X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=drivers%2Fmd%2Fdm-crypt.c;h=e412980763bded7f0de5f05c01e9a896e254ef7e;hb=f653398c86a1c104f0992bd788dd4bb065449be4;hp=6dbaeee48ced47b45d71b38a8fd6bc9cf5256730;hpb=48527fa7cf7fefb84e9fe03cddd08ddafc9f15f3;p=safe%2Fjmp%2Flinux-2.6

diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index 6dbaeee..e412980 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -1,11 +1,12 @@
 /*
  * Copyright (C) 2003 Christophe Saout <christophe@saout.de>
  * Copyright (C) 2004 Clemens Fruhwirth <clemens@endorphin.org>
- * Copyright (C) 2006 Red Hat, Inc. All rights reserved.
+ * Copyright (C) 2006-2008 Red Hat, Inc. All rights reserved.
  *
  * This file is released under the GPL.
  */
 
+#include <linux/completion.h>
 #include <linux/err.h>
 #include <linux/module.h>
 #include <linux/init.h>
@@ -22,28 +23,16 @@
 #include <asm/page.h>
 #include <asm/unaligned.h>
 
-#include "dm.h"
+#include <linux/device-mapper.h>
 
 #define DM_MSG_PREFIX "crypt"
 #define MESG_STR(x) x, sizeof(x)
 
 /*
- * per bio private data
- */
-struct crypt_io {
-	struct dm_target *target;
-	struct bio *base_bio;
-	struct bio *first_clone;
-	struct work_struct work;
-	atomic_t pending;
-	int error;
-	int post_process;
-};
-
-/*
  * context holding the current state of a multi-part conversion
  */
 struct convert_context {
+	struct completion restart;
 	struct bio *bio_in;
 	struct bio *bio_out;
 	unsigned int offset_in;
@@ -51,14 +40,36 @@ struct convert_context {
 	unsigned int idx_in;
 	unsigned int idx_out;
 	sector_t sector;
-	int write;
+	atomic_t pending;
+};
+
+/*
+ * per bio private data
+ */
+struct dm_crypt_io {
+	struct dm_target *target;
+	struct bio *base_bio;
+	struct work_struct work;
+
+	struct convert_context ctx;
+
+	atomic_t pending;
+	int error;
+	sector_t sector;
+	struct dm_crypt_io *base_io;
+};
+
+struct dm_crypt_request {
+	struct convert_context *ctx;
+	struct scatterlist sg_in;
+	struct scatterlist sg_out;
 };
 
 struct crypt_config;
 
 struct crypt_iv_operations {
 	int (*ctr)(struct crypt_config *cc, struct dm_target *ti,
-	           const char *opts);
+		   const char *opts);
 	void (*dtr)(struct crypt_config *cc);
 	const char *(*status)(struct crypt_config *cc);
 	int (*generator)(struct crypt_config *cc, u8 *iv, sector_t sector);
@@ -74,25 +85,48 @@ struct crypt_config {
 	sector_t start;
 
 	/*
-	 * pool for per bio private data and
-	 * for encryption buffer pages
+	 * pool for per bio private data, crypto requests and
+	 * encryption requeusts/buffer pages
 	 */
 	mempool_t *io_pool;
+	mempool_t *req_pool;
 	mempool_t *page_pool;
 	struct bio_set *bs;
 
+	struct workqueue_struct *io_queue;
+	struct workqueue_struct *crypt_queue;
+
 	/*
 	 * crypto related data
 	 */
 	struct crypt_iv_operations *iv_gen_ops;
 	char *iv_mode;
-	struct crypto_cipher *iv_gen_private;
+	union {
+		struct crypto_cipher *essiv_tfm;
+		int benbi_shift;
+	} iv_gen_private;
 	sector_t iv_offset;
 	unsigned int iv_size;
 
+	/*
+	 * Layout of each crypto request:
+	 *
+	 *   struct ablkcipher_request
+	 *      context
+	 *      padding
+	 *   struct dm_crypt_request
+	 *      padding
+	 *   IV
+	 *
+	 * The padding is added so that dm_crypt_request and the IV are
+	 * correctly aligned.
+	 */
+	unsigned int dmreq_start;
+	struct ablkcipher_request *req;
+
 	char cipher[CRYPTO_MAX_ALG_NAME];
 	char chainmode[CRYPTO_MAX_ALG_NAME];
-	struct crypto_blkcipher *tfm;
+	struct crypto_ablkcipher *tfm;
 	unsigned long flags;
 	unsigned int key_size;
 	u8 key[0];
@@ -102,13 +136,16 @@ struct crypt_config {
 #define MIN_POOL_PAGES 32
 #define MIN_BIO_PAGES  8
 
-static kmem_cache_t *_crypt_io_pool;
+static struct kmem_cache *_crypt_io_pool;
+
+static void clone_init(struct dm_crypt_io *, struct bio *);
+static void kcryptd_queue_crypt(struct dm_crypt_io *io);
 
 /*
  * Different IV generation algorithms:
  *
  * plain: the initial vector is the 32-bit little-endian version of the sector
- *        number, padded with zeros if neccessary.
+ *        number, padded with zeros if necessary.
  *
  * essiv: "encrypted sector|salt initial vector", the sector number is
  *        encrypted with the bulk cipher using a salt as key. The salt
@@ -117,6 +154,9 @@ static kmem_cache_t *_crypt_io_pool;
  * benbi: the 64-bit "big-endian 'narrow block'-count", starting at 1
  *        (needed for LRW-32-AES and possible other narrow block modes)
  *
+ * null: the initial vector is always zero.  Provides compatibility with
+ *       obsolete loop_fish2 devices.  Do not use for new devices.
+ *
  * plumb: unimplemented, see:
  * http://article.gmane.org/gmane.linux.kernel.device-mapper.dm-crypt/454
  */
@@ -130,7 +170,7 @@ static int crypt_iv_plain_gen(struct crypt_config *cc, u8 *iv, sector_t sector)
 }
 
 static int crypt_iv_essiv_ctr(struct crypt_config *cc, struct dm_target *ti,
-	                      const char *opts)
+			      const char *opts)
 {
 	struct crypto_cipher *essiv_tfm;
 	struct crypto_hash *hash_tfm;
@@ -160,7 +200,7 @@ static int crypt_iv_essiv_ctr(struct crypt_config *cc, struct dm_target *ti,
 		return -ENOMEM;
 	}
 
-	sg_set_buf(&sg, cc->key, cc->key_size);
+	sg_init_one(&sg, cc->key, cc->key_size);
 	desc.tfm = hash_tfm;
 	desc.flags = CRYPTO_TFM_REQ_MAY_SLEEP;
 	err = crypto_hash_digest(&desc, &sg, cc->key_size, salt);
@@ -168,6 +208,7 @@ static int crypt_iv_essiv_ctr(struct crypt_config *cc, struct dm_target *ti,
 
 	if (err) {
 		ti->error = "Error calculating hash in ESSIV";
+		kfree(salt);
 		return err;
 	}
 
@@ -179,9 +220,9 @@ static int crypt_iv_essiv_ctr(struct crypt_config *cc, struct dm_target *ti,
 		return PTR_ERR(essiv_tfm);
 	}
 	if (crypto_cipher_blocksize(essiv_tfm) !=
-	    crypto_blkcipher_ivsize(cc->tfm)) {
+	    crypto_ablkcipher_ivsize(cc->tfm)) {
 		ti->error = "Block size of ESSIV cipher does "
-			        "not match IV size of block cipher";
+			    "not match IV size of block cipher";
 		crypto_free_cipher(essiv_tfm);
 		kfree(salt);
 		return -EINVAL;
@@ -195,29 +236,29 @@ static int crypt_iv_essiv_ctr(struct crypt_config *cc, struct dm_target *ti,
 	}
 	kfree(salt);
 
-	cc->iv_gen_private = essiv_tfm;
+	cc->iv_gen_private.essiv_tfm = essiv_tfm;
 	return 0;
 }
 
 static void crypt_iv_essiv_dtr(struct crypt_config *cc)
 {
-	crypto_free_cipher(cc->iv_gen_private);
-	cc->iv_gen_private = NULL;
+	crypto_free_cipher(cc->iv_gen_private.essiv_tfm);
+	cc->iv_gen_private.essiv_tfm = NULL;
 }
 
 static int crypt_iv_essiv_gen(struct crypt_config *cc, u8 *iv, sector_t sector)
 {
 	memset(iv, 0, cc->iv_size);
 	*(u64 *)iv = cpu_to_le64(sector);
-	crypto_cipher_encrypt_one(cc->iv_gen_private, iv, iv);
+	crypto_cipher_encrypt_one(cc->iv_gen_private.essiv_tfm, iv, iv);
 	return 0;
 }
 
 static int crypt_iv_benbi_ctr(struct crypt_config *cc, struct dm_target *ti,
 			      const char *opts)
 {
-	unsigned int bs = crypto_blkcipher_blocksize(cc->tfm);
-	int log = long_log2(bs);
+	unsigned bs = crypto_ablkcipher_blocksize(cc->tfm);
+	int log = ilog2(bs);
 
 	/* we need to calculate how far we must shift the sector count
 	 * to get the cipher block count, we use this shift in _gen */
@@ -232,21 +273,30 @@ static int crypt_iv_benbi_ctr(struct crypt_config *cc, struct dm_target *ti,
 		return -EINVAL;
 	}
 
-	cc->iv_gen_private = (void *)(9 - log);
+	cc->iv_gen_private.benbi_shift = 9 - log;
 
 	return 0;
 }
 
 static void crypt_iv_benbi_dtr(struct crypt_config *cc)
 {
-	cc->iv_gen_private = NULL;
 }
 
 static int crypt_iv_benbi_gen(struct crypt_config *cc, u8 *iv, sector_t sector)
 {
+	__be64 val;
+
 	memset(iv, 0, cc->iv_size - sizeof(u64)); /* rest is cleared below */
-	put_unaligned(cpu_to_be64(((u64)sector << (u32)cc->iv_gen_private) + 1),
-		      (__be64 *)(iv + cc->iv_size - sizeof(u64)));
+
+	val = cpu_to_be64(((u64)sector << cc->iv_gen_private.benbi_shift) + 1);
+	put_unaligned(val, (__be64 *)(iv + cc->iv_size - sizeof(u64)));
+
+	return 0;
+}
+
+static int crypt_iv_null_gen(struct crypt_config *cc, u8 *iv, sector_t sector)
+{
+	memset(iv, 0, cc->iv_size);
 
 	return 0;
 }
@@ -267,165 +317,204 @@ static struct crypt_iv_operations crypt_iv_benbi_ops = {
 	.generator = crypt_iv_benbi_gen
 };
 
-static int
-crypt_convert_scatterlist(struct crypt_config *cc, struct scatterlist *out,
-                          struct scatterlist *in, unsigned int length,
-                          int write, sector_t sector)
+static struct crypt_iv_operations crypt_iv_null_ops = {
+	.generator = crypt_iv_null_gen
+};
+
+static void crypt_convert_init(struct crypt_config *cc,
+			       struct convert_context *ctx,
+			       struct bio *bio_out, struct bio *bio_in,
+			       sector_t sector)
 {
-	u8 iv[cc->iv_size];
-	struct blkcipher_desc desc = {
-		.tfm = cc->tfm,
-		.info = iv,
-		.flags = CRYPTO_TFM_REQ_MAY_SLEEP,
-	};
-	int r;
+	ctx->bio_in = bio_in;
+	ctx->bio_out = bio_out;
+	ctx->offset_in = 0;
+	ctx->offset_out = 0;
+	ctx->idx_in = bio_in ? bio_in->bi_idx : 0;
+	ctx->idx_out = bio_out ? bio_out->bi_idx : 0;
+	ctx->sector = sector + cc->iv_offset;
+	init_completion(&ctx->restart);
+}
+
+static struct dm_crypt_request *dmreq_of_req(struct crypt_config *cc,
+					     struct ablkcipher_request *req)
+{
+	return (struct dm_crypt_request *)((char *)req + cc->dmreq_start);
+}
+
+static struct ablkcipher_request *req_of_dmreq(struct crypt_config *cc,
+					       struct dm_crypt_request *dmreq)
+{
+	return (struct ablkcipher_request *)((char *)dmreq - cc->dmreq_start);
+}
+
+static int crypt_convert_block(struct crypt_config *cc,
+			       struct convert_context *ctx,
+			       struct ablkcipher_request *req)
+{
+	struct bio_vec *bv_in = bio_iovec_idx(ctx->bio_in, ctx->idx_in);
+	struct bio_vec *bv_out = bio_iovec_idx(ctx->bio_out, ctx->idx_out);
+	struct dm_crypt_request *dmreq;
+	u8 *iv;
+	int r = 0;
+
+	dmreq = dmreq_of_req(cc, req);
+	iv = (u8 *)ALIGN((unsigned long)(dmreq + 1),
+			 crypto_ablkcipher_alignmask(cc->tfm) + 1);
+
+	dmreq->ctx = ctx;
+	sg_init_table(&dmreq->sg_in, 1);
+	sg_set_page(&dmreq->sg_in, bv_in->bv_page, 1 << SECTOR_SHIFT,
+		    bv_in->bv_offset + ctx->offset_in);
+
+	sg_init_table(&dmreq->sg_out, 1);
+	sg_set_page(&dmreq->sg_out, bv_out->bv_page, 1 << SECTOR_SHIFT,
+		    bv_out->bv_offset + ctx->offset_out);
+
+	ctx->offset_in += 1 << SECTOR_SHIFT;
+	if (ctx->offset_in >= bv_in->bv_len) {
+		ctx->offset_in = 0;
+		ctx->idx_in++;
+	}
+
+	ctx->offset_out += 1 << SECTOR_SHIFT;
+	if (ctx->offset_out >= bv_out->bv_len) {
+		ctx->offset_out = 0;
+		ctx->idx_out++;
+	}
 
 	if (cc->iv_gen_ops) {
-		r = cc->iv_gen_ops->generator(cc, iv, sector);
+		r = cc->iv_gen_ops->generator(cc, iv, ctx->sector);
 		if (r < 0)
 			return r;
-
-		if (write)
-			r = crypto_blkcipher_encrypt_iv(&desc, out, in, length);
-		else
-			r = crypto_blkcipher_decrypt_iv(&desc, out, in, length);
-	} else {
-		if (write)
-			r = crypto_blkcipher_encrypt(&desc, out, in, length);
-		else
-			r = crypto_blkcipher_decrypt(&desc, out, in, length);
 	}
 
+	ablkcipher_request_set_crypt(req, &dmreq->sg_in, &dmreq->sg_out,
+				     1 << SECTOR_SHIFT, iv);
+
+	if (bio_data_dir(ctx->bio_in) == WRITE)
+		r = crypto_ablkcipher_encrypt(req);
+	else
+		r = crypto_ablkcipher_decrypt(req);
+
 	return r;
 }
 
-static void
-crypt_convert_init(struct crypt_config *cc, struct convert_context *ctx,
-                   struct bio *bio_out, struct bio *bio_in,
-                   sector_t sector, int write)
+static void kcryptd_async_done(struct crypto_async_request *async_req,
+			       int error);
+static void crypt_alloc_req(struct crypt_config *cc,
+			    struct convert_context *ctx)
 {
-	ctx->bio_in = bio_in;
-	ctx->bio_out = bio_out;
-	ctx->offset_in = 0;
-	ctx->offset_out = 0;
-	ctx->idx_in = bio_in ? bio_in->bi_idx : 0;
-	ctx->idx_out = bio_out ? bio_out->bi_idx : 0;
-	ctx->sector = sector + cc->iv_offset;
-	ctx->write = write;
+	if (!cc->req)
+		cc->req = mempool_alloc(cc->req_pool, GFP_NOIO);
+	ablkcipher_request_set_tfm(cc->req, cc->tfm);
+	ablkcipher_request_set_callback(cc->req, CRYPTO_TFM_REQ_MAY_BACKLOG |
+					CRYPTO_TFM_REQ_MAY_SLEEP,
+					kcryptd_async_done,
+					dmreq_of_req(cc, cc->req));
 }
 
 /*
  * Encrypt / decrypt data from one bio to another one (can be the same one)
  */
 static int crypt_convert(struct crypt_config *cc,
-                         struct convert_context *ctx)
+			 struct convert_context *ctx)
 {
-	int r = 0;
+	int r;
+
+	atomic_set(&ctx->pending, 1);
 
 	while(ctx->idx_in < ctx->bio_in->bi_vcnt &&
 	      ctx->idx_out < ctx->bio_out->bi_vcnt) {
-		struct bio_vec *bv_in = bio_iovec_idx(ctx->bio_in, ctx->idx_in);
-		struct bio_vec *bv_out = bio_iovec_idx(ctx->bio_out, ctx->idx_out);
-		struct scatterlist sg_in = {
-			.page = bv_in->bv_page,
-			.offset = bv_in->bv_offset + ctx->offset_in,
-			.length = 1 << SECTOR_SHIFT
-		};
-		struct scatterlist sg_out = {
-			.page = bv_out->bv_page,
-			.offset = bv_out->bv_offset + ctx->offset_out,
-			.length = 1 << SECTOR_SHIFT
-		};
-
-		ctx->offset_in += sg_in.length;
-		if (ctx->offset_in >= bv_in->bv_len) {
-			ctx->offset_in = 0;
-			ctx->idx_in++;
-		}
 
-		ctx->offset_out += sg_out.length;
-		if (ctx->offset_out >= bv_out->bv_len) {
-			ctx->offset_out = 0;
-			ctx->idx_out++;
+		crypt_alloc_req(cc, ctx);
+
+		atomic_inc(&ctx->pending);
+
+		r = crypt_convert_block(cc, ctx, cc->req);
+
+		switch (r) {
+		/* async */
+		case -EBUSY:
+			wait_for_completion(&ctx->restart);
+			INIT_COMPLETION(ctx->restart);
+			/* fall through*/
+		case -EINPROGRESS:
+			cc->req = NULL;
+			ctx->sector++;
+			continue;
+
+		/* sync */
+		case 0:
+			atomic_dec(&ctx->pending);
+			ctx->sector++;
+			cond_resched();
+			continue;
+
+		/* error */
+		default:
+			atomic_dec(&ctx->pending);
+			return r;
 		}
-
-		r = crypt_convert_scatterlist(cc, &sg_out, &sg_in, sg_in.length,
-		                              ctx->write, ctx->sector);
-		if (r < 0)
-			break;
-
-		ctx->sector++;
 	}
 
-	return r;
+	return 0;
 }
 
- static void dm_crypt_bio_destructor(struct bio *bio)
- {
-	struct crypt_io *io = bio->bi_private;
+static void dm_crypt_bio_destructor(struct bio *bio)
+{
+	struct dm_crypt_io *io = bio->bi_private;
 	struct crypt_config *cc = io->target->private;
 
 	bio_free(bio, cc->bs);
- }
+}
 
 /*
  * Generate a new unfragmented bio with the given size
  * This should never violate the device limitations
- * May return a smaller bio when running out of pages
+ * May return a smaller bio when running out of pages, indicated by
+ * *out_of_pages set to 1.
  */
-static struct bio *
-crypt_alloc_buffer(struct crypt_config *cc, unsigned int size,
-                   struct bio *base_bio, unsigned int *bio_vec_idx)
+static struct bio *crypt_alloc_buffer(struct dm_crypt_io *io, unsigned size,
+				      unsigned *out_of_pages)
 {
+	struct crypt_config *cc = io->target->private;
 	struct bio *clone;
 	unsigned int nr_iovecs = (size + PAGE_SIZE - 1) >> PAGE_SHIFT;
 	gfp_t gfp_mask = GFP_NOIO | __GFP_HIGHMEM;
-	unsigned int i;
-
-	if (base_bio) {
-		clone = bio_alloc_bioset(GFP_NOIO, base_bio->bi_max_vecs, cc->bs);
-		__bio_clone(clone, base_bio);
-	} else
-		clone = bio_alloc_bioset(GFP_NOIO, nr_iovecs, cc->bs);
+	unsigned i, len;
+	struct page *page;
 
+	clone = bio_alloc_bioset(GFP_NOIO, nr_iovecs, cc->bs);
 	if (!clone)
 		return NULL;
 
-	clone->bi_destructor = dm_crypt_bio_destructor;
-
-	/* if the last bio was not complete, continue where that one ended */
-	clone->bi_idx = *bio_vec_idx;
-	clone->bi_vcnt = *bio_vec_idx;
-	clone->bi_size = 0;
-	clone->bi_flags &= ~(1 << BIO_SEG_VALID);
-
-	/* clone->bi_idx pages have already been allocated */
-	size -= clone->bi_idx * PAGE_SIZE;
-
-	for (i = clone->bi_idx; i < nr_iovecs; i++) {
-		struct bio_vec *bv = bio_iovec_idx(clone, i);
+	clone_init(io, clone);
+	*out_of_pages = 0;
 
-		bv->bv_page = mempool_alloc(cc->page_pool, gfp_mask);
-		if (!bv->bv_page)
+	for (i = 0; i < nr_iovecs; i++) {
+		page = mempool_alloc(cc->page_pool, gfp_mask);
+		if (!page) {
+			*out_of_pages = 1;
 			break;
+		}
 
 		/*
 		 * if additional pages cannot be allocated without waiting,
 		 * return a partially allocated bio, the caller will then try
 		 * to allocate additional bios while submitting this partial bio
 		 */
-		if ((i - clone->bi_idx) == (MIN_BIO_PAGES - 1))
+		if (i == (MIN_BIO_PAGES - 1))
 			gfp_mask = (gfp_mask | __GFP_NOWARN) & ~__GFP_WAIT;
 
-		bv->bv_offset = 0;
-		if (size > PAGE_SIZE)
-			bv->bv_len = PAGE_SIZE;
-		else
-			bv->bv_len = size;
+		len = (size > PAGE_SIZE) ? PAGE_SIZE : size;
+
+		if (!bio_add_page(clone, page, len, 0)) {
+			mempool_free(page, cc->page_pool);
+			break;
+		}
 
-		clone->bi_size += bv->bv_len;
-		clone->bi_vcnt++;
-		size -= bv->bv_len;
+		size -= len;
 	}
 
 	if (!clone->bi_size) {
@@ -433,42 +522,15 @@ crypt_alloc_buffer(struct crypt_config *cc, unsigned int size,
 		return NULL;
 	}
 
-	/*
-	 * Remember the last bio_vec allocated to be able
-	 * to correctly continue after the splitting.
-	 */
-	*bio_vec_idx = clone->bi_vcnt;
-
 	return clone;
 }
 
-static void crypt_free_buffer_pages(struct crypt_config *cc,
-                                    struct bio *clone, unsigned int bytes)
+static void crypt_free_buffer_pages(struct crypt_config *cc, struct bio *clone)
 {
-	unsigned int i, start, end;
+	unsigned int i;
 	struct bio_vec *bv;
 
-	/*
-	 * This is ugly, but Jens Axboe thinks that using bi_idx in the
-	 * endio function is too dangerous at the moment, so I calculate the
-	 * correct position using bi_vcnt and bi_size.
-	 * The bv_offset and bv_len fields might already be modified but we
-	 * know that we always allocated whole pages.
-	 * A fix to the bi_idx issue in the kernel is in the works, so
-	 * we will hopefully be able to revert to the cleaner solution soon.
-	 */
-	i = clone->bi_vcnt - 1;
-	bv = bio_iovec_idx(clone, i);
-	end = (i << PAGE_SHIFT) + (bv->bv_offset + bv->bv_len) - clone->bi_size;
-	start = end - bytes;
-
-	start >>= PAGE_SHIFT;
-	if (!clone->bi_size)
-		end = clone->bi_vcnt;
-	else
-		end >>= PAGE_SHIFT;
-
-	for (i = start; i < end; i++) {
+	for (i = 0; i < clone->bi_vcnt; i++) {
 		bv = bio_iovec_idx(clone, i);
 		BUG_ON(!bv->bv_page);
 		mempool_free(bv->bv_page, cc->page_pool);
@@ -476,80 +538,97 @@ static void crypt_free_buffer_pages(struct crypt_config *cc,
 	}
 }
 
+static struct dm_crypt_io *crypt_io_alloc(struct dm_target *ti,
+					  struct bio *bio, sector_t sector)
+{
+	struct crypt_config *cc = ti->private;
+	struct dm_crypt_io *io;
+
+	io = mempool_alloc(cc->io_pool, GFP_NOIO);
+	io->target = ti;
+	io->base_bio = bio;
+	io->sector = sector;
+	io->error = 0;
+	io->base_io = NULL;
+	atomic_set(&io->pending, 0);
+
+	return io;
+}
+
+static void crypt_inc_pending(struct dm_crypt_io *io)
+{
+	atomic_inc(&io->pending);
+}
+
 /*
  * One of the bios was finished. Check for completion of
  * the whole request and correctly clean up the buffer.
+ * If base_io is set, wait for the last fragment to complete.
  */
-static void dec_pending(struct crypt_io *io, int error)
+static void crypt_dec_pending(struct dm_crypt_io *io)
 {
-	struct crypt_config *cc = (struct crypt_config *) io->target->private;
-
-	if (error < 0)
-		io->error = error;
+	struct crypt_config *cc = io->target->private;
+	struct bio *base_bio = io->base_bio;
+	struct dm_crypt_io *base_io = io->base_io;
+	int error = io->error;
 
 	if (!atomic_dec_and_test(&io->pending))
 		return;
 
-	if (io->first_clone)
-		bio_put(io->first_clone);
-
-	bio_endio(io->base_bio, io->base_bio->bi_size, io->error);
-
 	mempool_free(io, cc->io_pool);
+
+	if (likely(!base_io))
+		bio_endio(base_bio, error);
+	else {
+		if (error && !base_io->error)
+			base_io->error = error;
+		crypt_dec_pending(base_io);
+	}
 }
 
 /*
- * kcryptd:
+ * kcryptd/kcryptd_io:
  *
  * Needed because it would be very unwise to do decryption in an
  * interrupt context.
+ *
+ * kcryptd performs the actual encryption or decryption.
+ *
+ * kcryptd_io performs the IO submission.
+ *
+ * They must be separated as otherwise the final stages could be
+ * starved by new requests which can block in the first stages due
+ * to memory allocation.
  */
-static struct workqueue_struct *_kcryptd_workqueue;
-static void kcryptd_do_work(struct work_struct *work);
-
-static void kcryptd_queue_io(struct crypt_io *io)
-{
-	INIT_WORK(&io->work, kcryptd_do_work);
-	queue_work(_kcryptd_workqueue, &io->work);
-}
-
-static int crypt_endio(struct bio *clone, unsigned int done, int error)
+static void crypt_endio(struct bio *clone, int error)
 {
-	struct crypt_io *io = clone->bi_private;
+	struct dm_crypt_io *io = clone->bi_private;
 	struct crypt_config *cc = io->target->private;
-	unsigned read_io = bio_data_dir(clone) == READ;
+	unsigned rw = bio_data_dir(clone);
+
+	if (unlikely(!bio_flagged(clone, BIO_UPTODATE) && !error))
+		error = -EIO;
 
 	/*
-	 * free the processed pages, even if
-	 * it's only a partially completed write
+	 * free the processed pages
 	 */
-	if (!read_io)
-		crypt_free_buffer_pages(cc, clone, done);
-
-	/* keep going - not finished yet */
-	if (unlikely(clone->bi_size))
-		return 1;
+	if (rw == WRITE)
+		crypt_free_buffer_pages(cc, clone);
 
-	if (!read_io)
-		goto out;
+	bio_put(clone);
 
-	if (unlikely(!bio_flagged(clone, BIO_UPTODATE))) {
-		error = -EIO;
-		goto out;
+	if (rw == READ && !error) {
+		kcryptd_queue_crypt(io);
+		return;
 	}
 
-	bio_put(clone);
-	io->post_process = 1;
-	kcryptd_queue_io(io);
-	return 0;
+	if (unlikely(error))
+		io->error = error;
 
-out:
-	bio_put(clone);
-	dec_pending(io, error);
-	return error;
+	crypt_dec_pending(io);
 }
 
-static void clone_init(struct crypt_io *io, struct bio *clone)
+static void clone_init(struct dm_crypt_io *io, struct bio *clone)
 {
 	struct crypt_config *cc = io->target->private;
 
@@ -557,16 +636,16 @@ static void clone_init(struct crypt_io *io, struct bio *clone)
 	clone->bi_end_io  = crypt_endio;
 	clone->bi_bdev    = cc->dev->bdev;
 	clone->bi_rw      = io->base_bio->bi_rw;
+	clone->bi_destructor = dm_crypt_bio_destructor;
 }
 
-static void process_read(struct crypt_io *io)
+static void kcryptd_io_read(struct dm_crypt_io *io)
 {
 	struct crypt_config *cc = io->target->private;
 	struct bio *base_bio = io->base_bio;
 	struct bio *clone;
-	sector_t sector = base_bio->bi_sector - io->target->begin;
 
-	atomic_inc(&io->pending);
+	crypt_inc_pending(io);
 
 	/*
 	 * The block layer might modify the bvec array, so always
@@ -575,106 +654,228 @@ static void process_read(struct crypt_io *io)
 	 */
 	clone = bio_alloc_bioset(GFP_NOIO, bio_segments(base_bio), cc->bs);
 	if (unlikely(!clone)) {
-		dec_pending(io, -ENOMEM);
+		io->error = -ENOMEM;
+		crypt_dec_pending(io);
 		return;
 	}
 
 	clone_init(io, clone);
-	clone->bi_destructor = dm_crypt_bio_destructor;
 	clone->bi_idx = 0;
 	clone->bi_vcnt = bio_segments(base_bio);
 	clone->bi_size = base_bio->bi_size;
-	clone->bi_sector = cc->start + sector;
+	clone->bi_sector = cc->start + io->sector;
 	memcpy(clone->bi_io_vec, bio_iovec(base_bio),
 	       sizeof(struct bio_vec) * clone->bi_vcnt);
 
 	generic_make_request(clone);
 }
 
-static void process_write(struct crypt_io *io)
+static void kcryptd_io_write(struct dm_crypt_io *io)
+{
+	struct bio *clone = io->ctx.bio_out;
+	generic_make_request(clone);
+}
+
+static void kcryptd_io(struct work_struct *work)
+{
+	struct dm_crypt_io *io = container_of(work, struct dm_crypt_io, work);
+
+	if (bio_data_dir(io->base_bio) == READ)
+		kcryptd_io_read(io);
+	else
+		kcryptd_io_write(io);
+}
+
+static void kcryptd_queue_io(struct dm_crypt_io *io)
 {
 	struct crypt_config *cc = io->target->private;
-	struct bio *base_bio = io->base_bio;
-	struct bio *clone;
-	struct convert_context ctx;
-	unsigned remaining = base_bio->bi_size;
-	sector_t sector = base_bio->bi_sector - io->target->begin;
-	unsigned bvec_idx = 0;
 
-	atomic_inc(&io->pending);
+	INIT_WORK(&io->work, kcryptd_io);
+	queue_work(cc->io_queue, &io->work);
+}
+
+static void kcryptd_crypt_write_io_submit(struct dm_crypt_io *io,
+					  int error, int async)
+{
+	struct bio *clone = io->ctx.bio_out;
+	struct crypt_config *cc = io->target->private;
+
+	if (unlikely(error < 0)) {
+		crypt_free_buffer_pages(cc, clone);
+		bio_put(clone);
+		io->error = -EIO;
+		crypt_dec_pending(io);
+		return;
+	}
+
+	/* crypt_convert should have filled the clone bio */
+	BUG_ON(io->ctx.idx_out < clone->bi_vcnt);
+
+	clone->bi_sector = cc->start + io->sector;
+
+	if (async)
+		kcryptd_queue_io(io);
+	else
+		generic_make_request(clone);
+}
+
+static void kcryptd_crypt_write_convert(struct dm_crypt_io *io)
+{
+	struct crypt_config *cc = io->target->private;
+	struct bio *clone;
+	struct dm_crypt_io *new_io;
+	int crypt_finished;
+	unsigned out_of_pages = 0;
+	unsigned remaining = io->base_bio->bi_size;
+	sector_t sector = io->sector;
+	int r;
 
-	crypt_convert_init(cc, &ctx, NULL, base_bio, sector, 1);
+	/*
+	 * Prevent io from disappearing until this function completes.
+	 */
+	crypt_inc_pending(io);
+	crypt_convert_init(cc, &io->ctx, NULL, io->base_bio, sector);
 
 	/*
 	 * The allocated buffers can be smaller than the whole bio,
 	 * so repeat the whole process until all the data can be handled.
 	 */
 	while (remaining) {
-		clone = crypt_alloc_buffer(cc, base_bio->bi_size,
-					   io->first_clone, &bvec_idx);
+		clone = crypt_alloc_buffer(io, remaining, &out_of_pages);
 		if (unlikely(!clone)) {
-			dec_pending(io, -ENOMEM);
-			return;
+			io->error = -ENOMEM;
+			break;
 		}
 
-		ctx.bio_out = clone;
+		io->ctx.bio_out = clone;
+		io->ctx.idx_out = 0;
+
+		remaining -= clone->bi_size;
+		sector += bio_sectors(clone);
+
+		crypt_inc_pending(io);
+		r = crypt_convert(cc, &io->ctx);
+		crypt_finished = atomic_dec_and_test(&io->ctx.pending);
 
-		if (unlikely(crypt_convert(cc, &ctx) < 0)) {
-			crypt_free_buffer_pages(cc, clone, clone->bi_size);
-			bio_put(clone);
-			dec_pending(io, -EIO);
-			return;
+		/* Encryption was already finished, submit io now */
+		if (crypt_finished) {
+			kcryptd_crypt_write_io_submit(io, r, 0);
+
+			/*
+			 * If there was an error, do not try next fragments.
+			 * For async, error is processed in async handler.
+			 */
+			if (unlikely(r < 0))
+				break;
+
+			io->sector = sector;
 		}
 
-		clone_init(io, clone);
-		clone->bi_sector = cc->start + sector;
+		/*
+		 * Out of memory -> run queues
+		 * But don't wait if split was due to the io size restriction
+		 */
+		if (unlikely(out_of_pages))
+			congestion_wait(BLK_RW_ASYNC, HZ/100);
+
+		/*
+		 * With async crypto it is unsafe to share the crypto context
+		 * between fragments, so switch to a new dm_crypt_io structure.
+		 */
+		if (unlikely(!crypt_finished && remaining)) {
+			new_io = crypt_io_alloc(io->target, io->base_bio,
+						sector);
+			crypt_inc_pending(new_io);
+			crypt_convert_init(cc, &new_io->ctx, NULL,
+					   io->base_bio, sector);
+			new_io->ctx.idx_in = io->ctx.idx_in;
+			new_io->ctx.offset_in = io->ctx.offset_in;
 
-		if (!io->first_clone) {
 			/*
-			 * hold a reference to the first clone, because it
-			 * holds the bio_vec array and that can't be freed
-			 * before all other clones are released
+			 * Fragments after the first use the base_io
+			 * pending count.
 			 */
-			bio_get(clone);
-			io->first_clone = clone;
+			if (!io->base_io)
+				new_io->base_io = io;
+			else {
+				new_io->base_io = io->base_io;
+				crypt_inc_pending(io->base_io);
+				crypt_dec_pending(io);
+			}
+
+			io = new_io;
 		}
+	}
 
-		remaining -= clone->bi_size;
-		sector += bio_sectors(clone);
+	crypt_dec_pending(io);
+}
 
-		/* prevent bio_put of first_clone */
-		if (remaining)
-			atomic_inc(&io->pending);
+static void kcryptd_crypt_read_done(struct dm_crypt_io *io, int error)
+{
+	if (unlikely(error < 0))
+		io->error = -EIO;
 
-		generic_make_request(clone);
+	crypt_dec_pending(io);
+}
 
-		/* out of memory -> run queues */
-		if (remaining)
-			congestion_wait(bio_data_dir(clone), HZ/100);
-	}
+static void kcryptd_crypt_read_convert(struct dm_crypt_io *io)
+{
+	struct crypt_config *cc = io->target->private;
+	int r = 0;
+
+	crypt_inc_pending(io);
+
+	crypt_convert_init(cc, &io->ctx, io->base_bio, io->base_bio,
+			   io->sector);
+
+	r = crypt_convert(cc, &io->ctx);
+
+	if (atomic_dec_and_test(&io->ctx.pending))
+		kcryptd_crypt_read_done(io, r);
+
+	crypt_dec_pending(io);
 }
 
-static void process_read_endio(struct crypt_io *io)
+static void kcryptd_async_done(struct crypto_async_request *async_req,
+			       int error)
 {
+	struct dm_crypt_request *dmreq = async_req->data;
+	struct convert_context *ctx = dmreq->ctx;
+	struct dm_crypt_io *io = container_of(ctx, struct dm_crypt_io, ctx);
 	struct crypt_config *cc = io->target->private;
-	struct convert_context ctx;
 
-	crypt_convert_init(cc, &ctx, io->base_bio, io->base_bio,
-			   io->base_bio->bi_sector - io->target->begin, 0);
+	if (error == -EINPROGRESS) {
+		complete(&ctx->restart);
+		return;
+	}
+
+	mempool_free(req_of_dmreq(cc, dmreq), cc->req_pool);
+
+	if (!atomic_dec_and_test(&ctx->pending))
+		return;
 
-	dec_pending(io, crypt_convert(cc, &ctx));
+	if (bio_data_dir(io->base_bio) == READ)
+		kcryptd_crypt_read_done(io, error);
+	else
+		kcryptd_crypt_write_io_submit(io, error, 1);
 }
 
-static void kcryptd_do_work(struct work_struct *work)
+static void kcryptd_crypt(struct work_struct *work)
 {
-	struct crypt_io *io = container_of(work, struct crypt_io, work);
+	struct dm_crypt_io *io = container_of(work, struct dm_crypt_io, work);
 
-	if (io->post_process)
-		process_read_endio(io);
-	else if (bio_data_dir(io->base_bio) == READ)
-		process_read(io);
+	if (bio_data_dir(io->base_bio) == READ)
+		kcryptd_crypt_read_convert(io);
 	else
-		process_write(io);
+		kcryptd_crypt_write_convert(io);
+}
+
+static void kcryptd_queue_crypt(struct dm_crypt_io *io)
+{
+	struct crypt_config *cc = io->target->private;
+
+	INIT_WORK(&io->work, kcryptd_crypt);
+	queue_work(cc->crypt_queue, &io->work);
 }
 
 /*
@@ -728,7 +929,7 @@ static int crypt_set_key(struct crypt_config *cc, char *key)
 	cc->key_size = key_size; /* initial settings */
 
 	if ((!key_size && strcmp(key, "-")) ||
-	    (key_size && crypt_decode_key(cc->key, key, key_size) < 0))
+	   (key_size && crypt_decode_key(cc->key, key, key_size) < 0))
 		return -EINVAL;
 
 	set_bit(DM_CRYPT_KEY_VALID, &cc->flags);
@@ -750,7 +951,7 @@ static int crypt_wipe_key(struct crypt_config *cc)
 static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 {
 	struct crypt_config *cc;
-	struct crypto_blkcipher *tfm;
+	struct crypto_ablkcipher *tfm;
 	char *tmp;
 	char *cipher;
 	char *chainmode;
@@ -784,10 +985,10 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 
  	if (crypt_set_key(cc, argv[1])) {
 		ti->error = "Error decoding key";
-		goto bad1;
+		goto bad_cipher;
 	}
 
-	/* Compatiblity mode for old dm-crypt cipher strings */
+	/* Compatibility mode for old dm-crypt cipher strings */
 	if (!chainmode || (strcmp(chainmode, "plain") == 0 && !ivmode)) {
 		chainmode = "cbc";
 		ivmode = "plain";
@@ -795,19 +996,19 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 
 	if (strcmp(chainmode, "ecb") && !ivmode) {
 		ti->error = "This chaining mode requires an IV mechanism";
-		goto bad1;
+		goto bad_cipher;
 	}
 
-	if (snprintf(cc->cipher, CRYPTO_MAX_ALG_NAME, "%s(%s)", chainmode, 
-		     cipher) >= CRYPTO_MAX_ALG_NAME) {
+	if (snprintf(cc->cipher, CRYPTO_MAX_ALG_NAME, "%s(%s)",
+		     chainmode, cipher) >= CRYPTO_MAX_ALG_NAME) {
 		ti->error = "Chain mode + cipher name is too long";
-		goto bad1;
+		goto bad_cipher;
 	}
 
-	tfm = crypto_alloc_blkcipher(cc->cipher, 0, CRYPTO_ALG_ASYNC);
+	tfm = crypto_alloc_ablkcipher(cc->cipher, 0, 0);
 	if (IS_ERR(tfm)) {
 		ti->error = "Error allocating crypto tfm";
-		goto bad1;
+		goto bad_cipher;
 	}
 
 	strcpy(cc->cipher, cipher);
@@ -827,20 +1028,22 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 		cc->iv_gen_ops = &crypt_iv_essiv_ops;
 	else if (strcmp(ivmode, "benbi") == 0)
 		cc->iv_gen_ops = &crypt_iv_benbi_ops;
+	else if (strcmp(ivmode, "null") == 0)
+		cc->iv_gen_ops = &crypt_iv_null_ops;
 	else {
 		ti->error = "Invalid IV mode";
-		goto bad2;
+		goto bad_ivmode;
 	}
 
 	if (cc->iv_gen_ops && cc->iv_gen_ops->ctr &&
 	    cc->iv_gen_ops->ctr(cc, ti, ivopts) < 0)
-		goto bad2;
+		goto bad_ivmode;
 
-	cc->iv_size = crypto_blkcipher_ivsize(tfm);
+	cc->iv_size = crypto_ablkcipher_ivsize(tfm);
 	if (cc->iv_size)
 		/* at least a 64 bit sector number should fit in our buffer */
 		cc->iv_size = max(cc->iv_size,
-		                  (unsigned int)(sizeof(u64) / sizeof(u8)));
+				  (unsigned int)(sizeof(u64) / sizeof(u8)));
 	else {
 		if (cc->iv_gen_ops) {
 			DMWARN("Selected cipher does not support IVs");
@@ -853,42 +1056,56 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	cc->io_pool = mempool_create_slab_pool(MIN_IOS, _crypt_io_pool);
 	if (!cc->io_pool) {
 		ti->error = "Cannot allocate crypt io mempool";
-		goto bad3;
+		goto bad_slab_pool;
+	}
+
+	cc->dmreq_start = sizeof(struct ablkcipher_request);
+	cc->dmreq_start += crypto_ablkcipher_reqsize(tfm);
+	cc->dmreq_start = ALIGN(cc->dmreq_start, crypto_tfm_ctx_alignment());
+	cc->dmreq_start += crypto_ablkcipher_alignmask(tfm) &
+			   ~(crypto_tfm_ctx_alignment() - 1);
+
+	cc->req_pool = mempool_create_kmalloc_pool(MIN_IOS, cc->dmreq_start +
+			sizeof(struct dm_crypt_request) + cc->iv_size);
+	if (!cc->req_pool) {
+		ti->error = "Cannot allocate crypt request mempool";
+		goto bad_req_pool;
 	}
+	cc->req = NULL;
 
 	cc->page_pool = mempool_create_page_pool(MIN_POOL_PAGES, 0);
 	if (!cc->page_pool) {
 		ti->error = "Cannot allocate page mempool";
-		goto bad4;
+		goto bad_page_pool;
 	}
 
-	cc->bs = bioset_create(MIN_IOS, MIN_IOS, 4);
+	cc->bs = bioset_create(MIN_IOS, 0);
 	if (!cc->bs) {
 		ti->error = "Cannot allocate crypt bioset";
 		goto bad_bs;
 	}
 
-	if (crypto_blkcipher_setkey(tfm, cc->key, key_size) < 0) {
+	if (crypto_ablkcipher_setkey(tfm, cc->key, key_size) < 0) {
 		ti->error = "Error setting key";
-		goto bad5;
+		goto bad_device;
 	}
 
 	if (sscanf(argv[2], "%llu", &tmpll) != 1) {
 		ti->error = "Invalid iv_offset sector";
-		goto bad5;
+		goto bad_device;
 	}
 	cc->iv_offset = tmpll;
 
 	if (sscanf(argv[4], "%llu", &tmpll) != 1) {
 		ti->error = "Invalid device sector";
-		goto bad5;
+		goto bad_device;
 	}
 	cc->start = tmpll;
 
 	if (dm_get_device(ti, argv[3], cc->start, ti->len,
-	                  dm_table_get_mode(ti->table), &cc->dev)) {
+			  dm_table_get_mode(ti->table), &cc->dev)) {
 		ti->error = "Device lookup failed";
-		goto bad5;
+		goto bad_device;
 	}
 
 	if (ivmode && cc->iv_gen_ops) {
@@ -897,30 +1114,50 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 		cc->iv_mode = kmalloc(strlen(ivmode) + 1, GFP_KERNEL);
 		if (!cc->iv_mode) {
 			ti->error = "Error kmallocing iv_mode string";
-			goto bad5;
+			goto bad_ivmode_string;
 		}
 		strcpy(cc->iv_mode, ivmode);
 	} else
 		cc->iv_mode = NULL;
 
+	cc->io_queue = create_singlethread_workqueue("kcryptd_io");
+	if (!cc->io_queue) {
+		ti->error = "Couldn't create kcryptd io queue";
+		goto bad_io_queue;
+	}
+
+	cc->crypt_queue = create_singlethread_workqueue("kcryptd");
+	if (!cc->crypt_queue) {
+		ti->error = "Couldn't create kcryptd queue";
+		goto bad_crypt_queue;
+	}
+
+	ti->num_flush_requests = 1;
 	ti->private = cc;
 	return 0;
 
-bad5:
+bad_crypt_queue:
+	destroy_workqueue(cc->io_queue);
+bad_io_queue:
+	kfree(cc->iv_mode);
+bad_ivmode_string:
+	dm_put_device(ti, cc->dev);
+bad_device:
 	bioset_free(cc->bs);
 bad_bs:
 	mempool_destroy(cc->page_pool);
-bad4:
+bad_page_pool:
+	mempool_destroy(cc->req_pool);
+bad_req_pool:
 	mempool_destroy(cc->io_pool);
-bad3:
+bad_slab_pool:
 	if (cc->iv_gen_ops && cc->iv_gen_ops->dtr)
 		cc->iv_gen_ops->dtr(cc);
-bad2:
-	crypto_free_blkcipher(tfm);
-bad1:
+bad_ivmode:
+	crypto_free_ablkcipher(tfm);
+bad_cipher:
 	/* Must zero key material before freeing */
-	memset(cc, 0, sizeof(*cc) + cc->key_size * sizeof(u8));
-	kfree(cc);
+	kzfree(cc);
 	return -EINVAL;
 }
 
@@ -928,36 +1165,47 @@ static void crypt_dtr(struct dm_target *ti)
 {
 	struct crypt_config *cc = (struct crypt_config *) ti->private;
 
+	destroy_workqueue(cc->io_queue);
+	destroy_workqueue(cc->crypt_queue);
+
+	if (cc->req)
+		mempool_free(cc->req, cc->req_pool);
+
 	bioset_free(cc->bs);
 	mempool_destroy(cc->page_pool);
+	mempool_destroy(cc->req_pool);
 	mempool_destroy(cc->io_pool);
 
 	kfree(cc->iv_mode);
 	if (cc->iv_gen_ops && cc->iv_gen_ops->dtr)
 		cc->iv_gen_ops->dtr(cc);
-	crypto_free_blkcipher(cc->tfm);
+	crypto_free_ablkcipher(cc->tfm);
 	dm_put_device(ti, cc->dev);
 
 	/* Must zero key material before freeing */
-	memset(cc, 0, sizeof(*cc) + cc->key_size * sizeof(u8));
-	kfree(cc);
+	kzfree(cc);
 }
 
 static int crypt_map(struct dm_target *ti, struct bio *bio,
 		     union map_info *map_context)
 {
-	struct crypt_config *cc = ti->private;
-	struct crypt_io *io;
+	struct dm_crypt_io *io;
+	struct crypt_config *cc;
 
-	io = mempool_alloc(cc->io_pool, GFP_NOIO);
-	io->target = ti;
-	io->base_bio = bio;
-	io->first_clone = NULL;
-	io->error = io->post_process = 0;
-	atomic_set(&io->pending, 0);
-	kcryptd_queue_io(io);
+	if (unlikely(bio_empty_barrier(bio))) {
+		cc = ti->private;
+		bio->bi_bdev = cc->dev->bdev;
+		return DM_MAPIO_REMAPPED;
+	}
 
-	return 0;
+	io = crypt_io_alloc(ti, bio, bio->bi_sector - ti->begin);
+
+	if (bio_data_dir(io->base_bio) == READ)
+		kcryptd_queue_io(io);
+	else
+		kcryptd_queue_crypt(io);
+
+	return DM_MAPIO_SUBMITTED;
 }
 
 static int crypt_status(struct dm_target *ti, status_type_t type,
@@ -1050,9 +1298,32 @@ error:
 	return -EINVAL;
 }
 
+static int crypt_merge(struct dm_target *ti, struct bvec_merge_data *bvm,
+		       struct bio_vec *biovec, int max_size)
+{
+	struct crypt_config *cc = ti->private;
+	struct request_queue *q = bdev_get_queue(cc->dev->bdev);
+
+	if (!q->merge_bvec_fn)
+		return max_size;
+
+	bvm->bi_bdev = cc->dev->bdev;
+	bvm->bi_sector = cc->start + bvm->bi_sector - ti->begin;
+
+	return min(max_size, q->merge_bvec_fn(q, bvm, biovec));
+}
+
+static int crypt_iterate_devices(struct dm_target *ti,
+				 iterate_devices_callout_fn fn, void *data)
+{
+	struct crypt_config *cc = ti->private;
+
+	return fn(ti, cc->dev, cc->start, ti->len, data);
+}
+
 static struct target_type crypt_target = {
 	.name   = "crypt",
-	.version= {1, 3, 0},
+	.version = {1, 7, 0},
 	.module = THIS_MODULE,
 	.ctr    = crypt_ctr,
 	.dtr    = crypt_dtr,
@@ -1062,48 +1333,30 @@ static struct target_type crypt_target = {
 	.preresume = crypt_preresume,
 	.resume = crypt_resume,
 	.message = crypt_message,
+	.merge  = crypt_merge,
+	.iterate_devices = crypt_iterate_devices,
 };
 
 static int __init dm_crypt_init(void)
 {
 	int r;
 
-	_crypt_io_pool = kmem_cache_create("dm-crypt_io",
-	                                   sizeof(struct crypt_io),
-	                                   0, 0, NULL, NULL);
+	_crypt_io_pool = KMEM_CACHE(dm_crypt_io, 0);
 	if (!_crypt_io_pool)
 		return -ENOMEM;
 
-	_kcryptd_workqueue = create_workqueue("kcryptd");
-	if (!_kcryptd_workqueue) {
-		r = -ENOMEM;
-		DMERR("couldn't create kcryptd");
-		goto bad1;
-	}
-
 	r = dm_register_target(&crypt_target);
 	if (r < 0) {
 		DMERR("register failed %d", r);
-		goto bad2;
+		kmem_cache_destroy(_crypt_io_pool);
 	}
 
-	return 0;
-
-bad2:
-	destroy_workqueue(_kcryptd_workqueue);
-bad1:
-	kmem_cache_destroy(_crypt_io_pool);
 	return r;
 }
 
 static void __exit dm_crypt_exit(void)
 {
-	int r = dm_unregister_target(&crypt_target);
-
-	if (r < 0)
-		DMERR("unregister failed %d", r);
-
-	destroy_workqueue(_kcryptd_workqueue);
+	dm_unregister_target(&crypt_target);
 	kmem_cache_destroy(_crypt_io_pool);
 }