Merge tag 'for-5.16/dm-changes' of...

Merge tag 'for-5.16/dm-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm

Pull device mapper updates from Mike Snitzer:

 - Add DM core support for emitting audit events through the audit
   subsystem. Also enhance both the integrity and crypt targets to emit
   events to via dm-audit.

 - Various other simple code improvements and cleanups.

* tag 'for-5.16/dm-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
  dm table: log table creation error code
  dm: make workqueue names device-specific
  dm writecache: Make use of the helper macro kthread_run()
  dm crypt: Make use of the helper macro kthread_run()
  dm verity: use bvec_kmap_local in verity_for_bv_block
  dm log writes: use memcpy_from_bvec in log_writes_map
  dm integrity: use bvec_kmap_local in __journal_read_write
  dm integrity: use bvec_kmap_local in integrity_metadata
  dm: add add_disk() error handling
  dm: Remove redundant flush_workqueue() calls
  dm crypt: log aead integrity violations to audit subsystem
  dm integrity: log audit events for dm-integrity target
  dm: introduce audit event module for device mapper

drivers/md/Kconfig

@@ -610,6 +610,7 @@ config DM_INTEGRITY
 	select CRYPTO
 	select CRYPTO_SKCIPHER
 	select ASYNC_XOR
+	select DM_AUDIT if AUDIT
 	help
 	  This device-mapper target emulates a block device that has
 	  additional per-sector tags that can be used for storing
@@ -642,4 +643,13 @@ config DM_ZONED
 
 	  If unsure, say N.
 
+config DM_AUDIT
+	bool "DM audit events"
+	depends on AUDIT
+	help
+	  Generate audit events for device-mapper.
+
+	  Enables audit logging of several security relevant events in the
+	  particular device-mapper targets, especially the integrity target.
+
 endif # MD

drivers/md/Makefile

@@ -107,3 +107,7 @@ endif
 ifeq ($(CONFIG_DM_VERITY_VERIFY_ROOTHASH_SIG),y)
 dm-verity-objs			+= dm-verity-verify-sig.o
 endif
+
+ifeq ($(CONFIG_DM_AUDIT),y)
+dm-mod-objs			+= dm-audit.o
+endif

drivers/md/dm-audit.c

+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Creating audit records for mapped devices.
+ *
+ * Copyright (C) 2021 Fraunhofer AISEC. All rights reserved.
+ *
+ * Authors: Michael Weiß <michael.weiss@aisec.fraunhofer.de>
+ */
+
+#include <linux/audit.h>
+#include <linux/module.h>
+#include <linux/device-mapper.h>
+#include <linux/bio.h>
+#include <linux/blkdev.h>
+
+#include "dm-audit.h"
+#include "dm-core.h"
+
+static struct audit_buffer *dm_audit_log_start(int audit_type,
+					       const char *dm_msg_prefix,
+					       const char *op)
+{
+	struct audit_buffer *ab;
+
+	if (audit_enabled == AUDIT_OFF)
+		return NULL;
+
+	ab = audit_log_start(audit_context(), GFP_KERNEL, audit_type);
+	if (unlikely(!ab))
+		return NULL;
+
+	audit_log_format(ab, "module=%s op=%s", dm_msg_prefix, op);
+	return ab;
+}
+
+void dm_audit_log_ti(int audit_type, const char *dm_msg_prefix, const char *op,
+		     struct dm_target *ti, int result)
+{
+	struct audit_buffer *ab = NULL;
+	struct mapped_device *md = dm_table_get_md(ti->table);
+	int dev_major = dm_disk(md)->major;
+	int dev_minor = dm_disk(md)->first_minor;
+
+	switch (audit_type) {
+	case AUDIT_DM_CTRL:
+		ab = dm_audit_log_start(audit_type, dm_msg_prefix, op);
+		if (unlikely(!ab))
+			return;
+		audit_log_task_info(ab);
+		audit_log_format(ab, " dev=%d:%d error_msg='%s'", dev_major,
+				 dev_minor, !result ? ti->error : "success");
+		break;
+	case AUDIT_DM_EVENT:
+		ab = dm_audit_log_start(audit_type, dm_msg_prefix, op);
+		if (unlikely(!ab))
+			return;
+		audit_log_format(ab, " dev=%d:%d sector=?", dev_major,
+				 dev_minor);
+		break;
+	default: /* unintended use */
+		return;
+	}
+
+	audit_log_format(ab, " res=%d", result);
+	audit_log_end(ab);
+}
+EXPORT_SYMBOL_GPL(dm_audit_log_ti);
+
+void dm_audit_log_bio(const char *dm_msg_prefix, const char *op,
+		      struct bio *bio, sector_t sector, int result)
+{
+	struct audit_buffer *ab;
+	int dev_major = MAJOR(bio->bi_bdev->bd_dev);
+	int dev_minor = MINOR(bio->bi_bdev->bd_dev);
+
+	ab = dm_audit_log_start(AUDIT_DM_EVENT, dm_msg_prefix, op);
+	if (unlikely(!ab))
+		return;
+
+	audit_log_format(ab, " dev=%d:%d sector=%llu res=%d",
+			 dev_major, dev_minor, sector, result);
+	audit_log_end(ab);
+}
+EXPORT_SYMBOL_GPL(dm_audit_log_bio);

drivers/md/dm-audit.h

+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Creating audit records for mapped devices.
+ *
+ * Copyright (C) 2021 Fraunhofer AISEC. All rights reserved.
+ *
+ * Authors: Michael Weiß <michael.weiss@aisec.fraunhofer.de>
+ */
+
+#ifndef DM_AUDIT_H
+#define DM_AUDIT_H
+
+#include <linux/device-mapper.h>
+#include <linux/audit.h>
+
+#ifdef CONFIG_DM_AUDIT
+void dm_audit_log_bio(const char *dm_msg_prefix, const char *op,
+		      struct bio *bio, sector_t sector, int result);
+
+/*
+ * dm_audit_log_ti() is not intended to be used directly in dm modules,
+ * the wrapper functions below should be called by dm modules instead.
+ */
+void dm_audit_log_ti(int audit_type, const char *dm_msg_prefix, const char *op,
+		     struct dm_target *ti, int result);
+
+static inline void dm_audit_log_ctr(const char *dm_msg_prefix,
+				    struct dm_target *ti, int result)
+{
+	dm_audit_log_ti(AUDIT_DM_CTRL, dm_msg_prefix, "ctr", ti, result);
+}
+
+static inline void dm_audit_log_dtr(const char *dm_msg_prefix,
+				    struct dm_target *ti, int result)
+{
+	dm_audit_log_ti(AUDIT_DM_CTRL, dm_msg_prefix, "dtr", ti, result);
+}
+
+static inline void dm_audit_log_target(const char *dm_msg_prefix, const char *op,
+				       struct dm_target *ti, int result)
+{
+	dm_audit_log_ti(AUDIT_DM_EVENT, dm_msg_prefix, op, ti, result);
+}
+#else
+static inline void dm_audit_log_bio(const char *dm_msg_prefix, const char *op,
+				    struct bio *bio, sector_t sector,
+				    int result)
+{
+}
+static inline void dm_audit_log_target(const char *dm_msg_prefix,
+				       const char *op, struct dm_target *ti,
+				       int result)
+{
+}
+static inline void dm_audit_log_ctr(const char *dm_msg_prefix,
+				    struct dm_target *ti, int result)
+{
+}
+
+static inline void dm_audit_log_dtr(const char *dm_msg_prefix,
+				    struct dm_target *ti, int result)
+{
+}
+#endif
+
+#endif

drivers/md/dm-bufio.c

@@ -2082,7 +2082,6 @@ static void __exit dm_bufio_exit(void)
 	int bug = 0;
 
 	cancel_delayed_work_sync(&dm_bufio_cleanup_old_work);
-	flush_workqueue(dm_bufio_wq);
 	destroy_workqueue(dm_bufio_wq);
 
 	if (dm_bufio_client_count) {

drivers/md/dm-crypt.c

@@ -42,6 +42,8 @@
 
 #include <linux/device-mapper.h>
 
+#include "dm-audit.h"
+
 #define DM_MSG_PREFIX "crypt"
 
 /*
@@ -1363,8 +1365,12 @@ static int crypt_convert_block_aead(struct crypt_config *cc,
 
 	if (r == -EBADMSG) {
 		char b[BDEVNAME_SIZE];
-		DMERR_LIMIT("%s: INTEGRITY AEAD ERROR, sector %llu", bio_devname(ctx->bio_in, b),
-			    (unsigned long long)le64_to_cpu(*sector));
+		sector_t s = le64_to_cpu(*sector);
+
+		DMERR_LIMIT("%s: INTEGRITY AEAD ERROR, sector %llu",
+			    bio_devname(ctx->bio_in, b), s);
+		dm_audit_log_bio(DM_MSG_PREFIX, "integrity-aead",
+				 ctx->bio_in, s, 0);
 	}
 
 	if (!r && cc->iv_gen_ops && cc->iv_gen_ops->post)
@@ -2174,8 +2180,12 @@ static void kcryptd_async_done(struct crypto_async_request *async_req,
 
 	if (error == -EBADMSG) {
 		char b[BDEVNAME_SIZE];
-		DMERR_LIMIT("%s: INTEGRITY AEAD ERROR, sector %llu", bio_devname(ctx->bio_in, b),
-			    (unsigned long long)le64_to_cpu(*org_sector_of_dmreq(cc, dmreq)));
+		sector_t s = le64_to_cpu(*org_sector_of_dmreq(cc, dmreq));
+
+		DMERR_LIMIT("%s: INTEGRITY AEAD ERROR, sector %llu",
+			    bio_devname(ctx->bio_in, b), s);
+		dm_audit_log_bio(DM_MSG_PREFIX, "integrity-aead",
+				 ctx->bio_in, s, 0);
 		io->error = BLK_STS_PROTECTION;
 	} else if (error < 0)
 		io->error = BLK_STS_IOERR;
@@ -2735,6 +2745,8 @@ static void crypt_dtr(struct dm_target *ti)
 	dm_crypt_clients_n--;
 	crypt_calculate_pages_per_client();
 	spin_unlock(&dm_crypt_clients_lock);
+
+	dm_audit_log_dtr(DM_MSG_PREFIX, ti, 1);
 }
 
 static int crypt_ctr_ivmode(struct dm_target *ti, const char *ivmode)
@@ -3351,21 +3363,22 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	spin_lock_init(&cc->write_thread_lock);
 	cc->write_tree = RB_ROOT;
 
-	cc->write_thread = kthread_create(dmcrypt_write, cc, "dmcrypt_write/%s", devname);
+	cc->write_thread = kthread_run(dmcrypt_write, cc, "dmcrypt_write/%s", devname);
 	if (IS_ERR(cc->write_thread)) {
 		ret = PTR_ERR(cc->write_thread);
 		cc->write_thread = NULL;
 		ti->error = "Couldn't spawn write thread";
 		goto bad;
 	}
-	wake_up_process(cc->write_thread);
 
 	ti->num_flush_bios = 1;
 	ti->limit_swap_bios = true;
 
+	dm_audit_log_ctr(DM_MSG_PREFIX, ti, 1);
 	return 0;
 
 bad:
+	dm_audit_log_ctr(DM_MSG_PREFIX, ti, 0);
 	crypt_dtr(ti);
 	return ret;
 }

drivers/md/dm-integrity.c

@@ -23,6 +23,8 @@
 #include <linux/async_tx.h>
 #include <linux/dm-bufio.h>
 
+#include "dm-audit.h"
+
 #define DM_MSG_PREFIX "integrity"
 
 #define DEFAULT_INTERLEAVE_SECTORS	32768
@@ -539,6 +541,7 @@ static int sb_mac(struct dm_integrity_c *ic, bool wr)
 		}
 		if (memcmp((__u8 *)ic->sb + (1 << SECTOR_SHIFT) - size, result, size)) {
 			dm_integrity_io_error(ic, "superblock mac", -EILSEQ);
+			dm_audit_log_target(DM_MSG_PREFIX, "mac-superblock", ic->ti, 0);
 			return -EILSEQ;
 		}
 	}
@@ -876,8 +879,10 @@ static void rw_section_mac(struct dm_integrity_c *ic, unsigned section, bool wr)
 		if (likely(wr))
 			memcpy(&js->mac, result + (j * JOURNAL_MAC_PER_SECTOR), JOURNAL_MAC_PER_SECTOR);
 		else {
-			if (memcmp(&js->mac, result + (j * JOURNAL_MAC_PER_SECTOR), JOURNAL_MAC_PER_SECTOR))
+			if (memcmp(&js->mac, result + (j * JOURNAL_MAC_PER_SECTOR), JOURNAL_MAC_PER_SECTOR)) {
 				dm_integrity_io_error(ic, "journal mac", -EILSEQ);
+				dm_audit_log_target(DM_MSG_PREFIX, "mac-journal", ic->ti, 0);
+			}
 		}
 	}
 }
@@ -1765,7 +1770,7 @@ static void integrity_metadata(struct work_struct *w)
 			char *mem, *checksums_ptr;
 
 again:
-			mem = (char *)kmap_atomic(bv.bv_page) + bv.bv_offset;
+			mem = bvec_kmap_local(&bv);
 			pos = 0;
 			checksums_ptr = checksums;
 			do {
@@ -1775,17 +1780,22 @@ static void integrity_metadata(struct work_struct *w)
 				pos += ic->sectors_per_block << SECTOR_SHIFT;
 				sector += ic->sectors_per_block;
 			} while (pos < bv.bv_len && sectors_to_process && checksums != checksums_onstack);
-			kunmap_atomic(mem);
+			kunmap_local(mem);
 
 			r = dm_integrity_rw_tag(ic, checksums, &dio->metadata_block, &dio->metadata_offset,
 						checksums_ptr - checksums, dio->op == REQ_OP_READ ? TAG_CMP : TAG_WRITE);
 			if (unlikely(r)) {
 				if (r > 0) {
 					char b[BDEVNAME_SIZE];
-					DMERR_LIMIT("%s: Checksum failed at sector 0x%llx", bio_devname(bio, b),
-						    (sector - ((r + ic->tag_size - 1) / ic->tag_size)));
+					sector_t s;
+
+					s = sector - ((r + ic->tag_size - 1) / ic->tag_size);
+					DMERR_LIMIT("%s: Checksum failed at sector 0x%llx",
+						    bio_devname(bio, b), s);
 					r = -EILSEQ;
 					atomic64_inc(&ic->number_of_mismatches);
+					dm_audit_log_bio(DM_MSG_PREFIX, "integrity-checksum",
+							 bio, s, 0);
 				}
 				if (likely(checksums != checksums_onstack))
 					kfree(checksums);
@@ -1953,7 +1963,7 @@ static bool __journal_read_write(struct dm_integrity_io *dio, struct bio *bio,
 		n_sectors -= bv.bv_len >> SECTOR_SHIFT;
 		bio_advance_iter(bio, &bio->bi_iter, bv.bv_len);
 retry_kmap:
-		mem = kmap_atomic(bv.bv_page);
+		mem = bvec_kmap_local(&bv);
 		if (likely(dio->op == REQ_OP_WRITE))
 			flush_dcache_page(bv.bv_page);
 
@@ -1967,7 +1977,7 @@ static bool __journal_read_write(struct dm_integrity_io *dio, struct bio *bio,
 
 				if (unlikely(journal_entry_is_inprogress(je))) {
 					flush_dcache_page(bv.bv_page);
-					kunmap_atomic(mem);
+					kunmap_local(mem);
 
 					__io_wait_event(ic->copy_to_journal_wait, !journal_entry_is_inprogress(je));
 					goto retry_kmap;
@@ -1991,6 +2001,8 @@ static bool __journal_read_write(struct dm_integrity_io *dio, struct bio *bio,
 					if (unlikely(memcmp(checksums_onstack, journal_entry_tag(ic, je), ic->tag_size))) {
 						DMERR_LIMIT("Checksum failed when reading from journal, at sector 0x%llx",
 							    logical_sector);
+						dm_audit_log_bio(DM_MSG_PREFIX, "journal-checksum",
+								 bio, logical_sector, 0);
 					}
 				}
 #endif
@@ -2058,7 +2070,7 @@ static bool __journal_read_write(struct dm_integrity_io *dio, struct bio *bio,
 
 		if (unlikely(dio->op == REQ_OP_READ))
 			flush_dcache_page(bv.bv_page);
-		kunmap_atomic(mem);
+		kunmap_local(mem);
 	} while (n_sectors);
 
 	if (likely(dio->op == REQ_OP_WRITE)) {
@@ -2534,8 +2546,10 @@ static void do_journal_write(struct dm_integrity_c *ic, unsigned write_start,
 
 					integrity_sector_checksum(ic, sec + ((l - j) << ic->sb->log2_sectors_per_block),
 								  (char *)access_journal_data(ic, i, l), test_tag);
-					if (unlikely(memcmp(test_tag, journal_entry_tag(ic, je2), ic->tag_size)))
+					if (unlikely(memcmp(test_tag, journal_entry_tag(ic, je2), ic->tag_size))) {
 						dm_integrity_io_error(ic, "tag mismatch when replaying journal", -EILSEQ);
+						dm_audit_log_target(DM_MSG_PREFIX, "integrity-replay-journal", ic->ti, 0);
+					}
 				}
 
 				journal_entry_set_unused(je2);
@@ -4514,9 +4528,11 @@ static int dm_integrity_ctr(struct dm_target *ti, unsigned argc, char **argv)
 	if (ic->discard)
 		ti->num_discard_bios = 1;
 
+	dm_audit_log_ctr(DM_MSG_PREFIX, ti, 1);
 	return 0;
 
 bad:
+	dm_audit_log_ctr(DM_MSG_PREFIX, ti, 0);
 	dm_integrity_dtr(ti);
 	return r;
 }
@@ -4590,6 +4606,7 @@ static void dm_integrity_dtr(struct dm_target *ti)
 	free_alg(&ic->journal_mac_alg);
 
 	kfree(ic);
+	dm_audit_log_dtr(DM_MSG_PREFIX, ti, 1);
 }
 
 static struct target_type integrity_target = {

drivers/md/dm-log-writes.c

@@ -753,7 +753,7 @@ static int log_writes_map(struct dm_target *ti, struct bio *bio)
 	 */
 	bio_for_each_segment(bv, bio, iter) {
 		struct page *page;
-		void *src, *dst;
+		void *dst;
 
 		page = alloc_page(GFP_NOIO);
 		if (!page) {
@@ -765,11 +765,9 @@ static int log_writes_map(struct dm_target *ti, struct bio *bio)
 			return DM_MAPIO_KILL;
 		}
 
-		src = kmap_atomic(bv.bv_page);
 		dst = kmap_atomic(page);
-		memcpy(dst, src + bv.bv_offset, bv.bv_len);
+		memcpy_from_bvec(dst, &bv);
 		kunmap_atomic(dst);
-		kunmap_atomic(src);
 		block->vecs[i].bv_page = page;
 		block->vecs[i].bv_len = bv.bv_len;
 		block->vec_cnt++;

drivers/md/dm-table.c

@@ -706,7 +706,7 @@ int dm_table_add_target(struct dm_table *t, const char *type,
 
 	r = dm_split_args(&argc, &argv, params);
 	if (r) {
-		tgt->error = "couldn't split parameters (insufficient memory)";
+		tgt->error = "couldn't split parameters";
 		goto bad;
 	}
 
@@ -724,7 +724,7 @@ int dm_table_add_target(struct dm_table *t, const char *type,
 	return 0;
 
  bad:
-	DMERR("%s: %s: %s", dm_device_name(t->md), type, tgt->error);
+	DMERR("%s: %s: %s (%pe)", dm_device_name(t->md), type, tgt->error, ERR_PTR(r));
 	dm_put_target_type(tgt->type);
 	return r;
 }

drivers/md/dm-verity-target.c

@@ -428,14 +428,14 @@ int verity_for_bv_block(struct dm_verity *v, struct dm_verity_io *io,
 		unsigned len;
 		struct bio_vec bv = bio_iter_iovec(bio, *iter);
 
-		page = kmap_atomic(bv.bv_page);
+		page = bvec_kmap_local(&bv);
 		len = bv.bv_len;
 
 		if (likely(len >= todo))
 			len = todo;
 
-		r = process(v, io, page + bv.bv_offset, len);
-		kunmap_atomic(page);
+		r = process(v, io, page, len);
+		kunmap_local(page);
 
 		if (r < 0)
 			return r;

drivers/md/dm-writecache.c

@@ -2264,14 +2264,13 @@ static int writecache_ctr(struct dm_target *ti, unsigned argc, char **argv)
 
 	raw_spin_lock_init(&wc->endio_list_lock);
 	INIT_LIST_HEAD(&wc->endio_list);
-	wc->endio_thread = kthread_create(writecache_endio_thread, wc, "writecache_endio");
+	wc->endio_thread = kthread_run(writecache_endio_thread, wc, "writecache_endio");
 	if (IS_ERR(wc->endio_thread)) {
 		r = PTR_ERR(wc->endio_thread);
 		wc->endio_thread = NULL;
 		ti->error = "Couldn't spawn endio thread";
 		goto bad;
 	}
-	wake_up_process(wc->endio_thread);
 
 	/*
 	 * Parse the mode (pmem or ssd)
@@ -2493,14 +2492,13 @@ static int writecache_ctr(struct dm_target *ti, unsigned argc, char **argv)
 		wc->memory_map_size -= (uint64_t)wc->start_sector << SECTOR_SHIFT;
 
 		bio_list_init(&wc->flush_list);
-		wc->flush_thread = kthread_create(writecache_flush_thread, wc, "dm_writecache_flush");
+		wc->flush_thread = kthread_run(writecache_flush_thread, wc, "dm_writecache_flush");
 		if (IS_ERR(wc->flush_thread)) {
 			r = PTR_ERR(wc->flush_thread);
 			wc->flush_thread = NULL;
 			ti->error = "Couldn't spawn flush thread";
 			goto bad;
 		}
-		wake_up_process(wc->flush_thread);
 
 		r = calculate_memory_size(wc->memory_map_size, wc->block_size,
 					  &n_blocks, &n_metadata_blocks);

drivers/md/dm-zoned-target.c

@@ -967,7 +967,6 @@ static void dmz_dtr(struct dm_target *ti)
 	struct dmz_target *dmz = ti->private;
 	int i;
 
-	flush_workqueue(dmz->chunk_wq);
 	destroy_workqueue(dmz->chunk_wq);
 
 	for (i = 0; i < dmz->nr_ddevs; i++)

drivers/md/dm.c

@@ -1792,7 +1792,7 @@ static struct mapped_device *alloc_dev(int minor)
 
 	format_dev_t(md->name, MKDEV(_major, minor));
 
-	md->wq = alloc_workqueue("kdmflush", WQ_MEM_RECLAIM, 0);
+	md->wq = alloc_workqueue("kdmflush/%s", WQ_MEM_RECLAIM, 0, md->name);
 	if (!md->wq)
 		goto bad;
 

include/uapi/linux/audit.h

@@ -120,6 +120,8 @@
 #define AUDIT_EVENT_LISTENER	1335	/* Task joined multicast read socket */
 #define AUDIT_URINGOP		1336	/* io_uring operation */
 #define AUDIT_OPENAT2		1337	/* Record showing openat2 how args */
+#define AUDIT_DM_CTRL		1338	/* Device Mapper target control */
+#define AUDIT_DM_EVENT		1339	/* Device Mapper events */
 
 #define AUDIT_AVC		1400	/* SE Linux avc denial or grant */
 #define AUDIT_SELINUX_ERR	1401	/* Internal SE Linux Errors */