Concepts – VLSI System Design

VSDOpen2022 – Ground-level efforts to boost Semiconductor tapeouts and Skill Development

VSDOpen conference is an attempt to bring out some cutting-edge activities especially around open-source EDA with a special focus on skill development using open and proprietary tools. VSDOpen also focuses on milestones achieved by VSD in the past year, and some interesting projects which VSD will be working on in the next year. It’s like the VSD Annual Hands-on meeting where everyone is invited for free and allowed to rate us for our work 🙂

RISC-V waterfall diagram and hazards

The above waterfall diagram is representing a sequence of instructions that are fetched from memory and how they progress to the various stages of pipeline. In the above diagram you got program counter (P), fetch (F), decode (D), register read (R), execute (E) and register write (W). We fetch one instruction at a time. Potentially, you can fetch multiple instructions at a time, which would be a super-scalar architecture.

Characterize Level-shifter – It can’t get simpler than this

Well, that’s was just a demo of the powerful efabless platform. Do you want to know more about this powerful yet simple platform? Do you want to build complete SoC using this platform? We did an entire webinar on this, and here’s the copy of it.

Listen from CEO/architect himself on Machine learning

This webinar will talk about how do you do that using Machine Learning and Deep Learning techniques. Participants for this webinar range from students to some of service company India head and Program managers. So we will cover from basics to advanced + labs on cloud.

The beauty of slack-based timing ECO

A timing ECO should be power, performance and area aware and that was the crux of this webinar, where we discussed several strategies about how to do effective ECO as an expert .Slack based ECO is a beautiful strategy which helps you to achieve your timing target, while helping you to reduce on power and area

From VLSI to System Design (SoC) – The choice of SPI

SPI model is a master/slave model. There’s some SPI master which determines who gets to transmit and who gets to receive. The output from SPI master is called MOSI (Master Out Slave In). If you have 2 slaves, slave 1 and slave 2, as shown below, MOSI goes to all the slaves .Then you have another line MISO (Master In Slave Out). All the wires are connected, as shown in below image. Then you have a master only function called SCLK, which goes to all the slaves. Now also, there must be a slave select (SS) for S1 and a slave select for S2.

PPA (power, performance, area) card

A PPA card like the above, is something which every VLSI engineer should be carrying like a business card. Why? Right from RTL to synthesis to PNR to signoff, we do things like upsize, downsize, VT swap, and many more, and all these factors impacts or tweaks your design PPA in one way or the other.
Let’s take an example of ‘downsize’

Committed in 2011, delivered in 2018

the flowchart is what you need to understand just to be an expert in the field of VLSI and semiconductors. Every topic shown in above image is a field, and every topic has a beautiful physics behind it, which when blended with tools in a video course, becomes a master-piece

Facts – About below open-source EDA tool

Why “integrated”? Because at lower nodes, you have to integrate other parts of the flow. Sign-off (you can see power and timing buttons below), clock tree synthesis (you can see synthesis button) must be integrated, so we have a fully integrated PnR flow that we built from day one.

Wanna quick solution to identify overflows? – Use RISC-V branches

The final instruction which is ‘bne’ is ‘branch if not equal’ which is essentially means to branch to a different address location pointed by ‘overflow’, if t3 is not equal to t4, which is the case here. So, here’s where the overflow is detected and an error message will be printed out, informing ‘you are out of range’