Subject: Re: [vhdl-200x-perf] Re: Performance issues
From: Steve Casselman (sc@vcc.com)
Date: Tue Apr 29 2003 - 15:46:04 PDT
Would you get a race condition if you had unit delay? So this is kind of
what I was talking about before you'd have to be a real expert to know to
use a shared variable and then if you wanted to switch between accuracy and
speed you would have to change your code (or make it much more complicated).
It would be great if there was a simple switch or change of library that
told the simulator to swap in all the tricky fast stuff and then go back to
the slower more accurate models when you get near the end of your design
cycle. Some way where all the expert trade offs for speed vs accuracy are
done in the standard and the end user just has to specify the "fast" or
"accurate" system.
Steve
----- Original Message -----
From: "Stephen Bailey" <stephen@srbailey.com>
To: <vhdl-200x-perf@server.eda.org>
Sent: Tuesday, April 29, 2003 3:41 PM
Subject: Re: [vhdl-200x-perf] Re: Performance issues
> VHDL does have the equivalent. They are called shared variables. Also, I
think you would have a difficult time finding more than 1
> or 2 people from the VHDL user community that would want the Verilog race
problem introduced into VHDL.
>
> Nonetheless, the Performance group has a request for light-weight signals
which covers this area.
>
> -Steve Bailey
>
> > One reason Verilog is faster than VHDL is that has "blocking"
assignments
> > which propagate signal changes immediately to other modules rather than
> > "non-blocking" where the change is scheduled for the next delta or
later.
> > VHDL only has the later which always incurs an extra
scheduling/descheduling
> > overhead.
> >
> > You could speed VHDL up by adding a "blocking" assign.
> >
> > Note: the trade-off is speed vs. stability, blocking assigns lead to
> > intrinsically
> > racy behavior. The better way to get a speed up is to use parallel
> > processing
> > in which case you want the more stable behavior.
> >
> > Kev.
> >
> > --
> > National Semiconductor, Tel: (408) 721 3251
> > 2900 Semiconductor Drive, Mail Stop D3-500, Santa Clara, CA 95052-8090
> >
> >
> >
> >
>
This archive was generated by hypermail 2b28 : Tue Apr 29 2003 - 15:46:41 PDT