On Wed, 27 May 2026 16:20:00 GMT, Paul Sandoz <[email protected]> wrote:
>>> Hi @PaulSandoz , As per the following comment fix for JDK-8373574 is >>> included in this patch. [#28002 >>> (comment)](https://github.com/openjdk/jdk/pull/28002#issuecomment-3661203155) >>> >>> Patch has been regressed by @XiaohongGong on AARCH64 >> >> Great! would you mind closing down >> [JDK-8373574](https://bugs.openjdk.org/browse/JDK-8373574) as duplicate of >> the issue for this PR or update 8373574 to focus on a regression test if one >> is not added to this PR? > >> @PaulSandoz @jatin-bhateja I saw that there is already a PR for the template >> fuzzer, so I'd like to review that one too, so we can integrate the two >> things in close succession. As we have seen with the rest of the Vector API >> before: fuzzing revealed a good number of bugs that our hand-written tests >> did not cover. Integrating tests later means we discover bugs later, and >> that puts more pressure on the compiler team - I'd like to avoid that ;) >> >> #30997 > > Understood. We are getting very close to RDP1, after we still have time to > add more tests but not features. > > @jatin-bhateja is #30997 ready for testing with this PR? > @PaulSandoz @jatin-bhateja Is the goal really to push this before RDP1? And > the template testing only after? The template fuzzer will take a few weeks to > find hits. So then we basically have no time to fix before the release. I > don't want to be responsible for those bugs to end up on our compiler bug > dashboard. > > I would prefer if we could just push it after the fork/branch, add the > template fuzzer and give it some months to bake, and have enough time to fix > follow-up bugs. My hope is, with sufficiently good preliminary test results, that we might straddle RDP1, with the fuzzer going in quickly afterwards. Sherman is running tests with the two PRs, but it sounds like it will not be sufficient especially if you want to run the fuzzer in various configurations that stress the compiler? If that is the case then I agree with you, the responsible thing to do is integrate after RDP1, ideally soon afterwards to increase the soak time. @jatin-bhateja we need to ensure this PR does not cause quality issues with the release *and* additional work for Emanuel and other compiler engineers having to manage them later in the release cycle - as we get closer to the release it gets more expensive it is to fix them. ------------- PR Comment: https://git.openjdk.org/jdk/pull/28002#issuecomment-4564898041
