| commit | 3fa9fbd7dd1c1ab01dd5da120ea885cae719cd83 | [log] [tgz] |
|---|---|---|
| author | Benoit Jacob <benoitjacob@google.com> | Wed Apr 03 12:30:20 2024 -0400 |
| committer | GitHub <noreply@github.com> | Wed Apr 03 16:30:20 2024 +0000 |
| tree | d1da06e12383d880bf39b84005a8c7684d9854f3 | |
| parent | 2c88e49d184b621c763a5cfb4af693f8dcbc6a07 [diff] |
Integrate LLVM at llvm/llvm-project@a6d932bca8875198fbf34564cda8a8d1640cdcbc (#16944) IREE-side changes to adapt to MLIR changes: 1. `initializeOptions` changes to adapt to https://github.com/llvm/llvm-project/pull/87289 2. `enableFastMathMode` removal: https://github.com/llvm/llvm-project/pull/86578. 3. Bazel changes to adapt to https://github.com/llvm/llvm-project/pull/86819 IREE-side fixes for preexisting bugs revealed by a MLIR change: 1. `mlp_tosa` test fix: the shapes were inconsistent, used to accidentally work, until MLIR started catching it since https://github.com/llvm/llvm-project/pull/85798. See diagnostic in [87396](https://github.com/llvm/llvm-project/issues/87396#issuecomment-2032929368). FYI @MaheshRavishankar. IREE-side fixes accidentally lumped into this: 1. The `iree_copts.cmake` change: It just happens that my bleeding-edge Clang was updated and started diagnosing some code relying on C++20 semantics. Filed https://github.com/openxla/iree/issues/16946 as TODO. --------- Co-authored-by: Scott Todd <scott.todd0@gmail.com>
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.