getting vendor IDs

[Barry Bouwsma]

On Mon, 20 Dec 2004 17:45:40 -0800, Julian Elischer wrote:

> of vendor names vs USB IDs. taken from the USB website in a backdoor manner.

Thanks for that pointer.  Interesting, combined with the research I've
done on one particular vendor ID...  Also gives me something to do,
if nobody else wants to jump on it...


> sort -n |awk -F: '{printf "{ 0x%x, \"%s\" },\n", $1, $2 } '
                               ^^^^ 0x%04x seems to better match
the format of usbdevs...


Now, thanks to Warner, I have a better clue about usbdevs.h under
Free/NetBSD, to try and come up with a good common file for both,
without the two being too different, making later diffs ugly...

I may try to bring usbdevs into shape with the USB site vendor
list, separating the others for later consideration.  One of the
things I noticed in a diff was:
vendor HP               0x03f0  Hewlett Packard
vendor HP2              0xf003  Hewlett Packard
That looks to me more like an endian issue, than a separate vendor
ID, right?  (must dig for references, unless it's a real product
that HP botched up)

There seem to be a good number of USB vendors whose products show
up in other Vendors' products.  Apparently particularly so in digital
cameras, where you have the chip manufacturer, then the product ID
which can refer to a random camera manufacturer's product.  Or even
several different manufacturers' products.  F'rexample,
vendor JENOPTIK         0x0784  Jenoptik
Not on the USB website, but some googling makes it appear that this
is an ID used by SoundVision, whose chips appear in dozens of vendors'
cameras.  Jenoptik is only one of them.  My camera with this ID is
a completely different make.  (I had to use this for a quirk in
FreeBSD; NetBSD worked out-of-the-box with no quirks.)

In the case of the Jenoptik camera in the present usbdevs file, its
product ID apparently is shared by two other camera models from one
manufacturer, according to a site titled `Digital Camera Support for
UN*X', which also lists several Jenoptik cameras with entirely
different vendor IDs.  (But none with the SoundVision chip, go
figure.  Not that the list of cameras is anywhere near comprehensive.)

In such a case, the only thing I can see doing is to provide the
end-product description as part of the vendor product, for NetBSD.
Which will be rather a lot of work, but when you have the case of
some unfamiliar-to-consumer's chip in a well-known camera case,
I don't see an easy way to keep the descriptions short yet clear.

I suspect there are a bunch more of this, like I noted with SUSTEEN,
apparently being used in some SONYERICSSON product -- the former is
the registered vendor on the USB site.


Yeah, it's nitpicking.  It doesn't matter to D/FBSD, but apparently
does to NetBSD (and OpenBSD?  Haven't checked), so I'd rather do this
in a way agreeable to NetBSD...


Also, in the event that D/FBSD implement a pciconf-like command for
USB that would search through a database for detailed vendor/product
descriptions, apparently like NetBSD has in the kernel, to make up
for the noted deficiency with two of my USB devices, then this info
would be relevant to FreeBSD and DFly.


barry bouwsma
sheesh I ramble on and on and on and on