Some Eclipse Foundation services are deprecated, or will be soon. Please ensure you've read this important communication.
Bug 362161 - VM crash in Graphics_FillPolygon / gdiplus.dll when extending heap size
Summary: VM crash in Graphics_FillPolygon / gdiplus.dll when extending heap size
Status: CLOSED DUPLICATE of bug 301508
Alias: None
Product: Platform
Classification: Eclipse Project
Component: SWT (show other bugs)
Version: 4.2   Edit
Hardware: PC Windows NT
: P3 normal with 1 vote (vote)
Target Milestone: ---   Edit
Assignee: Platform-SWT-Inbox CLA
QA Contact:
URL:
Whiteboard:
Keywords:
: 383543 (view as bug list)
Depends on:
Blocks:
 
Reported: 2011-10-27 04:17 EDT by Jesper Skov CLA
Modified: 2012-09-07 09:59 EDT (History)
9 users (show)

See Also:

Attachments
Crash stack (26.06 KB, text/plain)
2011-10-27 04:18 EDT, Jesper Skov CLA 		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 Jesper Skov CLA 2011-10-27 04:17:31 EDT 
Build Identifier: 20110916-0149

When using Mylyn to display a Jira issue in Eclipse, it crashes with a VM error iff I have extended the heap to 1024MB. When using the default of 512MB, I cannot reproduce the problem.

It may (or may not) be related to bug #239905.

Reproducible: Always

Steps to Reproduce:
1. Change eclipse.ini so it uses -Xmx1024m.
2. Show Mylyn Task list, double click on a Jira issue to show it.
3. Eclipse crashes hard.
Comment 1 Jesper Skov CLA 2011-10-27 04:18:51 EDT 
Created attachment 206057 [details]
Crash stack
Comment 2 Felipe Heidrich CLA 2011-11-11 12:35:56 EST 
I suspect this is bug in the video driver / gdi+ integration.
Comment 3 Jesper Skov CLA 2012-04-30 05:59:32 EDT 
For what it is worth, I have also experienced the problem with JunoM6.

It happened when just opening Eclipse. It would crash immediately.
Than I lowered the Heap size to 800MB (from 1000MB), and now it runs without problems...
Comment 4 Tomasz Zarna CLA 2012-04-30 07:08:17 EDT 
(In reply to comment #3)
> For what it is worth, I have also experienced the problem with JunoM6.

Affirmative, I'm seeing this on the latest I-build for Juno (I20120429-2000). I've tried different VMs including Sun/Oracle's and IBM's from 1.6 to the latest 1.7. It was killing me until I applied Jesper's tip from comment 0.
Comment 5 Niko Stotz CLA 2012-06-28 03:27:06 EDT 
I still experience this bug with Juno final. I'm using a pretty standard Dell workstation, therefore I think this will cause quite some trouble when business projects start migrating to e4.

System specs:
WinXP Professional, Version 2002, SP3 (physical address extension enabled)
Dell OptiPlex 780
Graphics: Intel 4 Series Internal Chipset, Driver date 2009-06-25, Driver version 6.14.10.5082
Two monitors attached (one analog, one DVI)

The -Xmx limit is 512m on this system, anything above crashes the workbench as soon as it's displayed.
Comment 6 Tomasz Zarna CLA 2012-06-28 06:47:25 EDT 
(In reply to comment #4)
> It was killing me until I applied Jesper's tip from comment 0.

I'm taking that back, see bug 383543.
Comment 7 Steffen Pingel CLA 2012-07-01 15:17:52 EDT 
*** Bug 383543 has been marked as a duplicate of this bug. ***
Comment 8 Silenio Quarti CLA 2012-07-04 09:48:15 EDT 
Please take a look at these following bugs. They have similar stacks and they indicate it is a video driver problem.

bug#301508
bug#231368
bug#347765
Comment 9 Jesper Skov CLA 2012-09-07 05:40:26 EDT 
It does appear to be a video driver problem.

I have reduced XP video driver optimization 2-3 notches, and I don't see crashes any more.

Might as well close this issue, I guess...
Comment 10 Silenio Quarti CLA 2012-09-07 09:59:04 EDT 

*** This bug has been marked as a duplicate of bug 301508 ***