313086 – Deadlock when saving CUE

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

Bug 313086 - Deadlock when saving CUE

Summary: Deadlock when saving CUE

Status:	CLOSED DUPLICATE of bug 311863

Alias:	None

Product:	JDT
Classification:	Eclipse Project
Component:	Text (show other bugs)
Version:	3.6
Hardware:	PC Windows XP

Importance:	P3 normal (vote)
Target Milestone:	---
Assignee:	JDT-Text-Inbox
QA Contact:

URL:
Whiteboard:
Keywords:

Depends on:
Blocks:

Reported:	2010-05-17 05:16 EDT by Benno Baumgartner
Modified:	2010-05-17 11:45 EDT (History)
CC List:	3 users (show)

See Also:

Attachments
Thread Dump (29.50 KB, text/plain) 2010-05-17 05:16 EDT, Benno Baumgartner	no flags	Details
View All 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 Benno Baumgartner

2010-05-17 05:16:55 EDT

Created attachment 168698 [details]
Thread Dump

I20100429-1549

I'm not sure what I did, nothing fancy though, just edited some Java file in the CUE and saved:

"main" prio=6 tid=0x008f6400 nid=0x680 in Object.wait() [0x0012d000]
   java.lang.Thread.State: TIMED_WAITING (on object monitor)
        at java.lang.Object.wait(Native Method)
        at org.eclipse.core.internal.jobs.Semaphore.acquire(Semaphore.java:38)
        - locked <0x0b327008> (a org.eclipse.core.internal.jobs.Semaphore)
        at org.eclipse.core.internal.jobs.OrderedLock.doAcquire(OrderedLock.java:175)
        at org.eclipse.core.internal.jobs.OrderedLock.acquire(OrderedLock.java:109)
        at org.eclipse.core.internal.jobs.OrderedLock.acquire(OrderedLock.java:83)
        at org.eclipse.core.internal.resources.WorkManager.checkIn(WorkManager.java:118)
        at org.eclipse.core.internal.resources.Workspace.prepareOperation(Workspace.java:1914)
        at org.eclipse.core.internal.resources.Marker.setAttribute(Marker.java:233)
        at org.eclipse.core.internal.resources.Marker.setAttribute(Marker.java:222)
        at org.eclipse.ui.texteditor.MarkerUtilities.setIntAttribute(MarkerUtilities.java:334)
        at org.eclipse.ui.texteditor.MarkerUtilities.setCharStart(MarkerUtilities.java:308)
        at org.eclipse.ui.texteditor.BasicMarkerUpdater.updateMarker(BasicMarkerUpdater.java:75)
        at org.eclipse.ui.texteditor.AbstractMarkerAnnotationModel.updateMarker(AbstractMarkerAnnotationModel.java:511)
        at org.eclipse.ui.texteditor.AbstractMarkerAnnotationModel.updateMarkers(AbstractMarkerAnnotationModel.java:571)
        at org.eclipse.ui.texteditor.AbstractMarkerAnnotationModel.commit(AbstractMarkerAnnotationModel.java:636)
        at org.eclipse.core.internal.filebuffers.ResourceTextFileBuffer.commitFileBufferContent(ResourceTextFileBuffer.java:402)
        at org.eclipse.core.internal.filebuffers.ResourceFileBuffer.commit(ResourceFileBuffer.java:325)
        at org.eclipse.jdt.internal.ui.javaeditor.DocumentAdapter.save(DocumentAdapter.java:472)
        at org.eclipse.jdt.internal.ui.javaeditor.CompilationUnitDocumentProvider$5.run(CompilationUnitDocumentProvider.java:1579)
        at org.eclipse.core.runtime.SafeRunner.run(SafeRunner.java:42)
        at org.eclipse.jdt.internal.ui.javaeditor.CompilationUnitDocumentProvider.notifyPostSaveListeners(CompilationUnitDocumentProvider.java:1566)
        at org.eclipse.jdt.internal.ui.javaeditor.CompilationUnitDocumentProvider.commitWorkingCopy(CompilationUnitDocumentProvider.java:1363)

Had to kill eclipse after 10 minutes.

See attachment for full dump.

Comment 1 Dani Megert

2010-05-17 05:28:02 EDT

John, any idea what could be wrong? Could the changes made for bug 307282 cause this? We recently got similar bug reports, see bug 308235 and bug 312543.

Comment 2 Dani Megert

2010-05-17 05:41:55 EDT

Note that setting a marker attribute uses no scheduling rule ('null').

Comment 3 John Arthorne

2010-05-17 11:30:00 EDT

(In reply to comment #1)
> John, any idea what could be wrong? Could the changes made for bug 307282 cause
> this? We recently got similar bug reports, see bug 308235 and bug 312543.

Yes, I think this is caused by the fix for bug 307282. I didn't see the connection before, but I believe what we are seeing here is bug 311863, which was introduced by the fix for bug 307282. Luckily this is already fixed in RC1 so I think this can be marked as a duplicate.

Comment 4 James Blackburn

2010-05-17 11:39:30 EDT

(In reply to comment #3)
> Yes, I think this is caused by the fix for bug 307282.

Yes, I agree this is probably caused by 307282 (see bug 311863 comment 11). The fix in M7 unfortunately exposed this issue to callers of OrderedLock#acquire(). I haven't seen any similar issues since this was fixed (we added a test + I had a readily reproducible real life example here).

Comment 5 Dani Megert

2010-05-17 11:45:44 EDT


*** This bug has been marked as a duplicate of bug 311863 ***