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

v24.2

Browse Source
This commit is contained in:
Jay D Dee
2024-05-20 23:08:50 -04:00
parent 4f930574cc
commit 042d13d1e1
129 changed files with 835 additions and 538 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
algo/lyra2/allium-4way.c
View File

@@ -15,7 +15,7 @@
#include "algo/groestl/sph_groestl.h"
#endif
#if defined(__AVX512F__) && defined(__AVX512VL__) && defined(__AVX512DQ__) && defined(__AVX512BW__)
#if defined(SIMD512)
#define ALLIUM_16WAY 1
#elif defined(__AVX2__)
#define ALLIUM_8WAY 1

6
algo/lyra2/lyra2-gate.h
View File

@@ -5,7 +5,7 @@
#include <stdint.h>
#include "lyra2.h"
#if defined(__AVX512F__) && defined(__AVX512VL__) && defined(__AVX512DQ__) && defined(__AVX512BW__)
#if defined(SIMD512)
#define LYRA2REV3_16WAY 1
#elif defined(__AVX2__)
#define LYRA2REV3_8WAY 1
@@ -49,7 +49,7 @@ bool init_lyra2rev3_ctx();
//////////////////////////////////
#if defined(__AVX512F__) && defined(__AVX512VL__) && defined(__AVX512DQ__) && defined(__AVX512BW__)
#if defined(SIMD512)
#define LYRA2REV2_16WAY 1
#elif defined(__AVX2__)
#define LYRA2REV2_8WAY 1
@@ -108,7 +108,7 @@ bool lyra2h_thread_init();
/////////////////////////////////////////
#if defined(__AVX512F__) && defined(__AVX512VL__) && defined(__AVX512DQ__) && defined(__AVX512BW__)
#if defined(SIMD512)
#define PHI2_8WAY 1
#elif defined(__AVX2__) && defined(__AES__)
#define PHI2_4WAY 1

2
algo/lyra2/lyra2-hash-2way.c
View File

@@ -41,7 +41,7 @@
// lyra2z330, lyra2h,
#if defined(__AVX512F__) && defined(__AVX512VL__) && defined(__AVX512DQ__) && defined(__AVX512BW__)
#if defined(SIMD512)
/**
* Executes Lyra2 based on the G function from Blake2b. This version supports salts and passwords

2
algo/lyra2/lyra2.h
View File

@@ -59,7 +59,7 @@ int LYRA2Z( uint64_t*, void *K, uint64_t kLen, const void *pwd,
int LYRA2(void *K, int64_t kLen, const void *pwd, int32_t pwdlen, const void *salt, int32_t saltlen, int64_t timeCost, const int16_t nRows, const int16_t nCols);
#if defined(__AVX512F__) && defined(__AVX512VL__) && defined(__AVX512DQ__) && defined(__AVX512BW__)
#if defined(SIMD512)
int LYRA2RE_2WAY( void *K, uint64_t kLen, const void *pwd, uint64_t pwdlen,
uint64_t timeCost, uint64_t nRows, uint64_t nCols );

2
algo/lyra2/lyra2z-4way.c
View File

@@ -3,7 +3,7 @@
#include "lyra2.h"
#include "algo/blake/blake256-hash.h"
#if defined(__AVX512F__) && defined(__AVX512VL__) && defined(__AVX512DQ__) && defined(__AVX512BW__)
#if defined(SIMD512)
#define LYRA2Z_16WAY 1
#elif defined(__AVX2__)
#define LYRA2Z_8WAY 1

2
algo/lyra2/phi2-4way.c
View File

@@ -4,7 +4,7 @@
#include "algo/gost/sph_gost.h"
#include "algo/cubehash/cubehash_sse2.h"
#include "lyra2.h"
#if defined(__VAES__) && defined(__AVX512F__) && defined(__AVX512VL__) && defined(__AVX512DQ__) && defined(__AVX512BW__)
#if defined(__VAES__) && defined(SIMD512)
#include "algo/echo/echo-hash-4way.h"
#elif defined(__AES__)
#include "algo/echo/aes_ni/hash_api.h"

2
algo/lyra2/sponge-2way.c
View File

@@ -27,7 +27,7 @@
#include "lyra2.h"
#include "simd-utils.h"
#if defined(__AVX512F__) && defined(__AVX512VL__) && defined(__AVX512DQ__) && defined(__AVX512BW__)
#if defined(SIMD512)
inline void squeeze_2way( uint64_t *State, byte *Out, unsigned int len )
{

12
algo/lyra2/sponge.h
View File

@@ -43,7 +43,7 @@ static const uint64_t blake2b_IV[8] =
0x1f83d9abfb41bd6bULL, 0x5be0cd19137e2179ULL
};
#if defined(__AVX512F__) && defined(__AVX512VL__) && defined(__AVX512DQ__) && defined(__AVX512BW__)
#if defined(SIMD512)
#define G2W_4X64(a,b,c,d) \
a = _mm512_add_epi64( a, b ); \
@@ -150,13 +150,13 @@ static const uint64_t blake2b_IV[8] =
// returns void, all args updated
#define G_2X64(a,b,c,d) \
a = v128_add64( a, b ); \
d = v128_ror64( v128_xor( d, a), 32 ); \
d = v128_ror64xor( d, a, 32 ); \
c = v128_add64( c, d ); \
b = v128_ror64( v128_xor( b, c ), 24 ); \
b = v128_ror64xor( b, c, 24 ); \
a = v128_add64( a, b ); \
d = v128_ror64( v128_xor( d, a ), 16 ); \
d = v128_ror64xor( d, a, 16 ); \
c = v128_add64( c, d ); \
b = v128_ror64( v128_xor( b, c ), 63 );
b = v128_ror64xor( b, c, 63 );
#define LYRA_ROUND_AVX(s0,s1,s2,s3,s4,s5,s6,s7) \
{ \
@@ -222,7 +222,7 @@ static inline uint64_t rotr64( const uint64_t w, const unsigned c ){
G( r, 7, v[ 3], v[ 4], v[ 9], v[14] );
#if defined(__AVX512F__) && defined(__AVX512VL__) && defined(__AVX512DQ__) && defined(__AVX512BW__)
#if defined(SIMD512)
union _ovly_512
{