>From Julian
That being said performance of the PIC construct is very good in our
case.
Do you have any quantitative data on your call site performance vs number
of targets?
thx
mark___
mlvm-dev mailing list
mlvm-dev@openjdk.java.net
http://ma
> For your use case do you have the same situation as I do with 99%+ of call
> sites having
> less than 5 targets?
Yes.
We set the PIC<->megamorphic threshold to 5 in Golo, in which case we degrade
to a stable but slower map-based + invoker dispatch.
That being said performance of the PIC c
>From Julian
How is it different from Rémi's construct?
Performance would be the hope.
My position has been that with a decent pic api the jvm would be able to
optimize the
pic to a few test/branch instructions for the large majority of callsites.
For your use case do you have the same situa
How is it different from Rémi's construct?
https://code.google.com/p/jsr292-cookbook/source/browse/trunk/inlining-cache/src/jsr292/cookbook/icache/RT.java
(we use that pattern in Golo)
- Julien
> On 09 Mar 2015, at 18:51, Mark Roos wrote:
>
> I was thinking about a generic pic, easy to use bu
