GOSTSec
/
ccminer
mirror of https://github.com/GOSTSec/ccminer
1
0
Fork 0
Code Issues Releases Wiki Activity
Browse Source

lyra2: remove unused nonce param

pull/5/head
Tanguy Pruvot 7 years ago
parent
57f8f776fb
commit
a9357e1ec8
5 changed files with 28 additions and 28 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. 6
      lyra2/allium.cu
  2. 30
      lyra2/cuda_lyra2.cu
  3. 4
      lyra2/cuda_lyra2_sm2.cuh
  4. 12
      lyra2/cuda_lyra2_sm5.cuh
  5. 4
      lyra2/lyra2RE.cu

6
lyra2/allium.cu
Unescape View File

@ -30,7 +30,7 @@ extern void cubehash256_cpu_hash_32(int thr_id, uint32_t threads, uint32_t start
extern void skein256_cpu_init(int thr_id, uint32_t threads); extern void skein256_cpu_init(int thr_id, uint32_t threads);
extern void lyra2_cpu_init(int thr_id, uint32_t threads, uint64_t *d_matrix); extern void lyra2_cpu_init(int thr_id, uint32_t threads, uint64_t *d_matrix);
extern void lyra2_cpu_hash_32(int thr_id, uint32_t threads, uint32_t startNonce, uint64_t *d_outputHash, bool gtx750ti); extern void lyra2_cpu_hash_32(int thr_id, uint32_t threads, uint64_t *d_outputHash, bool gtx750ti);
extern void groestl256_cpu_init(int thr_id, uint32_t threads); extern void groestl256_cpu_init(int thr_id, uint32_t threads);
extern void groestl256_cpu_free(int thr_id); extern void groestl256_cpu_free(int thr_id);
@ -141,9 +141,9 @@ extern "C" int scanhash_allium(int thr_id, struct work* work, uint32_t max_nonce
//blake256_cpu_hash_80(thr_id, throughput, pdata[19], d_hash[thr_id], order++); //blake256_cpu_hash_80(thr_id, throughput, pdata[19], d_hash[thr_id], order++);
//keccak256_sm3_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], order++); //keccak256_sm3_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], order++);
blakeKeccak256_cpu_hash_80(thr_id, throughput, pdata[19], d_hash[thr_id], order++); blakeKeccak256_cpu_hash_80(thr_id, throughput, pdata[19], d_hash[thr_id], order++);
lyra2_cpu_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], gtx750ti); lyra2_cpu_hash_32(thr_id, throughput, d_hash[thr_id], gtx750ti);
cubehash256_cpu_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], order++); cubehash256_cpu_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], order++);
lyra2_cpu_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], gtx750ti); lyra2_cpu_hash_32(thr_id, throughput, d_hash[thr_id], gtx750ti);
skein256_cpu_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], order++); skein256_cpu_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], order++);
*hashes_done = pdata[19] - first_nonce + throughput; *hashes_done = pdata[19] - first_nonce + throughput;

30
lyra2/cuda_lyra2.cu
Unescape View File

@ -409,7 +409,7 @@ __constant__ uint2x4 blake2b_IV[2] = {
}; };
__global__ __launch_bounds__(64, 1) __global__ __launch_bounds__(64, 1)
void lyra2_gpu_hash_32_1(uint32_t threads, uint32_t startNounce, uint2 *g_hash) void lyra2_gpu_hash_32_1(uint32_t threads, uint2 *g_hash)
{ {
const uint32_t thread = (blockDim.x * blockIdx.x + threadIdx.x); const uint32_t thread = (blockDim.x * blockIdx.x + threadIdx.x);
@ -436,7 +436,7 @@ void lyra2_gpu_hash_32_1(uint32_t threads, uint32_t startNounce, uint2 *g_hash)
__global__ __global__
__launch_bounds__(TPB52, 1) __launch_bounds__(TPB52, 1)
void lyra2_gpu_hash_32_2(uint32_t threads, uint32_t startNounce, uint64_t *g_hash) void lyra2_gpu_hash_32_2(uint32_t threads, uint64_t *g_hash)
{ {
const uint32_t thread = blockDim.y * blockIdx.x + threadIdx.y; const uint32_t thread = blockDim.y * blockIdx.x + threadIdx.y;
@ -481,7 +481,7 @@ void lyra2_gpu_hash_32_2(uint32_t threads, uint32_t startNounce, uint64_t *g_has
} }
__global__ __launch_bounds__(64, 1) __global__ __launch_bounds__(64, 1)
void lyra2_gpu_hash_32_3(uint32_t threads, uint32_t startNounce, uint2 *g_hash) void lyra2_gpu_hash_32_3(uint32_t threads, uint2 *g_hash)
{ {
const uint32_t thread = blockDim.x * blockIdx.x + threadIdx.x; const uint32_t thread = blockDim.x * blockIdx.x + threadIdx.x;
@ -502,7 +502,7 @@ void lyra2_gpu_hash_32_3(uint32_t threads, uint32_t startNounce, uint2 *g_hash)
g_hash[thread + threads * 2] = state[0].z; g_hash[thread + threads * 2] = state[0].z;
g_hash[thread + threads * 3] = state[0].w; g_hash[thread + threads * 3] = state[0].w;
} //thread }
} }
#else #else
#if __CUDA_ARCH__ < 500 #if __CUDA_ARCH__ < 500
@ -510,9 +510,9 @@ void lyra2_gpu_hash_32_3(uint32_t threads, uint32_t startNounce, uint2 *g_hash)
/* for unsupported SM arch */ /* for unsupported SM arch */
__device__ void* DMatrix; __device__ void* DMatrix;
#endif #endif
__global__ void lyra2_gpu_hash_32_1(uint32_t threads, uint32_t startNounce, uint2 *g_hash) {} __global__ void lyra2_gpu_hash_32_1(uint32_t threads, uint2 *g_hash) {}
__global__ void lyra2_gpu_hash_32_2(uint32_t threads, uint32_t startNounce, uint64_t *g_hash) {} __global__ void lyra2_gpu_hash_32_2(uint32_t threads, uint64_t *g_hash) {}
__global__ void lyra2_gpu_hash_32_3(uint32_t threads, uint32_t startNounce, uint2 *g_hash) {} __global__ void lyra2_gpu_hash_32_3(uint32_t threads, uint2 *g_hash) {}
#endif #endif
__host__ __host__
@ -523,7 +523,7 @@ void lyra2_cpu_init(int thr_id, uint32_t threads, uint64_t *d_matrix)
} }
__host__ __host__
void lyra2_cpu_hash_32(int thr_id, uint32_t threads, uint32_t startNounce, uint64_t *d_hash, bool gtx750ti) void lyra2_cpu_hash_32(int thr_id, uint32_t threads, uint64_t *d_hash, bool gtx750ti)
{ {
int dev_id = device_map[thr_id % MAX_GPUS]; int dev_id = device_map[thr_id % MAX_GPUS];
@ -544,11 +544,11 @@ void lyra2_cpu_hash_32(int thr_id, uint32_t threads, uint32_t startNounce, uint6
if (cuda_arch[dev_id] >= 520) if (cuda_arch[dev_id] >= 520)
{ {
lyra2_gpu_hash_32_1 <<< grid2, block2 >>> (threads, startNounce, (uint2*)d_hash); lyra2_gpu_hash_32_1 <<< grid2, block2 >>> (threads, (uint2*)d_hash);
lyra2_gpu_hash_32_2 <<< grid1, block1, 24 * (8 - 0) * sizeof(uint2) * tpb >>> (threads, startNounce, d_hash); lyra2_gpu_hash_32_2 <<< grid1, block1, 24 * (8 - 0) * sizeof(uint2) * tpb >>> (threads, d_hash);
lyra2_gpu_hash_32_3 <<< grid2, block2 >>> (threads, startNounce, (uint2*)d_hash); lyra2_gpu_hash_32_3 <<< grid2, block2 >>> (threads, (uint2*)d_hash);
} }
else if (cuda_arch[dev_id] >= 500) else if (cuda_arch[dev_id] >= 500)
{ {
@ -561,12 +561,12 @@ void lyra2_cpu_hash_32(int thr_id, uint32_t threads, uint32_t startNounce, uint6
// suitable amount to adjust for 10warp // suitable amount to adjust for 10warp
shared_mem = 6144; shared_mem = 6144;
lyra2_gpu_hash_32_1_sm5 <<< grid2, block2 >>> (threads, startNounce, (uint2*)d_hash); lyra2_gpu_hash_32_1_sm5 <<< grid2, block2 >>> (threads, (uint2*)d_hash);
lyra2_gpu_hash_32_2_sm5 <<< grid1, block1, shared_mem >>> (threads, startNounce, (uint2*)d_hash); lyra2_gpu_hash_32_2_sm5 <<< grid1, block1, shared_mem >>> (threads, (uint2*)d_hash);
lyra2_gpu_hash_32_3_sm5 <<< grid2, block2 >>> (threads, startNounce, (uint2*)d_hash); lyra2_gpu_hash_32_3_sm5 <<< grid2, block2 >>> (threads, (uint2*)d_hash);
} }
else else
lyra2_gpu_hash_32_sm2 <<< grid3, block3 >>> (threads, startNounce, d_hash); lyra2_gpu_hash_32_sm2 <<< grid3, block3 >>> (threads, d_hash);
} }

4
lyra2/cuda_lyra2_sm2.cuh
Unescape View File

@ -131,7 +131,7 @@ void reduceDuplexRowSetup(const int rowIn, const int rowInOut, const int rowOut,
} }
__global__ __launch_bounds__(TPB30, 1) __global__ __launch_bounds__(TPB30, 1)
void lyra2_gpu_hash_32_sm2(uint32_t threads, uint32_t startNounce, uint64_t *g_hash) void lyra2_gpu_hash_32_sm2(uint32_t threads, uint64_t *g_hash)
{ {
uint32_t thread = (blockDim.x * blockIdx.x + threadIdx.x); uint32_t thread = (blockDim.x * blockIdx.x + threadIdx.x);
if (thread < threads) if (thread < threads)
@ -224,5 +224,5 @@ void lyra2_gpu_hash_32_sm2(uint32_t threads, uint32_t startNounce, uint64_t *g_h
#else #else
/* if __CUDA_ARCH__ < 200 .. host */ /* if __CUDA_ARCH__ < 200 .. host */
__global__ void lyra2_gpu_hash_32_sm2(uint32_t threads, uint32_t startNounce, uint64_t *g_hash) {} __global__ void lyra2_gpu_hash_32_sm2(uint32_t threads, uint64_t *g_hash) {}
#endif #endif

12
lyra2/cuda_lyra2_sm5.cuh
Unescape View File

@ -589,7 +589,7 @@ void reduceDuplexRowV50_8(const int rowInOut, uint2 state[4], const uint32_t thr
} }
__global__ __launch_bounds__(64, 1) __global__ __launch_bounds__(64, 1)
void lyra2_gpu_hash_32_1_sm5(uint32_t threads, uint32_t startNounce, uint2 *g_hash) void lyra2_gpu_hash_32_1_sm5(uint32_t threads, uint2 *g_hash)
{ {
const uint32_t thread = (blockDim.x * blockIdx.x + threadIdx.x); const uint32_t thread = (blockDim.x * blockIdx.x + threadIdx.x);
@ -622,7 +622,7 @@ void lyra2_gpu_hash_32_1_sm5(uint32_t threads, uint32_t startNounce, uint2 *g_ha
} }
__global__ __launch_bounds__(TPB50, 1) __global__ __launch_bounds__(TPB50, 1)
void lyra2_gpu_hash_32_2_sm5(uint32_t threads, uint32_t startNounce, uint2 *g_hash) void lyra2_gpu_hash_32_2_sm5(uint32_t threads, uint2 *g_hash)
{ {
const uint32_t thread = (blockDim.y * blockIdx.x + threadIdx.y); const uint32_t thread = (blockDim.y * blockIdx.x + threadIdx.y);
@ -662,7 +662,7 @@ void lyra2_gpu_hash_32_2_sm5(uint32_t threads, uint32_t startNounce, uint2 *g_ha
} }
__global__ __launch_bounds__(64, 1) __global__ __launch_bounds__(64, 1)
void lyra2_gpu_hash_32_3_sm5(uint32_t threads, uint32_t startNounce, uint2 *g_hash) void lyra2_gpu_hash_32_3_sm5(uint32_t threads, uint2 *g_hash)
{ {
const uint32_t thread = (blockDim.x * blockIdx.x + threadIdx.x); const uint32_t thread = (blockDim.x * blockIdx.x + threadIdx.x);
@ -687,7 +687,7 @@ void lyra2_gpu_hash_32_3_sm5(uint32_t threads, uint32_t startNounce, uint2 *g_ha
#else #else
/* if __CUDA_ARCH__ != 500 .. host */ /* if __CUDA_ARCH__ != 500 .. host */
__global__ void lyra2_gpu_hash_32_1_sm5(uint32_t threads, uint32_t startNounce, uint2 *g_hash) {} __global__ void lyra2_gpu_hash_32_1_sm5(uint32_t threads, uint2 *g_hash) {}
__global__ void lyra2_gpu_hash_32_2_sm5(uint32_t threads, uint32_t startNounce, uint2 *g_hash) {} __global__ void lyra2_gpu_hash_32_2_sm5(uint32_t threads, uint2 *g_hash) {}
__global__ void lyra2_gpu_hash_32_3_sm5(uint32_t threads, uint32_t startNounce, uint2 *g_hash) {} __global__ void lyra2_gpu_hash_32_3_sm5(uint32_t threads, uint2 *g_hash) {}
#endif #endif

4
lyra2/lyra2RE.cu
Unescape View File

@ -26,7 +26,7 @@ extern void skein256_cpu_hash_32(int thr_id, uint32_t threads, uint32_t startNon
extern void skein256_cpu_init(int thr_id, uint32_t threads); extern void skein256_cpu_init(int thr_id, uint32_t threads);
extern void lyra2_cpu_init(int thr_id, uint32_t threads, uint64_t *d_matrix); extern void lyra2_cpu_init(int thr_id, uint32_t threads, uint64_t *d_matrix);
extern void lyra2_cpu_hash_32(int thr_id, uint32_t threads, uint32_t startNonce, uint64_t *d_outputHash, bool gtx750ti); extern void lyra2_cpu_hash_32(int thr_id, uint32_t threads, uint64_t *d_outputHash, bool gtx750ti);
extern void groestl256_cpu_init(int thr_id, uint32_t threads); extern void groestl256_cpu_init(int thr_id, uint32_t threads);
extern void groestl256_cpu_free(int thr_id); extern void groestl256_cpu_free(int thr_id);
@ -130,7 +130,7 @@ extern "C" int scanhash_lyra2(int thr_id, struct work* work, uint32_t max_nonce,
//blake256_cpu_hash_80(thr_id, throughput, pdata[19], d_hash[thr_id], order++); //blake256_cpu_hash_80(thr_id, throughput, pdata[19], d_hash[thr_id], order++);
//keccak256_sm3_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], order++); //keccak256_sm3_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], order++);
blakeKeccak256_cpu_hash_80(thr_id, throughput, pdata[19], d_hash[thr_id], order++); blakeKeccak256_cpu_hash_80(thr_id, throughput, pdata[19], d_hash[thr_id], order++);
lyra2_cpu_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], gtx750ti); lyra2_cpu_hash_32(thr_id, throughput, d_hash[thr_id], gtx750ti);
skein256_cpu_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], order++); skein256_cpu_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], order++);
*hashes_done = pdata[19] - first_nonce + throughput; *hashes_done = pdata[19] - first_nonce + throughput;

Write Preview
Loading…
Cancel
Save