On Fri, 19 Jul 2024 20:32:23 GMT, Bradford Wetmore <wetm...@openjdk.org> wrote:
> I understand that the main purpose is to provide performance numbers and no > one will be watching it, but if it's not taking too much time, it's nice to > have the additional test for ensuring nothing went wrong when exercising the > code. Even if someone was watching the report, there is no baseline for comparison that would highlight a lower performance in the current code. So, the only criteria for this test to pass is that it does not timeout. In that case, it can be automatic rather than manual. ------------- PR Comment: https://git.openjdk.org/jdk/pull/20135#issuecomment-2247711960
