PCI-Arbiter Extension

[neumayk@kontron.de (Karl Neumayer)]

Hello Experts,

sorry that I arise my PCI-Arbiter question once again.

>I asked:
>  ... does anyone know a reasonable way how to extend the PCI-Arbiter,
>   which normally resides in the ChipSets north bridge ...
>
>Norman J. Rasmussen answered:
>   It is impossible unless the original PCI arbiter understands that a
cascaded 
>   arbiter is to be used.  Since arbitration is pipelined there are
conditions in 
>   which two devices can believe that they own the bus.
>
>   If the primary arbiter knows that a second level is supported, then the
first 
>   arbiter must not deassert one GNT# and assert a different GNT# on the same 
>   clock...
>   The first level arbiter must ALWAYS cause a dead clock between the
deassertion
>   of one GNT# and the assertion of the next GNT#.  If the arbiter never
deasserts
>   a GNT# until FRAME# is asserted, this can cause a hand system when a master 
>   removes its REQ# because it changed it mind...
>
>Tom Keaveny answered:
>   Would it be possible to run all of the non-bridge REQ#/GNT# pairs to
>   a standalone arbiter.  This arbiter would then communicate to the
>   bridge's embedded arbiter via one of the original REG#/GNT# pairs, on
>   behalf of the other agents. In this way, the original bridge only 
>   has one GNT# line of interest.
>
>   Dealing with the bridge's desire to access the bus is probably a bit
>   sticky (effectively an internal grant that can change on the fly),
>   and would likely vary according to the bridge...

I agree with them, because that was exactly my conclusion on that problem.
But for that, I still wonder how this problem is solved by the manufacturers
of CompactPCI CPU cards (which usually employ standard ChipSets, for example
the Intel 82430FX). As far as I know, CompactPCI uses 8 Slots which should
all be capable of busmastering.

I only have a short form of the CompactPCI specification, so I do not know if
it is mandatory or optional to have all 8 Slots beeing capable of busmastering.



Any comments appreciated
Thanks,
Karl
Õ