Subject: Re: toolchain/29170: -current tools cross-build fails on FreeBSD in lib/libc/gen/pw_scan.c
To: None <toolchain-manager@netbsd.org, gnats-admin@netbsd.org,>
From: James Chacon <jmc@NetBSD.org>
List: netbsd-bugs
Date: 01/31/2005 18:07:01
The following reply was made to PR toolchain/29170; it has been noted by GNATS.
From: James Chacon <jmc@NetBSD.org>
To: gnats-bugs@NetBSD.org
Cc: toolchain-manager@NetBSD.org, gnats-admin@NetBSD.org,
netbsd-bugs@NetBSD.org
Subject: Re: toolchain/29170: -current tools cross-build fails on FreeBSD in lib/libc/gen/pw_scan.c
Date: Mon, 31 Jan 2005 12:06:32 -0600
On Mon, Jan 31, 2005 at 11:35:02AM +0000, Lubomir Sedlacik wrote:
> The following reply was made to PR toolchain/29170; it has been noted by GNATS.
>
> From: Lubomir Sedlacik <salo@Xtrmntr.org>
> To: gnats-bugs@netbsd.org
> Cc:
> Subject: Re: toolchain/29170: -current tools cross-build fails on FreeBSD in lib/libc/gen/pw_scan.c
> Date: Mon, 31 Jan 2005 12:34:07 +0100
>
> --dMyqICaxQaaUjrCL
> Content-Type: text/plain; charset=us-ascii
> Content-Disposition: inline
> Content-Transfer-Encoding: quoted-printable
>
> proposed fix:
>
> Index: compat_defs.h
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
> RCS file: /cvsroot/src/tools/compat/compat_defs.h,v
> retrieving revision 1.45
> diff -u -r1.45 compat_defs.h
> --- compat_defs.h 11 Dec 2004 09:34:08 -0000 1.45
> +++ compat_defs.h 31 Jan 2005 11:30:20 -0000
> @@ -448,6 +448,13 @@
> #ifndef GID_MAX
> #define GID_MAX UID_MAX
> #endif
> +#ifndef LOGIN_NAME_MAX
> +# if defined(_POSIX_LOGIN_NAME_MAX)
> +# define LOGIN_NAME_MAX _POSIX_LOGIN_NAME_MAX
> +# else
> +# define LOGIN_NAME_MAX 9
> +# endif
> +#endif
> =20
> #ifndef UQUAD_MAX
> #define UQUAD_MAX ((u_quad_t)-1)
1. Can you make sure any attachments/patches are always sent in some form of
ascii only? Gnats ate part of this.
2. I noticed this on a solaris cross build as well and most definitly
we don't want to pick up any local definition of this so I'll fix it a
different way to always force the in-tree definition to be used.
James