Google Git
Sign in
opensecura / 3p / openxla / iree / 750de593b6f7cfdb43c5fa67a00666dac8ace431
commit750de593b6f7cfdb43c5fa67a00666dac8ace431[log] [tgz]
authorStella Laurenzo <stellaraccident@gmail.com>Mon Feb 14 11:57:35 2022 -0800
committerGitHub <noreply@github.com>Mon Feb 14 11:57:35 2022 -0800
treeb2c069e0f1546deb2b377ba0955bf82dd49675a6
parentb959830d762b0614d42a0cf3d2a623410cb3247a [diff]
Rewrite Python function argument packing in C++. (#8311)

* Rewrite Python function argument packing in C++.

* This is a first step towards implementing all hot path code in C++.
* With the Python code, I didn't have great performance baselines, but I estimate it was taking ~80-100us to do argument packing for a 3 item call. This patch:
  * Average of 8us total for argument packing for this case.
  * Average of 1us per DeviceArray.
  * Both numbers are for invocations with static reflection data (which is also now layered for easier access to it can be added at runtime if desired to make things faster). The dynamic case does a bit more work.
  * Once more of this is ported to C++, we can eliminate a lot more overhead by intercepting the tp_call protocol directly (vs doing list/dict manipulation).
* This also fixes a critical bug where the Python "DEVICE_VISIBLE" enumeration value was actually set to IREE_HAL_MEMORY_TYPE_HOST_VISIBLE. I believe the result was that all arguments were being set up in a non optimal way (but we were still getting a latency boost from the results). Attempting to make this fix had the effect of opening a bug farm of other issues setting up a buffer allocation, so I ended up adding a TODO and re-enabling HOST_VISIBLE. Will triage more once this lands, now that is is possible to actually disable HOST_VISIBLE.
* Need to check performance in a real setting. My maching is pretty noisy. I am seeing lower e2e numbers than before.
* Release gil around invoke and buffer copy
10 files changed
tree: b2c069e0f1546deb2b377ba0955bf82dd49675a6
  1. .github/
  2. benchmarks/
  3. bindings/
  4. build_tools/
  5. colab/
  6. docs/
  7. experimental/
  8. integrations/
  9. iree/
  10. llvm-external-projects/
  11. scripts/
  12. third_party/
  13. .bazelignore
  14. .bazelrc
  15. .bazelversion
  16. .clang-format
  17. .gitignore
  18. .gitmodules
  19. .pylintrc
  20. .style.yapf
  21. .yamllint.yml
  22. AUTHORS
  23. BUILD.bazel
  24. CMakeLists.txt
  25. configure_bazel.py
  26. CONTRIBUTING.md
  27. LICENSE
  28. README.md
  29. 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.

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.

Build Status

CI SystemBuild SystemPlatformArchitectureConfiguration / ComponentStatus
KokoroBazelLinuxx86-64kokoro status bazel/linux/x86-swiftshader/core
KokoroCMake & BazelLinuxx86-64 (swiftshader)Integrationskokoro status cmake-bazel/linux/x86-swiftshader
KokoroCMake & BazelLinuxx86-64 (turing)Integrationskokoro status cmake-bazel/linux/x86-turing
KokoroCMakeLinuxx86-64 (swiftshader)kokoro status cmake/linux/x86-swiftshader
KokoroCMakeLinuxx86-64 (swiftshader)asankokoro status cmake/linux/x86-swiftshader-asan
KokoroCMakeLinuxx86-64 (turing)kokoro status cmake/linux/x86-turing
KokoroCMakeAndroidarm64-v8aRuntime (build only)kokoro status cmake/android/arm64-v8a
KokoroCMakeBare Metalrisc-v-32Runtimekokoro status cmake/baremetal/riscv32
KokoroCMakeLinuxrisc-v-64Runtimekokoro status cmake/linux/riscv64
BuildkiteCMakeAndroidarm64-v8aRuntimebuildkite status iree-android-arm64-v8a
BuildKiteCMakeAndroidarm64-v8aRuntime Benchmarksbuildkite status iree-benchmark
BuildKiteCMakeLinuxx86-64Tracing + Standalone Runtimebuildkite status iree-build-configurations

Architecture Overview

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.