[Bug tree-optimization/24126] New: [4.1 Regression] tree-stdarg.c causes wrong code, not tracking va_list variable correctly

[pinskia at gcc dot gnu dot org]

Take the following code:
#include <stdarg.h>

int
foo (int a, ...)
{
  va_list ap;
  va_list ap1;
  va_list ap2;
  va_list ap3;
  va_list ap4;

  va_start (ap, a);
  int b;
  ap1 = ap;
  ap2 = ap1;

  do {
   ap1 = ap2;
    b = va_arg (ap1, int);
    ap2 = ap1;
  } while (b > 10);

  return a + b;



}

int
main ()
{
  if (foo (1, 2, 3) != 3)
    abort ();
  return 0;
}

---

With -O1, we get an abort as the argruments at not saved to the stack at all.

This is reduced/changed from 20000519-1.c which was exposed by RTH's patch:
        * c-common.c (handle_mode_attribute): When not modifying in place,
        create subtypes for enumerations.
        (sync_resolve_return): Use TYPE_MAIN_VARIANT.
        * gimplify.c (create_tmp_from_val): Likewise.

-- 
           Summary: [4.1 Regression] tree-stdarg.c causes wrong code, not
                    tracking va_list variable correctly
           Product: gcc
           Version: 4.1.0
            Status: UNCONFIRMED
          Keywords: wrong-code
          Severity: critical
          Priority: P2
         Component: tree-optimization
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: pinskia at gcc dot gnu dot org
                CC: gcc-bugs at gcc dot gnu dot org
GCC target triplet: powerpc-darwin,powerpc-aix


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=24126