[partial-reconfig] Boundary Scan vs. ICAP - which is faster?

[mlee@xxxxxxxxxxxxx]

Dear Friends,

It's me again, sorry to bother you.

Theoritically, internal configuration (thruough ICAP port) should be
faster than external configuration through JTAG port. Am I right?

I have a Partial Reconfiguration Design with Microblaze as the fixed
module and one reconfigurable (dynamic) module. I would like to partially
reconfigure this dynamic module using ICAP, but the BRAMs enclosed by the
microblaze module is too small to fit this entire partial bitstream.
Therefore, I am thinking to store the bitstream in PC, and then transfer
this partial bitstream through serial port segment by segment to
Microblaze. In another words, it's incremental reconfiguration of that
dynamic module.

So my question is, which one do you think is faster? To reconfigure the
module externally using JTAG, or to do this incremental ICAP
reconfiguration? And if BRAM were big enough to fit the entire partial
bitstream (so no need to transfer bitstream through serial port), do you
think then ICAP is faster than external reconfig.?

Thank you very much for your opinion and advice.


Peter Lee

___________________________
partial-reconfig mailing list
partial-reconfig@itee.uq.edu.au
Mailing List Archive : http://www.itee.uq.edu.au/~listarch/partial-reconfig/