On Sat, 23 Oct 2004, Jason Lunz wrote: > I'm having a very similar problem with a Dell GX270 desktop. According > to lspci, it appears to have 5 usb controllers, 4 uhci and 1 echi:
> The machine has no less than _8_ USB ports; 2 in front and 6 in back. > Each pair is connected to one of the UHCI controllers. As far as I can > tell, there are no physical ports wired up to the EHCI controller that > lspci claims is present. That's a normal arrangement. Each UHCI controller manages 2 ports and the EHCI controller is connected to all of them. > I attempted to eliminate this phantom EHCI > controller as a source of the problem by not loading the ehci module, > but it still happens either way. > > The random disconnect/lockup happens reproducibly on either 2.6.8.1 or > 2.6.9 (both vanilla). I'm only using a USB keyboard with a built in hub. > lsusb shows it like this: > > $ lsusb > Bus 004 Device 003: ID 05f3:0007 PI Engineering, Inc. Kinesis Advantage PRO MPC/USB > Keyboard > Bus 004 Device 002: ID 05f3:0081 PI Engineering, Inc. Kinesis Integrated Hub > On 2.6.8.1, the _only_ message in dmesg is something like: > > kernel: usb 4-1.2: USB disconnect, address 3 Pleas try turning on the USB debugging option in the kernel configuration. This should provide a lot more information. > Once this appears, USB is completely locked up. It may take minutes or > days to happen. I can discern no pattern. Everything else appears to be > fine. I can connect a PS/2 keyboard and use it to clean up and reboot > the system, which is necessary because running rmmod to unload uhci_hcd > hangs in uninterruptible sleep. Unplugging the keyboard and plugging it > back in has no effect. Any process that accesses /proc/bus/usb/004/003 > (or whichever one is appropriate, depending where it's plugged in) will > hang in uninterruptible sleep forever. There seems to be no way to > unwedge uhci at this point short of a reboot. > > In 2.6.9, everything's the same, but there are additional messages in > syslog preceding the disconnect: > > kernel: drivers/usb/input/hid-core.c: input irq status -84 received > last message repeated 319 times > kernel: usb 4-1.2: USB disconnect, address 3 > > I don't know if that means anything. Yes, it means that the HID driver was trying to communicate with the keyboard and not receiving any responses. It's what you would expect to see when a device disconnects. It's possible that this is a problem with the HID driver. The extra debugging information will help narrow things down. What also would help is if, when the lock-up occurs, you could get a stack trace by using Alt-SysRq-T. The khubd process will be of particular interest. Alan Stern ------------------------------------------------------- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-users
