[External] : Re: Introductions / JDWP JVM Crashes

[Chen Liang]

Hi Adam,
I have marked the bug as new for re-triage by the serviceability engineers: the closing was indeed weird, that we do not usually close a bug because we don't plan to fix it in the short term or low priority.  You can submit a patch for it.

For getting assigned to the JBS: for new contributors, the assignee is usually the sponsor of your patch.  Unfortunately I do not understand native code well enough to sponsor your patch, but I will forward this to serviceability-dev list and see if any engineer there can help.

I think you can create a PR with the title "6726901: JDWP: ReferenceType.GetValues crashes jvm in case non-static fields are passed" now.

P.S. When you reply to someone's reply to the list, remember to reply-all instead of just reply, so the list receives your reply email too.

Regards,
Chen Liang
________________________________
From: Adam Bruce <adam at adambruce.net>
Sent: Sunday, November 17, 2024 9:18 AM
To: Chen Liang <chen.l.liang at oracle.com>
Subject: [External] : Re: Introductions / JDWP JVM Crashes

Hi Chen,

Thanks for the detailed reply. I have found some matching bugs in JBS, however they have been marked as closed with a resolution of Won't Fix. e.g. https://bugs.openjdk.org/browse/JDK-6726901

Would I be allowed to create a PR for the bug? I believe it was marked as Won't Fix due to low priority.

Also, as I am not an Author, is there a process for getting assigned to a ticket in JBS, or shall I proceed with creating a PR anyway?

Cheers,
Adam
On Saturday, November 16th, 2024 at 11:34 PM, Chen Liang <chen.l.liang at oracle.com> wrote:
Hi Adam,
I believe serviceability-dev at openjdk.org is the correct list for jdwp and Java agents.  I know it takes care of agents, and a peek into its archives shows it takes care of jdwp as well.

For patches, you usually first submit a bug through https://bugs.java.com/<https://urldefense.com/v3/__https://bugs.java.com/__;!!ACWV5N9M2RV99hQ!LldezN2kHNtCbBX84HBVCy-BPKkcMrjPDwL3ULIgbnXfi6SBE5EWnuxhyb_paFBCj2cJAjKUz5d0AoBF$>; once you find no duplicate, you can create a new bug. (Note the component will be "core serviceability" which the serviceability-dev list manages.  It takes a day or two for the bug to be triaged and moved from incident JI-9XXXXXX to an official bug JDK-8XXXXXX that you can submit a patch for.  (Alternatively, you can ask on the serviceability-dev list, and contributors with bug system access may create issues for you instead).

Once you have a bug number, follow the process in https://openjdk.org/guide/.   Note that JDK project prefers to review ready-to-merge patches; so you should open a draft PR and move it out of draft if you want to open a PR with prototype work, otherwise the commit or review updates to the PR will spam the mailing lists.  Also if you haven't signed the Oracle Contributor Agreement, the automated process will ask you to, but it will take a few days to process, so best to sign it ahead at https://oca.opensource.oracle.com/.

Regards,
Chen Liang
________________________________
From: discuss <discuss-retn at openjdk.org> on behalf of Adam Bruce <adam at adambruce.net>
Sent: Saturday, November 16, 2024 12:24 PM
To: discuss at openjdk.org <discuss at openjdk.org>
Subject: Introductions / JDWP JVM Crashes

Hi everyone,

My name is Adam Bruce and I'm a software engineer with an interest in JVM
internals.

I'm currently working on a project involving JDWP and have identified
some bugs in the OpenJDK protocol implementation which lead to JVM crashes
due to segmentation faults.

I understand that this is not a mailing list for technical discussions, so
I was wondering if anybody could point me in the right direction for
reporting and discussing this issue (I don't believe there is a mailing list
specifically for JDWP or java agents).

Additionally, it would be great if someone could advise on the best way to
suggest a patch for these bugs as I have developed a fix for the issue locally
and would like to suggest the changes as possible solutions.

Many thanks,
Adam

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/serviceability-dev/attachments/20241118/6608eb25/attachment.htm>