Merge tag 'md/4.9-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/shli/md

Pull MD fixes from Shaohua Li:
 "There are several bug fixes queued:

   - fix raid5-cache recovery bugs

   - fix discard IO error handling for raid1/10

   - fix array sync writes bogus position to superblock

   - fix IO error handling for raid array with external metadata"

* tag 'md/4.9-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/shli/md:
  md: be careful not lot leak internal curr_resync value into metadata. -- (all)
  raid1: handle read error also in readonly mode
  raid5-cache: correct condition for empty metadata write
  md: report 'write_pending' state when array in sync
  md/raid5: write an empty meta-block when creating log super-block
  md/raid5: initialize next_checkpoint field before use
  RAID10: ignore discard error
  RAID1: ignore discard error

drivers/md/md.c

@@ -3887,10 +3887,10 @@ array_state_show(struct mddev *mddev, char *page)
 			st = read_auto;
 			break;
 		case 0:
-			if (mddev->in_sync)
-				st = clean;
-			else if (test_bit(MD_CHANGE_PENDING, &mddev->flags))
+			if (test_bit(MD_CHANGE_PENDING, &mddev->flags))
 				st = write_pending;
+			else if (mddev->in_sync)
+				st = clean;
 			else if (mddev->safemode)
 				st = active_idle;
 			else
@@ -8144,14 +8144,14 @@ void md_do_sync(struct md_thread *thread)
 
 	if (!test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery) &&
 	    !test_bit(MD_RECOVERY_INTR, &mddev->recovery) &&
-	    mddev->curr_resync > 2) {
+	    mddev->curr_resync > 3) {
 		mddev->curr_resync_completed = mddev->curr_resync;
 		sysfs_notify(&mddev->kobj, NULL, "sync_completed");
 	}
 	mddev->pers->sync_request(mddev, max_sectors, &skipped);
 
 	if (!test_bit(MD_RECOVERY_CHECK, &mddev->recovery) &&
-	    mddev->curr_resync > 2) {
+	    mddev->curr_resync > 3) {
 		if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) {
 			if (test_bit(MD_RECOVERY_INTR, &mddev->recovery)) {
 				if (mddev->curr_resync >= mddev->recovery_cp) {

drivers/md/raid1.c

@@ -403,11 +403,14 @@ static void raid1_end_write_request(struct bio *bio)
 	struct bio *to_put = NULL;
 	int mirror = find_bio_disk(r1_bio, bio);
 	struct md_rdev *rdev = conf->mirrors[mirror].rdev;
+	bool discard_error;
+
+	discard_error = bio->bi_error && bio_op(bio) == REQ_OP_DISCARD;
 
 	/*
 	 * 'one mirror IO has finished' event handler:
 	 */
-	if (bio->bi_error) {
+	if (bio->bi_error && !discard_error) {
 		set_bit(WriteErrorSeen,	&rdev->flags);
 		if (!test_and_set_bit(WantReplacement, &rdev->flags))
 			set_bit(MD_RECOVERY_NEEDED, &
@@ -444,7 +447,7 @@ static void raid1_end_write_request(struct bio *bio)
 
 		/* Maybe we can clear some bad blocks. */
 		if (is_badblock(rdev, r1_bio->sector, r1_bio->sectors,
-				&first_bad, &bad_sectors)) {
+				&first_bad, &bad_sectors) && !discard_error) {
 			r1_bio->bios[mirror] = IO_MADE_GOOD;
 			set_bit(R1BIO_MadeGood, &r1_bio->state);
 		}
@@ -2294,17 +2297,23 @@ static void handle_read_error(struct r1conf *conf, struct r1bio *r1_bio)
 	 * This is all done synchronously while the array is
 	 * frozen
 	 */
+
+	bio = r1_bio->bios[r1_bio->read_disk];
+	bdevname(bio->bi_bdev, b);
+	bio_put(bio);
+	r1_bio->bios[r1_bio->read_disk] = NULL;
+
 	if (mddev->ro == 0) {
 		freeze_array(conf, 1);
 		fix_read_error(conf, r1_bio->read_disk,
 			       r1_bio->sector, r1_bio->sectors);
 		unfreeze_array(conf);
-	} else
-		md_error(mddev, conf->mirrors[r1_bio->read_disk].rdev);
+	} else {
+		r1_bio->bios[r1_bio->read_disk] = IO_BLOCKED;
+	}
+
 	rdev_dec_pending(conf->mirrors[r1_bio->read_disk].rdev, conf->mddev);
 
-	bio = r1_bio->bios[r1_bio->read_disk];
-	bdevname(bio->bi_bdev, b);
 read_more:
 	disk = read_balance(conf, r1_bio, &max_sectors);
 	if (disk == -1) {
@@ -2315,11 +2324,6 @@ read_more:
 	} else {
 		const unsigned long do_sync
 			= r1_bio->master_bio->bi_opf & REQ_SYNC;
-		if (bio) {
-			r1_bio->bios[r1_bio->read_disk] =
-				mddev->ro ? IO_BLOCKED : NULL;
-			bio_put(bio);
-		}
 		r1_bio->read_disk = disk;
 		bio = bio_clone_mddev(r1_bio->master_bio, GFP_NOIO, mddev);
 		bio_trim(bio, r1_bio->sector - bio->bi_iter.bi_sector,

drivers/md/raid10.c

@@ -447,6 +447,9 @@ static void raid10_end_write_request(struct bio *bio)
 	struct r10conf *conf = r10_bio->mddev->private;
 	int slot, repl;
 	struct md_rdev *rdev = NULL;
+	bool discard_error;
+
+	discard_error = bio->bi_error && bio_op(bio) == REQ_OP_DISCARD;
 
 	dev = find_bio_disk(conf, r10_bio, bio, &slot, &repl);
 
@@ -460,7 +463,7 @@ static void raid10_end_write_request(struct bio *bio)
 	/*
 	 * this branch is our 'one mirror IO has finished' event handler:
 	 */
-	if (bio->bi_error) {
+	if (bio->bi_error && !discard_error) {
 		if (repl)
 			/* Never record new bad blocks to replacement,
 			 * just fail it.
@@ -503,7 +506,7 @@ static void raid10_end_write_request(struct bio *bio)
 		if (is_badblock(rdev,
 				r10_bio->devs[slot].addr,
 				r10_bio->sectors,
-				&first_bad, &bad_sectors)) {
+				&first_bad, &bad_sectors) && !discard_error) {
 			bio_put(bio);
 			if (repl)
 				r10_bio->devs[slot].repl_bio = IO_MADE_GOOD;

drivers/md/raid5-cache.c

@@ -1087,7 +1087,7 @@ static int r5l_recovery_log(struct r5l_log *log)
 	 * 1's seq + 10 and let superblock points to meta2. The same recovery will
 	 * not think meta 3 is a valid meta, because its seq doesn't match
 	 */
-	if (ctx.seq > log->last_cp_seq + 1) {
+	if (ctx.seq > log->last_cp_seq) {
 		int ret;
 
 		ret = r5l_log_write_empty_meta_block(log, ctx.pos, ctx.seq + 10);
@@ -1096,6 +1096,8 @@ static int r5l_recovery_log(struct r5l_log *log)
 		log->seq = ctx.seq + 11;
 		log->log_start = r5l_ring_add(log, ctx.pos, BLOCK_SECTORS);
 		r5l_write_super(log, ctx.pos);
+		log->last_checkpoint = ctx.pos;
+		log->next_checkpoint = ctx.pos;
 	} else {
 		log->log_start = ctx.pos;
 		log->seq = ctx.seq;
@@ -1154,6 +1156,7 @@ create:
 	if (create_super) {
 		log->last_cp_seq = prandom_u32();
 		cp = 0;
+		r5l_log_write_empty_meta_block(log, cp, log->last_cp_seq);
 		/*
 		 * Make sure super points to correct address. Log might have
 		 * data very soon. If super hasn't correct log tail address,
@@ -1168,6 +1171,7 @@ create:
 	if (log->max_free_space > RECLAIM_MAX_FREE_SPACE)
 		log->max_free_space = RECLAIM_MAX_FREE_SPACE;
 	log->last_checkpoint = cp;
+	log->next_checkpoint = cp;
 
 	__free_page(page);