Environment Setup

Prerequisites

Before proceeding with the Allo installation, please follow the instructions on the MLIR-AIE website to install the required Vitis and XRT environment. Stop when you reach the “Install IRON for AMD Ryzen™ AI AIE Application” section as we need a separate process to install MLIR-AIE under the Allo environment.

Install from Source

Please follow the general instructions in Install from Source to install the LLVM-19 project and the Allo package. In the following, we suppose you have already installed the LLVM-19 project and enable the allo conda environment.

Below are the exact commands to set up the environment:

Step 1

Activate the allo conda environment

conda activate allo

Step 2

We depend on the MLIR-AIE project to compile the Allo IR to AIE. Install release 1.0

# Install IRON library and mlir-aie from a wheel
python3 -m pip install mlir_aie -f https://github.com/Xilinx/mlir-aie/releases/expanded_assets/v1.0
# Install Peano from a llvm-aie wheel
python3 -m pip install https://github.com/Xilinx/llvm-aie/releases/download/nightly/llvm_aie-19.0.0.2025041501+b2a279c1-py3-none-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl

Warning

The mlir_aie wheel requires manylinux_2_35, and some systems (e.g., those with glibc 2.34, confirmed by ldd --version) do not meet this requirement. This results in an installation failure such as:

ERROR: mlir_aie-0.0.1.2025042204+24208c0-cp312-cp312-manylinux_2_35_x86_64.whl is not a supported wheel on this platform.

Step 3

Clone the mlir-aie repository and checkout to the commit corresponding to release 1.0

git clone https://github.com/Xilinx/mlir-aie.git
cd mlir-aie
git checkout 07320d6

Then, install python requirements, setup environment and add tools to PATHs (under mlir-aie)

# Install basic Python requirements
python3 -m pip install -r python/requirements.txt
# Install the pre-commit hooks defined in .pre-commit-config.yaml
pre-commit install
# Install MLIR Python Extras
HOST_MLIR_PYTHON_PACKAGE_PREFIX=aie python3 -m pip install -r python/requirements_extras.txt
# Install Torch for ML examples
python3 -m pip install -r python/requirements_ml.txt

source utils/env_setup.sh

Step 4

Clone the allo repository and install.

You may want to set up environment variables first to use a custom CMake and LLVM build. For example:

export PATH=/opt/cmake-3.31.5-linux-x86_64/bin:/opt/llvm-project-19.x/build/bin:$PATH
export LLVM_BUILD_DIR=/opt/llvm-project-19.x/build

Then clone the allo repository and install by running the following commands

git clone https://github.com/cornell-zhang/allo.git
cd allo
python3 -m pip install -v -e .

Note

See Internal Installation (Cornell) for Zhang Group students.

Step 5

Setup Vitis and XRT.

Note

See Internal Installation (Cornell) for Zhang Group students.

Lastly, you can verify the AIE backend by running the following command under the allo folder.

python3 tests/dataflow/aie/test_vector.py

Patches and Configuration

To use components from the MLIR-AIE toolchain as libraries:

Note

The instructions below are based on MLIR-AIE release v1.0, which corresponds to commit 07320d6. For compatibility, make sure to use this commit when copying the following components.

Clone and checkout the specific commit:

git clone https://github.com/Xilinx/mlir-aie.git
cd mlir-aie
git checkout 07320d6

  • To use external kernels as an AIE kernel library:

    export MLIR_AIE_EXTERNAL_KERNEL_DIR=/your/copied/path/aie_kernels

  • To use runtime_lib for the host:

    export RUNTIME_LIB_DIR=/your/copied/path/runtime_lib

If you run into issues when using aiecc.py such as:

error: expected ')' at end of argument list
declare void @llvm.memcpy.p0.p0.i64(ptr noalias writeonly captures(none), ptr noalias readonly captures(none), i64, i1 immarg) #1
                                                          ^

You can fix this by modifying downgrade_ir_for_peano in:

$MLIR_AIE_INSTALL_DIR/python/aie/compiler/aiecc/main.py

Update the function as follows:

Before:

def downgrade_ir_for_peano(llvmir):
    llvmir = llvmir.replace("getelementptr inbounds nuw", "getelementptr inbounds")
    return llvmir

After:

def downgrade_ir_for_peano(llvmir):
    llvmir = llvmir.replace("getelementptr inbounds nuw", "getelementptr inbounds")
    llvmir = llvmir.replace("captures(none)", "")
    return llvmir

Internal Installation (Cornell)

For Zhang Group students, please set up environment variables in Step 4 with the following commands.

export PATH=/opt/cmake-3.31.5-linux-x86_64/bin:/opt/llvm-project-19.x/build/bin:$PATH
export LLVM_BUILD_DIR=/opt/llvm-project-19.x/build

And set up Vitis and XRT in Step 5 by running the following commands.

source /opt/common/setupVitis.sh
source /opt/common/setupXRT.sh

Lastly, to verify the installation, you can run the following command:

python3 tests/dataflow/aie/test_vector.py

If the unit tests pass, then the installation is successful. Otherwise, please contact us for help.