On Mon, Nov 11, 2013 at 12:26:09PM +0100, Richard Biener wrote:
> The question is whether you for example want to handle
>
> a_2 = 1 + 0;
>
> at RTL expansion time? I'd say it's better to have
I think we already handle that just fine, there are tons of various
simplify_gen_* calls during expansion, and we know there the mode etc.
Just Joern hit a place which wasn't prepared to handle it properly,
so either we handle it as you are suggesting by forcing one of the
constants into a register, or we simplify the comparison and if it
simplifies into a constant, we transform it to something simpler.
Jakub
