Skip to main content

Enabling OpenGl kills the X windows

7 replies [Last post]
rah003
Offline
Joined: 2004-05-26
Points: 0

6uN-b11 on Linux ace2 2.6.24-7-generic #1 SMP Thu Feb 7 01:29:58 UTC 2008 i686 GNU/Linux

When opengl pipeline is enabled the x window freezes on startup ... starting the app with J2D_TRACE_LEVEL=4 and -Dsun.java2d.ppmoffscreen=False -Dsun.java2d.opengl=True
I get only the trace below before it freezes ... is this a known thing or is there anything more I can do to track down what the problem is?
Cheers,
Jan

[I] GLXGC_FindBestVisual: scn=0
[I] GLXGC_InitGLX
[I] OGLFuncs_OpenLibrary
[I] OGLFuncs_InitPlatformFuncs
[I] OGLFuncs_InitBaseFuncs
[I] OGLFuncs_InitExtFuncs
[I] GLXGC_InitGLX: client GLX version=1.4
[I] GLXGC_InitFBConfig: scn=0 vis=0x0
[V] candidate fbconfigs:
[V] id=0x29 db=0 alpha=0 depth=24 stencil=0 valid=true
[V] id=0x2a db=0 alpha=8 depth=24 stencil=0 valid=false (large depth)
[V] id=0x5f db=0 alpha=0 depth=24 stencil=0 valid=false (large depth)
[V] id=0x60 db=0 alpha=8 depth=24 stencil=0 valid=false (large depth)
[V] id=0x44 db=0 alpha=0 depth=24 stencil=0 valid=false (large depth)
[V] id=0x45 db=0 alpha=8 depth=24 stencil=0 valid=false (large depth)
[V] id=0x25 db=0 alpha=0 depth=24 stencil=8 valid=false (large depth)
[V] id=0x26 db=0 alpha=8 depth=24 stencil=8 valid=false (large depth)
[V] id=0x5b db=0 alpha=0 depth=24 stencil=8 valid=false (large depth)
[V] id=0x5c db=0 alpha=8 depth=24 stencil=8 valid=false (large depth)
[V] id=0x40 db=0 alpha=0 depth=24 stencil=8 valid=false (large depth)
[V] id=0x41 db=0 alpha=8 depth=24 stencil=8 valid=false (large depth)
[V] id=0x27 db=1 alpha=0 depth=24 stencil=0 valid=false (large depth)
[V] id=0x28 db=1 alpha=8 depth=24 stencil=0 valid=false (large depth)
[V] id=0x5d db=1 alpha=0 depth=24 stencil=0 valid=false (large depth)
[V] id=0x5e db=1 alpha=8 depth=24 stencil=0 valid=false (large depth)
[V] id=0x42 db=1 alpha=0 depth=24 stencil=0 valid=false (large depth)
[V] id=0x43 db=1 alpha=8 depth=24 stencil=0 valid=false (large depth)
[V] id=0x21 db=1 alpha=0 depth=24 stencil=8 valid=false (large depth)
[V] id=0x24 db=1 alpha=8 depth=24 stencil=8 valid=false (large depth)
[V] id=0x23 db=1 alpha=0 depth=24 stencil=8 valid=false (large depth)
[V] id=0x5a db=1 alpha=8 depth=24 stencil=8 valid=false (large depth)
[V] id=0x22 db=1 alpha=0 depth=24 stencil=8 valid=false (large depth)
[V] id=0x3f db=1 alpha=8 depth=24 stencil=8 valid=false (large depth)
[I] GLXGC_FindBestVisual: chose 0x29 as the best visual for screen 0
[I] GLXGraphicsConfig_getGLXConfigInfo
[I] GLXGC_InitFBConfig: scn=0 vis=0x29
[V] candidate fbconfigs:
[V] id=0x29 db=0 alpha=0 depth=24 stencil=0 valid=true
[I] OGLContext_IsExtensionAvailable: GL_ARB_fragment_shader=true
[I] OGLContext_IsExtensionAvailable: GL_ARB_multitexture=true
[I] OGLContext_IsExtensionAvailable: GL_ARB_texture_non_power_of_two=true
[I] OGLContext_IsExtensionAvailable: GL_ARB_texture_rectangle=true
[I] OGLContext_IsExtensionAvailable: GL_EXT_framebuffer_object=true
[I] OGLContext_IsFBObjectExtensionAvailable: fbobject supported
[I] OGLContext_IsLCDShaderSupportAvailable: LCD text shader supported
[I] OGLContext_IsBIOpShaderSupportAvailable: BufferedImageOp shader supported
[I] OGLContext_IsGradShaderSupportAvailable: Linear/RadialGradientPaint shader supported
[I] OGLContext_IsExtensionAvailable: GL_NV_fragment_program=true
[I] OGLContext_IsExtensionAvailable: GL_NV_fragment_program2=true
[I] GLXGraphicsConfig_getGLXConfigInfo: OpenGL version=2.1.2 NVIDIA 169.09
OpenGL pipeline enabled for default config on screen 0

Reply viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
linuxhippy
Offline
Joined: 2004-01-07
Points: 0

> I've read the proposal, however I can hardly voice
> opinion about something i know nothing about ...
> sorry :(
Was just kidding ;)
Thanks for spending the time and reading it.

linuxhippy
Offline
Joined: 2004-01-07
Points: 0

Does the problem also happen with the earlier series 100 drivers?
The 169 drivers where a port of many changes done on the windows branch, so it seems NVidia again broke the OpenGL pipeline.

By the way I started a proposal to implement an Xrender backend for Java2d which should work more reliable compared to the OpenGL pipeline over a large set of hardware.
If you're interested in it, please support it at: http://mail.openjdk.java.net/mailman/listinfo/challenge-discuss
(yes I know, its plain dumb lobbying^^)

lg Clemens

rah003
Offline
Joined: 2004-05-26
Points: 0

> Does the problem also happen with the earlier series
> 100 drivers?
> The 169 drivers where a port of many changes done on
> the windows branch, so it seems NVidia again broke
> the OpenGL pipeline.
Could be ... I haven't used older version of drivers with this graphic card so i can't say.
Strangely enough other applications (non-java) using opengl works with those dirvers just fine.

linuxhippy
Offline
Joined: 2004-01-07
Points: 0

Sun even sent them some automatic tests to be able to track down regressions which would break the Java2D-OpenGL pipeline. Seems they have not been run on the 169' drivers.

The problem is that the OpenGL pipeline stresses some uncommon and untested areas of OpenGL drivers, which these days are mostly used for 3D games.

Clemens

Btw. thanks for ignoring my XRender proposal ;)

rah003
Offline
Joined: 2004-05-26
Points: 0

> Sun even sent them some automatic tests to be able to track down regressions which would break the
> Java2D-OpenGL pipeline. Seems they have not been run on the 169' drivers.

Any place where i can get those tests and run them to at least figure out what function is broken?

> Btw. thanks for ignoring my XRender proposal ;)

I've read the proposal, however I can hardly voice opinion about something i know nothing about ... sorry :(

Chris Campbell

Hi Jan,

java2d@JAVADESKTOP.ORG wrote:
>> Sun even sent them some automatic tests to be able to track down regressions which would break the
>> Java2D-OpenGL pipeline. Seems they have not been run on the 169' drivers.
>
> Any place where i can get those tests and run them to at least figure out what function is broken?
>

Not yet, unfortunately. I don't think they'd be that helpful in this
case anyway, as something very fundamental seems to be broken.

Could you try the following (turning off FBO usage) to see if it helps?

% java -Dsun.java2d.opengl=True -Dsun.java2d.opengl.fbobject=false -jar
SwingSet2.jar

Thanks,
Chris

===========================================================================
To unsubscribe, send email to listserv@java.sun.com and include in the body
of the message "signoff JAVA2D-INTEREST". For general help, send email to
listserv@java.sun.com and include in the body of the message "help".

rah003
Offline
Joined: 2004-05-26
Points: 0

Hey Chris,
there is a difference - the X Windows now freeze when app window is half transparent instead of fully painted and soft restart of gdm (/etc/init.d/gdm restart) is enough to make X work again. Before there was still noise on the screen even after gdm restart and reboot was the only solution.
Let me know if there is anything else I can do to help to track it down.
Cheers,
Jan