Merge tag 'block-5.11-2020-12-23' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:
 "A few stragglers in here, but mostly just straight fixes. In
  particular:

   - Set of rnbd fixes for issues around changes for the merge window
     (Gioh, Jack, Md Haris Iqbal)

   - iocost tracepoint addition (Baolin)

   - Copyright/maintainers update (Christoph)

   - Remove old blk-mq fast path CPU warning (Daniel)

   - loop max_part fix (Josh)

   - Remote IPI threaded IRQ fix (Sebastian)

   - dasd stable fixes (Stefan)

   - bcache merge window fixup and style fixup (Yi, Zheng)"

* tag 'block-5.11-2020-12-23' of git://git.kernel.dk/linux-block:
  md/bcache: convert comma to semicolon
  bcache:remove a superfluous check in register_bcache
  block: update some copyrights
  block: remove a pointless self-reference in block_dev.c
  MAINTAINERS: add fs/block_dev.c to the block section
  blk-mq: Don't complete on a remote CPU in force threaded mode
  s390/dasd: fix list corruption of lcu list
  s390/dasd: fix list corruption of pavgroup group list
  s390/dasd: prevent inconsistent LCU device data
  s390/dasd: fix hanging device offline processing
  blk-iocost: Add iocg idle state tracepoint
  nbd: Respect max_part for all partition scans
  block/rnbd-clt: Does not request pdu to rtrs-clt
  block/rnbd-clt: Dynamically allocate sglist for rnbd_iu
  block/rnbd: Set write-back cache and fua same to the target device
  block/rnbd: Fix typos
  block/rnbd-srv: Protect dev session sysfs removal
  block/rnbd-clt: Fix possible memleak
  block/rnbd-clt: Get rid of warning regarding size argument in strlcpy
  blk-mq: Remove 'running from the wrong CPU' warning

18 files changed, 145 insertions, 94 deletions

diff --git a/MAINTAINERS b/MAINTAINERS
index ad0e34bf8453..32944ecc5733 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -3199,6 +3199,7 @@ S:	Maintained
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git
 F:	block/
 F:	drivers/block/
+F:	fs/block_dev.c
 F:	include/linux/blk*
 F:	kernel/trace/blktrace.c
 F:	lib/sbitmap.c
diff --git a/block/blk-iocost.c b/block/blk-iocost.c
index ffa418c0dcb1..ac6078a34939 100644
--- a/block/blk-iocost.c
+++ b/block/blk-iocost.c
@@ -2185,6 +2185,9 @@ static int ioc_check_iocgs(struct ioc *ioc, struct ioc_now *now)
 							    WEIGHT_ONE);
 			}
 
+			TRACE_IOCG_PATH(iocg_idle, iocg, now,
+					atomic64_read(&iocg->active_period),
+					atomic64_read(&ioc->cur_period), vtime);
 			__propagate_weights(iocg, 0, 0, false, now);
 			list_del_init(&iocg->active_list);
 		}
diff --git a/block/blk-mq.c b/block/blk-mq.c
index b09ce00cc6af..c338c9bc5a2c 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -650,6 +650,14 @@ static inline bool blk_mq_complete_need_ipi(struct request *rq)
 	if (!IS_ENABLED(CONFIG_SMP) ||
 	    !test_bit(QUEUE_FLAG_SAME_COMP, &rq->q->queue_flags))
 		return false;
+	/*
+	 * With force threaded interrupts enabled, raising softirq from an SMP
+	 * function call will always result in waking the ksoftirqd thread.
+	 * This is probably worse than completing the request on a different
+	 * cache domain.
+	 */
+	if (force_irqthreads)
+		return false;
 
 	/* same CPU or cache domain?  Complete locally */
 	if (cpu == rq->mq_ctx->cpu ||
@@ -1496,31 +1504,6 @@ static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx)
 	int srcu_idx;
 
 	/*
-	 * We should be running this queue from one of the CPUs that
-	 * are mapped to it.
-	 *
-	 * There are at least two related races now between setting
-	 * hctx->next_cpu from blk_mq_hctx_next_cpu() and running
-	 * __blk_mq_run_hw_queue():
-	 *
-	 * - hctx->next_cpu is found offline in blk_mq_hctx_next_cpu(),
-	 *   but later it becomes online, then this warning is harmless
-	 *   at all
-	 *
-	 * - hctx->next_cpu is found online in blk_mq_hctx_next_cpu(),
-	 *   but later it becomes offline, then the warning can't be
-	 *   triggered, and we depend on blk-mq timeout handler to
-	 *   handle dispatched requests to this hctx
-	 */
-	if (!cpumask_test_cpu(raw_smp_processor_id(), hctx->cpumask) &&
-		cpu_online(hctx->next_cpu)) {
-		printk(KERN_WARNING "run queue from wrong CPU %d, hctx %s\n",
-			raw_smp_processor_id(),
-			cpumask_empty(hctx->cpumask) ? "inactive": "active");
-		dump_stack();
-	}
-
-	/*
 	 * We can't run the queue inline with ints disabled. Ensure that
 	 * we catch bad users of this early.
 	 */
diff --git a/block/genhd.c b/block/genhd.c
index b84b8671e627..73faec438e49 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -1,6 +1,8 @@
 // SPDX-License-Identifier: GPL-2.0
 /*
  *  gendisk handling
+ *
+ * Portions Copyright (C) 2020 Christoph Hellwig
  */
 
 #include <linux/module.h>
diff --git a/block/partitions/core.c b/block/partitions/core.c
index deca253583bd..e7d776db803b 100644
--- a/block/partitions/core.c
+++ b/block/partitions/core.c
@@ -2,6 +2,7 @@
 /*
  * Copyright (C) 1991-1998  Linus Torvalds
  * Re-organised Feb 1998 Russell King
+ * Copyright (C) 2020 Christoph Hellwig
  */
 #include <linux/fs.h>
 #include <linux/slab.h>
diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
index 92f84ed0ba9e..6727358e147d 100644
--- a/drivers/block/nbd.c
+++ b/drivers/block/nbd.c
@@ -318,7 +318,8 @@ static int nbd_set_size(struct nbd_device *nbd, loff_t bytesize,
 	blk_queue_logical_block_size(nbd->disk->queue, blksize);
 	blk_queue_physical_block_size(nbd->disk->queue, blksize);
 
-	set_bit(GD_NEED_PART_SCAN, &nbd->disk->state);
+	if (max_part)
+		set_bit(GD_NEED_PART_SCAN, &nbd->disk->state);
 	if (!set_capacity_and_notify(nbd->disk, bytesize >> 9))
 		kobject_uevent(&nbd_to_dev(nbd)->kobj, KOBJ_CHANGE);
 	return 0;
@@ -1476,9 +1477,11 @@ static int nbd_open(struct block_device *bdev, fmode_t mode)
 		refcount_set(&nbd->config_refs, 1);
 		refcount_inc(&nbd->refs);
 		mutex_unlock(&nbd->config_lock);
-		set_bit(GD_NEED_PART_SCAN, &bdev->bd_disk->state);
+		if (max_part)
+			set_bit(GD_NEED_PART_SCAN, &bdev->bd_disk->state);
 	} else if (nbd_disconnected(nbd->config)) {
-		set_bit(GD_NEED_PART_SCAN, &bdev->bd_disk->state);
+		if (max_part)
+			set_bit(GD_NEED_PART_SCAN, &bdev->bd_disk->state);
 	}
 out:
 	mutex_unlock(&nbd_index_mutex);
diff --git a/drivers/block/rnbd/rnbd-clt-sysfs.c b/drivers/block/rnbd/rnbd-clt-sysfs.c
index a7caeedeb198..d4aa6bfc9555 100644
--- a/drivers/block/rnbd/rnbd-clt-sysfs.c
+++ b/drivers/block/rnbd/rnbd-clt-sysfs.c
@@ -432,7 +432,7 @@ void rnbd_clt_remove_dev_symlink(struct rnbd_clt_dev *dev)
 	 * i.e. rnbd_clt_unmap_dev_store() leading to a sysfs warning because
 	 * of sysfs link already was removed already.
 	 */
-	if (strlen(dev->blk_symlink_name) && try_module_get(THIS_MODULE)) {
+	if (dev->blk_symlink_name && try_module_get(THIS_MODULE)) {
 		sysfs_remove_link(rnbd_devs_kobj, dev->blk_symlink_name);
 		kfree(dev->blk_symlink_name);
 		module_put(THIS_MODULE);
@@ -521,7 +521,8 @@ static int rnbd_clt_add_dev_symlink(struct rnbd_clt_dev *dev)
 	return 0;
 
 out_err:
-	dev->blk_symlink_name[0] = '\0';
+	kfree(dev->blk_symlink_name);
+	dev->blk_symlink_name = NULL ;
 	return ret;
 }
 
diff --git a/drivers/block/rnbd/rnbd-clt.c b/drivers/block/rnbd/rnbd-clt.c
index a199b190c73d..96e3f9fe8241 100644
--- a/drivers/block/rnbd/rnbd-clt.c
+++ b/drivers/block/rnbd/rnbd-clt.c
@@ -88,6 +88,8 @@ static int rnbd_clt_set_dev_attr(struct rnbd_clt_dev *dev,
 	dev->discard_alignment	    = le32_to_cpu(rsp->discard_alignment);
 	dev->secure_discard	    = le16_to_cpu(rsp->secure_discard);
 	dev->rotational		    = rsp->rotational;
+	dev->wc 		    = !!(rsp->cache_policy & RNBD_WRITEBACK);
+	dev->fua		    = !!(rsp->cache_policy & RNBD_FUA);
 
 	dev->max_hw_sectors = sess->max_io_size / SECTOR_SIZE;
 	dev->max_segments = BMAX_SEGMENTS;
@@ -347,19 +349,26 @@ static struct rnbd_iu *rnbd_get_iu(struct rnbd_clt_session *sess,
 	struct rnbd_iu *iu;
 	struct rtrs_permit *permit;
 
+	iu = kzalloc(sizeof(*iu), GFP_KERNEL);
+	if (!iu) {
+		return NULL;
+	}
+
 	permit = rnbd_get_permit(sess, con_type,
 				  wait ? RTRS_PERMIT_WAIT :
 				  RTRS_PERMIT_NOWAIT);
-	if (unlikely(!permit))
+	if (unlikely(!permit)) {
+		kfree(iu);
 		return NULL;
-	iu = rtrs_permit_to_pdu(permit);
+	}
+
 	iu->permit = permit;
 	/*
 	 * 1st reference is dropped after finishing sending a "user" message,
 	 * 2nd reference is dropped after confirmation with the response is
 	 * returned.
 	 * 1st and 2nd can happen in any order, so the rnbd_iu should be
-	 * released (rtrs_permit returned to ibbtrs) only leased after both
+	 * released (rtrs_permit returned to rtrs) only after both
 	 * are finished.
 	 */
 	atomic_set(&iu->refcount, 2);
@@ -371,8 +380,10 @@ static struct rnbd_iu *rnbd_get_iu(struct rnbd_clt_session *sess,
 
 static void rnbd_put_iu(struct rnbd_clt_session *sess, struct rnbd_iu *iu)
 {
-	if (atomic_dec_and_test(&iu->refcount))
+	if (atomic_dec_and_test(&iu->refcount)) {
 		rnbd_put_permit(sess, iu->permit);
+		kfree(iu);
+	}
 }
 
 static void rnbd_softirq_done_fn(struct request *rq)
@@ -382,6 +393,7 @@ static void rnbd_softirq_done_fn(struct request *rq)
 	struct rnbd_iu *iu;
 
 	iu = blk_mq_rq_to_pdu(rq);
+	sg_free_table_chained(&iu->sgt, RNBD_INLINE_SG_CNT);
 	rnbd_put_permit(sess, iu->permit);
 	blk_mq_end_request(rq, errno_to_blk_status(iu->errno));
 }
@@ -475,7 +487,7 @@ static int send_msg_close(struct rnbd_clt_dev *dev, u32 device_id, bool wait)
 	iu->buf = NULL;
 	iu->dev = dev;
 
-	sg_mark_end(&iu->sglist[0]);
+	sg_alloc_table(&iu->sgt, 1, GFP_KERNEL);
 
 	msg.hdr.type	= cpu_to_le16(RNBD_MSG_CLOSE);
 	msg.device_id	= cpu_to_le32(device_id);
@@ -490,6 +502,7 @@ static int send_msg_close(struct rnbd_clt_dev *dev, u32 device_id, bool wait)
 		err = errno;
 	}
 
+	sg_free_table(&iu->sgt);
 	rnbd_put_iu(sess, iu);
 	return err;
 }
@@ -562,7 +575,8 @@ static int send_msg_open(struct rnbd_clt_dev *dev, bool wait)
 	iu->buf = rsp;
 	iu->dev = dev;
 
-	sg_init_one(iu->sglist, rsp, sizeof(*rsp));
+	sg_alloc_table(&iu->sgt, 1, GFP_KERNEL);
+	sg_init_one(iu->sgt.sgl, rsp, sizeof(*rsp));
 
 	msg.hdr.type	= cpu_to_le16(RNBD_MSG_OPEN);
 	msg.access_mode	= dev->access_mode;
@@ -570,7 +584,7 @@ static int send_msg_open(struct rnbd_clt_dev *dev, bool wait)
 
 	WARN_ON(!rnbd_clt_get_dev(dev));
 	err = send_usr_msg(sess->rtrs, READ, iu,
-			   &vec, sizeof(*rsp), iu->sglist, 1,
+			   &vec, sizeof(*rsp), iu->sgt.sgl, 1,
 			   msg_open_conf, &errno, wait);
 	if (err) {
 		rnbd_clt_put_dev(dev);
@@ -580,6 +594,7 @@ static int send_msg_open(struct rnbd_clt_dev *dev, bool wait)
 		err = errno;
 	}
 
+	sg_free_table(&iu->sgt);
 	rnbd_put_iu(sess, iu);
 	return err;
 }
@@ -608,7 +623,8 @@ static int send_msg_sess_info(struct rnbd_clt_session *sess, bool wait)
 	iu->buf = rsp;
 	iu->sess = sess;
 
-	sg_init_one(iu->sglist, rsp, sizeof(*rsp));
+	sg_alloc_table(&iu->sgt, 1, GFP_KERNEL);
+	sg_init_one(iu->sgt.sgl, rsp, sizeof(*rsp));
 
 	msg.hdr.type = cpu_to_le16(RNBD_MSG_SESS_INFO);
 	msg.ver      = RNBD_PROTO_VER_MAJOR;
@@ -624,7 +640,7 @@ static int send_msg_sess_info(struct rnbd_clt_session *sess, bool wait)
 		goto put_iu;
 	}
 	err = send_usr_msg(sess->rtrs, READ, iu,
-			   &vec, sizeof(*rsp), iu->sglist, 1,
+			   &vec, sizeof(*rsp), iu->sgt.sgl, 1,
 			   msg_sess_info_conf, &errno, wait);
 	if (err) {
 		rnbd_clt_put_sess(sess);
@@ -634,7 +650,7 @@ put_iu:
 	} else {
 		err = errno;
 	}
-
+	sg_free_table(&iu->sgt);
 	rnbd_put_iu(sess, iu);
 	return err;
 }
@@ -803,7 +819,7 @@ static struct rnbd_clt_session *alloc_sess(const char *sessname)
 	rnbd_init_cpu_qlists(sess->cpu_queues);
 
 	/*
-	 * That is simple percpu variable which stores cpu indeces, which are
+	 * That is simple percpu variable which stores cpu indices, which are
 	 * incremented on each access.  We need that for the sake of fairness
 	 * to wake up queues in a round-robin manner.
 	 */
@@ -1014,11 +1030,10 @@ static int rnbd_client_xfer_request(struct rnbd_clt_dev *dev,
 	 * See queue limits.
 	 */
 	if (req_op(rq) != REQ_OP_DISCARD)
-		sg_cnt = blk_rq_map_sg(dev->queue, rq, iu->sglist);
+		sg_cnt = blk_rq_map_sg(dev->queue, rq, iu->sgt.sgl);
 
 	if (sg_cnt == 0)
-		/* Do not forget to mark the end */
-		sg_mark_end(&iu->sglist[0]);
+		sg_mark_end(&iu->sgt.sgl[0]);
 
 	msg.hdr.type	= cpu_to_le16(RNBD_MSG_IO);
 	msg.device_id	= cpu_to_le32(dev->device_id);
@@ -1027,13 +1042,13 @@ static int rnbd_client_xfer_request(struct rnbd_clt_dev *dev,
 		.iov_base = &msg,
 		.iov_len  = sizeof(msg)
 	};
-	size = rnbd_clt_get_sg_size(iu->sglist, sg_cnt);
+	size = rnbd_clt_get_sg_size(iu->sgt.sgl, sg_cnt);
 	req_ops = (struct rtrs_clt_req_ops) {
 		.priv = iu,
 		.conf_fn = msg_io_conf,
 	};
 	err = rtrs_clt_request(rq_data_dir(rq), &req_ops, rtrs, permit,
-			       &vec, 1, size, iu->sglist, sg_cnt);
+			       &vec, 1, size, iu->sgt.sgl, sg_cnt);
 	if (unlikely(err)) {
 		rnbd_clt_err_rl(dev, "RTRS failed to transfer IO, err: %d\n",
 				 err);
@@ -1120,6 +1135,7 @@ static blk_status_t rnbd_queue_rq(struct blk_mq_hw_ctx *hctx,
 	struct rnbd_clt_dev *dev = rq->rq_disk->private_data;
 	struct rnbd_iu *iu = blk_mq_rq_to_pdu(rq);
 	int err;
+	blk_status_t ret = BLK_STS_IOERR;
 
 	if (unlikely(dev->dev_state != DEV_STATE_MAPPED))
 		return BLK_STS_IOERR;
@@ -1131,32 +1147,35 @@ static blk_status_t rnbd_queue_rq(struct blk_mq_hw_ctx *hctx,
 		return BLK_STS_RESOURCE;
 	}
 
+	iu->sgt.sgl = iu->first_sgl;
+	err = sg_alloc_table_chained(&iu->sgt,
+				     /* Even-if the request has no segment,
+				      * sglist must have one entry at least */
+				     blk_rq_nr_phys_segments(rq) ? : 1,
+				     iu->sgt.sgl,
+				     RNBD_INLINE_SG_CNT);
+	if (err) {
+		rnbd_clt_err_rl(dev, "sg_alloc_table_chained ret=%d\n", err);
+		rnbd_clt_dev_kick_mq_queue(dev, hctx, 10/*ms*/);
+		rnbd_put_permit(dev->sess, iu->permit);
+		return BLK_STS_RESOURCE;
+	}
+
 	blk_mq_start_request(rq);
 	err = rnbd_client_xfer_request(dev, rq, iu);
 	if (likely(err == 0))
 		return BLK_STS_OK;
 	if (unlikely(err == -EAGAIN || err == -ENOMEM)) {
 		rnbd_clt_dev_kick_mq_queue(dev, hctx, 10/*ms*/);
-		rnbd_put_permit(dev->sess, iu->permit);
-		return BLK_STS_RESOURCE;
+		ret = BLK_STS_RESOURCE;
 	}
-
+	sg_free_table_chained(&iu->sgt, RNBD_INLINE_SG_CNT);
 	rnbd_put_permit(dev->sess, iu->permit);
-	return BLK_STS_IOERR;
-}
-
-static int rnbd_init_request(struct blk_mq_tag_set *set, struct request *rq,
-			      unsigned int hctx_idx, unsigned int numa_node)
-{
-	struct rnbd_iu *iu = blk_mq_rq_to_pdu(rq);
-
-	sg_init_table(iu->sglist, BMAX_SEGMENTS);
-	return 0;
+	return ret;
 }
 
 static struct blk_mq_ops rnbd_mq_ops = {
 	.queue_rq	= rnbd_queue_rq,
-	.init_request	= rnbd_init_request,
 	.complete	= rnbd_softirq_done_fn,
 };
 
@@ -1170,7 +1189,7 @@ static int setup_mq_tags(struct rnbd_clt_session *sess)
 	tag_set->numa_node		= NUMA_NO_NODE;
 	tag_set->flags		= BLK_MQ_F_SHOULD_MERGE |
 				  BLK_MQ_F_TAG_QUEUE_SHARED;
-	tag_set->cmd_size		= sizeof(struct rnbd_iu);
+	tag_set->cmd_size	= sizeof(struct rnbd_iu) + RNBD_RDMA_SGL_SIZE;
 	tag_set->nr_hw_queues	= num_online_cpus();
 
 	return blk_mq_alloc_tag_set(tag_set);
@@ -1208,7 +1227,7 @@ find_and_get_or_create_sess(const char *sessname,
 	 */
 	sess->rtrs = rtrs_clt_open(&rtrs_ops, sessname,
 				   paths, path_cnt, port_nr,
-				   sizeof(struct rnbd_iu),
+				   0, /* Do not use pdu of rtrs */
 				   RECONNECT_DELAY, BMAX_SEGMENTS,
 				   BLK_MAX_SEGMENT_SIZE,
 				   MAX_RECONNECTS);
@@ -1305,7 +1324,7 @@ static void setup_request_queue(struct rnbd_clt_dev *dev)
 	blk_queue_max_segments(dev->queue, dev->max_segments);
 	blk_queue_io_opt(dev->queue, dev->sess->max_io_size);
 	blk_queue_virt_boundary(dev->queue, SZ_4K - 1);
-	blk_queue_write_cache(dev->queue, true, true);
+	blk_queue_write_cache(dev->queue, dev->wc, dev->fua);
 	dev->queue->queuedata = dev;
 }
 
@@ -1388,12 +1407,11 @@ static struct rnbd_clt_dev *init_dev(struct rnbd_clt_session *sess,
 		goto out_queues;
 	}
 
-	dev->pathname = kzalloc(strlen(pathname) + 1, GFP_KERNEL);
+	dev->pathname = kstrdup(pathname, GFP_KERNEL);
 	if (!dev->pathname) {
 		ret = -ENOMEM;
 		goto out_queues;
 	}
-	strlcpy(dev->pathname, pathname, strlen(pathname) + 1);
 
 	dev->clt_device_id	= ret;
 	dev->sess		= sess;
@@ -1529,13 +1547,13 @@ struct rnbd_clt_dev *rnbd_clt_map_device(const char *sessname,
 	}
 
 	rnbd_clt_info(dev,
-		       "map_device: Device mapped as %s (nsectors: %zu, logical_block_size: %d, physical_block_size: %d, max_write_same_sectors: %d, max_discard_sectors: %d, discard_granularity: %d, discard_alignment: %d, secure_discard: %d, max_segments: %d, max_hw_sectors: %d, rotational: %d)\n",
+		       "map_device: Device mapped as %s (nsectors: %zu, logical_block_size: %d, physical_block_size: %d, max_write_same_sectors: %d, max_discard_sectors: %d, discard_granularity: %d, discard_alignment: %d, secure_discard: %d, max_segments: %d, max_hw_sectors: %d, rotational: %d, wc: %d, fua: %d)\n",
 		       dev->gd->disk_name, dev->nsectors,
 		       dev->logical_block_size, dev->physical_block_size,
 		       dev->max_write_same_sectors, dev->max_discard_sectors,
 		       dev->discard_granularity, dev->discard_alignment,
 		       dev->secure_discard, dev->max_segments,
-		       dev->max_hw_sectors, dev->rotational);
+		       dev->max_hw_sectors, dev->rotational, dev->wc, dev->fua);
 
 	mutex_unlock(&dev->lock);
 
@@ -1667,7 +1685,7 @@ static void rnbd_destroy_sessions(void)
 	/*
 	 * Here at this point there is no any concurrent access to sessions
 	 * list and devices list:
-	 *   1. New session or device can'be be created - session sysfs files
+	 *   1. New session or device can't be created - session sysfs files
 	 *      are removed.
 	 *   2. Device or session can't be removed - module reference is taken
 	 *      into account in unmap device sysfs callback.
diff --git a/drivers/block/rnbd/rnbd-clt.h b/drivers/block/rnbd/rnbd-clt.h
index b193d5904050..537d499dad3b 100644
--- a/drivers/block/rnbd/rnbd-clt.h
+++ b/drivers/block/rnbd/rnbd-clt.h
@@ -44,6 +44,13 @@ struct rnbd_iu_comp {
 	int errno;
 };
 
+#ifdef CONFIG_ARCH_NO_SG_CHAIN
+#define RNBD_INLINE_SG_CNT 0
+#else
+#define RNBD_INLINE_SG_CNT 2
+#endif
+#define RNBD_RDMA_SGL_SIZE (sizeof(struct scatterlist) * RNBD_INLINE_SG_CNT)
+
 struct rnbd_iu {
 	union {
 		struct request *rq; /* for block io */
@@ -56,11 +63,12 @@ struct rnbd_iu {
 		/* use to send msg associated with a sess */
 		struct rnbd_clt_session *sess;
 	};
-	struct scatterlist	sglist[BMAX_SEGMENTS];
+	struct sg_table		sgt;
 	struct work_struct	work;
 	int			errno;
 	struct rnbd_iu_comp	comp;
 	atomic_t		refcount;
+	struct scatterlist	first_sgl[]; /* must be the last one */
 };
 
 struct rnbd_cpu_qlist {
@@ -112,6 +120,8 @@ struct rnbd_clt_dev {
 	enum rnbd_access_mode	access_mode;
 	bool			read_only;
 	bool			rotational;
+	bool			wc;
+	bool			fua;
 	u32			max_hw_sectors;
 	u32			max_write_same_sectors;
 	u32			max_discard_sectors;
diff --git a/drivers/block/rnbd/rnbd-proto.h b/drivers/block/rnbd/rnbd-proto.h
index ca166241452c..c1bc5c0fef71 100644
--- a/drivers/block/rnbd/rnbd-proto.h
+++ b/drivers/block/rnbd/rnbd-proto.h
@@ -108,6 +108,11 @@ struct rnbd_msg_close {
 	__le32		device_id;
 };
 
+enum rnbd_cache_policy {
+	RNBD_FUA = 1 << 0,
+	RNBD_WRITEBACK = 1 << 1,
+};
+
 /**
  * struct rnbd_msg_open_rsp - response message to RNBD_MSG_OPEN
  * @hdr:		message header
@@ -124,6 +129,7 @@ struct rnbd_msg_close {
  * @max_segments:	max segments hardware support in one transfer
  * @secure_discard:	supports secure discard
  * @rotation:		is a rotational disc?
+ * @cache_policy: 	support write-back caching or FUA?
  */
 struct rnbd_msg_open_rsp {
 	struct rnbd_msg_hdr	hdr;
@@ -139,7 +145,8 @@ struct rnbd_msg_open_rsp {
 	__le16			max_segments;
 	__le16			secure_discard;
 	u8			rotational;
-	u8			reserved[11];
+	u8			cache_policy;
+	u8			reserved[10];
 };
 
 /**
diff --git a/drivers/block/rnbd/rnbd-srv.c b/drivers/block/rnbd/rnbd-srv.c
index d1ee72ed8384..b8e44331e494 100644
--- a/drivers/block/rnbd/rnbd-srv.c
+++ b/drivers/block/rnbd/rnbd-srv.c
@@ -338,9 +338,10 @@ static int rnbd_srv_link_ev(struct rtrs_srv *rtrs,
 
 void rnbd_srv_sess_dev_force_close(struct rnbd_srv_sess_dev *sess_dev)
 {
+	mutex_lock(&sess_dev->sess->lock);
 	rnbd_srv_destroy_dev_session_sysfs(sess_dev);
+	mutex_unlock(&sess_dev->sess->lock);
 	sess_dev->keep_id = true;
-
 }
 
 static int process_msg_close(struct rtrs_srv *rtrs,
@@ -549,6 +550,7 @@ static void rnbd_srv_fill_msg_open_rsp(struct rnbd_msg_open_rsp *rsp,
 					struct rnbd_srv_sess_dev *sess_dev)
 {
 	struct rnbd_dev *rnbd_dev = sess_dev->rnbd_dev;
+	struct request_queue *q = bdev_get_queue(rnbd_dev->bdev);
 
 	rsp->hdr.type = cpu_to_le16(RNBD_MSG_OPEN_RSP);
 	rsp->device_id =
@@ -573,8 +575,12 @@ static void rnbd_srv_fill_msg_open_rsp(struct rnbd_msg_open_rsp *rsp,
 		cpu_to_le32(rnbd_dev_get_discard_alignment(rnbd_dev));
 	rsp->secure_discard =
 		cpu_to_le16(rnbd_dev_get_secure_discard(rnbd_dev));
-	rsp->rotational =
-		!blk_queue_nonrot(bdev_get_queue(rnbd_dev->bdev));
+	rsp->rotational = !blk_queue_nonrot(q);
+	rsp->cache_policy = 0;
+	if (test_bit(QUEUE_FLAG_WC, &q->queue_flags))
+		rsp->cache_policy |= RNBD_WRITEBACK;
+	if (blk_queue_fua(q))
+		rsp->cache_policy |= RNBD_FUA;
 }
 
 static struct rnbd_srv_sess_dev *
diff --git a/drivers/infiniband/ulp/rtrs/rtrs-clt.c b/drivers/infiniband/ulp/rtrs/rtrs-clt.c
index 560865f65dc4..67f86c405a26 100644
--- a/drivers/infiniband/ulp/rtrs/rtrs-clt.c
+++ b/drivers/infiniband/ulp/rtrs/rtrs-clt.c
@@ -157,12 +157,6 @@ void rtrs_clt_put_permit(struct rtrs_clt *clt, struct rtrs_permit *permit)
 }
 EXPORT_SYMBOL(rtrs_clt_put_permit);
 
-void *rtrs_permit_to_pdu(struct rtrs_permit *permit)
-{
-	return permit + 1;
-}
-EXPORT_SYMBOL(rtrs_permit_to_pdu);
-
 /**
  * rtrs_permit_to_clt_con() - returns RDMA connection pointer by the permit
  * @sess: client session pointer
diff --git a/drivers/infiniband/ulp/rtrs/rtrs.h b/drivers/infiniband/ulp/rtrs/rtrs.h
index 9af750f4d783..8738e90e715a 100644
--- a/drivers/infiniband/ulp/rtrs/rtrs.h
+++ b/drivers/infiniband/ulp/rtrs/rtrs.h
@@ -63,13 +63,6 @@ struct rtrs_clt *rtrs_clt_open(struct rtrs_clt_ops *ops,
 
 void rtrs_clt_close(struct rtrs_clt *sess);
 
-/**
- * rtrs_permit_to_pdu() - converts rtrs_permit to opaque pdu pointer
- * @permit: RTRS permit pointer, it associates the memory allocation for future
- *          RDMA operation.
- */
-void *rtrs_permit_to_pdu(struct rtrs_permit *permit);
-
 enum {
 	RTRS_PERMIT_NOWAIT = 0,
 	RTRS_PERMIT_WAIT   = 1,
diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c
index 0e06d721cd8e..a4752ac410dc 100644
--- a/drivers/md/bcache/super.c
+++ b/drivers/md/bcache/super.c
@@ -2535,8 +2535,6 @@ static ssize_t register_bcache(struct kobject *k, struct kobj_attribute *attr,
 			else
 				err = "device busy";
 			mutex_unlock(&bch_register_lock);
-			if (!IS_ERR(bdev))
-				bdput(bdev);
 			if (attr == &ksysfs_register_quiet)
 				goto done;
 		}
diff --git a/drivers/md/bcache/sysfs.c b/drivers/md/bcache/sysfs.c
index 554e3afc9b68..00a520c03f41 100644
--- a/drivers/md/bcache/sysfs.c
+++ b/drivers/md/bcache/sysfs.c
@@ -404,7 +404,7 @@ STORE(__cached_dev)
 		if (!env)
 			return -ENOMEM;
 		add_uevent_var(env, "DRIVER=bcache");
-		add_uevent_var(env, "CACHED_UUID=%pU", dc->sb.uuid),
+		add_uevent_var(env, "CACHED_UUID=%pU", dc->sb.uuid);
 		add_uevent_var(env, "CACHED_LABEL=%s", buf);
 		kobject_uevent_env(&disk_to_dev(dc->disk.disk)->kobj,
 				   KOBJ_CHANGE,
diff --git a/drivers/s390/block/dasd_alias.c b/drivers/s390/block/dasd_alias.c
index 99f86612f775..dc78a523a69f 100644
--- a/drivers/s390/block/dasd_alias.c
+++ b/drivers/s390/block/dasd_alias.c
@@ -256,7 +256,6 @@ void dasd_alias_disconnect_device_from_lcu(struct dasd_device *device)
 		return;
 	device->discipline->get_uid(device, &uid);
 	spin_lock_irqsave(&lcu->lock, flags);
-	list_del_init(&device->alias_list);
 	/* make sure that the workers don't use this device */
 	if (device == lcu->suc_data.device) {
 		spin_unlock_irqrestore(&lcu->lock, flags);
@@ -283,6 +282,7 @@ void dasd_alias_disconnect_device_from_lcu(struct dasd_device *device)
 
 	spin_lock_irqsave(&aliastree.lock, flags);
 	spin_lock(&lcu->lock);
+	list_del_init(&device->alias_list);
 	if (list_empty(&lcu->grouplist) &&
 	    list_empty(&lcu->active_devices) &&
 	    list_empty(&lcu->inactive_devices)) {
@@ -462,11 +462,19 @@ static int read_unit_address_configuration(struct dasd_device *device,
 	spin_unlock_irqrestore(&lcu->lock, flags);
 
 	rc = dasd_sleep_on(cqr);
-	if (rc && !suborder_not_supported(cqr)) {
+	if (!rc)
+		goto out;
+
+	if (suborder_not_supported(cqr)) {
+		/* suborder not supported or device unusable for IO */
+		rc = -EOPNOTSUPP;
+	} else {
+		/* IO failed but should be retried */
 		spin_lock_irqsave(&lcu->lock, flags);
 		lcu->flags |= NEED_UAC_UPDATE;
 		spin_unlock_irqrestore(&lcu->lock, flags);
 	}
+out:
 	dasd_sfree_request(cqr, cqr->memdev);
 	return rc;
 }
@@ -503,6 +511,14 @@ static int _lcu_update(struct dasd_device *refdev, struct alias_lcu *lcu)
 		return rc;
 
 	spin_lock_irqsave(&lcu->lock, flags);
+	/*
+	 * there is another update needed skip the remaining handling
+	 * the data might already be outdated
+	 * but especially do not add the device to an LCU with pending
+	 * update
+	 */
+	if (lcu->flags & NEED_UAC_UPDATE)
+		goto out;
 	lcu->pav = NO_PAV;
 	for (i = 0; i < MAX_DEVICES_PER_LCU; ++i) {
 		switch (lcu->uac->unit[i].ua_type) {
@@ -521,6 +537,7 @@ static int _lcu_update(struct dasd_device *refdev, struct alias_lcu *lcu)
 				 alias_list) {
 		_add_device_to_lcu(lcu, device, refdev);
 	}
+out:
 	spin_unlock_irqrestore(&lcu->lock, flags);
 	return 0;
 }
@@ -625,6 +642,7 @@ int dasd_alias_add_device(struct dasd_device *device)
 	}
 	if (lcu->flags & UPDATE_PENDING) {
 		list_move(&device->alias_list, &lcu->active_devices);
+		private->pavgroup = NULL;
 		_schedule_lcu_update(lcu, device);
 	}
 	spin_unlock_irqrestore(&lcu->lock, flags);
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 9e56ee1f2652..9293045e128c 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -1,9 +1,8 @@
 // SPDX-License-Identifier: GPL-2.0-only
 /*
- *  linux/fs/block_dev.c
- *
  *  Copyright (C) 1991, 1992  Linus Torvalds
  *  Copyright (C) 2001  Andrea Arcangeli <andrea@suse.de> SuSE
+ *  Copyright (C) 2016 - 2020 Christoph Hellwig
  */
 
 #include <linux/init.h>
diff --git a/include/trace/events/iocost.h b/include/trace/events/iocost.h
index 0b6869980ba2..e282ce02fa2d 100644
--- a/include/trace/events/iocost.h
+++ b/include/trace/events/iocost.h
@@ -11,7 +11,7 @@ struct ioc_gq;
 
 #include <linux/tracepoint.h>
 
-TRACE_EVENT(iocost_iocg_activate,
+DECLARE_EVENT_CLASS(iocost_iocg_state,
 
 	TP_PROTO(struct ioc_gq *iocg, const char *path, struct ioc_now *now,
 		u64 last_period, u64 cur_period, u64 vtime),
@@ -59,6 +59,20 @@ TRACE_EVENT(iocost_iocg_activate,
 	)
 );
 
+DEFINE_EVENT(iocost_iocg_state, iocost_iocg_activate,
+	TP_PROTO(struct ioc_gq *iocg, const char *path, struct ioc_now *now,
+		 u64 last_period, u64 cur_period, u64 vtime),
+
+	TP_ARGS(iocg, path, now, last_period, cur_period, vtime)
+);
+
+DEFINE_EVENT(iocost_iocg_state, iocost_iocg_idle,
+	TP_PROTO(struct ioc_gq *iocg, const char *path, struct ioc_now *now,
+		 u64 last_period, u64 cur_period, u64 vtime),
+
+	TP_ARGS(iocg, path, now, last_period, cur_period, vtime)
+);
+
 DECLARE_EVENT_CLASS(iocg_inuse_update,
 
 	TP_PROTO(struct ioc_gq *iocg, const char *path, struct ioc_now *now,