commit | 0d281b7e4823f93c7630d61a3b0784c67622b136 | [log] [tgz] |
---|---|---|
author | Lei Zhang <antiagainst@google.com> | Mon Jun 01 20:00:23 2020 -0400 |
committer | Lei Zhang <antiagainst@google.com> | Tue Jun 02 15:01:49 2020 -0400 |
tree | 942b3093e277a60f95c2bce5343191554600c6e6 | |
parent | 99fcd424092658bc9aa4d973d0cc0c6d682427ea [diff] |
[cmake] Set up cross compilation for IREE This commit adds cross compilation support to IREE's CMake build system. This requires a few non-trival changes due to how IREE structures the code. In IREE we auto-generate many C++ source code files during build process. Notably we use flatc to generate FlatBuffer include files from schemas and tablegen to generate many other compiler source files. And we even use iree-translate to compile models into IREE binary modules for testing... These binaries must be first be compiled on the host and then we can later use them to generate code for compiling towards the target architecture. This commit adds a iree_cross_compile.cmake file for a bunch of utilities for cross-compilation. If cross compilation mode is detected, we invoke CMake another time under another directory and feed in the configurations for host to compile the autogen tools first. The CMake invocation needs to track and drive the host compilation though; this is mainly done via file dependencies and custom commands. For the target compilation, we wire up the dependency there. Tested the following command and it worked: ```shell mkdir build-android cd build-android cmake .. \ -DCMAKE_TOOLCHAIN_FILE=$ANDROID_NDK/build/cmake/android.toolchain.cmake \ -DANDROID_ABI="arm64-v8a" -DANDROID_PLATFORM=android-24 \ -DIREE_BUILD_SAMPLES=OFF -DIREE_BUILD_COMPILER=OFF \ -DIREE_BUILD_TESTS=OFF -DIREE_HOST_C_COMPILER=`which clang` \ -DIREE_HOST_CXX_COMPILER=`which clang++` -GNinja ninja ```
IREE (Intermediate Representation Execution Environment, pronounced as “eerie”) is an MLIR-based end-to-end compiler that lowers ML models to a unified IR optimized for real-time mobile/edge inference against heterogeneous hardware accelerators. IREE also provides flexible deployment solutions for the compiled ML models.
IREE is still in its early phase. We have settled down on the overarching infrastructure and are actively improving various software components as well as project logisitics. It is still quite far from ready for everyday use and is made available without any support at the moment. With that said, we welcome any kind of feedback on any communication channels!
For development, IREE supports both Bazel and CMake on Windows and Linux. We are working on enabling macOS support. For deployment, IREE aims to additionally cover Android and iOS.
Please see the Getting Started pages on IREE's documentation hub to configure, compile, and run IREE in your favorite development environment!
IREE hosts all its documentation and project status dashboards on GitHub Pages. We are still building up the website; please feel free to create issues for the documentation you'd like to see!
We also have some public talks that explain IREE's concepts and architecture:
IREE adopts a holistic approach towards ML model compilation: the IR produced contains both the scheduling logic, required to communicate data dependencies to low-level parallel pipelined hardware/API like Vulkan, and the execution logic, encoding dense computation on the hardware in the form of hardware/API-specific binaries like SPIR-V.
The architecture of IREE is best illustrated by the following picture:
Being compilation-based means IREE does not have a traditional runtime that dispatches “ops” to their fat kernel implemenations. What IREE provides is a toolbox for different deployment scenarios. It scales from running generated code on a particular API (such as emitting C code calling external DSP kernels), to a HAL (Hardware Abstraction Layer) that allows the same generated code to target multiple APIs (like Vulkan and Direct3D 12), to a full VM allowing runtime model loading for flexible deployment options and heterogeneous execution.
IREE aims to
IREE is still at its early stage; we have lots of exciting future plans. Please check out the long-term design roadmap and short-term focus areas.
We use GitHub Projects to track various IREE components and GitHub Milestones for major features and quarterly plans. Please check out for updated information.
CI System | Build System | Platform | Component | Status |
---|---|---|---|---|
GitHub Actions | Bazel | Linux | Integrations | Workflow History |
GitHub Actions | Bazel | Linux | Other | Workflow History |
Kokoro | Bazel | Linux | Core + Bindings | |
Kokoro | CMake | Linux | Core + Bindings |
IREE is licensed under the terms of the Apache license. See LICENSE for more information.