Re: git: 98b88b839df1 - main - x11/cde: actually fix build on powerpc64le

From: Piotr Kubaj <pkubaj_at_anongoth.pl>
Date: Mon, 14 Feb 2022 12:08:23 UTC 
On 22-02-14 12:59:38, Jan Beich wrote:
> Piotr Kubaj <pkubaj@FreeBSD.org> writes:
> 
> > The branch main has been updated by pkubaj:
> >
> > URL: https://cgit.FreeBSD.org/ports/commit/?id=98b88b839df1a88f538b8bfdc74e9734b0df7ea3
> >
> > commit 98b88b839df1a88f538b8bfdc74e9734b0df7ea3
> > Author:     Piotr Kubaj <pkubaj@FreeBSD.org>
> > AuthorDate: 2022-02-14 11:38:50 +0000
> > Commit:     Piotr Kubaj <pkubaj@FreeBSD.org>
> > CommitDate: 2022-02-14 11:38:50 +0000
> >
> >     x11/cde: actually fix build on powerpc64le
> >     
> >     This port doesn't respect setting FLAGS via Makefile, so -DPOWERPC64LE goes unnoticed.
> >     
> >     Switch to ifdef __powerpc64__.
> [...]
> > -.if ${ARCH} == powerpc64le
> > -CPPFLAGS+=	-DPOWERPC64LE
> > -.endif
> [...]
> > -+#ifdef POWERPC64LE
> > ++#ifdef __powerpc64__
> >  +			tmpval = va_listval(va_arg(ap, va_listarg));
> >  +			va_copy(ap, tmpval);
> >  +#else
> >  +			va_copy(ap, va_listval(va_arg(ap, va_listarg)));
> >  +#endif
> 
> Do you intentionally ignore endianess here? __powerpc64__ is defined on
> both big (aka ARCH=powerpc64) and little (aka ARCH=powerpc64le).

Yes, powerpc64 has actually the same issue, but fixing it here causes segfaults (probably endianness-related) later on during build.