block: Require subsystems to explicitly allocate bio_set integrity mempool

MD and DM create a new bio_set for every metadevice. Each bio_set has an
integrity mempool attached regardless of whether the metadevice is
capable of passing integrity metadata. This is a waste of memory.

Instead we defer the allocation decision to MD and DM since we know at
metadevice creation time whether integrity passthrough is needed or not.

Automatic integrity mempool allocation can then be removed from
bioset_create() and we make an explicit integrity allocation for the
fs_bio_set.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reported-by: Zdenek Kabelac <zkabelac@redhat.com>
Acked-by: Mike Snitzer <snizer@redhat.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>

9 files changed, 35 insertions, 19 deletions

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index f50a7b9..416d4e2 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -55,6 +55,7 @@ struct dm_table {
 	struct dm_target *targets;
 
 	unsigned discards_supported:1;
+	unsigned integrity_supported:1;
 
 	/*
 	 * Indicates the rw permissions for the new logical
@@ -859,7 +860,7 @@ int dm_table_alloc_md_mempools(struct dm_table *t)
 		return -EINVAL;
 	}
 
-	t->mempools = dm_alloc_md_mempools(type);
+	t->mempools = dm_alloc_md_mempools(type, t->integrity_supported);
 	if (!t->mempools)
 		return -ENOMEM;
 
@@ -935,8 +936,10 @@ static int dm_table_prealloc_integrity(struct dm_table *t, struct mapped_device
 	struct dm_dev_internal *dd;
 
 	list_for_each_entry(dd, devices, list)
-		if (bdev_get_integrity(dd->dm_dev.bdev))
+		if (bdev_get_integrity(dd->dm_dev.bdev)) {
+			t->integrity_supported = 1;
 			return blk_integrity_register(dm_disk(md), NULL);
+		}
 
 	return 0;
 }
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index d22b990..8882070 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -2620,9 +2620,10 @@ int dm_noflush_suspending(struct dm_target *ti)
 }
 EXPORT_SYMBOL_GPL(dm_noflush_suspending);
 
-struct dm_md_mempools *dm_alloc_md_mempools(unsigned type)
+struct dm_md_mempools *dm_alloc_md_mempools(unsigned type, unsigned integrity)
 {
 	struct dm_md_mempools *pools = kmalloc(sizeof(*pools), GFP_KERNEL);
+	unsigned int pool_size = (type == DM_TYPE_BIO_BASED) ? 16 : MIN_IOS;
 
 	if (!pools)
 		return NULL;
@@ -2639,13 +2640,18 @@ struct dm_md_mempools *dm_alloc_md_mempools(unsigned type)
 	if (!pools->tio_pool)
 		goto free_io_pool_and_out;
 
-	pools->bs = (type == DM_TYPE_BIO_BASED) ?
-		    bioset_create(16, 0) : bioset_create(MIN_IOS, 0);
+	pools->bs = bioset_create(pool_size, 0);
 	if (!pools->bs)
 		goto free_tio_pool_and_out;
 
+	if (integrity && bioset_integrity_create(pools->bs, pool_size))
+		goto free_bioset_and_out;
+
 	return pools;
 
+free_bioset_and_out:
+	bioset_free(pools->bs);
+
 free_tio_pool_and_out:
 	mempool_destroy(pools->tio_pool);
 
diff --git a/drivers/md/dm.h b/drivers/md/dm.h
index 0c2dd5f..1aaf167 100644
--- a/drivers/md/dm.h
+++ b/drivers/md/dm.h
@@ -149,7 +149,7 @@ void dm_kcopyd_exit(void);
 /*
  * Mempool operations
  */
-struct dm_md_mempools *dm_alloc_md_mempools(unsigned type);
+struct dm_md_mempools *dm_alloc_md_mempools(unsigned type, unsigned integrity);
 void dm_free_md_mempools(struct dm_md_mempools *pools);
 
 #endif
diff --git a/drivers/md/linear.c b/drivers/md/linear.c
index 338804f8..abfb59a 100644
--- a/drivers/md/linear.c
+++ b/drivers/md/linear.c
@@ -210,8 +210,7 @@ static int linear_run (mddev_t *mddev)
 	blk_queue_merge_bvec(mddev->queue, linear_mergeable_bvec);
 	mddev->queue->backing_dev_info.congested_fn = linear_congested;
 	mddev->queue->backing_dev_info.congested_data = mddev;
-	md_integrity_register(mddev);
-	return 0;
+	return md_integrity_register(mddev);
 }
 
 static void free_conf(struct rcu_head *head)
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 86ba66c..1876761 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -1803,8 +1803,12 @@ int md_integrity_register(mddev_t *mddev)
 			mdname(mddev));
 		return -EINVAL;
 	}
-	printk(KERN_NOTICE "md: data integrity on %s enabled\n",
-		mdname(mddev));
+	printk(KERN_NOTICE "md: data integrity enabled on %s\n", mdname(mddev));
+	if (bioset_integrity_create(mddev->bio_set, BIO_POOL_SIZE)) {
+		printk(KERN_ERR "md: failed to create integrity pool for %s\n",
+		       mdname(mddev));
+		return -EINVAL;
+	}
 	return 0;
 }
 EXPORT_SYMBOL(md_integrity_register);
diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c
index 5e694b1..c358909 100644
--- a/drivers/md/multipath.c
+++ b/drivers/md/multipath.c
@@ -315,7 +315,7 @@ static int multipath_remove_disk(mddev_t *mddev, int number)
 			p->rdev = rdev;
 			goto abort;
 		}
-		md_integrity_register(mddev);
+		err = md_integrity_register(mddev);
 	}
 abort:
 
@@ -489,7 +489,10 @@ static int multipath_run (mddev_t *mddev)
 
 	mddev->queue->backing_dev_info.congested_fn = multipath_congested;
 	mddev->queue->backing_dev_info.congested_data = mddev;
-	md_integrity_register(mddev);
+
+	if (md_integrity_register(mddev))
+		goto out_free_conf;
+
 	return 0;
 
 out_free_conf:
diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c
index 95916fd..e86bf36 100644
--- a/drivers/md/raid0.c
+++ b/drivers/md/raid0.c
@@ -379,8 +379,7 @@ static int raid0_run(mddev_t *mddev)
 
 	blk_queue_merge_bvec(mddev->queue, raid0_mergeable_bvec);
 	dump_zones(mddev);
-	md_integrity_register(mddev);
-	return 0;
+	return md_integrity_register(mddev);
 }
 
 static int raid0_stop(mddev_t *mddev)
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 8f34ad5..c2a21ae5 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -1132,7 +1132,7 @@ static int raid1_remove_disk(mddev_t *mddev, int number)
 			p->rdev = rdev;
 			goto abort;
 		}
-		md_integrity_register(mddev);
+		err = md_integrity_register(mddev);
 	}
 abort:
 
@@ -2017,8 +2017,7 @@ static int run(mddev_t *mddev)
 
 	mddev->queue->backing_dev_info.congested_fn = raid1_congested;
 	mddev->queue->backing_dev_info.congested_data = mddev;
-	md_integrity_register(mddev);
-	return 0;
+	return md_integrity_register(mddev);
 }
 
 static int stop(mddev_t *mddev)
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index c0d0f5f..f7b6237 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -1188,7 +1188,7 @@ static int raid10_remove_disk(mddev_t *mddev, int number)
 			p->rdev = rdev;
 			goto abort;
 		}
-		md_integrity_register(mddev);
+		err = md_integrity_register(mddev);
 	}
 abort:
 
@@ -2343,7 +2343,10 @@ static int run(mddev_t *mddev)
 
 	if (conf->near_copies < conf->raid_disks)
 		blk_queue_merge_bvec(mddev->queue, raid10_mergeable_bvec);
-	md_integrity_register(mddev);
+
+	if (md_integrity_register(mddev))
+		goto out_free_conf;
+
 	return 0;
 
 out_free_conf: