block-5.15-2021-09-05

-----BEGIN PGP SIGNATURE-----
 
 iQJEBAABCAAuFiEEwPw5LcreJtl1+l5K99NY+ylx4KYFAmE1hQcQHGF4Ym9lQGtl
 cm5lbC5kawAKCRD301j7KXHgprjfEACdG+medwQOPpKNSoAvQmYyQnZRMbPjiruv
 A4nW2L6MKaExO59qQLVbBYHaH2+ng2UR/p5jNi2AKm+hrQEYllxlNvuCkRBIn97J
 r45R48mzBbHjR4kE3Fdu1mOFpBWOuU9JrtzHI+JF/Sl/qPIxKYNHf5E66T6l90Fz
 0hJkorAoVB7+hQYixdmkM9quZy11D5SY3aM+bG8r2uNjZTBEHMfmOen8o1giR0vC
 EOHzObuC6WLjLGQInNW+Cq2//vVVybQa79mhOUMp93z5nhDMtwUu7MH4B4kmGpix
 GLjDa1DukUZe7nGcnsRKmjjXQ+BpG6YF52Z2RfVZpWZn83t5c4YQsq++TPZ8KfpK
 4NAFFuSbGM/+QWwEiiyWu00syvpzrEJ4ZIJyZX3FYEeKyKWVRGHqlMDcS9LstYOk
 4OfgQUcJ7f/fXeedwi0OGJS1BLr6fi8RnazIafCNIIJLe1XIwTsNufPCNxWYqDAi
 0XhH+uYGD38VoUiR5JymZku6frwY4kxssA1khPPE5jWbzCZXiHprwwzaP4hBNNeZ
 c5cn9/1ZQSoTE3ebrX9pzTn5wRZwAL+iDhZ2SpLlN2Ji1BJ4EM9H8qFGj3U/CSM4
 OWKY2c0VwJYQUhjO4QDBx0MblJgNy8HsvmqGETuxUlk56j3Q1Mx3ViPV43amP9eM
 OM4mGige3Q==
 =4SCA
 -----END PGP SIGNATURE-----

Merge tag 'block-5.15-2021-09-05' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:
 "Was going to send this one in later this week, but given that -Werror
  is now enabled (or at least available), the mq-deadline fix really
  should go in for the folks hitting that.

   - Ensure dd_queued() is only there if needed (Geert)

   - Fix a kerneldoc warning for bio_alloc_kiocb()

   - BFQ fix for queue merging

   - loop locking fix (Tetsuo)"

* tag 'block-5.15-2021-09-05' of git://git.kernel.dk/linux-block:
  loop: reduce the loop_ctl_mutex scope
  bio: fix kerneldoc documentation for bio_alloc_kiocb()
  block, bfq: honor already-setup queue merges
  block/mq-deadline: Move dd_queued() to fix defined but not used warning

block/bfq-iosched.c

@@ -2662,6 +2662,15 @@ bfq_setup_merge(struct bfq_queue *bfqq, struct bfq_queue *new_bfqq)
 	 * are likely to increase the throughput.
 	 */
 	bfqq->new_bfqq = new_bfqq;
+	/*
+	 * The above assignment schedules the following redirections:
+	 * each time some I/O for bfqq arrives, the process that
+	 * generated that I/O is disassociated from bfqq and
+	 * associated with new_bfqq. Here we increases new_bfqq->ref
+	 * in advance, adding the number of processes that are
+	 * expected to be associated with new_bfqq as they happen to
+	 * issue I/O.
+	 */
 	new_bfqq->ref += process_refs;
 	return new_bfqq;
 }
@@ -2724,6 +2733,10 @@ bfq_setup_cooperator(struct bfq_data *bfqd, struct bfq_queue *bfqq,
 {
 	struct bfq_queue *in_service_bfqq, *new_bfqq;
 
+	/* if a merge has already been setup, then proceed with that first */
+	if (bfqq->new_bfqq)
+		return bfqq->new_bfqq;
+
 	/*
 	 * Check delayed stable merge for rotational or non-queueing
 	 * devs. For this branch to be executed, bfqq must not be
@@ -2825,9 +2838,6 @@ bfq_setup_cooperator(struct bfq_data *bfqd, struct bfq_queue *bfqq,
 	if (bfq_too_late_for_merging(bfqq))
 		return NULL;
 
-	if (bfqq->new_bfqq)
-		return bfqq->new_bfqq;
-
 	if (!io_struct || unlikely(bfqq == &bfqd->oom_bfqq))
 		return NULL;
 

block/bio.c

@@ -1688,7 +1688,7 @@ EXPORT_SYMBOL(bioset_init_from_src);
 /**
  * bio_alloc_kiocb - Allocate a bio from bio_set based on kiocb
  * @kiocb:	kiocb describing the IO
- * @nr_iovecs:	number of iovecs to pre-allocate
+ * @nr_vecs:	number of iovecs to pre-allocate
  * @bs:		bio_set to allocate from
  *
  * Description:

block/mq-deadline.c

@@ -270,12 +270,6 @@ deadline_move_request(struct deadline_data *dd, struct dd_per_prio *per_prio,
 	deadline_remove_request(rq->q, per_prio, rq);
 }
 
-/* Number of requests queued for a given priority level. */
-static u32 dd_queued(struct deadline_data *dd, enum dd_prio prio)
-{
-	return dd_sum(dd, inserted, prio) - dd_sum(dd, completed, prio);
-}
-
 /*
  * deadline_check_fifo returns 0 if there are no expired requests on the fifo,
  * 1 otherwise. Requires !list_empty(&dd->fifo_list[data_dir])
@@ -953,6 +947,12 @@ static int dd_async_depth_show(void *data, struct seq_file *m)
 	return 0;
 }
 
+/* Number of requests queued for a given priority level. */
+static u32 dd_queued(struct deadline_data *dd, enum dd_prio prio)
+{
+	return dd_sum(dd, inserted, prio) - dd_sum(dd, completed, prio);
+}
+
 static int dd_queued_show(void *data, struct seq_file *m)
 {
 	struct request_queue *q = data;

drivers/block/loop.c

@@ -2111,18 +2111,6 @@ int loop_register_transfer(struct loop_func_table *funcs)
 	return 0;
 }
 
-static int unregister_transfer_cb(int id, void *ptr, void *data)
-{
-	struct loop_device *lo = ptr;
-	struct loop_func_table *xfer = data;
-
-	mutex_lock(&lo->lo_mutex);
-	if (lo->lo_encryption == xfer)
-		loop_release_xfer(lo);
-	mutex_unlock(&lo->lo_mutex);
-	return 0;
-}
-
 int loop_unregister_transfer(int number)
 {
 	unsigned int n = number;
@@ -2130,9 +2118,20 @@ int loop_unregister_transfer(int number)
 
 	if (n == 0 || n >= MAX_LO_CRYPT || (xfer = xfer_funcs[n]) == NULL)
 		return -EINVAL;
+	/*
+	 * This function is called from only cleanup_cryptoloop().
+	 * Given that each loop device that has a transfer enabled holds a
+	 * reference to the module implementing it we should never get here
+	 * with a transfer that is set (unless forced module unloading is
+	 * requested). Thus, check module's refcount and warn if this is
+	 * not a clean unloading.
+	 */
+#ifdef CONFIG_MODULE_UNLOAD
+	if (xfer->owner && module_refcount(xfer->owner) != -1)
+		pr_err("Danger! Unregistering an in use transfer function.\n");
+#endif
 
 	xfer_funcs[n] = NULL;
-	idr_for_each(&loop_index_idr, &unregister_transfer_cb, xfer);
 	return 0;
 }
 
@@ -2323,8 +2322,9 @@ static int loop_add(int i)
 	} else {
 		err = idr_alloc(&loop_index_idr, lo, 0, 0, GFP_KERNEL);
 	}
+	mutex_unlock(&loop_ctl_mutex);
 	if (err < 0)
-		goto out_unlock;
+		goto out_free_dev;
 	i = err;
 
 	err = -ENOMEM;
@@ -2393,15 +2393,19 @@ static int loop_add(int i)
 	disk->events		= DISK_EVENT_MEDIA_CHANGE;
 	disk->event_flags	= DISK_EVENT_FLAG_UEVENT;
 	sprintf(disk->disk_name, "loop%d", i);
+	/* Make this loop device reachable from pathname. */
 	add_disk(disk);
+	/* Show this loop device. */
+	mutex_lock(&loop_ctl_mutex);
+	lo->idr_visible = true;
 	mutex_unlock(&loop_ctl_mutex);
 	return i;
 
 out_cleanup_tags:
 	blk_mq_free_tag_set(&lo->tag_set);
 out_free_idr:
+	mutex_lock(&loop_ctl_mutex);
 	idr_remove(&loop_index_idr, i);
-out_unlock:
 	mutex_unlock(&loop_ctl_mutex);
 out_free_dev:
 	kfree(lo);
@@ -2411,9 +2415,14 @@ out:
 
 static void loop_remove(struct loop_device *lo)
 {
+	/* Make this loop device unreachable from pathname. */
 	del_gendisk(lo->lo_disk);
 	blk_cleanup_disk(lo->lo_disk);
 	blk_mq_free_tag_set(&lo->tag_set);
+	mutex_lock(&loop_ctl_mutex);
+	idr_remove(&loop_index_idr, lo->lo_number);
+	mutex_unlock(&loop_ctl_mutex);
+	/* There is no route which can find this loop device. */
 	mutex_destroy(&lo->lo_mutex);
 	kfree(lo);
 }
@@ -2437,31 +2446,40 @@ static int loop_control_remove(int idx)
 		return -EINVAL;
 	}
 		
+	/* Hide this loop device for serialization. */
 	ret = mutex_lock_killable(&loop_ctl_mutex);
 	if (ret)
 		return ret;
-
 	lo = idr_find(&loop_index_idr, idx);
-	if (!lo) {
+	if (!lo || !lo->idr_visible)
 		ret = -ENODEV;
-		goto out_unlock_ctrl;
+	else
-	}
+		lo->idr_visible = false;
+	mutex_unlock(&loop_ctl_mutex);
+	if (ret)
+		return ret;
 
+	/* Check whether this loop device can be removed. */
 	ret = mutex_lock_killable(&lo->lo_mutex);
 	if (ret)
-		goto out_unlock_ctrl;
+		goto mark_visible;
 	if (lo->lo_state != Lo_unbound ||
 	    atomic_read(&lo->lo_refcnt) > 0) {
 		mutex_unlock(&lo->lo_mutex);
 		ret = -EBUSY;
-		goto out_unlock_ctrl;
+		goto mark_visible;
 	}
+	/* Mark this loop device no longer open()-able. */
 	lo->lo_state = Lo_deleting;
 	mutex_unlock(&lo->lo_mutex);
 
-	idr_remove(&loop_index_idr, lo->lo_number);
 	loop_remove(lo);
-out_unlock_ctrl:
+	return 0;
+
+mark_visible:
+	/* Show this loop device again. */
+	mutex_lock(&loop_ctl_mutex);
+	lo->idr_visible = true;
 	mutex_unlock(&loop_ctl_mutex);
 	return ret;
 }
@@ -2475,7 +2493,8 @@ static int loop_control_get_free(int idx)
 	if (ret)
 		return ret;
 	idr_for_each_entry(&loop_index_idr, lo, id) {
-		if (lo->lo_state == Lo_unbound)
+		/* Hitting a race results in creating a new loop device which is harmless. */
+		if (lo->idr_visible && data_race(lo->lo_state) == Lo_unbound)
 			goto found;
 	}
 	mutex_unlock(&loop_ctl_mutex);
@@ -2591,10 +2610,14 @@ static void __exit loop_exit(void)
 	unregister_blkdev(LOOP_MAJOR, "loop");
 	misc_deregister(&loop_misc);
 
-	mutex_lock(&loop_ctl_mutex);
+	/*
+	 * There is no need to use loop_ctl_mutex here, for nobody else can
+	 * access loop_index_idr when this module is unloading (unless forced
+	 * module unloading is requested). If this is not a clean unloading,
+	 * we have no means to avoid kernel crash.
+	 */
 	idr_for_each_entry(&loop_index_idr, lo, id)
 		loop_remove(lo);
-	mutex_unlock(&loop_ctl_mutex);
 
 	idr_destroy(&loop_index_idr);
 }

drivers/block/loop.h

@@ -68,6 +68,7 @@ struct loop_device {
 	struct blk_mq_tag_set	tag_set;
 	struct gendisk		*lo_disk;
 	struct mutex		lo_mutex;
+	bool			idr_visible;
 };
 
 struct loop_cmd {