Merge git://git.kernel.org/pub/scm/linux/kernel/git/agk/linux-2.6-dm

* git://git.kernel.org/pub/scm/linux/kernel/git/agk/linux-2.6-dm: (34 commits)
  dm table: set flush capability based on underlying devices
  dm crypt: optionally support discard requests
  dm raid: add md raid1 support
  dm raid: support metadata devices
  dm raid: add write_mostly parameter
  dm raid: add region_size parameter
  dm raid: improve table parameters documentation
  dm ioctl: forbid multiple device specifiers
  dm ioctl: introduce __get_dev_cell
  dm ioctl: fill in device parameters in more ioctls
  dm flakey: add corrupt_bio_byte feature
  dm flakey: add drop_writes
  dm flakey: support feature args
  dm flakey: use dm_target_offset and support discards
  dm table: share target argument parsing functions
  dm snapshot: skip reading origin when overwriting complete chunk
  dm: ignore merge_bvec for snapshots when safe
  dm table: clean dm_get_device and move exports
  dm raid: tidy includes
  dm ioctl: prevent empty message
  ...

15 files changed, 1351 insertions, 345 deletions

diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig
index 8420129fc5ee..f75a66e7d312 100644
--- a/drivers/md/Kconfig
+++ b/drivers/md/Kconfig
@@ -241,12 +241,13 @@ config DM_MIRROR
          needed for live data migration tools such as 'pvmove'.
 
 config DM_RAID
-       tristate "RAID 4/5/6 target (EXPERIMENTAL)"
+       tristate "RAID 1/4/5/6 target (EXPERIMENTAL)"
        depends on BLK_DEV_DM && EXPERIMENTAL
+       select MD_RAID1
        select MD_RAID456
        select BLK_DEV_MD
        ---help---
-	 A dm target that supports RAID4, RAID5 and RAID6 mappings
+	 A dm target that supports RAID1, RAID4, RAID5 and RAID6 mappings
 
 	 A RAID-5 set of N drives with a capacity of C MB per drive provides
 	 the capacity of C * (N - 1) MB, and protects against a failure
diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index bae6c4e23d3f..49da55c1528a 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -30,7 +30,6 @@
 #include <linux/device-mapper.h>
 
 #define DM_MSG_PREFIX "crypt"
-#define MESG_STR(x) x, sizeof(x)
 
 /*
  * context holding the current state of a multi-part conversion
@@ -239,7 +238,7 @@ static int crypt_iv_plain_gen(struct crypt_config *cc, u8 *iv,
 			      struct dm_crypt_request *dmreq)
 {
 	memset(iv, 0, cc->iv_size);
-	*(u32 *)iv = cpu_to_le32(dmreq->iv_sector & 0xffffffff);
+	*(__le32 *)iv = cpu_to_le32(dmreq->iv_sector & 0xffffffff);
 
 	return 0;
 }
@@ -248,7 +247,7 @@ static int crypt_iv_plain64_gen(struct crypt_config *cc, u8 *iv,
 				struct dm_crypt_request *dmreq)
 {
 	memset(iv, 0, cc->iv_size);
-	*(u64 *)iv = cpu_to_le64(dmreq->iv_sector);
+	*(__le64 *)iv = cpu_to_le64(dmreq->iv_sector);
 
 	return 0;
 }
@@ -415,7 +414,7 @@ static int crypt_iv_essiv_gen(struct crypt_config *cc, u8 *iv,
 	struct crypto_cipher *essiv_tfm = this_crypt_config(cc)->iv_private;
 
 	memset(iv, 0, cc->iv_size);
-	*(u64 *)iv = cpu_to_le64(dmreq->iv_sector);
+	*(__le64 *)iv = cpu_to_le64(dmreq->iv_sector);
 	crypto_cipher_encrypt_one(essiv_tfm, iv, iv);
 
 	return 0;
@@ -1575,11 +1574,17 @@ bad_mem:
 static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 {
 	struct crypt_config *cc;
-	unsigned int key_size;
+	unsigned int key_size, opt_params;
 	unsigned long long tmpll;
 	int ret;
+	struct dm_arg_set as;
+	const char *opt_string;
+
+	static struct dm_arg _args[] = {
+		{0, 1, "Invalid number of feature args"},
+	};
 
-	if (argc != 5) {
+	if (argc < 5) {
 		ti->error = "Not enough arguments";
 		return -EINVAL;
 	}
@@ -1648,6 +1653,30 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	}
 	cc->start = tmpll;
 
+	argv += 5;
+	argc -= 5;
+
+	/* Optional parameters */
+	if (argc) {
+		as.argc = argc;
+		as.argv = argv;
+
+		ret = dm_read_arg_group(_args, &as, &opt_params, &ti->error);
+		if (ret)
+			goto bad;
+
+		opt_string = dm_shift_arg(&as);
+
+		if (opt_params == 1 && opt_string &&
+		    !strcasecmp(opt_string, "allow_discards"))
+			ti->num_discard_requests = 1;
+		else if (opt_params) {
+			ret = -EINVAL;
+			ti->error = "Invalid feature arguments";
+			goto bad;
+		}
+	}
+
 	ret = -ENOMEM;
 	cc->io_queue = alloc_workqueue("kcryptd_io",
 				       WQ_NON_REENTRANT|
@@ -1682,9 +1711,16 @@ static int crypt_map(struct dm_target *ti, struct bio *bio,
 	struct dm_crypt_io *io;
 	struct crypt_config *cc;
 
-	if (bio->bi_rw & REQ_FLUSH) {
+	/*
+	 * If bio is REQ_FLUSH or REQ_DISCARD, just bypass crypt queues.
+	 * - for REQ_FLUSH device-mapper core ensures that no IO is in-flight
+	 * - for REQ_DISCARD caller must use flush if IO ordering matters
+	 */
+	if (unlikely(bio->bi_rw & (REQ_FLUSH | REQ_DISCARD))) {
 		cc = ti->private;
 		bio->bi_bdev = cc->dev->bdev;
+		if (bio_sectors(bio))
+			bio->bi_sector = cc->start + dm_target_offset(ti, bio->bi_sector);
 		return DM_MAPIO_REMAPPED;
 	}
 
@@ -1727,6 +1763,10 @@ static int crypt_status(struct dm_target *ti, status_type_t type,
 
 		DMEMIT(" %llu %s %llu", (unsigned long long)cc->iv_offset,
 				cc->dev->name, (unsigned long long)cc->start);
+
+		if (ti->num_discard_requests)
+			DMEMIT(" 1 allow_discards");
+
 		break;
 	}
 	return 0;
@@ -1770,12 +1810,12 @@ static int crypt_message(struct dm_target *ti, unsigned argc, char **argv)
 	if (argc < 2)
 		goto error;
 
-	if (!strnicmp(argv[0], MESG_STR("key"))) {
+	if (!strcasecmp(argv[0], "key")) {
 		if (!test_bit(DM_CRYPT_SUSPENDED, &cc->flags)) {
 			DMWARN("not suspended during key manipulation.");
 			return -EINVAL;
 		}
-		if (argc == 3 && !strnicmp(argv[1], MESG_STR("set"))) {
+		if (argc == 3 && !strcasecmp(argv[1], "set")) {
 			ret = crypt_set_key(cc, argv[2]);
 			if (ret)
 				return ret;
@@ -1783,7 +1823,7 @@ static int crypt_message(struct dm_target *ti, unsigned argc, char **argv)
 				ret = cc->iv_gen_ops->init(cc);
 			return ret;
 		}
-		if (argc == 2 && !strnicmp(argv[1], MESG_STR("wipe"))) {
+		if (argc == 2 && !strcasecmp(argv[1], "wipe")) {
 			if (cc->iv_gen_ops && cc->iv_gen_ops->wipe) {
 				ret = cc->iv_gen_ops->wipe(cc);
 				if (ret)
@@ -1823,7 +1863,7 @@ static int crypt_iterate_devices(struct dm_target *ti,
 
 static struct target_type crypt_target = {
 	.name   = "crypt",
-	.version = {1, 10, 0},
+	.version = {1, 11, 0},
 	.module = THIS_MODULE,
 	.ctr    = crypt_ctr,
 	.dtr    = crypt_dtr,
diff --git a/drivers/md/dm-flakey.c b/drivers/md/dm-flakey.c
index ea790623c30b..89f73ca22cfa 100644
--- a/drivers/md/dm-flakey.c
+++ b/drivers/md/dm-flakey.c
@@ -1,6 +1,6 @@
 /*
  * Copyright (C) 2003 Sistina Software (UK) Limited.
- * Copyright (C) 2004, 2010 Red Hat, Inc. All rights reserved.
+ * Copyright (C) 2004, 2010-2011 Red Hat, Inc. All rights reserved.
  *
  * This file is released under the GPL.
  */
@@ -15,6 +15,9 @@
 
 #define DM_MSG_PREFIX "flakey"
 
+#define all_corrupt_bio_flags_match(bio, fc)	\
+	(((bio)->bi_rw & (fc)->corrupt_bio_flags) == (fc)->corrupt_bio_flags)
+
 /*
  * Flakey: Used for testing only, simulates intermittent,
  * catastrophic device failure.
@@ -25,60 +28,189 @@ struct flakey_c {
 	sector_t start;
 	unsigned up_interval;
 	unsigned down_interval;
+	unsigned long flags;
+	unsigned corrupt_bio_byte;
+	unsigned corrupt_bio_rw;
+	unsigned corrupt_bio_value;
+	unsigned corrupt_bio_flags;
+};
+
+enum feature_flag_bits {
+	DROP_WRITES
 };
 
+static int parse_features(struct dm_arg_set *as, struct flakey_c *fc,
+			  struct dm_target *ti)
+{
+	int r;
+	unsigned argc;
+	const char *arg_name;
+
+	static struct dm_arg _args[] = {
+		{0, 6, "Invalid number of feature args"},
+		{1, UINT_MAX, "Invalid corrupt bio byte"},
+		{0, 255, "Invalid corrupt value to write into bio byte (0-255)"},
+		{0, UINT_MAX, "Invalid corrupt bio flags mask"},
+	};
+
+	/* No feature arguments supplied. */
+	if (!as->argc)
+		return 0;
+
+	r = dm_read_arg_group(_args, as, &argc, &ti->error);
+	if (r)
+		return r;
+
+	while (argc) {
+		arg_name = dm_shift_arg(as);
+		argc--;
+
+		/*
+		 * drop_writes
+		 */
+		if (!strcasecmp(arg_name, "drop_writes")) {
+			if (test_and_set_bit(DROP_WRITES, &fc->flags)) {
+				ti->error = "Feature drop_writes duplicated";
+				return -EINVAL;
+			}
+
+			continue;
+		}
+
+		/*
+		 * corrupt_bio_byte <Nth_byte> <direction> <value> <bio_flags>
+		 */
+		if (!strcasecmp(arg_name, "corrupt_bio_byte")) {
+			if (!argc)
+				ti->error = "Feature corrupt_bio_byte requires parameters";
+
+			r = dm_read_arg(_args + 1, as, &fc->corrupt_bio_byte, &ti->error);
+			if (r)
+				return r;
+			argc--;
+
+			/*
+			 * Direction r or w?
+			 */
+			arg_name = dm_shift_arg(as);
+			if (!strcasecmp(arg_name, "w"))
+				fc->corrupt_bio_rw = WRITE;
+			else if (!strcasecmp(arg_name, "r"))
+				fc->corrupt_bio_rw = READ;
+			else {
+				ti->error = "Invalid corrupt bio direction (r or w)";
+				return -EINVAL;
+			}
+			argc--;
+
+			/*
+			 * Value of byte (0-255) to write in place of correct one.
+			 */
+			r = dm_read_arg(_args + 2, as, &fc->corrupt_bio_value, &ti->error);
+			if (r)
+				return r;
+			argc--;
+
+			/*
+			 * Only corrupt bios with these flags set.
+			 */
+			r = dm_read_arg(_args + 3, as, &fc->corrupt_bio_flags, &ti->error);
+			if (r)
+				return r;
+			argc--;
+
+			continue;
+		}
+
+		ti->error = "Unrecognised flakey feature requested";
+		return -EINVAL;
+	}
+
+	if (test_bit(DROP_WRITES, &fc->flags) && (fc->corrupt_bio_rw == WRITE)) {
+		ti->error = "drop_writes is incompatible with corrupt_bio_byte with the WRITE flag set";
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 /*
- * Construct a flakey mapping: <dev_path> <offset> <up interval> <down interval>
+ * Construct a flakey mapping:
+ * <dev_path> <offset> <up interval> <down interval> [<#feature args> [<arg>]*]
+ *
+ *   Feature args:
+ *     [drop_writes]
+ *     [corrupt_bio_byte <Nth_byte> <direction> <value> <bio_flags>]
+ *
+ *   Nth_byte starts from 1 for the first byte.
+ *   Direction is r for READ or w for WRITE.
+ *   bio_flags is ignored if 0.
  */
 static int flakey_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 {
+	static struct dm_arg _args[] = {
+		{0, UINT_MAX, "Invalid up interval"},
+		{0, UINT_MAX, "Invalid down interval"},
+	};
+
+	int r;
 	struct flakey_c *fc;
-	unsigned long long tmp;
+	unsigned long long tmpll;
+	struct dm_arg_set as;
+	const char *devname;
 
-	if (argc != 4) {
-		ti->error = "dm-flakey: Invalid argument count";
+	as.argc = argc;
+	as.argv = argv;
+
+	if (argc < 4) {
+		ti->error = "Invalid argument count";
 		return -EINVAL;
 	}
 
-	fc = kmalloc(sizeof(*fc), GFP_KERNEL);
+	fc = kzalloc(sizeof(*fc), GFP_KERNEL);
 	if (!fc) {
-		ti->error = "dm-flakey: Cannot allocate linear context";
+		ti->error = "Cannot allocate linear context";
 		return -ENOMEM;
 	}
 	fc->start_time = jiffies;
 
-	if (sscanf(argv[1], "%llu", &tmp) != 1) {
-		ti->error = "dm-flakey: Invalid device sector";
+	devname = dm_shift_arg(&as);
+
+	if (sscanf(dm_shift_arg(&as), "%llu", &tmpll) != 1) {
+		ti->error = "Invalid device sector";
 		goto bad;
 	}
-	fc->start = tmp;
+	fc->start = tmpll;
 
-	if (sscanf(argv[2], "%u", &fc->up_interval) != 1) {
-		ti->error = "dm-flakey: Invalid up interval";
+	r = dm_read_arg(_args, &as, &fc->up_interval, &ti->error);
+	if (r)
 		goto bad;
-	}
 
-	if (sscanf(argv[3], "%u", &fc->down_interval) != 1) {
-		ti->error = "dm-flakey: Invalid down interval";
+	r = dm_read_arg(_args, &as, &fc->down_interval, &ti->error);
+	if (r)
 		goto bad;
-	}
 
 	if (!(fc->up_interval + fc->down_interval)) {
-		ti->error = "dm-flakey: Total (up + down) interval is zero";
+		ti->error = "Total (up + down) interval is zero";
 		goto bad;
 	}
 
 	if (fc->up_interval + fc->down_interval < fc->up_interval) {
-		ti->error = "dm-flakey: Interval overflow";
+		ti->error = "Interval overflow";
 		goto bad;
 	}
 
-	if (dm_get_device(ti, argv[0], dm_table_get_mode(ti->table), &fc->dev)) {
-		ti->error = "dm-flakey: Device lookup failed";
+	r = parse_features(&as, fc, ti);
+	if (r)
+		goto bad;
+
+	if (dm_get_device(ti, devname, dm_table_get_mode(ti->table), &fc->dev)) {
+		ti->error = "Device lookup failed";
 		goto bad;
 	}
 
 	ti->num_flush_requests = 1;
+	ti->num_discard_requests = 1;
 	ti->private = fc;
 	return 0;
 
@@ -99,7 +231,7 @@ static sector_t flakey_map_sector(struct dm_target *ti, sector_t bi_sector)
 {
 	struct flakey_c *fc = ti->private;
 
-	return fc->start + (bi_sector - ti->begin);
+	return fc->start + dm_target_offset(ti, bi_sector);
 }
 
 static void flakey_map_bio(struct dm_target *ti, struct bio *bio)
@@ -111,6 +243,25 @@ static void flakey_map_bio(struct dm_target *ti, struct bio *bio)
 		bio->bi_sector = flakey_map_sector(ti, bio->bi_sector);
 }
 
+static void corrupt_bio_data(struct bio *bio, struct flakey_c *fc)
+{
+	unsigned bio_bytes = bio_cur_bytes(bio);
+	char *data = bio_data(bio);
+
+	/*
+	 * Overwrite the Nth byte of the data returned.
+	 */
+	if (data && bio_bytes >= fc->corrupt_bio_byte) {
+		data[fc->corrupt_bio_byte - 1] = fc->corrupt_bio_value;
+
+		DMDEBUG("Corrupting data bio=%p by writing %u to byte %u "
+			"(rw=%c bi_rw=%lu bi_sector=%llu cur_bytes=%u)\n",
+			bio, fc->corrupt_bio_value, fc->corrupt_bio_byte,
+			(bio_data_dir(bio) == WRITE) ? 'w' : 'r',
+			bio->bi_rw, (unsigned long long)bio->bi_sector, bio_bytes);
+	}
+}
+
 static int flakey_map(struct dm_target *ti, struct bio *bio,
 		      union map_info *map_context)
 {
@@ -119,18 +270,71 @@ static int flakey_map(struct dm_target *ti, struct bio *bio,
 
 	/* Are we alive ? */
 	elapsed = (jiffies - fc->start_time) / HZ;
-	if (elapsed % (fc->up_interval + fc->down_interval) >= fc->up_interval)
+	if (elapsed % (fc->up_interval + fc->down_interval) >= fc->up_interval) {
+		/*
+		 * Flag this bio as submitted while down.
+		 */
+		map_context->ll = 1;
+
+		/*
+		 * Map reads as normal.
+		 */
+		if (bio_data_dir(bio) == READ)
+			goto map_bio;
+
+		/*
+		 * Drop writes?
+		 */
+		if (test_bit(DROP_WRITES, &fc->flags)) {
+			bio_endio(bio, 0);
+			return DM_MAPIO_SUBMITTED;
+		}
+
+		/*
+		 * Corrupt matching writes.
+		 */
+		if (fc->corrupt_bio_byte && (fc->corrupt_bio_rw == WRITE)) {
+			if (all_corrupt_bio_flags_match(bio, fc))
+				corrupt_bio_data(bio, fc);
+			goto map_bio;
+		}
+
+		/*
+		 * By default, error all I/O.
+		 */
 		return -EIO;
+	}
 
+map_bio:
 	flakey_map_bio(ti, bio);
 
 	return DM_MAPIO_REMAPPED;
 }
 
+static int flakey_end_io(struct dm_target *ti, struct bio *bio,
+			 int error, union map_info *map_context)
+{
+	struct flakey_c *fc = ti->private;
+	unsigned bio_submitted_while_down = map_context->ll;
+
+	/*
+	 * Corrupt successful READs while in down state.
+	 * If flags were specified, only corrupt those that match.
+	 */
+	if (!error && bio_submitted_while_down &&
+	    (bio_data_dir(bio) == READ) && (fc->corrupt_bio_rw == READ) &&
+	    all_corrupt_bio_flags_match(bio, fc))
+		corrupt_bio_data(bio, fc);
+
+	return error;
+}
+
 static int flakey_status(struct dm_target *ti, status_type_t type,
 			 char *result, unsigned int maxlen)
 {
+	unsigned sz = 0;
 	struct flakey_c *fc = ti->private;
+	unsigned drop_writes;
 
 	switch (type) {
 	case STATUSTYPE_INFO:
@@ -138,9 +342,22 @@ static int flakey_status(struct dm_target *ti, status_type_t type,
 		break;
 
 	case STATUSTYPE_TABLE:
-		snprintf(result, maxlen, "%s %llu %u %u", fc->dev->name,
-			 (unsigned long long)fc->start, fc->up_interval,
-			 fc->down_interval);
+		DMEMIT("%s %llu %u %u ", fc->dev->name,
+		       (unsigned long long)fc->start, fc->up_interval,
+		       fc->down_interval);
+
+		drop_writes = test_bit(DROP_WRITES, &fc->flags);
+		DMEMIT("%u ", drop_writes + (fc->corrupt_bio_byte > 0) * 5);
+
+		if (drop_writes)
+			DMEMIT("drop_writes ");
+
+		if (fc->corrupt_bio_byte)
+			DMEMIT("corrupt_bio_byte %u %c %u %u ",
+			       fc->corrupt_bio_byte,
+			       (fc->corrupt_bio_rw == WRITE) ? 'w' : 'r',
+			       fc->corrupt_bio_value, fc->corrupt_bio_flags);
+
 		break;
 	}
 	return 0;
@@ -177,11 +394,12 @@ static int flakey_iterate_devices(struct dm_target *ti, iterate_devices_callout_
 
 static struct target_type flakey_target = {
 	.name   = "flakey",
-	.version = {1, 1, 0},
+	.version = {1, 2, 0},
 	.module = THIS_MODULE,
 	.ctr    = flakey_ctr,
 	.dtr    = flakey_dtr,
 	.map    = flakey_map,
+	.end_io = flakey_end_io,
 	.status = flakey_status,
 	.ioctl	= flakey_ioctl,
 	.merge	= flakey_merge,
diff --git a/drivers/md/dm-io.c b/drivers/md/dm-io.c
index 2067288f61f9..ad2eba40e319 100644
--- a/drivers/md/dm-io.c
+++ b/drivers/md/dm-io.c
@@ -38,6 +38,8 @@ struct io {
 	struct dm_io_client *client;
 	io_notify_fn callback;
 	void *context;
+	void *vma_invalidate_address;
+	unsigned long vma_invalidate_size;
 } __attribute__((aligned(DM_IO_MAX_REGIONS)));
 
 static struct kmem_cache *_dm_io_cache;
@@ -116,6 +118,10 @@ static void dec_count(struct io *io, unsigned int region, int error)
 		set_bit(region, &io->error_bits);
 
 	if (atomic_dec_and_test(&io->count)) {
+		if (io->vma_invalidate_size)
+			invalidate_kernel_vmap_range(io->vma_invalidate_address,
+						     io->vma_invalidate_size);
+
 		if (io->sleeper)
 			wake_up_process(io->sleeper);
 
@@ -159,6 +165,9 @@ struct dpages {
 
 	unsigned context_u;
 	void *context_ptr;
+
+	void *vma_invalidate_address;
+	unsigned long vma_invalidate_size;
 };
 
 /*
@@ -377,6 +386,9 @@ static int sync_io(struct dm_io_client *client, unsigned int num_regions,
 	io->sleeper = current;
 	io->client = client;
 
+	io->vma_invalidate_address = dp->vma_invalidate_address;
+	io->vma_invalidate_size = dp->vma_invalidate_size;
+
 	dispatch_io(rw, num_regions, where, dp, io, 1);
 
 	while (1) {
@@ -415,13 +427,21 @@ static int async_io(struct dm_io_client *client, unsigned int num_regions,
 	io->callback = fn;
 	io->context = context;
 
+	io->vma_invalidate_address = dp->vma_invalidate_address;
+	io->vma_invalidate_size = dp->vma_invalidate_size;
+
 	dispatch_io(rw, num_regions, where, dp, io, 0);
 	return 0;
 }
 
-static int dp_init(struct dm_io_request *io_req, struct dpages *dp)
+static int dp_init(struct dm_io_request *io_req, struct dpages *dp,
+		   unsigned long size)
 {
 	/* Set up dpages based on memory type */
+
+	dp->vma_invalidate_address = NULL;
+	dp->vma_invalidate_size = 0;
+
 	switch (io_req->mem.type) {
 	case DM_IO_PAGE_LIST:
 		list_dp_init(dp, io_req->mem.ptr.pl, io_req->mem.offset);
@@ -432,6 +452,11 @@ static int dp_init(struct dm_io_request *io_req, struct dpages *dp)
 		break;
 
 	case DM_IO_VMA:
+		flush_kernel_vmap_range(io_req->mem.ptr.vma, size);
+		if ((io_req->bi_rw & RW_MASK) == READ) {
+			dp->vma_invalidate_address = io_req->mem.ptr.vma;
+			dp->vma_invalidate_size = size;
+		}
 		vm_dp_init(dp, io_req->mem.ptr.vma);
 		break;
 
@@ -460,7 +485,7 @@ int dm_io(struct dm_io_request *io_req, unsigned num_regions,
 	int r;
 	struct dpages dp;
 
-	r = dp_init(io_req, &dp);
+	r = dp_init(io_req, &dp, (unsigned long)where->count << SECTOR_SHIFT);
 	if (r)
 		return r;
 
diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c
index 4cacdad2270a..2e9a3ca37bdd 100644
--- a/drivers/md/dm-ioctl.c
+++ b/drivers/md/dm-ioctl.c
@@ -128,6 +128,24 @@ static struct hash_cell *__get_uuid_cell(const char *str)
 	return NULL;
 }
 
+static struct hash_cell *__get_dev_cell(uint64_t dev)
+{
+	struct mapped_device *md;
+	struct hash_cell *hc;
+
+	md = dm_get_md(huge_decode_dev(dev));
+	if (!md)
+		return NULL;
+
+	hc = dm_get_mdptr(md);
+	if (!hc) {
+		dm_put(md);
+		return NULL;
+	}
+
+	return hc;
+}
+
 /*-----------------------------------------------------------------
  * Inserting, removing and renaming a device.
  *---------------------------------------------------------------*/
@@ -718,25 +736,45 @@ static int dev_create(struct dm_ioctl *param, size_t param_size)
  */
 static struct hash_cell *__find_device_hash_cell(struct dm_ioctl *param)
 {
-	struct mapped_device *md;
-	void *mdptr = NULL;
+	struct hash_cell *hc = NULL;
 
-	if (*param->uuid)
-		return __get_uuid_cell(param->uuid);
+	if (*param->uuid) {
+		if (*param->name || param->dev)
+			return NULL;
 
-	if (*param->name)
-		return __get_name_cell(param->name);
+		hc = __get_uuid_cell(param->uuid);
+		if (!hc)
+			return NULL;
+	} else if (*param->name) {
+		if (param->dev)
+			return NULL;
 
-	md = dm_get_md(huge_decode_dev(param->dev));
-	if (!md)
-		goto out;
+		hc = __get_name_cell(param->name);
+		if (!hc)
+			return NULL;
+	} else if (param->dev) {
+		hc = __get_dev_cell(param->dev);
+		if (!hc)
+			return NULL;
+	} else
+		return NULL;
 
-	mdptr = dm_get_mdptr(md);
-	if (!mdptr)
-		dm_put(md);
+	/*
+	 * Sneakily write in both the name and the uuid
+	 * while we have the cell.
+	 */
+	strlcpy(param->name, hc->name, sizeof(param->name));
+	if (hc->uuid)
+		strlcpy(param->uuid, hc->uuid, sizeof(param->uuid));
+	else
+		param->uuid[0] = '\0';
 
-out:
-	return mdptr;
+	if (hc->new_map)
+		param->flags |= DM_INACTIVE_PRESENT_FLAG;
+	else
+		param->flags &= ~DM_INACTIVE_PRESENT_FLAG;
+
+	return hc;
 }
 
 static struct mapped_device *find_device(struct dm_ioctl *param)
@@ -746,24 +784,8 @@ static struct mapped_device *find_device(struct dm_ioctl *param)
 
 	down_read(&_hash_lock);
 	hc = __find_device_hash_cell(param);
-	if (hc) {
+	if (hc)
 		md = hc->md;
-
-		/*
-		 * Sneakily write in both the name and the uuid
-		 * while we have the cell.
-		 */
-		strlcpy(param->name, hc->name, sizeof(param->name));
-		if (hc->uuid)
-			strlcpy(param->uuid, hc->uuid, sizeof(param->uuid));
-		else
-			param->uuid[0] = '\0';
-
-		if (hc->new_map)
-			param->flags |= DM_INACTIVE_PRESENT_FLAG;
-		else
-			param->flags &= ~DM_INACTIVE_PRESENT_FLAG;
-	}
 	up_read(&_hash_lock);
 
 	return md;
@@ -1402,6 +1424,11 @@ static int target_message(struct dm_ioctl *param, size_t param_size)
 		goto out;
 	}
 
+	if (!argc) {
+		DMWARN("Empty message received.");
+		goto out;
+	}
+
 	table = dm_get_live_table(md);
 	if (!table)
 		goto out_argv;
diff --git a/drivers/md/dm-kcopyd.c b/drivers/md/dm-kcopyd.c
index 320401dec104..f82147029636 100644
--- a/drivers/md/dm-kcopyd.c
+++ b/drivers/md/dm-kcopyd.c
@@ -224,8 +224,6 @@ struct kcopyd_job {
 	unsigned int num_dests;
 	struct dm_io_region dests[DM_KCOPYD_MAX_REGIONS];
 
-	sector_t offset;
-	unsigned int nr_pages;
 	struct page_list *pages;
 
 	/*
@@ -380,7 +378,7 @@ static int run_io_job(struct kcopyd_job *job)
 		.bi_rw = job->rw,
 		.mem.type = DM_IO_PAGE_LIST,
 		.mem.ptr.pl = job->pages,
-		.mem.offset = job->offset,
+		.mem.offset = 0,
 		.notify.fn = complete_io,
 		.notify.context = job,
 		.client = job->kc->io_client,
@@ -397,10 +395,9 @@ static int run_io_job(struct kcopyd_job *job)
 static int run_pages_job(struct kcopyd_job *job)
 {
 	int r;
+	unsigned nr_pages = dm_div_up(job->dests[0].count, PAGE_SIZE >> 9);
 
-	job->nr_pages = dm_div_up(job->dests[0].count + job->offset,
-				  PAGE_SIZE >> 9);
-	r = kcopyd_get_pages(job->kc, job->nr_pages, &job->pages);
+	r = kcopyd_get_pages(job->kc, nr_pages, &job->pages);
 	if (!r) {
 		/* this job is ready for io */
 		push(&job->kc->io_jobs, job);
@@ -602,8 +599,6 @@ int dm_kcopyd_copy(struct dm_kcopyd_client *kc, struct dm_io_region *from,
 	job->num_dests = num_dests;
 	memcpy(&job->dests, dests, sizeof(*dests) * num_dests);
 
-	job->offset = 0;
-	job->nr_pages = 0;
 	job->pages = NULL;
 
 	job->fn = fn;
@@ -622,6 +617,37 @@ int dm_kcopyd_copy(struct dm_kcopyd_client *kc, struct dm_io_region *from,
 }
 EXPORT_SYMBOL(dm_kcopyd_copy);
 
+void *dm_kcopyd_prepare_callback(struct dm_kcopyd_client *kc,
+				 dm_kcopyd_notify_fn fn, void *context)
+{
+	struct kcopyd_job *job;
+
+	job = mempool_alloc(kc->job_pool, GFP_NOIO);
+
+	memset(job, 0, sizeof(struct kcopyd_job));
+	job->kc = kc;
+	job->fn = fn;
+	job->context = context;
+
+	atomic_inc(&kc->nr_jobs);
+
+	return job;
+}
+EXPORT_SYMBOL(dm_kcopyd_prepare_callback);
+
+void dm_kcopyd_do_callback(void *j, int read_err, unsigned long write_err)
+{
+	struct kcopyd_job *job = j;
+	struct dm_kcopyd_client *kc = job->kc;
+
+	job->read_err = read_err;
+	job->write_err = write_err;
+
+	push(&kc->complete_jobs, job);
+	wake(kc);
+}
+EXPORT_SYMBOL(dm_kcopyd_do_callback);
+
 /*
  * Cancels a kcopyd job, eg. someone might be deactivating a
  * mirror.
diff --git a/drivers/md/dm-log-userspace-base.c b/drivers/md/dm-log-userspace-base.c
index aa2e0c374ab3..1021c8986011 100644
--- a/drivers/md/dm-log-userspace-base.c
+++ b/drivers/md/dm-log-userspace-base.c
@@ -394,8 +394,7 @@ static int flush_by_group(struct log_c *lc, struct list_head *flush_list)
 			group[count] = fe->region;
 			count++;
 
-			list_del(&fe->list);
-			list_add(&fe->list, &tmp_list);
+			list_move(&fe->list, &tmp_list);
 
 			type = fe->type;
 			if (count >= MAX_FLUSH_GROUP_COUNT)
diff --git a/drivers/md/dm-log.c b/drivers/md/dm-log.c
index 948e3f4925bf..3b52bb72bd1f 100644
--- a/drivers/md/dm-log.c
+++ b/drivers/md/dm-log.c
@@ -197,15 +197,21 @@ EXPORT_SYMBOL(dm_dirty_log_destroy);
 #define MIRROR_DISK_VERSION 2
 #define LOG_OFFSET 2
 
-struct log_header {
-	uint32_t magic;
+struct log_header_disk {
+	__le32 magic;
 
 	/*
 	 * Simple, incrementing version. no backward
 	 * compatibility.
 	 */
+	__le32 version;
+	__le64 nr_regions;
+} __packed;
+
+struct log_header_core {
+	uint32_t magic;
 	uint32_t version;
-	sector_t nr_regions;
+	uint64_t nr_regions;
 };
 
 struct log_c {
@@ -239,10 +245,10 @@ struct log_c {
 	int log_dev_failed;
 	int log_dev_flush_failed;
 	struct dm_dev *log_dev;
-	struct log_header header;
+	struct log_header_core header;
 
 	struct dm_io_region header_location;
-	struct log_header *disk_header;
+	struct log_header_disk *disk_header;
 };
 
 /*
@@ -251,34 +257,34 @@ struct log_c {
  */
 static inline int log_test_bit(uint32_t *bs, unsigned bit)
 {
-	return test_bit_le(bit, (unsigned long *) bs) ? 1 : 0;
+	return test_bit_le(bit, bs) ? 1 : 0;
 }
 
 static inline void log_set_bit(struct log_c *l,
 			       uint32_t *bs, unsigned bit)
 {
-	__test_and_set_bit_le(bit, (unsigned long *) bs);
+	__set_bit_le(bit, bs);
 	l->touched_cleaned = 1;
 }
 
 static inline void log_clear_bit(struct log_c *l,
 				 uint32_t *bs, unsigned bit)
 {
-	__test_and_clear_bit_le(bit, (unsigned long *) bs);
+	__clear_bit_le(bit, bs);
 	l->touched_dirtied = 1;
 }
 
 /*----------------------------------------------------------------
  * Header IO
  *--------------------------------------------------------------*/
-static void header_to_disk(struct log_header *core, struct log_header *disk)
+static void header_to_disk(struct log_header_core *core, struct log_header_disk *disk)
 {
 	disk->magic = cpu_to_le32(core->magic);
 	disk->version = cpu_to_le32(core->version);
 	disk->nr_regions = cpu_to_le64(core->nr_regions);
 }
 
-static void header_from_disk(struct log_header *core, struct log_header *disk)
+static void header_from_disk(struct log_header_core *core, struct log_header_disk *disk)
 {
 	core->magic = le32_to_cpu(disk->magic);
 	core->version = le32_to_cpu(disk->version);
@@ -486,7 +492,7 @@ static int create_log_context(struct dm_dirty_log *log, struct dm_target *ti,
 	memset(lc->sync_bits, (sync == NOSYNC) ? -1 : 0, bitset_size);
 	lc->sync_count = (sync == NOSYNC) ? region_count : 0;
 
-	lc->recovering_bits = vmalloc(bitset_size);
+	lc->recovering_bits = vzalloc(bitset_size);
 	if (!lc->recovering_bits) {
 		DMWARN("couldn't allocate sync bitset");
 		vfree(lc->sync_bits);
@@ -498,7 +504,6 @@ static int create_log_context(struct dm_dirty_log *log, struct dm_target *ti,
 		kfree(lc);
 		return -ENOMEM;
 	}
-	memset(lc->recovering_bits, 0, bitset_size);
 	lc->sync_search = 0;
 	log->context = lc;
 
@@ -739,8 +744,7 @@ static int core_get_resync_work(struct dm_dirty_log *log, region_t *region)
 		return 0;
 
 	do {
-		*region = find_next_zero_bit_le(
-					     (unsigned long *) lc->sync_bits,
+		*region = find_next_zero_bit_le(lc->sync_bits,
 					     lc->region_count,
 					     lc->sync_search);
 		lc->sync_search = *region + 1;
diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index c3547016f0f1..5e0090ef4182 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -22,7 +22,6 @@
 #include <linux/atomic.h>
 
 #define DM_MSG_PREFIX "multipath"
-#define MESG_STR(x) x, sizeof(x)
 #define DM_PG_INIT_DELAY_MSECS 2000
 #define DM_PG_INIT_DELAY_DEFAULT ((unsigned) -1)
 
@@ -505,80 +504,29 @@ static void trigger_event(struct work_struct *work)
  *      <#paths> <#per-path selector args>
  *         [<path> [<arg>]* ]+ ]+
  *---------------------------------------------------------------*/
-struct param {
-	unsigned min;
-	unsigned max;
-	char *error;
-};
-
-static int read_param(struct param *param, char *str, unsigned *v, char **error)
-{
-	if (!str ||
-	    (sscanf(str, "%u", v) != 1) ||
-	    (*v < param->min) ||
-	    (*v > param->max)) {
-		*error = param->error;
-		return -EINVAL;
-	}
-
-	return 0;
-}
-