This is the mail archive of the libc-alpha@sources.redhat.com mailing list for the glibc project.


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

Re: gcc bug impacts glibc cvs


On Mon, Mar 12, 2001 at 02:18:07PM -0800, Ulrich Drepper wrote:
> Daniel Jacobowitz <dmj+@andrew.cmu.edu> writes:
> 
> > extern void *xyzzy __attribute__ ((__weak__));
> > int
> > main (void)
> > {
> >   return &xyzzy ? ((int)xyzzy) : 0;
> > }
> 
> This works just fine on x86 as well.  In fact I have never seen this
> problem on x86.

If you look at the generated assembler on both, you'll see why - I'm
betting that short forward branches are cheap on x86, while even
correctly predicted they're a bit harsher on PowerPC, so we go out of
our way to avoid them.

> > While it's certainly a bit too fragile of a test, could we at least use
> > this to autoconf for the bug until someone thinks of a better one?
> 
> If this is what exposes the problem, sure.

Shall I make up the patch, then?  Or you, Franz?

-- 
Daniel Jacobowitz                           Debian GNU/Linux Developer
Monta Vista Software                              Debian Security Team
                         "I am croutons!"


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