cuda::pipeline::producer_acquire

cuda::pipeline::producer_acquire#

2026-10-01

2 min read time

Applies to Linux and Windows

This page documents cuda::pipeline::producer_acquire, which blocks the current producer thread until the next pipeline stage is available.

Defined in header <cuda/pipeline>:

template <cuda::thread_scope Scope>
__host__ __device__
void cuda::pipeline<Scope>::producer_acquire();

Blocks the current thread until the next pipeline stage is available.

Note

  • If the calling thread is a consumer thread, the behavior is undefined.

  • If the pipeline is in a quitted state, the behavior is undefined.