RE: [sv-bc] streaming operator: svlog LRM

Hi Jonthan,

I am very sorry for the confusion. What I had actually meant was the
following:

Typdef Struct{
            logic a;
            logic b;
} myType
 
Typdef Struct{
            logic a;
            logic b;
            logic c;
            logic d;
} myType1

Module test( input myType in1, input myType in2, output myType out1
[1:0], output logic[3:0] out2, output myType1 out3);

            assign out1 = {>>{in1, in2}}; 
            assign out2 = {>>{in1, in2}};
            assign out3 = {>>{in1, in2}};
endmodule

What this would mean is that all assignments are 4 bit assignments
{in1.a, in1.b, in2.a, in2.b}. So, as you said all these three should be
perfectly fine, right?

I will also go through the link you have given and get back to you if I
have any doubts. 

Thanks a lot for your help.

Regards,
Kapil



-----Original Message-----
From: Jonathan Bromley [mailto:jonathan.bromley@doulos.com] 
Sent: Thursday, July 19, 2007 5:01 PM
To: Kapil Kaushik; sv-ec@eda-stds.org; sv-bc@eda-stds.org
Subject: RE: [sv-bc] streaming operator: svlog LRM

hi Kapil

In your example....

Typdef Struct{
            logic a;
            logic b;
} myType
 
Typdef Struct{
            logic a;
            logic b;
            logic c;
            logic d;
} myType

I'm guessing you meant these two typedefs to have different names?
 
 
  Module test( input myType in1, input myType in2, 

and in1, in2 should be of the two different types???

    output myType out1 [1:0], output logic[3:0] out2, output out3);

            assign out1 = {>>{in1, in2}}; 
            assign out2 = {>>{in1, in2}};
            assign out3 = {>>{in1, in2}};

I think that *all* these assignments are illegal, because 
the streaming concatenation {>>{in1,in2}} contains six bits -
it's effectively {in1.a,in1.b,in2.a,in2.b,in2.c,in2.d} -
and the targets are smaller than 6 bits.  However, if you
provided a 6-bit or larger target, then the assignments 
would be fine.  The bitstream resulting from the {>>{}} 
operation is automatically cast to the type of the target.
What you *cannot* do is use the bitstream in a situation 
where it has a type of its own.  For example, if A is an
integer variable,

   A = {>>{in1, in2}} + 1;

would be illegal.

Mantis item 1707 covers this area.  I've recently written
a new proposal for it; I'd be very interested to know if
that new proposal answers your concerns.  The whole
Mantis item is at
  http://www.eda-stds.org/svdb/view.php?id=1707
(log on as username guest, password guest)

and here is the modified LRM text as a PDF for convenience.
--
Jonathan Bromley, Consultant

DOULOS - Developing Design Know-how
VHDL * Verilog * SystemC * e * Perl * Tcl/Tk * Project Services

Doulos Ltd. Church Hatch, 22 Market Place, Ringwood, Hampshire, BH24
1AW, UK
Tel: +44 (0)1425 471223                   Email:
jonathan.bromley@doulos.com
Fax: +44 (0)1425 471573                           Web:
http://www.doulos.com

The contents of this message may contain personal views which
are not the views of Doulos Ltd., unless specifically stated.

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.