Interface to multiple MCUs

Hello!

I'd like to ask advice of DSP gurus in my doubt. In our design we have
TI's C6414 DSP on main board. This board has to communicate with multiple
sub-boards of the instrument. We want to have bidirectional communication.
At present we use DSP's McBSP in SPI mode and ARM9 MCU on sub-boards.
Little circuitry allows addressing, i.e. its possible to select single
slave for SPI master-to-slave communication. Additional circuitry used to
convey communication request from slaves to DSP. Delivery guaranteed by
software algorithm.

The solution is far from elegant and rather slow. We don't mind to place
more interface chips on boards. I ask community to suggest possible
solutions. I can roughly outline requirements as follows:
1. High throughput - at least not worse then SPI, i.e. over 8 Mbps.
2. Low latency.
3. Multiplexing ability.
4. Less wires. 

Any consideration are very appreciated.

"vk" <v.kazmirenko@gmail.com> writes:

> Hello!
>
> I'd like to ask advice of DSP gurus in my doubt. In our design we have
> TI's C6414 DSP on main board. This board has to communicate with multiple
> sub-boards of the instrument. We want to have bidirectional communication.
> At present we use DSP's McBSP in SPI mode and ARM9 MCU on sub-boards.
> Little circuitry allows addressing, i.e. its possible to select single
> slave for SPI master-to-slave communication. Additional circuitry used to
> convey communication request from slaves to DSP. Delivery guaranteed by
> software algorithm.
>
> The solution is far from elegant and rather slow. We don't mind to place
> more interface chips on boards. I ask community to suggest possible
> solutions. I can roughly outline requirements as follows:
> 1. High throughput - at least not worse then SPI, i.e. over 8 Mbps.
> 2. Low latency.
> 3. Multiplexing ability.
> 4. Less wires. 

At least some of the 64x family has an integrted PCI bus interface. That
would be my suggestion.
-- 
%  Randy Yates                  % "...the answer lies within your soul
%% Fuquay-Varina, NC            %       'cause no one knows which side
%%% 919-577-9882                %                   the coin will fall."
%%%% <yates@ieee.org>           %  'Big Wheels', *Out of the Blue*, ELO
http://www.digitalsignallabs.com

vk wrote:

> I'd like to ask advice of DSP gurus in my doubt. In our design we have
> TI's C6414 DSP on main board. This board has to communicate with multiple
> sub-boards of the instrument. We want to have bidirectional communication.
> At present we use DSP's McBSP in SPI mode and ARM9 MCU on sub-boards.
> Little circuitry allows addressing, i.e. its possible to select single
> slave for SPI master-to-slave communication. Additional circuitry used to
> convey communication request from slaves to DSP. Delivery guaranteed by
> software algorithm.
> 
> The solution is far from elegant and rather slow. We don't mind to place
> more interface chips on boards. I ask community to suggest possible
> solutions. I can roughly outline requirements as follows:
> 1. High throughput - at least not worse then SPI, i.e. over 8 Mbps.
> 2. Low latency.
> 3. Multiplexing ability.
> 4. Less wires. 

Consider the TDM interface similar to T1 or E1. All devices are on the 
same synchronous bus. Each of the slaves occupies the dedicated time 
slot. The operation is fairly straightforward and the hardware is very 
simple too.

Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com