Error - unable to use CUDA device even though it's not busy
Hello,

I am trying to run a simple example program from the "CUDA by Example" book. The source code is given here: http://developer.nvidia.com/cuda-example-introduction-general-purpose-gpu-programming and the code is "chapter03/simple_kernel_params.cu". I am able to compile the code on my Tesla C2070 but it gives me the following error when I try to run the code:


all CUDA-capable devices are busy or unavailable in simple_kernel_params.cu at line 26

Any idea on how to fix this? I am able to query the device just fine, and I am the only one trying to run any jobs on the device so I know it is not busy. Any help is appreciated.
Hello,



I am trying to run a simple example program from the "CUDA by Example" book. The source code is given here: http://developer.nvidia.com/cuda-example-introduction-general-purpose-gpu-programming and the code is "chapter03/simple_kernel_params.cu". I am able to compile the code on my Tesla C2070 but it gives me the following error when I try to run the code:





all CUDA-capable devices are busy or unavailable in simple_kernel_params.cu at line 26



Any idea on how to fix this? I am able to query the device just fine, and I am the only one trying to run any jobs on the device so I know it is not busy. Any help is appreciated.

#1
Posted 04/01/2012 11:00 PM   
What happens when you run the [url="http://developer.nvidia.com/cuda-cc-sdk-code-samples#deviceQuery"]device query example from the SDK[/url]?
What happens when you run the device query example from the SDK?

Always check return codes of CUDA calls for errors. Do not use __syncthreads() in conditional code unless the condition is guaranteed to evaluate identically for all threads of each block. Run your program under cuda-memcheck to detect stray memory accesses. If your kernel dies for larger problem sizes, it might exceed the runtime limit and trigger the watchdog timer.

#2
Posted 04/02/2012 09:57 AM   
Scroll To Top