-
Notifications
You must be signed in to change notification settings - Fork 733
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
OpenJDK java/util/Map/InPlaceOpsCollisions NullPointerException String.charAt #17922
Comments
Looks like a dup of #17774 |
Reproduced 3 times in a 30x Grinder job. |
3/30 is frequent enough to be a blocker. |
I haven't been able to reproduce the failure locally, running the test more than 60 times. |
I ran a 30x Grinder job with the latest nightly build from https://openj9-jenkins.osuosl.org/job/Build_JDK11_aarch64_linux_Nightly/498/, and got no failures. |
Another 30x Grinder job was also successful. |
Failure is highly intermittent, and hasn't been reproduced in a couple of weeks. I'll move it out to 0.43 for now. |
https://openj9-jenkins.osuosl.org/job/Test_openjdk11_j9_sanity.openjdk_aarch64_linux_OpenJDK11_testList_1/7/
java/util/stream/boottest/java.base/java/util/stream/SpinedBufferTest.java
|
I ran a 20x jdk_util_0 Grinder job, and got no failures. |
Another 20x Grinder job did not fail, either. |
I was able to reproduce it and get dumps with -Xdump option. https://hyc-runtimes-jenkins.swg-devops.com/job/Grinder/35030/
Unfortunately, jitdump was not available.
|
I got a jitdump generated on NullPointerException. https://hyc-runtimes-jenkins.swg-devops.com/job/Grinder_iteration_2/2478/ NullPointerException was thrown when loading arrayLength of
The
In the optimization phase, there was an
|
@hzongaro Please take a look at @Akira1Saitoh's comment above. It looks like an issue in optimization. |
Thank you for the investigation, @Akira1Saitoh and @knn-k - I will take a look. |
@jmesyou, may I ask you to take a look at this problem? I've been trying to reproduce the failure in Grinder runs with extra tracing of General Store Sinking in the jitdump, using the following
However, there's an internal grinder run that's currently in progress that has shown at least one |
FYI - on any future Grinder reruns, you might want to used the
|
It seems that I was mistaken about the usage of the
However, when support for the
Thus, my use of |
https://openj9-jenkins.osuosl.org/job/Test_openjdk11_j9_sanity.openjdk_aarch64_linux_OMR_testList_2/33/ https://openj9-jenkins.osuosl.org/job/Test_openjdk11_j9_sanity.openjdk_aarch64_linux_OMR_testList_1/34/ |
Moving to the backlog for now, as it's a rather intermittent failure. @jmesyou will continue investigating. |
Something similar to this (failures in
though it's a different platform. |
The IL that results after General Store Sinking that @Akira1Saitoh described above looks very similar to the IL in #17515 (comment). These are likely duplicates of one another. |
As I mentioned in #17515 (comment), the fix needs some further work. I'll have to move this out to 0.51. |
Oops! Missed that this was already assigned to "Future." |
It seems we haven't seen a recurrence of this problem on aarch64 in quite a while. I was unable to reproduce it with recent builds. The only problem that had been identified appeared to be the same problem reported in issue #17515, which has been fixed by OMR pull request eclipse-omr/omr#7573, so I will close this issue. |
https://openj9-jenkins.osuosl.org/job/Test_openjdk11_j9_sanity.openjdk_aarch64_linux_Nightly_testList_1/26
jdk_util_0
java/util/Map/InPlaceOpsCollisions.java
The text was updated successfully, but these errors were encountered: