commit | 315a00483259caeaeabb0b676f53bdbd4adeb23d | [log] [tgz] |
---|---|---|
author | Ben Vanik <benvanik@google.com> | Tue Nov 30 13:11:09 2021 -0800 |
committer | GitHub <noreply@github.com> | Tue Nov 30 13:11:09 2021 -0800 |
tree | 818887e83da512be87333fde452e308baac9f384 | |
parent | 4027ac99d97b7f11e7a9b8435773fb91909e631a [diff] |
Allowing heap buffers to be allocated with metadata split out. (#7774) Now if the host and local device iree_allocator_t match buffers will be allocated as a slab (single alloc for metadata + data) while if they differ the two will come from the specified allocators. This allows for custom statistics/pooling/etc for device buffers that does not route through the per-instance host allocator that may be used to scope the normal host-side allocations.
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.