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.
Build: NetBeans IDE 7.3 (Build 201302132200) VM: Java HotSpot(TM) 64-Bit Server VM, 20.14-b01-445, Java(TM) SE Runtime Environment, 1.6.0_41-b02-445-11M4107 OS: Mac OS X User Comments: alesak: deleted php file GUEST: On git change branch Maximum slowness yet reported was 21786 ms, average is 20943
Created attachment 132023 [details] nps snapshot
AWT-EventQueue is blocked on Node.destroy's call to Mutex.postWriteRequest. The Mutex lock is probably held by parsing thread, and especially HtmlHintsProvider. 21s in production warrants at least P2.
(In reply to comment #2) > AWT-EventQueue is blocked on Node.destroy's call to Mutex.postWriteRequest. The > Mutex lock is probably held by parsing thread, and especially > HtmlHintsProvider. Why do you think so? > > 21s in production warrants at least P2.
> > Mutex lock is probably held by parsing thread, and especially > > HtmlHintsProvider. > Why do you think so? By looking at the content of the npss file I came to the such conclusion. Do you have problems reading npss files?
Jardo, I have no problem with nps files, but in this case I do not see any code in the parsing thread wich would acquire the mutex.
(In reply to comment #2) > > AWT-EventQueue is blocked on Node.destroy's call to Mutex.postWriteRequest. See org.openide.nodes.FilterNode.originalDestroyed() 100.0 11,691 ms > > Mutex lock is probably held by parsing thread, and especially > > HtmlHintsProvider. > Why do you think so? Because it is the only thread that runs 100% of time. More info. The lock is held by thread that computes: org.netbeans.modules.csl.navigation.ElementNode$ElementChildren.createNodes() 11,691 ms by calling into org.netbeans.modules.html.editor.gsf.HtmlStructureItem.getNestedItems()11,691 ms which is then blocked by the parsing thread, see call to org.netbeans.modules.html.editor.gsf.HtmlStructureItem.runTask() 11,691 ms and below.
Thanks for the additional information Jardo. Passing to CSL then as the nodes implementation lives there and the behaviour where StructureItem-s create the children items lazily in getNestedItems() using the parsing task has been agreed with Svata.
*** Bug 226540 has been marked as a duplicate of this bug. ***
Seems as duplicate of issue #225925; in the current source, getNestedItems() is not called in ElementNode's constructor, but rather from Children.addNotify(). *** This bug has been marked as a duplicate of bug 225925 ***