invoke* on a MethodHandle

[Jeroen Frijters]

Apologies for top-posting, I'm using webmail at the moment.

My two cents as a non-typical VM (IKVM.NET) implementer are that I don't expect any difficulty implementing this and also that it makes a lot of sense for this to work.

Regards,
Jeroen

________________________________________
From: mlvm-dev-bounces at openjdk.java.net [mlvm-dev-bounces at openjdk.java.net] On Behalf Of John Rose [John.Rose at Sun.COM]
Sent: Saturday, February 07, 2009 11:49 PM
To: Rémi Forax
Cc: Da Vinci Machine Project
Subject: Re: invoke* on a MethodHandle

On Feb 5, 2009, at 2:17 AM, Rémi Forax wrote:

> It's exactly like the way ".class" is/was compile(d) by javac.
> Until 1.5, ".class" was emulated by generating the same private method
> on all toplevel class.
> If we don't have LDC MethodHandle, we (all language developers)
> will rely on a private method that create a MethodType and
> call a MethodHandles.find*.

Yes, I agree.  The open question is whether the idea is sufficiently
"natural" to the various implementations of JVM to make it (a) a
simple addition instead of (b) a big implementation problem.  For
HotSpot it is more like (a), because we have to convert MHs down to
methodOops anyway for other reasons.  I conjecture that is typical of
JVMs.

-- John
_______________________________________________
mlvm-dev mailing list
mlvm-dev at openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev