[CPU] Scalarize vector loads used in a scalar fashion (#14150) This PR makes sure that we scalarize the vector load that are used in a scalar fashion (i.e., vector.transfer_read + vector.extract) in matmuls. The LLVM backend is sometimes able to scalarize this vector load but not always. For cases where it's not scalarized, register pressure increases as the liveness of this vector load spans across all the unrolled iterations of the matmul. There are also some shuffles/permutations generated that don't make too much sense from the performance point of view.
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.