mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Alexandre Chartre <alexandre.chartre@oracle.com>
To: David Laight <david.laight.linux@gmail.com>
Cc: alexandre.chartre@oracle.com,
	Nathan Chancellor <nathan@kernel.org>,
	linux-kernel@vger.kernel.org, mingo@kernel.org,
	jpoimboe@kernel.org, peterz@infradead.org
Subject: Re: [PATCH v6 03/30] objtool: Disassemble code with libopcodes instead of running objdump
Date: Wed, 26 Nov 2025 11:29:32 +0100	[thread overview]
Message-ID: <c79ec945-d1e5-48e7-a510-4ffbcb8c580c@oracle.com> (raw)
In-Reply-To: <20251126100729.277f2854@pumpkin>


On 11/26/25 11:07, David Laight wrote:
> On Wed, 26 Nov 2025 10:00:36 +0100
> Alexandre Chartre <alexandre.chartre@oracle.com> wrote:
> 
>> On 11/25/25 19:16, Nathan Chancellor wrote:
>>> Hi Alexandre,
>>>
>>> On Fri, Nov 21, 2025 at 10:53:13AM +0100, Alexandre Chartre wrote:
>>>> objtool executes the objdump command to disassemble code. Use libopcodes
>>>> instead to have more control about the disassembly scope and output.
>>>> If libopcodes is not present then objtool is built without disassembly
>>>> support.
>>>>
>>>> Signed-off-by: Alexandre Chartre <alexandre.chartre@oracle.com>
>>> ...
>>>> diff --git a/tools/objtool/include/objtool/arch.h b/tools/objtool/include/objtool/arch.h
>>>> index d89f8b5ec14e3..18c0e69ee6170 100644
>>>> --- a/tools/objtool/include/objtool/arch.h
>>>> +++ b/tools/objtool/include/objtool/arch.h
>>>> @@ -103,4 +103,13 @@ bool arch_absolute_reloc(struct elf *elf, struct reloc *reloc);
>>>>    unsigned int arch_reloc_size(struct reloc *reloc);
>>>>    unsigned long arch_jump_table_sym_offset(struct reloc *reloc, struct reloc *table);
>>>>    
>>>> +#ifdef DISAS
>>>> +
>>>> +#include <bfd.h>
>>>
>>> This include of bfd.h breaks the build for me:
>>>
>>>     $ make -skj"$(nproc)" ARCH=x86_64 CROSS_COMPILE=x86_64-linux- clean defconfig bzImage
>>>     In file included from tools/objtool/include/objtool/arch.h:108,
>>>                      from check.c:14:
>>>     /usr/include/bfd.h:35:2: error: #error config.h must be included before this header
>>>        35 | #error config.h must be included before this header
>>>           |  ^~~~~
>>>     ...
>>>
>>> where my bfd.h has:
>>>
>>>     #ifndef __BFD_H_SEEN__
>>>     #define __BFD_H_SEEN__
>>>
>>>     /* PR 14072: Ensure that config.h is included first.  */
>>>     #if !defined PACKAGE && !defined PACKAGE_VERSION
>>>     #error config.h must be included before this header
>>>     #endif
>>
>> This check is effectively present in the bfd.h file generated from the
>> binutils source code. However it is not present in the bfd.h file provided
>> by the binutils RPM. I think this explained why we haven't seen this issue
>> so far.
>>
>> For history, this was introduced in 2012 for bug 14072. Then there was
>> complaints reported in bug 14243 and 15920. But it was decided not to
>> remove this change, and the suggested fix was to define PACKAGE when
>> including bfd.h.
>>
>> https://sourceware.org/bugzilla/show_bug.cgi?id=14072
>> https://sourceware.org/bugzilla/show_bug.cgi?id=14243
>> https://sourceware.org/bugzilla/show_bug.cgi?id=15920
>>
>> And Redhat has fixed the issue for the binutils RPM by removing this test:
>>
>> https://sourceware.org/bugzilla/show_bug.cgi?id=14243
>>
>>
>>> Something like this cures it for me but I am not sure if that is a
>>> proper fix or not since I see config.h in my binutils build folder has
>>> many other defines.
>>>
>>> Cheers,
>>> Nathan
>>>
>>> diff --git a/tools/objtool/Makefile b/tools/objtool/Makefile
>>> index df793ca6fc1a..96df4a73da23 100644
>>> --- a/tools/objtool/Makefile
>>> +++ b/tools/objtool/Makefile
>>> @@ -87,7 +87,7 @@ BUILD_DISAS := n
>>>    
>>>    ifeq ($(feature-libbfd),1)
>>>    	BUILD_DISAS := y
>>> -	OBJTOOL_CFLAGS += -DDISAS
>>> +	OBJTOOL_CFLAGS += -DDISAS -DPACKAGE="objtool-disas"
>>>    	OBJTOOL_LDFLAGS += -lopcodes
>>>    endif
>>>      
>>
>> This is the proper fix (as indicated in the binutils bugs), and this is
>> what the other kernel tools using bfd.h (bpf and perf) do. I will create
>> a patch with your suggestion.
> 
> ISTM that defining it just before including bfd.h is cleaner and more obvious.
> 

bfd.h is included at two different places: in arch.h and disas.c (and disas.c
includes arch.h). So we would need to ensure that PACKAGE always has the same
definition otherwise we will have a redefine error.

So I think it is simpler to have a global definition with -DPACKAGE.
Similarly, bpf has multiple includes and uses -DPACKAGE=bpf, while perf has
a single include and use a single define before the include.

alex.


  reply	other threads:[~2025-11-26 10:29 UTC|newest]

Thread overview: 110+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-11-21  9:53 [PATCH v6 00/30] objtool: Function validation tracing Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 01/30] objtool: Move disassembly functions to a separated file Alexandre Chartre
2025-11-24  9:12   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 02/30] objtool: Create disassembly context Alexandre Chartre
2025-11-24  9:12   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 03/30] objtool: Disassemble code with libopcodes instead of running objdump Alexandre Chartre
2025-11-24  9:12   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-25 18:16   ` [PATCH v6 03/30] " Nathan Chancellor
2025-11-26  9:00     ` Alexandre Chartre
2025-11-26 10:07       ` David Laight
2025-11-26 10:29         ` Alexandre Chartre [this message]
2025-12-09  6:58   ` Guenter Roeck
2025-12-09  7:51     ` Alexandre Chartre
2025-12-09 15:11       ` Guenter Roeck
2025-12-09 22:25       ` Maciej W. Rozycki
2025-12-10  5:08         ` Guenter Roeck
2025-12-10  7:53           ` Alexandre Chartre
2025-12-10 15:26             ` Guenter Roeck
2025-12-10 16:36               ` Alexandre Chartre
2025-12-10 18:42                 ` Guenter Roeck
2025-12-10 21:54                   ` Guenter Roeck
2025-12-15 16:00                     ` Alexandre Chartre
2025-12-15 16:46                       ` Guenter Roeck
2025-12-10  7:06         ` Alexandre Chartre
2025-12-10  8:56           ` Peter Zijlstra
2025-12-10  8:53         ` Peter Zijlstra
2025-12-10 12:01           ` Maciej W. Rozycki
2025-11-21  9:53 ` [PATCH v6 04/30] tool build: Remove annoying newline in build output Alexandre Chartre
2025-11-24  9:12   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 05/30] objtool: Print symbol during disassembly Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 06/30] objtool: Store instruction disassembly result Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 07/30] objtool: Disassemble instruction on warning or backtrace Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 08/30] objtool: Extract code to validate instruction from the validate branch loop Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 09/30] objtool: Record symbol name max length Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 10/30] objtool: Add option to trace function validation Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 11/30] objtool: Trace instruction state changes during " Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-12-01 20:23   ` [PATCH v6 11/30] " Nathan Chancellor
2025-12-02  1:30     ` Josh Poimboeuf
2025-12-02  8:34       ` Alexandre Chartre
2025-12-02 16:13         ` Josh Poimboeuf
2025-12-02 18:09           ` David Laight
2025-11-21  9:53 ` [PATCH v6 12/30] objtool: Improve register reporting " Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 13/30] objtool: Identify the different types of alternatives Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 14/30] objtool: Add functions to better name alternatives Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 15/30] objtool: Improve tracing of alternative instructions Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 16/30] objtool: Do not validate IBT for .return_sites and .call_sites Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 17/30] objtool: Add the --disas=<function-pattern> action Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 18/30] objtool: Preserve alternatives order Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 19/30] objtool: Print headers for alternatives Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 20/30] objtool: Disassemble group alternatives Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 21/30] objtool: Print addresses with alternative instructions Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 22/30] objtool: Disassemble exception table alternatives Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 23/30] objtool: Disassemble jump " Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 24/30] objtool: Fix address references in alternatives Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 25/30] objtool: Provide access to feature and flags of group alternatives Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 26/30] objtool: Function to get the name of a CPU feature Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-24 10:44   ` tip-bot2 for Alexandre Chartre
2025-11-24 11:59     ` Peter Zijlstra
2025-11-24 12:26       ` Borislav Petkov
2025-11-24 12:40         ` Alexandre Chartre
2025-11-24 19:05           ` david laight
2025-11-24 16:43         ` Alexandre Chartre
2025-11-24 20:14           ` Borislav Petkov
2025-11-24 19:49   ` [tip: objtool/core] objtool: Add " tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 27/30] objtool: Improve naming of group alternatives Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-24 10:44   ` tip-bot2 for Alexandre Chartre
2025-11-24 19:49   ` tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 28/30] objtool: Compact output for alternatives with one instruction Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-24 10:43   ` tip-bot2 for Alexandre Chartre
2025-11-24 19:49   ` tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 29/30] objtool: Add wide output for disassembly Alexandre Chartre
2025-11-24  9:11   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-24 10:43   ` tip-bot2 for Alexandre Chartre
2025-11-24 19:49   ` tip-bot2 for Alexandre Chartre
2025-11-21  9:53 ` [PATCH v6 30/30] objtool: Trim trailing NOPs in alternative Alexandre Chartre
2025-11-24  9:10   ` [tip: objtool/core] " tip-bot2 for Alexandre Chartre
2025-11-24 10:43   ` tip-bot2 for Alexandre Chartre
2025-11-24 19:49   ` tip-bot2 for Alexandre Chartre
2025-11-21 10:36 ` [PATCH v6 00/30] objtool: Function validation tracing Peter Zijlstra
2025-11-21 13:16   ` Alexandre Chartre
2025-11-21 13:51     ` Alexandre Chartre
2025-11-21 14:30       ` Peter Zijlstra
2025-11-24 11:04 ` Borislav Petkov
2025-11-24 11:39   ` Alexandre Chartre
2025-11-24 12:39     ` Borislav Petkov
2025-11-24 12:27   ` David Laight

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=c79ec945-d1e5-48e7-a510-4ffbcb8c580c@oracle.com \
    --to=alexandre.chartre@oracle.com \
    --cc=david.laight.linux@gmail.com \
    --cc=jpoimboe@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=nathan@kernel.org \
    --cc=peterz@infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®