AMD APP Profiler is a performance analysis tool that gathers data from the API run-time and GPU for OpenCL™ or DirectCompute applications.
This information can then be used by developers to discover the location of bottlenecks in the application and find ways to optimize the application’s performance. AMD APP Profiler can be used as a Microsoft Visual Studio plugin or command line tool.
- Measure the execution time of an OpenCL™/DirectCompute kernel.
- Query the hardware performance counters on an AMD Radeon™ graphics card.
- Compare multiple runs (sessions) of the same or different programs.
- Store the profile data for each run in a text file.
- Display the IL and ISA (hardware disassembly) code of the kernel for OpenCL™ kernels and DXASM code for DirectCompute kernels.
- Show a timeline (including data transfer and kernel dispatch) and an API trace for OpenCL™ programs.
- 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.
The following links provide an introduction to the AMD APP Profiler panels.
- APP Profiler Session Explorer
- APP Profiler Session
- APP Profiler Timeline
- APP Profiler Summary Pages
- APP Profiler Code Viewer
- APP Profiler Kernel Occupancy Viewer
- APP Profiler Settings
- APP Profiler Session Parameters
- Description of Configuration Files
- APP Profiler Kernel Occupancy
- CLPerfMarkerAMD Library