static cl_int queue_whirlpoolx_kernel(struct __clState *clState, struct _dev_blk_ctx *blk, __maybe_unused cl_uint threads)
{
uint64_t midblock[8], key[8] = { 0 }, tmp[8] = { 0 };
cl_ulong le_target;
cl_int status;
le_target = *(cl_ulong *)(blk->work->device_target + 24);
flip80(clState->cldata, blk->work->data);
memcpy(midblock, clState->cldata, 64);
// midblock = n, key = h
for (int i = 0; i < 10; ++i) {
tmp[0] = WHIRLPOOL_ROUND_CONSTANTS[i];
whirlpool_round(key, tmp);
tmp[0] = 0;
whirlpool_round(midblock, tmp);
for (int x = 0; x < 8; ++x) {
midblock[x] ^= key[x];
}
}
for (int i = 0; i < 8; ++i) {
midblock[i] ^= ((uint64_t *)(clState->cldata))[i];
}
status = clSetKernelArg(clState->kernel, 0, sizeof(cl_ulong8), (cl_ulong8 *)&midblock);
status |= clSetKernelArg(clState->kernel, 1, sizeof(cl_ulong), (void *)(((uint64_t *)clState->cldata) + 8));
status |= clSetKernelArg(clState->kernel, 2, sizeof(cl_ulong), (void *)(((uint64_t *)clState->cldata) + 9));
status |= clSetKernelArg(clState->kernel, 3, sizeof(cl_mem), (void *)&clState->outputBuffer);
status |= clSetKernelArg(clState->kernel, 4, sizeof(cl_ulong), (void *)&le_target);
return status;
}
void whirlpool_compress(uint8_t state[64], const uint8_t block[64]) {
const int NUM_ROUNDS = 10; // Any number from 0 to 32 is allowed
uint8_t tempState[64];
uint8_t tempBlock[64];
int i;
// Initialization
memcpy(tempState, state, 64);
for (i = 0; i < 64; i++)
tempBlock[i] = block[i] ^ state[i];
// Hashing rounds
uint8_t rcon[64];
memset(rcon + 8, 0, 56);
for (i = 0; i < NUM_ROUNDS; i++) {
int j;
for (j = 0; j < 8; j++)
rcon[j] = SBOX[(i << 3) | j];
whirlpool_round(tempState, rcon);
whirlpool_round(tempBlock, tempState);
}
// Final combining
for (i = 0; i < 64; i++)
state[i] ^= block[i] ^ tempBlock[i];
}
