| commit | 1d20679a16308c07836fb4a8f526804233bedbd3 | [log] [tgz] |
|---|---|---|
| author | Ben Vanik <ben.vanik@gmail.com> | Tue Aug 22 14:10:35 2023 -0700 |
| committer | GitHub <noreply@github.com> | Tue Aug 22 21:10:35 2023 +0000 |
| tree | 3ed40b4c3bf418b599efbe35349e2651e74a790d | |
| parent | a623a93d27cd75b7c306a8282704f10b5a58ad3a [diff] |
Enabling Vulkan buffer device addresses support in the runtime. (#14778) We'll need command buffer and compiler support to use it but this will let us see if there are any issues enabling the features on our targets. There's support for mixed SPIR-V programs using buffer device addresses and those not and VMFBs can contain both to allow for graceful fallback when not supported. SPIR-V executables with the `vulkan-spirv-fb-ptr` will get the support and future changes will add descriptor set layout flags (`IREE_HAL_DESCRIPTOR_SET_LAYOUT_FLAG_INDIRECT` or something) that allows for routing certain descriptors into the indirect parameter buffer. Progress on #13945.
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.