RFR: 8275775: VM.metaspace prints flag 'f' for classes that have non-trivial finalize()

[David Holmes]

On 2/11/2021 5:23 pm, Ioi Lam wrote:
> On Tue, 2 Nov 2021 06:45:55 GMT, Thomas Stuefe <stuefe at openjdk.org> wrote:
> 
>>> Thomas > "My feeling is no, since this adds a new command, so there can be no backward compat issues."
>>> :)
>>
>> Hah, that's embarrassing :-)
>>
>> I actually changed my mind with the asynchronous -Xlog discussions recently (https://bugs.openjdk.java.net/browse/JDK-8229517). CSR seemed to work well here, a good way to get more eyes for the user-facing command syntax. But I know that CSR is not the perfect process either (slow, arduous, and arguably new commands are not a backward compatibility issue).
> 
> I don't know about jcmd, but in general new interfaces should be discussed in CSR. We don't want to get stuck with an interface that's difficult to evolve (e.g., have forward compatibility issues).
> 
> https://wiki.openjdk.java.net/display/csr/CSR+FAQs

jcmd has been considered a diagnostic interface and therefore outside of 
the CSR purview. But that's not to say that we don't need to think 
carefully about any changes, whether new commands, or changes to 
existing ones.

David

> -------------
> 
> PR: https://git.openjdk.java.net/jdk/pull/6075
>