commit | 47a5f99157b55b0d847a1f47f597f69e2deede62 | [log] [tgz] |
---|---|---|
author | Quinn Dawkins <quinn.dawkins@gmail.com> | Sun May 12 14:28:33 2024 -0700 |
committer | GitHub <noreply@github.com> | Sun May 12 17:28:33 2024 -0400 |
tree | c5b44358af5f2d404dde4e7e1b0142312894b9dd | |
parent | d2dd9e23eff96787a66871239c32c627fc3da51b [diff] |
[Codegen][GPU] Move MFMA/WMMA constructors to interface method (#17356) This moves the builder for mfma/wmma ops to a method on the MMAIntrinsic interface. This removes the dependency on the AMDGPU dialect from the contraction lowering and instead makes it a detail of the mma attribute. This makes it easier to add more mma variants without having to update the distribution pattern each time, and in theory will allow external users to define their own attribute implementing the interface and generate code for it.
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.
Community meeting recordings: IREE YouTube channel
IREE is licensed under the terms of the Apache 2.0 License with LLVM Exceptions. See LICENSE for more information.