MIOpen

MIOpen is AMD's library for high-performance machine learning primitives.

You can find sources and binaries in our GitHub repository and our documentation on ROCm docs.

MIOpen supports these programming models (backends):

• HIP
• OpenCL (deprecated)

Building our documentation

To build the MIOpen documentation locally, run the following code from within the docs folder in our repository:

cd docs

pip3 install -r sphinx/requirements.txt

python3 -m sphinx -T -E -b html -d _build/doctrees -D language=en . _build/html

Installing MIOpen

To install MIOpen, you must first install these prerequisites:

• A ROCm-enabled platform
• A base software stack that includes either: *HIP (HIP and HCC libraries and header files)
  • OpenCL (OpenCL libraries and header files)--this is now deprecated
• ROCm CMake: provides CMake modules for common build tasks needed for the ROCm software stack
• Half: IEEE 754-based, half-precision floating-point library
• Boost: Version 1.79 is recommended, as older versions may need patches to work on newer systems
  • MIOpen uses boost-system and boost-filesystem packages to enable persistent kernel cache
• SQLite3: A reading and writing performance database
• lbzip2: A multi-threaded compress or decompress utility
• rocBLAS: AMD's library for Basic Linear Algebra Subprograms (BLAS) on the ROCm platform.
  • Minimum version branch for pre-ROCm 3.5 master-rocm-2.10
  • Minimum version branch for post-ROCm 3.5 master-rocm-3.5
• Multi-Level Intermediate Representation (MLIR) with its MIOpen dialect to support and complement kernel development
• Composable Kernel: A C++ templated device library for GEMM-like and reduction-like operators.

Installing with pre-built packages

You can install MIOpen on Ubuntu using apt-get install miopen-hip.

If using OpenCL, you can use apt-get install miopen-opencl (but this is not recommended, as OpenCL is deprecated).

Note that you can't install both backends on the same system simultaneously. If you want a different backend other than what currently exists, completely uninstall the existing backend prior to installing the new backend.

Installing with a kernels package

MIOpen provides an optional pre-compiled kernels package to reduce startup latency. These precompiled kernels comprise a select set of popular input configurations. We'll expand these kernels in future releases to include additional coverage.

Note that all compiled kernels are locally cached in the $HOME/.cache/miopen/ folder, so precompiled kernels reduce the startup latency only for the first run of a neural network. Precompiled kernels don't reduce startup time on subsequent runs.

To install the kernels package for your GPU architecture, use the following command:

apt-get install miopenkernels-<arch>-<num cu>

Where <arch> is the GPU architecture (e.g., gfx900, gfx906, gfx1030 ) and <num cu> is the number of CUs available in the GPU (e.g., 56, 64).

Note

Not installing these packages doesn't impact the functioning of MIOpen, since MIOpen compiles them on the target machine once you run the kernel. However, the compilation step may significantly increase the startup time for different operations.

The utils/install_precompiled_kernels.sh script provided as part of MIOpen automates the preceding process. It queries the user machine for the GPU architecture and then installs the appropriate package. You can invoke it using:

./utils/install_precompiled_kernels.sh

The preceding script depends on the rocminfo package to query the GPU architecture. Refer to Installing pre-compiled kernels for more information.

Installing dependencies

You can install dependencies using the install_deps.cmake script (cmake -P install_deps.cmake).

By default, this installs to /usr/local, but you can specify another location using the --prefix argument:

cmake -P install_deps.cmake --prefix <miopen-dependency-path>

An example CMake step is:

cmake -P install_deps.cmake --minimum --prefix /root/MIOpen/install_dir

You can use this prefix to specify the dependency path during the configuration phase using CMAKE_PREFIX_PATH.

MIOpen's HIP backend uses rocBLAS by default. You can install rocBLAS' minimum release using apt-get install rocblas. To disable rocBLAS, set the configuration flag -DMIOPEN_USE_ROCBLAS=Off. rocBLAS is not available with OpenCL.

Building MIOpen from source

You can build MIOpen form source with a HIP backend or an OpenCL backend.

HIP backend

First, create a build directory:

mkdir build; cd build;

Next, configure CMake. You can set the backend using the -DMIOPEN_BACKEND CMake variable.

Set the C++ compiler to clang++. For the HIP backend (ROCm 3.5 and later), run:

export CXX=<location-of-clang++-compiler>
cmake -DMIOPEN_BACKEND=HIP -DCMAKE_PREFIX_PATH="<hip-installed-path>;<rocm-installed-path>;<miopen-dependency-path>" ..

An example CMake step is:

export CXX=/opt/rocm/llvm/bin/clang++ && \
cmake -DMIOPEN_BACKEND=HIP -DCMAKE_PREFIX_PATH="/opt/rocm/;/opt/rocm/hip;/root/MIOpen/install_dir" ..

Note

When specifying the path for the CMAKE_PREFIX_PATH variable, do not use the tilde (~) shorthand to represent the home directory.

OpenCL backend

Note

OpenCL is deprecated. We recommend using a HIP backend and following the instructions listed in the preceding section.

First, run:

cmake -DMIOPEN_BACKEND=OpenCL ..

The preceding code assumes OpenCL is installed in one of the standard locations. If not, then manually set these CMake variables:

cmake -DMIOPEN_BACKEND=OpenCL -DMIOPEN_HIP_COMPILER=<hip-compiler-path> -DOPENCL_LIBRARIES=<opencl-library-path> -DOPENCL_INCLUDE_DIRS=<opencl-headers-path> ..

Here's an example dependency path for an environment in ROCm 3.5 and later:

cmake -DMIOPEN_BACKEND=OpenCL -DMIOPEN_HIP_COMPILER=/opt/rocm/llvm/bin/clang++ -DCMAKE_PREFIX_PATH="/opt/rocm/;/opt/rocm/hip;/root/MIOpen/install_dir" ..

Setting up locations

By default, the install location is set to /opt/rocm. You can change this using CMAKE_INSTALL_PREFIX:

cmake -DMIOPEN_BACKEND=HIP -DCMAKE_INSTALL_PREFIX=<miopen-installed-path> ..

System performance database and user database

The default path to the system performance database (System PerfDb) is miopen/share/miopen/db/ within the install location. The default path to the user performance database (User PerfDb) is ~/.config/miopen/. For development purposes, setting BUILD_DEV changes the default path to both database files to the source directory:

cmake -DMIOPEN_BACKEND=HIP -DBUILD_DEV=On ..

Database paths can be explicitly customized using the MIOPEN_SYSTEM_DB_PATH (System PerfDb) and MIOPEN_USER_DB_PATH (User PerfDb) CMake variables.

To learn more, refer to the performance database documentation.

Persistent program cache

By default, MIOpen caches device programs in the ~/.cache/miopen/ directory. Within the cache directory, there is a directory for each version of MIOpen. You can change the location of the cache directory during configuration using the -DMIOPEN_CACHE_DIR=<cache-directory-path> flag.

You can also disable the cache during runtime using the MIOPEN_DISABLE_CACHE=1 environmental variable.

For MIOpen version 2.3 and earlier

If the compiler changes, or you modify the kernels, then you must delete the cache for the MIOpen version in use (e.g., rm -rf ~/.cache/miopen/<miopen-version-number>). You can find more information in the cache documentation.

For MIOpen version 2.4 and later

MIOpen's kernel cache directory is versioned so that your cached kernels won't collide when upgrading from an earlier version.

Changing the CMake configuration

The configuration can be changed after running CMake (using ccmake):

ccmake .. or cmake-gui: cmake-gui ..

The ccmake program can be downloaded as a Linux package (cmake-curses-gui), but is not available on Windows.

Building the library

You can build the library from the build directory using the 'Release' configuration:

cmake --build . --config Release or make

You can install it using the 'install' target:

cmake --build . --config Release --target install or make install

This installs the library to the CMAKE_INSTALL_PREFIX path that you specified.

Building the driver

MIOpen provides an application-driver that you can use to run any layer in isolation, and measure library performance and verification.

You can build the driver using the MIOpenDriver target:

cmake --build . --config Release --target MIOpenDriver or make MIOpenDriver

To learn more, refer to the driver documentation.

Running the tests

You can run tests using the 'check' target:

cmake --build . --config Release --target check OR make check

To build and run a single test, use the following code:

cmake --build . --config Release --target test_tensor
./bin/test_tensor

Formatting the code

All the code is formatted using clang-format. To format a file, use:

clang-format-10 -style=file -i <path-to-source-file>

To format the code per commit, you can install githooks:

./.githooks/install

Storing large file using Git Large File Storage

Git Large File Storage (LFS) replaces large files, such as audio samples, videos, datasets, and graphics with text pointers inside Git, while storing the file contents on a remote server. In MIOpen, we use Gi LFS to store our large files, such as our kernel database files (*.kdb) that are normally > 0.5 GB.

You can install Git LFS using the following code:

sudo apt install git-lfs
git lfs install

In the Git repository where you want to use Git LFS, track the file type using the following code (if the file type has already been tracked, you can skip this step):

git lfs track "*.file_type"
git add .gitattributes

You can pull all or a single large file using:

git lfs pull --exclude=
or
git lfs pull --exclude= --include "filename"

Update the large files and push to GitHub using:

git add my_large_files
git commit -m "the message"
git push

Installing the dependencies manually

If you're using Ubuntu v16, you can install the Boost packages using:

sudo apt-get install libboost-dev
sudo apt-get install libboost-system-dev
sudo apt-get install libboost-filesystem-dev

Note

By default, MIOpen attempts to build with Boost statically linked libraries. If required, you can build with dynamically linked Boost libraries using the -DBoost_USE_STATIC_LIBS=Off flag during the configuration stage. However, this is not recommended.

You must install the half header from the half website.

Using Docker

The easiest way to build MIOpen is via Docker. You can build the top-level Docker file using:

docker build -t miopen-image .

Then, to enter the development environment, use docker run. For example:

docker run -it -v $HOME:/data --privileged --rm --device=/dev/kfd --device /dev/dri:/dev/dri:rw  --volume /dev/dri:/dev/dri:rw -v /var/lib/docker/:/var/lib/docker --group-add video --cap-add=SYS_PTRACE --security-opt seccomp=unconfined miopen-image

You can find prebuilt Docker images on ROCm's public Docker Hub.

Porting from cuDNN to MIOpen

Our porting guide highlights the key differences between cuDNN and MIOpen APIs.