GOSTcoin support for ccminer CUDA miner project, compatible with most nvidia cards
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 

156 lines
4.3 KiB

/**
* S3 Hash (Also called 3S - Used by 1Coin)
*/
extern "C" {
#include "sph/sph_skein.h"
#include "sph/sph_shavite.h"
#include "sph/sph_simd.h"
}
#include "miner.h"
#include "cuda_helper.h"
#include <stdint.h>
static uint32_t *d_hash[MAX_GPUS];
extern void x11_shavite512_cpu_init(int thr_id, uint32_t threads);
extern void x11_shavite512_cpu_hash_80(int thr_id, uint32_t threads, uint32_t startNounce, uint32_t *d_hash, int order);
extern void x11_shavite512_setBlock_80(void *pdata);
extern int x11_simd512_cpu_init(int thr_id, uint32_t threads);
extern void x11_simd512_cpu_hash_64(int thr_id, uint32_t threads, uint32_t startNounce, uint32_t *d_nonceVector, uint32_t *d_hash, int order);
extern void x11_simd512_cpu_free(int thr_id);
extern void quark_skein512_cpu_init(int thr_id, uint32_t threads);
extern void quark_skein512_cpu_hash_64(int thr_id, uint32_t threads, uint32_t startNounce, uint32_t *d_nonceVector, uint32_t *d_hash, int order);
/* CPU HASH */
extern "C" void s3hash(void *output, const void *input)
{
sph_shavite512_context ctx_shavite;
sph_simd512_context ctx_simd;
sph_skein512_context ctx_skein;
unsigned char hash[64];
sph_shavite512_init(&ctx_shavite);
sph_shavite512(&ctx_shavite, input, 80);
sph_shavite512_close(&ctx_shavite, (void*) hash);
sph_simd512_init(&ctx_simd);
sph_simd512(&ctx_simd, (const void*) hash, 64);
sph_simd512_close(&ctx_simd, (void*) hash);
sph_skein512_init(&ctx_skein);
sph_skein512(&ctx_skein, (const void*) hash, 64);
sph_skein512_close(&ctx_skein, (void*) hash);
memcpy(output, hash, 32);
}
static bool init[MAX_GPUS] = { 0 };
/* Main S3 entry point */
extern "C" int scanhash_s3(int thr_id, struct work* work, uint32_t max_nonce, unsigned long *hashes_done)
{
uint32_t *pdata = work->data;
uint32_t *ptarget = work->target;
const uint32_t first_nonce = pdata[19];
int intensity = 20; // 256*256*8*2;
#ifdef WIN32
// reduce by one the intensity on windows
intensity--;
#endif
uint32_t throughput = cuda_default_throughput(thr_id, 1 << intensity);
if (init[thr_id]) throughput = min(throughput, max_nonce - first_nonce);
if (opt_benchmark)
((uint32_t*)ptarget)[7] = 0xF;
if (!init[thr_id])
{
cudaSetDevice(device_map[thr_id]);
x11_shavite512_cpu_init(thr_id, throughput);
x11_simd512_cpu_init(thr_id, throughput);
quark_skein512_cpu_init(thr_id, throughput);
CUDA_CALL_OR_RET_X(cudaMalloc(&d_hash[thr_id], 16 * sizeof(uint32_t) * throughput), 0);
cuda_check_cpu_init(thr_id, throughput);
init[thr_id] = true;
}
uint32_t endiandata[20];
for (int k=0; k < 20; k++)
be32enc(&endiandata[k], pdata[k]);
x11_shavite512_setBlock_80((void*)endiandata);
cuda_check_cpu_setTarget(ptarget);
do {
const uint32_t Htarg = ptarget[7];
uint32_t foundNonce;
int order = 0;
x11_shavite512_cpu_hash_80(thr_id, throughput, pdata[19], d_hash[thr_id], order++);
x11_simd512_cpu_hash_64(thr_id, throughput, pdata[19], NULL, d_hash[thr_id], order++);
quark_skein512_cpu_hash_64(thr_id, throughput, pdata[19], NULL, d_hash[thr_id], order++);
*hashes_done = pdata[19] - first_nonce + throughput;
foundNonce = cuda_check_hash(thr_id, throughput, pdata[19], d_hash[thr_id]);
if (foundNonce != UINT32_MAX)
{
uint32_t vhash64[8];
be32enc(&endiandata[19], foundNonce);
s3hash(vhash64, endiandata);
if (vhash64[7] <= Htarg && fulltest(vhash64, ptarget)) {
int res = 1;
uint32_t secNonce = cuda_check_hash_suppl(thr_id, throughput, pdata[19], d_hash[thr_id], 1);
work_set_target_ratio(work, vhash64);
if (secNonce != 0) {
be32enc(&endiandata[19], secNonce);
s3hash(vhash64, endiandata);
if (bn_hash_target_ratio(vhash64, ptarget) > work->shareratio)
work_set_target_ratio(work, vhash64);
pdata[21] = secNonce;
res++;
}
pdata[19] = foundNonce;
return res;
} else {
applog(LOG_WARNING, "GPU #%d: result for nonce $%08X does not validate on CPU!", device_map[thr_id], foundNonce);
}
}
pdata[19] += throughput;
} while (pdata[19] < max_nonce && !work_restart[thr_id].restart);
*hashes_done = pdata[19] - first_nonce + 1;
return 0;
}
// cleanup
extern "C" void free_s3(int thr_id)
{
if (!init[thr_id])
return;
cudaSetDevice(device_map[thr_id]);
cudaFree(d_hash[thr_id]);
x11_simd512_cpu_free(thr_id);
cuda_check_cpu_free(thr_id);
init[thr_id] = false;
cudaDeviceSynchronize();
}