Google Git
Sign in
opensecura / 3p / openxla / iree / 0f4dd7350fe4e6c44149afbb3572e04b6572b298
commit0f4dd7350fe4e6c44149afbb3572e04b6572b298[log] [tgz]
authorBen Vanik <ben.vanik@gmail.com>Tue Sep 19 17:56:36 2023 -0700
committerGitHub <noreply@github.com>Tue Sep 19 20:56:36 2023 -0400
treef96e2da26cbd35c2a26a1a8f6ebf72bb74d4b96e
parent04259d04d6354cb9ba53451894db7f48aa2a9501 [diff]
Adding `--task_topology_cpu_ids=` flag. (#14969)

This allows for explicitly specifying one or more logical CPU ID sets
that should be used for task system topologies. The IDs are in the range
of [0, total_cpu_count) to match lscpu/lstopo on Linux and on a
flattened view of all cores on Windows (so with group 0 having 4 logical
CPUs an ID of 6 indicates group 1 cpu 1). When the flag is specified all
other topology flags are ignored as the NUMA node specified is implicit
in the absolute logical CPU ID.

For programmatic control (python/etc) the
`iree_task_topology_initialize_from_logical_cpu_set_string` helper is
exposed. A hosting application that already knows precisely which thread
affinities it wants can use the new
`iree_task_topology_initialize_from_thread_affinities` to directly
specify without any interpretation while still getting proper
constructive sharing masks when supported on the platform.

Examples:
`--task_topology_cpu_ids=0`: one topology with logical CPU 0 only
`--task_topology_cpu_ids=0,1`: one topology with logical CPU 0 and 1
`--task_topology_cpu_ids=0,1 task_topology_cpu_ids=32,33`: two
topologies with logical CPU 0,1 and 32,33 respectively
6 files changed
tree: f96e2da26cbd35c2a26a1a8f6ebf72bb74d4b96e
  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

License

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