Debugging CUDA Programs
The TotalView CUDA debugger is an integrated debugging tool capable of simultaneously debugging CUDA code that is running on the hosts host system and the NVIDIA® GPU. CUDA support is an extension to the standard version TotalView, and is capable of debugging 64-bit CUDA programs. Debugging 32-bit CUDA programs is currently not supported.
Supported major features:
*Debug CUDA application running directly on GPU hardware
*Set breakpoints, pause execution, and single step in GPU code
*View GPU variables in PTX registers, local, parameter, global, or shared memory
*Access runtime variables, such as threadIdx, blockIdx, blockDim, etc.
*Debug multiple GPU devices per process
*Support for the CUDA MemoryChecker
*Debug remote, distributed and clustered systems
*All host debugging features are supported, except for ReplayEngine
 
RELATED TOPICS 
Using the CUDA debugger
About the CUDA Debugger” in the Classic TotalView User Guide
The CLI dcuda command
dcuda in the Classic TotalView Reference Guide