| commit | c074522d0c2f0cc5c798f9f23c85f7314e799710 | [log] [tgz] |
|---|---|---|
| author | Geoffrey Martin-Noble <gcmn@google.com> | Tue Nov 08 17:52:49 2022 -0800 |
| committer | GitHub <noreply@github.com> | Wed Nov 09 01:52:49 2022 +0000 |
| tree | bce4815d12f70ab5959c0ece25ac0cedf105cff2 | |
| parent | 8ee57387050c1343f666559da7abd1a4b5dadcd8 [diff] |
Create Docker images for the latest software (#11083) This enables us to test against the latest compiler features and local dependencies. I'm giving the asan build a little bit of double duty here in that it will also catch issues specific to later compiler versions. The potential confusion or extra iteration time from this combination seems worth it for the compute/latency savings. The asan build seems a good candidate for using the latest compiler because sanitizers presumably catch more issues with later versions. We don't necessarily need to make guarantees that our build works with our oldest supported compiler version and sanitizers. It seems reasonable that people running sanitizer builds could update to the latest versions. This change also restructures the way we use Docker contexts. We have already accrued some duplication because of the difficulty factoring Docker files. Instead, this moves the docker context to include potentially any file in the repository, which allows reuse via scripts. We can continue to reduce duplication in future changes.
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.