| commit | 7b92a6d38599d3255b4513f1019c540a3401cfd3 | [log] [tgz] |
|---|---|---|
| author | Lei Zhang <antiagainst@gmail.com> | Mon Oct 30 19:32:44 2023 -0700 |
| committer | GitHub <noreply@github.com> | Mon Oct 30 19:32:44 2023 -0700 |
| tree | 88307b68421a4adc3e6a98216378e893fac1e595 | |
| parent | 546e372735935a76b9ab5aa90653b276aa80e403 [diff] |
[cuda] Avoid sorting when composing kernel arguments (#15325) This commit changes the graph command buffer to keep track of a fixed list of descriptor sets. This simplifies the logic of pushing descriptors/constants; we only perform kernel parameter serialization at the dispatch time. This means we don't pay the overhead if multiple push descriptor/constant commands are issued before dispatching. Also we don't need to sort descriptors anymore. Progress towards https://github.com/openxla/iree/issues/13245
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.