Re: PB3400 ethernet and Q1->Q3 upgrade

Re: PB3400 ethernet and Q1->Q3 upgrade

[Benjamin Herrenschmidt]

Humm... even weirder: Apple's code seems to set the base address of
ohare2 to the same base address as ohare. So either there's a second chip
cascaded there, or there's not and in this case, the eth interrupt is bit 28.

-- 
           Perso. e-mail: <mailto:bh40@calva.net>
           Work   e-mail: <mailto:benh@mipsys.com>
BenH.      Web   : <http://calvaweb.calvacom.fr/bh40/>


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: PB3400 ethernet and Q1->Q3 upgrade

[Shawn Ostermann]

> From: Benjamin Herrenschmidt <bh40@calva.net>
> 
> I'll hack a test interrupt handler for the DEC chip cascaded on this
> second ohare (at first, I'll simply put all the code in the dec driver,
> and later I'll fix all this up like I did for the wallstreet if
> necessary, but with only the ethernet using this interrupt, this may not
> be useful).

Lots of people are asking to get the modem working too (although it's
only 33.6), so it would be nice if we could get that supported.  I'm
hoping it will be a trivial change to whatever you're able to come up
with for the ethernet.

Thanks again for your help!

Shawn "The User" Ostermann
-------------------------------------------------------------------------
   Dr. Shawn Ostermann  -  Associate Professor  -  Ohio University
      140 Morton Hall, Ohio University, Athens, Ohio  45701-2979
 ostermann@cs.ohiou.edu -- FAX: (740)593-0406 -- Voice: (740)593-1242
    http://ace.cs.ohiou.edu/~osterman   http://jarok.cs.ohiou.edu

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: PB3400 ethernet and Q1->Q3 upgrade

[Benjamin Herrenschmidt]

On Tue, Sep 21, 1999, Shawn Ostermann <ostermann@cs.ohiou.edu> wrote:

>Unfortunately, I get in over my head very quickly when talking about
>hardware details.  Do you have enough information to look into those
>points?

I think I found where this mysterious ethernet interrupt is. I don't have
one of those machines to test with, but you can find interesting things
in darwin's source code in the drvPExpert package (platform expert,
contains the motherboard specific stuffs).

Basically, it appears that Apple checks for the presence of ohare2
(pci106b,7). When available, the ethernet irq is "fixed" (xor'ed with
0x18) and a special ohare2 handler is cascaded on the fifth irq of ohare
(bit 27 or 28, there's apparently a bug in the comment).
The eth itself is on the fourth or fifth interrupt of ohare 2 controller.
(I'm afraid: did they put an entire second ohare just for an interrupt
controller ? Or do they also use the SCC for the modem ? Do you have 2
media bays on this machine ?)

So it may be possible to implement this the same way I implemented the
gatwick interrupts for the wallstreet.

-- 
           Perso. e-mail: <mailto:bh40@calva.net>
           Work   e-mail: <mailto:benh@mipsys.com>
BenH.      Web   : <http://calvaweb.calvacom.fr/bh40/>


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/