Adding `iree.tensor.trace` support for printf debugging. (#16746) Any tensor level op prior to or during flow can be annotated with the `iree.tensor.trace` attribute to have `flow.tensor.trace` ops for all tensor operands and results generated by the pass. The attribute can either be a unit attr to have the trace key chosen automatically or a string attr to specify it. We run the pass once at the head of the pipeline prior to dispatch region formation and again once after, but users can also slice out IR at any phase, add the attributes, use iree-opt to run the pass, and pipe it back through the pipeline to continue compilation. This is printf debugging: it's not great, but it gets the job done.
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.