[deliverable/linux.git] / drivers / crypto / padlock-aes.c

/* 
 * Cryptographic API.
 *
 * Support for VIA PadLock hardware crypto engine.
 *
 * Copyright (c) 2004  Michal Ludvig <michal@logix.cz>
 *
 * Key expansion routine taken from crypto/aes.c
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * ---------------------------------------------------------------------------
 * Copyright (c) 2002, Dr Brian Gladman <brg@gladman.me.uk>, Worcester, UK.
 * All rights reserved.
 *
 * LICENSE TERMS
 *
 * The free distribution and use of this software in both source and binary
 * form is allowed (with or without changes) provided that:
 *
 *   1. distributions of this source code include the above copyright
 *      notice, this list of conditions and the following disclaimer;
 *
 *   2. distributions in binary form include the above copyright
 *      notice, this list of conditions and the following disclaimer
 *      in the documentation and/or other associated materials;
 *
 *   3. the copyright holder's name is not used to endorse products
 *      built using this software without specific written permission.
 *
 * ALTERNATIVELY, provided that this notice is retained in full, this product
 * may be distributed under the terms of the GNU General Public License (GPL),
 * in which case the provisions of the GPL apply INSTEAD OF those given above.
 *
 * DISCLAIMER
 *
 * This software is provided 'as is' with no explicit or implied warranties
 * in respect of its properties, including, but not limited to, correctness
 * and/or fitness for purpose.
 * ---------------------------------------------------------------------------
 */

#include <linux/module.h>
#include <linux/init.h>
#include <linux/types.h>
#include <linux/errno.h>
#include <linux/crypto.h>
#include <linux/interrupt.h>
#include <linux/kernel.h>
#include <asm/byteorder.h>
#include "padlock.h"

#define AES_MIN_KEY_SIZE	16	/* in uint8_t units */
#define AES_MAX_KEY_SIZE	32	/* ditto */
#define AES_BLOCK_SIZE		16	/* ditto */
#define AES_EXTENDED_KEY_SIZE	64	/* in uint32_t units */
#define AES_EXTENDED_KEY_SIZE_B	(AES_EXTENDED_KEY_SIZE * sizeof(uint32_t))

/* Whenever making any changes to the following
 * structure *make sure* you keep E, d_data
 * and cword aligned on 16 Bytes boundaries!!! */
struct aes_ctx {
	struct {
		struct cword encrypt;
		struct cword decrypt;
	} cword;
	u32 *D;
	int key_length;
	u32 E[AES_EXTENDED_KEY_SIZE]
		__attribute__ ((__aligned__(PADLOCK_ALIGNMENT)));
	u32 d_data[AES_EXTENDED_KEY_SIZE]
		__attribute__ ((__aligned__(PADLOCK_ALIGNMENT)));
};

/* ====== Key management routines ====== */

static inline uint32_t
generic_rotr32 (const uint32_t x, const unsigned bits)
{
	const unsigned n = bits % 32;
	return (x >> n) | (x << (32 - n));
}

static inline uint32_t
generic_rotl32 (const uint32_t x, const unsigned bits)
{
	const unsigned n = bits % 32;
	return (x << n) | (x >> (32 - n));
}

#define rotl generic_rotl32
#define rotr generic_rotr32

/*
 * #define byte(x, nr) ((unsigned char)((x) >> (nr*8))) 
 */
static inline uint8_t
byte(const uint32_t x, const unsigned n)
{
	return x >> (n << 3);
}

#define E_KEY ctx->E
#define D_KEY ctx->D

static uint8_t pow_tab[256];
static uint8_t log_tab[256];
static uint8_t sbx_tab[256];
static uint8_t isb_tab[256];
static uint32_t rco_tab[10];
static uint32_t ft_tab[4][256];
static uint32_t it_tab[4][256];

static uint32_t fl_tab[4][256];
static uint32_t il_tab[4][256];

static inline uint8_t
f_mult (uint8_t a, uint8_t b)
{
	uint8_t aa = log_tab[a], cc = aa + log_tab[b];

	return pow_tab[cc + (cc < aa ? 1 : 0)];
}

#define ff_mult(a,b)    (a && b ? f_mult(a, b) : 0)

#define f_rn(bo, bi, n, k)					\
    bo[n] =  ft_tab[0][byte(bi[n],0)] ^				\
             ft_tab[1][byte(bi[(n + 1) & 3],1)] ^		\
             ft_tab[2][byte(bi[(n + 2) & 3],2)] ^		\
             ft_tab[3][byte(bi[(n + 3) & 3],3)] ^ *(k + n)

#define i_rn(bo, bi, n, k)					\
    bo[n] =  it_tab[0][byte(bi[n],0)] ^				\
             it_tab[1][byte(bi[(n + 3) & 3],1)] ^		\
             it_tab[2][byte(bi[(n + 2) & 3],2)] ^		\
             it_tab[3][byte(bi[(n + 1) & 3],3)] ^ *(k + n)

#define ls_box(x)				\
    ( fl_tab[0][byte(x, 0)] ^			\
      fl_tab[1][byte(x, 1)] ^			\
      fl_tab[2][byte(x, 2)] ^			\
      fl_tab[3][byte(x, 3)] )

#define f_rl(bo, bi, n, k)					\
    bo[n] =  fl_tab[0][byte(bi[n],0)] ^				\
             fl_tab[1][byte(bi[(n + 1) & 3],1)] ^		\
             fl_tab[2][byte(bi[(n + 2) & 3],2)] ^		\
             fl_tab[3][byte(bi[(n + 3) & 3],3)] ^ *(k + n)

#define i_rl(bo, bi, n, k)					\
    bo[n] =  il_tab[0][byte(bi[n],0)] ^				\
             il_tab[1][byte(bi[(n + 3) & 3],1)] ^		\
             il_tab[2][byte(bi[(n + 2) & 3],2)] ^		\
             il_tab[3][byte(bi[(n + 1) & 3],3)] ^ *(k + n)

static void
gen_tabs (void)
{
	uint32_t i, t;
	uint8_t p, q;

	/* log and power tables for GF(2**8) finite field with
	   0x011b as modular polynomial - the simplest prmitive
	   root is 0x03, used here to generate the tables */

	for (i = 0, p = 1; i < 256; ++i) {
		pow_tab[i] = (uint8_t) p;
		log_tab[p] = (uint8_t) i;

		p ^= (p << 1) ^ (p & 0x80 ? 0x01b : 0);
	}

	log_tab[1] = 0;

	for (i = 0, p = 1; i < 10; ++i) {
		rco_tab[i] = p;

		p = (p << 1) ^ (p & 0x80 ? 0x01b : 0);
	}

	for (i = 0; i < 256; ++i) {
		p = (i ? pow_tab[255 - log_tab[i]] : 0);
		q = ((p >> 7) | (p << 1)) ^ ((p >> 6) | (p << 2));
		p ^= 0x63 ^ q ^ ((q >> 6) | (q << 2));
		sbx_tab[i] = p;
		isb_tab[p] = (uint8_t) i;
	}

	for (i = 0; i < 256; ++i) {
		p = sbx_tab[i];

		t = p;
		fl_tab[0][i] = t;
		fl_tab[1][i] = rotl (t, 8);
		fl_tab[2][i] = rotl (t, 16);
		fl_tab[3][i] = rotl (t, 24);

		t = ((uint32_t) ff_mult (2, p)) |
		    ((uint32_t) p << 8) |
		    ((uint32_t) p << 16) | ((uint32_t) ff_mult (3, p) << 24);

		ft_tab[0][i] = t;
		ft_tab[1][i] = rotl (t, 8);
		ft_tab[2][i] = rotl (t, 16);
		ft_tab[3][i] = rotl (t, 24);

		p = isb_tab[i];

		t = p;
		il_tab[0][i] = t;
		il_tab[1][i] = rotl (t, 8);
		il_tab[2][i] = rotl (t, 16);
		il_tab[3][i] = rotl (t, 24);

		t = ((uint32_t) ff_mult (14, p)) |
		    ((uint32_t) ff_mult (9, p) << 8) |
		    ((uint32_t) ff_mult (13, p) << 16) |
		    ((uint32_t) ff_mult (11, p) << 24);

		it_tab[0][i] = t;
		it_tab[1][i] = rotl (t, 8);
		it_tab[2][i] = rotl (t, 16);
		it_tab[3][i] = rotl (t, 24);
	}
}

#define star_x(x) (((x) & 0x7f7f7f7f) << 1) ^ ((((x) & 0x80808080) >> 7) * 0x1b)

#define imix_col(y,x)       \
    u   = star_x(x);        \
    v   = star_x(u);        \
    w   = star_x(v);        \
    t   = w ^ (x);          \
   (y)  = u ^ v ^ w;        \
   (y) ^= rotr(u ^ t,  8) ^ \
          rotr(v ^ t, 16) ^ \
          rotr(t,24)

/* initialise the key schedule from the user supplied key */

#define loop4(i)                                    \
{   t = rotr(t,  8); t = ls_box(t) ^ rco_tab[i];    \
    t ^= E_KEY[4 * i];     E_KEY[4 * i + 4] = t;    \
    t ^= E_KEY[4 * i + 1]; E_KEY[4 * i + 5] = t;    \
    t ^= E_KEY[4 * i + 2]; E_KEY[4 * i + 6] = t;    \
    t ^= E_KEY[4 * i + 3]; E_KEY[4 * i + 7] = t;    \
}

#define loop6(i)                                    \
{   t = rotr(t,  8); t = ls_box(t) ^ rco_tab[i];    \
    t ^= E_KEY[6 * i];     E_KEY[6 * i + 6] = t;    \
    t ^= E_KEY[6 * i + 1]; E_KEY[6 * i + 7] = t;    \
    t ^= E_KEY[6 * i + 2]; E_KEY[6 * i + 8] = t;    \
    t ^= E_KEY[6 * i + 3]; E_KEY[6 * i + 9] = t;    \
    t ^= E_KEY[6 * i + 4]; E_KEY[6 * i + 10] = t;   \
    t ^= E_KEY[6 * i + 5]; E_KEY[6 * i + 11] = t;   \
}

#define loop8(i)                                    \
{   t = rotr(t,  8); ; t = ls_box(t) ^ rco_tab[i];  \
    t ^= E_KEY[8 * i];     E_KEY[8 * i + 8] = t;    \
    t ^= E_KEY[8 * i + 1]; E_KEY[8 * i + 9] = t;    \
    t ^= E_KEY[8 * i + 2]; E_KEY[8 * i + 10] = t;   \
    t ^= E_KEY[8 * i + 3]; E_KEY[8 * i + 11] = t;   \
    t  = E_KEY[8 * i + 4] ^ ls_box(t);    \
    E_KEY[8 * i + 12] = t;                \
    t ^= E_KEY[8 * i + 5]; E_KEY[8 * i + 13] = t;   \
    t ^= E_KEY[8 * i + 6]; E_KEY[8 * i + 14] = t;   \
    t ^= E_KEY[8 * i + 7]; E_KEY[8 * i + 15] = t;   \
}

/* Tells whether the ACE is capable to generate
   the extended key for a given key_len. */
static inline int
aes_hw_extkey_available(uint8_t key_len)
{
	/* TODO: We should check the actual CPU model/stepping
	         as it's possible that the capability will be
	         added in the next CPU revisions. */
	if (key_len == 16)
		return 1;
	return 0;
}

static inline struct aes_ctx *aes_ctx(struct crypto_tfm *tfm)
{
	unsigned long addr = (unsigned long)crypto_tfm_ctx(tfm);
	unsigned long align = PADLOCK_ALIGNMENT;

	if (align <= crypto_tfm_ctx_alignment())
		align = 1;
	return (struct aes_ctx *)ALIGN(addr, align);
}

static int aes_set_key(struct crypto_tfm *tfm, const u8 *in_key,
		       unsigned int key_len, u32 *flags)
{
	struct aes_ctx *ctx = aes_ctx(tfm);
	const __le32 *key = (const __le32 *)in_key;
	uint32_t i, t, u, v, w;
	uint32_t P[AES_EXTENDED_KEY_SIZE];
	uint32_t rounds;

	if (key_len != 16 && key_len != 24 && key_len != 32) {
		*flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
		return -EINVAL;
	}

	ctx->key_length = key_len;

	/*
	 * If the hardware is capable of generating the extended key
	 * itself we must supply the plain key for both encryption
	 * and decryption.
	 */
	ctx->D = ctx->E;

	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]);

	/* Prepare control words. */
	memset(&ctx->cword, 0, sizeof(ctx->cword));

	ctx->cword.decrypt.encdec = 1;
	ctx->cword.encrypt.rounds = 10 + (key_len - 16) / 4;
	ctx->cword.decrypt.rounds = ctx->cword.encrypt.rounds;
	ctx->cword.encrypt.ksize = (key_len - 16) / 8;
	ctx->cword.decrypt.ksize = ctx->cword.encrypt.ksize;

	/* Don't generate extended keys if the hardware can do it. */
	if (aes_hw_extkey_available(key_len))
		return 0;

	ctx->D = ctx->d_data;
	ctx->cword.encrypt.keygen = 1;
	ctx->cword.decrypt.keygen = 1;

	switch (key_len) {
	case 16:
		t = E_KEY[3];
		for (i = 0; i < 10; ++i)
			loop4 (i);
		break;

	case 24:
		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] = 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;
	}

	D_KEY[0] = E_KEY[0];
	D_KEY[1] = E_KEY[1];
	D_KEY[2] = E_KEY[2];
	D_KEY[3] = E_KEY[3];

	for (i = 4; i < key_len + 24; ++i) {
		imix_col (D_KEY[i], E_KEY[i]);
	}

	/* PadLock needs a different format of the decryption key. */
	rounds = 10 + (key_len - 16) / 4;

	for (i = 0; i < rounds; i++) {
		P[((i + 1) * 4) + 0] = D_KEY[((rounds - i - 1) * 4) + 0];
		P[((i + 1) * 4) + 1] = D_KEY[((rounds - i - 1) * 4) + 1];
		P[((i + 1) * 4) + 2] = D_KEY[((rounds - i - 1) * 4) + 2];
		P[((i + 1) * 4) + 3] = D_KEY[((rounds - i - 1) * 4) + 3];
	}

	P[0] = E_KEY[(rounds * 4) + 0];
	P[1] = E_KEY[(rounds * 4) + 1];
	P[2] = E_KEY[(rounds * 4) + 2];
	P[3] = E_KEY[(rounds * 4) + 3];

	memcpy(D_KEY, P, AES_EXTENDED_KEY_SIZE_B);

	return 0;
}

/* ====== Encryption/decryption routines ====== */

/* These are the real call to PadLock. */
static inline void padlock_xcrypt_ecb(const u8 *input, u8 *output, void *key,
				      void *control_word, u32 count)
{
	asm volatile ("pushfl; popfl");		/* enforce key reload. */
	asm volatile (".byte 0xf3,0x0f,0xa7,0xc8"	/* rep xcryptecb */
		      : "+S"(input), "+D"(output)
		      : "d"(control_word), "b"(key), "c"(count));
}

static inline u8 *padlock_xcrypt_cbc(const u8 *input, u8 *output, void *key,
				     u8 *iv, void *control_word, u32 count)
{
	/* Enforce key reload. */
	asm volatile ("pushfl; popfl");
	/* rep xcryptcbc */
	asm volatile (".byte 0xf3,0x0f,0xa7,0xd0"
		      : "+S" (input), "+D" (output), "+a" (iv)
		      : "d" (control_word), "b" (key), "c" (count));
	return iv;
}

static void aes_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
{
	struct aes_ctx *ctx = aes_ctx(tfm);
	padlock_xcrypt_ecb(in, out, ctx->E, &ctx->cword.encrypt, 1);
}

static void aes_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
{
	struct aes_ctx *ctx = aes_ctx(tfm);
	padlock_xcrypt_ecb(in, out, ctx->D, &ctx->cword.decrypt, 1);
}

static unsigned int aes_encrypt_ecb(const struct cipher_desc *desc, u8 *out,
				    const u8 *in, unsigned int nbytes)
{
	struct aes_ctx *ctx = aes_ctx(desc->tfm);
	padlock_xcrypt_ecb(in, out, ctx->E, &ctx->cword.encrypt,
			   nbytes / AES_BLOCK_SIZE);
	return nbytes & ~(AES_BLOCK_SIZE - 1);
}

static unsigned int aes_decrypt_ecb(const struct cipher_desc *desc, u8 *out,
				    const u8 *in, unsigned int nbytes)
{
	struct aes_ctx *ctx = aes_ctx(desc->tfm);
	padlock_xcrypt_ecb(in, out, ctx->D, &ctx->cword.decrypt,
			   nbytes / AES_BLOCK_SIZE);
	return nbytes & ~(AES_BLOCK_SIZE - 1);
}

static unsigned int aes_encrypt_cbc(const struct cipher_desc *desc, u8 *out,
				    const u8 *in, unsigned int nbytes)
{
	struct aes_ctx *ctx = aes_ctx(desc->tfm);
	u8 *iv;

	iv = padlock_xcrypt_cbc(in, out, ctx->E, desc->info,
				&ctx->cword.encrypt, nbytes / AES_BLOCK_SIZE);
	memcpy(desc->info, iv, AES_BLOCK_SIZE);

	return nbytes & ~(AES_BLOCK_SIZE - 1);
}

static unsigned int aes_decrypt_cbc(const struct cipher_desc *desc, u8 *out,
				    const u8 *in, unsigned int nbytes)
{
	struct aes_ctx *ctx = aes_ctx(desc->tfm);
	padlock_xcrypt_cbc(in, out, ctx->D, desc->info, &ctx->cword.decrypt,
			   nbytes / AES_BLOCK_SIZE);
	return nbytes & ~(AES_BLOCK_SIZE - 1);
}

static struct crypto_alg aes_alg = {
	.cra_name		=	"aes",
	.cra_driver_name	=	"aes-padlock",
	.cra_priority		=	300,
	.cra_flags		=	CRYPTO_ALG_TYPE_CIPHER,
	.cra_blocksize		=	AES_BLOCK_SIZE,
	.cra_ctxsize		=	sizeof(struct aes_ctx),
	.cra_alignmask		=	PADLOCK_ALIGNMENT - 1,
	.cra_module		=	THIS_MODULE,
	.cra_list		=	LIST_HEAD_INIT(aes_alg.cra_list),
	.cra_u			=	{
		.cipher = {
			.cia_min_keysize	=	AES_MIN_KEY_SIZE,
			.cia_max_keysize	=	AES_MAX_KEY_SIZE,
			.cia_setkey	   	= 	aes_set_key,
			.cia_encrypt	 	=	aes_encrypt,
			.cia_decrypt	  	=	aes_decrypt,
			.cia_encrypt_ecb 	=	aes_encrypt_ecb,
			.cia_decrypt_ecb  	=	aes_decrypt_ecb,
			.cia_encrypt_cbc 	=	aes_encrypt_cbc,
			.cia_decrypt_cbc  	=	aes_decrypt_cbc,
		}
	}
};

int __init padlock_init_aes(void)
{
	printk(KERN_NOTICE PFX "Using VIA PadLock ACE for AES algorithm.\n");

	gen_tabs();
	return crypto_register_alg(&aes_alg);
}

void __exit padlock_fini_aes(void)
{
	crypto_unregister_alg(&aes_alg);
}
Commit	Line	Data
1da177e4 LT	1	/*
	2	* Cryptographic API.
	3	*
	4	* Support for VIA PadLock hardware crypto engine.
	5	*
	6	* Copyright (c) 2004 Michal Ludvig <michal@logix.cz>
	7	*
	8	* Key expansion routine taken from crypto/aes.c
	9	*
	10	* This program is free software; you can redistribute it and/or modify
	11	* it under the terms of the GNU General Public License as published by
	12	* the Free Software Foundation; either version 2 of the License, or
	13	* (at your option) any later version.
	14	*
	15	* ---------------------------------------------------------------------------
	16	* Copyright (c) 2002, Dr Brian Gladman <brg@gladman.me.uk>, Worcester, UK.
	17	* All rights reserved.
	18	*
	19	* LICENSE TERMS
	20	*
	21	* The free distribution and use of this software in both source and binary
	22	* form is allowed (with or without changes) provided that:
	23	*
	24	* 1. distributions of this source code include the above copyright
	25	* notice, this list of conditions and the following disclaimer;
	26	*
	27	* 2. distributions in binary form include the above copyright
	28	* notice, this list of conditions and the following disclaimer
	29	* in the documentation and/or other associated materials;
	30	*
	31	* 3. the copyright holder's name is not used to endorse products
	32	* built using this software without specific written permission.
	33	*
	34	* ALTERNATIVELY, provided that this notice is retained in full, this product
	35	* may be distributed under the terms of the GNU General Public License (GPL),
	36	* in which case the provisions of the GPL apply INSTEAD OF those given above.
	37	*
	38	* DISCLAIMER
	39	*
	40	* This software is provided 'as is' with no explicit or implied warranties
	41	* in respect of its properties, including, but not limited to, correctness
	42	* and/or fitness for purpose.
	43	* ---------------------------------------------------------------------------
	44	*/
	45
	46	#include <linux/module.h>
	47	#include <linux/init.h>
	48	#include <linux/types.h>
	49	#include <linux/errno.h>
	50	#include <linux/crypto.h>
	51	#include <linux/interrupt.h>
6789b2dc	52	#include <linux/kernel.h>
1da177e4 LT	53	#include <asm/byteorder.h>
	54	#include "padlock.h"
	55
	56	#define AES_MIN_KEY_SIZE 16 /* in uint8_t units */
	57	#define AES_MAX_KEY_SIZE 32 /* ditto */
	58	#define AES_BLOCK_SIZE 16 /* ditto */
	59	#define AES_EXTENDED_KEY_SIZE 64 /* in uint32_t units */
	60	#define AES_EXTENDED_KEY_SIZE_B (AES_EXTENDED_KEY_SIZE * sizeof(uint32_t))
	61
cc08632f ML	62	/* Whenever making any changes to the following
	63	* structure make sure you keep E, d_data
	64	* and cword aligned on 16 Bytes boundaries!!! */
1da177e4	65	struct aes_ctx {
6789b2dc HX	66	struct {
	67	struct cword encrypt;
	68	struct cword decrypt;
	69	} cword;
82062c72	70	u32 *D;
1da177e4	71	int key_length;
cc08632f ML	72	u32 E[AES_EXTENDED_KEY_SIZE]
	73	__attribute__ ((__aligned__(PADLOCK_ALIGNMENT)));
	74	u32 d_data[AES_EXTENDED_KEY_SIZE]
	75	__attribute__ ((__aligned__(PADLOCK_ALIGNMENT)));
1da177e4 LT	76	};
	77
	78	/* ====== Key management routines ====== */
	79
	80	static inline uint32_t
	81	generic_rotr32 (const uint32_t x, const unsigned bits)
	82	{
	83	const unsigned n = bits % 32;
	84	return (x >> n) \| (x << (32 - n));
	85	}
	86
	87	static inline uint32_t
	88	generic_rotl32 (const uint32_t x, const unsigned bits)
	89	{
	90	const unsigned n = bits % 32;
	91	return (x << n) \| (x >> (32 - n));
	92	}
	93
	94	#define rotl generic_rotl32
	95	#define rotr generic_rotr32
	96
	97	/*
	98	* #define byte(x, nr) ((unsigned char)((x) >> (nr*8)))
	99	*/
	100	static inline uint8_t
	101	byte(const uint32_t x, const unsigned n)
	102	{
	103	return x >> (n << 3);
	104	}
	105
1da177e4 LT	106	#define E_KEY ctx->E
	107	#define D_KEY ctx->D
	108
	109	static uint8_t pow_tab[256];
	110	static uint8_t log_tab[256];
	111	static uint8_t sbx_tab[256];
	112	static uint8_t isb_tab[256];
	113	static uint32_t rco_tab[10];
	114	static uint32_t ft_tab[4][256];
	115	static uint32_t it_tab[4][256];
	116
	117	static uint32_t fl_tab[4][256];
	118	static uint32_t il_tab[4][256];
	119
	120	static inline uint8_t
	121	f_mult (uint8_t a, uint8_t b)
	122	{
	123	uint8_t aa = log_tab[a], cc = aa + log_tab[b];
	124
	125	return pow_tab[cc + (cc < aa ? 1 : 0)];
	126	}
	127
	128	#define ff_mult(a,b) (a && b ? f_mult(a, b) : 0)
	129
	130	#define f_rn(bo, bi, n, k) \
	131	bo[n] = ft_tab[0][byte(bi[n],0)] ^ \
	132	ft_tab[1][byte(bi[(n + 1) & 3],1)] ^ \
	133	ft_tab[2][byte(bi[(n + 2) & 3],2)] ^ \
	134	ft_tab[3][byte(bi[(n + 3) & 3],3)] ^ *(k + n)
	135
	136	#define i_rn(bo, bi, n, k) \
	137	bo[n] = it_tab[0][byte(bi[n],0)] ^ \
	138	it_tab[1][byte(bi[(n + 3) & 3],1)] ^ \
	139	it_tab[2][byte(bi[(n + 2) & 3],2)] ^ \
	140	it_tab[3][byte(bi[(n + 1) & 3],3)] ^ *(k + n)
	141
	142	#define ls_box(x) \
	143	( fl_tab[0][byte(x, 0)] ^ \
	144	fl_tab[1][byte(x, 1)] ^ \
	145	fl_tab[2][byte(x, 2)] ^ \
	146	fl_tab[3][byte(x, 3)] )
	147
	148	#define f_rl(bo, bi, n, k) \
	149	bo[n] = fl_tab[0][byte(bi[n],0)] ^ \
	150	fl_tab[1][byte(bi[(n + 1) & 3],1)] ^ \
	151	fl_tab[2][byte(bi[(n + 2) & 3],2)] ^ \
	152	fl_tab[3][byte(bi[(n + 3) & 3],3)] ^ *(k + n)
	153
	154	#define i_rl(bo, bi, n, k) \
	155	bo[n] = il_tab[0][byte(bi[n],0)] ^ \
	156	il_tab[1][byte(bi[(n + 3) & 3],1)] ^ \
	157	il_tab[2][byte(bi[(n + 2) & 3],2)] ^ \
	158	il_tab[3][byte(bi[(n + 1) & 3],3)] ^ *(k + n)
	159
	160	static void
	161	gen_tabs (void)
	162	{
	163	uint32_t i, t;
	164	uint8_t p, q;
	165
	166	/* log and power tables for GF(2**8) finite field with
	167	0x011b as modular polynomial - the simplest prmitive
	168	root is 0x03, used here to generate the tables */
	169
170	for (i = 0, p = 1; i < 256; ++i) {
171	pow_tab[i] = (uint8_t) p;
172	log_tab[p] = (uint8_t) i;
173
174	p ^= (p << 1) ^ (p & 0x80 ? 0x01b : 0);
175	}
176
177	log_tab[1] = 0;
178
179	for (i = 0, p = 1; i < 10; ++i) {
180	rco_tab[i] = p;
181
182	p = (p << 1) ^ (p & 0x80 ? 0x01b : 0);
183	}
184
185	for (i = 0; i < 256; ++i) {
186	p = (i ? pow_tab[255 - log_tab[i]] : 0);
187	q = ((p >> 7) \| (p << 1)) ^ ((p >> 6) \| (p << 2));
188	p ^= 0x63 ^ q ^ ((q >> 6) \| (q << 2));
189	sbx_tab[i] = p;
190	isb_tab[p] = (uint8_t) i;
191	}
192
193	for (i = 0; i < 256; ++i) {
194	p = sbx_tab[i];
195
196	t = p;
197	fl_tab[0][i] = t;
198	fl_tab[1][i] = rotl (t, 8);
199	fl_tab[2][i] = rotl (t, 16);
200	fl_tab[3][i] = rotl (t, 24);
201
202	t = ((uint32_t) ff_mult (2, p)) \|
203	((uint32_t) p << 8) \|
204	((uint32_t) p << 16) \| ((uint32_t) ff_mult (3, p) << 24);
205
206	ft_tab[0][i] = t;
207	ft_tab[1][i] = rotl (t, 8);
208	ft_tab[2][i] = rotl (t, 16);
209	ft_tab[3][i] = rotl (t, 24);
210
211	p = isb_tab[i];
212
213	t = p;
214	il_tab[0][i] = t;
215	il_tab[1][i] = rotl (t, 8);
216	il_tab[2][i] = rotl (t, 16);
217	il_tab[3][i] = rotl (t, 24);
218
219	t = ((uint32_t) ff_mult (14, p)) \|
220	((uint32_t) ff_mult (9, p) << 8) \|
221	((uint32_t) ff_mult (13, p) << 16) \|
222	((uint32_t) ff_mult (11, p) << 24);
223
224	it_tab[0][i] = t;
225	it_tab[1][i] = rotl (t, 8);
226	it_tab[2][i] = rotl (t, 16);
227	it_tab[3][i] = rotl (t, 24);
228	}
229	}
230
231	#define star_x(x) (((x) & 0x7f7f7f7f) << 1) ^ ((((x) & 0x80808080) >> 7) * 0x1b)
232
233	#define imix_col(y,x) \
234	u = star_x(x); \
235	v = star_x(u); \
236	w = star_x(v); \
237	t = w ^ (x); \
238	(y) = u ^ v ^ w; \
239	(y) ^= rotr(u ^ t, 8) ^ \
240	rotr(v ^ t, 16) ^ \
241	rotr(t,24)
242
243	/* initialise the key schedule from the user supplied key */
244
245	#define loop4(i) \
246	{ t = rotr(t, 8); t = ls_box(t) ^ rco_tab[i]; \
247	t ^= E_KEY[4 * i]; E_KEY[4 * i + 4] = t; \
248	t ^= E_KEY[4 * i + 1]; E_KEY[4 * i + 5] = t; \
249	t ^= E_KEY[4 * i + 2]; E_KEY[4 * i + 6] = t; \
250	t ^= E_KEY[4 * i + 3]; E_KEY[4 * i + 7] = t; \
251	}
252
253	#define loop6(i) \
254	{ t = rotr(t, 8); t = ls_box(t) ^ rco_tab[i]; \
255	t ^= E_KEY[6 * i]; E_KEY[6 * i + 6] = t; \
256	t ^= E_KEY[6 * i + 1]; E_KEY[6 * i + 7] = t; \
257	t ^= E_KEY[6 * i + 2]; E_KEY[6 * i + 8] = t; \
258	t ^= E_KEY[6 * i + 3]; E_KEY[6 * i + 9] = t; \
259	t ^= E_KEY[6 * i + 4]; E_KEY[6 * i + 10] = t; \
260	t ^= E_KEY[6 * i + 5]; E_KEY[6 * i + 11] = t; \
261	}
262
263	#define loop8(i) \
264	{ t = rotr(t, 8); ; t = ls_box(t) ^ rco_tab[i]; \
265	t ^= E_KEY[8 * i]; E_KEY[8 * i + 8] = t; \
266	t ^= E_KEY[8 * i + 1]; E_KEY[8 * i + 9] = t; \
267	t ^= E_KEY[8 * i + 2]; E_KEY[8 * i + 10] = t; \
268	t ^= E_KEY[8 * i + 3]; E_KEY[8 * i + 11] = t; \
269	t = E_KEY[8 * i + 4] ^ ls_box(t); \
270	E_KEY[8 * i + 12] = t; \
271	t ^= E_KEY[8 * i + 5]; E_KEY[8 * i + 13] = t; \
272	t ^= E_KEY[8 * i + 6]; E_KEY[8 * i + 14] = t; \
273	t ^= E_KEY[8 * i + 7]; E_KEY[8 * i + 15] = t; \
274	}
275
276	/* Tells whether the ACE is capable to generate
277	the extended key for a given key_len. */
278	static inline int
279	aes_hw_extkey_available(uint8_t key_len)
280	{
281	/* TODO: We should check the actual CPU model/stepping
282	as it's possible that the capability will be
283	added in the next CPU revisions. */
284	if (key_len == 16)
285	return 1;
286	return 0;
287	}
288
6c2bb98b	289	static inline struct aes_ctx aes_ctx(struct crypto_tfm tfm)
6789b2dc	290	{
6c2bb98b	291	unsigned long addr = (unsigned long)crypto_tfm_ctx(tfm);
f10b7897 HX	292	unsigned long align = PADLOCK_ALIGNMENT;
	293
	294	if (align <= crypto_tfm_ctx_alignment())
	295	align = 1;
6c2bb98b	296	return (struct aes_ctx *)ALIGN(addr, align);
6789b2dc HX	297	}
6789b2dc HX	298
6c2bb98b HX	299	static int aes_set_key(struct crypto_tfm tfm, const u8 in_key,
6c2bb98b HX	300	unsigned int key_len, u32 *flags)
1da177e4	301	{
6c2bb98b	302	struct aes_ctx *ctx = aes_ctx(tfm);
06ace7a9	303	const __le32 key = (const __le32 )in_key;
1da177e4 LT	304	uint32_t i, t, u, v, w;
	305	uint32_t P[AES_EXTENDED_KEY_SIZE];
	306	uint32_t rounds;
	307
	308	if (key_len != 16 && key_len != 24 && key_len != 32) {
	309	*flags \|= CRYPTO_TFM_RES_BAD_KEY_LEN;
	310	return -EINVAL;
	311	}
	312
	313	ctx->key_length = key_len;
	314
6789b2dc HX	315	/*
	316	* If the hardware is capable of generating the extended key
	317	* itself we must supply the plain key for both encryption
	318	* and decryption.
	319	*/
82062c72	320	ctx->D = ctx->E;
1da177e4	321
06ace7a9 HX	322	E_KEY[0] = le32_to_cpu(key[0]);
	323	E_KEY[1] = le32_to_cpu(key[1]);
	324	E_KEY[2] = le32_to_cpu(key[2]);
	325	E_KEY[3] = le32_to_cpu(key[3]);
1da177e4	326
6789b2dc HX	327	/* Prepare control words. */
	328	memset(&ctx->cword, 0, sizeof(ctx->cword));
	329
	330	ctx->cword.decrypt.encdec = 1;
	331	ctx->cword.encrypt.rounds = 10 + (key_len - 16) / 4;
	332	ctx->cword.decrypt.rounds = ctx->cword.encrypt.rounds;
	333	ctx->cword.encrypt.ksize = (key_len - 16) / 8;
	334	ctx->cword.decrypt.ksize = ctx->cword.encrypt.ksize;
	335
1da177e4 LT	336	/* Don't generate extended keys if the hardware can do it. */
	337	if (aes_hw_extkey_available(key_len))
	338	return 0;
	339
6789b2dc HX	340	ctx->D = ctx->d_data;
	341	ctx->cword.encrypt.keygen = 1;
	342	ctx->cword.decrypt.keygen = 1;
	343
1da177e4 LT	344	switch (key_len) {
	345	case 16:
	346	t = E_KEY[3];
	347	for (i = 0; i < 10; ++i)
	348	loop4 (i);
	349	break;
	350
	351	case 24:
06ace7a9 HX	352	E_KEY[4] = le32_to_cpu(key[4]);
06ace7a9 HX	353	t = E_KEY[5] = le32_to_cpu(key[5]);
1da177e4 LT	354	for (i = 0; i < 8; ++i)
	355	loop6 (i);
	356	break;
	357
	358	case 32:
102d60a2 HX	359	E_KEY[4] = le32_to_cpu(key[4]);
	360	E_KEY[5] = le32_to_cpu(key[5]);
	361	E_KEY[6] = le32_to_cpu(key[6]);
	362	t = E_KEY[7] = le32_to_cpu(key[7]);
1da177e4 LT	363	for (i = 0; i < 7; ++i)
	364	loop8 (i);
	365	break;
	366	}
	367
	368	D_KEY[0] = E_KEY[0];
	369	D_KEY[1] = E_KEY[1];
	370	D_KEY[2] = E_KEY[2];
	371	D_KEY[3] = E_KEY[3];
	372
	373	for (i = 4; i < key_len + 24; ++i) {
	374	imix_col (D_KEY[i], E_KEY[i]);
	375	}
	376
	377	/* PadLock needs a different format of the decryption key. */
	378	rounds = 10 + (key_len - 16) / 4;
	379
	380	for (i = 0; i < rounds; i++) {
	381	P[((i + 1) * 4) + 0] = D_KEY[((rounds - i - 1) * 4) + 0];
	382	P[((i + 1) * 4) + 1] = D_KEY[((rounds - i - 1) * 4) + 1];
	383	P[((i + 1) * 4) + 2] = D_KEY[((rounds - i - 1) * 4) + 2];
	384	P[((i + 1) * 4) + 3] = D_KEY[((rounds - i - 1) * 4) + 3];
	385	}
	386
	387	P[0] = E_KEY[(rounds * 4) + 0];
	388	P[1] = E_KEY[(rounds * 4) + 1];
	389	P[2] = E_KEY[(rounds * 4) + 2];
	390	P[3] = E_KEY[(rounds * 4) + 3];
	391
	392	memcpy(D_KEY, P, AES_EXTENDED_KEY_SIZE_B);
	393
	394	return 0;
	395	}
	396
	397	/* ====== Encryption/decryption routines ====== */
	398
28e8c3ad	399	/* These are the real call to PadLock. */
6789b2dc HX	400	static inline void padlock_xcrypt_ecb(const u8 input, u8 output, void *key,
6789b2dc HX	401	void *control_word, u32 count)
1da177e4 LT	402	{
	403	asm volatile ("pushfl; popfl"); /* enforce key reload. */
	404	asm volatile (".byte 0xf3,0x0f,0xa7,0xc8" /* rep xcryptecb */
	405	: "+S"(input), "+D"(output)
	406	: "d"(control_word), "b"(key), "c"(count));
	407	}
	408
476df259 HX	409	static inline u8 padlock_xcrypt_cbc(const u8 input, u8 output, void key,
476df259 HX	410	u8 iv, void control_word, u32 count)
28e8c3ad HX	411	{
	412	/* Enforce key reload. */
	413	asm volatile ("pushfl; popfl");
	414	/* rep xcryptcbc */
	415	asm volatile (".byte 0xf3,0x0f,0xa7,0xd0"
	416	: "+S" (input), "+D" (output), "+a" (iv)
	417	: "d" (control_word), "b" (key), "c" (count));
476df259	418	return iv;
28e8c3ad HX	419	}
28e8c3ad HX	420
6c2bb98b	421	static void aes_encrypt(struct crypto_tfm tfm, u8 out, const u8 *in)
1da177e4	422	{
6c2bb98b	423	struct aes_ctx *ctx = aes_ctx(tfm);
6789b2dc	424	padlock_xcrypt_ecb(in, out, ctx->E, &ctx->cword.encrypt, 1);
1da177e4 LT	425	}
1da177e4 LT	426
6c2bb98b	427	static void aes_decrypt(struct crypto_tfm tfm, u8 out, const u8 *in)
1da177e4	428	{
6c2bb98b	429	struct aes_ctx *ctx = aes_ctx(tfm);
6789b2dc	430	padlock_xcrypt_ecb(in, out, ctx->D, &ctx->cword.decrypt, 1);
1da177e4 LT	431	}
1da177e4 LT	432
28e8c3ad HX	433	static unsigned int aes_encrypt_ecb(const struct cipher_desc desc, u8 out,
	434	const u8 *in, unsigned int nbytes)
	435	{
6c2bb98b	436	struct aes_ctx *ctx = aes_ctx(desc->tfm);
28e8c3ad HX	437	padlock_xcrypt_ecb(in, out, ctx->E, &ctx->cword.encrypt,
	438	nbytes / AES_BLOCK_SIZE);
	439	return nbytes & ~(AES_BLOCK_SIZE - 1);
	440	}
	441
	442	static unsigned int aes_decrypt_ecb(const struct cipher_desc desc, u8 out,
	443	const u8 *in, unsigned int nbytes)
	444	{
6c2bb98b	445	struct aes_ctx *ctx = aes_ctx(desc->tfm);
28e8c3ad HX	446	padlock_xcrypt_ecb(in, out, ctx->D, &ctx->cword.decrypt,
	447	nbytes / AES_BLOCK_SIZE);
	448	return nbytes & ~(AES_BLOCK_SIZE - 1);
	449	}
	450
	451	static unsigned int aes_encrypt_cbc(const struct cipher_desc desc, u8 out,
	452	const u8 *in, unsigned int nbytes)
	453	{
6c2bb98b	454	struct aes_ctx *ctx = aes_ctx(desc->tfm);
476df259 HX	455	u8 *iv;
	456
	457	iv = padlock_xcrypt_cbc(in, out, ctx->E, desc->info,
	458	&ctx->cword.encrypt, nbytes / AES_BLOCK_SIZE);
	459	memcpy(desc->info, iv, AES_BLOCK_SIZE);
	460
28e8c3ad HX	461	return nbytes & ~(AES_BLOCK_SIZE - 1);
	462	}
	463
	464	static unsigned int aes_decrypt_cbc(const struct cipher_desc desc, u8 out,
	465	const u8 *in, unsigned int nbytes)
	466	{
6c2bb98b	467	struct aes_ctx *ctx = aes_ctx(desc->tfm);
28e8c3ad HX	468	padlock_xcrypt_cbc(in, out, ctx->D, desc->info, &ctx->cword.decrypt,
	469	nbytes / AES_BLOCK_SIZE);
	470	return nbytes & ~(AES_BLOCK_SIZE - 1);
	471	}
	472
1da177e4 LT	473	static struct crypto_alg aes_alg = {
1da177e4 LT	474	.cra_name = "aes",
c8a19c91 HX	475	.cra_driver_name = "aes-padlock",
c8a19c91 HX	476	.cra_priority = 300,
1da177e4 LT	477	.cra_flags = CRYPTO_ALG_TYPE_CIPHER,
1da177e4 LT	478	.cra_blocksize = AES_BLOCK_SIZE,
fbdae9f3	479	.cra_ctxsize = sizeof(struct aes_ctx),
6789b2dc	480	.cra_alignmask = PADLOCK_ALIGNMENT - 1,
1da177e4 LT	481	.cra_module = THIS_MODULE,
	482	.cra_list = LIST_HEAD_INIT(aes_alg.cra_list),
	483	.cra_u = {
	484	.cipher = {
	485	.cia_min_keysize = AES_MIN_KEY_SIZE,
	486	.cia_max_keysize = AES_MAX_KEY_SIZE,
	487	.cia_setkey = aes_set_key,
	488	.cia_encrypt = aes_encrypt,
28e8c3ad HX	489	.cia_decrypt = aes_decrypt,
	490	.cia_encrypt_ecb = aes_encrypt_ecb,
	491	.cia_decrypt_ecb = aes_decrypt_ecb,
	492	.cia_encrypt_cbc = aes_encrypt_cbc,
	493	.cia_decrypt_cbc = aes_decrypt_cbc,
1da177e4 LT	494	}
	495	}
	496	};
	497
	498	int __init padlock_init_aes(void)
	499	{
	500	printk(KERN_NOTICE PFX "Using VIA PadLock ACE for AES algorithm.\n");
	501
	502	gen_tabs();
	503	return crypto_register_alg(&aes_alg);
	504	}
	505
	506	void __exit padlock_fini_aes(void)
	507	{
	508	crypto_unregister_alg(&aes_alg);
	509	}