| commit | d0a79c4a79b57e3ae38659af610c4c7d72e6d5b0 | [log] [tgz] |
|---|---|---|
| author | Geoffrey Martin-Noble <gcmn@google.com> | Mon Jan 31 17:46:32 2022 -0800 |
| committer | GitHub <noreply@github.com> | Mon Jan 31 17:46:32 2022 -0800 |
| tree | 41d2a7dbcc16276802edeff47d13d42865b4cf05 | |
| parent | 6996e22a60fcf50afcbca3ec0496f5afe5069592 [diff] |
Generate arm intrinsics for MMT4D (#7751) This passes all the e2e matmul tests (I added one specifically for the MMT4D shape, which I think was missing test coverage). The option to use intrinsics vs asm is added as a global flag to be used for *development* as we explore these two different paths. It defaults to using asm, since that version is actually producing the thing we want right now. Registration of Arm Neon dialects and passes is added unconditionally in a few places. I can look into making that conditional if that seems worthwhile. The generated asm is also not great. The LLVM shuffle + vdotq_s32 intrinsic is not getting switched into the version that uses lanes directly. I wonder if the issue has to do with no way to express poison at this level. Here's a gist showing a lowering down to asm of the lit test here: https://gist.github.com/GMNGeoffrey/02509944091560adf8150ceb2445cb27 In contrast to inline asm: https://gist.github.com/GMNGeoffrey/06c4bb92708f1d3d2bc173a4ceafe5dd Regardless, this isn't in a production path, so I think we can optimize later.
IREE (Intermediate Representation Execution Environment, pronounced as “eerie”) is an MLIR-based end-to-end compiler and runtime that lowers Machine Learning (ML) models to a unified IR that scales up to meet the needs of the datacenter and down to satisfy the constraints and special considerations of mobile and edge deployments.
See our website for project details, user guides, and instructions on building from source.
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!
| CI System | Build System | Platform | Architecture | Configuration / Component | Status |
|---|---|---|---|---|---|
| Kokoro | Bazel | Linux | x86-64 |  | |
| Kokoro | CMake & Bazel | Linux | x86-64 (swiftshader) | Integrations |  |
| Kokoro | CMake & Bazel | Linux | x86-64 (turing) | Integrations |  |
| Kokoro | CMake | Linux | x86-64 (swiftshader) |  | |
| Kokoro | CMake | Linux | x86-64 (swiftshader) | asan |  |
| Kokoro | CMake | Linux | x86-64 (turing) |  | |
| Kokoro | CMake | Android | arm64-v8a | Runtime (build only) |  |
| Kokoro | CMake | Bare Metal | risc-v-32 | Runtime |  |
| Kokoro | CMake | Linux | risc-v-64 | Runtime |  |
| Buildkite | CMake | Android | arm64-v8a | Runtime |  |
| BuildKite | CMake | Android | arm64-v8a | Runtime Benchmarks |  |
| BuildKite | CMake | Linux | x86-64 | Tracing + Standalone Runtime |  |
See our website for more information.
IREE is licensed under the terms of the Apache 2.0 License with LLVM Exceptions. See LICENSE for more information.