This is the mail archive of the
binutils@sourceware.org
mailing list for the binutils project.
RE: Binutils 2.26 released
- From: "Pierre Muller" <pierre dot muller at ics-cnrs dot unistra dot fr>
- To: "'Tristan Gingold'" <gingold at adacore dot com>, "'binutils'" <binutils at sourceware dot org>
- Date: Mon, 1 Feb 2016 15:07:39 +0100
- Subject: RE: Binutils 2.26 released
- Authentication-results: sourceware.org; auth=none
- References: <9F317B17-DC7C-42A1-A3BD-89E7A349FAE5 at adacore dot com> <56A8FC25 dot 6060503 at ubuntu dot com> <00cb01d15cf3$e9a8fef0$bcfafcd0$ at muller@ics-cnrs.unistra.fr> <3834D8F4-3992-47FD-9680-089D631A8DD8 at adacore dot com>
I am willing to submit a patch,
the problem is that I am not sure I understand the error...
There are many other AT_XXX macros in
/usr/include/sys/auxv.h
Thus I suppose that gcc does not generate an error
if the old macro and the new are the same,
otherwise I would have gotten an error on the first AT_NULL macro.
Thus it is only because in auxv.h there is:
#define AT_SUN_CAP_HW1 2009
followed by
#define AT_SUN_HWCAP AT_SUN_CAP_HW1 /* deprecated; for backward compat */
whereas in include/elf/common.h
#define AT_SUN_HWCAP 2009 /* Machine dependent hints about
processor capabilities. */
So should I restrict the change to AT_SU_HWCAP only,
or rather discard the whole AT_XXX macros?
Comments welcome,
Pierre
> -----Message d'origine-----
> De : binutils-owner@sourceware.org [mailto:binutils-
> owner@sourceware.org] De la part de Tristan Gingold
> Envoyé : lundi 1 février 2016 14:40
> À : Pierre Muller
> Cc : Matthias Klose; binutils
> Objet : Re: Binutils 2.26 released
>
>
> > On 01 Feb 2016, at 14:24, Pierre Muller <pierre.muller@ics-
> cnrs.unistra.fr> wrote:
> >
> > Hi all,
> >
> > in order to be able to compile successfully binutils-2.26 sources
> > on a Oracle Solaris machine,
> >
> > cat /etc/release
> > Oracle Solaris 11.2 X86
> > Copyright (c) 1983, 2014, Oracle and/or its affiliates. All rights
> > reserved.
> > Assembled 23 June 2014
> >
> > I needed to modify include/elf/common.h
> > (see patch below).
> >
> > Is this related to the fact that development was not set to false?
>
> Very very unlikely. You should submit your patch.
>
> Tristan.
>
> >
> > Pierre Muller
> >
> >>> I am pleased to announce binutils 2.26. You can get it from
> >> sourceware.org
> >>> and ftp.gnu.org in the usual places, e.g.
> >> ftp://ftp.gnu.org/gnu/binutils/.
> >>
> >> Thanks!
> >>
> >> However, the release doesn't set back the development mode to
> release.
> >> At least
> >> bfd/development.sh is still set to true, causing the date included
> in
> >> the
> >> sonames. Is anything else affected?
> >>
> >
> > In file included from
> > /export/home/pierre/gnu/binutils-2.26/ld/../bfd/elf-bfd.h:25:0,
> > from /export/home/pierre/gnu/binutils-
> 2.26/ld/ldlang.c:43:
> > /export/home/pierre/gnu/binutils-
> 2.26/ld/../include/elf/common.h:1010:0:
> > error: "AT_SUN_HWCAP" redefined [-Werror]
> > #define AT_SUN_HWCAP 2009 /* Machine dependent hints about
> > ^
> > In file included from /usr/include/dlfcn.h:15:0,
> > from /export/home/pierre/gnu/binutils-
> 2.26/ld/sysdep.h:80,
> > from /export/home/pierre/gnu/binutils-
> 2.26/ld/ldlang.c:21:
> > /usr/include/sys/auxv.h:121:0: note: this is the location of the
> previous
> > definition
> > #define AT_SUN_HWCAP AT_SUN_CAP_HW1 /* deprecated; for backward
> compat */
> > ^
> > cc1: all warnings being treated as errors
> >
> > This small change allowed me to complete compilation of all-binutils
> all-gas
> > all-ld.
> >
> > diff -ur binutils-2.26/include/elf/common.h
> > binutils-2.26-mod/include/elf/common.h
> > --- binutils-2.26/include/elf/common.h 2015-11-13 09:27:42.000000000
> +0100
> > +++ binutils-2.26-mod/include/elf/common.h 2016-01-27
> > 15:31:19.234233985 +0100
> > @@ -998,6 +998,7 @@
> > #define AT_L2_CACHESHAPE 36
> > #define AT_L3_CACHESHAPE 37
> >
> > +#ifndef AT_SUN_UID
> > #define AT_SUN_UID 2000 /* Effective user ID. */
> > #define AT_SUN_RUID 2001 /* Real user ID. */
> > #define AT_SUN_GID 2002 /* Effective group ID. */
> > @@ -1009,6 +1010,8 @@
> > #define AT_SUN_PLATFORM 2008 /* Platform name string. */
> > #define AT_SUN_HWCAP 2009 /* Machine dependent hints about
> > processor capabilities. */
> > +#endif
> > +/* These are only in kernel */
> > #define AT_SUN_IFLUSH 2010 /* Should flush icache? */
> > #define AT_SUN_CPU 2011 /* CPU name string. */
> > #define AT_SUN_EMUL_ENTRY 2012 /* COFF entry point address. */
> > Only in binutils-2.26-mod/include/elf: common.h~