multiGPU.c example is giving an error

[harbm]

Basic usage works fine, but I'm trying to run the multiGPU.c example but it gives the following error.

[xxxx@xxxxxxxxxx multiGPU]$ ./multiGPU
Found 2 devices, will launch 2 threads

Thread 0 - Launched
Thread 0 - Binding to device 0
Thread 1 - Launched
Thread 1 - Binding to device 1
Thread 0 - CUDA Error: exclusive-thread device already in use by a different thread (see Programmer's Guide)
Thread 1 - Allocating matrices
Thread 1 - Initializing CULA
Thread 1 - Calling culaSgeqrf
Thread 1 - Shutting down CULA

Thread 0 - Failed

edit: This using the premium (paid) version of cula dense R17, (for some reason this doesn't appear under my username). Running on a linux box with 2 kepler GPUs and with CUDA 5.0.