commit | 8e738a79d7abf688a0d93741c1a220117ebca28e | [log] [tgz] |
---|---|---|
author | Stella Laurenzo <laurenzo@google.com> | Sun Feb 19 12:53:09 2023 -0800 |
committer | GitHub <noreply@github.com> | Sun Feb 19 20:53:09 2023 +0000 |
tree | c8b55a9c64e32c15e61ff1d1e5e976486b72aa0c | |
parent | 96d959e01177e911ad3837e4799b530a7d931e16 [diff] |
Add a bring-your-own-LLVM build script. (#12274) This wraps the following into one reference flow (that we can run on CI): 1. Build and install LLVM+tools (llvm, lld, clang). 2. Build and install MLIR standalone against installed LLVM. 3. Build IREE against separately installed LLVM and MLIR. Supporting a multi-step like this may help division of labor in shops who have a compiler team focused on building an LLVM toolchain as it will enable them to build that apart from the MLIR churn. Folks would then layer a locally built MLIR on top of that and build IREE on top of the combination. Builds on #12256. Progress on #12231.
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!
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.