RFR: 8253049: Enhance itable_stub for AArch64 and x86_64
Vladimir Ivanov
vladimir.x.ivanov at oracle.com
Tue Sep 15 15:36:38 UTC 2020
>> And speaking of the overall approach (as it is implemented now), IMO
>> increased complexity doesn't worth it. If interface calls become a
>> bottleneck, the problem lies not in itable stub, but the overall design
>> which requires linear scan over itables. It's better to put the effort
>> there than micro-optimizing the stub.
>
> Indeed. When I first came to HotSpot after working on GCJ for years
> I was very surprised to see a linear scan used for interface dispatch.
FTR ErikÖ has been looking into rewriting virtual dispatch logic:
http://openjdk.java.net/jeps/8221828
Best regards,
Vladimir Ivanov
>
> The code improvements look to be fairly minor.
>
More information about the hotspot-compiler-dev
mailing list