Changes

From CDOT Wiki
Jump to: navigation, search

A-Team

16 bytes added, 1 April
Initial implementation
int i = blockIdx.x * blockDim.x + threadIdx.x;
int j = blockIdx.y * blockDim.y + threadIdx.y;
//matrix multiplication if (i < ni && j < nj) { float sum = 0.0f;
for (int k = 0; k < nk; k++)
sum += d_a[i * nk + k] * d_b[k * nj + j];
d_p[i * nj + j] = sum;
}
113
edits

Navigation menu