1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
|
$void __global__ TimesTwoKernel(cuDoubleComplex *A,
$ cuDoubleComplex *B,
$ int const N)
${
$ /* Calculate the global linear index, assuming a 1-d grid. */
$ int const i = blockDim.x * blockIdx.x + threadIdx.x;
$ if (i < N) {
$ /* B[i] = make_cuDoubleComplex(2.0,0.0) * A[i]; */
$ B[i] = cuCmul(make_cuDoubleComplex(2.0,0.0),A[i]);
$ }
$}
$void TimesTwo(cuDoubleComplex *A,
$ cuDoubleComplex *B,
$ int const N)
${
$ int const threadsPerBlock = 256;
$ int blocksPerGrid = (N + threadsPerBlock - 1) / threadsPerBlock;
$ TimesTwoKernel<<<blocksPerGrid, threadsPerBlock>>>(A, B, N);
$}
@function result = timestwo_complex(a)
n = numel(a)
# TimesTwo(gpu dcomplex[] a, gpu output dcomplex[n] result, int n);
end
|
