[Bug tree-optimization/37837] New: missed PRE

[Bug tree-optimization/37837] New: missed PRE

[rguenth at gcc dot gnu dot org]

Testcase where only one constant, not two alternative constants, are on the
entry of the threaded bb.

extern int printf (const char *, ...);

void
__attribute__((noinline))
test (double cn, int *neig)
{
  double tt, al[3];

  *neig = 3;

  if (__builtin_fabs(cn) > 1.)
    cn = -1.;

  tt = __builtin_atan2 (__builtin_sqrt (1. - __builtin_pow (cn, 2.)), cn)
       * 3.333333333333333e-1;

  al[0] = __builtin_cos (tt);
  al[1] = __builtin_cos (2.0943951023931944 + tt);
  al[2] = __builtin_cos (4.1887902047863879 + tt);

  if ((__builtin_fabs ( al[0] - al[1]) < 1.e-5)
      || (__builtin_fabs (al[0] - al[2]) < 1.e-5)
      || (__builtin_fabs (al[1] - al[2]) < 1.e-5))
    *neig = 2;
}

int
main ()
{
  int neig;

  test (-1.0, &neig);
  printf ("neig = %d\n", neig);
  if (neig != 2)
    __builtin_abort ();
  test (1.0, &neig);
  printf ("neig = %d\n", neig);
  if (neig != 2)
    __builtin_abort ();
  test (-2.0, &neig);
  printf ("neig = %d\n", neig);
  if (neig != 2)
    __builtin_abort ();
  return 0;  
}


cn == -1 is available.


-- 
           Summary: missed PRE
           Product: gcc
           Version: 4.4.0
            Status: UNCONFIRMED
          Keywords: missed-optimization
          Severity: enhancement
          Priority: P3
         Component: tree-optimization
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: rguenth at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37837

[Bug tree-optimization/37837] missed PRE

[rguenth at gcc dot gnu dot org]

------- Comment #1 from rguenth at gcc dot gnu dot org  2008-11-01 14:14 -------
Seems to work now.  DOM1 threads the jump, if disabled PRE correctly detects
the partial redundant calls and optimizes them.  But we miss a jump threading
pass after PRE which makes us end up with

<bb 2>:
  *neig = 3;
  if (ABS_EXPR <cn> > 1.0e+0)
    goto <bb 9>;
  else
    goto <bb 3>;

<bb 9>:
  prephitmp.33 = 1.5e+0;
  prephitmp.31 = 4.99999999999997002397833512077340856194496154785e-1;
  prephitmp.29 = -1.0e+0;
  prephitmp.17 = 5.00000000000000111022302462515654042363166809082e-1;
  goto <bb 4>;

...

<bb 4>:
  if (prephitmp.33 < 1.00000000000000008180305391403130954586231382564e-5)
    goto <bb 7>;
  else
    goto <bb 5>;

<bb 5>:
  if (ABS_EXPR <prephitmp.17 - prephitmp.31> <
1.00000000000000008180305391403130954586231382564e-5)
    goto <bb 7>;
  else
    goto <bb 6>;

<bb 6>:
  if (ABS_EXPR <prephitmp.29 - prephitmp.31> <
1.00000000000000008180305391403130954586231382564e-5)
    goto <bb 7>;
  else
    goto <bb 8>;

<bb 7>:
  *neig = 2;


note that VRP doesn't track floating-point ranges and thus does not thread
jumps here.


-- 

rguenth at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|                            |FIXED
   Target Milestone|---                         |4.4.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37837