Unified memory - in page error reading location

Accelerated Computing CUDA CUDA Programming and Performance
1

Hi, I have done some experiments with Unified Memory. I watched CUDACast #18: Unified Memory
CUDACast #18: Unified Memory - YouTube)
and tried to run the code from code-samples/dataElem_um_c++_2.cu at master · NVIDIA-developer-blog/code-samples · GitHub Unfortunately the code throws following exception:

Unhandled exception at 0x00007FF73AFB25AB in ShapeletDiscoveryParallel.exe: 0xC0000006: In page error reading location 0x0000000A00000008 (status code 0xC0000022).

My environment: Lenovo W530, Windows 8.1 Pro (64bit), NVIDIA Quadro K1000M, Driver version 340.84, Visual Studio 2012

I would be grateful if you could help me.

2

Hi Rodrigo,

I am experiencing a similar problem. Did you manage to get a solution?

Thanks,

Angus

3

Do you have any trouble if you run a Unified Memory cuda sample code?

The particular code for the parallel forall blog post doesn’t seem to be doing any cuda error checking.

You could also try running your code with cuda-memcheck