Feb. 21st, 2013

hw vs sw

Feb. 21st, 2013 10:51 pm
izard: (Default)
Just returned from a day trip to a customer in Osnabrueck.

I was there with a colleague from H/W design team, and we noticed an anti-pattern:
One of the issues we were dealing with was related to a problem in bringing up a freshly designed board with latest CPU revision. Booting the system starts with BIOS code that initialises the peripherals, so the problem could be caused by H/W (schematics, layout), or S/W (BIOS) or (likely) a combination of both factors.

However S/W guys were happy to blame H/W guys and ruled out any problems with S/W (having a plausible technical proof points), and H/W guys were certain that the layout and schematics are both fine, and problem must be in S/W (with even more plausible proofs). I am happy that I was not involved in this discussion, but I will follow it and my bet is that the root cause is related to layout :)

The bright spot for me was that today at last I was officially allowed to brief the customer with a new feature they waited for few years, and sketch a design changes for their s/w using this new feature of upcoming CPUs. Alas, I still have to wait until it becomes public so I can publish an article on habrahabr rather than describe it under NDA.

Profile

izard: (Default)
izard

September 2025

S M T W T F S
 1 23456
78910111213
14151617181920
21222324252627
282930    

Most Popular Tags

Page Summary

Style Credit

Expand Cut Tags

No cut tags
Page generated Sep. 2nd, 2025 08:11 pm
Powered by Dreamwidth Studios