Merge tag 'for-6.15/block-20250322' of git://git.kernel.dk/linux

Pull block updates from Jens Axboe:

 - Fixes for integrity handling

 - NVMe pull request via Keith:
      - Secure concatenation for TCP transport (Hannes)
      - Multipath sysfs visibility (Nilay)
      - Various cleanups (Qasim, Baruch, Wang, Chen, Mike, Damien, Li)
      - Correct use of 64-bit BARs for pci-epf target (Niklas)
      - Socket fix for selinux when used in containers (Peijie)

 - MD pull request via Yu:
      - fix recovery can preempt resync (Li Nan)
      - fix md-bitmap IO limit (Su Yue)
      - fix raid10 discard with REQ_NOWAIT (Xiao Ni)
      - fix raid1 memory leak (Zheng Qixing)
      - fix mddev uaf (Yu Kuai)
      - fix raid1,raid10 IO flags (Yu Kuai)
      - some refactor and cleanup (Yu Kuai)

 - Series cleaning up and fixing bugs in the bad block handling code

 - Improve support for write failure simulation in null_blk

 - Various lock ordering fixes

 - Fixes for locking for debugfs attributes

 - Various ublk related fixes and improvements

 - Cleanups for blk-rq-qos wait handling

 - blk-throttle fixes

 - Fixes for loop dio and sync handling

 - Fixes and cleanups for the auto-PI code

 - Block side support for hardware encryption keys in blk-crypto

 - Various cleanups and fixes

* tag 'for-6.15/block-20250322' of git://git.kernel.dk/linux: (105 commits)
  nvmet: replace max(a, min(b, c)) by clamp(val, lo, hi)
  nvme-tcp: fix selinux denied when calling sock_sendmsg
  nvmet: pci-epf: Always configure BAR0 as 64-bit
  nvmet: Remove duplicate uuid_copy
  nvme: zns: Simplify nvme_zone_parse_entry()
  nvmet: pci-epf: Remove redundant 'flush_workqueue()' calls
  nvmet-fc: Remove unused functions
  nvme-pci: remove stale comment
  nvme-fc: Utilise min3() to simplify queue count calculation
  nvme-multipath: Add visibility for queue-depth io-policy
  nvme-multipath: Add visibility for numa io-policy
  nvme-multipath: Add visibility for round-robin io-policy
  nvmet: add tls_concat and tls_key debugfs entries
  nvmet-tcp: support secure channel concatenation
  nvmet: Add 'sq' argument to alloc_ctrl_args
  nvme-fabrics: reset admin connection for secure concatenation
  nvme-tcp: request secure channel concatenation
  nvme-keyring: add nvme_tls_psk_refresh()
  nvme: add nvme_auth_derive_tls_psk()
  nvme: add nvme_auth_generate_digest()
  ...

10 files changed, 259 insertions, 175 deletions

diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index c05fe27a96b6..674527d770dc 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -45,8 +45,6 @@ enum {
 	Lo_deleting,
 };
 
-struct loop_func_table;
-
 struct loop_device {
 	int		lo_number;
 	loff_t		lo_offset;
@@ -54,7 +52,8 @@ struct loop_device {
 	int		lo_flags;
 	char		lo_file_name[LO_NAME_SIZE];
 
-	struct file *	lo_backing_file;
+	struct file	*lo_backing_file;
+	unsigned int	lo_min_dio_size;
 	struct block_device *lo_device;
 
 	gfp_t		old_gfp_mask;
@@ -169,29 +168,14 @@ static loff_t get_loop_size(struct loop_device *lo, struct file *file)
  * of backing device, and the logical block size of loop is bigger than that of
  * the backing device.
  */
-static bool lo_bdev_can_use_dio(struct loop_device *lo,
-		struct block_device *backing_bdev)
-{
-	unsigned int sb_bsize = bdev_logical_block_size(backing_bdev);
-
-	if (queue_logical_block_size(lo->lo_queue) < sb_bsize)
-		return false;
-	if (lo->lo_offset & (sb_bsize - 1))
-		return false;
-	return true;
-}
-
 static bool lo_can_use_dio(struct loop_device *lo)
 {
-	struct inode *inode = lo->lo_backing_file->f_mapping->host;
-
 	if (!(lo->lo_backing_file->f_mode & FMODE_CAN_ODIRECT))
 		return false;
-
-	if (S_ISBLK(inode->i_mode))
-		return lo_bdev_can_use_dio(lo, I_BDEV(inode));
-	if (inode->i_sb->s_bdev)
-		return lo_bdev_can_use_dio(lo, inode->i_sb->s_bdev);
+	if (queue_logical_block_size(lo->lo_queue) < lo->lo_min_dio_size)
+		return false;
+	if (lo->lo_offset & (lo->lo_min_dio_size - 1))
+		return false;
 	return true;
 }
 
@@ -205,20 +189,12 @@ static bool lo_can_use_dio(struct loop_device *lo)
  */
 static inline void loop_update_dio(struct loop_device *lo)
 {
-	bool dio_in_use = lo->lo_flags & LO_FLAGS_DIRECT_IO;
-
 	lockdep_assert_held(&lo->lo_mutex);
 	WARN_ON_ONCE(lo->lo_state == Lo_bound &&
 		     lo->lo_queue->mq_freeze_depth == 0);
 
-	if (lo->lo_backing_file->f_flags & O_DIRECT)
-		lo->lo_flags |= LO_FLAGS_DIRECT_IO;
 	if ((lo->lo_flags & LO_FLAGS_DIRECT_IO) && !lo_can_use_dio(lo))
 		lo->lo_flags &= ~LO_FLAGS_DIRECT_IO;
-
-	/* flush dirty pages before starting to issue direct I/O */
-	if ((lo->lo_flags & LO_FLAGS_DIRECT_IO) && !dio_in_use)
-		vfs_fsync(lo->lo_backing_file, 0);
 }
 
 /**
@@ -541,6 +517,28 @@ static void loop_reread_partitions(struct loop_device *lo)
 			__func__, lo->lo_number, lo->lo_file_name, rc);
 }
 
+static unsigned int loop_query_min_dio_size(struct loop_device *lo)
+{
+	struct file *file = lo->lo_backing_file;
+	struct block_device *sb_bdev = file->f_mapping->host->i_sb->s_bdev;
+	struct kstat st;
+
+	/*
+	 * Use the minimal dio alignment of the file system if provided.
+	 */
+	if (!vfs_getattr(&file->f_path, &st, STATX_DIOALIGN, 0) &&
+	    (st.result_mask & STATX_DIOALIGN))
+		return st.dio_offset_align;
+
+	/*
+	 * In a perfect world this wouldn't be needed, but as of Linux 6.13 only
+	 * a handful of file systems support the STATX_DIOALIGN flag.
+	 */
+	if (sb_bdev)
+		return bdev_logical_block_size(sb_bdev);
+	return SECTOR_SIZE;
+}
+
 static inline int is_loop_device(struct file *file)
 {
 	struct inode *i = file->f_mapping->host;
@@ -573,6 +571,17 @@ static int loop_validate_file(struct file *file, struct block_device *bdev)
 	return 0;
 }
 
+static void loop_assign_backing_file(struct loop_device *lo, struct file *file)
+{
+	lo->lo_backing_file = file;
+	lo->old_gfp_mask = mapping_gfp_mask(file->f_mapping);
+	mapping_set_gfp_mask(file->f_mapping,
+			lo->old_gfp_mask & ~(__GFP_IO | __GFP_FS));
+	if (lo->lo_backing_file->f_flags & O_DIRECT)
+		lo->lo_flags |= LO_FLAGS_DIRECT_IO;
+	lo->lo_min_dio_size = loop_query_min_dio_size(lo);
+}
+
 /*
  * loop_change_fd switched the backing store of a loopback device to
  * a new file. This is useful for operating system installers to free up
@@ -622,14 +631,18 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev,
 	if (get_loop_size(lo, file) != get_loop_size(lo, old_file))
 		goto out_err;
 
+	/*
+	 * We might switch to direct I/O mode for the loop device, write back
+	 * all dirty data the page cache now that so that the individual I/O
+	 * operations don't have to do that.
+	 */
+	vfs_fsync(file, 0);
+
 	/* and ... switch */
 	disk_force_media_change(lo->lo_disk);
 	memflags = blk_mq_freeze_queue(lo->lo_queue);
 	mapping_set_gfp_mask(old_file->f_mapping, lo->old_gfp_mask);
-	lo->lo_backing_file = file;
-	lo->old_gfp_mask = mapping_gfp_mask(file->f_mapping);
-	mapping_set_gfp_mask(file->f_mapping,
-			     lo->old_gfp_mask & ~(__GFP_IO|__GFP_FS));
+	loop_assign_backing_file(lo, file);
 	loop_update_dio(lo);
 	blk_mq_unfreeze_queue(lo->lo_queue, memflags);
 	partscan = lo->lo_flags & LO_FLAGS_PARTSCAN;
@@ -971,12 +984,11 @@ loop_set_status_from_info(struct loop_device *lo,
 	return 0;
 }
 
-static unsigned int loop_default_blocksize(struct loop_device *lo,
-		struct block_device *backing_bdev)
+static unsigned int loop_default_blocksize(struct loop_device *lo)
 {
-	/* In case of direct I/O, match underlying block size */
-	if ((lo->lo_backing_file->f_flags & O_DIRECT) && backing_bdev)
-		return bdev_logical_block_size(backing_bdev);
+	/* In case of direct I/O, match underlying minimum I/O size */
+	if (lo->lo_flags & LO_FLAGS_DIRECT_IO)
+		return lo->lo_min_dio_size;
 	return SECTOR_SIZE;
 }
 
@@ -994,7 +1006,7 @@ static void loop_update_limits(struct loop_device *lo, struct queue_limits *lim,
 		backing_bdev = inode->i_sb->s_bdev;
 
 	if (!bsize)
-		bsize = loop_default_blocksize(lo, backing_bdev);
+		bsize = loop_default_blocksize(lo);
 
 	loop_get_discard_config(lo, &granularity, &max_discard_sectors);
 
@@ -1019,7 +1031,6 @@ static int loop_configure(struct loop_device *lo, blk_mode_t mode,
 			  const struct loop_config *config)
 {
 	struct file *file = fget(config->fd);
-	struct address_space *mapping;
 	struct queue_limits lim;
 	int error;
 	loff_t size;
@@ -1055,8 +1066,6 @@ static int loop_configure(struct loop_device *lo, blk_mode_t mode,
 	if (error)
 		goto out_unlock;
 
-	mapping = file->f_mapping;
-
 	if ((config->info.lo_flags & ~LOOP_CONFIGURE_SETTABLE_FLAGS) != 0) {
 		error = -EINVAL;
 		goto out_unlock;
@@ -1088,9 +1097,7 @@ static int loop_configure(struct loop_device *lo, blk_mode_t mode,
 	set_disk_ro(lo->lo_disk, (lo->lo_flags & LO_FLAGS_READ_ONLY) != 0);
 
 	lo->lo_device = bdev;
-	lo->lo_backing_file = file;
-	lo->old_gfp_mask = mapping_gfp_mask(mapping);
-	mapping_set_gfp_mask(mapping, lo->old_gfp_mask & ~(__GFP_IO|__GFP_FS));
+	loop_assign_backing_file(lo, file);
 
 	lim = queue_limits_start_update(lo->lo_queue);
 	loop_update_limits(lo, &lim, config->block_size);
@@ -1099,6 +1106,13 @@ static int loop_configure(struct loop_device *lo, blk_mode_t mode,
 	if (error)
 		goto out_unlock;
 
+	/*
+	 * We might switch to direct I/O mode for the loop device, write back
+	 * all dirty data the page cache now that so that the individual I/O
+	 * operations don't have to do that.
+	 */
+	vfs_fsync(file, 0);
+
 	loop_update_dio(lo);
 	loop_sysfs_init(lo);
 
diff --git a/drivers/block/mtip32xx/mtip32xx.c b/drivers/block/mtip32xx/mtip32xx.c
index 95361099a2dc..0d619df03fa9 100644
--- a/drivers/block/mtip32xx/mtip32xx.c
+++ b/drivers/block/mtip32xx/mtip32xx.c
@@ -2056,7 +2056,7 @@ static void mtip_hw_submit_io(struct driver_data *dd, struct request *rq,
 	unsigned int nents;
 
 	/* Map the scatter list for DMA access */
-	nents = blk_rq_map_sg(hctx->queue, rq, command->sg);
+	nents = blk_rq_map_sg(rq, command->sg);
 	nents = dma_map_sg(&dd->pdev->dev, command->sg, nents, dma_dir);
 
 	prefetch(&port->flags);
diff --git a/drivers/block/null_blk/main.c b/drivers/block/null_blk/main.c
index da1ecbf988b8..3bb9cee0a9b5 100644
--- a/drivers/block/null_blk/main.c
+++ b/drivers/block/null_blk/main.c
@@ -473,6 +473,8 @@ NULLB_DEVICE_ATTR(shared_tags, bool, NULL);
 NULLB_DEVICE_ATTR(shared_tag_bitmap, bool, NULL);
 NULLB_DEVICE_ATTR(fua, bool, NULL);
 NULLB_DEVICE_ATTR(rotational, bool, NULL);
+NULLB_DEVICE_ATTR(badblocks_once, bool, NULL);
+NULLB_DEVICE_ATTR(badblocks_partial_io, bool, NULL);
 
 static ssize_t nullb_device_power_show(struct config_item *item, char *page)
 {
@@ -559,14 +561,14 @@ static ssize_t nullb_device_badblocks_store(struct config_item *item,
 		goto out;
 	/* enable badblocks */
 	cmpxchg(&t_dev->badblocks.shift, -1, 0);
-	if (buf[0] == '+')
-		ret = badblocks_set(&t_dev->badblocks, start,
-			end - start + 1, 1);
-	else
-		ret = badblocks_clear(&t_dev->badblocks, start,
-			end - start + 1);
-	if (ret == 0)
+	if (buf[0] == '+') {
+		if (badblocks_set(&t_dev->badblocks, start,
+				  end - start + 1, 1))
+			ret = count;
+	} else if (badblocks_clear(&t_dev->badblocks, start,
+				   end - start + 1)) {
 		ret = count;
+	}
 out:
 	kfree(orig);
 	return ret;
@@ -592,41 +594,43 @@ static ssize_t nullb_device_zone_offline_store(struct config_item *item,
 CONFIGFS_ATTR_WO(nullb_device_, zone_offline);
 
 static struct configfs_attribute *nullb_device_attrs[] = {
-	&nullb_device_attr_size,
+	&nullb_device_attr_badblocks,
+	&nullb_device_attr_badblocks_once,
+	&nullb_device_attr_badblocks_partial_io,
+	&nullb_device_attr_blocking,
+	&nullb_device_attr_blocksize,
+	&nullb_device_attr_cache_size,
 	&nullb_device_attr_completion_nsec,
-	&nullb_device_attr_submit_queues,
-	&nullb_device_attr_poll_queues,
+	&nullb_device_attr_discard,
+	&nullb_device_attr_fua,
 	&nullb_device_attr_home_node,
-	&nullb_device_attr_queue_mode,
-	&nullb_device_attr_blocksize,
-	&nullb_device_attr_max_sectors,
-	&nullb_device_attr_irqmode,
 	&nullb_device_attr_hw_queue_depth,
 	&nullb_device_attr_index,
-	&nullb_device_attr_blocking,
-	&nullb_device_attr_use_per_node_hctx,
-	&nullb_device_attr_power,
-	&nullb_device_attr_memory_backed,
-	&nullb_device_attr_discard,
+	&nullb_device_attr_irqmode,
+	&nullb_device_attr_max_sectors,
 	&nullb_device_attr_mbps,
-	&nullb_device_attr_cache_size,
-	&nullb_device_attr_badblocks,
-	&nullb_device_attr_zoned,
-	&nullb_device_attr_zone_size,
+	&nullb_device_attr_memory_backed,
+	&nullb_device_attr_no_sched,
+	&nullb_device_attr_poll_queues,
+	&nullb_device_attr_power,
+	&nullb_device_attr_queue_mode,
+	&nullb_device_attr_rotational,
+	&nullb_device_attr_shared_tag_bitmap,
+	&nullb_device_attr_shared_tags,
+	&nullb_device_attr_size,
+	&nullb_device_attr_submit_queues,
+	&nullb_device_attr_use_per_node_hctx,
+	&nullb_device_attr_virt_boundary,
+	&nullb_device_attr_zone_append_max_sectors,
 	&nullb_device_attr_zone_capacity,
-	&nullb_device_attr_zone_nr_conv,
-	&nullb_device_attr_zone_max_open,
+	&nullb_device_attr_zone_full,
 	&nullb_device_attr_zone_max_active,
-	&nullb_device_attr_zone_append_max_sectors,
-	&nullb_device_attr_zone_readonly,
+	&nullb_device_attr_zone_max_open,
+	&nullb_device_attr_zone_nr_conv,
 	&nullb_device_attr_zone_offline,
-	&nullb_device_attr_zone_full,
-	&nullb_device_attr_virt_boundary,
-	&nullb_device_attr_no_sched,
-	&nullb_device_attr_shared_tags,
-	&nullb_device_attr_shared_tag_bitmap,
-	&nullb_device_attr_fua,
-	&nullb_device_attr_rotational,
+	&nullb_device_attr_zone_readonly,
+	&nullb_device_attr_zone_size,
+	&nullb_device_attr_zoned,
 	NULL,
 };
 
@@ -704,16 +708,28 @@ nullb_group_drop_item(struct config_group *group, struct config_item *item)
 
 static ssize_t memb_group_features_show(struct config_item *item, char *page)
 {
-	return snprintf(page, PAGE_SIZE,
-			"badblocks,blocking,blocksize,cache_size,fua,"
-			"completion_nsec,discard,home_node,hw_queue_depth,"
-			"irqmode,max_sectors,mbps,memory_backed,no_sched,"
-			"poll_queues,power,queue_mode,shared_tag_bitmap,"
-			"shared_tags,size,submit_queues,use_per_node_hctx,"
-			"virt_boundary,zoned,zone_capacity,zone_max_active,"
-			"zone_max_open,zone_nr_conv,zone_offline,zone_readonly,"
-			"zone_size,zone_append_max_sectors,zone_full,"
-			"rotational\n");
+
+	struct configfs_attribute **entry;
+	char delimiter = ',';
+	size_t left = PAGE_SIZE;
+	size_t written = 0;
+	int ret;
+
+	for (entry = &nullb_device_attrs[0]; *entry && left > 0; entry++) {
+		if (!*(entry + 1))
+			delimiter = '\n';
+		ret = snprintf(page + written, left, "%s%c", (*entry)->ca_name,
+			       delimiter);
+		if (ret >= left) {
+			WARN_ONCE(1, "Too many null_blk features to print\n");
+			memzero_explicit(page, PAGE_SIZE);
+			return -ENOBUFS;
+		}
+		left -= ret;
+		written += ret;
+	}
+
+	return written;
 }
 
 CONFIGFS_ATTR_RO(memb_group_, features);
@@ -1249,25 +1265,37 @@ static int null_transfer(struct nullb *nullb, struct page *page,
 	return err;
 }
 
-static blk_status_t null_handle_rq(struct nullb_cmd *cmd)
+/*
+ * Transfer data for the given request. The transfer size is capped with the
+ * nr_sectors argument.
+ */
+static blk_status_t null_handle_data_transfer(struct nullb_cmd *cmd,
+					      sector_t nr_sectors)
 {
 	struct request *rq = blk_mq_rq_from_pdu(cmd);
 	struct nullb *nullb = cmd->nq->dev->nullb;
 	int err = 0;
 	unsigned int len;
 	sector_t sector = blk_rq_pos(rq);
+	unsigned int max_bytes = nr_sectors << SECTOR_SHIFT;
+	unsigned int transferred_bytes = 0;
 	struct req_iterator iter;
 	struct bio_vec bvec;
 
 	spin_lock_irq(&nullb->lock);
 	rq_for_each_segment(bvec, rq, iter) {
 		len = bvec.bv_len;
+		if (transferred_bytes + len > max_bytes)
+			len = max_bytes - transferred_bytes;
 		err = null_transfer(nullb, bvec.bv_page, len, bvec.bv_offset,
 				     op_is_write(req_op(rq)), sector,
 				     rq->cmd_flags & REQ_FUA);
 		if (err)
 			break;
 		sector += len >> SECTOR_SHIFT;
+		transferred_bytes += len;
+		if (transferred_bytes >= max_bytes)
+			break;
 	}
 	spin_unlock_irq(&nullb->lock);
 
@@ -1295,31 +1323,51 @@ static inline blk_status_t null_handle_throttled(struct nullb_cmd *cmd)
 	return sts;
 }
 
-static inline blk_status_t null_handle_badblocks(struct nullb_cmd *cmd,
-						 sector_t sector,
-						 sector_t nr_sectors)
+/*
+ * Check if the command should fail for the badblocks. If so, return
+ * BLK_STS_IOERR and return number of partial I/O sectors to be written or read,
+ * which may be less than the requested number of sectors.
+ *
+ * @cmd:        The command to handle.
+ * @sector:     The start sector for I/O.
+ * @nr_sectors: Specifies number of sectors to write or read, and returns the
+ *              number of sectors to be written or read.
+ */
+blk_status_t null_handle_badblocks(struct nullb_cmd *cmd, sector_t sector,
+				   unsigned int *nr_sectors)
 {
 	struct badblocks *bb = &cmd->nq->dev->badblocks;
-	sector_t first_bad;
-	int bad_sectors;
+	struct nullb_device *dev = cmd->nq->dev;
+	unsigned int block_sectors = dev->blocksize >> SECTOR_SHIFT;
+	sector_t first_bad, bad_sectors;
+	unsigned int partial_io_sectors = 0;
 
-	if (badblocks_check(bb, sector, nr_sectors, &first_bad, &bad_sectors))
-		return BLK_STS_IOERR;
+	if (!badblocks_check(bb, sector, *nr_sectors, &first_bad, &bad_sectors))
+		return BLK_STS_OK;
 
-	return BLK_STS_OK;
+	if (cmd->nq->dev->badblocks_once)
+		badblocks_clear(bb, first_bad, bad_sectors);
+
+	if (cmd->nq->dev->badblocks_partial_io) {
+		if (!IS_ALIGNED(first_bad, block_sectors))
+			first_bad = ALIGN_DOWN(first_bad, block_sectors);
+		if (sector < first_bad)
+			partial_io_sectors = first_bad - sector;
+	}
+	*nr_sectors = partial_io_sectors;
+
+	return BLK_STS_IOERR;
 }
 
-static inline blk_status_t null_handle_memory_backed(struct nullb_cmd *cmd,
-						     enum req_op op,
-						     sector_t sector,
-						     sector_t nr_sectors)
+blk_status_t null_handle_memory_backed(struct nullb_cmd *cmd, enum req_op op,
+				       sector_t sector, sector_t nr_sectors)
 {
 	struct nullb_device *dev = cmd->nq->dev;
 
 	if (op == REQ_OP_DISCARD)
 		return null_handle_discard(dev, sector, nr_sectors);
 
-	return null_handle_rq(cmd);
+	return null_handle_data_transfer(cmd, nr_sectors);
 }
 
 static void nullb_zero_read_cmd_buffer(struct nullb_cmd *cmd)
@@ -1366,18 +1414,19 @@ blk_status_t null_process_cmd(struct nullb_cmd *cmd, enum req_op op,
 			      sector_t sector, unsigned int nr_sectors)
 {
 	struct nullb_device *dev = cmd->nq->dev;
+	blk_status_t badblocks_ret = BLK_STS_OK;
 	blk_status_t ret;
 
-	if (dev->badblocks.shift != -1) {
-		ret = null_handle_badblocks(cmd, sector, nr_sectors);
+	if (dev->badblocks.shift != -1)
+		badblocks_ret = null_handle_badblocks(cmd, sector, &nr_sectors);
+
+	if (dev->memory_backed && nr_sectors) {
+		ret = null_handle_memory_backed(cmd, op, sector, nr_sectors);
 		if (ret != BLK_STS_OK)
 			return ret;
 	}
 
-	if (dev->memory_backed)
-		return null_handle_memory_backed(cmd, op, sector, nr_sectors);
-
-	return BLK_STS_OK;
+	return badblocks_ret;
 }
 
 static void null_handle_cmd(struct nullb_cmd *cmd, sector_t sector,
diff --git a/drivers/block/null_blk/null_blk.h b/drivers/block/null_blk/null_blk.h
index 6f9fe6171087..7bb6128dbaaf 100644
--- a/drivers/block/null_blk/null_blk.h
+++ b/drivers/block/null_blk/null_blk.h
@@ -63,6 +63,8 @@ struct nullb_device {
 	unsigned long flags; /* device flags */
 	unsigned int curr_cache;
 	struct badblocks badblocks;
+	bool badblocks_once;
+	bool badblocks_partial_io;
 
 	unsigned int nr_zones;
 	unsigned int nr_zones_imp_open;
@@ -131,6 +133,10 @@ blk_status_t null_handle_discard(struct nullb_device *dev, sector_t sector,
 				 sector_t nr_sectors);
 blk_status_t null_process_cmd(struct nullb_cmd *cmd, enum req_op op,
 			      sector_t sector, unsigned int nr_sectors);
+blk_status_t null_handle_badblocks(struct nullb_cmd *cmd, sector_t sector,
+				   unsigned int *nr_sectors);
+blk_status_t null_handle_memory_backed(struct nullb_cmd *cmd, enum req_op op,
+				       sector_t sector, sector_t nr_sectors);
 
 #ifdef CONFIG_BLK_DEV_ZONED
 int null_init_zoned_dev(struct nullb_device *dev, struct queue_limits *lim);
diff --git a/drivers/block/null_blk/zoned.c b/drivers/block/null_blk/zoned.c
index 0d5f9bf95229..4e5728f45989 100644
--- a/drivers/block/null_blk/zoned.c
+++ b/drivers/block/null_blk/zoned.c
@@ -353,6 +353,7 @@ static blk_status_t null_zone_write(struct nullb_cmd *cmd, sector_t sector,
 	struct nullb_device *dev = cmd->nq->dev;
 	unsigned int zno = null_zone_no(dev, sector);
 	struct nullb_zone *zone = &dev->zones[zno];
+	blk_status_t badblocks_ret = BLK_STS_OK;
 	blk_status_t ret;
 
 	trace_nullb_zone_op(cmd, zno, zone->cond);
@@ -412,9 +413,20 @@ static blk_status_t null_zone_write(struct nullb_cmd *cmd, sector_t sector,
 		zone->cond = BLK_ZONE_COND_IMP_OPEN;
 	}
 
-	ret = null_process_cmd(cmd, REQ_OP_WRITE, sector, nr_sectors);
-	if (ret != BLK_STS_OK)
-		goto unlock_zone;
+	if (dev->badblocks.shift != -1) {
+		badblocks_ret = null_handle_badblocks(cmd, sector, &nr_sectors);
+		if (badblocks_ret != BLK_STS_OK && !nr_sectors) {
+			ret = badblocks_ret;
+			goto unlock_zone;
+		}
+	}
+
+	if (dev->memory_backed) {
+		ret = null_handle_memory_backed(cmd, REQ_OP_WRITE, sector,
+						nr_sectors);
+		if (ret != BLK_STS_OK)
+			goto unlock_zone;
+	}
 
 	zone->wp += nr_sectors;
 	if (zone->wp == zone->start + zone->capacity) {
@@ -429,7 +441,7 @@ static blk_status_t null_zone_write(struct nullb_cmd *cmd, sector_t sector,
 		zone->cond = BLK_ZONE_COND_FULL;
 	}
 
-	ret = BLK_STS_OK;
+	ret = badblocks_ret;
 
 unlock_zone:
 	null_unlock_zone(dev, zone);
diff --git a/drivers/block/rnbd/rnbd-clt.c b/drivers/block/rnbd/rnbd-clt.c
index 82467ecde7ec..15627417f12e 100644
--- a/drivers/block/rnbd/rnbd-clt.c
+++ b/drivers/block/rnbd/rnbd-clt.c
@@ -1010,7 +1010,7 @@ static int rnbd_client_xfer_request(struct rnbd_clt_dev *dev,
 	 * See queue limits.
 	 */
 	if ((req_op(rq) != REQ_OP_DISCARD) && (req_op(rq) != REQ_OP_WRITE_ZEROES))
-		sg_cnt = blk_rq_map_sg(dev->queue, rq, iu->sgt.sgl);
+		sg_cnt = blk_rq_map_sg(rq, iu->sgt.sgl);
 
 	if (sg_cnt == 0)
 		sg_mark_end(&iu->sgt.sgl[0]);
diff --git a/drivers/block/sunvdc.c b/drivers/block/sunvdc.c
index 282f81616a78..2b33fb5b949b 100644
--- a/drivers/block/sunvdc.c
+++ b/drivers/block/sunvdc.c
@@ -485,7 +485,7 @@ static int __send_request(struct request *req)
 	}
 
 	sg_init_table(sg, port->ring_cookies);
-	nsg = blk_rq_map_sg(req->q, req, sg);
+	nsg = blk_rq_map_sg(req, sg);
 
 	len = 0;
 	for (i = 0; i < nsg; i++)
diff --git a/drivers/block/ublk_drv.c b/drivers/block/ublk_drv.c
index 6d7b26ab434f..c060da409ed8 100644
--- a/drivers/block/ublk_drv.c
+++ b/drivers/block/ublk_drv.c
@@ -73,11 +73,10 @@
 /* All UBLK_PARAM_TYPE_* should be included here */
 #define UBLK_PARAM_TYPE_ALL                                \
 	(UBLK_PARAM_TYPE_BASIC | UBLK_PARAM_TYPE_DISCARD | \
-	 UBLK_PARAM_TYPE_DEVT | UBLK_PARAM_TYPE_ZONED)
+	 UBLK_PARAM_TYPE_DEVT | UBLK_PARAM_TYPE_ZONED |    \
+	 UBLK_PARAM_TYPE_DMA_ALIGN)
 
 struct ublk_rq_data {
-	struct llist_node node;
-
 	struct kref ref;
 };
 
@@ -144,8 +143,6 @@ struct ublk_queue {
 	struct task_struct	*ubq_daemon;
 	char *io_cmd_buf;
 
-	struct llist_head	io_cmds;
-
 	unsigned long io_addr;	/* mapped vm address */
 	unsigned int max_io_sz;
 	bool force_abort;
@@ -494,15 +491,17 @@ static wait_queue_head_t ublk_idr_wq;	/* wait until one idr is freed */
 
 static DEFINE_MUTEX(ublk_ctl_mutex);
 
+
+#define UBLK_MAX_UBLKS UBLK_MINORS
+
 /*
- * Max ublk devices allowed to add
+ * Max unprivileged ublk devices allowed to add
  *
  * It can be extended to one per-user limit in future or even controlled
  * by cgroup.
  */
-#define UBLK_MAX_UBLKS UBLK_MINORS
-static unsigned int ublks_max = 64;
-static unsigned int ublks_added;	/* protected by ublk_ctl_mutex */
+static unsigned int unprivileged_ublks_max = 64;
+static unsigned int unprivileged_ublks_added; /* protected by ublk_ctl_mutex */
 
 static struct miscdevice ublk_misc;
 
@@ -573,6 +572,16 @@ static int ublk_validate_params(const struct ublk_device *ub)
 	else if (ublk_dev_is_zoned(ub))
 		return -EINVAL;
 
+	if (ub->params.types & UBLK_PARAM_TYPE_DMA_ALIGN) {
+		const struct ublk_param_dma_align *p = &ub->params.dma;
+
+		if (p->alignment >= PAGE_SIZE)
+			return -EINVAL;
+
+		if (!is_power_of_2(p->alignment + 1))
+			return -EINVAL;
+	}
+
 	return 0;
 }
 
@@ -1100,7 +1109,7 @@ static void ublk_complete_rq(struct kref *ref)
 }
 
 /*
- * Since __ublk_rq_task_work always fails requests immediately during
+ * Since ublk_rq_task_work_cb always fails requests immediately during
  * exiting, __ublk_fail_req() is only called from abort context during
  * exiting. So lock is unnecessary.
  *
@@ -1146,11 +1155,14 @@ static inline void __ublk_abort_rq(struct ublk_queue *ubq,
 		blk_mq_end_request(rq, BLK_STS_IOERR);
 }
 
-static inline void __ublk_rq_task_work(struct request *req,
-				       unsigned issue_flags)
+static void ublk_rq_task_work_cb(struct io_uring_cmd *cmd,
+				 unsigned int issue_flags)
 {
-	struct ublk_queue *ubq = req->mq_hctx->driver_data;
-	int tag = req->tag;
+	struct ublk_uring_cmd_pdu *pdu = ublk_get_uring_cmd_pdu(cmd);
+	struct ublk_queue *ubq = pdu->ubq;
+	int tag = pdu->tag;
+	struct request *req = blk_mq_tag_to_rq(
+		ubq->dev->tag_set.tags[ubq->q_id], tag);
 	struct ublk_io *io = &ubq->ios[tag];
 	unsigned int mapped_bytes;
 
@@ -1225,34 +1237,11 @@ static inline void __ublk_rq_task_work(struct request *req,
 	ubq_complete_io_cmd(io, UBLK_IO_RES_OK, issue_flags);
 }
 
-static inline void ublk_forward_io_cmds(struct ublk_queue *ubq,
-					unsigned issue_flags)
-{
-	struct llist_node *io_cmds = llist_del_all(&ubq->io_cmds);
-	struct ublk_rq_data *data, *tmp;
-
-	io_cmds = llist_reverse_order(io_cmds);
-	llist_for_each_entry_safe(data, tmp, io_cmds, node)
-		__ublk_rq_task_work(blk_mq_rq_from_pdu(data), issue_flags);
-}
-
-static void ublk_rq_task_work_cb(struct io_uring_cmd *cmd, unsigned issue_flags)
-{
-	struct ublk_uring_cmd_pdu *pdu = ublk_get_uring_cmd_pdu(cmd);
-	struct ublk_queue *ubq = pdu->ubq;
-
-	ublk_forward_io_cmds(ubq, issue_flags);
-}
-
 static void ublk_queue_cmd(struct ublk_queue *ubq, struct request *rq)
 {
-	struct ublk_rq_data *data = blk_mq_rq_to_pdu(rq);
+	struct ublk_io *io = &ubq->ios[rq->tag];
 
-	if (llist_add(&data->node, &ubq->io_cmds)) {
-		struct ublk_io *io = &ubq->ios[rq->tag];
-
-		io_uring_cmd_complete_in_task(io->cmd, ublk_rq_task_work_cb);
-	}
+	io_uring_cmd_complete_in_task(io->cmd, ublk_rq_task_work_cb);
 }
 
 static enum blk_eh_timer_return ublk_timeout(struct request *rq)
@@ -1445,7 +1434,7 @@ static void ublk_abort_queue(struct ublk_device *ub, struct ublk_queue *ubq)
 			struct request *rq;
 
 			/*
-			 * Either we fail the request or ublk_rq_task_work_fn
+			 * Either we fail the request or ublk_rq_task_work_cb
 			 * will do it
 			 */
 			rq = blk_mq_tag_to_rq(ub->tag_set.tags[ubq->q_id], i);
@@ -1911,10 +1900,9 @@ static int __ublk_ch_uring_cmd(struct io_uring_cmd *cmd,
 	return -EIOCBQUEUED;
 
  out:
-	io_uring_cmd_done(cmd, ret, 0, issue_flags);
 	pr_devel("%s: complete: cmd op %d, tag %d ret %x io_flags %x\n",
 			__func__, cmd_op, tag, ret, io->flags);
-	return -EIOCBQUEUED;
+	return ret;
 }
 
 static inline struct request *__ublk_check_and_get_req(struct ublk_device *ub,
@@ -1970,7 +1958,10 @@ static inline int ublk_ch_uring_cmd_local(struct io_uring_cmd *cmd,
 static void ublk_ch_uring_cmd_cb(struct io_uring_cmd *cmd,
 		unsigned int issue_flags)
 {
-	ublk_ch_uring_cmd_local(cmd, issue_flags);
+	int ret = ublk_ch_uring_cmd_local(cmd, issue_flags);
+
+	if (ret != -EIOCBQUEUED)
+		io_uring_cmd_done(cmd, ret, 0, issue_flags);
 }
 
 static int ublk_ch_uring_cmd(struct io_uring_cmd *cmd, unsigned int issue_flags)
@@ -2235,7 +2226,8 @@ static int ublk_add_chdev(struct ublk_device *ub)
 	if (ret)
 		goto fail;
 
-	ublks_added++;
+	if (ub->dev_info.flags & UBLK_F_UNPRIVILEGED_DEV)
+		unprivileged_ublks_added++;
 	return 0;
  fail:
 	put_device(dev);
@@ -2264,11 +2256,16 @@ static int ublk_add_tag_set(struct ublk_device *ub)
 
 static void ublk_remove(struct ublk_device *ub)
 {
+	bool unprivileged;
+
 	ublk_stop_dev(ub);
 	cancel_work_sync(&ub->nosrv_work);
 	cdev_device_del(&ub->cdev, &ub->cdev_dev);
+	unprivileged = ub->dev_info.flags & UBLK_F_UNPRIVILEGED_DEV;
 	ublk_put_device(ub);
-	ublks_added--;
+
+	if (unprivileged)
+		unprivileged_ublks_added--;
 }
 
 static struct ublk_device *ublk_get_device_from_id(int idx)
@@ -2343,6 +2340,9 @@ static int ublk_ctrl_start_dev(struct ublk_device *ub, struct io_uring_cmd *cmd)
 	if (ub->params.basic.attrs & UBLK_ATTR_ROTATIONAL)
 		lim.features |= BLK_FEAT_ROTATIONAL;
 
+	if (ub->params.types & UBLK_PARAM_TYPE_DMA_ALIGN)
+		lim.dma_alignment = ub->params.dma.alignment;
+
 	if (wait_for_completion_interruptible(&ub->completion) != 0)
 		return -EINTR;
 
@@ -2530,7 +2530,8 @@ static int ublk_ctrl_add_dev(struct io_uring_cmd *cmd)
 		return ret;
 
 	ret = -EACCES;
-	if (ublks_added >= ublks_max)
+	if ((info.flags & UBLK_F_UNPRIVILEGED_DEV) &&
+	    unprivileged_ublks_added >= unprivileged_ublks_max)
 		goto out_unlock;
 
 	ret = -ENOMEM;
@@ -3101,10 +3102,9 @@ static int ublk_ctrl_uring_cmd(struct io_uring_cmd *cmd,
 	if (ub)
 		ublk_put_device(ub);
  out:
-	io_uring_cmd_done(cmd, ret, 0, issue_flags);
 	pr_devel("%s: cmd done ret %d cmd_op %x, dev id %d qid %d\n",
 			__func__, ret, cmd->cmd_op, header->dev_id, header->queue_id);
-	return -EIOCBQUEUED;
+	return ret;
 }
 
 static const struct file_operations ublk_ctl_fops = {
@@ -3168,23 +3168,26 @@ static void __exit ublk_exit(void)
 module_init(ublk_init);
 module_exit(ublk_exit);
 
-static int ublk_set_max_ublks(const char *buf, const struct kernel_param *kp)
+static int ublk_set_max_unprivileged_ublks(const char *buf,
+					   const struct kernel_param *kp)
 {
 	return param_set_uint_minmax(buf, kp, 0, UBLK_MAX_UBLKS);
 }
 
-static int ublk_get_max_ublks(char *buf, const struct kernel_param *kp)
+static int ublk_get_max_unprivileged_ublks(char *buf,
+					   const struct kernel_param *kp)
 {
-	return sysfs_emit(buf, "%u\n", ublks_max);
+	return sysfs_emit(buf, "%u\n", unprivileged_ublks_max);
 }
 
-static const struct kernel_param_ops ublk_max_ublks_ops = {
-	.set = ublk_set_max_ublks,
-	.get = ublk_get_max_ublks,
+static const struct kernel_param_ops ublk_max_unprivileged_ublks_ops = {
+	.set = ublk_set_max_unprivileged_ublks,
+	.get = ublk_get_max_unprivileged_ublks,
 };
 
-module_param_cb(ublks_max, &ublk_max_ublks_ops, &ublks_max, 0644);
-MODULE_PARM_DESC(ublks_max, "max number of ublk devices allowed to add(default: 64)");
+module_param_cb(ublks_max, &ublk_max_unprivileged_ublks_ops,
+		&unprivileged_ublks_max, 0644);
+MODULE_PARM_DESC(ublks_max, "max number of unprivileged ublk devices allowed to add(default: 64)");
 
 MODULE_AUTHOR("Ming Lei <ming.lei@redhat.com>");
 MODULE_DESCRIPTION("Userspace block device");
diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
index 91cde76a4b3e..7cffea01d868 100644
--- a/drivers/block/virtio_blk.c
+++ b/drivers/block/virtio_blk.c
@@ -226,7 +226,7 @@ static int virtblk_map_data(struct blk_mq_hw_ctx *hctx, struct request *req,
 	if (unlikely(err))
 		return -ENOMEM;
 
-	return blk_rq_map_sg(hctx->queue, req, vbr->sg_table.sgl);
+	return blk_rq_map_sg(req, vbr->sg_table.sgl);
 }
 
 static void virtblk_cleanup_cmd(struct request *req)
diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c
index edcd08a9dcef..5babe575c288 100644
--- a/drivers/block/xen-blkfront.c
+++ b/drivers/block/xen-blkfront.c
@@ -751,7 +751,7 @@ static int blkif_queue_rw_req(struct request *req, struct blkfront_ring_info *ri
 	id = blkif_ring_get_request(rinfo, req, &final_ring_req);
 	ring_req = &rinfo->shadow[id].req;
 
-	num_sg = blk_rq_map_sg(req->q, req, rinfo->shadow[id].sg);
+	num_sg = blk_rq_map_sg(req, rinfo->shadow[id].sg);
 	num_grant = 0;
 	/* Calculate the number of grant used */
 	for_each_sg(rinfo->shadow[id].sg, sg, num_sg, i)