Google Git
Sign in
opensecura / 3p / openxla / iree / 9e20e68b8107fdf803d3848ca31888d3297c33de
commit9e20e68b8107fdf803d3848ca31888d3297c33de[log] [tgz]
authorMax191 <44243577+Max191@users.noreply.github.com>Mon Nov 04 14:50:55 2024 -0500
committerGitHub <noreply@github.com>Mon Nov 04 11:50:55 2024 -0800
treef9e1d715be745da00be381f6e9c7c242aa7fcb79
parenta757965b41885dfbc8e409b488724c1d4a4e3cc8 [diff]
[LLVMGPU] Enable IGEMM for convolutions by default (#19006)

This flips the `iree-codegen-llvmgpu-use-igemm` flag from false to true.
Convolutions that are aligned with MMA intrinsic shapes will now go down
the IGEMM path, using the TileAndFuse pipeline. Some tests are also
updated, since the default behavior has changed.

### Benchmarks ###
***Note**: These are simply some local comparative benchmarks, intended
to show the difference when flipping the flag. Do not read too much into
the exact numbers, since the benchmarking environment was probably not
ideal.

- Local benchmarks show improvement (~0.5ms) for int8 punet, and a
regression for int8 VAE (~1ms on batch size 1, ~20ms on batch size 8).
- Broader microbenchmark data of more convolution shapes shows general
improvement on average with IGEMM over the default path.
- The current model benchmarks have no tuning, and the regressions in
VAE should be recoverable with tuning for the IGEMM path.

---------

Signed-off-by: Max Dawkins <max.dawkins@gmail.com>
6 files changed
tree: f9e1d715be745da00be381f6e9c7c242aa7fcb79
  1. .github/
  2. build_tools/
  3. compiler/
  4. docs/
  5. experimental/
  6. integrations/
  7. lib/
  8. llvm-external-projects/
  9. runtime/
  10. samples/
  11. tests/
  12. third_party/
  13. tools/
  14. .bazel_to_cmake.cfg.py
  15. .bazelignore
  16. .bazelrc
  17. .bazelversion
  18. .clang-format
  19. .git-blame-ignore-revs
  20. .gitattributes
  21. .gitignore
  22. .gitmodules
  23. .pre-commit-config.yaml
  24. .yamllint.yml
  25. AUTHORS
  26. BUILD.bazel
  27. CITATION.cff
  28. CMakeLists.txt
  29. configure_bazel.py
  30. CONTRIBUTING.md
  31. LICENSE
  32. MAINTAINERS.md
  33. README.md
  34. RELEASING.md
  35. 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.

IREE Discord Status pre-commit OpenSSF Best Practices

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

Release status

PackageRelease status
GitHub release (stable)GitHub Release
GitHub release (nightly)GitHub Release
Python iree-compilerPyPI version
Python iree-runtimePyPI version

Build status

CI PkgCI

Host platformBuild status
LinuxCI - Linux x64 clang
CI - Linux arm64 clang
macOSCI - macOS x64 clang
WindowsCI - Windows x64 MSVC

For the full list of workflows see https://iree.dev/developers/general/github-actions/.

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

Community meeting recordings: IREE YouTube channel

  • 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.