Debian Bug report logs - #381720
Binutils screwed for i386 binaries on amd64.

Toggle useless messages

Message #5 received at submit@bugs.debian.org (full text, mbox, reply):

From: "Zephaniah E. Hull" <warp@aehallh.com>

To: submit@bugs.debian.org

Subject: Binutils screwed for i386 binaries on amd64.

Date: Sun, 6 Aug 2006 14:42:08 -0400

[Message part 1 (text/plain, inline)]

Package: binutils
Version: 2.17-2
Severity: important

When compiling the kernel on amd64, I'm ending up with:

  CC      arch/x86_64/boot/compressed/misc.o
objdump: arch/x86_64/boot/compressed/.tmp_misc.o: File format is ambiguous
objdump: Matching formats: elf32-i386 elf32-i386-freebsd
  OBJCOPY arch/x86_64/boot/compressed/vmlinux.bin
  GZIP    arch/x86_64/boot/compressed/vmlinux.bin.gz
  LD      arch/x86_64/boot/compressed/piggy.o
  LD      arch/x86_64/boot/compressed/vmlinux
  OBJCOPY arch/x86_64/boot/vmlinux.bin
objcopy: arch/x86_64/boot/compressed/vmlinux: File format is ambiguous
objcopy: Matching formats: elf32-i386 elf32-i386-freebsd 
make[2]: *** [arch/x86_64/boot/vmlinux.bin] Error 1 

This breaks all attempts to use objdump/objcopy on i386 binaries under
amd64, which breaks the kernel compilation quite solidly.

Zephaniah E. Hull.

-- 
	  1024D/E65A7801 Zephaniah E. Hull <warp@aehallh.com>
	   92ED 94E4 B1E6 3624 226D  5727 4453 008B E65A 7801
	    CCs of replies from mailing lists are requested.

Ken Thompson claims that he started developing UNIX so he could play
Space War, but the end product shows he was really much more interested
in cheating at Scrabble.
  -- Seen in the SDM.

[signature.asc (application/pgp-signature, inline)]

Message #10 received at 381720@bugs.debian.org (full text, mbox, reply):

From: "Zephaniah E. Hull" <warp@aehallh.com>

To: 381720@bugs.debian.org

Subject: Found it.

Date: Sun, 6 Aug 2006 15:13:39 -0400

[Message part 1 (text/plain, inline)]

The addition of i486-kfreebsd-gnu to configure-multi-stamp in 2.17-2
broke it.

Reverting that change seems to fix things.

Either that patch needs to be reverted, or binutils needs to be taught
how to tell the bloody difference.

Zephaniah E. Hull.

-- 
	  1024D/E65A7801 Zephaniah E. Hull <warp@aehallh.com>
	   92ED 94E4 B1E6 3624 226D  5727 4453 008B E65A 7801
	    CCs of replies from mailing lists are requested.

  Yes, Java is so bulletproofed that to a C programmer it feels like
being in a straightjacket, but it's a really comfy and warm
straightjacket, and the world would be a safer place if everyone was
straightjacketed most of the time.        -- Overheard in the SDM.

[signature.asc (application/pgp-signature, inline)]

Message #15 received at 381720-quiet@bugs.debian.org (full text, mbox, reply):

From: Petr Salinger <Petr.Salinger@seznam.cz>

To: "Zephaniah E. Hull" <warp@aehallh.com>

Cc: 381720-quiet@bugs.debian.org

Subject: manipulating i386 binaries on amd64

Date: Fri, 11 Aug 2006 11:20:14 +0200 (CEST)

> Either that patch needs to be reverted, or binutils needs to be taught
> how to tell the bloody difference.

Or do not left binutils(-multiarch) to guess format and directly tell 
which format they should use.

objdump -b elf32-i386
objcopy -I elf32-i386 -O elf32-i386

With plain binutils package, the guess is not ambiguous,
as there is only one matching format.

Petr

Message #20 received at 381720-quiet@bugs.debian.org (full text, mbox, reply):

From: "Zephaniah E. Hull" <warp@aehallh.com>

To: Petr Salinger <Petr.Salinger@seznam.cz>

Cc: 381720-quiet@bugs.debian.org

Subject: Re: manipulating i386 binaries on amd64

Date: Fri, 11 Aug 2006 16:55:55 -0400

[Message part 1 (text/plain, inline)]

On Fri, Aug 11, 2006 at 11:20:14AM +0200, Petr Salinger wrote:
> 
> >Either that patch needs to be reverted, or binutils needs to be taught
> >how to tell the bloody difference.
> 
> Or do not left binutils(-multiarch) to guess format and directly tell 
> which format they should use.
> 
> objdump -b elf32-i386
> objcopy -I elf32-i386 -O elf32-i386
> 
> With plain binutils package, the guess is not ambiguous,
> as there is only one matching format.

Which would be acceptable, if -multiarch did not override objdump and
objcopy, as it is if you have -multiarch installed for cross compiling
you break local kernel compiling.  Which is not exactly optimal.

Zephaniah E. Hull.

-- 
	  1024D/E65A7801 Zephaniah E. Hull <warp@aehallh.com>
	   92ED 94E4 B1E6 3624 226D  5727 4453 008B E65A 7801
	    CCs of replies from mailing lists are requested.

<cas> Mercury: gpm isn't a very good web browser.  fix it.

[signature.asc (application/pgp-signature, inline)]

Message #27 received at 381720@bugs.debian.org (full text, mbox, reply):

From: Jonathan Nieder <jrnieder@gmail.com>

To: 381720@bugs.debian.org

Subject: Re: manipulating i386 binaries on amd64

Date: Fri, 3 Jun 2011 18:13:56 -0500

reassign 381720 binutils-multiarch 2.17-2
quit

Petr Salinger wrote:

>> Either that patch needs to be reverted, or binutils needs to be taught
>> how to tell the bloody difference.
>
> Or do not left binutils(-multiarch) to guess format and directly
> tell which format they should use.
>
> objdump -b elf32-i386
> objcopy -I elf32-i386 -O elf32-i386
>
> With plain binutils package, the guess is not ambiguous,
> as there is only one matching format.

I'm not sure if kfreebsd can be detected in ELF headers yet (as it
ought to).  As Petr says, this only affects binutils-multiarch.
Reassigning to avoid distraction.

Message #38 received at 381720-done@bugs.debian.org (full text, mbox, reply):

From: Matthias Klose <doko@debian.org>

To: 381720-done@bugs.debian.org

Subject: closing old binutils issue, please recheck with recent binutils

Date: Thu, 18 Dec 2014 18:28:07 +0000

Control: tags -1 + wontfix moreinfo

Closing binutils bug reports reported for ancient versions
of binutils, or for outdated packages built with outdated
versions of binutils.  This doesn't mean that this issue
is fixed, but your help is needed to confirm the issue with
a recent binutils version (currently 2.24.90.20141209-1,
available in unstable).

Please reopen the issue if you think this still needs fixing,
adding all relevant files (including system libraries) in
a tarball attached to the bug report.

Send a report that this bug log contains spam.