Subject: [sv-ac] SV-AC draft 1 comments
From: Adam Krolnik (krolnik@lsil.com)
Date: Tue Apr 08 2003 - 08:51:15 PDT
Good morning;
I don't think I will be able to attend the meeting.
Please review these comments on the rest of the document.
Also, see my mail from April 7 labeled "Assertion technical issues."
Issue:
1. Usefulness of templates.
2. Dynamic variable context.
Thanks.
Adam Krolnik
Verification Mgr.
LSI Logic Corp.
Plano TX. 75074
From page 8.
first sentence, "note that variables referenced in a seq (sic)"
paragraph, "sequence s_20 _l ... A sequence can be referenced in a
property, an
assert statement or a cover statement.
Strike 'assert statement'
Strike a sequence (referred as a sub-sequence) can be defined within..."
Section 1.7.1 - update document with correct repetition operator names
and make sure
syntax matches name. Make sure description matches names replaced.
pg 11, starting with "A sequence can be repeated" is redundant from the
above example.
pg 12, update repetition name to goto repetition. Also, update
repetition afterward to
nonconsecutive repetition.
1.7.4 give a sample picture like was done in figure 1-5.
pg 19, sequence ts1 can complete in 'c ##1 d' This is the correct value.
pg 26, sect 1.8
what happens if you say,
(a, x = data) [*1:3]]; ## [1:2] y ##1 z == x
Which value of X do you get at the end? How would this work?
also, a [*1:3] ## [1:2] y, x = data ##1 z == x
What data would you get? when there are overlaps in a and y?
pg 29, 1-18 - Are sequences definable in properties?
pg 29, disabble (sic)
Why is first_match() applied to the top? And what happens when
implication is
there - same thing?
Recommend spelling out what happens with implication. Is this meaning
that if I have
an antecedent produce multiple matches that I can get multiple
passes/failures of the consequent?
What are the associativity rules for $first_match( a => b) ? Is
$first_match(a) => b equivalent?
pg 30, sequences declared inside are for local use only - this is not
true - hierarchical
name allows access.
pg 31 1.11.1 "is different than the desalination sequence, method
matched " What is that?
pg 33, first line "else fail_stat;" Semi is part of fail statement this
would mean that end ; would parse
and $display(); ; would be necessary too.
pg 37. What about the default clocking stuff - it's a shorthand for
defining the clock, but
there must be a clock in the block for it to work... What about
blocks without
clocks.
pg 36, can you have a label on an assert statement in a module? The BNF
in draft4 does not allow this.
r4_p : assert r4;
What about properties in functions? Are they out?
pg 40 - template contains "clock domain declarations" Nope, not in
assert statements BNF.
pg 1, Syn 1.2 - must include template declaration and template
instantiation in
concurrent_assertion_item if you want to use them. Should
procedural-assert_item
allow template instantiations ?
This archive was generated by hypermail 2b28 : Tue Apr 08 2003 - 08:54:22 PDT