Fix `fp16` feature on arm64: the proper feature name is `fullfp16`, not `fp16`. (#15479)

The symptom being fixed here is these warnings while building
`iree-test-deps` on arm64:

```
'+fp16' is not a recognized feature for this target (ignoring feature)
```

This was saying that bytecode modules that are meant to be compiled with
fp16 native arithmetic instructions were actually compiled without them,
falling back on slower paths converting to fp32.

The problem was that we were mixing up two concepts in the LLVM AArch64
target: "extension" vs "feature". The features are the individual,
mutually orthogonal things that we want to be working with here. The
feature name here is `fullfp16`. Extensions are a higher-level
construct, a group of features. Many extensions contain only one
feature, and in that case, most of the time, the extension name equals
the feature name. But `fp16` is an exception:


https://github.com/llvm/llvm-project/blob/26ab444e88fc8fdd554e5a9381a68b7b5e63b6fd/llvm/lib/Target/AArch64/AsmParser/AArch64AsmParser.cpp#L3639

I think the most important dimension here is orthogonality. Features are
mutually orthogonal, so they are what we want to reflect in our own CPU
feature bits.

The one place where we need to keep working with extensions is in Clang
copts, when building either native code or bitcode in ukernels. There,
the `-march=armv8.2-a+...` wants extensions, not features.
12 files changed
tree: 2e8e92ae66f8bb172c41730f47bd9eb6ec79fbd5
  1. .devcontainer/
  2. .github/
  3. build_tools/
  4. compiler/
  5. docs/
  6. experimental/
  7. integrations/
  8. lib/
  9. llvm-external-projects/
  10. runtime/
  11. samples/
  12. tests/
  13. third_party/
  14. tools/
  15. .bazel_to_cmake.cfg.py
  16. .bazelignore
  17. .bazelrc
  18. .bazelversion
  19. .clang-format
  20. .dockerignore
  21. .git-blame-ignore-revs
  22. .gitignore
  23. .gitmodules
  24. .yamllint.yml
  25. AUTHORS
  26. BUILD.bazel
  27. CITATION.cff
  28. CMakeLists.txt
  29. configure_bazel.py
  30. CONTRIBUTING.md
  31. LICENSE
  32. README.md
  33. WORKSPACE
README.md

IREE: Intermediate Representation Execution Environment

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.

CI Status

Project Status

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!

Communication Channels

Related Project Channels

  • MLIR topic within LLVM Discourse: IREE is enabled by and heavily relies on MLIR. IREE sometimes is referred to in certain MLIR discussions. Useful if you are also interested in MLIR evolution.

Architecture Overview

IREE Architecture IREE Architecture

See our website for more information.

Presentations and Talks

License

IREE is licensed under the terms of the Apache 2.0 License with LLVM Exceptions. See LICENSE for more information.