Hi,
> > driver code I need for the SMC91C111 driver aka EtherNAT driver for the
> > Falcon. Here's the patch - I could only test it so far on ARAnyM, meaning
> > the driver correctly fails to load because the hardware cannot be found.
> > Christian may be able to build the driver and test on his Falcon before I
> > get mine set up.
>
> Ah, a clone of drivers/net/smc91x.c. Is 91x going the way of the LANCE? ;-)
Well, I may split the Atari bits like fake interrupt handler off into a
separate file, but the EtherNAT really does have an interrupt so we can
safely assume the polling hack is going to go away eventually. Consider
that clone a temporary measure only.
Anyway, making a polled interrupt handler transparent to the
arch-independent driver might be possible with some changes to the arch
interrupt code only (i.e. certain interrupt numbers like Timer D get
special handling in that I don't just register the interrupt but set up
the timer at the same time). Doesn't get much cleaner that way.
> > driver is built or not. May I incude autoconf.h in the arch code? The
>
> Thy shall not include autoconf.h ;-)
Seemed to recall something to that effect :-)
> You mean <linux/config.h>? That one is auto-included these days.
> Yes, you can check for CONFIG_* in arch code. But it gets messy with modules
> (need to check for CONFIG_*_MODULE, too).
That may have been the problem. I'll add the module test as well then.
> Alternatively, you can put the platform device in the driver. That's
> what e.g. drivers/block/floppy.c does.
As device_initcall? May be cleaner indeed. I was glad I found the initcall
trick at all... Does that work on module load as well?
> > other one is a kconfig issue - I need to select MMIO either as module or
> > builtin, depending on how the EtherNAT driver is built. That does not
> > appear to be automatic yet.
>
> You mean MII? Doesn't `select MII' do the right thing?
Did fail (MII compiled in when driver was modular) at least once for me.
> > Signed-Off-By: <[EMAIL PROTECTED]>
> ^ ^ ^
> Signed-off-by: Michael Schmitz <[EMAIL PROTECTED]>??
> (That's what I changed in your other patch).
Sorry, that's what I meant to write. I should probably phase out that
address anyway and start using a linux-m68k.org one at some time.
OK, I'll move the platform init to the driver if possible, and I was
trying to track the smc91Cx driver closely anyway for a later merge. I'll
have to see about enabling the interrupts first though.
Cheers,
Michael
-
To unsubscribe from this list: send the line "unsubscribe linux-m68k" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
