I first came across ASI just before leaving IMC, when it was the Arapahoe working group. IMC was interested as the SE fabric chip had a striking resemblence to what would be needed as an ASI switch. I read all the specs for ASI and PCI-Express. IMC decided that the market wasn't at that point really ready for an ASI switch so opted to produce an interim PCI-Express/ASI bridge device, a decision which nearly everyone else in this market is pursuing as well. I left IMC at that point to begin my sabbatical, while they abandoned the comms market with the NPU/TMC/SE and morphed into a PCI-Express company IMC-Semi. They eventually closed, down selling this IP to IDT.
I next fell over ASI at Vitesse where just prior to another RIF and reorganization they assigned me to the ASI-SIG working groups working on producing both Fabric Management Specifications for ASI as well as associated technology demonstration implementation projects showing both ASI and the fabric management software working. This was targetted at a Linux implementation which certainly perked my interest. So much so that I've decided to continue trying to contribute a GPL implementation of ASI Fabric Management for Linux privately in my spare time.
This will be my interpretation of how ASI should be managed in the Linux style and is implemented solely from the publicly available core specifications of ASI and its protocols. This is not based on any of the work for the Fabric Management Work Groups whose work at this point is still under wraps.
I'm contributing this work because I want to help this technology and the people involved succeed. If any of them objects to this effort I will happily stop upon request. Please have the courtesy to contact me directly in this case, no need for cease and desist letters from lawyers.