cgd@netbsd.org (Chris G. Demetriou) writes:

> (1) what is the namespace for subsystem vendor IDs?  (I've heard "same
> as vendor IDs," but haven't had C&V quoted at me, so i'm
> skeptical. 8-)

I think the whole point to subsystem IDs is that if a vendor uses some
other chip, the chip will have the chip's vendor with the other id in
the subsystem slot.  This would make sense, and would allow for
instance NCR to use a DEC chip on their network card but still
identify their own product correctly.

> (2) what is the namespace for subsystem IDs?  I.e. is it per subsystem
> vendor, or per prod/vendor/subsystem_vendor?

This I know for certain.  There is a company I'm working with who uses
their PCI vendor ID in two ways:  one, as subsystem IDs for cards they
make that use a chip with a hard-coded vendor ID, and two, as a
primary vendor ID.  Since they have two different cases where the card
initially identifies itself as a base chip and are using the same
subsystem ID, I think subsystem IDs are in the same space as vendor
IDs.

> There's also the issue that, in the largest share of devices i've
> seen, if they're present at all they're completely ignorable.

Yes, but it might be "nice" to print out that de2 is a specific card
type, rather than a generic card type.

> There're only two cases i've heard of where paying attention to them
> are important, and of those, only one has a driver in the source tree.

Is the one that isn't in the tree the driver I wrote?  If so, rumor
has it that it'll be in the tree within 6 months or so...

--Michael