| commit | 0573f4f5be5ebf8f89686ac2cabe1d784c9e2112 | [log] [tgz] |
|---|---|---|
| author | Nicolas Vasilache <nicolasvasilache@users.noreply.github.com> | Tue Dec 06 22:03:15 2022 +0100 |
| committer | GitHub <noreply@github.com> | Tue Dec 06 21:03:15 2022 +0000 |
| tree | 30e9f50e716ac928a8555ce425106763c764f477 | |
| parent | 01cf67a9e21b83c9b81f1531b1d85e996c6d1476 [diff] |
Transform dialect bringup (#11422) Bringup a generalized form of the transform dialect: * Transform Dialect Bringup Branch * Improve tests/transform_dialect/cuda/reduction.mlir by making input a… (#11208) * [NFC] Refactor transform dialect matcher. (#11211) * Add back mistakenly removed execution test in matmul CPU (#11209) * [DONOTMERGETOMAIN] Drop a bunch of CI stuff we do not care about for … (#11215) * Move transform dialect builder file to Common (#11216) * Move the CPU to not use attributes. (#11219) * NFC - Apply minor API cleanups (#11238) * Add basic structured op matcher with fluent API (#11237) * Add another cuda reduction test with other sizes and always tiling th… (#11243) * Matcher API for inputs and inits/outputs (#11240) * Support reduction+elementwise in transform dialect strategies (#11288) * Refactor the matching code and move all the parameters into structs (#11222) * [DONOTMERGETOMAIN] Add a flag to disable tensor fusion in flow (#11314) * Add strategy+test for reduction(elementwise) (#11315) * Add matcher callback operations to the transform dialect (#11389) * Fixes related to IREEEraseHALDescriptorTypeFromMemRefOp * Revert "[DONOTMERGETOMAIN] Add a flag to disable tensor fusion in flow (#11314)" * Revert "[DONOTMERGETOMAIN] Drop a bunch of CI stuff we do not care about for … (#11215)" * Revert "Transform Dialect Bringup Branch" * Drop tests using --iree-disable-fusion-of-tensor-ops * Use C++ callbacks to match reductions inside the transform dialect (#11390) Co-authored-by: Oleksandr "Alex" Zinenko <zinenko@google.com> Co-authored-by: Nicolas Vasilache <nicolas.vasilache@gmail.com> Co-authored-by: Thomas Raoux <thomasraoux@google.com> Co-authored-by: Matthias Springer <springerm@google.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.