Enabling CUDA MemoryChecker Feature

You can detect global memory addressing violations and misaligned global memory accesses by enabling the CUDA MemoryChecker feature.

To enable the feature, use one of the following:

Select "Enable CUDA memory checking" from the Startup Parameters dialog box in TotalView, as shown in Figure 255.

 

Figure 255: Enabling CUDA memory checking from TotalView Startup Parameters

Pass the -cuda_memcheck option to the totalview command.

Set the TV::cuda_memcheck CLI state variable to true. For example:

dset TV::cuda_memcheck true

Note that global memory violations and misaligned global memory accesses will be detected only while the CUDA thread is running. Detection will not happen when single-stepping the CUDA thread.