[Codegen][ROCDL] Add support for nhwc convolution with mfma (#16579) This supports convolutions using the matmul pipeline by simply tiling the kernel dimensions to 1 and then folding them away, leaving a contraction along the remaining image/input channel dimensions. The config selection logic is kept the same as matmul for now, ignoring the additional dimensions. This only support nhwc convolutions, however supporting nchw convolutions is simply a matter of supporting matmul_transpose_a variants. This also does not support strided convolutions because this approach will fail to vectorize in such cases.
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.