On Mon, Jun 27, 2005 at 08:55:28AM +1000, matthew green wrote:
   > 
   >    I'm suggesting that we should create logical "bus" drivers to attach
   >    these devices to.  They wouldn't actually _do_ anything but act as
   >    a shim to set up bus_dma the right way; but they should be sufficient
   >    for that, no?
   > 
   > 
   > but how?  bus_dma is defined in <machine/bus.h> and afaik there is
   > no way to do what you are suggesting (without changing bus_dma.)
   
   I think Thor has in mind something like this (certainly, I do)
   
   pci0 at mainbus0
   bouncepci0 at pci0
   bouncepci0: using bounce buffers for transfers above 1Gb
   bce0 at bouncepci0
   
   bouncepci is a (MD?) bus, which provides implementations of the normal
   bus_dma interface that include internal bounce buffers.


it has to be an MD bus (without re-defining bus_dma).  that's what
i'm saying is a problem.

it may be best that we provide a set of routines to do bouncing but
that the broken devices themselves use directly rather than pushing
it off to the bus_dma level.


.mrg.