Jukka Salmi --> current-users (2005-06-15 17:51:39 +0200):
> Hello,
> 
> I just noticed that the uhubs on a i386 system are no longer identified:
> 
> $ diff dmesg.old dmesg.now
> 6c6
> < NetBSD 3.99.5 (GENERIC) #0: Mon May 23 16:57:01 CEST 2005
> ---
> > NetBSD 3.99.6 (GENERIC) #6: Wed Jun 15 08:53:22 CEST 2005
> 13c13
> < cpu0: AMD Athlon XP 2000+ (686-class), 1666.82 MHz, id 0x681
> ---
> > cpu0: AMD Athlon XP 2000+ (686-class), 1666.81 MHz, id 0x681
> 50c50
> < uhub0: VIA Technologies UHCI root hub, class 9/0, rev 1.00/1.00, addr 1
> ---
> > uhub0: vendor 0x1106 product 0x0000, class 9/0, rev 1.00/1.00, addr 1
> 56c56
> < uhub1: VIA Technologies UHCI root hub, class 9/0, rev 1.00/1.00, addr 1
> ---
> > uhub1: vendor 0x1106 product 0x0000, class 9/0, rev 1.00/1.00, addr 1
> 62c62
> < uhub2: VIA Technologies UHCI root hub, class 9/0, rev 1.00/1.00, addr 1
> ---
> > uhub2: vendor 0x1106 product 0x0000, class 9/0, rev 1.00/1.00, addr 1
> 70c70
> < uhub3: VIA Technologies EHCI root hub, class 9/0, rev 2.00/1.00, addr 1
> ---
> > uhub3: vendor 0x1106 product 0x0000, class 9/0, rev 2.00/1.00, addr 1

Another appearance of the same problem is with usbdevs(8):

$ usbdevs
addr 1: product 0x0000, vendor 0x1106
addr 1: product 0x0000, vendor 0x1106
addr 1: product 0x0000, vendor 0x1106
addr 1: product 0x0000, vendor 0x1106


> What could be the source of this problem?

It's [1]this commit which causes the problem; building a kernel with
revision 1.125 of src/sys/dev/usb/usb_subr.c fixes the problem:

$ usbdevs
addr 1: UHCI root hub, VIA Technologies
addr 1: UHCI root hub, VIA Technologies
addr 1: UHCI root hub, VIA Technologies
addr 1: EHCI root hub, VIA Technologies


Cheers, Jukka

[1] http://mail-index.netbsd.org/source-changes/2005/05/30/0019.html

-- 
bashian roulette:
$ ((RANDOM%6)) || rm -rf ~