-
Notifications
You must be signed in to change notification settings - Fork 729
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
JDK19 HCRLateAttachWorkload_previewEnabled_1_FAILED StackOverflowError: operating system stack overflow #15250
Comments
@JasonFengJ9 is this still failing |
The test has been excluded, launched an internal grinder
Similar errors were reported on hotspot JDK19 as well HCRLateAttachWorkload_previewEnabled_0 and HCRLateAttachWorkload_previewEnabled_1 failing with StackOverflow Errors |
The
RI JDK19 passed 10x internal grinder
FYI @tajila |
is it all platforms? |
Actually, this makes me think it may be a jit issue
@JasonFengJ9 Can you see if this reproduces with -Xint |
Launched 10x -Xint internal grinder |
Tried
@llxia @Mesbah-Alam any insights on why the test won't take the grinder option or follow the mode specified? |
From the console output, Mode100 did not get passed in:
This issue affects all system tests with Mode. It seems no Modes were passed in the system tests for the past month. The issue was introduced between Sep 18 (good build) - Sep 25 (bad build).
I suspect https://github.com/adoptium/aqa-tests/blob/d1c960685d0e4f043a11675608048eb7bd682f60/system/system.mk#L63 is incorrect. I think it is related to PR adoptium/aqa-tests#3972 |
Related: eclipse-openj9/openj9#15250 Signed-off-by: Lan Xia <[email protected]>
Related: eclipse-openj9/openj9#15250 Signed-off-by: Lan Xia <[email protected]> Signed-off-by: Lan Xia <[email protected]>
Related: eclipse-openj9/openj9#15250 Signed-off-by: Lan Xia <[email protected]> Signed-off-by: Lan Xia <[email protected]>
Related: eclipse-openj9/openj9#15250 Signed-off-by: Lan Xia <[email protected]> Signed-off-by: Lan Xia <[email protected]> Signed-off-by: Lan Xia <[email protected]>
@JasonFengJ9 the above issue is resolved via adoptium/aqa-tests#4052. Please try to rebase your branch and try again. Thanks |
|
I added https://hyc-runtimes-jenkins.swg-devops.com/job/Grinder/28807/tapResults/ |
With -Xtrace:trigger=method{java/lang/StackOverflowError,sysdump}
Result file |
I see something very strange here: The bytecodes for String::length() is:
but when I look at the runtime version its
So there are some extra bytecodes at the start |
The class is retransformed, the original bytecodes before the transformation is correct
|
I dont know much about this test, but if that transformation is intended then the StackOverflow makes sense. |
Based on the README, https://github.com/adoptium/aqa-systemtest/blob/0eb9f57f198fe194c070295c261e8bafeaed08aa/openjdk.test.debugging/docs/HCR_README.md The test randomly adds a method to String API, I think in this particular case there is nothing we can do about it because it is infinitely recursive. I think we can close and re-enable the test, with a different seed value we shouldn't see the failure. |
@JasonFengJ9 can you please unexclude the test |
|
The test is failing on all platforms, https://openj9-jenkins.osuosl.org/job/Pipeline-Build-Test-JDK19/194/ Example failure,
|
Created adoptium/aqa-tests#4121 |
|
Failure link
From an internal build
job/Test_openjdknext_j9_extended.system_aarch64_linux_Personal/1/tapResults/
(cent7-aarch64-6
):Rerun in Grinder - Change TARGET to run only the failed test targets.
Optional info
Failure output (captured from console output)
The text was updated successfully, but these errors were encountered: