This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.

Bug 128542 - Deadlock in ''Versioning long tasks'' thread
Summary: Deadlock in ''Versioning long tasks'' thread
Status: CLOSED FIXED
Alias: None
Product: ide
Classification: Unclassified
Component: Performance (show other bugs)
Version: 3.x
Hardware: PC Linux
: P2 blocker (vote)
Assignee: Jaroslav Tulach
URL: http://test4u.czech.sun.com/test4u-us...
Keywords: TEST
Depends on:
Blocks:
 
Reported: 2008-02-27 11:00 UTC by Alexander Kouznetsov
Modified: 2011-05-25 11:38 UTC (History)
0 users

See Also:
Issue Type: DEFECT
Exception Reporter:

Attachments
Thread dump (33.68 KB, text/plain)
2008-02-27 11:01 UTC, Alexander Kouznetsov 		Details
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Alexander Kouznetsov 2008-02-27 11:00:54 UTC 
Product Version         = NetBeans IDE Dev (Build 200802260005)
   Operating System        = Linux version 2.6.20-15-generic running on i386
   Java; VM; Vendor        = 1.6.0_04; Java HotSpot(TM) Client VM 10.0-b19; Sun Microsystems Inc.

The following deadlock happened during performance tests execution:

 Found one Java-level deadlock:
 =============================
 "AWT-EventQueue-2":
   waiting for ownable synchronizer 0xa29fb1d8, (a java.util.concurrent.locks.ReentrantLock$NonfairSync),
   which is held by "AWT-XAWT"
 "AWT-XAWT":
   waiting to lock monitor 0x080d9d14 (object 0xa45e5828, a org.netbeans.performance.test.guitracker.LoggingRepaintManager),
   which is held by "Versioning long tasks"
 "Versioning long tasks":
   waiting to lock monitor 0x08453c38 (object 0xa2b4cc88, a org.netbeans.core.windows.view.ui.MainWindow),
   which is held by "AWT-EventQueue-2"
Comment 1 Alexander Kouznetsov 2008-02-27 11:01:46 UTC 
Created attachment 57338 [details]
Thread dump
Comment 2 Maros Sandor 2008-02-27 12:07:23 UTC 
In versioning we are just calling java.awt.Component.repaint() as you can see in the thread dump and the deadlock seems
to be caused by performance code. Please evaluate.
Comment 3 Jaroslav Tulach 2008-03-25 17:10:11 UTC 
Nobody else wants to fix this?
Comment 4 Jaroslav Tulach 2008-03-25 17:17:34 UTC 
changeset:   75246:10f135d8cc7f
tag:         tip
user:        Jaroslav Tulach <jtulach@netbeans.org>
date:        Tue Mar 25 17:17:13 2008 +0100
summary:     #128542: Do not call super.addDirtyRegion while holding lock on ourself