.. _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