On Wed, 27 Jul 2022 12:30:43 GMT, Aleksey Shipilev <sh...@openjdk.org> wrote:
>>> ...I can take this over, unless you want to do it, Aleksey? >> >> I find it dubious to try and guess what GCs would do with non-strong refs, >> but feel free. Don't reassign the bug yet, just see how messy that would be? > >> > ...I can take this over, unless you want to do it, Aleksey? >> >> I find it dubious to try and guess what GCs would do with non-strong refs, >> but feel free. Don't reassign the bug yet, just see how messy that would be? > > On the other hand, this test is in tier2, so it makes lots of testing with > other GCs not clean. I would like to have this fix in, and *then* do any > followups that might make the test more targeted. > So how do we proceed @shipilev ? I open another issue or are you willing to > accept my changes into your issue? I prefer to commit the simpler (mine) version first, and the follow up with any more sophisticated attempt to fix it. > It then basically tests just that newly constructed SoftReference is not > cleared in the next moment after construction. Yes, but not really. There is still a 100ms sleep and reference processing involved, which somewhat verifies that the cache is not blown away immediately/accidentally. > Who is responsible to select the GC algorithm? CIs routinely test existing suites with different GCs. The PR body contains a sample reproducer how that happens. If you are doing the separate `@run` statements, you need to also check for `@requires vm.gc.XXX`, etc. ------------- PR: https://git.openjdk.org/jdk/pull/9533
