On Sat, Feb 21, 2015 at 10:38 AM, Keean Schupke <[email protected]> wrote:
> On 21 Feb 2015 13:39, "Matt Oliveri" <[email protected]> wrote:> But I > think William's right that there are better pathological cases > > > for your scheme. Once I get you to admit that the inlining is > > mandatory. > > The inlining does not seem mandatory because you can just specialise the > call instead, which I think is what shap was proposing. > Except that shap's specialization has the property that no legal specialization introduces allocation. We may determine after some explanation that your proposal has this property as well, but I don't see it yet, and Matt is correct that it is a requirement. shap
_______________________________________________ bitc-dev mailing list [email protected] http://www.coyotos.org/mailman/listinfo/bitc-dev
