[sv-bc] Mantis 1000 - evcd driver recording

From: Gordon Vreugdenhil <gordonv_at_.....>
Date: Mon Jun 16 2008 - 14:26:33 PDT
I've taken a closer look at Mantis 1000 and have a few
questions (I hope I didn't mess up the directionality here!).

1)  The old text was more clear that one first considered
     whether there was a stronger driver (strong range versus
     weak range) in determining the result.  The current text
     seems to imply that 2 weak input drivers could dominate a single
     strong output driver.  I am assuming that is not the
     intent.  So is the intent to first gather the subset of
     strongest drivers and then do the conflict resolution as
     described?  Is it purely the strongest drive, with no
     consideration of ranges?

2)  In the following, is the value an "h" or "H"?  It depends
     on whether you only count the strongest "1" or count all
     of the ones on the "winning" side. The comment about the
     "strongest driver" implies that you count all the drivers
     but that interpretation isn't obvious given the later
     comment that:
         It is not considered to be a conflict state when multiple
         drivers are driving different values but at different
         strengths.

     Example:

         module dut(a);
           inout    a;

           assign (supply0,supply1) a=1;
           assign (strong0,strong1) a=1;

         endmodule

         module tb(a);
           inout    a;

           dut dut(a);

           assign (strong0,strong1) a=0;

         endmodule

     If I change the "strong" 0 drive in tb to a 1, do I get a "1" or
     an "H" or an "h"?
     If I leave the "strong" 0 drive in tb there and add a strong 1
     drive, what do I get?


3)  Assuming that the result of the original example in (2) is "h",
     should the evcd be recording
         ph 0 7
     or
         ph 6 7
     since there is a strong0 active as well.

4) Minor nit -- I don't really like calling this section
    "Conflict states".  I'd really prefer to call this
    "Driver State Resolution" and prefix the state description
    by stating what constitutes a "conflict".

Gord
-- 
--------------------------------------------------------------------
Gordon Vreugdenhil                                503-685-0808
Model Technology (Mentor Graphics)                gordonv@model.com


-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
Received on Mon Jun 16 14:27:10 2008

This archive was generated by hypermail 2.1.8 : Mon Jun 16 2008 - 14:27:46 PDT