commit | 981f48768a5a27baac818b8afa506abaf4ba75ad | [log] [tgz] |
---|---|---|
author | Scott Todd <scotttodd@google.com> | Wed Nov 03 08:30:15 2021 -0700 |
committer | GitHub <noreply@github.com> | Wed Nov 03 08:30:15 2021 -0700 |
tree | 1274fba61e97d21e304fe1903fc98eda2eea969d | |
parent | b00b12fb6bdd301a85c15f084fccdfd275e0b64a [diff] |
Flip -iree-llvm-link-embedded to true by default. (#7372) Tying up loose ends on https://github.com/google/iree/issues/4717. The system linking path is more flexible for debugging and integration with platform-specific tooling, while the embedded linking path is more portable and is likely to have better performance. Summary of changes: * Remove explicit `--iree-llvm-link-embedded=true` from targets using `dylib-sync` * Remove test suites that were testing the embedded path explicitly (now duplicates of the default test suites) * Disambiguate `iree-llvm-system-linker-path`/`IREE_LLVMAOT_SYSTEM_LINKER_PATH` and `iree-llvm-embedded-linker-path`/`IREE_LLVMAOT_EMBEDDED_LINKER_PATH` flags/env-vars * Also remove docs referencing them. If the automatic discovery fails, error messages should give enough context for developers to debug on their own. * Include source build of `lld` in test targets that now depend on it * Smooth out linker tool discovery in `EmbeddedLinkerTool.cpp`
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.