harmony-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Salikh Zakirov (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HARMONY-3266) [drlvm][jit][opt] shutdown.TestNativeAllocation fails on debug OPT
Date Thu, 01 Mar 2007 15:45:50 GMT

    [ https://issues.apache.org/jira/browse/HARMONY-3266?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12476970

Salikh Zakirov commented on HARMONY-3266:

The test TestNativeAllocation is inherently incorrect. It does not passes on any of the competition
JREs, and is heavily relying on the current DRLVM thread stopping algorithm.
(which boils down to throwing ThreadDeath from each encountered safepoint, and thus is also
relying on a safepoint distribution in jitted code. This is likely cause of regression on
Jitrino only).

The test makes way too strong assumptions on what code is executed on the VM shutdown. It
should not make any.

> [drlvm][jit][opt] shutdown.TestNativeAllocation fails on debug OPT
> ------------------------------------------------------------------
>                 Key: HARMONY-3266
>                 URL: https://issues.apache.org/jira/browse/HARMONY-3266
>             Project: Harmony
>          Issue Type: Bug
>          Components: DRLVM
>         Environment: Any debug cfg
>            Reporter: Alexey Varlamov
>            Priority: Minor
> The debug build of Jitrino fails the smoke test shutdown.TestNativeAllocation (release
build PASSES). To reproduce:
> working_vm/build> ./lnx_ia32_gcc_debug/deploy/jdk/jre/bin/java -cp ./lnx_ia32_gcc_debug/semis/vm/_smoke.tests/classes
-Djava.library.path=./lnx_ia32_gcc_debug/semis/vm/_smoke.tests/natives/shutdown/ -Xem:opt
> Seems that finally block is not handled correctly upon shutdown.

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message