Merge master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6

68 files changed, 1022 insertions, 747 deletions

diff --git a/arch/i386/crypto/aes-i586-asm.S b/arch/i386/crypto/aes-i586-asm.S
index 7b73c67cb4e..911b15377f2 100644
--- a/arch/i386/crypto/aes-i586-asm.S
+++ b/arch/i386/crypto/aes-i586-asm.S
@@ -255,18 +255,17 @@ aes_enc_blk:
 	xor     8(%ebp),%r4
 	xor     12(%ebp),%r5
 
-	sub     $8,%esp           // space for register saves on stack
-	add     $16,%ebp          // increment to next round key
-	sub     $10,%r3          
-	je      4f              // 10 rounds for 128-bit key
-	add     $32,%ebp
-	sub     $2,%r3
-	je      3f              // 12 rounds for 128-bit key
-	add     $32,%ebp
-
-2:	fwd_rnd1( -64(%ebp) ,ft_tab)	// 14 rounds for 128-bit key
+	sub     $8,%esp		// space for register saves on stack
+	add     $16,%ebp	// increment to next round key
+	cmp     $12,%r3
+	jb      4f		// 10 rounds for 128-bit key
+	lea     32(%ebp),%ebp
+	je      3f		// 12 rounds for 192-bit key
+	lea     32(%ebp),%ebp
+
+2:	fwd_rnd1( -64(%ebp) ,ft_tab)	// 14 rounds for 256-bit key
 	fwd_rnd2( -48(%ebp) ,ft_tab)
-3:	fwd_rnd1( -32(%ebp) ,ft_tab)	// 12 rounds for 128-bit key
+3:	fwd_rnd1( -32(%ebp) ,ft_tab)	// 12 rounds for 192-bit key
 	fwd_rnd2( -16(%ebp) ,ft_tab)
 4:	fwd_rnd1(    (%ebp) ,ft_tab)	// 10 rounds for 128-bit key
 	fwd_rnd2( +16(%ebp) ,ft_tab)
@@ -334,18 +333,17 @@ aes_dec_blk:
 	xor     8(%ebp),%r4
 	xor     12(%ebp),%r5
 
-	sub     $8,%esp         // space for register saves on stack
-	sub     $16,%ebp        // increment to next round key
-	sub     $10,%r3          
-	je      4f              // 10 rounds for 128-bit key
-	sub     $32,%ebp
-	sub     $2,%r3
-	je      3f              // 12 rounds for 128-bit key
-	sub     $32,%ebp
+	sub     $8,%esp		// space for register saves on stack
+	sub     $16,%ebp	// increment to next round key
+	cmp     $12,%r3
+	jb      4f		// 10 rounds for 128-bit key
+	lea     -32(%ebp),%ebp
+	je      3f		// 12 rounds for 192-bit key
+	lea     -32(%ebp),%ebp
 
-2:	inv_rnd1( +64(%ebp), it_tab)	// 14 rounds for 128-bit key
+2:	inv_rnd1( +64(%ebp), it_tab)	// 14 rounds for 256-bit key
 	inv_rnd2( +48(%ebp), it_tab)
-3:	inv_rnd1( +32(%ebp), it_tab)	// 12 rounds for 128-bit key
+3:	inv_rnd1( +32(%ebp), it_tab)	// 12 rounds for 192-bit key
 	inv_rnd2( +16(%ebp), it_tab)
 4:	inv_rnd1(    (%ebp), it_tab)	// 10 rounds for 128-bit key
 	inv_rnd2( -16(%ebp), it_tab)
diff --git a/arch/i386/crypto/aes.c b/arch/i386/crypto/aes.c
index 88ee85c3b43..a50397b1d5c 100644
--- a/arch/i386/crypto/aes.c
+++ b/arch/i386/crypto/aes.c
@@ -36,6 +36,8 @@
  * Copyright (c) 2004 Red Hat, Inc., James Morris <jmorris@redhat.com>
  *
  */
+
+#include <asm/byteorder.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/init.h>
@@ -59,7 +61,6 @@ struct aes_ctx {
 };
 
 #define WPOLY 0x011b
-#define u32_in(x) le32_to_cpup((const __le32 *)(x))
 #define bytes2word(b0, b1, b2, b3)  \
 	(((u32)(b3) << 24) | ((u32)(b2) << 16) | ((u32)(b1) << 8) | (b0))
 
@@ -93,7 +94,6 @@ static u32 rcon_tab[RC_LENGTH];
 
 u32 ft_tab[4][256];
 u32 fl_tab[4][256];
-static u32 ls_tab[4][256];
 static u32 im_tab[4][256];
 u32 il_tab[4][256];
 u32 it_tab[4][256];
@@ -144,15 +144,6 @@ static void gen_tabs(void)
 		fl_tab[2][i] = upr(w, 2);
 		fl_tab[3][i] = upr(w, 3);
 		
-		/*
-		 * table for key schedule if fl_tab above is
-		 * not of the required form
-		 */
-		ls_tab[0][i] = w;
-		ls_tab[1][i] = upr(w, 1);
-		ls_tab[2][i] = upr(w, 2);
-		ls_tab[3][i] = upr(w, 3);
-		
 		b = fi(inv_affine((u8)i));
 		w = bytes2word(fe(b), f9(b), fd(b), fb(b));
 
@@ -393,13 +384,14 @@ aes_set_key(void *ctx_arg, const u8 *in_key, unsigned int key_len, u32 *flags)
 	int i;
 	u32 ss[8];
 	struct aes_ctx *ctx = ctx_arg;
+	const __le32 *key = (const __le32 *)in_key;
 
 	/* encryption schedule */
 	
-	ctx->ekey[0] = ss[0] = u32_in(in_key);
-	ctx->ekey[1] = ss[1] = u32_in(in_key + 4);
-	ctx->ekey[2] = ss[2] = u32_in(in_key + 8);
-	ctx->ekey[3] = ss[3] = u32_in(in_key + 12);
+	ctx->ekey[0] = ss[0] = le32_to_cpu(key[0]);
+	ctx->ekey[1] = ss[1] = le32_to_cpu(key[1]);
+	ctx->ekey[2] = ss[2] = le32_to_cpu(key[2]);
+	ctx->ekey[3] = ss[3] = le32_to_cpu(key[3]);
 
 	switch(key_len) {
 	case 16:
@@ -410,8 +402,8 @@ aes_set_key(void *ctx_arg, const u8 *in_key, unsigned int key_len, u32 *flags)
 		break;
 		
 	case 24:
-		ctx->ekey[4] = ss[4] = u32_in(in_key + 16);
-		ctx->ekey[5] = ss[5] = u32_in(in_key + 20);
+		ctx->ekey[4] = ss[4] = le32_to_cpu(key[4]);
+		ctx->ekey[5] = ss[5] = le32_to_cpu(key[5]);
 		for (i = 0; i < 7; i++)
 			ke6(ctx->ekey, i);
 		kel6(ctx->ekey, 7); 
@@ -419,10 +411,10 @@ aes_set_key(void *ctx_arg, const u8 *in_key, unsigned int key_len, u32 *flags)
 		break;
 
 	case 32:
-		ctx->ekey[4] = ss[4] = u32_in(in_key + 16);
-		ctx->ekey[5] = ss[5] = u32_in(in_key + 20);
-		ctx->ekey[6] = ss[6] = u32_in(in_key + 24);
-		ctx->ekey[7] = ss[7] = u32_in(in_key + 28);
+		ctx->ekey[4] = ss[4] = le32_to_cpu(key[4]);
+		ctx->ekey[5] = ss[5] = le32_to_cpu(key[5]);
+		ctx->ekey[6] = ss[6] = le32_to_cpu(key[6]);
+		ctx->ekey[7] = ss[7] = le32_to_cpu(key[7]);
 		for (i = 0; i < 6; i++)
 			ke8(ctx->ekey, i);
 		kel8(ctx->ekey, 6);
@@ -436,10 +428,10 @@ aes_set_key(void *ctx_arg, const u8 *in_key, unsigned int key_len, u32 *flags)
 	
 	/* decryption schedule */
 	
-	ctx->dkey[0] = ss[0] = u32_in(in_key);
-	ctx->dkey[1] = ss[1] = u32_in(in_key + 4);
-	ctx->dkey[2] = ss[2] = u32_in(in_key + 8);
-	ctx->dkey[3] = ss[3] = u32_in(in_key + 12);
+	ctx->dkey[0] = ss[0] = le32_to_cpu(key[0]);
+	ctx->dkey[1] = ss[1] = le32_to_cpu(key[1]);
+	ctx->dkey[2] = ss[2] = le32_to_cpu(key[2]);
+	ctx->dkey[3] = ss[3] = le32_to_cpu(key[3]);
 
 	switch (key_len) {
 	case 16:
@@ -450,8 +442,8 @@ aes_set_key(void *ctx_arg, const u8 *in_key, unsigned int key_len, u32 *flags)
 		break;
 		
 	case 24:
-		ctx->dkey[4] = ff(ss[4] = u32_in(in_key + 16));
-		ctx->dkey[5] = ff(ss[5] = u32_in(in_key + 20));
+		ctx->dkey[4] = ff(ss[4] = le32_to_cpu(key[4]));
+		ctx->dkey[5] = ff(ss[5] = le32_to_cpu(key[5]));
 		kdf6(ctx->dkey, 0);
 		for (i = 1; i < 7; i++)
 			kd6(ctx->dkey, i);
@@ -459,10 +451,10 @@ aes_set_key(void *ctx_arg, const u8 *in_key, unsigned int key_len, u32 *flags)
 		break;
 
 	case 32:
-		ctx->dkey[4] = ff(ss[4] = u32_in(in_key + 16));
-		ctx->dkey[5] = ff(ss[5] = u32_in(in_key + 20));
-		ctx->dkey[6] = ff(ss[6] = u32_in(in_key + 24));
-		ctx->dkey[7] = ff(ss[7] = u32_in(in_key + 28));
+		ctx->dkey[4] = ff(ss[4] = le32_to_cpu(key[4]));
+		ctx->dkey[5] = ff(ss[5] = le32_to_cpu(key[5]));
+		ctx->dkey[6] = ff(ss[6] = le32_to_cpu(key[6]));
+		ctx->dkey[7] = ff(ss[7] = le32_to_cpu(key[7]));
 		kdf8(ctx->dkey, 0);
 		for (i = 1; i < 6; i++)
 			kd8(ctx->dkey, i);
@@ -484,6 +476,8 @@ static inline void aes_decrypt(void *ctx, u8 *dst, const u8 *src)
 
 static struct crypto_alg aes_alg = {
 	.cra_name		=	"aes",
+	.cra_driver_name	=	"aes-i586",
+	.cra_priority		=	200,
 	.cra_flags		=	CRYPTO_ALG_TYPE_CIPHER,
 	.cra_blocksize		=	AES_BLOCK_SIZE,
 	.cra_ctxsize		=	sizeof(struct aes_ctx),
diff --git a/arch/x86_64/crypto/aes.c b/arch/x86_64/crypto/aes.c
index acfdaa28791..fb1b961a2e2 100644
--- a/arch/x86_64/crypto/aes.c
+++ b/arch/x86_64/crypto/aes.c
@@ -74,8 +74,6 @@ static inline u8 byte(const u32 x, const unsigned n)
 	return x >> (n << 3);
 }
 
-#define u32_in(x) le32_to_cpu(*(const __le32 *)(x))
-
 struct aes_ctx
 {
 	u32 key_length;
@@ -234,6 +232,7 @@ static int aes_set_key(void *ctx_arg, const u8 *in_key, unsigned int key_len,
 		       u32 *flags)
 {
 	struct aes_ctx *ctx = ctx_arg;
+	const __le32 *key = (const __le32 *)in_key;
 	u32 i, j, t, u, v, w;
 
 	if (key_len != 16 && key_len != 24 && key_len != 32) {
@@ -243,10 +242,10 @@ static int aes_set_key(void *ctx_arg, const u8 *in_key, unsigned int key_len,
 
 	ctx->key_length = key_len;
 
-	D_KEY[key_len + 24] = E_KEY[0] = u32_in(in_key);
-	D_KEY[key_len + 25] = E_KEY[1] = u32_in(in_key + 4);
-	D_KEY[key_len + 26] = E_KEY[2] = u32_in(in_key + 8);
-	D_KEY[key_len + 27] = E_KEY[3] = u32_in(in_key + 12);
+	D_KEY[key_len + 24] = E_KEY[0] = le32_to_cpu(key[0]);
+	D_KEY[key_len + 25] = E_KEY[1] = le32_to_cpu(key[1]);
+	D_KEY[key_len + 26] = E_KEY[2] = le32_to_cpu(key[2]);
+	D_KEY[key_len + 27] = E_KEY[3] = le32_to_cpu(key[3]);
 
 	switch (key_len) {
 	case 16:
@@ -256,17 +255,17 @@ static int aes_set_key(void *ctx_arg, const u8 *in_key, unsigned int key_len,
 		break;
 
 	case 24:
-		E_KEY[4] = u32_in(in_key + 16);
-		t = E_KEY[5] = u32_in(in_key + 20);
+		E_KEY[4] = le32_to_cpu(key[4]);
+		t = E_KEY[5] = le32_to_cpu(key[5]);
 		for (i = 0; i < 8; ++i)
 			loop6 (i);
 		break;
 
 	case 32:
-		E_KEY[4] = u32_in(in_key + 16);
-		E_KEY[5] = u32_in(in_key + 20);
-		E_KEY[6] = u32_in(in_key + 24);
-		t = E_KEY[7] = u32_in(in_key + 28);
+		E_KEY[4] = le32_to_cpu(key[4]);
+		E_KEY[5] = le32_to_cpu(key[5]);
+		E_KEY[6] = le32_to_cpu(key[6]);
+		t = E_KEY[7] = le32_to_cpu(key[7]);
 		for (i = 0; i < 7; ++i)
 			loop8(i);
 		break;
@@ -290,6 +289,8 @@ extern void aes_decrypt(void *ctx_arg, u8 *out, const u8 *in);
 
 static struct crypto_alg aes_alg = {
 	.cra_name		=	"aes",
+	.cra_driver_name	=	"aes-x86_64",
+	.cra_priority		=	200,
 	.cra_flags		=	CRYPTO_ALG_TYPE_CIPHER,
 	.cra_blocksize		=	AES_BLOCK_SIZE,
 	.cra_ctxsize		=	sizeof(struct aes_ctx),
diff --git a/crypto/Kconfig b/crypto/Kconfig
index 52e1d4108a9..c442f2e7ce4 100644
--- a/crypto/Kconfig
+++ b/crypto/Kconfig
@@ -157,7 +157,7 @@ config CRYPTO_SERPENT
 
 config CRYPTO_AES
 	tristate "AES cipher algorithms"
-	depends on CRYPTO && !(X86 || UML_X86)
+	depends on CRYPTO
 	help
 	  AES cipher algorithms (FIPS-197). AES uses the Rijndael 
 	  algorithm.
diff --git a/crypto/aes.c b/crypto/aes.c
index 5df92888ef5..0a6a5c14368 100644
--- a/crypto/aes.c
+++ b/crypto/aes.c
@@ -73,9 +73,6 @@ byte(const u32 x, const unsigned n)
 	return x >> (n << 3);
 }
 
-#define u32_in(x) le32_to_cpu(*(const u32 *)(x))
-#define u32_out(to, from) (*(u32 *)(to) = cpu_to_le32(from))
-
 struct aes_ctx {
 	int key_length;
 	u32 E[60];
@@ -256,6 +253,7 @@ static int
 aes_set_key(void *ctx_arg, const u8 *in_key, unsigned int key_len, u32 *flags)
 {
 	struct aes_ctx *ctx = ctx_arg;
+	const __le32 *key = (const __le32 *)in_key;
 	u32 i, t, u, v, w;
 
 	if (key_len != 16 && key_len != 24 && key_len != 32) {
@@ -265,10 +263,10 @@ aes_set_key(void *ctx_arg, const u8 *in_key, unsigned int key_len, u32 *flags)
 
 	ctx->key_length = key_len;
 
-	E_KEY[0] = u32_in (in_key);
-	E_KEY[1] = u32_in (in_key + 4);
-	E_KEY[2] = u32_in (in_key + 8);
-	E_KEY[3] = u32_in (in_key + 12);
+	E_KEY[0] = le32_to_cpu(key[0]);
+	E_KEY[1] = le32_to_cpu(key[1]);
+	E_KEY[2] = le32_to_cpu(key[2]);
+	E_KEY[3] = le32_to_cpu(key[3]);
 
 	switch (key_len) {
 	case 16:
@@ -278,17 +276,17 @@ aes_set_key(void *ctx_arg, const u8 *in_key, unsigned int key_len, u32 *flags)
 		break;
 
 	case 24:
-		E_KEY[4] = u32_in (in_key + 16);
-		t = E_KEY[5] = u32_in (in_key + 20);
+		E_KEY[4] = le32_to_cpu(key[4]);
+		t = E_KEY[5] = le32_to_cpu(key[5]);
 		for (i = 0; i < 8; ++i)
 			loop6 (i);
 		break;
 
 	case 32:
-		E_KEY[4] = u32_in (in_key + 16);
-		E_KEY[5] = u32_in (in_key + 20);
-		E_KEY[6] = u32_in (in_key + 24);
-		t = E_KEY[7] = u32_in (in_key + 28);
+		E_KEY[4] = le32_to_cpu(key[4]);
+		E_KEY[5] = le32_to_cpu(key[5]);
+		E_KEY[6] = le32_to_cpu(key[6]);
+		t = E_KEY[7] = le32_to_cpu(key[7]);
 		for (i = 0; i < 7; ++i)
 			loop8 (i);
 		break;
@@ -324,13 +322,15 @@ aes_set_key(void *ctx_arg, const u8 *in_key, unsigned int key_len, u32 *flags)
 static void aes_encrypt(void *ctx_arg, u8 *out, const u8 *in)
 {
 	const struct aes_ctx *ctx = ctx_arg;
+	const __le32 *src = (const __le32 *)in;
+	__le32 *dst = (__le32 *)out;
 	u32 b0[4], b1[4];
 	const u32 *kp = E_KEY + 4;
 
-	b0[0] = u32_in (in) ^ E_KEY[0];
-	b0[1] = u32_in (in + 4) ^ E_KEY[1];
-	b0[2] = u32_in (in + 8) ^ E_KEY[2];
-	b0[3] = u32_in (in + 12) ^ E_KEY[3];
+	b0[0] = le32_to_cpu(src[0]) ^ E_KEY[0];
+	b0[1] = le32_to_cpu(src[1]) ^ E_KEY[1];
+	b0[2] = le32_to_cpu(src[2]) ^ E_KEY[2];
+	b0[3] = le32_to_cpu(src[3]) ^ E_KEY[3];
 
 	if (ctx->key_length > 24) {
 		f_nround (b1, b0, kp);
@@ -353,10 +353,10 @@ static void aes_encrypt(void *ctx_arg, u8 *out, const u8 *in)
 	f_nround (b1, b0, kp);
 	f_lround (b0, b1, kp);
 
-	u32_out (out, b0[0]);
-	u32_out (out + 4, b0[1]);
-	u32_out (out + 8, b0[2]);
-	u32_out (out + 12, b0[3]);
+	dst[0] = cpu_to_le32(b0[0]);
+	dst[1] = cpu_to_le32(b0[1]);
+	dst[2] = cpu_to_le32(b0[2]);
+	dst[3] = cpu_to_le32(b0[3]);
 }
 
 /* decrypt a block of text */
@@ -377,14 +377,16 @@ static void aes_encrypt(void *ctx_arg, u8 *out, const u8 *in)
 static void aes_decrypt(void *ctx_arg, u8 *out, const u8 *in)
 {
 	const struct aes_ctx *ctx = ctx_arg;
+	const __le32 *src = (const __le32 *)in;
+	__le32 *dst = (__le32 *)out;
 	u32 b0[4], b1[4];
 	const int key_len = ctx->key_length;
 	const u32 *kp = D_KEY + key_len + 20;
 
-	b0[0] = u32_in (in) ^ E_KEY[key_len + 24];
-	b0[1] = u32_in (in + 4) ^ E_KEY[key_len + 25];
-	b0[2] = u32_in (in + 8) ^ E_KEY[key_len + 26];
-	b0[3] = u32_in (in + 12) ^ E_KEY[key_len + 27];
+	b0[0] = le32_to_cpu(src[0]) ^ E_KEY[key_len + 24];
+	b0[1] = le32_to_cpu(src[1]) ^ E_KEY[key_len + 25];
+	b0[2] = le32_to_cpu(src[2]) ^ E_KEY[key_len + 26];
+	b0[3] = le32_to_cpu(src[3]) ^ E_KEY[key_len + 27];
 
 	if (key_len > 24) {
 		i_nround (b1, b0, kp);
@@ -407,18 +409,21 @@ static void aes_decrypt(void *ctx_arg, u8 *out, const u8 *in)
 	i_nround (b1, b0, kp);
 	i_lround (b0, b1, kp);
 
-	u32_out (out, b0[0]);
-	u32_out (out + 4, b0[1]);
-	u32_out (out + 8, b0[2]);
-	u32_out (out + 12, b0[3]);
+	dst[0] = cpu_to_le32(b0[0]);
+	dst[1] = cpu_to_le32(b0[1]);
+	dst[2] = cpu_to_le32(b0[2]);
+	dst[3] = cpu_to_le32(b0[3]);
 }
 
 
 static struct crypto_alg aes_alg = {
 	.cra_name		=	"aes",
+	.cra_driver_name	=	"aes-generic",
+	.cra_priority		=	100,
 	.cra_flags		=	CRYPTO_ALG_TYPE_CIPHER,
 	.cra_blocksize		=	AES_BLOCK_SIZE,
 	.cra_ctxsize		=	sizeof(struct aes_ctx),
+	.cra_alignmask		=	3,
 	.cra_module		=	THIS_MODULE,
 	.cra_list		=	LIST_HEAD_INIT(aes_alg.cra_list),
 	.cra_u			=	{
diff --git a/crypto/anubis.c b/crypto/anubis.c
index 3925eb0133c..2c796bdb91a 100644
--- a/crypto/anubis.c
+++ b/crypto/anubis.c
@@ -32,8 +32,10 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/mm.h>
+#include <asm/byteorder.h>
 #include <asm/scatterlist.h>
 #include <linux/crypto.h>
+#include <linux/types.h>
 
 #define ANUBIS_MIN_KEY_SIZE	16
 #define ANUBIS_MAX_KEY_SIZE	40
@@ -461,8 +463,8 @@ static const u32 rc[] = {
 static int anubis_setkey(void *ctx_arg, const u8 *in_key,
 			 unsigned int key_len, u32 *flags)
 {
-
-	int N, R, i, pos, r;
+	const __be32 *key = (const __be32 *)in_key;
+	int N, R, i, r;
 	u32 kappa[ANUBIS_MAX_N];
 	u32 inter[ANUBIS_MAX_N];
 
@@ -483,13 +485,8 @@ static int anubis_setkey(void *ctx_arg, const u8 *in_key,
 	ctx->R = R = 8 + N;
 
 	/* * map cipher key to initial key state (mu): */
-		for (i = 0, pos = 0; i < N; i++, pos += 4) {
-		kappa[i] =
-			(in_key[pos    ] << 24) ^
-			(in_key[pos + 1] << 16) ^
-			(in_key[pos + 2] <<  8) ^
-			(in_key[pos + 3]      );
-	}
+	for (i = 0; i < N; i++)
+		kappa[i] = be32_to_cpu(key[i]);
 
 	/*
 	 * generate R + 1 round keys:
@@ -578,7 +575,9 @@ static int anubis_setkey(void *ctx_arg, const u8 *in_key,
 static void anubis_crypt(u32 roundKey[ANUBIS_MAX_ROUNDS + 1][4],
 		u8 *ciphertext, const u8 *plaintext, const int R)
 {
-	int i, pos, r;
+	const __be32 *src = (const __be32 *)plaintext;
+	__be32 *dst = (__be32 *)ciphertext;
+	int i, r;
 	u32 state[4];
 	u32 inter[4];
 
@@ -586,14 +585,8 @@ static void anubis_crypt(u32 roundKey[ANUBIS_MAX_ROUNDS + 1][4],
 	 * map plaintext block to cipher state (mu)
 	 * and add initial round key (sigma[K^0]):
 	 */
-	for (i = 0, pos = 0; i < 4; i++, pos += 4) {
-		state[i] =
-			(plaintext[pos    ] << 24) ^
-			(plaintext[pos + 1] << 16) ^
-			(plaintext[pos + 2] <<  8) ^
-			(plaintext[pos + 3]      ) ^
-			roundKey[0][i];
-	}
+	for (i = 0; i < 4; i++)
+		state[i] = be32_to_cpu(src[i]) ^ roundKey[0][i];
 
 	/*
 	 * R - 1 full rounds:
@@ -663,13 +656,8 @@ static void anubis_crypt(u32 roundKey[ANUBIS_MAX_ROUNDS + 1][4],
 	 * map cipher state to ciphertext block (mu^{-1}):
 	 */
 
-	for (i = 0, pos = 0; i < 4; i++, pos += 4) {
-		u32 w = inter[i];
-		ciphertext[pos    ] = (u8)(w >> 24);
-		ciphertext[pos + 1] = (u8)(w >> 16);
-		ciphertext[pos + 2] = (u8)(w >>  8);
-		ciphertext[pos + 3] = (u8)(w      );
-	}
+	for (i = 0; i < 4; i++)
+		dst[i] = cpu_to_be32(inter[i]);
 }
 
 static void anubis_encrypt(void *ctx_arg, u8 *dst, const u8 *src)
@@ -689,6 +677,7 @@ static struct crypto_alg anubis_alg = {
 	.cra_flags		=	CRYPTO_ALG_TYPE_CIPHER,
 	.cra_blocksize		=	ANUBIS_BLOCK_SIZE,
 	.cra_ctxsize		=	sizeof (struct anubis_ctx),
+	.cra_alignmask		=	3,
 	.cra_module		=	THIS_MODULE,
 	.cra_list		=	LIST_HEAD_INIT(anubis_alg.cra_list),
 	.cra_u			=	{ .cipher = {
diff --git a/crypto/api.c b/crypto/api.c
index 40ae42e9b6a..e26156f7183 100644
--- a/crypto/api.c
+++ b/crypto/api.c
@@ -3,6 +3,7 @@
  *
  * Copyright (c) 2002 James Morris <jmorris@intercode.com.au>
  * Copyright (c) 2002 David S. Miller (davem@redhat.com)
+ * Copyright (c) 2005 Herbert Xu <herbert@gondor.apana.org.au>
  *
  * Portions derived from Cryptoapi, by Alexander Kjeldaas <astor@fast.no>
  * and Nettle, by Niels Möller.
@@ -18,9 +19,11 @@
 #include <linux/init.h>
 #include <linux/crypto.h>
 #include <linux/errno.h>
+#include <linux/kernel.h>
 #include <linux/kmod.h>
 #include <linux/rwsem.h>
 #include <linux/slab.h>
+#include <linux/string.h>
 #include "internal.h"
 
 LIST_HEAD(crypto_alg_list);
@@ -39,6 +42,7 @@ static inline void crypto_alg_put(struct crypto_alg *alg)
 static struct crypto_alg *crypto_alg_lookup(const char *name)
 {
 	struct crypto_alg *q, *alg = NULL;
+	int best = -1;
 
 	if (!name)
 		return NULL;
@@ -46,11 +50,23 @@ static struct crypto_alg *crypto_alg_lookup(const char *name)
 	down_read(&crypto_alg_sem);
 	
 	list_for_each_entry(q, &crypto_alg_list, cra_list) {
-		if (!(strcmp(q->cra_name, name))) {
-			if (crypto_alg_get(q))
-				alg = q;
+		int exact, fuzzy;
+
+		exact = !strcmp(q->cra_driver_name, name);
+		fuzzy = !strcmp(q->cra_name, name);
+		if (!exact && !(fuzzy && q->cra_priority >