Omnitrace: Application Profiling, Tracing, and Analysis

Omnitrace is an AMD open source research project and is not supported as part of the ROCm software stack.

Documentation

The full documentation for omnitrace is available at amdresearch.github.io/omnitrace.

Quick Start

Omnitrace Settings

omnitrace-avail -Sd will provide a list of all the possible omnitrace settings, their current value, and a description of the setting when running an instrumented binary.

Omnitrace Executable

The omnitrace executable is used to instrument an existing binary.

omnitrace --help
omnitrace <omnitrace-options> -- <exe-or-library> <exe-options>

Binary Rewrite

Rewrite the text section of an executable or library with instrumentation:

omnitrace -o app.inst -- /path/to/app

In binary rewrite mode, if you also want instrumentation in the linked libraries, you must also rewrite those libraries. Example of rewriting the functions starting with "hip" with instrumentation in the amdhip64 library:

mkdir -p ./lib
omnitrace -R '^hip' -o ./lib/libamdhip64.so.4 -- /opt/rocm/lib/libamdhip64.so.4
export LD_LIBRARY_PATH=${PWD}/lib:${LD_LIBRARY_PATH}

Verify via ldd that your executable will load the instrumented library -- if you built your executable with an RPATH to the original library's directory, then prefixing LD_LIBRARY_PATH will have no effect.

Once you have rewritten your executable and/or libraries with instrumentation, you can just run the (instrumented) executable or exectuable which loads the instrumented libraries normally, e.g.:

./app.inst

If you want to re-define certain settings to new default in a binary rewrite, use the --env option. This omnitrace option will set the environment variable to the given value but will not override it. E.g. the default value of OMNITRACE_PERFETTO_BUFFER_SIZE_KB is 1024000 KB (1 GiB):

# buffer size defaults to 1024000
omnitrace -o app.inst -- /path/to/app
./app.inst

Passing --env OMNITRACE_PERFETTO_BUFFER_SIZE_KB=5120000 will change the default value in app.inst to 5120000 KiB (5 GiB):

# defaults to 5 GiB buffer size
omnitrace -o app.inst --env OMNITRACE_PERFETTO_BUFFER_SIZE_KB=5120000 -- /path/to/app
./app.inst

# override default 5 GiB buffer size to 200 MB
export OMNITRACE_PERFETTO_BUFFER_SIZE_KB=200000
./app.inst

Runtime Instrumentation

Runtime instrumentation will not only instrument the text section of the executable but also the text sections of the linked libraries. Thus, it may be useful to exclude those libraries via the -ME (module exclude) regex option or exclude specific functions with the -E regex option.

omnitrace -- /path/to/app
omnitrace -ME '^(libhsa-runtime64|libz\\.so)' -- /path/to/app
omnitrace -E 'rocr::atomic|rocr::core|rocr::HSA' --  /path/to/app

Visualizing Perfetto Results

Visit ui.perfetto.dev in your browser and open up the .proto file(s) created by omnitrace.

Merging the traces from rocprof and omnitrace

This section requires installing Julia.

Installing Julia

Julia is available via Linux package managers or may be available via a module. Debian-based distributions such as Ubuntu can run (as a super-user):

apt-get install julia

Once Julia is installed, install the necessary packages (this operation only needs to be performed once):

julia -e 'using Pkg; for name in ["JSON", "DataFrames", "Dates", "CSV", "Chain", "PrettyTables"]; Pkg.add(name); end'

Using rocprof externally for tracing is deprecated. The current version has built-in support for recording the GPU activity and HIP API calls. If you want to use an external rocprof, either configure CMake with -DOMNITRACE_USE_ROCTRACER=OFF or explicitly set OMNITRACE_ROCTRACER_ENABLED=OFF in the environment.

Use the omnitrace-merge.jl Julia script to merge rocprof and perfetto traces.

export OMNITRACE_USE_ROCTRACER=OFF
rocprof --hip-trace --roctx-trace --stats ./app.inst
omnitrace-merge.jl results.json omnitrace-app.inst-output/2021-09-02_01.03_PM/*.proto

Use Perfetto tracing with System Backend

Enable traced and perfetto in the background:

pkill traced
traced --background
perfetto --out ./omnitrace-perfetto.proto --txt -c ${OMNITRACE_ROOT}/share/omnitrace.cfg --background

Configure omnitrace to use the perfetto system backend:

export OMNITRACE_PERFETTO_BACKEND=system

And finally, execute your instrumented application. Either the binary rewritten application:

omnitrace -o ./myapp.inst -- ./myapp
./myapp.inst

Or with runtime instrumentation:

omnitrace -- ./myapp