RFR: 8331658: secondary_super_cache does not scale well: C1 [v2]

[Vladimir Ivanov]

On Thu, 6 Jun 2024 09:09:34 GMT, Andrew Haley <aph at openjdk.org> wrote:

> The end goal, I hope, is to remove the secondary_super_cache field altogether because it does nothing useful. Apart from anything else, we'd remove some cruft in the VM. So, C1, interpreter, and runtime is next.

Sure, I fully support complete removal of SSC. Thanks a lot for taking care of it! It looks like we are already on the same page. 

I just tried to make a point that the rest of the JVM (outside C2) seems much less bothered about performance of subtype checks. It doesn't mean these's no need to optimize those scenarios, but as part of SSC removal I'd prefer to prioritize simplicity of the implementation over peak performance.

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

PR Comment: https://git.openjdk.org/jdk/pull/19426#issuecomment-2153524330