# PYNQ_Composable_Pipeline

**Repository Path**: chenguoping76/PYNQ_Composable_Pipeline

## Basic Information

- **Project Name**: PYNQ_Composable_Pipeline
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: BSD-3-Clause
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2025-05-21
- **Last Updated**: 2025-05-21

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

![license](https://img.shields.io/github/license/Xilinx/PYNQ_Composable_Pipeline?color=g) ![python](https://github.com/Xilinx/PYNQ_Composable_Pipeline/workflows/Python/badge.svg) [![docs](https://readthedocs.org/projects/pynq-composable/badge/?version=latest)](https://pynq-composable.readthedocs.io/en/latest/?badge=latest)


# Composable Pipeline

The Composable pipeline is an overlay with a novel and clever architecture that allow us to adapt how the data flows between a series of IP cores.

The key characteristic of a composable overlay is an AXI4-Stream Switch, which plays the same role as an old [telephone switchboard](https://en.wikipedia.org/wiki/Telephone_switchboard). The AXI4-Stream Switch provides the runtime routing of how data flow from one IP core to another.

On top of that, the composable overlay includes Dynamic Function eXchange [DFX](https://www.xilinx.com/products/design-tools/vivado/implementation/dynamic-function-exchange.html) regions that brings new functionality to the design at runtime.

The combination of these two technologies plus a pythonic API, built on top of [pynq](http://www.pynq.io/), provide an unprecedented flexibility.

<p align="center">
<img src="./docs/source/images/layers.png" width="220"/>
</p>

The composable overlay architecture of a composable video overlay for the PYNQ-Z2 is shown in the image below

![](./pynq_composable/notebooks/img/cv-4pr.png)

## Supported Boards

* [PYNQ-Z1](boards/Pynq-Z1/README.md)
* [PYNQ-Z2](boards/Pynq-Z2/README.md)
* [PYNQ-ZU](boards/Pynq-ZU/README.md)
* [Kria KV260](boards/KV260/README.md), Vitis and Vivado 2020.2.2 are required.

> Note: PYNQ-Z1 is supported with the same composable overlay as PYNQ-Z2

To rebuild the composable pipeline you need Vitis and Vivado 2020.2. Navigate to one of the supported boards folder and run `make`, only steps for Linux are provided.

## Clone this repository

```sh
git clone https://github.com/Xilinx/PYNQ_Composable_Pipeline.git --recursive
```

Note that this project depends on the [Vitis Accelerated Libraries](https://github.com/Xilinx/Vitis_Libraries) and [PYNQ](https://github.com/Xilinx/PYNQ)

## Install composable pipeline on your board

```sh
git clone https://github.com/Xilinx/PYNQ_Composable_Pipeline
python3 -m pip install PYNQ_Composable_Pipeline/ --no-build-isolation
pynq-get-notebooks pynq_composable -p $PYNQ_JUPYTER_NOTEBOOKS
```

The notebooks will be delivered into the folder `/home/xilinx/jupyter_notebooks`

## Documentation

Find out more [documentation on Read The Docs](https://pynq-composable.readthedocs.io/en/latest/).

## Contributing

We welcome contributions, please review the [contributing](CONTRIBUTING.md) guidelines to contribute.

## Licenses

Copyright (C) 2021-2022 Xilinx, Inc

[SPDX-License-Identifier: BSD-3-Clause](LICENSE.md)

[Third-party license](THIRD_PARTY_LIC)

### Binary Files License

Pre-compiled binary files are not provided under an OSI-approved open source license, because Xilinx is incapable of providing 100% corresponding sources.

Binary files are provided under the following [license](boards/Pynq-Z2/LICENSE)

------------------------------------------------------
<p align="center">Copyright&copy; 2021-2022 Xilinx</p>