Merge tag 'for-5.12/dm-changes' of...

Merge tag 'for-5.12/dm-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm

Pull device mapper updates from Mike Snitzer:

 - Fix DM integrity's HMAC support to provide enhanced security of
   internal_hash and journal_mac capabilities.

 - Various DM writecache fixes to address performance, fix table output
   to match what was provided at table creation, fix writing beyond end
   of device when shrinking underlying data device, and a couple other
   small cleanups.

 - Add DM crypt support for using trusted keys.

 - Fix deadlock when swapping to DM crypt device by throttling number of
   in-flight REQ_SWAP bios. Implemented in DM core so that other
   bio-based targets can opt-in by setting ti->limit_swap_bios.

 - Fix various inverted logic bugs in the .iterate_devices callout
   functions that are used to assess if specific feature or capability
   is supported across all devices being combined/stacked by DM.

 - Fix DM era target bugs that exposed users to lost writes or memory
   leaks.

 - Add DM core support for passing through inline crypto support of
   underlying devices. Includes block/keyslot-manager changes that
   enable extending this support to DM.

 - Various small fixes and cleanups (spelling fixes, front padding
   calculation cleanup, cleanup conditional zoned support in targets,
   etc).

* tag 'for-5.12/dm-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm: (31 commits)
  dm: fix deadlock when swapping to encrypted device
  dm: simplify target code conditional on CONFIG_BLK_DEV_ZONED
  dm: set DM_TARGET_PASSES_CRYPTO feature for some targets
  dm: support key eviction from keyslot managers of underlying devices
  dm: add support for passing through inline crypto support
  block/keyslot-manager: Introduce functions for device mapper support
  block/keyslot-manager: Introduce passthrough keyslot manager
  dm era: only resize metadata in preresume
  dm era: Use correct value size in equality function of writeset tree
  dm era: Fix bitset memory leaks
  dm era: Verify the data block size hasn't changed
  dm era: Reinitialize bitset cache before digesting a new writeset
  dm era: Update in-core bitset after committing the metadata
  dm era: Recover committed writeset after crash
  dm writecache: use bdev_nr_sectors() instead of open-coded equivalent
  dm writecache: fix writing beyond end of underlying device when shrinking
  dm table: remove needless request_queue NULL pointer checks
  dm table: fix zoned iterate_devices based device capability checks
  dm table: fix DAX iterate_devices based device capability checks
  dm table: fix iterate_devices based device capability checks
  ...

Documentation/admin-guide/device-mapper/dm-crypt.rst

@@ -67,7 +67,7 @@ Parameters::
     the value passed in <key_size>.
 
 <key_type>
-    Either 'logon', 'user' or 'encrypted' kernel key type.
+    Either 'logon', 'user', 'encrypted' or 'trusted' kernel key type.
 
 <key_description>
     The kernel keyring key description crypt target should look for

Documentation/admin-guide/device-mapper/dm-integrity.rst

@@ -186,6 +186,17 @@ fix_padding
 	space-efficient. If this option is not present, large padding is
 	used - that is for compatibility with older kernels.
 
+fix_hmac
+	Improve security of internal_hash and journal_mac:
+
+	- the section number is mixed to the mac, so that an attacker can't
+	  copy sectors from one journal section to another journal section
+	- the superblock is protected by journal_mac
+	- a 16-byte salt stored in the superblock is mixed to the mac, so
+	  that the attacker can't detect that two disks have the same hmac
+	  key and also to disallow the attacker to move sectors from one
+	  disk to another
+
 legacy_recalculate
 	Allow recalculating of volumes with HMAC keys. This is disabled by
 	default for security reasons - an attacker could modify the volume,

block/blk-crypto.c

@@ -409,3 +409,4 @@ int blk_crypto_evict_key(struct request_queue *q,
 	 */
 	return blk_crypto_fallback_evict_key(key);
 }
+EXPORT_SYMBOL_GPL(blk_crypto_evict_key);

block/keyslot-manager.c

@@ -63,6 +63,11 @@ static inline void blk_ksm_hw_exit(struct blk_keyslot_manager *ksm)
 		pm_runtime_put_sync(ksm->dev);
 }
 
+static inline bool blk_ksm_is_passthrough(struct blk_keyslot_manager *ksm)
+{
+	return ksm->num_slots == 0;
+}
+
 /**
  * blk_ksm_init() - Initialize a keyslot manager
  * @ksm: The keyslot_manager to initialize.
@@ -234,6 +239,10 @@ blk_status_t blk_ksm_get_slot_for_key(struct blk_keyslot_manager *ksm,
 	int err;
 
 	*slot_ptr = NULL;
+
+	if (blk_ksm_is_passthrough(ksm))
+		return BLK_STS_OK;
+
 	down_read(&ksm->lock);
 	slot = blk_ksm_find_and_grab_keyslot(ksm, key);
 	up_read(&ksm->lock);
@@ -354,6 +363,16 @@ int blk_ksm_evict_key(struct blk_keyslot_manager *ksm,
 	struct blk_ksm_keyslot *slot;
 	int err = 0;
 
+	if (blk_ksm_is_passthrough(ksm)) {
+		if (ksm->ksm_ll_ops.keyslot_evict) {
+			blk_ksm_hw_enter(ksm);
+			err = ksm->ksm_ll_ops.keyslot_evict(ksm, key, -1);
+			blk_ksm_hw_exit(ksm);
+			return err;
+		}
+		return 0;
+	}
+
 	blk_ksm_hw_enter(ksm);
 	slot = blk_ksm_find_keyslot(ksm, key);
 	if (!slot)
@@ -389,6 +408,9 @@ void blk_ksm_reprogram_all_keys(struct blk_keyslot_manager *ksm)
 {
 	unsigned int slot;
 
+	if (blk_ksm_is_passthrough(ksm))
+		return;
+
 	/* This is for device initialization, so don't resume the device */
 	down_write(&ksm->lock);
 	for (slot = 0; slot < ksm->num_slots; slot++) {
@@ -430,3 +452,127 @@ void blk_ksm_unregister(struct request_queue *q)
 {
 	q->ksm = NULL;
 }
+
+/**
+ * blk_ksm_intersect_modes() - restrict supported modes by child device
+ * @parent: The keyslot manager for parent device
+ * @child: The keyslot manager for child device, or NULL
+ *
+ * Clear any crypto mode support bits in @parent that aren't set in @child.
+ * If @child is NULL, then all parent bits are cleared.
+ *
+ * Only use this when setting up the keyslot manager for a layered device,
+ * before it's been exposed yet.
+ */
+void blk_ksm_intersect_modes(struct blk_keyslot_manager *parent,
+			     const struct blk_keyslot_manager *child)
+{
+	if (child) {
+		unsigned int i;
+
+		parent->max_dun_bytes_supported =
+			min(parent->max_dun_bytes_supported,
+			    child->max_dun_bytes_supported);
+		for (i = 0; i < ARRAY_SIZE(child->crypto_modes_supported);
+		     i++) {
+			parent->crypto_modes_supported[i] &=
+				child->crypto_modes_supported[i];
+		}
+	} else {
+		parent->max_dun_bytes_supported = 0;
+		memset(parent->crypto_modes_supported, 0,
+		       sizeof(parent->crypto_modes_supported));
+	}
+}
+EXPORT_SYMBOL_GPL(blk_ksm_intersect_modes);
+
+/**
+ * blk_ksm_is_superset() - Check if a KSM supports a superset of crypto modes
+ *			   and DUN bytes that another KSM supports. Here,
+ *			   "superset" refers to the mathematical meaning of the
+ *			   word - i.e. if two KSMs have the *same* capabilities,
+ *			   they *are* considered supersets of each other.
+ * @ksm_superset: The KSM that we want to verify is a superset
+ * @ksm_subset: The KSM that we want to verify is a subset
+ *
+ * Return: True if @ksm_superset supports a superset of the crypto modes and DUN
+ *	   bytes that @ksm_subset supports.
+ */
+bool blk_ksm_is_superset(struct blk_keyslot_manager *ksm_superset,
+			 struct blk_keyslot_manager *ksm_subset)
+{
+	int i;
+
+	if (!ksm_subset)
+		return true;
+
+	if (!ksm_superset)
+		return false;
+
+	for (i = 0; i < ARRAY_SIZE(ksm_superset->crypto_modes_supported); i++) {
+		if (ksm_subset->crypto_modes_supported[i] &
+		    (~ksm_superset->crypto_modes_supported[i])) {
+			return false;
+		}
+	}
+
+	if (ksm_subset->max_dun_bytes_supported >
+	    ksm_superset->max_dun_bytes_supported) {
+		return false;
+	}
+
+	return true;
+}
+EXPORT_SYMBOL_GPL(blk_ksm_is_superset);
+
+/**
+ * blk_ksm_update_capabilities() - Update the restrictions of a KSM to those of
+ *				   another KSM
+ * @target_ksm: The KSM whose restrictions to update.
+ * @reference_ksm: The KSM to whose restrictions this function will update
+ *		   @target_ksm's restrictions to.
+ *
+ * Blk-crypto requires that crypto capabilities that were
+ * advertised when a bio was created continue to be supported by the
+ * device until that bio is ended. This is turn means that a device cannot
+ * shrink its advertised crypto capabilities without any explicit
+ * synchronization with upper layers. So if there's no such explicit
+ * synchronization, @reference_ksm must support all the crypto capabilities that
+ * @target_ksm does
+ * (i.e. we need blk_ksm_is_superset(@reference_ksm, @target_ksm) == true).
+ *
+ * Note also that as long as the crypto capabilities are being expanded, the
+ * order of updates becoming visible is not important because it's alright
+ * for blk-crypto to see stale values - they only cause blk-crypto to
+ * believe that a crypto capability isn't supported when it actually is (which
+ * might result in blk-crypto-fallback being used if available, or the bio being
+ * failed).
+ */
+void blk_ksm_update_capabilities(struct blk_keyslot_manager *target_ksm,
+				 struct blk_keyslot_manager *reference_ksm)
+{
+	memcpy(target_ksm->crypto_modes_supported,
+	       reference_ksm->crypto_modes_supported,
+	       sizeof(target_ksm->crypto_modes_supported));
+
+	target_ksm->max_dun_bytes_supported =
+				reference_ksm->max_dun_bytes_supported;
+}
+EXPORT_SYMBOL_GPL(blk_ksm_update_capabilities);
+
+/**
+ * blk_ksm_init_passthrough() - Init a passthrough keyslot manager
+ * @ksm: The keyslot manager to init
+ *
+ * Initialize a passthrough keyslot manager.
+ * Called by e.g. storage drivers to set up a keyslot manager in their
+ * request_queue, when the storage driver wants to manage its keys by itself.
+ * This is useful for inline encryption hardware that doesn't have the concept
+ * of keyslots, and for layered devices.
+ */
+void blk_ksm_init_passthrough(struct blk_keyslot_manager *ksm)
+{
+	memset(ksm, 0, sizeof(*ksm));
+	init_rwsem(&ksm->lock);
+}
+EXPORT_SYMBOL_GPL(blk_ksm_init_passthrough);

drivers/md/Kconfig

@@ -270,6 +270,7 @@ config DM_CRYPT
 	tristate "Crypt target support"
 	depends on BLK_DEV_DM
 	depends on (ENCRYPTED_KEYS || ENCRYPTED_KEYS=n)
+	depends on (TRUSTED_KEYS || TRUSTED_KEYS=n)
 	select CRYPTO
 	select CRYPTO_CBC
 	select CRYPTO_ESSIV

drivers/md/dm-core.h

@@ -13,6 +13,7 @@
 #include <linux/ktime.h>
 #include <linux/genhd.h>
 #include <linux/blk-mq.h>
+#include <linux/keyslot-manager.h>
 
 #include <trace/events/block.h>
 
@@ -102,6 +103,10 @@ struct mapped_device {
 	/* kobject and completion */
 	struct dm_kobject_holder kobj_holder;
 
+	int swap_bios;
+	struct semaphore swap_bios_semaphore;
+	struct mutex swap_bios_lock;
+
 	struct dm_stats stats;
 
 	/* for blk-mq request-based DM support */
@@ -162,6 +167,10 @@ struct dm_table {
 	void *event_context;
 
 	struct dm_md_mempools *mempools;
+
+#ifdef CONFIG_BLK_INLINE_ENCRYPTION
+	struct blk_keyslot_manager *ksm;
+#endif
 };
 
 static inline struct completion *dm_get_completion_from_kobject(struct kobject *kobj)

drivers/md/dm-crypt.c

@@ -37,6 +37,7 @@
 #include <linux/key-type.h>
 #include <keys/user-type.h>
 #include <keys/encrypted-type.h>
+#include <keys/trusted-type.h>
 
 #include <linux/device-mapper.h>
 
@@ -133,7 +134,7 @@ enum flags { DM_CRYPT_SUSPENDED, DM_CRYPT_KEY_VALID,
 	     DM_CRYPT_WRITE_INLINE };
 
 enum cipher_flags {
-	CRYPT_MODE_INTEGRITY_AEAD,	/* Use authenticated mode for cihper */
+	CRYPT_MODE_INTEGRITY_AEAD,	/* Use authenticated mode for cipher */
 	CRYPT_IV_LARGE_SECTORS,		/* Calculate IV from sector_size, not 512B sectors */
 	CRYPT_ENCRYPT_PREPROCESS,	/* Must preprocess data for encryption (elephant) */
 };
@@ -2436,7 +2437,6 @@ static int set_key_user(struct crypt_config *cc, struct key *key)
 	return 0;
 }
 
-#if defined(CONFIG_ENCRYPTED_KEYS) || defined(CONFIG_ENCRYPTED_KEYS_MODULE)
 static int set_key_encrypted(struct crypt_config *cc, struct key *key)
 {
 	const struct encrypted_key_payload *ekp;
@@ -2452,7 +2452,22 @@ static int set_key_encrypted(struct crypt_config *cc, struct key *key)
 
 	return 0;
 }
-#endif /* CONFIG_ENCRYPTED_KEYS */
+
+static int set_key_trusted(struct crypt_config *cc, struct key *key)
+{
+	const struct trusted_key_payload *tkp;
+
+	tkp = key->payload.data[0];
+	if (!tkp)
+		return -EKEYREVOKED;
+
+	if (cc->key_size != tkp->key_len)
+		return -EINVAL;
+
+	memcpy(cc->key, tkp->key, cc->key_size);
+
+	return 0;
+}
 
 static int crypt_set_keyring_key(struct crypt_config *cc, const char *key_string)
 {
@@ -2482,11 +2497,14 @@ static int crypt_set_keyring_key(struct crypt_config *cc, const char *key_string
 	} else if (!strncmp(key_string, "user:", key_desc - key_string + 1)) {
 		type = &key_type_user;
 		set_key = set_key_user;
-#if defined(CONFIG_ENCRYPTED_KEYS) || defined(CONFIG_ENCRYPTED_KEYS_MODULE)
-	} else if (!strncmp(key_string, "encrypted:", key_desc - key_string + 1)) {
+	} else if (IS_ENABLED(CONFIG_ENCRYPTED_KEYS) &&
+		   !strncmp(key_string, "encrypted:", key_desc - key_string + 1)) {
 		type = &key_type_encrypted;
 		set_key = set_key_encrypted;
-#endif
+	} else if (IS_ENABLED(CONFIG_TRUSTED_KEYS) &&
+	           !strncmp(key_string, "trusted:", key_desc - key_string + 1)) {
+		type = &key_type_trusted;
+		set_key = set_key_trusted;
 	} else {
 		return -EINVAL;
 	}
@@ -3116,7 +3134,6 @@ static int crypt_ctr_optional(struct dm_target *ti, unsigned int argc, char **ar
 }
 
 #ifdef CONFIG_BLK_DEV_ZONED
-
 static int crypt_report_zones(struct dm_target *ti,
 		struct dm_report_zones_args *args, unsigned int nr_zones)
 {
@@ -3127,7 +3144,8 @@ static int crypt_report_zones(struct dm_target *ti,
 	return blkdev_report_zones(cc->dev->bdev, sector, nr_zones,
 				   dm_report_zones_cb, args);
 }
-
+#else
+#define crypt_report_zones NULL
 #endif
 
 /*
@@ -3324,6 +3342,7 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	wake_up_process(cc->write_thread);
 
 	ti->num_flush_bios = 1;
+	ti->limit_swap_bios = true;
 
 	return 0;
 
@@ -3558,14 +3577,12 @@ static void crypt_io_hints(struct dm_target *ti, struct queue_limits *limits)
 
 static struct target_type crypt_target = {
 	.name   = "crypt",
-	.version = {1, 22, 0},
+	.version = {1, 23, 0},
 	.module = THIS_MODULE,
 	.ctr    = crypt_ctr,
 	.dtr    = crypt_dtr,
-#ifdef CONFIG_BLK_DEV_ZONED
 	.features = DM_TARGET_ZONED_HM,
 	.report_zones = crypt_report_zones,
-#endif
 	.map    = crypt_map,
 	.status = crypt_status,
 	.postsuspend = crypt_postsuspend,

drivers/md/dm-dust.c

@@ -130,7 +130,7 @@ static int dust_add_block(struct dust_device *dd, unsigned long long block,
 
 	dd->badblock_count++;
 	if (!dd->quiet_mode) {
-		DMINFO("%s: badblock added at block %llu with write fail count %hhu",
+		DMINFO("%s: badblock added at block %llu with write fail count %u",
 		       __func__, block, wr_fail_cnt);
 	}
 	spin_unlock_irqrestore(&dd->dust_lock, flags);

drivers/md/dm-era-target.c

@@ -47,6 +47,7 @@ struct writeset {
 static void writeset_free(struct writeset *ws)
 {
 	vfree(ws->bits);
+	ws->bits = NULL;
 }
 
 static int setup_on_disk_bitset(struct dm_disk_bitset *info,
@@ -71,8 +72,6 @@ static size_t bitset_size(unsigned nr_bits)
  */
 static int writeset_alloc(struct writeset *ws, dm_block_t nr_blocks)
 {
-	ws->md.nr_bits = nr_blocks;
-	ws->md.root = INVALID_WRITESET_ROOT;
 	ws->bits = vzalloc(bitset_size(nr_blocks));
 	if (!ws->bits) {
 		DMERR("%s: couldn't allocate in memory bitset", __func__);
@@ -85,12 +84,14 @@ static int writeset_alloc(struct writeset *ws, dm_block_t nr_blocks)
 /*
  * Wipes the in-core bitset, and creates a new on disk bitset.
  */
-static int writeset_init(struct dm_disk_bitset *info, struct writeset *ws)
+static int writeset_init(struct dm_disk_bitset *info, struct writeset *ws,
+			 dm_block_t nr_blocks)
 {
 	int r;
 
-	memset(ws->bits, 0, bitset_size(ws->md.nr_bits));
+	memset(ws->bits, 0, bitset_size(nr_blocks));
 
+	ws->md.nr_bits = nr_blocks;
 	r = setup_on_disk_bitset(info, ws->md.nr_bits, &ws->md.root);
 	if (r) {
 		DMERR("%s: setup_on_disk_bitset failed", __func__);
@@ -134,7 +135,7 @@ static int writeset_test_and_set(struct dm_disk_bitset *info,
 {
 	int r;
 
-	if (!test_and_set_bit(block, ws->bits)) {
+	if (!test_bit(block, ws->bits)) {
 		r = dm_bitset_set_bit(info, ws->md.root, block, &ws->md.root);
 		if (r) {
 			/* FIXME: fail mode */
@@ -388,7 +389,7 @@ static void ws_dec(void *context, const void *value)
 
 static int ws_eq(void *context, const void *value1, const void *value2)
 {
-	return !memcmp(value1, value2, sizeof(struct writeset_metadata));
+	return !memcmp(value1, value2, sizeof(struct writeset_disk));
 }
 
 /*----------------------------------------------------------------*/
@@ -564,6 +565,15 @@ static int open_metadata(struct era_metadata *md)
 	}
 
 	disk = dm_block_data(sblock);
+
+	/* Verify the data block size hasn't changed */
+	if (le32_to_cpu(disk->data_block_size) != md->block_size) {
+		DMERR("changing the data block size (from %u to %llu) is not supported",
+		      le32_to_cpu(disk->data_block_size), md->block_size);
+		r = -EINVAL;
+		goto bad;
+	}
+
 	r = dm_tm_open_with_sm(md->bm, SUPERBLOCK_LOCATION,
 			       disk->metadata_space_map_root,
 			       sizeof(disk->metadata_space_map_root),
@@ -575,10 +585,10 @@ static int open_metadata(struct era_metadata *md)
 
 	setup_infos(md);
 
-	md->block_size = le32_to_cpu(disk->data_block_size);
 	md->nr_blocks = le32_to_cpu(disk->nr_blocks);
 	md->current_era = le32_to_cpu(disk->current_era);
 
+	ws_unpack(&disk->current_writeset, &md->current_writeset->md);
 	md->writeset_tree_root = le64_to_cpu(disk->writeset_tree_root);
 	md->era_array_root = le64_to_cpu(disk->era_array_root);
 	md->metadata_snap = le64_to_cpu(disk->metadata_snap);
@@ -746,6 +756,12 @@ static int metadata_digest_lookup_writeset(struct era_metadata *md,
 	ws_unpack(&disk, &d->writeset);
 	d->value = cpu_to_le32(key);
 
+	/*
+	 * We initialise another bitset info to avoid any caching side effects
+	 * with the previous one.
+	 */
+	dm_disk_bitset_init(md->tm, &d->info);
+
 	d->nr_bits = min(d->writeset.nr_bits, md->nr_blocks);
 	d->current_bit = 0;
 	d->step = metadata_digest_transcribe_writeset;
@@ -759,12 +775,6 @@ static int metadata_digest_start(struct era_metadata *md, struct digest *d)
 		return 0;
 
 	memset(d, 0, sizeof(*d));
-
-	/*
-	 * We initialise another bitset info to avoid any caching side
-	 * effects with the previous one.
-	 */
-	dm_disk_bitset_init(md->tm, &d->info);
 	d->step = metadata_digest_lookup_writeset;
 
 	return 0;
@@ -802,6 +812,8 @@ static struct era_metadata *metadata_open(struct block_device *bdev,
 
 static void metadata_close(struct era_metadata *md)
 {
+	writeset_free(&md->writesets[0]);
+	writeset_free(&md->writesets[1]);
 	destroy_persistent_data_objects(md);
 	kfree(md);
 }
@@ -839,6 +851,7 @@ static int metadata_resize(struct era_metadata *md, void *arg)
 	r = writeset_alloc(&md->writesets[1], *new_size);
 	if (r) {
 		DMERR("%s: writeset_alloc failed for writeset 1", __func__);
+		writeset_free(&md->writesets[0]);
 		return r;
 	}
 
@@ -849,6 +862,8 @@ static int metadata_resize(struct era_metadata *md, void *arg)
 			    &value, &md->era_array_root);
 	if (r) {
 		DMERR("%s: dm_array_resize failed", __func__);
+		writeset_free(&md->writesets[0]);
+		writeset_free(&md->writesets[1]);
 		return r;
 	}
 
@@ -870,7 +885,6 @@ static int metadata_era_archive(struct era_metadata *md)
 	}
 
 	ws_pack(&md->current_writeset->md, &value);
-	md->current_writeset->md.root = INVALID_WRITESET_ROOT;
 
 	keys[0] = md->current_era;
 	__dm_bless_for_disk(&value);
@@ -882,6 +896,7 @@ static int metadata_era_archive(struct era_metadata *md)
 		return r;
 	}
 
+	md->current_writeset->md.root = INVALID_WRITESET_ROOT;
 	md->archived_writesets = true;
 
 	return 0;
@@ -898,7 +913,7 @@ static int metadata_new_era(struct era_metadata *md)
 	int r;
 	struct writeset *new_writeset = next_writeset(md);
 
-	r = writeset_init(&md->bitset_info, new_writeset);
+	r = writeset_init(&md->bitset_info, new_writeset, md->nr_blocks);
 	if (r) {
 		DMERR("%s: writeset_init failed", __func__);
 		return r;
@@ -951,7 +966,7 @@ static int metadata_commit(struct era_metadata *md)
 	int r;
 	struct dm_block *sblock;
 
-	if (md->current_writeset->md.root != SUPERBLOCK_LOCATION) {
+	if (md->current_writeset->md.root != INVALID_WRITESET_ROOT) {
 		r = dm_bitset_flush(&md->bitset_info, md->current_writeset->md.root,
 				    &md->current_writeset->md.root);
 		if (r) {
@@ -1225,8 +1240,10 @@ static void process_deferred_bios(struct era *era)
 	int r;
 	struct bio_list deferred_bios, marked_bios;
 	struct bio *bio;
+	struct blk_plug plug;
 	bool commit_needed = false;
 	bool failed = false;
+	struct writeset *ws = era->md->current_writeset;
 
 	bio_list_init(&deferred_bios);
 	bio_list_init(&marked_bios);
@@ -1236,9 +1253,11 @@ static void process_deferred_bios(struct era *era)
 	bio_list_init(&era->deferred_bios);
 	spin_unlock(&era->deferred_lock);
 
+	if (bio_list_empty(&deferred_bios))
+		return;
+
 	while ((bio = bio_list_pop(&deferred_bios))) {
-		r = writeset_test_and_set(&era->md->bitset_info,
-					  era->md->current_writeset,
+		r = writeset_test_and_set(&era->md->bitset_info, ws,
 					  get_block(era, bio));
 		if (r < 0) {
 			/*
@@ -1246,7 +1265,6 @@ static void process_deferred_bios(struct era *era)
 			 * FIXME: finish.
 			 */
 			failed = true;
-
 		} else if (r == 0)
 			commit_needed = true;
 
@@ -1262,9 +1280,19 @@ static void process_deferred_bios(struct era *era)
 	if (failed)
 		while ((bio = bio_list_pop(&marked_bios)))
 			bio_io_error(bio);
-	else
-		while ((bio = bio_list_pop(&marked_bios)))
+	else {
+		blk_start_plug(&plug);
+		while ((bio = bio_list_pop(&marked_bios))) {
+			/*
+			 * Only update the in-core writeset if the on-disk one
+			 * was updated too.
+			 */
+			if (commit_needed)
+				set_bit(get_block(era, bio), ws->bits);
 			submit_bio_noacct(bio);
+		}
+		blk_finish_plug(&plug);
+	}
 }
 
 static void process_rpc_calls(struct era *era)
@@ -1473,15 +1501,6 @@ static int era_ctr(struct dm_target *ti, unsigned argc, char **argv)
 	}
 	era->md = md;
 
-	era->nr_blocks = calc_nr_blocks(era);
-
-	r = metadata_resize(era->md, &era->nr_blocks);
-	if (r) {
-		ti->error = "couldn't resize metadata";
-		era_destroy(era);
-		return -ENOMEM;
-	}
-
 	era->wq = alloc_ordered_workqueue("dm-" DM_MSG_PREFIX, WQ_MEM_RECLAIM);
 	if (!era->wq) {
 		ti->error = "could not create workqueue for metadata object";
@@ -1556,16 +1575,24 @@ static int era_preresume(struct dm_target *ti)
 	dm_block_t new_size = calc_nr_blocks(era);
 
 	if (era->nr_blocks != new_size) {
-		r = in_worker1(era, metadata_resize, &new_size);
-		if (r)
+		r = metadata_resize(era->md, &new_size);
+		if (r) {
+			DMERR("%s: metadata_resize failed", __func__);
+			return r;
+		}
+
+		r = metadata_commit(era->md);
+		if (r) {
+			DMERR("%s: metadata_commit failed", __func__);
 			return r;
+		}
 
 		era->nr_blocks = new_size;
 	}
 
 	start_worker(era);
 
-	r = in_worker0(era, metadata_new_era);
+	r = in_worker0(era, metadata_era_rollover);
 	if (r) {
 		DMERR("%s: metadata_era_rollover failed", __func__);
 		return r;

drivers/md/dm-flakey.c

@@ -469,6 +469,8 @@ static int flakey_report_zones(struct dm_target *ti,
 	return blkdev_report_zones(fc->dev->bdev, sector, nr_zones,
 				   dm_report_zones_cb, args);
 }
+#else
+#define flakey_report_zones NULL
 #endif
 
 static int flakey_iterate_devices(struct dm_target *ti, iterate_devices_callout_fn fn, void *data)
@@ -481,10 +483,8 @@ static int flakey_iterate_devices(struct dm_target *ti, iterate_devices_callout_
 static struct target_type flakey_target = {
 	.name   = "flakey",
 	.version = {1, 5, 0},
-#ifdef CONFIG_BLK_DEV_ZONED
-	.features = DM_TARGET_ZONED_HM,
+	.features = DM_TARGET_ZONED_HM | DM_TARGET_PASSES_CRYPTO,
 	.report_zones = flakey_report_zones,
-#endif
 	.module = THIS_MODULE,
 	.ctr    = flakey_ctr,
 	.dtr    = flakey_dtr,

drivers/md/dm-integrity.c

@@ -40,6 +40,7 @@
 #define BITMAP_BLOCK_SIZE		4096	/* don't change it */
 #define BITMAP_FLUSH_INTERVAL		(10 * HZ)
 #define DISCARD_FILLER			0xf6
+#define SALT_SIZE			16
 
 /*
  * Warning - DEBUG_PRINT prints security-sensitive data to the log,
@@ -57,6 +58,7 @@
 #define SB_VERSION_2			2
 #define SB_VERSION_3			3
 #define SB_VERSION_4			4
+#define SB_VERSION_5			5
 #define SB_SECTORS			8
 #define MAX_SECTORS_PER_BLOCK		8
 
@@ -72,12 +74,15 @@ struct superblock {
 	__u8 log2_blocks_per_bitmap_bit;
 	__u8 pad[2];
 	__u64 recalc_sector;
+	__u8 pad2[8];
+	__u8 salt[SALT_SIZE];
 };
 
 #define SB_FLAG_HAVE_JOURNAL_MAC	0x1
 #define SB_FLAG_RECALCULATING		0x2
 #define SB_FLAG_DIRTY_BITMAP		0x4
 #define SB_FLAG_FIXED_PADDING		0x8
+#define SB_FLAG_FIXED_HMAC		0x10
 
 #define	JOURNAL_ENTRY_ROUNDUP		8
 
@@ -259,6 +264,7 @@ struct dm_integrity_c {
 	bool recalculate_flag;
 	bool discard;
 	bool fix_padding;
+	bool fix_hmac;
 	bool legacy_recalculate;
 
 	struct alg_spec internal_hash_alg;
@@ -389,8 +395,11 @@ static int dm_integrity_failed(struct dm_integrity_c *ic)
 
 static bool dm_integrity_disable_recalculate(struct dm_integrity_c *ic)
 {
-	if ((ic->internal_hash_alg.key || ic->journal_mac_alg.key) &&
-	    !ic->legacy_recalculate)
+	if (ic->legacy_recalculate)
+		return false;
+	if (!(ic->sb->flags & cpu_to_le32(SB_FLAG_FIXED_HMAC)) ?
+	    ic->internal_hash_alg.key || ic->journal_mac_alg.key :
+	    ic->internal_hash_alg.key && !ic->journal_mac_alg.key)
 		return true;
 	return false;
 }
@@ -477,7 +486,9 @@ static void wraparound_section(struct dm_integrity_c *ic, unsigned *sec_ptr)
 
 static void sb_set_version(struct dm_integrity_c *ic)
 {
-	if (ic->sb->flags & cpu_to_le32(SB_FLAG_FIXED_PADDING))
+	if (ic->sb->flags & cpu_to_le32(SB_FLAG_FIXED_HMAC))
+		ic->sb->version = SB_VERSION_5;
+	else if (ic->sb->flags & cpu_to_le32(SB_FLAG_FIXED_PADDING))
 		ic->sb->version = SB_VERSION_4;
 	else if (ic->mode == 'B' || ic->sb->flags & cpu_to_le32(SB_FLAG_DIRTY_BITMAP))
 		ic->sb->version = SB_VERSION_3;
@@ -487,10 +498,58 @@ static void sb_set_version(struct dm_integrity_c *ic)
 		ic->sb->version = SB_VERSION_1;
 }
 
+static int sb_mac(struct dm_integrity_c *ic, bool wr)
+{
+	SHASH_DESC_ON_STACK(desc, ic->journal_mac);
+	int r;
+	unsigned size = crypto_shash_digestsize(ic->journal_mac);
+
+	if (sizeof(struct superblock) + 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);
+		if (unlikely(r < 0)) {
+			dm_integrity_io_error(ic, "crypto_shash_final", r);
+			return r;
+		}
+	} else {
+		__u8 result[HASH_MAX_DIGESTSIZE];
+		r = crypto_shash_final(desc, result);
+		if (unlikely(r < 0)) {
+			dm_integrity_io_error(ic, "crypto_shash_final", r);
+			return r;
+		}
+		if (memcmp((__u8 *)ic->sb + (1 << SECTOR_SHIFT) - size, result, size)) {
+			dm_integrity_io_error(ic, "superblock mac", -EILSEQ);
+			return -EILSEQ;
+		}
+	}
+
+	return 0;
+}
+
 static int sync_rw_sb(struct dm_integrity_c *ic, int op, int op_flags)
 {
 	struct dm_io_request io_req;
 	struct dm_io_region io_loc;
+	int r;
 
 	io_req.bi_op = op;
 	io_req.bi_op_flags = op_flags;
@@ -502,10 +561,28 @@ static int sync_rw_sb(struct dm_integrity_c *ic, int op, int op_flags)
 	io_loc.sector = ic->start;
 	io_loc.count = SB_SECTORS;
 
-	if (op == REQ_OP_WRITE)
+	if (op == REQ_OP_WRITE) {
 		sb_set_version(ic);
+		if (ic->journal_mac && ic->sb->flags & cpu_to_le32(SB_FLAG_FIXED_HMAC)) {
+			r = sb_mac(ic, true);
+			if (unlikely(r))
+				return r;
+		}
+	}
 
-	return dm_io(&io_req, 1, &io_loc, NULL);
+	r = dm_io(&io_req, 1, &io_loc, NULL);
+	if (unlikely(r))
+		return r;
+
+	if (op == REQ_OP_READ) {
+		if (ic->mode != 'R' && ic->journal_mac && ic->sb->flags & cpu_to_le32(SB_FLAG_FIXED_HMAC)) {
+			r = sb_mac(ic, false);
+			if (unlikely(r))
+				return r;
+		}
+	}
+
+	return 0;
 }
 
 #define BITMAP_OP_TEST_ALL_SET		0
@@ -722,15 +799,32 @@ static void section_mac(struct dm_integrity_c *ic, unsigned section, __u8 result
 	desc->tfm = ic->journal_mac;
 
 	r = crypto_shash_init(desc);
-	if (unlikely(r)) {
+	if (unlikely(r < 0)) {
 		dm_integrity_io_error(ic, "crypto_shash_init", r);
 		goto err;
 	}
 
+	if (ic->sb->flags & cpu_to_le32(SB_FLAG_FIXED_HMAC)) {
+		uint64_t section_le;
+
+		r = crypto_shash_update(desc, (__u8 *)&ic->sb->salt, SALT_SIZE);
+		if (unlikely(r < 0)) {
+			dm_integrity_io_error(ic, "crypto_shash_update", r);
+			goto err;
+		}
+
+		section_le = cpu_to_le64(section);
+		r = crypto_shash_update(desc, (__u8 *)&section_le, sizeof section_le);
+		if (unlikely(r < 0)) {
+			dm_integrity_io_error(ic, "crypto_shash_update", r);
+			goto err;
+		}
+	}
+
 	for (j = 0; j < ic->journal_section_entries; j++) {
 		struct journal_entry *je = access_journal_entry(ic, section, j);
 		r = crypto_shash_update(desc, (__u8 *)&je->u.sector, sizeof je->u.sector);
-		if (unlikely(r)) {
+		if (unlikely(r < 0)) {
 			dm_integrity_io_error(ic, "crypto_shash_update", r);
 			goto err;
 		}
@@ -740,7 +834,7 @@ static void section_mac(struct dm_integrity_c *ic, unsigned section, __u8 result
 
 	if (likely(size <= JOURNAL_MAC_SIZE)) {
 		r = crypto_shash_final(desc, result);
-		if (unlikely(r)) {
+		if (unlikely(r < 0)) {
 			dm_integrity_io_error(ic, "crypto_shash_final", r);
 			goto err;
 		}
@@ -753,7 +847,7 @@ static void section_mac(struct dm_integrity_c *ic, unsigned section, __u8 result
 			goto err;
 		}
 		r = crypto_shash_final(desc, digest);
-		if (unlikely(r)) {
+		if (unlikely(r < 0)) {
 			dm_integrity_io_error(ic, "crypto_shash_final", r);
 			goto err;
 		}
@@ -1399,7 +1493,7 @@ static void flush_notify(unsigned long error, void *fr_)
 {
 	struct flush_request *fr = fr_;
 	if (unlikely(error != 0))
-		dm_integrity_io_error(fr->ic, "flusing disk cache", -EIO);
+		dm_integrity_io_error(fr->ic, "flushing disk cache", -EIO);
 	complete(&fr->comp);
 }
 
@@ -1556,6 +1650,14 @@ static void integrity_sector_checksum(struct dm_integrity_c *ic, sector_t sector
 		goto failed;
 	}
 
+	if (ic->sb->flags & cpu_to_le32(SB_FLAG_FIXED_HMAC)) {
+		r = crypto_shash_update(req, (__u8 *)&ic->sb->salt, SALT_SIZE);
+		if (unlikely(r < 0)) {
+			dm_integrity_io_error(ic, "crypto_shash_update", r);
+			goto failed;
+		}
+	}
+
 	r = crypto_shash_update(req, (const __u8 *)&sector_le, sizeof sector_le);
 	if (unlikely(r < 0)) {
 		dm_integrity_io_error(ic, "crypto_shash_update", r);
@@ -3149,6 +3251,7 @@ static void dm_integrity_status(struct dm_target *ti, status_type_t type,
 		arg_count += !!ic->journal_crypt_alg.alg_string;
 		arg_count += !!ic->journal_mac_alg.alg_string;
 		arg_count += (ic->sb->flags & cpu_to_le32(SB_FLAG_FIXED_PADDING)) != 0;
+		arg_count += (ic->sb->flags & cpu_to_le32(SB_FLAG_FIXED_HMAC)) != 0;
 		arg_count += ic->legacy_recalculate;
 		DMEMIT("%s %llu %u %c %u", ic->dev->name, ic->start,
 		       ic->tag_size, ic->mode, arg_count);
@@ -3173,6 +3276,8 @@ static void dm_integrity_status(struct dm_target *ti, status_type_t type,
 		}
 		if ((ic->sb->flags & cpu_to_le32(SB_FLAG_FIXED_PADDING)) != 0)
 			DMEMIT(" fix_padding");
+		if ((ic->sb->flags & cpu_to_le32(SB_FLAG_FIXED_HMAC)) != 0)
+			DMEMIT(" fix_hmac");
 		if (ic->legacy_recalculate)
 			DMEMIT(" legacy_recalculate");
 
@@ -3310,6 +3415,11 @@ static int initialize_superblock(struct dm_integrity_c *ic, unsigned journal_sec
 	if (!journal_sections)
 		journal_sections = 1;
 
+	if (ic->fix_hmac && (ic->internal_hash_alg.alg_string || ic->journal_mac_alg.alg_string)) {
+		ic->sb->flags |= cpu_to_le32(SB_FLAG_FIXED_HMAC);
+		get_random_bytes(ic->sb->salt, SALT_SIZE);
+	}
+
 	if (!ic->meta_dev) {
 		if (ic->fix_padding)
 			ic->sb->flags |= cpu_to_le32(SB_FLAG_FIXED_PADDING);
@@ -3804,7 +3914,7 @@ static int dm_integrity_ctr(struct dm_target *ti, unsigned argc, char **argv)
 	unsigned extra_args;
 	struct dm_arg_set as;
 	static const struct dm_arg _args[] = {
-		{0, 16, "Invalid number of feature args"},
+		{0, 17, "Invalid number of feature args"},
 	};
 	unsigned journal_sectors, interleave_sectors, buffer_sectors, journal_watermark, sync_msec;
 	bool should_write_sb;
@@ -3942,7 +4052,7 @@ static int dm_integrity_ctr(struct dm_target *ti, unsigned argc, char **argv)
 			if (r)
 				goto bad;
 		} else if (!strncmp(opt_string, "journal_mac:", strlen("journal_mac:"))) {
-			r = get_alg_and_key(opt_string, &ic->journal_mac_alg,  &ti->error,
+			r = get_alg_and_key(opt_string, &ic->journal_mac_alg, &ti->error,
 					    "Invalid journal_mac argument");
 			if (r)
 				goto bad;
@@ -3952,6 +4062,8 @@ static int dm_integrity_ctr(struct dm_target *ti, unsigned argc, char **argv)
 			ic->discard = true;
 		} else if (!strcmp(opt_string, "fix_padding")) {
 			ic->fix_padding = true;
+		} else if (!strcmp(opt_string, "fix_hmac")) {
+			ic->fix_hmac = true;
 		} else if (!strcmp(opt_string, "legacy_recalculate")) {
 			ic->legacy_recalculate = true;
 		} else {
@@ -4110,7 +4222,7 @@ static int dm_integrity_ctr(struct dm_target *ti, unsigned argc, char **argv)
 			should_write_sb = true;
 	}
 
-	if (!ic->sb->version || ic->sb->version > SB_VERSION_4) {
+	if (!ic->sb->version || ic->sb->version > SB_VERSION_5) {
 		r = -EINVAL;
 		ti->error = "Unknown version";
 		goto bad;
@@ -4442,7 +4554,7 @@ static void dm_integrity_dtr(struct dm_target *ti)
 
 static struct target_type integrity_target = {
 	.name			= "integrity",
-	.version		= {1, 6, 0},
+	.version		= {1, 7, 0},
 	.module			= THIS_MODULE,
 	.features		= DM_TARGET_SINGLETON | DM_TARGET_INTEGRITY,
 	.ctr			= dm_integrity_ctr,

drivers/md/dm-linear.c

@@ -146,6 +146,8 @@ static int linear_report_zones(struct dm_target *ti,
 	return blkdev_report_zones(lc->dev->bdev, sector, nr_zones,
 				   dm_report_zones_cb, args);
 }
+#else
+#define linear_report_zones NULL
 #endif
 
 static int linear_iterate_devices(struct dm_target *ti,
@@ -227,13 +229,9 @@ static int linear_dax_zero_page_range(struct dm_target *ti, pgoff_t pgoff,
 static struct target_type linear_target = {
 	.name   = "linear",
 	.version = {1, 4, 0},
-#ifdef CONFIG_BLK_DEV_ZONED
 	.features = DM_TARGET_PASSES_INTEGRITY | DM_TARGET_NOWAIT |
-		    DM_TARGET_ZONED_HM,
+		    DM_TARGET_ZONED_HM | DM_TARGET_PASSES_CRYPTO,
 	.report_zones = linear_report_zones,
-#else
-	.features = DM_TARGET_PASSES_INTEGRITY | DM_TARGET_NOWAIT,
-#endif
 	.module = THIS_MODULE,
 	.ctr    = linear_ctr,
 	.dtr    = linear_dtr,

drivers/md/dm-writecache.c

@@ -148,6 +148,7 @@ struct dm_writecache {
 	size_t metadata_sectors;
 	size_t n_blocks;
 	uint64_t seq_count;
+	sector_t data_device_sectors;
 	void *block_start;
 	struct wc_entry *entries;
 	unsigned block_size;
@@ -159,14 +160,22 @@ struct dm_writecache {
 	bool overwrote_committed:1;
 	bool memory_vmapped:1;
 
+	bool start_sector_set:1;
 	bool high_wm_percent_set:1;
 	bool low_wm_percent_set:1;
 	bool max_writeback_jobs_set:1;
 	bool autocommit_blocks_set:1;
 	bool autocommit_time_set:1;
+	bool max_age_set:1;
 	bool writeback_fua_set:1;
 	bool flush_on_suspend:1;
 	bool cleaner:1;
+	bool cleaner_set:1;
+
+	unsigned high_wm_percent_value;
+	unsigned low_wm_percent_value;
+	unsigned autocommit_time_value;
+	unsigned max_age_value;
 
 	unsigned writeback_all;
 	struct workqueue_struct *writeback_wq;
@@ -523,7 +532,7 @@ static void ssd_commit_superblock(struct dm_writecache *wc)
 
 	region.bdev = wc->ssd_dev->bdev;
 	region.sector = 0;
-	region.count = PAGE_SIZE;
+	region.count = PAGE_SIZE >> SECTOR_SHIFT;
 
 	if (unlikely(region.sector + region.count > wc->metadata_sectors))
 		region.count = wc->metadata_sectors - region.sector;
@@ -969,6 +978,8 @@ static void writecache_resume(struct dm_target *ti)
 
 	wc_lock(wc);
 
+	wc->data_device_sectors = bdev_nr_sectors(wc->dev->bdev);
+
 	if (WC_MODE_PMEM(wc)) {
 		persistent_memory_invalidate_cache(wc->memory_map, wc->memory_map_size);
 	} else {
@@ -1638,6 +1649,10 @@ static bool wc_add_block(struct writeback_struct *wb, struct wc_entry *e, gfp_t
 	void *address = memory_data(wc, e);
 
 	persistent_memory_flush_cache(address, block_size);
+
+	if (unlikely(bio_end_sector(&wb->bio) >= wc->data_device_sectors))
+		return true;
+
 	return bio_add_page(&wb->bio, persistent_memory_page(address),
 			    block_size, persistent_memory_page_offset(address)) != 0;
 }
@@ -1709,6 +1724,9 @@ static void __writecache_writeback_pmem(struct dm_writecache *wc, struct writeba
 		if (writecache_has_error(wc)) {
 			bio->bi_status = BLK_STS_IOERR;
 			bio_endio(bio);
+		} else if (unlikely(!bio_sectors(bio))) {
+			bio->bi_status = BLK_STS_OK;
+			bio_endio(bio);
 		} else {
 			submit_bio(bio);
 		}
@@ -1752,6 +1770,14 @@ static void __writecache_writeback_ssd(struct dm_writecache *wc, struct writebac
 			e = f;
 		}
 
+		if (unlikely(to.sector + to.count > wc->data_device_sectors)) {
+			if (to.sector >= wc->data_device_sectors) {
+				writecache_copy_endio(0, 0, c);
+				continue;
+			}
+			from.count = to.count = wc->data_device_sectors - to.sector;
+		}
+
 		dm_kcopyd_copy(wc->dm_kcopyd, &from, 1, &to, 0, writecache_copy_endio, c);
 
 		__writeback_throttle(wc, wbl);
@@ -2004,8 +2030,7 @@ static void writecache_dtr(struct dm_target *ti)
 	if (wc->ssd_dev)
 		dm_put_device(ti, wc->ssd_dev);
 
-	if (wc->entries)
-		vfree(wc->entries);
+	vfree(wc->entries);
 
 	if (wc->memory_map) {
 		if (WC_MODE_PMEM(wc))
@@ -2020,8 +2045,7 @@ static void writecache_dtr(struct dm_target *ti)
 	if (wc->dm_io)
 		dm_io_client_destroy(wc->dm_io);
 
-	if (wc->dirty_bitmap)
-		vfree(wc->dirty_bitmap);
+	vfree(wc->dirty_bitmap);
 
 	kfree(wc);
 }
@@ -2205,6 +2229,7 @@ static int writecache_ctr(struct dm_target *ti, unsigned argc, char **argv)
 			if (sscanf(string, "%llu%c", &start_sector, &dummy) != 1)
 				goto invalid_optional;
 			wc->start_sector = start_sector;
+			wc->start_sector_set = true;
 			if (wc->start_sector != start_sector ||
 			    wc->start_sector >= wc->memory_map_size >> SECTOR_SHIFT)
 				goto invalid_optional;
@@ -2214,6 +2239,7 @@ static int writecache_ctr(struct dm_target *ti, unsigned argc, char **argv)
 				goto invalid_optional;
 			if (high_wm_percent < 0 || high_wm_percent > 100)
 				goto invalid_optional;
+			wc->high_wm_percent_value = high_wm_percent;
 			wc->high_wm_percent_set = true;
 		} else if (!strcasecmp(string, "low_watermark") && opt_params >= 1) {
 			string = dm_shift_arg(&as), opt_params--;
@@ -2221,6 +2247,7 @@ static int writecache_ctr(struct dm_target *ti, unsigned argc, char **argv)
 				goto invalid_optional;
 			if (low_wm_percent < 0 || low_wm_percent > 100)
 				goto invalid_optional;
+			wc->low_wm_percent_value = low_wm_percent;
 			wc->low_wm_percent_set = true;
 		} else if (!strcasecmp(string, "writeback_jobs") && opt_params >= 1) {
 			string = dm_shift_arg(&as), opt_params--;
@@ -2240,6 +2267,7 @@ static int writecache_ctr(struct dm_target *ti, unsigned argc, char **argv)
 			if (autocommit_msecs > 3600000)
 				goto invalid_optional;
 			wc->autocommit_jiffies = msecs_to_jiffies(autocommit_msecs);
+			wc->autocommit_time_value = autocommit_msecs;
 			wc->autocommit_time_set = true;
 		} else if (!strcasecmp(string, "max_age") && opt_params >= 1) {
 			unsigned max_age_msecs;
@@ -2249,7 +2277,10 @@ static int writecache_ctr(struct dm_target *ti, unsigned argc, char **argv)
 			if (max_age_msecs > 86400000)
 				goto invalid_optional;
 			wc->max_age = msecs_to_jiffies(max_age_msecs);
+			wc->max_age_set = true;
+			wc->max_age_value = max_age_msecs;
 		} else if (!strcasecmp(string, "cleaner")) {
+			wc->cleaner_set = true;
 			wc->cleaner = true;
 		} else if (!strcasecmp(string, "fua")) {
 			if (WC_MODE_PMEM(wc)) {
@@ -2455,7 +2486,6 @@ static void writecache_status(struct dm_target *ti, status_type_t type,
 	struct dm_writecache *wc = ti->private;
 	unsigned extra_args;
 	unsigned sz = 0;
-	uint64_t x;
 
 	switch (type) {
 	case STATUSTYPE_INFO:
@@ -2467,11 +2497,11 @@ static void writecache_status(struct dm_target *ti, status_type_t type,
 		DMEMIT("%c %s %s %u ", WC_MODE_PMEM(wc) ? 'p' : 's',
 				wc->dev->name, wc->ssd_dev->name, wc->block_size);
 		extra_args = 0;
-		if (wc->start_sector)
+		if (wc->start_sector_set)
 			extra_args += 2;
-		if (wc->high_wm_percent_set && !wc->cleaner)
+		if (wc->high_wm_percent_set)
 			extra_args += 2;
-		if (wc->low_wm_percent_set && !wc->cleaner)
+		if (wc->low_wm_percent_set)
 			extra_args += 2;
 		if (wc->max_writeback_jobs_set)
 			extra_args += 2;
@@ -2479,37 +2509,29 @@ static void writecache_status(struct dm_target *ti, status_type_t type,
 			extra_args += 2;
 		if (wc->autocommit_time_set)
 			extra_args += 2;
-		if (wc->max_age != MAX_AGE_UNSPECIFIED)
+		if (wc->max_age_set)
 			extra_args += 2;
-		if (wc->cleaner)
+		if (wc->cleaner_set)
 			extra_args++;
 		if (wc->writeback_fua_set)
 			extra_args++;
 
 		DMEMIT("%u", extra_args);
-		if (wc->start_sector)
+		if (wc->start_sector_set)
 			DMEMIT(" start_sector %llu", (unsigned long long)wc->start_sector);
-		if (wc->high_wm_percent_set && !wc->cleaner) {
-			x = (uint64_t)wc->freelist_high_watermark * 100;
-			x += wc->n_blocks / 2;
-			do_div(x, (size_t)wc->n_blocks);
-			DMEMIT(" high_watermark %u", 100 - (unsigned)x);
-		}
-		if (wc->low_wm_percent_set && !wc->cleaner) {
-			x = (uint64_t)wc->freelist_low_watermark * 100;
-			x += wc->n_blocks / 2;
-			do_div(x, (size_t)wc->n_blocks);
-			DMEMIT(" low_watermark %u", 100 - (unsigned)x);
-		}
+		if (wc->high_wm_percent_set)
+			DMEMIT(" high_watermark %u", wc->high_wm_percent_value);
+		if (wc->low_wm_percent_set)
+			DMEMIT(" low_watermark %u", wc->low_wm_percent_value);
 		if (wc->max_writeback_jobs_set)
 			DMEMIT(" writeback_jobs %u", wc->max_writeback_jobs);
 		if (wc->autocommit_blocks_set)
 			DMEMIT(" autocommit_blocks %u", wc->autocommit_blocks);
 		if (wc->autocommit_time_set)
-			DMEMIT(" autocommit_time %u", jiffies_to_msecs(wc->autocommit_jiffies));
-		if (wc->max_age != MAX_AGE_UNSPECIFIED)
-			DMEMIT(" max_age %u", jiffies_to_msecs(wc->max_age));
-		if (wc->cleaner)
+			DMEMIT(" autocommit_time %u", wc->autocommit_time_value);
+		if (wc->max_age_set)
+			DMEMIT(" max_age %u", wc->max_age_value);
+		if (wc->cleaner_set)
 			DMEMIT(" cleaner");
 		if (wc->writeback_fua_set)
 			DMEMIT(" %sfua", wc->writeback_fua ? "" : "no");
@@ -2519,7 +2541,7 @@ static void writecache_status(struct dm_target *ti, status_type_t type,
 
 static struct target_type writecache_target = {
 	.name			= "writecache",
-	.version		= {1, 3, 0},
+	.version		= {1, 4, 0},
 	.module			= THIS_MODULE,
 	.ctr			= writecache_ctr,
 	.dtr			= writecache_dtr,

drivers/md/dm.c

@@ -28,6 +28,7 @@
 #include <linux/refcount.h>
 #include <linux/part_stat.h>
 #include <linux/blk-crypto.h>
+#include <linux/keyslot-manager.h>
 
 #define DM_MSG_PREFIX "core"
 
@@ -105,12 +106,16 @@ struct dm_io {
 	struct dm_target_io tio;
 };
 
+#define DM_TARGET_IO_BIO_OFFSET (offsetof(struct dm_target_io, clone))
+#define DM_IO_BIO_OFFSET \
+	(offsetof(struct dm_target_io, clone) + offsetof(struct dm_io, tio))
+
 void *dm_per_bio_data(struct bio *bio, size_t data_size)
 {
 	struct dm_target_io *tio = container_of(bio, struct dm_target_io, clone);
 	if (!tio->inside_dm_io)
-		return (char *)bio - offsetof(struct dm_target_io, clone) - data_size;
-	return (char *)bio - offsetof(struct dm_target_io, clone) - offsetof(struct dm_io, tio) - data_size;
+		return (char *)bio - DM_TARGET_IO_BIO_OFFSET - data_size;
+	return (char *)bio - DM_IO_BIO_OFFSET - data_size;
 }
 EXPORT_SYMBOL_GPL(dm_per_bio_data);
 
@@ -118,9 +123,9 @@ struct bio *dm_bio_from_per_bio_data(void *data, size_t data_size)
 {
 	struct dm_io *io = (struct dm_io *)((char *)data + data_size);
 	if (io->magic == DM_IO_MAGIC)
-		return (struct bio *)((char *)io + offsetof(struct dm_io, tio) + offsetof(struct dm_target_io, clone));
+		return (struct bio *)((char *)io + DM_IO_BIO_OFFSET);
 	BUG_ON(io->magic != DM_TIO_MAGIC);
-	return (struct bio *)((char *)io + offsetof(struct dm_target_io, clone));
+	return (struct bio *)((char *)io + DM_TARGET_IO_BIO_OFFSET);
 }
 EXPORT_SYMBOL_GPL(dm_bio_from_per_bio_data);
 
@@ -148,6 +153,16 @@ EXPORT_SYMBOL_GPL(dm_bio_get_target_bio_nr);
 #define DM_NUMA_NODE NUMA_NO_NODE
 static int dm_numa_node = DM_NUMA_NODE;
 
+#define DEFAULT_SWAP_BIOS	(8 * 1048576 / PAGE_SIZE)
+static int swap_bios = DEFAULT_SWAP_BIOS;
+static int get_swap_bios(void)
+{
+	int latch = READ_ONCE(swap_bios);
+	if (unlikely(latch <= 0))
+		latch = DEFAULT_SWAP_BIOS;
+	return latch;
+}
+
 /*
  * For mempools pre-allocation at the table loading time.
  */
@@ -969,6 +984,11 @@ void disable_write_zeroes(struct mapped_device *md)
 	limits->max_write_zeroes_sectors = 0;
 }
 
+static bool swap_bios_limit(struct dm_target *ti, struct bio *bio)
+{
+	return unlikely((bio->bi_opf & REQ_SWAP) != 0) && unlikely(ti->limit_swap_bios);
+}
+
 static void clone_endio(struct bio *bio)
 {
 	blk_status_t error = bio->bi_status;
@@ -1020,6 +1040,11 @@ static void clone_endio(struct bio *bio)
 		}
 	}
 
+	if (unlikely(swap_bios_limit(tio->ti, bio))) {
+		struct mapped_device *md = io->md;
+		up(&md->swap_bios_semaphore);
+	}
+
 	free_tio(tio);
 	dec_pending(io, error);
 }
@@ -1129,7 +1154,7 @@ static bool dm_dax_supported(struct dax_device *dax_dev, struct block_device *bd
 	if (!map)
 		goto out;
 
-	ret = dm_table_supports_dax(map, device_supports_dax, &blocksize);
+	ret = dm_table_supports_dax(map, device_not_dax_capable, &blocksize);
 
 out:
 	dm_put_live_table(md, srcu_idx);
@@ -1253,6 +1278,22 @@ void dm_accept_partial_bio(struct bio *bio, unsigned n_sectors)
 }
 EXPORT_SYMBOL_GPL(dm_accept_partial_bio);
 
+static noinline void __set_swap_bios_limit(struct mapped_device *md, int latch)
+{
+	mutex_lock(&md->swap_bios_lock);
+	while (latch < md->swap_bios) {
+		cond_resched();
+		down(&md->swap_bios_semaphore);
+		md->swap_bios--;
+	}
+	while (latch > md->swap_bios) {
+		cond_resched();
+		up(&md->swap_bios_semaphore);
+		md->swap_bios++;
+	}
+	mutex_unlock(&md->swap_bios_lock);
+}
+
 static blk_qc_t __map_bio(struct dm_target_io *tio)
 {
 	int r;
@@ -1272,6 +1313,14 @@ static blk_qc_t __map_bio(struct dm_target_io *tio)
 	atomic_inc(&io->io_count);
 	sector = clone->bi_iter.bi_sector;
 
+	if (unlikely(swap_bios_limit(ti, clone))) {
+		struct mapped_device *md = io->md;
+		int latch = get_swap_bios();
+		if (unlikely(latch != md->swap_bios))
+			__set_swap_bios_limit(md, latch);
+		down(&md->swap_bios_semaphore);
+	}
+
 	r = ti->type->map(ti, clone);
 	switch (r) {
 	case DM_MAPIO_SUBMITTED:
@@ -1282,10 +1331,18 @@ static blk_qc_t __map_bio(struct dm_target_io *tio)
 		ret = submit_bio_noacct(clone);
 		break;
 	case DM_MAPIO_KILL:
+		if (unlikely(swap_bios_limit(ti, clone))) {
+			struct mapped_device *md = io->md;
+			up(&md->swap_bios_semaphore);
+		}
 		free_tio(tio);
 		dec_pending(io, BLK_STS_IOERR);
 		break;
 	case DM_MAPIO_REQUEUE:
+		if (unlikely(swap_bios_limit(ti, clone))) {
+			struct mapped_device *md = io->md;
+			up(&md->swap_bios_semaphore);
+		}
 		free_tio(tio);
 		dec_pending(io, BLK_STS_DM_REQUEUE);
 		break;
@@ -1718,6 +1775,19 @@ static const struct dax_operations dm_dax_ops;
 
 static void dm_wq_work(struct work_struct *work);
 
+#ifdef CONFIG_BLK_INLINE_ENCRYPTION
+static void dm_queue_destroy_keyslot_manager(struct request_queue *q)
+{
+	dm_destroy_keyslot_manager(q->ksm);
+}
+
+#else /* CONFIG_BLK_INLINE_ENCRYPTION */
+
+static inline void dm_queue_destroy_keyslot_manager(struct request_queue *q)
+{
+}
+#endif /* !CONFIG_BLK_INLINE_ENCRYPTION */
+
 static void cleanup_mapped_device(struct mapped_device *md)
 {
 	if (md->wq)
@@ -1739,14 +1809,17 @@ static void cleanup_mapped_device(struct mapped_device *md)
 		put_disk(md->disk);
 	}
 
-	if (md->queue)
+	if (md->queue) {
+		dm_queue_destroy_keyslot_manager(md->queue);
 		blk_cleanup_queue(md->queue);
+	}
 
 	cleanup_srcu_struct(&md->io_barrier);
 
 	mutex_destroy(&md->suspend_lock);
 	mutex_destroy(&md->type_lock);
 	mutex_destroy(&md->table_devices_lock);
+	mutex_destroy(&md->swap_bios_lock);
 
 	dm_mq_cleanup_mapped_device(md);
 }
@@ -1814,6 +1887,10 @@ static struct mapped_device *alloc_dev(int minor)
 	init_waitqueue_head(&md->eventq);
 	init_completion(&md->kobj_holder.completion);
 
+	md->swap_bios = get_swap_bios();
+	sema_init(&md->swap_bios_semaphore, md->swap_bios);
+	mutex_init(&md->swap_bios_lock);
+
 	md->disk->major = _major;
 	md->disk->first_minor = minor;
 	md->disk->fops = &dm_blk_dops;
@@ -2849,8 +2926,8 @@ struct dm_md_mempools *dm_alloc_md_mempools(struct mapped_device *md, enum dm_qu
 	case DM_TYPE_BIO_BASED:
 	case DM_TYPE_DAX_BIO_BASED:
 		pool_size = max(dm_get_reserved_bio_based_ios(), min_pool_size);
-		front_pad = roundup(per_io_data_size, __alignof__(struct dm_target_io)) + offsetof(struct dm_target_io, clone);
-		io_front_pad = roundup(front_pad,  __alignof__(struct dm_io)) + offsetof(struct dm_io, tio);
+		front_pad = roundup(per_io_data_size, __alignof__(struct dm_target_io)) + DM_TARGET_IO_BIO_OFFSET;
+		io_front_pad = roundup(per_io_data_size,  __alignof__(struct dm_io)) + DM_IO_BIO_OFFSET;
 		ret = bioset_init(&pools->io_bs, pool_size, io_front_pad, 0);
 		if (ret)
 			goto out;
@@ -3097,6 +3174,9 @@ MODULE_PARM_DESC(reserved_bio_based_ios, "Reserved IOs in bio-based mempools");
 module_param(dm_numa_node, int, S_IRUGO | S_IWUSR);
 MODULE_PARM_DESC(dm_numa_node, "NUMA node for DM device memory allocations");
 
+module_param(swap_bios, int, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(swap_bios, "Maximum allowed inflight swap IOs");
+
 MODULE_DESCRIPTION(DM_NAME " driver");
 MODULE_AUTHOR("Joe Thornber <dm-devel@redhat.com>");
 MODULE_LICENSE("GPL");

drivers/md/dm.h

@@ -73,7 +73,7 @@ void dm_table_free_md_mempools(struct dm_table *t);
 struct dm_md_mempools *dm_table_get_md_mempools(struct dm_table *t);
 bool dm_table_supports_dax(struct dm_table *t, iterate_devices_callout_fn fn,
 			   int *blocksize);
-int device_supports_dax(struct dm_target *ti, struct dm_dev *dev,
+int device_not_dax_capable(struct dm_target *ti, struct dm_dev *dev,
 			   sector_t start, sector_t len, void *data);
 
 void dm_lock_md_type(struct mapped_device *md);

drivers/md/persistent-data/dm-btree-internal.h

@@ -100,7 +100,7 @@ struct dm_block *shadow_parent(struct shadow_spine *s);
 
 int shadow_has_parent(struct shadow_spine *s);
 
-int shadow_root(struct shadow_spine *s);
+dm_block_t shadow_root(struct shadow_spine *s);
 
 /*
  * Some inlines.

drivers/md/persistent-data/dm-btree-spine.c

@@ -235,7 +235,7 @@ int shadow_has_parent(struct shadow_spine *s)
 	return s->count >= 2;
 }
 
-int shadow_root(struct shadow_spine *s)
+dm_block_t shadow_root(struct shadow_spine *s)
 {
 	return s->root;
 }

include/linux/device-mapper.h

@@ -93,9 +93,18 @@ typedef int (*dm_message_fn) (struct dm_target *ti, unsigned argc, char **argv,
 
 typedef int (*dm_prepare_ioctl_fn) (struct dm_target *ti, struct block_device **bdev);
 
+#ifdef CONFIG_BLK_DEV_ZONED
 typedef int (*dm_report_zones_fn) (struct dm_target *ti,
 				   struct dm_report_zones_args *args,
 				   unsigned int nr_zones);
+#else
+/*
+ * Define dm_report_zones_fn so that targets can assign to NULL if
+ * CONFIG_BLK_DEV_ZONED disabled. Otherwise each target needs to do
+ * awkward #ifdefs in their target_type, etc.
+ */
+typedef int (*dm_report_zones_fn) (struct dm_target *dummy);
+#endif
 
 /*
  * These iteration functions are typically used to check (and combine)
@@ -187,9 +196,7 @@ struct target_type {
 	dm_status_fn status;
 	dm_message_fn message;
 	dm_prepare_ioctl_fn prepare_ioctl;
-#ifdef CONFIG_BLK_DEV_ZONED
 	dm_report_zones_fn report_zones;
-#endif
 	dm_busy_fn busy;
 	dm_iterate_devices_fn iterate_devices;
 	dm_io_hints_fn io_hints;
@@ -248,8 +255,13 @@ struct target_type {
 /*
  * Indicates that a target supports host-managed zoned block devices.
  */
+#ifdef CONFIG_BLK_DEV_ZONED
 #define DM_TARGET_ZONED_HM		0x00000040
 #define dm_target_supports_zoned_hm(type) ((type)->features & DM_TARGET_ZONED_HM)
+#else
+#define DM_TARGET_ZONED_HM		0x00000000
+#define dm_target_supports_zoned_hm(type) (false)
+#endif
 
 /*
  * A target handles REQ_NOWAIT
@@ -257,6 +269,12 @@ struct target_type {
 #define DM_TARGET_NOWAIT		0x00000080
 #define dm_target_supports_nowait(type) ((type)->features & DM_TARGET_NOWAIT)
 
+/*
+ * A target supports passing through inline crypto support.
+ */
+#define DM_TARGET_PASSES_CRYPTO		0x00000100
+#define dm_target_passes_crypto(type) ((type)->features & DM_TARGET_PASSES_CRYPTO)
+
 struct dm_target {
 	struct dm_table *table;
 	struct target_type *type;
@@ -325,6 +343,11 @@ struct dm_target {
 	 * whether or not its underlying devices have support.
 	 */
 	bool discards_supported:1;
+
+	/*
+	 * Set if we need to limit the number of in-flight bios when swapping.
+	 */
+	bool limit_swap_bios:1;
 };
 
 void *dm_per_bio_data(struct bio *bio, size_t data_size);
@@ -533,6 +556,11 @@ void dm_table_run_md_queue_async(struct dm_table *t);
 struct dm_table *dm_swap_table(struct mapped_device *md,
 			       struct dm_table *t);
 
+/*
+ * Table keyslot manager functions
+ */
+void dm_destroy_keyslot_manager(struct blk_keyslot_manager *ksm);
+
 /*
  * A wrapper around vmalloc.
  */

include/linux/keyslot-manager.h

@@ -106,4 +106,15 @@ void blk_ksm_reprogram_all_keys(struct blk_keyslot_manager *ksm);
 
 void blk_ksm_destroy(struct blk_keyslot_manager *ksm);
 
+void blk_ksm_intersect_modes(struct blk_keyslot_manager *parent,
+			     const struct blk_keyslot_manager *child);
+
+void blk_ksm_init_passthrough(struct blk_keyslot_manager *ksm);
+
+bool blk_ksm_is_superset(struct blk_keyslot_manager *ksm_superset,
+			 struct blk_keyslot_manager *ksm_subset);
+
+void blk_ksm_update_capabilities(struct blk_keyslot_manager *target_ksm,
+				 struct blk_keyslot_manager *reference_ksm);
+
 #endif /* __LINUX_KEYSLOT_MANAGER_H */

include/uapi/linux/dm-ioctl.h

@@ -272,9 +272,9 @@ enum {
 #define DM_DEV_SET_GEOMETRY	_IOWR(DM_IOCTL, DM_DEV_SET_GEOMETRY_CMD, struct dm_ioctl)
 
 #define DM_VERSION_MAJOR	4
-#define DM_VERSION_MINOR	43
+#define DM_VERSION_MINOR	44
 #define DM_VERSION_PATCHLEVEL	0
-#define DM_VERSION_EXTRA	"-ioctl (2020-10-01)"
+#define DM_VERSION_EXTRA	"-ioctl (2021-02-01)"
 
 /* Status bits */
 #define DM_READONLY_FLAG	(1 << 0) /* In/Out */