mirror of
https://github.com/JayDDee/cpuminer-opt.git
synced 2025-09-17 23:44:27 +00:00
v3.10.6
This commit is contained in:
Jay D Dee
@@ -1,7 +1,4 @@
|
||||
#include "x22i-gate.h"
|
||||
|
||||
#if defined(X22I_4WAY)
|
||||
|
||||
#include "algo/blake/blake-hash-4way.h"
|
||||
#include "algo/bmw/bmw-hash-4way.h"
|
||||
#include "algo/echo/aes_ni/hash_api.h"
|
||||
@@ -12,6 +9,7 @@
|
||||
#include "algo/luffa/luffa-hash-2way.h"
|
||||
#include "algo/cubehash/cube-hash-2way.h"
|
||||
#include "algo/shavite/shavite-hash-2way.h"
|
||||
#include "algo/shavite/sph_shavite.h"
|
||||
#include "algo/simd/simd-hash-2way.h"
|
||||
#include "algo/shavite/sph_shavite.h"
|
||||
#include "algo/hamsi/hamsi-hash-4way.h"
|
||||
@@ -25,6 +23,426 @@
|
||||
#include "algo/gost/sph_gost.h"
|
||||
#include "algo/swifftx/swifftx.h"
|
||||
|
||||
|
||||
#if defined(X22I_8WAY)
|
||||
|
||||
union _x22i_8way_ctx_overlay
|
||||
{
|
||||
blake512_8way_context blake;
|
||||
bmw512_8way_context bmw;
|
||||
hashState_groestl groestl;
|
||||
hashState_echo echo;
|
||||
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;
|
||||
hamsi512_8way_context hamsi;
|
||||
sph_fugue512_context fugue;
|
||||
shabal512_8way_context shabal;
|
||||
sph_whirlpool_context whirlpool;
|
||||
sha512_8way_context sha512;
|
||||
haval256_5_8way_context haval;
|
||||
sph_tiger_context tiger;
|
||||
sph_gost512_context gost;
|
||||
sha256_8way_context sha256;
|
||||
};
|
||||
typedef union _x22i_8way_ctx_overlay x22i_8way_ctx_overlay;
|
||||
|
||||
void x22i_8way_hash( void *output, const void *input )
|
||||
{
|
||||
uint64_t vhash[8*8] __attribute__ ((aligned (128)));
|
||||
uint64_t vhashA[8*8] __attribute__ ((aligned (64)));
|
||||
uint64_t vhashB[8*8] __attribute__ ((aligned (64)));
|
||||
uint64_t hash0[8*4] __attribute__ ((aligned (64)));
|
||||
uint64_t hash1[8*4] __attribute__ ((aligned (64)));
|
||||
uint64_t hash2[8*4] __attribute__ ((aligned (64)));
|
||||
uint64_t hash3[8*4] __attribute__ ((aligned (64)));
|
||||
uint64_t hash4[8*4] __attribute__ ((aligned (64)));
|
||||
uint64_t hash5[8*4] __attribute__ ((aligned (64)));
|
||||
uint64_t hash6[8*4] __attribute__ ((aligned (64)));
|
||||
uint64_t hash7[8*4] __attribute__ ((aligned (64)));
|
||||
|
||||
// unsigned char hash[64 * 4] __attribute__((aligned(64))) = {0};
|
||||
unsigned char hashA0[64] __attribute__((aligned(64))) = {0};
|
||||
unsigned char hashA1[64] __attribute__((aligned(32))) = {0};
|
||||
unsigned char hashA2[64] __attribute__((aligned(32))) = {0};
|
||||
unsigned char hashA3[64] __attribute__((aligned(32))) = {0};
|
||||
unsigned char hashA4[64] __attribute__((aligned(64))) = {0};
|
||||
unsigned char hashA5[64] __attribute__((aligned(32))) = {0};
|
||||
unsigned char hashA6[64] __attribute__((aligned(32))) = {0};
|
||||
unsigned char hashA7[64] __attribute__((aligned(32))) = {0};
|
||||
x22i_8way_ctx_overlay ctx;
|
||||
|
||||
blake512_8way_init( &ctx.blake );
|
||||
blake512_8way_update( &ctx.blake, input, 80 );
|
||||
blake512_8way_close( &ctx.blake, vhash );
|
||||
|
||||
bmw512_8way_init( &ctx.bmw );
|
||||
bmw512_8way_update( &ctx.bmw, vhash, 64 );
|
||||
bmw512_8way_close( &ctx.bmw, vhash );
|
||||
|
||||
dintrlv_8x64_512( hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7, vhash );
|
||||
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash0,
|
||||
(const char*)hash0, 512 );
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash1,
|
||||
(const char*)hash1, 512 );
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash2,
|
||||
(const char*)hash2, 512 );
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash3,
|
||||
(const char*)hash3, 512 );
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash4,
|
||||
(const char*)hash4, 512 );
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash5,
|
||||
(const char*)hash5, 512 );
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash6,
|
||||
(const char*)hash6, 512 );
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash7,
|
||||
(const char*)hash7, 512 );
|
||||
|
||||
intrlv_8x64_512( vhash, hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7 );
|
||||
|
||||
skein512_8way_init( &ctx.skein );
|
||||
skein512_8way_update( &ctx.skein, vhash, 64 );
|
||||
skein512_8way_close( &ctx.skein, vhash );
|
||||
|
||||
jh512_8way_init( &ctx.jh );
|
||||
jh512_8way_update( &ctx.jh, vhash, 64 );
|
||||
jh512_8way_close( &ctx.jh, vhash );
|
||||
|
||||
keccak512_8way_init( &ctx.keccak );
|
||||
keccak512_8way_update( &ctx.keccak, vhash, 64 );
|
||||
keccak512_8way_close( &ctx.keccak, vhash );
|
||||
|
||||
rintrlv_8x64_4x128( vhashA, vhashB, vhash, 512 );
|
||||
|
||||
luffa_4way_init( &ctx.luffa, 512 );
|
||||
luffa_4way_update_close( &ctx.luffa, vhashA, vhashA, 64 );
|
||||
luffa_4way_init( &ctx.luffa, 512 );
|
||||
luffa_4way_update_close( &ctx.luffa, vhashB, vhashB, 64 );
|
||||
|
||||
cube_4way_init( &ctx.cube, 512, 16, 32 );
|
||||
cube_4way_update_close( &ctx.cube, vhashA, vhashA, 64 );
|
||||
cube_4way_init( &ctx.cube, 512, 16, 32 );
|
||||
cube_4way_update_close( &ctx.cube, vhashB, vhashB, 64 );
|
||||
|
||||
dintrlv_4x128_512( hash0, hash1, hash2, hash3, vhashA );
|
||||
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhashB );
|
||||
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash0, 64 );
|
||||
sph_shavite512_close( &ctx.shavite, hash0 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash1, 64 );
|
||||
sph_shavite512_close( &ctx.shavite, hash1 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash2, 64 );
|
||||
sph_shavite512_close( &ctx.shavite, hash2 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash3, 64 );
|
||||
sph_shavite512_close( &ctx.shavite, hash3 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash4, 64 );
|
||||
sph_shavite512_close( &ctx.shavite, hash4 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash5, 64 );
|
||||
sph_shavite512_close( &ctx.shavite, hash5 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash6, 64 );
|
||||
sph_shavite512_close( &ctx.shavite, hash6 );
|
||||
sph_shavite512_init( &ctx.shavite );
|
||||
sph_shavite512( &ctx.shavite, hash7, 64 );
|
||||
sph_shavite512_close( &ctx.shavite, hash7 );
|
||||
|
||||
intrlv_4x128_512( vhashA, hash0, hash1, hash2, hash3 );
|
||||
intrlv_4x128_512( vhashB, hash4, hash5, hash6, hash7 );
|
||||
|
||||
simd_4way_init( &ctx.simd, 512 );
|
||||
simd_4way_update_close( &ctx.simd, vhashA, vhashA, 512 );
|
||||
simd_4way_init( &ctx.simd, 512 );
|
||||
simd_4way_update_close( &ctx.simd, vhashB, vhashB, 512 );
|
||||
|
||||
dintrlv_4x128_512( hash0, hash1, hash2, hash3, vhashA );
|
||||
dintrlv_4x128_512( hash4, hash5, hash6, hash7, vhashB );
|
||||
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence*)hash0,
|
||||
(const BitSequence*)hash0, 512 );
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence*)hash1,
|
||||
(const BitSequence*)hash1, 512 );
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence*)hash2,
|
||||
(const BitSequence*)hash2, 512 );
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence*)hash3,
|
||||
(const BitSequence*)hash3, 512 );
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence*)hash4,
|
||||
(const BitSequence*)hash4, 512 );
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence*)hash5,
|
||||
(const BitSequence*)hash5, 512 );
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence*)hash6,
|
||||
(const BitSequence*)hash6, 512 );
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence*)hash7,
|
||||
(const BitSequence*)hash7, 512 );
|
||||
|
||||
intrlv_8x64_512( vhash, hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7 );
|
||||
|
||||
hamsi512_8way_init( &ctx.hamsi );
|
||||
hamsi512_8way_update( &ctx.hamsi, vhash, 64 );
|
||||
hamsi512_8way_close( &ctx.hamsi, vhash );
|
||||
|
||||
dintrlv_8x64_512( hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7, vhash );
|
||||
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash0, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash0 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash1, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash1 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash2, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash2 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash3, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash3 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash4, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash4 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash5, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash5 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash6, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash6 );
|
||||
sph_fugue512_init( &ctx.fugue );
|
||||
sph_fugue512( &ctx.fugue, hash7, 64 );
|
||||
sph_fugue512_close( &ctx.fugue, hash7 );
|
||||
|
||||
intrlv_8x32_512( vhash, hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7 );
|
||||
|
||||
shabal512_8way_init( &ctx.shabal );
|
||||
shabal512_8way_update( &ctx.shabal, vhash, 64 );
|
||||
shabal512_8way_close( &ctx.shabal, vhash );
|
||||
|
||||
dintrlv_8x32_512( &hash0[8], &hash1[8], &hash2[8], &hash3[8],
|
||||
&hash4[8], &hash5[8], &hash6[8], &hash7[8], vhash );
|
||||
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, &hash0[8], 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, &hash0[16] );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, &hash1[8], 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, &hash1[16] );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, &hash2[8], 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, &hash2[16] );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, &hash3[8], 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, &hash3[16] );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, &hash4[8], 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, &hash4[16] );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, &hash5[8], 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, &hash5[16] );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, &hash6[8], 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, &hash6[16] );
|
||||
sph_whirlpool_init( &ctx.whirlpool );
|
||||
sph_whirlpool( &ctx.whirlpool, &hash7[8], 64 );
|
||||
sph_whirlpool_close( &ctx.whirlpool, &hash7[16] );
|
||||
|
||||
intrlv_8x64_512( vhash, &hash0[16], &hash1[16], &hash2[16], &hash3[16],
|
||||
&hash4[16], &hash5[16], &hash6[16], &hash7[16] );
|
||||
|
||||
sha512_8way_init( &ctx.sha512 );
|
||||
sha512_8way_update( &ctx.sha512, vhash, 64 );
|
||||
sha512_8way_close( &ctx.sha512, vhash );
|
||||
|
||||
dintrlv_8x64_512( &hash0[24], &hash1[24], &hash2[24], &hash3[24],
|
||||
&hash4[24], &hash5[24], &hash6[24], &hash7[24], vhash );
|
||||
|
||||
ComputeSingleSWIFFTX((unsigned char*)hash0, (unsigned char*)hashA0);
|
||||
ComputeSingleSWIFFTX((unsigned char*)hash1, (unsigned char*)hashA1);
|
||||
ComputeSingleSWIFFTX((unsigned char*)hash2, (unsigned char*)hashA2);
|
||||
ComputeSingleSWIFFTX((unsigned char*)hash3, (unsigned char*)hashA3);
|
||||
ComputeSingleSWIFFTX((unsigned char*)hash4, (unsigned char*)hashA4);
|
||||
ComputeSingleSWIFFTX((unsigned char*)hash5, (unsigned char*)hashA5);
|
||||
ComputeSingleSWIFFTX((unsigned char*)hash6, (unsigned char*)hashA6);
|
||||
ComputeSingleSWIFFTX((unsigned char*)hash7, (unsigned char*)hashA7);
|
||||
|
||||
intrlv_8x32_512( vhashA, hashA0, hashA1, hashA2, hashA3,
|
||||
hashA4, hashA5, hashA6, hashA7 );
|
||||
|
||||
memset( vhash, 0, 64*8 );
|
||||
|
||||
haval256_5_8way_init( &ctx.haval );
|
||||
haval256_5_8way_update( &ctx.haval, vhashA, 64 );
|
||||
haval256_5_8way_close( &ctx.haval, vhash );
|
||||
|
||||
dintrlv_8x32_512( hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7, vhash );
|
||||
|
||||
memset( hashA0, 0, 64 );
|
||||
memset( hashA1, 0, 64 );
|
||||
memset( hashA2, 0, 64 );
|
||||
memset( hashA3, 0, 64 );
|
||||
memset( hashA4, 0, 64 );
|
||||
memset( hashA5, 0, 64 );
|
||||
memset( hashA6, 0, 64 );
|
||||
memset( hashA7, 0, 64 );
|
||||
|
||||
sph_tiger_init(&ctx.tiger);
|
||||
sph_tiger (&ctx.tiger, (const void*) hash0, 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) hashA0);
|
||||
sph_tiger_init(&ctx.tiger);
|
||||
sph_tiger (&ctx.tiger, (const void*) hash1, 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) hashA1);
|
||||
sph_tiger_init(&ctx.tiger);
|
||||
sph_tiger (&ctx.tiger, (const void*) hash2, 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) hashA2);
|
||||
sph_tiger_init(&ctx.tiger);
|
||||
sph_tiger (&ctx.tiger, (const void*) hash3, 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) hashA3);
|
||||
sph_tiger_init(&ctx.tiger);
|
||||
sph_tiger (&ctx.tiger, (const void*) hash4, 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) hashA4);
|
||||
sph_tiger_init(&ctx.tiger);
|
||||
sph_tiger (&ctx.tiger, (const void*) hash5, 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) hashA5);
|
||||
sph_tiger_init(&ctx.tiger);
|
||||
sph_tiger (&ctx.tiger, (const void*) hash6, 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) hashA6);
|
||||
sph_tiger_init(&ctx.tiger);
|
||||
sph_tiger (&ctx.tiger, (const void*) hash7, 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) hashA7);
|
||||
|
||||
memset( hash0, 0, 64 );
|
||||
memset( hash1, 0, 64 );
|
||||
memset( hash2, 0, 64 );
|
||||
memset( hash3, 0, 64 );
|
||||
memset( hash4, 0, 64 );
|
||||
memset( hash5, 0, 64 );
|
||||
memset( hash6, 0, 64 );
|
||||
memset( hash7, 0, 64 );
|
||||
|
||||
intrlv_2x256( vhash, hashA0, hashA1, 256 );
|
||||
LYRA2RE_2WAY( vhash, 32, vhash, 32, 1, 4, 4 );
|
||||
dintrlv_2x256( hash0, hash1, vhash, 256 );
|
||||
intrlv_2x256( vhash, hashA2, hashA3, 256 );
|
||||
LYRA2RE_2WAY( vhash, 32, vhash, 32, 1, 4, 4 );
|
||||
dintrlv_2x256( hash2, hash3, vhash, 256 );
|
||||
intrlv_2x256( vhash, hashA4, hashA5, 256 );
|
||||
LYRA2RE_2WAY( vhash, 32, vhash, 32, 1, 4, 4 );
|
||||
dintrlv_2x256( hash4, hash5, vhash, 256 );
|
||||
intrlv_2x256( vhash, hashA6, hashA7, 256 );
|
||||
LYRA2RE_2WAY( vhash, 32, vhash, 32, 1, 4, 4 );
|
||||
dintrlv_2x256( hash6, hash7, vhash, 256 );
|
||||
|
||||
sph_gost512_init( &ctx.gost );
|
||||
sph_gost512 ( &ctx.gost, (const void*) hash0, 64 );
|
||||
sph_gost512_close( &ctx.gost, (void*) hash0 );
|
||||
sph_gost512_init( &ctx.gost );
|
||||
sph_gost512 ( &ctx.gost, (const void*) hash1, 64 );
|
||||
sph_gost512_close( &ctx.gost, (void*) hash1 );
|
||||
sph_gost512_init( &ctx.gost );
|
||||
sph_gost512 ( &ctx.gost, (const void*) hash2, 64 );
|
||||
sph_gost512_close( &ctx.gost, (void*) hash2 );
|
||||
sph_gost512_init( &ctx.gost );
|
||||
sph_gost512 ( &ctx.gost, (const void*) hash3, 64 );
|
||||
sph_gost512_close( &ctx.gost, (void*) hash3 );
|
||||
sph_gost512_init( &ctx.gost );
|
||||
sph_gost512 ( &ctx.gost, (const void*) hash4, 64 );
|
||||
sph_gost512_close( &ctx.gost, (void*) hash4 );
|
||||
sph_gost512_init( &ctx.gost );
|
||||
sph_gost512 ( &ctx.gost, (const void*) hash5, 64 );
|
||||
sph_gost512_close( &ctx.gost, (void*) hash5 );
|
||||
sph_gost512_init( &ctx.gost );
|
||||
sph_gost512 ( &ctx.gost, (const void*) hash6, 64 );
|
||||
sph_gost512_close( &ctx.gost, (void*) hash6 );
|
||||
sph_gost512_init( &ctx.gost );
|
||||
sph_gost512 ( &ctx.gost, (const void*) hash7, 64 );
|
||||
sph_gost512_close( &ctx.gost, (void*) hash7 );
|
||||
|
||||
intrlv_8x32_512( vhash, hash0, hash1, hash2, hash3,
|
||||
hash4, hash5, hash6, hash7 );
|
||||
|
||||
sha256_8way_init( &ctx.sha256 );
|
||||
sha256_8way_update( &ctx.sha256, vhash, 64 );
|
||||
sha256_8way_close( &ctx.sha256, output );
|
||||
}
|
||||
|
||||
int scanhash_x22i_8way( struct work* work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr )
|
||||
{
|
||||
uint32_t hash[8*16] __attribute__ ((aligned (128)));
|
||||
uint32_t vdata[24*8] __attribute__ ((aligned (64)));
|
||||
uint32_t lane_hash[8] __attribute__ ((aligned (64)));
|
||||
uint32_t *hash7 = &(hash[7<<3]);
|
||||
uint32_t *pdata = work->data;
|
||||
uint32_t *ptarget = work->target;
|
||||
const uint32_t first_nonce = pdata[19];
|
||||
__m512i *noncev = (__m512i*)vdata + 9; // aligned
|
||||
uint32_t n = first_nonce;
|
||||
const uint32_t last_nonce = max_nonce - 8;
|
||||
const int thr_id = mythr->id;
|
||||
const uint32_t Htarg = ptarget[7];
|
||||
|
||||
if (opt_benchmark)
|
||||
((uint32_t*)ptarget)[7] = 0x08ff;
|
||||
|
||||
InitializeSWIFFTX();
|
||||
|
||||
mm512_bswap32_intrlv80_8x64( vdata, pdata );
|
||||
do
|
||||
{
|
||||
*noncev = mm512_intrlv_blend_32( mm512_bswap_32(
|
||||
_mm512_set_epi32( n+7, 0, n+6, 0, n+5, 0, n+4, 0,
|
||||
n+3, 0, n+2, 0, n+1, 0, n, 0 ) ), *noncev );
|
||||
x22i_8way_hash( hash, vdata );
|
||||
|
||||
for ( int lane = 0; lane < 8; lane++ )
|
||||
if unlikely( ( hash7[ lane ] <= Htarg ) )
|
||||
{
|
||||
extr_lane_8x32( lane_hash, hash, lane, 256 );
|
||||
if ( likely( fulltest( lane_hash, ptarget ) && !opt_benchmark ) )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
}
|
||||
}
|
||||
n += 8;
|
||||
} while ( likely( ( n < last_nonce ) && !work_restart[thr_id].restart ) );
|
||||
|
||||
*hashes_done = n - first_nonce;
|
||||
return 0;
|
||||
}
|
||||
|
||||
|
||||
#elif defined(X22I_4WAY)
|
||||
|
||||
|
||||
union _x22i_4way_ctx_overlay
|
||||
{
|
||||
blake512_4way_context blake;
|
||||
|
||||
@@ -2,27 +2,39 @@
|
||||
|
||||
bool register_x22i_algo( algo_gate_t* gate )
|
||||
{
|
||||
#if defined (X22I_4WAY)
|
||||
#if defined (X22I_8WAY)
|
||||
gate->scanhash = (void*)&scanhash_x22i_8way;
|
||||
gate->hash = (void*)&x22i_8way_hash;
|
||||
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | AVX512_OPT;
|
||||
#elif defined (X22I_4WAY)
|
||||
gate->scanhash = (void*)&scanhash_x22i_4way;
|
||||
gate->hash = (void*)&x22i_4way_hash;
|
||||
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | SHA_OPT | AVX512_OPT;
|
||||
#else
|
||||
gate->scanhash = (void*)&scanhash_x22i;
|
||||
gate->hash = (void*)&x22i_hash;
|
||||
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | SHA_OPT | AVX512_OPT;
|
||||
#endif
|
||||
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | SHA_OPT;
|
||||
return true;
|
||||
};
|
||||
|
||||
bool register_x25x_algo( algo_gate_t* gate )
|
||||
{
|
||||
#if defined (X22I_4WAY)
|
||||
#if defined (X25X_8WAY)
|
||||
gate->scanhash = (void*)&scanhash_x25x_8way;
|
||||
gate->hash = (void*)&x25x_8way_hash;
|
||||
// gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | AVX512_OPT;
|
||||
#elif defined (X25X_4WAY)
|
||||
gate->scanhash = (void*)&scanhash_x25x_4way;
|
||||
gate->hash = (void*)&x25x_4way_hash;
|
||||
// gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | SHA_OPT | AVX512_OPT;
|
||||
#else
|
||||
gate->scanhash = (void*)&scanhash_x25x;
|
||||
gate->hash = (void*)&x25x_hash;
|
||||
// gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | SHA_OPT | AVX512_OPT;
|
||||
#endif
|
||||
gate->optimizations = SSE2_OPT | AES_OPT | AVX2_OPT | SHA_OPT;
|
||||
|
||||
return true;
|
||||
};
|
||||
|
||||
|
||||
@@ -6,30 +6,64 @@
|
||||
#include <stdint.h>
|
||||
#include <unistd.h>
|
||||
|
||||
#if defined(__AVX2__) && defined(__AES__)
|
||||
#define X22I_4WAY
|
||||
#if defined(__AVX512F__) && defined(__AVX512VL__) && defined(__AVX512DQ__) && defined(__AVX512BW__)
|
||||
#define X22I_8WAY 1
|
||||
#elif defined(__AVX2__) && defined(__AES__)
|
||||
#define X22I_4WAY 1
|
||||
#endif
|
||||
|
||||
bool register_x22i__algo( algo_gate_t* gate );
|
||||
bool register_x22i_algo( algo_gate_t* gate );
|
||||
|
||||
#if defined(X22I_4WAY)
|
||||
#if defined(X22I_8WAY)
|
||||
|
||||
void x22i_8way_hash( void *state, const void *input );
|
||||
int scanhash_x22i_8way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
#elif defined(X22I_4WAY)
|
||||
|
||||
void x22i_4way_hash( void *state, const void *input );
|
||||
int scanhash_x22i_4way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
void x25x_4way_hash( void *state, const void *input );
|
||||
int scanhash_x25x_4way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
#endif
|
||||
#else
|
||||
|
||||
void x22i_hash( void *state, const void *input );
|
||||
int scanhash_x22i( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
#endif
|
||||
|
||||
|
||||
// Big problems with x25x 8 way. It blows up just by increasing the
|
||||
// buffer sizes and nothing else. It may have to do with accessing 2 dim arrays.
|
||||
|
||||
//#if defined(__AVX512F__) && defined(__AVX512VL__) && defined(__AVX512DQ__) && defined(__AVX512BW__)
|
||||
// #define X25X_8WAY 1
|
||||
#if defined(__AVX2__) && defined(__AES__)
|
||||
#define X25X_4WAY 1
|
||||
#endif
|
||||
|
||||
bool register_x25i_algo( algo_gate_t* gate );
|
||||
|
||||
#if defined(X25X_8WAY)
|
||||
|
||||
void x25x_8way_hash( void *state, const void *input );
|
||||
int scanhash_x25x_8way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
#elif defined(X25X_4WAY)
|
||||
|
||||
void x25x_4way_hash( void *state, const void *input );
|
||||
int scanhash_x25x_4way( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
#else
|
||||
|
||||
void x25x_hash( void *state, const void *input );
|
||||
int scanhash_x25x( struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done, struct thr_info *mythr );
|
||||
|
||||
#endif
|
||||
|
||||
#endif // X22I_GATE_H__
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
#include "x22i-gate.h"
|
||||
|
||||
#if defined(X22I_4WAY)
|
||||
#if defined(X25X_4WAY)
|
||||
|
||||
#include "algo/blake/blake-hash-4way.h"
|
||||
#include "algo/bmw/bmw-hash-4way.h"
|
||||
@@ -88,276 +88,282 @@ void x25x_4way_hash( void *output, const void *input )
|
||||
unsigned char hash2[25][64] __attribute__((aligned(64))) = {0};
|
||||
unsigned char hash3[25][64] __attribute__((aligned(64))) = {0};
|
||||
uint64_t vhash[8*4] __attribute__ ((aligned (64)));
|
||||
unsigned char vhashA[24][64*4] __attribute__ ((aligned (64)));
|
||||
// Doubling the size of vhashX breaks everything. It may have something
|
||||
// to do with accessing arrays: vhashX vs vhashX[0] vs &vhash[0].
|
||||
// Changing notation did seem to allow the larger buffer but still resulted
|
||||
// in problems further along.
|
||||
// unsigned char vhashX[24][64*8] __attribute__ ((aligned (64)));
|
||||
unsigned char vhashX[24][64*4] __attribute__ ((aligned (64)));
|
||||
x25x_4way_ctx_overlay ctx __attribute__ ((aligned (64)));
|
||||
|
||||
blake512_4way_init( &ctx.blake );
|
||||
blake512_4way( &ctx.blake, input, 80 );
|
||||
blake512_4way_close( &ctx.blake, vhash );
|
||||
dintrlv_4x64_512( &hash0[0], &hash1[0], &hash2[0], &hash3[0], vhash );
|
||||
dintrlv_4x64_512( hash0[0], hash1[0], hash2[0], hash3[0], vhash );
|
||||
|
||||
bmw512_4way_init( &ctx.bmw );
|
||||
bmw512_4way( &ctx.bmw, vhash, 64 );
|
||||
bmw512_4way_close( &ctx.bmw, vhash );
|
||||
dintrlv_4x64_512( &hash0[1], &hash1[1], &hash2[1], &hash3[1], vhash );
|
||||
dintrlv_4x64_512( hash0[1], hash1[1], hash2[1], hash3[1], vhash );
|
||||
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)&hash0[2],
|
||||
(const char*)&hash0[1], 512 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash0[2],
|
||||
(const char*)hash0[1], 512 );
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)&hash1[2],
|
||||
(const char*)&hash1[1], 512 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash1[2],
|
||||
(const char*)hash1[1], 512 );
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)&hash2[2],
|
||||
(const char*)&hash2[1], 512 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash2[2],
|
||||
(const char*)hash2[1], 512 );
|
||||
init_groestl( &ctx.groestl, 64 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)&hash3[2],
|
||||
(const char*)&hash3[1], 512 );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash3[2],
|
||||
(const char*)hash3[1], 512 );
|
||||
|
||||
intrlv_4x64_512( vhash, &hash0[2], &hash1[2], &hash2[2], &hash3[2] );
|
||||
intrlv_4x64_512( vhash, hash0[2], hash1[2], hash2[2], hash3[2] );
|
||||
|
||||
skein512_4way_init( &ctx.skein );
|
||||
skein512_4way( &ctx.skein, vhash, 64 );
|
||||
skein512_4way_close( &ctx.skein, vhash );
|
||||
dintrlv_4x64_512( &hash0[3], &hash1[3], &hash2[3], &hash3[3], vhash );
|
||||
dintrlv_4x64_512( hash0[3], hash1[3], hash2[3], hash3[3], vhash );
|
||||
|
||||
jh512_4way_init( &ctx.jh );
|
||||
jh512_4way( &ctx.jh, vhash, 64 );
|
||||
jh512_4way_close( &ctx.jh, vhash );
|
||||
dintrlv_4x64_512( &hash0[4], &hash1[4], &hash2[4], &hash3[4], vhash );
|
||||
dintrlv_4x64_512( hash0[4], hash1[4], hash2[4], hash3[4], vhash );
|
||||
|
||||
keccak512_4way_init( &ctx.keccak );
|
||||
keccak512_4way( &ctx.keccak, vhash, 64 );
|
||||
keccak512_4way_close( &ctx.keccak, vhash );
|
||||
dintrlv_4x64_512( &hash0[5], &hash1[5], &hash2[5], &hash3[5], vhash );
|
||||
dintrlv_4x64_512( hash0[5], hash1[5], hash2[5], hash3[5], vhash );
|
||||
|
||||
init_luffa( &ctx.luffa, 512 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)&hash0[6],
|
||||
(const BitSequence*)&hash0[5], 64 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash0[6],
|
||||
(const BitSequence*)hash0[5], 64 );
|
||||
init_luffa( &ctx.luffa, 512 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)&hash1[6],
|
||||
(const BitSequence*)&hash1[5], 64 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash1[6],
|
||||
(const BitSequence*)hash1[5], 64 );
|
||||
init_luffa( &ctx.luffa, 512 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)&hash2[6],
|
||||
(const BitSequence*)&hash2[5], 64 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash2[6],
|
||||
(const BitSequence*)hash2[5], 64 );
|
||||
init_luffa( &ctx.luffa, 512 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)&hash3[6],
|
||||
(const BitSequence*)&hash3[5], 64 );
|
||||
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash3[6],
|
||||
(const BitSequence*)hash3[5], 64 );
|
||||
|
||||
cubehashInit( &ctx.cube, 512, 16, 32 );
|
||||
cubehashUpdateDigest( &ctx.cube, (byte*) &hash0[7],
|
||||
(const byte*)&hash0[6], 64 );
|
||||
cubehashUpdateDigest( &ctx.cube, (byte*) hash0[7],
|
||||
(const byte*)hash0[6], 64 );
|
||||
cubehashInit( &ctx.cube, 512, 16, 32 );
|
||||
cubehashUpdateDigest( &ctx.cube, (byte*) &hash1[7],
|
||||
(const byte*)&hash1[6], 64 );
|
||||
cubehashUpdateDigest( &ctx.cube, (byte*) hash1[7],
|
||||
(const byte*)hash1[6], 64 );
|
||||
cubehashInit( &ctx.cube, 512, 16, 32 );
|
||||
cubehashUpdateDigest( &ctx.cube, (byte*) &hash2[7],
|
||||
(const byte*)&hash2[6], 64 );
|
||||
cubehashUpdateDigest( &ctx.cube, (byte*) hash2[7],
|
||||
(const byte*)hash2[6], 64 );
|
||||
cubehashInit( &ctx.cube, 512, 16, 32 );
|
||||
cubehashUpdateDigest( &ctx.cube, (byte*) &hash3[7],
|
||||
(const byte*)&hash3[6], 64 );
|
||||
cubehashUpdateDigest( &ctx.cube, (byte*) hash3[7],
|
||||
(const byte*)hash3[6], 64 );
|
||||
|
||||
sph_shavite512_init(&ctx.shavite);
|
||||
sph_shavite512(&ctx.shavite, (const void*) &hash0[7], 64);
|
||||
sph_shavite512_close(&ctx.shavite, &hash0[8]);
|
||||
sph_shavite512(&ctx.shavite, (const void*) hash0[7], 64);
|
||||
sph_shavite512_close(&ctx.shavite, hash0[8]);
|
||||
sph_shavite512_init(&ctx.shavite);
|
||||
sph_shavite512(&ctx.shavite, (const void*) &hash1[7], 64);
|
||||
sph_shavite512_close(&ctx.shavite, &hash1[8]);
|
||||
sph_shavite512(&ctx.shavite, (const void*) hash1[7], 64);
|
||||
sph_shavite512_close(&ctx.shavite, hash1[8]);
|
||||
sph_shavite512_init(&ctx.shavite);
|
||||
sph_shavite512(&ctx.shavite, (const void*) &hash2[7], 64);
|
||||
sph_shavite512_close(&ctx.shavite, &hash2[8]);
|
||||
sph_shavite512(&ctx.shavite, (const void*) hash2[7], 64);
|
||||
sph_shavite512_close(&ctx.shavite, hash2[8]);
|
||||
sph_shavite512_init(&ctx.shavite);
|
||||
sph_shavite512(&ctx.shavite, (const void*) &hash3[7], 64);
|
||||
sph_shavite512_close(&ctx.shavite, &hash3[8]);
|
||||
sph_shavite512(&ctx.shavite, (const void*) hash3[7], 64);
|
||||
sph_shavite512_close(&ctx.shavite, hash3[8]);
|
||||
|
||||
init_sd( &ctx.simd, 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence*)&hash0[9],
|
||||
(const BitSequence*)&hash0[8], 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence*)hash0[9],
|
||||
(const BitSequence*)hash0[8], 512 );
|
||||
init_sd( &ctx.simd, 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence*)&hash1[9],
|
||||
(const BitSequence*)&hash1[8], 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence*)hash1[9],
|
||||
(const BitSequence*)hash1[8], 512 );
|
||||
init_sd( &ctx.simd, 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence*)&hash2[9],
|
||||
(const BitSequence*)&hash2[8], 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence*)hash2[9],
|
||||
(const BitSequence*)hash2[8], 512 );
|
||||
init_sd( &ctx.simd, 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence*)&hash3[9],
|
||||
(const BitSequence*)&hash3[8], 512 );
|
||||
update_final_sd( &ctx.simd, (BitSequence*)hash3[9],
|
||||
(const BitSequence*)hash3[8], 512 );
|
||||
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence*)&hash0[10],
|
||||
(const BitSequence*)&hash0[9], 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence*)hash0[10],
|
||||
(const BitSequence*)hash0[9], 512 );
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence*)&hash1[10],
|
||||
(const BitSequence*)&hash1[9], 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence*)hash1[10],
|
||||
(const BitSequence*)hash1[9], 512 );
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence*)&hash2[10],
|
||||
(const BitSequence*)&hash2[9], 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence*)hash2[10],
|
||||
(const BitSequence*)hash2[9], 512 );
|
||||
init_echo( &ctx.echo, 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence*)&hash3[10],
|
||||
(const BitSequence*)&hash3[9], 512 );
|
||||
update_final_echo ( &ctx.echo, (BitSequence*)hash3[10],
|
||||
(const BitSequence*)hash3[9], 512 );
|
||||
|
||||
intrlv_4x64_512( vhash, &hash0[10], &hash1[10], &hash2[10], &hash3[10] );
|
||||
intrlv_4x64_512( vhash, hash0[10], hash1[10], hash2[10], hash3[10] );
|
||||
|
||||
hamsi512_4way_init( &ctx.hamsi );
|
||||
hamsi512_4way( &ctx.hamsi, vhash, 64 );
|
||||
hamsi512_4way_close( &ctx.hamsi, vhash );
|
||||
dintrlv_4x64_512( &hash0[11], &hash1[11], &hash2[11], &hash3[11], vhash );
|
||||
dintrlv_4x64_512( hash0[11], hash1[11], hash2[11], hash3[11], vhash );
|
||||
|
||||
sph_fugue512_init(&ctx.fugue);
|
||||
sph_fugue512(&ctx.fugue, (const void*) &hash0[11], 64);
|
||||
sph_fugue512_close(&ctx.fugue, &hash0[12]);
|
||||
sph_fugue512(&ctx.fugue, (const void*) hash0[11], 64);
|
||||
sph_fugue512_close(&ctx.fugue, hash0[12]);
|
||||
sph_fugue512_init(&ctx.fugue);
|
||||
sph_fugue512(&ctx.fugue, (const void*) &hash1[11], 64);
|
||||
sph_fugue512_close(&ctx.fugue, &hash1[12]);
|
||||
sph_fugue512(&ctx.fugue, (const void*) hash1[11], 64);
|
||||
sph_fugue512_close(&ctx.fugue, hash1[12]);
|
||||
sph_fugue512_init(&ctx.fugue);
|
||||
sph_fugue512(&ctx.fugue, (const void*) &hash2[11], 64);
|
||||
sph_fugue512_close(&ctx.fugue, &hash2[12]);
|
||||
sph_fugue512(&ctx.fugue, (const void*) hash2[11], 64);
|
||||
sph_fugue512_close(&ctx.fugue, hash2[12]);
|
||||
sph_fugue512_init(&ctx.fugue);
|
||||
sph_fugue512(&ctx.fugue, (const void*) &hash3[11], 64);
|
||||
sph_fugue512_close(&ctx.fugue, &hash3[12]);
|
||||
sph_fugue512(&ctx.fugue, (const void*) hash3[11], 64);
|
||||
sph_fugue512_close(&ctx.fugue, hash3[12]);
|
||||
|
||||
intrlv_4x32_512( vhash, &hash0[12], &hash1[12], &hash2[12], &hash3[12] );
|
||||
intrlv_4x32_512( vhash, hash0[12], hash1[12], hash2[12], hash3[12] );
|
||||
|
||||
shabal512_4way_init( &ctx.shabal );
|
||||
shabal512_4way( &ctx.shabal, vhash, 64 );
|
||||
shabal512_4way_close( &ctx.shabal, vhash );
|
||||
dintrlv_4x32_512( &hash0[13], &hash1[13], &hash2[13], &hash3[13], vhash );
|
||||
dintrlv_4x32_512( hash0[13], hash1[13], hash2[13], hash3[13], vhash );
|
||||
|
||||
sph_whirlpool_init(&ctx.whirlpool);
|
||||
sph_whirlpool (&ctx.whirlpool, (const void*) &hash0[13], 64);
|
||||
sph_whirlpool_close(&ctx.whirlpool, &hash0[14]);
|
||||
sph_whirlpool (&ctx.whirlpool, (const void*) hash0[13], 64);
|
||||
sph_whirlpool_close(&ctx.whirlpool, hash0[14]);
|
||||
sph_whirlpool_init(&ctx.whirlpool);
|
||||
sph_whirlpool (&ctx.whirlpool, (const void*) &hash1[13], 64);
|
||||
sph_whirlpool_close(&ctx.whirlpool, &hash1[14]);
|
||||
sph_whirlpool (&ctx.whirlpool, (const void*) hash1[13], 64);
|
||||
sph_whirlpool_close(&ctx.whirlpool, hash1[14]);
|
||||
sph_whirlpool_init(&ctx.whirlpool);
|
||||
sph_whirlpool (&ctx.whirlpool, (const void*) &hash2[13], 64);
|
||||
sph_whirlpool_close(&ctx.whirlpool, &hash2[14]);
|
||||
sph_whirlpool (&ctx.whirlpool, (const void*) hash2[13], 64);
|
||||
sph_whirlpool_close(&ctx.whirlpool, hash2[14]);
|
||||
sph_whirlpool_init(&ctx.whirlpool);
|
||||
sph_whirlpool (&ctx.whirlpool, (const void*) &hash3[13], 64);
|
||||
sph_whirlpool_close(&ctx.whirlpool, &hash3[14]);
|
||||
sph_whirlpool (&ctx.whirlpool, (const void*) hash3[13], 64);
|
||||
sph_whirlpool_close(&ctx.whirlpool, hash3[14]);
|
||||
|
||||
intrlv_4x64_512( vhash, &hash0[14], &hash1[14], &hash2[14], &hash3[14] );
|
||||
intrlv_4x64_512( vhash, hash0[14], hash1[14], hash2[14], hash3[14] );
|
||||
|
||||
sha512_4way_init( &ctx.sha512 );
|
||||
sha512_4way( &ctx.sha512, vhash, 64 );
|
||||
sha512_4way_close( &ctx.sha512, vhash );
|
||||
dintrlv_4x64_512( &hash0[15], &hash1[15], &hash2[15], &hash3[15], vhash );
|
||||
dintrlv_4x64_512( hash0[15], hash1[15], hash2[15], hash3[15], vhash );
|
||||
|
||||
|
||||
ComputeSingleSWIFFTX((unsigned char*)&hash0[12], (unsigned char*)&hash0[16]);
|
||||
ComputeSingleSWIFFTX((unsigned char*)&hash1[12], (unsigned char*)&hash1[16]);
|
||||
ComputeSingleSWIFFTX((unsigned char*)&hash2[12], (unsigned char*)&hash2[16]);
|
||||
ComputeSingleSWIFFTX((unsigned char*)&hash3[12], (unsigned char*)&hash3[16]);
|
||||
ComputeSingleSWIFFTX((unsigned char*)hash0[12], (unsigned char*)hash0[16]);
|
||||
ComputeSingleSWIFFTX((unsigned char*)hash1[12], (unsigned char*)hash1[16]);
|
||||
ComputeSingleSWIFFTX((unsigned char*)hash2[12], (unsigned char*)hash2[16]);
|
||||
ComputeSingleSWIFFTX((unsigned char*)hash3[12], (unsigned char*)hash3[16]);
|
||||
|
||||
intrlv_4x32_512( &vhashA, &hash0[16], &hash1[16], &hash2[16], &hash3[16] );
|
||||
intrlv_4x32_512( vhashX[0], hash0[16], hash1[16], hash2[16], hash3[16] );
|
||||
|
||||
memset( vhash, 0, 64*4 );
|
||||
|
||||
haval256_5_4way_init( &ctx.haval );
|
||||
haval256_5_4way( &ctx.haval, vhashA, 64 );
|
||||
haval256_5_4way( &ctx.haval, vhashX[0], 64 );
|
||||
haval256_5_4way_close( &ctx.haval, vhash );
|
||||
dintrlv_4x32_512( &hash0[17], &hash1[17], &hash2[17], &hash3[17], vhash );
|
||||
dintrlv_4x32_512( hash0[17], hash1[17], hash2[17], hash3[17], vhash );
|
||||
|
||||
sph_tiger_init(&ctx.tiger);
|
||||
sph_tiger (&ctx.tiger, (const void*) &hash0[17], 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) &hash0[18]);
|
||||
sph_tiger (&ctx.tiger, (const void*) hash0[17], 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) hash0[18]);
|
||||
sph_tiger_init(&ctx.tiger);
|
||||
sph_tiger (&ctx.tiger, (const void*) &hash1[17], 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) &hash1[18]);
|
||||
sph_tiger (&ctx.tiger, (const void*) hash1[17], 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) hash1[18]);
|
||||
sph_tiger_init(&ctx.tiger);
|
||||
sph_tiger (&ctx.tiger, (const void*) &hash2[17], 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) &hash2[18]);
|
||||
sph_tiger (&ctx.tiger, (const void*) hash2[17], 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) hash2[18]);
|
||||
sph_tiger_init(&ctx.tiger);
|
||||
sph_tiger (&ctx.tiger, (const void*) &hash3[17], 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) &hash3[18]);
|
||||
sph_tiger (&ctx.tiger, (const void*) hash3[17], 64);
|
||||
sph_tiger_close(&ctx.tiger, (void*) hash3[18]);
|
||||
|
||||
LYRA2RE( (void*)&hash0[19], 32, (const void*)&hash0[18], 32,
|
||||
(const void*)&hash0[18], 32, 1, 4, 4 );
|
||||
LYRA2RE( (void*)&hash1[19], 32, (const void*)&hash1[18], 32,
|
||||
(const void*)&hash1[18], 32, 1, 4, 4 );
|
||||
LYRA2RE( (void*)&hash2[19], 32, (const void*)&hash2[18], 32,
|
||||
(const void*)&hash2[18], 32, 1, 4, 4 );
|
||||
LYRA2RE( (void*)&hash3[19], 32, (const void*)&hash3[18], 32,
|
||||
(const void*)&hash3[18], 32, 1, 4, 4 );
|
||||
LYRA2RE( (void*)hash0[19], 32, (const void*)hash0[18], 32,
|
||||
(const void*)hash0[18], 32, 1, 4, 4 );
|
||||
LYRA2RE( (void*)hash1[19], 32, (const void*)hash1[18], 32,
|
||||
(const void*)hash1[18], 32, 1, 4, 4 );
|
||||
LYRA2RE( (void*)hash2[19], 32, (const void*)hash2[18], 32,
|
||||
(const void*)hash2[18], 32, 1, 4, 4 );
|
||||
LYRA2RE( (void*)hash3[19], 32, (const void*)hash3[18], 32,
|
||||
(const void*)hash3[18], 32, 1, 4, 4 );
|
||||
|
||||
sph_gost512_init(&ctx.gost);
|
||||
sph_gost512 (&ctx.gost, (const void*) &hash0[19], 64);
|
||||
sph_gost512_close(&ctx.gost, (void*) &hash0[20]);
|
||||
sph_gost512 (&ctx.gost, (const void*) hash0[19], 64);
|
||||
sph_gost512_close(&ctx.gost, (void*) hash0[20]);
|
||||
sph_gost512_init(&ctx.gost);
|
||||
sph_gost512 (&ctx.gost, (const void*) &hash1[19], 64);
|
||||
sph_gost512_close(&ctx.gost, (void*) &hash1[20]);
|
||||
sph_gost512 (&ctx.gost, (const void*) hash1[19], 64);
|
||||
sph_gost512_close(&ctx.gost, (void*) hash1[20]);
|
||||
sph_gost512_init(&ctx.gost);
|
||||
sph_gost512 (&ctx.gost, (const void*) &hash2[19], 64);
|
||||
sph_gost512_close(&ctx.gost, (void*) &hash2[20]);
|
||||
sph_gost512 (&ctx.gost, (const void*) hash2[19], 64);
|
||||
sph_gost512_close(&ctx.gost, (void*) hash2[20]);
|
||||
sph_gost512_init(&ctx.gost);
|
||||
sph_gost512 (&ctx.gost, (const void*) &hash3[19], 64);
|
||||
sph_gost512_close(&ctx.gost, (void*) &hash3[20]);
|
||||
sph_gost512 (&ctx.gost, (const void*) hash3[19], 64);
|
||||
sph_gost512_close(&ctx.gost, (void*) hash3[20]);
|
||||
|
||||
intrlv_4x32_512( vhashA, &hash0[20], &hash1[20], &hash2[20], &hash3[20] );
|
||||
intrlv_4x32_512( vhashX[0], hash0[20], hash1[20], hash2[20], hash3[20] );
|
||||
memset( vhash, 0, 64*4 );
|
||||
|
||||
sha256_4way_init( &ctx.sha256 );
|
||||
sha256_4way( &ctx.sha256, vhashA, 64 );
|
||||
sha256_4way( &ctx.sha256, vhashX[0], 64 );
|
||||
sha256_4way_close( &ctx.sha256, vhash );
|
||||
dintrlv_4x32_512( &hash0[21], &hash1[21], &hash2[21], &hash3[21], vhash );
|
||||
dintrlv_4x32_512( hash0[21], hash1[21], hash2[21], hash3[21], vhash );
|
||||
|
||||
sph_panama_init(&ctx.panama);
|
||||
sph_panama (&ctx.panama, (const void*) &hash0[21], 64 );
|
||||
sph_panama_close(&ctx.panama, (void*) &hash0[22]);
|
||||
sph_panama (&ctx.panama, (const void*) hash0[21], 64 );
|
||||
sph_panama_close(&ctx.panama, (void*) hash0[22]);
|
||||
sph_panama_init(&ctx.panama);
|
||||
sph_panama (&ctx.panama, (const void*) &hash1[21], 64 );
|
||||
sph_panama_close(&ctx.panama, (void*) &hash1[22]);
|
||||
sph_panama (&ctx.panama, (const void*) hash1[21], 64 );
|
||||
sph_panama_close(&ctx.panama, (void*) hash1[22]);
|
||||
sph_panama_init(&ctx.panama);
|
||||
sph_panama (&ctx.panama, (const void*) &hash2[21], 64 );
|
||||
sph_panama_close(&ctx.panama, (void*) &hash2[22]);
|
||||
sph_panama (&ctx.panama, (const void*) hash2[21], 64 );
|
||||
sph_panama_close(&ctx.panama, (void*) hash2[22]);
|
||||
sph_panama_init(&ctx.panama);
|
||||
sph_panama (&ctx.panama, (const void*) &hash3[21], 64 );
|
||||
sph_panama_close(&ctx.panama, (void*) &hash3[22]);
|
||||
sph_panama (&ctx.panama, (const void*) hash3[21], 64 );
|
||||
sph_panama_close(&ctx.panama, (void*) hash3[22]);
|
||||
|
||||
laneHash(512, (const BitSequence*)&hash0[22], 512, (BitSequence*)&hash0[23]);
|
||||
laneHash(512, (const BitSequence*)&hash1[22], 512, (BitSequence*)&hash1[23]);
|
||||
laneHash(512, (const BitSequence*)&hash2[22], 512, (BitSequence*)&hash2[23]);
|
||||
laneHash(512, (const BitSequence*)&hash3[22], 512, (BitSequence*)&hash3[23]);
|
||||
laneHash(512, (const BitSequence*)hash0[22], 512, (BitSequence*)hash0[23]);
|
||||
laneHash(512, (const BitSequence*)hash1[22], 512, (BitSequence*)hash1[23]);
|
||||
laneHash(512, (const BitSequence*)hash2[22], 512, (BitSequence*)hash2[23]);
|
||||
laneHash(512, (const BitSequence*)hash3[22], 512, (BitSequence*)hash3[23]);
|
||||
|
||||
x25x_shuffle( hash0 );
|
||||
x25x_shuffle( hash1 );
|
||||
x25x_shuffle( hash2 );
|
||||
x25x_shuffle( hash3 );
|
||||
|
||||
intrlv_4x32_512( &vhashA[ 0], &hash0[ 0], &hash1[ 0], &hash2[ 0], &hash3[ 0] );
|
||||
intrlv_4x32_512( &vhashA[ 1], &hash0[ 1], &hash1[ 1], &hash2[ 1], &hash3[ 1] );
|
||||
intrlv_4x32_512( &vhashA[ 2], &hash0[ 2], &hash1[ 2], &hash2[ 2], &hash3[ 2] );
|
||||
intrlv_4x32_512( &vhashA[ 3], &hash0[ 3], &hash1[ 3], &hash2[ 3], &hash3[ 3] );
|
||||
intrlv_4x32_512( &vhashA[ 4], &hash0[ 4], &hash1[ 4], &hash2[ 4], &hash3[ 4] );
|
||||
intrlv_4x32_512( &vhashA[ 5], &hash0[ 5], &hash1[ 5], &hash2[ 5], &hash3[ 5] );
|
||||
intrlv_4x32_512( &vhashA[ 6], &hash0[ 6], &hash1[ 6], &hash2[ 6], &hash3[ 6] );
|
||||
intrlv_4x32_512( &vhashA[ 7], &hash0[ 7], &hash1[ 7], &hash2[ 7], &hash3[ 7] );
|
||||
intrlv_4x32_512( &vhashA[ 8], &hash0[ 8], &hash1[ 8], &hash2[ 8], &hash3[ 8] );
|
||||
intrlv_4x32_512( &vhashA[ 9], &hash0[ 9], &hash1[ 9], &hash2[ 9], &hash3[ 9] );
|
||||
intrlv_4x32_512( &vhashA[10], &hash0[10], &hash1[10], &hash2[10], &hash3[10] );
|
||||
intrlv_4x32_512( &vhashA[11], &hash0[11], &hash1[11], &hash2[11], &hash3[11] );
|
||||
intrlv_4x32_512( &vhashA[12], &hash0[12], &hash1[12], &hash2[12], &hash3[12] );
|
||||
intrlv_4x32_512( &vhashA[13], &hash0[13], &hash1[13], &hash2[13], &hash3[13] );
|
||||
intrlv_4x32_512( &vhashA[14], &hash0[14], &hash1[14], &hash2[14], &hash3[14] );
|
||||
intrlv_4x32_512( &vhashA[15], &hash0[15], &hash1[15], &hash2[15], &hash3[15] );
|
||||
intrlv_4x32_512( &vhashA[16], &hash0[16], &hash1[16], &hash2[16], &hash3[16] );
|
||||
intrlv_4x32_512( &vhashA[17], &hash0[17], &hash1[17], &hash2[17], &hash3[17] );
|
||||
intrlv_4x32_512( &vhashA[18], &hash0[18], &hash1[18], &hash2[18], &hash3[18] );
|
||||
intrlv_4x32_512( &vhashA[19], &hash0[19], &hash1[19], &hash2[19], &hash3[19] );
|
||||
intrlv_4x32_512( &vhashA[20], &hash0[20], &hash1[20], &hash2[20], &hash3[20] );
|
||||
intrlv_4x32_512( &vhashA[21], &hash0[21], &hash1[21], &hash2[21], &hash3[21] );
|
||||
intrlv_4x32_512( &vhashA[22], &hash0[22], &hash1[22], &hash2[22], &hash3[22] );
|
||||
intrlv_4x32_512( &vhashA[23], &hash0[23], &hash1[23], &hash2[23], &hash3[23] );
|
||||
intrlv_4x32_512( vhashX[ 0], hash0[ 0], hash1[ 0], hash2[ 0], hash3[ 0] );
|
||||
intrlv_4x32_512( vhashX[ 1], hash0[ 1], hash1[ 1], hash2[ 1], hash3[ 1] );
|
||||
intrlv_4x32_512( vhashX[ 2], hash0[ 2], hash1[ 2], hash2[ 2], hash3[ 2] );
|
||||
intrlv_4x32_512( vhashX[ 3], hash0[ 3], hash1[ 3], hash2[ 3], hash3[ 3] );
|
||||
intrlv_4x32_512( vhashX[ 4], hash0[ 4], hash1[ 4], hash2[ 4], hash3[ 4] );
|
||||
intrlv_4x32_512( vhashX[ 5], hash0[ 5], hash1[ 5], hash2[ 5], hash3[ 5] );
|
||||
intrlv_4x32_512( vhashX[ 6], hash0[ 6], hash1[ 6], hash2[ 6], hash3[ 6] );
|
||||
intrlv_4x32_512( vhashX[ 7], hash0[ 7], hash1[ 7], hash2[ 7], hash3[ 7] );
|
||||
intrlv_4x32_512( vhashX[ 8], hash0[ 8], hash1[ 8], hash2[ 8], hash3[ 8] );
|
||||
intrlv_4x32_512( vhashX[ 9], hash0[ 9], hash1[ 9], hash2[ 9], hash3[ 9] );
|
||||
intrlv_4x32_512( vhashX[10], hash0[10], hash1[10], hash2[10], hash3[10] );
|
||||
intrlv_4x32_512( vhashX[11], hash0[11], hash1[11], hash2[11], hash3[11] );
|
||||
intrlv_4x32_512( vhashX[12], hash0[12], hash1[12], hash2[12], hash3[12] );
|
||||
intrlv_4x32_512( vhashX[13], hash0[13], hash1[13], hash2[13], hash3[13] );
|
||||
intrlv_4x32_512( vhashX[14], hash0[14], hash1[14], hash2[14], hash3[14] );
|
||||
intrlv_4x32_512( vhashX[15], hash0[15], hash1[15], hash2[15], hash3[15] );
|
||||
intrlv_4x32_512( vhashX[16], hash0[16], hash1[16], hash2[16], hash3[16] );
|
||||
intrlv_4x32_512( vhashX[17], hash0[17], hash1[17], hash2[17], hash3[17] );
|
||||
intrlv_4x32_512( vhashX[18], hash0[18], hash1[18], hash2[18], hash3[18] );
|
||||
intrlv_4x32_512( vhashX[19], hash0[19], hash1[19], hash2[19], hash3[19] );
|
||||
intrlv_4x32_512( vhashX[20], hash0[20], hash1[20], hash2[20], hash3[20] );
|
||||
intrlv_4x32_512( vhashX[21], hash0[21], hash1[21], hash2[21], hash3[21] );
|
||||
intrlv_4x32_512( vhashX[22], hash0[22], hash1[22], hash2[22], hash3[22] );
|
||||
intrlv_4x32_512( vhashX[23], hash0[23], hash1[23], hash2[23], hash3[23] );
|
||||
|
||||
blake2s_4way_init( &ctx.blake2s, 32 );
|
||||
blake2s_4way_full_blocks( &ctx.blake2s, vhash, vhashA, 64*24 );
|
||||
|
||||
dintrlv_4x32( &hash0[24], &hash1[24], &hash2[24], &hash3[24], vhash, 256 );
|
||||
blake2s_4way_full_blocks( &ctx.blake2s, output, vhashX, 64*24 );
|
||||
/*
|
||||
dintrlv_4x32( hash0[24], hash1[24], hash2[24], hash3[24], vhash, 256 );
|
||||
|
||||
memcpy(output, &hash0[24], 32);
|
||||
memcpy(output+32, &hash1[24], 32);
|
||||
memcpy(output+64, &hash2[24], 32);
|
||||
memcpy(output+96, &hash3[24], 32);
|
||||
memcpy(output, hash0[24], 32);
|
||||
memcpy(output+32, hash1[24], 32);
|
||||
memcpy(output+64, hash2[24], 32);
|
||||
memcpy(output+96, hash3[24], 32);
|
||||
*/
|
||||
}
|
||||
|
||||
int scanhash_x25x_4way( struct work* work, uint32_t max_nonce,
|
||||
@@ -365,11 +371,14 @@ int scanhash_x25x_4way( struct work* work, uint32_t max_nonce,
|
||||
{
|
||||
uint32_t hash[4*16] __attribute__ ((aligned (64)));
|
||||
uint32_t vdata[24*4] __attribute__ ((aligned (64)));
|
||||
uint32_t lane_hash[8] __attribute__ ((aligned (32)));
|
||||
uint32_t *hash7 = &(hash[7<<2]);
|
||||
uint32_t *pdata = work->data;
|
||||
uint32_t *ptarget = work->target;
|
||||
const uint32_t first_nonce = pdata[19];
|
||||
__m256i *noncev = (__m256i*)vdata + 9; // aligned
|
||||
uint32_t n = first_nonce;
|
||||
const uint32_t last_nonce = max_nonce - 4;
|
||||
const int thr_id = mythr->id;
|
||||
const uint32_t Htarg = ptarget[7];
|
||||
|
||||
@@ -385,6 +394,16 @@ int scanhash_x25x_4way( struct work* work, uint32_t max_nonce,
|
||||
_mm256_set_epi32( n+3, 0, n+2, 0, n+1, 0, n, 0 ) ), *noncev );
|
||||
x25x_4way_hash( hash, vdata );
|
||||
|
||||
for ( int lane = 0; lane < 4; lane++ ) if ( hash7[lane] <= Htarg )
|
||||
{
|
||||
extr_lane_4x32( lane_hash, hash, lane, 256 );
|
||||
if ( fulltest( lane_hash, ptarget ) && !opt_benchmark )
|
||||
{
|
||||
pdata[19] = n + lane;
|
||||
submit_lane_solution( work, lane_hash, mythr, lane );
|
||||
}
|
||||
}
|
||||
/*
|
||||
for ( int i = 0; i < 4; i++ )
|
||||
if ( unlikely( (hash+(i<<3))[7] <= Htarg ) )
|
||||
if( likely( fulltest( hash+(i<<3), ptarget ) && !opt_benchmark ) )
|
||||
@@ -392,10 +411,11 @@ int scanhash_x25x_4way( struct work* work, uint32_t max_nonce,
|
||||
pdata[19] = n+i;
|
||||
submit_lane_solution( work, hash+(i<<3), mythr, i );
|
||||
}
|
||||
*/
|
||||
n += 4;
|
||||
} while ( likely( ( n < max_nonce - 4 ) && !work_restart[thr_id].restart ) );
|
||||
} while ( likely( ( n < last_nonce ) && !work_restart[thr_id].restart ) );
|
||||
|
||||
*hashes_done = n - first_nonce + 1;
|
||||
*hashes_done = n - first_nonce;
|
||||
return 0;
|
||||
}
|
||||
|
||||
|
||||
Reference in New Issue
Block a user