[sv-ec] RE: Resending : Clarification or Expedited resolution of Mantis ID 3657

As I understand it, even a singleton value is part of a range.

The BNF (using the simpler 2009 BNF, the 2012 BNF has the same intent in this context) is

bins_or_options ::=
coverage_option
| [ wildcard ] bins_keyword bin_identifier [ [ [ expression ] ] ] = { open_range_list } [ iff ( expression) ]
...

I think that as it stands, there really is a contradiction.

19.5.4 says,
When a bin definition includes an X or Z, it indicates that the bin count should only be incremented when the sampled value has an X or Z in the same bit positions, i.e., the comparison is done using ===.

19.5.7 says,
An implementation shall issue a warning under the following conditions:
...
3) If b yields a value with any x or z bits...

If a warning is issued for a bins element, the following rules shall apply:
- If an element of a bins covergroup_range_list or set_covergroup_expression is a singleton value b, the element shall not participate in the bins values.

Regards,
Shalom


From: owner-sv-ec@eda.org [mailto:owner-sv-ec@eda.org] On Behalf Of Arturo Salz
Sent: Monday, March 04, 2013 20:46
To: Anupam Singal; sv-ec@eda.org
Cc: Devraj Goyal; Arturo Salz
Subject: [sv-ec] RE: Resending : Clarification or Expedited resolution of Mantis ID 3657

Anupam,

The LRM does explicitly say that, look at the second bullet:

If a warning is issued for a bins element, the following rules shall apply:

-      If an element of a bins covergroup_range_list or set_covergroup_expression is a singleton value b, the element shall not participate in the bins values.

-      If an element of a bins covergroup_range_list is a range  [b1 : b2]  and  either  b1  or  b2  contains any x or z bits or every value in the range would generate a warning, then the element shall not participate in the bins values.

-      If an element of a bins covergroup_range_list is a range [b1 : b2] and there exists at least one value in the range for which a warning would not be issued, then the range shall be treated as containing the intersection of the values in the range and the values expressible by the effective type of e.

This could be made more clear.

                Arturo

From: Anupam Singal [mailto:sanupam@cadence.com]
Sent: Monday, March 04, 2013 3:55 AM
To: sv-ec@eda.org<mailto:sv-ec@eda.org>
Cc: Devraj Goyal; Arturo Salz
Subject: RE: Resending : Clarification or Expedited resolution of Mantis ID 3657

Hi Arturo ,

Thanks for your inputs for below . However still LRM should clarify further ..

Currently 19.5.7 does not explicitly mention that it is for range specifications only and also refers to
singleton values in the paragraph of warnings ..

While triple inequality may be correct in this case LRM must make further clarification :


i.                       that section 19.5.7 deals  only with  range specifications .

ii.                   Reference to singleton values as well in the section must be removed ..

Regards ,
Anupam Singal



From: Arturo Salz [mailto:Arturo.Salz@synopsys.com]
Sent: Friday, March 01, 2013 12:05 AM
To: Anupam Singal; sv-ec@eda.org<mailto:sv-ec@eda.org>
Cc: Devraj Goyal
Subject: RE: Resending : Clarification or Expedited resolution of Mantis ID 3657

Anupam,

I don't believe there is a contradiction. The statement in 19.5.7 is referring only to an range specification, hence, the warning is only required when the expression has X/Z bits in a range where it is ambiguous how to compute range/membership with expressions involving X/Z, but not in a singleton value, which uses === . This was introduced in Mantis 1313 - http://www.eda.org/svdb/view.php?id=1313.

                Arturo

From: owner-sv-ec@eda.org<mailto:owner-sv-ec@eda.org> [mailto:owner-sv-ec@eda.org] On Behalf Of Anupam Singal
Sent: Thursday, February 28, 2013 5:06 AM
To: sv-ec@eda.org<mailto:sv-ec@eda.org>
Cc: Devraj Goyal
Subject: [sv-ec] Resending : Clarification or Expedited resolution of Mantis ID 3657



There seems to be a conflict in the section 19.5.4 section ( Wildcard specification of coverage point bins)
and section 19.5.7 (Value resolution )  for a particular case of NON wild card bin having X or Z .


eg   for   bin[]=   { 4'b10XZ  }  ;

Q )     Should one check explicitly for X and Z using === operator as per  section 19.5.4

OR

   Issue a Warning as per section 19.5.7  Value resolution, sub-section b 3) as below


" ...
3) If b yields a value with any x or z bits.

This rule does not apply to wildcard bins because x
and z shall be treated as 0 and 1 as described above.
... "

The same had been captured earlier in Ref Mantis ID 3657 and is awaiting comments / resolution .

Could  someone clarify the above case above by way of comments /  clarification or expedite the resolution on this ID ?


Regards ,
Anupam Singal
sanupam@cadence.com<mailto:sanupam@cadence.com>
---------------------------------------------------------------------
Intel Israel (74) Limited

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.