NVIDIA® CUDA Profiler Tools Interface (CUPTI) for CUDA Toolkit 10.2 is now available

The NVIDIA® CUDA Profiler Tools Interface (CUPTI) is a library that enables the creation of profiling and tracing tools that target CUDA applications. CUPTI provides a set of APIs targeted at ISVs creating profilers and other performance optimization tools:

  • the Activity API,
  • the Callback API,
  • the Event API,
  • the Metric API, and
  • the Profiler API

Using these CUPTI APIs, independent software developers can create profiling tools which give insight into the CPU and GPU behavior of CUDA applications.

CUPTI for CUDA Toolkit 10.2 includes these improvements:

  • CUPTI allows tracing features for non-root and non-admin users on desktop platforms. Note that event and metric profiling continues to have security restrictions for non-root and non-admin users. More details about the issue and the solutions can be found on the GPU Counter Permissions Information page: https://developer.nvidia.com/nvidia-development-tools-solutions-ERR_NVGPUCTRPERM-permission-issue-performance-counters
  • Added support for tracing features on the virtual GPUs (vGPU).
  • CUPTI no longer turns off the performance characteristics of CUDA Graph when tracing the application.
  • CUPTI now shows memset nodes in the CUDA graph.
  • Fixed the incorrect timing issue for the asynchronous cuMemset/cudaMemset activity.
  • Several performance improvements are done in the tracing path.

For more information on CUPTI for CUDA Toolkit 10.2, including features, requirements, documentation and support, please visit https://developer.nvidia.com/CUPTI.

To download this version, get it as part of the CUDA Toolkit 10.2 from https://developer.nvidia.com/cuda-toolkit.