Re: [ng-spice-devel] CVS files removal


To ng-spice-devel@ieee.ing.uniroma1.it
From "Arno W. Peters" <a.w.peters@ieee.org>
Date Sun, 9 Jul 2000 12:44:56 +0200
Delivered-To mailing list ng-spice-devel@ieee.ing.uniroma1.it
In-Reply-To <Pine.LNX.3.96.1000709113216.8849A-100000@ieee.ing.uniroma1.it >; from pnenzi@ieee.ing.uniroma1.it on Sun, Jul 09, 2000 at 11:33:29AM +0200
Mailing-List contact ng-spice-devel-help@ieee.ing.uniroma1.it; run by ezmlm
References <Pine.LNX.3.96.1000709113216.8849A-100000@ieee.ing.uniroma1.it >
Reply-To ng-spice-devel@ieee.ing.uniroma1.it
User-Agent Mutt/1.2i

On Sun, Jul 09, 2000 at 11:33:29AM +0200, Paolo Nenzi wrote:

> I have seen your last commit and read the message. Can you explain better
> the dependence between the device "number" and the analysis code ?

Take for example src/analysis/ckttemp.c; it contains the following
function.  I have numbered the lines so I can reference them easier in
the explanation.


  1: int
  2: CKTtemp(CKTcircuit *ckt)
  3: {
  4:     int error;
  5:     int i;
  6:
  7:     ckt->CKTvt = CONSTKoverQ * ckt->CKTtemp;
  8:
  9:     for (i=0;i<DEVmaxnum;i++) {
 10:         if ( ((*DEVices[i]).DEVtemperature != NULL) &&
 11:                 (ckt->CKThead[i] != NULL) ){
 12:             error =
 13: (*((*DEVices[i]).DEVtemperature))(ckt->CKThead[i],ckt);
 14:             if(error) return(error);
 15:         }
 16:     }
 17:     return(OK);
 18: }


Now look at line 10/11 and line 13.  Every DEVices[i] has a
corresponding ckt->CKThead[i].  The latter holds the parameter
information for all devices of that type.  If I want to abstract
DEVices, I will also need to abstract ckt->CKThead.  So far I don't
know how to do that cleanly.

I hope that clarifies the CVS message.

Greetings,
-- 
Arno

PGP signature


Partial thread listing: