Some Eclipse Foundation services are deprecated, or will be soon. Please ensure you've read this important communication.

Bug 369917

Summary:

Deadlock in JavaReconciler while trying to lock DefaultClassLoader

Product:

[Eclipse Project] Equinox

Reporter:

Matthias Schmidt <schmidtm>

Component:

Framework

Assignee:

equinox.framework-inbox <equinox.framework-inbox>

Status:

CLOSED DUPLICATE

QA Contact:

Severity:

major

Priority:

CC:

daniel_megert, markus.kell.r, tjwatson

Version:

3.8.0 Juno

Target Milestone:

---

Hardware:

OS:

Windows XP

Whiteboard:

Attachments:

Description	Flags
Stacktrace of deadlock from JDK 1.7.0-b147	none

Description Matthias Schmidt

2012-01-27 06:49:03 EST

Build Identifier: Indigo Service Release 1, Build id: 20110916-0149

Every now and then, when I'm clicking in the editor pane of a java file inside a maven project eclipse deadlocks. The deadlock got detected by the JVM deadlock detection facility. Find a full stacktrace attached.

Reproducible: Sometimes

Steps to Reproduce:
Sorry guys, this happens every now and then (see above)

Comment 1 Matthias Schmidt

2012-01-27 06:51:01 EST

Created attachment 210174 [details]
Stacktrace of deadlock from JDK 1.7.0-b147

Comment 2 Dani Megert

2012-01-30 08:26:50 EST

Looks like a dup of bug 344030.

Comment 3 Thomas Watson

2012-01-30 09:02:33 EST

(In reply to comment #1)
> Created attachment 210174 [details]
> Stacktrace of deadlock from JDK 1.7.0-b147

Are you certain this is a Java 7 VM that you used?

> Full thread dump Java HotSpot(TM) Client VM (17.0-b17 mixed mode):
Some googling of 17.0-b17 indicates this is Java 6.

> at org.eclipse.osgi.baseadaptor.loader.ClasspathManager.findLocalClass_LockClassLoader(ClasspathManager.java:477)

The use of findLocalClass_LockClassLoader also indicates that this is Java 6 or earlier.

This seems a duplicate of bug212262 which can really only be fixed by the use of Java 7 and the utilization of the ClassLoader#registerAsParallelCapable method.

Comment 4 Matthias Schmidt

2012-01-31 05:10:12 EST

Sorry guys, my fault. Sure it is JDK6:

java version "1.6.0_21"
Java(TM) SE Runtime Environment (build 1.6.0_21-b07)
Java HotSpot(TM) Client VM (build 17.0-b17, mixed mode, sharing)

Comment 5 Thomas Watson

2012-01-31 08:47:11 EST

(In reply to comment #4)
> Sorry guys, my fault. Sure it is JDK6:
> 
> java version "1.6.0_21"
> Java(TM) SE Runtime Environment (build 1.6.0_21-b07)
> Java HotSpot(TM) Client VM (build 17.0-b17, mixed mode, sharing)

Thanks for clarifying.  I am going to duplicate this to bug212262.  Unfortunately there is nothing we can do to avoid this kind of deadlock with out help from the Java 7 VM.  We have tried many times to no avail.

*** This bug has been marked as a duplicate of bug 212262 ***