Merge tag 'for-6.11/block-20240722' of git://git.kernel.dk/linux

Pull more block updates from Jens Axboe:

 - MD fixes via Song:
     - md-cluster fixes (Heming Zhao)
     - raid1 fix (Mateusz Jończyk)

 - s390/dasd module description (Jeff)

 - Series cleaning up and hardening the blk-mq debugfs flag handling
   (John, Christoph)

 - blk-cgroup cleanup (Xiu)

 - Error polled IO attempts if backend doesn't support it (hexue)

 - Fix for an sbitmap hang (Yang)

* tag 'for-6.11/block-20240722' of git://git.kernel.dk/linux: (23 commits)
  blk-cgroup: move congestion_count to struct blkcg
  sbitmap: fix io hung due to race on sbitmap_word::cleared
  block: avoid polling configuration errors
  block: Catch possible entries missing from rqf_name[]
  block: Simplify definition of RQF_NAME()
  block: Use enum to define RQF_x bit indexes
  block: Catch possible entries missing from cmd_flag_name[]
  block: Catch possible entries missing from alloc_policy_name[]
  block: Catch possible entries missing from hctx_flag_name[]
  block: Catch possible entries missing from hctx_state_name[]
  block: Catch possible entries missing from blk_queue_flag_name[]
  block: Make QUEUE_FLAG_x as an enum
  block: Relocate BLK_MQ_MAX_DEPTH
  block: Relocate BLK_MQ_CPU_WORK_BATCH
  block: remove QUEUE_FLAG_STOPPED
  block: Add missing entry to hctx_flag_name[]
  block: Add zone write plugging entry to rqf_name[]
  block: Add missing entries from cmd_flag_name[]
  s390/dasd: fix error checks in dasd_copy_pair_store()
  s390/dasd: add missing MODULE_DESCRIPTION() macros
  ...

Author: torvalds

block/blk-cgroup.c

@@ -2182,12 +2182,13 @@ void blk_cgroup_bio_start(struct bio *bio)
 
 bool blk_cgroup_congested(void)
 {
-	struct cgroup_subsys_state *css;
+	struct blkcg *blkcg;
 	bool ret = false;
 
 	rcu_read_lock();
-	for (css = blkcg_css(); css; css = css->parent) {
-		if (atomic_read(&css->cgroup->congestion_count)) {
+	for (blkcg = css_to_blkcg(blkcg_css()); blkcg;
+	     blkcg = blkcg_parent(blkcg)) {
+		if (atomic_read(&blkcg->congestion_count)) {
 			ret = true;
 			break;
 		}

block/blk-cgroup.h

@@ -95,6 +95,8 @@ struct blkcg {
 	struct cgroup_subsys_state	css;
 	spinlock_t			lock;
 	refcount_t			online_pin;
+	/* If there is block congestion on this cgroup. */
+	atomic_t			congestion_count;
 
 	struct radix_tree_root		blkg_tree;
 	struct blkcg_gq	__rcu		*blkg_hint;
@@ -374,7 +376,7 @@ static inline void blkcg_use_delay(struct blkcg_gq *blkg)
 	if (WARN_ON_ONCE(atomic_read(&blkg->use_delay) < 0))
 		return;
 	if (atomic_add_return(1, &blkg->use_delay) == 1)
-		atomic_inc(&blkg->blkcg->css.cgroup->congestion_count);
+		atomic_inc(&blkg->blkcg->congestion_count);
 }
 
 static inline int blkcg_unuse_delay(struct blkcg_gq *blkg)
@@ -399,7 +401,7 @@ static inline int blkcg_unuse_delay(struct blkcg_gq *blkg)
 	if (old == 0)
 		return 0;
 	if (old == 1)
-		atomic_dec(&blkg->blkcg->css.cgroup->congestion_count);
+		atomic_dec(&blkg->blkcg->congestion_count);
 	return 1;
 }
 
@@ -418,7 +420,7 @@ static inline void blkcg_set_delay(struct blkcg_gq *blkg, u64 delay)
 
 	/* We only want 1 person setting the congestion count for this blkg. */
 	if (!old && atomic_try_cmpxchg(&blkg->use_delay, &old, -1))
-		atomic_inc(&blkg->blkcg->css.cgroup->congestion_count);
+		atomic_inc(&blkg->blkcg->congestion_count);
 
 	atomic64_set(&blkg->delay_nsec, delay);
 }
@@ -435,7 +437,7 @@ static inline void blkcg_clear_delay(struct blkcg_gq *blkg)
 
 	/* We only want 1 person clearing the congestion count for this blkg. */
 	if (old && atomic_try_cmpxchg(&blkg->use_delay, &old, 0))
-		atomic_dec(&blkg->blkcg->css.cgroup->congestion_count);
+		atomic_dec(&blkg->blkcg->congestion_count);
 }
 
 /**

block/blk-core.c

@@ -791,8 +791,11 @@ void submit_bio_noacct(struct bio *bio)
 		}
 	}
 
-	if (!(q->limits.features & BLK_FEAT_POLL))
+	if (!(q->limits.features & BLK_FEAT_POLL) &&
+			(bio->bi_opf & REQ_POLLED)) {
 		bio_clear_polled(bio);
+		goto not_supported;
+	}
 
 	switch (bio_op(bio)) {
 	case REQ_OP_READ:

block/blk-mq-debugfs.c

@@ -5,6 +5,7 @@
 
 #include <linux/kernel.h>
 #include <linux/blkdev.h>
+#include <linux/build_bug.h>
 #include <linux/debugfs.h>
 
 #include "blk.h"
@@ -79,7 +80,6 @@ static int queue_pm_only_show(void *data, struct seq_file *m)
 
 #define QUEUE_FLAG_NAME(name) [QUEUE_FLAG_##name] = #name
 static const char *const blk_queue_flag_name[] = {
-	QUEUE_FLAG_NAME(STOPPED),
 	QUEUE_FLAG_NAME(DYING),
 	QUEUE_FLAG_NAME(NOMERGES),
 	QUEUE_FLAG_NAME(SAME_COMP),
@@ -100,6 +100,7 @@ static int queue_state_show(void *data, struct seq_file *m)
 {
 	struct request_queue *q = data;
 
+	BUILD_BUG_ON(ARRAY_SIZE(blk_queue_flag_name) != QUEUE_FLAG_MAX);
 	blk_flags_show(m, q->queue_flags, blk_queue_flag_name,
 		       ARRAY_SIZE(blk_queue_flag_name));
 	seq_puts(m, "\n");
@@ -164,6 +165,7 @@ static int hctx_state_show(void *data, struct seq_file *m)
 {
 	struct blk_mq_hw_ctx *hctx = data;
 
+	BUILD_BUG_ON(ARRAY_SIZE(hctx_state_name) != BLK_MQ_S_MAX);
 	blk_flags_show(m, hctx->state, hctx_state_name,
 		       ARRAY_SIZE(hctx_state_name));
 	seq_puts(m, "\n");
@@ -181,10 +183,11 @@ static const char *const alloc_policy_name[] = {
 static const char *const hctx_flag_name[] = {
 	HCTX_FLAG_NAME(SHOULD_MERGE),
 	HCTX_FLAG_NAME(TAG_QUEUE_SHARED),
-	HCTX_FLAG_NAME(BLOCKING),
-	HCTX_FLAG_NAME(NO_SCHED),
 	HCTX_FLAG_NAME(STACKING),
 	HCTX_FLAG_NAME(TAG_HCTX_SHARED),
+	HCTX_FLAG_NAME(BLOCKING),
+	HCTX_FLAG_NAME(NO_SCHED),
+	HCTX_FLAG_NAME(NO_SCHED_BY_DEFAULT),
 };
 #undef HCTX_FLAG_NAME
 
@@ -193,6 +196,10 @@ static int hctx_flags_show(void *data, struct seq_file *m)
 	struct blk_mq_hw_ctx *hctx = data;
 	const int alloc_policy = BLK_MQ_FLAG_TO_ALLOC_POLICY(hctx->flags);
 
+	BUILD_BUG_ON(ARRAY_SIZE(hctx_flag_name) !=
+			BLK_MQ_F_ALLOC_POLICY_START_BIT);
+	BUILD_BUG_ON(ARRAY_SIZE(alloc_policy_name) != BLK_TAG_ALLOC_MAX);
+
 	seq_puts(m, "alloc_policy=");
 	if (alloc_policy < ARRAY_SIZE(alloc_policy_name) &&
 	    alloc_policy_name[alloc_policy])
@@ -223,12 +230,17 @@ static const char *const cmd_flag_name[] = {
 	CMD_FLAG_NAME(RAHEAD),
 	CMD_FLAG_NAME(BACKGROUND),
 	CMD_FLAG_NAME(NOWAIT),
-	CMD_FLAG_NAME(NOUNMAP),
 	CMD_FLAG_NAME(POLLED),
+	CMD_FLAG_NAME(ALLOC_CACHE),
+	CMD_FLAG_NAME(SWAP),
+	CMD_FLAG_NAME(DRV),
+	CMD_FLAG_NAME(FS_PRIVATE),
+	CMD_FLAG_NAME(ATOMIC),
+	CMD_FLAG_NAME(NOUNMAP),
 };
 #undef CMD_FLAG_NAME
 
-#define RQF_NAME(name) [ilog2((__force u32)RQF_##name)] = #name
+#define RQF_NAME(name) [__RQF_##name] = #name
 static const char *const rqf_name[] = {
 	RQF_NAME(STARTED),
 	RQF_NAME(FLUSH_SEQ),
@@ -243,6 +255,7 @@ static const char *const rqf_name[] = {
 	RQF_NAME(HASHED),
 	RQF_NAME(STATS),
 	RQF_NAME(SPECIAL_PAYLOAD),
+	RQF_NAME(ZONE_WRITE_PLUGGING),
 	RQF_NAME(TIMED_OUT),
 	RQF_NAME(RESV),
 };
@@ -268,6 +281,9 @@ int __blk_mq_debugfs_rq_show(struct seq_file *m, struct request *rq)
 	const enum req_op op = req_op(rq);
 	const char *op_str = blk_op_str(op);
 
+	BUILD_BUG_ON(ARRAY_SIZE(cmd_flag_name) != __REQ_NR_BITS);
+	BUILD_BUG_ON(ARRAY_SIZE(rqf_name) != __RQF_BITS);
+
 	seq_printf(m, "%p {.op=", rq);
 	if (strcmp(op_str, "UNKNOWN") == 0)
 		seq_printf(m, "%u", op);

block/blk-mq.h

@@ -36,6 +36,8 @@ enum {
 	BLK_MQ_TAG_MAX		= BLK_MQ_NO_TAG - 1,
 };
 
+#define BLK_MQ_CPU_WORK_BATCH	(8)
+
 typedef unsigned int __bitwise blk_insert_t;
 #define BLK_MQ_INSERT_AT_HEAD		((__force blk_insert_t)0x01)
 

drivers/md/md-cluster.c

@@ -15,6 +15,7 @@
 
 #define LVB_SIZE	64
 #define NEW_DEV_TIMEOUT 5000
+#define WAIT_DLM_LOCK_TIMEOUT (30 * HZ)
 
 struct dlm_lock_resource {
 	dlm_lockspace_t *ls;
@@ -56,6 +57,7 @@ struct resync_info {
 #define		MD_CLUSTER_ALREADY_IN_CLUSTER		6
 #define		MD_CLUSTER_PENDING_RECV_EVENT		7
 #define 	MD_CLUSTER_HOLDING_MUTEX_FOR_RECVD		8
+#define		MD_CLUSTER_WAITING_FOR_SYNC		9
 
 struct md_cluster_info {
 	struct mddev *mddev; /* the md device which md_cluster_info belongs to */
@@ -91,6 +93,7 @@ struct md_cluster_info {
 	sector_t sync_hi;
 };
 
+/* For compatibility, add the new msg_type at the end. */
 enum msg_type {
 	METADATA_UPDATED = 0,
 	RESYNCING,
@@ -100,6 +103,7 @@ enum msg_type {
 	BITMAP_NEEDS_SYNC,
 	CHANGE_CAPACITY,
 	BITMAP_RESIZE,
+	RESYNCING_START,
 };
 
 struct cluster_msg {
@@ -130,8 +134,13 @@ static int dlm_lock_sync(struct dlm_lock_resource *res, int mode)
 			0, sync_ast, res, res->bast);
 	if (ret)
 		return ret;
-	wait_event(res->sync_locking, res->sync_locking_done);
+	ret = wait_event_timeout(res->sync_locking, res->sync_locking_done,
+				WAIT_DLM_LOCK_TIMEOUT);
 	res->sync_locking_done = false;
+	if (!ret) {
+		pr_err("locking DLM '%s' timeout!\n", res->name);
+		return -EBUSY;
+	}
 	if (res->lksb.sb_status == 0)
 		res->mode = mode;
 	return res->lksb.sb_status;
@@ -455,6 +464,7 @@ static void process_suspend_info(struct mddev *mddev,
 		clear_bit(MD_RESYNCING_REMOTE, &mddev->recovery);
 		remove_suspend_info(mddev, slot);
 		set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
+		clear_bit(MD_CLUSTER_WAITING_FOR_SYNC, &cinfo->state);
 		md_wakeup_thread(mddev->thread);
 		return;
 	}
@@ -525,6 +535,7 @@ static int process_add_new_disk(struct mddev *mddev, struct cluster_msg *cmsg)
 		res = -1;
 	}
 	clear_bit(MD_CLUSTER_WAITING_FOR_NEWDISK, &cinfo->state);
+	set_bit(MD_CLUSTER_WAITING_FOR_SYNC, &cinfo->state);
 	return res;
 }
 
@@ -593,6 +604,9 @@ static int process_recvd_msg(struct mddev *mddev, struct cluster_msg *msg)
 	case CHANGE_CAPACITY:
 		set_capacity_and_notify(mddev->gendisk, mddev->array_sectors);
 		break;
+	case RESYNCING_START:
+		clear_bit(MD_CLUSTER_WAITING_FOR_SYNC, &mddev->cluster_info->state);
+		break;
 	case RESYNCING:
 		set_bit(MD_RESYNCING_REMOTE, &mddev->recovery);
 		process_suspend_info(mddev, le32_to_cpu(msg->slot),
@@ -743,15 +757,15 @@ static void unlock_comm(struct md_cluster_info *cinfo)
  */
 static int __sendmsg(struct md_cluster_info *cinfo, struct cluster_msg *cmsg)
 {
-	int error;
+	int error, unlock_error;
 	int slot = cinfo->slot_number - 1;
 
 	cmsg->slot = cpu_to_le32(slot);
 	/*get EX on Message*/
 	error = dlm_lock_sync(cinfo->message_lockres, DLM_LOCK_EX);
 	if (error) {
 		pr_err("md-cluster: failed to get EX on MESSAGE (%d)\n", error);
-		goto failed_message;
+		return error;
 	}
 
 	memcpy(cinfo->message_lockres->lksb.sb_lvbptr, (void *)cmsg,
@@ -781,14 +795,10 @@ static int __sendmsg(struct md_cluster_info *cinfo, struct cluster_msg *cmsg)
 	}
 
 failed_ack:
-	error = dlm_unlock_sync(cinfo->message_lockres);
-	if (unlikely(error != 0)) {
+	while ((unlock_error = dlm_unlock_sync(cinfo->message_lockres)))
 		pr_err("md-cluster: failed convert to NL on MESSAGE(%d)\n",
-			error);
-		/* in case the message can't be released due to some reason */
-		goto failed_ack;
-	}
-failed_message:
+			unlock_error);
+
 	return error;
 }
 
@@ -1343,6 +1353,23 @@ static void resync_info_get(struct mddev *mddev, sector_t *lo, sector_t *hi)
 	spin_unlock_irq(&cinfo->suspend_lock);
 }
 
+static int resync_status_get(struct mddev *mddev)
+{
+	struct md_cluster_info *cinfo = mddev->cluster_info;
+
+	return test_bit(MD_CLUSTER_WAITING_FOR_SYNC, &cinfo->state);
+}
+
+static int resync_start_notify(struct mddev *mddev)
+{
+	struct md_cluster_info *cinfo = mddev->cluster_info;
+	struct cluster_msg cmsg = {0};
+
+	cmsg.type = cpu_to_le32(RESYNCING_START);
+
+	return sendmsg(cinfo, &cmsg, 0);
+}
+
 static int resync_info_update(struct mddev *mddev, sector_t lo, sector_t hi)
 {
 	struct md_cluster_info *cinfo = mddev->cluster_info;
@@ -1577,6 +1604,8 @@ static const struct md_cluster_operations cluster_ops = {
 	.resync_start = resync_start,
 	.resync_finish = resync_finish,
 	.resync_info_update = resync_info_update,
+	.resync_start_notify = resync_start_notify,
+	.resync_status_get = resync_status_get,
 	.resync_info_get = resync_info_get,
 	.metadata_update_start = metadata_update_start,
 	.metadata_update_finish = metadata_update_finish,

drivers/md/md-cluster.h

@@ -14,6 +14,8 @@ struct md_cluster_operations {
 	int (*leave)(struct mddev *mddev);
 	int (*slot_number)(struct mddev *mddev);
 	int (*resync_info_update)(struct mddev *mddev, sector_t lo, sector_t hi);
+	int (*resync_start_notify)(struct mddev *mddev);
+	int (*resync_status_get)(struct mddev *mddev);
 	void (*resync_info_get)(struct mddev *mddev, sector_t *lo, sector_t *hi);
 	int (*metadata_update_start)(struct mddev *mddev);
 	int (*metadata_update_finish)(struct mddev *mddev);

drivers/md/md.c

@@ -8978,7 +8978,8 @@ void md_do_sync(struct md_thread *thread)
 	 * This will mean we have to start checking from the beginning again.
 	 *
 	 */
-
+	if (mddev_is_clustered(mddev))
+		md_cluster_ops->resync_start_notify(mddev);
 	do {
 		int mddev2_minor = -1;
 		mddev->curr_resync = MD_RESYNC_DELAYED;
@@ -9992,8 +9993,18 @@ static void check_sb_changes(struct mddev *mddev, struct md_rdev *rdev)
 			 */
 			if (rdev2->raid_disk == -1 && role != MD_DISK_ROLE_SPARE &&
 			    !(le32_to_cpu(sb->feature_map) &
-			      MD_FEATURE_RESHAPE_ACTIVE)) {
-				rdev2->saved_raid_disk = role;
+			      MD_FEATURE_RESHAPE_ACTIVE) &&
+			    !md_cluster_ops->resync_status_get(mddev)) {
+				/*
+				 * -1 to make raid1_add_disk() set conf->fullsync
+				 * to 1. This could avoid skipping sync when the
+				 * remote node is down during resyncing.
+				 */
+				if ((le32_to_cpu(sb->feature_map)
+				    & MD_FEATURE_RECOVERY_OFFSET))
+					rdev2->saved_raid_disk = -1;
+				else
+					rdev2->saved_raid_disk = role;
 				ret = remove_and_add_spares(mddev, rdev2);
 				pr_info("Activated spare: %pg\n",
 					rdev2->bdev);

drivers/md/raid1.c

@@ -680,6 +680,7 @@ static int choose_slow_rdev(struct r1conf *conf, struct r1bio *r1_bio,
 		len = r1_bio->sectors;
 		read_len = raid1_check_read_range(rdev, this_sector, &len);
 		if (read_len == r1_bio->sectors) {
+			*max_sectors = read_len;
 			update_read_sectors(conf, disk, this_sector, read_len);
 			return disk;
 		}

drivers/s390/block/dasd_devmap.c

@@ -2248,13 +2248,19 @@ static ssize_t dasd_copy_pair_store(struct device *dev,
 
 	/* allocate primary devmap if needed */
 	prim_devmap = dasd_find_busid(prim_busid);
-	if (IS_ERR(prim_devmap))
+	if (IS_ERR(prim_devmap)) {
 		prim_devmap = dasd_add_busid(prim_busid, DASD_FEATURE_DEFAULT);
+		if (IS_ERR(prim_devmap))
+			return PTR_ERR(prim_devmap);
+	}
 
 	/* allocate secondary devmap if needed */
 	sec_devmap = dasd_find_busid(sec_busid);
-	if (IS_ERR(sec_devmap))
+	if (IS_ERR(sec_devmap)) {
 		sec_devmap = dasd_add_busid(sec_busid, DASD_FEATURE_DEFAULT);
+		if (IS_ERR(sec_devmap))
+			return PTR_ERR(sec_devmap);
+	}
 
 	/* setting copy relation is only allowed for offline secondary */
 	if (sec_devmap->device)