Compile on Severn, runs on RH9, segfaults on Severn

[Michael K. Johnson]

On Wed, Sep 10, 2003 at 01:13:25PM +0200, Julio Sanchez wrote:
> On Tue, 09 Sep 2003 16:41:44 +0200, Julio Sanchez wrote:
> 
> > I checked that too.  No, it is being set correctly and makes no difference
> > on the result.
> 
> What I have done is to downgrade to kernel-2.4.20-20.9 and both examples
> run correctly now.

Can you please try kernel-2.4.22-1.2040.nptl from RHN?  There have
been several fixes to exec-shield there.  I know turning off exec-shield
didn't fix it, but it's still possible that it is the culprit.

> So it is the kernel or the kernel triggers it elsewhere (glibc?).
> 
> Do I report it in bugzilla under kernel?  Is there anything I can do it to
> diagnose it better?  Transcode is too large a thing that depends on too
> many libraries to be sure.

Is it linking against any object files that you neither recompiled
yourself nor obtained as an integral part of Severn?  (I.e. did 
you get any of those "too many libraries" in object form?)

michaelkjohnson

 "He that composes himself is wiser than he that composes a book."
 Linux Application Development                     -- Ben Franklin
 http://people.redhat.com/johnsonm/lad/