- Leverages the primitives (Span<T> and Memory<T>) and low-allocation APIs.
- Provides an efficient programming model while freeing the developers from having to manage buffers.
- A pipeline is like a
Stream
that pushes data to you rather than having you pull. One chunk of code feeds data into a pipeline, and another chunk of code awaits data to pull from the pipeline. - When writing to a pipeline, the caller allocates memory from the pipeline directly.
- Pipelines formalize transferring ownership of buffers so that the readers can flow buffers up the stack without copying.
- Supports moving data without copying.
This repository has been archived by the owner on Aug 2, 2023. It is now read-only.