RFR: 8342540: InterfaceCalls micro-benchmark gives misleading results [v2]

[Andrew Haley]

On Fri, 18 Oct 2024 14:05:48 GMT, Aleksey Shipilev <shade at openjdk.org> wrote:

> Is there even a point to do non-randomized test then?

That's a very interesting question, and one that has been occupying me since I discovered this problem a few days ago.

I have no idea how often well-predicted megamorphic calls occur. I can speculate that the "typical" megamorphic case lies somewhere between these extremes, but that is all. It may well be that normal behaviour is chaotic, but I strongly suspect that the cases are unlikely to be equally probable, as they are here. So, it is possible that an utterly unpredictable access pattern is just as unrealistic as a perfectly predictable one.

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

PR Comment: https://git.openjdk.org/jdk/pull/21581#issuecomment-2422827754