On Mon, Jun 17, 2013 at 01:44:25PM -0400, Jason Merrill wrote:
> On 06/17/2013 12:54 PM, Joseph S. Myers wrote:
> >I suppose it's OK to fix the regression, though really we should be
> >eliminating these early convert_to_* optimizations (optimizing later on
> >GIMPLE if possible) rather than adding to them.
>
> My thought as well. How hard is it to fix this in gimple fold?
PRs for it are open for almost 3 years now, I think Kai has some patch, but
it hasn't been posted yet, so it is unclear if it would make 4.9.
In any case, it isn't probably something that can be backported to release
branches, while perhaps this simple convert.c change is.
Once we have it in GIMPLE, sure, the change in convert.c can be happily
reverted and also those optimizations removed from there too.
Jakub
