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 252383 - CodeCompletion took 28877 ms.
Summary: CodeCompletion took 28877 ms.
Status: RESOLVED FIXED
Alias: None
Product: java
Classification: Unclassified
Component: Project (show other bugs)
Version: 8.1
Hardware: All All
: P3 normal (vote)
Assignee: Tomas Zezula
URL:
Keywords: PERFORMANCE
Depends on:
Blocks:
 
Reported: 2015-05-13 15:14 UTC by NukemBy
Modified: 2015-07-30 01:25 UTC (History)
1 user (show)

See Also:
Issue Type: DEFECT
Exception Reporter: 216071

Attachments
nps snapshot (256.98 KB, application/nps)
2015-05-13 15:14 UTC, NukemBy 		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 NukemBy 2015-05-13 15:14:25 UTC 
This issue was reported manually by mentlicher.
It already has 2 duplicates 


Build: NetBeans IDE Dev (Build 1)
VM: Java HotSpot(TM) 64-Bit Server VM, 24.72-b04, Java(TM) SE Runtime Environment, 1.7.0_72-b14
OS: Windows Server 2008 R2

User Comments:
NukemBy: Trying to add breakpoint on exception: typing NullP in "New Breakpoint" window

NukemBy: Trying to add breakpoint on exception - wait for autosuggest after typing "NullP"



Maximum slowness yet reported was 28877 ms, average is 21286
Comment 1 NukemBy 2015-05-13 15:14:29 UTC 
Created attachment 153633 [details]
nps snapshot
Comment 2 Martin Entlicher 2015-05-13 15:18:46 UTC 
Parsing takes a really long time. SourceForBinaryQuery is slow.
Comment 3 Svata Dedic 2015-07-24 08:53:51 UTC 
delegates to j2seplaetform/project.
Comment 4 Tomas Zezula 2015-07-29 16:29:36 UTC 
There are 3 things:
1st) The warm up of the global libraries may not work if it's the first call to the SourceForBinaryQuery - fixed
2nd) The LibraryManagers were not cached - fixed
3rd) Extremely slow *monitorenter* instruction - not sure why. The monitor is uncontended or at least no contention visible in sampling. It can be spin-then-park if a small invisible contention exist or biased locking stack annotations or something else. If it happen again I we will need hotspot lock log.

Fixed jet-main 9a36c922ba8f
Comment 5 Quality Engineering 2015-07-30 01:25:20 UTC 
Integrated into 'main-silver', will be available in build *201507300002* on http://bits.netbeans.org/dev/nightly/ (upload may still be in progress)

Changeset: http://hg.netbeans.org/main-silver/rev/9a36c922ba8f
User: Tomas Zezula <tzezula@netbeans.org>
Log: #252383:CodeCompletion took 28877 ms.