| commit | 668c020cc04d66a60b31a6cda46ebdf3b854e952 | [log] [tgz] |
|---|---|---|
| author | Han-Chung Wang <hanhan0912@gmail.com> | Wed Nov 01 21:31:35 2023 -0700 |
| committer | GitHub <noreply@github.com> | Thu Nov 02 04:31:35 2023 +0000 |
| tree | dabac42b07385ddabc4ee07c8e2e2c8cbbbcdc3c | |
| parent | 77a8c5529e9d2f3a6a93270576c037c8337558d7 [diff] |
Cast tensor.empty type to TypeConverter's type during materialization. (#15375) It is a bug introduced by https://github.com/openxla/iree/commit/03d655a99408f43bd5d960172b15fb0d5823e633 VMVX has a different assumption because it generally supports dynamic cases. I.e., it treats all the matmul as dynamic shapes even they are static inputs. This could make inferred shape and TypeConverter's target shape different in terms of dynamism. Casting it to the target type to make dialect conversion happy.
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.