Google Git
Sign in
opensecura / 3p / openxla / iree / 0a6cdf0d5bd07f53b673fd1c516ea29e89045fcd
commit0a6cdf0d5bd07f53b673fd1c516ea29e89045fcd[log] [tgz]
authorManish Gupta <manigupta@google.com>Tue Nov 08 22:12:02 2022 -0800
committerGitHub <noreply@github.com>Tue Nov 08 22:12:02 2022 -0800
treec6b459bbdad897041fae4f89199aabd33ad626d2
parenta230101548fb67d3ff7a1f122e1a0dd6b865410c [diff]
Add support for GEMM e2e Test For CUDA backend on F16 input (#10842)

This pull request adds support for testing e2e GEMM on a CUDA backend for F16 input and F16 accumulation.

- Functional testing of F16 GEMMs requires setting the right tolerance value to ensure correctness checks.
- Tolerance-based testing is fragile and finding tolerances is hard. 
- Instead we fill the buffers as small integers centered at zero and test for equivalence.

This pull requests adds `e2e_matmul_direct_f16_gpu_large_LLVMGPUMatmulTensorCore_cuda_cuda` e2e GEMM test to CUDA backend. We have not total four e2e GEMM tests on CUDA backend.

```bash
manigupta@manigupta-gpu-a100 ~/cpu_machine_workspace/repos/iree/iree_tree_1/iree-build-debug $ ctest -j96 -R e2e_matmul.*cuda

1/4 Test #41: iree/tests/e2e/matmul/e2e_matmul_direct_f32_gpu_large_LLVMGPUMatmulTensorCore_cuda_cuda ...   Passed    2.83 sec
2/4 Test #42: iree/tests/e2e/matmul/e2e_matmul_direct_f16_gpu_large_LLVMGPUMatmulTensorCore_cuda_cuda ...   Passed    3.09 sec
3/4 Test #40: iree/tests/e2e/matmul/e2e_matmul_direct_f32_gpu_large_LLVMGPUMatmulSimt_cuda_cuda .........   Passed    3.38 sec
4/4 Test #43: iree/tests/e2e/matmul/e2e_matmul_direct_f32_large_split_k_cuda_cuda .......................   Passed    4.43 sec
```
10 files changed
tree: c6b459bbdad897041fae4f89199aabd33ad626d2
  1. .github/
  2. benchmarks/
  3. build_tools/
  4. compiler/
  5. docs/
  6. experimental/
  7. integrations/
  8. llvm-external-projects/
  9. runtime/
  10. samples/
  11. tests/
  12. third_party/
  13. tools/
  14. .bazelignore
  15. .bazelrc
  16. .bazelversion
  17. .clang-format
  18. .dockerignore
  19. .gitignore
  20. .gitmodules
  21. .pylintrc
  22. .style.yapf
  23. .yamllint.yml
  24. AUTHORS
  25. BUILD.bazel
  26. CITATION.cff
  27. CMakeLists.txt
  28. configure_bazel.py
  29. CONTRIBUTING.md
  30. LICENSE
  31. README.md
  32. 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.