Re: PCI -> PCMCIA Adapter woes - patch

In message: <[email protected]>
            John Utz <[email protected]> writes:
: If the chip has a lot of it's circuitry in parallel with that pin, which
: likely if it's multifunction, because that pin is designed to do more
: things than usual??. 

Actually, just saw this part.  Lemme explain a little.  The bits in
question route a GPIO pin somewhere.  MF pin 0 is the only ping that
can be routed to INTA#, if I'm reading things correctly.  Ditto for
INTB#.  And those are the only way to do so.

However, I just checked the big fat file of datasheets (rather than
just one).  Here's what I found:
	1211	   2 is INTA#	(called MFUNC0)
	1221	   2 is INTA#	(called MFUNC0)
	1251b	   1 is INTB#	(called IRQMUX0) 2 is IRQ2
	1450	   1 is INTB#	(called IRQMUX0) 2 is IRQ2
	4450	   2 is INTA#	(called MFUNC0)
	1031	   not present
	1130	   not present
so it looks like there's some non-uniformity here.  Now I just wish I
could get more real datasheets on some of these things.  Maybe I
should make contact with an internal engineer at TI I've talked to in
the past from time to time.

This argues for a special sysctl/boot hint.  For the 1251b and 1450 we
should do nothing, it looks like.

Warner


To Unsubscribe: send mail to [email protected]
with "unsubscribe freebsd-mobile" in the body of the message