commit | 44c41871477acf1044c717609080a3b64c053b08 | [log] [tgz] |
---|---|---|
author | Stella Laurenzo <stellaraccident@gmail.com> | Wed Jan 19 19:39:08 2022 -0800 |
committer | GitHub <noreply@github.com> | Wed Jan 19 19:39:08 2022 -0800 |
tree | 0d930c56d910d3349e213b4c47959cc3a0b31275 | |
parent | d81b5812e443e35dc6bc1ddbdc0c8df54eefb9d2 [diff] |
Integrate llvm-project at d5db25fb1d0863f4f82445f2812bb64793429583 (#8137) * Reset third_party/llvm-project: d5db25fb1d0863f4f82445f2812bb64793429583 (2022-01-18 18:48:43 -0800): [mlir][textmate] Add support for function visibility * Delete dialect parse/print hooks that are now auto-generated. * Remove redundant generated dialect parse/print from iree_input. * Bump mhlo to pickup changes for https://reviews.llvm.org/D117524. * Handle auto-generated print/parse attr/type conflicts. * Disables auto generated for dialects that do meaningful things. Deletes those that are just boilerplate. * Renames tablegen FLOW_Dialect -> Flow_Dialect since the generated class names are derived from that and need to match. * Hard fork iree-dialects to integrations/tensorflow. We should have done this from the beginning and then updated it explicitly on frontend upgrades (which is what out of tree is doing). Effectively, this directory is an island within the repo and must be isolated. This broke as changes to make it compatible with LLVM head don't work for the frontend version. * Fix remaining build errors. * Fix Codegen lit tests. * Fix more lit tests. * Fix yapf error. * Fix last failing lit test. * Fix failing python lit tests. * Fix pydm.functional_if printer to print a space before the region start. (this used to happen implicitly) Co-authored-by: MaheshRavishankar <ravishankarm@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.