Subject: Minutes Verilog++ 9th Committee Meeting
From: David Kelf (davek@co-design.com)
Date: Mon Oct 08 2001 - 10:55:34 PDT
Verilog++ 9th Committee Meeting
October 8th, 2001
DaveK taking minutes
Attendees
(aaaaaaaaa) Vassilios Gerousis *
(arar-aaaa) Dave Kelf *
(aaa--aaaa) John Sanguinetti *
(--ra-aaa-) John Emmitt
(--a-aa---) Dennis Brophy
(aaaaaaaaa) Stu Sutherland *
(----aaaaa) David Knapp
(raaar-aaa) Tom Fitzpatrick (Peter Flake) *
(-aaaaaaa-) Phil Moorby
(-aaaaaaaa) Anders Nordstrom
(-a-aaaaaa) Cliff Cummings
(aa-aaaaaa) Simon Davidmann *
(---aaaa--) Harry Foster
(aa-a-a-aaa) Stefen Boyd *
(aaaaaaaaa) David Smith *
(-a-a--aa-) Mike McNamara
(aa) Kevin Cameron *
(-a) Andy Tsay
Quick discussion around the name - go with Verilog-ACE for now.
Stu's doc. Many people have read a good portion of the documentation.
PeterF has sent an email with feedback on the whole doc.
Discussion on interfaces. David S - would like to extend the sense of
implicit types that already exists in Verilog - useful for analog flow and
other applications. Various issues with port types - different
configurations might require different port types. The lack of this might
require users to deal with more complexity than necessary. PeterF thinks
this could be solved with parameters or by having required types in an
interface to allow the selection of the required type for the configuration
in use. Stu - would type parameter help where it could be driven from
module using connection. PeterF - looks like there are a number of
solutions to this that should be examined.
Review Stu's doc Sections 1 - 6.
Simon - change system level design to architectural level - more
appropriate for advancing hardware design. Use Verilog-ACE as a name for
now (Verilog ACellerra Extensions).
PeterF - Bit widths (int, long int) should be defined as respective exact
widths. Some discussion around char and the fact that it has been defined
as different bit lengths. PeterF char should be left implementation
dependent - a la C.
PeterF - mask/unmask removed? 2 state/4state should be used. Stu mentioned
that masked/unmasked not used elsewhere in doc.
DaveK - does timescale need to be considered. PeterF - suggests that
default be set to 1s for compatibility. Stu - concern about how compatible
this would be for existing tool chains/libraries. Suggests referring to the
IEEE1364 nomenclature.
DavidS - Concern about the use of Scalar. Lose scalar where it is used in
the datatype heading. Check for use of scalar elsewhere to see if it is
consistent.
Vassilios - Interfaces section still needs to be done. Co-Design
responsibility.
DavidS - Does timescale need to be more closely defined. PeterF - defined
in section 12. Could do with a bit more clarification with regard to hierarchy.
Glossary of terms required - PeterF to look at.
Various other minor clarifications noted.
PeterF - initialization of variables should be included. On IEEE committee
apparently there were complaints about this due to implementation. Works if
these are treated like an initial block. Should be included.
Stu - is intent correct in force/assign rewrite. Yes it is but check
Peter's email regarding writing variables through ports.
DavidR - Attribute should have default of bit 1. PeterF to check on the use
of attributes for modport within interfaces.
Moving forward - Final sections (7 onwards) to be reviewed next meeting.
Stu will get Draft 2 ready for the Nov 5th meeting.
LAST CHANCE FOR FEEDBACK TO BE INCLUDED BEFORE OCT 22ND MEETING - PLEASE
EVERYONE PROVIDE FEEDBACK BEFORE THIS MEETING, PREFERABLY BY EMAIL
NEXT MEETING 9 AM pst, Monday 22nd Oct
405 244 5555 code 3715
Future meetings
5th Nov
face 2 face in week of Nov 12th
______________________________
Dave Kelf
VP Marketing
Co-Design Automation, Inc.
Tel: 1 877 6 CODESIGN ext 404
Mobile: 1 617 571 9883
Fax: 1 781 662 2281
Email: davek@co-design.com
Web: www.co-design.com
www.superlog.org
Latest News:
http://www.co-design.com/news/index.htm
"Faster, Smarter Verilog"
______________________________
This archive was generated by hypermail 2b28 : Tue Oct 09 2001 - 05:45:42 PDT