Agile Hardware @ DAC – Part 1

Good news and better news…

The good news is that I’ll be part of an agile hardware panel discussion at DAC this year in San Francisco. The discussion happens at 10:30am on June 9th at the Moscone Center and it’s happening thanks to Randy Smith and folks at Sonics. For more info, check out the conference website.

The better news is that this is the first of two (repeat: two) exciting chances at DAC to build the agile hardware community (I’ll point you to the other in part two as the details are finalized). Needless to say, two chances to reach out at a conference like DAC is a great opportunity so if you’re a fan of agile hardware, a practitioner, someone who wants to know more or somebody who wants to get out and publicly disagree with the whole idea, I hope to see you there!

-neil

Goodbye AgileSoC.com… For A While

On Friday I decided I could use a break from the blog so I’m going to step away for a bit. Not exactly sure for how long, but it might be a while. Or maybe not. We’ll see. Meanwhile, you’ll still be able to find me at neil.johnson@agilesoc.com for questions, comment and discussion re: agile hardware. Later.

-neil

The Key To Test-driven Development of RTL

For me, this is a very exciting post because I think I’ve made some pretty important headway regarding TDD for hardware designers.

My big side project as of late has been a real pilot dedicated to using TDD to write RTL. I’ve blogged about some of the things I’ve learned already but the big eye-opener that I haven’t talked about yet is how TDD helps us with design partitioning and testability. Continue reading

Now Is Not The Time To Be Practical

They call them comfort zones for a reason, it’s because they’re comfortable.

I don’t like being in my comfort zone because when I get comfortable I fail to recognize when a change is necessary. I start framing uncomfortable change as impractical, then concoct seemingly logical arguments to avoid it. Trouble is, some of these arguments aren’t logical at all. They’re just my way of avoiding the pain that comes with taking the chance.

Some may think that becoming an agile development team is impractical; that the approach won’t work for them, that it’s too big a risk or it’ll be all pain and no gain. Where agile becomes impractical, it’ll be watered down or avoided.

As an example, suppose someone suggests XP as a model for an agile transition. Amoung other things, XP brings with it a series of primary practices. If XP doesn’t fit into your comfort zone, there’s any number of practical reasons for why you might avoid any – or all – of these practices (listed in bold). Continue reading

Agile IC Methodology From Sonics

Here’s a marketing video posted yesterday from Sonics about an Agile IC Methodology they’re promoting in the run-up to ARM TechCon. Normally I might not post stuff like this directly on AgileSoC.com but I’m making an exception because one of the people in the video is me… along with some other guys that are far more recognizable 🙂 .

If you watch the video, I hope you’ll take a few minutes to comment. I’m interested to hear what people think about it and the fact that a few different folks within the industry are coming together to promote agile development.

[youtube_sc url=9xVfIqcY1Qw width=640 height=480]

-neil

Now What?

You’ve started doing TDD or unit testing your systemverilog code with SVUnit, your defect rate is down and you’re producing better code. You’re at the point where your experience could benefit the rest of your team but you’re not sure how to get the point across.

What do you do? Continue reading

Stepping Through an RTL Unit Test

Pretty much every testbench I’ve ever built, used or seen has a free-running clock that’s driven within a while or forever loop. Not much can happen without the clock in a synchronous design so defining the clock logic is usually the first and most obvious thing we do as verification engineers.

Screen Shot 2014-09-22 at 1.51.14 PM

Assuming your design-under-test is synchronous to the positive edge (they almost always are), testbench components usually do their work somewhere off the posedge to avoid races. With even a simple testbench these days, there will be several components, each with their own thread, pushing or pulling data from various interfaces on the DUT. The free-running clock is that steady bass drum beat that holds everything together. Continue reading