commit | 769036f0e2c501500d3901d2c5059598912a8b4d | [log] [tgz] |
---|---|---|
author | Ben Vanik <benvanik@google.com> | Tue Feb 01 15:34:28 2022 -0800 |
committer | GitHub <noreply@github.com> | Tue Feb 01 15:34:28 2022 -0800 |
tree | 0d6b38a46b9a5c00f8cb1084e251200d88c29f87 | |
parent | e838b4aadaaed0d6f84203cfb7f20ce33e632fa3 [diff] |
Adding iree_wait_source_t and cleaning up iree_wait_handle_t. (#8224) Wait sources add an abstraction between the platform primitives such that core code can pass them around. Wait sources can be implemented with the platform wait handles or with custom types (such as HAL semaphores) and enable deferred handle creation. In this way we can pass around the wait sources and only if/when the consumer of them wants to perform a system wait will we need to materialize the wait handle. This also works to move more out from behind the IREE_ENABLE_THREADING flag: wait handles are not related to threading.
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.