United StatesChange Country, Oracle Worldwide Web Sites Communities I am a... I want to...
Bug ID: JDK-6341149 OGL: Enabling OGL on winXPpro with GeForce4 MX440 crashed VM with 81.85 drivers
JDK-6341149 : OGL: Enabling OGL on winXPpro with GeForce4 MX440 crashed VM with 81.85 drivers

Details
Type:
Bug
Submit Date:
2005-10-25
Status:
Closed
Updated Date:
2010-04-03
Project Name:
JDK
Resolved Date:
2005-11-17
Component:
client-libs
OS:
windows_xp
Sub-Component:
2d
CPU:
x86
Priority:
P3
Resolution:
Not an Issue
Affected Versions:
6
Fixed Versions:

Related Reports

Sub Tasks

Description
Java Version: 1.6.0-ea-b57
Machine: ladybird.sfbay with WindowsXPpro sp2
Vedio Card: GeForce 4 MX4400 with AGP8X
Driver Version: 81.85 Release Date: Oct20,2005
Dell Precision 330 Pentium4

Problems:
I just installed a video card driver by downloading it from nvidia web site,
when I run Java2Demo with the newest driver, and mustang recent build b57,
the VM is crashed when closing the app.
It can be reproducible all times.
GeForce 4 was OK with the older drivers, may it be a driver bug?
If you don't enable OGL, it is OK too with the new driver.

There is a bug for ATI card, 5066318.
Since this one is only happened with the new driver, decide to file a new bug.

To reproduce:
1. go to nvidia.com to download drivers for GeForce 4
   the version: 81.85 with release date: oct20, 2005
2. Using mustang recent build to launch java2demo with ogl enabled
   bin\java -jar -Dsun.java2d.opengl=True demo\jfc\Java2D\*.jar
3.close the app, you will see the crashed happens right way.

See attached log for the crashed.

                                    

Comments
EVALUATION

This is obviously a regression in the latest Windows driver release (81.85)
from Nvidia.  I have filed a new bug report (195804) against Nvidia.  Apparently
their Linux/Solaris driver team has been running our automated JavaOGLBAT
suite in their nightly testing, but the Windows driver team has not.  If they
had, they would have caught this regression before the driver was released.
I've contacted Nvidia and we will work to make sure that our testsuite is
run on the Windows side of the house in the near future to prevent further
regressions.  I will leave this bug report open until we have verified that
the bug has been fixed in a future driver release from Nvidia.
                                     
2005-10-27
EVALUATION

This driver bug has been fixed by Nvidia in their most recent driver release
(81.94, available since 11/14/2005).  I will close this as "not a defect" since
it is not a bug in the JDK, and is no longer reproducable with the latest
drivers from Nvidia.
                                     
2005-11-17



Hardware and Software, Engineered to Work Together