harmony-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Pavel Afremov (JIRA)" <j...@apache.org>
Subject [jira] Updated: (HARMONY-5103) [drlvm] Crash in DeleteGlobalRef
Date Fri, 09 Nov 2007 19:10:50 GMT

     [ https://issues.apache.org/jira/browse/HARMONY-5103?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Pavel Afremov updated HARMONY-5103:
-----------------------------------

    Attachment: HARMONY-5103.patch

Fix consist in elimination jni_env usage in DeleteGlobalRef.

> [drlvm] Crash in DeleteGlobalRef
> --------------------------------
>
>                 Key: HARMONY-5103
>                 URL: https://issues.apache.org/jira/browse/HARMONY-5103
>             Project: Harmony
>          Issue Type: Bug
>          Components: DRLVM
>         Environment: x86-64
>            Reporter: Pavel Afremov
>         Attachments: HARMONY-5103.patch
>
>
> During evaluating HARMONY-2772 crash on following stack was reproduced:
> SIGSEGV in VM code.
> Stack trace:
>   0: JNIEnv_External::DeleteGlobalRef(_jobject*) (/nfs/ims/proj/drl/mrt2/users/pnafremo/work/H64/drlvm/vm/include/jni.h:678)
>   1: jthread_vm_detach (/nfs/ims/proj/drl/mrt2/users/pnafremo/work/H64/drlvm/vm/vmcore/src/thread/thread_java_basic.cpp:419)
>   2: jthread_wrapper_start_proc(void*) (/nfs/ims/proj/drl/mrt2/users/pnafremo/work/H64/drlvm/vm/vmcore/src/thread/thread_java_basic.cpp:172)
>   3: start_thread (??:-1)
> <end of stack trace>
> Segmentation fault
> As evaluation shows the source of the crash is in jni_env usage after deallocation in
vm_detach.
> If jni_env, after deallocation but before usage, is nullified in other thread, sigsegv
happen.

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


Mime
View raw message