PATCH: PR middle-end/22284: [4.1 Regression] ia64 exception handling broken

H. J. Lu hjl@lucon.org
Tue Jul 26 00:46:00 GMT 2005


On Mon, Jul 25, 2005 at 05:13:20PM -0700, Geoff Keating wrote:
> 
> Hi HJ, Paul,
> 
> From <http://gcc.gnu.org/ml/gcc-patches/2005-07/msg01510.html>:
> >2005-07-22  H.J. Lu  <hongjiu.lu@intel.com>
> >
> >        PR middle-end/22284
> >        * libsupc++/eh_personality.cc (PERSONALITY_FUNCTION): Revert
> >        the change to info.ttype_base.
> 
> It seems possible that, although perhaps correct for IA64, applying  
> this patch will break the ARM EABI.
> 
> HJ, do you know if it breaks ARM?
> 
> Paul, could you look at this patch and at the part of your patch  
> which it's reversing, and determine if that change is really needed  
> for proper behaviour on ARM?
> 
> If it does break ARM to remove the change, then the change should  
> probably be conditional on __ARM_EABI_UNWINDER__ or something.
> 

info.ttype_base is only used by get_ttype_entry, which isn't used when
__ARM_EABI_UNWINDER__ is defined. I don't know why Paul made the change
for ARM EABI. I will be very surprised if it makes a difference for
ARM EABI.


H.J.



More information about the Gcc-patches mailing list