| commit | 97a13940bcbb8ab21b8a42044af20cf9dec33c3a | [log] [tgz] |
|---|---|---|
| author | Ben Vanik <ben.vanik@gmail.com> | Wed Oct 25 09:16:10 2023 -0700 |
| committer | GitHub <noreply@github.com> | Wed Oct 25 09:16:10 2023 -0700 |
| tree | 4a7c7866c47a47f0d0d7b3f07dc37a3df9bd3d53 | |
| parent | 446f96eb896dcc5a1d29ff0af3363d8df1fe7021 [diff] |
Adding `vm.br_table` op. (#15286) This implements a 0-based dense branch table that can be lowered from `cf.switch` which is a weird op but what we have and what comes from `scf.index_switch`. The exact encoding may change in the future to make interpreting faster but as structured it should be good for basic JITs. The initial usage of this will come from entirely dense (or mostly dense) `scf.index_switch` ops so only the most basic optimization for offseting the base of the table is implemented. There's no canonicalization/folding as `cf.switch` has most of that but we should probably add some of the same handlers in the future.
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.