block: use the holder as indication for exclusive opens

The current interface for exclusive opens is rather confusing as it
requires both the FMODE_EXCL flag and a holder.  Remove the need to pass
FMODE_EXCL and just key off the exclusive open off a non-NULL holder.

For blkdev_put this requires adding the holder argument, which provides
better debug checking that only the holder actually releases the hold,
but at the same time allows removing the now superfluous mode argument.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Acked-by: Christian Brauner <brauner@kernel.org>
Acked-by: David Sterba <dsterba@suse.com>		[btrfs]
Acked-by: Jack Wang <jinpu.wang@ionos.com>		[rnbd]
Link: https://lore.kernel.org/r/20230608110258.189493-16-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

3 files changed, 30 insertions, 29 deletions

diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c
index 4a2aed047aec..7022fea396f2 100644
--- a/drivers/md/bcache/super.c
+++ b/drivers/md/bcache/super.c
@@ -1369,7 +1369,7 @@ static void cached_dev_free(struct closure *cl)
 		put_page(virt_to_page(dc->sb_disk));
 
 	if (!IS_ERR_OR_NULL(dc->bdev))
-		blkdev_put(dc->bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL);
+		blkdev_put(dc->bdev, bcache_kobj);
 
 	wake_up(&unregister_wait);
 
@@ -2218,7 +2218,7 @@ void bch_cache_release(struct kobject *kobj)
 		put_page(virt_to_page(ca->sb_disk));
 
 	if (!IS_ERR_OR_NULL(ca->bdev))
-		blkdev_put(ca->bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL);
+		blkdev_put(ca->bdev, bcache_kobj);
 
 	kfree(ca);
 	module_put(THIS_MODULE);
@@ -2359,7 +2359,7 @@ static int register_cache(struct cache_sb *sb, struct cache_sb_disk *sb_disk,
 		 * call blkdev_put() to bdev in bch_cache_release(). So we
 		 * explicitly call blkdev_put() here.
 		 */
-		blkdev_put(bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL);
+		blkdev_put(bdev, bcache_kobj);
 		if (ret == -ENOMEM)
 			err = "cache_alloc(): -ENOMEM";
 		else if (ret == -EPERM)
@@ -2461,7 +2461,7 @@ static void register_bdev_worker(struct work_struct *work)
 	if (!dc) {
 		fail = true;
 		put_page(virt_to_page(args->sb_disk));
-		blkdev_put(args->bdev, FMODE_READ | FMODE_WRITE | FMODE_EXCL);
+		blkdev_put(args->bdev, bcache_kobj);
 		goto out;
 	}
 
@@ -2491,7 +2491,7 @@ static void register_cache_worker(struct work_struct *work)
 	if (!ca) {
 		fail = true;
 		put_page(virt_to_page(args->sb_disk));
-		blkdev_put(args->bdev, FMODE_READ | FMODE_WRITE | FMODE_EXCL);
+		blkdev_put(args->bdev, bcache_kobj);
 		goto out;
 	}
 
@@ -2558,8 +2558,7 @@ static ssize_t register_bcache(struct kobject *k, struct kobj_attribute *attr,
 
 	ret = -EINVAL;
 	err = "failed to open device";
-	bdev = blkdev_get_by_path(strim(path),
-				  FMODE_READ|FMODE_WRITE|FMODE_EXCL,
+	bdev = blkdev_get_by_path(strim(path), FMODE_READ | FMODE_WRITE,
 				  bcache_kobj, NULL);
 	if (IS_ERR(bdev)) {
 		if (bdev == ERR_PTR(-EBUSY)) {
@@ -2648,7 +2647,7 @@ async_done:
 out_put_sb_page:
 	put_page(virt_to_page(sb_disk));
 out_blkdev_put:
-	blkdev_put(bdev, FMODE_READ | FMODE_WRITE | FMODE_EXCL);
+	blkdev_put(bdev, register_bcache);
 out_free_sb:
 	kfree(sb);
 out_free_path:
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 246b8f028a98..b16e37362c5a 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -746,7 +746,7 @@ static struct table_device *open_table_device(struct mapped_device *md,
 		return ERR_PTR(-ENOMEM);
 	refcount_set(&td->count, 1);
 
-	bdev = blkdev_get_by_dev(dev, mode | FMODE_EXCL, _dm_claim_ptr, NULL);
+	bdev = blkdev_get_by_dev(dev, mode, _dm_claim_ptr, NULL);
 	if (IS_ERR(bdev)) {
 		r = PTR_ERR(bdev);
 		goto out_free_td;
@@ -771,7 +771,7 @@ static struct table_device *open_table_device(struct mapped_device *md,
 	return td;
 
 out_blkdev_put:
-	blkdev_put(bdev, mode | FMODE_EXCL);
+	blkdev_put(bdev, _dm_claim_ptr);
 out_free_td:
 	kfree(td);
 	return ERR_PTR(r);
@@ -784,7 +784,7 @@ static void close_table_device(struct table_device *td, struct mapped_device *md
 {
 	if (md->disk->slave_dir)
 		bd_unlink_disk_holder(td->dm_dev.bdev, md->disk);
-	blkdev_put(td->dm_dev.bdev, td->dm_dev.mode | FMODE_EXCL);
+	blkdev_put(td->dm_dev.bdev, _dm_claim_ptr);
 	put_dax(td->dm_dev.dax_dev);
 	list_del(&td->list);
 	kfree(td);
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 159197dd7b6d..dad4a5539f9f 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -2449,7 +2449,10 @@ static void rdev_delayed_delete(struct work_struct *ws)
 
 void md_autodetect_dev(dev_t dev);
 
-static void export_rdev(struct md_rdev *rdev)
+/* just for claiming the bdev */
+static struct md_rdev claim_rdev;
+
+static void export_rdev(struct md_rdev *rdev, struct mddev *mddev)
 {
 	pr_debug("md: export_rdev(%pg)\n", rdev->bdev);
 	md_rdev_clear(rdev);
@@ -2457,7 +2460,7 @@ static void export_rdev(struct md_rdev *rdev)
 	if (test_bit(AutoDetected, &rdev->flags))
 		md_autodetect_dev(rdev->bdev->bd_dev);
 #endif
-	blkdev_put(rdev->bdev, FMODE_READ | FMODE_WRITE | FMODE_EXCL);
+	blkdev_put(rdev->bdev, mddev->major_version == -2 ? &claim_rdev : rdev);
 	rdev->bdev = NULL;
 	kobject_put(&rdev->kobj);
 }
@@ -2485,7 +2488,7 @@ static void md_kick_rdev_from_array(struct md_rdev *rdev)
 	INIT_WORK(&rdev->del_work, rdev_delayed_delete);
 	kobject_get(&rdev->kobj);
 	queue_work(md_rdev_misc_wq, &rdev->del_work);
-	export_rdev(rdev);
+	export_rdev(rdev, rdev->mddev);
 }
 
 static void export_array(struct mddev *mddev)
@@ -3612,6 +3615,7 @@ int md_rdev_init(struct md_rdev *rdev)
 	return badblocks_init(&rdev->badblocks, 0);
 }
 EXPORT_SYMBOL_GPL(md_rdev_init);
+
 /*
  * Import a device. If 'super_format' >= 0, then sanity check the superblock
  *
@@ -3624,7 +3628,6 @@ EXPORT_SYMBOL_GPL(md_rdev_init);
  */
 static struct md_rdev *md_import_device(dev_t newdev, int super_format, int super_minor)
 {
-	static struct md_rdev claim_rdev; /* just for claiming the bdev */
 	struct md_rdev *rdev;
 	sector_t size;
 	int err;
@@ -3640,8 +3643,7 @@ static struct md_rdev *md_import_device(dev_t newdev, int super_format, int supe
 	if (err)
 		goto out_clear_rdev;
 
-	rdev->bdev = blkdev_get_by_dev(newdev,
-			FMODE_READ | FMODE_WRITE | FMODE_EXCL,
+	rdev->bdev = blkdev_get_by_dev(newdev, FMODE_READ | FMODE_WRITE,
 			super_format == -2 ? &claim_rdev : rdev, NULL);
 	if (IS_ERR(rdev->bdev)) {
 		pr_warn("md: could not open device unknown-block(%u,%u).\n",
@@ -3679,7 +3681,7 @@ static struct md_rdev *md_import_device(dev_t newdev, int super_format, int supe
 	return rdev;
 
 out_blkdev_put:
-	blkdev_put(rdev->bdev, FMODE_READ | FMODE_WRITE | FMODE_EXCL);
+	blkdev_put(rdev->bdev, super_format == -2 ? &claim_rdev : rdev);
 out_clear_rdev:
 	md_rdev_clear(rdev);
 out_free_rdev:
@@ -4560,7 +4562,7 @@ new_dev_store(struct mddev *mddev, const char *buf, size_t len)
 	err = bind_rdev_to_array(rdev, mddev);
  out:
 	if (err)
-		export_rdev(rdev);
+		export_rdev(rdev, mddev);
 	mddev_unlock(mddev);
 	if (!err)
 		md_new_event();
@@ -6498,7 +6500,7 @@ static void autorun_devices(int part)
 			rdev_for_each_list(rdev, tmp, &candidates) {
 				list_del_init(&rdev->same_set);
 				if (bind_rdev_to_array(rdev, mddev))
-					export_rdev(rdev);
+					export_rdev(rdev, mddev);
 			}
 			autorun_array(mddev);
 			mddev_unlock(mddev);
@@ -6508,7 +6510,7 @@ static void autorun_devices(int part)
 		 */
 		rdev_for_each_list(rdev, tmp, &candidates) {
 			list_del_init(&rdev->same_set);
-			export_rdev(rdev);
+			export_rdev(rdev, mddev);
 		}
 		mddev_put(mddev);
 	}
@@ -6696,13 +6698,13 @@ int md_add_new_disk(struct mddev *mddev, struct mdu_disk_info_s *info)
 				pr_warn("md: %pg has different UUID to %pg\n",
 					rdev->bdev,
 					rdev0->bdev);
-				export_rdev(rdev);
+				export_rdev(rdev, mddev);
 				return -EINVAL;
 			}
 		}
 		err = bind_rdev_to_array(rdev, mddev);
 		if (err)
-			export_rdev(rdev);
+			export_rdev(rdev, mddev);
 		return err;
 	}
 
@@ -6746,7 +6748,7 @@ int md_add_new_disk(struct mddev *mddev, struct mdu_disk_info_s *info)
 			/* This was a hot-add request, but events doesn't
 			 * match, so reject it.
 			 */
-			export_rdev(rdev);
+			export_rdev(rdev, mddev);
 			return -EINVAL;
 		}
 
@@ -6772,7 +6774,7 @@ int md_add_new_disk(struct mddev *mddev, struct mdu_disk_info_s *info)
 				}
 			}
 			if (has_journal || mddev->bitmap) {
-				export_rdev(rdev);
+				export_rdev(rdev, mddev);
 				return -EBUSY;
 			}
 			set_bit(Journal, &rdev->flags);
@@ -6787,7 +6789,7 @@ int md_add_new_disk(struct mddev *mddev, struct mdu_disk_info_s *info)
 				/* --add initiated by this node */
 				err = md_cluster_ops->add_new_disk(mddev, rdev);
 				if (err) {
-					export_rdev(rdev);
+					export_rdev(rdev, mddev);
 					return err;
 				}
 			}
@@ -6797,7 +6799,7 @@ int md_add_new_disk(struct mddev *mddev, struct mdu_disk_info_s *info)
 		err = bind_rdev_to_array(rdev, mddev);
 
 		if (err)
-			export_rdev(rdev);
+			export_rdev(rdev, mddev);
 
 		if (mddev_is_clustered(mddev)) {
 			if (info->state & (1 << MD_DISK_CANDIDATE)) {
@@ -6860,7 +6862,7 @@ int md_add_new_disk(struct mddev *mddev, struct mdu_disk_info_s *info)
 
 		err = bind_rdev_to_array(rdev, mddev);
 		if (err) {
-			export_rdev(rdev);
+			export_rdev(rdev, mddev);
 			return err;
 		}
 	}
@@ -6985,7 +6987,7 @@ static int hot_add_disk(struct mddev *mddev, dev_t dev)
 	return 0;
 
 abort_export:
-	export_rdev(rdev);
+	export_rdev(rdev, mddev);
 	return err;
 }