On Sun, 28 Feb 2021 07:20:19 GMT, Aleksey Shipilev <sh...@openjdk.org> wrote:
>> Attila Szegedi has refreshed the contents of this pull request, and previous >> commits have been removed. The incremental views will show differences >> compared to the previous content of the PR. > > test/jdk/jdk/dynalink/TypeConverterFactoryMemoryLeakTest.java line 61: > >> 59: // With explicit GC calls succeeds in 11-12 iterations depending on >> GC used. >> 60: // 1000 should be a safe upper limit after which we can consider it >> failed. >> 61: private static final int MAX_ITERATIONS = 1000; > > Ah, here is a test trivia. Some configurations do not have either ZGC or > Shenandoah. So you need to check GC availabilty before adding `@run`. For > consistency, checking every GC availability is even better. Usually done by > splitting the `@test` blocks, and adding `@requires` tags: > https://github.com/openjdk/jdk/blob/master/test/hotspot/jtreg/serviceability/dcmd/gc/HeapDumpCompressedTest.java#L37-L107 Thanks, that's an excellent suggestion! Implemented it now. ------------- PR: https://git.openjdk.java.net/jdk/pull/2617
