Merge branch 'for-linus' of git://neil.brown.name/md

* 'for-linus' of git://neil.brown.name/md:
  md: don't retry recovery of raid1 that fails due to error on source drive.
  md: Allow md devices to be created by name.
  md: make devices disappear when they are no longer needed.
  md: centralise all freeing of an 'mddev' in 'md_free'
  md: move allocation of ->queue from mddev_find to md_probe
  md: need another print_sb for mdp_superblock_1
  md: use list_for_each_entry macro directly
  md: raid0: make hash_spacing and preshift sector-based.
  md: raid0: Represent the size of strip zones in sectors.
  md: raid0 create_strip_zones(): Add KERN_INFO/KERN_ERR to printk's.
  md: raid0 create_strip_zones(): Make two local variables sector-based.
  md: raid0: Represent zone->zone_offset in sectors.
  md: raid0: Represent device offset in sectors.
  md: raid0_make_request(): Replace local variable block by sector.
  md: raid0_make_request(): Remove local variable chunk_size.
  md: raid0_make_request(): Replace chunksize_bits by chunksect_bits.
  md: use sysfs_notify_dirent to notify changes to md/sync_action.
  md: fix bitmap-on-external-file bug.

drivers/md/bitmap.c

@@ -215,7 +215,6 @@ static struct page *read_sb_page(mddev_t *mddev, long offset,
 	/* choose a good rdev and read the page from there */
 
 	mdk_rdev_t *rdev;
-	struct list_head *tmp;
 	sector_t target;
 
 	if (!page)
@@ -223,7 +222,7 @@ static struct page *read_sb_page(mddev_t *mddev, long offset,
 	if (!page)
 		return ERR_PTR(-ENOMEM);
 
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		if (! test_bit(In_sync, &rdev->flags)
 		    || test_bit(Faulty, &rdev->flags))
 			continue;
@@ -964,7 +963,9 @@ static int bitmap_init_from_disk(struct bitmap *bitmap, sector_t start)
 				 */
 				page = bitmap->sb_page;
 				offset = sizeof(bitmap_super_t);
-				read_sb_page(bitmap->mddev, bitmap->offset,
+				if (!file)
+					read_sb_page(bitmap->mddev,
+						     bitmap->offset,
 						     page,
 						     index, count);
 			} else if (file) {

drivers/md/faulty.c

@@ -283,7 +283,6 @@ static int reconfig(mddev_t *mddev, int layout, int chunk_size)
 static int run(mddev_t *mddev)
 {
 	mdk_rdev_t *rdev;
-	struct list_head *tmp;
 	int i;
 
 	conf_t *conf = kmalloc(sizeof(*conf), GFP_KERNEL);
@@ -296,7 +295,7 @@ static int run(mddev_t *mddev)
 	}
 	conf->nfaults = 0;
 
-	rdev_for_each(rdev, tmp, mddev)
+	list_for_each_entry(rdev, &mddev->disks, same_set)
 		conf->rdev = rdev;
 
 	mddev->array_sectors = mddev->size * 2;

drivers/md/linear.c

@@ -105,7 +105,6 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks)
 	int i, nb_zone, cnt;
 	sector_t min_sectors;
 	sector_t curr_sector;
-	struct list_head *tmp;
 
 	conf = kzalloc (sizeof (*conf) + raid_disks*sizeof(dev_info_t),
 			GFP_KERNEL);
@@ -115,7 +114,7 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks)
 	cnt = 0;
 	conf->array_sectors = 0;
 
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		int j = rdev->raid_disk;
 		dev_info_t *disk = conf->disks + j;
 

drivers/md/multipath.c

@@ -408,7 +408,6 @@ static int multipath_run (mddev_t *mddev)
 	int disk_idx;
 	struct multipath_info *disk;
 	mdk_rdev_t *rdev;
-	struct list_head *tmp;
 
 	if (mddev->level != LEVEL_MULTIPATH) {
 		printk("multipath: %s: raid level not set to multipath IO (%d)\n",
@@ -441,7 +440,7 @@ static int multipath_run (mddev_t *mddev)
 	}
 
 	conf->working_disks = 0;
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		disk_idx = rdev->raid_disk;
 		if (disk_idx < 0 ||
 		    disk_idx >= mddev->raid_disks)

drivers/md/raid1.c

@@ -1016,12 +1016,16 @@ static void error(mddev_t *mddev, mdk_rdev_t *rdev)
 	 * else mark the drive as failed
 	 */
 	if (test_bit(In_sync, &rdev->flags)
-	    && (conf->raid_disks - mddev->degraded) == 1)
+	    && (conf->raid_disks - mddev->degraded) == 1) {
 		/*
 		 * Don't fail the drive, act as though we were just a
-		 * normal single drive
+		 * normal single drive.
+		 * However don't try a recovery from this drive as
+		 * it is very likely to fail.
 		 */
+		mddev->recovery_disabled = 1;
 		return;
+	}
 	if (test_and_clear_bit(In_sync, &rdev->flags)) {
 		unsigned long flags;
 		spin_lock_irqsave(&conf->device_lock, flags);
@@ -1919,7 +1923,6 @@ static int run(mddev_t *mddev)
 	int i, j, disk_idx;
 	mirror_info_t *disk;
 	mdk_rdev_t *rdev;
-	struct list_head *tmp;
 
 	if (mddev->level != 1) {
 		printk("raid1: %s: raid level not set to mirroring (%d)\n",
@@ -1964,7 +1967,7 @@ static int run(mddev_t *mddev)
 	spin_lock_init(&conf->device_lock);
 	mddev->queue->queue_lock = &conf->device_lock;
 
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		disk_idx = rdev->raid_disk;
 		if (disk_idx >= mddev->raid_disks
 		    || disk_idx < 0)

drivers/md/raid10.c

@@ -2025,7 +2025,6 @@ static int run(mddev_t *mddev)
 	int i, disk_idx;
 	mirror_info_t *disk;
 	mdk_rdev_t *rdev;
-	struct list_head *tmp;
 	int nc, fc, fo;
 	sector_t stride, size;
 
@@ -2108,7 +2107,7 @@ static int run(mddev_t *mddev)
 	spin_lock_init(&conf->device_lock);
 	mddev->queue->queue_lock = &conf->device_lock;
 
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		disk_idx = rdev->raid_disk;
 		if (disk_idx >= mddev->raid_disks
 		    || disk_idx < 0)

drivers/md/raid5.c

@@ -3998,7 +3998,6 @@ static int run(mddev_t *mddev)
 	int raid_disk, memory;
 	mdk_rdev_t *rdev;
 	struct disk_info *disk;
-	struct list_head *tmp;
 	int working_disks = 0;
 
 	if (mddev->level != 5 && mddev->level != 4 && mddev->level != 6) {
@@ -4108,7 +4107,7 @@ static int run(mddev_t *mddev)
 
 	pr_debug("raid5: run(%s) called.\n", mdname(mddev));
 
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		raid_disk = rdev->raid_disk;
 		if (raid_disk >= conf->raid_disks
 		    || raid_disk < 0)
@@ -4533,7 +4532,6 @@ static int raid5_start_reshape(mddev_t *mddev)
 {
 	raid5_conf_t *conf = mddev_to_conf(mddev);
 	mdk_rdev_t *rdev;
-	struct list_head *rtmp;
 	int spares = 0;
 	int added_devices = 0;
 	unsigned long flags;
@@ -4541,7 +4539,7 @@ static int raid5_start_reshape(mddev_t *mddev)
 	if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery))
 		return -EBUSY;
 
-	rdev_for_each(rdev, rtmp, mddev)
+	list_for_each_entry(rdev, &mddev->disks, same_set)
 		if (rdev->raid_disk < 0 &&
 		    !test_bit(Faulty, &rdev->flags))
 			spares++;
@@ -4563,7 +4561,7 @@ static int raid5_start_reshape(mddev_t *mddev)
 	/* Add some new drives, as many as will fit.
 	 * We know there are enough to make the newly sized array work.
 	 */
-	rdev_for_each(rdev, rtmp, mddev)
+	list_for_each_entry(rdev, &mddev->disks, same_set)
 		if (rdev->raid_disk < 0 &&
 		    !test_bit(Faulty, &rdev->flags)) {
 			if (raid5_add_disk(mddev, rdev) == 0) {

fs/block_dev.c

@@ -1005,6 +1005,7 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 	}
 
 	lock_kernel();
+ restart:
 
 	ret = -ENXIO;
 	disk = get_gendisk(bdev->bd_dev, &partno);
@@ -1025,6 +1026,19 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 
 			if (disk->fops->open) {
 				ret = disk->fops->open(bdev, mode);
+				if (ret == -ERESTARTSYS) {
+					/* Lost a race with 'disk' being
+					 * deleted, try again.
+					 * See md.c
+					 */
+					disk_put_part(bdev->bd_part);
+					bdev->bd_part = NULL;
+					module_put(disk->fops->owner);
+					put_disk(disk);
+					bdev->bd_disk = NULL;
+					mutex_unlock(&bdev->bd_mutex);
+					goto restart;
+				}
 				if (ret)
 					goto out_clear;
 			}

include/linux/raid/md_k.h

@@ -137,6 +137,9 @@ struct mddev_s
 	struct gendisk			*gendisk;
 
 	struct kobject			kobj;
+	int				hold_active;
+#define	UNTIL_IOCTL	1
+#define	UNTIL_STOP	2
 
 	/* Superblock information */
 	int				major_version,
@@ -215,6 +218,9 @@ struct mddev_s
 #define	MD_RECOVERY_FROZEN	9
 
 	unsigned long			recovery;
+	int				recovery_disabled; /* if we detect that recovery
+							    * will always fail, set this
+							    * so we don't loop trying */
 
 	int				in_sync;	/* know to not need resync */
 	struct mutex			reconfig_mutex;
@@ -244,6 +250,9 @@ struct mddev_s
 	struct sysfs_dirent		*sysfs_state;	/* handle for 'array_state'
 							 * file in sysfs.
 							 */
+	struct sysfs_dirent		*sysfs_action;  /* handle for 'sync_action' */
+
+	struct work_struct del_work;	/* used for delayed sysfs removal */
 
 	spinlock_t			write_lock;
 	wait_queue_head_t		sb_wait;	/* for waiting on superblock updates */
@@ -334,17 +343,14 @@ static inline char * mdname (mddev_t * mddev)
  * iterates through some rdev ringlist. It's safe to remove the
  * current 'rdev'. Dont touch 'tmp' though.
  */
-#define rdev_for_each_list(rdev, tmp, list)				\
-									\
-	for ((tmp) = (list).next;					\
-		(rdev) = (list_entry((tmp), mdk_rdev_t, same_set)),	\
-			(tmp) = (tmp)->next, (tmp)->prev != &(list)	\
-		; )
+#define rdev_for_each_list(rdev, tmp, head)				\
+	list_for_each_entry_safe(rdev, tmp, head, same_set)
+
 /*
  * iterates through the 'same array disks' ringlist
  */
 #define rdev_for_each(rdev, tmp, mddev)				\
-	rdev_for_each_list(rdev, tmp, (mddev)->disks)
+	list_for_each_entry_safe(rdev, tmp, &((mddev)->disks), same_set)
 
 #define rdev_for_each_rcu(rdev, mddev)				\
 	list_for_each_entry_rcu(rdev, &((mddev)->disks), same_set)

include/linux/raid/md_p.h

@@ -194,6 +194,8 @@ static inline __u64 md_event(mdp_super_t *sb) {
 	return (ev<<32)| sb->events_lo;
 }
 
+#define MD_SUPERBLOCK_1_TIME_SEC_MASK ((1ULL<<40) - 1)
+
 /*
  * The version-1 superblock :
  * All numeric fields are little-endian.

include/linux/raid/raid0.h

@@ -5,9 +5,9 @@
 
 struct strip_zone
 {
-	sector_t zone_offset;	/* Zone offset in md_dev */
-	sector_t dev_offset;	/* Zone offset in real dev */
-	sector_t size;		/* Zone size */
+	sector_t zone_start;	/* Zone offset in md_dev (in sectors) */
+	sector_t dev_start;	/* Zone offset in real dev (in sectors) */
+	sector_t sectors;	/* Zone size in sectors */
 	int nb_dev;		/* # of devices attached to the zone */
 	mdk_rdev_t **dev;	/* Devices attached to the zone */
 };
@@ -19,8 +19,8 @@ struct raid0_private_data
 	mdk_rdev_t **devlist; /* lists of rdevs, pointed to by strip_zone->dev */
 	int nr_strip_zones;
 
-	sector_t hash_spacing;
-	int preshift;			/* shift this before divide by hash_spacing */
+	sector_t spacing;
+	int sector_shift; /* shift this before divide by spacing */
 };
 
 typedef struct raid0_private_data raid0_conf_t;