Re: [testsuite] XFAIL gfortran.dg/nint_2.f90 only for -O0

[Janis Johnson <janis187 at us dot ibm dot com>]

On Thu, 2008-05-22 at 12:41 +0900, Tim Prince wrote:
> Janis Johnson wrote:
> > This test fails on some targets due to a bug in the math library, which
> > is out of our control.  It has been XFAILed for those targets, but that
> > makes one torture run XFAIL and the rest XPASS, cluttering up the test
> > summary.  Use the new test directive dg-xfail-run-if to XFAIL only the
> > option that is expected to fail, allowing the rest of the torture
> > options to PASS.
> > 
> > Tested on powerpc64-linux with -m32/-m64, checked in on trunk.
> > 
> > 2008-05-21  Janis Johnson  <janis187@us.ibm.com>
> > 
> > 	* gfortran.dg/nint_2.f90: XFAIL only when using -O0.
> > 
> > Index: gcc/testsuite/gfortran.dg/nint_2.f90
> > ===================================================================
> > --- gcc/testsuite/gfortran.dg/nint_2.f90	(revision 135729)
> > +++ gcc/testsuite/gfortran.dg/nint_2.f90	(working copy)
> > @@ -3,7 +3,8 @@
> >  ! PR 31202
> >  ! http://gcc.gnu.org/ml/fortran/2005-04/msg00139.html
> >  !
> > -! { dg-do run { xfail powerpc-ibm-aix* powerpc*-*-linux* *-*-mingw* } }
> > +! { dg-do run }
> > +! { dg-xfail-run-if "PR 33271, math library bug" { powerpc-ibm-aix powerpc*-*-linux* *-*-mingw* } { "-O0" } { "" } }
> >    real(kind=8) :: a
> >    integer(kind=8) :: i1, i2
> >    real :: b
> > 
> > 
> > 
> > --- AV & Spam Filtering by M+Guardian - Risk Free Email (TM) ---
> > 
> > 
> cygwin targets are among those which FAIL.  I know there is precedent for 
> FAILing cygwin where others are XFAILed for the same test, but I'd like to 
> know if we are being asked to stop any efforts for cygwin.

I simply moved the existing XFAIL list.  If it should include cygwin,
feel free to add it.

Janis