On Fri, 28 Aug 2026 01:21:43 GMT, Guanqiang Han <[email protected]> wrote:
>> This change has broken a number of tests that involve the use of long >> options: >> - sun/misc/UnsafeMemoryAccessWarnings.java >> - java/lang/reflect/Field/mutateFinals/cli/CommandLineTest.java >> - java/lang/Object/FinalizationOption.java >> - java/foreign/enablenativeaccess/TestEnableNativeAccess.java >> >> I will file a bug, but this change may need to be backed out. >> >> EDIT: I see now these are all tests that were modified by this PR. Filed >> [JDK-8391315](https://bugs.openjdk.org/browse/JDK-8391315) > > Hi @dholmes-ora , Thanks for reporting this. > > I took a closer look, and these messages seem to come from negative test > cases that intentionally pass missing or invalid option values and verify the > resulting error output. The tests themselves should still pass overall, and I > have also verified them locally. > > Could you check whether these tests are actually reported as failed in your > tier 5 run, or whether these are just the expected stderr messages captured > from the negative cases? Thanks. Hello @hgqxjj, > Could you check whether these tests are actually reported as failed in your > tier 5 run, or whether these are just the expected stderr messages captured > from the negative cases? I've now added a comment to JDK-8391315 which explains why these tests are failing. ------------- PR Comment: https://git.openjdk.org/jdk/pull/31928#issuecomment-5448664645
