Dump scheduling statistics for collecting dispatch numbers (#12757) Use option `--iree-scheduling-dump-statistics-format` to dump scheduling statistics at stream level, which contains the number of dispatches, when compiling the benchmark suites. https://github.com/openxla/iree/blob/97779d7f494660f88864b035475ec77a1e54c6c8/compiler/src/iree/compiler/Dialect/Stream/Transforms/Passes.cpp#L312 This allows us to collect and track the changes in the number of dispatches, which is a requested feature for a long time. It seems fine to upload this metric to https://perf.iree.dev, as now we use filters to better organize the metrics (and the dashboard performance looks fine). I still want to have a better way to structure the metrics, with the database later... The benchmark tool will be updated later to gather these information. --------- Co-authored-by: Geoffrey Martin-Noble <gcmn@google.com>
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.