Minutes for P1500 Working Group Meeting
June 24, 2003

Attendees

Karim Arabi  
Luis Basto
Dwayne Burek
Tapan Chakraborty
Wu-Tung Cheng
Chen-Huan Chiang
Vivek Chickermane
Mike Collins
Grady Giles
Alan Hales  
Rohit Kapur
Brion Keller
Teresa McLaurin
Mike Ricchetti
Francisco da Silva
Paul Soong
Lee Whetsel
Yervant Zorian

Scribe

Mike Collins

Agenda

  1. Introduction
  2. Task Force Updates
  3. Compliancy Rules for Partially Wrapped Cores
  4. Balloting Process
  5. Essential Patent identification
  6. Future Meetings and Misc.

Minutes

1. Introduction

Yervant provide opening remarks and described the main topics to be discussed in this WG meeting.

2. Task Force Updates

Various task forces provided their brief status updates:

a. Scalable Architecture Task Force; Lee Whetsel

The scalable architecture has been comprehensively reviewed and is quite stable. The consensus of the Task Force is that the last two chapters on compliancy rules could use some more review.

b. Documentation Task Force; Francisco da Silva

The draft D0.7 will be published today (June 24th) and forwarded to the reviewers.

The plug and play chapter is new as is the example reference chapter. There are some thoughts about adding to the examples.

The partially-wrapped cores material is postponed, until the completion of its discussion by the Working Group.

Review forms have been generated from IEEE database.

1450.6 and 1500 drafts will go out together. The reviewers will have 6 weeks to provide their review forms.

c. Compliancy Definition and Information Model Task Force; Karim Arabi

Compliancy has a few remaining issues and were discussed extensively. The main problem is with partially wrapped cores. The remainder of the meeting was used to discuss issues related to dealing with partially wrapped cores and a few concerns about unwrapped compliant.

3. Compliancy Rules for Partially Wrapped Cores

To summarize the discussion some of the philosophies about compliancy are as follows:

(Architecture Philosophy?); The idea is that the standard is a specification for architecture. It has been crafted to be as flexible as possible. The business of dealing with partially wrapped or un-wrapped cores is an implementation detail that should be handled by EDA tools.

The alternative is (the realist philosophy??) that a lot of the business is concerned with unwrapped or partially wrapped cores. If the standard does not address these issues it is liable to be irrelevant.

The major concern was with partially wrapped cores.

Grady emailed a document about difficulties with CTL when trying to wrap partially wrapped cores. And Rohit replied with some thoughts. The issue was how to deal with the dangling signals that need to come from the wrapper elements that aren't there to the wrapper elements that are included by the core provider. It was necessary to clarify how CTL might handle this case and if the things that are in place now to do this.

Examples were sighted of partially wrapped cores from ARM that typically come with a wrapper and no WIR or bypass register. In addition, cores that may be wrapped for 1149.1 either partially or completely may be candidates to be integrated into P1500 wrappers. Several participants thought that differential and bi-directional signals would likely come with the wrapper cell pre-installed. And, there was considerable discussion of shared functions; that is, wrapper logic that is part of the functional logic.

Rohit expressed the desire to give it "A good hard day" and look at what happens. He is optimistic that CTL can have a way to describe the operation of the "dangling signals" such that this is doable with a modification of the CTL.

This created an action item and an add hoc group headed by Rohit Kapur that includes but is not limitied to:

They are expected to have several meetings to come up with a full proposal to the Working Group to be presented in mid July time frame.

It was agreed that the Working Group meets to cover the remaining review process. This is mainly to discuss the last chapter of the D0.7.

4. Ballot Process

Yervant described the ballot invitation process and informed the WG that the ballot invitation will start soon. Grady mentioned that the membership to IEEE Standards Association is a must to participate in the balloting process. They are expected to have a meeting around the 14th or 15th of July to report on the activity.

5. Essential Patent Identification

Yervant asked if there is any newly identified patents that the WG members feel that are essential to P1500. There were no new patents since last meeting.

The previous identified four patent holding companies have been approached by Yervant to inform us if their patents are essential.

6. Future Meetings Timeline and Misc

The WG will meet on a weekly basis (Thursday 2 hour conference calls) to review the last chapter.

Partially wrapped core ad-hoc team to prepare a proposal for the Working Group. WG meeting will be held on July 22.

Reviewers of D0.7 will report back by the first week in August.

The Documentation Task Force will continue to incorporate any changes from the above activities to D0.8, which will be the document for publication.

Meeting adjourned at 12pm (PDT)