| commit | 10b3e1e8b1b11b9eb67f31f4a2fa469c09054ff8 | [log] [tgz] |
|---|---|---|
| author | Geoffrey Martin-Noble <gcmn@google.com> | Tue Oct 27 10:04:35 2020 -0700 |
| committer | GitHub <noreply@github.com> | Tue Oct 27 10:04:35 2020 -0700 |
| tree | 0bd4239c78753f633796cd91495e6295e1395823 | |
| parent | 1cece448eef33e8c1f953788757ed74c8ac7103c [diff] |
Use LLVM Bazel BUILD files from google/llvm-bazel (#3544) Adopts build files from https://github.com/google/llvm-bazel. This is another attempt at https://github.com/google/iree/pull/2794 now that https://github.com/google/llvm-bazel/pull/60 added Windows support. These make our LLVM build independent of TensorFlow. Switching to them does slightly increase maintenance burden on our team because the TF team is not assisting in maintaining them. It is not a guarantee that llvm-bazel will already have a commit containing the up to date workflows. However, it does have an advertised tag for each LLVM commit that is the current best-effort for what builds, a workflow to automatically pull in new LLVM commits and a build pipeline that notifies me as soon as the build breaks (independent of any integration into Google source control). I've been mostly able to stay ahead of the integrates into Google source control. I will be expanding our build cop docs to cover more of the workflows here as I am buildcopping this week, but the general workflow is to run `./scripts/git/update_llvm_dependent_submodules.sh` in lieu of the update old `update_tf_submodule.sh` script. A GitHub action will also create those PRs automatically to be merged.
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 logistics. 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 implementations. 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 in the early stages of development and not yet ready for broad adoption. Check out the long-term design roadmap to get a sense of where we're headed.
We plan on a quarterly basis using OKRs. Review our latest objectives to get a sense of what we're up to in the near term.
We use GitHub Projects to track progress on IREE components and specific efforts. We use GitHub Milestones to track the work associated with plans for each quarter.
| CI System | Build System | Platform | Architecture | Component | Status |
|---|---|---|---|---|---|
| Kokoro | Bazel | Linux | x86 | Core |  |
| Kokoro | Bazel | Linux | x86 | Bindings |  |
| Kokoro | Bazel | Linux | x86-swiftshader | Integrations |  |
| Kokoro | Bazel | Linux | x86-turing | Integrations |  |
| Kokoro | CMake | Linux | x86-swiftshader | Core + Bindings |  |
| Kokoro | CMake | Linux | x86-turing | Core + Bindings |  |
| Kokoro | CMake | Android | arm64-v8a | Runtime (build only) |  |
| BuildKite | CMake | Android | arm64-v8a | Runtime |  |
IREE is licensed under the terms of the Apache license. See LICENSE for more information.