commit | 164a60ef1558069bc5a343544a321981ef1b0b72 | [log] [tgz] |
---|---|---|
author | Nirvedh Meshram <96096277+nirvedhmeshram@users.noreply.github.com> | Thu Oct 10 18:12:52 2024 -0500 |
committer | GitHub <noreply@github.com> | Thu Oct 10 18:12:52 2024 -0500 |
tree | 0fb5aa5360dbe203ff81b54bc1fd577a673bc2a7 | |
parent | 9aa11e553a34614c81efbc53985e8e38ec692466 [diff] |
[ROCM] Disable mixed precision fma instructions that cause numeric issues (#18753) Scalar mixed precision fma instructions are rare in codegen as we dont generate something like that from the MLIR side but a multiplication followed by a truncation can result in them and unfortunately they have issues with the zeroing semantics that the llvm backend has trouble with. See this issue https://github.com/iree-org/iree/issues/18746 We can opt out of this feature which is what this PR does. We also update the punet regression suite outputs which were affected by this bug with the new results.
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
Package | Release status |
---|---|
GitHub release (stable) | |
GitHub release (nightly) | |
Python iree-compiler | |
Python iree-runtime |
Host platform | Build status |
---|---|
Linux | |
macOS | |
Windows |
For the full list of workflows see https://iree.dev/developers/general/github-actions/.
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.