Introduction

The Radeon Compute Profiler (RCP) is a performance analysis tool that gathers data from the API run-time and GPU for OpenCL™ and ROCm/HSA applications. This information can be used by developers to discover bottlenecks in the application and to find ways to optimize the application’s performance.

Features of the Radeon Compute Profiler include:

  • Measure the execution time of an OpenCL or ROCm/HSA kernel.
  • Query the hardware performance counters on an AMD Radeon™ graphics card.
  • Use the AMDTActivityLogger API to trace and measure the execution of segments in the program.
  • Store the profile data for each run in a text file.
  • Display the IL/HSAIL and ISA (hardware disassembly) code of the kernel for OpenCL kernels.
  • Show a timeline (including data transfer and kernel dispatch) and an API trace for OpenCL or HSA programs (when used in conjunction with the CodeXL user interface).
  • Calculate and display kernel occupancy info, which estimates the number of in-flight wavefronts on a compute unit as a percentage of the theoretical maximum number of wavefronts that the compute unit can support.