Merge tag 'md-3.7' of git://neil.brown.name/md

Pull md updates from NeilBrown:
 - "discard" support, some dm-raid improvements and other assorted bits
   and pieces.

* tag 'md-3.7' of git://neil.brown.name/md: (29 commits)
  md: refine reporting of resync/reshape delays.
  md/raid5: be careful not to resize_stripes too big.
  md: make sure manual changes to recovery checkpoint are saved.
  md/raid10: use correct limit variable
  md: writing to sync_action should clear the read-auto state.
  Subject: [PATCH] md:change resync_mismatches to atomic64_t to avoid races
  md/raid5: make sure to_read and to_write never go negative.
  md: When RAID5 is dirty, force reconstruct-write instead of read-modify-write.
  md/raid5: protect debug message against NULL derefernce.
  md/raid5: add some missing locking in handle_failed_stripe.
  MD: raid5 avoid unnecessary zero page for trim
  MD: raid5 trim support
  md/bitmap:Don't use IS_ERR to judge alloc_page().
  md/raid1: Don't release reference to device while handling read error.
  raid: replace list_for_each_continue_rcu with new interface
  add further __init annotations to crypto/xor.c
  DM RAID: Fix for "sync" directive ineffectiveness
  DM RAID: Fix comparison of index and quantity for "rebuild" parameter
  DM RAID: Add rebuild capability for RAID10
  DM RAID: Move 'rebuild' checking code to its own function
  ...

Documentation/device-mapper/dm-raid.txt

@@ -132,3 +132,12 @@ Here we can see the RAID type is raid4, there are 5 devices - all of
 which are 'A'live, and the array is 2/490221568 complete with recovery.
 Faulty or missing devices are marked 'D'.  Devices that are out-of-sync
 are marked 'a'.
+
+
+Version History
+---------------
+1.0.0	Initial version.  Support for RAID 4/5/6
+1.1.0	Added support for RAID 1
+1.2.0	Handle creation of arrays that contain failed devices.
+1.3.0	Added support for RAID 10
+1.3.1	Allow device replacement/rebuild for RAID 10

crypto/xor.c

@@ -56,11 +56,11 @@ xor_blocks(unsigned int src_count, unsigned int bytes, void *dest, void **srcs)
 EXPORT_SYMBOL(xor_blocks);
 
 /* Set of all registered templates.  */
-static struct xor_block_template *template_list;
+static struct xor_block_template *__initdata template_list;
 
 #define BENCH_SIZE (PAGE_SIZE)
 
-static void
+static void __init
 do_xor_speed(struct xor_block_template *tmpl, void *b1, void *b2)
 {
 	int speed;

drivers/md/bitmap.c

@@ -163,20 +163,17 @@ static struct md_rdev *next_active_rdev(struct md_rdev *rdev, struct mddev *mdde
 	 * As devices are only added or removed when raid_disk is < 0 and
 	 * nr_pending is 0 and In_sync is clear, the entries we return will
 	 * still be in the same position on the list when we re-enter
-	 * list_for_each_continue_rcu.
+	 * list_for_each_entry_continue_rcu.
 	 */
-	struct list_head *pos;
 	rcu_read_lock();
 	if (rdev == NULL)
 		/* start at the beginning */
-		pos = &mddev->disks;
+		rdev = list_entry_rcu(&mddev->disks, struct md_rdev, same_set);
 	else {
 		/* release the previous rdev and start from there. */
 		rdev_dec_pending(rdev, mddev);
-		pos = &rdev->same_set;
 	}
-	list_for_each_continue_rcu(pos, &mddev->disks) {
-		rdev = list_entry(pos, struct md_rdev, same_set);
+	list_for_each_entry_continue_rcu(rdev, &mddev->disks, same_set) {
 		if (rdev->raid_disk >= 0 &&
 		    !test_bit(Faulty, &rdev->flags)) {
 			/* this is a usable devices */
@@ -473,14 +470,10 @@ static int bitmap_new_disk_sb(struct bitmap *bitmap)
 {
 	bitmap_super_t *sb;
 	unsigned long chunksize, daemon_sleep, write_behind;
-	int err = -EINVAL;
 
 	bitmap->storage.sb_page = alloc_page(GFP_KERNEL);
-	if (IS_ERR(bitmap->storage.sb_page)) {
-		err = PTR_ERR(bitmap->storage.sb_page);
-		bitmap->storage.sb_page = NULL;
-		return err;
-	}
+	if (bitmap->storage.sb_page == NULL)
+		return -ENOMEM;
 	bitmap->storage.sb_page->index = 0;
 
 	sb = kmap_atomic(bitmap->storage.sb_page);

drivers/md/dm-raid.c

@@ -337,6 +337,84 @@ static int validate_region_size(struct raid_set *rs, unsigned long region_size)
 	return 0;
 }
 
+/*
+ * validate_rebuild_devices
+ * @rs
+ *
+ * Determine if the devices specified for rebuild can result in a valid
+ * usable array that is capable of rebuilding the given devices.
+ *
+ * Returns: 0 on success, -EINVAL on failure.
+ */
+static int validate_rebuild_devices(struct raid_set *rs)
+{
+	unsigned i, rebuild_cnt = 0;
+	unsigned rebuilds_per_group, copies, d;
+
+	if (!(rs->print_flags & DMPF_REBUILD))
+		return 0;
+
+	for (i = 0; i < rs->md.raid_disks; i++)
+		if (!test_bit(In_sync, &rs->dev[i].rdev.flags))
+			rebuild_cnt++;
+
+	switch (rs->raid_type->level) {
+	case 1:
+		if (rebuild_cnt >= rs->md.raid_disks)
+			goto too_many;
+		break;
+	case 4:
+	case 5:
+	case 6:
+		if (rebuild_cnt > rs->raid_type->parity_devs)
+			goto too_many;
+		break;
+	case 10:
+		copies = raid10_md_layout_to_copies(rs->md.layout);
+		if (rebuild_cnt < copies)
+			break;
+
+		/*
+		 * It is possible to have a higher rebuild count for RAID10,
+		 * as long as the failed devices occur in different mirror
+		 * groups (i.e. different stripes).
+		 *
+		 * Right now, we only allow for "near" copies.  When other
+		 * formats are added, we will have to check those too.
+		 *
+		 * When checking "near" format, make sure no adjacent devices
+		 * have failed beyond what can be handled.  In addition to the
+		 * simple case where the number of devices is a multiple of the
+		 * number of copies, we must also handle cases where the number
+		 * of devices is not a multiple of the number of copies.
+		 * E.g.    dev1 dev2 dev3 dev4 dev5
+		 *          A    A    B    B    C
+		 *          C    D    D    E    E
+		 */
+		rebuilds_per_group = 0;
+		for (i = 0; i < rs->md.raid_disks * copies; i++) {
+			d = i % rs->md.raid_disks;
+			if (!test_bit(In_sync, &rs->dev[d].rdev.flags) &&
+			    (++rebuilds_per_group >= copies))
+				goto too_many;
+			if (!((i + 1) % copies))
+				rebuilds_per_group = 0;
+		}
+		break;
+	default:
+		DMERR("The rebuild parameter is not supported for %s",
+		      rs->raid_type->name);
+		rs->ti->error = "Rebuild not supported for this RAID type";
+		return -EINVAL;
+	}
+
+	return 0;
+
+too_many:
+	rs->ti->error = "Too many rebuild devices specified";
+	return -EINVAL;
+}
+
 /*
  * Possible arguments are...
  *	<chunk_size> [optional_args]
@@ -365,7 +443,7 @@ static int parse_raid_params(struct raid_set *rs, char **argv,
 {
 	char *raid10_format = "near";
 	unsigned raid10_copies = 2;
-	unsigned i, rebuild_cnt = 0;
+	unsigned i;
 	unsigned long value, region_size = 0;
 	sector_t sectors_per_dev = rs->ti->len;
 	sector_t max_io_len;
@@ -461,31 +539,7 @@ static int parse_raid_params(struct raid_set *rs, char **argv,
 
 		/* Parameters that take a numeric value are checked here */
 		if (!strcasecmp(key, "rebuild")) {
-			rebuild_cnt++;
-
-			switch (rs->raid_type->level) {
-			case 1:
-				if (rebuild_cnt >= rs->md.raid_disks) {
-					rs->ti->error = "Too many rebuild devices specified";
-					return -EINVAL;
-				}
-				break;
-			case 4:
-			case 5:
-			case 6:
-				if (rebuild_cnt > rs->raid_type->parity_devs) {
-					rs->ti->error = "Too many rebuild devices specified for given RAID type";
-					return -EINVAL;
-				}
-				break;
-			case 10:
-			default:
-				DMERR("The rebuild parameter is not supported for %s", rs->raid_type->name);
-				rs->ti->error = "Rebuild not supported for this RAID type";
-				return -EINVAL;
-			}
-
-			if (value > rs->md.raid_disks) {
+			if (value >= rs->md.raid_disks) {
 				rs->ti->error = "Invalid rebuild index given";
 				return -EINVAL;
 			}
@@ -608,6 +662,9 @@ static int parse_raid_params(struct raid_set *rs, char **argv,
 	}
 	rs->md.dev_sectors = sectors_per_dev;
 
+	if (validate_rebuild_devices(rs))
+		return -EINVAL;
+
 	/* Assume there are no metadata devices until the drives are parsed */
 	rs->md.persistent = 0;
 	rs->md.external = 1;
@@ -960,6 +1017,19 @@ static int analyse_superblocks(struct dm_target *ti, struct raid_set *rs)
 
 	freshest = NULL;
 	rdev_for_each_safe(rdev, tmp, mddev) {
+		/*
+		 * Skipping super_load due to DMPF_SYNC will cause
+		 * the array to undergo initialization again as
+		 * though it were new.  This is the intended effect
+		 * of the "sync" directive.
+		 *
+		 * When reshaping capability is added, we must ensure
+		 * that the "sync" directive is disallowed during the
+		 * reshape.
+		 */
+		if (rs->print_flags & DMPF_SYNC)
+			continue;
+
 		if (!rdev->meta_bdev)
 			continue;
 
@@ -1360,7 +1430,7 @@ static void raid_resume(struct dm_target *ti)
 
 static struct target_type raid_target = {
 	.name = "raid",
-	.version = {1, 3, 0},
+	.version = {1, 3, 1},
 	.module = THIS_MODULE,
 	.ctr = raid_ctr,
 	.dtr = raid_dtr,

drivers/md/linear.c

@@ -138,6 +138,7 @@ static struct linear_conf *linear_conf(struct mddev *mddev, int raid_disks)
 	struct linear_conf *conf;
 	struct md_rdev *rdev;
 	int i, cnt;
+	bool discard_supported = false;
 
 	conf = kzalloc (sizeof (*conf) + raid_disks*sizeof(struct dev_info),
 			GFP_KERNEL);
@@ -171,6 +172,8 @@ static struct linear_conf *linear_conf(struct mddev *mddev, int raid_disks)
 		conf->array_sectors += rdev->sectors;
 		cnt++;
 
+		if (blk_queue_discard(bdev_get_queue(rdev->bdev)))
+			discard_supported = true;
 	}
 	if (cnt != raid_disks) {
 		printk(KERN_ERR "md/linear:%s: not enough drives present. Aborting!\n",
@@ -178,6 +181,11 @@ static struct linear_conf *linear_conf(struct mddev *mddev, int raid_disks)
 		goto out;
 	}
 
+	if (!discard_supported)
+		queue_flag_clear_unlocked(QUEUE_FLAG_DISCARD, mddev->queue);
+	else
+		queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, mddev->queue);
+
 	/*
 	 * Here we calculate the device offsets.
 	 */
@@ -244,7 +252,9 @@ static int linear_add(struct mddev *mddev, struct md_rdev *rdev)
 	if (!newconf)
 		return -ENOMEM;
 
-	oldconf = rcu_dereference(mddev->private);
+	oldconf = rcu_dereference_protected(mddev->private,
+					    lockdep_is_held(
+						    &mddev->reconfig_mutex));
 	mddev->raid_disks++;
 	rcu_assign_pointer(mddev->private, newconf);
 	md_set_array_sectors(mddev, linear_size(mddev, 0, 0));
@@ -256,7 +266,10 @@ static int linear_add(struct mddev *mddev, struct md_rdev *rdev)
 
 static int linear_stop (struct mddev *mddev)
 {
-	struct linear_conf *conf = mddev->private;
+	struct linear_conf *conf =
+		rcu_dereference_protected(mddev->private,
+					  lockdep_is_held(
+						  &mddev->reconfig_mutex));
 
 	/*
 	 * We do not require rcu protection here since
@@ -326,6 +339,14 @@ static void linear_make_request(struct mddev *mddev, struct bio *bio)
 	bio->bi_sector = bio->bi_sector - start_sector
 		+ tmp_dev->rdev->data_offset;
 	rcu_read_unlock();
+
+	if (unlikely((bio->bi_rw & REQ_DISCARD) &&
+		     !blk_queue_discard(bdev_get_queue(bio->bi_bdev)))) {
+		/* Just ignore it */
+		bio_endio(bio, 0);
+		return;
+	}
+
 	generic_make_request(bio);
 }
 

drivers/md/md.h

@@ -282,7 +282,7 @@ struct mddev {
 
 	sector_t			resync_max_sectors; /* may be set by personality */
 
-	sector_t			resync_mismatches; /* count of sectors where
+	atomic64_t			resync_mismatches; /* count of sectors where
 							    * parity/replica mismatch found
 							    */
 
@@ -540,12 +540,13 @@ static inline void sysfs_unlink_rdev(struct mddev *mddev, struct md_rdev *rdev)
 	list_for_each_entry_rcu(rdev, &((mddev)->disks), same_set)
 
 struct md_thread {
-	void			(*run) (struct mddev *mddev);
+	void			(*run) (struct md_thread *thread);
 	struct mddev		*mddev;
 	wait_queue_head_t	wqueue;
 	unsigned long           flags;
 	struct task_struct	*tsk;
 	unsigned long		timeout;
+	void			*private;
 };
 
 #define THREAD_WAKEUP  0
@@ -584,7 +585,7 @@ static inline void safe_put_page(struct page *p)
 extern int register_md_personality(struct md_personality *p);
 extern int unregister_md_personality(struct md_personality *p);
 extern struct md_thread *md_register_thread(
-	void (*run)(struct mddev *mddev),
+	void (*run)(struct md_thread *thread),
 	struct mddev *mddev,
 	const char *name);
 extern void md_unregister_thread(struct md_thread **threadp);
@@ -603,7 +604,7 @@ extern void md_super_write(struct mddev *mddev, struct md_rdev *rdev,
 extern void md_super_wait(struct mddev *mddev);
 extern int sync_page_io(struct md_rdev *rdev, sector_t sector, int size, 
 			struct page *page, int rw, bool metadata_op);
-extern void md_do_sync(struct mddev *mddev);
+extern void md_do_sync(struct md_thread *thread);
 extern void md_new_event(struct mddev *mddev);
 extern int md_allow_write(struct mddev *mddev);
 extern void md_wait_for_blocked_rdev(struct md_rdev *rdev, struct mddev *mddev);

drivers/md/multipath.c

@@ -335,8 +335,9 @@ static int multipath_remove_disk(struct mddev *mddev, struct md_rdev *rdev)
  *	3.	Performs writes following reads for array syncronising.
  */
 
-static void multipathd (struct mddev *mddev)
+static void multipathd(struct md_thread *thread)
 {
+	struct mddev *mddev = thread->mddev;
 	struct multipath_bh *mp_bh;
 	struct bio *bio;
 	unsigned long flags;

drivers/md/raid0.c

@@ -88,6 +88,7 @@ static int create_strip_zones(struct mddev *mddev, struct r0conf **private_conf)
 	char b[BDEVNAME_SIZE];
 	char b2[BDEVNAME_SIZE];
 	struct r0conf *conf = kzalloc(sizeof(*conf), GFP_KERNEL);
+	bool discard_supported = false;
 
 	if (!conf)
 		return -ENOMEM;
@@ -195,6 +196,9 @@ static int create_strip_zones(struct mddev *mddev, struct r0conf **private_conf)
 		if (!smallest || (rdev1->sectors < smallest->sectors))
 			smallest = rdev1;
 		cnt++;
+
+		if (blk_queue_discard(bdev_get_queue(rdev1->bdev)))
+			discard_supported = true;
 	}
 	if (cnt != mddev->raid_disks) {
 		printk(KERN_ERR "md/raid0:%s: too few disks (%d of %d) - "
@@ -272,6 +276,11 @@ static int create_strip_zones(struct mddev *mddev, struct r0conf **private_conf)
 	blk_queue_io_opt(mddev->queue,
 			 (mddev->chunk_sectors << 9) * mddev->raid_disks);
 
+	if (!discard_supported)
+		queue_flag_clear_unlocked(QUEUE_FLAG_DISCARD, mddev->queue);
+	else
+		queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, mddev->queue);
+
 	pr_debug("md/raid0:%s: done.\n", mdname(mddev));
 	*private_conf = conf;
 
@@ -423,6 +432,7 @@ static int raid0_run(struct mddev *mddev)
 		return -EINVAL;
 	blk_queue_max_hw_sectors(mddev->queue, mddev->chunk_sectors);
 	blk_queue_max_write_same_sectors(mddev->queue, mddev->chunk_sectors);
+	blk_queue_max_discard_sectors(mddev->queue, mddev->chunk_sectors);
 
 	/* if private is not null, we are here after takeover */
 	if (mddev->private == NULL) {
@@ -510,7 +520,7 @@ static void raid0_make_request(struct mddev *mddev, struct bio *bio)
 		sector_t sector = bio->bi_sector;
 		struct bio_pair *bp;
 		/* Sanity check -- queue functions should prevent this happening */
-		if (bio->bi_vcnt != 1 ||
+		if ((bio->bi_vcnt != 1 && bio->bi_vcnt != 0) ||
 		    bio->bi_idx != 0)
 			goto bad_map;
 		/* This is a one page bio that upper layers
@@ -536,6 +546,13 @@ static void raid0_make_request(struct mddev *mddev, struct bio *bio)
 	bio->bi_sector = sector_offset + zone->dev_start +
 		tmp_dev->data_offset;
 
+	if (unlikely((bio->bi_rw & REQ_DISCARD) &&
+		     !blk_queue_discard(bdev_get_queue(bio->bi_bdev)))) {
+		/* Just ignore it */
+		bio_endio(bio, 0);
+		return;
+	}
+
 	generic_make_request(bio);
 	return;
 

drivers/md/raid1.c

@@ -333,9 +333,10 @@ static void raid1_end_read_request(struct bio *bio, int error)
 		spin_unlock_irqrestore(&conf->device_lock, flags);
 	}
 
-	if (uptodate)
+	if (uptodate) {
 		raid_end_bio_io(r1_bio);
-	else {
+		rdev_dec_pending(conf->mirrors[mirror].rdev, conf->mddev);
+	} else {
 		/*
 		 * oops, read error:
 		 */
@@ -349,9 +350,8 @@ static void raid1_end_read_request(struct bio *bio, int error)
 			(unsigned long long)r1_bio->sector);
 		set_bit(R1BIO_ReadError, &r1_bio->state);
 		reschedule_retry(r1_bio);
+		/* don't drop the reference on read_disk yet */
 	}
-
-	rdev_dec_pending(conf->mirrors[mirror].rdev, conf->mddev);
 }
 
 static void close_write(struct r1bio *r1_bio)
@@ -781,6 +781,11 @@ static void flush_pending_writes(struct r1conf *conf)
 		while (bio) { /* submit pending writes */
 			struct bio *next = bio->bi_next;
 			bio->bi_next = NULL;
+			if (unlikely((bio->bi_rw & REQ_DISCARD) &&
+			    !blk_queue_discard(bdev_get_queue(bio->bi_bdev))))
+				/* Just ignore it */
+				bio_endio(bio, 0);
+			else
 				generic_make_request(bio);
 			bio = next;
 		}
@@ -994,6 +999,8 @@ static void make_request(struct mddev *mddev, struct bio * bio)
 	const int rw = bio_data_dir(bio);
 	const unsigned long do_sync = (bio->bi_rw & REQ_SYNC);
 	const unsigned long do_flush_fua = (bio->bi_rw & (REQ_FLUSH | REQ_FUA));
+	const unsigned long do_discard = (bio->bi_rw
+					  & (REQ_DISCARD | REQ_SECURE));
 	struct md_rdev *blocked_rdev;
 	struct blk_plug_cb *cb;
 	struct raid1_plug_cb *plug = NULL;
@@ -1295,7 +1302,7 @@ static void make_request(struct mddev *mddev, struct bio * bio)
 				   conf->mirrors[i].rdev->data_offset);
 		mbio->bi_bdev = conf->mirrors[i].rdev->bdev;
 		mbio->bi_end_io	= raid1_end_write_request;
-		mbio->bi_rw = WRITE | do_flush_fua | do_sync;
+		mbio->bi_rw = WRITE | do_flush_fua | do_sync | do_discard;
 		mbio->bi_private = r1_bio;
 
 		atomic_inc(&r1_bio->remaining);
@@ -1549,6 +1556,8 @@ static int raid1_add_disk(struct mddev *mddev, struct md_rdev *rdev)
 		clear_bit(Unmerged, &rdev->flags);
 	}
 	md_integrity_add_rdev(rdev, mddev);
+	if (blk_queue_discard(bdev_get_queue(rdev->bdev)))
+		queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, mddev->queue);
 	print_conf(conf);
 	return err;
 }
@@ -1867,7 +1876,7 @@ static int process_checks(struct r1bio *r1_bio)
 		} else
 			j = 0;
 		if (j >= 0)
-			mddev->resync_mismatches += r1_bio->sectors;
+			atomic64_add(r1_bio->sectors, &mddev->resync_mismatches);
 		if (j < 0 || (test_bit(MD_RECOVERY_CHECK, &mddev->recovery)
 			      && test_bit(BIO_UPTODATE, &sbio->bi_flags))) {
 			/* No need to write to this device. */
@@ -2220,6 +2229,7 @@ static void handle_read_error(struct r1conf *conf, struct r1bio *r1_bio)
 		unfreeze_array(conf);
 	} else
 		md_error(mddev, conf->mirrors[r1_bio->read_disk].rdev);
+	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);
@@ -2285,8 +2295,9 @@ static void handle_read_error(struct r1conf *conf, struct r1bio *r1_bio)
 	}
 }
 
-static void raid1d(struct mddev *mddev)
+static void raid1d(struct md_thread *thread)
 {
+	struct mddev *mddev = thread->mddev;
 	struct r1bio *r1_bio;
 	unsigned long flags;
 	struct r1conf *conf = mddev->private;
@@ -2783,6 +2794,7 @@ static int run(struct mddev *mddev)
 	int i;
 	struct md_rdev *rdev;
 	int ret;
+	bool discard_supported = false;
 
 	if (mddev->level != 1) {
 		printk(KERN_ERR "md/raid1:%s: raid level not set to mirroring (%d)\n",
@@ -2812,6 +2824,8 @@ static int run(struct mddev *mddev)
 			continue;
 		disk_stack_limits(mddev->gendisk, rdev->bdev,
 				  rdev->data_offset << 9);
+		if (blk_queue_discard(bdev_get_queue(rdev->bdev)))
+			discard_supported = true;
 	}
 
 	mddev->degraded = 0;
@@ -2846,6 +2860,13 @@ static int run(struct mddev *mddev)
 		mddev->queue->backing_dev_info.congested_fn = raid1_congested;
 		mddev->queue->backing_dev_info.congested_data = mddev;
 		blk_queue_merge_bvec(mddev->queue, raid1_mergeable_bvec);
+
+		if (discard_supported)
+			queue_flag_set_unlocked(QUEUE_FLAG_DISCARD,
+						mddev->queue);
+		else
+			queue_flag_clear_unlocked(QUEUE_FLAG_DISCARD,
+						  mddev->queue);
 	}
 
 	ret =  md_integrity_register(mddev);

drivers/md/raid10.c

@@ -911,6 +911,11 @@ static void flush_pending_writes(struct r10conf *conf)
 		while (bio) { /* submit pending writes */
 			struct bio *next = bio->bi_next;
 			bio->bi_next = NULL;
+			if (unlikely((bio->bi_rw & REQ_DISCARD) &&
+			    !blk_queue_discard(bdev_get_queue(bio->bi_bdev))))
+				/* Just ignore it */
+				bio_endio(bio, 0);
+			else
 				generic_make_request(bio);
 			bio = next;
 		}
@@ -1050,6 +1055,44 @@ static sector_t choose_data_offset(struct r10bio *r10_bio,
 		return rdev->new_data_offset;
 }
 
+struct raid10_plug_cb {
+	struct blk_plug_cb	cb;
+	struct bio_list		pending;
+	int			pending_cnt;
+};
+
+static void raid10_unplug(struct blk_plug_cb *cb, bool from_schedule)
+{
+	struct raid10_plug_cb *plug = container_of(cb, struct raid10_plug_cb,
+						   cb);
+	struct mddev *mddev = plug->cb.data;
+	struct r10conf *conf = mddev->private;
+	struct bio *bio;
+
+	if (from_schedule) {
+		spin_lock_irq(&conf->device_lock);
+		bio_list_merge(&conf->pending_bio_list, &plug->pending);
+		conf->pending_count += plug->pending_cnt;
+		spin_unlock_irq(&conf->device_lock);
+		md_wakeup_thread(mddev->thread);
+		kfree(plug);
+		return;
+	}
+
+	/* we aren't scheduling, so we can do the write-out directly. */
+	bio = bio_list_get(&plug->pending);
+	bitmap_unplug(mddev->bitmap);
+	wake_up(&conf->wait_barrier);
+
+	while (bio) { /* submit pending writes */
+		struct bio *next = bio->bi_next;
+		bio->bi_next = NULL;
+		generic_make_request(bio);
+		bio = next;
+	}
+	kfree(plug);
+}
+
 static void make_request(struct mddev *mddev, struct bio * bio)
 {
 	struct r10conf *conf = mddev->private;
@@ -1061,8 +1104,12 @@ static void make_request(struct mddev *mddev, struct bio * bio)
 	const int rw = bio_data_dir(bio);
 	const unsigned long do_sync = (bio->bi_rw & REQ_SYNC);
 	const unsigned long do_fua = (bio->bi_rw & REQ_FUA);
+	const unsigned long do_discard = (bio->bi_rw
+					  & (REQ_DISCARD | REQ_SECURE));
 	unsigned long flags;
 	struct md_rdev *blocked_rdev;
+	struct blk_plug_cb *cb;
+	struct raid10_plug_cb *plug = NULL;
 	int sectors_handled;
 	int max_sectors;
 	int sectors;
@@ -1081,7 +1128,7 @@ static void make_request(struct mddev *mddev, struct bio * bio)
 			 || conf->prev.near_copies < conf->prev.raid_disks))) {
 		struct bio_pair *bp;
 		/* Sanity check -- queue functions should prevent this happening */
-		if (bio->bi_vcnt != 1 ||
+		if ((bio->bi_vcnt != 1 && bio->bi_vcnt != 0) ||
 		    bio->bi_idx != 0)
 			goto bad_map;
 		/* This is a one page bio that upper layers
@@ -1410,15 +1457,26 @@ static void make_request(struct mddev *mddev, struct bio * bio)
 						      conf->mirrors[d].rdev));
 		mbio->bi_bdev = conf->mirrors[d].rdev->bdev;
 		mbio->bi_end_io	= raid10_end_write_request;
-		mbio->bi_rw = WRITE | do_sync | do_fua;
+		mbio->bi_rw = WRITE | do_sync | do_fua | do_discard;
 		mbio->bi_private = r10_bio;
 
 		atomic_inc(&r10_bio->remaining);
+
+		cb = blk_check_plugged(raid10_unplug, mddev, sizeof(*plug));
+		if (cb)
+			plug = container_of(cb, struct raid10_plug_cb, cb);
+		else
+			plug = NULL;
 		spin_lock_irqsave(&conf->device_lock, flags);
+		if (plug) {
+			bio_list_add(&plug->pending, mbio);
+			plug->pending_cnt++;
+		} else {
 			bio_list_add(&conf->pending_bio_list, mbio);
 			conf->pending_count++;
+		}
 		spin_unlock_irqrestore(&conf->device_lock, flags);
-		if (!mddev_check_plugged(mddev))
+		if (!plug)
 			md_wakeup_thread(mddev->thread);
 
 		if (!r10_bio->devs[i].repl_bio)
@@ -1439,7 +1497,7 @@ static void make_request(struct mddev *mddev, struct bio * bio)
 					   conf->mirrors[d].replacement));
 		mbio->bi_bdev = conf->mirrors[d].replacement->bdev;
 		mbio->bi_end_io	= raid10_end_write_request;
-		mbio->bi_rw = WRITE | do_sync | do_fua;
+		mbio->bi_rw = WRITE | do_sync | do_fua | do_discard;
 		mbio->bi_private = r10_bio;
 
 		atomic_inc(&r10_bio->remaining);
@@ -1638,7 +1696,7 @@ static int raid10_spare_active(struct mddev *mddev)
 			   && !test_bit(Faulty, &tmp->rdev->flags)
 			   && !test_and_set_bit(In_sync, &tmp->rdev->flags)) {
 			count++;
-			sysfs_notify_dirent(tmp->rdev->sysfs_state);
+			sysfs_notify_dirent_safe(tmp->rdev->sysfs_state);
 		}
 	}
 	spin_lock_irqsave(&conf->device_lock, flags);
@@ -1725,6 +1783,9 @@ static int raid10_add_disk(struct mddev *mddev, struct md_rdev *rdev)
 		clear_bit(Unmerged, &rdev->flags);
 	}
 	md_integrity_add_rdev(rdev, mddev);
+	if (blk_queue_discard(bdev_get_queue(rdev->bdev)))
+		queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, mddev->queue);
+
 	print_conf(conf);
 	return err;
 }
@@ -1952,7 +2013,7 @@ static void sync_request_write(struct mddev *mddev, struct r10bio *r10_bio)
 					break;
 			if (j == vcnt)
 				continue;
-			mddev->resync_mismatches += r10_bio->sectors;
+			atomic64_add(r10_bio->sectors, &mddev->resync_mismatches);
 			if (test_bit(MD_RECOVERY_CHECK, &mddev->recovery))
 				/* Don't fix anything. */
 				continue;
@@ -2673,8 +2734,9 @@ static void handle_write_completed(struct r10conf *conf, struct r10bio *r10_bio)
 	}
 }
 
-static void raid10d(struct mddev *mddev)
+static void raid10d(struct md_thread *thread)
 {
+	struct mddev *mddev = thread->mddev;
 	struct r10bio *r10_bio;
 	unsigned long flags;
 	struct r10conf *conf = mddev->private;
@@ -3158,7 +3220,7 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr,
 				else {
 					bad_sectors -= (sector - first_bad);
 					if (max_sync > bad_sectors)
-						max_sync = max_sync;
+						max_sync = bad_sectors;
 					continue;
 				}
 			}
@@ -3482,6 +3544,7 @@ static int run(struct mddev *mddev)
 	sector_t size;
 	sector_t min_offset_diff = 0;
 	int first = 1;
+	bool discard_supported = false;
 
 	if (mddev->private == NULL) {
 		conf = setup_conf(mddev);
@@ -3498,6 +3561,8 @@ static int run(struct mddev *mddev)
 
 	chunk_size = mddev->chunk_sectors << 9;
 	if (mddev->queue) {
+		blk_queue_max_discard_sectors(mddev->queue,
+					      mddev->chunk_sectors);
 		blk_queue_io_min(mddev->queue, chunk_size);
 		if (conf->geo.raid_disks % conf->geo.near_copies)
 			blk_queue_io_opt(mddev->queue, chunk_size * conf->geo.raid_disks);
@@ -3543,8 +3608,16 @@ static int run(struct mddev *mddev)
 					  rdev->data_offset << 9);
 
 		disk->head_position = 0;
+
+		if (blk_queue_discard(bdev_get_queue(rdev->bdev)))
+			discard_supported = true;
 	}
 
+	if (discard_supported)
+		queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, mddev->queue);
+	else
+		queue_flag_clear_unlocked(QUEUE_FLAG_DISCARD, mddev->queue);
+
 	/* need to check that every block has at least one working mirror */
 	if (!enough(conf, -1)) {
 		printk(KERN_ERR "md/raid10:%s: not enough operational mirrors.\n",

drivers/md/raid5.h

@@ -298,6 +298,7 @@ enum r5dev_flags {
 	R5_WantReplace, /* We need to update the replacement, we have read
 			 * data in, and now is a good time to write it out.
 			 */
+	R5_Discard,	/* Discard the stripe */
 };
 
 /*