commit | 9d5456390e26220b4a5affa57079b4253a64f973 | [log] [tgz] |
---|---|---|
author | Stella Laurenzo <laurenzo@google.com> | Tue Apr 12 17:56:53 2022 -0700 |
committer | GitHub <noreply@github.com> | Tue Apr 12 17:56:53 2022 -0700 |
tree | db451c69d33fa3bfda02d88c81a34c8087fe04fd | |
parent | 4f03b4c690d24e4350e6c1500925b9fdb31f664d [diff] |
Add a sript to fetch and trim the CUDA toolkit deps needed to build. (#8862) The package and directory structure was a bit more complicated than presumed and I opted to fetch and use nvidia's download script to manage it. This script then trims to just the license and files we need, bringing the cost from ~100MiB to ~1MiB. We will just add this to the base docker image and then extend our CMake logic to probe when the magic env var is set. CMake updates will be made separately once images are rolled. ``` root@129bf88cd357:/# du -h /usr/local/iree_cuda_deps/ 464K /usr/local/iree_cuda_deps/nvvm/libdevice 468K /usr/local/iree_cuda_deps/nvvm 776K /usr/local/iree_cuda_deps/include 1.3M /usr/local/iree_cuda_deps/ ``` Progress on #8847
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.