Google Git
Sign in
opensecura / 3p / openxla / iree / 1423d09736a8ffddedc673238d6a5b4033b7962e
commit1423d09736a8ffddedc673238d6a5b4033b7962e[log] [tgz]
authorbjacob <benoitjacob@google.com>Tue May 30 20:19:17 2023 -0400
committerGitHub <noreply@github.com>Tue May 30 20:19:17 2023 -0400
tree2b4bd035dc127137e6fe856de7427e686eef909e
parent041b4e87c6fc9ec75b1612a68f5c24f4851f3028 [diff]
Enable AVX2+FMA in e2e matmul + ukernels test; support comma-separated CPU features. (#13837)

The e2e matmul test covering llvmcpu-ukernels didn't exercise any
interesting architecture code path: since it didn't specify any CPU
features, it was falling back on the architecture-default, which, on
x86-64, would be SSE2, but we don't have a dedicated code path for that,
so it was falling back on scalar code.

This makes it use the AVX2+FMA path. Note, the trace-runner program used
by this trace-runner test, `iree-e2e-matmul-test.c`, will detect CPU
features and skip the test if the machine doesn't support these
features. Note, this is the true reason why this needs to be explicitly
labelled as "CPU features" and can't just be folded into copts --- we
need the CPU features called out explicitly in the yaml trace so the
trace runner can perform that check.

This required plumbing some support for multiple comma-separated CPU
features.

This is the last little stop on the route towards #13804 before we
finally bring in arm64 into the picture, making this officially
multi-arch...
4 files changed
tree: 2b4bd035dc127137e6fe856de7427e686eef909e
  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. .gitignore
  22. .gitmodules
  23. .pylintrc
  24. .style.yapf
  25. .yamllint.yml
  26. AUTHORS
  27. BUILD.bazel
  28. CITATION.cff
  29. CMakeLists.txt
  30. configure_bazel.py
  31. CONTRIBUTING.md
  32. LICENSE
  33. README.md
  34. 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

  • 2021-06-09: IREE Runtime Design Tech Talk (recording and slides)
  • 2020-08-20: IREE CodeGen: MLIR Open Design Meeting Presentation (recording and slides)
  • 2020-03-18: Interactive HAL IR Walkthrough (recording)
  • 2020-01-31: End-to-end MLIR Workflow in IREE: MLIR Open Design Meeting Presentation (recording and slides)

License

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