| commit | f8f299694a81a43a98b2361b2cde95adf0bfe53c | [log] [tgz] |
|---|---|---|
| author | Ben Vanik <ben.vanik@gmail.com> | Tue Jul 09 13:09:35 2024 -0700 |
| committer | GitHub <noreply@github.com> | Tue Jul 09 13:09:35 2024 -0700 |
| tree | e0175598ce533cc8d016946a447cf3c65964e621 | |
| parent | 94ecb8bde107ff49ba141cec0fb7468d8c450c76 [diff] |
Retaining binding tables and plumbing indirect cmds in local-task. (#17838) The task system doesn't currently support resubmission of task graphs (as scheduling the graphs happens in-place in the task graph structure) so it is emulated as with a deferred command buffer as with other targets. Async targets need to retain the binding table contents until the submission completes (as the resource set in the command buffer won't) and a helper was added to allow that to be done efficiently in the future (today it's still slow).
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.
Community meeting recordings: IREE YouTube channel
IREE is licensed under the terms of the Apache 2.0 License with LLVM Exceptions. See LICENSE for more information.