On Sun, Sep 29, 2013 at 07:40:18AM +0200, frantisek holop wrote: > hmm, on Sat, Sep 28, 2013 at 12:25:27PM +0100, Stuart Henderson said that > > On 2013/09/28 11:34, frantisek holop wrote: > > > hi there, > > > > > > after inserting this D-LINK DWA-125 rev A3 usb wifi, i get: > > > > > > ugen1 at uhub0 port 6 "Ralink 11n Adapter" rev 2.00/1.01 addr 5 > > > > > > which is kind of true, as it is supposed to be > > > powered by Ralink RT5370. but according to > > > /sys/dev/usb/usbdevs, 0x2001 is DLINK and 0x3c19 is not even in there. > > > > Normally, strings for USB devices come from the device itself, iirc the > > kernel > > strings are used when the device has no string. > > > > > to add more to my confusion usbdevs -v shows: > > > > > > $ usbdevs -v | grep Ralink > > > port 2 addr 2: high speed, power 450 mA, config 1, 802.11 n > > > WLAN(0x01a2), Ralink(0x0411), rev 1.01, iSerialNumber 1.0 > > > port 6 addr 5: high speed, power 450 mA, config 1, 11n Adapter(0x3c19), > > > Ralink(0x2001), rev 1.01, iSerialNumber 1.0 > > > > > > what is that other device? > > > > Any more clues from lsusb -v ? > > well, the other device, that is of course my working, > other usb wifi dongle.. what threw me completely off was > that 0x0411 is Melco, the device itself is a Buffalo, > and it is shown as Ralink... > > lsusb cleared that one up: > > Bus 000 Device 002: ID 0411:01a2 BUFFALO INC. (formerly MelCo., Inc.) > WLI-UC-GNM Wireless LAN Adapter > Device Descriptor: > idVendor 0x0411 BUFFALO INC. (formerly MelCo., Inc.) > idProduct 0x01a2 WLI-UC-GNM Wireless LAN Adapter > iManufacturer 1 Ralink > iProduct 2 802.11 n WLAN > iSerial 3 1.0 > > > the D-Link device shows up as: > > Bus 000 Device 005: ID 2001:3c19 D-Link Corp. > Device Descriptor: > idVendor 0x2001 D-Link Corp. > idProduct 0x3c19 > iManufacturer 1 Ralink > iProduct 2 11n Adapter > iSerial 3 1.0 > > > so iManufacturer and iProduct override usbdevs device ids. i personally > find this kind of confusing, and i think it dilutes the value of dmesgs. > usbdevs(8) output is outright lying to me. i think it should make a > disctinction between usbdevs id and iManufacturer. in my case, i have 2 > totally different devices but according to dmesg they are basically the > same, "Ralink 11n Adapter" and "Ralink 802.11 n WLAN". > > for what it's worth here is a small patch for usbdevs:
You obviously didn't try compiling this, it will break all of the references to MELCO devices in the kernel. While the usb forum's usb.if lists the vid as buffalo now I don't see any compelling reason to change it, buffalo is a subsidiary of melco. > > > Index: usbdevs > =================================================================== > RCS file: /cvs/src/sys/dev/usb/usbdevs,v > retrieving revision 1.605 > diff -u -p -r1.605 usbdevs > --- usbdevs 5 Sep 2013 19:42:09 -0000 1.605 > +++ usbdevs 29 Sep 2013 05:34:07 -0000 > @@ -68,7 +68,7 @@ vendor QUANTA2 0x0408 Quanta > vendor NEC 0x0409 NEC > vendor KODAK 0x040a Eastman Kodak > vendor VIA 0x040d VIA > -vendor MELCO 0x0411 Melco > +vendor BUFFALO 0x0411 Buffalo > vendor LEADTEK 0x0413 Leadtek > vendor CREATIVE 0x041e Creative Labs > vendor NOKIA2 0x0421 Nokia > @@ -1457,6 +1457,7 @@ product DLINK RT2570 0x3c00 RT2570 > product DLINK DUBE100B1 0x3c05 DUB-E100 rev B1 > product DLINK RT2870 0x3c09 RT2870 > product DLINK RT3072 0x3c0a RT3072 > +product DLINK DWA125A3 0x3c19 DWA-125 rev A3 > product DLINK DSB650C 0x4000 10Mbps Ethernet > product DLINK DSB650TX1 0x4001 10/100 Ethernet > product DLINK DSB650TX 0x4002 10/100 Ethernet > > > -f > -- > monday: in christian countries, the day after the footbal.
