| commit | fc93e9138b9a0bad4cd5280bb597bc906c519ccf | [log] [tgz] |
|---|---|---|
| author | NatashaKnk <natashaknk@google.com> | Mon Jun 05 13:13:59 2023 -0700 |
| committer | GitHub <noreply@github.com> | Mon Jun 05 13:13:59 2023 -0700 |
| tree | 9ae6e9894781ca6abb0f82c86dab2edd5cb8d214 | |
| parent | 571f28a87ba01482c8b24b87e4d2a1909ca61f1c [diff] |
Add LinalgExt TypePropagation pattern that handles i1 inputs/outputs (#13936)
This pattern functions under several assumptions:
* Input ("Updates") element type matches output element type.
* There is only one input/output -- although listed as variadic,
LinalgExt doesn't actually support variadic inputs/outputs. The input
list must always be length 2 (updates + indices) and the output list
must be always of length 1.
* We are only handling illegal types for updates and outputs. This is
because an input/output matrix of type i1 (for example) is a reasonable
situation, but an index matrix of an illegal type is an indication that
something else is going wrong (at least, I personally can't think of an
example where that would be the case)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.