commit | 49c88c693e34a089a9956d505f9bcca45d1ef837 | [log] [tgz] |
---|---|---|
author | Ben Vanik <benvanik@google.com> | Mon Dec 13 12:18:55 2021 -0800 |
committer | Ben Vanik <benvanik@google.com> | Tue Jan 04 10:19:07 2022 -0800 |
tree | 2a9f9b7e951bfd17bd518be686d8036fb47d4ae7 | |
parent | 48fb79ca9b246306098d2726c30dc853c808d9aa [diff] |
Giving iree_hal_allocator_allocate_buffer initial_data. This removes the need for a large number of the mapping/write_data calls. On implementations where providing initial data is cheap and doing anything else is extremely expensive (metal/webgpu) this saves us needing to stage copies. Existing implementations still do the same thing as before only now at least have the ability to do something better.
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.