> 

re: faxes

people have probably gathered by now that i'm really busy on too many
things right now.  however it is ok to call me up and call me a hoser
if i promised to do something for you and flaked complete.y.

>   Well, I have nearly all the info I need to hack up an if_ie driver.
>    (Got data from Intel today)

cool.

>   Except for one thing... where in Sun OBIO memory is the Intel chip
> mapped? Or does it have special dual port memory somewhere?
> 


well look at OBIO_INTEL_ETHER
thats where the register(s) are accessible.  so you do have to use the
obio_alloc() stuff to do that.

on the sun, the intel chip does DVMA (dma to virtual addresses).
however the intel (and the amd) only address like 16mb of memory, so
the architecture  specifies that you can do dvma to the upper 16mb of
the address space.  So you don't have to worry about this, there is a
dvma_alloc() routine provided.  So you use the registers to control
the intel and point it to the lump of memory and it is effectively
then dual ported.  see leattach() to see how all this works.  it
should be nearly identical in the intel case.

>   Also, a question:
>   
>   #define OBIO_IOX_BUS      0x001A0000
>   #define OBIO_DES          0x001C0000
> 
>   #define OBIO_IO_BUS_SIZE      0x1000000			
> 
>   These two things don't seem compatible (hard to fit a megabyte into 128k..)
>   I take it that IOX_BUS != IO_BUS_SIZE. What are these things then?
> 
>   Also... ideas about VME bus info? I take it that we will have a vme
> bus device akin to obio. Sun part #s are welcome.
> 
>    :!mcr!:            |  "Elegant and extremely rapid for calculation are the 
>    Michael Richardson | techniques of Young tableaux. They also have the merit
>  mcr@ccs.carleton.ca /of being fun to play with." - p.47 Intro to Quarks&Partons
>  mcr@sandelman.ocunix.on.ca / +1 613 729-5409 / +1 613 788-2600 3853 (work)
> 
> 
> 
> 
> 
>   


------------------------------------------------------------------------------