[sv-bc] Issue Review

From: Gordon Vreugdenhil <gordonv_at_.....>
Date: Thu Apr 20 2006 - 10:32:16 PDT
Here are my comments and classifications of the issues that were
in my review buckets.


1172, major, high
    Issue: 15.1, 15.6, A.7.5.3: scalar_timing_check_expressions has redundancies
    Comment: Various simulators have different interpretations in the
             presence of X and Z values.

1173, (enhancement), feature, normal
    Issue: variable width floating point in Verilog 200X

1174, (enhancement), minor, high
    Issue: Allow instance array connected to data array
    Comment: This is a pretty natural enhancement.  It should
             probably address multidimensional cases as well
             as various edge conditions involving size
             mismatches.

1175, (enhancement), feature, normal
    Issue: Add field widths to print formats
    Comment: I doubt there will be anything contentious in this enhancment,
             but a full proposal is needed.

1176, (enhancement), feature, immediate
    Issue: Proposal for Extending Verilog Data Types
    Comment: I think this should be closed as addressed by P1800 and
             any remaining enhancements be filed as a new request.

1179, (enhancement), text, low
    Issue: Add Quick Reference
    Comment: I'm not sure this belongs in an LRM.

1180-1188: (enhancement), feature, low
    Comment: This entire set is Verilog-AMS integration related.
             All of this should be addressed during the main
             Verilog-AMS integration.


1264, major, normal
    Issue: 8.13.1,2: "default:value" description unclear
    Comment: This is related to Mantis 1265

1265, (clarification), minor, normal
    Issue: 8.13.1: unclear sentence - what is "more arrays" ?

1266, (clarification), text low
    Issue: 8.16 Operator overloading syntax - at most 2 arguments
    Comment: No error here; just a suggestion to make the grammar and
             terminology more precise.

1267, (enhancement), feature, low
    Issue: ifx-elsex enhancement, assign x to outputs if condition is x
    Comment: The requested feature can be expressed in other forms in
             the language so I think we can wait with this


1268, major, normal
    Issue: search order for data object and scope names is unclear and inaccurate
    Comment: This issue interacts with other name resolution issues
             that have been raised on the reflector.  Any comprehensive
             name resolution proposal should deal with the issues here
             as well.

1271, (clarification), trivial, immediate
    Issue: A.6.2: delay_or_event_control should be optional in blocking_assignment
    Comment: The suggested grammar change is not strictly necessary (and in fact
             would create reduce/reduce conflicts if directly implemented) but
             does avoid an easy to make error in reading the grammar.  I think
             that we should have an email vote on it and if anyone opposes the
             change, just close the issue as "won't fix"


1272, trivial, immediate
    Issue: Syntax 10-3: &&& should be all red
    Commnet: This is a duplicate of 929.  1272 should be closed
             and 929 should be an email vote.

1280, major, high
    Issue: Wrong replication example at end of 3.8
    Comment: Due to the syntax change, inconsistent examples lead
             to quite a bit of confusion.  I think that Greg's
             suggested expansion is correct.


1291, feature, immediate
    Issue: range and signedness are parallel attributes of every packed dimension
    Comment: Brad has suggested that we close this as "won't fix".  I agree.

1292, minor, normal
    Issue: clarify compiler directive placement (non-)restrictions
    Comment: most of the clarification is likely "minor" but
             there are interactions with `protect that might
             argue for bumping this to major.  I left it
             as minor but it wouldn't take much for me to agree
             that this is major.


Gord.

-- 
--------------------------------------------------------------------
Gordon Vreugdenhil                                503-685-0808
Model Technology (Mentor Graphics)                gordonv@model.com
Received on Thu Apr 20 10:32:20 2006

This archive was generated by hypermail 2.1.8 : Thu Apr 20 2006 - 10:32:33 PDT