RFR: 8349146: [REDO] Implement a better allocator for downcalls [v8]

[Maurizio Cimadamore]

On Fri, 2 May 2025 14:12:42 GMT, Per Minborg <pminborg at openjdk.org> wrote:

> The performance before this PR can be seen in the "confined" benchmarks above. In those benchmarks, a regular `Arena.ofConfined` is created upon every invocation.

Yes - but I was referring to the performance of the native call -- not that of the allocation per se (e.g. `CallOverheadByValue`). If I'm not mistaken the numbers of that benchmark all refer to new code (well, except when we pass by pointer, in which case there's no allocation -- both before and after this PR).

E.g. it would be useful to also see how the by-value call has improved before/after this PR, in addition to looking at the difference between byValue/byRef in this PR.

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

PR Comment: https://git.openjdk.org/jdk/pull/24829#issuecomment-2847747932