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

35
algo/x14/polytimos-4way.c
View File

@@ -9,8 +9,7 @@
#include "algo/skein/skein-hash-4way.h"
#include "algo/shabal/shabal-hash-4way.h"
#include "algo/fugue//sph_fugue.h"
#include "algo/luffa/sse2/luffa_for_sse2.h"
//#include "algo/shabal/sph_shabal.h"
#include "algo/luffa/luffa-hash-2way.h"
#include "algo/gost/sph_gost.h"
#include "algo/echo/aes_ni/hash_api.h"
@@ -18,7 +17,7 @@ typedef struct {
skein512_4way_context skein;
shabal512_4way_context shabal;
hashState_echo echo;
hashState_luffa luffa;
luffa_2way_context luffa;
sph_fugue512_context fugue;
sph_gost512_context gost;
} poly_4way_ctx_holder;
@@ -27,12 +26,12 @@ poly_4way_ctx_holder poly_4way_ctx;
void init_polytimos_4way_ctx()
{
skein512_4way_init( &poly_4way_ctx.skein );
shabal512_4way_init( &poly_4way_ctx.shabal );
init_echo( &poly_4way_ctx.echo, 512 );
init_luffa( &poly_4way_ctx.luffa, 512 );
sph_fugue512_init( &poly_4way_ctx.fugue );
sph_gost512_init( &poly_4way_ctx.gost );
skein512_4way_init( &poly_4way_ctx.skein );
shabal512_4way_init( &poly_4way_ctx.shabal );
init_echo( &poly_4way_ctx.echo, 512 );
luffa_2way_init( &poly_4way_ctx.luffa, 512 );
sph_fugue512_init( &poly_4way_ctx.fugue );
sph_gost512_init( &poly_4way_ctx.gost );
}
void polytimos_4way_hash( void *output, const void *input )
@@ -67,17 +66,13 @@ void polytimos_4way_hash( void *output, const void *input )
update_final_echo( &ctx.echo, (BitSequence *)hash3,
(const BitSequence *) hash3, 512 );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash0,
(const BitSequence*)hash0, 64 );
memcpy( &ctx.luffa, &poly_4way_ctx.luffa, sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash1,
(const BitSequence*)hash1, 64 );
memcpy( &ctx.luffa, &poly_4way_ctx.luffa, sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash2,
(const BitSequence*)hash2, 64 );
memcpy( &ctx.luffa, &poly_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 );
sph_fugue512( &ctx.fugue, hash0, 64 );
sph_fugue512_close( &ctx.fugue, hash0 );

2
algo/x14/polytimos.c
View File

@@ -8,7 +8,7 @@
#include "algo/skein/sph_skein.h"
#include "algo/echo/sph_echo.h"
#include "algo/fugue//sph_fugue.h"
#include "algo/luffa/sse2/luffa_for_sse2.h"
#include "algo/luffa/luffa_for_sse2.h"
#include "algo/shabal/sph_shabal.h"
#include "algo/gost/sph_gost.h"
#ifndef NO_AES_NI

53
algo/x14/x14-4way.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"
#include "algo/echo/sph_echo.h"
#include "algo/hamsi/hamsi-hash-4way.h"
@@ -29,10 +29,10 @@ 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;
hamsi512_4way_context hamsi;
sph_fugue512_context fugue;
@@ -45,15 +45,14 @@ void init_x14_4way_ctx()
{
blake512_4way_init( &x14_4way_ctx.blake );
bmw512_4way_init( &x14_4way_ctx.bmw );
sph_bmw512_init( &x14_4way_ctx.bmw );
init_groestl( &x14_4way_ctx.groestl, 64 );
skein512_4way_init( &x14_4way_ctx.skein );
jh512_4way_init( &x14_4way_ctx.jh );
keccak512_4way_init( &x14_4way_ctx.keccak );
init_luffa( &x14_4way_ctx.luffa, 512 );
luffa_2way_init( &x14_4way_ctx.luffa, 512 );
cubehashInit( &x14_4way_ctx.cube, 512, 16, 32 );
sph_shavite512_init( &x14_4way_ctx.shavite );
init_sd( &x14_4way_ctx.simd, 512 );
simd_2way_init( &x14_4way_ctx.simd, 512 );
init_echo( &x14_4way_ctx.echo, 512 );
hamsi512_4way_init( &x14_4way_ctx.hamsi );
sph_fugue512_init( &x14_4way_ctx.fugue );
@@ -109,17 +108,13 @@ void x14_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, &x14_4way_ctx.luffa, sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash1,
(const BitSequence*)hash1, 64 );
memcpy( &ctx.luffa, &x14_4way_ctx.luffa, sizeof(hashState_luffa) );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash2,
(const BitSequence*)hash2, 64 );
memcpy( &ctx.luffa, &x14_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 );
// 8 Cubehash
cubehashUpdateDigest( &ctx.cube, (byte*)hash0, (const byte*) hash0, 64 );
@@ -147,17 +142,13 @@ void x14_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, &x14_4way_ctx.simd, sizeof(hashState_sd) );
update_final_sd( &ctx.simd, (BitSequence *)hash1,
(const BitSequence *)hash1, 512 );
memcpy( &ctx.simd, &x14_4way_ctx.simd, sizeof(hashState_sd) );
update_final_sd( &ctx.simd, (BitSequence *)hash2,
(const BitSequence *)hash2, 512 );
memcpy( &ctx.simd, &x14_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_init( &ctx.simd, 512 );
simd_2way_update_close( &ctx.simd, vhash, vhash, 512 );
mm256_deinterleave_2x128( hash2, hash3, vhash, 512 );
// 11 Echo
update_final_echo( &ctx.echo, (BitSequence *)hash0,
@@ -173,10 +164,10 @@ void x14_4way_hash( void *state, const void *input )
(const BitSequence *) hash3, 512 );
// 12 Hamsi parallel 4way 32 bit
mm_interleave_4x32( vhash, hash0, hash1, hash2, hash3, 512 );
mm256_interleave_4x64( vhash, hash0, hash1, hash2, hash3, 512 );
hamsi512_4way( &ctx.hamsi, vhash, 64 );
hamsi512_4way_close( &ctx.hamsi, vhash );
mm_deinterleave_4x32( hash0, hash1, hash2, hash3, vhash, 512 );
mm256_deinterleave_4x64( hash0, hash1, hash2, hash3, vhash, 512 );
// 13 Fugue serial
sph_fugue512( &ctx.fugue, hash0, 64 );

4
algo/x14/x14.c
View File

@@ -20,9 +20,9 @@
#include "algo/fugue/sph_fugue.h"
#include "algo/shabal/sph_shabal.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/echo/sse2/sph_echo.h"
#include "algo/blake/sse2/blake.c"
#include "algo/bmw/sse2/bmw.c"