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

* 'for-linus' of git://neil.brown.name/md: (39 commits)
  md/raid5: correctly update sync_completed when we reach max_resync
  md/raid5: add missing call to schedule() after prepare_to_wait()
  md/linear: use call_rcu to free obsolete 'conf' structures.
  md linear: Protecting mddev with rcu locks to avoid races
  md: Move check for bitmap presence to personality code.
  md: remove chunksize rounding from common code.
  md: raid0/linear: ensure device sizes are rounded to chunk size.
  md: move assignment of ->utime so that it never gets skipped.
  md: Push down reconstruction log message to personality code.
  md: merge reconfig and check_reshape methods.
  md: remove unnecessary arguments from ->reconfig method.
  md: raid5: check stripe cache is large enough in start_reshape
  md: raid0: chunk_sectors cleanups.
  md: fix some comments.
  md/raid5: Use is_power_of_2() in raid5_reconfig()/raid6_reconfig().
  md: convert conf->chunk_size and conf->prev_chunk to sectors.
  md: Convert mddev->new_chunk to sectors.
  md: Make mddev->chunk_size sector-based.
  md: raid0 :Enables chunk size other than powers of 2.
  md: prepare for non-power-of-two chunk sizes
  ...

drivers/md/faulty.c

@@ -255,14 +255,14 @@ static void status(struct seq_file *seq, mddev_t *mddev)
 }
 
 
-static int reconfig(mddev_t *mddev, int layout, int chunk_size)
+static int reshape(mddev_t *mddev)
 {
-	int mode = layout & ModeMask;
-	int count = layout >> ModeShift;
+	int mode = mddev->new_layout & ModeMask;
+	int count = mddev->new_layout >> ModeShift;
 	conf_t *conf = mddev->private;
 
-	if (chunk_size != -1)
-		return -EINVAL;
+	if (mddev->new_layout < 0)
+		return 0;
 
 	/* new layout */
 	if (mode == ClearFaults)
@@ -279,6 +279,7 @@ static int reconfig(mddev_t *mddev, int layout, int chunk_size)
 		atomic_set(&conf->counters[mode], count);
 	} else
 		return -EINVAL;
+	mddev->new_layout = -1;
 	mddev->layout = -1; /* makes sure further changes come through */
 	return 0;
 }
@@ -298,8 +299,12 @@ static int run(mddev_t *mddev)
 {
 	mdk_rdev_t *rdev;
 	int i;
+	conf_t *conf;
+
+	if (md_check_no_bitmap(mddev))
+		return -EINVAL;
 
-	conf_t *conf = kmalloc(sizeof(*conf), GFP_KERNEL);
+	conf = kmalloc(sizeof(*conf), GFP_KERNEL);
 	if (!conf)
 		return -ENOMEM;
 
@@ -315,7 +320,7 @@ static int run(mddev_t *mddev)
 	md_set_array_sectors(mddev, faulty_size(mddev, 0, 0));
 	mddev->private = conf;
 
-	reconfig(mddev, mddev->layout, -1);
+	reshape(mddev);
 
 	return 0;
 }
@@ -338,7 +343,7 @@ static struct mdk_personality faulty_personality =
 	.run		= run,
 	.stop		= stop,
 	.status		= status,
-	.reconfig	= reconfig,
+	.check_reshape	= reshape,
 	.size		= faulty_size,
 };
 

drivers/md/linear.h

@@ -3,27 +3,19 @@
 
 struct dev_info {
 	mdk_rdev_t	*rdev;
-	sector_t	num_sectors;
-	sector_t	start_sector;
+	sector_t	end_sector;
 };
 
 typedef struct dev_info dev_info_t;
 
 struct linear_private_data
 {
-	struct linear_private_data *prev;	/* earlier version */
-	dev_info_t		**hash_table;
-	sector_t		spacing;
 	sector_t		array_sectors;
-	int			sector_shift;	/* shift before dividing
-						 * by spacing
-						 */
 	dev_info_t		disks[0];
+	struct rcu_head		rcu;
 };
 
 
 typedef struct linear_private_data linear_conf_t;
 
-#define mddev_to_conf(mddev) ((linear_conf_t *) mddev->private)
-
 #endif

drivers/md/md.h

@@ -29,13 +29,6 @@
 typedef struct mddev_s mddev_t;
 typedef struct mdk_rdev_s mdk_rdev_t;
 
-/*
- * options passed in raidrun:
- */
-
-/* Currently this must fit in an 'int' */
-#define MAX_CHUNK_SIZE (1<<30)
-
 /*
  * MD's 'extended' device
  */
@@ -145,7 +138,7 @@ struct mddev_s
 	int 				external;	/* metadata is
 							 * managed externally */
 	char				metadata_type[17]; /* externally set*/
-	int				chunk_size;
+	int				chunk_sectors;
 	time_t				ctime, utime;
 	int				level, layout;
 	char				clevel[16];
@@ -166,7 +159,8 @@ struct mddev_s
 	 * If reshape_position is MaxSector, then no reshape is happening (yet).
 	 */
 	sector_t			reshape_position;
-	int				delta_disks, new_level, new_layout, new_chunk;
+	int				delta_disks, new_level, new_layout;
+	int				new_chunk_sectors;
 
 	struct mdk_thread_s		*thread;	/* management thread */
 	struct mdk_thread_s		*sync_thread;	/* doing resync or reconstruct */
@@ -325,7 +319,6 @@ struct mdk_personality
 	int (*check_reshape) (mddev_t *mddev);
 	int (*start_reshape) (mddev_t *mddev);
 	void (*finish_reshape) (mddev_t *mddev);
-	int (*reconfig) (mddev_t *mddev, int layout, int chunk_size);
 	/* quiesce moves between quiescence states
 	 * 0 - fully active
 	 * 1 - no new requests allowed
@@ -437,5 +430,6 @@ extern void md_new_event(mddev_t *mddev);
 extern int md_allow_write(mddev_t *mddev);
 extern void md_wait_for_blocked_rdev(mdk_rdev_t *rdev, mddev_t *mddev);
 extern void md_set_array_sectors(mddev_t *mddev, sector_t array_sectors);
+extern int md_check_no_bitmap(mddev_t *mddev);
 
 #endif /* _MD_MD_H */

drivers/md/multipath.c

@@ -58,7 +58,7 @@ static void multipath_reschedule_retry (struct multipath_bh *mp_bh)
 {
 	unsigned long flags;
 	mddev_t *mddev = mp_bh->mddev;
-	multipath_conf_t *conf = mddev_to_conf(mddev);
+	multipath_conf_t *conf = mddev->private;
 
 	spin_lock_irqsave(&conf->device_lock, flags);
 	list_add(&mp_bh->retry_list, &conf->retry_list);
@@ -75,7 +75,7 @@ static void multipath_reschedule_retry (struct multipath_bh *mp_bh)
 static void multipath_end_bh_io (struct multipath_bh *mp_bh, int err)
 {
 	struct bio *bio = mp_bh->master_bio;
-	multipath_conf_t *conf = mddev_to_conf(mp_bh->mddev);
+	multipath_conf_t *conf = mp_bh->mddev->private;
 
 	bio_endio(bio, err);
 	mempool_free(mp_bh, conf->pool);
@@ -85,7 +85,7 @@ static void multipath_end_request(struct bio *bio, int error)
 {
 	int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
 	struct multipath_bh * mp_bh = (struct multipath_bh *)(bio->bi_private);
-	multipath_conf_t *conf = mddev_to_conf(mp_bh->mddev);
+	multipath_conf_t *conf = mp_bh->mddev->private;
 	mdk_rdev_t *rdev = conf->multipaths[mp_bh->path].rdev;
 
 	if (uptodate)
@@ -107,7 +107,7 @@ static void multipath_end_request(struct bio *bio, int error)
 
 static void unplug_slaves(mddev_t *mddev)
 {
-	multipath_conf_t *conf = mddev_to_conf(mddev);
+	multipath_conf_t *conf = mddev->private;
 	int i;
 
 	rcu_read_lock();
@@ -138,7 +138,7 @@ static void multipath_unplug(struct request_queue *q)
 static int multipath_make_request (struct request_queue *q, struct bio * bio)
 {
 	mddev_t *mddev = q->queuedata;
-	multipath_conf_t *conf = mddev_to_conf(mddev);
+	multipath_conf_t *conf = mddev->private;
 	struct multipath_bh * mp_bh;
 	struct multipath_info *multipath;
 	const int rw = bio_data_dir(bio);
@@ -180,7 +180,7 @@ static int multipath_make_request (struct request_queue *q, struct bio * bio)
 
 static void multipath_status (struct seq_file *seq, mddev_t *mddev)
 {
-	multipath_conf_t *conf = mddev_to_conf(mddev);
+	multipath_conf_t *conf = mddev->private;
 	int i;
 	
 	seq_printf (seq, " [%d/%d] [", conf->raid_disks,
@@ -195,7 +195,7 @@ static void multipath_status (struct seq_file *seq, mddev_t *mddev)
 static int multipath_congested(void *data, int bits)
 {
 	mddev_t *mddev = data;
-	multipath_conf_t *conf = mddev_to_conf(mddev);
+	multipath_conf_t *conf = mddev->private;
 	int i, ret = 0;
 
 	rcu_read_lock();
@@ -220,7 +220,7 @@ static int multipath_congested(void *data, int bits)
  */
 static void multipath_error (mddev_t *mddev, mdk_rdev_t *rdev)
 {
-	multipath_conf_t *conf = mddev_to_conf(mddev);
+	multipath_conf_t *conf = mddev->private;
 
 	if (conf->working_disks <= 1) {
 		/*
@@ -367,7 +367,7 @@ static void multipathd (mddev_t *mddev)
 	struct multipath_bh *mp_bh;
 	struct bio *bio;
 	unsigned long flags;
-	multipath_conf_t *conf = mddev_to_conf(mddev);
+	multipath_conf_t *conf = mddev->private;
 	struct list_head *head = &conf->retry_list;
 
 	md_check_recovery(mddev);
@@ -421,6 +421,9 @@ static int multipath_run (mddev_t *mddev)
 	struct multipath_info *disk;
 	mdk_rdev_t *rdev;
 
+	if (md_check_no_bitmap(mddev))
+		return -EINVAL;
+
 	if (mddev->level != LEVEL_MULTIPATH) {
 		printk("multipath: %s: raid level not set to multipath IO (%d)\n",
 		       mdname(mddev), mddev->level);
@@ -531,7 +534,7 @@ out:
 
 static int multipath_stop (mddev_t *mddev)
 {
-	multipath_conf_t *conf = mddev_to_conf(mddev);
+	multipath_conf_t *conf = mddev->private;
 
 	md_unregister_thread(mddev->thread);
 	mddev->thread = NULL;

drivers/md/multipath.h

@@ -18,12 +18,6 @@ struct multipath_private_data {
 
 typedef struct multipath_private_data multipath_conf_t;
 
-/*
- * this is the only point in the RAID code where we violate
- * C type safety. mddev->private is an 'opaque' pointer.
- */
-#define mddev_to_conf(mddev) ((multipath_conf_t *) mddev->private)
-
 /*
  * this is our 'private' 'collective' MULTIPATH buffer head.
  * it contains information about what kind of IO operations were started

drivers/md/raid0.h

@@ -3,26 +3,18 @@
 
 struct strip_zone
 {
-	sector_t zone_start;	/* Zone offset in md_dev (in sectors) */
+	sector_t zone_end;	/* Start of the next zone (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 */
 };
 
 struct raid0_private_data
 {
-	struct strip_zone **hash_table; /* Table of indexes into strip_zone */
 	struct strip_zone *strip_zone;
 	mdk_rdev_t **devlist; /* lists of rdevs, pointed to by strip_zone->dev */
 	int nr_strip_zones;
-
-	sector_t spacing;
-	int sector_shift; /* shift this before divide by spacing */
 };
 
 typedef struct raid0_private_data raid0_conf_t;
 
-#define mddev_to_conf(mddev) ((raid0_conf_t *) mddev->private)
-
 #endif

drivers/md/raid1.c

@@ -182,7 +182,7 @@ static void put_all_bios(conf_t *conf, r1bio_t *r1_bio)
 
 static void free_r1bio(r1bio_t *r1_bio)
 {
-	conf_t *conf = mddev_to_conf(r1_bio->mddev);
+	conf_t *conf = r1_bio->mddev->private;
 
 	/*
 	 * Wake up any possible resync thread that waits for the device
@@ -196,7 +196,7 @@ static void free_r1bio(r1bio_t *r1_bio)
 
 static void put_buf(r1bio_t *r1_bio)
 {
-	conf_t *conf = mddev_to_conf(r1_bio->mddev);
+	conf_t *conf = r1_bio->mddev->private;
 	int i;
 
 	for (i=0; i<conf->raid_disks; i++) {
@@ -214,7 +214,7 @@ static void reschedule_retry(r1bio_t *r1_bio)
 {
 	unsigned long flags;
 	mddev_t *mddev = r1_bio->mddev;
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 
 	spin_lock_irqsave(&conf->device_lock, flags);
 	list_add(&r1_bio->retry_list, &conf->retry_list);
@@ -253,7 +253,7 @@ static void raid_end_bio_io(r1bio_t *r1_bio)
  */
 static inline void update_head_pos(int disk, r1bio_t *r1_bio)
 {
-	conf_t *conf = mddev_to_conf(r1_bio->mddev);
+	conf_t *conf = r1_bio->mddev->private;
 
 	conf->mirrors[disk].head_position =
 		r1_bio->sector + (r1_bio->sectors);
@@ -264,7 +264,7 @@ static void raid1_end_read_request(struct bio *bio, int error)
 	int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
 	r1bio_t * r1_bio = (r1bio_t *)(bio->bi_private);
 	int mirror;
-	conf_t *conf = mddev_to_conf(r1_bio->mddev);
+	conf_t *conf = r1_bio->mddev->private;
 
 	mirror = r1_bio->read_disk;
 	/*
@@ -309,7 +309,7 @@ static void raid1_end_write_request(struct bio *bio, int error)
 	int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
 	r1bio_t * r1_bio = (r1bio_t *)(bio->bi_private);
 	int mirror, behind = test_bit(R1BIO_BehindIO, &r1_bio->state);
-	conf_t *conf = mddev_to_conf(r1_bio->mddev);
+	conf_t *conf = r1_bio->mddev->private;
 	struct bio *to_put = NULL;
 
 
@@ -541,7 +541,7 @@ static int read_balance(conf_t *conf, r1bio_t *r1_bio)
 
 static void unplug_slaves(mddev_t *mddev)
 {
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 	int i;
 
 	rcu_read_lock();
@@ -573,7 +573,7 @@ static void raid1_unplug(struct request_queue *q)
 static int raid1_congested(void *data, int bits)
 {
 	mddev_t *mddev = data;
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 	int i, ret = 0;
 
 	rcu_read_lock();
@@ -772,7 +772,7 @@ do_sync_io:
 static int make_request(struct request_queue *q, struct bio * bio)
 {
 	mddev_t *mddev = q->queuedata;
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 	mirror_info_t *mirror;
 	r1bio_t *r1_bio;
 	struct bio *read_bio;
@@ -991,7 +991,7 @@ static int make_request(struct request_queue *q, struct bio * bio)
 
 static void status(struct seq_file *seq, mddev_t *mddev)
 {
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 	int i;
 
 	seq_printf(seq, " [%d/%d] [", conf->raid_disks,
@@ -1010,7 +1010,7 @@ static void status(struct seq_file *seq, mddev_t *mddev)
 static void error(mddev_t *mddev, mdk_rdev_t *rdev)
 {
 	char b[BDEVNAME_SIZE];
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 
 	/*
 	 * If it is not operational, then we have already marked it as dead
@@ -1214,7 +1214,7 @@ static void end_sync_write(struct bio *bio, int error)
 	int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
 	r1bio_t * r1_bio = (r1bio_t *)(bio->bi_private);
 	mddev_t *mddev = r1_bio->mddev;
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 	int i;
 	int mirror=0;
 
@@ -1248,7 +1248,7 @@ static void end_sync_write(struct bio *bio, int error)
 
 static void sync_request_write(mddev_t *mddev, r1bio_t *r1_bio)
 {
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 	int i;
 	int disks = conf->raid_disks;
 	struct bio *bio, *wbio;
@@ -1562,7 +1562,7 @@ static void raid1d(mddev_t *mddev)
 	r1bio_t *r1_bio;
 	struct bio *bio;
 	unsigned long flags;
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 	struct list_head *head = &conf->retry_list;
 	int unplug=0;
 	mdk_rdev_t *rdev;
@@ -1585,7 +1585,7 @@ static void raid1d(mddev_t *mddev)
 		spin_unlock_irqrestore(&conf->device_lock, flags);
 
 		mddev = r1_bio->mddev;
-		conf = mddev_to_conf(mddev);
+		conf = mddev->private;
 		if (test_bit(R1BIO_IsSync, &r1_bio->state)) {
 			sync_request_write(mddev, r1_bio);
 			unplug = 1;
@@ -1706,7 +1706,7 @@ static int init_resync(conf_t *conf)
 
 static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, int go_faster)
 {
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 	r1bio_t *r1_bio;
 	struct bio *bio;
 	sector_t max_sector, nr_sectors;
@@ -2052,6 +2052,10 @@ static int run(mddev_t *mddev)
 		goto out_free_conf;
 	}
 
+	if (mddev->recovery_cp != MaxSector)
+		printk(KERN_NOTICE "raid1: %s is not clean"
+		       " -- starting background reconstruction\n",
+		       mdname(mddev));
 	printk(KERN_INFO 
 		"raid1: raid set %s active with %d out of %d mirrors\n",
 		mdname(mddev), mddev->raid_disks - mddev->degraded, 
@@ -2087,7 +2091,7 @@ out:
 
 static int stop(mddev_t *mddev)
 {
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 	struct bitmap *bitmap = mddev->bitmap;
 	int behind_wait = 0;
 
@@ -2155,16 +2159,16 @@ static int raid1_reshape(mddev_t *mddev)
 	mempool_t *newpool, *oldpool;
 	struct pool_info *newpoolinfo;
 	mirror_info_t *newmirrors;
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 	int cnt, raid_disks;
 	unsigned long flags;
 	int d, d2, err;
 
 	/* Cannot change chunk_size, layout, or level */
-	if (mddev->chunk_size != mddev->new_chunk ||
+	if (mddev->chunk_sectors != mddev->new_chunk_sectors ||
 	    mddev->layout != mddev->new_layout ||
 	    mddev->level != mddev->new_level) {
-		mddev->new_chunk = mddev->chunk_size;
+		mddev->new_chunk_sectors = mddev->chunk_sectors;
 		mddev->new_layout = mddev->layout;
 		mddev->new_level = mddev->level;
 		return -EINVAL;
@@ -2252,7 +2256,7 @@ static int raid1_reshape(mddev_t *mddev)
 
 static void raid1_quiesce(mddev_t *mddev, int state)
 {
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 
 	switch(state) {
 	case 1:

drivers/md/raid1.h

@@ -63,12 +63,6 @@ struct r1_private_data_s {
 
 typedef struct r1_private_data_s conf_t;
 
-/*
- * this is the only point in the RAID code where we violate
- * C type safety. mddev->private is an 'opaque' pointer.
- */
-#define mddev_to_conf(mddev) ((conf_t *) mddev->private)
-
 /*
  * this is our 'private' RAID1 bio.
  *

drivers/md/raid10.c

@@ -188,7 +188,7 @@ static void put_all_bios(conf_t *conf, r10bio_t *r10_bio)
 
 static void free_r10bio(r10bio_t *r10_bio)
 {
-	conf_t *conf = mddev_to_conf(r10_bio->mddev);
+	conf_t *conf = r10_bio->mddev->private;
 
 	/*
 	 * Wake up any possible resync thread that waits for the device
@@ -202,7 +202,7 @@ static void free_r10bio(r10bio_t *r10_bio)
 
 static void put_buf(r10bio_t *r10_bio)
 {
-	conf_t *conf = mddev_to_conf(r10_bio->mddev);
+	conf_t *conf = r10_bio->mddev->private;
 
 	mempool_free(r10_bio, conf->r10buf_pool);
 
@@ -213,7 +213,7 @@ static void reschedule_retry(r10bio_t *r10_bio)
 {
 	unsigned long flags;
 	mddev_t *mddev = r10_bio->mddev;
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 
 	spin_lock_irqsave(&conf->device_lock, flags);
 	list_add(&r10_bio->retry_list, &conf->retry_list);
@@ -245,7 +245,7 @@ static void raid_end_bio_io(r10bio_t *r10_bio)
  */
 static inline void update_head_pos(int slot, r10bio_t *r10_bio)
 {
-	conf_t *conf = mddev_to_conf(r10_bio->mddev);
+	conf_t *conf = r10_bio->mddev->private;
 
 	conf->mirrors[r10_bio->devs[slot].devnum].head_position =
 		r10_bio->devs[slot].addr + (r10_bio->sectors);
@@ -256,7 +256,7 @@ static void raid10_end_read_request(struct bio *bio, int error)
 	int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
 	r10bio_t * r10_bio = (r10bio_t *)(bio->bi_private);
 	int slot, dev;
-	conf_t *conf = mddev_to_conf(r10_bio->mddev);
+	conf_t *conf = r10_bio->mddev->private;
 
 
 	slot = r10_bio->read_slot;
@@ -297,7 +297,7 @@ static void raid10_end_write_request(struct bio *bio, int error)
 	int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
 	r10bio_t * r10_bio = (r10bio_t *)(bio->bi_private);
 	int slot, dev;
-	conf_t *conf = mddev_to_conf(r10_bio->mddev);
+	conf_t *conf = r10_bio->mddev->private;
 
 	for (slot = 0; slot < conf->copies; slot++)
 		if (r10_bio->devs[slot].bio == bio)
@@ -461,7 +461,7 @@ static int raid10_mergeable_bvec(struct request_queue *q,
 	mddev_t *mddev = q->queuedata;
 	sector_t sector = bvm->bi_sector + get_start_sect(bvm->bi_bdev);
 	int max;
-	unsigned int chunk_sectors = mddev->chunk_size >> 9;
+	unsigned int chunk_sectors = mddev->chunk_sectors;
 	unsigned int bio_sectors = bvm->bi_size >> 9;
 
 	max =  (chunk_sectors - ((sector & (chunk_sectors - 1)) + bio_sectors)) << 9;
@@ -596,7 +596,7 @@ rb_out:
 
 static void unplug_slaves(mddev_t *mddev)
 {
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 	int i;
 
 	rcu_read_lock();
@@ -628,7 +628,7 @@ static void raid10_unplug(struct request_queue *q)
 static int raid10_congested(void *data, int bits)
 {
 	mddev_t *mddev = data;
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 	int i, ret = 0;
 
 	rcu_read_lock();
@@ -788,7 +788,7 @@ static void unfreeze_array(conf_t *conf)
 static int make_request(struct request_queue *q, struct bio * bio)
 {
 	mddev_t *mddev = q->queuedata;
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 	mirror_info_t *mirror;
 	r10bio_t *r10_bio;
 	struct bio *read_bio;
@@ -981,11 +981,11 @@ static int make_request(struct request_queue *q, struct bio * bio)
 
 static void status(struct seq_file *seq, mddev_t *mddev)
 {
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 	int i;
 
 	if (conf->near_copies < conf->raid_disks)
-		seq_printf(seq, " %dK chunks", mddev->chunk_size/1024);
+		seq_printf(seq, " %dK chunks", mddev->chunk_sectors / 2);
 	if (conf->near_copies > 1)
 		seq_printf(seq, " %d near-copies", conf->near_copies);
 	if (conf->far_copies > 1) {
@@ -1006,7 +1006,7 @@ static void status(struct seq_file *seq, mddev_t *mddev)
 static void error(mddev_t *mddev, mdk_rdev_t *rdev)
 {
 	char b[BDEVNAME_SIZE];
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 
 	/*
 	 * If it is not operational, then we have already marked it as dead
@@ -1215,7 +1215,7 @@ abort:
 static void end_sync_read(struct bio *bio, int error)
 {
 	r10bio_t * r10_bio = (r10bio_t *)(bio->bi_private);
-	conf_t *conf = mddev_to_conf(r10_bio->mddev);
+	conf_t *conf = r10_bio->mddev->private;
 	int i,d;
 
 	for (i=0; i<conf->copies; i++)
@@ -1253,7 +1253,7 @@ static void end_sync_write(struct bio *bio, int error)
 	int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
 	r10bio_t * r10_bio = (r10bio_t *)(bio->bi_private);
 	mddev_t *mddev = r10_bio->mddev;
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 	int i,d;
 
 	for (i = 0; i < conf->copies; i++)
@@ -1300,7 +1300,7 @@ static void end_sync_write(struct bio *bio, int error)
  */
 static void sync_request_write(mddev_t *mddev, r10bio_t *r10_bio)
 {
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 	int i, first;
 	struct bio *tbio, *fbio;
 
@@ -1400,7 +1400,7 @@ done:
 
 static void recovery_request_write(mddev_t *mddev, r10bio_t *r10_bio)
 {
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 	int i, d;
 	struct bio *bio, *wbio;
 
@@ -1549,7 +1549,7 @@ static void raid10d(mddev_t *mddev)
 	r10bio_t *r10_bio;
 	struct bio *bio;
 	unsigned long flags;
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 	struct list_head *head = &conf->retry_list;
 	int unplug=0;
 	mdk_rdev_t *rdev;
@@ -1572,7 +1572,7 @@ static void raid10d(mddev_t *mddev)
 		spin_unlock_irqrestore(&conf->device_lock, flags);
 
 		mddev = r10_bio->mddev;
-		conf = mddev_to_conf(mddev);
+		conf = mddev->private;
 		if (test_bit(R10BIO_IsSync, &r10_bio->state)) {
 			sync_request_write(mddev, r10_bio);
 			unplug = 1;
@@ -1680,7 +1680,7 @@ static int init_resync(conf_t *conf)
 
 static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, int go_faster)
 {
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 	r10bio_t *r10_bio;
 	struct bio *biolist = NULL, *bio;
 	sector_t max_sector, nr_sectors;
@@ -2026,7 +2026,7 @@ static sector_t
 raid10_size(mddev_t *mddev, sector_t sectors, int raid_disks)
 {
 	sector_t size;
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 
 	if (!raid_disks)
 		raid_disks = mddev->raid_disks;
@@ -2050,9 +2050,10 @@ static int run(mddev_t *mddev)
 	int nc, fc, fo;
 	sector_t stride, size;
 
-	if (mddev->chunk_size < PAGE_SIZE) {
+	if (mddev->chunk_sectors < (PAGE_SIZE >> 9) ||
+	    !is_power_of_2(mddev->chunk_sectors)) {
 		printk(KERN_ERR "md/raid10: chunk size must be "
-		       "at least PAGE_SIZE(%ld).\n", PAGE_SIZE);
+		       "at least PAGE_SIZE(%ld) and be a power of 2.\n", PAGE_SIZE);
 		return -EINVAL;
 	}
 
@@ -2095,8 +2096,8 @@ static int run(mddev_t *mddev)
 	conf->far_copies = fc;
 	conf->copies = nc*fc;
 	conf->far_offset = fo;
-	conf->chunk_mask = (sector_t)(mddev->chunk_size>>9)-1;
-	conf->chunk_shift = ffz(~mddev->chunk_size) - 9;
+	conf->chunk_mask = mddev->chunk_sectors - 1;
+	conf->chunk_shift = ffz(~mddev->chunk_sectors);
 	size = mddev->dev_sectors >> conf->chunk_shift;
 	sector_div(size, fc);
 	size = size * conf->raid_disks;
@@ -2185,6 +2186,10 @@ static int run(mddev_t *mddev)
 		goto out_free_conf;
 	}
 
+	if (mddev->recovery_cp != MaxSector)
+		printk(KERN_NOTICE "raid10: %s is not clean"
+		       " -- starting background reconstruction\n",
+		       mdname(mddev));
 	printk(KERN_INFO
 		"raid10: raid set %s active with %d out of %d devices\n",
 		mdname(mddev), mddev->raid_disks - mddev->degraded,
@@ -2204,7 +2209,8 @@ static int run(mddev_t *mddev)
 	 * maybe...
 	 */
 	{
-		int stripe = conf->raid_disks * (mddev->chunk_size / PAGE_SIZE);
+		int stripe = conf->raid_disks *
+			((mddev->chunk_sectors << 9) / PAGE_SIZE);
 		stripe /= conf->near_copies;
 		if (mddev->queue->backing_dev_info.ra_pages < 2* stripe)
 			mddev->queue->backing_dev_info.ra_pages = 2* stripe;
@@ -2227,7 +2233,7 @@ out:
 
 static int stop(mddev_t *mddev)
 {
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 
 	raise_barrier(conf, 0);
 	lower_barrier(conf);
@@ -2245,7 +2251,7 @@ static int stop(mddev_t *mddev)
 
 static void raid10_quiesce(mddev_t *mddev, int state)
 {
-	conf_t *conf = mddev_to_conf(mddev);
+	conf_t *conf = mddev->private;
 
 	switch(state) {
 	case 1:

drivers/md/raid10.h

@@ -61,12 +61,6 @@ struct r10_private_data_s {
 
 typedef struct r10_private_data_s conf_t;
 
-/*
- * this is the only point in the RAID code where we violate
- * C type safety. mddev->private is an 'opaque' pointer.
- */
-#define mddev_to_conf(mddev) ((conf_t *) mddev->private)
-
 /*
  * this is our 'private' RAID10 bio.
  *

drivers/md/raid5.h

@@ -334,7 +334,8 @@ struct raid5_private_data {
 	struct hlist_head	*stripe_hashtbl;
 	mddev_t			*mddev;
 	struct disk_info	*spare;
-	int			chunk_size, level, algorithm;
+	int			chunk_sectors;
+	int			level, algorithm;
 	int			max_degraded;
 	int			raid_disks;
 	int			max_nr_stripes;
@@ -350,7 +351,8 @@ struct raid5_private_data {
 	 */
 	sector_t		reshape_safe;
 	int			previous_raid_disks;
-	int			prev_chunk, prev_algo;
+	int			prev_chunk_sectors;
+	int			prev_algo;
 	short			generation; /* increments with every reshape */
 	unsigned long		reshape_checkpoint; /* Time we last updated
 						     * metadata */
@@ -408,8 +410,6 @@ struct raid5_private_data {
 
 typedef struct raid5_private_data raid5_conf_t;
 
-#define mddev_to_conf(mddev) ((raid5_conf_t *) mddev->private)
-
 /*
  * Our supported algorithms
  */

include/linux/raid/md_p.h

@@ -232,7 +232,7 @@ struct mdp_superblock_1 {
 	__le64	reshape_position;	/* next address in array-space for reshape */
 	__le32	delta_disks;	/* change in number of raid_disks		*/
 	__le32	new_layout;	/* new layout					*/
-	__le32	new_chunk;	/* new chunk size (bytes)			*/
+	__le32	new_chunk;	/* new chunk size (512byte sectors)		*/
 	__u8	pad1[128-124];	/* set to 0 when written */
 
 	/* constant this-device information - 64 bytes */