You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@harmony.apache.org by "Eugene S. Ostrovsky (JIRA)" <ji...@apache.org> on 2007/05/25 17:18:16 UTC
[jira] Commented: (HARMONY-3982) [drlvm][jvmti] JVMTI
GetAllStackTraces, GetThreadListStackTraces functions work incorrectly
[ https://issues.apache.org/jira/browse/HARMONY-3982?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12499114 ]
Eugene S. Ostrovsky commented on HARMONY-3982:
----------------------------------------------
The reason of the failure is in GetStackTrace() jvmti function.
By spec it must return JVMTI_ERROR_ILLEGAL_ARGUMENT (103) for any value of start_depth argument if thread's stack depth is 0.
Finalizer threads often have empty stack because they are native threads attached to vm.
RI contradicts the spec and return success in such conditions.
> [drlvm][jvmti] JVMTI GetAllStackTraces, GetThreadListStackTraces functions work incorrectly
> -------------------------------------------------------------------------------------------
>
> Key: HARMONY-3982
> URL: https://issues.apache.org/jira/browse/HARMONY-3982
> Project: Harmony
> Issue Type: Bug
> Components: DRLVM
> Reporter: Elena Sayapina
> Attachments: H3982-GetStackTrace-for-empty-stack.patch
>
>
> 2 JVMTI tests from DRLVM Validation test suite (http://issues.apache.org/jira/browse/HARMONY-3206) failed on Harmony-r541420:
> It test correct work of GetAllStackTraces and GetThreadListStackTraces functions .
> vm.jvmti.funcs.GetAllStackTraces.GetAllStackTraces0101.GetAllStackTraces0101 output:
> java version "1.5.0"
> pre-alpha : not complete or compatible
> svn = r541420, (May 25 2007), Windows/ia32/msvc 1310, debug build
> http://incubator.apache.org/harmony
>
> check: Agent_OnLoad was called with options = NULL
> check: Agent_OnLoad was called with reserved = NULL
>
> -------------------------------------------------
>
> test GetAllStackTraces0101 is started
> {
> native: GetPhase result = 0 (must be zero)
> native: current phase is 4 (must be 4 (LIVE-phase))
> native: GetMethodName result = 0 (must be zero)
> native: method name is special_method
> native: signature name is ()V
> native: generic name is (null)
> native: GetAllStackTraces result = 103 (must be zero)
> native: stack_info ptr is 00000000 (must be zero)
> native: thread_count is 0
>
> Test of function GetAllStackTraces0101 : failed
>
> } /* test GetAllStackTraces0101 is finished */
>
> TestHarness:RunNegDRL analizeExit(): Check log for message passed
> Status is -1 (expected positive)
> vm.jvmti.funcs.GetThreadListStackTraces.GetThreadListStackTraces0101.GetThreadListStackTraces0101 output:
> java version "1.5.0"
> pre-alpha : not complete or compatible
> svn = r541420, (May 25 2007), Windows/ia32/msvc 1310, debug build
> http://incubator.apache.org/harmony
>
> check: Agent_OnLoad was called with options = NULL
> check: Agent_OnLoad was called with reserved = NULL
>
> -------------------------------------------------
>
> test GetThreadListStackTraces0101 is started
> {
> native: GetPhase result = 0 (must be zero)
> native: current phase is 4 (must be 4 (LIVE-phase))
> native: GetMethodName result = 0 (must be zero)
> native: method name is special_method
> native: signature name is ()V
> native: generic name is (null)
> native: GetAllThreads result = 0 (must be zero)
> native: GetAllThreads tc = 4 (must >= zero)
> native: GetThreadListStackTraces result = 103 (must be zero)
>
> Test of function GetThreadListStackTraces0101 : failed
>
> } /* test GetThreadListStackTraces0101 is finished */
>
> TestHarness:RunNegDRL analizeExit(): Check log for message passed
> Status is -1 (expected positive)
>
> Note that these tests passed on Harmony-r540603.
> It looks like regression of HARMONY-643 caused by HARMONY-3698 fix.
> Test sources to reproduce the failure can be found in attachment to HARMONY-643.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.