Google Git
Sign in
opensecura / 3p / openxla / iree / 038d2bada1dfaaee7db51b2d1d2ee6472aee970b
commit038d2bada1dfaaee7db51b2d1d2ee6472aee970b[log] [tgz]
authorScott Todd <scotttodd@google.com>Thu Aug 10 13:54:08 2023 -0700
committerGitHub <noreply@github.com>Thu Aug 10 13:54:08 2023 -0700
treee6a8788bc054d055bc57f45a49807490df0044a2
parent1ef5acf3398ce6c673fcd74137f4ca667e215e1b [diff]
Work around scalar support bug in webgpu compilation. (#14629)

See https://github.com/openxla/iree/issues/11054#issuecomment-1670260451
- this `multiple_results.mlir` test program uses scalars, and
https://github.com/openxla/iree/pull/13711 changed the code path that
scalar dispatches go down. This alternate code path does not appear to
have the same SPIR-V -> WGSL compatibility patches as the tensor path,
so we see compilation errors with messages like `entry point 'd0'
references multiple variables that use the same resource binding`. We
should fix that (and add tests so it doesn't regress again), but I want
to keep webgpu runtime debugging work unblocked.

This modified program exhibits the same runtime-related issue as
reported on https://github.com/openxla/iree/issues/13809 /
https://github.com/openxla/iree/pull/14163:
* The first invocation incorrectly returns 
  ```
  2xf32=2.23 2.45
  2xf32=0 0
  ```
* The second invocation correctly returns
  ```
  2xf32=1.23 1.45
  2xf32=2.23 2.45
  ```
2 files changed
tree: e6a8788bc054d055bc57f45a49807490df0044a2
  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

  • 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.