dylanyunlong/project_6
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
900ae0b1efc19e46cd313eb3ba7eb694fa596f6b
project_6/cccl_upstream/docs/libcudacxx/extended_api/synchronization_primitives.rst
muh-bot 2a7ca101d7 feat(cccl): integrate missing CCCL directories — python/, ci/, .agent/, docs/, test/ 
Sparse-checkout from NVIDIA/cccl main branch to complete cccl_upstream:

Added:
- python/cuda_cccl/ (226 files) — Python bindings for device-level algorithms
  Critical for muh toolchain: cuda.compute.reduce_into, scan, radix_sort, etc.
  Includes 204 .py files with full test coverage for all 27 algorithms
- ci/ (163 files) — Build/test infrastructure
  build_cub.sh, test_cub.sh, build_and_test_targets.sh, matrix.yaml
  Directly maps to our [INFRA-CI] and [INFRA-BUILD] items
- .agent/skills/ (7 files) — NVIDIA's own agent skills for CCCL
  cccl-style/SKILL.md, cccl-test/SKILL.md, sass-diff/SKILL.md
- docs/ (491 files) — Official CCCL documentation
  CI references, CMake guides, Python compute docs, libcudacxx PTX docs
- test/ (12 files) — Top-level integration tests (cuda_smoke, stdpar)
- Root configs: .clang-format, .clang-tidy, CONTRIBUTING.md, pyproject.toml
- CLAUDE.md symlink → AGENTS.md (NVIDIA's standard)

cccl_upstream now mirrors full NVIDIA/cccl structure:
  Before: 42M (cub + thrust + libcudacxx + cudax + c + examples + benchmarks)
  After:  53M (+python +ci +docs +.agent +test +configs)

This completes the CCCL base needed for:
- [muh-bench] items: ci/util/build_and_test_targets.sh for targeted builds
- [CCCL-verify] items: python/cuda_cccl/tests/ as reference implementations
- [CCCL-test] items: ci/test_cub.sh, ci/test_thrust.sh
- Agent workflow: .agent/skills/ for consistent style and test patterns
2026-08-07 02:34:33 +00:00

138 lines
4.5 KiB
ReStructuredText
Raw Blame History

.. _libcudacxx-extended-api-synchronization:
Synchronization Primitives
===========================
.. toctree::
:hidden:
:maxdepth: 1
synchronization_primitives/atomic
synchronization_primitives/atomic_ref
synchronization_primitives/latch
synchronization_primitives/barrier
synchronization_primitives/counting_semaphore
synchronization_primitives/binary_semaphore
synchronization_primitives/pipeline
.. rubric:: Atomics
.. list-table::
:widths: 25 45 30 30
:header-rows: 1
* - **Header**
- **Content**
- **CCCL Availability**
- **CUDA Toolkit Availability**
* - :ref:`cuda::atomic <libcudacxx-extended-api-synchronization-atomic>`
- System-wide `std::atomic <https://en.cppreference.com/w/cpp/atomic/atomic>`_ objects and operations
- libcu++ 1.0.0 / CCCL 2.0.0
- CUDA 10.2
* - :ref:`cuda::atomic_ref <libcudacxx-extended-api-synchronization-atomic-ref>`
- System-wide `std::atomic_ref <https://en.cppreference.com/w/cpp/atomic/atomic_ref>`_ objects and operations
- libcu++ 1.7.0 / CCCL 2.0.0
- CUDA 11.6
.. rubric:: Latches
.. list-table::
:widths: 25 45 30 30
:header-rows: 1
* - **Header**
- **Content**
- **CCCL Availability**
- **CUDA Toolkit Availability**
* - :ref:`cuda::latch <libcudacxx-extended-api-synchronization-latch>`
- System-wide `std::latch <https://en.cppreference.com/w/cpp/thread/latch>`_ single-phase asynchronous
thread coordination mechanism
- libcu++ 1.1.0 / CCCL 2.0.0
- CUDA 11.0
.. rubric:: Barriers
.. list-table::
:widths: 25 45 30 30
:header-rows: 1
* - **Header**
- **Content**
- **CCCL Availability**
- **CUDA Toolkit Availability**
* - :ref:`cuda::barrier <libcudacxx-extended-api-synchronization-barrier>`
- System wide `std::barrier <https://en.cppreference.com/w/cpp/thread/barrier>`_ multi-phase asynchronous
thread coordination mechanism
- libcu++ 1.1.0 / CCCL 2.0.0
- CUDA 11.0
.. rubric:: Semaphores
.. list-table::
:widths: 25 45 30 30
:header-rows: 1
* - **Header**
- **Content**
- **CCCL Availability**
- **CUDA Toolkit Availability**
* - :ref:`cuda::counting_semaphore <libcudacxx-extended-api-synchronization-counting-semaphore>`
- System wide `std::counting_semaphore <https://en.cppreference.com/w/cpp/thread/counting_semaphore>`_
primitive for constraining concurrent access
- libcu++ 1.1.0 / CCCL 2.0.0
- CUDA 11.0
* - :ref:`cuda::binary_semaphore <libcudacxx-extended-api-synchronization-counting-semaphore>`
- System wide `std::binary_semaphore <https://en.cppreference.com/w/cpp/thread/counting_semaphore>`_
primitive for mutual exclusion
- libcu++ 1.1.0 / CCCL 2.0.0
- CUDA 11.0
.. rubric:: Pipelines
The pipeline library is included in the CUDA Toolkit, but is not part of the open source libcu++ distribution.
.. list-table::
:widths: 25 45 30 30
:header-rows: 1
* - **Header**
- **Content**
- **CCCL Availability**
- **CUDA Toolkit Availability**
* - :ref:`cuda::pipeline <libcudacxx-extended-api-synchronization-pipeline>`
- Coordination mechanism for sequencing asynchronous operations
- libcu++ 1.2.0 / CCCL 2.0.0
- CUDA 11.1
* - :ref:`cuda::pipeline_shared_state <libcudacxx-extended-api-synchronization-pipeline-pipeline-shared-state>`
- :ref:`cuda::pipeline <libcudacxx-extended-api-synchronization-pipeline>` shared state object
- libcu++ 1.1.0 / CCCL 2.0.0
- CUDA 11.0
* - :ref:`cuda::pipeline_role <libcudacxx-extended-api-synchronization-pipeline-pipeline-role>`
- Defines producer/consumer role for a thread participating in a *pipeline*
- libcu++ 1.1.0 / CCCL 2.0.0
- CUDA 11.0
* - :ref:`cuda::make_pipeline <libcudacxx-extended-api-synchronization-pipeline-pipeline-role>`
- Creates a :ref:`cuda::pipeline <libcudacxx-extended-api-synchronization-pipeline>`
- libcu++ 1.1.0 / CCCL 2.0.0
- CUDA 11.0
* - :ref:`cuda::pipeline_consumer_wait_prior <libcudacxx-extended-api-synchronization-pipeline-pipeline-consumer-wait-prior>`
- Blocks the current thread until all operations committed up to a prior *pipeline stage* complete
- libcu++ 1.1.0 / CCCL 2.0.0
- CUDA 11.0
* - :ref:`cuda::pipeline_producer_commit <libcudacxx-extended-api-synchronization-pipeline-pipeline-producer-commit>`
- Binds operations previously issued by the current thread to a :ref:`cuda::barrier <libcudacxx-extended-api-synchronization-barrier>`
- libcu++ 1.1.0 / CCCL 2.0.0
- CUDA 11.0
Reference in New Issue View Git Blame Copy Permalink