Merge tag 'for-7.2/dm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm

Pull device mapper fixes from Mikulas Patocka:

 - dm-log: fix overflow on 32-bit machines

 - dm-era: fix out of bounds memory access; fix crashes on invalid args

 - dm-verity: fix buffer overflow in forward error correction

 - dm-thin: fix misbehavior on I/O failures

 - dm-pcache: fix NULL pointer dereference on invalid arguments

 - dm-inlinecrypt: fix memory leak on error handling

 - dm-integrity: fix ignoring the 'fix_hmac' option on device open

 - dm: don't store the keyring in memory for a long term

 - 12 miscellaneous fixes for bugs found by Claude Opus 4.6

* tag 'for-7.2/dm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm: (23 commits)
  dm thin metadata: fix superblock refcount leak on snapshot shadow failure
  dm-stats: fix dm_jiffies_to_msec64
  dm-stats: fix merge accounting
  dm-bufio: fix wrong count calculation in dm_bufio_issue_discard
  dm-verity: make error counter atomic
  dm-verity: increase sprintf buffer size
  dm-verity: fix a possible NULL pointer dereference
  dm-verity: avoid double increment of &use_bh_wq_enabled
  dm-ioctl: fix a possible overflow in list_version_get_info
  dm_early_create: fix freeing used table on dm_resume failure
  dm-integrity: fix a bug if the bio is out of limits
  dm-integrity: don't increment hash_offset twice
  dm-integrity: fix leaking uninitialized kernel memory
  dm-integrity: fix the 'fix_hmac' option
  dm era: fix error code propagation in era_ctr()
  dm era: fix NULL pointer dereference in metadata_open()
  dm: avoid leaking the caller's thread keyring via the table device file
  dm-inlinecrypt: Fix an error handling path in inlinecrypt_ctr()
  dm-pcache: reject option groups without values
  dm thin metadata: fix metadata snapshot consistency on commit failure
  ...

drivers/md/dm-bufio.c

@@ -2238,7 +2238,9 @@ int dm_bufio_issue_discard(struct dm_bufio_client *c, sector_t block, sector_t c
 	struct dm_io_region io_reg = {
 		.bdev = c->bdev,
 		.sector = block_to_sector(c, block),
-		.count = block_to_sector(c, count),
+		.count = likely(c->sectors_per_block_bits >= 0) ?
+			count << c->sectors_per_block_bits :
+			count * (c->block_size >> SECTOR_SHIFT),
 	};
 
 	if (WARN_ON_ONCE(dm_bufio_in_request()))

drivers/md/dm-era-target.c

@@ -810,8 +810,10 @@ static struct era_metadata *metadata_open(struct block_device *bdev,
 	int r;
 	struct era_metadata *md = kzalloc_obj(*md);
 
-	if (!md)
-		return NULL;
+	if (!md) {
+		DMERR("could not allocate metadata struct");
+		return ERR_PTR(-ENOMEM);
+	}
 
 	md->bdev = bdev;
 	md->block_size = block_size;
@@ -1229,6 +1231,7 @@ static dm_block_t get_block(struct era *era, struct bio *bio)
 static void remap_to_origin(struct era *era, struct bio *bio)
 {
 	bio_set_dev(bio, era->origin_dev->bdev);
+	bio->bi_iter.bi_sector = dm_target_offset(era->ti, bio->bi_iter.bi_sector);
 }
 
 /*
@@ -1486,7 +1489,7 @@ static int era_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	if (r) {
 		ti->error = "Error opening metadata device";
 		era_destroy(era);
-		return -EINVAL;
+		return r;
 	}
 
 	r = dm_get_device(ti, argv[1], BLK_OPEN_READ | BLK_OPEN_WRITE,
@@ -1494,7 +1497,7 @@ static int era_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	if (r) {
 		ti->error = "Error opening data device";
 		era_destroy(era);
-		return -EINVAL;
+		return r;
 	}
 
 	r = sscanf(argv[2], "%u%c", &era->sectors_per_block, &dummy);
@@ -1508,7 +1511,7 @@ static int era_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	if (r) {
 		ti->error = "could not set max io len";
 		era_destroy(era);
-		return -EINVAL;
+		return r;
 	}
 
 	if (!valid_block_size(era->sectors_per_block)) {
@@ -1560,7 +1563,7 @@ static void era_dtr(struct dm_target *ti)
 static int era_map(struct dm_target *ti, struct bio *bio)
 {
 	struct era *era = ti->private;
-	dm_block_t block = get_block(era, bio);
+	dm_block_t block;
 
 	/*
 	 * All bios get remapped to the origin device.  We do this now, but
@@ -1568,6 +1571,7 @@ static int era_map(struct dm_target *ti, struct bio *bio)
 	 * block is marked in this era.
 	 */
 	remap_to_origin(era, bio);
+	block = get_block(era, bio);
 
 	/*
 	 * REQ_PREFLUSH bios carry no data, so we're not interested in them.

drivers/md/dm-inlinecrypt.c

@@ -347,7 +347,8 @@ static int inlinecrypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	err = get_key_size(&argv[1]);
 	if (err < 0) {
 		ti->error = "Cannot parse key size";
-		return -EINVAL;
+		err = -EINVAL;
+		goto bad;
 	}
 	ctx->key_size = err;
 
@@ -398,6 +399,7 @@ static int inlinecrypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	if (ctx->iv_offset & ((ctx->sector_size >> SECTOR_SHIFT) - 1)) {
 		ti->error = "Wrong alignment of iv_offset sector";
 		err = -EINVAL;
+		goto bad;
 	}
 
 	ctx->max_dun = (ctx->iv_offset + ti->len - 1) >>

drivers/md/dm-integrity.c

@@ -1480,9 +1480,6 @@ static int dm_integrity_rw_tag(struct dm_integrity_c *ic, unsigned char *tag, se
 			*metadata_offset = 0;
 		}
 
-		if (unlikely(!is_power_of_2(ic->tag_size)))
-			hash_offset = (hash_offset + to_copy) % ic->tag_size;
-
 		total_size -= to_copy;
 	} while (unlikely(total_size));
 
@@ -2523,6 +2520,9 @@ static int dm_integrity_map_inline(struct dm_integrity_io *dio, bool from_map)
 	if (unlikely((bio->bi_opf & REQ_PREFLUSH) != 0))
 		return DM_MAPIO_REMAPPED;
 
+	if (unlikely(!dm_integrity_check_limits(ic, bio->bi_iter.bi_sector, bio)))
+		return DM_MAPIO_KILL;
+
 retry:
 	if (!dio->integrity_payload) {
 		unsigned digest_size, extra_size;
@@ -2587,10 +2587,6 @@ static int dm_integrity_map_inline(struct dm_integrity_io *dio, bool from_map)
 
 	dio->bio_details.bi_iter = bio->bi_iter;
 
-	if (unlikely(!dm_integrity_check_limits(ic, bio->bi_iter.bi_sector, bio))) {
-		return DM_MAPIO_KILL;
-	}
-
 	bio->bi_iter.bi_sector += ic->start + SB_SECTORS;
 
 	bip = bio_integrity_alloc(bio, GFP_NOIO, 1);
@@ -2606,7 +2602,7 @@ static int dm_integrity_map_inline(struct dm_integrity_io *dio, bool from_map)
 			struct bio_vec bv = bio_iter_iovec(bio, dio->bio_details.bi_iter);
 			const char *mem = integrity_kmap(ic, bv.bv_page);
 			if (ic->tag_size < ic->tuple_size)
-				memset(dio->integrity_payload + pos + ic->tag_size, 0, ic->tuple_size - ic->tuple_size);
+				memset(dio->integrity_payload + pos + ic->tag_size, 0, ic->tuple_size - ic->tag_size);
 			integrity_sector_checksum(ic, &dio->ahash_req, dio->bio_details.bi_iter.bi_sector, mem, bv.bv_offset, dio->integrity_payload + pos);
 			integrity_kunmap(ic, mem);
 			pos += ic->tuple_size;
@@ -5130,6 +5126,20 @@ static int dm_integrity_ctr(struct dm_target *ti, unsigned int argc, char **argv
 		ti->error = "Journal mac mismatch";
 		goto bad;
 	}
+	if (ic->fix_hmac && !(ic->sb->flags & cpu_to_le32(SB_FLAG_FIXED_HMAC)) && ic->journal_mac_alg.key_string) {
+		/*
+		 * If this happens, it may be either because someone tampered
+		 * with the device or it may be due to a bug in the
+		 * integritysetup tool.
+		 *
+		 * In the latter case, upgrade to integritysetup 2.8.7 and use
+		 * the argument --integrity-legacy-hmac when using the open
+		 * command.
+		 */
+		r = -EINVAL;
+		ti->error = "fix_hmac is on the command line but not in the superblock";
+		goto bad;
+	}
 
 	get_provided_data_sectors(ic);
 	if (!ic->provided_data_sectors) {

drivers/md/dm-ioctl.c

@@ -785,7 +785,7 @@ static void list_version_get_info(struct target_type *tt, void *param)
 	struct vers_iter *info = param;
 
 	/* Check space - it might have changed since the first iteration */
-	if ((char *)info->vers + sizeof(tt->version) + strlen(tt->name) + 1 > info->end) {
+	if ((char *)info->vers + sizeof(struct dm_target_versions) + strlen(tt->name) + 1 > info->end) {
 		info->flags = DM_BUFFER_FULL_FLAG;
 		return;
 	}
@@ -2473,7 +2473,7 @@ int __init dm_early_create(struct dm_ioctl *dmi,
 	/* resume device */
 	r = dm_resume(md);
 	if (r)
-		goto err_destroy_table;
+		goto err_hash_remove;
 
 	DMINFO("%s (%s) is ready", md->disk->disk_name, dmi->name);
 	dm_put(md);

drivers/md/dm-log.c

@@ -425,6 +425,9 @@ static int create_log_context(struct dm_dirty_log *log, struct dm_target *ti,
 	 */
 	bitset_size = dm_round_up(region_count, BITS_PER_LONG);
 	bitset_size >>= BYTE_SHIFT;
+	/* Handle dm_round_up rollover on 32-bit systems */
+	if (!bitset_size)
+		bitset_size = 1UL << (BITS_PER_LONG - BYTE_SHIFT);
 
 	lc->bitset_uint32_count = bitset_size / sizeof(*lc->clean_bits);
 

drivers/md/dm-pcache/dm_pcache.c

@@ -168,6 +168,10 @@ static int parse_cache_opts(struct dm_pcache *pcache, struct dm_arg_set *as,
 		argc--;
 
 		if (!strcmp(arg, "cache_mode")) {
+			if (!argc) {
+				*error = "Missing value for cache_mode";
+				return -EINVAL;
+			}
 			arg = dm_shift_arg(as);
 			if (!strcmp(arg, "writeback")) {
 				opts->cache_mode = PCACHE_CACHE_MODE_WRITEBACK;
@@ -177,6 +181,10 @@ static int parse_cache_opts(struct dm_pcache *pcache, struct dm_arg_set *as,
 			}
 			argc--;
 		} else if (!strcmp(arg, "data_crc")) {
+			if (!argc) {
+				*error = "Missing value for data_crc";
+				return -EINVAL;
+			}
 			arg = dm_shift_arg(as);
 			if (!strcmp(arg, "true")) {
 				opts->data_crc = true;

drivers/md/dm-stats.c

@@ -692,10 +692,8 @@ void dm_stats_account_io(struct dm_stats *stats, unsigned long bi_rw,
 		 */
 		last = raw_cpu_ptr(stats->last);
 		stats_aux->merged =
-			(bi_sector == (READ_ONCE(last->last_sector) &&
-				       ((bi_rw == WRITE) ==
-					(READ_ONCE(last->last_rw) == WRITE))
-				       ));
+			bi_sector == READ_ONCE(last->last_sector) &&
+				(bi_rw == WRITE) == (READ_ONCE(last->last_rw) == WRITE);
 		WRITE_ONCE(last->last_sector, end_sector);
 		WRITE_ONCE(last->last_rw, bi_rw);
 	} else
@@ -842,10 +840,10 @@ static unsigned long long dm_jiffies_to_msec64(struct dm_stat *s, unsigned long
 		result = jiffies_to_msecs(j & 0x3fffff);
 	if (j >= 1 << 22) {
 		mult = jiffies_to_msecs(1 << 22);
-		result += (unsigned long long)mult * (unsigned long long)jiffies_to_msecs((j >> 22) & 0x3fffff);
+		result += (unsigned long long)mult * ((j >> 22) & 0x3fffff);
 	}
 	if (j >= 1ULL << 44)
-		result += (unsigned long long)mult * (unsigned long long)mult * (unsigned long long)jiffies_to_msecs(j >> 44);
+		result += (unsigned long long)mult * (unsigned long long)(1 << 22) * (j >> 44);
 
 	return result;
 }

drivers/md/dm-thin-metadata.c

@@ -186,6 +186,7 @@ struct dm_pool_metadata {
 	uint32_t time;
 	dm_block_t root;
 	dm_block_t details_root;
+	dm_block_t held_root;
 	struct list_head thin_devices;
 	uint64_t trans_id;
 	unsigned long flags;
@@ -748,6 +749,7 @@ static int __open_metadata(struct dm_pool_metadata *pmd)
 	 */
 	pmd->root = le64_to_cpu(disk_super->data_mapping_root);
 	pmd->details_root = le64_to_cpu(disk_super->device_details_root);
+	pmd->held_root = le64_to_cpu(disk_super->held_root);
 
 	__setup_btree_details(pmd);
 	dm_bm_unlock(sblock);
@@ -838,6 +840,7 @@ static int __begin_transaction(struct dm_pool_metadata *pmd)
 	pmd->time = le32_to_cpu(disk_super->time);
 	pmd->root = le64_to_cpu(disk_super->data_mapping_root);
 	pmd->details_root = le64_to_cpu(disk_super->device_details_root);
+	pmd->held_root = le64_to_cpu(disk_super->held_root);
 	pmd->trans_id = le64_to_cpu(disk_super->trans_id);
 	pmd->flags = le32_to_cpu(disk_super->flags);
 	pmd->data_block_size = le32_to_cpu(disk_super->data_block_size);
@@ -928,6 +931,7 @@ static int __commit_transaction(struct dm_pool_metadata *pmd)
 	disk_super->time = cpu_to_le32(pmd->time);
 	disk_super->data_mapping_root = cpu_to_le64(pmd->root);
 	disk_super->device_details_root = cpu_to_le64(pmd->details_root);
+	disk_super->held_root = cpu_to_le64(pmd->held_root);
 	disk_super->trans_id = cpu_to_le64(pmd->trans_id);
 	disk_super->flags = cpu_to_le32(pmd->flags);
 
@@ -1333,9 +1337,14 @@ static int __reserve_metadata_snap(struct dm_pool_metadata *pmd)
 {
 	int r, inc;
 	struct thin_disk_superblock *disk_super;
-	struct dm_block *copy, *sblock;
+	struct dm_block *copy;
 	dm_block_t held_root;
 
+	if (pmd->held_root) {
+		DMWARN("Pool metadata snapshot already exists: release this before taking another.");
+		return -EBUSY;
+	}
+
 	/*
 	 * We commit to ensure the btree roots which we increment in a
 	 * moment are up to date.
@@ -1353,22 +1362,16 @@ static int __reserve_metadata_snap(struct dm_pool_metadata *pmd)
 	dm_sm_inc_block(pmd->metadata_sm, THIN_SUPERBLOCK_LOCATION);
 	r = dm_tm_shadow_block(pmd->tm, THIN_SUPERBLOCK_LOCATION,
 			       &sb_validator, &copy, &inc);
-	if (r)
+	if (r) {
+		dm_sm_dec_block(pmd->metadata_sm, THIN_SUPERBLOCK_LOCATION);
 		return r;
+	}
 
 	BUG_ON(!inc);
 
 	held_root = dm_block_location(copy);
 	disk_super = dm_block_data(copy);
 
-	if (le64_to_cpu(disk_super->held_root)) {
-		DMWARN("Pool metadata snapshot already exists: release this before taking another.");
-
-		dm_tm_dec(pmd->tm, held_root);
-		dm_tm_unlock(pmd->tm, copy);
-		return -EBUSY;
-	}
-
 	/*
 	 * Wipe the spacemap since we're not publishing this.
 	 */
@@ -1384,18 +1387,8 @@ static int __reserve_metadata_snap(struct dm_pool_metadata *pmd)
 	dm_tm_inc(pmd->tm, le64_to_cpu(disk_super->device_details_root));
 	dm_tm_unlock(pmd->tm, copy);
 
-	/*
-	 * Write the held root into the superblock.
-	 */
-	r = superblock_lock(pmd, &sblock);
-	if (r) {
-		dm_tm_dec(pmd->tm, held_root);
-		return r;
-	}
+	pmd->held_root = held_root;
 
-	disk_super = dm_block_data(sblock);
-	disk_super->held_root = cpu_to_le64(held_root);
-	dm_bm_unlock(sblock);
 	return 0;
 }
 
@@ -1415,18 +1408,10 @@ static int __release_metadata_snap(struct dm_pool_metadata *pmd)
 {
 	int r;
 	struct thin_disk_superblock *disk_super;
-	struct dm_block *sblock, *copy;
+	struct dm_block *copy;
 	dm_block_t held_root;
 
-	r = superblock_lock(pmd, &sblock);
-	if (r)
-		return r;
-
-	disk_super = dm_block_data(sblock);
-	held_root = le64_to_cpu(disk_super->held_root);
-	disk_super->held_root = cpu_to_le64(0);
-
-	dm_bm_unlock(sblock);
+	held_root = pmd->held_root;
 
 	if (!held_root) {
 		DMWARN("No pool metadata snapshot found: nothing to release.");
@@ -1437,13 +1422,15 @@ static int __release_metadata_snap(struct dm_pool_metadata *pmd)
 	if (r)
 		return r;
 
+	pmd->held_root = 0;
+
 	disk_super = dm_block_data(copy);
 	dm_btree_del(&pmd->info, le64_to_cpu(disk_super->data_mapping_root));
 	dm_btree_del(&pmd->details_info, le64_to_cpu(disk_super->device_details_root));
-	dm_sm_dec_block(pmd->metadata_sm, held_root);
-
 	dm_tm_unlock(pmd->tm, copy);
 
+	dm_sm_dec_block(pmd->metadata_sm, held_root);
+
 	return 0;
 }
 
@@ -1462,19 +1449,7 @@ int dm_pool_release_metadata_snap(struct dm_pool_metadata *pmd)
 static int __get_metadata_snap(struct dm_pool_metadata *pmd,
 			       dm_block_t *result)
 {
-	int r;
-	struct thin_disk_superblock *disk_super;
-	struct dm_block *sblock;
-
-	r = dm_bm_read_lock(pmd->bm, THIN_SUPERBLOCK_LOCATION,
-			    &sb_validator, &sblock);
-	if (r)
-		return r;
-
-	disk_super = dm_block_data(sblock);
-	*result = le64_to_cpu(disk_super->held_root);
-
-	dm_bm_unlock(sblock);
+	*result = pmd->held_root;
 
 	return 0;
 }

drivers/md/dm-verity-fec.c

@@ -220,7 +220,7 @@ static int fec_read_bufs(struct dm_verity *v, struct dm_verity_io *io,
 				     PTR_ERR(bbuf));
 
 			/* assume the block is corrupted */
-			if (neras && *neras <= v->fec->roots)
+			if (neras && *neras < v->fec->roots)
 				fio->erasures[(*neras)++] = i;
 
 			continue;
@@ -238,7 +238,7 @@ static int fec_read_bufs(struct dm_verity *v, struct dm_verity_io *io,
 			 * skip if we have already found the theoretical
 			 * maximum number (i.e. fec->roots) of erasures
 			 */
-			if (neras && *neras <= v->fec->roots &&
+			if (neras && *neras < v->fec->roots &&
 			    fec_is_erasure(v, io, want_digest, bbuf))
 				fio->erasures[(*neras)++] = i;
 		}

drivers/md/dm-verity-fec.h

@@ -47,7 +47,7 @@ struct dm_verity_fec {
 /* per-bio data */
 struct dm_verity_fec_io {
 	struct rs_control *rs;	/* Reed-Solomon state */
-	int erasures[DM_VERITY_FEC_MAX_ROOTS + 1]; /* erasures for decode_rs8 */
+	int erasures[DM_VERITY_FEC_MAX_ROOTS]; /* erasures for decode_rs8 */
 	u8 *output;		/* buffer for corrected output */
 	unsigned int level;		/* recursion level */
 	unsigned int nbufs;		/* number of buffers allocated */

drivers/md/dm-verity-loadpin.c

@@ -70,7 +70,7 @@ bool dm_verity_loadpin_is_bdev_trusted(struct block_device *bdev)
 
 	table = dm_get_live_table(md, &srcu_idx);
 
-	if (table->num_targets != 1)
+	if (!table || table->num_targets != 1)
 		goto out;
 
 	ti = dm_table_get_target(table, 0);

drivers/md/dm-verity-target.c

@@ -26,7 +26,7 @@
 
 #define DM_MSG_PREFIX			"verity"
 
-#define DM_VERITY_ENV_LENGTH		42
+#define DM_VERITY_ENV_LENGTH		46
 #define DM_VERITY_ENV_VAR_NAME		"DM_VERITY_ERR_BLOCK_NR"
 
 #define DM_VERITY_DEFAULT_PREFETCH_SIZE	262144
@@ -180,14 +180,16 @@ static int verity_handle_err(struct dm_verity *v, enum verity_block_type type,
 	char *envp[] = { verity_env, NULL };
 	const char *type_str = "";
 	struct mapped_device *md = dm_table_get_md(v->ti->table);
+	int ce;
 
 	/* Corruption should be visible in device status in all modes */
 	v->hash_failed = true;
 
-	if (v->corrupted_errs >= DM_VERITY_MAX_CORRUPTED_ERRS)
-		goto out;
-
-	v->corrupted_errs++;
+	ce = atomic_read(&v->corrupted_errs);
+	do {
+		if (ce >= DM_VERITY_MAX_CORRUPTED_ERRS)
+			goto out;
+	} while (!atomic_try_cmpxchg(&v->corrupted_errs, &ce, ce + 1));
 
 	switch (type) {
 	case DM_VERITY_BLOCK_TYPE_DATA:
@@ -203,7 +205,7 @@ static int verity_handle_err(struct dm_verity *v, enum verity_block_type type,
 	DMERR_LIMIT("%s: %s block %llu is corrupted", v->data_dev->name,
 		    type_str, block);
 
-	if (v->corrupted_errs == DM_VERITY_MAX_CORRUPTED_ERRS) {
+	if (ce + 1 == DM_VERITY_MAX_CORRUPTED_ERRS) {
 		DMERR("%s: reached maximum errors", v->data_dev->name);
 		dm_audit_log_target(DM_MSG_PREFIX, "max-corrupted-errors", v->ti, 0);
 	}
@@ -1262,6 +1264,8 @@ static int verity_parse_opt_args(struct dm_arg_set *as, struct dm_verity *v,
 			continue;
 
 		} else if (!strcasecmp(arg_name, DM_VERITY_OPT_TASKLET_VERIFY)) {
+			if (v->use_bh_wq)
+				continue;
 			v->use_bh_wq = true;
 			static_branch_inc(&use_bh_wq_enabled);
 			continue;

drivers/md/dm-verity.h

@@ -68,7 +68,7 @@ struct dm_verity {
 	unsigned int digest_size;	/* digest size for the current hash algorithm */
 	enum verity_mode mode;	/* mode for handling verification errors */
 	enum verity_mode error_mode;/* mode for handling I/O errors */
-	unsigned int corrupted_errs;/* Number of errors for corrupted blocks */
+	atomic_t corrupted_errs;/* Number of errors for corrupted blocks */
 
 	struct workqueue_struct *verify_wq;
 

drivers/md/dm.c

@@ -735,7 +735,16 @@ static struct table_device *open_table_device(struct mapped_device *md,
 		return ERR_PTR(-ENOMEM);
 	refcount_set(&td->count, 1);
 
-	bdev_file = bdev_file_open_by_dev(dev, mode, _dm_claim_ptr, NULL);
+	/*
+	 * Open the backing device with kernel rather than caller
+	 * credentials. Otherwise the caller's credentials would be
+	 * pinned in bdev_file->f_cred until the table device is closed.
+	 * That would keep the caller's thread keyring alive long beyond the
+	 * lifetime of the caller, breaking userspace expectation (e.g.
+	 * cryptsetup(8) leaking the LUKS volume key).
+	 */
+	scoped_with_kernel_creds()
+		bdev_file = bdev_file_open_by_dev(dev, mode, _dm_claim_ptr, NULL);
 	if (IS_ERR(bdev_file)) {
 		r = PTR_ERR(bdev_file);
 		goto out_free_td;