bug? Binding OpenGL buffers to different h threads OpenGL buffer binding bug

Had this subject under another thread but haven’t receive any answer to my question.
I decided to create a new thread and to publish the problem under a more adequate topic.

I have the following problem:
In one process just one host thread is allowed to bind OpenGL buffers.
Trying to bind a buffer to another host thread leads to the freeze of the computer.
I suppose it’s a bug in CUDA.

Or is this fact described somewhere in the CUDA programming guide already?

By the way the driver version is (download name: 169.21_forceware_winxp_64bit_international_whql.exe)
for Windows XP 64 bit edition and I was using the runtime API from the CUDA SDK 1.1. Graphics card is a nVidia 8800 GTS.
Toolkit (nvcc.exe, etc.) was versioned 1.1 and the assembly the runtime API calls were made from was generated by the Visual CPP Compiler,
but I don’t think that it matters, because I don’t suppose that this error is caused by the compilers!
CPU: Intel Core2 Duo X6800 2.93 GHz

Here is a DirectShow renderer in which the error occurs.
It’s a VS2005 project for windows.

DirectShow filters must be registered to use them by calling regsvr32.exe and passing the name of the DS filter (regsvr32.exe .ax).

The filtern than can be used with GraphEdit (graphedt.exe / graphedt_64.exe). Note that a 64 bit DS filter can be used by graphedt_64.exe and a 32 bit filter by graphedt.exe. Both version can be registered in the system at the same time.
The GraphEdit is a part of the Microsoft Platform SDK.

You will find the filter in the GraphEdit registered under the name “DS_CUDA_RENDERER”.
(The filter supports the format RGB32 only.)

Thanks in advance for your response.