RFR: JDK-8324169 : Remove macOS from ProblemList for CycleDMIMage.java when fix is available for JDK-8312518

Harshitha Onkar honkar at openjdk.org
Thu Jan 18 22:28:32 UTC 2024


On Thu, 18 Jan 2024 21:55:42 GMT, Phil Race <prr at openjdk.org> wrote:

>> CycleDMIMage.java passes on macOS after [JDK-8312518](https://bugs.openjdk.org/browse/JDK-8312518), [PR#17358](https://github.com/openjdk/jdk/pull/17358). It is de-problemlisted for macOS only.
>> 
>> 
>> CI testing looks good.
>
> test/jdk/ProblemList.txt line 436:
> 
>> 434: java/awt/Frame/DisposeParentGC/DisposeParentGC.java 8079786 macosx-all
>> 435: 
>> 436: java/awt/GraphicsDevice/DisplayModes/CycleDMImage.java 7099223,8274106 macosx-all,linux-all,windows-all
> 
> it seems odd to remove macOS from the list when one of the bugs here is specifically against macOS - https://bugs.openjdk.org/browse/JDK-8274106
> If the test still fails on some retina displays, then this needs to remain problem listed.
> If not, then remove the reference to 8274106
> 
> Wait 7099223 is also about macOS.
> So why is this test problem listed on all platforms ?
> If it passes always on macOS we also should close out those bugs.
> Please check it on all platforms and de-list it for all platforms that pass and if it does fail on Windows or Linux, create a bug for that platform's problems.

There are couple of JBS issues related to CycleDMImage.java
[JDK-7099223](https://bugs.openjdk.org/browse/JDK-7099223) - listed for all platforms
[JDK-8312459](https://bugs.openjdk.org/browse/JDK-8312459) - subtask to problemlist on macOS but the JBS number itself is not included in problemlist.
[JDK-8274106](https://bugs.openjdk.org/browse/JDK-8274106) - The one you mentioned earlier.

> Please check it on all platforms and de-list it for all platforms that pass and if it does fail on Windows or Linux, create a bug for that platform's problems.

Checking it again on all platforms.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/17493#discussion_r1458039811


More information about the client-libs-dev mailing list