On Tue, 4 Nov 2025 20:37:26 GMT, Johannes Döbler <[email protected]> wrote:
> is a bit confusing I definitely get your point on this. :) I went in circles setting up the PR for this same reason. However, the benchmark(s) ultimately are testing ArrayList.addAll() in three different scenarios: ArrayList, SingletonSet, and LinkedList. The inner class magic was suggested to simplify the benchmarking by allowing parameterized size on the ArrayList flavor, while truthfully reporting the fixed size of SingletonSet. I'm not sure there's a great solution here, but splitting up benchmarks that all call ArrayList.addAll() isn't intuitive. ------------- PR Review Comment: https://git.openjdk.org/jdk/pull/28116#discussion_r2492047856
