md/raid1: Allocate spare to store replacement devices and their bios.

In RAID1, a replacement is much like a normal device, so we just
double the size of the relevant arrays and look at all possible
devices for reads and writes.

This means that the array looks like it is now double the size in some
way - we need to be careful about that.
In particular, we checking if the array is still degraded while
creating a recovery request we need to only consider the first 'half'
- i.e. the real (non-replacement) devices.

Signed-off-by: NeilBrown <neilb@suse.de>

diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index ab8113cff7c9f853cfc4b1a9fec9b86e6756fd48..e7768e37f36a363d810d22ee44f84f1aea7a8f4b 100644 (file)
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -135,7 +135,7 @@ out_free_pages:
                        put_page(r1_bio->bios[j]->bi_io_vec[i].bv_page);
        j = -1;
 out_free_bio:
-       while ( ++j < pi->raid_disks )
+       while (++j < pi->raid_disks)
                bio_put(r1_bio->bios[j]);
        r1bio_pool_free(r1_bio, data);
        return NULL;
@@ -164,7 +164,7 @@ static void put_all_bios(struct r1conf *conf, struct r1bio *r1_bio)
 {
        int i;
 
-       for (i = 0; i < conf->raid_disks; i++) {
+       for (i = 0; i < conf->raid_disks * 2; i++) {
                struct bio **bio = r1_bio->bios + i;
                if (!BIO_SPECIAL(*bio))
                        bio_put(*bio);
@@ -185,7 +185,7 @@ static void put_buf(struct r1bio *r1_bio)
        struct r1conf *conf = r1_bio->mddev->private;
        int i;
 
-       for (i=0; i<conf->raid_disks; i++) {
+       for (i = 0; i < conf->raid_disks * 2; i++) {
                struct bio *bio = r1_bio->bios[i];
                if (bio->bi_end_io)
                        rdev_dec_pending(conf->mirrors[i].rdev, r1_bio->mddev);
@@ -280,11 +280,11 @@ static int find_bio_disk(struct r1bio *r1_bio, struct bio *bio)
        struct r1conf *conf = r1_bio->mddev->private;
        int raid_disks = conf->raid_disks;
 
-       for (mirror = 0; mirror < raid_disks; mirror++)
+       for (mirror = 0; mirror < raid_disks * 2; mirror++)
                if (r1_bio->bios[mirror] == bio)
                        break;
 
-       BUG_ON(mirror == raid_disks);
+       BUG_ON(mirror == raid_disks * 2);
        update_head_pos(mirror, r1_bio);
 
        return mirror;
@@ -506,7 +506,7 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect
                start_disk = conf->last_used;
        }
 
-       for (i = 0 ; i < conf->raid_disks ; i++) {
+       for (i = 0 ; i < conf->raid_disks * 2 ; i++) {
                sector_t dist;
                sector_t first_bad;
                int bad_sectors;
@@ -975,7 +975,7 @@ read_again:
         */
        plugged = mddev_check_plugged(mddev);
 
-       disks = conf->raid_disks;
+       disks = conf->raid_disks * 2;
  retry_write:
        blocked_rdev = NULL;
        rcu_read_lock();
@@ -989,7 +989,8 @@ read_again:
                }
                r1_bio->bios[i] = NULL;
                if (!rdev || test_bit(Faulty, &rdev->flags)) {
-                       set_bit(R1BIO_Degraded, &r1_bio->state);
+                       if (i < conf->raid_disks)
+                               set_bit(R1BIO_Degraded, &r1_bio->state);
                        continue;
                }
 
@@ -1493,7 +1494,7 @@ static int fix_sync_read_error(struct r1bio *r1_bio)
                                }
                        }
                        d++;
-                       if (d == conf->raid_disks)
+                       if (d == conf->raid_disks * 2)
                                d = 0;
                } while (!success && d != r1_bio->read_disk);
 
@@ -1510,7 +1511,7 @@ static int fix_sync_read_error(struct r1bio *r1_bio)
                               mdname(mddev),
                               bdevname(bio->bi_bdev, b),
                               (unsigned long long)r1_bio->sector);
-                       for (d = 0; d < conf->raid_disks; d++) {
+                       for (d = 0; d < conf->raid_disks * 2; d++) {
                                rdev = conf->mirrors[d].rdev;
                                if (!rdev || test_bit(Faulty, &rdev->flags))
                                        continue;
@@ -1536,7 +1537,7 @@ static int fix_sync_read_error(struct r1bio *r1_bio)
                /* write it back and re-read */
                while (d != r1_bio->read_disk) {
                        if (d == 0)
-                               d = conf->raid_disks;
+                               d = conf->raid_disks * 2;
                        d--;
                        if (r1_bio->bios[d]->bi_end_io != end_sync_read)
                                continue;
@@ -1551,7 +1552,7 @@ static int fix_sync_read_error(struct r1bio *r1_bio)
                d = start;
                while (d != r1_bio->read_disk) {
                        if (d == 0)
-                               d = conf->raid_disks;
+                               d = conf->raid_disks * 2;
                        d--;
                        if (r1_bio->bios[d]->bi_end_io != end_sync_read)
                                continue;
@@ -1584,7 +1585,7 @@ static int process_checks(struct r1bio *r1_bio)
        int primary;
        int i;
 
-       for (primary = 0; primary < conf->raid_disks; primary++)
+       for (primary = 0; primary < conf->raid_disks * 2; primary++)
                if (r1_bio->bios[primary]->bi_end_io == end_sync_read &&
                    test_bit(BIO_UPTODATE, &r1_bio->bios[primary]->bi_flags)) {
                        r1_bio->bios[primary]->bi_end_io = NULL;
@@ -1592,7 +1593,7 @@ static int process_checks(struct r1bio *r1_bio)
                        break;
                }
        r1_bio->read_disk = primary;
-       for (i = 0; i < conf->raid_disks; i++) {
+       for (i = 0; i < conf->raid_disks * 2; i++) {
                int j;
                int vcnt = r1_bio->sectors >> (PAGE_SHIFT- 9);
                struct bio *pbio = r1_bio->bios[primary];
@@ -1656,7 +1657,7 @@ static void sync_request_write(struct mddev *mddev, struct r1bio *r1_bio)
 {
        struct r1conf *conf = mddev->private;
        int i;
-       int disks = conf->raid_disks;
+       int disks = conf->raid_disks * 2;
        struct bio *bio, *wbio;
 
        bio = r1_bio->bios[r1_bio->read_disk];
@@ -1737,7 +1738,7 @@ static void fix_read_error(struct r1conf *conf, int read_disk,
                                success = 1;
                        else {
                                d++;
-                               if (d == conf->raid_disks)
+                               if (d == conf->raid_disks * 2)
                                        d = 0;
                        }
                } while (!success && d != read_disk);
@@ -1753,7 +1754,7 @@ static void fix_read_error(struct r1conf *conf, int read_disk,
                start = d;
                while (d != read_disk) {
                        if (d==0)
-                               d = conf->raid_disks;
+                               d = conf->raid_disks * 2;
                        d--;
                        rdev = conf->mirrors[d].rdev;
                        if (rdev &&
@@ -1765,7 +1766,7 @@ static void fix_read_error(struct r1conf *conf, int read_disk,
                while (d != read_disk) {
                        char b[BDEVNAME_SIZE];
                        if (d==0)
-                               d = conf->raid_disks;
+                               d = conf->raid_disks * 2;
                        d--;
                        rdev = conf->mirrors[d].rdev;
                        if (rdev &&
@@ -1887,7 +1888,7 @@ static void handle_sync_write_finished(struct r1conf *conf, struct r1bio *r1_bio
 {
        int m;
        int s = r1_bio->sectors;
-       for (m = 0; m < conf->raid_disks ; m++) {
+       for (m = 0; m < conf->raid_disks * 2 ; m++) {
                struct md_rdev *rdev = conf->mirrors[m].rdev;
                struct bio *bio = r1_bio->bios[m];
                if (bio->bi_end_io == NULL)
@@ -1909,7 +1910,7 @@ static void handle_sync_write_finished(struct r1conf *conf, struct r1bio *r1_bio
 static void handle_write_finished(struct r1conf *conf, struct r1bio *r1_bio)
 {
        int m;
-       for (m = 0; m < conf->raid_disks ; m++)
+       for (m = 0; m < conf->raid_disks * 2 ; m++)
                if (r1_bio->bios[m] == IO_MADE_GOOD) {
                        struct md_rdev *rdev = conf->mirrors[m].rdev;
                        rdev_clear_badblocks(rdev,
@@ -2184,7 +2185,7 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, int *skipp
        r1_bio->state = 0;
        set_bit(R1BIO_IsSync, &r1_bio->state);
 
-       for (i=0; i < conf->raid_disks; i++) {
+       for (i = 0; i < conf->raid_disks * 2; i++) {
                struct md_rdev *rdev;
                bio = r1_bio->bios[i];
 
@@ -2203,7 +2204,8 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, int *skipp
                rdev = rcu_dereference(conf->mirrors[i].rdev);
                if (rdev == NULL ||
                    test_bit(Faulty, &rdev->flags)) {
-                       still_degraded = 1;
+                       if (i < conf->raid_disks)
+                               still_degraded = 1;
                } else if (!test_bit(In_sync, &rdev->flags)) {
                        bio->bi_rw = WRITE;
                        bio->bi_end_io = end_sync_write;
@@ -2254,7 +2256,7 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, int *skipp
                 * need to mark them bad on all write targets
                 */
                int ok = 1;
-               for (i = 0 ; i < conf->raid_disks ; i++)
+               for (i = 0 ; i < conf->raid_disks * 2 ; i++)
                        if (r1_bio->bios[i]->bi_end_io == end_sync_write) {
                                struct md_rdev *rdev =
                                        rcu_dereference(conf->mirrors[i].rdev);
@@ -2323,7 +2325,7 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, int *skipp
                                len = sync_blocks<<9;
                }
 
-               for (i=0 ; i < conf->raid_disks; i++) {
+               for (i = 0 ; i < conf->raid_disks * 2; i++) {
                        bio = r1_bio->bios[i];
                        if (bio->bi_end_io) {
                                page = bio->bi_io_vec[bio->bi_vcnt].bv_page;
@@ -2356,7 +2358,7 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, int *skipp
         */
        if (test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery)) {
                atomic_set(&r1_bio->remaining, read_targets);
-               for (i=0; i<conf->raid_disks; i++) {
+               for (i = 0; i < conf->raid_disks * 2; i++) {
                        bio = r1_bio->bios[i];
                        if (bio->bi_end_io == end_sync_read) {
                                md_sync_acct(bio->bi_bdev, nr_sectors);
@@ -2393,7 +2395,8 @@ static struct r1conf *setup_conf(struct mddev *mddev)
        if (!conf)
                goto abort;
 
-       conf->mirrors = kzalloc(sizeof(struct mirror_info)*mddev->raid_disks,
+       conf->mirrors = kzalloc(sizeof(struct mirror_info)
+                               * mddev->raid_disks * 2,
                                 GFP_KERNEL);
        if (!conf->mirrors)
                goto abort;
@@ -2405,7 +2408,7 @@ static struct r1conf *setup_conf(struct mddev *mddev)
        conf->poolinfo = kzalloc(sizeof(*conf->poolinfo), GFP_KERNEL);
        if (!conf->poolinfo)
                goto abort;
-       conf->poolinfo->raid_disks = mddev->raid_disks;
+       conf->poolinfo->raid_disks = mddev->raid_disks * 2;
        conf->r1bio_pool = mempool_create(NR_RAID1_BIOS, r1bio_pool_alloc,
                                          r1bio_pool_free,
                                          conf->poolinfo);
@@ -2438,7 +2441,7 @@ static struct r1conf *setup_conf(struct mddev *mddev)
        conf->recovery_disabled = mddev->recovery_disabled - 1;
 
        conf->last_used = -1;
-       for (i = 0; i < conf->raid_disks; i++) {
+       for (i = 0; i < conf->raid_disks * 2; i++) {
 
                disk = conf->mirrors + i;
 
@@ -2665,7 +2668,7 @@ static int raid1_reshape(struct mddev *mddev)
        if (!newpoolinfo)
                return -ENOMEM;
        newpoolinfo->mddev = mddev;
-       newpoolinfo->raid_disks = raid_disks;
+       newpoolinfo->raid_disks = raid_disks * 2;
 
        newpool = mempool_create(NR_RAID1_BIOS, r1bio_pool_alloc,
                                 r1bio_pool_free, newpoolinfo);
@@ -2673,7 +2676,8 @@ static int raid1_reshape(struct mddev *mddev)
                kfree(newpoolinfo);
                return -ENOMEM;
        }
-       newmirrors = kzalloc(sizeof(struct mirror_info) * raid_disks, GFP_KERNEL);
+       newmirrors = kzalloc(sizeof(struct mirror_info) * raid_disks * 2,
+                            GFP_KERNEL);
        if (!newmirrors) {
                kfree(newpoolinfo);
                mempool_destroy(newpool);

diff --git a/drivers/md/raid1.h b/drivers/md/raid1.h
index c732b6cce935283a6adb826a641bb023accf081e..80ded139314cf8a649729ce8e22915ca11bbd17f 100644 (file)
--- a/drivers/md/raid1.h
+++ b/drivers/md/raid1.h
@@ -12,6 +12,9 @@ struct mirror_info {
  * pool was allocated for, so they know how much to allocate and free.
  * mddev->raid_disks cannot be used, as it can change while a pool is active
  * These two datums are stored in a kmalloced struct.
+ * The 'raid_disks' here is twice the raid_disks in r1conf.
+ * This allows space for each 'real' device can have a replacement in the
+ * second half of the array.
  */
 
 struct pool_info {
@@ -21,7 +24,9 @@ struct pool_info {
 
 struct r1conf {
        struct mddev            *mddev;
-       struct mirror_info              *mirrors;
+       struct mirror_info      *mirrors;       /* twice 'raid_disks' to
+                                                * allow for replacements.
+                                                */
        int                     raid_disks;
 
        /* When choose the best device for a read (read_balance())