WPF GUI freeze in hardware rendering condition

[idata]

I'm experiencing a GUI freeze problem, related to hard vs. software UI rendering.

The context : 2 WPF apps (.NET 3.5 SP1) running on a dual-display XP Embedded machine (DirectX 9.0c).

At some point, one of these app freezes. But only the rendering : the GUI stays active (message pump is active, the application is responsive), but the rendering is not done. The freeze can last seconds or forever. The 2nd app keeps running normally. The bug occurs only in a dual-display configuration, the 2nd app being a kind of catalyst (more video memory usage...)

Using http://windowsclient.net/wpf/perf/wpf-perf-tool.aspx# perforator Perforator, I see that, during a freeze, Video Memory Usage is continuously going up & down in spikes (cf : http://www.hostingpics.net/viewer.php?id=628634WPFperforatorfreezeunfreeze5.jpg screenshot). During that time, one native thread is consuming 100% of a CPU (the stack-trace involves WPFgfx, DirectX, GDI, and video driver)

Usually, when the app unfreeze, it's fully software rendered (purple tint with Perforator). Also, disabling hardware rendering immediately unfreeze the app.

I cannot reproduce the bug when hardware rendering is disabled.

I'm suspecting the video driver, as I cannot reproduce it on another hardware. It's said to be up-to-date (v. 6.14.10.4926 on a Mobile Intel(R) 945 Express Chipset). See driver information report below.

Can anyone help ?

Thanks

Intel(R) Graphics Media Accelerator Driver for Mobile Report

Report Date: 04/05/2011Report Time[hr:mm:ss]: 14:35:06Driver Version: 6.14.10.4926Operating System: Windows XP* Professional, Service Pack 3 (5.1.2600)Default Language: FrenchDirectX* Version: 9.0Physical Memory: 2039 MBMinimum Graphics Memory: 8 MBMaximum Graphics Memory: 64 MBGraphics Memory in Use: 38 MBProcessor: x86 family 6 Model 14 Stepping 12Processor Speed: 1666 MHZVendor ID: 8086Device ID: 27AEDevice Revision: 03* Accelerator Information *Accelerator in Use: Mobile Intel(R) 945 Express Chipset FamilyVideo BIOS: 1585.0Current Graphics Mode: 1024 x 768 True Color (60 Hz)* Devices Connected to the Graphics Accelerator *Active Notebook Displays: 1* Notebook *Monitor Name: Digital Flat Panel (1024x768)Display Type: DigitalGamma Value: 3,54DDC2 Protocol: SupportedMaximum Image Size: Horizontal: Not Available Vertical: Not AvailableMonitor Supported Modes:1024 x 768 (60 Hz)Display Power Management Support: Standby Mode: Not Supported Suspend Mode: Not Supported Active Off Mode: SupportedRaw EDID:00 ff ff ff ff ff ff 00 36 7f 03 00 03 00 00 0000 0c 01 03 80 ff ff ff 2f 00 00 a0 57 49 9b 2610 48 4f 00 00 00 01 01 01 01 01 01 01 01 01 0101 01 01 01 01 01 64 19 00 40 41 00 26 30 18 8836 00 00 00 00 00 00 18 00 00 00 fd 00 00 3c 00eb 07 00 00 00 00 00 00 00 00 00 00 00 00 00 0000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 6a* Other names and brands are the property of their respective owners.

[idata]

Are there some tools I can use to diagnose driver problems ? or to investigate around the fall-back to software rendering ?

[idata]

Hi mde

have you find a solution for your problem.

I have the same problem on a dual Display

The graphic card is an intel graphics media accelerator hd.

best regards

[idata]

Did you find a solution to this? I am facing the same problem. I know the reasons; do not know how to fix it. The Intel driver is wrongly picking the monitor driver as "Digital Flat Panel" instead of PnP driver. Do not know where to change it?

[idata]

I have the exact same problem with my WPF application which uses BitmapCache on visuals.

The GPU memory consumption remains steady for a while, but then after a number of hours it starts to creep and doesn't stop.

The symptom I have is: the UI stops updating but the UI thread is still running as per normal.

Eventually the Operating System will become completely unresponsive.

I've had a senior developer in Microsoft be kind enough to profile my application and they confirmed this was not a problem with the application (i.e. no memory leaks or scary code).

If I turn on BitmapCache on 6 or more visuals (thumnail images which I scroll left-right with code), the GPU memory consumption won't let the application run for more than 3 hours. If I turn BitmapCache on only 1 visual (main grid), the application will run for about 24 - 36 hours.

These embedded graphics chipsets / drivers are a bit nasty.

[idata]

I might point out, when I reported this problem 4 months ago (/message/145805# 145805 http://communities.intel.com/message/145805# 145805) Intel washed their hands because I was using Windows XP Embedded also (any excuse will do on a tough problem).

I've since migrated to Windows XP Professional (because I had to eliminate this as a potential factor) and as suspected, the problem remains.