commit | 9240f1c789f7f3ac2d62701ef1c6c9856587c509 | [log] [tgz] |
---|---|---|
author | Lei Zhang <antiagainst@google.com> | Wed Mar 23 16:14:24 2022 -0400 |
committer | GitHub <noreply@github.com> | Wed Mar 23 16:14:24 2022 -0400 |
tree | 1ecc0032ceee551cac542bf637b65c1c9116aea5 | |
parent | 5e5cbd43802d701ccd62c8dcc0e6042c5f77652b [diff] |
Integrate llvm-project at 9816fada1667ecc3343d285295c95848826693af (#8612) * Reset third_party/llvm-project: 9816fada1667ecc3343d285295c95848826693af (2022-03-18 10:50:55 +0100): [bazel] Port a954ade8ed41 * Cherry-picked https://github.com/llvm/llvm-project/commit/3b74aac29c97b587c30d940e00b338af384ede59 additionally to fix bufferization issues * TensorFlow commit: c9ddfac348b1f423e1463ae78b046514a8b03a48 * Cherry-picked https://github.com/tensorflow/tensorflow/commit/9642bbdbb6b27443d339b714f233272eb74ca637 additionally to fix leftover `dump()` calls for debugging * MHLO commit: 467cd37703dc0c4195ce6351617ef320bb60e927 * Updated external model registration * Updated FuncOp usages * The attribute carrying type information now should be `function_type`, instead of `type`. * `mlir::FuncOp` is now in `mlir::func::FuncOp`. * `builtin.func` in assembly should be `func.func` now. * Updated MHLO attribute assembly for `comparison_direction`, `precision`, etc. * Fixed TFL/StripMetadata.cpp to avoid dropping `tf_saved_model` attributes for ABI information
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.