Simple matrix-vector multiplication kernel

MatrixVectorMultiplyKernel.cu

__global__ void multiplyKernel(float *c, const float *a, const float *b, const int size) {
    int index = threadIdx.x + blockIdx.x * blockDim.x;
 
	c[index] = 0;
 
	for (int j = 0; j < size; ++j)
		c[index] += a[index * size + j] * b[index];
}