Re: [Fwd: VCS vs Verilog XL -- Differences in the Handling of Real Numbers]

From: Steven Sharp (sharp@cadence.com)
Date: Mon Oct 14 2002 - 11:45:42 PDT

  • Next message: Shalom.Bresticker@motorola.com: "Re: Blame it on IEEE!"

    Precedence: bulk

    I just did a quick test with Verilog-XL 3.4. It appears that X bits fed
    into $bitstoreal get treated as 0 (which is standard behavior in all similar
    situations). If all the bits are X or 0, then XL produces a result of 0.0,
    not 1.0. So I don't know what they are talking about.

    As far as comparisons with NaN (which can be gotten in ways other than
    $bitstoreal, like computing 0.0/0.0), I consider the Verilog-XL behavior
    to be correct. I believe that it matches the usual C behavior also.

    Steven Sharp
    sharp@cadence.com

    This archive was generated by hypermail 2.1.4 : Mon Oct 14 2002 - 11:47:24 PDT and
    sponsored by Boyd Technology, Inc.