Re: Exception Handling and _NS_COUNT_HACK

[Richard Frith-Macdonald]

On 29 Oct 2008, at 03:25, Samantha Rahn wrote:

> On Tue, Oct 28, 2008 at 6:25 AM, Richard Frith-Macdonald
> <richard@tiptree.demon.co.uk> wrote:
> > Maybe the simplest thing to do right now would be to add a  
> > configure option
> > to disable stack traces, so that anyone whose gcc has this bug in  
> > stackframe
> > checking can stop it being used at all.
>
> Great idea -- would solve my problem.

Try the current code in svn ... I changed the behavior of the  
GNUSTEP_STACK_TRACE environment variable slightly so that, as well as  
turning full traces on, it can be used to turn off the generation of  
the return addresses by setting it to NO.
This seems a bit ore flexible than a configure-time option.

However, turning off this feature seems like a poor/temporary fix, and  
it would still be good to figure out why the signal handler seems not  
to be getting called ... a failure in the signal handling mechanism  
could have repercussions anywhere else that a signal handler is called.