Merge branch 'for-linus' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:
 "A collection of fixes from the last few weeks that should go into the
  current series.  This contains:

   - Various fixes for the per-blkcg policy data, fixing regressions
     since 4.1.  From Arianna and Tejun

   - Code cleanup for bcache closure macros from me.  Really just
     flushing this out, it's been sitting in another branch for months

   - FIELD_SIZEOF cleanup from Maninder Singh

   - bio integrity oops fix from Mike

   - Timeout regression fix for blk-mq from Ming Lei"

* 'for-linus' of git://git.kernel.dk/linux-block:
  blk-mq: set default timeout as 30 seconds
  NVMe: Reread partitions on metadata formats
  bcache: don't embed 'return' statements in closure macros
  blkcg: fix blkcg_policy_data allocation bug
  blkcg: implement all_blkcgs list
  blkcg: blkcg_css_alloc() should grab blkcg_pol_mutex while iterating blkcg_policy[]
  blkcg: allow blkcg_pol_mutex to be grabbed from cgroup [file] methods
  block/blk-cgroup.c: free per-blkcg data when freeing the blkcg
  block: use FIELD_SIZEOF to calculate size of a field
  bio integrity: do not assume bio_integrity_pool exists if bioset exists

10 files changed, 113 insertions, 79 deletions

diff --git a/block/bio-integrity.c b/block/bio-integrity.c
index 0436c21db7f2..719b7152aed1 100644
--- a/block/bio-integrity.c
+++ b/block/bio-integrity.c
@@ -51,7 +51,7 @@ struct bio_integrity_payload *bio_integrity_alloc(struct bio *bio,
 	unsigned long idx = BIO_POOL_NONE;
 	unsigned inline_vecs;
 
-	if (!bs) {
+	if (!bs || !bs->bio_integrity_pool) {
 		bip = kmalloc(sizeof(struct bio_integrity_payload) +
 			      sizeof(struct bio_vec) * nr_vecs, gfp_mask);
 		inline_vecs = nr_vecs;
@@ -104,7 +104,7 @@ void bio_integrity_free(struct bio *bio)
 		kfree(page_address(bip->bip_vec->bv_page) +
 		      bip->bip_vec->bv_offset);
 
-	if (bs) {
+	if (bs && bs->bio_integrity_pool) {
 		if (bip->bip_slab != BIO_POOL_NONE)
 			bvec_free(bs->bvec_integrity_pool, bip->bip_vec,
 				  bip->bip_slab);
diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c
index 9f97da52d006..9da02c021ebe 100644
--- a/block/blk-cgroup.c
+++ b/block/blk-cgroup.c
@@ -29,6 +29,14 @@
 
 #define MAX_KEY_LEN 100
 
+/*
+ * blkcg_pol_mutex protects blkcg_policy[] and policy [de]activation.
+ * blkcg_pol_register_mutex nests outside of it and synchronizes entire
+ * policy [un]register operations including cgroup file additions /
+ * removals.  Putting cgroup file registration outside blkcg_pol_mutex
+ * allows grabbing it from cgroup callbacks.
+ */
+static DEFINE_MUTEX(blkcg_pol_register_mutex);
 static DEFINE_MUTEX(blkcg_pol_mutex);
 
 struct blkcg blkcg_root;
@@ -38,6 +46,8 @@ struct cgroup_subsys_state * const blkcg_root_css = &blkcg_root.css;
 
 static struct blkcg_policy *blkcg_policy[BLKCG_MAX_POLS];
 
+static LIST_HEAD(all_blkcgs);		/* protected by blkcg_pol_mutex */
+
 static bool blkcg_policy_enabled(struct request_queue *q,
 				 const struct blkcg_policy *pol)
 {
@@ -453,20 +463,7 @@ static int blkcg_reset_stats(struct cgroup_subsys_state *css,
 	struct blkcg_gq *blkg;
 	int i;
 
-	/*
-	 * XXX: We invoke cgroup_add/rm_cftypes() under blkcg_pol_mutex
-	 * which ends up putting cgroup's internal cgroup_tree_mutex under
-	 * it; however, cgroup_tree_mutex is nested above cgroup file
-	 * active protection and grabbing blkcg_pol_mutex from a cgroup
-	 * file operation creates a possible circular dependency.  cgroup
-	 * internal locking is planned to go through further simplification
-	 * and this issue should go away soon.  For now, let's trylock
-	 * blkcg_pol_mutex and restart the write on failure.
-	 *
-	 * http://lkml.kernel.org/g/5363C04B.4010400@oracle.com
-	 */
-	if (!mutex_trylock(&blkcg_pol_mutex))
-		return restart_syscall();
+	mutex_lock(&blkcg_pol_mutex);
 	spin_lock_irq(&blkcg->lock);
 
 	/*
@@ -822,8 +819,17 @@ static void blkcg_css_free(struct cgroup_subsys_state *css)
 {
 	struct blkcg *blkcg = css_to_blkcg(css);
 
-	if (blkcg != &blkcg_root)
+	mutex_lock(&blkcg_pol_mutex);
+	list_del(&blkcg->all_blkcgs_node);
+	mutex_unlock(&blkcg_pol_mutex);
+
+	if (blkcg != &blkcg_root) {
+		int i;
+
+		for (i = 0; i < BLKCG_MAX_POLS; i++)
+			kfree(blkcg->pd[i]);
 		kfree(blkcg);
+	}
 }
 
 static struct cgroup_subsys_state *
@@ -833,6 +839,8 @@ blkcg_css_alloc(struct cgroup_subsys_state *parent_css)
 	struct cgroup_subsys_state *ret;
 	int i;
 
+	mutex_lock(&blkcg_pol_mutex);
+
 	if (!parent_css) {
 		blkcg = &blkcg_root;
 		goto done;
@@ -875,14 +883,17 @@ done:
 #ifdef CONFIG_CGROUP_WRITEBACK
 	INIT_LIST_HEAD(&blkcg->cgwb_list);
 #endif
+	list_add_tail(&blkcg->all_blkcgs_node, &all_blkcgs);
+
+	mutex_unlock(&blkcg_pol_mutex);
 	return &blkcg->css;
 
 free_pd_blkcg:
 	for (i--; i >= 0; i--)
 		kfree(blkcg->pd[i]);
-
 free_blkcg:
 	kfree(blkcg);
+	mutex_unlock(&blkcg_pol_mutex);
 	return ret;
 }
 
@@ -1037,10 +1048,8 @@ int blkcg_activate_policy(struct request_queue *q,
 			  const struct blkcg_policy *pol)
 {
 	LIST_HEAD(pds);
-	LIST_HEAD(cpds);
 	struct blkcg_gq *blkg;
 	struct blkg_policy_data *pd, *nd;
-	struct blkcg_policy_data *cpd, *cnd;
 	int cnt = 0, ret;
 
 	if (blkcg_policy_enabled(q, pol))
@@ -1053,10 +1062,7 @@ int blkcg_activate_policy(struct request_queue *q,
 		cnt++;
 	spin_unlock_irq(q->queue_lock);
 
-	/*
-	 * Allocate per-blkg and per-blkcg policy data
-	 * for all existing blkgs.
-	 */
+	/* allocate per-blkg policy data for all existing blkgs */
 	while (cnt--) {
 		pd = kzalloc_node(pol->pd_size, GFP_KERNEL, q->node);
 		if (!pd) {
@@ -1064,15 +1070,6 @@ int blkcg_activate_policy(struct request_queue *q,
 			goto out_free;
 		}
 		list_add_tail(&pd->alloc_node, &pds);
-
-		if (!pol->cpd_size)
-			continue;
-		cpd = kzalloc_node(pol->cpd_size, GFP_KERNEL, q->node);
-		if (!cpd) {
-			ret = -ENOMEM;
-			goto out_free;
-		}
-		list_add_tail(&cpd->alloc_node, &cpds);
 	}
 
 	/*
@@ -1082,32 +1079,17 @@ int blkcg_activate_policy(struct request_queue *q,
 	spin_lock_irq(q->queue_lock);
 
 	list_for_each_entry(blkg, &q->blkg_list, q_node) {
-		if (WARN_ON(list_empty(&pds)) ||
-		    WARN_ON(pol->cpd_size && list_empty(&cpds))) {
+		if (WARN_ON(list_empty(&pds))) {
 			/* umm... this shouldn't happen, just abort */
 			ret = -ENOMEM;
 			goto out_unlock;
 		}
-		cpd = list_first_entry(&cpds, struct blkcg_policy_data,
-				       alloc_node);
-		list_del_init(&cpd->alloc_node);
 		pd = list_first_entry(&pds, struct blkg_policy_data, alloc_node);
 		list_del_init(&pd->alloc_node);
 
 		/* grab blkcg lock too while installing @pd on @blkg */
 		spin_lock(&blkg->blkcg->lock);
 
-		if (!pol->cpd_size)
-			goto no_cpd;
-		if (!blkg->blkcg->pd[pol->plid]) {
-			/* Per-policy per-blkcg data */
-			blkg->blkcg->pd[pol->plid] = cpd;
-			cpd->plid = pol->plid;
-			pol->cpd_init_fn(blkg->blkcg);
-		} else { /* must free it as it has already been extracted */
-			kfree(cpd);
-		}
-no_cpd:
 		blkg->pd[pol->plid] = pd;
 		pd->blkg = blkg;
 		pd->plid = pol->plid;
@@ -1124,8 +1106,6 @@ out_free:
 	blk_queue_bypass_end(q);
 	list_for_each_entry_safe(pd, nd, &pds, alloc_node)
 		kfree(pd);
-	list_for_each_entry_safe(cpd, cnd, &cpds, alloc_node)
-		kfree(cpd);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(blkcg_activate_policy);
@@ -1162,8 +1142,6 @@ void blkcg_deactivate_policy(struct request_queue *q,
 
 		kfree(blkg->pd[pol->plid]);
 		blkg->pd[pol->plid] = NULL;
-		kfree(blkg->blkcg->pd[pol->plid]);
-		blkg->blkcg->pd[pol->plid] = NULL;
 
 		spin_unlock(&blkg->blkcg->lock);
 	}
@@ -1182,11 +1160,13 @@ EXPORT_SYMBOL_GPL(blkcg_deactivate_policy);
  */
 int blkcg_policy_register(struct blkcg_policy *pol)
 {
+	struct blkcg *blkcg;
 	int i, ret;
 
 	if (WARN_ON(pol->pd_size < sizeof(struct blkg_policy_data)))
 		return -EINVAL;
 
+	mutex_lock(&blkcg_pol_register_mutex);
 	mutex_lock(&blkcg_pol_mutex);
 
 	/* find an empty slot */
@@ -1195,19 +1175,49 @@ int blkcg_policy_register(struct blkcg_policy *pol)
 		if (!blkcg_policy[i])
 			break;
 	if (i >= BLKCG_MAX_POLS)
-		goto out_unlock;
+		goto err_unlock;
 
-	/* register and update blkgs */
+	/* register @pol */
 	pol->plid = i;
-	blkcg_policy[i] = pol;
+	blkcg_policy[pol->plid] = pol;
+
+	/* allocate and install cpd's */
+	if (pol->cpd_size) {
+		list_for_each_entry(blkcg, &all_blkcgs, all_blkcgs_node) {
+			struct blkcg_policy_data *cpd;
+
+			cpd = kzalloc(pol->cpd_size, GFP_KERNEL);
+			if (!cpd) {
+				mutex_unlock(&blkcg_pol_mutex);
+				goto err_free_cpds;
+			}
+
+			blkcg->pd[pol->plid] = cpd;
+			cpd->plid = pol->plid;
+			pol->cpd_init_fn(blkcg);
+		}
+	}
+
+	mutex_unlock(&blkcg_pol_mutex);
 
 	/* everything is in place, add intf files for the new policy */
 	if (pol->cftypes)
 		WARN_ON(cgroup_add_legacy_cftypes(&blkio_cgrp_subsys,
 						  pol->cftypes));
-	ret = 0;
-out_unlock:
+	mutex_unlock(&blkcg_pol_register_mutex);
+	return 0;
+
+err_free_cpds:
+	if (pol->cpd_size) {
+		list_for_each_entry(blkcg, &all_blkcgs, all_blkcgs_node) {
+			kfree(blkcg->pd[pol->plid]);
+			blkcg->pd[pol->plid] = NULL;
+		}
+	}
+	blkcg_policy[pol->plid] = NULL;
+err_unlock:
 	mutex_unlock(&blkcg_pol_mutex);
+	mutex_unlock(&blkcg_pol_register_mutex);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(blkcg_policy_register);
@@ -1220,7 +1230,9 @@ EXPORT_SYMBOL_GPL(blkcg_policy_register);
  */
 void blkcg_policy_unregister(struct blkcg_policy *pol)
 {
-	mutex_lock(&blkcg_pol_mutex);
+	struct blkcg *blkcg;
+
+	mutex_lock(&blkcg_pol_register_mutex);
 
 	if (WARN_ON(blkcg_policy[pol->plid] != pol))
 		goto out_unlock;
@@ -1229,9 +1241,19 @@ void blkcg_policy_unregister(struct blkcg_policy *pol)
 	if (pol->cftypes)
 		cgroup_rm_cftypes(pol->cftypes);
 
-	/* unregister and update blkgs */
+	/* remove cpds and unregister */
+	mutex_lock(&blkcg_pol_mutex);
+
+	if (pol->cpd_size) {
+		list_for_each_entry(blkcg, &all_blkcgs, all_blkcgs_node) {
+			kfree(blkcg->pd[pol->plid]);
+			blkcg->pd[pol->plid] = NULL;
+		}
+	}
 	blkcg_policy[pol->plid] = NULL;
-out_unlock:
+
 	mutex_unlock(&blkcg_pol_mutex);
+out_unlock:
+	mutex_unlock(&blkcg_pol_register_mutex);
 }
 EXPORT_SYMBOL_GPL(blkcg_policy_unregister);
diff --git a/block/blk-core.c b/block/blk-core.c
index 82819e68f58b..627ed0c593fb 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -3370,7 +3370,7 @@ EXPORT_SYMBOL(blk_post_runtime_resume);
 int __init blk_dev_init(void)
 {
 	BUILD_BUG_ON(__REQ_NR_BITS > 8 *
-			sizeof(((struct request *)0)->cmd_flags));
+			FIELD_SIZEOF(struct request, cmd_flags));
 
 	/* used for unplugging and affects IO latency/throughput - HIGHPRI */
 	kblockd_workqueue = alloc_workqueue("kblockd",
diff --git a/block/blk-mq.c b/block/blk-mq.c
index f53779692c77..7d842db59699 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -1998,7 +1998,7 @@ struct request_queue *blk_mq_init_allocated_queue(struct blk_mq_tag_set *set,
 		goto err_hctxs;
 
 	setup_timer(&q->timeout, blk_mq_rq_timer, (unsigned long) q);
-	blk_queue_rq_timeout(q, set->timeout ? set->timeout : 30000);
+	blk_queue_rq_timeout(q, set->timeout ? set->timeout : 30 * HZ);
 
 	q->nr_queues = nr_cpu_ids;
 	q->nr_hw_queues = set->nr_hw_queues;
diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c
index d1d6141920d3..7920c2741b47 100644
--- a/drivers/block/nvme-core.c
+++ b/drivers/block/nvme-core.c
@@ -2108,8 +2108,17 @@ static void nvme_alloc_ns(struct nvme_dev *dev, unsigned nsid)
 		goto out_free_disk;
 
 	add_disk(ns->disk);
-	if (ns->ms)
-		revalidate_disk(ns->disk);
+	if (ns->ms) {
+		struct block_device *bd = bdget_disk(ns->disk, 0);
+		if (!bd)
+			return;
+		if (blkdev_get(bd, FMODE_READ, NULL)) {
+			bdput(bd);
+			return;
+		}
+		blkdev_reread_part(bd);
+		blkdev_put(bd, FMODE_READ);
+	}
 	return;
  out_free_disk:
 	kfree(disk);
diff --git a/drivers/md/bcache/closure.h b/drivers/md/bcache/closure.h
index a08e3eeac3c5..79a6d63e8ed3 100644
--- a/drivers/md/bcache/closure.h
+++ b/drivers/md/bcache/closure.h
@@ -320,7 +320,6 @@ static inline void closure_wake_up(struct closure_waitlist *list)
 do {									\
 	set_closure_fn(_cl, _fn, _wq);					\
 	closure_sub(_cl, CLOSURE_RUNNING + 1);				\
-	return;								\
 } while (0)
 
 /**
@@ -349,7 +348,6 @@ do {									\
 do {									\
 	set_closure_fn(_cl, _fn, _wq);					\
 	closure_queue(_cl);						\
-	return;								\
 } while (0)
 
 /**
@@ -365,7 +363,6 @@ do {									\
 do {									\
 	set_closure_fn(_cl, _destructor, NULL);				\
 	closure_sub(_cl, CLOSURE_RUNNING - CLOSURE_DESTRUCTOR + 1);	\
-	return;								\
 } while (0)
 
 /**
diff --git a/drivers/md/bcache/io.c b/drivers/md/bcache/io.c
index cb64e64a4789..bf6a9ca18403 100644
--- a/drivers/md/bcache/io.c
+++ b/drivers/md/bcache/io.c
@@ -105,6 +105,7 @@ void bch_generic_make_request(struct bio *bio, struct bio_split_pool *p)
 	} while (n != bio);
 
 	continue_at(&s->cl, bch_bio_submit_split_done, NULL);
+	return;
 submit:
 	generic_make_request(bio);
 }
diff --git a/drivers/md/bcache/journal.c b/drivers/md/bcache/journal.c
index ce64fc851251..418607a6ba33 100644
--- a/drivers/md/bcache/journal.c
+++ b/drivers/md/bcache/journal.c
@@ -592,12 +592,14 @@ static void journal_write_unlocked(struct closure *cl)
 
 	if (!w->need_write) {
 		closure_return_with_destructor(cl, journal_write_unlock);
+		return;
 	} else if (journal_full(&c->journal)) {
 		journal_reclaim(c);
 		spin_unlock(&c->journal.lock);
 
 		btree_flush_write(c);
 		continue_at(cl, journal_write, system_wq);
+		return;
 	}
 
 	c->journal.blocks_free -= set_blocks(w->data, block_bytes(c));
diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c
index 4afb2d26b148..f292790997d7 100644
--- a/drivers/md/bcache/request.c
+++ b/drivers/md/bcache/request.c
@@ -88,8 +88,10 @@ static void bch_data_insert_keys(struct closure *cl)
 	if (journal_ref)
 		atomic_dec_bug(journal_ref);
 
-	if (!op->insert_data_done)
+	if (!op->insert_data_done) {
 		continue_at(cl, bch_data_insert_start, op->wq);
+		return;
+	}
 
 	bch_keylist_free(&op->insert_keys);
 	closure_return(cl);
@@ -216,8 +218,10 @@ static void bch_data_insert_start(struct closure *cl)
 		/* 1 for the device pointer and 1 for the chksum */
 		if (bch_keylist_realloc(&op->insert_keys,
 					3 + (op->csum ? 1 : 0),
-					op->c))
+					op->c)) {
 			continue_at(cl, bch_data_insert_keys, op->wq);
+			return;
+		}
 
 		k = op->insert_keys.top;
 		bkey_init(k);
@@ -255,6 +259,7 @@ static void bch_data_insert_start(struct closure *cl)
 
 	op->insert_data_done = true;
 	continue_at(cl, bch_data_insert_keys, op->wq);
+	return;
 err:
 	/* bch_alloc_sectors() blocks if s->writeback = true */
 	BUG_ON(op->writeback);
@@ -576,8 +581,10 @@ static void cache_lookup(struct closure *cl)
 	ret = bch_btree_map_keys(&s->op, s->iop.c,
 				 &KEY(s->iop.inode, bio->bi_iter.bi_sector, 0),
 				 cache_lookup_fn, MAP_END_KEY);
-	if (ret == -EAGAIN)
+	if (ret == -EAGAIN) {
 		continue_at(cl, cache_lookup, bcache_wq);
+		return;
+	}
 
 	closure_return(cl);
 }
@@ -1085,6 +1092,7 @@ static void flash_dev_make_request(struct request_queue *q, struct bio *bio)
 		continue_at_nobarrier(&s->cl,
 				      flash_dev_nodata,
 				      bcache_wq);
+		return;
 	} else if (rw) {
 		bch_keybuf_check_overlapping(&s->iop.c->moving_gc_keys,
 					&KEY(d->id, bio->bi_iter.bi_sector, 0),
diff --git a/include/linux/blk-cgroup.h b/include/linux/blk-cgroup.h
index 58cfab80dd70..1b62d768c7df 100644
--- a/include/linux/blk-cgroup.h
+++ b/include/linux/blk-cgroup.h
@@ -47,6 +47,7 @@ struct blkcg {
 
 	struct blkcg_policy_data	*pd[BLKCG_MAX_POLS];
 
+	struct list_head		all_blkcgs_node;
 #ifdef CONFIG_CGROUP_WRITEBACK
 	struct list_head		cgwb_list;
 #endif
@@ -88,18 +89,12 @@ struct blkg_policy_data {
  * Policies that need to keep per-blkcg data which is independent
  * from any request_queue associated to it must specify its size
  * with the cpd_size field of the blkcg_policy structure and
- * embed a blkcg_policy_data in it. blkcg core allocates
- * policy-specific per-blkcg structures lazily the first time
- * they are actually needed, so it handles them together with
- * blkgs. cpd_init() is invoked to let each policy handle
- * per-blkcg data.
+ * embed a blkcg_policy_data in it.  cpd_init() is invoked to let
+ * each policy handle per-blkcg data.
  */
 struct blkcg_policy_data {
 	/* the policy id this per-policy data belongs to */
 	int				plid;
-
-	/* used during policy activation */
-	struct list_head		alloc_node;
 };
 
 /* association between a blk cgroup and a request queue */