11 files changed, 219 insertions, 158 deletions

diff --git a/block/blk-core.c b/block/blk-core.c
index c00e0bd..853f927 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -693,11 +693,20 @@ blk_init_queue_node(request_fn_proc *rfn, spinlock_t *lock, int node_id)
 	if (!uninit_q)
 		return NULL;
 
+	uninit_q->flush_rq = kzalloc(sizeof(struct request), GFP_KERNEL);
+	if (!uninit_q->flush_rq)
+		goto out_cleanup_queue;
+
 	q = blk_init_allocated_queue(uninit_q, rfn, lock);
 	if (!q)
-		blk_cleanup_queue(uninit_q);
-
+		goto out_free_flush_rq;
 	return q;
+
+out_free_flush_rq:
+	kfree(uninit_q->flush_rq);
+out_cleanup_queue:
+	blk_cleanup_queue(uninit_q);
+	return NULL;
 }
 EXPORT_SYMBOL(blk_init_queue_node);
 
@@ -1127,7 +1136,7 @@ static struct request *blk_old_get_request(struct request_queue *q, int rw,
 struct request *blk_get_request(struct request_queue *q, int rw, gfp_t gfp_mask)
 {
 	if (q->mq_ops)
-		return blk_mq_alloc_request(q, rw, gfp_mask, false);
+		return blk_mq_alloc_request(q, rw, gfp_mask);
 	else
 		return blk_old_get_request(q, rw, gfp_mask);
 }
@@ -1278,6 +1287,11 @@ void __blk_put_request(struct request_queue *q, struct request *req)
 	if (unlikely(!q))
 		return;
 
+	if (q->mq_ops) {
+		blk_mq_free_request(req);
+		return;
+	}
+
 	blk_pm_put_request(req);
 
 	elv_completed_request(q, req);
diff --git a/block/blk-exec.c b/block/blk-exec.c
index bbfc072..c68613b 100644
--- a/block/blk-exec.c
+++ b/block/blk-exec.c
@@ -65,7 +65,7 @@ void blk_execute_rq_nowait(struct request_queue *q, struct gendisk *bd_disk,
 	 * be resued after dying flag is set
 	 */
 	if (q->mq_ops) {
-		blk_mq_insert_request(q, rq, true);
+		blk_mq_insert_request(q, rq, at_head, true);
 		return;
 	}
 
diff --git a/block/blk-flush.c b/block/blk-flush.c
index 9288aaf..66e2b69 100644
--- a/block/blk-flush.c
+++ b/block/blk-flush.c
@@ -130,20 +130,26 @@ static void blk_flush_restore_request(struct request *rq)
 	blk_clear_rq_complete(rq);
 }
 
-static void mq_flush_data_run(struct work_struct *work)
+static void mq_flush_run(struct work_struct *work)
 {
 	struct request *rq;
 
-	rq = container_of(work, struct request, mq_flush_data);
+	rq = container_of(work, struct request, mq_flush_work);
 
 	memset(&rq->csd, 0, sizeof(rq->csd));
 	blk_mq_run_request(rq, true, false);
 }
 
-static void blk_mq_flush_data_insert(struct request *rq)
+static bool blk_flush_queue_rq(struct request *rq)
 {
-	INIT_WORK(&rq->mq_flush_data, mq_flush_data_run);
-	kblockd_schedule_work(rq->q, &rq->mq_flush_data);
+	if (rq->q->mq_ops) {
+		INIT_WORK(&rq->mq_flush_work, mq_flush_run);
+		kblockd_schedule_work(rq->q, &rq->mq_flush_work);
+		return false;
+	} else {
+		list_add_tail(&rq->queuelist, &rq->q->queue_head);
+		return true;
+	}
 }
 
 /**
@@ -187,12 +193,7 @@ static bool blk_flush_complete_seq(struct request *rq, unsigned int seq,
 
 	case REQ_FSEQ_DATA:
 		list_move_tail(&rq->flush.list, &q->flush_data_in_flight);
-		if (q->mq_ops)
-			blk_mq_flush_data_insert(rq);
-		else {
-			list_add(&rq->queuelist, &q->queue_head);
-			queued = true;
-		}
+		queued = blk_flush_queue_rq(rq);
 		break;
 
 	case REQ_FSEQ_DONE:
@@ -216,9 +217,6 @@ static bool blk_flush_complete_seq(struct request *rq, unsigned int seq,
 	}
 
 	kicked = blk_kick_flush(q);
-	/* blk_mq_run_flush will run queue */
-	if (q->mq_ops)
-		return queued;
 	return kicked | queued;
 }
 
@@ -230,10 +228,9 @@ static void flush_end_io(struct request *flush_rq, int error)
 	struct request *rq, *n;
 	unsigned long flags = 0;
 
-	if (q->mq_ops) {
-		blk_mq_free_request(flush_rq);
+	if (q->mq_ops)
 		spin_lock_irqsave(&q->mq_flush_lock, flags);
-	}
+
 	running = &q->flush_queue[q->flush_running_idx];
 	BUG_ON(q->flush_pending_idx == q->flush_running_idx);
 
@@ -263,49 +260,14 @@ static void flush_end_io(struct request *flush_rq, int error)
 	 * kblockd.
 	 */
 	if (queued || q->flush_queue_delayed) {
-		if (!q->mq_ops)
-			blk_run_queue_async(q);
-		else
-		/*
-		 * This can be optimized to only run queues with requests
-		 * queued if necessary.
-		 */
-			blk_mq_run_queues(q, true);
+		WARN_ON(q->mq_ops);
+		blk_run_queue_async(q);
 	}
 	q->flush_queue_delayed = 0;
 	if (q->mq_ops)
 		spin_unlock_irqrestore(&q->mq_flush_lock, flags);
 }
 
-static void mq_flush_work(struct work_struct *work)
-{
-	struct request_queue *q;
-	struct request *rq;
-
-	q = container_of(work, struct request_queue, mq_flush_work);
-
-	/* We don't need set REQ_FLUSH_SEQ, it's for consistency */
-	rq = blk_mq_alloc_request(q, WRITE_FLUSH|REQ_FLUSH_SEQ,
-		__GFP_WAIT|GFP_ATOMIC, true);
-	rq->cmd_type = REQ_TYPE_FS;
-	rq->end_io = flush_end_io;
-
-	blk_mq_run_request(rq, true, false);
-}
-
-/*
- * We can't directly use q->flush_rq, because it doesn't have tag and is not in
- * hctx->rqs[]. so we must allocate a new request, since we can't sleep here,
- * so offload the work to workqueue.
- *
- * Note: we assume a flush request finished in any hardware queue will flush
- * the whole disk cache.
- */
-static void mq_run_flush(struct request_queue *q)
-{
-	kblockd_schedule_work(q, &q->mq_flush_work);
-}
-
 /**
  * blk_kick_flush - consider issuing flush request
  * @q: request_queue being kicked
@@ -340,19 +302,31 @@ static bool blk_kick_flush(struct request_queue *q)
 	 * different from running_idx, which means flush is in flight.
 	 */
 	q->flush_pending_idx ^= 1;
+
 	if (q->mq_ops) {
-		mq_run_flush(q);
-		return true;
+		struct blk_mq_ctx *ctx = first_rq->mq_ctx;
+		struct blk_mq_hw_ctx *hctx = q->mq_ops->map_queue(q, ctx->cpu);
+
+		blk_mq_rq_init(hctx, q->flush_rq);
+		q->flush_rq->mq_ctx = ctx;
+
+		/*
+		 * Reuse the tag value from the fist waiting request,
+		 * with blk-mq the tag is generated during request
+		 * allocation and drivers can rely on it being inside
+		 * the range they asked for.
+		 */
+		q->flush_rq->tag = first_rq->tag;
+	} else {
+		blk_rq_init(q, q->flush_rq);
 	}
 
-	blk_rq_init(q, &q->flush_rq);
-	q->flush_rq.cmd_type = REQ_TYPE_FS;
-	q->flush_rq.cmd_flags = WRITE_FLUSH | REQ_FLUSH_SEQ;
-	q->flush_rq.rq_disk = first_rq->rq_disk;
-	q->flush_rq.end_io = flush_end_io;
+	q->flush_rq->cmd_type = REQ_TYPE_FS;
+	q->flush_rq->cmd_flags = WRITE_FLUSH | REQ_FLUSH_SEQ;
+	q->flush_rq->rq_disk = first_rq->rq_disk;
+	q->flush_rq->end_io = flush_end_io;
 
-	list_add_tail(&q->flush_rq.queuelist, &q->queue_head);
-	return true;
+	return blk_flush_queue_rq(q->flush_rq);
 }
 
 static void flush_data_end_io(struct request *rq, int error)
@@ -558,5 +532,4 @@ EXPORT_SYMBOL(blkdev_issue_flush);
 void blk_mq_init_flush(struct request_queue *q)
 {
 	spin_lock_init(&q->mq_flush_lock);
-	INIT_WORK(&q->mq_flush_work, mq_flush_work);
 }
diff --git a/block/blk-lib.c b/block/blk-lib.c
index 2da76c9..97a733c 100644
--- a/block/blk-lib.c
+++ b/block/blk-lib.c
@@ -119,6 +119,14 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
 
 		atomic_inc(&bb.done);
 		submit_bio(type, bio);
+
+		/*
+		 * We can loop for a long time in here, if someone does
+		 * full device discards (like mkfs). Be nice and allow
+		 * us to schedule out to avoid softlocking if preempt
+		 * is disabled.
+		 */
+		cond_resched();
 	}
 	blk_finish_plug(&plug);
 
diff --git a/block/blk-merge.c b/block/blk-merge.c
index 8f8adaa..6c583f9 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -21,6 +21,16 @@ static unsigned int __blk_recalc_rq_segments(struct request_queue *q,
 	if (!bio)
 		return 0;
 
+	/*
+	 * This should probably be returning 0, but blk_add_request_payload()
+	 * (Christoph!!!!)
+	 */
+	if (bio->bi_rw & REQ_DISCARD)
+		return 1;
+
+	if (bio->bi_rw & REQ_WRITE_SAME)
+		return 1;
+
 	fbio = bio;
 	cluster = blk_queue_cluster(q);
 	seg_size = 0;
@@ -161,30 +171,60 @@ new_segment:
 	*bvprv = *bvec;
 }
 
-/*
- * map a request to scatterlist, return number of sg entries setup. Caller
- * must make sure sg can hold rq->nr_phys_segments entries
- */
-int blk_rq_map_sg(struct request_queue *q, struct request *rq,
-		  struct scatterlist *sglist)
+static int __blk_bios_map_sg(struct request_queue *q, struct bio *bio,
+			     struct scatterlist *sglist,
+			     struct scatterlist **sg)
 {
 	struct bio_vec bvec, bvprv = { NULL };
-	struct req_iterator iter;
-	struct scatterlist *sg;
+	struct bvec_iter iter;
 	int nsegs, cluster;
 
 	nsegs = 0;
 	cluster = blk_queue_cluster(q);
 
-	/*
-	 * for each bio in rq
-	 */
-	sg = NULL;
-	rq_for_each_segment(bvec, rq, iter) {
-		__blk_segment_map_sg(q, &bvec, sglist, &bvprv, &sg,
-				     &nsegs, &cluster);
-	} /* segments in rq */
+	if (bio->bi_rw & REQ_DISCARD) {
+		/*
+		 * This is a hack - drivers should be neither modifying the
+		 * biovec, nor relying on bi_vcnt - but because of
+		 * blk_add_request_payload(), a discard bio may or may not have
+		 * a payload we need to set up here (thank you Christoph) and
+		 * bi_vcnt is really the only way of telling if we need to.
+		 */
+
+		if (bio->bi_vcnt)
+			goto single_segment;
+
+		return 0;
+	}
+
+	if (bio->bi_rw & REQ_WRITE_SAME) {
+single_segment:
+		*sg = sglist;
+		bvec = bio_iovec(bio);
+		sg_set_page(*sg, bvec.bv_page, bvec.bv_len, bvec.bv_offset);
+		return 1;
+	}
+
+	for_each_bio(bio)
+		bio_for_each_segment(bvec, bio, iter)
+			__blk_segment_map_sg(q, &bvec, sglist, &bvprv, sg,
+					     &nsegs, &cluster);
 
+	return nsegs;
+}
+
+/*
+ * map a request to scatterlist, return number of sg entries setup. Caller
+ * must make sure sg can hold rq->nr_phys_segments entries
+ */
+int blk_rq_map_sg(struct request_queue *q, struct request *rq,
+		  struct scatterlist *sglist)
+{
+	struct scatterlist *sg = NULL;
+	int nsegs = 0;
+
+	if (rq->bio)
+		nsegs = __blk_bios_map_sg(q, rq->bio, sglist, &sg);
 
 	if (unlikely(rq->cmd_flags & REQ_COPY_USER) &&
 	    (blk_rq_bytes(rq) & q->dma_pad_mask)) {
@@ -230,20 +270,13 @@ EXPORT_SYMBOL(blk_rq_map_sg);
 int blk_bio_map_sg(struct request_queue *q, struct bio *bio,
 		   struct scatterlist *sglist)
 {
-	struct bio_vec bvec, bvprv = { NULL };
-	struct scatterlist *sg;
-	int nsegs, cluster;
-	struct bvec_iter iter;
-
-	nsegs = 0;
-	cluster = blk_queue_cluster(q);
-
-	sg = NULL;
-	bio_for_each_segment(bvec, bio, iter) {
-		__blk_segment_map_sg(q, &bvec, sglist, &bvprv, &sg,
-				     &nsegs, &cluster);
-	} /* segments in bio */
+	struct scatterlist *sg = NULL;
+	int nsegs;
+	struct bio *next = bio->bi_next;
+	bio->bi_next = NULL;
 
+	nsegs = __blk_bios_map_sg(q, bio, sglist, &sg);
+	bio->bi_next = next;
 	if (sg)
 		sg_mark_end(sg);
 
diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c
index 5d70edc..83ae96c 100644
--- a/block/blk-mq-tag.c
+++ b/block/blk-mq-tag.c
@@ -184,7 +184,7 @@ void blk_mq_free_tags(struct blk_mq_tags *tags)
 ssize_t blk_mq_tag_sysfs_show(struct blk_mq_tags *tags, char *page)
 {
 	char *orig_page = page;
-	int cpu;
+	unsigned int cpu;
 
 	if (!tags)
 		return 0;
diff --git a/block/blk-mq.c b/block/blk-mq.c
index 57039fc..1fa9dd1 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -226,15 +226,14 @@ static struct request *blk_mq_alloc_request_pinned(struct request_queue *q,
 	return rq;
 }
 
-struct request *blk_mq_alloc_request(struct request_queue *q, int rw,
-		gfp_t gfp, bool reserved)
+struct request *blk_mq_alloc_request(struct request_queue *q, int rw, gfp_t gfp)
 {
 	struct request *rq;
 
 	if (blk_mq_queue_enter(q))
 		return NULL;
 
-	rq = blk_mq_alloc_request_pinned(q, rw, gfp, reserved);
+	rq = blk_mq_alloc_request_pinned(q, rw, gfp, false);
 	if (rq)
 		blk_mq_put_ctx(rq->mq_ctx);
 	return rq;
@@ -258,7 +257,7 @@ EXPORT_SYMBOL(blk_mq_alloc_reserved_request);
 /*
  * Re-init and set pdu, if we have it
  */
-static void blk_mq_rq_init(struct blk_mq_hw_ctx *hctx, struct request *rq)
+void blk_mq_rq_init(struct blk_mq_hw_ctx *hctx, struct request *rq)
 {
 	blk_rq_init(hctx->queue, rq);
 
@@ -305,7 +304,7 @@ static void blk_mq_bio_endio(struct request *rq, struct bio *bio, int error)
 		bio_endio(bio, error);
 }
 
-void blk_mq_complete_request(struct request *rq, int error)
+void blk_mq_end_io(struct request *rq, int error)
 {
 	struct bio *bio = rq->bio;
 	unsigned int bytes = 0;
@@ -330,48 +329,55 @@ void blk_mq_complete_request(struct request *rq, int error)
 	else
 		blk_mq_free_request(rq);
 }
+EXPORT_SYMBOL(blk_mq_end_io);
 
-void __blk_mq_end_io(struct request *rq, int error)
-{
-	if (!blk_mark_rq_complete(rq))
-		blk_mq_complete_request(rq, error);
-}
-
-static void blk_mq_end_io_remote(void *data)
+static void __blk_mq_complete_request_remote(void *data)
 {
 	struct request *rq = data;
 
-	__blk_mq_end_io(rq, rq->errors);
+	rq->q->softirq_done_fn(rq);
 }
 
-/*
- * End IO on this request on a multiqueue enabled driver. We'll either do
- * it directly inline, or punt to a local IPI handler on the matching
- * remote CPU.
- */
-void blk_mq_end_io(struct request *rq, int error)
+void __blk_mq_complete_request(struct request *rq)
 {
 	struct blk_mq_ctx *ctx = rq->mq_ctx;
 	int cpu;
 
-	if (!ctx->ipi_redirect)
-		return __blk_mq_end_io(rq, error);
+	if (!ctx->ipi_redirect) {
+		rq->q->softirq_done_fn(rq);
+		return;
+	}
 
 	cpu = get_cpu();
 	if (cpu != ctx->cpu && cpu_online(ctx->cpu)) {
-		rq->errors = error;
-		rq->csd.func = blk_mq_end_io_remote;
+		rq->csd.func = __blk_mq_complete_request_remote;
 		rq->csd.info = rq;
 		rq->csd.flags = 0;
 		__smp_call_function_single(ctx->cpu, &rq->csd, 0);
 	} else {
-		__blk_mq_end_io(rq, error);
+		rq->q->softirq_done_fn(rq);
 	}
 	put_cpu();
 }
-EXPORT_SYMBOL(blk_mq_end_io);
 
-static void blk_mq_start_request(struct request *rq)
+/**
+ * blk_mq_complete_request - end I/O on a request
+ * @rq:		the request being processed
+ *
+ * Description:
+ *	Ends all I/O on a request. It does not handle partial completions.
+ *	The actual completion happens out-of-order, through a IPI handler.
+ **/
+void blk_mq_complete_request(struct request *rq)
+{
+	if (unlikely(blk_should_fake_timeout(rq->q)))
+		return;
+	if (!blk_mark_rq_complete(rq))
+		__blk_mq_complete_request(rq);
+}
+EXPORT_SYMBOL(blk_mq_complete_request);
+
+static void blk_mq_start_request(struct request *rq, bool last)
 {
 	struct request_queue *q = rq->q;
 
@@ -384,6 +390,25 @@ static void blk_mq_start_request(struct request *rq)
 	 */
 	rq->deadline = jiffies + q->rq_timeout;
 	set_bit(REQ_ATOM_STARTED, &rq->atomic_flags);
+
+	if (q->dma_drain_size && blk_rq_bytes(rq)) {
+		/*
+		 * Make sure space for the drain appears.  We know we can do
+		 * this because max_hw_segments has been adjusted to be one
+		 * fewer than the device can handle.
+		 */
+		rq->nr_phys_segments++;
+	}
+
+	/*
+	 * Flag the last request in the series so that drivers know when IO
+	 * should be kicked off, if they don't do it on a per-request basis.
+	 *
+	 * Note: the flag isn't the only condition drivers should do kick off.
+	 * If drive is busy, the last request might not have the bit set.
+	 */
+	if (last)
+		rq->cmd_flags |= REQ_END;
 }
 
 static void blk_mq_requeue_request(struct request *rq)
@@ -392,6 +417,11 @@ static void blk_mq_requeue_request(struct request *rq)
 
 	trace_block_rq_requeue(q, rq);
 	clear_bit(REQ_ATOM_STARTED, &rq->atomic_flags);
+
+	rq->cmd_flags &= ~REQ_END;
+
+	if (q->dma_drain_size && blk_rq_bytes(rq))
+		rq->nr_phys_segments--;
 }
 
 struct blk_mq_timeout_data {
@@ -559,19 +589,8 @@ static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx)
 
 		rq = list_first_entry(&rq_list, struct request, queuelist);
 		list_del_init(&rq->queuelist);
-		blk_mq_start_request(rq);
 
-		/*
-		 * Last request in the series. Flag it as such, this
-		 * enables drivers to know when IO should be kicked off,
-		 * if they don't do it on a per-request basis.
-		 *
-		 * Note: the flag isn't the only condition drivers
-		 * should do kick off. If drive is busy, the last
-		 * request might not have the bit set.
-		 */
-		if (list_empty(&rq_list))
-			rq->cmd_flags |= REQ_END;
+		blk_mq_start_request(rq, list_empty(&rq_list));
 
 		ret = q->mq_ops->queue_rq(hctx, rq);
 		switch (ret) {
@@ -589,8 +608,8 @@ static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx)
 			break;
 		default:
 			pr_err("blk-mq: bad return on queue: %d\n", ret);
-			rq->errors = -EIO;
 		case BLK_MQ_RQ_QUEUE_ERROR:
+			rq->errors = -EIO;
 			blk_mq_end_io(rq, rq->errors);
 			break;
 		}
@@ -693,13 +712,16 @@ static void blk_mq_work_fn(struct work_struct *work)
 }
 
 static void __blk_mq_insert_request(struct blk_mq_hw_ctx *hctx,
-				    struct request *rq)
+				    struct request *rq, bool at_head)
 {
 	struct blk_mq_ctx *ctx = rq->mq_ctx;
 
 	trace_block_rq_insert(hctx->queue, rq);
 
-	list_add_tail(&rq->queuelist, &ctx->rq_list);
+	if (at_head)
+		list_add(&rq->queuelist, &ctx->rq_list);
+	else
+		list_add_tail(&rq->queuelist, &ctx->rq_list);
 	blk_mq_hctx_mark_pending(hctx, ctx);
 
 	/*
@@ -709,7 +731,7 @@ static void __blk_mq_insert_request(struct blk_mq_hw_ctx *hctx,
 }
 
 void blk_mq_insert_request(struct request_queue *q, struct request *rq,
-			   bool run_queue)
+			   bool at_head, bool run_queue)
 {
 	struct blk_mq_hw_ctx *hctx;
 	struct blk_mq_ctx *ctx, *current_ctx;
@@ -728,7 +750,7 @@ void blk_mq_insert_request(struct request_queue *q, struct request *rq,
 			rq->mq_ctx = ctx;
 		}
 		spin_lock(&ctx->lock);
-		__blk_mq_insert_request(hctx, rq);
+		__blk_mq_insert_request(hctx, rq, at_head);
 		spin_unlock(&ctx->lock);
 
 		blk_mq_put_ctx(current_ctx);
@@ -760,7 +782,7 @@ void blk_mq_run_request(struct request *rq, bool run_queue, bool async)
 
 	/* ctx->cpu might be offline */
 	spin_lock(&ctx->lock);
-	__blk_mq_insert_request(hctx, rq);
+	__blk_mq_insert_request(hctx, rq, false);
 	spin_unlock(&ctx->lock);
 
 	blk_mq_put_ctx(current_ctx);
@@ -798,7 +820,7 @@ static void blk_mq_insert_requests(struct request_queue *q,
 		rq = list_first_entry(list, struct request, queuelist);
 		list_del_init(&rq->queuelist);
 		rq->mq_ctx = ctx;
-		__blk_mq_insert_request(hctx, rq);
+		__blk_mq_insert_request(hctx, rq, false);
 	}
 	spin_unlock(&ctx->lock);
 
@@ -888,6 +910,11 @@ static void blk_mq_make_request(struct request_queue *q, struct bio *bio)
 
 	blk_queue_bounce(q, &bio);
 
+	if (bio_integrity_enabled(bio) && bio_integrity_prep(bio)) {
+		bio_endio(bio, -EIO);
+		return;
+	}
+
 	if (use_plug && blk_attempt_plug_merge(q, bio, &request_count))
 		return;
 
@@ -950,7 +977,7 @@ static void blk_mq_make_request(struct request_queue *q, struct bio *bio)
 		__blk_mq_free_request(hctx, ctx, rq);
 	else {
 		blk_mq_bio_to_request(rq, bio);
-		__blk_mq_insert_request(hctx, rq);
+		__blk_mq_insert_request(hctx, rq, false);
 	}
 
 	spin_unlock(&ctx->lock);
@@ -1309,15 +1336,6 @@ struct request_queue *blk_mq_init_queue(struct blk_mq_reg *reg,
 		reg->queue_depth = BLK_MQ_MAX_DEPTH;
 	}
 
-	/*
-	 * Set aside a tag for flush requests.  It will only be used while
-	 * another flush request is in progress but outside the driver.
-	 *
-	 * TODO: only allocate if flushes are supported
-	 */
-	reg->queue_depth++;
-	reg->reserved_tags++;
-
 	if (reg->queue_depth < (reg->reserved_tags + BLK_MQ_TAG_MIN))
 		return ERR_PTR(-EINVAL);
 
@@ -1360,17 +1378,27 @@ struct request_queue *blk_mq_init_queue(struct blk_mq_reg *reg,
 	q->mq_ops = reg->ops;
 	q->queue_flags |= QUEUE_FLAG_MQ_DEFAULT;
 
+	q->sg_reserved_size = INT_MAX;
+
 	blk_queue_make_request(q, blk_mq_make_request);
 	blk_queue_rq_timed_out(q, reg->ops->timeout);
 	if (reg->timeout)
 		blk_queue_rq_timeout(q, reg->timeout);
 
+	if (reg->ops->complete)
+		blk_queue_softirq_done(q, reg->ops->complete);
+
 	blk_mq_init_flush(q);
 	blk_mq_init_cpu_queues(q, reg->nr_hw_queues);
 
-	if (blk_mq_init_hw_queues(q, reg, driver_data))
+	q->flush_rq = kzalloc(round_up(sizeof(struct request) + reg->cmd_size,
+				cache_line_size()), GFP_KERNEL);
+	if (!q->flush_rq)
 		goto err_hw;
 
+	if (blk_mq_init_hw_queues(q, reg, driver_data))
+		goto err_flush_rq;
+
 	blk_mq_map_swqueue(q);
 
 	mutex_lock(&all_q_mutex);
@@ -1378,6 +1406,9 @@ struct request_queue *blk_mq_init_queue(struct blk_mq_reg *reg,
 	mutex_unlock(&all_q_mutex);
 
 	return q;
+
+err_flush_rq:
+	kfree(q->flush_rq);
 err_hw:
 	kfree(q->mq_map);
 err_map:
diff --git a/block/blk-mq.h b/block/blk-mq.h
index 5c39179..ed0035c 100644
--- a/block/blk-mq.h
+++ b/block/blk-mq.h
@@ -22,13 +22,13 @@ struct blk_mq_ctx {
 	struct kobject		kobj;
 };
 
-void __blk_mq_end_io(struct request *rq, int error);
-void blk_mq_complete_request(struct request *rq, int error);
+void __blk_mq_complete_request(struct request *rq);
 void blk_mq_run_request(struct request *rq, bool run_queue, bool async);
 void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async);
 void blk_mq_init_flush(struct request_queue *q);
 void blk_mq_drain_queue(struct request_queue *q);
 void blk_mq_free_queue(struct request_queue *q);
+void blk_mq_rq_init(struct blk_mq_hw_ctx *hctx, struct request *rq);
 
 /*
  * CPU hotplug helpers
diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c
index 8095c4a..7500f87 100644
--- a/block/blk-sysfs.c
+++ b/block/blk-sysfs.c
@@ -549,6 +549,8 @@ static void blk_release_queue(struct kobject *kobj)
 	if (q->mq_ops)
 		blk_mq_free_queue(q);
 
+	kfree(q->flush_rq);
+
 	blk_trace_shutdown(q);
 
 	bdi_destroy(&q->backing_dev_info);
diff --git a/block/blk-timeout.c b/block/blk-timeout.c
index bba81c9..d96f7061 100644
--- a/block/blk-timeout.c
+++ b/block/blk-timeout.c
@@ -91,7 +91,7 @@ static void blk_rq_timed_out(struct request *req)
 	case BLK_EH_HANDLED:
 		/* Can we use req->errors here? */
 		if (q->mq_ops)
-			blk_mq_complete_request(req, req->errors);
+			__blk_mq_complete_request(req);
 		else
 			__blk_complete_request(req);
 		break;
diff --git a/block/blk.h b/block/blk.h
index c90e1d8..d23b415 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -113,7 +113,7 @@ static inline struct request *__elv_next_request(struct request_queue *q)
 			q->flush_queue_delayed = 1;
 			return NULL;
 		}
-		if (unlikely(blk_queue_dying(q)) ||
+		if (unlikely(blk_queue_bypass(q)) ||
 		    !q->elevator->type->ops.elevator_dispatch_fn(q, 0))
 			return NULL;
 	}