[Open-graphics] Nascent XP10 bridge

[Kenneth Ostby]

Timothy Normand Miller:
>This is more of a stating point than a proposed design and request for
>comment.  Some more thought needs to be put into how other blocks
>communicate with this (fifos, etc.), and we need to work out the
>issues associated with the two-cycle latency of any signals crossing
>the bridge (busy signals, etc.).
>

Hum, I played somewhat around with the idea of a bus protocol this
weekend, but I didn't get around to formalizing it into a mail. However,
I had some other idea that might be interesting regarding multi-byte
transmissions. Something that stroke me as interesting would be to skip
transferring the count in a own byte, and instead assert a
continuous signal, leaving the guest side to increase the local memory
address by the word size for each data word it receives/sends. It
however, allows us skip sending a single word describing the amount of
data to read / write. 

Another idea I had was to use some of the control signal to specify a
partition of the Spartan3 chip we want to communicate with. The
practical implications of this is to extend the address space by X bits,
but it might be useful if we want to separate the address ranges on the
chips. I.e., having different 2^32  address domains. It also makes sense
in the case where we want to extend with daughter boards, allowing
addressing in a separate domain.

Hope it somewhat made sense, 
Kenneth


-- 
Life on the earth might be expensive, but it 
includes an annual free trip around the earth.

Kenneth Østby
http://langly.org