fox@tcnet.ch wrote:

> I hope Itsumi is still reading here :-)

Well, NetBSD is still my fun :-)

> uhci0 at pci0 dev 9 function 2: VIA Technologies VT83C572 USB Controller 
> (rev. 0x02)
> uhci0: can't map i/o space

This is onboard VIA uhci and it can't be used anyway.

> ohci0 at pci0 dev 10 function 0: NEC USB Host Controller (rev. 0x41)
> ohci0: interrupting at irq 9
> ohci0: OHCI version 1.0
> usb0 at ohci0: USB revision 1.0

I guess this part should work.
Which port did you plug devices to?
It looks ochi0 is used for first three ports on the card.

> ohci1 at pci0 dev 10 function 1: NEC USB Host Controller (rev. 0x41)
> ohci1: interrupting at irq 0
> ohci1: OHCI version 1.0
 :
> NEC USB Host Controller (USB serial bus, interface 0x20, revision 0x02) 
> at pci0 dev 10 function 2 not configured

You could add "ehci* at pci? dev ? function ?" to enable
USB 2.0 ports, but anyway the firmware doesn't map interrupts
for multi function PCI devices so these the second ohci (at function 1)
and ehci (at function 2) won't work.

I guess these devices use PCI interrupt lines other than INT A,
but I don't know such interrupt lines are actually connected
into the interrupt controller on Qubes. When I checked Linux sources
last time there was no statement about these interrupt lines, so
I'm afraid we have to try-and-errors to see which interrupt
line can be used for each function of the card.

Could you post debug messages by "options PCI_CONFIG_DUMP" first?
Note it's so large that you might also have to have
"options MSGBUFSIZE=65536" or something.

> i did run /dev/MAKEDEV scsibus

The default MAKEDEV(8) creates scsibus[0-3] so this is not needed.
---
Izumi Tsutsui