Re: Verilog-AMS Committee Meeting Minutes - Jan 11 2007

Martin -
Sorry I missed the meeting.

You didn't list all the typos in the minutes, so I'll mention
a few things I noticed.

Page 221, math functions: it's $log10, not $log (there's
a $log in digital that does something else).


I don't understand the exception at the end of this
sentence from pages 224-225:
  For each % character (except %m and %%) which appears
  in a string, a corresponding expression shall follow
  the string in the argument list, except a null argument.


Page 225: %s for converting sequences of 8 bits into 
ascii characters.  What does SV do for printing actual
strings?  (This %s is useful for regs in 1364, which
are 8-bit-per-character arrays.)


Page 226: $fopen in 1364-2005 has a second form,
  fd = $fopen( "file_name", type );

Is this not supported in AMS?  I recall something about
how the reason that $fopen was in the AMS LRM 2.0 was
that it differed from that in 1364-1995, whereas the
other file functions ($fprint) were the same.


The $rdist functions need some thought.  Per a recent
e-mail on one of the SV lists, $random can be called
without a seed (in which case the simulator sets up
an internal variable to track it) -- but *cannot* have
a parameter as a seed; this is different from what we
have specified.

Also, what happens if "global" or "instance" is not
specified, which is the default (what happens in a
Monte-Carlo analysis in that case)?  What does it mean
"for each instance that references this value"?  Does
the simulator have to track every parameter in every
instance that depends, however indirectly, on the
random value, or does this only apply to instances
that set a parameter directly from the random function?

I just reassigned a mantis entry to you:
http://www.eda-stds.org/mantis/bug_view_page.php?bug_id=0000920
(be sure to read my bugnote).

-Geoffrey

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