Re: compiling glibc-2.3.1 with gcc-3.2.1 for pentium >= II

[Rene Rebe <rene dot rebe at gmx dot net>]

Hi.

Yes - thanks. We already found out and a quick fix is:

+/* __i686 is used in assembler code below as normal text within a statement.
+   When optimizing for pentium2 or pentium3, gcc defines __i686 as cpp macro
+   which destroys the code below. Glibc is only using the cpp macro __i686__,
+   so we can savely undefine __i686 here. - Clifford Wolf */
+#undef __i686

Clifford, have you  already sent the mail to glibc-bugs mailing-list?

On: Wed, 8 Jan 2003 12:28:14 -0800,
    Joe Buck <jbuck@synopsys.com> wrote:
> 
> On Jan  7, 2003, Rene Rebe <rene.rebe@gmx.net> wrote:
> > 
> > > It is a gcc issue, since a Athlon optimised gcc-3.2.1 is able to
> > > compile the same preprocessed file ...
> 
> ... meaning that you should look at code that is #ifdefed out on the
> Athlon.  This *is* sysdep.h, after all.
> 
> On Wed, Jan 08, 2003 at 03:35:46PM -0200, Alexandre Oliva wrote:
> > It's very unlikely to be a GCC issue.  You're feeding GCC an assembly
> > source file, and it's the assembler that is complaining about ignored
> > characters.  It could be a problem in the preprocessor, for sure, but
> > then you'd have to save the preprocessed assembly generated by a
> > working and broken preprocessor and compare them for the bug report to
> > be useful.
> 
> Grep for __i686 in the input to GCC, then note that __i686 is a macro.
> 

- René

--  
René Rebe - Europe/Germany/Berlin
e-mail:   rene.rebe@gmx.net, rene@rocklinux.org
web:      www.rocklinux.org, drocklinux.dyndns.org/rene/

Anyone sending unwanted advertising e-mail to this address will be
charged $25 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.