Merge tag 'io_uring-5.9-2020-08-23' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:

 - NVMe pull request from Sagi:
       - nvme completion rework from Christoph and Chao that mostly came
         from a bit of divergence of how we classify errors related to
         pathing/retry etc.
       - nvmet passthru fixes from Chaitanya
       - minor nvmet fixes from Amit and I
       - mpath round-robin path selection fix from Martin
       - ignore noiob for zoned devices from Keith
       - minor nvme-fc fix from Tianjia"

 - BFQ cgroup leak fix (Dmitry)

 - block layer MAINTAINERS addition (Geert)

 - fix null_blk FUA checking (Hou)

 - get_max_io_size() size fix (Keith)

 - fix block page_is_mergeable() for compound pages (Matthew)

 - discard granularity fixes (Ming)

 - IO scheduler ordering fix (Ming)

 - misc fixes

* tag 'io_uring-5.9-2020-08-23' of git://git.kernel.dk/linux-block: (31 commits)
  null_blk: fix passing of REQ_FUA flag in null_handle_rq
  nvmet: Disable keep-alive timer when kato is cleared to 0h
  nvme: redirect commands on dying queue
  nvme: just check the status code type in nvme_is_path_error
  nvme: refactor command completion
  nvme: rename and document nvme_end_request
  nvme: skip noiob for zoned devices
  nvme-pci: fix PRP pool size
  nvme-pci: Use u32 for nvme_dev.q_depth and nvme_queue.q_depth
  nvme: Use spin_lock_irq() when taking the ctrl->lock
  nvmet: call blk_mq_free_request() directly
  nvmet: fix oops in pt cmd execution
  nvmet: add ns tear down label for pt-cmd handling
  nvme: multipath: round-robin: eliminate "fallback" variable
  nvme: multipath: round-robin: fix single non-optimized path case
  nvme-fc: Fix wrong return value in __nvme_fc_init_request()
  nvmet-passthru: Reject commands with non-sgl flags set
  nvmet: fix a memory leak
  blkcg: fix memleak for iolatency
  MAINTAINERS: Add missing header files to BLOCK LAYER section
  ...

4 files changed, 44 insertions, 25 deletions

diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index 33cc6558f9d2..5a9915101362 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -878,6 +878,7 @@ static void loop_config_discard(struct loop_device *lo)
 	struct file *file = lo->lo_backing_file;
 	struct inode *inode = file->f_mapping->host;
 	struct request_queue *q = lo->lo_queue;
+	u32 granularity, max_discard_sectors;
 
 	/*
 	 * If the backing device is a block device, mirror its zeroing
@@ -890,11 +891,10 @@ static void loop_config_discard(struct loop_device *lo)
 		struct request_queue *backingq;
 
 		backingq = bdev_get_queue(inode->i_bdev);
-		blk_queue_max_discard_sectors(q,
-			backingq->limits.max_write_zeroes_sectors);
 
-		blk_queue_max_write_zeroes_sectors(q,
-			backingq->limits.max_write_zeroes_sectors);
+		max_discard_sectors = backingq->limits.max_write_zeroes_sectors;
+		granularity = backingq->limits.discard_granularity ?:
+			queue_physical_block_size(backingq);
 
 	/*
 	 * We use punch hole to reclaim the free space used by the
@@ -903,23 +903,26 @@ static void loop_config_discard(struct loop_device *lo)
 	 * useful information.
 	 */
 	} else if (!file->f_op->fallocate || lo->lo_encrypt_key_size) {
-		q->limits.discard_granularity = 0;
-		q->limits.discard_alignment = 0;
-		blk_queue_max_discard_sectors(q, 0);
-		blk_queue_max_write_zeroes_sectors(q, 0);
+		max_discard_sectors = 0;
+		granularity = 0;
 
 	} else {
-		q->limits.discard_granularity = inode->i_sb->s_blocksize;
-		q->limits.discard_alignment = 0;
-
-		blk_queue_max_discard_sectors(q, UINT_MAX >> 9);
-		blk_queue_max_write_zeroes_sectors(q, UINT_MAX >> 9);
+		max_discard_sectors = UINT_MAX >> 9;
+		granularity = inode->i_sb->s_blocksize;
 	}
 
-	if (q->limits.max_write_zeroes_sectors)
+	if (max_discard_sectors) {
+		q->limits.discard_granularity = granularity;
+		blk_queue_max_discard_sectors(q, max_discard_sectors);
+		blk_queue_max_write_zeroes_sectors(q, max_discard_sectors);
 		blk_queue_flag_set(QUEUE_FLAG_DISCARD, q);
-	else
+	} else {
+		q->limits.discard_granularity = 0;
+		blk_queue_max_discard_sectors(q, 0);
+		blk_queue_max_write_zeroes_sectors(q, 0);
 		blk_queue_flag_clear(QUEUE_FLAG_DISCARD, q);
+	}
+	q->limits.discard_alignment = 0;
 }
 
 static void loop_unprepare_queue(struct loop_device *lo)
diff --git a/drivers/block/null_blk_main.c b/drivers/block/null_blk_main.c
index 47a9dad880af..d74443a9c8fa 100644
--- a/drivers/block/null_blk_main.c
+++ b/drivers/block/null_blk_main.c
@@ -1147,7 +1147,7 @@ static int null_handle_rq(struct nullb_cmd *cmd)
 		len = bvec.bv_len;
 		err = null_transfer(nullb, bvec.bv_page, len, bvec.bv_offset,
 				     op_is_write(req_op(rq)), sector,
-				     req_op(rq) & REQ_FUA);
+				     rq->cmd_flags & REQ_FUA);
 		if (err) {
 			spin_unlock_irq(&nullb->lock);
 			return err;
diff --git a/drivers/block/rnbd/rnbd-srv.c b/drivers/block/rnbd/rnbd-srv.c
index 0fb94843a495..e1bc8b4cd592 100644
--- a/drivers/block/rnbd/rnbd-srv.c
+++ b/drivers/block/rnbd/rnbd-srv.c
@@ -148,7 +148,8 @@ static int process_rdma(struct rtrs_srv *sess,
 	/* Generate bio with pages pointing to the rdma buffer */
 	bio = rnbd_bio_map_kern(data, sess_dev->rnbd_dev->ibd_bio_set, datalen, GFP_KERNEL);
 	if (IS_ERR(bio)) {
-		rnbd_srv_err(sess_dev, "Failed to generate bio, err: %ld\n", PTR_ERR(bio));
+		err = PTR_ERR(bio);
+		rnbd_srv_err(sess_dev, "Failed to generate bio, err: %d\n", err);
 		goto sess_dev_put;
 	}
 
diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
index 63b213e00b37..b2e48dac1ebd 100644
--- a/drivers/block/virtio_blk.c
+++ b/drivers/block/virtio_blk.c
@@ -126,16 +126,31 @@ static int virtblk_setup_discard_write_zeroes(struct request *req, bool unmap)
 	if (!range)
 		return -ENOMEM;
 
-	__rq_for_each_bio(bio, req) {
-		u64 sector = bio->bi_iter.bi_sector;
-		u32 num_sectors = bio->bi_iter.bi_size >> SECTOR_SHIFT;
-
-		range[n].flags = cpu_to_le32(flags);
-		range[n].num_sectors = cpu_to_le32(num_sectors);
-		range[n].sector = cpu_to_le64(sector);
-		n++;
+	/*
+	 * Single max discard segment means multi-range discard isn't
+	 * supported, and block layer only runs contiguity merge like
+	 * normal RW request. So we can't reply on bio for retrieving
+	 * each range info.
+	 */
+	if (queue_max_discard_segments(req->q) == 1) {
+		range[0].flags = cpu_to_le32(flags);
+		range[0].num_sectors = cpu_to_le32(blk_rq_sectors(req));
+		range[0].sector = cpu_to_le64(blk_rq_pos(req));
+		n = 1;
+	} else {
+		__rq_for_each_bio(bio, req) {
+			u64 sector = bio->bi_iter.bi_sector;
+			u32 num_sectors = bio->bi_iter.bi_size >> SECTOR_SHIFT;
+
+			range[n].flags = cpu_to_le32(flags);
+			range[n].num_sectors = cpu_to_le32(num_sectors);
+			range[n].sector = cpu_to_le64(sector);
+			n++;
+		}
 	}
 
+	WARN_ON_ONCE(n != segments);
+
 	req->special_vec.bv_page = virt_to_page(range);
 	req->special_vec.bv_offset = offset_in_page(range);
 	req->special_vec.bv_len = sizeof(*range) * segments;