[sv-cc] FW: Proposal 528 does not agree with 473 -- please fix!

 


--------------------------------------------------------- 
James H. Vellenga                            978-262-6381 
Engineering Director                   (FAX) 978-262-6636 
Cadence Design Systems, Inc.         vellenga@cadence.com 
270 Billerica Rd 
Chelmsford, MA 01824-4179 
"We all work with partial information." 
---------------------------------------------------------- 
  

-----Original Message-----
From: Jim Vellenga 
Sent: Wednesday, May 04, 2005 10:14 AM
To: Francoise Martinolle
Subject: RE: Proposal 528 does not agree with 473 -- please fix!

Point taken.

So I would suggest the following changes:

1) Put a green note in front of Note 3 to indicate that
this version supersedes the change approved as part of
Item 473.

2) To me, "the class defn that the class obj points to
when the query is made" is technically correct.  Somewhere
in the implementation there has to be a pointer from the
object back to the class definition.  But is this
going to be meaningful to the VPI application developer?
I'm wondering if it would mean more to say "the class
defn used to create the class obj," or something like
that.

3) "the class defn that was used to create the class
var" likewise seems misleading.  A class var may in fact
be static, and often will be.  The
static class var will be created some time prior to
running VPI, so in this sense the statement is correct,
but as written suggests that the class var was created
dynamically.

In addition, the class var can in fact be created
dynamically as part of a class obj.  In this case,
the class defn that the class var is declared with
may differ from the class defn used to create the
class obj -- and with it the class var.  So in
this case, the "class defn used to create the class
var" would be the class defn of the class obj rather
than that of the class var itself.  What we really
want to return is the class defn of the class var
itself.

So I would recommend going back to the wording of
473 for this part:

"For a class var, vpiClassDefn returns the class defn
with which the class var was declared in the SystemVerilog
source text."

4) When the last sentence of the first paragraph talks
about "The difference," it's not clear (until you read
the example itself) what difference is being talked
about.  I recommend changing the sentence to

"The difference between the two usages of vpiClassDefn
can be seen in the example below:"

5) I'd move the third sentence before the second sentence,
so that both the sentences that talk about vpiClassDefn
for a class var are together.

Regards,
Jim V.

--------------------------------------------------------- 
James H. Vellenga                            978-262-6381 
Engineering Director                   (FAX) 978-262-6636 
Cadence Design Systems, Inc.         vellenga@cadence.com 
270 Billerica Rd 
Chelmsford, MA 01824-4179 
"We all work with partial information." 
---------------------------------------------------------- 
  
 

] -----Original Message-----
] From: Francoise Martinolle 
] Sent: Tuesday, May 03, 2005 11:10 PM
] To: Jim Vellenga; Francoise Martinolle
] Subject: RE: Proposal 528 does not agree with 473 -- please fix!
] 
]  
] The reason is that I do not agree with the changes made in 473 once I 
] added the class obj.
] 
] 
] -----Original Message-----
] From: Jim Vellenga [mailto:vellenga] 
] Sent: Tuesday, May 03, 2005 3:50 PM
] To: Francoise Martinolle
] Subject: Proposal 528 does not agree with 473 -- please fix!
] Importance: High
] 
] Francoise,
] 
] Your updated proposal for 528 still does not agree with the 
] change for 473. 
] 
] Regards,
] Jim V.
] 
] --------------------------------------------------------- 
] James H. Vellenga                            978-262-6381 
] Engineering Director                   (FAX) 978-262-6636 
] Cadence Design Systems, Inc.         vellenga@cadence.com 
] 270 Billerica Rd
] Chelmsford, MA 01824-4179
] "We all work with partial information." 
] ---------------------------------------------------------- 
]   
] 
] -----Original Message-----
] From: Jim Vellenga
] Sent: Thursday, April 28, 2005 8:59 AM
] To: Francoise Martinolle
] Cc: Charlie Dawson
] Subject: RE: [sv-cc] uploaded proposals for 465 and 528
] 
] Francoise,'
] 
] In 528, the changes you propose to 32.22 Note 3 conflict with 
] those that the
] committee already approved as part of Item 473.  Please 
] change your proposal
] to conform to what is in 473.
] 
] Regards,
] Jim Vellenga
] 
] --------------------------------------------------------- 
] James H. Vellenga                            978-262-6381 
] Engineering Director                   (FAX) 978-262-6636 
] Cadence Design Systems, Inc.         vellenga@cadence.com 
] 270 Billerica Rd
] Chelmsford, MA 01824-4179
] "We all work with partial information." 
] ---------------------------------------------------------- 
]   
]  
] 
] ] -----Original Message-----
] ] From: owner-sv-cc@eda.org [mailto:owner-sv-cc@eda.org] On 
] ] Behalf Of Francoise Martinolle
] ] Sent: Wednesday, April 27, 2005 11:22 PM
] ] To: sv-cc@eda.org
] ] Subject: [sv-cc] uploaded proposals for 465 and 528
] ] 
] ]  
] ] 
] 
] 
]