diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2008-01-25 08:38:25 -0800 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2008-01-25 08:38:25 -0800 |
commit | eba0e319c12fb098d66316a8eafbaaa9174a07c3 (patch) | |
tree | b2703117db9e36bb3510654efd55361f61c54742 /arch | |
parent | df8dc74e8a383eaf2d9b44b80a71ec6f0e52b42e (diff) | |
parent | 15e7b4452b72ae890f2fcb027b4c4fa63a1c9a7a (diff) |
Merge git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6
* git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: (125 commits)
[CRYPTO] twofish: Merge common glue code
[CRYPTO] hifn_795x: Fixup container_of() usage
[CRYPTO] cast6: inline bloat--
[CRYPTO] api: Set default CRYPTO_MINALIGN to unsigned long long
[CRYPTO] tcrypt: Make xcbc available as a standalone test
[CRYPTO] xcbc: Remove bogus hash/cipher test
[CRYPTO] xcbc: Fix algorithm leak when block size check fails
[CRYPTO] tcrypt: Zero axbuf in the right function
[CRYPTO] padlock: Only reset the key once for each CBC and ECB operation
[CRYPTO] api: Include sched.h for cond_resched in scatterwalk.h
[CRYPTO] salsa20-asm: Remove unnecessary dependency on CRYPTO_SALSA20
[CRYPTO] tcrypt: Add select of AEAD
[CRYPTO] salsa20: Add x86-64 assembly version
[CRYPTO] salsa20_i586: Salsa20 stream cipher algorithm (i586 version)
[CRYPTO] gcm: Introduce rfc4106
[CRYPTO] api: Show async type
[CRYPTO] chainiv: Avoid lock spinning where possible
[CRYPTO] seqiv: Add select AEAD in Kconfig
[CRYPTO] scatterwalk: Handle zero nbytes in scatterwalk_map_and_copy
[CRYPTO] null: Allow setkey on digest_null
...
Diffstat (limited to 'arch')
-rw-r--r-- | arch/s390/crypto/aes_s390.c | 227 | ||||
-rw-r--r-- | arch/x86/crypto/Makefile | 12 | ||||
-rw-r--r-- | arch/x86/crypto/aes-i586-asm_32.S | 89 | ||||
-rw-r--r-- | arch/x86/crypto/aes-x86_64-asm_64.S | 68 | ||||
-rw-r--r-- | arch/x86/crypto/aes_32.c | 515 | ||||
-rw-r--r-- | arch/x86/crypto/aes_64.c | 336 | ||||
-rw-r--r-- | arch/x86/crypto/aes_glue.c | 57 | ||||
-rw-r--r-- | arch/x86/crypto/salsa20-i586-asm_32.S | 1114 | ||||
-rw-r--r-- | arch/x86/crypto/salsa20-x86_64-asm_64.S | 920 | ||||
-rw-r--r-- | arch/x86/crypto/salsa20_glue.c | 129 | ||||
-rw-r--r-- | arch/x86/crypto/twofish_64.c | 97 | ||||
-rw-r--r-- | arch/x86/crypto/twofish_glue.c (renamed from arch/x86/crypto/twofish_32.c) | 8 |
12 files changed, 2509 insertions, 1063 deletions
diff --git a/arch/s390/crypto/aes_s390.c b/arch/s390/crypto/aes_s390.c index 512669691ad..46c97058ebe 100644 --- a/arch/s390/crypto/aes_s390.c +++ b/arch/s390/crypto/aes_s390.c @@ -6,6 +6,7 @@ * s390 Version: * Copyright IBM Corp. 2005,2007 * Author(s): Jan Glauber (jang@de.ibm.com) + * Sebastian Siewior (sebastian@breakpoint.cc> SW-Fallback * * Derived from "crypto/aes_generic.c" * @@ -16,17 +17,13 @@ * */ +#include <crypto/aes.h> #include <crypto/algapi.h> +#include <linux/err.h> #include <linux/module.h> #include <linux/init.h> #include "crypt_s390.h" -#define AES_MIN_KEY_SIZE 16 -#define AES_MAX_KEY_SIZE 32 - -/* data block size for all key lengths */ -#define AES_BLOCK_SIZE 16 - #define AES_KEYLEN_128 1 #define AES_KEYLEN_192 2 #define AES_KEYLEN_256 4 @@ -39,45 +36,89 @@ struct s390_aes_ctx { long enc; long dec; int key_len; + union { + struct crypto_blkcipher *blk; + struct crypto_cipher *cip; + } fallback; }; -static int aes_set_key(struct crypto_tfm *tfm, const u8 *in_key, - unsigned int key_len) +/* + * Check if the key_len is supported by the HW. + * Returns 0 if it is, a positive number if it is not and software fallback is + * required or a negative number in case the key size is not valid + */ +static int need_fallback(unsigned int key_len) { - struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); - u32 *flags = &tfm->crt_flags; - switch (key_len) { case 16: if (!(keylen_flag & AES_KEYLEN_128)) - goto fail; + return 1; break; case 24: if (!(keylen_flag & AES_KEYLEN_192)) - goto fail; - + return 1; break; case 32: if (!(keylen_flag & AES_KEYLEN_256)) - goto fail; + return 1; break; default: - goto fail; + return -1; break; } + return 0; +} + +static int setkey_fallback_cip(struct crypto_tfm *tfm, const u8 *in_key, + unsigned int key_len) +{ + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + int ret; + + sctx->fallback.blk->base.crt_flags &= ~CRYPTO_TFM_REQ_MASK; + sctx->fallback.blk->base.crt_flags |= (tfm->crt_flags & + CRYPTO_TFM_REQ_MASK); + + ret = crypto_cipher_setkey(sctx->fallback.cip, in_key, key_len); + if (ret) { + tfm->crt_flags &= ~CRYPTO_TFM_RES_MASK; + tfm->crt_flags |= (sctx->fallback.blk->base.crt_flags & + CRYPTO_TFM_RES_MASK); + } + return ret; +} + +static int aes_set_key(struct crypto_tfm *tfm, const u8 *in_key, + unsigned int key_len) +{ + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + u32 *flags = &tfm->crt_flags; + int ret; + + ret = need_fallback(key_len); + if (ret < 0) { + *flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; + return -EINVAL; + } sctx->key_len = key_len; - memcpy(sctx->key, in_key, key_len); - return 0; -fail: - *flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; - return -EINVAL; + if (!ret) { + memcpy(sctx->key, in_key, key_len); + return 0; + } + + return setkey_fallback_cip(tfm, in_key, key_len); } static void aes_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in) { const struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + if (unlikely(need_fallback(sctx->key_len))) { + crypto_cipher_encrypt_one(sctx->fallback.cip, out, in); + return; + } + switch (sctx->key_len) { case 16: crypt_s390_km(KM_AES_128_ENCRYPT, &sctx->key, out, in, @@ -98,6 +139,11 @@ static void aes_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in) { const struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + if (unlikely(need_fallback(sctx->key_len))) { + crypto_cipher_decrypt_one(sctx->fallback.cip, out, in); + return; + } + switch (sctx->key_len) { case 16: crypt_s390_km(KM_AES_128_DECRYPT, &sctx->key, out, in, @@ -114,6 +160,29 @@ static void aes_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in) } } +static int fallback_init_cip(struct crypto_tfm *tfm) +{ + const char *name = tfm->__crt_alg->cra_name; + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + + sctx->fallback.cip = crypto_alloc_cipher(name, 0, + CRYPTO_ALG_ASYNC | CRYPTO_ALG_NEED_FALLBACK); + + if (IS_ERR(sctx->fallback.cip)) { + printk(KERN_ERR "Error allocating fallback algo %s\n", name); + return PTR_ERR(sctx->fallback.blk); + } + + return 0; +} + +static void fallback_exit_cip(struct crypto_tfm *tfm) +{ + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + + crypto_free_cipher(sctx->fallback.cip); + sctx->fallback.cip = NULL; +} static struct crypto_alg aes_alg = { .cra_name = "aes", @@ -125,6 +194,8 @@ static struct crypto_alg aes_alg = { .cra_ctxsize = sizeof(struct s390_aes_ctx), .cra_module = THIS_MODULE, .cra_list = LIST_HEAD_INIT(aes_alg.cra_list), + .cra_init = fallback_init_cip, + .cra_exit = fallback_exit_cip, .cra_u = { .cipher = { .cia_min_keysize = AES_MIN_KEY_SIZE, @@ -136,10 +207,70 @@ static struct crypto_alg aes_alg = { } }; +static int setkey_fallback_blk(struct crypto_tfm *tfm, const u8 *key, + unsigned int len) +{ + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + unsigned int ret; + + sctx->fallback.blk->base.crt_flags &= ~CRYPTO_TFM_REQ_MASK; + sctx->fallback.blk->base.crt_flags |= (tfm->crt_flags & + CRYPTO_TFM_REQ_MASK); + + ret = crypto_blkcipher_setkey(sctx->fallback.blk, key, len); + if (ret) { + tfm->crt_flags &= ~CRYPTO_TFM_RES_MASK; + tfm->crt_flags |= (sctx->fallback.blk->base.crt_flags & + CRYPTO_TFM_RES_MASK); + } + return ret; +} + +static int fallback_blk_dec(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + unsigned int ret; + struct crypto_blkcipher *tfm; + struct s390_aes_ctx *sctx = crypto_blkcipher_ctx(desc->tfm); + + tfm = desc->tfm; + desc->tfm = sctx->fallback.blk; + + ret = crypto_blkcipher_decrypt_iv(desc, dst, src, nbytes); + + desc->tfm = tfm; + return ret; +} + +static int fallback_blk_enc(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + unsigned int ret; + struct crypto_blkcipher *tfm; + struct s390_aes_ctx *sctx = crypto_blkcipher_ctx(desc->tfm); + + tfm = desc->tfm; + desc->tfm = sctx->fallback.blk; + + ret = crypto_blkcipher_encrypt_iv(desc, dst, src, nbytes); + + desc->tfm = tfm; + return ret; +} + static int ecb_aes_set_key(struct crypto_tfm *tfm, const u8 *in_key, unsigned int key_len) { struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + int ret; + + ret = need_fallback(key_len); + if (ret > 0) { + sctx->key_len = key_len; + return setkey_fallback_blk(tfm, in_key, key_len); + } switch (key_len) { case 16: @@ -188,6 +319,9 @@ static int ecb_aes_encrypt(struct blkcipher_desc *desc, struct s390_aes_ctx *sctx = crypto_blkcipher_ctx(desc->tfm); struct blkcipher_walk walk; + if (unlikely(need_fallback(sctx->key_len))) + return fallback_blk_enc(desc, dst, src, nbytes); + blkcipher_walk_init(&walk, dst, src, nbytes); return ecb_aes_crypt(desc, sctx->enc, sctx->key, &walk); } @@ -199,10 +333,37 @@ static int ecb_aes_decrypt(struct blkcipher_desc *desc, struct s390_aes_ctx *sctx = crypto_blkcipher_ctx(desc->tfm); struct blkcipher_walk walk; + if (unlikely(need_fallback(sctx->key_len))) + return fallback_blk_dec(desc, dst, src, nbytes); + blkcipher_walk_init(&walk, dst, src, nbytes); return ecb_aes_crypt(desc, sctx->dec, sctx->key, &walk); } +static int fallback_init_blk(struct crypto_tfm *tfm) +{ + const char *name = tfm->__crt_alg->cra_name; + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + + sctx->fallback.blk = crypto_alloc_blkcipher(name, 0, + CRYPTO_ALG_ASYNC | CRYPTO_ALG_NEED_FALLBACK); + + if (IS_ERR(sctx->fallback.blk)) { + printk(KERN_ERR "Error allocating fallback algo %s\n", name); + return PTR_ERR(sctx->fallback.blk); + } + + return 0; +} + +static void fallback_exit_blk(struct crypto_tfm *tfm) +{ + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + + crypto_free_blkcipher(sctx->fallback.blk); + sctx->fallback.blk = NULL; +} + static struct crypto_alg ecb_aes_alg = { .cra_name = "ecb(aes)", .cra_driver_name = "ecb-aes-s390", @@ -214,6 +375,8 @@ static struct crypto_alg ecb_aes_alg = { .cra_type = &crypto_blkcipher_type, .cra_module = THIS_MODULE, .cra_list = LIST_HEAD_INIT(ecb_aes_alg.cra_list), + .cra_init = fallback_init_blk, + .cra_exit = fallback_exit_blk, .cra_u = { .blkcipher = { .min_keysize = AES_MIN_KEY_SIZE, @@ -229,6 +392,13 @@ static int cbc_aes_set_key(struct crypto_tfm *tfm, const u8 *in_key, unsigned int key_len) { struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + int ret; + + ret = need_fallback(key_len); + if (ret > 0) { + sctx->key_len = key_len; + return setkey_fallback_blk(tfm, in_key, key_len); + } switch (key_len) { case 16: @@ -283,6 +453,9 @@ static int cbc_aes_encrypt(struct blkcipher_desc *desc, struct s390_aes_ctx *sctx = crypto_blkcipher_ctx(desc->tfm); struct blkcipher_walk walk; + if (unlikely(need_fallback(sctx->key_len))) + return fallback_blk_enc(desc, dst, src, nbytes); + blkcipher_walk_init(&walk, dst, src, nbytes); return cbc_aes_crypt(desc, sctx->enc, sctx->iv, &walk); } @@ -294,6 +467,9 @@ static int cbc_aes_decrypt(struct blkcipher_desc *desc, struct s390_aes_ctx *sctx = crypto_blkcipher_ctx(desc->tfm); struct blkcipher_walk walk; + if (unlikely(need_fallback(sctx->key_len))) + return fallback_blk_dec(desc, dst, src, nbytes); + blkcipher_walk_init(&walk, dst, src, nbytes); return cbc_aes_crypt(desc, sctx->dec, sctx->iv, &walk); } @@ -309,6 +485,8 @@ static struct crypto_alg cbc_aes_alg = { .cra_type = &crypto_blkcipher_type, .cra_module = THIS_MODULE, .cra_list = LIST_HEAD_INIT(cbc_aes_alg.cra_list), + .cra_init = fallback_init_blk, + .cra_exit = fallback_exit_blk, .cra_u = { .blkcipher = { .min_keysize = AES_MIN_KEY_SIZE, @@ -336,14 +514,10 @@ static int __init aes_init(void) return -EOPNOTSUPP; /* z9 109 and z9 BC/EC only support 128 bit key length */ - if (keylen_flag == AES_KEYLEN_128) { - aes_alg.cra_u.cipher.cia_max_keysize = AES_MIN_KEY_SIZE; - ecb_aes_alg.cra_u.blkcipher.max_keysize = AES_MIN_KEY_SIZE; - cbc_aes_alg.cra_u.blkcipher.max_keysize = AES_MIN_KEY_SIZE; + if (keylen_flag == AES_KEYLEN_128) printk(KERN_INFO "aes_s390: hardware acceleration only available for" "128 bit keys\n"); - } ret = crypto_register_alg(&aes_alg); if (ret) @@ -382,4 +556,3 @@ MODULE_ALIAS("aes"); MODULE_DESCRIPTION("Rijndael (AES) Cipher Algorithm"); MODULE_LICENSE("GPL"); - diff --git a/arch/x86/crypto/Makefile b/arch/x86/crypto/Makefile index 46bb609e244..3874c2de540 100644 --- a/arch/x86/crypto/Makefile +++ b/arch/x86/crypto/Makefile @@ -4,12 +4,16 @@ obj-$(CONFIG_CRYPTO_AES_586) += aes-i586.o obj-$(CONFIG_CRYPTO_TWOFISH_586) += twofish-i586.o +obj-$(CONFIG_CRYPTO_SALSA20_586) += salsa20-i586.o obj-$(CONFIG_CRYPTO_AES_X86_64) += aes-x86_64.o obj-$(CONFIG_CRYPTO_TWOFISH_X86_64) += twofish-x86_64.o +obj-$(CONFIG_CRYPTO_SALSA20_X86_64) += salsa20-x86_64.o -aes-i586-y := aes-i586-asm_32.o aes_32.o -twofish-i586-y := twofish-i586-asm_32.o twofish_32.o +aes-i586-y := aes-i586-asm_32.o aes_glue.o +twofish-i586-y := twofish-i586-asm_32.o twofish_glue.o +salsa20-i586-y := salsa20-i586-asm_32.o salsa20_glue.o -aes-x86_64-y := aes-x86_64-asm_64.o aes_64.o -twofish-x86_64-y := twofish-x86_64-asm_64.o twofish_64.o +aes-x86_64-y := aes-x86_64-asm_64.o aes_glue.o +twofish-x86_64-y := twofish-x86_64-asm_64.o twofish_glue.o +salsa20-x86_64-y := salsa20-x86_64-asm_64.o salsa20_glue.o diff --git a/arch/x86/crypto/aes-i586-asm_32.S b/arch/x86/crypto/aes-i586-asm_32.S index f942f0c8f63..1093bede3e0 100644 --- a/arch/x86/crypto/aes-i586-asm_32.S +++ b/arch/x86/crypto/aes-i586-asm_32.S @@ -46,9 +46,9 @@ #define in_blk 16 /* offsets in crypto_tfm structure */ -#define ekey (crypto_tfm_ctx_offset + 0) -#define nrnd (crypto_tfm_ctx_offset + 256) -#define dkey (crypto_tfm_ctx_offset + 260) +#define klen (crypto_tfm_ctx_offset + 0) +#define ekey (crypto_tfm_ctx_offset + 4) +#define dkey (crypto_tfm_ctx_offset + 244) // register mapping for encrypt and decrypt subroutines @@ -221,8 +221,8 @@ .global aes_enc_blk -.extern ft_tab -.extern fl_tab +.extern crypto_ft_tab +.extern crypto_fl_tab .align 4 @@ -236,7 +236,7 @@ aes_enc_blk: 1: push %ebx mov in_blk+4(%esp),%r2 push %esi - mov nrnd(%ebp),%r3 // number of rounds + mov klen(%ebp),%r3 // key size push %edi #if ekey != 0 lea ekey(%ebp),%ebp // key pointer @@ -255,26 +255,26 @@ aes_enc_blk: sub $8,%esp // space for register saves on stack add $16,%ebp // increment to next round key - cmp $12,%r3 + cmp $24,%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 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) - fwd_rnd1( +32(%ebp) ,ft_tab) - fwd_rnd2( +48(%ebp) ,ft_tab) - fwd_rnd1( +64(%ebp) ,ft_tab) - fwd_rnd2( +80(%ebp) ,ft_tab) - fwd_rnd1( +96(%ebp) ,ft_tab) - fwd_rnd2(+112(%ebp) ,ft_tab) - fwd_rnd1(+128(%ebp) ,ft_tab) - fwd_rnd2(+144(%ebp) ,fl_tab) // last round uses a different table +2: fwd_rnd1( -64(%ebp), crypto_ft_tab) // 14 rounds for 256-bit key + fwd_rnd2( -48(%ebp), crypto_ft_tab) +3: fwd_rnd1( -32(%ebp), crypto_ft_tab) // 12 rounds for 192-bit key + fwd_rnd2( -16(%ebp), crypto_ft_tab) +4: fwd_rnd1( (%ebp), crypto_ft_tab) // 10 rounds for 128-bit key + fwd_rnd2( +16(%ebp), crypto_ft_tab) + fwd_rnd1( +32(%ebp), crypto_ft_tab) + fwd_rnd2( +48(%ebp), crypto_ft_tab) + fwd_rnd1( +64(%ebp), crypto_ft_tab) + fwd_rnd2( +80(%ebp), crypto_ft_tab) + fwd_rnd1( +96(%ebp), crypto_ft_tab) + fwd_rnd2(+112(%ebp), crypto_ft_tab) + fwd_rnd1(+128(%ebp), crypto_ft_tab) + fwd_rnd2(+144(%ebp), crypto_fl_tab) // last round uses a different table // move final values to the output array. CAUTION: the // order of these assigns rely on the register mappings @@ -297,8 +297,8 @@ aes_enc_blk: .global aes_dec_blk -.extern it_tab -.extern il_tab +.extern crypto_it_tab +.extern crypto_il_tab .align 4 @@ -312,14 +312,11 @@ aes_dec_blk: 1: push %ebx mov in_blk+4(%esp),%r2 push %esi - mov nrnd(%ebp),%r3 // number of rounds + mov klen(%ebp),%r3 // key size push %edi #if dkey != 0 lea dkey(%ebp),%ebp // key pointer #endif - mov %r3,%r0 - shl $4,%r0 - add %r0,%ebp // input four columns and xor in first round key @@ -333,27 +330,27 @@ aes_dec_blk: xor 12(%ebp),%r5 sub $8,%esp // space for register saves on stack - sub $16,%ebp // increment to next round key - cmp $12,%r3 + add $16,%ebp // increment to next round key + cmp $24,%r3 jb 4f // 10 rounds for 128-bit key - lea -32(%ebp),%ebp + 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 256-bit key - inv_rnd2( +48(%ebp), it_tab) -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) - inv_rnd1( -32(%ebp), it_tab) - inv_rnd2( -48(%ebp), it_tab) - inv_rnd1( -64(%ebp), it_tab) - inv_rnd2( -80(%ebp), it_tab) - inv_rnd1( -96(%ebp), it_tab) - inv_rnd2(-112(%ebp), it_tab) - inv_rnd1(-128(%ebp), it_tab) - inv_rnd2(-144(%ebp), il_tab) // last round uses a different table + lea 32(%ebp),%ebp + +2: inv_rnd1( -64(%ebp), crypto_it_tab) // 14 rounds for 256-bit key + inv_rnd2( -48(%ebp), crypto_it_tab) +3: inv_rnd1( -32(%ebp), crypto_it_tab) // 12 rounds for 192-bit key + inv_rnd2( -16(%ebp), crypto_it_tab) +4: inv_rnd1( (%ebp), crypto_it_tab) // 10 rounds for 128-bit key + inv_rnd2( +16(%ebp), crypto_it_tab) + inv_rnd1( +32(%ebp), crypto_it_tab) + inv_rnd2( +48(%ebp), crypto_it_tab) + inv_rnd1( +64(%ebp), crypto_it_tab) + inv_rnd2( +80(%ebp), crypto_it_tab) + inv_rnd1( +96(%ebp), crypto_it_tab) + inv_rnd2(+112(%ebp), crypto_it_tab) + inv_rnd1(+128(%ebp), crypto_it_tab) + inv_rnd2(+144(%ebp), crypto_il_tab) // last round uses a different table // move final values to the output array. CAUTION: the // order of these assigns rely on the register mappings diff --git a/arch/x86/crypto/aes-x86_64-asm_64.S b/arch/x86/crypto/aes-x86_64-asm_64.S index 26b40de4d0b..a120f526c3d 100644 --- a/arch/x86/crypto/aes-x86_64-asm_64.S +++ b/arch/x86/crypto/aes-x86_64-asm_64.S @@ -8,10 +8,10 @@ * including this sentence is retained in full. */ -.extern aes_ft_tab -.extern aes_it_tab -.extern aes_fl_tab -.extern aes_il_tab +.extern crypto_ft_tab +.extern crypto_it_tab +.extern crypto_fl_tab +.extern crypto_il_tab .text @@ -56,13 +56,13 @@ .align 8; \ FUNC: movq r1,r2; \ movq r3,r4; \ - leaq BASE+KEY+52(r8),r9; \ + leaq BASE+KEY+48+4(r8),r9; \ movq r10,r11; \ movl (r7),r5 ## E; \ movl 4(r7),r1 ## E; \ movl 8(r7),r6 ## E; \ movl 12(r7),r7 ## E; \ - movl BASE(r8),r10 ## E; \ + movl BASE+0(r8),r10 ## E; \ xorl -48(r9),r5 ## E; \ xorl -44(r9),r1 ## E; \ xorl -40(r9),r6 ## E; \ @@ -154,37 +154,37 @@ FUNC: movq r1,r2; \ /* void aes_enc_blk(stuct crypto_tfm *tfm, u8 *out, const u8 *in) */ entry(aes_enc_blk,0,enc128,enc192) - encrypt_round(aes_ft_tab,-96) - encrypt_round(aes_ft_tab,-80) -enc192: encrypt_round(aes_ft_tab,-64) - encrypt_round(aes_ft_tab,-48) -enc128: encrypt_round(aes_ft_tab,-32) - encrypt_round(aes_ft_tab,-16) - encrypt_round(aes_ft_tab, 0) - encrypt_round(aes_ft_tab, 16) - encrypt_round(aes_ft_tab, 32) - encrypt_round(aes_ft_tab, 48) - encrypt_round(aes_ft_tab, 64) - encrypt_round(aes_ft_tab, 80) - encrypt_round(aes_ft_tab, 96) - encrypt_final(aes_fl_tab,112) + encrypt_round(crypto_ft_tab,-96) + encrypt_round(crypto_ft_tab,-80) +enc192: encrypt_round(crypto_ft_tab,-64) + encrypt_round(crypto_ft_tab,-48) +enc128: encrypt_round(crypto_ft_tab,-32) + encrypt_round(crypto_ft_tab,-16) + encrypt_round(crypto_ft_tab, 0) + encrypt_round(crypto_ft_tab, 16) + encrypt_round(crypto_ft_tab, 32) + encrypt_round(crypto_ft_tab, 48) + encrypt_round(crypto_ft_tab, 64) + encrypt_round(crypto_ft_tab, 80) + encrypt_round(crypto_ft_tab, 96) + encrypt_final(crypto_fl_tab,112) return /* void aes_dec_blk(struct crypto_tfm *tfm, u8 *out, const u8 *in) */ entry(aes_dec_blk,240,dec128,dec192) - decrypt_round(aes_it_tab,-96) - decrypt_round(aes_it_tab,-80) -dec192: decrypt_round(aes_it_tab,-64) - decrypt_round(aes_it_tab,-48) -dec128: decrypt_round(aes_it_tab,-32) - decrypt_round(aes_it_tab,-16) - decrypt_round(aes_it_tab, 0) - decrypt_round(aes_it_tab, 16) - decrypt_round(aes_it_tab, 32) - decrypt_round(aes_it_tab, 48) - decrypt_round(aes_it_tab, 64) - decrypt_round(aes_it_tab, 80) - decrypt_round(aes_it_tab, 96) - decrypt_final(aes_il_tab,112) + decrypt_round(crypto_it_tab,-96) + decrypt_round(crypto_it_tab,-80) +dec192: decrypt_round(crypto_it_tab,-64) + decrypt_round(crypto_it_tab,-48) +dec128: decrypt_round(crypto_it_tab,-32) + decrypt_round(crypto_it_tab,-16) + decrypt_round(crypto_it_tab, 0) + decrypt_round(crypto_it_tab, 16) + decrypt_round(crypto_it_tab, 32) + decrypt_round(crypto_it_tab, 48) + decrypt_round(crypto_it_tab, 64) + decrypt_round(crypto_it_tab, 80) + decrypt_round(crypto_it_tab, 96) + decrypt_final(crypto_il_tab,112) return diff --git a/arch/x86/crypto/aes_32.c b/arch/x86/crypto/aes_32.c deleted file mode 100644 index 49aad9397f1..00000000000 --- a/arch/x86/crypto/aes_32.c +++ /dev/null @@ -1,515 +0,0 @@ -/* - * - * Glue Code for optimized 586 assembler version of AES - * - * Copyright (c) 2002, Dr Brian Gladman <>, 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. - * - * Copyright (c) 2003, Adam J. Richter <adam@yggdrasil.com> (conversion to - * 2.5 API). - * Copyright (c) 2003, 2004 Fruhwirth Clemens <clemens@endorphin.org> - * 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> -#include <linux/types.h> -#include <linux/crypto.h> -#include <linux/linkage.h> - -asmlinkage void aes_enc_blk(struct crypto_tfm *tfm, u8 *dst, const u8 *src); -asmlinkage void aes_dec_blk(struct crypto_tfm *tfm, u8 *dst, const u8 *src); - -#define AES_MIN_KEY_SIZE 16 -#define AES_MAX_KEY_SIZE 32 -#define AES_BLOCK_SIZE 16 -#define AES_KS_LENGTH 4 * AES_BLOCK_SIZE -#define RC_LENGTH 29 - -struct aes_ctx { - u32 ekey[AES_KS_LENGTH]; - u32 rounds; - u32 dkey[AES_KS_LENGTH]; -}; - -#define WPOLY 0x011b -#define bytes2word(b0, b1, b2, b3) \ - (((u32)(b3) << 24) | ((u32)(b2) << 16) | ((u32)(b1) << 8) | (b0)) - -/* define the finite field multiplies required for Rijndael */ -#define f2(x) ((x) ? pow[log[x] + 0x19] : 0) -#define f3(x) ((x) ? pow[log[x] + 0x01] : 0) -#define f9(x) ((x) ? pow[log[x] + 0xc7] : 0) -#define fb(x) ((x) ? pow[log[x] + 0x68] : 0) -#define fd(x) ((x) ? pow[log[x] + 0xee] : 0) -#define fe(x) ((x) ? pow[log[x] + 0xdf] : 0) -#define fi(x) ((x) ? pow[255 - log[x]]: 0) - -static inline u32 upr(u32 x, int n) -{ - return (x << 8 * n) | (x >> (32 - 8 * n)); -} - -static inline u8 bval(u32 x, int n) -{ - return x >> 8 * n; -} - -/* The forward and inverse affine transformations used in the S-box */ -#define fwd_affine(x) \ - (w = (u32)x, w ^= (w<<1)^(w<<2)^(w<<3)^(w<<4), 0x63^(u8)(w^(w>>8))) - -#define inv_affine(x) \ - (w = (u32)x, w = (w<<1)^(w<<3)^(w<<6), 0x05^(u8)(w^(w>>8))) - -static u32 rcon_tab[RC_LENGTH]; - -u32 ft_tab[4][256]; -u32 fl_tab[4][256]; -static u32 im_tab[4][256]; -u32 il_tab[4][256]; -u32 it_tab[4][256]; - -static void gen_tabs(void) -{ - u32 i, w; - u8 pow[512], log[256]; - - /* - * log and power tables for GF(2^8) finite field with - * WPOLY as modular polynomial - the simplest primitive - * root is 0x03, used here to generate the tables. - */ - i = 0; w = 1; - - do { - pow[i] = (u8)w; - pow[i + 255] = (u8)w; - log[w] = (u8)i++; - w ^= (w << 1) ^ (w & 0x80 ? WPOLY : 0); - } while (w != 1); - - for(i = 0, w = 1; i < RC_LENGTH; ++i) { - rcon_tab[i] = bytes2word(w, 0, 0, 0); - w = f2(w); - } - - for(i = 0; i < 256; ++i) { - u8 b; - - b = fwd_affine(fi((u8)i)); - w = bytes2word(f2(b), b, b, f3(b)); - - /* tables for a normal encryption round */ - ft_tab[0][i] = w; - ft_tab[1][i] = upr(w, 1); - ft_tab[2][i] = upr(w, 2); - ft_tab[3][i] = upr(w, 3); - w = bytes2word(b, 0, 0, 0); - - /* - * tables for last encryption round - * (may also be used in the key schedule) - */ - fl_tab[0][i] = w; - fl_tab[1][i] = upr(w, 1); - fl_tab[2][i] = upr(w, 2); - fl_tab[3][i] = upr(w, 3); - - b = fi(inv_affine((u8)i)); - w = bytes2word(fe(b), f9(b), fd(b), fb(b)); - - /* tables for the inverse mix column operation */ - im_tab[0][b] = w; - im_tab[1][b] = upr(w, 1); - im_tab[2][b] = upr(w, 2); - im_tab[3][b] = upr(w, 3); - - /* tables for a normal decryption round */ - it_tab[0][i] = w; - it_tab[1][i] = upr(w,1); - it_tab[2][i] = upr(w,2); - it_tab[3][i] = upr(w,3); - - w = bytes2word(b, 0, 0, 0); - - /* tables for last decryption round */ - il_tab[0][i] = w; - il_tab[1][i] = upr(w,1); - il_tab[2][i] = upr(w,2); - il_tab[3][i] = upr(w,3); - } -} - -#define four_tables(x,tab,vf,rf,c) \ -( tab[0][bval(vf(x,0,c),rf(0,c))] ^ \ - tab[1][bval(vf(x,1,c),rf(1,c))] ^ \ - tab[2][bval(vf(x,2,c),rf(2,c))] ^ \ - tab[3][bval(vf(x,3,c),rf(3,c))] \ -) - -#define vf1(x,r,c) (x) -#define rf1(r,c) (r) -#define rf2(r,c) ((r-c)&3) - -#define inv_mcol(x) four_tables(x,im_tab,vf1,rf1,0) -#define ls_box(x,c) four_tables(x,fl_tab,vf1,rf2,c) - -#define ff(x) inv_mcol(x) - -#define ke4(k,i) \ -{ \ - k[4*(i)+4] = ss[0] ^= ls_box(ss[3],3) ^ rcon_tab[i]; \ - k[4*(i)+5] = ss[1] ^= ss[0]; \ - k[4*(i)+6] = ss[2] ^= ss[1]; \ - k[4*(i)+7] = ss[3] ^= ss[2]; \ -} - -#define kel4(k,i) \ -{ \ - k[4*(i)+4] = ss[0] ^= ls_box(ss[3],3) ^ rcon_tab[i]; \ - k[4*(i)+5] = ss[1] ^= ss[0]; \ - k[4*(i)+6] = ss[2] ^= ss[1]; k[4*(i)+7] = ss[3] ^= ss[2]; \ -} - -#define ke6(k,i) \ -{ \ - k[6*(i)+ 6] = ss[0] ^= ls_box(ss[5],3) ^ rcon_tab[i]; \ - k[6*(i)+ 7] = ss[1] ^= ss[0]; \ - k[6*(i)+ 8] = ss[2] ^= ss[1]; \ - k[6*(i)+ 9] = ss[3] ^= ss[2]; \ - k[6*(i)+10] = ss[4] ^= ss[3]; \ - k[6*(i)+11] = ss[5] ^= ss[4]; \ -} - -#define kel6(k,i) \ -{ \ - k[6*(i)+ 6] = ss[0] ^= ls_box(ss[5],3) ^ rcon_tab[i]; \ - k[6*(i)+ 7] = ss[1] ^= ss[0]; \ - k[6*(i)+ 8] = ss[2] ^= ss[1]; \ - k[6*(i)+ 9] = ss[3] ^= ss[2]; \ -} - -#define ke8(k,i) \ -{ \ - k[8*(i)+ 8] = ss[0] ^= ls_box(ss[7],3) ^ rcon_tab[i]; \ - k[8*(i)+ 9] = ss[1] ^= ss[0]; \ - k[8*(i)+10] = ss[2] ^= ss[1]; \ - k[8*(i)+11] = ss[3] ^= ss[2]; \ - k[8*(i)+12] = ss[4] ^= ls_box(ss[3],0); \ - k[8*(i)+13] = ss[5] ^= ss[4]; \ - k[8*(i)+14] = ss[6] ^= ss[5]; \ - k[8*(i)+15] = ss[7] ^= ss[6]; \ -} - -#define kel8(k,i) \ -{ \ - k[8*(i)+ 8] = ss[0] ^= ls_box(ss[7],3) ^ rcon_tab[i]; \ - k[8*(i)+ 9] = ss[1] ^= ss[0]; \ - k[8*(i)+10] = ss[2] ^= ss[1]; \ - k[8*(i)+11] = ss[3] ^= ss[2]; \ -} - -#define kdf4(k,i) \ -{ \ - ss[0] = ss[0] ^ ss[2] ^ ss[1] ^ ss[3]; \ - ss[1] = ss[1] ^ ss[3]; \ - ss[2] = ss[2] ^ ss[3]; \ - ss[3] = ss[3]; \ - ss[4] = ls_box(ss[(i+3) % 4], 3) ^ rcon_tab[i]; \ - ss[i % 4] ^= ss[4]; \ - ss[4] ^= k[4*(i)]; \ - k[4*(i)+4] = ff(ss[4]); \ - ss[4] ^= k[4*(i)+1]; \ - k[4*(i)+5] = ff(ss[4]); \ - ss[4] ^= k[4*(i)+2]; \ - k[4*(i)+6] = ff(ss[4]); \ - ss[4] ^= k[4*(i)+3]; \ - k[4*(i)+7] = ff(ss[4]); \ -} - -#define kd4(k,i) \ -{ \ - ss[4] = ls_box(ss[(i+3) % 4], 3) ^ rcon_tab[i]; \ - ss[i % 4] ^= ss[4]; \ - ss[4] = ff(ss[4]); \ - k[4*(i)+4] = ss[4] ^= k[4*(i)]; \ - k[4*(i)+5] = ss[4] ^= k[4*(i)+1]; \ - k[4*(i)+6] = ss[4] ^= k[4*(i)+2]; \ - k[4*(i)+7] = ss[4] ^= k[4*(i)+3]; \ -} - -#define kdl4(k,i) \ -{ \ - ss[4] = ls_box(ss[(i+3) % 4], 3) ^ rcon_tab[i]; \ - ss[i % 4] ^= ss[4]; \ - k[4*(i)+4] = (ss[0] ^= ss[1]) ^ ss[2] ^ ss[3]; \ - k[4*(i)+5] = ss[1] ^ ss[3]; \ - k[4*(i)+6] = ss[0]; \ - k[4*(i)+7] = ss[1]; \ -} - -#define kdf6(k,i) \ -{ \ - ss[0] ^= ls_box(ss[5],3) ^ rcon_tab[i]; \ - k[6*(i)+ 6] = ff(ss[0]); \ - ss[1] ^= ss[0]; \ - k[6*(i)+ 7] = ff(ss[1]); \ - ss[2] ^= ss[1]; \ - k[6*(i)+ 8] = ff(ss[2]); \ - ss[3] ^= ss[2]; \ - k[6*(i)+ 9] = ff(ss[3]); \ - ss[4] ^= ss[3]; \ - k[6*(i)+10] = ff(ss[4]); \ - ss[5] ^= ss[4]; \ - k[6*(i)+11] = ff(ss[5]); \ -} - -#define kd6(k,i) \ -{ \ - ss[6] = ls_box(ss[5],3) ^ rcon_tab[i]; \ - ss[0] ^= ss[6]; ss[6] = ff(ss[6]); \ - k[6*(i)+ 6] = ss[6] ^= k[6*(i)]; \ - ss[1] ^= ss[0]; \ - k[6*(i)+ 7] = ss[6] ^= k[6*(i)+ 1]; \ - ss[2] ^= ss[1]; \ - k[6*(i)+ 8] = ss[6] ^= k[6*(i)+ 2]; \ - ss[3] ^= ss[2]; \ - k[6*(i)+ 9] = ss[6] ^= k[6*(i)+ 3]; \ - ss[4] ^= ss[3]; \ - k[6*(i)+10] = ss[6] ^= k[6*(i)+ 4]; \ - ss[5] ^= ss[4]; \ - k[6*(i)+11] = ss[6] ^= k[6*(i)+ 5]; \ -} - -#define kdl6(k,i) \ -{ \ - ss[0] ^= ls_box(ss[5],3) ^ rcon_tab[i]; \ - k[6*(i)+ 6] = ss[0]; \ - ss[1] ^= ss[0]; \ - k[6*(i)+ 7] = ss[1]; \ - ss[2] ^= ss[1]; \ - k[6*(i)+ 8] = ss[2]; \ - ss[3] ^= ss[2]; \ - k[6*(i)+ 9] = ss[3]; \ -} - -#define kdf8(k,i) \ -{ \ - ss[0] ^= ls_box(ss[7],3) ^ rcon_tab[i]; \ - k[8*(i)+ 8] = ff(ss[0]); \ - ss[1] ^= ss[0]; \ - k[8*(i)+ 9] = ff(ss[1]); \ - ss[2] ^= ss[1]; \ - k[8*(i)+10] = ff(ss[2]); \ - ss[3] ^= ss[2]; \ - k[8*(i)+11] = ff(ss[3]); \ - ss[4] ^= ls_box(ss[3],0); \ - k[8*(i)+12] = ff(ss[4]); \ - ss[5] ^= ss[4]; \ - k[8*(i)+13] = ff(ss[5]); \ - ss[6] ^= ss[5]; \ - k[8*(i)+14] = ff(ss[6]); \ - ss[7] ^= ss[6]; \ - k[8*(i)+15] = ff(ss[7]); \ -} - -#define kd8(k,i) \ -{ \ - u32 __g = ls_box(ss[7],3) ^ rcon_tab[i]; \ - ss[0] ^= __g; \ - __g = ff(__g); \ |