commit | 43a4b766e0dc44abd7a880b55044dc14085d06e0 | [log] [tgz] |
---|---|---|
author | MaheshRavishankar <1663364+MaheshRavishankar@users.noreply.github.com> | Tue Apr 05 10:22:13 2022 -0700 |
committer | GitHub <noreply@github.com> | Tue Apr 05 10:22:13 2022 -0700 |
tree | 8f1e8a2c8feb733407be300f52ff6fa0b654f678 | |
parent | 5b3ccb008d407759b9770c232006fd1d1c6cb818 [diff] |
Add a pass to initialize all remaining empty tensors to zero-filled tensors (#8749) After dispatch region formation, all remaining linalg.init_tensor ops are converted to flow.tensor.splat ops as a fallback. Ideally this should never happen and all linalg.init_tensor get fused with their uses. But in cases where a front-end generates code where the init_tensor escapes, this provides a safety net for it. The code is still correct, but potentially uses more memory and is slower. Fixes #8717
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.