This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH]: Don't XFAIL 21674.cc on darwin
On Sun, Dec 14, 2008 at 08:25:40PM +0100, Paolo Bonzini wrote:
>
> >> Hum, the idea is fine; I'm unfamiliar with the construct. I've seen it
> >> spelled this way:
> >>
> >> target { ! { powerpc*-*-darwin* hppa*-*-hpux* ia64-*-hpux* } }
> >>
> >> not not sure about that spelling. Ok if you're sure or find someone to
> >> confirm it.
> >
> > Mike,
> > The testsuite/22_locale/ctype/is/char/2.cc testcase uses this form
> > of a negative match...
> >
> > // { dg-do run { xfail *-*-![linux]* } }
> >
> >
> > ...so unless that testcase is broken, the use of 'xfail *-*-![darwin]*'
> > should be equally valid.
>
> I agree that the syntax seemed weird and I had never seen it, and that
> I'd rather find the spelling similar to Mike's proposed one in other
> testcases, and use it. Maybe fixing the other example you found, too. :-)
>
> Paolo
Paolo,
I haven't found any examples which show that xfail can be used in
combination with target? Are you sure that combination is valid? I think
we shouldn't avoid running this test on darwin since it does pass and would
check that the behvaior isn't regressed in some later darwin libc.
Jack