On Fri, 5 Nov 1999, Bill Sommerfeld wrote:

> I took a look at the code, and found one suspicious smoking gun.
> 
> This probably isn't the problem you're trying to track down, but it
> looks like _isa_bus_dmamap_sync() in arch/arm32/isa/isa_machdep.c,
> (which does bounce-buffering for ISA DMA) should be, but isn't,
> calling _bus_dmamap_sync() (in arch/arm32/arm32/bus.c) at an
> appropriate time to do cache flushes and the like.
> 
> See bus_dmamap_sync(9) for information on this kernel internal
> interface..
Hi,
  Aren't we talking about the fxp driver on PCI ? If so then we are not
going near the ISA stuff and if this is run on an EBSA285 there is no ISA
bus.

I'll look at the isa dma sync issue but I don't think this is related to
the etherexpress problem.

Cheers,
				Mark