File: omp_device_managed_memory_alloc.c

package info (click to toggle)
llvm-toolchain-15 1%3A15.0.6-4
  • links: PTS, VCS
  • area: main
  • in suites: bookworm
  • size: 1,554,644 kB
  • sloc: cpp: 5,922,452; ansic: 1,012,136; asm: 674,362; python: 191,568; objc: 73,855; f90: 42,327; lisp: 31,913; pascal: 11,973; javascript: 10,144; sh: 9,421; perl: 7,447; ml: 5,527; awk: 3,523; makefile: 2,520; xml: 885; cs: 573; fortran: 567
file content (28 lines) | stat: -rw-r--r-- 662 bytes parent folder | download | duplicates (5) 
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
27
28
 
// RUN: %libomptarget-compile-run-and-check-nvptx64-nvidia-cuda
// REQUIRES: nvptx64-nvidia-cuda

#include <omp.h>
#include <stdio.h>

int main() {
  const int N = 64;

  // Allocates device managed memory that is shared between the host and device.
  int *shared_ptr =
      omp_alloc(N * sizeof(int), llvm_omp_target_shared_mem_alloc);

#pragma omp target teams distribute parallel for is_device_ptr(shared_ptr)
  for (int i = 0; i < N; ++i) {
    shared_ptr[i] = 1;
  }

  int sum = 0;
  for (int i = 0; i < N; ++i)
    sum += shared_ptr[i];

  // CHECK: PASS
  if (sum == N)
    printf("PASS\n");

  omp_free(shared_ptr, llvm_omp_target_shared_mem_alloc);
}