[llvm-bugs] [Bug 50733] New: [OPENMP 4.5]

via llvm-bugs llvm-bugs at lists.llvm.org
Wed Jun 16 01:06:35 PDT 2021


https://bugs.llvm.org/show_bug.cgi?id=50733

            Bug ID: 50733
           Summary: [OPENMP 4.5]
           Product: clang
           Version: 10.0
          Hardware: PC
                OS: Linux
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: OpenMP
          Assignee: unassignedclangbugs at nondot.org
          Reporter: xiao.liu at compiler-dev.com
                CC: llvm-bugs at lists.llvm.org

The testcase is,

#include <stdio.h>
#define M 10
int b = 10;
int main()
{
  int a[M], i;
  #pragma omp parallel
  {
    #pragma omp for linear(b:2)
    for (i = 0; i < M; i++) {
      a[i] = b;
    }
  }
  for (i = 0; i < M; i++) {
    printf("%d  ", a[i]);
  }
  printf("\n%d\n", b);
  return 0;
}

The result is,

10  12  14  16  18  20  22  24  26  28  
final b: 30


See OpenMP 4.5 specification, 2.15.3.7

"When a linear clause is specified on a construct,(...),The value corresponding
to the sequentially last iteration of the associated loop(s) is assigned to the
original list item."

I think that the final value of 'b' should be 28. And if there is some
assignment to 'b' in the loop, like as,

    #pragma omp for linear(b:2)
    for (i = 0; i < M; i++) {
      a[i] = b;
      b += 2;
    }

The final value of 'b' should be calculate as, (28+2)=30, instead of
(10+M*2)=30.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20210616/b3c61e2d/attachment.html>


More information about the llvm-bugs mailing list