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

v3.8.1

Browse Source
This commit is contained in:
Jay D Dee
2018-02-07 16:38:45 -05:00
parent 54b8fd7362
commit a28daca3ce
83 changed files with 5153 additions and 3924 deletions
Download Patch File Download Diff File Expand all files Collapse all files

49
algo/x11/c11-4way.c
View File

@@ -12,10 +12,10 @@
#include "algo/skein/skein-hash-4way.h"
#include "algo/jh/jh-hash-4way.h"
#include "algo/keccak/keccak-hash-4way.h"
#include "algo/luffa/sse2/luffa_for_sse2.h"
#include "algo/luffa/luffa-hash-2way.h"
#include "algo/cubehash/sse2/cubehash_sse2.h"
#include "algo/shavite/sph_shavite.h"
#include "algo/simd/sse2/nist.h"
#include "algo/simd/simd-hash-2way.h"
#include "algo/echo/aes_ni/hash_api.h"
typedef struct {
@@ -25,10 +25,10 @@ typedef struct {
skein512_4way_context skein;
jh512_4way_context jh;
keccak512_4way_context keccak;
hashState_luffa luffa;
luffa_2way_context luffa;
cubehashParam cube;
sph_shavite512_context shavite;
hashState_sd simd;
simd_2way_context simd;
hashState_echo echo;
} c11_4way_ctx_holder;
@@ -42,10 +42,10 @@ void init_c11_4way_ctx()
skein512_4way_init( &c11_4way_ctx.skein );
jh512_4way_init( &c11_4way_ctx.jh );
keccak512_4way_init( &c11_4way_ctx.keccak );
init_luffa( &c11_4way_ctx.luffa, 512 );
luffa_2way_init( &c11_4way_ctx.luffa, 512 );
cubehashInit( &c11_4way_ctx.cube, 512, 16, 32 );
sph_shavite512_init( &c11_4way_ctx.shavite );
init_sd( &c11_4way_ctx.simd, 512 );
simd_2way_init( &c11_4way_ctx.simd, 512 );
init_echo( &c11_4way_ctx.echo, 512 );
}
@@ -56,6 +56,7 @@ void c11_4way_hash( void *state, const void *input )
uint64_t hash2[8] __attribute__ ((aligned (64)));
uint64_t hash3[8] __attribute__ ((aligned (64)));
uint64_t vhash[8*4] __attribute__ ((aligned (64)));
uint64_t vhashB[8*2] __attribute__ ((aligned (64)));
c11_4way_ctx_holder ctx;
memcpy( &ctx, &c11_4way_ctx, sizeof(c11_4way_ctx) );
@@ -98,17 +99,13 @@ void c11_4way_hash( void *state, const void *input )
mm256_deinterleave_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
// 7 Luffa
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash0,
(const BitSequence*)hash0, 64 );
memcpy( &ctx.luffa, &c11_4way_ctx.luffa, sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash1,
(const BitSequence*)hash1, 64 );
memcpy( &ctx.luffa, &c11_4way_ctx.luffa, sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash2,
(const BitSequence*)hash2, 64 );
memcpy( &ctx.luffa, &c11_4way_ctx.luffa, sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash3,
(const BitSequence*)hash3, 64 );
mm256_interleave_2x128( vhash, hash0, hash1, 512 );
mm256_interleave_2x128( vhashB, hash2, hash3, 512 );
luffa_2way_update_close( &ctx.luffa, vhash, vhash, 64 );
luffa_2way_init( &ctx.luffa, 512 );
luffa_2way_update_close( &ctx.luffa, vhashB, vhashB, 64 );
mm256_deinterleave_2x128( hash0, hash1, vhash, 512 );
mm256_deinterleave_2x128( hash2, hash3, vhashB, 512 );
// 8 Cubehash
cubehashUpdateDigest( &ctx.cube, (byte*)hash0, (const byte*) hash0, 64 );
@@ -136,17 +133,13 @@ void c11_4way_hash( void *state, const void *input )
sph_shavite512_close( &ctx.shavite, hash3 );
// 10 Simd
update_final_sd( &ctx.simd, (BitSequence *)hash0,
(const BitSequence *)hash0, 512 );
memcpy( &ctx.simd, &c11_4way_ctx.simd, sizeof(hashState_sd) );
update_final_sd( &ctx.simd, (BitSequence *)hash1,
(const BitSequence *)hash1, 512 );
memcpy( &ctx.simd, &c11_4way_ctx.simd, sizeof(hashState_sd) );
update_final_sd( &ctx.simd, (BitSequence *)hash2,
(const BitSequence *)hash2, 512 );
memcpy( &ctx.simd, &c11_4way_ctx.simd, sizeof(hashState_sd) );
update_final_sd( &ctx.simd, (BitSequence *)hash3,
(const BitSequence *)hash3, 512 );
mm256_interleave_2x128( vhash, hash0, hash1, 512 );
mm256_interleave_2x128( vhashB, hash2, hash3, 512 );
simd_2way_update_close( &ctx.simd, vhash, vhash, 512 );
simd_2way_init( &ctx.simd, 512 );
simd_2way_update_close( &ctx.simd, vhashB, vhashB, 512 );
mm256_deinterleave_2x128( hash0, hash1, vhash, 512 );
mm256_deinterleave_2x128( hash2, hash3, vhashB, 512 );
// 11 Echo
update_final_echo( &ctx.echo, (BitSequence *)hash0,

4
algo/x11/c11.c
View File

@@ -22,9 +22,9 @@
#include "algo/echo/aes_ni/hash_api.h"
#endif
#include "algo/luffa/sse2/luffa_for_sse2.h"
#include "algo/luffa/luffa_for_sse2.h"
#include "algo/cubehash/sse2/cubehash_sse2.h"
#include "algo/simd/sse2/nist.h"
#include "algo/simd/nist.h"
#include "algo/blake/sse2/blake.c"
#include "algo/keccak/sse2/keccak.c"
#include "algo/bmw/sse2/bmw.c"

24
algo/x11/timetravel-4way.c
View File

@@ -12,7 +12,7 @@
#include "algo/skein/skein-hash-4way.h"
#include "algo/jh/jh-hash-4way.h"
#include "algo/keccak/keccak-hash-4way.h"
#include "algo/luffa/sse2/luffa_for_sse2.h"
#include "algo/luffa/luffa-hash-2way.h"
#include "algo/cubehash/sse2/cubehash_sse2.h"
static __thread uint32_t s_ntime = UINT32_MAX;
@@ -25,7 +25,7 @@ typedef struct {
skein512_4way_context skein;
jh512_4way_context jh;
keccak512_4way_context keccak;
hashState_luffa luffa;
luffa_2way_context luffa;
cubehashParam cube;
} tt8_4way_ctx_holder;
@@ -39,7 +39,7 @@ void init_tt8_4way_ctx()
skein512_4way_init( &tt8_4way_ctx.skein );
jh512_4way_init( &tt8_4way_ctx.jh );
keccak512_4way_init( &tt8_4way_ctx.keccak );
init_luffa( &tt8_4way_ctx.luffa, 512 );
luffa_2way_init( &tt8_4way_ctx.luffa, 512 );
cubehashInit( &tt8_4way_ctx.cube, 512, 16, 32 );
};
@@ -139,17 +139,13 @@ void timetravel_4way_hash(void *output, const void *input)
case 6:
mm256_deinterleave_4x64( hash0, hash1, hash2, hash3,
vhashA, dataLen<<3 );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash0,
(const BitSequence *)hash0, dataLen );
memcpy( &ctx.luffa, &tt8_4way_ctx.luffa, sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash1,
(const BitSequence*)hash1, dataLen );
memcpy( &ctx.luffa, &tt8_4way_ctx.luffa, sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash2,
(const BitSequence*)hash2, dataLen );
memcpy( &ctx.luffa, &tt8_4way_ctx.luffa, sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash3,
(const BitSequence*)hash3, dataLen );
mm256_interleave_2x128( vhashA, hash0, hash1, dataLen<<3 );
luffa_2way_update_close( &ctx.luffa, vhashA, vhashA, dataLen );
mm256_deinterleave_2x128( hash0, hash1, vhashA, dataLen<<3 );
mm256_interleave_2x128( vhashA, hash2, hash3, dataLen<<3 );
luffa_2way_init( &ctx.luffa, 512 );
luffa_2way_update_close( &ctx.luffa, vhashA, vhashA, dataLen );
mm256_deinterleave_2x128( hash2, hash3, vhashA, dataLen<<3 );
if ( i != 7 )
mm256_interleave_4x64( vhashB,
hash0, hash1, hash2, hash3, dataLen<<3 );

2
algo/x11/timetravel.c
View File

@@ -9,7 +9,7 @@
#include "algo/jh/sph_jh.h"
#include "algo/keccak/sph_keccak.h"
#include "algo/skein/sph_skein.h"
#include "algo/luffa/sse2/luffa_for_sse2.h"
#include "algo/luffa/luffa_for_sse2.h"
#include "algo/cubehash/sse2/cubehash_sse2.h"
#ifdef NO_AES_NI
#include "algo/groestl/sph_groestl.h"

48
algo/x11/timetravel10-4way.c
View File

@@ -12,10 +12,10 @@
#include "algo/skein/skein-hash-4way.h"
#include "algo/jh/jh-hash-4way.h"
#include "algo/keccak/keccak-hash-4way.h"
#include "algo/luffa/sse2/luffa_for_sse2.h"
#include "algo/luffa/luffa-hash-2way.h"
#include "algo/cubehash/sse2/cubehash_sse2.h"
#include "algo/shavite/sph_shavite.h"
#include "algo/simd/sse2/nist.h"
#include "algo/simd/simd-hash-2way.h"
static __thread uint32_t s_ntime = UINT32_MAX;
static __thread int permutation[TT10_FUNC_COUNT] = { 0 };
@@ -27,10 +27,10 @@ typedef struct {
skein512_4way_context skein;
jh512_4way_context jh;
keccak512_4way_context keccak;
hashState_luffa luffa;
luffa_2way_context luffa;
cubehashParam cube;
sph_shavite512_context shavite;
hashState_sd simd;
simd_2way_context simd;
} tt10_4way_ctx_holder;
tt10_4way_ctx_holder tt10_4way_ctx __attribute__ ((aligned (64)));
@@ -43,10 +43,10 @@ void init_tt10_4way_ctx()
skein512_4way_init( &tt10_4way_ctx.skein );
jh512_4way_init( &tt10_4way_ctx.jh );
keccak512_4way_init( &tt10_4way_ctx.keccak );
init_luffa( &tt10_4way_ctx.luffa, 512 );
luffa_2way_init( &tt10_4way_ctx.luffa, 512 );
cubehashInit( &tt10_4way_ctx.cube, 512, 16, 32 );
sph_shavite512_init( &tt10_4way_ctx.shavite );
init_sd( &tt10_4way_ctx.simd, 512 );
simd_2way_init( &tt10_4way_ctx.simd, 512 );
};
void timetravel10_4way_hash(void *output, const void *input)
@@ -145,17 +145,13 @@ void timetravel10_4way_hash(void *output, const void *input)
case 6:
mm256_deinterleave_4x64( hash0, hash1, hash2, hash3,
vhashA, dataLen<<3 );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash0,
(const BitSequence *)hash0, dataLen );
memcpy( &ctx.luffa, &tt10_4way_ctx.luffa, sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash1,
(const BitSequence*)hash1, dataLen );
memcpy( &ctx.luffa, &tt10_4way_ctx.luffa, sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash2,
(const BitSequence*)hash2, dataLen );
memcpy( &ctx.luffa, &tt10_4way_ctx.luffa, sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash3,
(const BitSequence*)hash3, dataLen );
mm256_interleave_2x128( vhashA, hash0, hash1, dataLen<<3 );
luffa_2way_update_close( &ctx.luffa, vhashA, vhashA, dataLen );
mm256_deinterleave_2x128( hash0, hash1, vhashA, dataLen<<3 );
mm256_interleave_2x128( vhashA, hash2, hash3, dataLen<<3 );
luffa_2way_init( &ctx.luffa, 512 );
luffa_2way_update_close( &ctx.luffa, vhashA, vhashA, dataLen );
mm256_deinterleave_2x128( hash2, hash3, vhashA, dataLen<<3 );
if ( i != 9 )
mm256_interleave_4x64( vhashB,
hash0, hash1, hash2, hash3, dataLen<<3 );
@@ -199,17 +195,13 @@ void timetravel10_4way_hash(void *output, const void *input)
case 9:
mm256_deinterleave_4x64( hash0, hash1, hash2, hash3,
vhashA, dataLen<<3 );
update_final_sd( &ctx.simd, (BitSequence *)hash0,
(const BitSequence *)hash0, dataLen<<3 );
memcpy( &ctx.simd, &tt10_4way_ctx.simd, sizeof ctx.simd );
update_final_sd( &ctx.simd, (BitSequence *)hash1,
(const BitSequence *)hash1, dataLen<<3 );
memcpy( &ctx.simd, &tt10_4way_ctx.simd, sizeof ctx.simd );
update_final_sd( &ctx.simd, (BitSequence *)hash2,
(const BitSequence *)hash2, dataLen<<3 );
memcpy( &ctx.simd, &tt10_4way_ctx.simd, sizeof ctx.simd );
update_final_sd( &ctx.simd, (BitSequence *)hash3,
(const BitSequence *)hash3, dataLen<<3 );
mm256_interleave_2x128( vhashA, hash0, hash1, dataLen<<3 );
simd_2way_update_close( &ctx.simd, vhashA, vhashA, dataLen<<3 );
mm256_deinterleave_2x128( hash0, hash1, vhashA, dataLen<<3 );
mm256_interleave_2x128( vhashA, hash2, hash3, dataLen<<3 );
simd_2way_init( &ctx.simd, 512 );
simd_2way_update_close( &ctx.simd, vhashA, vhashA, dataLen<<3 );
mm256_deinterleave_2x128( hash2, hash3, vhashA, dataLen<<3 );
if ( i != 9 )
mm256_interleave_4x64( vhashB,
hash0, hash1, hash2, hash3, dataLen<<3 );

4
algo/x11/timetravel10.c
View File

@@ -8,10 +8,10 @@
#include "algo/jh/sph_jh.h"
#include "algo/keccak/sph_keccak.h"
#include "algo/skein/sph_skein.h"
#include "algo/luffa/sse2/luffa_for_sse2.h"
#include "algo/luffa/luffa_for_sse2.h"
#include "algo/cubehash/sse2/cubehash_sse2.h"
#include "algo/shavite/sph_shavite.h"
#include "algo/simd/sse2/nist.h"
#include "algo/simd/nist.h"
#ifdef NO_AES_NI
#include "algo/groestl/sph_groestl.h"

54
algo/x11/x11-4way.c
View File

@@ -5,17 +5,16 @@
#include <string.h>
#include <stdint.h>
#include "algo/blake/blake-hash-4way.h"
#include "algo/bmw/bmw-hash-4way.h"
#include "algo/groestl/aes_ni/hash-groestl.h"
#include "algo/skein/skein-hash-4way.h"
#include "algo/jh/jh-hash-4way.h"
#include "algo/keccak/keccak-hash-4way.h"
#include "algo/luffa/sse2/luffa_for_sse2.h"
#include "algo/luffa/luffa-hash-2way.h"
#include "algo/cubehash/sse2/cubehash_sse2.h"
#include "algo/shavite/sph_shavite.h"
#include "algo/simd/sse2/nist.h"
#include "algo/simd/simd-hash-2way.h"
#include "algo/echo/aes_ni/hash_api.h"
typedef struct {
@@ -25,10 +24,10 @@ typedef struct {
skein512_4way_context skein;
jh512_4way_context jh;
keccak512_4way_context keccak;
hashState_luffa luffa;
luffa_2way_context luffa;
cubehashParam cube;
sph_shavite512_context shavite;
hashState_sd simd;
simd_2way_context simd;
hashState_echo echo;
} x11_4way_ctx_holder;
@@ -42,10 +41,10 @@ void init_x11_4way_ctx()
skein512_4way_init( &x11_4way_ctx.skein );
jh512_4way_init( &x11_4way_ctx.jh );
keccak512_4way_init( &x11_4way_ctx.keccak );
init_luffa( &x11_4way_ctx.luffa, 512 );
luffa_2way_init( &x11_4way_ctx.luffa, 512 );
cubehashInit( &x11_4way_ctx.cube, 512, 16, 32 );
sph_shavite512_init( &x11_4way_ctx.shavite );
init_sd( &x11_4way_ctx.simd, 512 );
simd_2way_init( &x11_4way_ctx.simd, 512 );
init_echo( &x11_4way_ctx.echo, 512 );
}
@@ -56,6 +55,8 @@ void x11_4way_hash( void *state, const void *input )
uint64_t hash2[8] __attribute__ ((aligned (64)));
uint64_t hash3[8] __attribute__ ((aligned (64)));
uint64_t vhash[8*4] __attribute__ ((aligned (64)));
uint64_t vhashB[8*2] __attribute__ ((aligned (64)));
x11_4way_ctx_holder ctx;
memcpy( &ctx, &x11_4way_ctx, sizeof(x11_4way_ctx) );
@@ -94,21 +95,16 @@ void x11_4way_hash( void *state, const void *input )
keccak512_4way( &ctx.keccak, vhash, 64 );
keccak512_4way_close( &ctx.keccak, vhash );
// Serial
mm256_deinterleave_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
// 7 Luffa
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash0,
(const BitSequence*)hash0, 64 );
memcpy( &ctx.luffa, &x11_4way_ctx.luffa, sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash1,
(const BitSequence*)hash1, 64 );
memcpy( &ctx.luffa, &x11_4way_ctx.luffa, sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash2,
(const BitSequence*)hash2, 64 );
memcpy( &ctx.luffa, &x11_4way_ctx.luffa, sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash3,
(const BitSequence*)hash3, 64 );
// 7 Luffa parallel 2 way 128 bit
mm256_interleave_2x128( vhash, hash0, hash1, 512 );
mm256_interleave_2x128( vhashB, hash2, hash3, 512 );
luffa_2way_update_close( &ctx.luffa, vhash, vhash, 64 );
luffa_2way_init( &ctx.luffa, 512 );
luffa_2way_update_close( &ctx.luffa, vhashB, vhashB, 64 );
mm256_deinterleave_2x128( hash0, hash1, vhash, 512 );
mm256_deinterleave_2x128( hash2, hash3, vhashB, 512 );
// 8 Cubehash
cubehashUpdateDigest( &ctx.cube, (byte*)hash0, (const byte*) hash0, 64 );
@@ -136,17 +132,13 @@ void x11_4way_hash( void *state, const void *input )
sph_shavite512_close( &ctx.shavite, hash3 );
// 10 Simd
update_final_sd( &ctx.simd, (BitSequence *)hash0,
(const BitSequence *)hash0, 512 );
memcpy( &ctx.simd, &x11_4way_ctx.simd, sizeof(hashState_sd) );
update_final_sd( &ctx.simd, (BitSequence *)hash1,
(const BitSequence *)hash1, 512 );
memcpy( &ctx.simd, &x11_4way_ctx.simd, sizeof(hashState_sd) );
update_final_sd( &ctx.simd, (BitSequence *)hash2,
(const BitSequence *)hash2, 512 );
memcpy( &ctx.simd, &x11_4way_ctx.simd, sizeof(hashState_sd) );
update_final_sd( &ctx.simd, (BitSequence *)hash3,
(const BitSequence *)hash3, 512 );
mm256_interleave_2x128( vhash, hash0, hash1, 512 );
mm256_interleave_2x128( vhashB, hash2, hash3, 512 );
simd_2way_update_close( &ctx.simd, vhash, vhash, 512 );
simd_2way_init( &ctx.simd, 512 );
simd_2way_update_close( &ctx.simd, vhashB, vhashB, 512 );
mm256_deinterleave_2x128( hash0, hash1, vhash, 512 );
mm256_deinterleave_2x128( hash2, hash3, vhashB, 512 );
// 11 Echo
update_final_echo( &ctx.echo, (BitSequence *)hash0,

6
algo/x11/x11.c
View File

@@ -10,10 +10,8 @@
#include "algo/jh/sph_jh.h"
#include "algo/keccak/sph_keccak.h"
#include "algo/skein/sph_skein.h"
#include "algo/luffa/sph_luffa.h"
#include "algo/cubehash/sph_cubehash.h"
#include "algo/shavite/sph_shavite.h"
#include "algo/simd/sph_simd.h"
#include "algo/echo/sph_echo.h"
#ifndef NO_AES_NI
@@ -21,9 +19,9 @@
#include "algo/echo/aes_ni/hash_api.h"
#endif
#include "algo/luffa/sse2/luffa_for_sse2.h"
#include "algo/luffa/luffa_for_sse2.h"
#include "algo/cubehash/sse2/cubehash_sse2.h"
#include "algo/simd/sse2/nist.h"
#include "algo/simd/nist.h"
#include "algo/blake/sse2/blake.c"
#include "algo/keccak/sse2/keccak.c"
#include "algo/bmw/sse2/bmw.c"

53
algo/x11/x11evo-4way.c
View File

@@ -11,15 +11,12 @@
#include "algo/skein/skein-hash-4way.h"
#include "algo/jh/jh-hash-4way.h"
#include "algo/keccak/keccak-hash-4way.h"
#include "algo/luffa/sph_luffa.h"
#include "algo/cubehash/sph_cubehash.h"
#include "algo/shavite/sph_shavite.h"
#include "algo/simd/sph_simd.h"
#include "algo/groestl/aes_ni/hash-groestl.h"
#include "algo/echo/aes_ni/hash_api.h"
#include "algo/luffa/sse2/luffa_for_sse2.h"
#include "algo/luffa/luffa-hash-2way.h"
#include "algo/cubehash/sse2/cubehash_sse2.h"
#include "algo/simd/sse2/nist.h"
#include "algo/simd/simd-hash-2way.h"
typedef struct {
blake512_4way_context blake;
@@ -28,10 +25,10 @@ typedef struct {
skein512_4way_context skein;
jh512_4way_context jh;
keccak512_4way_context keccak;
hashState_luffa luffa;
luffa_2way_context luffa;
cubehashParam cube;
sph_shavite512_context shavite;
hashState_sd simd;
simd_2way_context simd;
hashState_echo echo;
} x11evo_4way_ctx_holder;
@@ -45,10 +42,11 @@ void init_x11evo_4way_ctx()
skein512_4way_init( &x11evo_4way_ctx.skein );
jh512_4way_init( &x11evo_4way_ctx.jh );
keccak512_4way_init( &x11evo_4way_ctx.keccak );
luffa_2way_init( &x11evo_4way_ctx.luffa, 512 );
init_luffa( &x11evo_4way_ctx.luffa, 512 );
cubehashInit( &x11evo_4way_ctx.cube, 512, 16, 32 );
sph_shavite512_init( &x11evo_4way_ctx.shavite );
init_sd( &x11evo_4way_ctx.simd, 512 );
simd_2way_init( &x11evo_4way_ctx.simd, 512 );
init_echo( &x11evo_4way_ctx.echo, 512 );
}
@@ -142,20 +140,13 @@ void x11evo_4way_hash( void *state, const void *input )
case 6:
mm256_deinterleave_4x64( hash0, hash1, hash2, hash3,
vhash, 64<<3 );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash0,
(const BitSequence*)hash0, 64 );
memcpy( &ctx.luffa, &x11evo_4way_ctx.luffa,
sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash1,
(const BitSequence*)hash1, 64 );
memcpy( &ctx.luffa, &x11evo_4way_ctx.luffa,
sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash2,
(const BitSequence*)hash2, 64 );
memcpy( &ctx.luffa, &x11evo_4way_ctx.luffa,
sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash3,
(const BitSequence*)hash3, 64 );
mm256_interleave_2x128( vhash, hash0, hash1, 64<<3 );
luffa_2way_update_close( &ctx.luffa, vhash, vhash, 64 );
mm256_deinterleave_2x128( hash0, hash1, vhash, 64<<3 );
mm256_interleave_2x128( vhash, hash2, hash3, 64<<3 );
luffa_2way_init( &ctx.luffa, 512 );
luffa_2way_update_close( &ctx.luffa, vhash, vhash, 64 );
mm256_deinterleave_2x128( hash2, hash3, vhash, 64<<3 );
if ( i < len-1 )
mm256_interleave_4x64( vhash,
hash0, hash1, hash2, hash3, 64<<3 );
@@ -202,17 +193,13 @@ void x11evo_4way_hash( void *state, const void *input )
case 9:
mm256_deinterleave_4x64( hash0, hash1, hash2, hash3,
vhash, 64<<3 );
update_final_sd( &ctx.simd, (BitSequence *)hash0,
(const BitSequence *)hash0, 512 );
memcpy( &ctx.simd, &x11evo_4way_ctx.simd, sizeof(hashState_sd) );
update_final_sd( &ctx.simd, (BitSequence *)hash1,
(const BitSequence *)hash1, 512 );
memcpy( &ctx.simd, &x11evo_4way_ctx.simd, sizeof(hashState_sd) );
update_final_sd( &ctx.simd, (BitSequence *)hash2,
(const BitSequence *)hash2, 512 );
memcpy( &ctx.simd, &x11evo_4way_ctx.simd, sizeof(hashState_sd) );
update_final_sd( &ctx.simd, (BitSequence *)hash3,
(const BitSequence *)hash3, 512 );
mm256_interleave_2x128( vhash, hash0, hash1, 64<<3 );
simd_2way_update_close( &ctx.simd, vhash, vhash, 64<<3 );
mm256_deinterleave_2x128( hash0, hash1, vhash, 64<<3 );
mm256_interleave_2x128( vhash, hash2, hash3, 64<<3 );
simd_2way_init( &ctx.simd, 512 );
simd_2way_update_close( &ctx.simd, vhash, vhash, 64<<3 );
mm256_deinterleave_2x128( hash2, hash3, vhash, 64<<3 );
if ( i < len-1 )
mm256_interleave_4x64( vhash,
hash0, hash1, hash2, hash3, 64<<3 );

4
algo/x11/x11evo.c
View File

@@ -22,9 +22,9 @@
#include "algo/echo/aes_ni/hash_api.h"
#endif
#include "algo/luffa/sse2/luffa_for_sse2.h"
#include "algo/luffa/luffa_for_sse2.h"
#include "algo/cubehash/sse2/cubehash_sse2.h"
#include "algo/simd/sse2/nist.h"
#include "algo/simd/nist.h"
typedef struct {
#ifdef NO_AES_NI

48
algo/x11/x11gost-4way.c
View File

@@ -13,10 +13,10 @@
#include "algo/jh/jh-hash-4way.h"
#include "algo/keccak/keccak-hash-4way.h"
#include "algo/gost/sph_gost.h"
#include "algo/luffa/sse2/luffa_for_sse2.h"
#include "algo/luffa/luffa-hash-2way.h"
#include "algo/cubehash/sse2/cubehash_sse2.h"
#include "algo/shavite/sph_shavite.h"
#include "algo/simd/sse2/nist.h"
#include "algo/simd/simd-hash-2way.h"
#include "algo/echo/aes_ni/hash_api.h"
typedef struct {
@@ -27,10 +27,10 @@ typedef struct {
jh512_4way_context jh;
keccak512_4way_context keccak;
sph_gost512_context gost;
hashState_luffa luffa;
luffa_2way_context luffa;
cubehashParam cube;
sph_shavite512_context shavite;
hashState_sd simd;
simd_2way_context simd;
hashState_echo echo;
} x11gost_4way_ctx_holder;
@@ -45,10 +45,10 @@ void init_x11gost_4way_ctx()
jh512_4way_init( &x11gost_4way_ctx.jh );
keccak512_4way_init( &x11gost_4way_ctx.keccak );
sph_gost512_init( &x11gost_4way_ctx.gost );
init_luffa( &x11gost_4way_ctx.luffa, 512 );
luffa_2way_init( &x11gost_4way_ctx.luffa, 512 );
cubehashInit( &x11gost_4way_ctx.cube, 512, 16, 32 );
sph_shavite512_init( &x11gost_4way_ctx.shavite );
init_sd( &x11gost_4way_ctx.simd, 512 );
simd_2way_init( &x11gost_4way_ctx.simd, 512 );
init_echo( &x11gost_4way_ctx.echo, 512 );
}
@@ -59,6 +59,7 @@ void x11gost_4way_hash( void *state, const void *input )
uint64_t hash2[8] __attribute__ ((aligned (64)));
uint64_t hash3[8] __attribute__ ((aligned (64)));
uint64_t vhash[8*4] __attribute__ ((aligned (64)));
x11gost_4way_ctx_holder ctx;
memcpy( &ctx, &x11gost_4way_ctx, sizeof(x11gost_4way_ctx) );
@@ -109,17 +110,13 @@ void x11gost_4way_hash( void *state, const void *input )
sph_gost512( &ctx.gost, hash3, 64 );
sph_gost512_close( &ctx.gost, hash3 );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash0,
(const BitSequence*)hash0, 64 );
memcpy( &ctx.luffa, &x11gost_4way_ctx.luffa, sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash1,
(const BitSequence*)hash1, 64 );
memcpy( &ctx.luffa, &x11gost_4way_ctx.luffa, sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash2,
(const BitSequence*)hash2, 64 );
memcpy( &ctx.luffa, &x11gost_4way_ctx.luffa, sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash3,
(const BitSequence*)hash3, 64 );
mm256_interleave_2x128( vhash, hash0, hash1, 512 );
luffa_2way_update_close( &ctx.luffa, vhash, vhash, 64 );
mm256_deinterleave_2x128( hash0, hash1, vhash, 512 );
mm256_interleave_2x128( vhash, hash2, hash3, 512 );
luffa_2way_init( &ctx.luffa, 512 );
luffa_2way_update_close( &ctx.luffa, vhash, vhash, 64 );
mm256_deinterleave_2x128( hash2, hash3, vhash, 512 );
cubehashUpdateDigest( &ctx.cube, (byte*)hash0, (const byte*) hash0, 64 );
memcpy( &ctx.cube, &x11gost_4way_ctx.cube, sizeof(cubehashParam) );
@@ -144,17 +141,12 @@ void x11gost_4way_hash( void *state, const void *input )
sph_shavite512( &ctx.shavite, hash3, 64 );
sph_shavite512_close( &ctx.shavite, hash3 );
update_final_sd( &ctx.simd, (BitSequence *)hash0,
(const BitSequence *)hash0, 512 );
memcpy( &ctx.simd, &x11gost_4way_ctx.simd, sizeof(hashState_sd) );
update_final_sd( &ctx.simd, (BitSequence *)hash1,
(const BitSequence *)hash1, 512 );
memcpy( &ctx.simd, &x11gost_4way_ctx.simd, sizeof(hashState_sd) );
update_final_sd( &ctx.simd, (BitSequence *)hash2,
(const BitSequence *)hash2, 512 );
memcpy( &ctx.simd, &x11gost_4way_ctx.simd, sizeof(hashState_sd) );
update_final_sd( &ctx.simd, (BitSequence *)hash3,
(const BitSequence *)hash3, 512 );
mm256_interleave_2x128( vhash, hash0, hash1, 512 );
simd_2way_update_close( &ctx.simd, vhash, vhash, 512 );
mm256_deinterleave_2x128( hash0, hash1, vhash, 512 );
mm256_interleave_2x128( vhash, hash2, hash3, 512 );
simd_2way_update_close( &ctx.simd, vhash, vhash, 512 );
mm256_deinterleave_2x128( hash2, hash3, vhash, 512 );
update_final_echo( &ctx.echo, (BitSequence *)hash0,
(const BitSequence *) hash0, 512 );

4
algo/x11/x11gost.c
View File

@@ -10,9 +10,9 @@
#include "algo/shavite/sph_shavite.h"
#include "algo/echo/sph_echo.h"
#include "algo/luffa/sse2/luffa_for_sse2.h"
#include "algo/luffa/luffa_for_sse2.h"
#include "algo/cubehash/sse2/cubehash_sse2.h"
#include "algo/simd/sse2/nist.h"
#include "algo/simd/nist.h"
#include "algo/blake/sse2/blake.c"
#include "algo/keccak/sse2/keccak.c"
#include "algo/bmw/sse2/bmw.c"