This is the mail archive of the
gcc-prs@gcc.gnu.org
mailing list for the GCC project.
Re: optimization/10239: [m68k] switch jumptable causes assembler error
- From: Gunther Nikl <gni at gecko dot de>
- To: nobody at gcc dot gnu dot org
- Cc: gcc-prs at gcc dot gnu dot org,
- Date: 12 May 2003 12:06:02 -0000
- Subject: Re: optimization/10239: [m68k] switch jumptable causes assembler error
- Reply-to: Gunther Nikl <gni at gecko dot de>
The following reply was made to PR optimization/10239; it has been noted by GNATS.
From: Gunther Nikl <gni@gecko.de>
To: Dara Hazeghi <dhazeghi@yahoo.com>
Cc: gcc-gnats@gcc.gnu.org
Subject: Re: optimization/10239: [m68k] switch jumptable causes assembler error
Date: Mon, 12 May 2003 13:53:01 +0200
On Fri, May 09, 2003 at 05:21:11PM -0700, Dara Hazeghi wrote:
> with binutils 2.13.1 and gcc 3.3 branch and mainline cvs (20030508),
> I cannot repeat the bug you reported, ie no complaints from the
> assembler.
I have used an older binutils version: 2.9.1. You are right, that 2.13.1
doesn't complain BUT that doesn't mean thats better... GAS simply emits
zeros where the older gas complained. I would like to see compatibilty
with older versions.
> Can you verify whether you still encounter this bug in current gcc 3.3
> branch?
The bug is still present in the second 3.3 prerelease: "3.3 20030509
(prerelease)". You should look at the generated code with "-O" and with
"-O -fgcse". GCC shouldn't emit references to (local) symbols that aren't
present in the first place. I call that broken.
Gunther