Skip to content

Instantly share code, notes, and snippets.

@zhanwenchen

zhanwenchen/magma-readme.md

Forked from Birch-san/magma-readme.md
Last active August 9, 2023 20:53
Show Gist options
  • Save zhanwenchen/ada29fd8a44ee69d3f766a95ac317b21 to your computer and use it in GitHub Desktop.
Save zhanwenchen/ada29fd8a44ee69d3f766a95ac317b21 to your computer and use it in GitHub Desktop.
Download ZIP

Revisions

  1. zhanwenchen revised this gist Aug 9, 2023. 1 changed file with 1 addition and 1 deletion.
    2 changes: 1 addition & 1 deletion magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -53,7 +53,7 @@ export CONDA_LIB=${CONDA_PREFIX}/lib
    # be careful here; they didn't accept sm_89 so I had to round it down to major version, sm_80
    make clean && rm -rf build/

    TARGETARCH=amd64 cmake -H. -Bbuild -DUSE_FORTRAN=OFF -DGPU_TARGET="Ampere" -DBUILD_SHARED_LIBS=OFF -DCMAKE_CXX_FLAGS="-fPIC" -DCMAKE_C_FLAGS="-fPIC" -DMKLROOT=${CONDA_PREFIX} -DCUDA_NVCC_FLAGS="-Xfatbin;-compress-all;-DHAVE_CUBLAS;-std=c++11;--threads=0;" -GNinja
    TARGETARCH=amd64 cmake -H. -Bbuild -DUSE_FORTRAN=OFF -DGPU_TARGET="Ampere" -DBUILD_SHARED_LIBS=ON -DCMAKE_CXX_FLAGS="-fPIC" -DCMAKE_C_FLAGS="-fPIC" -DMKLROOT=${CONDA_PREFIX} -DCUDA_NVCC_FLAGS="-Xfatbin;-compress-all;-DHAVE_CUBLAS;-std=c++11;--threads=0;" -GNinja

    sudo mkdir /usr/local/magma/
    ```
  2. zhanwenchen revised this gist Aug 9, 2023. 1 changed file with 0 additions and 1 deletion.
    1 change: 0 additions & 1 deletion magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -49,7 +49,6 @@ make generate
    export LD_LIBRARY_PATH="${LD_LIBRARY_PATH:+${LD_LIBRARY_PATH}:}${CONDA_PREFIX}/lib:/usr/local/cuda/targets/x86_64-linux/lib"
    export CUDA_DIR="/usr/local/cuda-12.2"
    export CONDA_LIB=${CONDA_PREFIX}/lib
    #export LD_PRELOAD="${CONDA_PREFIX}/lib/libiomp5.so"

    # be careful here; they didn't accept sm_89 so I had to round it down to major version, sm_80
    make clean && rm -rf build/
  3. zhanwenchen revised this gist Aug 9, 2023. 1 changed file with 2 additions and 0 deletions.
    2 changes: 2 additions & 0 deletions magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -72,12 +72,14 @@ Install Magma:
    ```bash
    cp build/include/* ${CONDA_PREFIX}/include/
    cp build/lib/*.so ${CONDA_PREFIX}/lib/
    cp build/lib/*.a ${CONDA_PREFIX}/lib/
    cp build/lib/pkgconfig/*.pc ${CONDA_PREFIX}/lib/pkgconfig/
    sudo cp /usr/local/magma/include/* ${CONDA_PREFIX}/include/
    sudo cp /usr/local/magma/lib/pkgconfig/*.pc ${CONDA_PREFIX}/lib/pkgconfig/

    sudo cp build/include/* /usr/local/include/
    sudo cp build/lib/*.so /usr/local/lib/
    sudo cp build/lib/*.a /usr/local/lib/
    sudo cp build/lib/pkgconfig/*.pc /usr/lib/pkgconfig
    sudo cp /usr/local/magma/include/* /usr/local/include/
    sudo cp /usr/local/magma/lib/pkgconfig/*.pc /usr/lib/pkgconfig
  4. zhanwenchen revised this gist Aug 9, 2023. No changes.
  5. zhanwenchen revised this gist Aug 9, 2023. 1 changed file with 1 addition and 1 deletion.
    2 changes: 1 addition & 1 deletion magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -65,7 +65,7 @@ mkl_intel_thread
    Build Magma:
    ```bash
    #sudo cmake --build build --target install && rm -r ./build # buildkit"
    cmake --build build -j $(nproc) --target install
    sudo cmake --build build -j $(nproc) --target install
    ```

    Install Magma:
  6. zhanwenchen revised this gist Aug 9, 2023. 1 changed file with 1 addition and 2 deletions.
    3 changes: 1 addition & 2 deletions magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -57,15 +57,14 @@ make clean && rm -rf build/
    TARGETARCH=amd64 cmake -H. -Bbuild -DUSE_FORTRAN=OFF -DGPU_TARGET="Ampere" -DBUILD_SHARED_LIBS=OFF -DCMAKE_CXX_FLAGS="-fPIC" -DCMAKE_C_FLAGS="-fPIC" -DMKLROOT=${CONDA_PREFIX} -DCUDA_NVCC_FLAGS="-Xfatbin;-compress-all;-DHAVE_CUBLAS;-std=c++11;--threads=0;" -GNinja

    sudo mkdir /usr/local/magma/

    sudo cmake --build build --target install && rm -r ./build # buildkit"
    ```

    mkl_intel_thread


    Build Magma:
    ```bash
    #sudo cmake --build build --target install && rm -r ./build # buildkit"
    cmake --build build -j $(nproc) --target install
    ```

  7. zhanwenchen revised this gist Aug 9, 2023. 1 changed file with 1 addition and 1 deletion.
    2 changes: 1 addition & 1 deletion magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -58,7 +58,7 @@ TARGETARCH=amd64 cmake -H. -Bbuild -DUSE_FORTRAN=OFF -DGPU_TARGET="Ampere" -DBUI

    sudo mkdir /usr/local/magma/

    cmake --build build --target install && rm -r ./build # buildkit"
    sudo cmake --build build --target install && rm -r ./build # buildkit"
    ```

    mkl_intel_thread
  8. zhanwenchen revised this gist Aug 9, 2023. 1 changed file with 0 additions and 5 deletions.
    5 changes: 0 additions & 5 deletions magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -39,8 +39,6 @@ conda install -c intel intelpython3_full python=3.10 mkl-dpcpp mkl-include cmake

    conda install astunparse expecttest hypothesis numpy psutil pyyaml requests setuptools typing-extensions sympy filelock networkx jinja2 fsspec
    pip install types-dataclasses


    ```

    Configure Magma (see [README](https://bitbucket.org/icl/magma/src/master/README)):
    @@ -56,9 +54,6 @@ export CONDA_LIB=${CONDA_PREFIX}/lib
    # be careful here; they didn't accept sm_89 so I had to round it down to major version, sm_80
    make clean && rm -rf build/

    #TARGETARCH=amd64 cmake -H. -Bbuild -DUSE_FORTRAN=OFF -DGPU_TARGET="Ampere" -DCMAKE_CUDA_COMPILER="$CUDA_DIR/bin/nvcc" -DMKLROOT=${CONDA_PREFIX} -DLIBS="${CONDA_LIB}/libmkl_intel_lp64.so;${CONDA_LIB}/libmkl_intel_thread.so;${CONDA_LIB}/libmkl_core.so;${CONDA_LIB}/libiomp5.so;-lm;-ldl" -DLAPACK_LIBRARIES=${CONDA_LIB}/libmkl_intel_lp64.so;${CONDA_LIB}/libmkl_intel_thread.so;/${CONDA_LIB}/libmkl_core.so;${CONDA_LIB}/libiomp5.so;-lm;-ldl;-lm;-ldl" -DCUDA_NVCC_FLAGS="-Xfatbin;-compress-all;-DHAVE_CUBLAS;-std=c++11;--threads=0;" -GNinja


    TARGETARCH=amd64 cmake -H. -Bbuild -DUSE_FORTRAN=OFF -DGPU_TARGET="Ampere" -DBUILD_SHARED_LIBS=OFF -DCMAKE_CXX_FLAGS="-fPIC" -DCMAKE_C_FLAGS="-fPIC" -DMKLROOT=${CONDA_PREFIX} -DCUDA_NVCC_FLAGS="-Xfatbin;-compress-all;-DHAVE_CUBLAS;-std=c++11;--threads=0;" -GNinja

    sudo mkdir /usr/local/magma/
  9. zhanwenchen revised this gist Aug 9, 2023. 1 changed file with 3 additions and 8 deletions.
    11 changes: 3 additions & 8 deletions magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -32,18 +32,13 @@ echo "deb [signed-by=/usr/share/keyrings/oneapi-archive-keyring.gpg] https://apt

    sudo apt update

    #sudo apt install intel-basekit intel-aikit

    # Source the Intel oneAPI
    #. /opt/intel/oneapi/setvars.sh

    conda create -n ietrans
    conda activate ietrans
    conda config --env --add channels intel
    # As of writing, intel python 3.11 has not been released yet.
    conda install intelpython3_full python=3.10 mkl-dpcpp mkl-include cmake ninja intel-extension-for-pytorch intel-openmp intel-fortran-rt dpcpp-cpp-rt
    conda install -c intel intelpython3_full python=3.10 mkl-dpcpp mkl-include cmake ninja intel-extension-for-pytorch intel-openmp intel-fortran-rt dpcpp-cpp-rt

    conda install astunparse expecttest hypothesis numpy psutil pyyaml requests setuptools types-dataclasses typing-extensions sympy filelock networkx jinja2 fsspec
    conda install astunparse expecttest hypothesis numpy psutil pyyaml requests setuptools typing-extensions sympy filelock networkx jinja2 fsspec
    pip install types-dataclasses


    ```
  10. zhanwenchen revised this gist Aug 9, 2023. 1 changed file with 5 additions and 4 deletions.
    9 changes: 5 additions & 4 deletions magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -85,10 +85,11 @@ cp build/include/* ${CONDA_PREFIX}/include/
    cp build/lib/*.so ${CONDA_PREFIX}/lib/
    cp build/lib/pkgconfig/*.pc ${CONDA_PREFIX}/lib/pkgconfig/
    sudo cp /usr/local/magma/include/* ${CONDA_PREFIX}/include/
    #sudo cp /usr/local/magma/lib/*.so ${CONDA_PREFIX}/lib/
    sudo cp /usr/local/magma/lib/pkgconfig/*.pc ${CONDA_PREFIX}/lib/pkgconfig/

    #sudo cp build/target/include/* /usr/local/include/
    #sudo cp build/target/lib/*.so /usr/local/lib/
    #sudo cp build/target/lib/pkgconfig/*.pc /usr/local/lib/pkgconfig/
    sudo cp build/include/* /usr/local/include/
    sudo cp build/lib/*.so /usr/local/lib/
    sudo cp build/lib/pkgconfig/*.pc /usr/lib/pkgconfig
    sudo cp /usr/local/magma/include/* /usr/local/include/
    sudo cp /usr/local/magma/lib/pkgconfig/*.pc /usr/lib/pkgconfig
    ```
  11. zhanwenchen revised this gist Aug 9, 2023. 1 changed file with 1 addition and 6 deletions.
    7 changes: 1 addition & 6 deletions magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -50,15 +50,10 @@ conda install astunparse expecttest hypothesis numpy psutil pyyaml requests setu

    Configure Magma (see [README](https://bitbucket.org/icl/magma/src/master/README)):
    ```bash
    #export LD_LIBRARY_PATH=${HOME}/anaconda3/lib:${LD_LIBRARY_PATH}
    #export LD_PRELOAD=${HOME}/anaconda3/lib/libmkl_core.so:${HOME}/anaconda3/lib/libmkl_sequential.so


    echo -e "GPU_TARGET = sm_86\nBACKEND = cuda\nFORT = false" > make.inc
    make generate


    export LD_LIBRARY_PATH="${CONDA_PREFIX}/lib:/usr/local/cuda/targets/x86_64-linux/lib"
    export LD_LIBRARY_PATH="${LD_LIBRARY_PATH:+${LD_LIBRARY_PATH}:}${CONDA_PREFIX}/lib:/usr/local/cuda/targets/x86_64-linux/lib"
    export CUDA_DIR="/usr/local/cuda-12.2"
    export CONDA_LIB=${CONDA_PREFIX}/lib
    #export LD_PRELOAD="${CONDA_PREFIX}/lib/libiomp5.so"
  12. zhanwenchen revised this gist Aug 9, 2023. No changes.
  13. zhanwenchen revised this gist Aug 9, 2023. 1 changed file with 10 additions and 7 deletions.
    17 changes: 10 additions & 7 deletions magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -86,11 +86,14 @@ cmake --build build -j $(nproc) --target install

    Install Magma:
    ```bash
    cp build/target/include/* ${CONDA_PREFIX}/include/
    cp build/target/lib/*.so ${CONDA_PREFIX}/lib/
    cp build/target/lib/pkgconfig/*.pc ${CONDA_PREFIX}/lib/pkgconfig/

    sudo cp build/target/include/* /usr/local/include/
    sudo cp build/target/lib/*.so /usr/local/lib/
    sudo cp build/target/lib/pkgconfig/*.pc /usr/local/lib/pkgconfig/
    cp build/include/* ${CONDA_PREFIX}/include/
    cp build/lib/*.so ${CONDA_PREFIX}/lib/
    cp build/lib/pkgconfig/*.pc ${CONDA_PREFIX}/lib/pkgconfig/
    sudo cp /usr/local/magma/include/* ${CONDA_PREFIX}/include/
    #sudo cp /usr/local/magma/lib/*.so ${CONDA_PREFIX}/lib/
    sudo cp /usr/local/magma/lib/pkgconfig/*.pc ${CONDA_PREFIX}/lib/pkgconfig/

    #sudo cp build/target/include/* /usr/local/include/
    #sudo cp build/target/lib/*.so /usr/local/lib/
    #sudo cp build/target/lib/pkgconfig/*.pc /usr/local/lib/pkgconfig/
    ```
  14. zhanwenchen revised this gist Aug 9, 2023. 1 changed file with 14 additions and 4 deletions.
    18 changes: 14 additions & 4 deletions magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -61,13 +61,19 @@ make generate
    export LD_LIBRARY_PATH="${CONDA_PREFIX}/lib:/usr/local/cuda/targets/x86_64-linux/lib"
    export CUDA_DIR="/usr/local/cuda-12.2"
    export CONDA_LIB=${CONDA_PREFIX}/lib
    export LD_PRELOAD="${CONDA_PREFIX}/lib/libiomp5.so"
    #export LD_PRELOAD="${CONDA_PREFIX}/lib/libiomp5.so"

    # be careful here; they didn't accept sm_89 so I had to round it down to major version, sm_80
    make clean && rm -rf build && \
    cmake -DGPU_TARGET=Ampere -DCMAKE_CUDA_COMPILER="$CUDA_DIR/bin/nvcc" -DMKLROOT=${CONDA_PREFIX} -DLIBS="/localtmp/pct4et/conda_envs/ietrans/lib/libmkl_intel_lp64.so;/localtmp/pct4et/conda_envs/ietrans/lib/libmkl_intel_thread.so;/localtmp/pct4et/conda_envs/ietrans/lib/libmkl_core.so;/localtmp/pct4et/conda_envs/ietrans/lib/libiomp5.so;-lm;-ldl" -DLAPACK_LIBRARIES="/localtmp/pct4et/conda_envs/ietrans/lib/libmkl_intel_lp64.so;/localtmp/pct4et/conda_envs/ietrans/lib/libmkl_intel_thread.so;/localtmp/pct4et/conda_envs/ietrans/lib/libmkl_core.so;/localtmp/pct4et/conda_envs/ietrans/lib/libiomp5.so;-lm;-ldl;-lm;-ldl" . -Bbuild
    make clean && rm -rf build/


    #TARGETARCH=amd64 cmake -H. -Bbuild -DUSE_FORTRAN=OFF -DGPU_TARGET="Ampere" -DCMAKE_CUDA_COMPILER="$CUDA_DIR/bin/nvcc" -DMKLROOT=${CONDA_PREFIX} -DLIBS="${CONDA_LIB}/libmkl_intel_lp64.so;${CONDA_LIB}/libmkl_intel_thread.so;${CONDA_LIB}/libmkl_core.so;${CONDA_LIB}/libiomp5.so;-lm;-ldl" -DLAPACK_LIBRARIES=${CONDA_LIB}/libmkl_intel_lp64.so;${CONDA_LIB}/libmkl_intel_thread.so;/${CONDA_LIB}/libmkl_core.so;${CONDA_LIB}/libiomp5.so;-lm;-ldl;-lm;-ldl" -DCUDA_NVCC_FLAGS="-Xfatbin;-compress-all;-DHAVE_CUBLAS;-std=c++11;--threads=0;" -GNinja


    TARGETARCH=amd64 cmake -H. -Bbuild -DUSE_FORTRAN=OFF -DGPU_TARGET="Ampere" -DBUILD_SHARED_LIBS=OFF -DCMAKE_CXX_FLAGS="-fPIC" -DCMAKE_C_FLAGS="-fPIC" -DMKLROOT=${CONDA_PREFIX} -DCUDA_NVCC_FLAGS="-Xfatbin;-compress-all;-DHAVE_CUBLAS;-std=c++11;--threads=0;" -GNinja

    sudo mkdir /usr/local/magma/

    cmake --build build --target install && rm -r ./build # buildkit"
    ```

    mkl_intel_thread
    @@ -80,6 +86,10 @@ cmake --build build -j $(nproc) --target install

    Install Magma:
    ```bash
    cp build/target/include/* ${CONDA_PREFIX}/include/
    cp build/target/lib/*.so ${CONDA_PREFIX}/lib/
    cp build/target/lib/pkgconfig/*.pc ${CONDA_PREFIX}/lib/pkgconfig/

    sudo cp build/target/include/* /usr/local/include/
    sudo cp build/target/lib/*.so /usr/local/lib/
    sudo cp build/target/lib/pkgconfig/*.pc /usr/local/lib/pkgconfig/
  15. zhanwenchen revised this gist Aug 9, 2023. 1 changed file with 3 additions and 1 deletion.
    4 changes: 3 additions & 1 deletion magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -41,7 +41,9 @@ conda create -n ietrans
    conda activate ietrans
    conda config --env --add channels intel
    # As of writing, intel python 3.11 has not been released yet.
    conda install intelpython3_full python=3.10 mkl-dpcpp mkl-include cmake ninja intel-extension-for-pytorch intel-openmp intel-fortran-rt dpcpp-cpp-rt
    conda install intelpython3_full python=3.10 mkl-dpcpp mkl-include cmake ninja intel-extension-for-pytorch intel-openmp intel-fortran-rt dpcpp-cpp-rt

    conda install astunparse expecttest hypothesis numpy psutil pyyaml requests setuptools types-dataclasses typing-extensions sympy filelock networkx jinja2 fsspec


    ```
  16. zhanwenchen revised this gist Aug 7, 2023. 1 changed file with 7 additions and 2 deletions.
    9 changes: 7 additions & 2 deletions magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -56,16 +56,21 @@ echo -e "GPU_TARGET = sm_86\nBACKEND = cuda\nFORT = false" > make.inc
    make generate



    export LD_LIBRARY_PATH="${CONDA_PREFIX}/lib:/usr/local/cuda/targets/x86_64-linux/lib"
    export CUDA_DIR="/usr/local/cuda-12.2"
    export CONDA_LIB=${CONDA_PREFIX}/lib
    export LD_PRELOAD="${CONDA_PREFIX}/lib/libiomp5.so"

    # be careful here; they didn't accept sm_89 so I had to round it down to major version, sm_80
    make clean && rm -rf build && \
    cmake -DGPU_TARGET=Ampere -DCMAKE_CUDA_COMPILER="$CUDA_DIR/bin/nvcc" -DMKLROOT=${CONDA_PREFIX} -DCMAKE_INSTALL_PREFIX=build/target -DCMAKE_C_FLAGS="-liomp5 -fopenmp -lpthread -L${CONDA_PREFIX}/lib" -DCMAKE_CXX_FLAGS="-liomp5 -fopenmp -lpthread -L${CONDA_PREFIX}/lib" -DCMAKE_Fortran_FLAGS="-I${MKLROOT}/include -liomp5 -L${MKLROOT}/lib/ -lmkl_core -lmkl_gnu_thread -lmkl_gf_lp64" -DOpenMP_C_LIB_NAMES="libiomp5" -DOpenMP_CXX_LIB_NAMES="libiomp5" -DOpenMP_libiomp5_LIBRARY=${CONDA_PREFIX}/lib/libiomp5.so" . -Bbuild
    cmake -DGPU_TARGET=Ampere -DCMAKE_CUDA_COMPILER="$CUDA_DIR/bin/nvcc" -DMKLROOT=${CONDA_PREFIX} -DLIBS="/localtmp/pct4et/conda_envs/ietrans/lib/libmkl_intel_lp64.so;/localtmp/pct4et/conda_envs/ietrans/lib/libmkl_intel_thread.so;/localtmp/pct4et/conda_envs/ietrans/lib/libmkl_core.so;/localtmp/pct4et/conda_envs/ietrans/lib/libiomp5.so;-lm;-ldl" -DLAPACK_LIBRARIES="/localtmp/pct4et/conda_envs/ietrans/lib/libmkl_intel_lp64.so;/localtmp/pct4et/conda_envs/ietrans/lib/libmkl_intel_thread.so;/localtmp/pct4et/conda_envs/ietrans/lib/libmkl_core.so;/localtmp/pct4et/conda_envs/ietrans/lib/libiomp5.so;-lm;-ldl;-lm;-ldl" . -Bbuild


    ```

    mkl_intel_thread


    Build Magma:
    ```bash
    cmake --build build -j $(nproc) --target install
  17. zhanwenchen revised this gist Aug 7, 2023. 1 changed file with 2 additions and 1 deletion.
    3 changes: 2 additions & 1 deletion magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -62,7 +62,8 @@ export CUDA_DIR="/usr/local/cuda-12.2"
    export LD_PRELOAD="${CONDA_PREFIX}/lib/libiomp5.so"

    # be careful here; they didn't accept sm_89 so I had to round it down to major version, sm_80
    make clean && rm -rf build && cmake -DGPU_TARGET=Ampere -DCMAKE_CUDA_COMPILER="$CUDA_DIR/bin/nvcc" -DMKLROOT=${CONDA_PREFIX} -DCMAKE_INSTALL_PREFIX=build/target -DOpenMP_C_FLAGS= -DCMAKE_C_FLAGS="-liomp5 -lpthread -L${CONDA_PREFIX}/lib" -DCMAKE_CXX_FLAGS="-liomp5 -lpthread -L${CONDA_PREFIX}/lib" -DCMAKE_Fortran_FLAGS="-I${MKLROOT}/include -liomp5 -L${MKLROOT}/lib/ -lmkl_core -lmkl_gnu_thread -lmkl_gf_lp64" -DOpenMP_CXX_FLAGS="-qopenmp" -DOpenMP_CXX_LIB_NAMES="libiomp5" -DOpenMP_libiomp5_LIBRARY=${CONDA_PREFIX}/lib/libiomp5.so -DOpenMP_C_FLAGS="-qopenmp" -DOpenMP_C_LIB_NAMES="libiomp5" . -Bbuild
    make clean && rm -rf build && \
    cmake -DGPU_TARGET=Ampere -DCMAKE_CUDA_COMPILER="$CUDA_DIR/bin/nvcc" -DMKLROOT=${CONDA_PREFIX} -DCMAKE_INSTALL_PREFIX=build/target -DCMAKE_C_FLAGS="-liomp5 -fopenmp -lpthread -L${CONDA_PREFIX}/lib" -DCMAKE_CXX_FLAGS="-liomp5 -fopenmp -lpthread -L${CONDA_PREFIX}/lib" -DCMAKE_Fortran_FLAGS="-I${MKLROOT}/include -liomp5 -L${MKLROOT}/lib/ -lmkl_core -lmkl_gnu_thread -lmkl_gf_lp64" -DOpenMP_C_LIB_NAMES="libiomp5" -DOpenMP_CXX_LIB_NAMES="libiomp5" -DOpenMP_libiomp5_LIBRARY=${CONDA_PREFIX}/lib/libiomp5.so" . -Bbuild
    ```
    Build Magma:
  18. zhanwenchen revised this gist Aug 7, 2023. 1 changed file with 11 additions and 10 deletions.
    21 changes: 11 additions & 10 deletions magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -32,20 +32,16 @@ echo "deb [signed-by=/usr/share/keyrings/oneapi-archive-keyring.gpg] https://apt

    sudo apt update

    sudo apt install intel-basekit intel-aikit
    #sudo apt install intel-basekit intel-aikit

    # Source the Intel oneAPI
    . /opt/intel/oneapi/setvars.sh
    #. /opt/intel/oneapi/setvars.sh

    conda create -n ietrans
    conda activate ietrans
    conda config --env --add channels intel
    # As of writing, intel python 3.11 has not been released yet.
    conda install intelpython3_core python=3.10 mkl mkl-include
    export LD_LIBRARY_PATH=${CONDA_PREFIX}/lib
    export MKLROOT=${CONDA_PREFIX}
    export BLASLDFLAGS=-Wl,--no-as-needed -L$(MKLROOT)/lib -lmkl_intel_ilp64 \
    -lmkl_core -lmkl_intel_thread -ldl -lpthread -liomp5
    conda install intelpython3_full python=3.10 mkl-dpcpp mkl-include cmake ninja intel-extension-for-pytorch intel-openmp intel-fortran-rt dpcpp-cpp-rt


    ```
    @@ -58,10 +54,15 @@ Configure Magma (see [README](https://bitbucket.org/icl/magma/src/master/README)

    echo -e "GPU_TARGET = sm_86\nBACKEND = cuda\nFORT = false" > make.inc
    make generate
    export CUDA_DIR=/usr/local/cuda-12.2
    export LD_PRELOAD=



    export LD_LIBRARY_PATH="${CONDA_PREFIX}/lib:/usr/local/cuda/targets/x86_64-linux/lib"
    export CUDA_DIR="/usr/local/cuda-12.2"
    export LD_PRELOAD="${CONDA_PREFIX}/lib/libiomp5.so"

    # be careful here; they didn't accept sm_89 so I had to round it down to major version, sm_80
    cmake -DGPU_TARGET=Ampere -DCMAKE_CUDA_COMPILER="$CUDA_DIR/bin/nvcc" -DCMAKE_INSTALL_PREFIX=build/target . -Bbuild
    make clean && rm -rf build && cmake -DGPU_TARGET=Ampere -DCMAKE_CUDA_COMPILER="$CUDA_DIR/bin/nvcc" -DMKLROOT=${CONDA_PREFIX} -DCMAKE_INSTALL_PREFIX=build/target -DOpenMP_C_FLAGS= -DCMAKE_C_FLAGS="-liomp5 -lpthread -L${CONDA_PREFIX}/lib" -DCMAKE_CXX_FLAGS="-liomp5 -lpthread -L${CONDA_PREFIX}/lib" -DCMAKE_Fortran_FLAGS="-I${MKLROOT}/include -liomp5 -L${MKLROOT}/lib/ -lmkl_core -lmkl_gnu_thread -lmkl_gf_lp64" -DOpenMP_CXX_FLAGS="-qopenmp" -DOpenMP_CXX_LIB_NAMES="libiomp5" -DOpenMP_libiomp5_LIBRARY=${CONDA_PREFIX}/lib/libiomp5.so -DOpenMP_C_FLAGS="-qopenmp" -DOpenMP_C_LIB_NAMES="libiomp5" . -Bbuild
    ```

    Build Magma:
  19. zhanwenchen revised this gist Aug 7, 2023. 1 changed file with 18 additions and 2 deletions.
    20 changes: 18 additions & 2 deletions magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -32,18 +32,34 @@ echo "deb [signed-by=/usr/share/keyrings/oneapi-archive-keyring.gpg] https://apt

    sudo apt update

    sudo apt install intel-basekit
    sudo apt install intel-basekit intel-aikit

    # Source the Intel oneAPI
    . /opt/intel/oneapi/setvars.sh

    conda create -n ietrans
    conda activate ietrans
    conda config --env --add channels intel
    # As of writing, intel python 3.11 has not been released yet.
    conda install intelpython3_core python=3.10 mkl mkl-include
    export LD_LIBRARY_PATH=${CONDA_PREFIX}/lib
    export MKLROOT=${CONDA_PREFIX}
    export BLASLDFLAGS=-Wl,--no-as-needed -L$(MKLROOT)/lib -lmkl_intel_ilp64 \
    -lmkl_core -lmkl_intel_thread -ldl -lpthread -liomp5


    ```

    Configure Magma (see [README](https://bitbucket.org/icl/magma/src/master/README)):
    ```bash
    #export LD_LIBRARY_PATH=${HOME}/anaconda3/lib:${LD_LIBRARY_PATH}
    #export LD_PRELOAD=${HOME}/anaconda3/lib/libmkl_core.so:${HOME}/anaconda3/lib/libmkl_sequential.so


    echo -e "GPU_TARGET = sm_86\nBACKEND = cuda\nFORT = false" > make.inc
    make generate
    CUDA_DIR=/usr/local/cuda-12.2
    export CUDA_DIR=/usr/local/cuda-12.2
    export LD_PRELOAD=
    # be careful here; they didn't accept sm_89 so I had to round it down to major version, sm_80
    cmake -DGPU_TARGET=Ampere -DCMAKE_CUDA_COMPILER="$CUDA_DIR/bin/nvcc" -DCMAKE_INSTALL_PREFIX=build/target . -Bbuild
    ```
  20. zhanwenchen revised this gist Aug 7, 2023. 1 changed file with 3 additions and 0 deletions.
    3 changes: 3 additions & 0 deletions magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -34,6 +34,9 @@ sudo apt update

    sudo apt install intel-basekit

    # Source the Intel oneAPI
    . /opt/intel/oneapi/setvars.sh

    ```

    Configure Magma (see [README](https://bitbucket.org/icl/magma/src/master/README)):
  21. zhanwenchen revised this gist Aug 7, 2023. 1 changed file with 7 additions and 4 deletions.
    11 changes: 7 additions & 4 deletions magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -22,19 +22,22 @@ Install a Fortran compiler:
    sudo apt-get install gfortran
    ```

    Install Intel BLAS and LAPACK (https://www.intel.com/content/www/us/en/docs/oneapi/installation-guide-linux/2023-0/apt.html)
    Install Intel BLAS and LAPACK (https://www.intel.com/content/www/us/en/docs/oneapi/installation-guide-linux/2023-2/apt.html#GUID-186C17A8-4183-4BC3-B367-01331B1B74AF)
    ```bash
    wget -O- https://apt.repos.intel.com/intel-gpg-keys/GPG-PUB-KEY-INTEL-SW-PRODUCTS.PUB | gpg --dearmor | sudo tee /usr/share/keyrings/oneapi-archive-keyring.gpg > /dev/null
    --2023-08-07 00:02:33-- https://apt.repos.intel.com/intel-gpg-keys/GPG-PUB-KEY-INTEL-SW-PRODUCTS.PUB

    # Type sudo password

    echo "deb [signed-by=/usr/share/keyrings/oneapi-archive-keyring.gpg] https://apt.repos.intel.com/oneapi all main" | sudo tee /etc/apt/sources.list.d/oneAPI.list

    sudo apt update

    sudo apt install intel-basekit

    ```

    Configure Magma (see [README](https://bitbucket.org/icl/magma/src/master/README)):
    ```bash
    sudo apt install intel-basekit

    echo -e "GPU_TARGET = sm_86\nBACKEND = cuda\nFORT = false" > make.inc
    make generate
    CUDA_DIR=/usr/local/cuda-12.2
  22. zhanwenchen revised this gist Aug 7, 2023. 1 changed file with 13 additions and 10 deletions.
    23 changes: 13 additions & 10 deletions magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -1,12 +1,4 @@
    I wrote these instructions as part of ["installing PyTorch with CUDA 12.1.1"](https://gist.github.com/Birch-san/211f31f8d901dadd1025398fa1a603b8).
    I extracted them into this separate gist, because I realised there's a _much_ easier way to install magma for CUDA 12.1.1:
    https://anaconda.org/pytorch/magma-cuda121

    There's a conda package!

    ```bash
    conda install -c pytorch magma-cuda121
    ```
    I wrote these instructions as part of ["installing PyTorch with CUDA 12.1.1"](https://gist.github.com/Birch-san/211f31f8d901dadd1025398fa1a603b8).

    Anyway, if you still need to compile from source… here's how:

    @@ -30,13 +22,24 @@ Install a Fortran compiler:
    sudo apt-get install gfortran
    ```

    Install Intel BLAS and LAPACK (https://www.intel.com/content/www/us/en/docs/oneapi/installation-guide-linux/2023-0/apt.html)
    ```bash
    wget -O- https://apt.repos.intel.com/intel-gpg-keys/GPG-PUB-KEY-INTEL-SW-PRODUCTS.PUB | gpg --dearmor | sudo tee /usr/share/keyrings/oneapi-archive-keyring.gpg > /dev/null
    --2023-08-07 00:02:33-- https://apt.repos.intel.com/intel-gpg-keys/GPG-PUB-KEY-INTEL-SW-PRODUCTS.PUB

    # Type sudo password

    ```

    Configure Magma (see [README](https://bitbucket.org/icl/magma/src/master/README)):
    ```bash
    sudo apt install intel-basekit

    echo -e "GPU_TARGET = sm_86\nBACKEND = cuda\nFORT = false" > make.inc
    make generate
    CUDA_DIR=/usr/local/cuda-12.2
    # be careful here; they didn't accept sm_89 so I had to round it down to major version, sm_80
    cmake -DGPU_TARGET=sm_86 -DCMAKE_CUDA_COMPILER="$CUDA_DIR/bin/nvcc" -DCMAKE_INSTALL_PREFIX=build/target . -Bbuild
    cmake -DGPU_TARGET=Ampere -DCMAKE_CUDA_COMPILER="$CUDA_DIR/bin/nvcc" -DCMAKE_INSTALL_PREFIX=build/target . -Bbuild
    ```

    Build Magma:
  23. zhanwenchen revised this gist Aug 7, 2023. 1 changed file with 3 additions and 3 deletions.
    6 changes: 3 additions & 3 deletions magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -32,11 +32,11 @@ sudo apt-get install gfortran

    Configure Magma (see [README](https://bitbucket.org/icl/magma/src/master/README)):
    ```bash
    echo -e "GPU_TARGET = sm_89\nBACKEND = cuda\nFORT = false" > make.inc
    echo -e "GPU_TARGET = sm_86\nBACKEND = cuda\nFORT = false" > make.inc
    make generate
    CUDA_DIR=/usr/local/cuda-12.1
    CUDA_DIR=/usr/local/cuda-12.2
    # be careful here; they didn't accept sm_89 so I had to round it down to major version, sm_80
    cmake -DGPU_TARGET=sm_80 -DCMAKE_CUDA_COMPILER="$CUDA_DIR/bin/nvcc" -DCMAKE_INSTALL_PREFIX=build/target . -Bbuild
    cmake -DGPU_TARGET=sm_86 -DCMAKE_CUDA_COMPILER="$CUDA_DIR/bin/nvcc" -DCMAKE_INSTALL_PREFIX=build/target . -Bbuild
    ```

    Build Magma:
  24. @Birch-san Birch-san created this gist Apr 27, 2023.
    52 changes: 52 additions & 0 deletions magma-readme.md
    View file
    Original file line number Diff line number Diff line change
    @@ -0,0 +1,52 @@
    I wrote these instructions as part of ["installing PyTorch with CUDA 12.1.1"](https://gist.github.com/Birch-san/211f31f8d901dadd1025398fa1a603b8).
    I extracted them into this separate gist, because I realised there's a _much_ easier way to install magma for CUDA 12.1.1:
    https://anaconda.org/pytorch/magma-cuda121

    There's a conda package!

    ```bash
    conda install -c pytorch magma-cuda121
    ```

    Anyway, if you still need to compile from source… here's how:

    ## Build Magma from source

    This is a dependency of PyTorch, which is sensitive to CUDA version.

    Clone [Magma](https://bitbucket.org/icl/magma/src/master/):
    ```bash
    git clone --depth 1 https://bitbucket.org/icl/magma.git
    cd magma
    ```

    Look up your [CUDA Architecture](https://en.wikipedia.org/wiki/CUDA#GPUs_supported), e.g. 4090 has architecture 8.9. Use this to determine the two `GPU_TARGET` variables below.

    **Activate your conda environment, if you haven't done so already.**

    Install a Fortran compiler:

    ```bash
    sudo apt-get install gfortran
    ```

    Configure Magma (see [README](https://bitbucket.org/icl/magma/src/master/README)):
    ```bash
    echo -e "GPU_TARGET = sm_89\nBACKEND = cuda\nFORT = false" > make.inc
    make generate
    CUDA_DIR=/usr/local/cuda-12.1
    # be careful here; they didn't accept sm_89 so I had to round it down to major version, sm_80
    cmake -DGPU_TARGET=sm_80 -DCMAKE_CUDA_COMPILER="$CUDA_DIR/bin/nvcc" -DCMAKE_INSTALL_PREFIX=build/target . -Bbuild
    ```

    Build Magma:
    ```bash
    cmake --build build -j $(nproc) --target install
    ```

    Install Magma:
    ```bash
    sudo cp build/target/include/* /usr/local/include/
    sudo cp build/target/lib/*.so /usr/local/lib/
    sudo cp build/target/lib/pkgconfig/*.pc /usr/local/lib/pkgconfig/
    ```