Paul Koning wrote:
> 
> >>>>> "yodaiken" == yodaiken  <[EMAIL PROTECTED]> writes:
> 
>  yodaiken> It's not just the money, but the interface and the hosting
>  yodaiken> system.  What I really want is to be able to invoke a state
>  yodaiken> machine gennerator dynamically, have it generate a circuit,
>  yodaiken> upload the circuit to the hardware, all on line. Even
>  yodaiken> ignoring the horrors of NT, the tools are not designed with
>  yodaiken> this kind of work in mind -- although I'm told that Xilinx
>  yodaiken> has some good tools of this sort hidden away.
> 
> Interesting notion.  I remember early Xilinx app notes that talked
> about reprogramming the device on the fly.  (The example was an
> instrumentation recorder with a CPLD not big enough to hold both
> playback and record logic, so it would load whichever you needed
> depending on what you were doing...)
> 
> As for tools, I remember that some people at DEC's Palo Alto research
> center developed their own Xilinx tools because the stock ones did not
> do a good (high density) job on the sort of designs they were working
> on.  I wonder if those were ever made generally available.
> 

You might want to look at http://users.ox.ac.uk/~adrian/harptop.html .
It's a bit dated now. And we generated Xilinx XNF and used their old
tools for the back end. 
But the Harp boards are all about reconfigurability: designed before Tom
Kean's XC6000 chips were released and then killed by Xilinx :-(

ael
--
Dr A E Lawrence (from home)
--- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR
echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED]
----
For more information on Real-Time Linux see:
http://www.rtlinux.org/~rtlinux/

Reply via email to