Merge tag 'dm-4.4-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm

Pull device mapper updates from Mike Snitzer:
 "Smaller set of DM changes for this merge.  I've based these changes on
  Jens' for-4.4/reservations branch because the associated DM changes
  required it.

   - Revert a dm-multipath change that caused a regression for
     unprivledged users (e.g. kvm guests) that issued ioctls when a
     multipath device had no available paths.

   - Include Christoph's refactoring of DM's ioctl handling and add
     support for passing through persistent reservations with DM
     multipath.

   - All other changes are very simple cleanups"

* tag 'dm-4.4-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
  dm switch: simplify conditional in alloc_region_table()
  dm delay: document that offsets are specified in sectors
  dm delay: capitalize the start of an delay_ctr() error message
  dm delay: Use DM_MAPIO macros instead of open-coded equivalents
  dm linear: remove redundant target name from error messages
  dm persistent data: eliminate unnecessary return values
  dm: eliminate unused "bioset" process for each bio-based DM device
  dm: convert ffs to __ffs
  dm: drop NULL test before kmem_cache_destroy() and mempool_destroy()
  dm: add support for passing through persistent reservations
  dm: refactor ioctl handling
  Revert "dm mpath: fix stalls when handling invalid ioctls"
  dm: initialize non-blk-mq queue data before queue is used

1 files changed, 182 insertions, 27 deletions

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 485760ebba76..32440ad5f684 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -24,6 +24,7 @@
 #include <linux/ktime.h>
 #include <linux/elevator.h> /* for rq_end_sector() */
 #include <linux/blk-mq.h>
+#include <linux/pr.h>
 
 #include <trace/events/block.h>
 
@@ -555,18 +556,16 @@ static int dm_blk_getgeo(struct block_device *bdev, struct hd_geometry *geo)
 	return dm_get_geometry(md, geo);
 }
 
-static int dm_blk_ioctl(struct block_device *bdev, fmode_t mode,
-			unsigned int cmd, unsigned long arg)
+static int dm_get_live_table_for_ioctl(struct mapped_device *md,
+		struct dm_target **tgt, struct block_device **bdev,
+		fmode_t *mode, int *srcu_idx)
 {
-	struct mapped_device *md = bdev->bd_disk->private_data;
-	int srcu_idx;
 	struct dm_table *map;
-	struct dm_target *tgt;
-	int r = -ENOTTY;
+	int r;
 
 retry:
-	map = dm_get_live_table(md, &srcu_idx);
-
+	r = -ENOTTY;
+	map = dm_get_live_table(md, srcu_idx);
 	if (!map || !dm_table_get_size(map))
 		goto out;
 
@@ -574,8 +573,9 @@ retry:
 	if (dm_table_get_num_targets(map) != 1)
 		goto out;
 
-	tgt = dm_table_get_target(map, 0);
-	if (!tgt->type->ioctl)
+	*tgt = dm_table_get_target(map, 0);
+
+	if (!(*tgt)->type->prepare_ioctl)
 		goto out;
 
 	if (dm_suspended_md(md)) {
@@ -583,16 +583,46 @@ retry:
 		goto out;
 	}
 
-	r = tgt->type->ioctl(tgt, cmd, arg);
+	r = (*tgt)->type->prepare_ioctl(*tgt, bdev, mode);
+	if (r < 0)
+		goto out;
 
-out:
-	dm_put_live_table(md, srcu_idx);
+	return r;
 
+out:
+	dm_put_live_table(md, *srcu_idx);
 	if (r == -ENOTCONN) {
 		msleep(10);
 		goto retry;
 	}
+	return r;
+}
+
+static int dm_blk_ioctl(struct block_device *bdev, fmode_t mode,
+			unsigned int cmd, unsigned long arg)
+{
+	struct mapped_device *md = bdev->bd_disk->private_data;
+	struct dm_target *tgt;
+	int srcu_idx, r;
+
+	r = dm_get_live_table_for_ioctl(md, &tgt, &bdev, &mode, &srcu_idx);
+	if (r < 0)
+		return r;
 
+	if (r > 0) {
+		/*
+		 * Target determined this ioctl is being issued against
+		 * a logical partition of the parent bdev; so extra
+		 * validation is needed.
+		 */
+		r = scsi_verify_blk_ioctl(NULL, cmd);
+		if (r)
+			goto out;
+	}
+
+	r =  __blkdev_driver_ioctl(bdev, mode, cmd, arg);
+out:
+	dm_put_live_table(md, srcu_idx);
 	return r;
 }
 
@@ -1734,8 +1764,6 @@ static void dm_make_request(struct request_queue *q, struct bio *bio)
 
 	map = dm_get_live_table(md, &srcu_idx);
 
-	blk_queue_split(q, &bio, q->bio_split);
-
 	generic_start_io_acct(rw, bio_sectors(bio), &dm_disk(md)->part0);
 
 	/* if we're suspended, we have to queue this io for later */
@@ -2198,6 +2226,13 @@ static void dm_init_md_queue(struct mapped_device *md)
 	 * This queue is new, so no concurrency on the queue_flags.
 	 */
 	queue_flag_clear_unlocked(QUEUE_FLAG_STACKABLE, md->queue);
+
+	/*
+	 * Initialize data that will only be used by a non-blk-mq DM queue
+	 * - must do so here (in alloc_dev callchain) before queue is used
+	 */
+	md->queue->queuedata = md;
+	md->queue->backing_dev_info.congested_data = md;
 }
 
 static void dm_init_old_md_queue(struct mapped_device *md)
@@ -2208,10 +2243,7 @@ static void dm_init_old_md_queue(struct mapped_device *md)
 	/*
 	 * Initialize aspects of queue that aren't relevant for blk-mq
 	 */
-	md->queue->queuedata = md;
 	md->queue->backing_dev_info.congested_fn = dm_any_congested;
-	md->queue->backing_dev_info.congested_data = md;
-
 	blk_queue_bounce_limit(md->queue, BLK_BOUNCE_ANY);
 }
 
@@ -2221,10 +2253,8 @@ static void cleanup_mapped_device(struct mapped_device *md)
 		destroy_workqueue(md->wq);
 	if (md->kworker_task)
 		kthread_stop(md->kworker_task);
-	if (md->io_pool)
-		mempool_destroy(md->io_pool);
-	if (md->rq_pool)
-		mempool_destroy(md->rq_pool);
+	mempool_destroy(md->io_pool);
+	mempool_destroy(md->rq_pool);
 	if (md->bs)
 		bioset_free(md->bs);
 
@@ -2759,6 +2789,12 @@ int dm_setup_md_queue(struct mapped_device *md)
 	case DM_TYPE_BIO_BASED:
 		dm_init_old_md_queue(md);
 		blk_queue_make_request(md->queue, dm_make_request);
+		/*
+		 * DM handles splitting bios as needed.  Free the bio_split bioset
+		 * since it won't be used (saves 1 process per bio-based DM device).
+		 */
+		bioset_free(md->queue->bio_split);
+		md->queue->bio_split = NULL;
 		break;
 	}
 
@@ -3505,11 +3541,8 @@ void dm_free_md_mempools(struct dm_md_mempools *pools)
 	if (!pools)
 		return;
 
-	if (pools->io_pool)
-		mempool_destroy(pools->io_pool);
-
-	if (pools->rq_pool)
-		mempool_destroy(pools->rq_pool);
+	mempool_destroy(pools->io_pool);
+	mempool_destroy(pools->rq_pool);
 
 	if (pools->bs)
 		bioset_free(pools->bs);
@@ -3517,11 +3550,133 @@ void dm_free_md_mempools(struct dm_md_mempools *pools)
 	kfree(pools);
 }
 
+static int dm_pr_register(struct block_device *bdev, u64 old_key, u64 new_key,
+		u32 flags)
+{
+	struct mapped_device *md = bdev->bd_disk->private_data;
+	const struct pr_ops *ops;
+	struct dm_target *tgt;
+	fmode_t mode;
+	int srcu_idx, r;
+
+	r = dm_get_live_table_for_ioctl(md, &tgt, &bdev, &mode, &srcu_idx);
+	if (r < 0)
+		return r;
+
+	ops = bdev->bd_disk->fops->pr_ops;
+	if (ops && ops->pr_register)
+		r = ops->pr_register(bdev, old_key, new_key, flags);
+	else
+		r = -EOPNOTSUPP;
+
+	dm_put_live_table(md, srcu_idx);
+	return r;
+}
+
+static int dm_pr_reserve(struct block_device *bdev, u64 key, enum pr_type type,
+		u32 flags)
+{
+	struct mapped_device *md = bdev->bd_disk->private_data;
+	const struct pr_ops *ops;
+	struct dm_target *tgt;
+	fmode_t mode;
+	int srcu_idx, r;
+
+	r = dm_get_live_table_for_ioctl(md, &tgt, &bdev, &mode, &srcu_idx);
+	if (r < 0)
+		return r;
+
+	ops = bdev->bd_disk->fops->pr_ops;
+	if (ops && ops->pr_reserve)
+		r = ops->pr_reserve(bdev, key, type, flags);
+	else
+		r = -EOPNOTSUPP;
+
+	dm_put_live_table(md, srcu_idx);
+	return r;
+}
+
+static int dm_pr_release(struct block_device *bdev, u64 key, enum pr_type type)
+{
+	struct mapped_device *md = bdev->bd_disk->private_data;
+	const struct pr_ops *ops;
+	struct dm_target *tgt;
+	fmode_t mode;
+	int srcu_idx, r;
+
+	r = dm_get_live_table_for_ioctl(md, &tgt, &bdev, &mode, &srcu_idx);
+	if (r < 0)
+		return r;
+
+	ops = bdev->bd_disk->fops->pr_ops;
+	if (ops && ops->pr_release)
+		r = ops->pr_release(bdev, key, type);
+	else
+		r = -EOPNOTSUPP;
+
+	dm_put_live_table(md, srcu_idx);
+	return r;
+}
+
+static int dm_pr_preempt(struct block_device *bdev, u64 old_key, u64 new_key,
+		enum pr_type type, bool abort)
+{
+	struct mapped_device *md = bdev->bd_disk->private_data;
+	const struct pr_ops *ops;
+	struct dm_target *tgt;
+	fmode_t mode;
+	int srcu_idx, r;
+
+	r = dm_get_live_table_for_ioctl(md, &tgt, &bdev, &mode, &srcu_idx);
+	if (r < 0)
+		return r;
+
+	ops = bdev->bd_disk->fops->pr_ops;
+	if (ops && ops->pr_preempt)
+		r = ops->pr_preempt(bdev, old_key, new_key, type, abort);
+	else
+		r = -EOPNOTSUPP;
+
+	dm_put_live_table(md, srcu_idx);
+	return r;
+}
+
+static int dm_pr_clear(struct block_device *bdev, u64 key)
+{
+	struct mapped_device *md = bdev->bd_disk->private_data;
+	const struct pr_ops *ops;
+	struct dm_target *tgt;
+	fmode_t mode;
+	int srcu_idx, r;
+
+	r = dm_get_live_table_for_ioctl(md, &tgt, &bdev, &mode, &srcu_idx);
+	if (r < 0)
+		return r;
+
+	ops = bdev->bd_disk->fops->pr_ops;
+	if (ops && ops->pr_clear)
+		r = ops->pr_clear(bdev, key);
+	else
+		r = -EOPNOTSUPP;
+
+	dm_put_live_table(md, srcu_idx);
+	return r;
+}
+
+static const struct pr_ops dm_pr_ops = {
+	.pr_register	= dm_pr_register,
+	.pr_reserve	= dm_pr_reserve,
+	.pr_release	= dm_pr_release,
+	.pr_preempt	= dm_pr_preempt,
+	.pr_clear	= dm_pr_clear,
+};
+
 static const struct block_device_operations dm_blk_dops = {
 	.open = dm_blk_open,
 	.release = dm_blk_close,
 	.ioctl = dm_blk_ioctl,
 	.getgeo = dm_blk_getgeo,
+	.pr_ops = &dm_pr_ops,
 	.owner = THIS_MODULE
 };