[foreign-jextract] RFR: 8261906: Improve jextract support for virtual functions [v4]

[Maurizio Cimadamore]

On Wed, 3 Mar 2021 13:54:48 GMT, Jorn Vernee <jvernee at openjdk.org> wrote:

> Maybe we can just do both? Otherwise, I think out of these 2 I prefer the first, since it's the more primitive variant of the 2 (i.e. no need to create an intermediate wrapper instance).

I have no strong preference, but the currently implemented option does have the advantage of not requiring an intermediate allocation for the wrapper (although not 100% how that matters in practice: I'd expect the object to be escape-analyzed if used only once).

Having two is problematic in terms of adding the required overloads (e.g. we'd need to make up a new name for both options, which might be hard to decipher).

It also occurs to me, that, with the currently implemented option, one could still do this:

func f = header::f;
E.g. if a functional interface is expected, it should be possible to instantiate it using a method reference.

-------------

PR: https://git.openjdk.java.net/panama-foreign/pull/456