// clang-format off
// This test verifies that Xteam reduction kernel is generated by default and the
// thread_limit clause is honored. The blocksize is rounded down to the nearest
// power of 2.
// 
// RUN: %libomptarget-compile-generic
// RUN: env LIBOMPTARGET_KERNEL_TRACE=1 %libomptarget-run-generic 2>&1 | %fcheck-generic

// UNSUPPORTED: nvptx64-nvidia-cuda
// UNSUPPORTED: nvptx64-nvidia-cuda-LTO
// UNSUPPORTED: aarch64-unknown-linux-gnu
// UNSUPPORTED: aarch64-unknown-linux-gnu-LTO
// UNSUPPORTED: x86_64-unknown-linux-gnu
// UNSUPPORTED: x86_64-unknown-linux-gnu-LTO

// clang-format on
#include <stdio.h>

int main() {
  int N = 100;

  double a[N];

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

  double sum1;
  sum1 = 0;

#pragma omp target teams distribute parallel for reduction(+ : sum1) thread_limit(130)
  for (int j = 0; j < N; j = j + 1)
    sum1 += a[j];

  int rc = sum1 != 4950;
  return rc;
}
// clang-format off
/// CHECK: DEVID:[[S:[ ]*]][[DEVID:[0-9]+]] SGN:8
/// CHECK: teamsXthrds:([[S:[ ]*]][[NUM_TEAMS:[0-9]+]]X 128)