- Fix zone state management race in DM zoned target by eliminating

the unnecessary DMZ_ACTIVE state.
 
 - A couple fixes for issues the DM snapshot target's optional discard
   support added during first week of the 5.3 merge.
 
 - Increase default size of outstanding IO that is allowed for a each
   dm-kcopyd client and introduce tunable to allow user adjust.
 
 - Update DM core to use printk ratelimiting functions rather than
   duplicate them and in doing so fix an issue where DMDEBUG_LIMIT() rate
   limited KERN_DEBUG messages had excessive "callbacks suppressed"
   messages.
 -----BEGIN PGP SIGNATURE-----
 
 iQFHBAABCAAxFiEEJfWUX4UqZ4x1O2wixSPxCi2dA1oFAl0w2IsTHHNuaXR6ZXJA
 cmVkaGF0LmNvbQAKCRDFI/EKLZ0DWgPeCACMtDQrHqMTOT7OPDRxSJjZixefzL32
 lFi31mjjEb7GoxiS3dBepdJmQiUwROINdGLIGTfBAlH05b/8fgFgE6iCGZ9uzad4
 0PNe9q7pbtfQDLXx+mVMjEdK6P/ilmVFXCW8VQpAAeUFL+dwXYHHIbmQZ/rahOz5
 8nn6wGBQ/LRRcbV0hBHNXQymIXPxMweMxO3usSuKbfhe7JjRwslThGbZ4KVwjCwl
 sLl5mEWXwTKUemGXsXFbCbtH/rnZpbaiAkBedT0oV8g8atRBeQyj0vk48htincj7
 Uv6xGjJGXuqUkcvQnTx3C1fk3lH5xJb5MTL3WEN0g6fOmyJd1sMd0gd/
 =4Rpg
 -----END PGP SIGNATURE-----

Merge tag 'for-5.3/dm-changes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm

Pull more device mapper updates from Mike Snitzer:

 - Fix zone state management race in DM zoned target by eliminating the
   unnecessary DMZ_ACTIVE state.

 - A couple fixes for issues the DM snapshot target's optional discard
   support added during first week of the 5.3 merge.

 - Increase default size of outstanding IO that is allowed for a each
   dm-kcopyd client and introduce tunable to allow user adjust.

 - Update DM core to use printk ratelimiting functions rather than
   duplicate them and in doing so fix an issue where DMDEBUG_LIMIT()
   rate limited KERN_DEBUG messages had excessive "callbacks suppressed"
   messages.

* tag 'for-5.3/dm-changes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
  dm: use printk ratelimiting functions
  dm kcopyd: Increase default sub-job size to 512KB
  dm snapshot: fix oversights in optional discard support
  dm zoned: fix zone state management race

drivers/md/dm-kcopyd.c

@@ -28,10 +28,27 @@
 
 #include "dm-core.h"
 
-#define SUB_JOB_SIZE	128
 #define SPLIT_COUNT	8
 #define MIN_JOBS	8
-#define RESERVE_PAGES	(DIV_ROUND_UP(SUB_JOB_SIZE << SECTOR_SHIFT, PAGE_SIZE))
+
+#define DEFAULT_SUB_JOB_SIZE_KB 512
+#define MAX_SUB_JOB_SIZE_KB     1024
+
+static unsigned kcopyd_subjob_size_kb = DEFAULT_SUB_JOB_SIZE_KB;
+
+module_param(kcopyd_subjob_size_kb, uint, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(kcopyd_subjob_size_kb, "Sub-job size for dm-kcopyd clients");
+
+static unsigned dm_get_kcopyd_subjob_size(void)
+{
+	unsigned sub_job_size_kb;
+
+	sub_job_size_kb = __dm_get_module_param(&kcopyd_subjob_size_kb,
+						DEFAULT_SUB_JOB_SIZE_KB,
+						MAX_SUB_JOB_SIZE_KB);
+
+	return sub_job_size_kb << 1;
+}
 
 /*-----------------------------------------------------------------
  * Each kcopyd client has its own little pool of preallocated
@@ -41,6 +58,7 @@ struct dm_kcopyd_client {
 	struct page_list *pages;
 	unsigned nr_reserved_pages;
 	unsigned nr_free_pages;
+	unsigned sub_job_size;
 
 	struct dm_io_client *io_client;
 
@@ -693,8 +711,8 @@ static void segment_complete(int read_err, unsigned long write_err,
 		progress = job->progress;
 		count = job->source.count - progress;
 		if (count) {
-			if (count > SUB_JOB_SIZE)
-				count = SUB_JOB_SIZE;
+			if (count > kc->sub_job_size)
+				count = kc->sub_job_size;
 
 			job->progress += count;
 		}
@@ -821,7 +839,7 @@ void dm_kcopyd_copy(struct dm_kcopyd_client *kc, struct dm_io_region *from,
 	job->master_job = job;
 	job->write_offset = 0;
 
-	if (job->source.count <= SUB_JOB_SIZE)
+	if (job->source.count <= kc->sub_job_size)
 		dispatch_job(job);
 	else {
 		job->progress = 0;
@@ -888,6 +906,7 @@ int kcopyd_cancel(struct kcopyd_job *job, int block)
 struct dm_kcopyd_client *dm_kcopyd_client_create(struct dm_kcopyd_throttle *throttle)
 {
 	int r;
+	unsigned reserve_pages;
 	struct dm_kcopyd_client *kc;
 
 	kc = kzalloc(sizeof(*kc), GFP_KERNEL);
@@ -912,9 +931,12 @@ struct dm_kcopyd_client *dm_kcopyd_client_create(struct dm_kcopyd_throttle *thro
 		goto bad_workqueue;
 	}
 
+	kc->sub_job_size = dm_get_kcopyd_subjob_size();
+	reserve_pages = DIV_ROUND_UP(kc->sub_job_size << SECTOR_SHIFT, PAGE_SIZE);
+
 	kc->pages = NULL;
 	kc->nr_reserved_pages = kc->nr_free_pages = 0;
-	r = client_reserve_pages(kc, RESERVE_PAGES);
+	r = client_reserve_pages(kc, reserve_pages);
 	if (r)
 		goto bad_client_pages;
 

drivers/md/dm-snap.c

@@ -2072,6 +2072,12 @@ static int snapshot_merge_map(struct dm_target *ti, struct bio *bio)
 		return DM_MAPIO_REMAPPED;
 	}
 
+	if (unlikely(bio_op(bio) == REQ_OP_DISCARD)) {
+		/* Once merging, discards no longer effect change */
+		bio_endio(bio);
+		return DM_MAPIO_SUBMITTED;
+	}
+
 	chunk = sector_to_chunk(s->store, bio->bi_iter.bi_sector);
 
 	down_write(&s->lock);
@@ -2331,6 +2337,8 @@ static void snapshot_io_hints(struct dm_target *ti, struct queue_limits *limits)
 	if (snap->discard_zeroes_cow) {
 		struct dm_snapshot *snap_src = NULL, *snap_dest = NULL;
 
+		down_read(&_origins_lock);
+
 		(void) __find_snapshots_sharing_cow(snap, &snap_src, &snap_dest, NULL);
 		if (snap_src && snap_dest)
 			snap = snap_src;
@@ -2338,6 +2346,8 @@ static void snapshot_io_hints(struct dm_target *ti, struct queue_limits *limits)
 		/* All discards are split on chunk_size boundary */
 		limits->discard_granularity = snap->store->chunk_size;
 		limits->max_discard_sectors = snap->store->chunk_size;
+
+		up_read(&_origins_lock);
 	}
 }
 

drivers/md/dm-zoned-metadata.c

@@ -1601,30 +1601,6 @@ struct dm_zone *dmz_get_zone_for_reclaim(struct dmz_metadata *zmd)
 	return zone;
 }
 
-/*
- * Activate a zone (increment its reference count).
- */
-void dmz_activate_zone(struct dm_zone *zone)
-{
-	set_bit(DMZ_ACTIVE, &zone->flags);
-	atomic_inc(&zone->refcount);
-}
-
-/*
- * Deactivate a zone. This decrement the zone reference counter
- * and clears the active state of the zone once the count reaches 0,
- * indicating that all BIOs to the zone have completed. Returns
- * true if the zone was deactivated.
- */
-void dmz_deactivate_zone(struct dm_zone *zone)
-{
-	if (atomic_dec_and_test(&zone->refcount)) {
-		WARN_ON(!test_bit(DMZ_ACTIVE, &zone->flags));
-		clear_bit_unlock(DMZ_ACTIVE, &zone->flags);
-		smp_mb__after_atomic();
-	}
-}
-
 /*
  * Get the zone mapping a chunk, if the chunk is mapped already.
  * If no mapping exist and the operation is WRITE, a zone is

drivers/md/dm-zoned.h

@@ -115,7 +115,6 @@ enum {
 	DMZ_BUF,
 
 	/* Zone internal state */
-	DMZ_ACTIVE,
 	DMZ_RECLAIM,
 	DMZ_SEQ_WRITE_ERR,
 };
@@ -128,7 +127,6 @@ enum {
 #define dmz_is_empty(z)		((z)->wp_block == 0)
 #define dmz_is_offline(z)	test_bit(DMZ_OFFLINE, &(z)->flags)
 #define dmz_is_readonly(z)	test_bit(DMZ_READ_ONLY, &(z)->flags)
-#define dmz_is_active(z)	test_bit(DMZ_ACTIVE, &(z)->flags)
 #define dmz_in_reclaim(z)	test_bit(DMZ_RECLAIM, &(z)->flags)
 #define dmz_seq_write_err(z)	test_bit(DMZ_SEQ_WRITE_ERR, &(z)->flags)
 
@@ -188,8 +186,30 @@ void dmz_unmap_zone(struct dmz_metadata *zmd, struct dm_zone *zone);
 unsigned int dmz_nr_rnd_zones(struct dmz_metadata *zmd);
 unsigned int dmz_nr_unmap_rnd_zones(struct dmz_metadata *zmd);
 
-void dmz_activate_zone(struct dm_zone *zone);
-void dmz_deactivate_zone(struct dm_zone *zone);
+/*
+ * Activate a zone (increment its reference count).
+ */
+static inline void dmz_activate_zone(struct dm_zone *zone)
+{
+	atomic_inc(&zone->refcount);
+}
+
+/*
+ * Deactivate a zone. This decrement the zone reference counter
+ * indicating that all BIOs to the zone have completed when the count is 0.
+ */
+static inline void dmz_deactivate_zone(struct dm_zone *zone)
+{
+	atomic_dec(&zone->refcount);
+}
+
+/*
+ * Test if a zone is active, that is, has a refcount > 0.
+ */
+static inline bool dmz_is_active(struct dm_zone *zone)
+{
+	return atomic_read(&zone->refcount);
+}
 
 int dmz_lock_zone_reclaim(struct dm_zone *zone);
 void dmz_unlock_zone_reclaim(struct dm_zone *zone);

include/linux/device-mapper.h

@@ -529,29 +529,20 @@ void *dm_vcalloc(unsigned long nmemb, unsigned long elem_size);
  *---------------------------------------------------------------*/
 #define DM_NAME "device-mapper"
 
-#define DM_RATELIMIT(pr_func, fmt, ...)					\
-do {									\
-	static DEFINE_RATELIMIT_STATE(rs, DEFAULT_RATELIMIT_INTERVAL,	\
-				      DEFAULT_RATELIMIT_BURST);		\
-									\
-	if (__ratelimit(&rs))						\
-		pr_func(DM_FMT(fmt), ##__VA_ARGS__);			\
-} while (0)
-
 #define DM_FMT(fmt) DM_NAME ": " DM_MSG_PREFIX ": " fmt "\n"
 
 #define DMCRIT(fmt, ...) pr_crit(DM_FMT(fmt), ##__VA_ARGS__)
 
 #define DMERR(fmt, ...) pr_err(DM_FMT(fmt), ##__VA_ARGS__)
-#define DMERR_LIMIT(fmt, ...) DM_RATELIMIT(pr_err, fmt, ##__VA_ARGS__)
+#define DMERR_LIMIT(fmt, ...) pr_err_ratelimited(DM_FMT(fmt), ##__VA_ARGS__)
 #define DMWARN(fmt, ...) pr_warn(DM_FMT(fmt), ##__VA_ARGS__)
-#define DMWARN_LIMIT(fmt, ...) DM_RATELIMIT(pr_warn, fmt, ##__VA_ARGS__)
+#define DMWARN_LIMIT(fmt, ...) pr_warn_ratelimited(DM_FMT(fmt), ##__VA_ARGS__)
 #define DMINFO(fmt, ...) pr_info(DM_FMT(fmt), ##__VA_ARGS__)
-#define DMINFO_LIMIT(fmt, ...) DM_RATELIMIT(pr_info, fmt, ##__VA_ARGS__)
+#define DMINFO_LIMIT(fmt, ...) pr_info_ratelimited(DM_FMT(fmt), ##__VA_ARGS__)
 
 #ifdef CONFIG_DM_DEBUG
 #define DMDEBUG(fmt, ...) printk(KERN_DEBUG DM_FMT(fmt), ##__VA_ARGS__)
-#define DMDEBUG_LIMIT(fmt, ...) DM_RATELIMIT(pr_debug, fmt, ##__VA_ARGS__)
+#define DMDEBUG_LIMIT(fmt, ...) pr_debug_ratelimited(DM_FMT(fmt), ##__VA_ARGS__)
 #else
 #define DMDEBUG(fmt, ...) no_printk(fmt, ##__VA_ARGS__)
 #define DMDEBUG_LIMIT(fmt, ...) no_printk(fmt, ##__VA_ARGS__)