O0 cleanup_cfg vs debuggability

[James E Wilson]

On Fri, 2004-08-13 at 01:39, Olivier Hainque wrote:
>  I'm not clear why a mere inhibition of the edge forwarding optimization in
>  non-optimizing compilations doesn't seem to be an option.

Yes, I think that would be OK also.  I was just trying to offer more
alternatives, as you suggested that this change might be too drastic. 
If programs suddenly start running slower at -O0, some people will
notice and complain, but I doubt that we really have to worry about this
unless there are very significant performance changes, and that seems
rather unlikely.
-- 
Jim Wilson, GNU Tools Support, http://www.SpecifixInc.com