popov/cpuminer-opt-gpu
1
0
Fork 0
mirror of https://github.com/JayDDee/cpuminer-opt.git synced 2025-09-17 23:44:27 +00:00
Code Issues Packages Projects Releases Wiki Activity

v3.11.0

Browse Source
This commit is contained in:
Jay D Dee
2020-01-02 23:54:08 -05:00
parent 241bc26767
commit 3572cb53c4
118 changed files with 7030 additions and 1575 deletions
Download Patch File Download Diff File Expand all files Collapse all files

178
algo/x16/x16r-4way.c
View File

@@ -26,6 +26,11 @@
#include "algo/shabal/shabal-hash-4way.h"
#include "algo/whirlpool/sph_whirlpool.h"
#include "algo/sha/sha-hash-4way.h"
#if defined(__VAES__)
#include "algo/groestl/groestl512-hash-4way.h"
#include "algo/shavite/shavite-hash-4way.h"
#include "algo/echo/echo-hash-4way.h"
#endif
static __thread uint32_t s_ntime = UINT32_MAX;
static __thread char hashOrder[X16R_HASH_FUNC_COUNT + 1] = { 0 };
@@ -36,20 +41,26 @@ union _x16r_8way_context_overlay
{
blake512_8way_context blake;
bmw512_8way_context bmw;
hashState_groestl groestl;
skein512_8way_context skein;
jh512_8way_context jh;
keccak512_8way_context keccak;
luffa_4way_context luffa;
cube_4way_context cube;
sph_shavite512_context shavite;
simd_4way_context simd;
hashState_echo echo;
hamsi512_8way_context hamsi;
sph_fugue512_context fugue;
shabal512_8way_context shabal;
sph_whirlpool_context whirlpool;
sha512_8way_context sha512;
#if defined(__VAES__)
groestl512_4way_context groestl;
shavite512_4way_context shavite;
echo_4way_context echo;
#else
hashState_groestl groestl;
sph_shavite512_context shavite;
hashState_echo echo;
#endif
} __attribute__ ((aligned (64)));
typedef union _x16r_8way_context_overlay x16r_8way_context_overlay;
@@ -115,31 +126,42 @@ void x16r_8way_hash( void* output, const void* input )
hash7, vhash );
break;
case GROESTL:
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash0,
(const char*)in0, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash1,
(const char*)in1, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash2,
(const char*)in2, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash3,
(const char*)in3, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash4,
(const char*)in4, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash5,
(const char*)in5, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash6,
(const char*)in6, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash7,
(const char*)in7, size<<3 );
break;
#if defined(__VAES__)
intrlv_4x128( vhash, in0, in1, in2, in3, size<<3 );
groestl512_4way_init( &ctx.groestl, 64 );
groestl512_4way_update_close( &ctx.groestl, vhash, vhash, size<<3 );
dintrlv_4x128_512( hash0, hash1, hash2, hash3, vhash );
intrlv_4x128( vhash, in4, in5, in6, in7, size<<3 );
groestl512_4way_init( &ctx.groestl, 64 );
groestl512_4way_update_close( &ctx.groestl, vhash, vhash, size<<3 );
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
#else
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash0,
(const char*)in0, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash1,
(const char*)in1, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash2,
(const char*)in2, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash3,
(const char*)in3, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash4,
(const char*)in4, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash5,
(const char*)in5, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash6,
(const char*)in6, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash7,
(const char*)in7, size<<3 );
#endif
break;
case SKEIN:
skein512_8way_init( &ctx.skein );
if ( i == 0 )
@@ -203,6 +225,16 @@ void x16r_8way_hash( void* output, const void* input )
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
break;
case SHAVITE:
#if defined(__VAES__)
intrlv_4x128( vhash, in0, in1, in2, in3, size<<3 );
shavite512_4way_init( &ctx.shavite );
shavite512_4way_update_close( &ctx.shavite, vhash, vhash, 64 );
dintrlv_4x128_512( hash0, hash1, hash2, hash3, vhash );
intrlv_4x128( vhash, in4, in5, in6, in7, size<<3 );
shavite512_4way_init( &ctx.shavite );
shavite512_4way_update_close( &ctx.shavite, vhash, vhash, 64 );
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
#else
sph_shavite512_init( &ctx.shavite );
sph_shavite512( &ctx.shavite, in0, size );
sph_shavite512_close( &ctx.shavite, hash0 );
@@ -227,7 +259,8 @@ void x16r_8way_hash( void* output, const void* input )
sph_shavite512_init( &ctx.shavite );
sph_shavite512( &ctx.shavite, in7, size );
sph_shavite512_close( &ctx.shavite, hash7 );
break;
#endif
break;
case SIMD:
intrlv_4x128( vhash, in0, in1, in2, in3, size<<3 );
simd_4way_init( &ctx.simd, 512 );
@@ -239,31 +272,42 @@ void x16r_8way_hash( void* output, const void* input )
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
break;
case ECHO:
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash0,
(const BitSequence*)in0, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash1,
(const BitSequence*)in1, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash2,
(const BitSequence*)in2, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash3,
(const BitSequence*)in3, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash4,
(const BitSequence*)in4, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash5,
(const BitSequence*)in5, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash6,
(const BitSequence*)in6, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash7,
(const BitSequence*)in7, size<<3 );
break;
#if defined(__VAES__)
intrlv_4x128( vhash, in0, in1, in2, in3, size<<3 );
echo_4way_init( &ctx.echo, 512 );
echo_4way_update_close( &ctx.echo, vhash, vhash, 512 );
dintrlv_4x128_512( hash0, hash1, hash2, hash3, vhash );
intrlv_4x128( vhash, in4, in5, in6, in7, size<<3 );
echo_4way_init( &ctx.echo, 512 );
echo_4way_update_close( &ctx.echo, vhash, vhash, 512 );
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
#else
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash0,
(const BitSequence*)in0, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash1,
(const BitSequence*)in1, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash2,
(const BitSequence*)in2, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash3,
(const BitSequence*)in3, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash4,
(const BitSequence*)in4, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash5,
(const BitSequence*)in5, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash6,
(const BitSequence*)in6, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash7,
(const BitSequence*)in7, size<<3 );
#endif
break;
case HAMSI:
intrlv_8x64( vhash, in0, in1, in2, in3, in4, in5, in6, in7,
size<<3 );
@@ -463,11 +507,11 @@ void x16r_4way_hash( void* output, const void* input )
case BLAKE:
blake512_4way_init( &ctx.blake );
if ( i == 0 )
blake512_4way( &ctx.blake, input, size );
blake512_4way_update( &ctx.blake, input, size );
else
{
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
blake512_4way( &ctx.blake, vhash, size );
blake512_4way_update( &ctx.blake, vhash, size );
}
blake512_4way_close( &ctx.blake, vhash );
dintrlv_4x64_512( hash0, hash1, hash2, hash3, vhash );
@@ -475,11 +519,11 @@ void x16r_4way_hash( void* output, const void* input )
case BMW:
bmw512_4way_init( &ctx.bmw );
if ( i == 0 )
bmw512_4way( &ctx.bmw, input, size );
bmw512_4way_update( &ctx.bmw, input, size );
else
{
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
bmw512_4way( &ctx.bmw, vhash, size );
bmw512_4way_update( &ctx.bmw, vhash, size );
}
bmw512_4way_close( &ctx.bmw, vhash );
dintrlv_4x64_512( hash0, hash1, hash2, hash3, vhash );
@@ -501,11 +545,11 @@ void x16r_4way_hash( void* output, const void* input )
case SKEIN:
skein512_4way_init( &ctx.skein );
if ( i == 0 )
skein512_4way( &ctx.skein, input, size );
skein512_4way_update( &ctx.skein, input, size );
else
{
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
skein512_4way( &ctx.skein, vhash, size );
skein512_4way_update( &ctx.skein, vhash, size );
}
skein512_4way_close( &ctx.skein, vhash );
dintrlv_4x64_512( hash0, hash1, hash2, hash3, vhash );
@@ -513,11 +557,11 @@ void x16r_4way_hash( void* output, const void* input )
case JH:
jh512_4way_init( &ctx.jh );
if ( i == 0 )
jh512_4way( &ctx.jh, input, size );
jh512_4way_update( &ctx.jh, input, size );
else
{
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
jh512_4way( &ctx.jh, vhash, size );
jh512_4way_update( &ctx.jh, vhash, size );
}
jh512_4way_close( &ctx.jh, vhash );
dintrlv_4x64_512( hash0, hash1, hash2, hash3, vhash );
@@ -525,11 +569,11 @@ void x16r_4way_hash( void* output, const void* input )
case KECCAK:
keccak512_4way_init( &ctx.keccak );
if ( i == 0 )
keccak512_4way( &ctx.keccak, input, size );
keccak512_4way_update( &ctx.keccak, input, size );
else
{
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
keccak512_4way( &ctx.keccak, vhash, size );
keccak512_4way_update( &ctx.keccak, vhash, size );
}
keccak512_4way_close( &ctx.keccak, vhash );
dintrlv_4x64_512( hash0, hash1, hash2, hash3, vhash );
@@ -599,7 +643,7 @@ void x16r_4way_hash( void* output, const void* input )
case HAMSI:
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
hamsi512_4way_init( &ctx.hamsi );
hamsi512_4way( &ctx.hamsi, vhash, size );
hamsi512_4way_update( &ctx.hamsi, vhash, size );
hamsi512_4way_close( &ctx.hamsi, vhash );
dintrlv_4x64_512( hash0, hash1, hash2, hash3, vhash );
break;
@@ -620,7 +664,7 @@ void x16r_4way_hash( void* output, const void* input )
case SHABAL:
intrlv_4x32( vhash, in0, in1, in2, in3, size<<3 );
shabal512_4way_init( &ctx.shabal );
shabal512_4way( &ctx.shabal, vhash, size );
shabal512_4way_update( &ctx.shabal, vhash, size );
shabal512_4way_close( &ctx.shabal, vhash );
dintrlv_4x32_512( hash0, hash1, hash2, hash3, vhash );
break;
@@ -641,7 +685,7 @@ void x16r_4way_hash( void* output, const void* input )
case SHA_512:
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
sha512_4way_init( &ctx.sha512 );
sha512_4way( &ctx.sha512, vhash, size );
sha512_4way_update( &ctx.sha512, vhash, size );
sha512_4way_close( &ctx.sha512, vhash );
dintrlv_4x64_512( hash0, hash1, hash2, hash3, vhash );
break;

19
algo/x16/x16r-gate.c
View File

@@ -44,7 +44,7 @@ bool register_x16r_algo( algo_gate_t* gate )
gate->scanhash = (void*)&scanhash_x16r;
gate->hash = (void*)&x16r_hash;
#endif
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | AVX512_OPT;
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | AVX512_OPT | VAES_OPT;
x16_r_s_getAlgoString = (void*)&x16r_getAlgoString;
opt_target_factor = 256.0;
return true;
@@ -62,7 +62,7 @@ bool register_x16rv2_algo( algo_gate_t* gate )
gate->scanhash = (void*)&scanhash_x16rv2;
gate->hash = (void*)&x16rv2_hash;
#endif
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | AVX512_OPT;
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | AVX512_OPT | VAES_OPT;
x16_r_s_getAlgoString = (void*)&x16r_getAlgoString;
opt_target_factor = 256.0;
return true;
@@ -80,7 +80,7 @@ bool register_x16s_algo( algo_gate_t* gate )
gate->scanhash = (void*)&scanhash_x16r;
gate->hash = (void*)&x16r_hash;
#endif
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | AVX512_OPT;
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | AVX512_OPT | VAES_OPT;
x16_r_s_getAlgoString = (void*)&x16s_getAlgoString;
opt_target_factor = 256.0;
return true;
@@ -215,7 +215,7 @@ bool register_x16rt_algo( algo_gate_t* gate )
gate->scanhash = (void*)&scanhash_x16rt;
gate->hash = (void*)&x16rt_hash;
#endif
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | AVX512_OPT;
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | AVX512_OPT | VAES_OPT;
opt_target_factor = 256.0;
return true;
};
@@ -232,7 +232,7 @@ bool register_x16rt_veil_algo( algo_gate_t* gate )
gate->scanhash = (void*)&scanhash_x16rt;
gate->hash = (void*)&x16rt_hash;
#endif
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | AVX512_OPT;
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | AVX512_OPT | VAES_OPT;
gate->build_extraheader = (void*)&veil_build_extraheader;
opt_target_factor = 256.0;
return true;
@@ -262,17 +262,20 @@ bool register_x21s_algo( algo_gate_t* gate )
gate->scanhash = (void*)&scanhash_x21s_8way;
gate->hash = (void*)&x21s_8way_hash;
gate->miner_thread_init = (void*)&x21s_8way_thread_init;
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | AVX512_OPT;
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | AVX512_OPT
| VAES_OPT;
#elif defined (X21S_4WAY)
gate->scanhash = (void*)&scanhash_x21s_4way;
gate->hash = (void*)&x21s_4way_hash;
gate->miner_thread_init = (void*)&x21s_4way_thread_init;
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | SHA_OPT | AVX512_OPT;
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | SHA_OPT
| AVX512_OPT | VAES_OPT;
#else
gate->scanhash = (void*)&scanhash_x21s;
gate->hash = (void*)&x21s_hash;
gate->miner_thread_init = (void*)&x21s_thread_init;
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | SHA_OPT | AVX512_OPT;
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | SHA_OPT
| AVX512_OPT | VAES_OPT;
#endif
// gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | SHA_OPT | AVX512_OPT;
x16_r_s_getAlgoString = (void*)&x16s_getAlgoString;

82
algo/x16/x16rt-4way.c
View File

@@ -20,6 +20,11 @@
#include "algo/shabal/shabal-hash-4way.h"
#include "algo/whirlpool/sph_whirlpool.h"
#include "algo/sha/sha-hash-4way.h"
#if defined(__VAES__)
#include "algo/groestl/groestl512-hash-4way.h"
#include "algo/shavite/shavite-hash-4way.h"
#include "algo/echo/echo-hash-4way.h"
#endif
static __thread uint32_t s_ntime = UINT32_MAX;
static __thread char hashOrder[X16R_HASH_FUNC_COUNT + 1] = { 0 };
@@ -30,20 +35,26 @@ union _x16rt_8way_context_overlay
{
blake512_8way_context blake;
bmw512_8way_context bmw;
hashState_groestl groestl;
skein512_8way_context skein;
jh512_8way_context jh;
keccak512_8way_context keccak;
luffa_4way_context luffa;
cube_4way_context cube;
sph_shavite512_context shavite;
simd_4way_context simd;
hashState_echo echo;
hamsi512_8way_context hamsi;
sph_fugue512_context fugue;
shabal512_8way_context shabal;
sph_whirlpool_context whirlpool;
sha512_8way_context sha512;
#if defined(__VAES__)
groestl512_4way_context groestl;
shavite512_4way_context shavite;
echo_4way_context echo;
#else
hashState_groestl groestl;
sph_shavite512_context shavite;
hashState_echo echo;
#endif
} __attribute__ ((aligned (64)));
typedef union _x16rt_8way_context_overlay x16rt_8way_context_overlay;
@@ -109,6 +120,16 @@ void x16rt_8way_hash( void* output, const void* input )
hash7, vhash );
break;
case GROESTL:
#if defined(__VAES__)
intrlv_4x128( vhash, in0, in1, in2, in3, size<<3 );
groestl512_4way_init( &ctx.groestl, 64 );
groestl512_4way_update_close( &ctx.groestl, vhash, vhash, size<<3 );
dintrlv_4x128_512( hash0, hash1, hash2, hash3, vhash );
intrlv_4x128( vhash, in4, in5, in6, in7, size<<3 );
groestl512_4way_init( &ctx.groestl, 64 );
groestl512_4way_update_close( &ctx.groestl, vhash, vhash, size<<3 );
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
#else
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash0,
(const char*)in0, size<<3 );
@@ -133,7 +154,8 @@ void x16rt_8way_hash( void* output, const void* input )
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash7,
(const char*)in7, size<<3 );
break;
#endif
break;
case SKEIN:
skein512_8way_init( &ctx.skein );
if ( i == 0 )
@@ -197,6 +219,16 @@ void x16rt_8way_hash( void* output, const void* input )
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
break;
case SHAVITE:
#if defined(__VAES__)
intrlv_4x128( vhash, in0, in1, in2, in3, size<<3 );
shavite512_4way_init( &ctx.shavite );
shavite512_4way_update_close( &ctx.shavite, vhash, vhash, 64 );
dintrlv_4x128_512( hash0, hash1, hash2, hash3, vhash );
intrlv_4x128( vhash, in4, in5, in6, in7, size<<3 );
shavite512_4way_init( &ctx.shavite );
shavite512_4way_update_close( &ctx.shavite, vhash, vhash, 64 );
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
#else
sph_shavite512_init( &ctx.shavite );
sph_shavite512( &ctx.shavite, in0, size );
sph_shavite512_close( &ctx.shavite, hash0 );
@@ -221,7 +253,8 @@ void x16rt_8way_hash( void* output, const void* input )
sph_shavite512_init( &ctx.shavite );
sph_shavite512( &ctx.shavite, in7, size );
sph_shavite512_close( &ctx.shavite, hash7 );
break;
#endif
break;
case SIMD:
intrlv_4x128( vhash, in0, in1, in2, in3, size<<3 );
simd_4way_init( &ctx.simd, 512 );
@@ -233,6 +266,16 @@ void x16rt_8way_hash( void* output, const void* input )
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
break;
case ECHO:
#if defined(__VAES__)
intrlv_4x128( vhash, in0, in1, in2, in3, size<<3 );
echo_4way_init( &ctx.echo, 512 );
echo_4way_update_close( &ctx.echo, vhash, vhash, 512 );
dintrlv_4x128_512( hash0, hash1, hash2, hash3, vhash );
intrlv_4x128( vhash, in4, in5, in6, in7, size<<3 );
echo_4way_init( &ctx.echo, 512 );
echo_4way_update_close( &ctx.echo, vhash, vhash, 512 );
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
#else
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash0,
(const BitSequence*)in0, size<<3 );
@@ -257,7 +300,8 @@ void x16rt_8way_hash( void* output, const void* input )
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash7,
(const BitSequence*)in7, size<<3 );
break;
#endif
break;
case HAMSI:
intrlv_8x64( vhash, in0, in1, in2, in3, in4, in5, in6, in7,
size<<3 );
@@ -458,11 +502,11 @@ void x16rt_4way_hash( void* output, const void* input )
case BLAKE:
blake512_4way_init( &ctx.blake );
if ( i == 0 )
blake512_4way( &ctx.blake, input, size );
blake512_4way_update( &ctx.blake, input, size );
else
{
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
blake512_4way( &ctx.blake, vhash, size );
blake512_4way_update( &ctx.blake, vhash, size );
}
blake512_4way_close( &ctx.blake, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
@@ -470,11 +514,11 @@ void x16rt_4way_hash( void* output, const void* input )
case BMW:
bmw512_4way_init( &ctx.bmw );
if ( i == 0 )
bmw512_4way( &ctx.bmw, input, size );
bmw512_4way_update( &ctx.bmw, input, size );
else
{
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
bmw512_4way( &ctx.bmw, vhash, size );
bmw512_4way_update( &ctx.bmw, vhash, size );
}
bmw512_4way_close( &ctx.bmw, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
@@ -496,11 +540,11 @@ void x16rt_4way_hash( void* output, const void* input )
case SKEIN:
skein512_4way_init( &ctx.skein );
if ( i == 0 )
skein512_4way( &ctx.skein, input, size );
skein512_4way_update( &ctx.skein, input, size );
else
{
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
skein512_4way( &ctx.skein, vhash, size );
skein512_4way_update( &ctx.skein, vhash, size );
}
skein512_4way_close( &ctx.skein, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
@@ -508,11 +552,11 @@ void x16rt_4way_hash( void* output, const void* input )
case JH:
jh512_4way_init( &ctx.jh );
if ( i == 0 )
jh512_4way( &ctx.jh, input, size );
jh512_4way_update( &ctx.jh, input, size );
else
{
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
jh512_4way( &ctx.jh, vhash, size );
jh512_4way_update( &ctx.jh, vhash, size );
}
jh512_4way_close( &ctx.jh, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
@@ -520,11 +564,11 @@ void x16rt_4way_hash( void* output, const void* input )
case KECCAK:
keccak512_4way_init( &ctx.keccak );
if ( i == 0 )
keccak512_4way( &ctx.keccak, input, size );
keccak512_4way_update( &ctx.keccak, input, size );
else
{
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
keccak512_4way( &ctx.keccak, vhash, size );
keccak512_4way_update( &ctx.keccak, vhash, size );
}
keccak512_4way_close( &ctx.keccak, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
@@ -594,7 +638,7 @@ void x16rt_4way_hash( void* output, const void* input )
case HAMSI:
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
hamsi512_4way_init( &ctx.hamsi );
hamsi512_4way( &ctx.hamsi, vhash, size );
hamsi512_4way_update( &ctx.hamsi, vhash, size );
hamsi512_4way_close( &ctx.hamsi, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
break;
@@ -615,7 +659,7 @@ void x16rt_4way_hash( void* output, const void* input )
case SHABAL:
intrlv_4x32( vhash, in0, in1, in2, in3, size<<3 );
shabal512_4way_init( &ctx.shabal );
shabal512_4way( &ctx.shabal, vhash, size );
shabal512_4way_update( &ctx.shabal, vhash, size );
shabal512_4way_close( &ctx.shabal, vhash );
dintrlv_4x32( hash0, hash1, hash2, hash3, vhash, 512 );
break;
@@ -636,7 +680,7 @@ void x16rt_4way_hash( void* output, const void* input )
case SHA_512:
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
sha512_4way_init( &ctx.sha512 );
sha512_4way( &ctx.sha512, vhash, size );
sha512_4way_update( &ctx.sha512, vhash, size );
sha512_4way_close( &ctx.sha512, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
break;

83
algo/x16/x16rv2-4way.c
View File

@@ -27,6 +27,11 @@
#include "algo/whirlpool/sph_whirlpool.h"
#include "algo/sha/sha-hash-4way.h"
#include "algo/tiger/sph_tiger.h"
#if defined(__VAES__)
#include "algo/groestl/groestl512-hash-4way.h"
#include "algo/shavite/shavite-hash-4way.h"
#include "algo/echo/echo-hash-4way.h"
#endif
static __thread uint32_t s_ntime = UINT32_MAX;
static __thread char hashOrder[X16R_HASH_FUNC_COUNT + 1] = { 0 };
@@ -37,21 +42,30 @@ union _x16rv2_8way_context_overlay
{
blake512_8way_context blake;
bmw512_8way_context bmw;
hashState_groestl groestl;
// hashState_groestl groestl;
skein512_8way_context skein;
jh512_8way_context jh;
keccak512_8way_context keccak;
luffa_4way_context luffa;
cube_4way_context cube;
sph_shavite512_context shavite;
// sph_shavite512_context shavite;
simd_4way_context simd;
hashState_echo echo;
// hashState_echo echo;
hamsi512_8way_context hamsi;
sph_fugue512_context fugue;
shabal512_8way_context shabal;
sph_whirlpool_context whirlpool;
sha512_8way_context sha512;
sph_tiger_context tiger;
#if defined(__VAES__)
groestl512_4way_context groestl;
shavite512_4way_context shavite;
echo_4way_context echo;
#else
hashState_groestl groestl;
sph_shavite512_context shavite;
hashState_echo echo;
#endif
} __attribute__ ((aligned (64)));
typedef union _x16rv2_8way_context_overlay x16rv2_8way_context_overlay;
@@ -117,6 +131,16 @@ void x16rv2_8way_hash( void* output, const void* input )
hash7, vhash );
break;
case GROESTL:
#if defined(__VAES__)
intrlv_4x128( vhash, in0, in1, in2, in3, size<<3 );
groestl512_4way_init( &ctx.groestl, 64 );
groestl512_4way_update_close( &ctx.groestl, vhash, vhash, size<<3 );
dintrlv_4x128_512( hash0, hash1, hash2, hash3, vhash );
intrlv_4x128( vhash, in4, in5, in6, in7, size<<3 );
groestl512_4way_init( &ctx.groestl, 64 );
groestl512_4way_update_close( &ctx.groestl, vhash, vhash, size<<3 );
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
#else
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash0,
(const char*)in0, size<<3 );
@@ -141,7 +165,8 @@ void x16rv2_8way_hash( void* output, const void* input )
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash7,
(const char*)in7, size<<3 );
break;
#endif
break;
case SKEIN:
skein512_8way_init( &ctx.skein );
if ( i == 0 )
@@ -258,6 +283,16 @@ void x16rv2_8way_hash( void* output, const void* input )
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
break;
case SHAVITE:
#if defined(__VAES__)
intrlv_4x128( vhash, in0, in1, in2, in3, size<<3 );
shavite512_4way_init( &ctx.shavite );
shavite512_4way_update_close( &ctx.shavite, vhash, vhash, 64 );
dintrlv_4x128_512( hash0, hash1, hash2, hash3, vhash );
intrlv_4x128( vhash, in4, in5, in6, in7, size<<3 );
shavite512_4way_init( &ctx.shavite );
shavite512_4way_update_close( &ctx.shavite, vhash, vhash, 64 );
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
#else
sph_shavite512_init( &ctx.shavite );
sph_shavite512( &ctx.shavite, in0, size );
sph_shavite512_close( &ctx.shavite, hash0 );
@@ -282,7 +317,8 @@ void x16rv2_8way_hash( void* output, const void* input )
sph_shavite512_init( &ctx.shavite );
sph_shavite512( &ctx.shavite, in7, size );
sph_shavite512_close( &ctx.shavite, hash7 );
break;
#endif
break;
case SIMD:
intrlv_4x128( vhash, in0, in1, in2, in3, size<<3 );
simd_4way_init( &ctx.simd, 512 );
@@ -294,6 +330,16 @@ void x16rv2_8way_hash( void* output, const void* input )
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
break;
case ECHO:
#if defined(__VAES__)
intrlv_4x128( vhash, in0, in1, in2, in3, size<<3 );
echo_4way_init( &ctx.echo, 512 );
echo_4way_update_close( &ctx.echo, vhash, vhash, 512 );
dintrlv_4x128_512( hash0, hash1, hash2, hash3, vhash );
intrlv_4x128( vhash, in4, in5, in6, in7, size<<3 );
echo_4way_init( &ctx.echo, 512 );
echo_4way_update_close( &ctx.echo, vhash, vhash, 512 );
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
#else
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash0,
(const BitSequence*)in0, size<<3 );
@@ -318,7 +364,8 @@ void x16rv2_8way_hash( void* output, const void* input )
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash7,
(const BitSequence*)in7, size<<3 );
break;
#endif
break;
case HAMSI:
intrlv_8x64( vhash, in0, in1, in2, in3, in4, in5, in6, in7,
size<<3 );
@@ -553,11 +600,11 @@ void x16rv2_4way_hash( void* output, const void* input )
case BLAKE:
blake512_4way_init( &ctx.blake );
if ( i == 0 )
blake512_4way( &ctx.blake, input, size );
blake512_4way_update( &ctx.blake, input, size );
else
{
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
blake512_4way( &ctx.blake, vhash, size );
blake512_4way_update( &ctx.blake, vhash, size );
}
blake512_4way_close( &ctx.blake, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
@@ -565,11 +612,11 @@ void x16rv2_4way_hash( void* output, const void* input )
case BMW:
bmw512_4way_init( &ctx.bmw );
if ( i == 0 )
bmw512_4way( &ctx.bmw, input, size );
bmw512_4way_update( &ctx.bmw, input, size );
else
{
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
bmw512_4way( &ctx.bmw, vhash, size );
bmw512_4way_update( &ctx.bmw, vhash, size );
}
bmw512_4way_close( &ctx.bmw, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
@@ -591,11 +638,11 @@ void x16rv2_4way_hash( void* output, const void* input )
case SKEIN:
skein512_4way_init( &ctx.skein );
if ( i == 0 )
skein512_4way( &ctx.skein, input, size );
skein512_4way_update( &ctx.skein, input, size );
else
{
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
skein512_4way( &ctx.skein, vhash, size );
skein512_4way_update( &ctx.skein, vhash, size );
}
skein512_4way_close( &ctx.skein, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
@@ -603,11 +650,11 @@ void x16rv2_4way_hash( void* output, const void* input )
case JH:
jh512_4way_init( &ctx.jh );
if ( i == 0 )
jh512_4way( &ctx.jh, input, size );
jh512_4way_update( &ctx.jh, input, size );
else
{
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
jh512_4way( &ctx.jh, vhash, size );
jh512_4way_update( &ctx.jh, vhash, size );
}
jh512_4way_close( &ctx.jh, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
@@ -631,7 +678,7 @@ void x16rv2_4way_hash( void* output, const void* input )
intrlv_4x64( vhash, hash0, hash1, hash2, hash3, 512 );
keccak512_4way_init( &ctx.keccak );
keccak512_4way( &ctx.keccak, vhash, 64 );
keccak512_4way_update( &ctx.keccak, vhash, 64 );
keccak512_4way_close( &ctx.keccak, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
break;
@@ -721,7 +768,7 @@ void x16rv2_4way_hash( void* output, const void* input )
case HAMSI:
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
hamsi512_4way_init( &ctx.hamsi );
hamsi512_4way( &ctx.hamsi, vhash, size );
hamsi512_4way_update( &ctx.hamsi, vhash, size );
hamsi512_4way_close( &ctx.hamsi, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
break;
@@ -742,7 +789,7 @@ void x16rv2_4way_hash( void* output, const void* input )
case SHABAL:
intrlv_4x32( vhash, in0, in1, in2, in3, size<<3 );
shabal512_4way_init( &ctx.shabal );
shabal512_4way( &ctx.shabal, vhash, size );
shabal512_4way_update( &ctx.shabal, vhash, size );
shabal512_4way_close( &ctx.shabal, vhash );
dintrlv_4x32( hash0, hash1, hash2, hash3, vhash, 512 );
break;
@@ -779,7 +826,7 @@ void x16rv2_4way_hash( void* output, const void* input )
intrlv_4x64( vhash, hash0, hash1, hash2, hash3, 512 );
sha512_4way_init( &ctx.sha512 );
sha512_4way( &ctx.sha512, vhash, 64 );
sha512_4way_update( &ctx.sha512, vhash, 64 );
sha512_4way_close( &ctx.sha512, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
break;

183
algo/x16/x21s-4way.c
View File

@@ -30,6 +30,11 @@
#include "algo/tiger/sph_tiger.h"
#include "algo/gost/sph_gost.h"
#include "algo/lyra2/lyra2.h"
#if defined(__VAES__)
#include "algo/groestl/groestl512-hash-4way.h"
#include "algo/shavite/shavite-hash-4way.h"
#include "algo/echo/echo-hash-4way.h"
#endif
#if defined(__SHA__)
#include <openssl/sha.h>
#endif
@@ -45,15 +50,12 @@ union _x21s_8way_context_overlay
{
blake512_8way_context blake;
bmw512_8way_context bmw;
hashState_groestl groestl;
skein512_8way_context skein;
jh512_8way_context jh;
keccak512_8way_context keccak;
luffa_4way_context luffa;
cube_4way_context cube;
sph_shavite512_context shavite;
simd_4way_context simd;
hashState_echo echo;
hamsi512_8way_context hamsi;
sph_fugue512_context fugue;
shabal512_8way_context shabal;
@@ -63,6 +65,15 @@ union _x21s_8way_context_overlay
sph_tiger_context tiger;
sph_gost512_context gost;
sha256_8way_context sha256;
#if defined(__VAES__)
groestl512_4way_context groestl;
shavite512_4way_context shavite;
echo_4way_context echo;
#else
hashState_groestl groestl;
sph_shavite512_context shavite;
hashState_echo echo;
#endif
} __attribute__ ((aligned (64)));
typedef union _x21s_8way_context_overlay x21s_8way_context_overlay;
@@ -128,31 +139,42 @@ void x21s_8way_hash( void* output, const void* input )
hash7, vhash );
break;
case GROESTL:
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash0,
(const char*)in0, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash1,
(const char*)in1, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash2,
(const char*)in2, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash3,
(const char*)in3, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash4,
(const char*)in4, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash5,
(const char*)in5, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash6,
(const char*)in6, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash7,
(const char*)in7, size<<3 );
break;
#if defined(__VAES__)
intrlv_4x128( vhash, in0, in1, in2, in3, size<<3 );
groestl512_4way_init( &ctx.groestl, 64 );
groestl512_4way_update_close( &ctx.groestl, vhash, vhash, size<<3 );
dintrlv_4x128_512( hash0, hash1, hash2, hash3, vhash );
intrlv_4x128( vhash, in4, in5, in6, in7, size<<3 );
groestl512_4way_init( &ctx.groestl, 64 );
groestl512_4way_update_close( &ctx.groestl, vhash, vhash, size<<3 );
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
#else
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash0,
(const char*)in0, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash1,
(const char*)in1, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash2,
(const char*)in2, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash3,
(const char*)in3, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash4,
(const char*)in4, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash5,
(const char*)in5, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash6,
(const char*)in6, size<<3 );
init_groestl( &ctx.groestl, 64 );
update_and_final_groestl( &ctx.groestl, (char*)hash7,
(const char*)in7, size<<3 );
#endif
break;
case SKEIN:
skein512_8way_init( &ctx.skein );
if ( i == 0 )
@@ -216,6 +238,16 @@ void x21s_8way_hash( void* output, const void* input )
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
break;
case SHAVITE:
#if defined(__VAES__)
intrlv_4x128( vhash, in0, in1, in2, in3, size<<3 );
shavite512_4way_init( &ctx.shavite );
shavite512_4way_update_close( &ctx.shavite, vhash, vhash, 64 );
dintrlv_4x128_512( hash0, hash1, hash2, hash3, vhash );
intrlv_4x128( vhash, in4, in5, in6, in7, size<<3 );
shavite512_4way_init( &ctx.shavite );
shavite512_4way_update_close( &ctx.shavite, vhash, vhash, 64 );
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
#else
sph_shavite512_init( &ctx.shavite );
sph_shavite512( &ctx.shavite, in0, size );
sph_shavite512_close( &ctx.shavite, hash0 );
@@ -240,7 +272,8 @@ void x21s_8way_hash( void* output, const void* input )
sph_shavite512_init( &ctx.shavite );
sph_shavite512( &ctx.shavite, in7, size );
sph_shavite512_close( &ctx.shavite, hash7 );
break;
#endif
break;
case SIMD:
intrlv_4x128( vhash, in0, in1, in2, in3, size<<3 );
simd_4way_init( &ctx.simd, 512 );
@@ -252,31 +285,43 @@ void x21s_8way_hash( void* output, const void* input )
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
break;
case ECHO:
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash0,
(const BitSequence*)in0, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash1,
(const BitSequence*)in1, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash2,
(const BitSequence*)in2, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash3,
(const BitSequence*)in3, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash4,
(const BitSequence*)in4, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash5,
(const BitSequence*)in5, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash6,
(const BitSequence*)in6, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash7,
(const BitSequence*)in7, size<<3 );
break;
#if defined(__VAES__)
intrlv_4x128( vhash, in0, in1, in2, in3, size<<3 );
echo_4way_init( &ctx.echo, 512 );
echo_4way_update_close( &ctx.echo, vhash, vhash, 512 );
dintrlv_4x128_512( hash0, hash1, hash2, hash3, vhash );
intrlv_4x128( vhash, in4, in5, in6, in7, size<<3 );
echo_4way_init( &ctx.echo, 512 );
echo_4way_update_close( &ctx.echo, vhash, vhash, 512 );
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhash );
#else
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash0,
(const BitSequence*)in0, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash1,
(const BitSequence*)in1, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash2,
(const BitSequence*)in2, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash3,
(const BitSequence*)in3, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash4,
(const BitSequence*)in4, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash5,
(const BitSequence*)in5, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash6,
(const BitSequence*)in6, size<<3 );
init_echo( &ctx.echo, 512 );
update_final_echo ( &ctx.echo, (BitSequence *)hash7,
(const BitSequence*)in7, size<<3 );
#endif
break;
case HAMSI:
intrlv_8x64( vhash, in0, in1, in2, in3, in4, in5, in6, in7,
size<<3 );
@@ -578,11 +623,11 @@ void x21s_4way_hash( void* output, const void* input )
case BLAKE:
blake512_4way_init( &ctx.blake );
if ( i == 0 )
blake512_4way( &ctx.blake, input, size );
blake512_4way_update( &ctx.blake, input, size );
else
{
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
blake512_4way( &ctx.blake, vhash, size );
blake512_4way_update( &ctx.blake, vhash, size );
}
blake512_4way_close( &ctx.blake, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
@@ -590,11 +635,11 @@ void x21s_4way_hash( void* output, const void* input )
case BMW:
bmw512_4way_init( &ctx.bmw );
if ( i == 0 )
bmw512_4way( &ctx.bmw, input, size );
bmw512_4way_update( &ctx.bmw, input, size );
else
{
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
bmw512_4way( &ctx.bmw, vhash, size );
bmw512_4way_update( &ctx.bmw, vhash, size );
}
bmw512_4way_close( &ctx.bmw, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
@@ -616,11 +661,11 @@ void x21s_4way_hash( void* output, const void* input )
case SKEIN:
skein512_4way_init( &ctx.skein );
if ( i == 0 )
skein512_4way( &ctx.skein, input, size );
skein512_4way_update( &ctx.skein, input, size );
else
{
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
skein512_4way( &ctx.skein, vhash, size );
skein512_4way_update( &ctx.skein, vhash, size );
}
skein512_4way_close( &ctx.skein, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
@@ -628,11 +673,11 @@ void x21s_4way_hash( void* output, const void* input )
case JH:
jh512_4way_init( &ctx.jh );
if ( i == 0 )
jh512_4way( &ctx.jh, input, size );
jh512_4way_update( &ctx.jh, input, size );
else
{
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
jh512_4way( &ctx.jh, vhash, size );
jh512_4way_update( &ctx.jh, vhash, size );
}
jh512_4way_close( &ctx.jh, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
@@ -640,11 +685,11 @@ void x21s_4way_hash( void* output, const void* input )
case KECCAK:
keccak512_4way_init( &ctx.keccak );
if ( i == 0 )
keccak512_4way( &ctx.keccak, input, size );
keccak512_4way_update( &ctx.keccak, input, size );
else
{
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
keccak512_4way( &ctx.keccak, vhash, size );
keccak512_4way_update( &ctx.keccak, vhash, size );
}
keccak512_4way_close( &ctx.keccak, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
@@ -714,7 +759,7 @@ void x21s_4way_hash( void* output, const void* input )
case HAMSI:
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
hamsi512_4way_init( &ctx.hamsi );
hamsi512_4way( &ctx.hamsi, vhash, size );
hamsi512_4way_update( &ctx.hamsi, vhash, size );
hamsi512_4way_close( &ctx.hamsi, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
break;
@@ -735,7 +780,7 @@ void x21s_4way_hash( void* output, const void* input )
case SHABAL:
intrlv_4x32( vhash, in0, in1, in2, in3, size<<3 );
shabal512_4way_init( &ctx.shabal );
shabal512_4way( &ctx.shabal, vhash, size );
shabal512_4way_update( &ctx.shabal, vhash, size );
shabal512_4way_close( &ctx.shabal, vhash );
dintrlv_4x32( hash0, hash1, hash2, hash3, vhash, 512 );
break;
@@ -756,7 +801,7 @@ void x21s_4way_hash( void* output, const void* input )
case SHA_512:
intrlv_4x64( vhash, in0, in1, in2, in3, size<<3 );
sha512_4way_init( &ctx.sha512 );
sha512_4way( &ctx.sha512, vhash, size );
sha512_4way_update( &ctx.sha512, vhash, size );
sha512_4way_close( &ctx.sha512, vhash );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
break;
@@ -767,7 +812,7 @@ void x21s_4way_hash( void* output, const void* input )
intrlv_4x32( vhash, hash0, hash1, hash2, hash3, 512 );
haval256_5_4way_init( &ctx.haval );
haval256_5_4way( &ctx.haval, vhash, 64 );
haval256_5_4way_update( &ctx.haval, vhash, 64 );
haval256_5_4way_close( &ctx.haval, vhash );
dintrlv_4x32( hash0, hash1, hash2, hash3, vhash, 512 );
@@ -831,7 +876,7 @@ void x21s_4way_hash( void* output, const void* input )
intrlv_4x32( vhash, hash0, hash1, hash2, hash3, 512 );
sha256_4way_init( &ctx.sha256 );
sha256_4way( &ctx.sha256, vhash, 64 );
sha256_4way_update( &ctx.sha256, vhash, 64 );
sha256_4way_close( &ctx.sha256, vhash );
dintrlv_4x32( output, output+32, output+64,output+96, vhash, 256 );