On 08/18/2011 10:30 AM, Richard Guenther wrote:
Ok.
Great. Thus I'm going ahead and committing the patch to fix the regression.
It looks like we might want to elimiate abs_hwi alltogether in favor
of absu_hwi?
Indeed, you are totally right. But I'd rather work on that as a followup
patch (maybe with Sebastian' help?) because it doesn't seem completely
trivial: I'm mostly worried by the existing uses of abs_hwi *inside*
hwint.c itself not but those two remaining instances outside of it
(which would be possible to replace immediately).
Paolo.
