Merge tag 'for-linus-20190407' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:

 - Fixups for the pf/pcd queue handling (YueHaibing)

 - Revert of the three direct issue changes as they have been proven to
   cause an issue with dm-mpath (Bart)

 - Plug rq_count reset fix (Dongli)

 - io_uring double free in fileset registration error handling (me)

 - Make null_blk handle bad numa node passed in (John)

 - BFQ ifdef fix (Konstantin)

 - Flush queue leak fix (Shenghui)

 - Plug trace fix (Yufen)

* tag 'for-linus-20190407' of git://git.kernel.dk/linux-block:
  xsysace: Fix error handling in ace_setup
  null_blk: prevent crash from bad home_node value
  block: Revert v5.0 blk_mq_request_issue_directly() changes
  paride/pcd: Fix potential NULL pointer dereference and mem leak
  blk-mq: do not reset plug->rq_count before the list is sorted
  paride/pf: Fix potential NULL pointer dereference
  io_uring: fix double free in case of fileset regitration failure
  blk-mq: add trace block plug and unplug for multiple queues
  block: use blk_free_flush_queue() to free hctx->fq in blk_mq_init_hctx
  block/bfq: fix ifdef for CONFIG_BFQ_GROUP_IOSCHED=y

11 files changed, 110 insertions, 75 deletions

diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
index 4c592496a16a..fac188dd78fa 100644
--- a/block/bfq-iosched.c
+++ b/block/bfq-iosched.c
@@ -674,7 +674,7 @@ static bool bfq_symmetric_scenario(struct bfq_data *bfqd)
 	 * at least two nodes.
 	 */
 	return !(varied_queue_weights || multiple_classes_busy
-#ifdef BFQ_GROUP_IOSCHED_ENABLED
+#ifdef CONFIG_BFQ_GROUP_IOSCHED
 	       || bfqd->num_groups_with_pending_reqs > 0
 #endif
 		);
diff --git a/block/bfq-wf2q.c b/block/bfq-wf2q.c
index 63311d1ff1ed..a11bef75483d 100644
--- a/block/bfq-wf2q.c
+++ b/block/bfq-wf2q.c
@@ -1012,7 +1012,7 @@ static void __bfq_activate_entity(struct bfq_entity *entity,
 		entity->on_st = true;
 	}
 
-#ifdef BFQ_GROUP_IOSCHED_ENABLED
+#ifdef CONFIG_BFQ_GROUP_IOSCHED
 	if (!bfq_entity_to_bfqq(entity)) { /* bfq_group */
 		struct bfq_group *bfqg =
 			container_of(entity, struct bfq_group, entity);
diff --git a/block/blk-core.c b/block/blk-core.c
index 4673ebe42255..a55389ba8779 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -1245,8 +1245,6 @@ static int blk_cloned_rq_check_limits(struct request_queue *q,
  */
 blk_status_t blk_insert_cloned_request(struct request_queue *q, struct request *rq)
 {
-	blk_qc_t unused;
-
 	if (blk_cloned_rq_check_limits(q, rq))
 		return BLK_STS_IOERR;
 
@@ -1262,7 +1260,7 @@ blk_status_t blk_insert_cloned_request(struct request_queue *q, struct request *
 	 * bypass a potential scheduler on the bottom device for
 	 * insert.
 	 */
-	return blk_mq_try_issue_directly(rq->mq_hctx, rq, &unused, true, true);
+	return blk_mq_request_issue_directly(rq, true);
 }
 EXPORT_SYMBOL_GPL(blk_insert_cloned_request);
 
diff --git a/block/blk-mq-sched.c b/block/blk-mq-sched.c
index 40905539afed..aa6bc5c02643 100644
--- a/block/blk-mq-sched.c
+++ b/block/blk-mq-sched.c
@@ -423,10 +423,12 @@ void blk_mq_sched_insert_requests(struct blk_mq_hw_ctx *hctx,
 		 * busy in case of 'none' scheduler, and this way may save
 		 * us one extra enqueue & dequeue to sw queue.
 		 */
-		if (!hctx->dispatch_busy && !e && !run_queue_async)
+		if (!hctx->dispatch_busy && !e && !run_queue_async) {
 			blk_mq_try_issue_list_directly(hctx, list);
-		else
-			blk_mq_insert_requests(hctx, ctx, list);
+			if (list_empty(list))
+				return;
+		}
+		blk_mq_insert_requests(hctx, ctx, list);
 	}
 
 	blk_mq_run_hw_queue(hctx, run_queue_async);
diff --git a/block/blk-mq.c b/block/blk-mq.c
index 3ff3d7b49969..a9354835cf51 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -1711,11 +1711,12 @@ void blk_mq_flush_plug_list(struct blk_plug *plug, bool from_schedule)
 	unsigned int depth;
 
 	list_splice_init(&plug->mq_list, &list);
-	plug->rq_count = 0;
 
 	if (plug->rq_count > 2 && plug->multiple_queues)
 		list_sort(NULL, &list, plug_rq_cmp);
 
+	plug->rq_count = 0;
+
 	this_q = NULL;
 	this_hctx = NULL;
 	this_ctx = NULL;
@@ -1800,74 +1801,76 @@ static blk_status_t __blk_mq_issue_directly(struct blk_mq_hw_ctx *hctx,
 	return ret;
 }
 
-blk_status_t blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx,
+static blk_status_t __blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx,
 						struct request *rq,
 						blk_qc_t *cookie,
-						bool bypass, bool last)
+						bool bypass_insert, bool last)
 {
 	struct request_queue *q = rq->q;
 	bool run_queue = true;
-	blk_status_t ret = BLK_STS_RESOURCE;
-	int srcu_idx;
-	bool force = false;
 
-	hctx_lock(hctx, &srcu_idx);
 	/*
-	 * hctx_lock is needed before checking quiesced flag.
+	 * RCU or SRCU read lock is needed before checking quiesced flag.
 	 *
-	 * When queue is stopped or quiesced, ignore 'bypass', insert
-	 * and return BLK_STS_OK to caller, and avoid driver to try to
-	 * dispatch again.
+	 * When queue is stopped or quiesced, ignore 'bypass_insert' from
+	 * blk_mq_request_issue_directly(), and return BLK_STS_OK to caller,
+	 * and avoid driver to try to dispatch again.
 	 */
-	if (unlikely(blk_mq_hctx_stopped(hctx) || blk_queue_quiesced(q))) {
+	if (blk_mq_hctx_stopped(hctx) || blk_queue_quiesced(q)) {
 		run_queue = false;
-		bypass = false;
-		goto out_unlock;
+		bypass_insert = false;
+		goto insert;
 	}
 
-	if (unlikely(q->elevator && !bypass))
-		goto out_unlock;
+	if (q->elevator && !bypass_insert)
+		goto insert;
 
 	if (!blk_mq_get_dispatch_budget(hctx))
-		goto out_unlock;
+		goto insert;
 
 	if (!blk_mq_get_driver_tag(rq)) {
 		blk_mq_put_dispatch_budget(hctx);
-		goto out_unlock;
+		goto insert;
 	}
 
-	/*
-	 * Always add a request that has been through
-	 *.queue_rq() to the hardware dispatch list.
-	 */
-	force = true;
-	ret = __blk_mq_issue_directly(hctx, rq, cookie, last);
-out_unlock:
+	return __blk_mq_issue_directly(hctx, rq, cookie, last);
+insert:
+	if (bypass_insert)
+		return BLK_STS_RESOURCE;
+
+	blk_mq_request_bypass_insert(rq, run_queue);
+	return BLK_STS_OK;
+}
+
+static void blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx,
+		struct request *rq, blk_qc_t *cookie)
+{
+	blk_status_t ret;
+	int srcu_idx;
+
+	might_sleep_if(hctx->flags & BLK_MQ_F_BLOCKING);
+
+	hctx_lock(hctx, &srcu_idx);
+
+	ret = __blk_mq_try_issue_directly(hctx, rq, cookie, false, true);
+	if (ret == BLK_STS_RESOURCE || ret == BLK_STS_DEV_RESOURCE)
+		blk_mq_request_bypass_insert(rq, true);
+	else if (ret != BLK_STS_OK)
+		blk_mq_end_request(rq, ret);
+
+	hctx_unlock(hctx, srcu_idx);
+}
+
+blk_status_t blk_mq_request_issue_directly(struct request *rq, bool last)
+{
+	blk_status_t ret;
+	int srcu_idx;
+	blk_qc_t unused_cookie;
+	struct blk_mq_hw_ctx *hctx = rq->mq_hctx;
+
+	hctx_lock(hctx, &srcu_idx);
+	ret = __blk_mq_try_issue_directly(hctx, rq, &unused_cookie, true, last);
 	hctx_unlock(hctx, srcu_idx);
-	switch (ret) {
-	case BLK_STS_OK:
-		break;
-	case BLK_STS_DEV_RESOURCE:
-	case BLK_STS_RESOURCE:
-		if (force) {
-			blk_mq_request_bypass_insert(rq, run_queue);
-			/*
-			 * We have to return BLK_STS_OK for the DM
-			 * to avoid livelock. Otherwise, we return
-			 * the real result to indicate whether the
-			 * request is direct-issued successfully.
-			 */
-			ret = bypass ? BLK_STS_OK : ret;
-		} else if (!bypass) {
-			blk_mq_sched_insert_request(rq, false,
-						    run_queue, false);
-		}
-		break;
-	default:
-		if (!bypass)
-			blk_mq_end_request(rq, ret);
-		break;
-	}
 
 	return ret;
 }
@@ -1875,20 +1878,22 @@ out_unlock:
 void blk_mq_try_issue_list_directly(struct blk_mq_hw_ctx *hctx,
 		struct list_head *list)
 {
-	blk_qc_t unused;
-	blk_status_t ret = BLK_STS_OK;
-
 	while (!list_empty(list)) {
+		blk_status_t ret;
 		struct request *rq = list_first_entry(list, struct request,
 				queuelist);
 
 		list_del_init(&rq->queuelist);
-		if (ret == BLK_STS_OK)
-			ret = blk_mq_try_issue_directly(hctx, rq, &unused,
-							false,
+		ret = blk_mq_request_issue_directly(rq, list_empty(list));
+		if (ret != BLK_STS_OK) {
+			if (ret == BLK_STS_RESOURCE ||
+					ret == BLK_STS_DEV_RESOURCE) {
+				blk_mq_request_bypass_insert(rq,
 							list_empty(list));
-		else
-			blk_mq_sched_insert_request(rq, false, true, false);
+				break;
+			}
+			blk_mq_end_request(rq, ret);
+		}
 	}
 
 	/*
@@ -1896,7 +1901,7 @@ void blk_mq_try_issue_list_directly(struct blk_mq_hw_ctx *hctx,
 	 * the driver there was more coming, but that turned out to
 	 * be a lie.
 	 */
-	if (ret != BLK_STS_OK && hctx->queue->mq_ops->commit_rqs)
+	if (!list_empty(list) && hctx->queue->mq_ops->commit_rqs)
 		hctx->queue->mq_ops->commit_rqs(hctx);
 }
 
@@ -2003,19 +2008,21 @@ static blk_qc_t blk_mq_make_request(struct request_queue *q, struct bio *bio)
 			plug->rq_count--;
 		}
 		blk_add_rq_to_plug(plug, rq);
+		trace_block_plug(q);
 
 		blk_mq_put_ctx(data.ctx);
 
 		if (same_queue_rq) {
 			data.hctx = same_queue_rq->mq_hctx;
+			trace_block_unplug(q, 1, true);
 			blk_mq_try_issue_directly(data.hctx, same_queue_rq,
-					&cookie, false, true);
+					&cookie);
 		}
 	} else if ((q->nr_hw_queues > 1 && is_sync) || (!q->elevator &&
 			!data.hctx->dispatch_busy)) {
 		blk_mq_put_ctx(data.ctx);
 		blk_mq_bio_to_request(rq, bio);
-		blk_mq_try_issue_directly(data.hctx, rq, &cookie, false, true);
+		blk_mq_try_issue_directly(data.hctx, rq, &cookie);
 	} else {
 		blk_mq_put_ctx(data.ctx);
 		blk_mq_bio_to_request(rq, bio);
@@ -2332,7 +2339,7 @@ static int blk_mq_init_hctx(struct request_queue *q,
 	return 0;
 
  free_fq:
-	kfree(hctx->fq);
+	blk_free_flush_queue(hctx->fq);
  exit_hctx:
 	if (set->ops->exit_hctx)
 		set->ops->exit_hctx(hctx, hctx_idx);
diff --git a/block/blk-mq.h b/block/blk-mq.h
index d704fc7766f4..423ea88ab6fb 100644
--- a/block/blk-mq.h
+++ b/block/blk-mq.h
@@ -70,10 +70,8 @@ void blk_mq_request_bypass_insert(struct request *rq, bool run_queue);
 void blk_mq_insert_requests(struct blk_mq_hw_ctx *hctx, struct blk_mq_ctx *ctx,
 				struct list_head *list);
 
-blk_status_t blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx,
-						struct request *rq,
-						blk_qc_t *cookie,
-						bool bypass, bool last);
+/* Used by blk_insert_cloned_request() to issue request directly */
+blk_status_t blk_mq_request_issue_directly(struct request *rq, bool last);
 void blk_mq_try_issue_list_directly(struct blk_mq_hw_ctx *hctx,
 				    struct list_head *list);
 
diff --git a/drivers/block/null_blk_main.c b/drivers/block/null_blk_main.c
index 417a9f15c116..d7ac09c092f2 100644
--- a/drivers/block/null_blk_main.c
+++ b/drivers/block/null_blk_main.c
@@ -1748,6 +1748,11 @@ static int __init null_init(void)
 		return -EINVAL;
 	}
 
+	if (g_home_node != NUMA_NO_NODE && g_home_node >= nr_online_nodes) {
+		pr_err("null_blk: invalid home_node value\n");
+		g_home_node = NUMA_NO_NODE;
+	}
+
 	if (g_queue_mode == NULL_Q_RQ) {
 		pr_err("null_blk: legacy IO path no longer available\n");
 		return -EINVAL;
diff --git a/drivers/block/paride/pcd.c b/drivers/block/paride/pcd.c
index 377a694dc228..6d415b20fb70 100644
--- a/drivers/block/paride/pcd.c
+++ b/drivers/block/paride/pcd.c
@@ -314,6 +314,7 @@ static void pcd_init_units(void)
 		disk->queue = blk_mq_init_sq_queue(&cd->tag_set, &pcd_mq_ops,
 						   1, BLK_MQ_F_SHOULD_MERGE);
 		if (IS_ERR(disk->queue)) {
+			put_disk(disk);
 			disk->queue = NULL;
 			continue;
 		}
@@ -750,6 +751,8 @@ static int pcd_detect(void)
 
 	printk("%s: No CD-ROM drive found\n", name);
 	for (unit = 0, cd = pcd; unit < PCD_UNITS; unit++, cd++) {
+		if (!cd->disk)
+			continue;
 		blk_cleanup_queue(cd->disk->queue);
 		cd->disk->queue = NULL;
 		blk_mq_free_tag_set(&cd->tag_set);
@@ -1010,8 +1013,14 @@ static int __init pcd_init(void)
 	pcd_probe_capabilities();
 
 	if (register_blkdev(major, name)) {
-		for (unit = 0, cd = pcd; unit < PCD_UNITS; unit++, cd++)
+		for (unit = 0, cd = pcd; unit < PCD_UNITS; unit++, cd++) {
+			if (!cd->disk)
+				continue;
+
+			blk_cleanup_queue(cd->disk->queue);
+			blk_mq_free_tag_set(&cd->tag_set);
 			put_disk(cd->disk);
+		}
 		return -EBUSY;
 	}
 
@@ -1032,6 +1041,9 @@ static void __exit pcd_exit(void)
 	int unit;
 
 	for (unit = 0, cd = pcd; unit < PCD_UNITS; unit++, cd++) {
+		if (!cd->disk)
+			continue;
+
 		if (cd->present) {
 			del_gendisk(cd->disk);
 			pi_release(cd->pi);
diff --git a/drivers/block/paride/pf.c b/drivers/block/paride/pf.c
index 103b617cdc31..35e6e271b219 100644
--- a/drivers/block/paride/pf.c
+++ b/drivers/block/paride/pf.c
@@ -762,6 +762,8 @@ static int pf_detect(void)
 
 	printk("%s: No ATAPI disk detected\n", name);
 	for (pf = units, unit = 0; unit < PF_UNITS; pf++, unit++) {
+		if (!pf->disk)
+			continue;
 		blk_cleanup_queue(pf->disk->queue);
 		pf->disk->queue = NULL;
 		blk_mq_free_tag_set(&pf->tag_set);
@@ -1029,8 +1031,13 @@ static int __init pf_init(void)
 	pf_busy = 0;
 
 	if (register_blkdev(major, name)) {
-		for (pf = units, unit = 0; unit < PF_UNITS; pf++, unit++)
+		for (pf = units, unit = 0; unit < PF_UNITS; pf++, unit++) {
+			if (!pf->disk)
+				continue;
+			blk_cleanup_queue(pf->disk->queue);
+			blk_mq_free_tag_set(&pf->tag_set);
 			put_disk(pf->disk);
+		}
 		return -EBUSY;
 	}
 
@@ -1051,6 +1058,9 @@ static void __exit pf_exit(void)
 	int unit;
 	unregister_blkdev(major, name);
 	for (pf = units, unit = 0; unit < PF_UNITS; pf++, unit++) {
+		if (!pf->disk)
+			continue;
+
 		if (pf->present)
 			del_gendisk(pf->disk);
 
diff --git a/drivers/block/xsysace.c b/drivers/block/xsysace.c
index 87ccef4bd69e..32a21b8d1d85 100644
--- a/drivers/block/xsysace.c
+++ b/drivers/block/xsysace.c
@@ -1090,6 +1090,8 @@ static int ace_setup(struct ace_device *ace)
 	return 0;
 
 err_read:
+	/* prevent double queue cleanup */
+	ace->gd->queue = NULL;
 	put_disk(ace->gd);
 err_alloc_disk:
 	blk_cleanup_queue(ace->queue);
diff --git a/fs/io_uring.c b/fs/io_uring.c
index bbdbd56cf2ac..07d6ef195d05 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -2215,6 +2215,7 @@ static int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg,
 			fput(ctx->user_files[i]);
 
 		kfree(ctx->user_files);
+		ctx->user_files = NULL;
 		ctx->nr_user_files = 0;
 		return ret;
 	}