For the last while, the RapidWright team has made RapidWright installable in python3 with a simple pip install rapidwright (see for more details). This means this repository is no longer a convenient way to install RapidWright in Python. However, I am leaving it up for posterity's sake, and in case the examples are helpful for anyone.
This repository provides a simple way to get started using RapidWright via Python3.
RapidWright is written in Java, and as such is normally used in Java environments.
However, setting up the RapidWright environment in Java can be cumbersome.
Additionally many people (especially those first getting started with RapidWright) prefer Python to java, thus the makers of RapidWright have come out with a way to use the RapidWright API in a Python3 environment using JPype1.
The JPype1 Python library exposes Java APIs to Python as if they were native.
This repository packages the setup instructions found here into a setup.sh script for convenience.
In addition, this repository contains a few tutorials written in Python3 to help new users get started.
Before getting started with this repository, it is highly recommended to familiarize yourself with the RapidWright Documentation, particularly the introduction through the "Implementation Basics" section.
To setup the RapidWright Python environment, run make in the top directory.
This will download, install, and setup everything needed for the RapidWright environment.
After the environment has been setup, one can enter the associated Python environment by running source activate and leave the environment by running deactivate.
As the RapidWright for Python Instructions state, you can run rapidwright.py in interactive mode by running python -i rapidwright.py.
This allows you to explore the various APIs. In addition, running this script in interactive mode will allow you to use tab-complete on the RapidWright Java Classes.
for example, after instantiating a AWS_F1 device with device = Device.getDevice(Device.AWS_F1), one can tab complete device. to get the following:
>>> device.
device.AWS_F1 device.getClass( device.getSLRByConfigOrderIndex(
device.DEVICE_FILE_VERSION device.getClockRegion( device.getSLRs(
device.FRAMEWORK_NAME device.getClockRegionFromTile( device.getSeries(
device.FRAMEWORK_NAME_AND_VERSION device.getClockRegions( device.getSite(
device.KCU105 device.getColumns( device.getSiteFromPackagePin(
device.PYNQ_Z1 device.getDevice( device.getSitePin(
device.QUIET_MESSAGE device.getDeviceName( device.getSiteTypeCount(
device.RAPIDWRIGHT_MINOR_VERSION device.getDeviceVersion( device.getTile(
device.RAPIDWRIGHT_QUARTER_VERSION device.getFamilyType( device.getTileTypeCount(
device.RAPIDWRIGHT_VERSION device.getMasterSLR( device.getTiles(
device.RAPIDWRIGHT_YEAR_VERSION device.getName( device.getWire(
device.RW_QUIET_MESSAGE device.getNode( device.hashCode(
device.a( device.getNumOfClockRegionRows( device.notify(
device.equals( device.getNumOfClockRegionsColumns( device.notifyAll(
device.getActivePackage( device.getNumOfSLRs( device.quietReflectiveAccessWarning(
device.getAllCompatibleSites( device.getPIP( device.releaseDeviceReferences(
device.getAllSitesOfType( device.getPackage( device.setActivePackage(
device.getAllTiles( device.getPackages( device.toString(
device.getArchitecture( device.getRows( device.wait(
device.getAvailableDevices( device.getSLR(
>>> device.
It can also be helpful when using RapidWright to refer to the RapidWright JavaDocs.
There are four tutorials found in the tutorials directory. These tutorials focus on four potential use cases of RapidWright. These use cases are:
- Using RapidWright to extract information of a device.
- Using RapidWright to extract information of a design.
- Using RapidWright to create a design from scratch.
- Using RapidWright to route a design.
These tutorials are meant to be simple and straight forward. They do not represent everything that RapidWright can be used for. Additional, more advanced tutorials can be found here. These additional tutorials are mostly written for the java version of RapidWright, and so additional work may be required to do them in Python.
This tutorial shows how to use RapidWright to extract some basic information of a device. While this example can be easily replicated in Vivado using a couple tcl commands, it is often desirable to use RapidWright to get device information for the performance advantage and ease of post processing in languages such as Java or Python.
This tutorial shows how to use RapidWright to extract some basic information of a design. It also does some post-processing (plotting the info in matplotlib) to show some of the advantages of using RapidWright with Python.
This tutorial shows how to make a simple design in RapidWright. Note that this tutorial outputs a .dcp file. To get a bitstream one must open the resulting .dcp file in Vivado and use the 'write_bitstream' command.
This tutorial is based on Lesson1 of the RapidWright tutorial found here: https://github.com/Xilinx/RapidWright/blob/master/src/com/xilinx/rapidwright/examples/Lesson1.java
RWRoute is a new open-source, timing-based router written in RapidWright. It provides a substantial speedup to routing when compared to Vivado's router, with a sacrifice in the critical path delay. This may be useful for applications such as quick design iterations.
This program not as much of a tutorial as much as it is a simple wrapper over the RWRouter main function. To use simply run 'python rwroute <input.dcp> <output.dcp>'. Note that RWRoute currently only works with UltraScale+ devices.
This file also contains the code of RWRoute.main() converted from Java into Python. This code is commented out, and can be used to understand and manipulate the main function of RWRoute.