Mailing-List: contact commits-help@harmony.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@harmony.apache.org
Message-ID: <33519411.1198686763038.JavaMail.jira@brutus>
Date: Wed, 26 Dec 2007 08:32:43 -0800 (PST)
From: "Gregory Shimansky (JIRA)" <jira@apache.org>
To: commits@harmony.apache.org
Subject: [jira] Commented: (HARMONY-4937) [drlvm][thread]
 suspend_enable()/suspend_disable() race condition detector
In-Reply-To: <12486346.1192191230602.JavaMail.jira@brutus>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/HARMONY-4937?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12554426 ] 

Gregory Shimansky commented on HARMONY-4937:
--------------------------------------------

With this patch C-unit thread tests don't build on Linux with the following error:

       [cc] ../obj/thread_unit_test_utils.o: In function `jthread_get_native_thread':
       [cc] /nfs/ims/proj/drl/mrt2/users/gregory/suse2/trunk/working_vm/vm/vmcore/include/thread_manager.h:382: undefined reference to `jthread_get_tm_data'
       [cc] ../obj/thread_unit_test_utils.o: In function `jthread_get_vm_thread_from_java':
       [cc] /nfs/ims/proj/drl/mrt2/users/gregory/suse2/trunk/working_vm/vm/vmcore/include/thread_manager.h:393: undefined reference to `jthread_get_tm_data'
       [cc] collect2: ld returned 1 exit status


> [drlvm][thread] suspend_enable()/suspend_disable() race condition detector
> --------------------------------------------------------------------------
>
>                 Key: HARMONY-4937
>                 URL: https://issues.apache.org/jira/browse/HARMONY-4937
>             Project: Harmony
>          Issue Type: Sub-task
>          Components: DRLVM
>            Reporter: Pavel Rebriy
>            Assignee: Gregory Shimansky
>         Attachments: H4937.patch, H4937_test.patch, H4937_test002.patch, test_native_exp_sync.c, test_native_exp_sync.c
>
>
> This is an experimental issue. Its goal is to define race condition in present suspend_disable/suspend_enable mechanism and to develop a new one.
> The issue is implemented as cunit test which based on DRLVM. Test works in the following way:
> Main thread (see function test_exp_sync()) starts 1 tested thread (see function test_thread_proc()), several GC requester threads (see function test_gc_request_thread_proc()) and several JAVA requester threads (see function test_java_request_thread_proc()). Tested thread makes suspend_disable()/suspend_enable() in a loop. Requester threads ask GC or JAVA suspend()/resume() tested thread properly. Main thread controls the execution.
> By default test is base on hythread functionality. If define TESTED_RUN is changed to 1, tested functionality will be used. Tested functionality is base on algorithm which describes in HARMONY-4908.
> To check the test need to complete the following steps:
> 1. Apply the patch:
>      $ cd <harmony>/working_vm
>      $ patch -p1 -i test.patch
> 2. Build DRLVM:
>      $ cd build
>      $ ./build.bat
> 3. Build and run cunit test:
>      $ ./build.bat cunit.test
> The result of running will be in <harmony>/working_vm/build/win_ia32_msvc_debug/tests/cunit.tests/results.

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