MLAutoScheduler

MLAutoScheduler is a tool designed specifically for optimizing machine learning workloads. It leverages MLIR transformations to achieve the best schedule for a given ML program. It uses the coreAutoScheduler repo to implement the core functionalities of the autoscheduler.

Getting Started

This is an example of how you may give instructions on setting up your project locally. To get a local copy up and running follow these simple example steps.

Prerequisites:

Required

1. CMake: version 3.20 or greater.
2. Ninja.
3. Gcc : version 13.2.
4. [Gxx]: version 13.2.

Build:

1. Building MLIR :

   git clone https://github.com/llvm/llvm-project.git
   git checkout release/18.x
   mkdir llvm-project/build
   cd llvm-project/build
   cmake -G Ninja ../llvm \
   -DLLVM_ENABLE_PROJECTS="clang;llvm;mlir;openmp" \
   -DLLVM_BUILD_EXAMPLES=ON \
   -DLLVM_TARGETS_TO_BUILD="Native;NVPTX;AMDGPU" \
   -DCMAKE_BUILD_TYPE=Release \
   -DLLVM_ENABLE_ASSERTIONS=ON \
   
   cmake --build . --target check-mlir

   For more detalis follow instructions at https://mlir.llvm.org/getting_started/

2. Clone the repo

   git clone https://github.com/Modern-Compilers-Lab/MLAutoScheduler

3. Clone the submodules

   git submodule update --init --recursive

4. Build

    mkdir build
    cd build/
    cmake .. -DMLIR_DIR={Path to llvm folder}/build/lib/cmake/mlir
    cmake --build . -j

5. Add env variables :

   export LLVM_PATH={Path to llvm folder}
   export SHARED_LIBS=export SHARED_LIBS="${LLVM_PATH}/build/lib/libmlir_runner_utils.so,${LLVM_PATH}/build/lib/libmlir_c_runner_utils.so,${LLVM_PATH}/build/lib/libomp.so"
   export AS_VERBOSE=1 

6. Run

    bin/AutoSchedulerML ../benchmarks/{name of the benchmark}.mlir