GOSTSec/ccminer
1
0
Fork 0
mirror of https://github.com/GOSTSec/ccminer synced 2025-08-26 13:51:51 +00:00
Code Issues Releases Wiki Activity

skunk: proper indent and trim spaces

Browse Source
This commit is contained in:
Tanguy Pruvot 2017-08-14 03:55:49 +02:00
parent dfdf4b0691
commit 6055af1fa9
3 changed files with 29 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

31
skunk/cuda_skunk.cu
View File

@ -1,6 +1,7 @@
/** /**
* skein + cube + fugue (to fix) * skein + cube + fugue merged kernel, based on krnlx work
* Based on krnlx work *
* based on alexis78 sib kernels, final touch by tpruvot
*/ */
#include <miner.h> #include <miner.h>
@ -34,7 +35,8 @@ static void rrounds(uint32_t *x){
SWAP(x[ 4], x[12]);x[ 4] ^= x[20];x[12] ^= x[28];SWAP(x[ 5], x[13]);x[ 5] ^= x[21];x[13] ^= x[29]; SWAP(x[ 4], x[12]);x[ 4] ^= x[20];x[12] ^= x[28];SWAP(x[ 5], x[13]);x[ 5] ^= x[21];x[13] ^= x[29];
SWAP(x[ 6], x[14]);x[ 6] ^= x[22];x[14] ^= x[30];SWAP(x[ 7], x[15]);x[ 7] ^= x[23];x[15] ^= x[31]; SWAP(x[ 6], x[14]);x[ 6] ^= x[22];x[14] ^= x[30];SWAP(x[ 7], x[15]);x[ 7] ^= x[23];x[15] ^= x[31];
/* "swap x_1jk0m with x_1jk1m" */ /* "swap x_1jk0m with x_1jk1m" */
SWAP(x[16], x[18]); SWAP(x[17], x[19]); SWAP(x[20], x[22]); SWAP(x[21], x[23]);SWAP(x[24], x[26]); SWAP(x[25], x[27]); SWAP(x[28], x[30]); SWAP(x[29], x[31]); SWAP(x[16], x[18]); SWAP(x[17], x[19]); SWAP(x[20], x[22]); SWAP(x[21], x[23]);
SWAP(x[24], x[26]); SWAP(x[25], x[27]); SWAP(x[28], x[30]); SWAP(x[29], x[31]);
/* "add x_0jklm into x_1jklm modulo 2^32 rotate x_0jklm upwards by 11 bits" */ /* "add x_0jklm into x_1jklm modulo 2^32 rotate x_0jklm upwards by 11 bits" */
x[16] = x[16] + x[ 0]; x[ 0] = ROTL32(x[ 0],11);x[17] = x[17] + x[ 1];x[ 1] = ROTL32(x[ 1],11); x[16] = x[16] + x[ 0]; x[ 0] = ROTL32(x[ 0],11);x[17] = x[17] + x[ 1];x[ 1] = ROTL32(x[ 1],11);
x[18] = x[18] + x[ 2]; x[ 2] = ROTL32(x[ 2],11);x[19] = x[19] + x[ 3];x[ 3] = ROTL32(x[ 3],11); x[18] = x[18] + x[ 2]; x[ 2] = ROTL32(x[ 2],11);x[19] = x[19] + x[ 3];x[ 3] = ROTL32(x[ 3],11);
@ -50,7 +52,8 @@ static void rrounds(uint32_t *x){
SWAP(x[ 8], x[12]); x[ 8] ^= x[24]; x[12] ^= x[28]; SWAP(x[ 9], x[13]); x[ 9] ^= x[25]; x[13] ^= x[29]; SWAP(x[ 8], x[12]); x[ 8] ^= x[24]; x[12] ^= x[28]; SWAP(x[ 9], x[13]); x[ 9] ^= x[25]; x[13] ^= x[29];
SWAP(x[10], x[14]); x[10] ^= x[26]; x[14] ^= x[30]; SWAP(x[11], x[15]); x[11] ^= x[27]; x[15] ^= x[31]; SWAP(x[10], x[14]); x[10] ^= x[26]; x[14] ^= x[30]; SWAP(x[11], x[15]); x[11] ^= x[27]; x[15] ^= x[31];
/* "swap x_1jkl0 with x_1jkl1" */ /* "swap x_1jkl0 with x_1jkl1" */
SWAP(x[16], x[17]); SWAP(x[18], x[19]); SWAP(x[20], x[21]); SWAP(x[22], x[23]);SWAP(x[24], x[25]); SWAP(x[26], x[27]); SWAP(x[28], x[29]); SWAP(x[30], x[31]); SWAP(x[16], x[17]); SWAP(x[18], x[19]); SWAP(x[20], x[21]); SWAP(x[22], x[23]);
SWAP(x[24], x[25]); SWAP(x[26], x[27]); SWAP(x[28], x[29]); SWAP(x[30], x[31]);
} }
} }
@ -430,7 +433,7 @@ void skunk_gpu_hash_80(uint32_t threads, uint32_t startNounce, uint64_t *output6
TFBIG_4o_UI2(17); TFBIG_4o_UI2(17);
TFBIG_ADDKEY_UI2(p[0], p[1], p[2], p[3], p[4], p[5], p[6], p[7], h, t, 18); TFBIG_ADDKEY_UI2(p[0], p[1], p[2], p[3], p[4], p[5], p[6], p[7], h, t, 18);
//cubehash // cubehash512
uint32_t x[32] = { uint32_t x[32] = {
0x2AEA2A61, 0x50F494D4, 0x2D538B8B, 0x4167D83E, 0x2AEA2A61, 0x50F494D4, 0x2D538B8B, 0x4167D83E,
0x3FEE2313, 0xC701CF8C, 0xCC39968E, 0x50AC5695, 0x3FEE2313, 0xC701CF8C, 0xCC39968E, 0x50AC5695,
@ -470,6 +473,7 @@ void skunk_gpu_hash_80(uint32_t threads, uint32_t startNounce, uint64_t *output6
for (int i = 0;i < 10;++i) for (int i = 0;i < 10;++i)
rrounds(x); rrounds(x);
// fugue512
uint32_t Hash[16]; uint32_t Hash[16];
#pragma unroll 16 #pragma unroll 16
for(int i = 0; i < 16; i++) for(int i = 0; i < 16; i++)
@ -488,6 +492,7 @@ void skunk_gpu_hash_80(uint32_t threads, uint32_t startNounce, uint64_t *output6
FUGUE512_3(Hash[0x9], Hash[0xA], Hash[0xB]); FUGUE512_3(Hash[0x9], Hash[0xA], Hash[0xB]);
FUGUE512_3(Hash[0xC], Hash[0xD], Hash[0xE]); FUGUE512_3(Hash[0xC], Hash[0xD], Hash[0xE]);
FUGUE512_3(Hash[0xF], 0U, 512U); FUGUE512_3(Hash[0xF], 0U, 512U);
//#pragma unroll 16 //#pragma unroll 16
for (uint32_t i = 0; i < 32; i+=2){ for (uint32_t i = 0; i < 32; i+=2){
mROR3; mROR3;
@ -514,10 +519,14 @@ void skunk_gpu_hash_80(uint32_t threads, uint32_t startNounce, uint64_t *output6
} }
S[ 4] ^= S[ 0]; S[ 9] ^= S[ 0]; S[18] ^= S[ 0]; S[27] ^= S[ 0]; S[ 4] ^= S[ 0]; S[ 9] ^= S[ 0]; S[18] ^= S[ 0]; S[27] ^= S[ 0];
S[ 0] = cuda_swab32(S[ 1]); S[ 1] = cuda_swab32(S[ 2]); S[ 2] = cuda_swab32(S[ 3]); S[ 3] = cuda_swab32(S[ 4]); S[ 0] = cuda_swab32(S[ 1]); S[ 1] = cuda_swab32(S[ 2]);
S[ 4] = cuda_swab32(S[ 9]); S[ 5] = cuda_swab32(S[10]); S[ 6] = cuda_swab32(S[11]); S[ 7] = cuda_swab32(S[12]); S[ 2] = cuda_swab32(S[ 3]); S[ 3] = cuda_swab32(S[ 4]);
S[ 8] = cuda_swab32(S[18]); S[ 9] = cuda_swab32(S[19]); S[10] = cuda_swab32(S[20]); S[11] = cuda_swab32(S[21]); S[ 4] = cuda_swab32(S[ 9]); S[ 5] = cuda_swab32(S[10]);
S[12] = cuda_swab32(S[27]); S[13] = cuda_swab32(S[28]); S[14] = cuda_swab32(S[29]); S[15] = cuda_swab32(S[30]); S[ 6] = cuda_swab32(S[11]); S[ 7] = cuda_swab32(S[12]);
S[ 8] = cuda_swab32(S[18]); S[ 9] = cuda_swab32(S[19]);
S[10] = cuda_swab32(S[20]); S[11] = cuda_swab32(S[21]);
S[12] = cuda_swab32(S[27]); S[13] = cuda_swab32(S[28]);
S[14] = cuda_swab32(S[29]); S[15] = cuda_swab32(S[30]);
uint64_t *outpHash = &output64[thread<<3]; uint64_t *outpHash = &output64[thread<<3];
*(uint2x4*)&outpHash[ 0] = *(uint2x4*)&S[ 0]; *(uint2x4*)&outpHash[ 0] = *(uint2x4*)&S[ 0];
@ -621,7 +630,9 @@ void skunk_setBlock_80(int thr_id, void *pdata)
p[2] += p[3]; p[2] += p[3];
p[4] += p[5]; p[6] += p[7]; p[4] += p[5]; p[6] += p[7];
p[3] = ROTL64(p[3], 36) ^ p[2]; p[5] = ROTL64(p[5], 19) ^ p[4]; p[7] = ROTL64(p[7], 37) ^ p[6]; p[3] = ROTL64(p[3], 36) ^ p[2];
p[5] = ROTL64(p[5], 19) ^ p[4];
p[7] = ROTL64(p[7], 37) ^ p[6];
p[4] += p[7]; p[6] += p[5]; p[4] += p[7]; p[6] += p[5];
p[7] = ROTL64(p[7], 27) ^ p[4]; p[7] = ROTL64(p[7], 27) ^ p[4];

4
skunk/cuda_skunk_streebog.cu
View File

@ -1,12 +1,12 @@
/* /*
* Streebog GOST R 34.10-2012 CUDA implementation. * Streebog GOST R 34.10-2012 stripped CUDA implementation for final hash
* *
* https://tools.ietf.org/html/rfc6986 * https://tools.ietf.org/html/rfc6986
* https://en.wikipedia.org/wiki/Streebog * https://en.wikipedia.org/wiki/Streebog
* *
* ==========================(LICENSE BEGIN)============================ * ==========================(LICENSE BEGIN)============================
* *
* @author Tanguy Pruvot - 2015 * @author Tanguy Pruvot - 2017
* @author Alexis Provos - 2016 * @author Alexis Provos - 2016
*/ */

1
skunk/skunk.cu
View File

@ -156,7 +156,6 @@ extern "C" int scanhash_skunk(int thr_id, struct work* work, uint32_t max_nonce,
uint32_t secNonce = work->nonces[1] = startNounce + h_resNonce[1]; uint32_t secNonce = work->nonces[1] = startNounce + h_resNonce[1];
be32enc(&endiandata[19], secNonce); be32enc(&endiandata[19], secNonce);
skunk_hash(vhash, endiandata); skunk_hash(vhash, endiandata);
work->nonces[1] = secNonce;
if (bn_hash_target_ratio(vhash, ptarget) > work->shareratio[0]) { if (bn_hash_target_ratio(vhash, ptarget) > work->shareratio[0]) {
work_set_target_ratio(work, vhash); work_set_target_ratio(work, vhash);
xchg(work->nonces[1], work->nonces[0]); xchg(work->nonces[1], work->nonces[0]);