Peter Clifton wrote:
> On Thu, 2008-01-24 at 15:50 +0000, Chris Jones wrote:
>> David Bisset is the expert on the current expansion boards, but as far 
>> as I am aware Ethernet controllers and so on live on the system bus. A 
>> little address decoding should allow your application to live on the 
>> bus, too.
> 
> The application required uncommitted IO lines to be driven as PWM
> signals, although I guess since the system bus will be lots faster than
> my PWM signals, some addressable latch register could probably do what I
> need. The only thing to check, was that I could ensure I got reasonable
> latency on the bus.

The finest way to get your PWM signals via the bus would be to put the 
PWM-generation logic in a CPLD hanging off the bus, in much the same way 
as the Balloon board's FPGA/CPLD hangs off the bus already, thus 
avoiding such frequent (and real-time-critical) bus accesses. The only 
reason to do this would be if the Samosa connector didn't have enough 
wires on it to carry your PWM outputs, I guess.

> The PWM signals are to be generated from the FPGA, so it all sounds a
> little dodgy though... presumably the CPU is "master" of the bus
> normally?

Yes. As far as I know there's no mechanism other than DMA for anything 
else to take control of the bus.

Chris

-- 
Chris Jones - chris@martin-jones.com
Martin-Jones Technology Ltd, makers of Solidlights
148 Catharine Street, Cambridge, CB1 3AR, UK
Phone +44 (0) 1223 655611 Fax +44 (0) 870 112 3908
http://www.solidlights.co.uk/