From 5c5028ee594ce5f907ca6ad1c32cca6a15098464 Mon Sep 17 00:00:00 2001
From: Keith Busch <kbusch@kernel.org>
Date: Mon, 20 Oct 2025 13:47:15 -0700
Subject: block: rename min_segment_size

Despite its name, the block layer is fine with segments smaller that the
"min_segment_size" limit. The value is an optimization limit indicating
the largest segment that can be used without considering boundary
limits. Smaller segments can take a fast path, so give it a name that
reflects that: max_fast_segment_size.

Signed-off-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 include/linux/blkdev.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux/blkdev.h')

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 70b671a9a7f7..99be263b31ab 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -378,7 +378,7 @@ struct queue_limits {
 	unsigned int		max_sectors;
 	unsigned int		max_user_sectors;
 	unsigned int		max_segment_size;
-	unsigned int		min_segment_size;
+	unsigned int		max_fast_segment_size;
 	unsigned int		physical_block_size;
 	unsigned int		logical_block_size;
 	unsigned int		alignment_offset;
-- 
cgit v1.2.3


From fdb9aed869f34d776298b3a8197909eb820e4d0d Mon Sep 17 00:00:00 2001
From: Damien Le Moal <dlemoal@kernel.org>
Date: Wed, 5 Nov 2025 06:22:38 +0900
Subject: block: introduce disk_report_zone()

Commit b76b840fd933 ("dm: Fix dm-zoned-reclaim zone write pointer
alignment") introduced an indirect call for the callback function of a
report zones executed with blkdev_report_zones(). This is necessary so
that the function disk_zone_wplug_sync_wp_offset() can be called to
refresh a zone write plug zone write pointer offset after a write error.
However, this solution makes following the path of a zone information
harder to understand.

Clean this up by introducing the new blk_report_zones_args structure to
define a zone report callback and its private data and introduce the
helper function disk_report_zone() which calls both
disk_zone_wplug_sync_wp_offset() and the zone report user callback
function for all zones of a zone report. This helper function must be
called by all block device drivers that implement the report zones
block operation in order to correctly report a zone information.

All block device drivers supporting the report_zones block operation are
updated to use this new scheme.

Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 block/blk-zoned.c                 | 79 +++++++++++++++++++++------------------
 drivers/block/null_blk/null_blk.h |  3 +-
 drivers/block/null_blk/zoned.c    |  4 +-
 drivers/block/ublk_drv.c          |  4 +-
 drivers/block/virtio_blk.c        | 11 +++---
 drivers/block/zloop.c             |  4 +-
 drivers/md/dm-zone.c              | 54 ++++++++++++++------------
 drivers/md/dm.h                   |  3 +-
 drivers/nvme/host/core.c          |  5 +--
 drivers/nvme/host/multipath.c     |  4 +-
 drivers/nvme/host/nvme.h          |  2 +-
 drivers/nvme/host/zns.c           | 10 ++---
 drivers/scsi/sd.h                 |  2 +-
 drivers/scsi/sd_zbc.c             | 20 ++++------
 include/linux/blkdev.h            |  7 +++-
 include/linux/device-mapper.h     | 10 ++++-
 16 files changed, 120 insertions(+), 102 deletions(-)

(limited to 'include/linux/blkdev.h')

diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index 345a99c0b031..de3524c17f67 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -114,30 +114,16 @@ const char *blk_zone_cond_str(enum blk_zone_cond zone_cond)
 }
 EXPORT_SYMBOL_GPL(blk_zone_cond_str);
 
-struct disk_report_zones_cb_args {
-	struct gendisk	*disk;
-	report_zones_cb	user_cb;
-	void		*user_data;
+/*
+ * Zone report arguments for block device drivers report_zones operation.
+ * @cb: report_zones_cb callback for each reported zone.
+ * @data: Private data passed to report_zones_cb.
+ */
+struct blk_report_zones_args {
+	report_zones_cb cb;
+	void		*data;
 };
 
-static void disk_zone_wplug_sync_wp_offset(struct gendisk *disk,
-					   struct blk_zone *zone);
-
-static int disk_report_zones_cb(struct blk_zone *zone, unsigned int idx,
-				void *data)
-{
-	struct disk_report_zones_cb_args *args = data;
-	struct gendisk *disk = args->disk;
-
-	if (disk->zone_wplugs_hash)
-		disk_zone_wplug_sync_wp_offset(disk, zone);
-
-	if (!args->user_cb)
-		return 0;
-
-	return args->user_cb(zone, idx, args->user_data);
-}
-
 /**
  * blkdev_report_zones - Get zones information
  * @bdev:	Target block device
@@ -161,10 +147,9 @@ int blkdev_report_zones(struct block_device *bdev, sector_t sector,
 			unsigned int nr_zones, report_zones_cb cb, void *data)
 {
 	struct gendisk *disk = bdev->bd_disk;
-	struct disk_report_zones_cb_args args = {
-		.disk = disk,
-		.user_cb = cb,
-		.user_data = data,
+	struct blk_report_zones_args args = {
+		.cb = cb,
+		.data = data,
 	};
 
 	if (!bdev_is_zoned(bdev) || WARN_ON_ONCE(!disk->fops->report_zones))
@@ -173,8 +158,7 @@ int blkdev_report_zones(struct block_device *bdev, sector_t sector,
 	if (!nr_zones || sector >= get_capacity(disk))
 		return 0;
 
-	return disk->fops->report_zones(disk, sector, nr_zones,
-					disk_report_zones_cb, &args);
+	return disk->fops->report_zones(disk, sector, nr_zones, &args);
 }
 EXPORT_SYMBOL_GPL(blkdev_report_zones);
 
@@ -692,15 +676,32 @@ static void disk_zone_wplug_sync_wp_offset(struct gendisk *disk,
 	disk_put_zone_wplug(zwplug);
 }
 
-static int disk_zone_sync_wp_offset(struct gendisk *disk, sector_t sector)
+/**
+ * disk_report_zone - Report one zone
+ * @disk:	Target disk
+ * @zone:	The zone to report
+ * @idx:	The index of the zone in the overall zone report
+ * @args:	report zones callback and data
+ *
+ * Description:
+ *    Helper function for block device drivers to report one zone of a zone
+ *    report initiated with blkdev_report_zones(). The zone being reported is
+ *    specified by @zone and used to update, if necessary, the zone write plug
+ *    information for the zone. If @args specifies a user callback function,
+ *    this callback is executed.
+ */
+int disk_report_zone(struct gendisk *disk, struct blk_zone *zone,
+		     unsigned int idx, struct blk_report_zones_args *args)
 {
-	struct disk_report_zones_cb_args args = {
-		.disk = disk,
-	};
+	if (disk->zone_wplugs_hash)
+		disk_zone_wplug_sync_wp_offset(disk, zone);
+
+	if (args && args->cb)
+		return args->cb(zone, idx, args->data);
 
-	return disk->fops->report_zones(disk, sector, 1,
-					disk_report_zones_cb, &args);
+	return 0;
 }
+EXPORT_SYMBOL_GPL(disk_report_zone);
 
 static void blk_zone_reset_bio_endio(struct bio *bio)
 {
@@ -1786,6 +1787,10 @@ int blk_revalidate_disk_zones(struct gendisk *disk)
 	sector_t capacity = get_capacity(disk);
 	struct blk_revalidate_zone_args args = { };
 	unsigned int memflags, noio_flag;
+	struct blk_report_zones_args rep_args = {
+		.cb = blk_revalidate_zone_cb,
+		.data = &args,
+	};
 	int ret = -ENOMEM;
 
 	if (WARN_ON_ONCE(!blk_queue_is_zoned(q)))
@@ -1817,8 +1822,7 @@ int blk_revalidate_disk_zones(struct gendisk *disk)
 		return ret;
 	}
 
-	ret = disk->fops->report_zones(disk, 0, UINT_MAX,
-				       blk_revalidate_zone_cb, &args);
+	ret = disk->fops->report_zones(disk, 0, UINT_MAX, &rep_args);
 	if (!ret) {
 		pr_warn("%s: No zones reported\n", disk->disk_name);
 		ret = -ENODEV;
@@ -1863,6 +1867,7 @@ EXPORT_SYMBOL_GPL(blk_revalidate_disk_zones);
 int blk_zone_issue_zeroout(struct block_device *bdev, sector_t sector,
 			   sector_t nr_sects, gfp_t gfp_mask)
 {
+	struct gendisk *disk = bdev->bd_disk;
 	int ret;
 
 	if (WARN_ON_ONCE(!bdev_is_zoned(bdev)))
@@ -1878,7 +1883,7 @@ int blk_zone_issue_zeroout(struct block_device *bdev, sector_t sector,
 	 * pointer. Undo this using a report zone to update the zone write
 	 * pointer to the correct current value.
 	 */
-	ret = disk_zone_sync_wp_offset(bdev->bd_disk, sector);
+	ret = disk->fops->report_zones(disk, sector, 1, NULL);
 	if (ret != 1)
 		return ret < 0 ? ret : -EIO;
 
diff --git a/drivers/block/null_blk/null_blk.h b/drivers/block/null_blk/null_blk.h
index 7bb6128dbaaf..6c4c4bbe7dad 100644
--- a/drivers/block/null_blk/null_blk.h
+++ b/drivers/block/null_blk/null_blk.h
@@ -143,7 +143,8 @@ int null_init_zoned_dev(struct nullb_device *dev, struct queue_limits *lim);
 int null_register_zoned_dev(struct nullb *nullb);
 void null_free_zoned_dev(struct nullb_device *dev);
 int null_report_zones(struct gendisk *disk, sector_t sector,
-		      unsigned int nr_zones, report_zones_cb cb, void *data);
+		      unsigned int nr_zones,
+		      struct blk_report_zones_args *args);
 blk_status_t null_process_zoned_cmd(struct nullb_cmd *cmd, enum req_op op,
 				    sector_t sector, sector_t nr_sectors);
 size_t null_zone_valid_read_len(struct nullb *nullb,
diff --git a/drivers/block/null_blk/zoned.c b/drivers/block/null_blk/zoned.c
index 4e5728f45989..6a93b12a06ff 100644
--- a/drivers/block/null_blk/zoned.c
+++ b/drivers/block/null_blk/zoned.c
@@ -191,7 +191,7 @@ void null_free_zoned_dev(struct nullb_device *dev)
 }
 
 int null_report_zones(struct gendisk *disk, sector_t sector,
-		unsigned int nr_zones, report_zones_cb cb, void *data)
+		unsigned int nr_zones, struct blk_report_zones_args *args)
 {
 	struct nullb *nullb = disk->private_data;
 	struct nullb_device *dev = nullb->dev;
@@ -225,7 +225,7 @@ int null_report_zones(struct gendisk *disk, sector_t sector,
 		blkz.capacity = zone->capacity;
 		null_unlock_zone(dev, zone);
 
-		error = cb(&blkz, i, data);
+		error = disk_report_zone(disk, &blkz, i, args);
 		if (error)
 			return error;
 	}
diff --git a/drivers/block/ublk_drv.c b/drivers/block/ublk_drv.c
index 96e07763cd28..97cc4bc0a6ce 100644
--- a/drivers/block/ublk_drv.c
+++ b/drivers/block/ublk_drv.c
@@ -367,7 +367,7 @@ static void *ublk_alloc_report_buffer(struct ublk_device *ublk,
 }
 
 static int ublk_report_zones(struct gendisk *disk, sector_t sector,
-		      unsigned int nr_zones, report_zones_cb cb, void *data)
+		      unsigned int nr_zones, struct blk_report_zones_args *args)
 {
 	struct ublk_device *ub = disk->private_data;
 	unsigned int zone_size_sectors = disk->queue->limits.chunk_sectors;
@@ -430,7 +430,7 @@ free_req:
 			if (!zone->len)
 				break;
 
-			ret = cb(zone, i, data);
+			ret = disk_report_zone(disk, zone, i, args);
 			if (ret)
 				goto out;
 
diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
index f061420dfb10..a5e97f03dbf0 100644
--- a/drivers/block/virtio_blk.c
+++ b/drivers/block/virtio_blk.c
@@ -584,7 +584,8 @@ out:
 
 static int virtblk_parse_zone(struct virtio_blk *vblk,
 			       struct virtio_blk_zone_descriptor *entry,
-			       unsigned int idx, report_zones_cb cb, void *data)
+			       unsigned int idx,
+			       struct blk_report_zones_args *args)
 {
 	struct blk_zone zone = { };
 
@@ -650,12 +651,12 @@ static int virtblk_parse_zone(struct virtio_blk *vblk,
 	 * The callback below checks the validity of the reported
 	 * entry data, no need to further validate it here.
 	 */
-	return cb(&zone, idx, data);
+	return disk_report_zone(vblk->disk, &zone, idx, args);
 }
 
 static int virtblk_report_zones(struct gendisk *disk, sector_t sector,
-				 unsigned int nr_zones, report_zones_cb cb,
-				 void *data)
+				 unsigned int nr_zones,
+				 struct blk_report_zones_args *args)
 {
 	struct virtio_blk *vblk = disk->private_data;
 	struct virtio_blk_zone_report *report;
@@ -693,7 +694,7 @@ static int virtblk_report_zones(struct gendisk *disk, sector_t sector,
 
 		for (i = 0; i < nz && zone_idx < nr_zones; i++) {
 			ret = virtblk_parse_zone(vblk, &report->zones[i],
-						 zone_idx, cb, data);
+						 zone_idx, args);
 			if (ret)
 				goto fail_report;
 
diff --git a/drivers/block/zloop.c b/drivers/block/zloop.c
index a423228e201b..92be9f0af00a 100644
--- a/drivers/block/zloop.c
+++ b/drivers/block/zloop.c
@@ -647,7 +647,7 @@ static int zloop_open(struct gendisk *disk, blk_mode_t mode)
 }
 
 static int zloop_report_zones(struct gendisk *disk, sector_t sector,
-		unsigned int nr_zones, report_zones_cb cb, void *data)
+		unsigned int nr_zones, struct blk_report_zones_args *args)
 {
 	struct zloop_device *zlo = disk->private_data;
 	struct blk_zone blkz = {};
@@ -687,7 +687,7 @@ static int zloop_report_zones(struct gendisk *disk, sector_t sector,
 
 		mutex_unlock(&zone->lock);
 
-		ret = cb(&blkz, i, data);
+		ret = disk_report_zone(disk, &blkz, i, args);
 		if (ret)
 			return ret;
 	}
diff --git a/drivers/md/dm-zone.c b/drivers/md/dm-zone.c
index 78e17dd4d01b..984fb621b0e9 100644
--- a/drivers/md/dm-zone.c
+++ b/drivers/md/dm-zone.c
@@ -17,33 +17,26 @@
  * For internal zone reports bypassing the top BIO submission path.
  */
 static int dm_blk_do_report_zones(struct mapped_device *md, struct dm_table *t,
-				  sector_t sector, unsigned int nr_zones,
-				  report_zones_cb cb, void *data)
+				  unsigned int nr_zones,
+				  struct dm_report_zones_args *args)
 {
-	struct gendisk *disk = md->disk;
-	int ret;
-	struct dm_report_zones_args args = {
-		.next_sector = sector,
-		.orig_data = data,
-		.orig_cb = cb,
-	};
-
 	do {
 		struct dm_target *tgt;
+		int ret;
 
-		tgt = dm_table_find_target(t, args.next_sector);
+		tgt = dm_table_find_target(t, args->next_sector);
 		if (WARN_ON_ONCE(!tgt->type->report_zones))
 			return -EIO;
 
-		args.tgt = tgt;
-		ret = tgt->type->report_zones(tgt, &args,
-					      nr_zones - args.zone_idx);
+		args->tgt = tgt;
+		ret = tgt->type->report_zones(tgt, args,
+					      nr_zones - args->zone_idx);
 		if (ret < 0)
 			return ret;
-	} while (args.zone_idx < nr_zones &&
-		 args.next_sector < get_capacity(disk));
+	} while (args->zone_idx < nr_zones &&
+		 args->next_sector < get_capacity(md->disk));
 
-	return args.zone_idx;
+	return args->zone_idx;
 }
 
 /*
@@ -52,7 +45,8 @@ static int dm_blk_do_report_zones(struct mapped_device *md, struct dm_table *t,
  * generally implemented by targets using dm_report_zones().
  */
 int dm_blk_report_zones(struct gendisk *disk, sector_t sector,
-			unsigned int nr_zones, report_zones_cb cb, void *data)
+			unsigned int nr_zones,
+			struct blk_report_zones_args *args)
 {
 	struct mapped_device *md = disk->private_data;
 	struct dm_table *map;
@@ -76,9 +70,14 @@ int dm_blk_report_zones(struct gendisk *disk, sector_t sector,
 		map = zone_revalidate_map;
 	}
 
-	if (map)
-		ret = dm_blk_do_report_zones(md, map, sector, nr_zones, cb,
-					     data);
+	if (map) {
+		struct dm_report_zones_args dm_args = {
+			.disk = md->disk,
+			.next_sector = sector,
+			.rep_args = args,
+		};
+		ret = dm_blk_do_report_zones(md, map, nr_zones, &dm_args);
+	}
 
 	if (put_table)
 		dm_put_live_table(md, srcu_idx);
@@ -113,7 +112,9 @@ static int dm_report_zones_cb(struct blk_zone *zone, unsigned int idx,
 	}
 
 	args->next_sector = zone->start + zone->len;
-	return args->orig_cb(zone, args->zone_idx++, args->orig_data);
+
+	return disk_report_zone(args->disk, zone, args->zone_idx++,
+				args->rep_args);
 }
 
 /*
@@ -492,10 +493,15 @@ int dm_zone_get_reset_bitmap(struct mapped_device *md, struct dm_table *t,
 			     sector_t sector, unsigned int nr_zones,
 			     unsigned long *need_reset)
 {
+	struct dm_report_zones_args args = {
+		.disk = md->disk,
+		.next_sector = sector,
+		.cb = dm_zone_need_reset_cb,
+		.data = need_reset,
+	};
 	int ret;
 
-	ret = dm_blk_do_report_zones(md, t, sector, nr_zones,
-				     dm_zone_need_reset_cb, need_reset);
+	ret = dm_blk_do_report_zones(md, t, nr_zones, &args);
 	if (ret != nr_zones) {
 		DMERR("Get %s zone reset bitmap failed\n",
 		      md->disk->disk_name);
diff --git a/drivers/md/dm.h b/drivers/md/dm.h
index 245f52b59215..7a795979ec72 100644
--- a/drivers/md/dm.h
+++ b/drivers/md/dm.h
@@ -109,7 +109,8 @@ void dm_finalize_zone_settings(struct dm_table *t, struct queue_limits *lim);
 void dm_zone_endio(struct dm_io *io, struct bio *clone);
 #ifdef CONFIG_BLK_DEV_ZONED
 int dm_blk_report_zones(struct gendisk *disk, sector_t sector,
-			unsigned int nr_zones, report_zones_cb cb, void *data);
+			unsigned int nr_zones,
+			struct blk_report_zones_args *args);
 bool dm_is_zone_write(struct mapped_device *md, struct bio *bio);
 int dm_zone_get_reset_bitmap(struct mapped_device *md, struct dm_table *t,
 			     sector_t sector, unsigned int nr_zones,
diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
index fa4181d7de73..c0fe50fb7b08 100644
--- a/drivers/nvme/host/core.c
+++ b/drivers/nvme/host/core.c
@@ -2599,10 +2599,9 @@ static void nvme_configure_opal(struct nvme_ctrl *ctrl, bool was_suspended)
 
 #ifdef CONFIG_BLK_DEV_ZONED
 static int nvme_report_zones(struct gendisk *disk, sector_t sector,
-		unsigned int nr_zones, report_zones_cb cb, void *data)
+		unsigned int nr_zones, struct blk_report_zones_args *args)
 {
-	return nvme_ns_report_zones(disk->private_data, sector, nr_zones, cb,
-			data);
+	return nvme_ns_report_zones(disk->private_data, sector, nr_zones, args);
 }
 #else
 #define nvme_report_zones	NULL
diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c
index 543e17aead12..0b7ac0735bd0 100644
--- a/drivers/nvme/host/multipath.c
+++ b/drivers/nvme/host/multipath.c
@@ -576,7 +576,7 @@ static int nvme_ns_head_get_unique_id(struct gendisk *disk, u8 id[16],
 
 #ifdef CONFIG_BLK_DEV_ZONED
 static int nvme_ns_head_report_zones(struct gendisk *disk, sector_t sector,
-		unsigned int nr_zones, report_zones_cb cb, void *data)
+		unsigned int nr_zones, struct blk_report_zones_args *args)
 {
 	struct nvme_ns_head *head = disk->private_data;
 	struct nvme_ns *ns;
@@ -585,7 +585,7 @@ static int nvme_ns_head_report_zones(struct gendisk *disk, sector_t sector,
 	srcu_idx = srcu_read_lock(&head->srcu);
 	ns = nvme_find_path(head);
 	if (ns)
-		ret = nvme_ns_report_zones(ns, sector, nr_zones, cb, data);
+		ret = nvme_ns_report_zones(ns, sector, nr_zones, args);
 	srcu_read_unlock(&head->srcu, srcu_idx);
 	return ret;
 }
diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h
index 102fae6a231c..928c748ccbd1 100644
--- a/drivers/nvme/host/nvme.h
+++ b/drivers/nvme/host/nvme.h
@@ -1108,7 +1108,7 @@ struct nvme_zone_info {
 };
 
 int nvme_ns_report_zones(struct nvme_ns *ns, sector_t sector,
-		unsigned int nr_zones, report_zones_cb cb, void *data);
+		unsigned int nr_zones, struct blk_report_zones_args *args);
 int nvme_query_zone_info(struct nvme_ns *ns, unsigned lbaf,
 		struct nvme_zone_info *zi);
 void nvme_update_zone_info(struct nvme_ns *ns, struct queue_limits *lim,
diff --git a/drivers/nvme/host/zns.c b/drivers/nvme/host/zns.c
index cce4c5b55aa9..deea2dbef5b8 100644
--- a/drivers/nvme/host/zns.c
+++ b/drivers/nvme/host/zns.c
@@ -148,8 +148,8 @@ static void *nvme_zns_alloc_report_buffer(struct nvme_ns *ns,
 
 static int nvme_zone_parse_entry(struct nvme_ns *ns,
 				 struct nvme_zone_descriptor *entry,
-				 unsigned int idx, report_zones_cb cb,
-				 void *data)
+				 unsigned int idx,
+				 struct blk_report_zones_args *args)
 {
 	struct nvme_ns_head *head = ns->head;
 	struct blk_zone zone = { };
@@ -169,11 +169,11 @@ static int nvme_zone_parse_entry(struct nvme_ns *ns,
 	else
 		zone.wp = nvme_lba_to_sect(head, le64_to_cpu(entry->wp));
 
-	return cb(&zone, idx, data);
+	return disk_report_zone(ns->disk, &zone, idx, args);
 }
 
 int nvme_ns_report_zones(struct nvme_ns *ns, sector_t sector,
-		unsigned int nr_zones, report_zones_cb cb, void *data)
+		unsigned int nr_zones, struct blk_report_zones_args *args)
 {
 	struct nvme_zone_report *report;
 	struct nvme_command c = { };
@@ -213,7 +213,7 @@ int nvme_ns_report_zones(struct nvme_ns *ns, sector_t sector,
 
 		for (i = 0; i < nz && zone_idx < nr_zones; i++) {
 			ret = nvme_zone_parse_entry(ns, &report->entries[i],
-						    zone_idx, cb, data);
+						    zone_idx, args);
 			if (ret)
 				goto out_free;
 			zone_idx++;
diff --git a/drivers/scsi/sd.h b/drivers/scsi/sd.h
index 36382eca941c..574af8243016 100644
--- a/drivers/scsi/sd.h
+++ b/drivers/scsi/sd.h
@@ -240,7 +240,7 @@ blk_status_t sd_zbc_setup_zone_mgmt_cmnd(struct scsi_cmnd *cmd,
 unsigned int sd_zbc_complete(struct scsi_cmnd *cmd, unsigned int good_bytes,
 			     struct scsi_sense_hdr *sshdr);
 int sd_zbc_report_zones(struct gendisk *disk, sector_t sector,
-		unsigned int nr_zones, report_zones_cb cb, void *data);
+		unsigned int nr_zones, struct blk_report_zones_args *args);
 
 #else /* CONFIG_BLK_DEV_ZONED */
 
diff --git a/drivers/scsi/sd_zbc.c b/drivers/scsi/sd_zbc.c
index a8db66428f80..56e455fb5add 100644
--- a/drivers/scsi/sd_zbc.c
+++ b/drivers/scsi/sd_zbc.c
@@ -35,8 +35,7 @@ static bool sd_zbc_is_gap_zone(const u8 buf[64])
  * @buf: SCSI zone descriptor.
  * @idx: Index of the zone relative to the first zone reported by the current
  *	sd_zbc_report_zones() call.
- * @cb: Callback function pointer.
- * @data: Second argument passed to @cb.
+ * @args: report zones arguments (callback, etc)
  *
  * Return: Value returned by @cb.
  *
@@ -44,12 +43,11 @@ static bool sd_zbc_is_gap_zone(const u8 buf[64])
  * call @cb(blk_zone, @data).
  */
 static int sd_zbc_parse_report(struct scsi_disk *sdkp, const u8 buf[64],
-			       unsigned int idx, report_zones_cb cb, void *data)
+			unsigned int idx, struct blk_report_zones_args *args)
 {
 	struct scsi_device *sdp = sdkp->device;
 	struct blk_zone zone = { 0 };
 	sector_t start_lba, gran;
-	int ret;
 
 	if (WARN_ON_ONCE(sd_zbc_is_gap_zone(buf)))
 		return -EINVAL;
@@ -87,11 +85,7 @@ static int sd_zbc_parse_report(struct scsi_disk *sdkp, const u8 buf[64],
 	else
 		zone.wp = logical_to_sectors(sdp, get_unaligned_be64(&buf[24]));
 
-	ret = cb(&zone, idx, data);
-	if (ret)
-		return ret;
-
-	return 0;
+	return disk_report_zone(sdkp->disk, &zone, idx, args);
 }
 
 /**
@@ -217,14 +211,14 @@ static inline sector_t sd_zbc_zone_sectors(struct scsi_disk *sdkp)
  * @disk: Disk to report zones for.
  * @sector: Start sector.
  * @nr_zones: Maximum number of zones to report.
- * @cb: Callback function called to report zone information.
- * @data: Second argument passed to @cb.
+ * @args: Callback arguments.
  *
  * Called by the block layer to iterate over zone information. See also the
  * disk->fops->report_zones() calls in block/blk-zoned.c.
  */
 int sd_zbc_report_zones(struct gendisk *disk, sector_t sector,
-			unsigned int nr_zones, report_zones_cb cb, void *data)
+			unsigned int nr_zones,
+			struct blk_report_zones_args *args)
 {
 	struct scsi_disk *sdkp = scsi_disk(disk);
 	sector_t lba = sectors_to_logical(sdkp->device, sector);
@@ -283,7 +277,7 @@ int sd_zbc_report_zones(struct gendisk *disk, sector_t sector,
 			}
 
 			ret = sd_zbc_parse_report(sdkp, buf + offset, zone_idx,
-						  cb, data);
+						  args);
 			if (ret)
 				goto out;
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 99be263b31ab..2f75fb15f55f 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -38,6 +38,7 @@ struct blk_flush_queue;
 struct kiocb;
 struct pr_ops;
 struct rq_qos;
+struct blk_report_zones_args;
 struct blk_queue_stats;
 struct blk_stat_callback;
 struct blk_crypto_profile;
@@ -432,6 +433,9 @@ struct queue_limits {
 typedef int (*report_zones_cb)(struct blk_zone *zone, unsigned int idx,
 			       void *data);
 
+int disk_report_zone(struct gendisk *disk, struct blk_zone *zone,
+		     unsigned int idx, struct blk_report_zones_args *args);
+
 #define BLK_ALL_ZONES  ((unsigned int)-1)
 int blkdev_report_zones(struct block_device *bdev, sector_t sector,
 		unsigned int nr_zones, report_zones_cb cb, void *data);
@@ -1662,7 +1666,8 @@ struct block_device_operations {
 	/* this callback is with swap_lock and sometimes page table lock held */
 	void (*swap_slot_free_notify) (struct block_device *, unsigned long);
 	int (*report_zones)(struct gendisk *, sector_t sector,
-			unsigned int nr_zones, report_zones_cb cb, void *data);
+			    unsigned int nr_zones,
+			    struct blk_report_zones_args *args);
 	char *(*devnode)(struct gendisk *disk, umode_t *mode);
 	/* returns the length of the identifier or a negative errno: */
 	int (*get_unique_id)(struct gendisk *disk, u8 id[16],
diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h
index 84fdc3a6a19a..38f625af6ab4 100644
--- a/include/linux/device-mapper.h
+++ b/include/linux/device-mapper.h
@@ -538,12 +538,18 @@ void dm_submit_bio_remap(struct bio *clone, struct bio *tgt_clone);
 #ifdef CONFIG_BLK_DEV_ZONED
 struct dm_report_zones_args {
 	struct dm_target *tgt;
+	struct gendisk *disk;
 	sector_t next_sector;
 
-	void *orig_data;
-	report_zones_cb orig_cb;
 	unsigned int zone_idx;
 
+	/* for block layer ->report_zones */
+	struct blk_report_zones_args *rep_args;
+
+	/* for internal users */
+	report_zones_cb cb;
+	void *data;
+
 	/* must be filled by ->report_zones before calling dm_report_zones_cb */
 	sector_t start;
 };
-- 
cgit v1.2.3


From 6e945ffb6555705cf20b1fcdc21a139911562995 Mon Sep 17 00:00:00 2001
From: Damien Le Moal <dlemoal@kernel.org>
Date: Wed, 5 Nov 2025 06:22:40 +0900
Subject: block: use zone condition to determine conventional zones

The conv_zones_bitmap field of struct gendisk is used to define a bitmap
to identify the conventional zones of a zoned block device. The bit for
a zone is set in this bitmap if the zone is a conventional one, that is,
if the zone type is BLK_ZONE_TYPE_CONVENTIONAL. For such zone, this
always corresponds to the zone condition BLK_ZONE_COND_NOT_WP.
In other words, conv_zones_bitmap tracks a single condition of the
zones of a zoned block device.

In preparation for tracking more zone conditions, change
conv_zones_bitmap into an array of zone conditions, using 1 byte per
zone. This increases the memory usage from 1 bit per zone to 1 byte per
zone, that is, from 16 KiB to about 100 KiB for a 30 TB SMR HDD with 256
MiB zones. This is a trade-off to allow fast cached report zones later
on top of this change.

Rename the conv_zones_bitmap field of struct gendisk to zones_cond. Add
a blk_revalidate_zone_cond() function to initialize the zones_cond array
of a disk during device scan and to update it on device revalidation.
Move the allocation of the zones_cond array to
disk_revalidate_zone_resources(), making sure that this array is always
allocated, even for devices that do not need zone write plugs (zone
resources), to ensure that bdev_zone_is_seq() can be re-implemented to
use the zone condition array in place of the conv zones bitmap.

Finally, the function bdev_zone_is_seq() is rewritten to use a test on
the condition of the target zone.

Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 block/blk-zoned.c      | 153 ++++++++++++++++++++++++++++++++++---------------
 include/linux/blkdev.h |  37 +++---------
 2 files changed, 117 insertions(+), 73 deletions(-)

(limited to 'include/linux/blkdev.h')

diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index d4fc87b0be6b..f62862274f9a 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -114,6 +114,33 @@ const char *blk_zone_cond_str(enum blk_zone_cond zone_cond)
 }
 EXPORT_SYMBOL_GPL(blk_zone_cond_str);
 
+/**
+ * bdev_zone_is_seq - check if a sector belongs to a sequential write zone
+ * @bdev:       block device to check
+ * @sector:     sector number
+ *
+ * Check if @sector on @bdev is contained in a sequential write required zone.
+ */
+bool bdev_zone_is_seq(struct block_device *bdev, sector_t sector)
+{
+	struct gendisk *disk = bdev->bd_disk;
+	unsigned int zno = disk_zone_no(disk, sector);
+	bool is_seq = false;
+	u8 *zones_cond;
+
+	if (!bdev_is_zoned(bdev))
+		return false;
+
+	rcu_read_lock();
+	zones_cond = rcu_dereference(disk->zones_cond);
+	if (zones_cond && zno < disk->nr_zones)
+		is_seq = zones_cond[zno] != BLK_ZONE_COND_NOT_WP;
+	rcu_read_unlock();
+
+	return is_seq;
+}
+EXPORT_SYMBOL_GPL(bdev_zone_is_seq);
+
 /*
  * Zone report arguments for block device drivers report_zones operation.
  * @cb: report_zones_cb callback for each reported zone.
@@ -1458,22 +1485,16 @@ static void disk_destroy_zone_wplugs_hash_table(struct gendisk *disk)
 	disk->zone_wplugs_hash_bits = 0;
 }
 
-static unsigned int disk_set_conv_zones_bitmap(struct gendisk *disk,
-					       unsigned long *bitmap)
+static void disk_set_zones_cond_array(struct gendisk *disk, u8 *zones_cond)
 {
-	unsigned int nr_conv_zones = 0;
 	unsigned long flags;
 
 	spin_lock_irqsave(&disk->zone_wplugs_lock, flags);
-	if (bitmap)
-		nr_conv_zones = bitmap_weight(bitmap, disk->nr_zones);
-	bitmap = rcu_replace_pointer(disk->conv_zones_bitmap, bitmap,
-				     lockdep_is_held(&disk->zone_wplugs_lock));
+	zones_cond = rcu_replace_pointer(disk->zones_cond, zones_cond,
+				lockdep_is_held(&disk->zone_wplugs_lock));
 	spin_unlock_irqrestore(&disk->zone_wplugs_lock, flags);
 
-	kfree_rcu_mightsleep(bitmap);
-
-	return nr_conv_zones;
+	kfree_rcu_mightsleep(zones_cond);
 }
 
 void disk_free_zone_resources(struct gendisk *disk)
@@ -1497,7 +1518,7 @@ void disk_free_zone_resources(struct gendisk *disk)
 	mempool_destroy(disk->zone_wplugs_pool);
 	disk->zone_wplugs_pool = NULL;
 
-	disk_set_conv_zones_bitmap(disk, NULL);
+	disk_set_zones_cond_array(disk, NULL);
 	disk->zone_capacity = 0;
 	disk->last_zone_capacity = 0;
 	disk->nr_zones = 0;
@@ -1516,12 +1537,31 @@ static inline bool disk_need_zone_resources(struct gendisk *disk)
 		queue_emulates_zone_append(disk->queue);
 }
 
+struct blk_revalidate_zone_args {
+	struct gendisk	*disk;
+	u8		*zones_cond;
+	unsigned int	nr_zones;
+	unsigned int	nr_conv_zones;
+	unsigned int	zone_capacity;
+	unsigned int	last_zone_capacity;
+	sector_t	sector;
+};
+
 static int disk_revalidate_zone_resources(struct gendisk *disk,
-					  unsigned int nr_zones)
+				struct blk_revalidate_zone_args *args)
 {
 	struct queue_limits *lim = &disk->queue->limits;
 	unsigned int pool_size;
 
+	args->disk = disk;
+	args->nr_zones =
+		DIV_ROUND_UP_ULL(get_capacity(disk), lim->chunk_sectors);
+
+	/* Cached zone conditions: 1 byte per zone */
+	args->zones_cond = kzalloc(args->nr_zones, GFP_NOIO);
+	if (!args->zones_cond)
+		return -ENOMEM;
+
 	if (!disk_need_zone_resources(disk))
 		return 0;
 
@@ -1531,7 +1571,8 @@ static int disk_revalidate_zone_resources(struct gendisk *disk,
 	 */
 	pool_size = max(lim->max_open_zones, lim->max_active_zones);
 	if (!pool_size)
-		pool_size = min(BLK_ZONE_WPLUG_DEFAULT_POOL_SIZE, nr_zones);
+		pool_size =
+			min(BLK_ZONE_WPLUG_DEFAULT_POOL_SIZE, args->nr_zones);
 
 	if (!disk->zone_wplugs_hash)
 		return disk_alloc_zone_resources(disk, pool_size);
@@ -1539,15 +1580,6 @@ static int disk_revalidate_zone_resources(struct gendisk *disk,
 	return 0;
 }
 
-struct blk_revalidate_zone_args {
-	struct gendisk	*disk;
-	unsigned long	*conv_zones_bitmap;
-	unsigned int	nr_zones;
-	unsigned int	zone_capacity;
-	unsigned int	last_zone_capacity;
-	sector_t	sector;
-};
-
 /*
  * Update the disk zone resources information and device queue limits.
  * The disk queue is frozen when this is executed.
@@ -1556,7 +1588,7 @@ static int disk_update_zone_resources(struct gendisk *disk,
 				      struct blk_revalidate_zone_args *args)
 {
 	struct request_queue *q = disk->queue;
-	unsigned int nr_seq_zones, nr_conv_zones;
+	unsigned int nr_seq_zones;
 	unsigned int pool_size, memflags;
 	struct queue_limits lim;
 	int ret = 0;
@@ -1566,24 +1598,24 @@ static int disk_update_zone_resources(struct gendisk *disk,
 	memflags = blk_mq_freeze_queue(q);
 
 	disk->nr_zones = args->nr_zones;
-	disk->zone_capacity = args->zone_capacity;
-	disk->last_zone_capacity = args->last_zone_capacity;
-	nr_conv_zones =
-		disk_set_conv_zones_bitmap(disk, args->conv_zones_bitmap);
-	if (nr_conv_zones >= disk->nr_zones) {
+	if (args->nr_conv_zones >= disk->nr_zones) {
 		pr_warn("%s: Invalid number of conventional zones %u / %u\n",
-			disk->disk_name, nr_conv_zones, disk->nr_zones);
+			disk->disk_name, args->nr_conv_zones, disk->nr_zones);
 		ret = -ENODEV;
 		goto unfreeze;
 	}
 
+	disk->zone_capacity = args->zone_capacity;
+	disk->last_zone_capacity = args->last_zone_capacity;
+	disk_set_zones_cond_array(disk, args->zones_cond);
+
 	/*
-	 * Some devices can advertize zone resource limits that are larger than
+	 * Some devices can advertise zone resource limits that are larger than
 	 * the number of sequential zones of the zoned block device, e.g. a
 	 * small ZNS namespace. For such case, assume that the zoned device has
 	 * no zone resource limits.
 	 */
-	nr_seq_zones = disk->nr_zones - nr_conv_zones;
+	nr_seq_zones = disk->nr_zones - args->nr_conv_zones;
 	if (lim.max_open_zones >= nr_seq_zones)
 		lim.max_open_zones = 0;
 	if (lim.max_active_zones >= nr_seq_zones)
@@ -1624,6 +1656,44 @@ unfreeze:
 	return ret;
 }
 
+static int blk_revalidate_zone_cond(struct blk_zone *zone, unsigned int idx,
+				    struct blk_revalidate_zone_args *args)
+{
+	enum blk_zone_cond cond = zone->cond;
+
+	/* Check that the zone condition is consistent with the zone type. */
+	switch (cond) {
+	case BLK_ZONE_COND_NOT_WP:
+		if (zone->type != BLK_ZONE_TYPE_CONVENTIONAL)
+			goto invalid_condition;
+		break;
+	case BLK_ZONE_COND_IMP_OPEN:
+	case BLK_ZONE_COND_EXP_OPEN:
+	case BLK_ZONE_COND_CLOSED:
+	case BLK_ZONE_COND_EMPTY:
+	case BLK_ZONE_COND_FULL:
+	case BLK_ZONE_COND_OFFLINE:
+	case BLK_ZONE_COND_READONLY:
+		if (zone->type != BLK_ZONE_TYPE_SEQWRITE_REQ)
+			goto invalid_condition;
+		break;
+	default:
+		pr_warn("%s: Invalid zone condition 0x%X\n",
+			args->disk->disk_name, cond);
+		return -ENODEV;
+	}
+
+	args->zones_cond[idx] = cond;
+
+	return 0;
+
+invalid_condition:
+	pr_warn("%s: Invalid zone condition 0x%x for type 0x%x\n",
+		args->disk->disk_name, cond, zone->type);
+
+	return -ENODEV;
+}
+
 static int blk_revalidate_conv_zone(struct blk_zone *zone, unsigned int idx,
 				    struct blk_revalidate_zone_args *args)
 {
@@ -1638,17 +1708,7 @@ static int blk_revalidate_conv_zone(struct blk_zone *zone, unsigned int idx,
 	if (disk_zone_is_last(disk, zone))
 		args->last_zone_capacity = zone->capacity;
 
-	if (!disk_need_zone_resources(disk))
-		return 0;
-
-	if (!args->conv_zones_bitmap) {
-		args->conv_zones_bitmap =
-			bitmap_zalloc(args->nr_zones, GFP_NOIO);
-		if (!args->conv_zones_bitmap)
-			return -ENOMEM;
-	}
-
-	set_bit(idx, args->conv_zones_bitmap);
+	args->nr_conv_zones++;
 
 	return 0;
 }
@@ -1746,6 +1806,11 @@ static int blk_revalidate_zone_cb(struct blk_zone *zone, unsigned int idx,
 		return -ENODEV;
 	}
 
+	/* Check zone condition */
+	ret = blk_revalidate_zone_cond(zone, idx, args);
+	if (ret)
+		return ret;
+
 	/* Check zone type */
 	switch (zone->type) {
 	case BLK_ZONE_TYPE_CONVENTIONAL:
@@ -1813,10 +1878,8 @@ int blk_revalidate_disk_zones(struct gendisk *disk)
 	 * Ensure that all memory allocations in this context are done as if
 	 * GFP_NOIO was specified.
 	 */
-	args.disk = disk;
-	args.nr_zones = (capacity + zone_sectors - 1) >> ilog2(zone_sectors);
 	noio_flag = memalloc_noio_save();
-	ret = disk_revalidate_zone_resources(disk, args.nr_zones);
+	ret = disk_revalidate_zone_resources(disk, &args);
 	if (ret) {
 		memalloc_noio_restore(noio_flag);
 		return ret;
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 2f75fb15f55f..53bcfbc2f68f 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -196,7 +196,7 @@ struct gendisk {
 	unsigned int		nr_zones;
 	unsigned int		zone_capacity;
 	unsigned int		last_zone_capacity;
-	unsigned long __rcu	*conv_zones_bitmap;
+	u8 __rcu		*zones_cond;
 	unsigned int		zone_wplugs_hash_bits;
 	atomic_t		nr_zone_wplugs;
 	spinlock_t		zone_wplugs_lock;
@@ -925,12 +925,20 @@ static inline unsigned int bdev_zone_capacity(struct block_device *bdev,
 {
 	return disk_zone_capacity(bdev->bd_disk, pos);
 }
+
+bool bdev_zone_is_seq(struct block_device *bdev, sector_t sector);
+
 #else /* CONFIG_BLK_DEV_ZONED */
 static inline unsigned int disk_nr_zones(struct gendisk *disk)
 {
 	return 0;
 }
 
+static inline bool bdev_zone_is_seq(struct block_device *bdev, sector_t sector)
+{
+	return false;
+}
+
 static inline bool bio_needs_zone_write_plugging(struct bio *bio)
 {
 	return false;
@@ -1533,33 +1541,6 @@ static inline bool bdev_is_zone_aligned(struct block_device *bdev,
 	return bdev_is_zone_start(bdev, sector);
 }
 
-/**
- * bdev_zone_is_seq - check if a sector belongs to a sequential write zone
- * @bdev:	block device to check
- * @sector:	sector number
- *
- * Check if @sector on @bdev is contained in a sequential write required zone.
- */
-static inline bool bdev_zone_is_seq(struct block_device *bdev, sector_t sector)
-{
-	bool is_seq = false;
-
-#if IS_ENABLED(CONFIG_BLK_DEV_ZONED)
-	if (bdev_is_zoned(bdev)) {
-		struct gendisk *disk = bdev->bd_disk;
-		unsigned long *bitmap;
-
-		rcu_read_lock();
-		bitmap = rcu_dereference(disk->conv_zones_bitmap);
-		is_seq = !bitmap ||
-			!test_bit(disk_zone_no(disk, sector), bitmap);
-		rcu_read_unlock();
-	}
-#endif
-
-	return is_seq;
-}
-
 int blk_zone_issue_zeroout(struct block_device *bdev, sector_t sector,
 			   sector_t nr_sects, gfp_t gfp_mask);
 
-- 
cgit v1.2.3


From f2284eec5053df271c78e687672247922bcee881 Mon Sep 17 00:00:00 2001
From: Damien Le Moal <dlemoal@kernel.org>
Date: Wed, 5 Nov 2025 06:22:43 +0900
Subject: block: introduce blkdev_get_zone_info()

Introduce the function blkdev_get_zone_info() to obtain a single zone
information from cached zone data, that is, either from the zone write
plug for the target zone if it exists and from the disk zones_cond
array otherwise.

Since sequential zones that do not have a zone write plug are either
full, empty or in a bad state (read-only or offline), the zone write
pointer can be inferred from the zone condition cached in the disk
zones_cond array. For sequential zones that have a zone write plug, the
zone condition and zone write pointer are obtained from the condition
and write pointer offset managed with the zone write plug. This allows
obtaining the information for a zone much more quickly than having to
execute a report zones command on the device.

blkdev_get_zone_info() falls back to using a regular zone report if the
target zone is flagged as needing an update with the
BLK_ZONE_WPLUG_NEED_WP_UPDATE flag, or if the target device does not
use zone write plugs (i.e. a device mapper device). In this case, the
new function blkdev_report_zone_fallback() is used and the zone
condition is reported consistantly with the cahced report, that is, the
BLK_ZONE_COND_ACTIVE condition is used in place of the implicit open,
explicit open and closed conditions. This is achieved by adding the
.report_active field to struct blk_report_zones_args and by having
disk_report_zone() sets the correct zone condition if .report_active is
true.

In preparation for using blkdev_get_zone_info() in upcoming file systems
changes, also export this function as a GPL symbol.

Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 block/blk-zoned.c      | 141 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/blkdev.h |   3 ++
 2 files changed, 144 insertions(+)

(limited to 'include/linux/blkdev.h')

diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index 9ce7570c91e1..d98babfe49df 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -203,6 +203,7 @@ EXPORT_SYMBOL_GPL(bdev_zone_is_seq);
 struct blk_report_zones_args {
 	report_zones_cb cb;
 	void		*data;
+	bool		report_active;
 };
 
 static int blkdev_do_report_zones(struct block_device *bdev, sector_t sector,
@@ -820,6 +821,23 @@ static void disk_zone_wplug_sync_wp_offset(struct gendisk *disk,
 int disk_report_zone(struct gendisk *disk, struct blk_zone *zone,
 		     unsigned int idx, struct blk_report_zones_args *args)
 {
+	if (args->report_active) {
+		/*
+		 * If we come here, then this is a report zones as a fallback
+		 * for a cached report. So collapse the implicit open, explicit
+		 * open and closed conditions into the active zone condition.
+		 */
+		switch (zone->cond) {
+		case BLK_ZONE_COND_IMP_OPEN:
+		case BLK_ZONE_COND_EXP_OPEN:
+		case BLK_ZONE_COND_CLOSED:
+			zone->cond = BLK_ZONE_COND_ACTIVE;
+			break;
+		default:
+			break;
+		}
+	}
+
 	if (disk->zone_wplugs_hash)
 		disk_zone_wplug_sync_wp_offset(disk, zone);
 
@@ -830,6 +848,129 @@ int disk_report_zone(struct gendisk *disk, struct blk_zone *zone,
 }
 EXPORT_SYMBOL_GPL(disk_report_zone);
 
+static int blkdev_report_zone_cb(struct blk_zone *zone, unsigned int idx,
+				 void *data)
+{
+	memcpy(data, zone, sizeof(struct blk_zone));
+	return 0;
+}
+
+static int blkdev_report_zone_fallback(struct block_device *bdev,
+				       sector_t sector, struct blk_zone *zone)
+{
+	struct blk_report_zones_args args = {
+		.cb = blkdev_report_zone_cb,
+		.data = zone,
+		.report_active = true,
+	};
+
+	return blkdev_do_report_zones(bdev, sector, 1, &args);
+}
+
+/**
+ * blkdev_get_zone_info - Get a single zone information from cached data
+ * @bdev:   Target block device
+ * @sector: Sector contained by the target zone
+ * @zone:   zone structure to return the zone information
+ *
+ * Description:
+ *    Get the zone information for the zone containing @sector using the zone
+ *    write plug of the target zone, if one exist, or the disk zone condition
+ *    array otherwise. The zone condition may be reported as being
+ *    the BLK_ZONE_COND_ACTIVE condition for a zone that is in the implicit
+ *    open, explicit open or closed condition.
+ *
+ *    Returns 0 on success and a negative error code on failure.
+ */
+int blkdev_get_zone_info(struct block_device *bdev, sector_t sector,
+			 struct blk_zone *zone)
+{
+	struct gendisk *disk = bdev->bd_disk;
+	sector_t zone_sectors = bdev_zone_sectors(bdev);
+	struct blk_zone_wplug *zwplug;
+	unsigned long flags;
+	u8 *zones_cond;
+
+	if (!bdev_is_zoned(bdev))
+		return -EOPNOTSUPP;
+
+	if (sector >= get_capacity(disk))
+		return -EINVAL;
+
+	memset(zone, 0, sizeof(*zone));
+	sector = ALIGN_DOWN(sector, zone_sectors);
+
+	rcu_read_lock();
+	zones_cond = rcu_dereference(disk->zones_cond);
+	if (!disk->zone_wplugs_hash || !zones_cond) {
+		rcu_read_unlock();
+		return blkdev_report_zone_fallback(bdev, sector, zone);
+	}
+	zone->cond = zones_cond[disk_zone_no(disk, sector)];
+	rcu_read_unlock();
+
+	zone->start = sector;
+	zone->len = zone_sectors;
+
+	/*
+	 * If this is a conventional zone, we do not have a zone write plug and
+	 * can report the zone immediately.
+	 */
+	if (zone->cond == BLK_ZONE_COND_NOT_WP) {
+		zone->type = BLK_ZONE_TYPE_CONVENTIONAL;
+		zone->capacity = zone_sectors;
+		zone->wp = ULLONG_MAX;
+		return 0;
+	}
+
+	/*
+	 * This is a sequential write required zone. If the zone is read-only or
+	 * offline, only set the zone write pointer to an invalid value and
+	 * report the zone.
+	 */
+	zone->type = BLK_ZONE_TYPE_SEQWRITE_REQ;
+	if (disk_zone_is_last(disk, zone))
+		zone->capacity = disk->last_zone_capacity;
+	else
+		zone->capacity = disk->zone_capacity;
+
+	if (zone->cond == BLK_ZONE_COND_READONLY ||
+	    zone->cond == BLK_ZONE_COND_OFFLINE) {
+		zone->wp = ULLONG_MAX;
+		return 0;
+	}
+
+	/*
+	 * If the zone does not have a zone write plug, it is either full or
+	 * empty, as we otherwise would have a zone write plug for it. In this
+	 * case, set the write pointer accordingly and report the zone.
+	 * Otherwise, if we have a zone write plug, use it.
+	 */
+	zwplug = disk_get_zone_wplug(disk, sector);
+	if (!zwplug) {
+		if (zone->cond == BLK_ZONE_COND_FULL)
+			zone->wp = ULLONG_MAX;
+		else
+			zone->wp = sector;
+		return 0;
+	}
+
+	spin_lock_irqsave(&zwplug->lock, flags);
+	if (zwplug->flags & BLK_ZONE_WPLUG_NEED_WP_UPDATE) {
+		spin_unlock_irqrestore(&zwplug->lock, flags);
+		disk_put_zone_wplug(zwplug);
+		return blkdev_report_zone_fallback(bdev, sector, zone);
+	}
+	zone->cond = zwplug->cond;
+	zone->wp = sector + zwplug->wp_offset;
+	spin_unlock_irqrestore(&zwplug->lock, flags);
+
+	disk_put_zone_wplug(zwplug);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(blkdev_get_zone_info);
+
 static void blk_zone_reset_bio_endio(struct bio *bio)
 {
 	struct gendisk *disk = bio->bi_bdev->bd_disk;
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 53bcfbc2f68f..03a594b4dfbc 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -436,6 +436,9 @@ typedef int (*report_zones_cb)(struct blk_zone *zone, unsigned int idx,
 int disk_report_zone(struct gendisk *disk, struct blk_zone *zone,
 		     unsigned int idx, struct blk_report_zones_args *args);
 
+int blkdev_get_zone_info(struct block_device *bdev, sector_t sector,
+			 struct blk_zone *zone);
+
 #define BLK_ALL_ZONES  ((unsigned int)-1)
 int blkdev_report_zones(struct block_device *bdev, sector_t sector,
 		unsigned int nr_zones, report_zones_cb cb, void *data);
-- 
cgit v1.2.3


From 31f0656a4ab712edf2888eabcc0664197a4a938e Mon Sep 17 00:00:00 2001
From: Damien Le Moal <dlemoal@kernel.org>
Date: Wed, 5 Nov 2025 06:22:44 +0900
Subject: block: introduce blkdev_report_zones_cached()

Introduce the function blkdev_report_zones_cached() to provide a fast
report zone built using the blkdev_get_zone_info() function, which gets
zone information from a disk zones_cond array or zone write plugs.
For a large capacity SMR drive, such fast report zone can be completed
in a few milliseconds compared to several seconds completion times
when the report zone is obtained from the device.

The zone report is built in the same manner as with the regular
blkdev_report_zones() function, that is, the first zone reported is the
one containing the specified start sector and the report is limited to
the specified number of zones (nr_zones argument). The information for
each zone in the report is obtained using blkdev_get_zone_info().

For zoned devices that do not use zone write plug resources,
using blkdev_get_zone_info() is inefficient as the zone report would
be very slow, generated one zone at a time. To avoid this,
blkdev_report_zones_cached() falls back to calling
blkdev_do_report_zones() to execute a regular zone report. In this case,
the .report_active field of struct blk_report_zones_args is set to true
to report zone conditions using the BLK_ZONE_COND_ACTIVE condition in
place of the implicit open, explicit open and closed conditions.

Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 block/blk-zoned.c      | 88 ++++++++++++++++++++++++++++++++++++++++++--------
 include/linux/blkdev.h |  2 ++
 2 files changed, 77 insertions(+), 13 deletions(-)

(limited to 'include/linux/blkdev.h')

diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index d98babfe49df..bbd105b11843 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -74,6 +74,19 @@ struct blk_zone_wplug {
 	enum blk_zone_cond	cond;
 };
 
+static inline bool disk_need_zone_resources(struct gendisk *disk)
+{
+	/*
+	 * All request-based zoned devices need zone resources so that the
+	 * block layer can automatically handle write BIO plugging. BIO-based
+	 * device drivers (e.g. DM devices) are normally responsible for
+	 * handling zone write ordering and do not need zone resources, unless
+	 * the driver requires zone append emulation.
+	 */
+	return queue_is_mq(disk->queue) ||
+		queue_emulates_zone_append(disk->queue);
+}
+
 static inline unsigned int disk_zone_wplugs_hash_size(struct gendisk *disk)
 {
 	return 1U << disk->zone_wplugs_hash_bits;
@@ -971,6 +984,68 @@ int blkdev_get_zone_info(struct block_device *bdev, sector_t sector,
 }
 EXPORT_SYMBOL_GPL(blkdev_get_zone_info);
 
+/**
+ * blkdev_report_zones_cached - Get cached zones information
+ * @bdev:     Target block device
+ * @sector:   Sector from which to report zones
+ * @nr_zones: Maximum number of zones to report
+ * @cb:       Callback function called for each reported zone
+ * @data:     Private data for the callback function
+ *
+ * Description:
+ *    Similar to blkdev_report_zones() but instead of calling into the low level
+ *    device driver to get the zone report from the device, use
+ *    blkdev_get_zone_info() to generate the report from the disk zone write
+ *    plugs and zones condition array. Since calling this function without a
+ *    callback does not make sense, @cb must be specified.
+ */
+int blkdev_report_zones_cached(struct block_device *bdev, sector_t sector,
+			unsigned int nr_zones, report_zones_cb cb, void *data)
+{
+	struct gendisk *disk = bdev->bd_disk;
+	sector_t capacity = get_capacity(disk);
+	sector_t zone_sectors = bdev_zone_sectors(bdev);
+	unsigned int idx = 0;
+	struct blk_zone zone;
+	int ret;
+
+	if (!cb || !bdev_is_zoned(bdev) ||
+	    WARN_ON_ONCE(!disk->fops->report_zones))
+		return -EOPNOTSUPP;
+
+	if (!nr_zones || sector >= capacity)
+		return 0;
+
+	/*
+	 * If we do not have any zone write plug resources, fallback to using
+	 * the regular zone report.
+	 */
+	if (!disk_need_zone_resources(disk)) {
+		struct blk_report_zones_args args = {
+			.cb = cb,
+			.data = data,
+			.report_active = true,
+		};
+
+		return blkdev_do_report_zones(bdev, sector, nr_zones, &args);
+	}
+
+	for (sector = ALIGN_DOWN(sector, zone_sectors);
+	     sector < capacity && idx < nr_zones;
+	     sector += zone_sectors, idx++) {
+		ret = blkdev_get_zone_info(bdev, sector, &zone);
+		if (ret)
+			return ret;
+
+		ret = cb(&zone, idx, data);
+		if (ret)
+			return ret;
+	}
+
+	return idx;
+}
+EXPORT_SYMBOL_GPL(blkdev_report_zones_cached);
+
 static void blk_zone_reset_bio_endio(struct bio *bio)
 {
 	struct gendisk *disk = bio->bi_bdev->bd_disk;
@@ -1781,19 +1856,6 @@ void disk_free_zone_resources(struct gendisk *disk)
 	disk->nr_zones = 0;
 }
 
-static inline bool disk_need_zone_resources(struct gendisk *disk)
-{
-	/*
-	 * All mq zoned devices need zone resources so that the block layer
-	 * can automatically handle write BIO plugging. BIO-based device drivers
-	 * (e.g. DM devices) are normally responsible for handling zone write
-	 * ordering and do not need zone resources, unless the driver requires
-	 * zone append emulation.
-	 */
-	return queue_is_mq(disk->queue) ||
-		queue_emulates_zone_append(disk->queue);
-}
-
 struct blk_revalidate_zone_args {
 	struct gendisk	*disk;
 	u8		*zones_cond;
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 03a594b4dfbc..f0ab02e0a673 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -442,6 +442,8 @@ int blkdev_get_zone_info(struct block_device *bdev, sector_t sector,
 #define BLK_ALL_ZONES  ((unsigned int)-1)
 int blkdev_report_zones(struct block_device *bdev, sector_t sector,
 		unsigned int nr_zones, report_zones_cb cb, void *data);
+int blkdev_report_zones_cached(struct block_device *bdev, sector_t sector,
+		unsigned int nr_zones, report_zones_cb cb, void *data);
 int blkdev_zone_mgmt(struct block_device *bdev, enum req_op op,
 		sector_t sectors, sector_t nr_sectors);
 int blk_revalidate_disk_zones(struct gendisk *disk);
-- 
cgit v1.2.3


From 15638d52cbcf6e969f4a5e2757b118355db583f3 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Wed, 5 Nov 2025 14:52:15 -0500
Subject: block: fix cached zone reporting after zone append was used

No zone plugs are allocated when a zone is opened by calling Zone Append
on it.  This makes the cached zone reporting report incorrectly empty
zones if the file system is unmounted and report zones is called after
that, e.g. by xfstests test cases using the scratch device.

Fix this by recording if zone append was used on a device, and disable
cached reporting for the device until a ZONE_RESET_ALL happens that
guarantees all zones are empty.

We could probably do even better using a per-zone flag, but the practical
use cache for zone reporting after the initial mount are rather limited,
so let's keep things simple for now.

Fixes: 31f0656a4ab7 ("block: introduce blkdev_report_zones_cached()")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 block/blk-zoned.c      | 26 +++++++++++++++++++++-----
 include/linux/blkdev.h |  1 +
 2 files changed, 22 insertions(+), 5 deletions(-)

(limited to 'include/linux/blkdev.h')

diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index a0ce17e2143f..c5226bcaaa94 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -899,6 +899,19 @@ static int blkdev_report_zone_fallback(struct block_device *bdev,
 	return blkdev_do_report_zones(bdev, sector, 1, &args);
 }
 
+/*
+ * For devices that natively support zone append operations, we do not use zone
+ * write plugging for zone append writes, which makes the zone condition
+ * tracking invalid once zone append was used.  In that case fall back to a
+ * regular report zones to get correct information.
+ */
+static inline bool blkdev_has_cached_report_zones(struct block_device *bdev)
+{
+	return disk_need_zone_resources(bdev->bd_disk) &&
+		(bdev_emulates_zone_append(bdev) ||
+		 !test_bit(GD_ZONE_APPEND_USED, &bdev->bd_disk->state));
+}
+
 /**
  * blkdev_get_zone_info - Get a single zone information from cached data
  * @bdev:   Target block device
@@ -932,6 +945,9 @@ int blkdev_get_zone_info(struct block_device *bdev, sector_t sector,
 	memset(zone, 0, sizeof(*zone));
 	sector = ALIGN_DOWN(sector, zone_sectors);
 
+	if (!blkdev_has_cached_report_zones(bdev))
+		return blkdev_report_zone_fallback(bdev, sector, zone);
+
 	rcu_read_lock();
 	zones_cond = rcu_dereference(disk->zones_cond);
 	if (!disk->zone_wplugs_hash || !zones_cond) {
@@ -1035,11 +1051,7 @@ int blkdev_report_zones_cached(struct block_device *bdev, sector_t sector,
 	if (!nr_zones || sector >= capacity)
 		return 0;
 
-	/*
-	 * If we do not have any zone write plug resources, fallback to using
-	 * the regular zone report.
-	 */
-	if (!disk_need_zone_resources(disk)) {
+	if (!blkdev_has_cached_report_zones(bdev)) {
 		struct blk_report_zones_args args = {
 			.cb = cb,
 			.data = data,
@@ -1115,6 +1127,7 @@ static void blk_zone_reset_all_bio_endio(struct bio *bio)
 	for (sector = 0; sector < capacity;
 	     sector += bdev_zone_sectors(bio->bi_bdev))
 		disk_zone_set_cond(disk, sector, BLK_ZONE_COND_EMPTY);
+	clear_bit(GD_ZONE_APPEND_USED, &disk->state);
 }
 
 static void blk_zone_finish_bio_endio(struct bio *bio)
@@ -1474,6 +1487,9 @@ static void blk_zone_wplug_handle_native_zone_append(struct bio *bio)
 	struct blk_zone_wplug *zwplug;
 	unsigned long flags;
 
+	if (!test_bit(GD_ZONE_APPEND_USED, &disk->state))
+		set_bit(GD_ZONE_APPEND_USED, &disk->state);
+
 	/*
 	 * We have native support for zone append operations, so we are not
 	 * going to handle @bio through plugging. However, we may already have a
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index f0ab02e0a673..6a498aa7f7e7 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -173,6 +173,7 @@ struct gendisk {
 #define GD_ADDED			4
 #define GD_SUPPRESS_PART_SCAN		5
 #define GD_OWNS_QUEUE			6
+#define GD_ZONE_APPEND_USED		7
 
 	struct mutex open_mutex;	/* open/close mutex */
 	unsigned open_partitions;	/* number of open partitions */
-- 
cgit v1.2.3


From 25976c314f6596254c9b1e2291d94393b7d5ae81 Mon Sep 17 00:00:00 2001
From: Damien Le Moal <dlemoal@kernel.org>
Date: Fri, 7 Nov 2025 15:38:44 +0900
Subject: block: introduce bdev_zone_start()

Introduce the function bdev_zone_start() as a more explicit (and clear)
replacement for ALIGN_DOWN() to get the start sector of a zone
containing a particular sector of a zoned block device.

Use this new helper in blkdev_get_zone_info() and
blkdev_report_zones_cached().

Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 block/blk-zoned.c      | 4 ++--
 include/linux/blkdev.h | 6 ++++++
 2 files changed, 8 insertions(+), 2 deletions(-)

(limited to 'include/linux/blkdev.h')

diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index b580d59ce210..3791755bc6ad 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -950,7 +950,7 @@ int blkdev_get_zone_info(struct block_device *bdev, sector_t sector,
 		return -EINVAL;
 
 	memset(zone, 0, sizeof(*zone));
-	sector = ALIGN_DOWN(sector, zone_sectors);
+	sector = bdev_zone_start(bdev, sector);
 
 	if (!blkdev_has_cached_report_zones(bdev))
 		return blkdev_report_zone_fallback(bdev, sector, zone);
@@ -1068,7 +1068,7 @@ int blkdev_report_zones_cached(struct block_device *bdev, sector_t sector,
 		return blkdev_do_report_zones(bdev, sector, nr_zones, &args);
 	}
 
-	for (sector = ALIGN_DOWN(sector, zone_sectors);
+	for (sector = bdev_zone_start(bdev, sector);
 	     sector < capacity && idx < nr_zones;
 	     sector += zone_sectors, idx++) {
 		ret = blkdev_get_zone_info(bdev, sector, &zone);
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 6a498aa7f7e7..2fff8a80dbd2 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1522,6 +1522,12 @@ static inline sector_t bdev_zone_sectors(struct block_device *bdev)
 	return q->limits.chunk_sectors;
 }
 
+static inline sector_t bdev_zone_start(struct block_device *bdev,
+				       sector_t sector)
+{
+	return sector & ~(bdev_zone_sectors(bdev) - 1);
+}
+
 static inline sector_t bdev_offset_from_zone_start(struct block_device *bdev,
 						   sector_t sector)
 {
-- 
cgit v1.2.3


From 935a20d1bebf6236076785fac3ff81e3931834e9 Mon Sep 17 00:00:00 2001
From: Bart Van Assche <bvanassche@acm.org>
Date: Fri, 14 Nov 2025 13:04:07 -0800
Subject: block: Remove queue freezing from several sysfs store callbacks

Freezing the request queue from inside sysfs store callbacks may cause a
deadlock in combination with the dm-multipath driver and the
queue_if_no_path option. Additionally, freezing the request queue slows
down system boot on systems where sysfs attributes are set synchronously.

Fix this by removing the blk_mq_freeze_queue() / blk_mq_unfreeze_queue()
calls from the store callbacks that do not strictly need these callbacks.
Add the __data_racy annotation to request_queue.rq_timeout to suppress
KCSAN data race reports about the rq_timeout reads.

This patch may cause a small delay in applying the new settings.

For all the attributes affected by this patch, I/O will complete
correctly whether the old or the new value of the attribute is used.

This patch affects the following sysfs attributes:
* io_poll_delay
* io_timeout
* nomerges
* read_ahead_kb
* rq_affinity

Here is an example of a deadlock triggered by running test srp/002
if this patch is not applied:

task:multipathd
Call Trace:
 <TASK>
 __schedule+0x8c1/0x1bf0
 schedule+0xdd/0x270
 schedule_preempt_disabled+0x1c/0x30
 __mutex_lock+0xb89/0x1650
 mutex_lock_nested+0x1f/0x30
 dm_table_set_restrictions+0x823/0xdf0
 __bind+0x166/0x590
 dm_swap_table+0x2a7/0x490
 do_resume+0x1b1/0x610
 dev_suspend+0x55/0x1a0
 ctl_ioctl+0x3a5/0x7e0
 dm_ctl_ioctl+0x12/0x20
 __x64_sys_ioctl+0x127/0x1a0
 x64_sys_call+0xe2b/0x17d0
 do_syscall_64+0x96/0x3a0
 entry_SYSCALL_64_after_hwframe+0x4b/0x53
 </TASK>
task:(udev-worker)
Call Trace:
 <TASK>
 __schedule+0x8c1/0x1bf0
 schedule+0xdd/0x270
 blk_mq_freeze_queue_wait+0xf2/0x140
 blk_mq_freeze_queue_nomemsave+0x23/0x30
 queue_ra_store+0x14e/0x290
 queue_attr_store+0x23e/0x2c0
 sysfs_kf_write+0xde/0x140
 kernfs_fop_write_iter+0x3b2/0x630
 vfs_write+0x4fd/0x1390
 ksys_write+0xfd/0x230
 __x64_sys_write+0x76/0xc0
 x64_sys_call+0x276/0x17d0
 do_syscall_64+0x96/0x3a0
 entry_SYSCALL_64_after_hwframe+0x4b/0x53
 </TASK>

Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Nilay Shroff <nilay@linux.ibm.com>
Cc: Martin Wilck <mwilck@suse.com>
Cc: Benjamin Marzinski <bmarzins@redhat.com>
Cc: stable@vger.kernel.org
Fixes: af2814149883 ("block: freeze the queue in queue_attr_store")
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Nilay Shroff <nilay@linux.ibm.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 block/blk-sysfs.c      | 26 ++++++++------------------
 include/linux/blkdev.h |  2 +-
 2 files changed, 9 insertions(+), 19 deletions(-)

(limited to 'include/linux/blkdev.h')

diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c
index 76c47fe9b8d6..8684c57498cc 100644
--- a/block/blk-sysfs.c
+++ b/block/blk-sysfs.c
@@ -143,21 +143,22 @@ queue_ra_store(struct gendisk *disk, const char *page, size_t count)
 {
 	unsigned long ra_kb;
 	ssize_t ret;
-	unsigned int memflags;
 	struct request_queue *q = disk->queue;
 
 	ret = queue_var_store(&ra_kb, page, count);
 	if (ret < 0)
 		return ret;
 	/*
-	 * ->ra_pages is protected by ->limits_lock because it is usually
-	 * calculated from the queue limits by queue_limits_commit_update.
+	 * The ->ra_pages change below is protected by ->limits_lock because it
+	 * is usually calculated from the queue limits by
+	 * queue_limits_commit_update().
+	 *
+	 * bdi->ra_pages reads are not serialized against bdi->ra_pages writes.
+	 * Use WRITE_ONCE() to write bdi->ra_pages once.
 	 */
 	mutex_lock(&q->limits_lock);
-	memflags = blk_mq_freeze_queue(q);
-	disk->bdi->ra_pages = ra_kb >> (PAGE_SHIFT - 10);
+	WRITE_ONCE(disk->bdi->ra_pages, ra_kb >> (PAGE_SHIFT - 10));
 	mutex_unlock(&q->limits_lock);
-	blk_mq_unfreeze_queue(q, memflags);
 
 	return ret;
 }
@@ -375,21 +376,18 @@ static ssize_t queue_nomerges_store(struct gendisk *disk, const char *page,
 				    size_t count)
 {
 	unsigned long nm;
-	unsigned int memflags;
 	struct request_queue *q = disk->queue;
 	ssize_t ret = queue_var_store(&nm, page, count);
 
 	if (ret < 0)
 		return ret;
 
-	memflags = blk_mq_freeze_queue(q);
 	blk_queue_flag_clear(QUEUE_FLAG_NOMERGES, q);
 	blk_queue_flag_clear(QUEUE_FLAG_NOXMERGES, q);
 	if (nm == 2)
 		blk_queue_flag_set(QUEUE_FLAG_NOMERGES, q);
 	else if (nm)
 		blk_queue_flag_set(QUEUE_FLAG_NOXMERGES, q);
-	blk_mq_unfreeze_queue(q, memflags);
 
 	return ret;
 }
@@ -409,7 +407,6 @@ queue_rq_affinity_store(struct gendisk *disk, const char *page, size_t count)
 #ifdef CONFIG_SMP
 	struct request_queue *q = disk->queue;
 	unsigned long val;
-	unsigned int memflags;
 
 	ret = queue_var_store(&val, page, count);
 	if (ret < 0)
@@ -421,7 +418,6 @@ queue_rq_affinity_store(struct gendisk *disk, const char *page, size_t count)
 	 * are accessed individually using atomic test_bit operation. So we
 	 * don't grab any lock while updating these flags.
 	 */
-	memflags = blk_mq_freeze_queue(q);
 	if (val == 2) {
 		blk_queue_flag_set(QUEUE_FLAG_SAME_COMP, q);
 		blk_queue_flag_set(QUEUE_FLAG_SAME_FORCE, q);
@@ -432,7 +428,6 @@ queue_rq_affinity_store(struct gendisk *disk, const char *page, size_t count)
 		blk_queue_flag_clear(QUEUE_FLAG_SAME_COMP, q);
 		blk_queue_flag_clear(QUEUE_FLAG_SAME_FORCE, q);
 	}
-	blk_mq_unfreeze_queue(q, memflags);
 #endif
 	return ret;
 }
@@ -446,11 +441,9 @@ static ssize_t queue_poll_delay_store(struct gendisk *disk, const char *page,
 static ssize_t queue_poll_store(struct gendisk *disk, const char *page,
 				size_t count)
 {
-	unsigned int memflags;
 	ssize_t ret = count;
 	struct request_queue *q = disk->queue;
 
-	memflags = blk_mq_freeze_queue(q);
 	if (!(q->limits.features & BLK_FEAT_POLL)) {
 		ret = -EINVAL;
 		goto out;
@@ -459,7 +452,6 @@ static ssize_t queue_poll_store(struct gendisk *disk, const char *page,
 	pr_info_ratelimited("writes to the poll attribute are ignored.\n");
 	pr_info_ratelimited("please use driver specific parameters instead.\n");
 out:
-	blk_mq_unfreeze_queue(q, memflags);
 	return ret;
 }
 
@@ -472,7 +464,7 @@ static ssize_t queue_io_timeout_show(struct gendisk *disk, char *page)
 static ssize_t queue_io_timeout_store(struct gendisk *disk, const char *page,
 				  size_t count)
 {
-	unsigned int val, memflags;
+	unsigned int val;
 	int err;
 	struct request_queue *q = disk->queue;
 
@@ -480,9 +472,7 @@ static ssize_t queue_io_timeout_store(struct gendisk *disk, const char *page,
 	if (err || val == 0)
 		return -EINVAL;
 
-	memflags = blk_mq_freeze_queue(q);
 	blk_queue_rq_timeout(q, msecs_to_jiffies(val));
-	blk_mq_unfreeze_queue(q, memflags);
 
 	return count;
 }
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 2fff8a80dbd2..cb4ba09959ee 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -495,7 +495,7 @@ struct request_queue {
 	 */
 	unsigned long		queue_flags;
 
-	unsigned int		rq_timeout;
+	unsigned int __data_racy rq_timeout;
 
 	unsigned int		queue_depth;
 
-- 
cgit v1.2.3


From d0c98769ee7d5db8d699a270690639cde1766cd4 Mon Sep 17 00:00:00 2001
From: Fengnan Chang <fengnanchang@gmail.com>
Date: Fri, 28 Nov 2025 16:53:13 +0800
Subject: blk-mq: use array manage hctx map instead of xarray

After commit 4e5cc99e1e48 ("blk-mq: manage hctx map via xarray"), we use
an xarray instead of array to store hctx, but in poll mode, each time
in blk_mq_poll, we need use xa_load to find corresponding hctx, this
introduce some costs. In my test, xa_load may cost 3.8% cpu.

This patch revert previous change, eliminates the overhead of xa_load
and can result in a 3% performance improvement.

Signed-off-by: Fengnan Chang <changfengnan@bytedance.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 block/blk-mq-tag.c     |  2 +-
 block/blk-mq.c         | 58 ++++++++++++++++++++++++++++++++------------------
 block/blk-mq.h         |  2 +-
 include/linux/blk-mq.h |  3 ++-
 include/linux/blkdev.h |  2 +-
 5 files changed, 42 insertions(+), 25 deletions(-)

(limited to 'include/linux/blkdev.h')

diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c
index 5b664dbdf655..33946cdb5716 100644
--- a/block/blk-mq-tag.c
+++ b/block/blk-mq-tag.c
@@ -499,7 +499,7 @@ void blk_mq_queue_tag_busy_iter(struct request_queue *q, busy_tag_iter_fn *fn,
 	int srcu_idx;
 
 	/*
-	 * __blk_mq_update_nr_hw_queues() updates nr_hw_queues and hctx_table
+	 * __blk_mq_update_nr_hw_queues() updates nr_hw_queues and queue_hw_ctx
 	 * while the queue is frozen. So we can use q_usage_counter to avoid
 	 * racing with it.
 	 */
diff --git a/block/blk-mq.c b/block/blk-mq.c
index f2650c97a75e..1ef81110eb8a 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -730,7 +730,7 @@ struct request *blk_mq_alloc_request_hctx(struct request_queue *q,
 	 * If not tell the caller that it should skip this queue.
 	 */
 	ret = -EXDEV;
-	data.hctx = xa_load(&q->hctx_table, hctx_idx);
+	data.hctx = q->queue_hw_ctx[hctx_idx];
 	if (!blk_mq_hw_queue_mapped(data.hctx))
 		goto out_queue_exit;
 	cpu = cpumask_first_and(data.hctx->cpumask, cpu_online_mask);
@@ -3946,8 +3946,6 @@ static void blk_mq_exit_hctx(struct request_queue *q,
 			blk_free_flush_queue_callback);
 	hctx->fq = NULL;
 
-	xa_erase(&q->hctx_table, hctx_idx);
-
 	spin_lock(&q->unused_hctx_lock);
 	list_add(&hctx->hctx_list, &q->unused_hctx_list);
 	spin_unlock(&q->unused_hctx_lock);
@@ -3989,14 +3987,8 @@ static int blk_mq_init_hctx(struct request_queue *q,
 				hctx->numa_node))
 		goto exit_hctx;
 
-	if (xa_insert(&q->hctx_table, hctx_idx, hctx, GFP_KERNEL))
-		goto exit_flush_rq;
-
 	return 0;
 
- exit_flush_rq:
-	if (set->ops->exit_request)
-		set->ops->exit_request(set, hctx->fq->flush_rq, hctx_idx);
  exit_hctx:
 	if (set->ops->exit_hctx)
 		set->ops->exit_hctx(hctx, hctx_idx);
@@ -4385,7 +4377,7 @@ void blk_mq_release(struct request_queue *q)
 		kobject_put(&hctx->kobj);
 	}
 
-	xa_destroy(&q->hctx_table);
+	kfree(q->queue_hw_ctx);
 
 	/*
 	 * release .mq_kobj and sw queue's kobject now because
@@ -4529,26 +4521,44 @@ static struct blk_mq_hw_ctx *blk_mq_alloc_and_init_hctx(
 static void __blk_mq_realloc_hw_ctxs(struct blk_mq_tag_set *set,
 				     struct request_queue *q)
 {
-	struct blk_mq_hw_ctx *hctx;
-	unsigned long i, j;
+	int i, j, end;
+	struct blk_mq_hw_ctx **hctxs = q->queue_hw_ctx;
+
+	if (q->nr_hw_queues < set->nr_hw_queues) {
+		struct blk_mq_hw_ctx **new_hctxs;
+
+		new_hctxs = kcalloc_node(set->nr_hw_queues,
+				       sizeof(*new_hctxs), GFP_KERNEL,
+				       set->numa_node);
+		if (!new_hctxs)
+			return;
+		if (hctxs)
+			memcpy(new_hctxs, hctxs, q->nr_hw_queues *
+			       sizeof(*hctxs));
+		q->queue_hw_ctx = new_hctxs;
+		kfree(hctxs);
+		hctxs = new_hctxs;
+	}
 
 	for (i = 0; i < set->nr_hw_queues; i++) {
 		int old_node;
 		int node = blk_mq_get_hctx_node(set, i);
-		struct blk_mq_hw_ctx *old_hctx = xa_load(&q->hctx_table, i);
+		struct blk_mq_hw_ctx *old_hctx = hctxs[i];
 
 		if (old_hctx) {
 			old_node = old_hctx->numa_node;
 			blk_mq_exit_hctx(q, set, old_hctx, i);
 		}
 
-		if (!blk_mq_alloc_and_init_hctx(set, q, i, node)) {
+		hctxs[i] = blk_mq_alloc_and_init_hctx(set, q, i, node);
+		if (!hctxs[i]) {
 			if (!old_hctx)
 				break;
 			pr_warn("Allocate new hctx on node %d fails, fallback to previous one on node %d\n",
 					node, old_node);
-			hctx = blk_mq_alloc_and_init_hctx(set, q, i, old_node);
-			WARN_ON_ONCE(!hctx);
+			hctxs[i] = blk_mq_alloc_and_init_hctx(set, q, i,
+					old_node);
+			WARN_ON_ONCE(!hctxs[i]);
 		}
 	}
 	/*
@@ -4557,13 +4567,21 @@ static void __blk_mq_realloc_hw_ctxs(struct blk_mq_tag_set *set,
 	 */
 	if (i != set->nr_hw_queues) {
 		j = q->nr_hw_queues;
+		end = i;
 	} else {
 		j = i;
+		end = q->nr_hw_queues;
 		q->nr_hw_queues = set->nr_hw_queues;
 	}
 
-	xa_for_each_start(&q->hctx_table, j, hctx, j)
-		blk_mq_exit_hctx(q, set, hctx, j);
+	for (; j < end; j++) {
+		struct blk_mq_hw_ctx *hctx = hctxs[j];
+
+		if (hctx) {
+			blk_mq_exit_hctx(q, set, hctx, j);
+			hctxs[j] = NULL;
+		}
+	}
 }
 
 static void blk_mq_realloc_hw_ctxs(struct blk_mq_tag_set *set,
@@ -4599,8 +4617,6 @@ int blk_mq_init_allocated_queue(struct blk_mq_tag_set *set,
 	INIT_LIST_HEAD(&q->unused_hctx_list);
 	spin_lock_init(&q->unused_hctx_lock);
 
-	xa_init(&q->hctx_table);
-
 	blk_mq_realloc_hw_ctxs(set, q);
 	if (!q->nr_hw_queues)
 		goto err_hctxs;
@@ -5187,7 +5203,7 @@ int blk_mq_poll(struct request_queue *q, blk_qc_t cookie,
 {
 	if (!blk_mq_can_poll(q))
 		return 0;
-	return blk_hctx_poll(q, xa_load(&q->hctx_table, cookie), iob, flags);
+	return blk_hctx_poll(q, q->queue_hw_ctx[cookie], iob, flags);
 }
 
 int blk_rq_poll(struct request *rq, struct io_comp_batch *iob,
diff --git a/block/blk-mq.h b/block/blk-mq.h
index c4fccdeb5441..80a3f0c2bce7 100644
--- a/block/blk-mq.h
+++ b/block/blk-mq.h
@@ -84,7 +84,7 @@ static inline struct blk_mq_hw_ctx *blk_mq_map_queue_type(struct request_queue *
 							  enum hctx_type type,
 							  unsigned int cpu)
 {
-	return xa_load(&q->hctx_table, q->tag_set->map[type].mq_map[cpu]);
+	return q->queue_hw_ctx[q->tag_set->map[type].mq_map[cpu]];
 }
 
 static inline enum hctx_type blk_mq_get_hctx_type(blk_opf_t opf)
diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h
index b54506b3b76d..9208ff90ae16 100644
--- a/include/linux/blk-mq.h
+++ b/include/linux/blk-mq.h
@@ -1016,7 +1016,8 @@ static inline void *blk_mq_rq_to_pdu(struct request *rq)
 }
 
 #define queue_for_each_hw_ctx(q, hctx, i)				\
-	xa_for_each(&(q)->hctx_table, (i), (hctx))
+	for ((i) = 0; (i) < (q)->nr_hw_queues &&			\
+	     ({ hctx = (q)->queue_hw_ctx[i]; 1; }); (i)++)
 
 #define hctx_for_each_ctx(hctx, ctx, i)					\
 	for ((i) = 0; (i) < (hctx)->nr_ctx &&				\
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index cb4ba09959ee..6195f89648db 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -503,7 +503,7 @@ struct request_queue {
 
 	/* hw dispatch queues */
 	unsigned int		nr_hw_queues;
-	struct xarray		hctx_table;
+	struct blk_mq_hw_ctx	**queue_hw_ctx;
 
 	struct percpu_ref	q_usage_counter;
 	struct lock_class_key	io_lock_cls_key;
-- 
cgit v1.2.3


From 89e1fb7ceffd898505ad7fa57acec0585bfaa2cc Mon Sep 17 00:00:00 2001
From: Fengnan Chang <fengnanchang@gmail.com>
Date: Fri, 28 Nov 2025 16:53:14 +0800
Subject: blk-mq: fix potential uaf for 'queue_hw_ctx'

This is just apply Kuai's patch in [1] with mirror changes.

blk_mq_realloc_hw_ctxs() will free the 'queue_hw_ctx'(e.g. undate
submit_queues through configfs for null_blk), while it might still be
used from other context(e.g. switch elevator to none):

t1					t2
elevator_switch
 blk_mq_unquiesce_queue
  blk_mq_run_hw_queues
   queue_for_each_hw_ctx
    // assembly code for hctx = (q)->queue_hw_ctx[i]
    mov    0x48(%rbp),%rdx -> read old queue_hw_ctx

					__blk_mq_update_nr_hw_queues
					 blk_mq_realloc_hw_ctxs
					  hctxs = q->queue_hw_ctx
					  q->queue_hw_ctx = new_hctxs
					  kfree(hctxs)
    movslq %ebx,%rax
    mov    (%rdx,%rax,8),%rdi ->uaf

This problem was found by code review, and I comfirmed that the concurrent
scenario do exist(specifically 'q->queue_hw_ctx' can be changed during
blk_mq_run_hw_queues()), however, the uaf problem hasn't been repoduced yet
without hacking the kernel.

Sicne the queue is freezed in __blk_mq_update_nr_hw_queues(), fix the
problem by protecting 'queue_hw_ctx' through rcu where it can be accessed
without grabbing 'q_usage_counter'.

[1] https://lore.kernel.org/all/20220225072053.2472431-1-yukuai3@huawei.com/

Signed-off-by: Yu Kuai <yukuai3@huawei.com>
Signed-off-by: Fengnan Chang <changfengnan@bytedance.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 block/blk-mq.c         |  7 ++++++-
 include/linux/blk-mq.h | 13 ++++++++++++-
 include/linux/blkdev.h |  2 +-
 3 files changed, 19 insertions(+), 3 deletions(-)

(limited to 'include/linux/blkdev.h')

diff --git a/block/blk-mq.c b/block/blk-mq.c
index 1ef81110eb8a..4e96bb246247 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -4535,7 +4535,12 @@ static void __blk_mq_realloc_hw_ctxs(struct blk_mq_tag_set *set,
 		if (hctxs)
 			memcpy(new_hctxs, hctxs, q->nr_hw_queues *
 			       sizeof(*hctxs));
-		q->queue_hw_ctx = new_hctxs;
+		rcu_assign_pointer(q->queue_hw_ctx, new_hctxs);
+		/*
+		 * Make sure reading the old queue_hw_ctx from other
+		 * context concurrently won't trigger uaf.
+		 */
+		synchronize_rcu_expedited();
 		kfree(hctxs);
 		hctxs = new_hctxs;
 	}
diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h
index 9208ff90ae16..eb7254b3dddd 100644
--- a/include/linux/blk-mq.h
+++ b/include/linux/blk-mq.h
@@ -1015,9 +1015,20 @@ static inline void *blk_mq_rq_to_pdu(struct request *rq)
 	return rq + 1;
 }
 
+static inline struct blk_mq_hw_ctx *queue_hctx(struct request_queue *q, int id)
+{
+	struct blk_mq_hw_ctx *hctx;
+
+	rcu_read_lock();
+	hctx = rcu_dereference(q->queue_hw_ctx)[id];
+	rcu_read_unlock();
+
+	return hctx;
+}
+
 #define queue_for_each_hw_ctx(q, hctx, i)				\
 	for ((i) = 0; (i) < (q)->nr_hw_queues &&			\
-	     ({ hctx = (q)->queue_hw_ctx[i]; 1; }); (i)++)
+	     ({ hctx = queue_hctx((q), i); 1; }); (i)++)
 
 #define hctx_for_each_ctx(hctx, ctx, i)					\
 	for ((i) = 0; (i) < (hctx)->nr_ctx &&				\
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 6195f89648db..72e34acd439c 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -503,7 +503,7 @@ struct request_queue {
 
 	/* hw dispatch queues */
 	unsigned int		nr_hw_queues;
-	struct blk_mq_hw_ctx	**queue_hw_ctx;
+	struct blk_mq_hw_ctx * __rcu *queue_hw_ctx;
 
 	struct percpu_ref	q_usage_counter;
 	struct lock_class_key	io_lock_cls_key;
-- 
cgit v1.2.3