ROCm Systems Profiler documentation

ROCm Systems Profiler is designed for the high-level profiling and comprehensive tracing of applications running on the CPU or the CPU and GPU. It supports dynamic binary instrumentation, call-stack sampling, and various other features for determining which function and line number are currently executing. To learn more, see What is ROCm Systems Profiler?

ROCm Systems Profiler is open source and hosted at ROCm/rocprofiler-systems. It is the successor to ROCm/omnitrace.

Install

• Quick start

• ROCm Systems Profiler installation

Use the following topics to learn more about the advantages of ROCm Systems Profiler in application profiling, how it supports performance analysis, and how to leverage its capabilities in practice:

How to

• Configuring and validating the ROCm Systems Profiler environment

• Configuring runtime options

• Sampling the call stack

• Instrumenting and rewriting a binary application

• Performing causal profiling

• Understanding the ROCm Systems Profiler output

• Profiling Python scripts

• Using the ROCm Systems Profiler API

• General tips for using ROCm Systems Profiler

Conceptual

• Data collection modes

• Features and use cases

Reference

• Development guide

• Glossary

• API library

• Class member functions

• Globals

• Classes, structures, and interfaces

Tutorials

• GitHub examples

• Video tutorials

To contribute to the documentation, refer to Contributing to ROCm.

You can find licensing information on the Licensing page.