[Bug middle-end/47092] [4.6/4.7 Regression] gfortran.fortran-torture/execute/forall_4.f90 is VERY slow with -fgraphite-identity

[spop at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47092

Sebastian Pop <spop at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #5 from Sebastian Pop <spop at gcc dot gnu.org> 2011-07-27 18:12:14 UTC ---
This is a duplicate of PR47594: in this PR we generate a very long loop

  for (scat_3=0;scat_3<=4294967295*scat_1+3;scat_3++) {
    S17(scat_1,scat_3);
  }

that looks like the code gen error in PR47594:
http://gcc.gnu.org/ml/gcc-patches/2011-07/msg02177.html

    for (scat_3=0;scat_3<=4294967295*scat_1+T_51-1;scat_3++) {
      S6(scat_1,scat_3);
    }

This bug is due to the niter analysis that generates an unsigned
expression for the upper bound.

*** This bug has been marked as a duplicate of bug 47594 ***