RE: [sv-bc] A new proposal uploaded for issue 216

I fell for that also, but the example is correct and NC is only
incremented once. Read my earlier post
http://www.eda.org/sv-bc/hm/3101.html for a suggested improvement in the
example.


> -----Original Message-----
> From: owner-sv-bc@eda.org [mailto:owner-sv-bc@eda.org] On Behalf Of
Greg
> Jaxon
> Sent: Tuesday, May 03, 2005 10:57 AM
> To: Brad.Pierce@synopsys.com
> Cc: sv-bc@eda.org
> Subject: Re: [sv-bc] A new proposal uploaded for issue 216
> 
> http://www.eda.org/svdb/file_download.php?file_id=978&type=bug says:
> 
> the following declares three variables GEN[0].my_var1, GEN[1].my_var1
and
> GEN[2].my_var1
> 
> 
> 
>               `define NC 0
> 
>                   for (genvar I = 0; I < 3; I++) begin:GEN
> 
>               `incr NC
> 
>                        var my_var``NC;
> 
>                   end
> 
> --------------------
> 
> Do you have any prior text supporting the notion that generate loops
> are unrolled BEFORE macro substitution occurs?   It must then be the
> case that macro definitions cannot be substituted into the loop bounds
> expressions of generated for loops.   Considering that only the
keyword
> "genvar" was needed to cause this profound inversion of the macro
> processing
> order, I think I have to object that this is too ambiguous for my
taste.
> 
> Greg Jaxon
>