>Number:         32354
>Category:       kern
>Synopsis:       Recent patch breaks usb for some device/controller
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Wed Dec 21 15:35:00 +0000 2005
>Originator:     Mark A. Wicks
>Release:        NetBSD 3.0 (and current)
>Organization:

>Environment:
System: NetBSD gaspra.kettering.edu 3.0 NetBSD 3.0 (GASPRA - Customized) #3: Tue Dec 20 16:28:11 EST 2005 mwicks@gaspra.kettering.edu:/var/tmp/build/obj/sys/arch/i386/compile/GASPRA-3.0 i386
Architecture: i386
Machine: i386
>Description:
	A recent patch to usb_subr.c (revision 1.122.2.1 in netbsd-3
or revision 1.130 in current) breaks certain USB device/controller
combinations.  The patch submitter noted that he expected certain
buggy devices would break, but I have a scanner that works on some
computers and not others with this patch.  I don't think the scanner
itself is the problem.

Here are the symptoms: The "scanimage" command from sane-backends
hangs. The command "scanimage -L" completes the first time, but hangs
the second time. I can provide additional information if somebody
tells me what information would be useful.

>How-To-Repeat:
Here's the hardware combination that doesn't work:

uhci0 at pci0 dev 16 function 0: VIA Technologies VT83C572 USB Controller (rev. 0x81)
usb0 at uhci0: USB revision 1.0

Controller /dev/usb4:
addr 1: high speed, self powered, config 1, EHCI root hub(0x0000), VIA Technologie(0x1106), rev 1.00
 port 4 addr 2: high speed, self powered, config 1, EPSON Scanner(0x011e), EPSON(0x04b8), rev 1.00

But on another computer with

uhci0 at pci0 dev 7 function 2: Intel 82371AB (PIIX4) USB Host Controller (rev. 0x01)

everything works perfectly.

>Fix:
   Reverse the patch to usb_subr.c (1.122.2.1 in netbsd-3 or 1.130 in -current)

>Unformatted:
 combinations