commit | 2126488001ce8eed5318f315df043e40a8ab0cd4 | [log] [tgz] |
---|---|---|
author | Ben Vanik <ben.vanik@gmail.com> | Thu Jan 12 09:47:56 2023 -0800 |
committer | GitHub <noreply@github.com> | Thu Jan 12 09:47:56 2023 -0800 |
tree | 3a595bbff2580efb257bcb311e0766b6b8dd99fa | |
parent | 20d746a7877f45c1060a54096875fda7cb06c9ab [diff] |
Adding iree_hal_device_replace_allocator API. (#11809) This allows for wrapping base device allocators in shims that can specialize allocation logic. It's a dangerous method but for drivers that have specific requirements on their base allocators such as internal dependencies on device handles (Vulkan/CUDA/etc) it's difficult to allow for pure allocator injection on device creation. Users can now inject their own custom device allocators when managing devices themselves and as we add supported upstream ones we can use them in tools based on the new `--device_allocator=` flag.
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.