| commit | 96be7cd59b127e1c7581a0263c8f0569d06f093e | [log] [tgz] |
|---|---|---|
| author | Ben Vanik <benvanik@google.com> | Mon Jan 31 14:48:23 2022 -0800 |
| committer | Ben Vanik <benvanik@google.com> | Tue Feb 01 11:32:58 2022 -0800 |
| tree | 8985e2f42dafd9c13bc69d4618684697abae0078 | |
| parent | c226d92f1318b9958d8ea6c2b742c1af6454ef20 [diff] |
Propagating errors from task system calls/dispatches. Waits still need work that will come with follow-on refactorings. Failures of calls and dispatch workgroups will now make their way to the parent scope and properly discard all subsequent tasks. The scope retains the first failure encountered so that it can be returned back to user layers (propagated to HAL semaphores, the event loop, etc). Progress on #4026.
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.