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

v23.5

Browse Source
This commit is contained in:
Jay D Dee
2023-10-25 20:36:20 -04:00
parent 31c4dedf59
commit 160608cce5
180 changed files with 10318 additions and 13097 deletions
Download Patch File Download Diff File Expand all files Collapse all files

44
algo/x22/x22i.c
View File

@@ -18,7 +18,11 @@
#include "algo/keccak/sph_keccak.h"
#include "algo/cubehash/cubehash_sse2.h"
#include "algo/shavite/sph_shavite.h"
#include "algo/simd/nist.h"
#if defined(__aarch64__)
#include "algo/simd/sph_simd.h"
#else
#include "algo/simd/nist.h"
#endif
#include "algo/hamsi/sph_hamsi.h"
#include "algo/shabal/sph_shabal.h"
#include "algo/whirlpool/sph_whirlpool.h"
@@ -29,11 +33,7 @@
#include "algo/lyra2/lyra2.h"
#include "algo/gost/sph_gost.h"
#include "algo/swifftx/swifftx.h"
#if defined(__aarch64__)
#include "algo/luffa/sph_luffa.h"
#else
#include "algo/luffa/luffa_for_sse2.h"
#endif
#include "algo/luffa/luffa_for_sse2.h"
union _x22i_context_overlay
{
@@ -51,14 +51,14 @@ union _x22i_context_overlay
sph_jh512_context jh;
sph_keccak512_context keccak;
sph_skein512_context skein;
#if defined(__aarch64__)
sph_luffa512_context luffa;
#else
hashState_luffa luffa;
#endif
cubehashParam cube;
sph_shavite512_context shavite;
#if defined(__aarch64__)
sph_simd512_context simd;
#else
hashState_sd simd;
#endif
sph_hamsi512_context hamsi;
sph_shabal512_context shabal;
sph_whirlpool_context whirlpool;
@@ -107,27 +107,23 @@ int x22i_hash( void *output, const void *input, int thrid )
if ( work_restart[thrid].restart ) return 0;
#if defined(__aarch64__)
sph_luffa512_init(&ctx.luffa );
sph_luffa512(&ctx.luffa, (const void*) hash, 64);
sph_luffa512_close(&ctx.luffa, hash);
#else
init_luffa( &ctx.luffa, 512 );
update_and_final_luffa( &ctx.luffa, (BitSequence*)hash,
(const BitSequence*)hash, 64 );
#endif
luffa_full( &ctx.luffa, hash, 512, hash, 64 );
cubehashInit( &ctx.cube, 512, 16, 32 );
cubehashUpdateDigest( &ctx.cube, (byte*) hash,
(const byte*)hash, 64 );
cubehashUpdateDigest( &ctx.cube, hash, hash, 64 );
sph_shavite512_init(&ctx.shavite);
sph_shavite512(&ctx.shavite, (const void*) hash, 64);
sph_shavite512_close(&ctx.shavite, hash);
init_sd( &ctx.simd, 512 );
update_final_sd( &ctx.simd, (BitSequence*)hash,
(const BitSequence*)hash, 512 );
#if defined(__aarch64__)
sph_simd512_init(&ctx.simd );
sph_simd512(&ctx.simd, (const void*) hash, 64);
sph_simd512_close(&ctx.simd, hash);
#else
simd_full( &ctx.simd, (BitSequence *)hash,
(const BitSequence *)hash, 512 );
#endif
#if defined(__AES__)
init_echo( &ctx.echo, 512 );

20
algo/x22/x25x-4way.c
View File

@@ -572,11 +572,11 @@ int scanhash_x25x_8way( struct work *work, uint32_t max_nonce,
const __m512i eight = _mm512_set1_epi64( 8 );
if ( bench ) ptarget[7] = 0x08ff;
edata[0] = mm128_swap64_32( casti_m128i( pdata, 0 ) );
edata[1] = mm128_swap64_32( casti_m128i( pdata, 1 ) );
edata[2] = mm128_swap64_32( casti_m128i( pdata, 2 ) );
edata[3] = mm128_swap64_32( casti_m128i( pdata, 3 ) );
edata[4] = mm128_swap64_32( casti_m128i( pdata, 4 ) );
edata[0] = v128_swap64_32( casti_v128( pdata, 0 ) );
edata[1] = v128_swap64_32( casti_v128( pdata, 1 ) );
edata[2] = v128_swap64_32( casti_v128( pdata, 2 ) );
edata[3] = v128_swap64_32( casti_v128( pdata, 3 ) );
edata[4] = v128_swap64_32( casti_v128( pdata, 4 ) );
mm512_intrlv80_8x64( vdata, edata );
*noncev = _mm512_add_epi32( *noncev, _mm512_set_epi32(
@@ -930,11 +930,11 @@ int scanhash_x25x_4way( struct work* work, uint32_t max_nonce,
if ( bench ) ptarget[7] = 0x08ff;
edata[0] = mm128_swap64_32( casti_m128i( pdata, 0 ) );
edata[1] = mm128_swap64_32( casti_m128i( pdata, 1 ) );
edata[2] = mm128_swap64_32( casti_m128i( pdata, 2 ) );
edata[3] = mm128_swap64_32( casti_m128i( pdata, 3 ) );
edata[4] = mm128_swap64_32( casti_m128i( pdata, 4 ) );
edata[0] = v128_swap64_32( casti_v128( pdata, 0 ) );
edata[1] = v128_swap64_32( casti_v128( pdata, 1 ) );
edata[2] = v128_swap64_32( casti_v128( pdata, 2 ) );
edata[3] = v128_swap64_32( casti_v128( pdata, 3 ) );
edata[4] = v128_swap64_32( casti_v128( pdata, 4 ) );
mm256_intrlv80_4x64( vdata, edata );
*noncev = _mm256_add_epi32( *noncev, _mm256_set_epi32(

42
algo/x22/x25x.c
View File

@@ -18,7 +18,11 @@
#include "algo/keccak/sph_keccak.h"
#include "algo/cubehash/cubehash_sse2.h"
#include "algo/shavite/sph_shavite.h"
#include "algo/simd/nist.h"
#if defined(__aarch64__)
#include "algo/simd/sph_simd.h"
#else
#include "algo/simd/nist.h"
#endif
#include "algo/hamsi/sph_hamsi.h"
#include "algo/shabal/sph_shabal.h"
#include "algo/whirlpool/sph_whirlpool.h"
@@ -32,11 +36,7 @@
#include "algo/blake/sph-blake2s.h"
#include "algo/panama/sph_panama.h"
#include "algo/lanehash/lane.h"
#if defined(__aarch64__)
#include "algo/luffa/sph_luffa.h"
#else
#include "algo/luffa/luffa_for_sse2.h"
#endif
#include "algo/luffa/luffa_for_sse2.h"
union _x25x_context_overlay
{
@@ -54,14 +54,14 @@ union _x25x_context_overlay
sph_jh512_context jh;
sph_keccak512_context keccak;
sph_skein512_context skein;
#if defined(__aarch64__)
sph_luffa512_context luffa;
#else
hashState_luffa luffa;
#endif
cubehashParam cube;
sph_shavite512_context shavite;
#if defined(__aarch64__)
sph_simd512_context simd;
#else
hashState_sd simd;
#endif
sph_hamsi512_context hamsi;
sph_shabal512_context shabal;
sph_whirlpool_context whirlpool;
@@ -112,27 +112,23 @@ int x25x_hash( void *output, const void *input, int thrid )
if ( work_restart[thrid].restart ) return 0;
#if defined(__aarch64__)
sph_luffa512_init(&ctx.luffa );
sph_luffa512(&ctx.luffa, (const void*) hash, 64);
sph_luffa512_close(&ctx.luffa, hash);
#else
init_luffa( &ctx.luffa, 512 );
update_and_final_luffa( &ctx.luffa, (BitSequence*)&hash[6],
(const BitSequence*)&hash[5], 64 );
#endif
luffa_full( &ctx.luffa, &hash[6], 512, &hash[5], 64 );
cubehashInit( &ctx.cube, 512, 16, 32 );
cubehashUpdateDigest( &ctx.cube, (byte*) &hash[7],
(const byte*)&hash[6], 64 );
cubehashUpdateDigest( &ctx.cube, &hash[7], &hash[6], 64 );
sph_shavite512_init(&ctx.shavite);
sph_shavite512(&ctx.shavite, (const void*) &hash[7], 64);
sph_shavite512_close(&ctx.shavite, &hash[8]);
init_sd( &ctx.simd, 512 );
update_final_sd( &ctx.simd, (BitSequence*)&hash[9],
(const BitSequence*)&hash[8], 512 );
#if defined(__aarch64__)
sph_simd512(&ctx.simd, (const void*) &hash[8], 64);
sph_simd512_close(&ctx.simd, &hash[9] );
#else
update_final_sd( &ctx.simd, (BitSequence *)&hash[9],
(const BitSequence *)&hash[8], 512 );
#endif
#if defined(__AES__)
init_echo( &ctx.echo, 512 );