alliance-support '2000
SCR and ALU3


MICHON Christian (xianyen@yahoo.com)
Mon, 26 Jun 2000 23:32:29 -0700 (PDT)

Hi, I tried (again) to make use of amg but this time with P&R trial (SCR). I managed through sed to do a temp fix of the vss/vdd line swap. Please use the Makefile attached to reproduce my experiments. I realized through the tutorials that in Alliance only ALU1/2 are used, at of course a cost in terms of die area. So I wanted to try with a 3rd metal. Went inside the sources of SCR to understand a bit, and understood that there is a SCR_L3MODE env variable that controls this... But, while doing this, a few problems occurred: 1- the sxlib cells once placed are inverted every row, without taking care of DRUC nwell rules. 2- the central power stripe should be alternate too, but remains regular (as if with only ALU1/2). This is detected at LVS. I'm not always getting the same results (looks random to me), but I can confirm DRC/LVS problems with this Makefile. And the die size is smaller by almost 50%... Did I trigger some hidden option not debugged yet ? Is there another way to force SCR to use ALU3 ? I saw somewhere (Ludo's page ?) that an OCR is on its way... I hope to see it available soon. Overall, your tools are quite **impressive**. With best regards, Christian. __________________________________________________ Do You Yahoo!? Get Yahoo! Mail - Free email you can access from anywhere! http://mail.yahoo.com/

 



Alliance Web Site © 1997, 2002 ASIM/LIP6/UPMC, page maintained by Czo [Olivier Sirol] , last updated on 22 March 2001.