RFR: 8263158: [TESTBUG] FAILED: runtime/cds/appcds/OldClassTest.java

David Holmes david.holmes at oracle.com
Mon Mar 8 20:44:18 UTC 2021


On 9/03/2021 5:52 am, Daniel D.Daugherty wrote:
> On Mon, 8 Mar 2021 19:38:58 GMT, Ioi Lam <iklam at openjdk.org> wrote:
> 
>>> Thanks @iklam for the investigation, it makes sense.
>>>
>>> As far as [I know](https://openjdk.java.net/jtreg/faq.html#4.29), `@run driver` is the same as `@run main` with the exception that any VM options specified on the command line will not be used when running the specified class. That's why I did this change.
>>>
>>> AgentVM and OtherVM are how jtreg starts the JVM(a brand new? or retrieve from the pool?), they should not affect this test. i.e. `@run main xxx` should also work for this test case.
>>
>> I think this qualifies as a "trivial change" under the HotSpot project rules. The changes can be integrated with just one reviewer.
> 
> @iklam - It seems that @dholmes-ora has some concerns about this fix. We might
> want to hold off on integrating until he has a chance to read your comments in the
> bug report.

I am okay with proceeding with the current fix at this time, but I'm 
still not really clear on where the fault lies. If an @module tag 
implies the test must run in othervm mode then it seems to me that jtreg 
should be doing something about ensuring that happens - either @driver 
should work regardless of jtreg general mode, or else it should trigger 
an error from jtreg.

Thanks,
David

> -------------
> 
> PR: https://git.openjdk.java.net/jdk/pull/2866
> 


More information about the hotspot-runtime-dev mailing list