[PATCH v5 02/20] gas: move code for object attribute parsing into obj-elf-attr.c

[Jan Beulich]

On 10.07.2025 13:38, Matthieu Longo wrote:
> On 2025-07-10 10:48, Richard Earnshaw (lists) wrote:
>> On 09/07/2025 11:12, Jan Beulich wrote:
>>> On 09.07.2025 12:02, Richard Earnshaw (lists) wrote:
>>>> On 07/07/2025 17:49, Matthieu Longo wrote:
>>>>> --- /dev/null
>>>>> +++ b/gas/config/obj-elf-attr.h
>>>>> @@ -0,0 +1,52 @@
>>>>> +/* Object attributes parsing.
>>>>> +   Copyright (C) 2025 Free Software Foundation, Inc.
>>>>> +
>>>>> +   This file is part of GAS, the GNU Assembler.
>>>>> +
>>>>> +   GAS is free software; you can redistribute it and/or modify
>>>>> +   it under the terms of the GNU General Public License as published by
>>>>> +   the Free Software Foundation; either version 3, or (at your option)
>>>>> +   any later version.
>>>>> +
>>>>> +   GAS is distributed in the hope that it will be useful,
>>>>> +   but WITHOUT ANY WARRANTY; without even the implied warranty of
>>>>> +   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
>>>>> +   GNU General Public License for more details.
>>>>> +
>>>>> +   You should have received a copy of the GNU General Public License
>>>>> +   along with GAS; see the file COPYING.  If not, write to the Free
>>>>> +   Software Foundation, 51 Franklin Street - Fifth Floor, Boston, MA
>>>>> +   02110-1301, USA.  */
>>>>> +
>>>>> +#ifndef _OBJ_ELF_ATTR_H
>>>>> +#define _OBJ_ELF_ATTR_H
>>>>> +
>>>>> +#include "as.h"
>>>>> +#include "bfd/elf-bfd.h"
>>>>> +
>>>>> +/* The target supports Object Attributes v1.  */
>>>>> +#if OBJ_ELF \
>>>>
>>>> I don't think you need to test OBJ_ELF now, this file is only built when that is true.
>>>
>>> This being a header file, don't things depend on where the file is included
>>> from?
>>>
>>> Jan
>>
>> The issue is moot if the TC_OBJ_ATTR definition is moved to the individual tc-<arch>.h files.  At that point you probably do need to test OBJ_ELF again as those headers are definitely used by multiple obj formats.
>>
>> R.
> 
> I am a bit confused about what the next step is for me.
> 
> As Richard stated it, if I move the definitions into tc-<arch>.h, then I 
> won't be able to determine whether the target is ELF or not because 
> OBJ_ELF is defined into obj-elf.h
> 
> In this case, it looks to me that I will need to use OBJ_MAYBE_ELF instead.
> This seems to be set from gas/configure.ac,
> 
>      case $fmt in
>        aout)	AC_DEFINE(OBJ_MAYBE_AOUT, 1,    [a.out support?])   ;;
>        coff)	AC_DEFINE(OBJ_MAYBE_COFF, 1,    [COFF support?])    ;;
>        ecoff)	AC_DEFINE(OBJ_MAYBE_ECOFF, 1,   [ECOFF support?])   ;;
>        elf)	AC_DEFINE(OBJ_MAYBE_ELF, 1,     [ELF support?])     ;;
>        generic)	AC_DEFINE(OBJ_MAYBE_GENERIC, 1, [generic support?]) ;;
>        som)	AC_DEFINE(OBJ_MAYBE_SOM, 1,     [SOM support?])     ;;
>      esac
> 
> but the value is not set in build-dir/gas/config.h when it is compiled 
> for target aarch64-none-linux-gnu.
> 
> /* ELF support? */
> /* #undef OBJ_MAYBE_ELF */
> 
> Unless you see a solution to detect whether the target is ELF from 
> tc-<arch>.h, I don't see how I will be able to move the definitions into 
> tc-<arch>.h

Include order is targ-env.h -> te-*.h -> obj-format.h -> targ-cpu.h -> tc-*.h.
How else would various tc-*.h use OBJ_* and/or OBJ_MAYBE_* checks?

Jan