 			   struct bio *bio, gfp_t gfp_mask);