ML SW toolchain based on IREE

Clone this repo:

Branches

  1. f291657 sw/vec_iree: Fix ukernel breakage due to upstream macro removal by Lun Dong · 3 weeks ago master
  2. 4384db6 Update conv1x1 mlir due to upstream change on tosa.rescale shift attribute by Lun Dong · 3 weeks ago
  3. 9dcda29 Fix a breakage due to IREE upstream renaming by Lun Dong · 5 weeks ago
  4. 163e63f Turn off unused-function warning in GCC by Cindy Liu · 7 weeks ago
  5. f3f8e49 Update FileCheck path to llvm-16-tools by Cindy Liu · 9 weeks ago

Springbok Model Execution Repository

This project contains ML model codegen and execution libraries/examples. The artifacts built from this project is targeted for RISC-V 32-bit baremetal machine (Springbok or Kelvin).

Prerequisite

If you get this project from Project Shodan manifest, you are all set. If not, you need to have following projects as well to build the project successfully.

This project needs to be at <dir>/sw/vec_iree

Code structure

  • audio_prep: Audio preprocessing libarary
  • build_tools: Utility scripts for the project
  • cmake: CMake Macros for the project
  • device: Device HAL driver library
  • model_util: Runtime utility library for model execution
  • native_log: BSP support for the Linux userspace emulation
  • risp4ml: Vision preprocessing library (Reduced ISP for ML)
  • samples: Codegen and execution of ML models based on IREE
    • branch_mul: Branch multiplication example
    • microbenchmarks: Currently only contain a convolution 1x1 example
    • simple_vec_mul: Point-wise vector multiplication examples
  • ssd_postprocess: Vision postprocessing library for Single-Shot Detectors (SSD)
  • ukernel: Custom microkernel library for VMVX and LLVM

Build the project

If you are running from Project Shodan repo, you can build the artifacts with m iree. If you download this project alone, you need to download/build IREE's host compiler, then go to the top level directory.

For Springbok

  cmake -B <output dir> -G Ninja \
     -DCMAKE_TOOLCHAIN_FILE="$(realpath sw/vec_iree/cmake/riscv_iree.cmake)" \
     -DCMAKE_BUILD_TYPE=MinSizeRel \
     -DIREE_HOST_BIN_DIR="$(IREE_COMPILER_DIR)/install/bin" \
     -DRISCV_TOOLCHAIN_ROOT=$(TOOLCHAINRV32_PATH) \
     <dir>/sw/vec_iree
  cmake --build <output dir> --target all

For Kelvin

  cmake -B <output dir> -G Ninja \
     -DCMAKE_TOOLCHAIN_FILE="$(realpath sw/vec_iree/cmake/riscv_iree_gcc.cmake)" \
     -DCMAKE_BUILD_TYPE=MinSizeRel \
     -DIREE_HOST_BIN_DIR="$(IREE_COMPILER_DIR)/install/bin" \
     -DRISCV_TOOLCHAIN_ROOT=$(TOOLCHAINRV32_PATH) \
     <dir>/sw/vec_iree
  cmake --build <output dir> --target all

Run the executables

You can run RENODE emulation with the built executables. In Shodan repo, sim_springbok for Renode simulation on Springbok; sim_kelvin for Renode simulation on Kelvin. Out-of-repo emulation is possible but requires more steps.

Test the executables

In Shodan, this project utilizes LLVM lit and FileCheck to test the ML executable performance. The tests are defined in the *_test.run files under <output dir>. To run the test, at the Shodan top level directory

lit --path <Filecheck dir> -a <output dir>

Filecheck in Debian testing is under /usr/lib/llvm-16/bin

Default is renode tests for Springbok. Add-D RUNNER=kelvin to enable renode tests for Kelvin. Add-D RUNNER=kelvin_hw to enable renode tests for Kelvin HW.