[ibis-macro] Re: A possible clarification BIRD for IBIS v5.1
- From: "Walter Katz" <wkatz@xxxxxxxxxx>
- To: <Arpad_Muranyi@xxxxxxxxxx>, "'IBIS-ATM'" <ibis-macro@xxxxxxxxxxxxx>
- Date: Tue, 18 Sep 2012 20:11:42 -0400 (EDT)
Arpad,
There is nothing in the spec that says all of the <value>s (including
<default value> - the first <value>) need to be unique. Nor is there
anything in the spec that says the <default value> needs to be one of the
<value>s. My interpretation of the spec is that the following two are
equivalent:
(Format List 0 0 1 2 3)
(Format List 0 1 2 3)
And the following two are equivalent:
(Format List 1 0 1 2 3)
(Format List 1 0 2 3)
Walter
From: ibis-macro-bounce@xxxxxxxxxxxxx
[mailto:ibis-macro-bounce@xxxxxxxxxxxxx] On Behalf Of Muranyi, Arpad
Sent: Tuesday, September 18, 2012 7:44 PM
To: 'IBIS-ATM'
Subject: [ibis-macro] A possible clarification BIRD for IBIS v5.1
Hello everyone,
I may have discovered a problem in the IBIS v5.1 specification
which may need a clarification BIRD.
I ran across this question when I was working with an AMI model
from a customer. They had a bunch of List parameters in their
.ami file, written so that the first value was always a repeat
of another value in the rest of the list:
(ParName (Usage In)
(Type Integer)
(Default 0)
(Format List 0 0 1 2 3)
(Description "A parameter that does miracles.")
)
Pg. 177 in the v5.1 specification says this:
List <default value> <value> <value> <value> ... <value>
This defines a discrete set of values from which the user may select one
value
This really doesn't say anything on what the syntax of <default value>
is. I do remember that we had a discussion on this in the ATM meeting
and if I remember correctly, we decided that the first value of the list
is the default, but it is part of the list and didn't have to be repeated,
but I can't find anything on it in the v5.1 spec.
Later, on pg. 182 I see this:
For AMI parameter Types "Range", "Increment" and "Steps" <min value>, <max
value> does not imply slow and fast corners, and the user may select any
value provided by these parameters regardless of what corner is used for
the simulation. If the user does not make a selection for parameter types
"Range", "List", "Increment" and "Steps", the EDA tool shall automatically
use the value defined by Default, if it exists, or the <typ value>
otherwise (regardless of what corner is used for the simulation).
This is even worse, because it applies "<typ value>", a left over from
v5.0
to "List", which was changed to have "<default value>".
Unless someone can help me find the rules on how lists are supposed
to be typed, I think we need to write a clarification BIRD, because
the way I read the specification, it could be interpreted at least
two different ways.
Thanks,
Arpad
=====================================================================
Other related posts:
