Subject: Re: Status of NetBSD-current vax
To: Matthias Scheler <tron@NetBSD.org>
From: James Chacon <jmc@NetBSD.org>
List: port-vax
Date: 04/05/2005 11:17:09
On Tue, Apr 05, 2005 at 10:22:59AM +0100, Matthias Scheler wrote:
> On Mon, Apr 04, 2005 at 08:17:30PM -0600, Michael L. Hitch wrote:
> > > cross building NetBSD-current vax on a NetBSD 2.0.2 system fails like this:
> > >
> > > #   compile  identd/identd.o
> > > /scratch/tron/toolsdir/bin/vax--netbsdelf-gcc -O2  -Wall -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -Wno-sign-compare -Wno-traditional -Wno-uninitialized -Wreturn-type -Wswitch -Wshadow -Wcast-qual -Wwrite-strings -std=c99  -Werror   -I/src/NetBSD-current/src/sys/dist/ipf -DWITH_IPF -DWITH_PF  -nostdinc -isystem /scratch/tron/destdir/usr/include  -c    /src/NetBSD-current/src/libexec/identd/identd.c
> > > cc1: unknown C standard `c99'
> > >
> > > *** Failed target:  identd.o
> > >
> > > Is this a known issue?
> > 
> >   I'm not sure if it's a "known" issue, but it appears to be due to WARNS
> > being set to 4 in the Makefile, which adds the -std=c99 option - which I
> > would presume is not supported by GCC 2.95 used on the vax.
> 
> Ah, that's where the flag comes from. I'll try this patch later:

No...The identd Makefile should be patched to just use WARNS=3

*nothing* else in the tree currently uses WARNS=4 and special casing the
.mk file just means it has to be found later if/when vax updates it's gcc.

If it's gonna be special cased, do it in the identd Makefile.

James

> 
> Index: bsd.sys.mk
> ===================================================================
> RCS file: /cvsroot/src/share/mk/bsd.sys.mk,v
> retrieving revision 1.115
> diff -u -r1.115 bsd.sys.mk
> --- bsd.sys.mk  20 Feb 2005 01:31:44 -0000      1.115
> +++ bsd.sys.mk  5 Apr 2005 09:22:23 -0000
> @@ -26,7 +26,7 @@
>  .if ${WARNS} > 2
>  CFLAGS+=       -Wcast-qual -Wwrite-strings
>  .endif
> -.if ${WARNS} > 3
> +.if ${WARNS} > 3 && ${MACHINE_ARCH} != "vax"
>  CFLAGS+=       -std=c99
>  .endif
>  .endif
> 
> 	Kind regards
> 
> -- 
> Matthias Scheler                                  http://scheler.de/~matthias/
> 
>