diff options
author | Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> | 2013-10-23 18:36:10 +0300 |
---|---|---|
committer | Jussi Kivilinna <jussi.kivilinna@iki.fi> | 2013-10-23 18:36:10 +0300 |
commit | 2fd83faa876d0be91ab7884b1a9eaa7793559eb9 (patch) | |
tree | bb0111c6e9c86ad003aac73764a726b406629265 /cipher/cast5.c | |
parent | 0b39fce7e3ce6761d6bd5195d093ec6857edb7c2 (diff) | |
download | libgcrypt-2fd83faa876d0be91ab7884b1a9eaa7793559eb9.tar.gz |
Enable assembler optimizations on earlier ARM cores
* cipher/blowfish-armv6.S => cipher/blowfish-arm.S: adapt to pre-armv6 CPUs.
* cipher/blowfish.c: enable assembly on armv4/armv5 little-endian CPUs.
* cipher/camellia-armv6.S => cipher/camellia-arm.S: adapt to pre-armv6 CPUs.
* cipher/camellia.c, cipher-camellia-glue.c: enable assembly on armv4/armv5
little-endian CPUs.
* cipher/cast5-armv6.S => cipher/cast5-arm.S: adapt to pre-armv6 CPUs.
* cipher/cast5.c: enable assembly on armv4/armv5 little-endian CPUs.
* cipher/rijndael-armv6.S => cipher/rijndael-arm.S: adapt to pre-armv6 CPUs.
* cipher/rijndael.c: enable assembly on armv4/armv5 little-endian CPUs.
* cipher/twofish-armv6.S => cipher/twofish-arm.S: adapt to pre-armv6 CPUs.
* cipher/twofish.c: enable assembly on armv4/armv5 little-endian CPUs.
--
Our ARMv6 assembly optimized code can be easily adapted to earlier CPUs.
The only incompatible place is rev instruction used to do byte swapping.
Replace it on <= ARMv6 with a series of 4 instructions.
Compare:
ECB/Stream CBC CFB OFB CTR
--------------- --------------- --------------- --------------- ---------------
AES 620ms 610ms 650ms 680ms 620ms 630ms 660ms 660ms 630ms 630ms
CAMELLIA128 720ms 720ms 780ms 790ms 770ms 760ms 780ms 780ms 770ms 760ms
CAMELLIA256 910ms 910ms 970ms 970ms 960ms 950ms 970ms 970ms 960ms 950ms
CAST5 820ms 820ms 930ms 920ms 890ms 860ms 930ms 920ms 880ms 890ms
BLOWFISH 550ms 560ms 650ms 660ms 630ms 600ms 660ms 650ms 610ms 620ms
ECB/Stream CBC CFB OFB CTR
--------------- --------------- --------------- --------------- ---------------
AES 130ms 140ms 180ms 200ms 160ms 170ms 190ms 200ms 170ms 170ms
CAMELLIA128 150ms 160ms 210ms 220ms 200ms 190ms 210ms 220ms 190ms 190ms
CAMELLIA256 180ms 180ms 260ms 240ms 240ms 230ms 250ms 250ms 230ms 230ms
CAST5 170ms 160ms 270ms 120ms 240ms 130ms 260ms 270ms 130ms 120ms
BLOWFISH 160ms 150ms 260ms 110ms 230ms 120ms 250ms 260ms 110ms 120ms
Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
[ jk: in camellia.h and twofish.c, USE_ARMV6_ASM => USE_ARM_ASM ]
[ jk: fix blowfish-arm.S when __ARM_FEATURE_UNALIGNED defined ]
[ jk: in twofish.S remove defined(HAVE_ARM_ARCH_V6) ]
[ jk: ARMv6 => ARM in comments ]
Diffstat (limited to 'cipher/cast5.c')
-rw-r--r-- | cipher/cast5.c | 46 |
1 files changed, 23 insertions, 23 deletions
diff --git a/cipher/cast5.c b/cipher/cast5.c index 92d9af8c..8c016d7c 100644 --- a/cipher/cast5.c +++ b/cipher/cast5.c @@ -52,11 +52,11 @@ # define USE_AMD64_ASM 1 #endif -/* USE_ARMV6_ASM indicates whether to use ARMv6 assembly code. */ -#undef USE_ARMV6_ASM -#if defined(HAVE_ARM_ARCH_V6) && defined(__ARMEL__) +/* USE_ARM_ASM indicates whether to use ARM assembly code. */ +#undef USE_ARM_ASM +#if defined(__ARMEL__) # ifdef HAVE_COMPATIBLE_GCC_ARM_PLATFORM_AS -# define USE_ARMV6_ASM 1 +# define USE_ARM_ASM 1 # endif #endif @@ -65,7 +65,7 @@ typedef struct { u32 Km[16]; byte Kr[16]; -#ifdef USE_ARMV6_ASM +#ifdef USE_ARM_ASM u32 Kr_arm_enc[16 / sizeof(u32)]; u32 Kr_arm_dec[16 / sizeof(u32)]; #endif @@ -400,35 +400,35 @@ decrypt_block (void *context, byte *outbuf, const byte *inbuf) return /*burn_stack*/ (2*8); } -#elif defined(USE_ARMV6_ASM) +#elif defined(USE_ARM_ASM) -/* ARMv6 assembly implementations of CAST5. */ -extern void _gcry_cast5_armv6_encrypt_block(CAST5_context *c, byte *outbuf, +/* ARM assembly implementations of CAST5. */ +extern void _gcry_cast5_arm_encrypt_block(CAST5_context *c, byte *outbuf, const byte *inbuf); -extern void _gcry_cast5_armv6_decrypt_block(CAST5_context *c, byte *outbuf, +extern void _gcry_cast5_arm_decrypt_block(CAST5_context *c, byte *outbuf, const byte *inbuf); /* These assembly implementations process two blocks in parallel. */ -extern void _gcry_cast5_armv6_ctr_enc(CAST5_context *ctx, byte *out, +extern void _gcry_cast5_arm_ctr_enc(CAST5_context *ctx, byte *out, const byte *in, byte *ctr); -extern void _gcry_cast5_armv6_cbc_dec(CAST5_context *ctx, byte *out, +extern void _gcry_cast5_arm_cbc_dec(CAST5_context *ctx, byte *out, const byte *in, byte *iv); -extern void _gcry_cast5_armv6_cfb_dec(CAST5_context *ctx, byte *out, +extern void _gcry_cast5_arm_cfb_dec(CAST5_context *ctx, byte *out, const byte *in, byte *iv); static void do_encrypt_block (CAST5_context *context, byte *outbuf, const byte *inbuf) { - _gcry_cast5_armv6_encrypt_block (context, outbuf, inbuf); + _gcry_cast5_arm_encrypt_block (context, outbuf, inbuf); } static void do_decrypt_block (CAST5_context *context, byte *outbuf, const byte *inbuf) { - _gcry_cast5_armv6_decrypt_block (context, outbuf, inbuf); + _gcry_cast5_arm_decrypt_block (context, outbuf, inbuf); } static unsigned int @@ -447,7 +447,7 @@ decrypt_block (void *context, byte *outbuf, const byte *inbuf) return /*burn_stack*/ (10*4); } -#else /*USE_ARMV6_ASM*/ +#else /*USE_ARM_ASM*/ #define F1(D,m,r) ( (I = ((m) + (D))), (I=rol(I,(r))), \ (((s1[I >> 24] ^ s2[(I>>16)&0xff]) - s3[(I>>8)&0xff]) + s4[I&0xff]) ) @@ -556,7 +556,7 @@ decrypt_block (void *context, byte *outbuf, const byte *inbuf) return /*burn_stack*/ (20+4*sizeof(void*)); } -#endif /*!USE_ARMV6_ASM*/ +#endif /*!USE_ARM_ASM*/ /* Bulk encryption of complete blocks in CTR mode. This function is only @@ -592,12 +592,12 @@ _gcry_cast5_ctr_enc(void *context, unsigned char *ctr, void *outbuf_arg, /* Use generic code to handle smaller chunks... */ /* TODO: use caching instead? */ } -#elif defined(USE_ARMV6_ASM) +#elif defined(USE_ARM_ASM) { /* Process data in 2 block chunks. */ while (nblocks >= 2) { - _gcry_cast5_armv6_ctr_enc(ctx, outbuf, inbuf, ctr); + _gcry_cast5_arm_ctr_enc(ctx, outbuf, inbuf, ctr); nblocks -= 2; outbuf += 2 * CAST5_BLOCKSIZE; @@ -660,12 +660,12 @@ _gcry_cast5_cbc_dec(void *context, unsigned char *iv, void *outbuf_arg, /* Use generic code to handle smaller chunks... */ } -#elif defined(USE_ARMV6_ASM) +#elif defined(USE_ARM_ASM) { /* Process data in 2 block chunks. */ while (nblocks >= 2) { - _gcry_cast5_armv6_cbc_dec(ctx, outbuf, inbuf, iv); + _gcry_cast5_arm_cbc_dec(ctx, outbuf, inbuf, iv); nblocks -= 2; outbuf += 2 * CAST5_BLOCKSIZE; @@ -722,12 +722,12 @@ _gcry_cast5_cfb_dec(void *context, unsigned char *iv, void *outbuf_arg, /* Use generic code to handle smaller chunks... */ } -#elif defined(USE_ARMV6_ASM) +#elif defined(USE_ARM_ASM) { /* Process data in 2 block chunks. */ while (nblocks >= 2) { - _gcry_cast5_armv6_cfb_dec(ctx, outbuf, inbuf, iv); + _gcry_cast5_arm_cfb_dec(ctx, outbuf, inbuf, iv); nblocks -= 2; outbuf += 2 * CAST5_BLOCKSIZE; @@ -936,7 +936,7 @@ do_cast_setkey( CAST5_context *c, const byte *key, unsigned keylen ) for(i=0; i < 16; i++ ) c->Kr[i] = k[i] & 0x1f; -#ifdef USE_ARMV6_ASM +#ifdef USE_ARM_ASM for (i = 0; i < 4; i++) { byte Kr_arm[4]; |