| commit | 39576d0a6e72c3e2060abddf8f5a58479da400eb | [log] [tgz] |
|---|---|---|
| author | Lei Zhang <antiagainst@gmail.com> | Tue Dec 19 13:07:29 2023 -0800 |
| committer | GitHub <noreply@github.com> | Tue Dec 19 13:07:29 2023 -0800 |
| tree | 4e0a807a636cfd07ae80db8b985100ac94d2f939 | |
| parent | d29890c9060984c93c6a1c8c543082887819a6d0 [diff] |
[spirv] Trim provided executable target before linking (#15888) This commit adds a pass to trim the provided target environment to the minimal required one to run the compiled spirv.module op. This increases the chance of linking different executable variants together when giving multiple targets. For the same provided target, it may actually generate more exectuables, given that now each dispatch may see a different set of capability requirements. So we may need to be smarter when linking executables--simply relying on exact target attribute match, which is the current state, is not the best way. But this pass alone should be a good cleanup.
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.