Chisel Bootcamp
Register for this two-day Chisel Bootcamp to learn how to design hardware with a highly productive and parameterizable Scala-embedded hardware description language, Chisel, from the University of California-Berkeley experts.
Early Registration Fee is $10 students, $40 non-students. Prices go up on 9/2/2019 to $15 students, $45 non-students.
Space is limited; please register early at https://www2.cs.duke.edu/chiselbootcamp/registration.php
CHISEL BOOTCAMP DAY 1 - Chisel Friday 9/6/2019 9am-5:30pm
Chisel is a Scala-embedded hardware description language developed at UC Berkeley. This bootcamp will get you up to speed on Chisel and show you how to design simple hardware using Chisel.
Friday presenter: Chick Markley
(8:30am Installing Bootcamp Essentials)
9:00am Intro to Chisel & the Bootcamp
9:15am Intro to Scala
10:00am Combinational Logic
11:00am Control Flow & Sequential Logic
12:00pm DSP Example
12:30pm Lunch (Provided - please email Lisa Wu Wills to indicate any dietary restrictions)
1:30pm Parameters
2:30pm Functions
3:30pm Types and More
4:30pm Further Topics
5:00pm Saturday Session Preview
CHISEL BOOTCAMP DAY 2 - RocketChip Generator Saturday 9/7/2019 9am-12pm
RocketChip Generator is an open-source RISC-V SoC generator written in Chisel and developed at UC Berkeley. This session will show you how the Chisel and RocketChip toolchains can be used to design, simulate, and deploy custom hardware. We will walk through a simple design example of a custom accelerator and how it is integrated into the RocketChip ecosystem.
Saturday Presenters: Albert Ou, Jerry Zhao, and Sagar Karandikar
The Chisel Bootcamp is hosted by Duke University Assistant Professor of Computer Science and Electrical & Computer Engineering Lisa Wu Wills, of the APEX Lab @ Duke. The goal of the APEX Lab @ Duke is to provide research scientists within and outside of the computer science/engineering discipline with systems to accelerate how they do research analysis. We want to make designing, deploying, and using custom hardware orders of magnitude simpler while demonstrating application-driven programmable hardware accelerated systems that are orders of magnitude more efficient.