md: add possibility to change data-offset for devices.

When reshaping we can avoid costly intermediate backup by
changing the 'start' address of the array on the device
(if there is enough room).

So as a first step, allow such a change to be requested
through sysfs, and recorded in v1.x metadata.

(As we didn't previous check that all 'pad' fields were zero,
 we need a new FEATURE flag for this.
 A (belatedly) check that all remaining 'pad' fields are
 zero to avoid a repeat of this)

The new data offset must be requested separately for each device.
This allows each to have a different change in the data offset.
This is not likely to be used often but as data_offset can be
set per-device, new_data_offset should be too.

This patch also removes the 'acknowledged' arg to rdev_set_badblocks as
it is never used and never will be.  At the same time we add a new
arg ('in_new') which is currently always zero but will be used more
soon.

When a reshape finishes we will need to update the data_offset
and rdev->sectors.  So provide an exported function to do that.
Signed-off-by: NeilBrown <neilb@suse.de>

drivers/md/md.h

@@ -55,6 +55,7 @@ struct md_rdev {
 	int		sb_loaded;
 	__u64		sb_events;
 	sector_t	data_offset;	/* start of data in array */
+	sector_t	new_data_offset;/* only relevant while reshaping */
 	sector_t 	sb_start;	/* offset of the super block (in 512byte sectors) */
 	int		sb_size;	/* bytes in the superblock */
 	int		preferred_minor;	/* autorun support */
@@ -193,8 +194,9 @@ static inline int is_badblock(struct md_rdev *rdev, sector_t s, int sectors,
 	return 0;
 }
 extern int rdev_set_badblocks(struct md_rdev *rdev, sector_t s, int sectors,
-			      int acknowledged);
-extern int rdev_clear_badblocks(struct md_rdev *rdev, sector_t s, int sectors);
+			      int is_new);
+extern int rdev_clear_badblocks(struct md_rdev *rdev, sector_t s, int sectors,
+				int is_new);
 extern void md_ack_all_badblocks(struct badblocks *bb);
 
 struct mddev {
@@ -592,6 +594,7 @@ extern void md_write_start(struct mddev *mddev, struct bio *bi);
 extern void md_write_end(struct mddev *mddev);
 extern void md_done_sync(struct mddev *mddev, int blocks, int ok);
 extern void md_error(struct mddev *mddev, struct md_rdev *rdev);
+extern void md_finish_reshape(struct mddev *mddev);
 
 extern int mddev_congested(struct mddev *mddev, int bits);
 extern void md_flush_request(struct mddev *mddev, struct bio *bio);

drivers/md/raid1.c

@@ -2024,7 +2024,7 @@ static void handle_sync_write_finished(struct r1conf *conf, struct r1bio *r1_bio
 			continue;
 		if (test_bit(BIO_UPTODATE, &bio->bi_flags) &&
 		    test_bit(R1BIO_MadeGood, &r1_bio->state)) {
-			rdev_clear_badblocks(rdev, r1_bio->sector, s);
+			rdev_clear_badblocks(rdev, r1_bio->sector, s, 0);
 		}
 		if (!test_bit(BIO_UPTODATE, &bio->bi_flags) &&
 		    test_bit(R1BIO_WriteError, &r1_bio->state)) {
@@ -2044,7 +2044,7 @@ static void handle_write_finished(struct r1conf *conf, struct r1bio *r1_bio)
 			struct md_rdev *rdev = conf->mirrors[m].rdev;
 			rdev_clear_badblocks(rdev,
 					     r1_bio->sector,
-					     r1_bio->sectors);
+					     r1_bio->sectors, 0);
 			rdev_dec_pending(rdev, conf->mddev);
 		} else if (r1_bio->bios[m] != NULL) {
 			/* This drive got a write error.  We need to

drivers/md/raid10.c

@@ -2480,7 +2480,7 @@ static void handle_write_completed(struct r10conf *conf, struct r10bio *r10_bio)
 				rdev_clear_badblocks(
 					rdev,
 					r10_bio->devs[m].addr,
-					r10_bio->sectors);
+					r10_bio->sectors, 0);
 			} else {
 				if (!rdev_set_badblocks(
 					    rdev,
@@ -2496,7 +2496,7 @@ static void handle_write_completed(struct r10conf *conf, struct r10bio *r10_bio)
 				rdev_clear_badblocks(
 					rdev,
 					r10_bio->devs[m].addr,
-					r10_bio->sectors);
+					r10_bio->sectors, 0);
 			} else {
 				if (!rdev_set_badblocks(
 					    rdev,
@@ -2515,7 +2515,7 @@ static void handle_write_completed(struct r10conf *conf, struct r10bio *r10_bio)
 				rdev_clear_badblocks(
 					rdev,
 					r10_bio->devs[m].addr,
-					r10_bio->sectors);
+					r10_bio->sectors, 0);
 				rdev_dec_pending(rdev, conf->mddev);
 			} else if (bio != NULL &&
 				   !test_bit(BIO_UPTODATE, &bio->bi_flags)) {
@@ -2532,7 +2532,7 @@ static void handle_write_completed(struct r10conf *conf, struct r10bio *r10_bio)
 				rdev_clear_badblocks(
 					rdev,
 					r10_bio->devs[m].addr,
-					r10_bio->sectors);
+					r10_bio->sectors, 0);
 				rdev_dec_pending(rdev, conf->mddev);
 			}
 		}

drivers/md/raid5.c

@@ -3561,7 +3561,7 @@ static void handle_stripe(struct stripe_head *sh)
 			if (test_and_clear_bit(R5_MadeGood, &dev->flags)) {
 				rdev = conf->disks[i].rdev;
 				rdev_clear_badblocks(rdev, sh->sector,
-						     STRIPE_SECTORS);
+						     STRIPE_SECTORS, 0);
 				rdev_dec_pending(rdev, conf->mddev);
 			}
 			if (test_and_clear_bit(R5_MadeGoodRepl, &dev->flags)) {
@@ -3570,7 +3570,7 @@ static void handle_stripe(struct stripe_head *sh)
 					/* rdev have been moved down */
 					rdev = conf->disks[i].rdev;
 				rdev_clear_badblocks(rdev, sh->sector,
-						     STRIPE_SECTORS);
+						     STRIPE_SECTORS, 0);
 				rdev_dec_pending(rdev, conf->mddev);
 			}
 		}
@@ -5505,10 +5505,14 @@ static int raid5_start_reshape(struct mddev *mddev)
 	if (!check_stripe_cache(mddev))
 		return -ENOSPC;
 
-	rdev_for_each(rdev, mddev)
+	rdev_for_each(rdev, mddev) {
+		/* Don't support changing data_offset yet */
+		if (rdev->new_data_offset != rdev->data_offset)
+			return -EINVAL;
 		if (!test_bit(In_sync, &rdev->flags)
 		    && !test_bit(Faulty, &rdev->flags))
 			spares++;
+	}
 
 	if (spares - mddev->degraded < mddev->delta_disks - conf->max_degraded)
 		/* Not enough devices even to make a degraded array

include/linux/raid/md_p.h

@@ -233,7 +233,10 @@ struct mdp_superblock_1 {
 	__le32	delta_disks;	/* change in number of raid_disks		*/
 	__le32	new_layout;	/* new layout					*/
 	__le32	new_chunk;	/* new chunk size (512byte sectors)		*/
-	__u8	pad1[128-124];	/* set to 0 when written */
+	__le32  new_offset;	/* signed number to add to data_offset in new
+				 * layout.  0 == no-change.  This can be
+				 * different on each device in the array.
+				 */
 
 	/* constant this-device information - 64 bytes */
 	__le64	data_offset;	/* sector start of data, often 0 */
@@ -285,11 +288,14 @@ struct mdp_superblock_1 {
 					    * of devices, but is going
 					    * backwards anyway.
 					    */
+#define	MD_FEATURE_NEW_OFFSET		64 /* new_offset must be honoured */
 #define	MD_FEATURE_ALL			(MD_FEATURE_BITMAP_OFFSET	\
 					|MD_FEATURE_RECOVERY_OFFSET	\
 					|MD_FEATURE_RESHAPE_ACTIVE	\
 					|MD_FEATURE_BAD_BLOCKS		\
 					|MD_FEATURE_REPLACEMENT		\
-					|MD_FEATURE_RESHAPE_BACKWARDS)
+					|MD_FEATURE_RESHAPE_BACKWARDS	\
+					|MD_FEATURE_NEW_OFFSET		\
+					)
 
 #endif