Mailing-List: contact commits-help@harmony.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@harmony.apache.org
Message-ID: <1610201226.1207314206318.JavaMail.jira@brutus>
Date: Fri, 4 Apr 2008 06:03:26 -0700 (PDT)
From: "Alexei Fedotov (JIRA)" <jira@apache.org>
To: commits@harmony.apache.org
Subject: [jira] Commented: (HARMONY-5644) [drlvm][jni] using a component
 manger for VM startup
In-Reply-To: <2059504365.1206387446229.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-5644?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12585514#action_12585514 ] 

Alexei Fedotov commented on HARMONY-5644:
-----------------------------------------

Hello Pavel,

1. I'm ok with any naming. The only thing I grok is that "Intel" is no longer valid. :-) Feel free to update as you like.
2. I have not caught Global_Env note: I have not touched it with my changes yet. It is not yet understood how to refactor it since it calls C++ constructors, which a component manger is pure C.
3. VM uses component manger pool in terms of memory allocation. Instead of using a pool kept in a static global variable now the dedicated pool is returned for VM from a component manager.


> [drlvm][jni] using a component manger for VM startup
> ----------------------------------------------------
>
>                 Key: HARMONY-5644
>                 URL: https://issues.apache.org/jira/browse/HARMONY-5644
>             Project: Harmony
>          Issue Type: Improvement
>          Components: DRLVM
>            Reporter: Alexei Fedotov
>            Assignee: Pavel Pervov
>         Attachments: create_java_vm_via_compmgr.patch, create_java_vm_via_compmgr_1.patch, create_java_vm_via_compmgr_2.patch, create_java_vm_via_compmgr_3.patch
>
>
> Unifying pool usage on VM startup is a step for providing multi-VM environment and VM internal state serialization. Strangely, massive changes started to work before any runtime fixes, so I'm currently writing a stress test to check the patch in depth.

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