Martin Husemann <martin@duskware.de> writes:

> Maybe I don't get the point, but IMHO: when the card needs to be mapped at
> a special address in order to work, this should be stated in the CIS entry.
> If this is not in the CIS, the CIS is broken and the card should get a
> quirk entry.

It's not so simple: the problem seems to be related more to the PCMCIA
*controller* than to the cards themselves. There are too many
``broken'' cards to consider the problem comes from them only.

Furthermore, I feel reluctant to add CIS quirks (although that's what
I did) just to make the card work on the i386 while it works perfectly
in my Amiga. CIS quirks should not be used to circumvent problems
pertaining to a specific controller.

> Any default will fail for someone.

This may be true. That's why we should investigate the idea of
detecting the PCMCIA address bus size: 0x300 fits within 12 bits
whereas 0x400 does not.

But again, I don't have enough evidence and experience with the i386
to find TRT to do here alone.

  Aymeric