Merge: RHEL10.2 Update MD RAID to upstream 2nd release

MR: https://gitlab.com/redhat/centos-stream/src/kernel/centos-stream-10/-/merge_requests/1681

JIRA: https://issues.redhat.com/browse/RHEL-123668

BREW: https://brewweb.engineering.redhat.com/brew/taskinfo?taskID=69335300

Upstream Status: Commits are found in Linus's git tree

Signed-off-by: Nigel Croxon <ncroxon@redhat.com>

Approved-by: Phil Auld <pauld@redhat.com>
Approved-by: Xiao Ni <xni@redhat.com>
Approved-by: CKI KWF Bot <cki-ci-bot+kwf-gitlab-com@redhat.com>

Merged-by: CKI GitLab Kmaint Pipeline Bot <26919896-cki-kmaint-pipeline-bot@users.noreply.gitlab.com>

Author: CKI KWF Bot

crypto/async_tx/async_pq.c

@@ -119,7 +119,7 @@ do_sync_gen_syndrome(struct page **blocks, unsigned int *offsets, int disks,
 	for (i = 0; i < disks; i++) {
 		if (blocks[i] == NULL) {
 			BUG_ON(i > disks - 3); /* P or Q can't be zero */
-			srcs[i] = (void*)raid6_empty_zero_page;
+			srcs[i] = raid6_get_zero_page();
 		} else {
 			srcs[i] = page_address(blocks[i]) + offsets[i];
 

crypto/async_tx/async_raid6_recov.c

@@ -414,7 +414,7 @@ async_raid6_2data_recov(int disks, size_t bytes, int faila, int failb,
 		async_tx_quiesce(&submit->depend_tx);
 		for (i = 0; i < disks; i++)
 			if (blocks[i] == NULL)
-				ptrs[i] = (void *) raid6_empty_zero_page;
+				ptrs[i] = raid6_get_zero_page();
 			else
 				ptrs[i] = page_address(blocks[i]) + offs[i];
 
@@ -497,7 +497,7 @@ async_raid6_datap_recov(int disks, size_t bytes, int faila,
 		async_tx_quiesce(&submit->depend_tx);
 		for (i = 0; i < disks; i++)
 			if (blocks[i] == NULL)
-				ptrs[i] = (void*)raid6_empty_zero_page;
+				ptrs[i] = raid6_get_zero_page();
 			else
 				ptrs[i] = page_address(blocks[i]) + offs[i];
 

drivers/md/dm-raid.c

@@ -438,7 +438,7 @@ static bool rs_is_reshapable(struct raid_set *rs)
 /* Return true, if raid set in @rs is recovering */
 static bool rs_is_recovering(struct raid_set *rs)
 {
-	return rs->md.recovery_cp < rs->md.dev_sectors;
+	return rs->md.resync_offset < rs->md.dev_sectors;
 }
 
 /* Return true, if raid set in @rs is reshaping */
@@ -768,7 +768,7 @@ static struct raid_set *raid_set_alloc(struct dm_target *ti, struct raid_type *r
 	rs->md.layout = raid_type->algorithm;
 	rs->md.new_layout = rs->md.layout;
 	rs->md.delta_disks = 0;
-	rs->md.recovery_cp = MaxSector;
+	rs->md.resync_offset = MaxSector;
 
 	for (i = 0; i < raid_devs; i++)
 		md_rdev_init(&rs->dev[i].rdev);
@@ -912,7 +912,7 @@ static int parse_dev_params(struct raid_set *rs, struct dm_arg_set *as)
 		rs->md.external = 0;
 		rs->md.persistent = 1;
 		rs->md.major_version = 2;
-	} else if (rebuild && !rs->md.recovery_cp) {
+	} else if (rebuild && !rs->md.resync_offset) {
 		/*
 		 * Without metadata, we will not be able to tell if the array
 		 * is in-sync or not - we must assume it is not.  Therefore,
@@ -1695,20 +1695,20 @@ static void rs_setup_recovery(struct raid_set *rs, sector_t dev_sectors)
 {
 	/* raid0 does not recover */
 	if (rs_is_raid0(rs))
-		rs->md.recovery_cp = MaxSector;
+		rs->md.resync_offset = MaxSector;
 	/*
 	 * A raid6 set has to be recovered either
 	 * completely or for the grown part to
 	 * ensure proper parity and Q-Syndrome
 	 */
 	else if (rs_is_raid6(rs))
-		rs->md.recovery_cp = dev_sectors;
+		rs->md.resync_offset = dev_sectors;
 	/*
 	 * Other raid set types may skip recovery
 	 * depending on the 'nosync' flag.
 	 */
 	else
-		rs->md.recovery_cp = test_bit(__CTR_FLAG_NOSYNC, &rs->ctr_flags)
+		rs->md.resync_offset = test_bit(__CTR_FLAG_NOSYNC, &rs->ctr_flags)
 				     ? MaxSector : dev_sectors;
 }
 
@@ -2143,7 +2143,7 @@ static void super_sync(struct mddev *mddev, struct md_rdev *rdev)
 	sb->events = cpu_to_le64(mddev->events);
 
 	sb->disk_recovery_offset = cpu_to_le64(rdev->recovery_offset);
-	sb->array_resync_offset = cpu_to_le64(mddev->recovery_cp);
+	sb->array_resync_offset = cpu_to_le64(mddev->resync_offset);
 
 	sb->level = cpu_to_le32(mddev->level);
 	sb->layout = cpu_to_le32(mddev->layout);
@@ -2334,18 +2334,18 @@ static int super_init_validation(struct raid_set *rs, struct md_rdev *rdev)
 	}
 
 	if (!test_bit(__CTR_FLAG_NOSYNC, &rs->ctr_flags))
-		mddev->recovery_cp = le64_to_cpu(sb->array_resync_offset);
+		mddev->resync_offset = le64_to_cpu(sb->array_resync_offset);
 
 	/*
 	 * During load, we set FirstUse if a new superblock was written.
 	 * There are two reasons we might not have a superblock:
 	 * 1) The raid set is brand new - in which case, all of the
 	 *    devices must have their In_sync bit set.	Also,
-	 *    recovery_cp must be 0, unless forced.
+	 *    resync_offset must be 0, unless forced.
 	 * 2) This is a new device being added to an old raid set
 	 *    and the new device needs to be rebuilt - in which
 	 *    case the In_sync bit will /not/ be set and
-	 *    recovery_cp must be MaxSector.
+	 *    resync_offset must be MaxSector.
 	 * 3) This is/are a new device(s) being added to an old
 	 *    raid set during takeover to a higher raid level
 	 *    to provide capacity for redundancy or during reshape
@@ -2390,8 +2390,8 @@ static int super_init_validation(struct raid_set *rs, struct md_rdev *rdev)
 			      new_devs > 1 ? "s" : "");
 			return -EINVAL;
 		} else if (!test_bit(__CTR_FLAG_REBUILD, &rs->ctr_flags) && rs_is_recovering(rs)) {
-			DMERR("'rebuild' specified while raid set is not in-sync (recovery_cp=%llu)",
-			      (unsigned long long) mddev->recovery_cp);
+			DMERR("'rebuild' specified while raid set is not in-sync (resync_offset=%llu)",
+			      (unsigned long long) mddev->resync_offset);
 			return -EINVAL;
 		} else if (rs_is_reshaping(rs)) {
 			DMERR("'rebuild' specified while raid set is being reshaped (reshape_position=%llu)",
@@ -2700,11 +2700,11 @@ static int rs_adjust_data_offsets(struct raid_set *rs)
 	}
 out:
 	/*
-	 * Raise recovery_cp in case data_offset != 0 to
+	 * Raise resync_offset in case data_offset != 0 to
 	 * avoid false recovery positives in the constructor.
 	 */
-	if (rs->md.recovery_cp < rs->md.dev_sectors)
-		rs->md.recovery_cp += rs->dev[0].rdev.data_offset;
+	if (rs->md.resync_offset < rs->md.dev_sectors)
+		rs->md.resync_offset += rs->dev[0].rdev.data_offset;
 
 	/* Adjust data offsets on all rdevs but on any raid4/5/6 journal device */
 	rdev_for_each(rdev, &rs->md) {
@@ -2759,15 +2759,15 @@ static int rs_setup_takeover(struct raid_set *rs)
 	}
 
 	clear_bit(MD_ARRAY_FIRST_USE, &mddev->flags);
-	mddev->recovery_cp = MaxSector;
+	mddev->resync_offset = MaxSector;
 
 	while (d--) {
 		rdev = &rs->dev[d].rdev;
 
 		if (test_bit(d, (void *) rs->rebuild_disks)) {
 			clear_bit(In_sync, &rdev->flags);
 			clear_bit(Faulty, &rdev->flags);
-			mddev->recovery_cp = rdev->recovery_offset = 0;
+			mddev->resync_offset = rdev->recovery_offset = 0;
 			/* Bitmap has to be created when we do an "up" takeover */
 			set_bit(MD_ARRAY_FIRST_USE, &mddev->flags);
 		}
@@ -3225,7 +3225,7 @@ static int raid_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 			if (r)
 				goto bad;
 
-			rs_setup_recovery(rs, rs->md.recovery_cp < rs->md.dev_sectors ? rs->md.recovery_cp : rs->md.dev_sectors);
+			rs_setup_recovery(rs, rs->md.resync_offset < rs->md.dev_sectors ? rs->md.resync_offset : rs->md.dev_sectors);
 		} else {
 			/* This is no size change or it is shrinking, update size and record in superblocks */
 			r = rs_set_dev_and_array_sectors(rs, rs->ti->len, false);
@@ -3449,7 +3449,7 @@ static sector_t rs_get_progress(struct raid_set *rs, unsigned long recovery,
 
 	} else {
 		if (state == st_idle && !test_bit(MD_RECOVERY_INTR, &recovery))
-			r = mddev->recovery_cp;
+			r = mddev->resync_offset;
 		else
 			r = mddev->curr_resync_completed;
 
@@ -4073,15 +4073,15 @@ static int raid_preresume(struct dm_target *ti)
 		int chunksize = to_bytes(rs->requested_bitmap_chunk_sectors) ?: mddev->bitmap_info.chunksize;
 
 		r = mddev->bitmap_ops->resize(mddev, mddev->dev_sectors,
-					      chunksize, false);
+					      chunksize);
 		if (r)
 			DMERR("Failed to resize bitmap");
 	}
 
 	/* Check for any resize/reshape on @rs and adjust/initiate */
-	if (mddev->recovery_cp && mddev->recovery_cp < MaxSector) {
+	if (mddev->resync_offset && mddev->resync_offset < MaxSector) {
 		set_bit(MD_RECOVERY_REQUESTED, &mddev->recovery);
-		mddev->resync_min = mddev->recovery_cp;
+		mddev->resync_min = mddev->resync_offset;
 		if (test_bit(RT_FLAG_RS_GROW, &rs->runtime_flags))
 			mddev->resync_max_sectors = mddev->dev_sectors;
 	}

drivers/md/md-bitmap.c

@@ -232,20 +232,19 @@ 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);
-}
-
-static bool bitmap_enabled(struct mddev *mddev)
-{
-	struct bitmap *bitmap = mddev->bitmap;
+	struct bitmap *bitmap = data;
 
-	if (!bitmap)
-		return false;
+	if (!flush)
+		return true;
 
-	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 +483,8 @@ 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 +1244,7 @@ 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 +1788,9 @@ 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 +1839,6 @@ 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)
@@ -1987,12 +1977,12 @@ static void bitmap_dirty_bits(struct mddev *mddev, unsigned long s,
 
 		md_bitmap_set_memory_bits(bitmap, sec, 1);
 		md_bitmap_file_set_bit(bitmap, sec);
-		if (sec < bitmap->mddev->recovery_cp)
+		if (sec < bitmap->mddev->resync_offset)
 			/* We are asserting that the array is dirty,
-			 * so move the recovery_cp address back so
+			 * so move the resync_offset address back so
 			 * that it is obvious that it is dirty
 			 */
-			bitmap->mddev->recovery_cp = sec;
+			bitmap->mddev->resync_offset = sec;
 	}
 }
 
@@ -2060,9 +2050,6 @@ 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 +2063,6 @@ 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",
@@ -2258,7 +2242,7 @@ static int bitmap_load(struct mddev *mddev)
 	    || bitmap->events_cleared == mddev->events)
 		/* no need to keep dirty bits to optimise a
 		 * re-add of a missing device */
-		start = mddev->recovery_cp;
+		start = mddev->resync_offset;
 
 	mutex_lock(&mddev->bitmap_info.mutex);
 	err = md_bitmap_init_from_disk(bitmap, start);
@@ -2593,15 +2577,14 @@ static int __bitmap_resize(struct bitmap *bitmap, sector_t blocks,
 	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
@@ -2990,7 +2973,8 @@ 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,
 };
@@ -3026,6 +3010,8 @@ 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)