FC4 Kernel Panic Under VMware
It's Cold!

Sun Microsystems Launches OpenSPARC Project

Wow... Sun announced today that they'll be open sourcing the SPARC design database and associated verification environment:

Today, Sun also announced plans to publish specifications for the UltraSPARC-based chip, including the source of the design expressed in Verilog, a verification suite and simulation models, instruction set architecture specification (UltraSPARC Architecture 2005) and a Solaris OS port. The goal is to enable community members to build on proven technology at a markedly lower cost and to innovate freely. The source code will be released under an Open Source Initiative (OSI)-approved open source license.

http://www.sun.com/smi/Press/sunflash/2005-12/sunflash.20051206.4.html

Also check out "Open Source Hardware?".

This ties in nicely with the comments I've made earlier on this blog describing the benefits to be obtained from open-sourcing hardware verification environments.

But what does it really mean to open source the entire SPARC core and associated verification environment?  If the core is like many of the designs I've worked on in the past the RTL will mean very little without a good amount of documentation, a build environment, run scripts, synthesis scripts, EDA tools/licenses, circuit layouts, etc.  The verification environment would need a similar ecosystem.  For these reasons it's not likely that someone off the street would get much use from the code.  My guess is that it would take a team of at least 50 people and $30,000,000 to get off the ground. One question would be whether any derivative works based on the SPARC core would also need to be open-sourced... Most companies would frown on a requirement to open-source their entire RTL database because they had picked up components from Sun's release.

A group that wouldn't be constrained by licensing issues would be the folks in academia.  Groups from several universities could band together to share the work of any new development effort.  The project would be a great learning experience for students who would otherwise never get exposure to the large scale projects typical in industry.  Sun would also benefit by creating a community of developers far larger than they could ever employ on their own.

It's all fascinating stuff.  So - what do you think?  Is open source good for hardware design and verification?  Or is the idea doomed to failure?

Comments