commit | 64bda351975b968331d84d1c3fe587ba0116b239 | [log] [tgz] |
---|---|---|
author | Geoffrey Martin-Noble <gcmn@google.com> | Tue Nov 23 13:32:20 2021 -0800 |
committer | GitHub <noreply@github.com> | Tue Nov 23 16:32:20 2021 -0500 |
tree | 031e10d72da5f9b2d16f81655558251efeb4deb0 | |
parent | 964721192150677dc9b0811780841d7b18d07d5f [diff] |
Set CPUs to max frequency before running benchmarks (#7728) The aim is that this will reduce variance in benchmarks. Notably, this reduces our ability to capture important real-world scenarios, like running across threads with different clock speeds. We are deliberately making a tradeoff right now to get variance under control and we may revisit this in the future. Pinning to max clock speed is safe and consistent only because the benchmark phones are on cooling plates so thermal throttling won't be a problem. Only the Pixel phone is pinned because we don't have the Samsung phones rooted (because Samsung makes this a PITA and doing so caused the phone to randomly reboot all the time).
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.