| 1 | // RUN: %libomptarget-compilexx-run-and-check-generic |
|---|---|
| 2 | |
| 3 | // REQUIRES: gpu |
| 4 | |
| 5 | #include <cassert> |
| 6 | #include <iostream> |
| 7 | |
| 8 | int main(int argc, char *argv[]) { |
| 9 | constexpr const int num_threads = 64, N = 128; |
| 10 | int array[num_threads] = {0}; |
| 11 | |
| 12 | #pragma omp parallel for |
| 13 | for (int i = 0; i < num_threads; ++i) { |
| 14 | int tmp[N]; |
| 15 | |
| 16 | for (int j = 0; j < N; ++j) { |
| 17 | tmp[j] = i; |
| 18 | } |
| 19 | |
| 20 | #pragma omp target teams distribute parallel for map(tofrom : tmp) |
| 21 | for (int j = 0; j < N; ++j) { |
| 22 | tmp[j] += j; |
| 23 | } |
| 24 | |
| 25 | for (int j = 0; j < N; ++j) { |
| 26 | array[i] += tmp[j]; |
| 27 | } |
| 28 | } |
| 29 | |
| 30 | // Verify |
| 31 | for (int i = 0; i < num_threads; ++i) { |
| 32 | const int ref = (0 + N - 1) * N / 2 + i * N; |
| 33 | assert(array[i] == ref); |
| 34 | } |
| 35 | |
| 36 | std::cout << "PASS\n"; |
| 37 | |
| 38 | return 0; |
| 39 | } |
| 40 | |
| 41 | // CHECK: PASS |
| 42 |