Merge tag 'for-5.8/dm-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm

Pull device mapper fixes from Mike Snitzer:

 - A request-based DM fix to not use a waitqueue to wait for blk-mq IO
   completion because doing so is racey.

 - A couple more DM zoned target fixes to address issues introduced
   during the 5.8 cycle.

 - A DM core fix to use proper interface to cleanup DM's static flush
   bio.

 - A DM core fix to prevent mm recursion during memory allocation needed
   by dm_kobject_uevent.

* tag 'for-5.8/dm-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
  dm: use noio when sending kobject event
  dm zoned: Fix zone reclaim trigger
  dm zoned: fix unused but set variable warnings
  dm writecache: reject asynchronous pmem devices
  dm: use bio_uninit instead of bio_disassociate_blkg
  dm: do not use waitqueue for request-based DM

Author: torvalds

drivers/md/dm-rq.c

@@ -146,10 +146,6 @@ static void rq_end_stats(struct mapped_device *md, struct request *orig)
  */
 static void rq_completed(struct mapped_device *md)
 {
-	/* nudge anyone waiting on suspend queue */
-	if (unlikely(wq_has_sleeper(&md->wait)))
-		wake_up(&md->wait);
-
 	/*
 	 * dm_put() must be at the end of this function. See the comment above
 	 */

drivers/md/dm-writecache.c

@@ -2266,6 +2266,12 @@ static int writecache_ctr(struct dm_target *ti, unsigned argc, char **argv)
 	}
 
 	if (WC_MODE_PMEM(wc)) {
+		if (!dax_synchronous(wc->ssd_dev->dax_dev)) {
+			r = -EOPNOTSUPP;
+			ti->error = "Asynchronous persistent memory not supported as pmem cache";
+			goto bad;
+		}
+
 		r = persistent_memory_claim(wc);
 		if (r) {
 			ti->error = "Unable to map persistent memory for cache";

drivers/md/dm-zoned-metadata.c

@@ -2217,8 +2217,15 @@ struct dm_zone *dmz_alloc_zone(struct dmz_metadata *zmd, unsigned int dev_idx,
 {
 	struct list_head *list;
 	struct dm_zone *zone;
-	int i = 0;
+	int i;
+
+	/* Schedule reclaim to ensure free zones are available */
+	if (!(flags & DMZ_ALLOC_RECLAIM)) {
+		for (i = 0; i < zmd->nr_devs; i++)
+			dmz_schedule_reclaim(zmd->dev[i].reclaim);
+	}
 
+	i = 0;
 again:
 	if (flags & DMZ_ALLOC_CACHE)
 		list = &zmd->unmap_cache_list;

drivers/md/dm-zoned-reclaim.c

@@ -456,6 +456,8 @@ static unsigned int dmz_reclaim_percentage(struct dmz_reclaim *zrc)
 		nr_zones = dmz_nr_rnd_zones(zmd, zrc->dev_idx);
 		nr_unmap = dmz_nr_unmap_rnd_zones(zmd, zrc->dev_idx);
 	}
+	if (nr_unmap <= 1)
+		return 0;
 	return nr_unmap * 100 / nr_zones;
 }
 
@@ -501,7 +503,7 @@ static void dmz_reclaim_work(struct work_struct *work)
 {
 	struct dmz_reclaim *zrc = container_of(work, struct dmz_reclaim, work.work);
 	struct dmz_metadata *zmd = zrc->metadata;
-	unsigned int p_unmap, nr_unmap_rnd = 0, nr_rnd = 0;
+	unsigned int p_unmap;
 	int ret;
 
 	if (dmz_dev_is_dying(zmd))
@@ -527,9 +529,6 @@ static void dmz_reclaim_work(struct work_struct *work)
 		zrc->kc_throttle.throttle = min(75U, 100U - p_unmap / 2);
 	}
 
-	nr_unmap_rnd = dmz_nr_unmap_rnd_zones(zmd, zrc->dev_idx);
-	nr_rnd = dmz_nr_rnd_zones(zmd, zrc->dev_idx);
-
 	DMDEBUG("(%s/%u): Reclaim (%u): %s, %u%% free zones (%u/%u cache %u/%u random)",
 		dmz_metadata_label(zmd), zrc->dev_idx,
 		zrc->kc_throttle.throttle,

drivers/md/dm-zoned-target.c

@@ -400,15 +400,7 @@ static void dmz_handle_bio(struct dmz_target *dmz, struct dm_chunk_work *cw,
 		dm_per_bio_data(bio, sizeof(struct dmz_bioctx));
 	struct dmz_metadata *zmd = dmz->metadata;
 	struct dm_zone *zone;
-	int i, ret;
-
-	/*
-	 * Write may trigger a zone allocation. So make sure the
-	 * allocation can succeed.
-	 */
-	if (bio_op(bio) == REQ_OP_WRITE)
-		for (i = 0; i < dmz->nr_ddevs; i++)
-			dmz_schedule_reclaim(dmz->dev[i].reclaim);
+	int ret;
 
 	dmz_lock_metadata(zmd);
 

drivers/md/dm.c

@@ -12,6 +12,7 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/mutex.h>
+#include <linux/sched/mm.h>
 #include <linux/sched/signal.h>
 #include <linux/blkpg.h>
 #include <linux/bio.h>
@@ -654,28 +655,6 @@ static void free_tio(struct dm_target_io *tio)
 	bio_put(&tio->clone);
 }
 
-static bool md_in_flight_bios(struct mapped_device *md)
-{
-	int cpu;
-	struct hd_struct *part = &dm_disk(md)->part0;
-	long sum = 0;
-
-	for_each_possible_cpu(cpu) {
-		sum += part_stat_local_read_cpu(part, in_flight[0], cpu);
-		sum += part_stat_local_read_cpu(part, in_flight[1], cpu);
-	}
-
-	return sum != 0;
-}
-
-static bool md_in_flight(struct mapped_device *md)
-{
-	if (queue_is_mq(md->queue))
-		return blk_mq_queue_inflight(md->queue);
-	else
-		return md_in_flight_bios(md);
-}
-
 u64 dm_start_time_ns_from_clone(struct bio *bio)
 {
 	struct dm_target_io *tio = container_of(bio, struct dm_target_io, clone);
@@ -1465,9 +1444,6 @@ static int __send_empty_flush(struct clone_info *ci)
 	BUG_ON(bio_has_data(ci->bio));
 	while ((ti = dm_table_get_target(ci->map, target_nr++)))
 		__send_duplicate_bios(ci, ti, ti->num_flush_bios, NULL);
-
-	bio_disassociate_blkg(ci->bio);
-
 	return 0;
 }
 
@@ -1655,6 +1631,7 @@ static blk_qc_t __split_and_process_bio(struct mapped_device *md,
 		ci.bio = &flush_bio;
 		ci.sector_count = 0;
 		error = __send_empty_flush(&ci);
+		bio_uninit(ci.bio);
 		/* dec_pending submits any data associated with flush */
 	} else if (op_is_zone_mgmt(bio_op(bio))) {
 		ci.bio = bio;
@@ -1729,6 +1706,7 @@ static blk_qc_t __process_bio(struct mapped_device *md, struct dm_table *map,
 		ci.bio = &flush_bio;
 		ci.sector_count = 0;
 		error = __send_empty_flush(&ci);
+		bio_uninit(ci.bio);
 		/* dec_pending submits any data associated with flush */
 	} else {
 		struct dm_target_io *tio;
@@ -2470,15 +2448,29 @@ void dm_put(struct mapped_device *md)
 }
 EXPORT_SYMBOL_GPL(dm_put);
 
-static int dm_wait_for_completion(struct mapped_device *md, long task_state)
+static bool md_in_flight_bios(struct mapped_device *md)
+{
+	int cpu;
+	struct hd_struct *part = &dm_disk(md)->part0;
+	long sum = 0;
+
+	for_each_possible_cpu(cpu) {
+		sum += part_stat_local_read_cpu(part, in_flight[0], cpu);
+		sum += part_stat_local_read_cpu(part, in_flight[1], cpu);
+	}
+
+	return sum != 0;
+}
+
+static int dm_wait_for_bios_completion(struct mapped_device *md, long task_state)
 {
 	int r = 0;
 	DEFINE_WAIT(wait);
 
-	while (1) {
+	while (true) {
 		prepare_to_wait(&md->wait, &wait, task_state);
 
-		if (!md_in_flight(md))
+		if (!md_in_flight_bios(md))
 			break;
 
 		if (signal_pending_state(task_state, current)) {
@@ -2493,6 +2485,28 @@ static int dm_wait_for_completion(struct mapped_device *md, long task_state)
 	return r;
 }
 
+static int dm_wait_for_completion(struct mapped_device *md, long task_state)
+{
+	int r = 0;
+
+	if (!queue_is_mq(md->queue))
+		return dm_wait_for_bios_completion(md, task_state);
+
+	while (true) {
+		if (!blk_mq_queue_inflight(md->queue))
+			break;
+
+		if (signal_pending_state(task_state, current)) {
+			r = -EINTR;
+			break;
+		}
+
+		msleep(5);
+	}
+
+	return r;
+}
+
 /*
  * Process the deferred bios
  */
@@ -2926,17 +2940,25 @@ EXPORT_SYMBOL_GPL(dm_internal_resume_fast);
 int dm_kobject_uevent(struct mapped_device *md, enum kobject_action action,
 		       unsigned cookie)
 {
+	int r;
+	unsigned noio_flag;
 	char udev_cookie[DM_COOKIE_LENGTH];
 	char *envp[] = { udev_cookie, NULL };
 
+	noio_flag = memalloc_noio_save();
+
 	if (!cookie)
-		return kobject_uevent(&disk_to_dev(md->disk)->kobj, action);
+		r = kobject_uevent(&disk_to_dev(md->disk)->kobj, action);
 	else {
 		snprintf(udev_cookie, DM_COOKIE_LENGTH, "%s=%u",
 			 DM_COOKIE_ENV_VAR_NAME, cookie);
-		return kobject_uevent_env(&disk_to_dev(md->disk)->kobj,
-					  action, envp);
+		r = kobject_uevent_env(&disk_to_dev(md->disk)->kobj,
+				       action, envp);
 	}
+
+	memalloc_noio_restore(noio_flag);
+
+	return r;
 }
 
 uint32_t dm_next_uevent_seq(struct mapped_device *md)