This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: ICE in gcc/libgcc2.c:566 (gcc trunk)

From: "H. J. Lu" <hjl at lucon dot org>
To: Hanno Meyer-Thurow <h dot mth at web dot de>
Cc: Ian Lance Taylor <iant at google dot com>, gcc at gcc dot gnu dot org
Date: Sun, 11 Feb 2007 21:23:57 -0800
Subject: Re: ICE in gcc/libgcc2.c:566 (gcc trunk)
References: <20070207221818.99e72471.h.mth@web.de> <m3iredmtg1.fsf@localhost.localdomain> <20070211171115.e8e80f40.h.mth@web.de> <20070211210940.GB16058@lucon.org>

On Sun, Feb 11, 2007 at 01:09:40PM -0800, H. J. Lu wrote:
> On Sun, Feb 11, 2007 at 05:11:15PM +0100, Hanno Meyer-Thurow wrote:
> > On 07 Feb 2007 15:36:14 -0800
> > Ian Lance Taylor <iant@google.com> wrote:
> > 
> > > Can anybody else out there recreate this on their x86_64 system?
> > 
> > Not that I could not recreate the segfault but I found a way to hide the
> > segfault. Instead of optimizing with nocona
> > 
> > 	make \
> > 		LDFLAGS="-Wl,-O1 -Wl,--as-needed -Wl,--sort-common" \
> > 		STAGE1_CFLAGS="-O" \
> > 		LIBPATH="/usr/lib/gcj-4.3.0_alpha20070209/lib64" \
> > 		BOOT_CFLAGS=" -O2 -mtune=nocona -march=nocona -pipe -msse3 -mfpmath=sse" \
> > 		bootstrap
> > 
> 
> -mtune=nocona seems to miscompile the stage3 compiler. I got
> 

I opened a bug:

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30770


H.J.

References:
- Re: ICE in gcc/libgcc2.c:566 (gcc trunk)
  - From: Hanno Meyer-Thurow
- Re: ICE in gcc/libgcc2.c:566 (gcc trunk)
  - From: Ian Lance Taylor
- Re: ICE in gcc/libgcc2.c:566 (gcc trunk)
  - From: Hanno Meyer-Thurow
- Re: ICE in gcc/libgcc2.c:566 (gcc trunk)
  - From: H. J. Lu

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]