GOSTSec
/
ccminer-gostd-lite
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
Browse Source

decred: and even faster implementation by Alexis 

optimized for the 9xx and more recent, same results on the 750 Ti
+ restore second nonce support not present in nicehash published version

Better on linux at least...
master
Tanguy Pruvot 9 years ago
parent
7e490693e0
commit
c643b3b900
1 changed files with 396 additions and 281 deletions
Whitespace
Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL
Unified View
Diff Options
Show Stats Download Patch File Download Diff File
  1. 661
      Algo256/decred.cu

661
Algo256/decred.cu
Unescape View File

@ -2,7 +2,10 @@
* Blake-256 Decred 180-Bytes input Cuda Kernel (Tested on SM 5/5.2) * Blake-256 Decred 180-Bytes input Cuda Kernel (Tested on SM 5/5.2)
* *
* Tanguy Pruvot - Feb 2016 * Tanguy Pruvot - Feb 2016
* Alexis Provos - Mar 2016 *
* Merged 8-round blake (XVC) tweaks
* Further improved by: ~2.72%
* Alexis Provos - Jun 2016
*/ */
#include <stdint.h> #include <stdint.h>
@ -14,7 +17,11 @@ extern "C" {
} }
/* threads per block */ /* threads per block */
#define TPB 640 #define TPB 768
#define NPT 192
#define maxResults 8
/* max count of found nonces in one call */
#define NBN 2
/* hash by cpu with blake 256 */ /* hash by cpu with blake 256 */
extern "C" void decred_hash(void *output, const void *input) extern "C" void decred_hash(void *output, const void *input)
@ -32,166 +39,298 @@ extern "C" void decred_hash(void *output, const void *input)
#ifdef __INTELLISENSE__ #ifdef __INTELLISENSE__
#define __byte_perm(x, y, b) x #define __byte_perm(x, y, b) x
#define atomicInc(p, max) (*p)
#endif #endif
__constant__ uint32_t _ALIGN(16) c_data[32]; __constant__ uint32_t c_m[3];
__constant__ uint32_t _ALIGN(16) c_h[ 2]; __constant__ uint32_t _ALIGN(8) c_h[2];
__constant__ uint32_t _ALIGN(16) c_xors[215]; __constant__ uint32_t _ALIGN(32) c_v[16];
__constant__ uint32_t _ALIGN(32) c_x[90];
/* 16 adapters max */ /* Buffers of candidate nonce(s) */
static uint32_t *d_resNonce[MAX_GPUS]; static uint32_t *d_resNonce[MAX_GPUS];
static uint32_t *h_resNonce[MAX_GPUS]; static uint32_t *h_resNonce[MAX_GPUS];
/* macro bodies */ __device__ __forceinline__
#define pxorGS(a,b,c,d) { \ uint32_t ROR8(const uint32_t a) {
v[a]+= c_xors[i++] + v[b]; \ return __byte_perm(a, 0, 0x0321);
v[d] = __byte_perm(v[d] ^ v[a], 0, 0x1032); \ }
__device__ __forceinline__
uint32_t ROL16(const uint32_t a) {
return __byte_perm(a, 0, 0x1032);
}
__device__ __forceinline__
uint32_t xor3x(uint32_t a, uint32_t b, uint32_t c) {
uint32_t result;
#if __CUDA_ARCH__ >= 500 && CUDA_VERSION >= 7050
asm ("lop3.b32 %0, %1, %2, %3, 0x96;" : "=r"(result) : "r"(a), "r"(b),"r"(c)); //0x96 = 0xF0 ^ 0xCC ^ 0xAA
#else
result = a^b^c;
#endif
return result;
}
#define GSn(a,b,c,d,x,y) { \
v[a]+= x + v[b]; \
v[d] = ROL16(v[d] ^ v[a]); \
v[c]+= v[d]; \ v[c]+= v[d]; \
v[b] = ROTR32(v[b] ^ v[c], 12); \ v[b] = ROTR32(v[b] ^ v[c], 12); \
v[a]+= c_xors[i++] + v[b]; \ v[a]+= y + v[b]; \
v[d] = __byte_perm(v[d] ^ v[a], 0, 0x0321); \ v[d] = ROR8(v[d] ^ v[a]); \
v[c]+= v[d]; \ v[c]+= v[d]; \
v[b] = ROTR32(v[b] ^ v[c], 7); \ v[b] = ROTR32(v[b] ^ v[c], 7); \
} }
#define pxorGS2(a,b,c,d,a1,b1,c1,d1) {\ #define GSn3(a,b,c,d,x,y, a1,b1,c1,d1,x1,y1, a2,b2,c2,d2,x2,y2) { \
v[ a]+= c_xors[i++] + v[ b]; v[a1]+= c_xors[i++] + v[b1]; \ v[ a]+= x + v[ b]; v[a1]+= x1 + v[b1]; v[a2]+= x2 + v[b2];\
v[ d] = __byte_perm(v[ d] ^ v[ a], 0, 0x1032); v[d1] = __byte_perm(v[d1] ^ v[a1], 0, 0x1032); \ v[ d] = ROL16(v[ d] ^ v[ a]); v[d1] = ROL16(v[d1] ^ v[a1]); v[d2] = ROL16(v[d2] ^ v[a2]);\
v[ c]+= v[ d]; v[c1]+= v[d1]; \ v[ c]+= v[ d]; v[c1]+= v[d1]; v[c2]+= v[d2];\
v[ b] = ROTR32(v[ b] ^ v[ c], 12); v[b1] = ROTR32(v[b1] ^ v[c1], 12); \ v[ b] = ROTR32(v[ b] ^ v[ c], 12); v[b1] = ROTR32(v[b1] ^ v[c1], 12); v[b2] = ROTR32(v[b2] ^ v[c2], 12);\
v[ a]+= c_xors[i++] + v[ b]; v[a1]+= c_xors[i++] + v[b1]; \ v[ a]+= y + v[ b]; v[a1]+= y1 + v[b1]; v[a2]+= y2 + v[b2];\
v[ d] = __byte_perm(v[ d] ^ v[ a], 0, 0x0321); v[d1] = __byte_perm(v[d1] ^ v[a1], 0, 0x0321); \ v[ d] = ROR8(v[ d] ^ v[ a]); v[d1] = ROR8(v[d1] ^ v[a1]); v[d2] = ROR8(v[d2] ^ v[a2]);\
v[ c]+= v[ d]; v[c1]+= v[d1]; \ v[ c]+= v[ d]; v[c1]+= v[d1]; v[c2]+= v[d2];\
v[ b] = ROTR32(v[ b] ^ v[ c], 7); v[b1] = ROTR32(v[b1] ^ v[c1], 7); \ v[ b] = ROTR32(v[ b] ^ v[ c], 7); v[b1] = ROTR32(v[b1] ^ v[c1], 7); v[b2] = ROTR32(v[b2] ^ v[c2], 7);\
} }
#define pxory1GS2(a,b,c,d,a1,b1,c1,d1) { \ #define GSn4(a,b,c,d,x,y, a1,b1,c1,d1,x1,y1, a2,b2,c2,d2,x2,y2, a3,b3,c3,d3,x3,y3) { \
v[ a]+= c_xors[i++] + v[ b]; v[a1]+= c_xors[i++] + v[b1]; \ v[ a]+= x + v[ b]; v[a1]+= x1 + v[b1]; v[a2]+= x2 + v[b2]; v[a3]+= x3 + v[b3]; \
v[ d] = __byte_perm(v[ d] ^ v[ a], 0, 0x1032); v[d1] = __byte_perm(v[d1] ^ v[a1], 0, 0x1032); \ v[ d] = ROL16(v[ d] ^ v[ a]); v[d1] = ROL16(v[d1] ^ v[a1]); v[d2] = ROL16(v[d2] ^ v[a2]); v[d3] = ROL16(v[d3] ^ v[a3]); \
v[ c]+= v[ d]; v[c1]+= v[d1]; \ v[ c]+= v[ d]; v[c1]+= v[d1]; v[c2]+= v[d2]; v[c3]+= v[d3]; \
v[ b] = ROTR32(v[ b] ^ v[ c], 12); v[b1] = ROTR32(v[b1] ^ v[c1], 12); \ v[ b] = ROTR32(v[ b] ^ v[ c], 12); v[b1] = ROTR32(v[b1] ^ v[c1], 12); v[b2] = ROTR32(v[b2] ^ v[c2], 12); v[b3] = ROTR32(v[b3] ^ v[c3], 12); \
v[ a]+= c_xors[i++] + v[ b]; v[a1]+= (c_xors[i++]^nonce) + v[b1]; \ v[ a]+= y + v[ b]; v[a1]+= y1 + v[b1]; v[a2]+= y2 + v[b2]; v[a3]+= y3 + v[b3]; \
v[ d] = __byte_perm(v[ d] ^ v[ a], 0, 0x0321); v[d1] = __byte_perm(v[d1] ^ v[a1], 0, 0x0321); \ v[ d] = ROR8(v[ d] ^ v[ a]); v[d1] = ROR8(v[d1] ^ v[a1]); v[d2] = ROR8(v[d2] ^ v[a2]); v[d3] = ROR8(v[d3] ^ v[a3]); \
v[ c]+= v[ d]; v[c1]+= v[d1]; \ v[ c]+= v[ d]; v[c1]+= v[d1]; v[c2]+= v[d2]; v[c3]+= v[d3]; \
v[ b] = ROTR32(v[ b] ^ v[ c], 7); v[b1] = ROTR32(v[b1] ^ v[c1], 7); \ v[ b] = ROTR32(v[ b] ^ v[ c], 7); v[b1] = ROTR32(v[b1] ^ v[c1], 7); v[b2] = ROTR32(v[b2] ^ v[c2], 7); v[b3] = ROTR32(v[b3] ^ v[c3], 7); \
}
#define pxory0GS2(a,b,c,d,a1,b1,c1,d1) { \
v[ a]+= c_xors[i++] + v[ b]; v[a1]+= c_xors[i++] + v[b1]; \
v[ d] = __byte_perm(v[ d] ^ v[ a], 0, 0x1032); v[d1] = __byte_perm(v[d1] ^ v[a1], 0, 0x1032); \
v[ c]+= v[ d]; v[c1]+= v[d1]; \
v[ b] = ROTR32(v[ b] ^ v[ c], 12); v[b1] = ROTR32(v[b1] ^ v[c1], 12); \
v[ a]+= (c_xors[i++]^nonce) + v[ b]; v[a1]+= c_xors[i++] + v[b1]; \
v[ d] = __byte_perm(v[ d] ^ v[ a], 0, 0x0321); v[d1] = __byte_perm(v[d1] ^ v[a1], 0, 0x0321); \
v[ c]+= v[ d]; v[c1]+= v[d1]; \
v[ b] = ROTR32(v[ b] ^ v[ c], 7); v[b1] = ROTR32(v[b1] ^ v[c1], 7); \
}
#define pxorx1GS2(a,b,c,d,a1,b1,c1,d1) { \
v[ a]+= c_xors[i++] + v[ b]; v[a1]+= (c_xors[i++]^nonce) + v[b1]; \
v[ d] = __byte_perm(v[ d] ^ v[ a], 0, 0x1032); v[d1] = __byte_perm(v[d1] ^ v[a1], 0, 0x1032); \
v[ c]+= v[ d]; v[c1]+= v[d1]; \
v[ b] = ROTR32(v[ b] ^ v[ c], 12); v[b1] = ROTR32(v[b1] ^ v[c1], 12); \
v[ a]+= c_xors[i++] + v[ b]; v[a1]+= c_xors[i++] + v[b1]; \
v[ d] = __byte_perm(v[ d] ^ v[ a], 0, 0x0321); v[d1] = __byte_perm(v[d1] ^ v[a1], 0, 0x0321); \
v[ c]+= v[ d]; v[c1]+= v[d1]; \
v[ b] = ROTR32(v[ b] ^ v[ c], 7); v[b1] = ROTR32(v[b1] ^ v[c1], 7); \
}
#define pxorx0GS2(a,b,c,d,a1,b1,c1,d1) { \
v[ a]+= (c_xors[i++]^nonce) + v[ b]; v[a1]+= c_xors[i++] + v[b1]; \
v[ d] = __byte_perm(v[ d] ^ v[ a], 0, 0x1032); v[d1] = __byte_perm(v[d1] ^ v[a1], 0, 0x1032); \
v[ c]+= v[ d]; v[c1]+= v[d1]; \
v[ b] = ROTR32(v[ b] ^ v[ c], 12); v[b1] = ROTR32(v[b1] ^ v[c1], 12); \
v[ a]+= c_xors[i++] + v[ b]; v[a1]+= c_xors[i++] + v[b1]; \
v[ d] = __byte_perm(v[ d] ^ v[ a], 0, 0x0321); v[d1] = __byte_perm(v[d1] ^ v[a1], 0, 0x0321); \
v[ c]+= v[ d]; v[c1]+= v[d1]; \
v[ b] = ROTR32(v[ b] ^ v[ c], 7); v[b1] = ROTR32(v[b1] ^ v[c1], 7); \
} }
__global__ __launch_bounds__(TPB,1) __global__ __launch_bounds__(TPB,1)
void decred_gpu_hash_nonce(const uint32_t threads, const uint32_t startNonce, uint32_t *resNonce, const uint32_t highTarget) void decred_gpu_hash_nonce(const uint32_t threads, const uint32_t startNonce, uint32_t *resNonce)
{ {
const uint32_t thread = blockDim.x * blockIdx.x + threadIdx.x; uint64_t m3 = startNonce + blockDim.x * blockIdx.x + threadIdx.x;
const uint32_t step = gridDim.x * blockDim.x;
const uint64_t maxNonce = startNonce + threads;
const uint32_t z[16] = {
0x243F6A88, 0x85A308D3, 0x13198A2E, 0x03707344,
0xA4093822, 0x299F31D0, 0x082EFA98, 0xEC4E6C89,
0x452821E6, 0x38D01377, 0xBE5466CF, 0x34E90C6C,
0xC0AC29B7, 0xC97C50DD, 0x3F84D5B5, 0xB5470917
};
if (thread < threads)
{
uint32_t v[16]; uint32_t v[16];
uint32_t m[16];
#pragma unroll #pragma unroll
for(int i=0;i<16;i+=4){ for(int i=0;i<3;i++) {
*(uint4*)&v[i] = *(uint4*)&c_data[ i]; m[i] = c_m[i];
}
m[13] = 0x80000001;
m[15] = 0x000005a0;
const uint32_t m130 = z[12] ^ m[13];
const uint32_t m131 = m[13] ^ z[ 6];
const uint32_t m132 = z[15] ^ m[13];
const uint32_t m133 = z[ 3] ^ m[13];
const uint32_t m134 = z[ 4] ^ m[13];
const uint32_t m135 = z[14] ^ m[13];
const uint32_t m136 = m[13] ^ z[11];
const uint32_t m137 = m[13] ^ z[ 7];
const uint32_t m138 = m[13] ^ z[ 0];
volatile uint32_t m150 = z[14] ^ m[15];
volatile uint32_t m151 = z[ 9] ^ m[15];
volatile uint32_t m152 = m[15] ^ z[13];
volatile uint32_t m153 = m[15] ^ z[ 8];
const uint32_t m154 = z[10] ^ m[15];
const uint32_t m155 = z[ 1] ^ m[15];
const uint32_t m156 = m[15] ^ z[ 4];
const uint32_t m157 = z[ 6] ^ m[15];
const uint32_t m158 = m[15] ^ z[11];
const uint32_t h7 = c_h[ 0];
for( ; m3<maxNonce ; m3+=step) {
m[ 3] = m3;
#pragma unroll 16
for(int i=0; i<16; i++) {
v[i] = c_v[i];
} }
const uint32_t nonce = startNonce + thread; uint32_t xors[16];
int i=0; uint32_t i = 0;
v[ 1]+= (nonce ^ 0x13198A2E);
v[13] = __byte_perm(v[13] ^ v[1], 0, 0x0321); // round 1 { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 } partial
v[ 9]+= v[13]; xors[ 5] = z[2] ^ m[3];
v[ 5] = ROTR32(v[5] ^ v[9], 7); xors[ 9] = c_x[i++]; xors[10] = c_x[i++];
xors[11] = z[15];
v[ 1]+= c_xors[i++];// + v[ 6]; xors[12] = c_x[i++]; xors[13] = c_x[i++];
v[ 0]+= v[5]; xors[14] = m130;
v[12] = __byte_perm(v[12] ^ v[ 1], 0, 0x1032); v[15] = __byte_perm(v[15] ^ v[ 0], 0, 0x1032); xors[15] = m150;
v[11]+= v[12]; v[10]+= v[15];
v[ 6] = ROTR32(v[ 6] ^ v[11], 12); v[ 5] = ROTR32(v[5] ^ v[10], 12); v[ 1] += xors[ 5]; v[13] = ROR8(v[13] ^ v[1]);
v[ 1]+= c_xors[i++] + v[ 6]; v[ 0]+= c_xors[i++] + v[ 5]; v[ 9] += v[13]; v[ 5] = ROTR32(v[5] ^ v[9], 7);
v[12] = __byte_perm(v[12] ^ v[ 1], 0, 0x0321); v[15] = __byte_perm(v[15] ^ v[ 0], 0, 0x0321); v[ 0] += v[5]; v[15] = ROL16(v[15] ^ v[0]);
v[11]+= v[12]; v[10]+= v[15]; v[10] += v[15]; v[ 5] = ROTR32(v[5] ^ v[10], 12);
v[ 6] = ROTR32(v[ 6] ^ v[11], 7); v[ 5] = ROTR32(v[ 5] ^ v[10], 7); v[ 0] += xors[12] + v[5]; v[15] = ROR8(v[15] ^ v[0]);
v[10] += v[15]; v[ 5] = ROTR32(v[5] ^ v[10], 7);
pxorGS2( 2, 7, 8,13, 3, 4, 9,14);
// { 14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3 }, GSn3(1, 6,11,12, xors[ 9], xors[13], 2, 7, 8,13, xors[10], xors[14], 3, 4, 9,14, xors[11], xors[15]);
pxorGS2( 0, 4, 8,12, 1, 5, 9,13);pxorGS2( 2, 6,10,14, 3, 7,11,15);pxorGS2( 0, 5,10,15, 1, 6,11,12);pxory1GS2( 2, 7, 8,13, 3, 4, 9,14);
// { 11, 8, 12, 0, 5, 2, 15, 13, 10, 14, 3, 6, 7, 1, 9, 4 }, // round 2 { 14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3 }
pxorGS2( 0, 4, 8,12, 1, 5, 9,13);pxorGS2( 2, 6,10,14, 3, 7,11,15);pxorx1GS2( 0, 5,10,15, 1, 6,11,12);pxorGS2( 2, 7, 8,13, 3, 4, 9,14); xors[ 0] = z[10]; xors[ 1] = c_x[i++]; xors[ 2] = c_x[i++]; xors[ 3] = m131;
// { 7, 9, 3, 1, 13, 12, 11, 14, 2, 6, 5, 10, 4, 0, 15, 8 }, xors[ 8] = m[ 1]^z[12]; xors[ 9] = m[ 0]^z[ 2]; xors[10] = c_x[i++]; xors[11] = c_x[i++];
pxorx1GS2( 0, 4, 8,12, 1, 5, 9,13);pxorGS2( 2, 6,10,14, 3, 7,11,15);pxorGS2( 0, 5,10,15, 1, 6,11,12);pxorGS2( 2, 7, 8,13, 3, 4, 9,14); xors[ 4] = c_x[i++]; xors[ 5] = c_x[i++]; xors[ 6] = m151; xors[ 7] = c_x[i++];
// { 9, 0, 5, 7, 2, 4, 10, 15, 14, 1, 11, 12, 6, 8, 3, 13 }, xors[12] = c_x[i++]; xors[13] = z[ 0]^m[ 2]; xors[14] = c_x[i++]; xors[15] = z[ 5]^m[ 3];
pxorGS2( 0, 4, 8,12, 1, 5, 9,13);pxorGS2( 2, 6,10,14, 3, 7,11,15);pxorGS2( 0, 5,10,15, 1, 6,11,12);pxorx1GS2( 2, 7, 8,13, 3, 4, 9,14);
// { 2, 12, 6, 10, 0, 11, 8, 3, 4, 13, 7, 5, 15, 14, 1, 9 }, GSn4(0, 4, 8,12, xors[ 0], xors[ 4], 1, 5, 9,13, xors[ 1], xors[ 5], 2, 6,10,14, xors[ 2], xors[ 6], 3, 7,11,15, xors[ 3], xors[ 7]);
pxorGS2( 0, 4, 8,12, 1, 5, 9,13);pxory1GS2( 2, 6,10,14, 3, 7,11,15);pxorGS2( 0, 5,10,15, 1, 6,11,12);pxorGS2( 2, 7, 8,13, 3, 4, 9,14); GSn4(0, 5,10,15, xors[ 8], xors[12], 1, 6,11,12, xors[ 9], xors[13], 2, 7, 8,13, xors[10], xors[14], 3, 4, 9,14, xors[11], xors[15]);
// { 12, 5, 1, 15, 14, 13, 4, 10, 0, 7, 6, 3, 9, 2, 8, 11 },
pxorGS2( 0, 4, 8,12, 1, 5, 9,13);pxorGS2( 2, 6,10,14, 3, 7,11,15);pxory1GS2( 0, 5,10,15, 1, 6,11,12);pxorGS2( 2, 7, 8,13, 3, 4, 9,14); // round 3 { 11, 8, 12, 0, 5, 2, 15, 13, 10, 14, 3, 6, 7, 1, 9, 4 }
// { 13, 11, 7, 14, 12, 1, 3, 9, 5, 0, 15, 4, 8, 6, 2, 10 }, xors[ 0] = c_x[i++]; xors[ 1] = c_x[i++]; xors[ 2] = c_x[i++]; xors[ 3] = m152;
pxorGS2( 0, 4, 8,12, 1, 5, 9,13);pxorx1GS2( 2, 6,10,14, 3, 7,11,15);pxorGS2( 0, 5,10,15, 1, 6,11,12);pxorGS2( 2, 7, 8,13, 3, 4, 9,14); xors[ 8] = c_x[i++]; xors[ 9] = m[ 3]^z[ 6]; xors[10] = c_x[i++]; xors[11] = c_x[i++];
//{ 6, 15, 14, 9, 11, 3, 0, 8, 12, 2, 13, 7, 1, 4, 10, 5 }, xors[ 4] = c_x[i++]; xors[ 5] = z[12]^m[ 0]; xors[ 6] = z[ 5]^m[ 2]; xors[ 7] = m132;
pxorGS2( 0, 4, 8,12, 1, 5, 9,13);pxory0GS2( 2, 6,10,14, 3, 7,11,15);pxorGS2( 0, 5,10,15, 1, 6,11,12);pxorGS2( 2, 7, 8,13, 3, 4, 9,14); xors[12] = z[10]; xors[13] = c_x[i++]; xors[14] = z[ 7]^m[ 1]; xors[15] = c_x[i++];
//{ 10, 2, 8, 4, 7, 6, 1, 5, 15, 11, 9, 14, 3, 12, 13 , 0 },
pxorGS2( 0, 4, 8,12, 1, 5, 9,13);pxorGS2( 2, 6,10,14, 3, 7,11,15);pxorGS2( 0, 5,10,15, 1, 6,11,12);pxorx0GS2( 2, 7, 8,13, 3, 4, 9,14); GSn4(0, 4, 8,12, xors[ 0], xors[ 4], 1, 5, 9,13, xors[ 1], xors[ 5], 2, 6,10,14, xors[ 2], xors[ 6], 3, 7,11,15, xors[ 3], xors[ 7]);
//{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 }, GSn4(0, 5,10,15, xors[ 8], xors[12], 1, 6,11,12, xors[ 9], xors[13], 2, 7, 8,13, xors[10], xors[14], 3, 4, 9,14, xors[11], xors[15]);
pxory1GS2( 0, 4, 8,12, 1, 5, 9,13);pxorGS2( 2, 6,10,14, 3, 7,11,15);pxorGS2( 0, 5,10,15, 1, 6,11,12);pxorGS2( 2, 7, 8,13, 3, 4, 9,14);
// { 14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3 }, // round 4 { 7, 9, 3, 1, 13, 12, 11, 14, 2, 6, 5, 10, 4, 0, 15, 8 }
pxorGS2( 0, 4, 8,12, 1, 5, 9,13);pxorGS2( 2, 6,10,14, 3, 7,11,15);pxorGS2( 0, 5,10,15, 1, 6,11,12);pxory1GS2( 2, 7, 8,13, 3, 4, 9,14); xors[ 0] = c_x[i++]; xors[ 1] = m[ 3]^z[ 1]; xors[ 2] = m130; xors[ 3] = c_x[i++];
// { 11, 8, 12, 0, 5, 2, 15, 13, 10, 14, 3, 6, 7, 1, 9, 4 }, xors[ 8] = m[ 2]^z[ 6]; xors[ 9] = c_x[i++]; xors[10] = c_x[i++]; xors[11] = m153;
pxorGS2( 0, 4, 8,12, 1, 5, 9,13);pxorGS2( 2, 6,10,14, 3, 7,11,15);pxorx1GS2( 0, 5,10,15, 1, 6,11,12);pxorGS2( 2, 7, 8,13, 3, 4, 9,14); xors[ 4] = c_x[i++]; xors[ 5] = z[ 3]^m[ 1]; xors[ 6] = c_x[i++]; xors[ 7] = z[11];
//{ 7, 9, 3, 1, 13, 12, 11, 14, 2, 6, 5, 10, 4, 0, 15, 8 } xors[12] = c_x[i++]; xors[13] = c_x[i++]; xors[14] = z[ 4]^m[ 0]; xors[15] = c_x[i++];
pxorx1GS2( 0, 4, 8,12, 1, 5, 9,13);pxorGS2( 2, 6,10,14, 3, 7,11,15);pxorGS2( 0, 5,10,15, 1, 6,11,12);pxorGS( 2, 7, 8,13);
GSn4(0, 4, 8,12, xors[ 0], xors[ 4], 1, 5, 9,13, xors[ 1], xors[ 5], 2, 6,10,14, xors[ 2], xors[ 6], 3, 7,11,15, xors[ 3], xors[ 7]);
if ((c_h[1]^v[15]) == v[7]){ GSn4(0, 5,10,15, xors[ 8], xors[12], 1, 6,11,12, xors[ 9], xors[13], 2, 7, 8,13, xors[10], xors[14], 3, 4, 9,14, xors[11], xors[15]);
v[3] += c_xors[i++] + v[4];
v[14] = __byte_perm(v[14] ^ v[3], 0, 0x1032); // round 5 { 9, 0, 5, 7, 2, 4, 10, 15, 14, 1, 11, 12, 6, 8, 3, 13 }
v[9] += v[14]; xors[ 0] = c_x[i++]; xors[ 1] = c_x[i++]; xors[ 2] = m[ 2]^z[ 4]; xors[ 3] = c_x[i++];
v[4] = ROTR32(v[4] ^ v[9], 12); xors[ 8] = z[ 1]; xors[ 9] = c_x[i++]; xors[10] = c_x[i++]; xors[11] = m[ 3]^z[13];
v[3] += c_xors[i++] + v[4]; xors[ 4] = z[ 9]^m[ 0]; xors[ 5] = c_x[i++]; xors[ 6] = c_x[i++]; xors[ 7] = m154;
v[14] = __byte_perm(v[14] ^ v[3], 0, 0x0321); xors[12] = z[14]^m[ 1]; xors[13] = c_x[i++]; xors[14] = c_x[i++]; xors[15] = m133;
if(cuda_swab32((c_h[0]^v[6]^v[14])) <= highTarget) {
atomicMin(&resNonce[0], nonce); GSn4(0, 4, 8,12, xors[ 0], xors[ 4], 1, 5, 9,13, xors[ 1], xors[ 5], 2, 6,10,14, xors[ 2], xors[ 6], 3, 7,11,15, xors[ 3], xors[ 7]);
GSn4(0, 5,10,15, xors[ 8], xors[12], 1, 6,11,12, xors[ 9], xors[13], 2, 7, 8,13, xors[10], xors[14], 3, 4, 9,14, xors[11], xors[15]);
// round 6 { 2, 12, 6, 10, 0, 11, 8, 3, 4, 13, 7, 5, 15, 14, 1, 9 }
xors[ 0] = m[ 2]^z[12]; xors[ 1] = c_x[i++]; xors[ 2] = m[ 0]^z[11]; xors[ 3] = c_x[i++];
xors[ 8] = c_x[i++]; xors[ 9] = c_x[i++]; xors[10] = m150; xors[11] = m[ 1]^z[ 9];
xors[ 4] = c_x[i++]; xors[ 5] = c_x[i++]; xors[ 6] = c_x[i++]; xors[ 7] = z[ 8]^m[ 3];
xors[12] = m134; xors[13] = c_x[i++]; xors[14] = z[15]; xors[15] = c_x[i++];
GSn4(0, 4, 8,12, xors[ 0], xors[ 4], 1, 5, 9,13, xors[ 1], xors[ 5], 2, 6,10,14, xors[ 2], xors[ 6], 3, 7,11,15, xors[ 3], xors[ 7]);
GSn4(0, 5,10,15, xors[ 8], xors[12], 1, 6,11,12, xors[ 9], xors[13], 2, 7, 8,13, xors[10], xors[14], 3, 4, 9,14, xors[11], xors[15]);
// round 7 { 12, 5, 1, 15, 14, 13, 4, 10, 0, 7, 6, 3, 9, 2, 8, 11 }
xors[ 0] = c_x[i++]; xors[ 1] = m[ 1]^z[15]; xors[ 2] = z[13]; xors[ 3] = c_x[i++];
xors[ 8] = m[ 0]^z[ 7]; xors[ 9] = c_x[i++]; xors[10] = c_x[i++]; xors[11] = c_x[i++];
xors[ 4] = c_x[i++]; xors[ 5] = m155; xors[ 6] = m135; xors[ 7] = c_x[i++];
xors[12] = c_x[i++]; xors[13] = z[ 6]^m[ 3]; xors[14] = z[ 9]^m[ 2]; xors[15] = c_x[i++];
GSn4(0, 4, 8,12, xors[ 0], xors[ 4], 1, 5, 9,13, xors[ 1], xors[ 5], 2, 6,10,14, xors[ 2], xors[ 6], 3, 7,11,15, xors[ 3], xors[ 7]);
GSn4(0, 5,10,15, xors[ 8], xors[12], 1, 6,11,12, xors[ 9], xors[13], 2, 7, 8,13, xors[10], xors[14], 3, 4, 9,14, xors[11], xors[15]);
// round 8 { 13, 11, 7, 14, 12, 1, 3, 9, 5, 0, 15, 4, 8, 6, 2, 10 }
xors[ 0] = m136; xors[ 1] = c_x[i++]; xors[ 2] = c_x[i++]; xors[ 3] = m[ 3]^z[ 9];
xors[ 8] = c_x[i++]; xors[ 9] = m156; xors[10] = c_x[i++]; xors[11] = m[ 2]^z[10];
xors[ 4] = c_x[i++]; xors[ 5] = z[ 7]; xors[ 6] = z[12]^m[ 1]; xors[ 7] = c_x[i++];
xors[12] = z[ 5]^m[ 0]; xors[13] = c_x[i++]; xors[14] = c_x[i++]; xors[15] = c_x[i++];
GSn4(0, 4, 8,12, xors[ 0], xors[ 4], 1, 5, 9,13, xors[ 1], xors[ 5], 2, 6,10,14, xors[ 2], xors[ 6], 3, 7,11,15, xors[ 3], xors[ 7]);
GSn4(0, 5,10,15, xors[ 8], xors[12], 1, 6,11,12, xors[ 9], xors[13], 2, 7, 8,13, xors[10], xors[14], 3, 4, 9,14, xors[11], xors[15]);
// round 9 { 6, 15, 14, 9, 11, 3, 0, 8, 12, 2, 13, 7, 1, 4, 10, 5 }
xors[ 0] = c_x[i++]; xors[ 1] = z[ 9]; xors[ 2] = c_x[i++]; xors[ 3] = m[ 0]^z[ 8];
xors[ 8] = c_x[i++]; xors[ 9] = m137; xors[10] = m[ 1]^z[ 4]; xors[11] = c_x[i++];
xors[ 4] = m157; xors[ 5] = c_x[i++]; xors[ 6] = z[11]^m[ 3]; xors[ 7] = c_x[i++];
xors[12] = z[12]^m[ 2]; xors[13] = c_x[i++]; xors[14] = c_x[i++]; xors[15] = c_x[i++];
GSn4(0, 4, 8,12, xors[ 0], xors[ 4], 1, 5, 9,13, xors[ 1], xors[ 5], 2, 6,10,14, xors[ 2], xors[ 6], 3, 7,11,15, xors[ 3], xors[ 7]);
GSn4(0, 5,10,15, xors[ 8], xors[12], 1, 6,11,12, xors[ 9], xors[13], 2, 7, 8,13, xors[10], xors[14], 3, 4, 9,14, xors[11], xors[15]);
// round 10 { 10, 2, 8, 4, 7, 6, 1, 5, 15, 11, 9, 14, 3, 12, 13 , 0 }
xors[ 0] = c_x[i++]; xors[ 1] = c_x[i++]; xors[ 2] = c_x[i++]; xors[ 3] = m[ 1]^z[ 5];
xors[ 8] = m158; xors[ 9] = c_x[i++]; xors[10] = m[ 3]^z[12]; xors[11] = m138;
xors[ 4] = z[10]^m[ 2]; xors[ 5] = c_x[i++]; xors[ 6] = c_x[i++]; xors[ 7] = c_x[i++];
xors[12] = c_x[i++]; xors[13] = z[ 9]; xors[14] = c_x[i++]; xors[15] = z[13]^m[ 0];
GSn4(0, 4, 8,12, xors[ 0], xors[ 4], 1, 5, 9,13, xors[ 1], xors[ 5], 2, 6,10,14, xors[ 2], xors[ 6], 3, 7,11,15, xors[ 3], xors[ 7]);
GSn4(0, 5,10,15, xors[ 8], xors[12], 1, 6,11,12, xors[ 9], xors[13], 2, 7, 8,13, xors[10], xors[14], 3, 4, 9,14, xors[11], xors[15]);
// round 11 { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 }
xors[ 0] = m[ 0]^z[ 1]; xors[ 1] = m[ 2]^z[ 3]; xors[ 2] = c_x[i++]; xors[ 3] = c_x[i++];
xors[ 8] = c_x[i++]; xors[ 9] = c_x[ 0]; xors[10] = c_x[ 1]; xors[11] = z[15];
xors[ 4] = z[ 0]^m[ 1]; xors[ 5] = z[ 2]^m[ 3]; xors[ 6] = c_x[i++]; xors[ 7] = c_x[i++];
xors[12] = c_x[ 2]; xors[13] = c_x[ 3]; xors[14] = m130; xors[15] = m150;
GSn4(0, 4, 8,12, xors[ 0], xors[ 4], 1, 5, 9,13, xors[ 1], xors[ 5], 2, 6,10,14, xors[ 2], xors[ 6], 3, 7,11,15, xors[ 3], xors[ 7]);
GSn4(0, 5,10,15, xors[ 8], xors[12], 1, 6,11,12, xors[ 9], xors[13], 2, 7, 8,13, xors[10], xors[14], 3, 4, 9,14, xors[11], xors[15]);
//i=90
i=4;
// round 12 { 14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3 }
xors[ 0] = z[10]; xors[ 1] = c_x[i++]; xors[ 2] = c_x[i++]; xors[ 3] = m131;
xors[ 8] = m[ 1]^z[12]; xors[ 9] = m[ 0]^z[ 2]; xors[10] = c_x[i++]; xors[11] = c_x[i++];
xors[ 4] = c_x[i++]; xors[ 5] = c_x[i++]; xors[ 6] = m151; xors[ 7] = c_x[i++];
xors[12] = c_x[i++]; xors[13] = z[ 0]^m[ 2]; xors[14] = c_x[i++]; xors[15] = z[ 5]^m[ 3];
GSn4(0, 4, 8,12, xors[ 0], xors[ 4], 1, 5, 9,13, xors[ 1], xors[ 5], 2, 6,10,14, xors[ 2], xors[ 6], 3, 7,11,15, xors[ 3], xors[ 7]);
GSn4(0, 5,10,15, xors[ 8], xors[12], 1, 6,11,12, xors[ 9], xors[13], 2, 7, 8,13, xors[10], xors[14], 3, 4, 9,14, xors[11], xors[15]);
// round 13 { 11, 8, 12, 0, 5, 2, 15, 13, 10, 14, 3, 6, 7, 1, 9, 4 }
xors[ 0] = c_x[i++]; xors[ 1] = c_x[i++]; xors[ 2] = c_x[i++]; xors[ 3] = m152;
xors[ 8] = c_x[i++]; xors[ 9] = m[ 3]^z[ 6]; xors[10] = c_x[i++]; xors[11] = c_x[i++];
xors[ 4] = c_x[i++]; xors[ 5] = z[12]^m[ 0]; xors[ 6] = z[ 5]^m[ 2]; xors[ 7] = m132;
xors[12] = z[10]; xors[13] = c_x[i++]; xors[14] = z[ 7]^m[ 1]; xors[15] = c_x[i++];
GSn4(0, 4, 8,12, xors[ 0], xors[ 4], 1, 5, 9,13, xors[ 1], xors[ 5], 2, 6,10,14, xors[ 2], xors[ 6], 3, 7,11,15, xors[ 3], xors[ 7]);
GSn4(0, 5,10,15, xors[ 8], xors[12], 1, 6,11,12, xors[ 9], xors[13], 2, 7, 8,13, xors[10], xors[14], 3, 4, 9,14, xors[11], xors[15]);
// round 14 { 7, 9, 3, 1, 13, 12, 11, 14, 2, 6, 5, 10, 4, 0, 15, 8 }
xors[ 0] = c_x[i++]; xors[ 1] = m[ 3]^z[ 1]; xors[ 2] = m130; xors[ 3] = c_x[i++];
xors[ 8] = m[ 2]^z[ 6]; i++; xors[10] = c_x[i++];
xors[ 4] = c_x[i++]; xors[ 5] = z[ 3]^m[ 1]; xors[ 6] = c_x[i++]; xors[ 7] = z[11];
xors[12] = c_x[i++]; xors[14] = z[ 4]^m[ 0];
GSn4(0, 4, 8,12, xors[ 0], xors[ 4], 1, 5, 9,13, xors[ 1], xors[ 5], 2, 6,10,14, xors[ 2], xors[ 6], 3, 7,11,15, xors[ 3], xors[ 7]);
v[ 0]+= xors[ 8] + v[ 5];
v[ 2]+= xors[10] + v[ 7];
v[15] = ROL16(v[15] ^ v[ 0]);
v[13] = ROL16(v[13] ^ v[ 2]);
v[10]+= v[15];
v[ 8]+= v[13];
v[ 5] = ROTR32(v[ 5] ^ v[10], 12);
v[ 7] = ROTR32(v[ 7] ^ v[ 8], 12);
v[ 0]+= xors[12] + v[ 5];
v[ 2]+= xors[14] + v[ 7];
v[15] = ROTR32(v[15] ^ v[ 0],1);
v[13] = ROR8(v[13] ^ v[ 2]);
v[ 8]+= v[13];
if(xor3x(v[ 7],h7,v[ 8])==v[15]) {
uint32_t pos = atomicInc(&resNonce[0], UINT32_MAX)+1;
if(pos < maxResults)
resNonce[pos] = m[3];
return; return;
} }
} }
}
} }
__host__ __host__
void decred_cpu_setBlock_52(const uint32_t *input){ void decred_cpu_setBlock_52(const int thr_id,const uint32_t *input, const uint32_t *pend)
/* {
Precompute everything possible and pass it on constant memory const uint32_t z[16] = {
*/ 0x243F6A88UL, 0x85A308D3UL, 0x13198A2EUL, 0x03707344UL,
const sph_u32 _ALIGN(64) z[16] = { 0xA4093822UL, 0x299F31D0UL, 0x082EFA98UL, 0xEC4E6C89UL,
SPH_C32(0x243F6A88), SPH_C32(0x85A308D3), SPH_C32(0x13198A2E), SPH_C32(0x03707344), SPH_C32(0xA4093822), SPH_C32(0x299F31D0), SPH_C32(0x082EFA98), SPH_C32(0xEC4E6C89), 0x452821E6UL, 0x38D01377UL, 0xBE5466CFUL, 0x34E90C6CUL,
SPH_C32(0x452821E6), SPH_C32(0x38D01377), SPH_C32(0xBE5466CF), SPH_C32(0x34E90C6C), SPH_C32(0xC0AC29B7), SPH_C32(0xC97C50DD), SPH_C32(0x3F84D5B5), SPH_C32(0xB5470917) 0xC0AC29B7UL, 0xC97C50DDUL, 0x3F84D5B5UL, 0xB5470917UL
}; };
int i=0; sph_u32 _ALIGN(64) v[16];
sph_u32 _ALIGN(64) preXOR[215];
sph_u32 _ALIGN(64) data[16];
sph_u32 _ALIGN(64) m[16]; sph_u32 _ALIGN(64) m[16];
sph_u32 _ALIGN(64) h[ 2]; sph_u32 _ALIGN(64) h[ 2];
@ -200,11 +339,12 @@ void decred_cpu_setBlock_52(const uint32_t *input){
sph_blake256_init(&ctx); sph_blake256_init(&ctx);
sph_blake256(&ctx, input, 128); sph_blake256(&ctx, input, 128);
data[ 0] = ctx.H[0]; data[ 1] = ctx.H[1]; v[ 0] = ctx.H[0]; v[ 1] = ctx.H[1];
data[ 2] = ctx.H[2]; data[ 3] = ctx.H[3]; v[ 2] = ctx.H[2]; v[ 3] = ctx.H[3];
data[ 4] = ctx.H[4]; data[ 5] = ctx.H[5]; v[ 4] = ctx.H[4]; v[ 5] = ctx.H[5];
data[ 8] = ctx.H[6]; data[12] = swab32(input[35]); v[ 8] = ctx.H[6]; v[12] = swab32(input[35]);
data[13] = ctx.H[7]; v[13] = ctx.H[7];
// pre swab32 // pre swab32
m[ 0] = swab32(input[32]); m[ 1] = swab32(input[33]); m[ 0] = swab32(input[32]); m[ 1] = swab32(input[33]);
m[ 2] = swab32(input[34]); m[ 3] = 0; m[ 2] = swab32(input[34]); m[ 3] = 0;
@ -213,124 +353,71 @@ void decred_cpu_setBlock_52(const uint32_t *input){
m[ 8] = swab32(input[40]); m[ 9] = swab32(input[41]); m[ 8] = swab32(input[40]); m[ 9] = swab32(input[41]);
m[10] = swab32(input[42]); m[11] = swab32(input[43]); m[10] = swab32(input[42]); m[11] = swab32(input[43]);
m[12] = swab32(input[44]); m[13] = 0x80000001; m[12] = swab32(input[44]); m[13] = 0x80000001;
m[14] = 0; m[15] = 0x000005a0; m[14] = 0;
m[15] = 0x000005a0;
h[ 0] = data[ 8];
h[ 1] = data[13]; CUDA_SAFE_CALL(cudaMemcpyToSymbol(c_m, m, 3*sizeof(uint32_t), 0, cudaMemcpyHostToDevice));
CUDA_SAFE_CALL(cudaMemcpyToSymbol(c_h,h, 8, 0, cudaMemcpyHostToDevice)); h[ 0] = v[ 8];
h[ 1] = v[13];
data[ 0]+= (m[ 0] ^ z[1]) + data[ 4];
data[12] = SPH_ROTR32(z[4] ^ SPH_C32(0x5A0) ^ data[ 0], 16); v[ 0]+= (m[ 0] ^ z[1]) + v[ 4];
v[12] = SPH_ROTR32(z[4] ^ SPH_C32(0x5A0) ^ v[ 0], 16);
data[ 8] = z[0]+data[12];
data[ 4] = SPH_ROTR32(data[ 4] ^ data[ 8], 12); v[ 8] = z[0]+v[12];
data[ 0]+= (m[ 1] ^ z[0]) + data[ 4]; v[ 4] = SPH_ROTR32(v[ 4] ^ v[ 8], 12);
data[12] = SPH_ROTR32(data[12] ^ data[ 0],8); v[ 0]+= (m[ 1] ^ z[0]) + v[ 4];
data[ 8]+= data[12]; v[12] = SPH_ROTR32(v[12] ^ v[ 0],8);
data[ 4] = SPH_ROTR32(data[ 4] ^ data[ 8], 7); v[ 8]+= v[12];
v[ 4] = SPH_ROTR32(v[ 4] ^ v[ 8], 7);
data[ 1]+= (m[ 2] ^ z[3]) + data[ 5];
data[13] = SPH_ROTR32((z[5] ^ SPH_C32(0x5A0)) ^ data[ 1], 16); v[ 1]+= (m[ 2] ^ z[3]) + v[ 5];
data[ 9] = z[1]+data[13]; v[13] = SPH_ROTR32((z[5] ^ SPH_C32(0x5A0)) ^ v[ 1], 16);
data[ 5] = SPH_ROTR32(data[ 5] ^ data[ 9], 12); v[ 9] = z[1]+v[13];
data[ 1]+= data[ 5]; //+nonce ^ ... v[ 5] = SPH_ROTR32(v[ 5] ^ v[ 9], 12);
v[ 1]+= v[ 5]; //+nonce ^ ...
data[ 2]+= (m[ 4] ^ z[5]) + h[ 0];
data[14] = SPH_ROTR32(z[6] ^ data[ 2],16); v[ 2]+= (m[ 4] ^ z[5]) + h[ 0];
data[10] = z[2] + data[14]; v[14] = SPH_ROTR32(z[6] ^ v[ 2],16);
data[ 6] = SPH_ROTR32(h[ 0] ^ data[10], 12); v[10] = z[2] + v[14];
data[ 2]+= (m[ 5] ^ z[4]) + data[ 6]; v[ 6] = SPH_ROTR32(h[ 0] ^ v[10], 12);
data[14] = SPH_ROTR32(data[14] ^ data[ 2], 8); v[ 2]+= (m[ 5] ^ z[4]) + v[ 6];
data[10]+= data[14]; v[14] = SPH_ROTR32(v[14] ^ v[ 2], 8);
data[ 6] = SPH_ROTR32(data[ 6] ^ data[10], 7); v[10]+= v[14];
v[ 6] = SPH_ROTR32(v[ 6] ^ v[10], 7);
data[ 3]+= (m[ 6] ^ z[7]) + h[ 1];
data[15] = SPH_ROTR32(z[7] ^ data[ 3],16); v[ 3]+= (m[ 6] ^ z[7]) + h[ 1];
data[11] = z[3] + data[15]; v[15] = SPH_ROTR32(z[7] ^ v[ 3],16);
data[ 7] = SPH_ROTR32(h[ 1] ^ data[11], 12); v[11] = z[3] + v[15];
data[ 3]+= (m[ 7] ^ z[6]) + data[ 7]; v[ 7] = SPH_ROTR32(h[ 1] ^ v[11], 12);
data[15] = SPH_ROTR32(data[15] ^ data[ 3],8); v[ 3]+= (m[ 7] ^ z[6]) + v[ 7];
data[11]+= data[15]; v[15] = SPH_ROTR32(v[15] ^ v[ 3],8);
data[ 7] = SPH_ROTR32(data[11] ^ data[ 7], 7); v[11]+= v[15];
data[ 0]+= m[ 8] ^ z[9]; v[ 7] = SPH_ROTR32(v[11] ^ v[ 7], 7);
v[ 0]+= m[ 8] ^ z[9];
CUDA_SAFE_CALL(cudaMemcpyToSymbol(c_data, data, 64, 0, cudaMemcpyHostToDevice));
CUDA_SAFE_CALL(cudaMemcpyToSymbol(c_v, v,16*sizeof(uint32_t), 0, cudaMemcpyHostToDevice));
#define precalcXORGS(x,y) { \
preXOR[i++]= (m[x] ^ z[y]); \ h[ 0] = SPH_ROTL32(h[ 1], 7); //align the rotation with v[7] v[15];
preXOR[i++]= (m[y] ^ z[x]); \ CUDA_SAFE_CALL(cudaMemcpyToSymbol(c_h, h, 1*sizeof(uint32_t), 0, cudaMemcpyHostToDevice));
}
#define precalcXORGS2(x,y,x1,y1){\ uint32_t x[90];
preXOR[i++] = (m[ x] ^ z[ y]);\ int i=0;
preXOR[i++] = (m[x1] ^ z[y1]);\
preXOR[i++] = (m[ y] ^ z[ x]);\ x[i++] = m[10]^z[11]; x[i++] = m[12]^z[13]; x[i++] = m[ 9]^z[ 8]; x[i++] = z[10]^m[11]; x[i++] = m[ 4]^z[ 8]; x[i++] = m[ 9]^z[15]; x[i++] = m[11]^z[ 7]; x[i++] = m[ 5]^z[ 3];
preXOR[i++] = (m[y1] ^ z[x1]);\ x[i++] = z[14]^m[10]; x[i++] = z[ 4]^m[ 8]; x[i++] = z[13]^m[ 6]; x[i++] = z[ 1]^m[12]; x[i++] = z[11]^m[ 7]; x[i++] = m[11]^z[ 8]; x[i++] = m[12]^z[ 0]; x[i++] = m[ 5]^z[ 2];
} x[i++] = m[10]^z[14]; x[i++] = m[ 7]^z[ 1]; x[i++] = m[ 9]^z[ 4]; x[i++] = z[11]^m[ 8]; x[i++] = z[ 3]^m[ 6]; x[i++] = z[ 9]^m[ 4]; x[i++] = m[ 7]^z[ 9]; x[i++] = m[11]^z[14];
precalcXORGS(10,11); x[i++] = m[ 5]^z[10]; x[i++] = m[ 4]^z[ 0]; x[i++] = z[ 7]^m[ 9]; x[i++] = z[13]^m[12]; x[i++] = z[ 2]^m[ 6]; x[i++] = z[ 5]^m[10]; x[i++] = z[15]^m[ 8]; x[i++] = m[ 9]^z[ 0];
preXOR[ 0]+=data[ 6]; x[i++] = m[ 5]^z[ 7]; x[i++] = m[10]^z[15]; x[i++] = m[11]^z[12]; x[i++] = m[ 6]^z[ 8]; x[i++] = z[ 5]^m[ 7]; x[i++] = z[ 2]^m[ 4]; x[i++] = z[11]^m[12]; x[i++] = z[ 6]^m[ 8];
preXOR[i++] = (m[9] ^ z[8]); x[i++] = m[ 6]^z[10]; x[i++] = m[ 8]^z[ 3]; x[i++] = m[ 4]^z[13]; x[i++] = m[ 7]^z[ 5]; x[i++] = z[ 2]^m[12]; x[i++] = z[ 6]^m[10]; x[i++] = z[ 0]^m[11]; x[i++] = z[ 7]^m[ 5];
precalcXORGS2(12,13,14,15); x[i++] = z[ 1]^m[ 9]; x[i++] = m[12]^z[ 5]; x[i++] = m[ 4]^z[10]; x[i++] = m[ 6]^z[ 3]; x[i++] = m[ 9]^z[ 2]; x[i++] = m[ 8]^z[11]; x[i++] = z[12]^m[ 5]; x[i++] = z[ 4]^m[10];
precalcXORGS2(14,10, 4, 8); x[i++] = z[ 0]^m[ 7]; x[i++] = z[ 8]^m[11]; x[i++] = m[ 7]^z[14]; x[i++] = m[12]^z[ 1]; x[i++] = m[ 5]^z[ 0]; x[i++] = m[ 8]^z[ 6]; x[i++] = z[13]^m[11]; x[i++] = z[ 3]^m[ 9];
precalcXORGS2( 9,15,13, 6); x[i++] = z[15]^m[ 4]; x[i++] = z[ 8]^m[ 6]; x[i++] = z[ 2]^m[10]; x[i++] = m[ 6]^z[15]; x[i++] = m[11]^z[ 3]; x[i++] = m[12]^z[ 2]; x[i++] = m[10]^z[ 5]; x[i++] = z[14]^m[ 9];
precalcXORGS2( 1,12, 0, 2); x[i++] = z[ 0]^m[ 8]; x[i++] = z[13]^m[ 7]; x[i++] = z[ 1]^m[ 4]; x[i++] = z[10]^m[ 5]; x[i++] = m[10]^z[ 2]; x[i++] = m[ 8]^z[ 4]; x[i++] = m[ 7]^z[ 6]; x[i++] = m[ 9]^z[14];
precalcXORGS2(11, 7, 5, 3); x[i++] = z[ 8]^m[ 4]; x[i++] = z[ 7]^m[ 6]; x[i++] = z[ 1]^m[ 5]; x[i++] = z[15]^m[11]; x[i++] = z[ 3]^m[12]; x[i++] = m[ 4]^z[ 5]; x[i++] = m[ 6]^z[ 7]; x[i++] = m[ 8]^z[ 9];
x[i++] = z[ 4]^m[ 5]; x[i++] = z[ 6]^m[ 7];
precalcXORGS2(11, 8,12, 0);
precalcXORGS2( 5, 2,15,13); CUDA_SAFE_CALL(cudaMemcpyToSymbol(c_x, x, i*sizeof(uint32_t), 0, cudaMemcpyHostToDevice));
precalcXORGS2(10,14, 3, 6);
precalcXORGS2( 7, 1, 9, 4);
precalcXORGS2( 7, 9, 3, 1);
precalcXORGS2(13,12,11,14);
precalcXORGS2( 2, 6, 5,10);
precalcXORGS2( 4, 0,15, 8);
precalcXORGS2( 9, 0, 5, 7);
precalcXORGS2( 2, 4,10,15);
precalcXORGS2(14, 1,11,12);
precalcXORGS2( 6, 8, 3,13);
precalcXORGS2( 2,12, 6,10);
precalcXORGS2( 0,11, 8, 3);
precalcXORGS2( 4,13, 7, 5);
precalcXORGS2(15,14, 1, 9);
precalcXORGS2(12, 5, 1,15);
precalcXORGS2(14,13, 4,10);
precalcXORGS2( 0, 7, 6, 3);
precalcXORGS2( 9, 2, 8,11);
precalcXORGS2(13,11, 7,14);
precalcXORGS2(12, 1, 3, 9);
precalcXORGS2( 5, 0,15, 4);
precalcXORGS2( 8, 6, 2,10);
precalcXORGS2( 6,15,14, 9);
precalcXORGS2(11, 3, 0, 8);
precalcXORGS2(12, 2,13, 7);
precalcXORGS2( 1, 4,10, 5);
precalcXORGS2(10, 2, 8, 4);
precalcXORGS2( 7, 6, 1, 5);
precalcXORGS2(15,11, 9,14);
precalcXORGS2( 3,12,13, 0);
precalcXORGS2( 0, 1, 2, 3);
precalcXORGS2( 4, 5, 6, 7);
precalcXORGS2( 8, 9,10,11);
precalcXORGS2(12,13,14,15);
precalcXORGS2(14,10, 4, 8);
precalcXORGS2( 9,15,13, 6);
precalcXORGS2( 1,12, 0, 2);
precalcXORGS2(11, 7, 5, 3);
precalcXORGS2(11, 8,12, 0);
precalcXORGS2( 5, 2,15,13);
precalcXORGS2(10,14, 3, 6);
precalcXORGS2( 7, 1, 9, 4);
precalcXORGS2( 7, 9, 3, 1);
precalcXORGS2(13,12,11,14);
precalcXORGS2( 2, 6, 5,10);
precalcXORGS( 4, 0);
precalcXORGS(15, 8);
CUDA_SAFE_CALL(cudaMemcpyToSymbol(c_xors, preXOR, 215*sizeof(uint32_t), 0, cudaMemcpyHostToDevice));
} }
/* ############################################################################################################################### */ /* ############################################################################################################################### */
@ -349,7 +436,6 @@ extern "C" int scanhash_decred(int thr_id, struct work* work, uint32_t max_nonce
uint32_t *pnonce = &pdata[DCR_NONCE_OFT32]; uint32_t *pnonce = &pdata[DCR_NONCE_OFT32];
const uint32_t first_nonce = *pnonce; const uint32_t first_nonce = *pnonce;
const uint32_t targetHigh = (opt_benchmark?0x1ULL:ptarget[6]);
const int dev_id = device_map[thr_id]; const int dev_id = device_map[thr_id];
int intensity = (device_sm[dev_id] > 500 && !is_windows()) ? 29 : 25; int intensity = (device_sm[dev_id] > 500 && !is_windows()) ? 29 : 25;
if (device_sm[dev_id] < 350) intensity = 22; if (device_sm[dev_id] < 350) intensity = 22;
@ -357,15 +443,13 @@ extern "C" int scanhash_decred(int thr_id, struct work* work, uint32_t max_nonce
uint32_t throughput = cuda_default_throughput(thr_id, 1U << intensity); uint32_t throughput = cuda_default_throughput(thr_id, 1U << intensity);
if (init[thr_id]) throughput = min(throughput, max_nonce - first_nonce); if (init[thr_id]) throughput = min(throughput, max_nonce - first_nonce);
const dim3 grid((throughput + TPB-1)/(TPB)); const dim3 grid((throughput + (NPT*TPB)-1)/(NPT*TPB));
const dim3 block(TPB); const dim3 block(TPB);
int rc = 0;
if (opt_benchmark) { if (opt_benchmark) {
ptarget[6] = swab32(0xff); ptarget[6] = swab32(0xff);
} }
if (!init[thr_id]) { if (!init[thr_id]){
cudaSetDevice(dev_id); cudaSetDevice(dev_id);
if (opt_cudaschedule == -1 && gpu_threads == 1) { if (opt_cudaschedule == -1 && gpu_threads == 1) {
cudaDeviceReset(); cudaDeviceReset();
@ -375,36 +459,67 @@ extern "C" int scanhash_decred(int thr_id, struct work* work, uint32_t max_nonce
CUDA_LOG_ERROR(); CUDA_LOG_ERROR();
} }
CUDA_CALL_OR_RET_X(cudaMalloc(&d_resNonce[thr_id], sizeof(uint32_t)), -1); CUDA_CALL_OR_RET_X(cudaMalloc(&d_resNonce[thr_id], maxResults*sizeof(uint32_t)), -1);
CUDA_CALL_OR_RET_X(cudaMallocHost(&h_resNonce[thr_id], sizeof(uint32_t)), -1); CUDA_CALL_OR_RET_X(cudaMallocHost(&h_resNonce[thr_id], maxResults*sizeof(uint32_t)), -1);
init[thr_id] = true; init[thr_id] = true;
} }
cudaMemset(d_resNonce[thr_id], 0xff, sizeof(uint32_t));
memcpy(endiandata, pdata, 180); memcpy(endiandata, pdata, 180);
decred_cpu_setBlock_52(endiandata); decred_cpu_setBlock_52(thr_id, endiandata, &pdata[32]);
h_resNonce[thr_id][0] = 1;
do { do {
if (h_resNonce[thr_id][0])
cudaMemset(d_resNonce[thr_id], 0x00, sizeof(uint32_t));
// GPU HASH // GPU HASH
decred_gpu_hash_nonce <<<grid, block>>> (throughput, (*pnonce), d_resNonce[thr_id], targetHigh); decred_gpu_hash_nonce <<<grid, block>>> (throughput, (*pnonce), d_resNonce[thr_id]);
cudaThreadSynchronize();
cudaMemcpy(h_resNonce[thr_id], d_resNonce[thr_id], sizeof(uint32_t), cudaMemcpyDeviceToHost); cudaMemcpy(h_resNonce[thr_id], d_resNonce[thr_id], sizeof(uint32_t), cudaMemcpyDeviceToHost);
if (h_resNonce[thr_id][0] != UINT32_MAX) {
rc = 1; if (h_resNonce[thr_id][0])
// work_set_target_ratio(work, vhashcpu); {
cudaMemcpy(h_resNonce[thr_id], d_resNonce[thr_id], (h_resNonce[thr_id][0]+1)*sizeof(uint32_t), cudaMemcpyDeviceToHost);
for(uint32_t i=1; i <= h_resNonce[thr_id][0]; i++)
{
uint32_t vhash64[8];
be32enc(&endiandata[DCR_NONCE_OFT32], h_resNonce[thr_id][i]);
decred_hash(vhash64, endiandata);
if (vhash64[6] <= ptarget[6] && fulltest(vhash64, ptarget))
{
int rc = 1;
work_set_target_ratio(work, vhash64);
*hashes_done = (*pnonce) - first_nonce + throughput; *hashes_done = (*pnonce) - first_nonce + throughput;
work->nonces[0] = swab32(h_resNonce[thr_id][0]); work->nonces[0] = swab32(h_resNonce[thr_id][i]);
// search for another nonce
for(uint32_t j=i+1; j <= h_resNonce[thr_id][0]; j++)
{
be32enc(&endiandata[DCR_NONCE_OFT32], h_resNonce[thr_id][j]);
decred_hash(vhash64, endiandata);
if (vhash64[6] <= ptarget[6] && fulltest(vhash64, ptarget)){
work->nonces[1] = swab32(h_resNonce[thr_id][j]);
if(!opt_quiet)
gpulog(LOG_NOTICE, thr_id, "second nonce found %u / %08x - %u / %08x", i, work->nonces[0], j, work->nonces[1]);
if(bn_hash_target_ratio(vhash64, ptarget) > work->shareratio) {
work_set_target_ratio(work, vhash64);
xchg(work->nonces[1], work->nonces[0]);
}
rc = 2;
break;
}
}
*pnonce = work->nonces[0]; *pnonce = work->nonces[0];
return 1; return rc;
}
}
} }
*pnonce += throughput; *pnonce += throughput;
} while (!work_restart[thr_id].restart && (uint64_t)max_nonce > (uint64_t)throughput + (uint64_t)(*pnonce)); } while (!work_restart[thr_id].restart && max_nonce > (uint64_t)throughput + (*pnonce));
*hashes_done = (*pnonce) - first_nonce; *hashes_done = (*pnonce) - first_nonce;
MyStreamSynchronize(NULL, 0, device_map[thr_id]); MyStreamSynchronize(NULL, 0, device_map[thr_id]);
return rc; return 0;
} }
// cleanup // cleanup

Write Preview
Loading…
Cancel
Save