Re: errata/54: 5.6.6 Port connections - inaccurate description

From: Brad Pierce (Brad.Pierce@synopsys.com)
Date: Tue Nov 26 2002 - 18:40:00 PST

  • Next message: Shalom Bresticker: "Re: errata/54: 5.6.6 Port connections - inaccurate description"

    Precedence: bulk

    The following reply was made to PR errata/54; it has been noted by GNATS.

    From: "Brad Pierce" <Brad.Pierce@synopsys.com>
    To: <etf-bugs@boyd.com>
    Cc:
    Subject: Re: errata/54: 5.6.6 Port connections - inaccurate description
    Date: Tue, 26 Nov 2002 18:39:14 -0800

     In the note at the end of section 12.3.8, the LRM says --
     
        "A port that is declared as input (output) but used as
         an output (input) or inout may be coerced to inout.
         If not coerced to inout, a warning has to be issued."
     
     And after issuing the warning, then what? Unless it's intended
     that the message be a fatal error, something more must still
     happen. What is the correct behavior if the port is not coerced
     to inout?
     
     Moreover, contrary to the note, directions are not really declared
     on ports, but instead are declared on the port identifiers within
     port expressions. If the port expression is a concatenation, but
     the port identifiers within it are declared with mixed directions,
     such as half input and half output, what is the direction of the
     port?
     
         module m (.p({a,b})) ;
         input a;
         output b;
         not(b,a);
         endmodule
     
     To me, it seems like either p is being declared here as an inout
     (no coercion involved) or this is an error.
     
     Yet I don't see any place in the standard that speaks to this.
     And the function acc_fetch_direction() would return "accMixedIo"
     for port p. According to Table 139, there are four kinds of
     port direction --
     
        Input only
        Output only
        Bidirectional (input and output)
        A concatenation of input ports and output ports
     
     But I don't see this fourth kind of port direction mentioned
     anywhere else in the LRM.
     
     -- Brad
     
     

    This archive was generated by hypermail 2.1.4 : Tue Nov 26 2002 - 18:40:49 PST and
    sponsored by Boyd Technology, Inc.