RFR: 8259627: Potential memory leaks in JVMTI after JDK-8227745

[Richard Reingruber]

On Wed, 13 Jan 2021 08:42:59 GMT, Thomas Stuefe <stuefe at openjdk.org> wrote:

>> This change eliminates memory leaks in the JVMTI implementation reported by SonarCloud.
>> 
>> The leaks occur when the Java heap is exhausted.
>
> Looks good.

> 
> 
> > @shipilev thanks for doing the analysis and reporting the issues. I found 2 leaks. Do you see more that could be related to JDK-8227745 ([40f847e](https://github.com/openjdk/jdk/commit/40f847e2))?
> > I wanted to do a SonarCloud scan myself but was uncertain about the requested permissions, me using the service for work, etc. Would it be possible to publish the SC report?
> 
> I think those are only two instances. I am only aware of [this one](https://sonarcloud.io/project/issues?fileUuids=AXaE0amt8L9hkQskFrwX&id=jdk&resolved=false&types=BUG). I meant to set up my own some time later.

Excellent! I googled for public scans but failed. Now I see that scans are listed on the "Explore" page of SonarCloud. Cool the tool found the leaks!

Thanks for the review also.

> 
> 
> Looks good.

Thanks for looking!

-------------

PR: https://git.openjdk.java.net/jdk/pull/2055