I've been struggling with PCMCIA on the Alchemy boards for a bit now,
and thought maybe I'd inquire here for folks with more expertise.

Basically, I have stuff working well enough to parse CIS, etc.

Where things go south is when using certain devices (e.g.
if_ne_pcmcia.c).  In this particular case it looks like a stuck
interrupt -- basically what I think is happening is that
pcmcia_socket_enable() is called *after* pcmcia_intr_establish() by the
driver.

As soon as I enable interrupts, it looks like an interrupt is pending on
the device.  I've just (in theory) reset the board, so this shouldn't be
happening, I'd think.  What else is interesting is that if_ne_pcmcia
does this *before* marking the device enabled in its structures, so it
refuses to ack the interrupt.  Now its stuck. :-(

The other thought I have is that maybe CIS is requesting a different VCC
or VPP than the VS pins.  I don't see anything in the code that would
seem to indicate support for power related CIS tuples, or in the chip
structure to allow them to be changed.  Am I missing something?

Any advice here would be appreciated.  TIA.

-- 
Garrett D'Amore, Principal Software Engineer
Tadpole Computer / Computing Technologies Division,
General Dynamics C4 Systems
http://www.tadpolecomputer.com/
Phone: 951 325-2134  Fax: 951 325-2191