RE: Config facts & Dangerous Precedent - was: potential command line option

From: Brophy, Dennis (dennisb@model.com)
Date: Mon May 02 2005 - 10:18:24 PDT

  • Next message: Brophy, Dennis: "RE: 1364 mantis issues"

    Steven,

      Encouraging implementors to bring interpretation issues before a
    standards group for clarification is hopeful, but there is often no
    conflict until there is a second interpretation that is in conflict with
    the first.

      Certainly, the LRMs have numerous instances of ambiguities, but the
    first to interpret the ambiguity, in the absence of anything but the
    LRM, seems sufficient. What I notice is the problems of interpretation
    come when the second or successive implementations come to market. At
    one point in time Si2 had played a role of arbitrator for interpretation
    of gate-level Verilog implementations. A group of ASIC vendors paid to
    have a testing lab put in place that would take ASIC cells and run them
    against an implementation (for a fee) comparing results to what the ASIC
    vendors had expected to be the results. (I would guess this was
    probably something close to or equal to XL.) I think this activity had
    far greater impact and value than if these issues had been arbitrated in
    the standards committee. The point being that there may be other venues
    to arbitrate conflict besides the standards group.

    -Dennis

    -----Original Message-----
    From: owner-btf@boyd.com [mailto:owner-btf@boyd.com] On Behalf Of Steven
    Sharp
    Sent: Wednesday, April 27, 2005 1:02 PM
    To: btf@boyd.com; etf@boyd.com; sv-bc@eda.org; sv-ec@eda.org;
    cliffc@sunburst-design.com
    Subject: Re: Config facts & Dangerous Precedent - was: potential command
    line option

    Cliff wrote:
    >Now we have to determine what constitutes a valid "attempt to get the
    >ambiguity clarified before implementing."

    This part of my comment was less about the history of this issue, and
    more trying to promote better practices in the future.

    Despite our efforts, there will be ambiguities in the P1800 LRM. If
    everyone goes off and implements their own interpretation, then we won't
    have a standard language. We need to encourage implementors to bring
    such things to the standards group for clarification before
    implementing.

    Steven Sharp
    sharp@cadence.com

    This archive was generated by hypermail 2.1.4 : Mon May 02 2005 - 09:55:16 PDT and
    sponsored by Boyd Technology, Inc.