OpenROAD Tutorial @ DAC 2020

We are excited to see you at Tutorial 9: Developing Industrial Strength EDA Tools Using the OpenDB Open-source Database and the OpenROAD Framework. This email is to give you some important information about the tutorial.

Day: Monday July 20, 2020

Time: 1:30 PM

How to join: log in to the virtual event platform for DAC, navigate to the agenda and select Monday 1:30PM. Click on the session and join the live meeting.

Exercises on GitHub: https://github.com/The-OpenROAD-Project/DAC-2020-Tutorial

Preparation

During the tutorial, we will be covering how to get started with OpenROAD. However, that’s not all. We will be presenting how to develop tools using OpenROAD infrastructure, and how to contribute to its master codebase. For full participation in the tutorial, we ask you to prepare your machine as detailed below.

Supported Platforms

If you are not using any of the two platforms, we suggest that you download VirtualBox (https://www.virtualbox.org/) and install a supported platform. Alternatively, you can install Docker (https://docs.docker.com/get-docker/) and pull an official image.

Prerequisites

OpenROAD tools only need Tcl to be installed.

  • CentOS: sudo yum install tcl
  • Ubuntu: sudo apt install tcl

To use the full automation features of OpenROAD flow, you need to install:

  • Python 3
    • CentOS: sudo yum install python3
    • Ubuntu: sudo yum install python3
  • Time
    • CentOS: sudo yum install time
    • Ubuntu: sudo yum install time
  • KLayout

Download OpenROAD Tools

Download pre-built tools from: https://github.com/The-OpenROAD-Project/OpenROAD/releases

After you download, extract the downloaded file and add it to the PATH environment variable using: export PATH=</path/to/openroad/bin>:</path/to/openroad>:$PATH

Test OpenROAD Tools

From the terminal screen, type in:

$ openroad

| OpenROAD 0.9.0 bfb1291039

| This program is licensed under the BSD-3 license. See the LICENSE file for details.

| Components of the program may be licensed under more restrictive licenses which must be
| honored.

%

If you see the above OpenROAD message, that means tools are working properly.

Development Dependencies

If you are planning to dive into OpenROAD data models using C++ or Python, you will need to install development dependencies to be able to build the tool from sources.

Development dependencies are documented in this Dockerfile: https://github.com/The-OpenROAD-Project/OpenROAD/blob/master/Dockerfile

During The Session

Our tutorial will consist of multiple parts where we give presentations, show demos and also give you the opportunity to try the tools yourself. If you come to the session with all dependencies installed, this will guarantee that you get the most of the tutorial.

We will also have some of our main contributors who can answer your questions during the session. This is a great opportunity to build more EDA tools on top of OpenROAD.

We look forward to seeing you and having a great interactive session.

OpenROAD Team

aspyrou@eng.ucsd.edu

abdelrahman_hosny@brown.edu

abk-openroad@eng.ucsd.edu