Increase the PotentialValuesState analysis max value count. (#13862) 32 was too conservative and very easily hit in even moderately sized programs. This would result in passes that use this state like iree-stream-elide-timepoints stopping after the first few timepoints. This eats a bit more memory but is a drop in the bucket (4k instead of 512b per state entry, of which there are only dozens). Progress on #13637 (LLM's with control flow and lots of values were hitting this).
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.