Integrated: 8307783: runtime/reflect/ReflectOutOfMemoryError.java timed out
Daniel D. Daugherty
dcubed at openjdk.org
Sat May 20 16:43:57 UTC 2023
On Sat, 20 May 2023 16:32:10 GMT, Leonid Mesnik <lmesnik at openjdk.org> wrote:
>> A trivial fix to increase a timeout:
>> [JDK-8307783](https://bugs.openjdk.org/browse/JDK-8307783) runtime/reflect/ReflectOutOfMemoryError.java timed out
>>
>> Trivial fixes to ProblemList tests:
>> [JDK-8308468](https://bugs.openjdk.org/browse/JDK-8308468) ProblemList containers/docker/TestMemoryAwareness.java on linux-x64
>> [JDK-8308470](https://bugs.openjdk.org/browse/JDK-8308470) ProblemList javax/management/remote/mandatory/connection/BrokenConnectionTest.java on linux-all
>> [JDK-8308471](https://bugs.openjdk.org/browse/JDK-8308471) ProblemList javax/management/remote/mandatory/loading/MissingClassTest.java on windows-x64 w/ loom
>> [JDK-8308472](https://bugs.openjdk.org/browse/JDK-8308472) ProblemList javax/management/remote/mandatory/loading/RMIDownloadTest.java on windows-x64 w/ loom
>> [JDK-8308473](https://bugs.openjdk.org/browse/JDK-8308473) ProblemList java/lang/instrument/NativeMethodPrefixAgent.java with loom
>
> One small proposal inline.
@lmesnik - Thanks for fast review! Especially for a Saturday AM...
> test/hotspot/jtreg/runtime/reflect/ReflectOutOfMemoryError.java line 28:
>
>> 26: * @bug 8297977
>> 27: * @summary Test that throwing OOM from reflected method gets InvocationTargetException
>> 28: * @run main/othervm/timeout=150 ReflectOutOfMemoryError
>
> Doesn't it make a sense to exclude this test from tier1, if it might took more than 2 minutes to complete?
It's not that simple. Here's my comment from the bug:
https://bugs.openjdk.org/browse/JDK-8307783?focusedCommentId=14582995&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-14582995
In jdk-21+24-2010 (which has finished Tier[1-7]), there are 79 passing runs of
this test and the times range from 31s -> 7m 37s. The slowest is on linux-arch64,
but the fastest linux-aarch64 is 44s so it's not simply that linux-aarch64 is slow.
-------------
PR Comment: https://git.openjdk.org/jdk/pull/14070#issuecomment-1555947880
PR Review Comment: https://git.openjdk.org/jdk/pull/14070#discussion_r1199632653
More information about the hotspot-runtime-dev
mailing list