From e841b6608fdbbbf5665cbee399df982ae7112dec Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Cheen=20Hau=2C=20=E4=BF=8A=E8=B1=AA?= <33478814+chtanch@users.noreply.github.com> Date: Wed, 7 Feb 2024 21:14:15 +0800 Subject: [PATCH] Enhance gpu doc with PIP install oneAPI (#10109) * Add pip install oneapi instructions * Fixes * Add instruction for oneapi2023 * Runtime config * Fixes * Remove "Currently, oneAPI installed with .. " * Add pip package version for oneAPI 2024 * Reviewer comments * Fix errors --- .../source/doc/LLM/Overview/install_gpu.md | 72 +++++++++++++++++-- 1 file changed, 68 insertions(+), 4 deletions(-) diff --git a/docs/readthedocs/source/doc/LLM/Overview/install_gpu.md b/docs/readthedocs/source/doc/LLM/Overview/install_gpu.md index 261c0a302e3..2cbfc70de36 100644 --- a/docs/readthedocs/source/doc/LLM/Overview/install_gpu.md +++ b/docs/readthedocs/source/doc/LLM/Overview/install_gpu.md @@ -172,6 +172,38 @@ BigDL-LLM for GPU supports on Linux has been verified on: Intel® oneAPI Base Toolkit 2024.0 installation methods: .. tabs:: + .. tab:: PIP installer + + Step 1: Install oneAPI in a user-defined folder, e.g., ``~/intel/oneapi``. + + .. code-block:: bash + + export PYTHONUSERBASE=~/intel/oneapi + pip install dpcpp-cpp-rt==2024.0.2 mkl-dpcpp==2024.0.0 onednn==2024.0.0 --user + + .. note:: + + The oneAPI packages are visible in ``pip list`` only if ``PYTHONUSERBASE`` is properly set. + + Step 2: Configure your working conda environment (e.g. with name ``llm``) to append ~/intel/oneapi/lib to environment variable LD_LIBRARY_PATH + + .. code-block:: bash + + conda env config vars set LD_LIBRARY_PATH=$LD_LIBRARY_PATH:~/intel/oneapi/lib -n llm + + .. note:: + You can view the configured environment variables for ``llm`` by running ``conda env config vars list -n llm``. + You can continue with activating the conda environment ``llm`` and installing ``bigdl-llm``. + + .. note:: + + You can uninstall the package by simply deleting the package folder, and unsetting the configuration of your working conda environment (e.g., with name ``llm``). Thus, you are recommended to not install other pip packages in this folder. + + .. code-block:: bash + + rm -r ~/intel/oneapi + conda env config vars unset LD_LIBRARY_PATH -n llm + .. tab:: APT installer Step 1: Set up repository @@ -232,6 +264,38 @@ BigDL-LLM for GPU supports on Linux has been verified on: Intel® oneAPI Base Toolkit 2023.2 installation methods: .. tabs:: + .. tab:: PIP installer + + Step 1: Install oneAPI in a user-defined folder, e.g., ``~/intel/oneapi`` + + .. code-block:: bash + + export PYTHONUSERBASE=~/intel/oneapi + pip install dpcpp-cpp-rt==2023.2.0 mkl-dpcpp==2023.2.0 onednn-cpu-dpcpp-gpu-dpcpp==2023.2.0 --user + + .. note:: + + The oneAPI packages are visible in ``pip list`` only if ``PYTHONUSERBASE`` is properly set. + + Step 2: Configure your working conda environment (e.g. with name ``llm``) to append ~/intel/oneapi/lib to environment variable LD_LIBRARY_PATH + + .. code-block:: bash + + conda env config vars set LD_LIBRARY_PATH=$LD_LIBRARY_PATH:~/intel/oneapi/lib -n llm + + .. note:: + You can view the configured environment variables for ``llm`` by running ``conda env config vars list -n llm``. + You can continue with activating the conda environment ``llm`` and installing ``bigdl-llm``. + + .. note:: + + You can uninstall the package by simply deleting the package folder, and unsetting the configuration of your working conda environment (e.g., with name ``llm``). Thus, you are recommended to not install other pip packages in this folder. + + .. code-block:: bash + + rm -r ~/intel/oneapi + conda env config vars unset LD_LIBRARY_PATH -n llm + .. tab:: APT installer Step 1: Set up repository @@ -397,7 +461,7 @@ To use GPU acceleration on Linux, several environment variables are required or .. code-block:: bash - # Required step. Configure oneAPI environment variables + # Required step for APT or offline installed oneAPI. Configure oneAPI environment variables. Skip this step for pip-installed oneAPI since LD_LIBRARY_PATH has already been configured. source /opt/intel/oneapi/setvars.sh # Recommended Environment Variables @@ -410,8 +474,7 @@ To use GPU acceleration on Linux, several environment variables are required or .. code-block:: bash - # Required step. Configure oneAPI environment variables - source /opt/intel/oneapi/setvars.sh + # Required step for APT or offline installed oneAPI. Configure oneAPI environment variables. Skip this step for pip-installed oneAPI since LD_LIBRARY_PATH has already been configured. # Recommended Environment Variables export LD_PRELOAD=${LD_PRELOAD}:${CONDA_PREFIX}/lib/libtcmalloc.so @@ -457,5 +520,6 @@ Error: libmkl_sycl_blas.so.4: cannot open shared object file: No such file or di The reason for such errors is that oneAPI has not been initialized properly before running BigDL-LLM code or before importing IPEX package. -* Step 1: Make sure you execute setvars.sh of oneAPI Base Toolkit before running BigDL-LLM code. +* Step 1: For oneAPI installed using APT or Offline Installer, make sure you execute setvars.sh of oneAPI Base Toolkit before running BigDL-LLM code. +For PIP-installed oneAPI, run ``echo $LD_LIBRARY_PATH``. If the output does not contain ``/lib``, check `Prerequisites`_. to install oneAPI with PIP installer again * Step 2: Make sure you install matching versions of BigDL-LLM/pytorch/IPEX and oneAPI Base Toolkit. BigDL-LLM with PyTorch 2.1 should be used with oneAPI Base Toolkit version 2024.0. BigDL-LLM with PyTorch 2.0 should be used with oneAPI Base Toolkit version 2023.2.