| commit | 65680c69ff1433129be76bf37b028521d30cc5ac | [log] [tgz] |
|---|---|---|
| author | Kunwar Grover <groverkss@gmail.com> | Fri Jan 19 23:38:53 2024 +0530 |
| committer | GitHub <noreply@github.com> | Fri Jan 19 23:38:53 2024 +0530 |
| tree | ee0d2a57537f15b7bfc736d8c6622de079f0f343 | |
| parent | d71c14708be2b0ae07e8567b4737b0803f4acee9 [diff] |
[VectorDistribution] Add patterns for distributing transfer_read/transfer_write (#16115) This patch adds distribution patterns for transfer_read/transfer_write and lowers them to vector.load/vector.store per thread. These distribution patterns do the lowering for transfer_read/transfer_write in one-shot, which is different from how transfer_read/transfer_write are lowered in upstream mlir. The upstream patterns unroll one dimension at a time and apply the patterns recursively. We do this lowering in one-shot because we have the layout attribute which defines the iteration space for the lowering.
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.