openjdk team mailing list archive

[Erich Schubert <erich@xxxxxxxxxx>]

Hi,
I also just noticed another issue with the accessibility functionality.
ELKI (which uses Apache Batik), will not properly terminate when this
is enabled.
If the AtkWrapper is disabled, it quits just fine.

The only thread that involves non-OpenJDK code in the stack is this:
"Batik CleanerThread" #21 daemon prio=5 os_prio=0
tid=0x00007f2a44131800 nid=0x3dd1 in Object.wait()
[0x00007f2a651b5000]
   java.lang.Thread.State: WAITING (on object monitor)
    at java.lang.Object.wait(Native Method)
    - waiting on <0x00000000870fd3e0> (a java.lang.ref.ReferenceQueue$Lock)
    at java.lang.ref.ReferenceQueue.remove(ReferenceQueue.java:143)
    - locked <0x00000000870fd3e0> (a java.lang.ref.ReferenceQueue$Lock)
    at java.lang.ref.ReferenceQueue.remove(ReferenceQueue.java:164)
    at org.apache.batik.util.CleanerThread.run(Unknown Source)

If I disable AtkWrapper, everything is fine.
I haven't verified with Batik example if this also happens there.

Best regards,
Erich

On Mon, Jul 18, 2016 at 1:47 PM, Sebastian Humenda <shumenda@xxxxxx> wrote:
> found 787955 8u91-b14-3
> stop
>
> Erich Schubert schrieb am 17.07.2016, 11:30 +0200:
>>This seems to be fixed for me. I could not reproduce anymore.
> I'm still experiencing the same issue. I've uploaded a core dump on [^1]. It is
> actually enough to start the application, in this case MediathekView, alt+tab
> out of the window and into it again. After a short while the application
> crashes. The crash did not produce a log file though. If there's anything I can
> do to help, please let me know.
>
> Thanks
> Sebastian
> [^1]: http://wwwpub.zih.tu-dresden.de/~s7369555/mdv-core-8u91-b14-3.xz