editor's blog
Subscribe Now

Fast and Accurate?

It’s a song as old as time: if the tool takes too long to run, then run a less accurate version early on, and just use the accurate version at the end.

In other words, speed comes at the expense of accuracy.

So I have to say that I was a bit confused and suspicious when Atrenta claimed that their new Fast Lint didn’t sacrifice any accuracy. A conversation with their Mike Gianfagna cleared the picture up a bit.

You see, Atrenta says that they live and breathe by the accuracy of Spyglass. They say that their customers refer to designs being “Spyglass clean.” With that kind of equity in the brand, accuracy is probably the last thing they want to trade off. And you could practically hear them scrambling to explain when I suggested that they must be giving up accuracy.

So here’s the deal. They have their normal version of Spyglass that actually does synthesis under the hood in order to find things that the customer’s synthesis tool may not like. They refer to this as structural linting. Of course, synthesis takes time.

Then they released Advanced Lint, which adds formal proofs to the collection of things to be checked. This doesn’t increase the accuracy of things they were checking before; it adds a whole new list of things to check for. But formal analysis takes even more time.

And they started getting feedback that designers really couldn’t check their designs as often as they’d like because it just took too long.

So what they found was that they could take the 5000 or so rules that they have and structure them. There’s a first category that can be run quickly because they don’t need any synthesis. They grouped these together into Fast Lint. Then there’s the structural lint portion that does synthesis but not formal; and then there’s Advanced Lint that does the whole shebang.

Of course, they then adjusted their interface to reflect this new structure.

So Fast Lint is intended to be run each time the design is checked in. In fact, you can script things so that it’s automatically done at check-in, and check-in can be rejected if it fails. Structural Lint can be performed on a less frequent basis to give more of a thorough check. And Advanced Lint can be done, if at no other time, for sign-off. But probably before then: if you’re ready for sign-off, it better be clean.

So none of the tests lose accuracy; it’s just a matter of which tests you run when.

You can find more in their release

Leave a Reply

featured blogs
May 19, 2022
The current challenge in custom/mixed-signal design is to have a fast and silicon-accurate methodology. In this blog series, we are exploring the Custom IC Design Flow and Methodology stages. This... ...
May 19, 2022
Learn about the AI chip design breakthroughs and case studies discussed at SNUG Silicon Valley 2022, including autonomous PPA optimization using DSO.ai. The post Key Highlights from SNUG 2022: AI Is Fast Forwarding Chip Design appeared first on From Silicon To Software....
May 12, 2022
By Shelly Stalnaker Every year, the editors of Elektronik in Germany compile a list of the most interesting and innovative… ...
Apr 29, 2022
What do you do if someone starts waving furiously at you, seemingly delighted to see you, but you fear they are being overenthusiastic?...

featured video

EdgeQ Creates Big Connections with a Small Chip

Sponsored by Cadence Design Systems

Find out how EdgeQ delivered the world’s first 5G base station on a chip using Cadence’s logic simulation, digital implementation, timing and power signoff, synthesis, and physical verification signoff tools.

Click here for more information

featured paper

Intel Agilex FPGAs Deliver Game-Changing Flexibility & Agility for the Data-Centric World

Sponsored by Intel

The new Intel® Agilex™ FPGA is more than the latest programmable logic offering—it brings together revolutionary innovation in multiple areas of Intel technology leadership to create new opportunities to derive value and meaning from this transformation from edge to data center. Want to know more? Start with this white paper.

Click to read more

featured chalk talk

Introducing Vivado ML Editions

Sponsored by Xilinx

There are many ways that machine learning can help improve our IC designs, but when it comes to quality of results and design iteration - it’s a game changer. In this episode of Chalk Talk, Amelia Dalton chats with Nick Ni from Xilinx about the benefits of machine learning design optimization, what hierarchical module-based compilation brings to the table, and why extending a module design into an end-to-end flow can make all the difference in your next IC design.

Click here for more information about Vivado ML