Subject: Verilog++ 14th Committee Meeting Minutes
From: Dave Kelf (davek@co-design.com)
Date: Mon Dec 10 2001 - 11:46:23 PST
Verilog++ 14th Committee Meeting
December 10th, 2001
Vassilios leading. DaveK taking minutes
Attendees
(aaa-aaaaaaaaaa) Vassilios Gerousis *
(aa-aaarar-aaaa) Dave Kelf *
(----aaa--aaaa) John Sanguinetti
(aaaaa--a-aa---) Dennis Brophy (part) *
(aaaaaaaaaaaaaa) Stu Sutherland *
(--------aaaaa) David Knapp
(aaaaaraaar-aaa) Tom Fitzpatrick *
(arraa-aaaaaaa-) Phil Moorby *
(-aaaa-aaaaaaaa) Anders Nordstrom
(aaa--a-aaaaaa) Cliff Cummings *
(aaaraaa-aaaaaa) Simon Davidmann *
(aaa-aaaaa-a-aaa) Stefen Boyd *
(aaaaaaaaaaaaaa) David Smith *
(-aaa--a-a--aa-) Mike McNamara
(aaaaaaa) Kevin Cameron *
(-a-aa-a) Andy Tsay
(aa) Alec Stanculescu *
(a) Adam Krolnik *
Key a = attended, - = not attended,
r = represented by someone else, * = attended this meeting
No questions on minutes from previous meeting
Actions reviewed and most have been completed and, where appropriate, will
be included in the next draft - uncompleted actions from previous meetings are:
Kevin - Extern Proposal - he will resend.
Peter - Example on adding wires on page 57 of draft 3.
Cliff - to provide list of items to be deprecated
Question from Alec - Have we considered mechanisms to interface with
foreign languages, for example VHDL. In general this has not been considered.
Issue with doubling of emails for some people on the reflector - to be
check by Vassilios.
Also note that DaveK and Vassilios will work on web site this week.
BNF Review
- Generally OK.
- Peter F had one or two errors reported via email.
- Adam has done some work on the BNF and sent it to everyone.
- Some name differences between SystemVerilog and Verilog2K1 BNFs. This
should be fixed.
- Cliff and Stu will add BNF changes, after a comparison with the IEEE
Verilog 2K1 BNF, to Adam who can check it. After this point we will
continue to review it in meetings. Alec will also run it through some of
his in-house tools. Vassilios to contact the IEEE to get permission to use
existing Verilog2K1 BNF, and for the appropriate files.
Actions: Vassilios, Stu, Cliff, Adam, and Alec
Cliff's FSM examples
- Cliff sent out his statemachine examples coded using SystemVerilog
syntax. Cliff's summary is that code example takes same code volume as
straight Verilog and had other problem like not being able to code one hot
machines, etc. Proposes removing the statemachine syntax.
- Agreement on changing state keyword.
- Peter looked at one example and noted that the code volume is infact
smaller. Cliff pointed out that enumerated types can give you much the same
capability and code compaction.
- Alec discussed UDLI and the benefits provided there. Also mentioned that
we should be looking at efficient coding and FSM encoding implementation.
- Simon pointed out that maybe the problem is that these constructs are
designed for more abstract machine representations.
- Stefen - question is how do we get async reset in a statemachine. Maybe
transition between async versus sync is creating this issue. Alec suggested
features for async reset.
- Peter - hierarchical state machines are supposed to be handled better,
and Peter did provide an example. Cliff would be happy to review this.
Adam - notes that syntax does not help the user set up the output of the
machine - just the transisition coding. Alec - depends on Mealy versus
Moore machine. Adam - yes - engineers just consider the outputs, regardless
of the machine style.
Action: Adam to explain point of view on outputs in email.
Alec - similar issue to async resets - also should consider multiple
machines working off different clocks - more important for high level
machines. Alec to provide Peter with a UDLI manual.
Peter will consider async reset and make a stronger case for the existing
syntax. Also will consider one hot encoding and hierarchical machines.
Action: Peter F to provide examples to show power of existing syntax.
Should look at Cliff's examples.
Interface Section
Peter provided a verbal discussion on interfaces at the face to face
meeting four weeks ago, which was very helpful for everyone at the meeting.
If his explanation can be captured in words, this would be fine for the manual.
Action: Peter, or someone from Co-Design, to look at providing explanation.
Co-Design folks need to look at this internally to figure out how this can
be done.
Stu's email comments
Just prior to the meeting Stu sent out a few questions in an email that
needed some answers. From the emails, Stu to check flip flop coding, Peter
to reword change example. Other clarifications provided in meeting.
The next meeting will be held on January 7th, 2002. Time will be 9:00am
PST, 12:00pm EDT, 5:00pm GMT, 6:00pm Central Europe. Usual call in number -
405 244 5555 access code 3715.
Happy holidays everyone.
______________________________
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 : Mon Dec 10 2001 - 17:28:58 PST