This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug tree-optimization/66997] New: outer loop reduction fails to parallelize with graphite


https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66997

            Bug ID: 66997
           Summary: outer loop reduction fails to parallelize with
                    graphite
           Product: gcc
           Version: 6.0
            Status: UNCONFIRMED
          Severity: enhancement
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: vries at gcc dot gnu.org
  Target Milestone: ---

When running autopar.exp with
--target_board=unix/-ffast-math/-floop-parallelize-all, we run into:
...
FAIL: gcc.dg/autopar/uns-outer-6.c scan-tree-dump-times optimized "loopfn" 4
FAIL: gcc.dg/autopar/uns-outer-6.c scan-tree-dump-times parloops "parallelizing
outer loop" 1
...

[ We can drop -ffast-math once
https://gcc.gnu.org/ml/gcc-patches/2015-07/msg01863.html has been committed. ] 

relevant function in uns-outer-6.c:
...
void __attribute__((noinline))
parloop (int N)
{
  int i, j;
  unsigned int sum;

  /* Outer loop reduction, outerloop is parallelized.  */
  sum=0;
  for (i = 0; i < N; i++)
    {
      for (j = 0; j < N; j++)
        y[i]=x[i][j];
      sum += y[i];
    }
  g_sum = sum;
}
...

we find in the parloops dump:
...
Trying loop 3 as candidate
loop 3 is not innermost
loop is not parallel according to graphite
...


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]