| commit | 876ac79da67e165f7c55e9771fcedcec65b673ec | [log] [tgz] |
|---|---|---|
| author | Stella Laurenzo <laurenzo@google.com> | Wed May 03 18:31:39 2023 -0700 |
| committer | GitHub <noreply@github.com> | Wed May 03 18:31:39 2023 -0700 |
| tree | f65c3e11c3aa67982a11e515f27097ec280393da | |
| parent | 784b1220580a28254324cc1878e8ae6cb3783017 [diff] |
Remove the old low-level compiler C-API. (#13395) In place of the old C-API, we add two new unstable interop functions which allow the embedding API to be used to interface with the MLIR C-API: * ireeCompilerSessionGetContext * ireeCompilerInvocationImportModule Updates the compiler API test to demonstrate the mixed use of the embedding API and the MLIR C-API for directly compiling an in-memory module. I wasn't going to prioritize doing this soon, but it turns out that it is blocking further work on plugins, as it would force us to try to shim IREE's plugin mechanism into generic use of the MLIR C-API. This also removes the corresponding Python APIs which depended on these features. The tests for these have been disabled and have been high cost to keep working. We can bring back a better defined Python API if there is ever demand. Implements and fixes #13215.
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.