Merge tag 'md-6.18-20250909' of gitolite.kernel.org:pub/scm/linux/kernel/git/mdraid/linux into for-6.18/block

Pull MD changes from Yu Kuai:

"Redundant data is used to enhance data fault tolerance, and the storage
 method for redundant data vary depending on the RAID levels. And it's
 important to maintain the consistency of redundant data.

 Bitmap is used to record which data blocks have been synchronized and
 which ones need to be resynchronized or recovered. Each bit in the
 bitmap represents a segment of data in the array. When a bit is set,
 it indicates that the multiple redundant copies of that data segment
 may not be consistent. Data synchronization can be performed based on
 the bitmap after power failure or readding a disk. If there is no
 bitmap, a full disk synchronization is required.

 Due to known performance issues with md-bitmap and the unreasonable
 implementations:

 - self-managed IO submitting like filemap_write_page();
 - global spin_lock

 I have decided not to continue optimizing based on the current bitmap
 implementation, this new bitmap is invented without locking from IO fast
 path and can be used with fast disks.

 Key features for the new bitmap:
  - IO fastpath is lockless, if user issues lots of write IO to the same
    bitmap bit in a short time, only the first write has additional
    overhead to update bitmap bit, no additional overhead for the
    following writes;
  - support only resync or recover written data, means in the case
    creating new array or replacing with a new disk, there is no need to
    do a full disk resync/recovery;"

* tag 'md-6.18-20250909' of gitolite.kernel.org:pub/scm/linux/kernel/git/mdraid/linux: (24 commits)
  md/md-llbitmap: introduce new lockless bitmap
  md/md-bitmap: make method bitmap_ops->daemon_work optional
  md: add a new recovery_flag MD_RECOVERY_LAZY_RECOVER
  md/md-bitmap: add a new method blocks_synced() in bitmap_operations
  md/md-bitmap: add a new method skip_sync_blocks() in bitmap_operations
  md/md-bitmap: delay registration of bitmap_ops until creating bitmap
  md/md-bitmap: add a new sysfs api bitmap_type
  md: add a new mddev field 'bitmap_id'
  md/md-bitmap: support discard for bitmap ops
  md: factor out a helper raid_is_456()
  md: add a new parameter 'offset' to md_super_write()
  md/md-bitmap: introduce CONFIG_MD_BITMAP
  md: check before referencing mddev->bitmap_ops
  md/dm-raid: check before referencing mddev->bitmap_ops
  md/raid5: check before referencing mddev->bitmap_ops
  md/raid10: check before referencing mddev->bitmap_ops
  md/raid1: check before referencing mddev->bitmap_ops
  md/raid1: check bitmap before behind write
  md/md-bitmap: handle the case bitmap is not enabled before end_sync()
  md/md-bitmap: handle the case bitmap is not enabled before start_sync()
  ...

Documentation/admin-guide/md.rst

@@ -347,6 +347,54 @@ All md devices contain:
      active-idle
          like active, but no writes have been seen for a while (safe_mode_delay).
 
+  consistency_policy
+     This indicates how the array maintains consistency in case of unexpected
+     shutdown. It can be:
+
+     none
+       Array has no redundancy information, e.g. raid0, linear.
+
+     resync
+       Full resync is performed and all redundancy is regenerated when the
+       array is started after unclean shutdown.
+
+     bitmap
+       Resync assisted by a write-intent bitmap.
+
+     journal
+       For raid4/5/6, journal device is used to log transactions and replay
+       after unclean shutdown.
+
+     ppl
+       For raid5 only, Partial Parity Log is used to close the write hole and
+       eliminate resync.
+
+     The accepted values when writing to this file are ``ppl`` and ``resync``,
+     used to enable and disable PPL.
+
+  uuid
+     This indicates the UUID of the array in the following format:
+     xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx
+
+  bitmap_type
+     [RW] When read, this file will display the current and available
+     bitmap for this array. The currently active bitmap will be enclosed
+     in [] brackets. Writing an bitmap name or ID to this file will switch
+     control of this array to that new bitmap. Note that writing a new
+     bitmap for created array is forbidden.
+
+     none
+         No bitmap
+     bitmap
+         The default internal bitmap
+     llbitmap
+         The lockless internal bitmap
+
+If bitmap_type is not none, then additional bitmap attributes bitmap/xxx or
+llbitmap/xxx will be created after md device KOBJ_CHANGE event.
+
+If bitmap_type is bitmap, then the md device will also contain:
+
   bitmap/location
      This indicates where the write-intent bitmap for the array is
      stored.
@@ -401,35 +449,23 @@ All md devices contain:
      once the array becomes non-degraded, and this fact has been
      recorded in the metadata.
 
-  consistency_policy
-     This indicates how the array maintains consistency in case of unexpected
-     shutdown. It can be:
+If bitmap_type is llbitmap, then the md device will also contain:
 
-     none
-       Array has no redundancy information, e.g. raid0, linear.
+  llbitmap/bits
+     This is read-only, show status of bitmap bits, the number of each
+     value.
 
-     resync
-       Full resync is performed and all redundancy is regenerated when the
-       array is started after unclean shutdown.
+  llbitmap/metadata
+     This is read-only, show bitmap metadata, include chunksize, chunkshift,
+     chunks, offset and daemon_sleep.
 
-     bitmap
-       Resync assisted by a write-intent bitmap.
-
-     journal
-       For raid4/5/6, journal device is used to log transactions and replay
-       after unclean shutdown.
-
-     ppl
-       For raid5 only, Partial Parity Log is used to close the write hole and
-       eliminate resync.
-
-     The accepted values when writing to this file are ``ppl`` and ``resync``,
-     used to enable and disable PPL.
-
-  uuid
-     This indicates the UUID of the array in the following format:
-     xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx
+  llbitmap/daemon_sleep
+     This is read-write, time in seconds that daemon function will be
+     triggered to clear dirty bits.
 
+  llbitmap/barrier_idle
+     This is read-write, time in seconds that page barrier will be idled,
+     means dirty bits in the page will be cleared.
 
 As component devices are added to an md array, they appear in the ``md``
 directory as new directories named::

drivers/md/Kconfig

@@ -37,6 +37,32 @@ config BLK_DEV_MD
 
 	  If unsure, say N.
 
+config MD_BITMAP
+	bool "MD RAID bitmap support"
+	default y
+	depends on BLK_DEV_MD
+	help
+	  If you say Y here, support for the write intent bitmap will be
+	  enabled. The bitmap can be used to optimize resync speed after power
+	  failure or readding a disk, limiting it to recorded dirty sectors in
+	  bitmap.
+
+	  This feature can be added to existing MD array or MD array can be
+	  created with bitmap via mdadm(8).
+
+	  If unsure, say Y.
+
+config MD_LLBITMAP
+	bool "MD RAID lockless bitmap support"
+	depends on BLK_DEV_MD
+	help
+	  If you say Y here, support for the lockless write intent bitmap will
+	  be enabled.
+
+	  Note, this is an experimental feature.
+
+	  If unsure, say N.
+
 config MD_AUTODETECT
 	bool "Autodetect RAID arrays during kernel boot"
 	depends on BLK_DEV_MD=y
@@ -54,6 +80,7 @@ config MD_AUTODETECT
 config MD_BITMAP_FILE
 	bool "MD bitmap file support (deprecated)"
 	default y
+	depends on MD_BITMAP
 	help
 	  If you say Y here, support for write intent bitmaps in files on an
 	  external file system is enabled.  This is an alternative to the internal
@@ -174,6 +201,7 @@ config MD_RAID456
 
 config MD_CLUSTER
 	tristate "Cluster Support for MD"
+	select MD_BITMAP
 	depends on BLK_DEV_MD
 	depends on DLM
 	default n
@@ -393,6 +421,7 @@ config DM_RAID
        select MD_RAID1
        select MD_RAID10
        select MD_RAID456
+       select MD_BITMAP
        select BLK_DEV_MD
 	help
 	 A dm target that supports RAID1, RAID10, RAID4, RAID5 and RAID6 mappings

drivers/md/Makefile

@@ -27,7 +27,9 @@ dm-clone-y	+= dm-clone-target.o dm-clone-metadata.o
 dm-verity-y	+= dm-verity-target.o
 dm-zoned-y	+= dm-zoned-target.o dm-zoned-metadata.o dm-zoned-reclaim.o
 
-md-mod-y	+= md.o md-bitmap.o
+md-mod-y	+= md.o
+md-mod-$(CONFIG_MD_BITMAP)	+= md-bitmap.o
+md-mod-$(CONFIG_MD_LLBITMAP)	+= md-llbitmap.o
 raid456-y	+= raid5.o raid5-cache.o raid5-ppl.o
 linear-y       += md-linear.o
 

drivers/md/dm-raid.c

@@ -3953,9 +3953,11 @@ static int __load_dirty_region_bitmap(struct raid_set *rs)
 	    !test_and_set_bit(RT_FLAG_RS_BITMAP_LOADED, &rs->runtime_flags)) {
 		struct mddev *mddev = &rs->md;
 
-		r = mddev->bitmap_ops->load(mddev);
-		if (r)
-			DMERR("Failed to load bitmap");
+		if (md_bitmap_enabled(mddev, false)) {
+			r = mddev->bitmap_ops->load(mddev);
+			if (r)
+				DMERR("Failed to load bitmap");
+		}
 	}
 
 	return r;
@@ -4070,10 +4072,12 @@ static int raid_preresume(struct dm_target *ti)
 	       mddev->bitmap_info.chunksize != to_bytes(rs->requested_bitmap_chunk_sectors)))) {
 		int chunksize = to_bytes(rs->requested_bitmap_chunk_sectors) ?: mddev->bitmap_info.chunksize;
 
-		r = mddev->bitmap_ops->resize(mddev, mddev->dev_sectors,
-					      chunksize, false);
-		if (r)
-			DMERR("Failed to resize bitmap");
+		if (md_bitmap_enabled(mddev, false)) {
+			r = mddev->bitmap_ops->resize(mddev, mddev->dev_sectors,
+						      chunksize);
+			if (r)
+				DMERR("Failed to resize bitmap");
+		}
 	}
 
 	/* Check for any resize/reshape on @rs and adjust/initiate */

drivers/md/md-bitmap.c

@@ -34,15 +34,6 @@
 #include "md-bitmap.h"
 #include "md-cluster.h"
 
-#define BITMAP_MAJOR_LO 3
-/* version 4 insists the bitmap is in little-endian order
- * with version 3, it is host-endian which is non-portable
- * Version 5 is currently set only for clustered devices
- */
-#define BITMAP_MAJOR_HI 4
-#define BITMAP_MAJOR_CLUSTERED 5
-#define	BITMAP_MAJOR_HOSTENDIAN 3
-
 /*
  * in-memory bitmap:
  *
@@ -224,6 +215,8 @@ struct bitmap {
 	int cluster_slot;
 };
 
+static struct workqueue_struct *md_bitmap_wq;
+
 static int __bitmap_resize(struct bitmap *bitmap, sector_t blocks,
 			   int chunksize, bool init);
 
@@ -232,20 +225,19 @@ static inline char *bmname(struct bitmap *bitmap)
 	return bitmap->mddev ? mdname(bitmap->mddev) : "mdX";
 }
 
-static bool __bitmap_enabled(struct bitmap *bitmap)
+static bool bitmap_enabled(void *data, bool flush)
 {
-	return bitmap->storage.filemap &&
-	       !test_bit(BITMAP_STALE, &bitmap->flags);
-}
+	struct bitmap *bitmap = data;
 
-static bool bitmap_enabled(struct mddev *mddev)
-{
-	struct bitmap *bitmap = mddev->bitmap;
+	if (!flush)
+		return true;
 
-	if (!bitmap)
-		return false;
-
-	return __bitmap_enabled(bitmap);
+	/*
+	 * If caller want to flush bitmap pages to underlying disks, check if
+	 * there are cached pages in filemap.
+	 */
+	return !test_bit(BITMAP_STALE, &bitmap->flags) &&
+	       bitmap->storage.filemap != NULL;
 }
 
 /*
@@ -484,7 +476,8 @@ static int __write_sb_page(struct md_rdev *rdev, struct bitmap *bitmap,
 			return -EINVAL;
 	}
 
-	md_super_write(mddev, rdev, sboff + ps, (int)min(size, bitmap_limit), page);
+	md_write_metadata(mddev, rdev, sboff + ps, (int)min(size, bitmap_limit),
+			  page, 0);
 	return 0;
 }
 
@@ -1244,7 +1237,7 @@ static void __bitmap_unplug(struct bitmap *bitmap)
 	int dirty, need_write;
 	int writing = 0;
 
-	if (!__bitmap_enabled(bitmap))
+	if (!bitmap_enabled(bitmap, true))
 		return;
 
 	/* look at each page to see if there are any set bits that need to be
@@ -1788,15 +1781,9 @@ static bool __bitmap_start_sync(struct bitmap *bitmap, sector_t offset,
 				sector_t *blocks, bool degraded)
 {
 	bitmap_counter_t *bmc;
-	bool rv;
+	bool rv = false;
 
-	if (bitmap == NULL) {/* FIXME or bitmap set as 'failed' */
-		*blocks = 1024;
-		return true; /* always resync if no bitmap */
-	}
 	spin_lock_irq(&bitmap->counts.lock);
-
-	rv = false;
 	bmc = md_bitmap_get_counter(&bitmap->counts, offset, blocks, 0);
 	if (bmc) {
 		/* locked */
@@ -1845,10 +1832,6 @@ static void __bitmap_end_sync(struct bitmap *bitmap, sector_t offset,
 	bitmap_counter_t *bmc;
 	unsigned long flags;
 
-	if (bitmap == NULL) {
-		*blocks = 1024;
-		return;
-	}
 	spin_lock_irqsave(&bitmap->counts.lock, flags);
 	bmc = md_bitmap_get_counter(&bitmap->counts, offset, blocks, 0);
 	if (bmc == NULL)
@@ -2060,9 +2043,6 @@ static void bitmap_start_behind_write(struct mddev *mddev)
 	struct bitmap *bitmap = mddev->bitmap;
 	int bw;
 
-	if (!bitmap)
-		return;
-
 	atomic_inc(&bitmap->behind_writes);
 	bw = atomic_read(&bitmap->behind_writes);
 	if (bw > bitmap->behind_writes_used)
@@ -2076,9 +2056,6 @@ static void bitmap_end_behind_write(struct mddev *mddev)
 {
 	struct bitmap *bitmap = mddev->bitmap;
 
-	if (!bitmap)
-		return;
-
 	if (atomic_dec_and_test(&bitmap->behind_writes))
 		wake_up(&bitmap->behind_wait);
 	pr_debug("dec write-behind count %d/%lu\n",
@@ -2593,15 +2570,14 @@ err:
 	return ret;
 }
 
-static int bitmap_resize(struct mddev *mddev, sector_t blocks, int chunksize,
-			 bool init)
+static int bitmap_resize(struct mddev *mddev, sector_t blocks, int chunksize)
 {
 	struct bitmap *bitmap = mddev->bitmap;
 
 	if (!bitmap)
 		return 0;
 
-	return __bitmap_resize(bitmap, blocks, chunksize, init);
+	return __bitmap_resize(bitmap, blocks, chunksize, false);
 }
 
 static ssize_t
@@ -2990,12 +2966,19 @@ static struct attribute *md_bitmap_attrs[] = {
 	&max_backlog_used.attr,
 	NULL
 };
-const struct attribute_group md_bitmap_group = {
+
+static struct attribute_group md_bitmap_group = {
 	.name = "bitmap",
 	.attrs = md_bitmap_attrs,
 };
 
 static struct bitmap_operations bitmap_ops = {
+	.head = {
+		.type	= MD_BITMAP,
+		.id	= ID_BITMAP,
+		.name	= "bitmap",
+	},
+
 	.enabled		= bitmap_enabled,
 	.create			= bitmap_create,
 	.resize			= bitmap_resize,
@@ -3013,6 +2996,9 @@ static struct bitmap_operations bitmap_ops = {
 
 	.start_write		= bitmap_start_write,
 	.end_write		= bitmap_end_write,
+	.start_discard		= bitmap_start_write,
+	.end_discard		= bitmap_end_write,
+
 	.start_sync		= bitmap_start_sync,
 	.end_sync		= bitmap_end_sync,
 	.cond_end_sync		= bitmap_cond_end_sync,
@@ -3026,9 +3012,22 @@ static struct bitmap_operations bitmap_ops = {
 	.copy_from_slot		= bitmap_copy_from_slot,
 	.set_pages		= bitmap_set_pages,
 	.free			= md_bitmap_free,
+
+	.group			= &md_bitmap_group,
 };
 
-void mddev_set_bitmap_ops(struct mddev *mddev)
+int md_bitmap_init(void)
 {
-	mddev->bitmap_ops = &bitmap_ops;
+	md_bitmap_wq = alloc_workqueue("md_bitmap", WQ_MEM_RECLAIM | WQ_UNBOUND,
+				       0);
+	if (!md_bitmap_wq)
+		return -ENOMEM;
+
+	return register_md_submodule(&bitmap_ops.head);
+}
+
+void md_bitmap_exit(void)
+{
+	destroy_workqueue(md_bitmap_wq);
+	unregister_md_submodule(&bitmap_ops.head);
 }

drivers/md/md-bitmap.h

@@ -9,10 +9,26 @@
 
 #define BITMAP_MAGIC 0x6d746962
 
+/*
+ * version 3 is host-endian order, this is deprecated and not used for new
+ * array
+ */
+#define BITMAP_MAJOR_LO		3
+#define BITMAP_MAJOR_HOSTENDIAN	3
+/* version 4 is little-endian order, the default value */
+#define BITMAP_MAJOR_HI		4
+/* version 5 is only used for cluster */
+#define BITMAP_MAJOR_CLUSTERED	5
+/* version 6 is only used for lockless bitmap */
+#define BITMAP_MAJOR_LOCKLESS	6
+
 /* use these for bitmap->flags and bitmap->sb->state bit-fields */
 enum bitmap_state {
-	BITMAP_STALE	   = 1,  /* the bitmap file is out of date or had -EIO */
+	BITMAP_STALE	   = 1, /* the bitmap file is out of date or had -EIO */
 	BITMAP_WRITE_ERROR = 2, /* A write error has occurred */
+	BITMAP_FIRST_USE   = 3, /* llbitmap is just created */
+	BITMAP_CLEAN       = 4, /* llbitmap is created with assume_clean */
+	BITMAP_DAEMON_BUSY = 5, /* llbitmap daemon is not finished after daemon_sleep */
 	BITMAP_HOSTENDIAN  =15,
 };
 
@@ -61,11 +77,15 @@ struct md_bitmap_stats {
 	struct file	*file;
 };
 
+typedef void (md_bitmap_fn)(struct mddev *mddev, sector_t offset,
+			    unsigned long sectors);
+
 struct bitmap_operations {
-	bool (*enabled)(struct mddev *mddev);
+	struct md_submodule_head head;
+
+	bool (*enabled)(void *data, bool flush);
 	int (*create)(struct mddev *mddev);
-	int (*resize)(struct mddev *mddev, sector_t blocks, int chunksize,
-		      bool init);
+	int (*resize)(struct mddev *mddev, sector_t blocks, int chunksize);
 
 	int (*load)(struct mddev *mddev);
 	void (*destroy)(struct mddev *mddev);
@@ -80,10 +100,13 @@ struct bitmap_operations {
 	void (*end_behind_write)(struct mddev *mddev);
 	void (*wait_behind_writes)(struct mddev *mddev);
 
-	void (*start_write)(struct mddev *mddev, sector_t offset,
-			    unsigned long sectors);
-	void (*end_write)(struct mddev *mddev, sector_t offset,
-			  unsigned long sectors);
+	md_bitmap_fn *start_write;
+	md_bitmap_fn *end_write;
+	md_bitmap_fn *start_discard;
+	md_bitmap_fn *end_discard;
+
+	sector_t (*skip_sync_blocks)(struct mddev *mddev, sector_t offset);
+	bool (*blocks_synced)(struct mddev *mddev, sector_t offset);
 	bool (*start_sync)(struct mddev *mddev, sector_t offset,
 			   sector_t *blocks, bool degraded);
 	void (*end_sync)(struct mddev *mddev, sector_t offset, sector_t *blocks);
@@ -101,9 +124,75 @@ struct bitmap_operations {
 			      sector_t *hi, bool clear_bits);
 	void (*set_pages)(void *data, unsigned long pages);
 	void (*free)(void *data);
+
+	struct attribute_group *group;
 };
 
 /* the bitmap API */
-void mddev_set_bitmap_ops(struct mddev *mddev);
+static inline bool md_bitmap_registered(struct mddev *mddev)
+{
+	return mddev->bitmap_ops != NULL;
+}
+
+static inline bool md_bitmap_enabled(struct mddev *mddev, bool flush)
+{
+	/* bitmap_ops must be registered before creating bitmap. */
+	if (!md_bitmap_registered(mddev))
+		return false;
+
+	if (!mddev->bitmap)
+		return false;
+
+	return mddev->bitmap_ops->enabled(mddev->bitmap, flush);
+}
+
+static inline bool md_bitmap_start_sync(struct mddev *mddev, sector_t offset,
+					sector_t *blocks, bool degraded)
+{
+	/* always resync if no bitmap */
+	if (!md_bitmap_enabled(mddev, false)) {
+		*blocks = 1024;
+		return true;
+	}
+
+	return mddev->bitmap_ops->start_sync(mddev, offset, blocks, degraded);
+}
+
+static inline void md_bitmap_end_sync(struct mddev *mddev, sector_t offset,
+				      sector_t *blocks)
+{
+	if (!md_bitmap_enabled(mddev, false)) {
+		*blocks = 1024;
+		return;
+	}
+
+	mddev->bitmap_ops->end_sync(mddev, offset, blocks);
+}
+
+#ifdef CONFIG_MD_BITMAP
+int md_bitmap_init(void);
+void md_bitmap_exit(void);
+#else
+static inline int md_bitmap_init(void)
+{
+	return 0;
+}
+static inline void md_bitmap_exit(void)
+{
+}
+#endif
+
+#ifdef CONFIG_MD_LLBITMAP
+int md_llbitmap_init(void);
+void md_llbitmap_exit(void);
+#else
+static inline int md_llbitmap_init(void)
+{
+	return 0;
+}
+static inline void md_llbitmap_exit(void)
+{
+}
+#endif
 
 #endif

drivers/md/md-cluster.c

@@ -630,7 +630,7 @@ static int process_recvd_msg(struct mddev *mddev, struct cluster_msg *msg)
 		if (le64_to_cpu(msg->high) != mddev->pers->size(mddev, 0, 0))
 			ret = mddev->bitmap_ops->resize(mddev,
 							le64_to_cpu(msg->high),
-							0, false);
+							0);
 		break;
 	default:
 		ret = -1;

drivers/md/md.c

@@ -94,7 +94,6 @@ static struct workqueue_struct *md_wq;
  * workqueue whith reconfig_mutex grabbed.
  */
 static struct workqueue_struct *md_misc_wq;
-struct workqueue_struct *md_bitmap_wq;
 
 static int remove_and_add_spares(struct mddev *mddev,
 				 struct md_rdev *this);
@@ -677,8 +676,64 @@ static void active_io_release(struct percpu_ref *ref)
 
 static void no_op(struct percpu_ref *r) {}
 
+static bool mddev_set_bitmap_ops(struct mddev *mddev)
+{
+	struct bitmap_operations *old = mddev->bitmap_ops;
+	struct md_submodule_head *head;
+
+	if (mddev->bitmap_id == ID_BITMAP_NONE ||
+	    (old && old->head.id == mddev->bitmap_id))
+		return true;
+
+	xa_lock(&md_submodule);
+	head = xa_load(&md_submodule, mddev->bitmap_id);
+
+	if (!head) {
+		pr_warn("md: can't find bitmap id %d\n", mddev->bitmap_id);
+		goto err;
+	}
+
+	if (head->type != MD_BITMAP) {
+		pr_warn("md: invalid bitmap id %d\n", mddev->bitmap_id);
+		goto err;
+	}
+
+	mddev->bitmap_ops = (void *)head;
+	xa_unlock(&md_submodule);
+
+	if (!mddev_is_dm(mddev) && mddev->bitmap_ops->group) {
+		if (sysfs_create_group(&mddev->kobj, mddev->bitmap_ops->group))
+			pr_warn("md: cannot register extra bitmap attributes for %s\n",
+				mdname(mddev));
+		else
+			/*
+			 * Inform user with KOBJ_CHANGE about new bitmap
+			 * attributes.
+			 */
+			kobject_uevent(&mddev->kobj, KOBJ_CHANGE);
+	}
+	return true;
+
+err:
+	xa_unlock(&md_submodule);
+	return false;
+}
+
+static void mddev_clear_bitmap_ops(struct mddev *mddev)
+{
+	if (!mddev_is_dm(mddev) && mddev->bitmap_ops &&
+	    mddev->bitmap_ops->group)
+		sysfs_remove_group(&mddev->kobj, mddev->bitmap_ops->group);
+
+	mddev->bitmap_ops = NULL;
+}
+
 int mddev_init(struct mddev *mddev)
 {
+	if (!IS_ENABLED(CONFIG_MD_BITMAP))
+		mddev->bitmap_id = ID_BITMAP_NONE;
+	else
+		mddev->bitmap_id = ID_BITMAP;
 
 	if (percpu_ref_init(&mddev->active_io, active_io_release,
 			    PERCPU_REF_ALLOW_REINIT, GFP_KERNEL))
@@ -713,7 +768,6 @@ int mddev_init(struct mddev *mddev)
 	mddev->resync_min = 0;
 	mddev->resync_max = MaxSector;
 	mddev->level = LEVEL_NONE;
-	mddev_set_bitmap_ops(mddev);
 
 	INIT_WORK(&mddev->sync_work, md_start_sync);
 	INIT_WORK(&mddev->del_work, mddev_delayed_delete);
@@ -1020,15 +1074,26 @@ static void super_written(struct bio *bio)
 		wake_up(&mddev->sb_wait);
 }
 
-void md_super_write(struct mddev *mddev, struct md_rdev *rdev,
-		   sector_t sector, int size, struct page *page)
+/**
+ * md_write_metadata - write metadata to underlying disk, including
+ * array superblock, badblocks, bitmap superblock and bitmap bits.
+ * @mddev:	the array to write
+ * @rdev:	the underlying disk to write
+ * @sector:	the offset to @rdev
+ * @size:	the length of the metadata
+ * @page:	the metadata
+ * @offset:	the offset to @page
+ *
+ * Write @size bytes of @page start from @offset, to @sector of @rdev, Increment
+ * mddev->pending_writes before returning, and decrement it on completion,
+ * waking up sb_wait. Caller must call md_super_wait() after issuing io to all
+ * rdev. If an error occurred, md_error() will be called, and the @rdev will be
+ * kicked out from @mddev.
+ */
+void md_write_metadata(struct mddev *mddev, struct md_rdev *rdev,
+		       sector_t sector, int size, struct page *page,
+		       unsigned int offset)
 {
-	/* write first size bytes of page to sector of rdev
-	 * Increment mddev->pending_writes before returning
-	 * and decrement it on completion, waking up sb_wait
-	 * if zero is reached.
-	 * If an error occurred, call md_error
-	 */
 	struct bio *bio;
 
 	if (!page)
@@ -1046,7 +1111,7 @@ void md_super_write(struct mddev *mddev, struct md_rdev *rdev,
 	atomic_inc(&rdev->nr_pending);
 
 	bio->bi_iter.bi_sector = sector;
-	__bio_add_page(bio, page, size, 0);
+	__bio_add_page(bio, page, size, offset);
 	bio->bi_private = rdev;
 	bio->bi_end_io = super_written;
 
@@ -1356,6 +1421,9 @@ static u64 md_bitmap_events_cleared(struct mddev *mddev)
 	struct md_bitmap_stats stats;
 	int err;
 
+	if (!md_bitmap_enabled(mddev, false))
+		return 0;
+
 	err = mddev->bitmap_ops->get_stats(mddev->bitmap, &stats);
 	if (err)
 		return 0;
@@ -1653,8 +1721,8 @@ super_90_rdev_size_change(struct md_rdev *rdev, sector_t num_sectors)
 	if ((u64)num_sectors >= (2ULL << 32) && rdev->mddev->level >= 1)
 		num_sectors = (sector_t)(2ULL << 32) - 2;
 	do {
-		md_super_write(rdev->mddev, rdev, rdev->sb_start, rdev->sb_size,
-		       rdev->sb_page);
+		md_write_metadata(rdev->mddev, rdev, rdev->sb_start,
+				  rdev->sb_size, rdev->sb_page, 0);
 	} while (md_super_wait(rdev->mddev) < 0);
 	return num_sectors;
 }
@@ -2302,8 +2370,8 @@ super_1_rdev_size_change(struct md_rdev *rdev, sector_t num_sectors)
 	sb->super_offset = cpu_to_le64(rdev->sb_start);
 	sb->sb_csum = calc_sb_1_csum(sb);
 	do {
-		md_super_write(rdev->mddev, rdev, rdev->sb_start, rdev->sb_size,
-			       rdev->sb_page);
+		md_write_metadata(rdev->mddev, rdev, rdev->sb_start,
+				  rdev->sb_size, rdev->sb_page, 0);
 	} while (md_super_wait(rdev->mddev) < 0);
 	return num_sectors;
 
@@ -2313,13 +2381,15 @@ static int
 super_1_allow_new_offset(struct md_rdev *rdev,
 			 unsigned long long new_offset)
 {
+	struct mddev *mddev = rdev->mddev;
+
 	/* All necessary checks on new >= old have been done */
 	if (new_offset >= rdev->data_offset)
 		return 1;
 
 	/* with 1.0 metadata, there is no metadata to tread on
 	 * so we can always move back */
-	if (rdev->mddev->minor_version == 0)
+	if (mddev->minor_version == 0)
 		return 1;
 
 	/* otherwise we must be sure not to step on
@@ -2331,8 +2401,7 @@ super_1_allow_new_offset(struct md_rdev *rdev,
 	if (rdev->sb_start + (32+4)*2 > new_offset)
 		return 0;
 
-	if (!rdev->mddev->bitmap_info.file) {
-		struct mddev *mddev = rdev->mddev;
+	if (md_bitmap_registered(mddev) && !mddev->bitmap_info.file) {
 		struct md_bitmap_stats stats;
 		int err;
 
@@ -2804,24 +2873,24 @@ repeat:
 
 	mddev_add_trace_msg(mddev, "md md_update_sb");
 rewrite:
-	mddev->bitmap_ops->update_sb(mddev->bitmap);
+	if (md_bitmap_enabled(mddev, false))
+		mddev->bitmap_ops->update_sb(mddev->bitmap);
 	rdev_for_each(rdev, mddev) {
 		if (rdev->sb_loaded != 1)
 			continue; /* no noise on spare devices */
 
 		if (!test_bit(Faulty, &rdev->flags)) {
-			md_super_write(mddev,rdev,
-				       rdev->sb_start, rdev->sb_size,
-				       rdev->sb_page);
+			md_write_metadata(mddev, rdev, rdev->sb_start,
+					  rdev->sb_size, rdev->sb_page, 0);
 			pr_debug("md: (write) %pg's sb offset: %llu\n",
 				 rdev->bdev,
 				 (unsigned long long)rdev->sb_start);
 			rdev->sb_events = mddev->events;
 			if (rdev->badblocks.size) {
-				md_super_write(mddev, rdev,
-					       rdev->badblocks.sector,
-					       rdev->badblocks.size << 9,
-					       rdev->bb_page);
+				md_write_metadata(mddev, rdev,
+						  rdev->badblocks.sector,
+						  rdev->badblocks.size << 9,
+						  rdev->bb_page, 0);
 				rdev->badblocks.size = 0;
 			}
 
@@ -4149,6 +4218,86 @@ new_level_store(struct mddev *mddev, const char *buf, size_t len)
 static struct md_sysfs_entry md_new_level =
 __ATTR(new_level, 0664, new_level_show, new_level_store);
 
+static ssize_t
+bitmap_type_show(struct mddev *mddev, char *page)
+{
+	struct md_submodule_head *head;
+	unsigned long i;
+	ssize_t len = 0;
+
+	if (mddev->bitmap_id == ID_BITMAP_NONE)
+		len += sprintf(page + len, "[none] ");
+	else
+		len += sprintf(page + len, "none ");
+
+	xa_lock(&md_submodule);
+	xa_for_each(&md_submodule, i, head) {
+		if (head->type != MD_BITMAP)
+			continue;
+
+		if (mddev->bitmap_id == head->id)
+			len += sprintf(page + len, "[%s] ", head->name);
+		else
+			len += sprintf(page + len, "%s ", head->name);
+	}
+	xa_unlock(&md_submodule);
+
+	len += sprintf(page + len, "\n");
+	return len;
+}
+
+static ssize_t
+bitmap_type_store(struct mddev *mddev, const char *buf, size_t len)
+{
+	struct md_submodule_head *head;
+	enum md_submodule_id id;
+	unsigned long i;
+	int err = 0;
+
+	xa_lock(&md_submodule);
+
+	if (mddev->bitmap_ops) {
+		err = -EBUSY;
+		goto out;
+	}
+
+	if (cmd_match(buf, "none")) {
+		mddev->bitmap_id = ID_BITMAP_NONE;
+		goto out;
+	}
+
+	xa_for_each(&md_submodule, i, head) {
+		if (head->type == MD_BITMAP && cmd_match(buf, head->name)) {
+			mddev->bitmap_id = head->id;
+			goto out;
+		}
+	}
+
+	err = kstrtoint(buf, 10, &id);
+	if (err)
+		goto out;
+
+	if (id == ID_BITMAP_NONE) {
+		mddev->bitmap_id = id;
+		goto out;
+	}
+
+	head = xa_load(&md_submodule, id);
+	if (head && head->type == MD_BITMAP) {
+		mddev->bitmap_id = id;
+		goto out;
+	}
+
+	err = -ENOENT;
+
+out:
+	xa_unlock(&md_submodule);
+	return err ? err : len;
+}
+
+static struct md_sysfs_entry md_bitmap_type =
+__ATTR(bitmap_type, 0664, bitmap_type_show, bitmap_type_store);
+
 static ssize_t
 layout_show(struct mddev *mddev, char *page)
 {
@@ -4680,6 +4829,9 @@ bitmap_store(struct mddev *mddev, const char *buf, size_t len)
 	unsigned long chunk, end_chunk;
 	int err;
 
+	if (!md_bitmap_enabled(mddev, false))
+		return len;
+
 	err = mddev_lock(mddev);
 	if (err)
 		return err;
@@ -5752,6 +5904,7 @@ __ATTR(serialize_policy, S_IRUGO | S_IWUSR, serialize_policy_show,
 static struct attribute *md_default_attrs[] = {
 	&md_level.attr,
 	&md_new_level.attr,
+	&md_bitmap_type.attr,
 	&md_layout.attr,
 	&md_raid_disks.attr,
 	&md_uuid.attr,
@@ -5801,7 +5954,6 @@ static const struct attribute_group md_redundancy_group = {
 
 static const struct attribute_group *md_attr_groups[] = {
 	&md_default_group,
-	&md_bitmap_group,
 	NULL,
 };
 
@@ -6133,6 +6285,26 @@ static void md_safemode_timeout(struct timer_list *t)
 
 static int start_dirty_degraded;
 
+static int md_bitmap_create(struct mddev *mddev)
+{
+	if (mddev->bitmap_id == ID_BITMAP_NONE)
+		return -EINVAL;
+
+	if (!mddev_set_bitmap_ops(mddev))
+		return -ENOENT;
+
+	return mddev->bitmap_ops->create(mddev);
+}
+
+static void md_bitmap_destroy(struct mddev *mddev)
+{
+	if (!md_bitmap_registered(mddev))
+		return;
+
+	mddev->bitmap_ops->destroy(mddev);
+	mddev_clear_bitmap_ops(mddev);
+}
+
 int md_run(struct mddev *mddev)
 {
 	int err;
@@ -6299,7 +6471,7 @@ int md_run(struct mddev *mddev)
 	}
 	if (err == 0 && pers->sync_request &&
 	    (mddev->bitmap_info.file || mddev->bitmap_info.offset)) {
-		err = mddev->bitmap_ops->create(mddev);
+		err = md_bitmap_create(mddev);
 		if (err)
 			pr_warn("%s: failed to create bitmap (%d)\n",
 				mdname(mddev), err);
@@ -6372,7 +6544,7 @@ bitmap_abort:
 		pers->free(mddev, mddev->private);
 	mddev->private = NULL;
 	put_pers(pers);
-	mddev->bitmap_ops->destroy(mddev);
+	md_bitmap_destroy(mddev);
 abort:
 	bioset_exit(&mddev->io_clone_set);
 exit_sync_set:
@@ -6392,10 +6564,12 @@ int do_md_run(struct mddev *mddev)
 	if (err)
 		goto out;
 
-	err = mddev->bitmap_ops->load(mddev);
-	if (err) {
-		mddev->bitmap_ops->destroy(mddev);
-		goto out;
+	if (md_bitmap_registered(mddev)) {
+		err = mddev->bitmap_ops->load(mddev);
+		if (err) {
+			md_bitmap_destroy(mddev);
+			goto out;
+		}
 	}
 
 	if (mddev_is_clustered(mddev))
@@ -6546,7 +6720,8 @@ static void __md_stop_writes(struct mddev *mddev)
 		mddev->pers->quiesce(mddev, 0);
 	}
 
-	mddev->bitmap_ops->flush(mddev);
+	if (md_bitmap_enabled(mddev, true))
+		mddev->bitmap_ops->flush(mddev);
 
 	if (md_is_rdwr(mddev) &&
 	    ((!mddev->in_sync && !mddev_is_clustered(mddev)) ||
@@ -6573,7 +6748,8 @@ EXPORT_SYMBOL_GPL(md_stop_writes);
 
 static void mddev_detach(struct mddev *mddev)
 {
-	mddev->bitmap_ops->wait_behind_writes(mddev);
+	if (md_bitmap_enabled(mddev, false))
+		mddev->bitmap_ops->wait_behind_writes(mddev);
 	if (mddev->pers && mddev->pers->quiesce && !is_md_suspended(mddev)) {
 		mddev->pers->quiesce(mddev, 1);
 		mddev->pers->quiesce(mddev, 0);
@@ -6589,7 +6765,7 @@ static void __md_stop(struct mddev *mddev)
 {
 	struct md_personality *pers = mddev->pers;
 
-	mddev->bitmap_ops->destroy(mddev);
+	md_bitmap_destroy(mddev);
 	mddev_detach(mddev);
 	spin_lock(&mddev->lock);
 	mddev->pers = NULL;
@@ -7307,6 +7483,9 @@ static int set_bitmap_file(struct mddev *mddev, int fd)
 {
 	int err = 0;
 
+	if (!md_bitmap_registered(mddev))
+		return -EINVAL;
+
 	if (mddev->pers) {
 		if (!mddev->pers->quiesce || !mddev->thread)
 			return -EBUSY;
@@ -7363,16 +7542,16 @@ static int set_bitmap_file(struct mddev *mddev, int fd)
 	err = 0;
 	if (mddev->pers) {
 		if (fd >= 0) {
-			err = mddev->bitmap_ops->create(mddev);
+			err = md_bitmap_create(mddev);
 			if (!err)
 				err = mddev->bitmap_ops->load(mddev);
 
 			if (err) {
-				mddev->bitmap_ops->destroy(mddev);
+				md_bitmap_destroy(mddev);
 				fd = -1;
 			}
 		} else if (fd < 0) {
-			mddev->bitmap_ops->destroy(mddev);
+			md_bitmap_destroy(mddev);
 		}
 	}
 
@@ -7679,12 +7858,12 @@ static int update_array_info(struct mddev *mddev, mdu_array_info_t *info)
 				mddev->bitmap_info.default_offset;
 			mddev->bitmap_info.space =
 				mddev->bitmap_info.default_space;
-			rv = mddev->bitmap_ops->create(mddev);
+			rv = md_bitmap_create(mddev);
 			if (!rv)
 				rv = mddev->bitmap_ops->load(mddev);
 
 			if (rv)
-				mddev->bitmap_ops->destroy(mddev);
+				md_bitmap_destroy(mddev);
 		} else {
 			struct md_bitmap_stats stats;
 
@@ -7710,7 +7889,7 @@ static int update_array_info(struct mddev *mddev, mdu_array_info_t *info)
 				put_cluster_ops(mddev);
 				mddev->safemode_delay = DEFAULT_SAFEMODE_DELAY;
 			}
-			mddev->bitmap_ops->destroy(mddev);
+			md_bitmap_destroy(mddev);
 			mddev->bitmap_info.offset = 0;
 		}
 	}
@@ -8491,6 +8670,9 @@ static void md_bitmap_status(struct seq_file *seq, struct mddev *mddev)
 	unsigned long chunk_kb;
 	int err;
 
+	if (!md_bitmap_enabled(mddev, false))
+		return;
+
 	err = mddev->bitmap_ops->get_stats(mddev->bitmap, &stats);
 	if (err)
 		return;
@@ -8873,18 +9055,24 @@ EXPORT_SYMBOL_GPL(md_submit_discard_bio);
 static void md_bitmap_start(struct mddev *mddev,
 			    struct md_io_clone *md_io_clone)
 {
+	md_bitmap_fn *fn = unlikely(md_io_clone->rw == STAT_DISCARD) ?
+			   mddev->bitmap_ops->start_discard :
+			   mddev->bitmap_ops->start_write;
+
 	if (mddev->pers->bitmap_sector)
 		mddev->pers->bitmap_sector(mddev, &md_io_clone->offset,
 					   &md_io_clone->sectors);
 
-	mddev->bitmap_ops->start_write(mddev, md_io_clone->offset,
-				       md_io_clone->sectors);
+	fn(mddev, md_io_clone->offset, md_io_clone->sectors);
 }
 
 static void md_bitmap_end(struct mddev *mddev, struct md_io_clone *md_io_clone)
 {
-	mddev->bitmap_ops->end_write(mddev, md_io_clone->offset,
-				     md_io_clone->sectors);
+	md_bitmap_fn *fn = unlikely(md_io_clone->rw == STAT_DISCARD) ?
+			   mddev->bitmap_ops->end_discard :
+			   mddev->bitmap_ops->end_write;
+
+	fn(mddev, md_io_clone->offset, md_io_clone->sectors);
 }
 
 static void md_end_clone_io(struct bio *bio)
@@ -8893,7 +9081,7 @@ static void md_end_clone_io(struct bio *bio)
 	struct bio *orig_bio = md_io_clone->orig_bio;
 	struct mddev *mddev = md_io_clone->mddev;
 
-	if (bio_data_dir(orig_bio) == WRITE && mddev->bitmap)
+	if (bio_data_dir(orig_bio) == WRITE && md_bitmap_enabled(mddev, false))
 		md_bitmap_end(mddev, md_io_clone);
 
 	if (bio->bi_status && !orig_bio->bi_status)
@@ -8920,9 +9108,10 @@ static void md_clone_bio(struct mddev *mddev, struct bio **bio)
 	if (blk_queue_io_stat(bdev->bd_disk->queue))
 		md_io_clone->start_time = bio_start_io_acct(*bio);
 
-	if (bio_data_dir(*bio) == WRITE && mddev->bitmap) {
+	if (bio_data_dir(*bio) == WRITE && md_bitmap_enabled(mddev, false)) {
 		md_io_clone->offset = (*bio)->bi_iter.bi_sector;
 		md_io_clone->sectors = bio_sectors(*bio);
+		md_io_clone->rw = op_stat_group(bio_op(*bio));
 		md_bitmap_start(mddev, md_io_clone);
 	}
 
@@ -8944,7 +9133,7 @@ void md_free_cloned_bio(struct bio *bio)
 	struct bio *orig_bio = md_io_clone->orig_bio;
 	struct mddev *mddev = md_io_clone->mddev;
 
-	if (bio_data_dir(orig_bio) == WRITE && mddev->bitmap)
+	if (bio_data_dir(orig_bio) == WRITE && md_bitmap_enabled(mddev, false))
 		md_bitmap_end(mddev, md_io_clone);
 
 	if (bio->bi_status && !orig_bio->bi_status)
@@ -9010,6 +9199,39 @@ static sector_t md_sync_max_sectors(struct mddev *mddev,
 	}
 }
 
+/*
+ * If lazy recovery is requested and all rdevs are in sync, select the rdev with
+ * the higest index to perfore recovery to build initial xor data, this is the
+ * same as old bitmap.
+ */
+static bool mddev_select_lazy_recover_rdev(struct mddev *mddev)
+{
+	struct md_rdev *recover_rdev = NULL;
+	struct md_rdev *rdev;
+	bool ret = false;
+
+	rcu_read_lock();
+	rdev_for_each_rcu(rdev, mddev) {
+		if (rdev->raid_disk < 0)
+			continue;
+
+		if (test_bit(Faulty, &rdev->flags) ||
+		    !test_bit(In_sync, &rdev->flags))
+			break;
+
+		if (!recover_rdev || recover_rdev->raid_disk < rdev->raid_disk)
+			recover_rdev = rdev;
+	}
+
+	if (recover_rdev) {
+		clear_bit(In_sync, &recover_rdev->flags);
+		ret = true;
+	}
+
+	rcu_read_unlock();
+	return ret;
+}
+
 static sector_t md_sync_position(struct mddev *mddev, enum sync_action action)
 {
 	sector_t start = 0;
@@ -9041,6 +9263,14 @@ static sector_t md_sync_position(struct mddev *mddev, enum sync_action action)
 				start = rdev->recovery_offset;
 		rcu_read_unlock();
 
+		/*
+		 * If there are no spares, and raid456 lazy initial recover is
+		 * requested.
+		 */
+		if (test_bit(MD_RECOVERY_LAZY_RECOVER, &mddev->recovery) &&
+		    start == MaxSector && mddev_select_lazy_recover_rdev(mddev))
+			start = 0;
+
 		/* If there is a bitmap, we need to make sure all
 		 * writes that started before we added a spare
 		 * complete before we start doing a recovery.
@@ -9061,19 +9291,12 @@ static sector_t md_sync_position(struct mddev *mddev, enum sync_action action)
 
 static bool sync_io_within_limit(struct mddev *mddev)
 {
-	int io_sectors;
-
 	/*
 	 * For raid456, sync IO is stripe(4k) per IO, for other levels, it's
 	 * RESYNC_PAGES(64k) per IO.
 	 */
-	if (mddev->level == 4 || mddev->level == 5 || mddev->level == 6)
-		io_sectors = 8;
-	else
-		io_sectors = 128;
-
 	return atomic_read(&mddev->recovery_active) <
-		io_sectors * sync_io_depth(mddev);
+	       (raid_is_456(mddev) ? 8 : 128) * sync_io_depth(mddev);
 }
 
 #define SYNC_MARKS	10
@@ -9278,6 +9501,12 @@ void md_do_sync(struct md_thread *thread)
 		if (test_bit(MD_RECOVERY_INTR, &mddev->recovery))
 			break;
 
+		if (mddev->bitmap_ops && mddev->bitmap_ops->skip_sync_blocks) {
+			sectors = mddev->bitmap_ops->skip_sync_blocks(mddev, j);
+			if (sectors)
+				goto update;
+		}
+
 		sectors = mddev->pers->sync_request(mddev, j, max_sectors,
 						    &skipped);
 		if (sectors == 0) {
@@ -9293,6 +9522,7 @@ void md_do_sync(struct md_thread *thread)
 		if (test_bit(MD_RECOVERY_INTR, &mddev->recovery))
 			break;
 
+update:
 		j += sectors;
 		if (j > max_sectors)
 			/* when skipping, extra large numbers can be returned. */
@@ -9602,6 +9832,7 @@ static bool md_choose_sync_action(struct mddev *mddev, int *spares)
 
 		set_bit(MD_RECOVERY_RESHAPE, &mddev->recovery);
 		clear_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
+		clear_bit(MD_RECOVERY_LAZY_RECOVER, &mddev->recovery);
 		return true;
 	}
 
@@ -9610,6 +9841,7 @@ static bool md_choose_sync_action(struct mddev *mddev, int *spares)
 		remove_spares(mddev, NULL);
 		set_bit(MD_RECOVERY_SYNC, &mddev->recovery);
 		clear_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
+		clear_bit(MD_RECOVERY_LAZY_RECOVER, &mddev->recovery);
 		return true;
 	}
 
@@ -9619,7 +9851,7 @@ static bool md_choose_sync_action(struct mddev *mddev, int *spares)
 	 * re-add.
 	 */
 	*spares = remove_and_add_spares(mddev, NULL);
-	if (*spares) {
+	if (*spares || test_bit(MD_RECOVERY_LAZY_RECOVER, &mddev->recovery)) {
 		clear_bit(MD_RECOVERY_SYNC, &mddev->recovery);
 		clear_bit(MD_RECOVERY_CHECK, &mddev->recovery);
 		clear_bit(MD_RECOVERY_REQUESTED, &mddev->recovery);
@@ -9677,7 +9909,7 @@ static void md_start_sync(struct work_struct *ws)
 	 * We are adding a device or devices to an array which has the bitmap
 	 * stored on all devices. So make sure all bitmap pages get written.
 	 */
-	if (spares)
+	if (spares && md_bitmap_enabled(mddev, true))
 		mddev->bitmap_ops->write_all(mddev);
 
 	name = test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery) ?
@@ -9765,7 +9997,7 @@ static void unregister_sync_thread(struct mddev *mddev)
  */
 void md_check_recovery(struct mddev *mddev)
 {
-	if (mddev->bitmap)
+	if (md_bitmap_enabled(mddev, false) && mddev->bitmap_ops->daemon_work)
 		mddev->bitmap_ops->daemon_work(mddev);
 
 	if (signal_pending(current)) {
@@ -9832,6 +10064,7 @@ void md_check_recovery(struct mddev *mddev)
 			}
 
 			clear_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
+			clear_bit(MD_RECOVERY_LAZY_RECOVER, &mddev->recovery);
 			clear_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 			clear_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags);
 
@@ -9942,6 +10175,7 @@ void md_reap_sync_thread(struct mddev *mddev)
 	clear_bit(MD_RECOVERY_RESHAPE, &mddev->recovery);
 	clear_bit(MD_RECOVERY_REQUESTED, &mddev->recovery);
 	clear_bit(MD_RECOVERY_CHECK, &mddev->recovery);
+	clear_bit(MD_RECOVERY_LAZY_RECOVER, &mddev->recovery);
 	/*
 	 * We call mddev->cluster_ops->update_size here because sync_size could
 	 * be changed by md_update_sb, and MD_RECOVERY_RESHAPE is cleared,
@@ -10089,8 +10323,16 @@ static void md_geninit(void)
 
 static int __init md_init(void)
 {
-	int ret = -ENOMEM;
+	int ret = md_bitmap_init();
 
+	if (ret)
+		return ret;
+
+	ret = md_llbitmap_init();
+	if (ret)
+		goto err_bitmap;
+
+	ret = -ENOMEM;
 	md_wq = alloc_workqueue("md", WQ_MEM_RECLAIM, 0);
 	if (!md_wq)
 		goto err_wq;
@@ -10099,11 +10341,6 @@ static int __init md_init(void)
 	if (!md_misc_wq)
 		goto err_misc_wq;
 
-	md_bitmap_wq = alloc_workqueue("md_bitmap", WQ_MEM_RECLAIM | WQ_UNBOUND,
-				       0);
-	if (!md_bitmap_wq)
-		goto err_bitmap_wq;
-
 	ret = __register_blkdev(MD_MAJOR, "md", md_probe);
 	if (ret < 0)
 		goto err_md;
@@ -10122,12 +10359,13 @@ static int __init md_init(void)
 err_mdp:
 	unregister_blkdev(MD_MAJOR, "md");
 err_md:
-	destroy_workqueue(md_bitmap_wq);
-err_bitmap_wq:
 	destroy_workqueue(md_misc_wq);
 err_misc_wq:
 	destroy_workqueue(md_wq);
 err_wq:
+	md_llbitmap_exit();
+err_bitmap:
+	md_bitmap_exit();
 	return ret;
 }
 
@@ -10145,7 +10383,7 @@ static void check_sb_changes(struct mddev *mddev, struct md_rdev *rdev)
 		ret = mddev->pers->resize(mddev, le64_to_cpu(sb->size));
 		if (ret)
 			pr_info("md-cluster: resize failed\n");
-		else
+		else if (md_bitmap_enabled(mddev, false))
 			mddev->bitmap_ops->update_sb(mddev->bitmap);
 	}
 
@@ -10433,8 +10671,8 @@ static __exit void md_exit(void)
 	spin_unlock(&all_mddevs_lock);
 
 	destroy_workqueue(md_misc_wq);
-	destroy_workqueue(md_bitmap_wq);
 	destroy_workqueue(md_wq);
+	md_bitmap_exit();
 }
 
 subsys_initcall(md_init);

drivers/md/md.h

@@ -26,7 +26,7 @@
 enum md_submodule_type {
 	MD_PERSONALITY = 0,
 	MD_CLUSTER,
-	MD_BITMAP, /* TODO */
+	MD_BITMAP,
 };
 
 enum md_submodule_id {
@@ -38,8 +38,9 @@ enum md_submodule_id {
 	ID_RAID6	= 6,
 	ID_RAID10	= 10,
 	ID_CLUSTER,
-	ID_BITMAP,	/* TODO */
-	ID_LLBITMAP,	/* TODO */
+	ID_BITMAP,
+	ID_LLBITMAP,
+	ID_BITMAP_NONE,
 };
 
 struct md_submodule_head {
@@ -565,6 +566,7 @@ struct mddev {
 	struct percpu_ref		writes_pending;
 	int				sync_checkers;	/* # of threads checking writes_pending */
 
+	enum md_submodule_id		bitmap_id;
 	void				*bitmap; /* the bitmap for the device */
 	struct bitmap_operations	*bitmap_ops;
 	struct {
@@ -665,6 +667,8 @@ enum recovery_flags {
 	MD_RECOVERY_RESHAPE,
 	/* remote node is running resync thread */
 	MD_RESYNCING_REMOTE,
+	/* raid456 lazy initial recover */
+	MD_RECOVERY_LAZY_RECOVER,
 };
 
 enum md_ro_state {
@@ -796,7 +800,6 @@ struct md_sysfs_entry {
 	ssize_t (*show)(struct mddev *, char *);
 	ssize_t (*store)(struct mddev *, const char *, size_t);
 };
-extern const struct attribute_group md_bitmap_group;
 
 static inline struct kernfs_node *sysfs_get_dirent_safe(struct kernfs_node *sd, char *name)
 {
@@ -873,6 +876,7 @@ struct md_io_clone {
 	unsigned long	start_time;
 	sector_t	offset;
 	unsigned long	sectors;
+	enum stat_group	rw;
 	struct bio	bio_clone;
 };
 
@@ -909,8 +913,9 @@ void md_account_bio(struct mddev *mddev, struct bio **bio);
 void md_free_cloned_bio(struct bio *bio);
 
 extern bool __must_check md_flush_request(struct mddev *mddev, struct bio *bio);
-extern void md_super_write(struct mddev *mddev, struct md_rdev *rdev,
-			   sector_t sector, int size, struct page *page);
+void md_write_metadata(struct mddev *mddev, struct md_rdev *rdev,
+		       sector_t sector, int size, struct page *page,
+		       unsigned int offset);
 extern int md_super_wait(struct mddev *mddev);
 extern int sync_page_io(struct md_rdev *rdev, sector_t sector, int size,
 		struct page *page, blk_opf_t opf, bool metadata_op);
@@ -1013,7 +1018,6 @@ struct mdu_array_info_s;
 struct mdu_disk_info_s;
 
 extern int mdp_major;
-extern struct workqueue_struct *md_bitmap_wq;
 void md_autostart_arrays(int part);
 int md_set_array_info(struct mddev *mddev, struct mdu_array_info_s *info);
 int md_add_new_disk(struct mddev *mddev, struct mdu_disk_info_s *info);
@@ -1034,6 +1038,12 @@ static inline bool mddev_is_dm(struct mddev *mddev)
 	return !mddev->gendisk;
 }
 
+static inline bool raid_is_456(struct mddev *mddev)
+{
+	return mddev->level == ID_RAID4 || mddev->level == ID_RAID5 ||
+	       mddev->level == ID_RAID6;
+}
+
 static inline void mddev_trace_remap(struct mddev *mddev, struct bio *bio,
 		sector_t sector)
 {

drivers/md/raid1-10.c

@@ -140,7 +140,7 @@ static inline bool raid1_add_bio_to_plug(struct mddev *mddev, struct bio *bio,
 	 * If bitmap is not enabled, it's safe to submit the io directly, and
 	 * this can get optimal performance.
 	 */
-	if (!mddev->bitmap_ops->enabled(mddev)) {
+	if (!md_bitmap_enabled(mddev, true)) {
 		raid1_submit_write(bio);
 		return true;
 	}

drivers/md/raid1.c

@@ -1366,7 +1366,8 @@ static void raid1_read_request(struct mddev *mddev, struct bio *bio,
 				    (unsigned long long)r1_bio->sector,
 				    mirror->rdev->bdev);
 
-	if (test_bit(WriteMostly, &mirror->rdev->flags)) {
+	if (test_bit(WriteMostly, &mirror->rdev->flags) &&
+	    md_bitmap_enabled(mddev, false)) {
 		/*
 		 * Reading from a write-mostly device must take care not to
 		 * over-take any writes that are 'behind'
@@ -1452,6 +1453,30 @@ retry:
 	return true;
 }
 
+static void raid1_start_write_behind(struct mddev *mddev, struct r1bio *r1_bio,
+				     struct bio *bio)
+{
+	unsigned long max_write_behind = mddev->bitmap_info.max_write_behind;
+	struct md_bitmap_stats stats;
+	int err;
+
+	/* behind write rely on bitmap, see bitmap_operations */
+	if (!md_bitmap_enabled(mddev, false))
+		return;
+
+	err = mddev->bitmap_ops->get_stats(mddev->bitmap, &stats);
+	if (err)
+		return;
+
+	/* Don't do behind IO if reader is waiting, or there are too many. */
+	if (!stats.behind_wait && stats.behind_writes < max_write_behind)
+		alloc_behind_master_bio(r1_bio, bio);
+
+	if (test_bit(R1BIO_BehindIO, &r1_bio->state))
+		mddev->bitmap_ops->start_behind_write(mddev);
+
+}
+
 static void raid1_write_request(struct mddev *mddev, struct bio *bio,
 				int max_write_sectors)
 {
@@ -1612,22 +1637,8 @@ static void raid1_write_request(struct mddev *mddev, struct bio *bio,
 			continue;
 
 		if (first_clone) {
-			unsigned long max_write_behind =
-				mddev->bitmap_info.max_write_behind;
-			struct md_bitmap_stats stats;
-			int err;
-
-			/* do behind I/O ?
-			 * Not if there are too many, or cannot
-			 * allocate memory, or a reader on WriteMostly
-			 * is waiting for behind writes to flush */
-			err = mddev->bitmap_ops->get_stats(mddev->bitmap, &stats);
-			if (!err && write_behind && !stats.behind_wait &&
-			    stats.behind_writes < max_write_behind)
-				alloc_behind_master_bio(r1_bio, bio);
-
-			if (test_bit(R1BIO_BehindIO, &r1_bio->state))
-				mddev->bitmap_ops->start_behind_write(mddev);
+			if (write_behind)
+				raid1_start_write_behind(mddev, r1_bio, bio);
 			first_clone = 0;
 		}
 
@@ -2057,7 +2068,7 @@ static void abort_sync_write(struct mddev *mddev, struct r1bio *r1_bio)
 
 	/* make sure these bits don't get cleared. */
 	do {
-		mddev->bitmap_ops->end_sync(mddev, s, &sync_blocks);
+		md_bitmap_end_sync(mddev, s, &sync_blocks);
 		s += sync_blocks;
 		sectors_to_go -= sync_blocks;
 	} while (sectors_to_go > 0);
@@ -2804,12 +2815,13 @@ static sector_t raid1_sync_request(struct mddev *mddev, sector_t sector_nr,
 		 * We can find the current addess in mddev->curr_resync
 		 */
 		if (mddev->curr_resync < max_sector) /* aborted */
-			mddev->bitmap_ops->end_sync(mddev, mddev->curr_resync,
-						    &sync_blocks);
+			md_bitmap_end_sync(mddev, mddev->curr_resync,
+					   &sync_blocks);
 		else /* completed sync */
 			conf->fullsync = 0;
 
-		mddev->bitmap_ops->close_sync(mddev);
+		if (md_bitmap_enabled(mddev, false))
+			mddev->bitmap_ops->close_sync(mddev);
 		close_sync(conf);
 
 		if (mddev_is_clustered(mddev)) {
@@ -2829,7 +2841,7 @@ static sector_t raid1_sync_request(struct mddev *mddev, sector_t sector_nr,
 	/* before building a request, check if we can skip these blocks..
 	 * This call the bitmap_start_sync doesn't actually record anything
 	 */
-	if (!mddev->bitmap_ops->start_sync(mddev, sector_nr, &sync_blocks, true) &&
+	if (!md_bitmap_start_sync(mddev, sector_nr, &sync_blocks, true) &&
 	    !conf->fullsync && !test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery)) {
 		/* We can skip this block, and probably several more */
 		*skipped = 1;
@@ -2846,10 +2858,11 @@ static sector_t raid1_sync_request(struct mddev *mddev, sector_t sector_nr,
 	/* we are incrementing sector_nr below. To be safe, we check against
 	 * sector_nr + two times RESYNC_SECTORS
 	 */
-
-	mddev->bitmap_ops->cond_end_sync(mddev, sector_nr,
-		mddev_is_clustered(mddev) &&
-		(sector_nr + 2 * RESYNC_SECTORS > conf->cluster_sync_high));
+	if (md_bitmap_enabled(mddev, false))
+		mddev->bitmap_ops->cond_end_sync(mddev, sector_nr,
+			mddev_is_clustered(mddev) &&
+			(sector_nr + 2 * RESYNC_SECTORS >
+			 conf->cluster_sync_high));
 
 	if (raise_barrier(conf, sector_nr))
 		return 0;
@@ -3004,8 +3017,8 @@ static sector_t raid1_sync_request(struct mddev *mddev, sector_t sector_nr,
 		if (len == 0)
 			break;
 		if (sync_blocks == 0) {
-			if (!mddev->bitmap_ops->start_sync(mddev, sector_nr,
-						&sync_blocks, still_degraded) &&
+			if (!md_bitmap_start_sync(mddev, sector_nr,
+						  &sync_blocks, still_degraded) &&
 			    !conf->fullsync &&
 			    !test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery))
 				break;
@@ -3324,15 +3337,17 @@ static int raid1_resize(struct mddev *mddev, sector_t sectors)
 	 * worth it.
 	 */
 	sector_t newsize = raid1_size(mddev, sectors, 0);
-	int ret;
 
 	if (mddev->external_size &&
 	    mddev->array_sectors > newsize)
 		return -EINVAL;
 
-	ret = mddev->bitmap_ops->resize(mddev, newsize, 0, false);
-	if (ret)
-		return ret;
+	if (md_bitmap_enabled(mddev, false)) {
+		int ret = mddev->bitmap_ops->resize(mddev, newsize, 0);
+
+		if (ret)
+			return ret;
+	}
 
 	md_set_array_sectors(mddev, newsize);
 	if (sectors > mddev->dev_sectors &&

drivers/md/raid10.c

@@ -3221,15 +3221,13 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr,
 
 		if (mddev->curr_resync < max_sector) { /* aborted */
 			if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery))
-				mddev->bitmap_ops->end_sync(mddev,
-							    mddev->curr_resync,
-							    &sync_blocks);
+				md_bitmap_end_sync(mddev, mddev->curr_resync,
+						   &sync_blocks);
 			else for (i = 0; i < conf->geo.raid_disks; i++) {
 				sector_t sect =
 					raid10_find_virt(conf, mddev->curr_resync, i);
 
-				mddev->bitmap_ops->end_sync(mddev, sect,
-							    &sync_blocks);
+				md_bitmap_end_sync(mddev, sect, &sync_blocks);
 			}
 		} else {
 			/* completed sync */
@@ -3249,7 +3247,8 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr,
 			}
 			conf->fullsync = 0;
 		}
-		mddev->bitmap_ops->close_sync(mddev);
+		if (md_bitmap_enabled(mddev, false))
+			mddev->bitmap_ops->close_sync(mddev);
 		close_sync(conf);
 		*skipped = 1;
 		return sectors_skipped;
@@ -3351,9 +3350,8 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr,
 			 * we only need to recover the block if it is set in
 			 * the bitmap
 			 */
-			must_sync = mddev->bitmap_ops->start_sync(mddev, sect,
-								  &sync_blocks,
-								  true);
+			must_sync = md_bitmap_start_sync(mddev, sect,
+							 &sync_blocks, true);
 			if (sync_blocks < max_sync)
 				max_sync = sync_blocks;
 			if (!must_sync &&
@@ -3396,9 +3394,8 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr,
 				}
 			}
 
-			must_sync = mddev->bitmap_ops->start_sync(mddev, sect,
-						&sync_blocks, still_degraded);
-
+			md_bitmap_start_sync(mddev, sect, &sync_blocks,
+					     still_degraded);
 			any_working = 0;
 			for (j=0; j<conf->copies;j++) {
 				int k;
@@ -3570,13 +3567,13 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr,
 		 * safety reason, which ensures curr_resync_completed is
 		 * updated in bitmap_cond_end_sync.
 		 */
-		mddev->bitmap_ops->cond_end_sync(mddev, sector_nr,
+		if (md_bitmap_enabled(mddev, false))
+			mddev->bitmap_ops->cond_end_sync(mddev, sector_nr,
 					mddev_is_clustered(mddev) &&
 					(sector_nr + 2 * RESYNC_SECTORS > conf->cluster_sync_high));
 
-		if (!mddev->bitmap_ops->start_sync(mddev, sector_nr,
-						   &sync_blocks,
-						   mddev->degraded) &&
+		if (!md_bitmap_start_sync(mddev, sector_nr, &sync_blocks,
+					  mddev->degraded) &&
 		    !conf->fullsync && !test_bit(MD_RECOVERY_REQUESTED,
 						 &mddev->recovery)) {
 			/* We can skip this block */
@@ -4225,7 +4222,6 @@ static int raid10_resize(struct mddev *mddev, sector_t sectors)
 	 */
 	struct r10conf *conf = mddev->private;
 	sector_t oldsize, size;
-	int ret;
 
 	if (mddev->reshape_position != MaxSector)
 		return -EBUSY;
@@ -4239,9 +4235,12 @@ static int raid10_resize(struct mddev *mddev, sector_t sectors)
 	    mddev->array_sectors > size)
 		return -EINVAL;
 
-	ret = mddev->bitmap_ops->resize(mddev, size, 0, false);
-	if (ret)
-		return ret;
+	if (md_bitmap_enabled(mddev, false)) {
+		int ret = mddev->bitmap_ops->resize(mddev, size, 0);
+
+		if (ret)
+			return ret;
+	}
 
 	md_set_array_sectors(mddev, size);
 	if (sectors > mddev->dev_sectors &&
@@ -4507,8 +4506,9 @@ static int raid10_start_reshape(struct mddev *mddev)
 		oldsize = raid10_size(mddev, 0, 0);
 		newsize = raid10_size(mddev, 0, conf->geo.raid_disks);
 
-		if (!mddev_is_clustered(mddev)) {
-			ret = mddev->bitmap_ops->resize(mddev, newsize, 0, false);
+		if (!mddev_is_clustered(mddev) &&
+		    md_bitmap_enabled(mddev, false)) {
+			ret = mddev->bitmap_ops->resize(mddev, newsize, 0);
 			if (ret)
 				goto abort;
 			else
@@ -4530,13 +4530,14 @@ static int raid10_start_reshape(struct mddev *mddev)
 			    MD_FEATURE_RESHAPE_ACTIVE)) || (oldsize == newsize))
 			goto out;
 
-		ret = mddev->bitmap_ops->resize(mddev, newsize, 0, false);
+		/* cluster can't be setup without bitmap */
+		ret = mddev->bitmap_ops->resize(mddev, newsize, 0);
 		if (ret)
 			goto abort;
 
 		ret = mddev->cluster_ops->resize_bitmaps(mddev, newsize, oldsize);
 		if (ret) {
-			mddev->bitmap_ops->resize(mddev, oldsize, 0, false);
+			mddev->bitmap_ops->resize(mddev, oldsize, 0);
 			goto abort;
 		}
 	}

drivers/md/raid5.c

@@ -4097,7 +4097,8 @@ static int handle_stripe_dirtying(struct r5conf *conf,
 				  int disks)
 {
 	int rmw = 0, rcw = 0, i;
-	sector_t resync_offset = conf->mddev->resync_offset;
+	struct mddev *mddev = conf->mddev;
+	sector_t resync_offset = mddev->resync_offset;
 
 	/* Check whether resync is now happening or should start.
 	 * If yes, then the array is dirty (after unclean shutdown or
@@ -4116,6 +4117,12 @@ static int handle_stripe_dirtying(struct r5conf *conf,
 		pr_debug("force RCW rmw_level=%u, resync_offset=%llu sh->sector=%llu\n",
 			 conf->rmw_level, (unsigned long long)resync_offset,
 			 (unsigned long long)sh->sector);
+	} else if (mddev->bitmap_ops && mddev->bitmap_ops->blocks_synced &&
+		   !mddev->bitmap_ops->blocks_synced(mddev, sh->sector)) {
+		/* The initial recover is not done, must read everything */
+		rcw = 1; rmw = 2;
+		pr_debug("force RCW by lazy recovery, sh->sector=%llu\n",
+			 sh->sector);
 	} else for (i = disks; i--; ) {
 		/* would I have to read this buffer for read_modify_write */
 		struct r5dev *dev = &sh->dev[i];
@@ -4148,7 +4155,7 @@ static int handle_stripe_dirtying(struct r5conf *conf,
 	set_bit(STRIPE_HANDLE, &sh->state);
 	if ((rmw < rcw || (rmw == rcw && conf->rmw_level == PARITY_PREFER_RMW)) && rmw > 0) {
 		/* prefer read-modify-write, but need to get some data */
-		mddev_add_trace_msg(conf->mddev, "raid5 rmw %llu %d",
+		mddev_add_trace_msg(mddev, "raid5 rmw %llu %d",
 				sh->sector, rmw);
 
 		for (i = disks; i--; ) {
@@ -4227,8 +4234,8 @@ static int handle_stripe_dirtying(struct r5conf *conf,
 					set_bit(STRIPE_DELAYED, &sh->state);
 			}
 		}
-		if (rcw && !mddev_is_dm(conf->mddev))
-			blk_add_trace_msg(conf->mddev->gendisk->queue,
+		if (rcw && !mddev_is_dm(mddev))
+			blk_add_trace_msg(mddev->gendisk->queue,
 				"raid5 rcw %llu %d %d %d",
 				(unsigned long long)sh->sector, rcw, qread,
 				test_bit(STRIPE_DELAYED, &sh->state));
@@ -4698,10 +4705,21 @@ static void analyse_stripe(struct stripe_head *sh, struct stripe_head_state *s)
 			}
 		} else if (test_bit(In_sync, &rdev->flags))
 			set_bit(R5_Insync, &dev->flags);
-		else if (sh->sector + RAID5_STRIPE_SECTORS(conf) <= rdev->recovery_offset)
-			/* in sync if before recovery_offset */
-			set_bit(R5_Insync, &dev->flags);
-		else if (test_bit(R5_UPTODATE, &dev->flags) &&
+		else if (sh->sector + RAID5_STRIPE_SECTORS(conf) <=
+			 rdev->recovery_offset) {
+			/*
+			 * in sync if:
+			 *  - normal IO, or
+			 *  - resync IO that is not lazy recovery
+			 *
+			 * For lazy recovery, we have to mark the rdev without
+			 * In_sync as failed, to build initial xor data.
+			 */
+			if (!test_bit(STRIPE_SYNCING, &sh->state) ||
+			    !test_bit(MD_RECOVERY_LAZY_RECOVER,
+				      &conf->mddev->recovery))
+				set_bit(R5_Insync, &dev->flags);
+		} else if (test_bit(R5_UPTODATE, &dev->flags) &&
 			 test_bit(R5_Expanded, &dev->flags))
 			/* If we've reshaped into here, we assume it is Insync.
 			 * We will shortly update recovery_offset to make
@@ -6492,11 +6510,12 @@ static inline sector_t raid5_sync_request(struct mddev *mddev, sector_t sector_n
 		}
 
 		if (mddev->curr_resync < max_sector) /* aborted */
-			mddev->bitmap_ops->end_sync(mddev, mddev->curr_resync,
-						    &sync_blocks);
+			md_bitmap_end_sync(mddev, mddev->curr_resync,
+					   &sync_blocks);
 		else /* completed sync */
 			conf->fullsync = 0;
-		mddev->bitmap_ops->close_sync(mddev);
+		if (md_bitmap_enabled(mddev, false))
+			mddev->bitmap_ops->close_sync(mddev);
 
 		return 0;
 	}
@@ -6525,8 +6544,7 @@ static inline sector_t raid5_sync_request(struct mddev *mddev, sector_t sector_n
 	}
 	if (!test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery) &&
 	    !conf->fullsync &&
-	    !mddev->bitmap_ops->start_sync(mddev, sector_nr, &sync_blocks,
-					   true) &&
+	    !md_bitmap_start_sync(mddev, sector_nr, &sync_blocks, true) &&
 	    sync_blocks >= RAID5_STRIPE_SECTORS(conf)) {
 		/* we can skip this block, and probably more */
 		do_div(sync_blocks, RAID5_STRIPE_SECTORS(conf));
@@ -6535,7 +6553,8 @@ static inline sector_t raid5_sync_request(struct mddev *mddev, sector_t sector_n
 		return sync_blocks * RAID5_STRIPE_SECTORS(conf);
 	}
 
-	mddev->bitmap_ops->cond_end_sync(mddev, sector_nr, false);
+	if (md_bitmap_enabled(mddev, false))
+		mddev->bitmap_ops->cond_end_sync(mddev, sector_nr, false);
 
 	sh = raid5_get_active_stripe(conf, NULL, sector_nr,
 				     R5_GAS_NOBLOCK);
@@ -6557,9 +6576,7 @@ static inline sector_t raid5_sync_request(struct mddev *mddev, sector_t sector_n
 			still_degraded = true;
 	}
 
-	mddev->bitmap_ops->start_sync(mddev, sector_nr, &sync_blocks,
-				      still_degraded);
-
+	md_bitmap_start_sync(mddev, sector_nr, &sync_blocks, still_degraded);
 	set_bit(STRIPE_SYNC_REQUESTED, &sh->state);
 	set_bit(STRIPE_HANDLE, &sh->state);
 
@@ -6763,7 +6780,8 @@ static void raid5d(struct md_thread *thread)
 			/* Now is a good time to flush some bitmap updates */
 			conf->seq_flush++;
 			spin_unlock_irq(&conf->device_lock);
-			mddev->bitmap_ops->unplug(mddev, true);
+			if (md_bitmap_enabled(mddev, true))
+				mddev->bitmap_ops->unplug(mddev, true);
 			spin_lock_irq(&conf->device_lock);
 			conf->seq_write = conf->seq_flush;
 			activate_bit_delay(conf, conf->temp_inactive_list);
@@ -8312,7 +8330,6 @@ static int raid5_resize(struct mddev *mddev, sector_t sectors)
 	 */
 	sector_t newsize;
 	struct r5conf *conf = mddev->private;
-	int ret;
 
 	if (raid5_has_log(conf) || raid5_has_ppl(conf))
 		return -EINVAL;
@@ -8322,9 +8339,12 @@ static int raid5_resize(struct mddev *mddev, sector_t sectors)
 	    mddev->array_sectors > newsize)
 		return -EINVAL;
 
-	ret = mddev->bitmap_ops->resize(mddev, sectors, 0, false);
-	if (ret)
-		return ret;
+	if (md_bitmap_enabled(mddev, false)) {
+		int ret = mddev->bitmap_ops->resize(mddev, sectors, 0);
+
+		if (ret)
+			return ret;
+	}
 
 	md_set_array_sectors(mddev, newsize);
 	if (sectors > mddev->dev_sectors &&