Merge tag 'for-6.7/dm-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm

Pull device mapper updates from Mike Snitzer:

 - Update DM core to directly call the map function for both the linear
   and stripe targets; which are provided by DM core

 - Various updates to use new safer string functions

 - Update DM core to respect REQ_NOWAIT flag in normal bios so that
   memory allocations are always attempted with GFP_NOWAIT

 - Add Mikulas Patocka to MAINTAINERS as a DM maintainer!

 - Improve DM delay target's handling of short delays (< 50ms) by using
   a kthread to check expiration of IOs rather than timers and a wq

 - Update the DM error target so that it works with zoned storage. This
   helps xfstests to provide proper IO error handling coverage when
   testing a filesystem with native zoned storage support

 - Update both DM crypt and integrity targets to improve performance by
   using crypto_shash_digest() rather than init+update+final sequence

 - Fix DM crypt target by backfilling missing memory allocation
   accounting for compound pages

* tag 'for-6.7/dm-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
  dm crypt: account large pages in cc->n_allocated_pages
  dm integrity: use crypto_shash_digest() in sb_mac()
  dm crypt: use crypto_shash_digest() in crypt_iv_tcw_whitening()
  dm error: Add support for zoned block devices
  dm delay: for short delays, use kthread instead of timers and wq
  MAINTAINERS: add Mikulas Patocka as a DM maintainer
  dm: respect REQ_NOWAIT flag in normal bios issued to DM
  dm: enhance alloc_multiple_bios() to be more versatile
  dm: make __send_duplicate_bios return unsigned int
  dm log userspace: replace deprecated strncpy with strscpy
  dm ioctl: replace deprecated strncpy with strscpy_pad
  dm crypt: replace open-coded kmemdup_nul
  dm cache metadata: replace deprecated strncpy with strscpy
  dm: shortcut the calls to linear_map and stripe_map

Author: torvalds

MAINTAINERS

@@ -6028,6 +6028,7 @@ F:	include/linux/devm-helpers.h
 DEVICE-MAPPER  (LVM)
 M:	Alasdair Kergon <[email protected]>
 M:	Mike Snitzer <[email protected]>
+M:	Mikulas Patocka <[email protected]>
 M:	[email protected]
 L:	[email protected]
 S:	Maintained

drivers/md/dm-cache-metadata.c

@@ -597,7 +597,7 @@ static void read_superblock_fields(struct dm_cache_metadata *cmd,
 	cmd->discard_nr_blocks = to_dblock(le64_to_cpu(disk_super->discard_nr_blocks));
 	cmd->data_block_size = le32_to_cpu(disk_super->data_block_size);
 	cmd->cache_blocks = to_cblock(le32_to_cpu(disk_super->cache_blocks));
-	strncpy(cmd->policy_name, disk_super->policy_name, sizeof(cmd->policy_name));
+	strscpy(cmd->policy_name, disk_super->policy_name, sizeof(cmd->policy_name));
 	cmd->policy_version[0] = le32_to_cpu(disk_super->policy_version[0]);
 	cmd->policy_version[1] = le32_to_cpu(disk_super->policy_version[1]);
 	cmd->policy_version[2] = le32_to_cpu(disk_super->policy_version[2]);
@@ -707,7 +707,7 @@ static int __commit_transaction(struct dm_cache_metadata *cmd,
 	disk_super->discard_block_size = cpu_to_le64(cmd->discard_block_size);
 	disk_super->discard_nr_blocks = cpu_to_le64(from_dblock(cmd->discard_nr_blocks));
 	disk_super->cache_blocks = cpu_to_le32(from_cblock(cmd->cache_blocks));
-	strncpy(disk_super->policy_name, cmd->policy_name, sizeof(disk_super->policy_name));
+	strscpy(disk_super->policy_name, cmd->policy_name, sizeof(disk_super->policy_name));
 	disk_super->policy_version[0] = cpu_to_le32(cmd->policy_version[0]);
 	disk_super->policy_version[1] = cpu_to_le32(cmd->policy_version[1]);
 	disk_super->policy_version[2] = cpu_to_le32(cmd->policy_version[2]);
@@ -1726,7 +1726,7 @@ static int write_hints(struct dm_cache_metadata *cmd, struct dm_cache_policy *po
 	    (strlen(policy_name) > sizeof(cmd->policy_name) - 1))
 		return -EINVAL;
 
-	strncpy(cmd->policy_name, policy_name, sizeof(cmd->policy_name));
+	strscpy(cmd->policy_name, policy_name, sizeof(cmd->policy_name));
 	memcpy(cmd->policy_version, policy_version, sizeof(cmd->policy_version));
 
 	hint_size = dm_cache_policy_get_hint_size(policy);

drivers/md/dm-crypt.c

@@ -652,13 +652,7 @@ static int crypt_iv_tcw_whitening(struct crypt_config *cc,
 	/* calculate crc32 for every 32bit part and xor it */
 	desc->tfm = tcw->crc32_tfm;
 	for (i = 0; i < 4; i++) {
-		r = crypto_shash_init(desc);
-		if (r)
-			goto out;
-		r = crypto_shash_update(desc, &buf[i * 4], 4);
-		if (r)
-			goto out;
-		r = crypto_shash_final(desc, &buf[i * 4]);
+		r = crypto_shash_digest(desc, &buf[i * 4], 4, &buf[i * 4]);
 		if (r)
 			goto out;
 	}
@@ -1699,11 +1693,17 @@ static struct bio *crypt_alloc_buffer(struct dm_crypt_io *io, unsigned int size)
 		order = min(order, remaining_order);
 
 		while (order > 0) {
+			if (unlikely(percpu_counter_read_positive(&cc->n_allocated_pages) +
+					(1 << order) > dm_crypt_pages_per_client))
+				goto decrease_order;
 			pages = alloc_pages(gfp_mask
 				| __GFP_NOMEMALLOC | __GFP_NORETRY | __GFP_NOWARN | __GFP_COMP,
 				order);
-			if (likely(pages != NULL))
+			if (likely(pages != NULL)) {
+				percpu_counter_add(&cc->n_allocated_pages, 1 << order);
 				goto have_pages;
+			}
+decrease_order:
 			order--;
 		}
 
@@ -1741,10 +1741,13 @@ static void crypt_free_buffer_pages(struct crypt_config *cc, struct bio *clone)
 
 	if (clone->bi_vcnt > 0) { /* bio_for_each_folio_all crashes with an empty bio */
 		bio_for_each_folio_all(fi, clone) {
-			if (folio_test_large(fi.folio))
+			if (folio_test_large(fi.folio)) {
+				percpu_counter_sub(&cc->n_allocated_pages,
+						1 << folio_order(fi.folio));
 				folio_put(fi.folio);
-			else
+			} else {
 				mempool_free(&fi.folio->page, &cc->page_pool);
+			}
 		}
 	}
 }
@@ -2859,10 +2862,9 @@ static int crypt_ctr_auth_cipher(struct crypt_config *cc, char *cipher_api)
 	if (!start || !end || ++start > end)
 		return -EINVAL;
 
-	mac_alg = kzalloc(end - start + 1, GFP_KERNEL);
+	mac_alg = kmemdup_nul(start, end - start, GFP_KERNEL);
 	if (!mac_alg)
 		return -ENOMEM;
-	strncpy(mac_alg, start, end - start);
 
 	mac = crypto_alloc_ahash(mac_alg, 0, CRYPTO_ALG_ALLOCATES_MEMORY);
 	kfree(mac_alg);

drivers/md/dm-delay.c

@@ -13,6 +13,7 @@
 #include <linux/blkdev.h>
 #include <linux/bio.h>
 #include <linux/slab.h>
+#include <linux/kthread.h>
 
 #include <linux/device-mapper.h>
 
@@ -31,6 +32,7 @@ struct delay_c {
 	struct workqueue_struct *kdelayd_wq;
 	struct work_struct flush_expired_bios;
 	struct list_head delayed_bios;
+	struct task_struct *worker;
 	atomic_t may_delay;
 
 	struct delay_class read;
@@ -66,6 +68,44 @@ static void queue_timeout(struct delay_c *dc, unsigned long expires)
 	mutex_unlock(&dc->timer_lock);
 }
 
+static inline bool delay_is_fast(struct delay_c *dc)
+{
+	return !!dc->worker;
+}
+
+static void flush_delayed_bios_fast(struct delay_c *dc, bool flush_all)
+{
+	struct dm_delay_info *delayed, *next;
+
+	mutex_lock(&delayed_bios_lock);
+	list_for_each_entry_safe(delayed, next, &dc->delayed_bios, list) {
+		if (flush_all || time_after_eq(jiffies, delayed->expires)) {
+			struct bio *bio = dm_bio_from_per_bio_data(delayed,
+						sizeof(struct dm_delay_info));
+			list_del(&delayed->list);
+			dm_submit_bio_remap(bio, NULL);
+			delayed->class->ops--;
+		}
+	}
+	mutex_unlock(&delayed_bios_lock);
+}
+
+static int flush_worker_fn(void *data)
+{
+	struct delay_c *dc = data;
+
+	while (1) {
+		flush_delayed_bios_fast(dc, false);
+		if (unlikely(list_empty(&dc->delayed_bios))) {
+			set_current_state(TASK_INTERRUPTIBLE);
+			schedule();
+		} else
+			cond_resched();
+	}
+
+	return 0;
+}
+
 static void flush_bios(struct bio *bio)
 {
 	struct bio *n;
@@ -78,7 +118,7 @@ static void flush_bios(struct bio *bio)
 	}
 }
 
-static struct bio *flush_delayed_bios(struct delay_c *dc, int flush_all)
+static struct bio *flush_delayed_bios(struct delay_c *dc, bool flush_all)
 {
 	struct dm_delay_info *delayed, *next;
 	unsigned long next_expires = 0;
@@ -115,7 +155,10 @@ static void flush_expired_bios(struct work_struct *work)
 	struct delay_c *dc;
 
 	dc = container_of(work, struct delay_c, flush_expired_bios);
-	flush_bios(flush_delayed_bios(dc, 0));
+	if (delay_is_fast(dc))
+		flush_delayed_bios_fast(dc, false);
+	else
+		flush_bios(flush_delayed_bios(dc, false));
 }
 
 static void delay_dtr(struct dm_target *ti)
@@ -131,8 +174,11 @@ static void delay_dtr(struct dm_target *ti)
 		dm_put_device(ti, dc->write.dev);
 	if (dc->flush.dev)
 		dm_put_device(ti, dc->flush.dev);
+	if (dc->worker)
+		kthread_stop(dc->worker);
 
-	mutex_destroy(&dc->timer_lock);
+	if (!delay_is_fast(dc))
+		mutex_destroy(&dc->timer_lock);
 
 	kfree(dc);
 }
@@ -175,6 +221,7 @@ static int delay_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 {
 	struct delay_c *dc;
 	int ret;
+	unsigned int max_delay;
 
 	if (argc != 3 && argc != 6 && argc != 9) {
 		ti->error = "Requires exactly 3, 6 or 9 arguments";
@@ -188,16 +235,14 @@ static int delay_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	}
 
 	ti->private = dc;
-	timer_setup(&dc->delay_timer, handle_delayed_timer, 0);
-	INIT_WORK(&dc->flush_expired_bios, flush_expired_bios);
 	INIT_LIST_HEAD(&dc->delayed_bios);
-	mutex_init(&dc->timer_lock);
 	atomic_set(&dc->may_delay, 1);
 	dc->argc = argc;
 
 	ret = delay_class_ctr(ti, &dc->read, argv);
 	if (ret)
 		goto bad;
+	max_delay = dc->read.delay;
 
 	if (argc == 3) {
 		ret = delay_class_ctr(ti, &dc->write, argv);
@@ -206,6 +251,8 @@ static int delay_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 		ret = delay_class_ctr(ti, &dc->flush, argv);
 		if (ret)
 			goto bad;
+		max_delay = max(max_delay, dc->write.delay);
+		max_delay = max(max_delay, dc->flush.delay);
 		goto out;
 	}
 
@@ -216,19 +263,37 @@ static int delay_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 		ret = delay_class_ctr(ti, &dc->flush, argv + 3);
 		if (ret)
 			goto bad;
+		max_delay = max(max_delay, dc->flush.delay);
 		goto out;
 	}
 
 	ret = delay_class_ctr(ti, &dc->flush, argv + 6);
 	if (ret)
 		goto bad;
+	max_delay = max(max_delay, dc->flush.delay);
 
 out:
-	dc->kdelayd_wq = alloc_workqueue("kdelayd", WQ_MEM_RECLAIM, 0);
-	if (!dc->kdelayd_wq) {
-		ret = -EINVAL;
-		DMERR("Couldn't start kdelayd");
-		goto bad;
+	if (max_delay < 50) {
+		/*
+		 * In case of small requested delays, use kthread instead of
+		 * timers and workqueue to achieve better latency.
+		 */
+		dc->worker = kthread_create(&flush_worker_fn, dc,
+					    "dm-delay-flush-worker");
+		if (IS_ERR(dc->worker)) {
+			ret = PTR_ERR(dc->worker);
+			goto bad;
+		}
+	} else {
+		timer_setup(&dc->delay_timer, handle_delayed_timer, 0);
+		INIT_WORK(&dc->flush_expired_bios, flush_expired_bios);
+		mutex_init(&dc->timer_lock);
+		dc->kdelayd_wq = alloc_workqueue("kdelayd", WQ_MEM_RECLAIM, 0);
+		if (!dc->kdelayd_wq) {
+			ret = -EINVAL;
+			DMERR("Couldn't start kdelayd");
+			goto bad;
+		}
 	}
 
 	ti->num_flush_bios = 1;
@@ -260,7 +325,10 @@ static int delay_bio(struct delay_c *dc, struct delay_class *c, struct bio *bio)
 	list_add_tail(&delayed->list, &dc->delayed_bios);
 	mutex_unlock(&delayed_bios_lock);
 
-	queue_timeout(dc, expires);
+	if (delay_is_fast(dc))
+		wake_up_process(dc->worker);
+	else
+		queue_timeout(dc, expires);
 
 	return DM_MAPIO_SUBMITTED;
 }
@@ -270,8 +338,13 @@ static void delay_presuspend(struct dm_target *ti)
 	struct delay_c *dc = ti->private;
 
 	atomic_set(&dc->may_delay, 0);
-	del_timer_sync(&dc->delay_timer);
-	flush_bios(flush_delayed_bios(dc, 1));
+
+	if (delay_is_fast(dc))
+		flush_delayed_bios_fast(dc, true);
+	else {
+		del_timer_sync(&dc->delay_timer);
+		flush_bios(flush_delayed_bios(dc, true));
+	}
 }
 
 static void delay_resume(struct dm_target *ti)
@@ -356,7 +429,7 @@ static int delay_iterate_devices(struct dm_target *ti,
 
 static struct target_type delay_target = {
 	.name	     = "delay",
-	.version     = {1, 3, 0},
+	.version     = {1, 4, 0},
 	.features    = DM_TARGET_PASSES_INTEGRITY,
 	.module      = THIS_MODULE,
 	.ctr	     = delay_ctr,

drivers/md/dm-integrity.c

@@ -493,42 +493,32 @@ static int sb_mac(struct dm_integrity_c *ic, bool wr)
 {
 	SHASH_DESC_ON_STACK(desc, ic->journal_mac);
 	int r;
-	unsigned int size = crypto_shash_digestsize(ic->journal_mac);
+	unsigned int mac_size = crypto_shash_digestsize(ic->journal_mac);
+	__u8 *sb = (__u8 *)ic->sb;
+	__u8 *mac = sb + (1 << SECTOR_SHIFT) - mac_size;
 
-	if (sizeof(struct superblock) + size > 1 << SECTOR_SHIFT) {
+	if (sizeof(struct superblock) + mac_size > 1 << SECTOR_SHIFT) {
 		dm_integrity_io_error(ic, "digest is too long", -EINVAL);
 		return -EINVAL;
 	}
 
 	desc->tfm = ic->journal_mac;
 
-	r = crypto_shash_init(desc);
-	if (unlikely(r < 0)) {
-		dm_integrity_io_error(ic, "crypto_shash_init", r);
-		return r;
-	}
-
-	r = crypto_shash_update(desc, (__u8 *)ic->sb, (1 << SECTOR_SHIFT) - size);
-	if (unlikely(r < 0)) {
-		dm_integrity_io_error(ic, "crypto_shash_update", r);
-		return r;
-	}
-
 	if (likely(wr)) {
-		r = crypto_shash_final(desc, (__u8 *)ic->sb + (1 << SECTOR_SHIFT) - size);
+		r = crypto_shash_digest(desc, sb, mac - sb, mac);
 		if (unlikely(r < 0)) {
-			dm_integrity_io_error(ic, "crypto_shash_final", r);
+			dm_integrity_io_error(ic, "crypto_shash_digest", r);
 			return r;
 		}
 	} else {
-		__u8 result[HASH_MAX_DIGESTSIZE];
+		__u8 actual_mac[HASH_MAX_DIGESTSIZE];
 
-		r = crypto_shash_final(desc, result);
+		r = crypto_shash_digest(desc, sb, mac - sb, actual_mac);
 		if (unlikely(r < 0)) {
-			dm_integrity_io_error(ic, "crypto_shash_final", r);
+			dm_integrity_io_error(ic, "crypto_shash_digest", r);
 			return r;
 		}
-		if (memcmp((__u8 *)ic->sb + (1 << SECTOR_SHIFT) - size, result, size)) {
+		if (memcmp(mac, actual_mac, mac_size)) {
 			dm_integrity_io_error(ic, "superblock mac", -EILSEQ);
 			dm_audit_log_target(DM_MSG_PREFIX, "mac-superblock", ic->ti, 0);
 			return -EILSEQ;

drivers/md/dm-ioctl.c

@@ -1295,8 +1295,8 @@ static void retrieve_status(struct dm_table *table,
 		spec->status = 0;
 		spec->sector_start = ti->begin;
 		spec->length = ti->len;
-		strncpy(spec->target_type, ti->type->name,
-			sizeof(spec->target_type) - 1);
+		strscpy_pad(spec->target_type, ti->type->name,
+			sizeof(spec->target_type));
 
 		outptr += sizeof(struct dm_target_spec);
 		remaining = len - (outptr - outbuf);

drivers/md/dm-linear.c

@@ -85,7 +85,7 @@ static sector_t linear_map_sector(struct dm_target *ti, sector_t bi_sector)
 	return lc->start + dm_target_offset(ti, bi_sector);
 }
 
-static int linear_map(struct dm_target *ti, struct bio *bio)
+int linear_map(struct dm_target *ti, struct bio *bio)
 {
 	struct linear_c *lc = ti->private;
 

drivers/md/dm-log-userspace-base.c

@@ -224,7 +224,7 @@ static int userspace_ctr(struct dm_dirty_log *log, struct dm_target *ti,
 
 	lc->usr_argc = argc;
 
-	strncpy(lc->uuid, argv[0], DM_UUID_LEN);
+	strscpy(lc->uuid, argv[0], sizeof(lc->uuid));
 	argc--;
 	argv++;
 	spin_lock_init(&lc->flush_lock);

drivers/md/dm-stripe.c

@@ -268,7 +268,7 @@ static int stripe_map_range(struct stripe_c *sc, struct bio *bio,
 	return DM_MAPIO_SUBMITTED;
 }
 
-static int stripe_map(struct dm_target *ti, struct bio *bio)
+int stripe_map(struct dm_target *ti, struct bio *bio)
 {
 	struct stripe_c *sc = ti->private;
 	uint32_t stripe;