Merge branch 'v3.1-rc10' into for-3.2/core

Conflicts:
	block/blk-core.c
	include/linux/blkdev.h

Signed-off-by: Jens Axboe <axboe@kernel.dk>

11 files changed, 116 insertions, 58 deletions

diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index 49da55c1528a..8c2a000cf3f5 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -1698,6 +1698,8 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
         }
 
         ti->num_flush_requests = 1;
+        ti->discard_zeroes_data_unsupported = 1;
+
         return 0;
 
 bad:
diff --git a/drivers/md/dm-flakey.c b/drivers/md/dm-flakey.c
index 89f73ca22cfa..f84c08029b21 100644
--- a/drivers/md/dm-flakey.c
+++ b/drivers/md/dm-flakey.c
@@ -81,8 +81,10 @@ static int parse_features(struct dm_arg_set *as, struct flakey_c *fc,
                  * corrupt_bio_byte <Nth_byte> <direction> <value> <bio_flags>
                  */
                 if (!strcasecmp(arg_name, "corrupt_bio_byte")) {
-                        if (!argc)
+                        if (!argc) {
                                 ti->error = "Feature corrupt_bio_byte requires parameters";
+                                return -EINVAL;
+                        }
 
                         r = dm_read_arg(_args + 1, as, &fc->corrupt_bio_byte, &ti->error);
                         if (r)
diff --git a/drivers/md/dm-raid.c b/drivers/md/dm-raid.c
index a002dd85db1e..86df8b2cf927 100644
--- a/drivers/md/dm-raid.c
+++ b/drivers/md/dm-raid.c
@@ -449,7 +449,7 @@ static int parse_raid_params(struct raid_set *rs, char **argv,
                                 rs->ti->error = "write_mostly option is only valid for RAID1";
                                 return -EINVAL;
                         }
-                        if (value > rs->md.raid_disks) {
+                        if (value >= rs->md.raid_disks) {
                                 rs->ti->error = "Invalid write_mostly drive index given";
                                 return -EINVAL;
                         }
diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 986b8754bb08..bc04518e9d8b 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -1238,14 +1238,15 @@ static void dm_table_set_integrity(struct dm_table *t)
                 return;
 
         template_disk = dm_table_get_integrity_disk(t, true);
-        if (!template_disk &&
-            blk_integrity_is_initialized(dm_disk(t->md))) {
+        if (template_disk)
+                blk_integrity_register(dm_disk(t->md),
+                                       blk_get_integrity(template_disk));
+        else if (blk_integrity_is_initialized(dm_disk(t->md)))
                 DMWARN("%s: device no longer has a valid integrity profile",
                        dm_device_name(t->md));
-                return;
-        }
-        blk_integrity_register(dm_disk(t->md),
-                               blk_get_integrity(template_disk));
+        else
+                DMWARN("%s: unable to establish an integrity profile",
+                       dm_device_name(t->md));
 }
 
 static int device_flush_capable(struct dm_target *ti, struct dm_dev *dev,
@@ -1282,6 +1283,22 @@ static bool dm_table_supports_flush(struct dm_table *t, unsigned flush)
         return 0;
 }
 
+static bool dm_table_discard_zeroes_data(struct dm_table *t)
+{
+        struct dm_target *ti;
+        unsigned i = 0;
+
+        /* Ensure that all targets supports discard_zeroes_data. */
+        while (i < dm_table_get_num_targets(t)) {
+                ti = dm_table_get_target(t, i++);
+
+                if (ti->discard_zeroes_data_unsupported)
+                        return 0;
+        }
+
+        return 1;
+}
+
 void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
                                struct queue_limits *limits)
 {
@@ -1304,6 +1321,9 @@ void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
         }
         blk_queue_flush(q, flush);
 
+        if (!dm_table_discard_zeroes_data(t))
+                q->limits.discard_zeroes_data = 0;
+
         dm_table_set_integrity(t);
 
         /*
diff --git a/drivers/md/linear.h b/drivers/md/linear.h
index 0ce29b61605a..2f2da05b2ce9 100644
--- a/drivers/md/linear.h
+++ b/drivers/md/linear.h
@@ -10,9 +10,9 @@ typedef struct dev_info dev_info_t;
 
 struct linear_private_data
 {
+        struct rcu_head         rcu;
         sector_t                array_sectors;
         dev_info_t              disks[0];
-        struct rcu_head         rcu;
 };
 
 
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 5c2178562c96..8f52d4eb78a0 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -61,6 +61,11 @@
 static void autostart_arrays(int part);
 #endif
 
+/* pers_list is a list of registered personalities protected
+ * by pers_lock.
+ * pers_lock does extra service to protect accesses to
+ * mddev->thread when the mutex cannot be held.
+ */
 static LIST_HEAD(pers_list);
 static DEFINE_SPINLOCK(pers_lock);
 
@@ -735,7 +740,12 @@ static void mddev_unlock(mddev_t * mddev)
         } else
                 mutex_unlock(&mddev->reconfig_mutex);
 
+        /* was we've dropped the mutex we need a spinlock to
+         * make sur the thread doesn't disappear
+         */
+        spin_lock(&pers_lock);
         md_wakeup_thread(mddev->thread);
+        spin_unlock(&pers_lock);
 }
 
 static mdk_rdev_t * find_rdev_nr(mddev_t *mddev, int nr)
@@ -844,7 +854,7 @@ void md_super_write(mddev_t *mddev, mdk_rdev_t *rdev,
         bio->bi_end_io = super_written;
 
         atomic_inc(&mddev->pending_writes);
-        submit_bio(REQ_WRITE | REQ_SYNC | REQ_FLUSH | REQ_FUA, bio);
+        submit_bio(WRITE_FLUSH_FUA, bio);
 }
 
 void md_super_wait(mddev_t *mddev)
@@ -1134,8 +1144,11 @@ static int super_90_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version
                         ret = 0;
         }
         rdev->sectors = rdev->sb_start;
+        /* Limit to 4TB as metadata cannot record more than that */
+        if (rdev->sectors >= (2ULL << 32))
+                rdev->sectors = (2ULL << 32) - 2;
 
-        if (rdev->sectors < sb->size * 2 && sb->level > 1)
+        if (rdev->sectors < ((sector_t)sb->size) * 2 && sb->level >= 1)
                 /* "this cannot possibly happen" ... */
                 ret = -EINVAL;
 
@@ -1169,7 +1182,7 @@ static int super_90_validate(mddev_t *mddev, mdk_rdev_t *rdev)
                 mddev->clevel[0] = 0;
                 mddev->layout = sb->layout;
                 mddev->raid_disks = sb->raid_disks;
-                mddev->dev_sectors = sb->size * 2;
+                mddev->dev_sectors = ((sector_t)sb->size) * 2;
                 mddev->events = ev1;
                 mddev->bitmap_info.offset = 0;
                 mddev->bitmap_info.default_offset = MD_SB_BYTES >> 9;
@@ -1411,6 +1424,11 @@ super_90_rdev_size_change(mdk_rdev_t *rdev, sector_t num_sectors)
         rdev->sb_start = calc_dev_sboffset(rdev);
         if (!num_sectors || num_sectors > rdev->sb_start)
                 num_sectors = rdev->sb_start;
+        /* Limit to 4TB as metadata cannot record more than that.
+         * 4TB == 2^32 KB, or 2*2^32 sectors.
+         */
+        if (num_sectors >= (2ULL << 32))
+                num_sectors = (2ULL << 32) - 2;
         md_super_write(rdev->mddev, rdev, rdev->sb_start, rdev->sb_size,
                        rdev->sb_page);
         md_super_wait(rdev->mddev);
@@ -1734,6 +1752,11 @@ static void super_1_sync(mddev_t *mddev, mdk_rdev_t *rdev)
         sb->level = cpu_to_le32(mddev->level);
         sb->layout = cpu_to_le32(mddev->layout);
 
+        if (test_bit(WriteMostly, &rdev->flags))
+                sb->devflags |= WriteMostly1;
+        else
+                sb->devflags &= ~WriteMostly1;
+
         if (mddev->bitmap && mddev->bitmap_info.file == NULL) {
                 sb->bitmap_offset = cpu_to_le32((__u32)mddev->bitmap_info.offset);
                 sb->feature_map = cpu_to_le32(MD_FEATURE_BITMAP_OFFSET);
@@ -2557,7 +2580,10 @@ state_store(mdk_rdev_t *rdev, const char *buf, size_t len)
         int err = -EINVAL;
         if (cmd_match(buf, "faulty") && rdev->mddev->pers) {
                 md_error(rdev->mddev, rdev);
-                err = 0;
+                if (test_bit(Faulty, &rdev->flags))
+                        err = 0;
+                else
+                        err = -EBUSY;
         } else if (cmd_match(buf, "remove")) {
                 if (rdev->raid_disk >= 0)
                         err = -EBUSY;
@@ -2580,7 +2606,7 @@ state_store(mdk_rdev_t *rdev, const char *buf, size_t len)
                 err = 0;
         } else if (cmd_match(buf, "-blocked")) {
                 if (!test_bit(Faulty, &rdev->flags) &&
-                    test_bit(BlockedBadBlocks, &rdev->flags)) {
+                    rdev->badblocks.unacked_exist) {
                         /* metadata handler doesn't understand badblocks,
                          * so we need to fail the device
                          */
@@ -5979,6 +6005,8 @@ static int set_disk_faulty(mddev_t *mddev, dev_t dev)
                 return -ENODEV;
 
         md_error(mddev, rdev);
+        if (!test_bit(Faulty, &rdev->flags))
+                return -EBUSY;
         return 0;
 }
 
@@ -6407,11 +6435,18 @@ mdk_thread_t *md_register_thread(void (*run) (mddev_t *), mddev_t *mddev,
         return thread;
 }
 
-void md_unregister_thread(mdk_thread_t *thread)
+void md_unregister_thread(mdk_thread_t **threadp)
 {
+        mdk_thread_t *thread = *threadp;
         if (!thread)
                 return;
         dprintk("interrupting MD-thread pid %d\n", task_pid_nr(thread->tsk));
+        /* Locking ensures that mddev_unlock does not wake_up a
+         * non-existent thread
+         */
+        spin_lock(&pers_lock);
+        *threadp = NULL;
+        spin_unlock(&pers_lock);
 
         kthread_stop(thread->tsk);
         kfree(thread);
@@ -7318,8 +7353,7 @@ static void reap_sync_thread(mddev_t *mddev)
         mdk_rdev_t *rdev;
 
         /* resync has finished, collect result */
-        md_unregister_thread(mddev->sync_thread);
-        mddev->sync_thread = NULL;
+        md_unregister_thread(&mddev->sync_thread);
         if (!test_bit(MD_RECOVERY_INTR, &mddev->recovery) &&
             !test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery)) {
                 /* success...*/
diff --git a/drivers/md/md.h b/drivers/md/md.h
index bd47847cf7ca..1509a3eb9ae1 100644
--- a/drivers/md/md.h
+++ b/drivers/md/md.h
@@ -560,7 +560,7 @@ extern int register_md_personality(struct mdk_personality *p);
 extern int unregister_md_personality(struct mdk_personality *p);
 extern mdk_thread_t * md_register_thread(void (*run) (mddev_t *mddev),
                                 mddev_t *mddev, const char *name);
-extern void md_unregister_thread(mdk_thread_t *thread);
+extern void md_unregister_thread(mdk_thread_t **threadp);
 extern void md_wakeup_thread(mdk_thread_t *thread);
 extern void md_check_recovery(mddev_t *mddev);
 extern void md_write_start(mddev_t *mddev, struct bio *bi);
diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c
index 407cb5691425..618dd9e22513 100644
--- a/drivers/md/multipath.c
+++ b/drivers/md/multipath.c
@@ -514,8 +514,7 @@ static int multipath_stop (mddev_t *mddev)
 {
         multipath_conf_t *conf = mddev->private;
 
-        md_unregister_thread(mddev->thread);
-        mddev->thread = NULL;
+        md_unregister_thread(&mddev->thread);
         blk_sync_queue(mddev->queue); /* the unplug fn references 'conf'*/
         mempool_destroy(conf->pool);
         kfree(conf->multipaths);
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 97f2a5f977b1..d4ddfa627301 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -1099,12 +1099,11 @@ read_again:
                 bio_list_add(&conf->pending_bio_list, mbio);
                 spin_unlock_irqrestore(&conf->device_lock, flags);
         }
-        r1_bio_write_done(r1_bio);
-
-        /* In case raid1d snuck in to freeze_array */
-        wake_up(&conf->wait_barrier);
-
+        /* Mustn't call r1_bio_write_done before this next test,
+         * as it could result in the bio being freed.
+         */
         if (sectors_handled < (bio->bi_size >> 9)) {
+                r1_bio_write_done(r1_bio);
                 /* We need another r1_bio.  It has already been counted
                  * in bio->bi_phys_segments
                  */
@@ -1117,6 +1116,11 @@ read_again:
                 goto retry_write;
         }
 
+        r1_bio_write_done(r1_bio);
+
+        /* In case raid1d snuck in to freeze_array */
+        wake_up(&conf->wait_barrier);
+
         if (do_sync || !bitmap || !plugged)
                 md_wakeup_thread(mddev->thread);
 }
@@ -2556,8 +2560,7 @@ static int stop(mddev_t *mddev)
         raise_barrier(conf);
         lower_barrier(conf);
 
-        md_unregister_thread(mddev->thread);
-        mddev->thread = NULL;
+        md_unregister_thread(&mddev->thread);
         if (conf->r1bio_pool)
                 mempool_destroy(conf->r1bio_pool);
         kfree(conf->mirrors);
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 04b625e1cb60..ea5fc0b6a84c 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -337,6 +337,21 @@ static void close_write(r10bio_t *r10_bio)
         md_write_end(r10_bio->mddev);
 }
 
+static void one_write_done(r10bio_t *r10_bio)
+{
+        if (atomic_dec_and_test(&r10_bio->remaining)) {
+                if (test_bit(R10BIO_WriteError, &r10_bio->state))
+                        reschedule_retry(r10_bio);
+                else {
+                        close_write(r10_bio);
+                        if (test_bit(R10BIO_MadeGood, &r10_bio->state))
+                                reschedule_retry(r10_bio);
+                        else
+                                raid_end_bio_io(r10_bio);
+                }
+        }
+}
+
 static void raid10_end_write_request(struct bio *bio, int error)
 {
         int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
@@ -387,17 +402,7 @@ static void raid10_end_write_request(struct bio *bio, int error)
          * Let's see if all mirrored write operations have finished
          * already.
          */
-        if (atomic_dec_and_test(&r10_bio->remaining)) {
-                if (test_bit(R10BIO_WriteError, &r10_bio->state))
-                        reschedule_retry(r10_bio);
-                else {
-                        close_write(r10_bio);
-                        if (test_bit(R10BIO_MadeGood, &r10_bio->state))
-                                reschedule_retry(r10_bio);
-                        else
-                                raid_end_bio_io(r10_bio);
-                }
-        }
+        one_write_done(r10_bio);
         if (dec_rdev)
                 rdev_dec_pending(conf->mirrors[dev].rdev, conf->mddev);
 }
@@ -1125,20 +1130,12 @@ retry_write:
                 spin_unlock_irqrestore(&conf->device_lock, flags);
         }
 
-        if (atomic_dec_and_test(&r10_bio->remaining)) {
-                /* This matches the end of raid10_end_write_request() */
-                bitmap_endwrite(r10_bio->mddev->bitmap, r10_bio->sector,
-                                r10_bio->sectors,
-                                !test_bit(R10BIO_Degraded, &r10_bio->state),
-                                0);
-                md_write_end(mddev);
-                raid_end_bio_io(r10_bio);
-        }
-
-        /* In case raid10d snuck in to freeze_array */
-        wake_up(&conf->wait_barrier);
+        /* Don't remove the bias on 'remaining' (one_write_done) until
+         * after checking if we need to go around again.
+         */
 
         if (sectors_handled < (bio->bi_size >> 9)) {
+                one_write_done(r10_bio);
                 /* We need another r10_bio.  It has already been counted
                  * in bio->bi_phys_segments.
                  */
@@ -1152,6 +1149,10 @@ retry_write:
                 r10_bio->state = 0;
                 goto retry_write;
         }
+        one_write_done(r10_bio);
+
+        /* In case raid10d snuck in to freeze_array */
+        wake_up(&conf->wait_barrier);
 
         if (do_sync || !mddev->bitmap || !plugged)
                 md_wakeup_thread(mddev->thread);
@@ -2951,7 +2952,7 @@ static int run(mddev_t *mddev)
         return 0;
 
 out_free_conf:
-        md_unregister_thread(mddev->thread);
+        md_unregister_thread(&mddev->thread);
         if (conf->r10bio_pool)
                 mempool_destroy(conf->r10bio_pool);
         safe_put_page(conf->tmppage);
@@ -2969,8 +2970,7 @@ static int stop(mddev_t *mddev)
         raise_barrier(conf, 0);
         lower_barrier(conf);
 
-        md_unregister_thread(mddev->thread);
-        mddev->thread = NULL;
+        md_unregister_thread(&mddev->thread);
         blk_sync_queue(mddev->queue); /* the unplug fn references 'conf'*/
         if (conf->r10bio_pool)
                 mempool_destroy(conf->r10bio_pool);
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 96b7f6a1b6f2..83f2c44e170f 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -3336,7 +3336,7 @@ static void handle_stripe(struct stripe_head *sh)
 
 finish:
         /* wait for this device to become unblocked */
-        if (unlikely(s.blocked_rdev))
+        if (conf->mddev->external && unlikely(s.blocked_rdev))
                 md_wait_for_blocked_rdev(s.blocked_rdev, conf->mddev);
 
         if (s.handle_bad_blocks)
@@ -4939,8 +4939,7 @@ static int run(mddev_t *mddev)
 
         return 0;
 abort:
-        md_unregister_thread(mddev->thread);
-        mddev->thread = NULL;
+        md_unregister_thread(&mddev->thread);
         if (conf) {
                 print_raid5_conf(conf);
                 free_conf(conf);
@@ -4954,8 +4953,7 @@ static int stop(mddev_t *mddev)
 {
         raid5_conf_t *conf = mddev->private;
 
-        md_unregister_thread(mddev->thread);
-        mddev->thread = NULL;
+        md_unregister_thread(&mddev->thread);
         if (mddev->queue)
                 mddev->queue->backing_dev_info.congested_fn = NULL;
         free_conf(conf);