Merge branch 'md-next' of...

Merge branch 'md-next' of https://git.kernel.org/pub/scm/linux/kernel/git/song/md into for-6.4/block

Pull MD updates from Song:

"- md/bitmap: Optimal last page size, by Jon Derrick
 - Various raid10 fixes, by Yu Kuai and Li Nan
 - md: add error_handlers for raid0 and linear, by Mariusz Tkaczyk"

* 'md-next' of https://git.kernel.org/pub/scm/linux/kernel/git/song/md:
  md/raid5: remove unused working_disks variable
  md/raid10: don't call bio_start_io_acct twice for bio which experienced read error
  md/raid10: fix memleak of md thread
  md/raid10: fix memleak for 'conf->bio_split'
  md/raid10: fix leak of 'r10bio->remaining' for recovery
  md/raid10: don't BUG_ON() in raise_barrier()
  md: fix soft lockup in status_resync
  md: add error_handlers for raid0 and linear
  md: Use optimal I/O size for last bitmap page
  md: Fix types in sb writer
  md: Move sb writer loop to its own function
  md/raid10: Fix typo in comment (replacment -> replacement)
  md: make kobj_type structures constant
  md/raid10: fix null-ptr-deref in raid10_sync_request
  md/raid10: fix task hung in raid10d

drivers/md/md-bitmap.c

@@ -209,76 +209,99 @@ static struct md_rdev *next_active_rdev(struct md_rdev *rdev, struct mddev *mdde
 	return NULL;
 }
 
-static int write_sb_page(struct bitmap *bitmap, struct page *page, int wait)
+static unsigned int optimal_io_size(struct block_device *bdev,
+				    unsigned int last_page_size,
+				    unsigned int io_size)
+{
+	if (bdev_io_opt(bdev) > bdev_logical_block_size(bdev))
+		return roundup(last_page_size, bdev_io_opt(bdev));
+	return io_size;
+}
+
+static unsigned int bitmap_io_size(unsigned int io_size, unsigned int opt_size,
+				   sector_t start, sector_t boundary)
+{
+	if (io_size != opt_size &&
+	    start + opt_size / SECTOR_SIZE <= boundary)
+		return opt_size;
+	if (start + io_size / SECTOR_SIZE <= boundary)
+		return io_size;
+
+	/* Overflows boundary */
+	return 0;
+}
+
+static int __write_sb_page(struct md_rdev *rdev, struct bitmap *bitmap,
+			   struct page *page)
 {
-	struct md_rdev *rdev;
 	struct block_device *bdev;
 	struct mddev *mddev = bitmap->mddev;
 	struct bitmap_storage *store = &bitmap->storage;
+	sector_t offset = mddev->bitmap_info.offset;
+	sector_t ps, sboff, doff;
+	unsigned int size = PAGE_SIZE;
+	unsigned int opt_size = PAGE_SIZE;
+
+	bdev = (rdev->meta_bdev) ? rdev->meta_bdev : rdev->bdev;
+	if (page->index == store->file_pages - 1) {
+		unsigned int last_page_size = store->bytes & (PAGE_SIZE - 1);
+
+		if (last_page_size == 0)
+			last_page_size = PAGE_SIZE;
+		size = roundup(last_page_size, bdev_logical_block_size(bdev));
+		opt_size = optimal_io_size(bdev, last_page_size, size);
+	}
+
+	ps = page->index * PAGE_SIZE / SECTOR_SIZE;
+	sboff = rdev->sb_start + offset;
+	doff = rdev->data_offset;
+
+	/* Just make sure we aren't corrupting data or metadata */
+	if (mddev->external) {
+		/* Bitmap could be anywhere. */
+		if (sboff + ps > doff &&
+		    sboff < (doff + mddev->dev_sectors + PAGE_SIZE / SECTOR_SIZE))
+			return -EINVAL;
+	} else if (offset < 0) {
+		/* DATA  BITMAP METADATA  */
+		size = bitmap_io_size(size, opt_size, offset + ps, 0);
+		if (size == 0)
+			/* bitmap runs in to metadata */
+			return -EINVAL;
+
+		if (doff + mddev->dev_sectors > sboff)
+			/* data runs in to bitmap */
+			return -EINVAL;
+	} else if (rdev->sb_start < rdev->data_offset) {
+		/* METADATA BITMAP DATA */
+		size = bitmap_io_size(size, opt_size, sboff + ps, doff);
+		if (size == 0)
+			/* bitmap runs in to data */
+			return -EINVAL;
+	} else {
+		/* DATA METADATA BITMAP - no problems */
+	}
 
-restart:
-	rdev = NULL;
-	while ((rdev = next_active_rdev(rdev, mddev)) != NULL) {
-		int size = PAGE_SIZE;
-		loff_t offset = mddev->bitmap_info.offset;
+	md_super_write(mddev, rdev, sboff + ps, (int) size, page);
+	return 0;
+}
 
-		bdev = (rdev->meta_bdev) ? rdev->meta_bdev : rdev->bdev;
+static int write_sb_page(struct bitmap *bitmap, struct page *page, int wait)
+{
+	struct md_rdev *rdev;
+	struct mddev *mddev = bitmap->mddev;
+	int ret;
 
-		if (page->index == store->file_pages-1) {
-			int last_page_size = store->bytes & (PAGE_SIZE-1);
-			if (last_page_size == 0)
-				last_page_size = PAGE_SIZE;
-			size = roundup(last_page_size,
-				       bdev_logical_block_size(bdev));
-		}
-		/* Just make sure we aren't corrupting data or
-		 * metadata
-		 */
-		if (mddev->external) {
-			/* Bitmap could be anywhere. */
-			if (rdev->sb_start + offset + (page->index
-						       * (PAGE_SIZE/512))
-			    > rdev->data_offset
-			    &&
-			    rdev->sb_start + offset
-			    < (rdev->data_offset + mddev->dev_sectors
-			     + (PAGE_SIZE/512)))
-				goto bad_alignment;
-		} else if (offset < 0) {
-			/* DATA  BITMAP METADATA  */
-			if (offset
-			    + (long)(page->index * (PAGE_SIZE/512))
-			    + size/512 > 0)
-				/* bitmap runs in to metadata */
-				goto bad_alignment;
-			if (rdev->data_offset + mddev->dev_sectors
-			    > rdev->sb_start + offset)
-				/* data runs in to bitmap */
-				goto bad_alignment;
-		} else if (rdev->sb_start < rdev->data_offset) {
-			/* METADATA BITMAP DATA */
-			if (rdev->sb_start
-			    + offset
-			    + page->index*(PAGE_SIZE/512) + size/512
-			    > rdev->data_offset)
-				/* bitmap runs in to data */
-				goto bad_alignment;
-		} else {
-			/* DATA METADATA BITMAP - no problems */
+	do {
+		rdev = NULL;
+		while ((rdev = next_active_rdev(rdev, mddev)) != NULL) {
+			ret = __write_sb_page(rdev, bitmap, page);
+			if (ret)
+				return ret;
 		}
-		md_super_write(mddev, rdev,
-			       rdev->sb_start + offset
-			       + page->index * (PAGE_SIZE/512),
-			       size,
-			       page);
-	}
+	} while (wait && md_super_wait(mddev) < 0);
 
-	if (wait && md_super_wait(mddev) < 0)
-		goto restart;
 	return 0;
-
- bad_alignment:
-	return -EINVAL;
 }
 
 static void md_bitmap_file_kick(struct bitmap *bitmap);

drivers/md/md-linear.c

@@ -223,7 +223,8 @@ static bool linear_make_request(struct mddev *mddev, struct bio *bio)
 		     bio_sector < start_sector))
 		goto out_of_bounds;
 
-	if (unlikely(is_mddev_broken(tmp_dev->rdev, "linear"))) {
+	if (unlikely(is_rdev_broken(tmp_dev->rdev))) {
+		md_error(mddev, tmp_dev->rdev);
 		bio_io_error(bio);
 		return true;
 	}
@@ -270,6 +271,16 @@ static void linear_status (struct seq_file *seq, struct mddev *mddev)
 	seq_printf(seq, " %dk rounding", mddev->chunk_sectors / 2);
 }
 
+static void linear_error(struct mddev *mddev, struct md_rdev *rdev)
+{
+	if (!test_and_set_bit(MD_BROKEN, &mddev->flags)) {
+		char *md_name = mdname(mddev);
+
+		pr_crit("md/linear%s: Disk failure on %pg detected, failing array.\n",
+			md_name, rdev->bdev);
+	}
+}
+
 static void linear_quiesce(struct mddev *mddev, int state)
 {
 }
@@ -286,6 +297,7 @@ static struct md_personality linear_personality =
 	.hot_add_disk	= linear_add,
 	.size		= linear_size,
 	.quiesce	= linear_quiesce,
+	.error_handler	= linear_error,
 };
 
 static int __init linear_init (void)

drivers/md/md.c

@@ -78,7 +78,7 @@
 static LIST_HEAD(pers_list);
 static DEFINE_SPINLOCK(pers_lock);
 
-static struct kobj_type md_ktype;
+static const struct kobj_type md_ktype;
 
 struct md_cluster_operations *md_cluster_ops;
 EXPORT_SYMBOL(md_cluster_ops);
@@ -3597,7 +3597,7 @@ static const struct sysfs_ops rdev_sysfs_ops = {
 	.show		= rdev_attr_show,
 	.store		= rdev_attr_store,
 };
-static struct kobj_type rdev_ktype = {
+static const struct kobj_type rdev_ktype = {
 	.release	= rdev_free,
 	.sysfs_ops	= &rdev_sysfs_ops,
 	.default_groups	= rdev_default_groups,
@@ -5555,7 +5555,7 @@ static const struct sysfs_ops md_sysfs_ops = {
 	.show	= md_attr_show,
 	.store	= md_attr_store,
 };
-static struct kobj_type md_ktype = {
+static const struct kobj_type md_ktype = {
 	.release	= md_kobj_release,
 	.sysfs_ops	= &md_sysfs_ops,
 	.default_groups	= md_attr_groups,
@@ -7974,6 +7974,9 @@ void md_error(struct mddev *mddev, struct md_rdev *rdev)
 		return;
 	mddev->pers->error_handler(mddev, rdev);
 
+	if (mddev->pers->level == 0 || mddev->pers->level == LEVEL_LINEAR)
+		return;
+
 	if (mddev->degraded && !test_bit(MD_BROKEN, &mddev->flags))
 		set_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
 	sysfs_notify_dirent_safe(rdev->sysfs_state);
@@ -8029,16 +8032,16 @@ static int status_resync(struct seq_file *seq, struct mddev *mddev)
 	} else if (resync > max_sectors) {
 		resync = max_sectors;
 	} else {
-		resync -= atomic_read(&mddev->recovery_active);
-		if (resync < MD_RESYNC_ACTIVE) {
-			/*
-			 * Resync has started, but the subtraction has
-			 * yielded one of the special values. Force it
-			 * to active to ensure the status reports an
-			 * active resync.
-			 */
+		res = atomic_read(&mddev->recovery_active);
+		/*
+		 * Resync has started, but the subtraction has overflowed or
+		 * yielded one of the special values. Force it to active to
+		 * ensure the status reports an active resync.
+		 */
+		if (resync < res || resync - res < MD_RESYNC_ACTIVE)
 			resync = MD_RESYNC_ACTIVE;
-		}
+		else
+			resync -= res;
 	}
 
 	if (resync == MD_RESYNC_NONE) {

drivers/md/md.h

@@ -790,15 +790,9 @@ extern void mddev_destroy_serial_pool(struct mddev *mddev, struct md_rdev *rdev,
 struct md_rdev *md_find_rdev_nr_rcu(struct mddev *mddev, int nr);
 struct md_rdev *md_find_rdev_rcu(struct mddev *mddev, dev_t dev);
 
-static inline bool is_mddev_broken(struct md_rdev *rdev, const char *md_type)
+static inline bool is_rdev_broken(struct md_rdev *rdev)
 {
-	if (!disk_live(rdev->bdev->bd_disk)) {
-		if (!test_and_set_bit(MD_BROKEN, &rdev->mddev->flags))
-			pr_warn("md: %s: %s array has a missing/failed member\n",
-				mdname(rdev->mddev), md_type);
-		return true;
-	}
-	return false;
+	return !disk_live(rdev->bdev->bd_disk);
 }
 
 static inline void rdev_dec_pending(struct md_rdev *rdev, struct mddev *mddev)

drivers/md/raid0.c

@@ -569,8 +569,9 @@ static bool raid0_make_request(struct mddev *mddev, struct bio *bio)
 		return true;
 	}
 
-	if (unlikely(is_mddev_broken(tmp_dev, "raid0"))) {
+	if (unlikely(is_rdev_broken(tmp_dev))) {
 		bio_io_error(bio);
+		md_error(mddev, tmp_dev);
 		return true;
 	}
 
@@ -592,6 +593,16 @@ static void raid0_status(struct seq_file *seq, struct mddev *mddev)
 	return;
 }
 
+static void raid0_error(struct mddev *mddev, struct md_rdev *rdev)
+{
+	if (!test_and_set_bit(MD_BROKEN, &mddev->flags)) {
+		char *md_name = mdname(mddev);
+
+		pr_crit("md/raid0%s: Disk failure on %pg detected, failing array.\n",
+			md_name, rdev->bdev);
+	}
+}
+
 static void *raid0_takeover_raid45(struct mddev *mddev)
 {
 	struct md_rdev *rdev;
@@ -767,6 +778,7 @@ static struct md_personality raid0_personality=
 	.size		= raid0_size,
 	.takeover	= raid0_takeover,
 	.quiesce	= raid0_quiesce,
+	.error_handler	= raid0_error,
 };
 
 static int __init raid0_init (void)

drivers/md/raid10.c

@@ -952,7 +952,9 @@ static void flush_pending_writes(struct r10conf *conf)
 static void raise_barrier(struct r10conf *conf, int force)
 {
 	write_seqlock_irq(&conf->resync_lock);
-	BUG_ON(force && !conf->barrier);
+
+	if (WARN_ON_ONCE(force && !conf->barrier))
+		force = false;
 
 	/* Wait until no block IO is waiting (unless 'force') */
 	wait_event_barrier(conf, force || !conf->nr_waiting);
@@ -995,11 +997,15 @@ static bool stop_waiting_barrier(struct r10conf *conf)
 	    (!bio_list_empty(&bio_list[0]) || !bio_list_empty(&bio_list[1])))
 		return true;
 
-	/* move on if recovery thread is blocked by us */
-	if (conf->mddev->thread->tsk == current &&
-	    test_bit(MD_RECOVERY_RUNNING, &conf->mddev->recovery) &&
-	    conf->nr_queued > 0)
+	/*
+	 * move on if io is issued from raid10d(), nr_pending is not released
+	 * from original io(see handle_read_error()). All raise barrier is
+	 * blocked until this io is done.
+	 */
+	if (conf->mddev->thread->tsk == current) {
+		WARN_ON_ONCE(atomic_read(&conf->nr_pending) == 0);
 		return true;
+	}
 
 	return false;
 }
@@ -1244,7 +1250,8 @@ static void raid10_read_request(struct mddev *mddev, struct bio *bio,
 	}
 	slot = r10_bio->read_slot;
 
-	if (blk_queue_io_stat(bio->bi_bdev->bd_disk->queue))
+	if (!r10_bio->start_time &&
+	    blk_queue_io_stat(bio->bi_bdev->bd_disk->queue))
 		r10_bio->start_time = bio_start_io_acct(bio);
 	read_bio = bio_alloc_clone(rdev->bdev, bio, gfp, &mddev->bio_set);
 
@@ -1574,6 +1581,7 @@ static void __make_request(struct mddev *mddev, struct bio *bio, int sectors)
 	r10_bio->sector = bio->bi_iter.bi_sector;
 	r10_bio->state = 0;
 	r10_bio->read_slot = -1;
+	r10_bio->start_time = 0;
 	memset(r10_bio->devs, 0, sizeof(r10_bio->devs[0]) *
 			conf->geo.raid_disks);
 
@@ -1626,7 +1634,7 @@ static void raid10_end_discard_request(struct bio *bio)
 		/*
 		 * raid10_remove_disk uses smp_mb to make sure rdev is set to
 		 * replacement before setting replacement to NULL. It can read
-		 * rdev first without barrier protect even replacment is NULL
+		 * rdev first without barrier protect even replacement is NULL
 		 */
 		smp_rmb();
 		rdev = conf->mirrors[dev].rdev;
@@ -2609,11 +2617,22 @@ static void recovery_request_write(struct mddev *mddev, struct r10bio *r10_bio)
 {
 	struct r10conf *conf = mddev->private;
 	int d;
-	struct bio *wbio, *wbio2;
+	struct bio *wbio = r10_bio->devs[1].bio;
+	struct bio *wbio2 = r10_bio->devs[1].repl_bio;
+
+	/* Need to test wbio2->bi_end_io before we call
+	 * submit_bio_noacct as if the former is NULL,
+	 * the latter is free to free wbio2.
+	 */
+	if (wbio2 && !wbio2->bi_end_io)
+		wbio2 = NULL;
 
 	if (!test_bit(R10BIO_Uptodate, &r10_bio->state)) {
 		fix_recovery_read_error(r10_bio);
-		end_sync_request(r10_bio);
+		if (wbio->bi_end_io)
+			end_sync_request(r10_bio);
+		if (wbio2)
+			end_sync_request(r10_bio);
 		return;
 	}
 
@@ -2622,14 +2641,6 @@ static void recovery_request_write(struct mddev *mddev, struct r10bio *r10_bio)
 	 * and submit the write request
 	 */
 	d = r10_bio->devs[1].devnum;
-	wbio = r10_bio->devs[1].bio;
-	wbio2 = r10_bio->devs[1].repl_bio;
-	/* Need to test wbio2->bi_end_io before we call
-	 * submit_bio_noacct as if the former is NULL,
-	 * the latter is free to free wbio2.
-	 */
-	if (wbio2 && !wbio2->bi_end_io)
-		wbio2 = NULL;
 	if (wbio->bi_end_io) {
 		atomic_inc(&conf->mirrors[d].rdev->nr_pending);
 		md_sync_acct(conf->mirrors[d].rdev->bdev, bio_sectors(wbio));
@@ -2978,9 +2989,13 @@ static void handle_read_error(struct mddev *mddev, struct r10bio *r10_bio)
 		md_error(mddev, rdev);
 
 	rdev_dec_pending(rdev, mddev);
-	allow_barrier(conf);
 	r10_bio->state = 0;
 	raid10_read_request(mddev, r10_bio->master_bio, r10_bio);
+	/*
+	 * allow_barrier after re-submit to ensure no sync io
+	 * can be issued while regular io pending.
+	 */
+	allow_barrier(conf);
 }
 
 static void handle_write_completed(struct r10conf *conf, struct r10bio *r10_bio)
@@ -3289,10 +3304,6 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr,
 	sector_t chunk_mask = conf->geo.chunk_mask;
 	int page_idx = 0;
 
-	if (!mempool_initialized(&conf->r10buf_pool))
-		if (init_resync(conf))
-			return 0;
-
 	/*
 	 * Allow skipping a full rebuild for incremental assembly
 	 * of a clean array, like RAID1 does.
@@ -3308,6 +3319,10 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr,
 		return mddev->dev_sectors - sector_nr;
 	}
 
+	if (!mempool_initialized(&conf->r10buf_pool))
+		if (init_resync(conf))
+			return 0;
+
  skipped:
 	max_sector = mddev->dev_sectors;
 	if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery) ||
@@ -4004,6 +4019,20 @@ static int setup_geo(struct geom *geo, struct mddev *mddev, enum geo_type new)
 	return nc*fc;
 }
 
+static void raid10_free_conf(struct r10conf *conf)
+{
+	if (!conf)
+		return;
+
+	mempool_exit(&conf->r10bio_pool);
+	kfree(conf->mirrors);
+	kfree(conf->mirrors_old);
+	kfree(conf->mirrors_new);
+	safe_put_page(conf->tmppage);
+	bioset_exit(&conf->bio_split);
+	kfree(conf);
+}
+
 static struct r10conf *setup_conf(struct mddev *mddev)
 {
 	struct r10conf *conf = NULL;
@@ -4086,13 +4115,7 @@ static struct r10conf *setup_conf(struct mddev *mddev)
 	return conf;
 
  out:
-	if (conf) {
-		mempool_exit(&conf->r10bio_pool);
-		kfree(conf->mirrors);
-		safe_put_page(conf->tmppage);
-		bioset_exit(&conf->bio_split);
-		kfree(conf);
-	}
+	raid10_free_conf(conf);
 	return ERR_PTR(err);
 }
 
@@ -4129,6 +4152,9 @@ static int raid10_run(struct mddev *mddev)
 	if (!conf)
 		goto out;
 
+	mddev->thread = conf->thread;
+	conf->thread = NULL;
+
 	if (mddev_is_clustered(conf->mddev)) {
 		int fc, fo;
 
@@ -4141,9 +4167,6 @@ static int raid10_run(struct mddev *mddev)
 		}
 	}
 
-	mddev->thread = conf->thread;
-	conf->thread = NULL;
-
 	if (mddev->queue) {
 		blk_queue_max_write_zeroes_sectors(mddev->queue, 0);
 		blk_queue_io_min(mddev->queue, mddev->chunk_sectors << 9);
@@ -4283,10 +4306,7 @@ static int raid10_run(struct mddev *mddev)
 
 out_free_conf:
 	md_unregister_thread(&mddev->thread);
-	mempool_exit(&conf->r10bio_pool);
-	safe_put_page(conf->tmppage);
-	kfree(conf->mirrors);
-	kfree(conf);
+	raid10_free_conf(conf);
 	mddev->private = NULL;
 out:
 	return -EIO;
@@ -4294,15 +4314,7 @@ static int raid10_run(struct mddev *mddev)
 
 static void raid10_free(struct mddev *mddev, void *priv)
 {
-	struct r10conf *conf = priv;
-
-	mempool_exit(&conf->r10bio_pool);
-	safe_put_page(conf->tmppage);
-	kfree(conf->mirrors);
-	kfree(conf->mirrors_old);
-	kfree(conf->mirrors_new);
-	bioset_exit(&conf->bio_split);
-	kfree(conf);
+	raid10_free_conf(priv);
 }
 
 static void raid10_quiesce(struct mddev *mddev, int quiesce)

drivers/md/raid5.c

@@ -7716,7 +7716,6 @@ static void raid5_set_io_opt(struct r5conf *conf)
 static int raid5_run(struct mddev *mddev)
 {
 	struct r5conf *conf;
-	int working_disks = 0;
 	int dirty_parity_disks = 0;
 	struct md_rdev *rdev;
 	struct md_rdev *journal_dev = NULL;
@@ -7912,10 +7911,8 @@ static int raid5_run(struct mddev *mddev)
 			pr_warn("md: cannot handle concurrent replacement and reshape.\n");
 			goto abort;
 		}
-		if (test_bit(In_sync, &rdev->flags)) {
-			working_disks++;
+		if (test_bit(In_sync, &rdev->flags))
 			continue;
-		}
 		/* This disc is not fully in-sync.  However if it
 		 * just stored parity (beyond the recovery_offset),
 		 * when we don't need to be concerned about the