This is the mail archive of the
mailing list for the GCC project.
Re: Getting glibfortran to build on newlib targets aka. missing stdint.h (PR 14325)
- From: Paul Brook <paul at codesourcery dot com>
- To: Hans-Peter Nilsson <hp at bitrange dot com>
- Cc: gcc-patches at gcc dot gnu dot org,Paolo Bonzini <paolo dot bonzini at lu dot unisi dot ch>, fortran at gcc dot gnu dot org
- Date: Tue, 9 Nov 2004 15:39:03 +0000
- Subject: Re: Getting glibfortran to build on newlib targets aka. missing stdint.h (PR 14325)
- Organization: CodeSourcery
- References: <Pine.BSF.email@example.com> <418F3766.firstname.lastname@example.org> <Pine.BSF.email@example.com>
On Monday 08 November 2004 20:07, Hans-Peter Nilsson wrote:
> On Mon, 8 Nov 2004, Paolo Bonzini wrote:
> > > + # $1=typename, $2=alt1, $3=alt2, $4=cachevarsizename, $5=sizeof,
> > > $6=msgfrag. + AC_DEFUN([LIBGFOR_INTTYPE], [
> > > + AC_CHECK_TYPES([$1],,
> > > + [AC_CHECK_TYPES([$2],
> > > + [ac_cv_gforlib_$1=$2],
> > > + [AC_CHECK_SIZEOF([$3])
> > > + if test "$$4" != $5; then
> > Quite ugly...
> You're quite welcome, as long as we move forward.
> > I'd rather not have the fourth parameter and replace $$4 with
> > $ac_cv_sizeof_AS_TR_CPP([$3])
> Good suggestion. I'd like to hear from someone with approval
> rights whether the whole idea of autoconfing this is ok with
I'm ok with the idea of come sort of autoconf check, although I'm not sure
about the details of the implementation you suggest.
Hardcoding C typenames seem somewhat fragile, looping over [QHSDT]Imode types
may be better.
Also a prominent comment making it clear this is a kludge that can go away
once GCC provides stdint.h.