USB UHCI broken again w/ visor

USB verbose debug is ON, using driver usb-uhci, on an alpha, kernel 2.4.9, new batteries in the thing. This worked with 2.4.7. What happened? It seems like every other kernel version it gets broken again, and I can't sync my visor.

Aug 28 01:08:53 draal kernel: usb-uhci.c: $Revision: 1.259 $ time 12:40:16 Aug 25 2001 
Aug 28 01:08:53 draal kernel: usb-uhci.c: High bandwidth mode enabled 
Aug 28 01:08:53 draal kernel: usb-uhci.c: USB UHCI at I/O 0x8080, IRQ 18 
Aug 28 01:08:53 draal kernel: usb-uhci.c: Detected 2 ports 
Aug 28 01:08:53 draal kernel: usb.c: new USB bus registered, assigned bus number 1 
Aug 28 01:08:53 draal kernel: usb.c: kmalloc IF fffffc000fbef0a0, numif 1 
Aug 28 01:08:53 draal kernel: usb.c: new device strings: Mfr=0, Product=2, SerialNumber=1 
Aug 28 01:08:53 draal kernel: usb.c: USB device number 1 default language ID 0x0 
Aug 28 01:08:53 draal kernel: Product: USB UHCI Root Hub 
Aug 28 01:08:53 draal kernel: SerialNumber: 8080 
Aug 28 01:08:53 draal kernel: hub.c: USB hub found 
Aug 28 01:08:53 draal kernel: hub.c: 2 ports detected 
Aug 28 01:08:53 draal kernel: hub.c: standalone hub 
Aug 28 01:08:53 draal kernel: hub.c: ganged power switching 
Aug 28 01:08:53 draal kernel: hub.c: global over-current protection 
Aug 28 01:08:53 draal kernel: hub.c: power on to power good time: 2ms 
Aug 28 01:08:53 draal kernel: hub.c: hub controller current requirement: 0mA 
Aug 28 01:08:53 draal kernel: hub.c: port removable status: RR 
Aug 28 01:08:53 draal kernel: hub.c: local power source is good 
Aug 28 01:08:53 draal kernel: hub.c: no over-current condition exists 
Aug 28 01:08:53 draal kernel: hub.c: enabling power on all ports 
Aug 28 01:08:53 draal kernel: usb.c: hub driver claimed interface fffffc000fbef0a0 
Aug 28 01:08:53 draal kernel: usb-uhci.c: v1.251:USB Universal Host Controller Interface driver 
Aug 28 01:08:53 draal kernel: hub.c: port 1 connection change 
Aug 28 01:08:53 draal kernel: hub.c: port 1, portstatus 100, change 3, 12 Mb/s 
Aug 28 01:08:53 draal kernel: hub.c: port 2 connection change 
Aug 28 01:08:53 draal kernel: hub.c: port 2, portstatus 100, change 3, 12 Mb/s 
Aug 28 01:08:53 draal kernel: hub.c: port 1 enable change, status 100 
Aug 28 01:08:53 draal kernel: hub.c: port 2 enable change, status 100 
Aug 28 01:09:01 draal kernel: hub.c: port 1 connection change 
Aug 28 01:09:01 draal kernel: hub.c: port 1, portstatus 101, change 1, 12 Mb/s 
Aug 28 01:09:01 draal kernel: hub.c: port 1, portstatus 103, change 0, 12 Mb/s 
Aug 28 01:09:01 draal kernel: hub.c: USB new device connect on bus1/1, assigned device number 2 
Aug 28 01:09:04 draal kernel: usb_control/bulk_msg: timeout 
Aug 28 01:09:04 draal kernel: usb.c: USB device not accepting new address=2 (error=-60) 
Aug 28 01:09:04 draal kernel: hub.c: port 1, portstatus 103, change 0, 12 Mb/s 
Aug 28 01:09:04 draal kernel: hub.c: USB new device connect on bus1/1, assigned device number 3 
Aug 28 01:09:07 draal kernel: usb_control/bulk_msg: timeout 
Aug 28 01:09:07 draal kernel: usb.c: USB device not accepting new address=3 (error=-60) 
Aug 28 01:09:34 draal kernel: hub.c: port 1 connection change 
Aug 28 01:09:34 draal kernel: hub.c: port 1, portstatus 100, change 3, 12 Mb/s 
Aug 28 01:09:34 draal kernel: hub.c: port 1 enable change, status 100 
Aug 28 01:15:22 draal kernel: hub.c: port 1 connection change 
Aug 28 01:15:22 draal kernel: hub.c: port 1, portstatus 101, change 1, 12 Mb/s 
Aug 28 01:15:22 draal kernel: hub.c: port 1, portstatus 103, change 0, 12 Mb/s 
Aug 28 01:15:22 draal kernel: hub.c: USB new device connect on bus1/1, assigned device number 4 
Aug 28 01:15:25 draal kernel: usb_control/bulk_msg: timeout 

Please cc me as I am not on this list.

Cheers,
-- Bob

Bob McElrath ([email protected]) Univ. of Wisconsin at Madison, Department of Physics

It looks like your uhci controller is not getting interrupts anymore. Does the value of /proc/interrupts for the usb-uhci driver get incremented?

Does any other usb devices work on this system?

thanks,

greg k-h

No, it seems stuck at a non-zero value.

(0)<mcelrath@draal:/home/mcelrath/minijam> cat /proc/interrupts

  1:     224376          XT-PIC   keyboard
  2:          0          XT-PIC   cascade
  8:  250012322             RTC  +timer
 11:    3019583          XT-PIC   PAS16
 12:    1891610          XT-PIC   PS/2 Mouse
 14:      60292          XT-PIC  +ide0
 16:     650152       CABRIOLET   BusLogic BT-958
 18:      23085       CABRIOLET   usb-uhci
 19:     304381       CABRIOLET   eth0
 20:          0       CABRIOLET   isa-cascade

ERR: 0

The non-zero value is probably because I was using another device.

Yes, I have a SanDisk ImageMage (SDDR-12) that sorta-works with the usb-storage driver.

The device times out though after transferring a few files. I don't know if the problem is with the usb-storage driver, usb-uhci, or the device itself. But this problem has existed for a long time, and OHCI users don't report this problem. (using usb-storage with this device and my little program to transfer files to it:
http://www.sourceforge.net/projects/mj-tools/) I haven't been able to track this bug down yet...

The behaviour of the usb-storage stuff is unchanged from 2.4.7, but the visor definitely did work with 2.4.7, where it doesn't now.

Thanks,
-- Bob

P.S. please CC me as I am not on linux-kernel.

Bob McElrath ([email protected]) Univ. of Wisconsin at Madison, Department of Physics

But the usb-storage code did change from 2.4.7-2.4.9 while the visor code hasn't :)

So I'd try unloading the usb system (or rebooting if it's compiled into the kernel), and just trying a visor sync without running the usb-storage code to try to narrow down the potential problem.

thanks,

greg k-h

I tried the visor first, and saw this behavior. (without even insmoding the usb-storage driver) I tried it several times, with the same results. Only this morning before sending my message did I try the usb-storage to see if it was broken too.

I've rmmod'ed everything and re-insmoded everything (except usb-storage) with the same results.

Maybe related:
Why would /proc/bus/usb be always empty?

I'll reboot to 2.4.7 again when I get home tonight and try it.

Cheers,
-- Bob

Bob McElrath ([email protected]) Univ. of Wisconsin at Madison, Department of Physics

And usb-storage worked?
Maybe it's a problem in your visor. Does a soft reset of it fix it?

Did you mount usbdevfs there? See http://www.linux-usb.org/FAQ.html#gs3

thanks,

greg k-h

Yes. *sigh* Never again will I buy this palm/visor volatile RAM non-rechargable B&W no memory protection piece of crap. I wonder how well those Agendas work...

Thanks. Sorry for bothering you.

No. I thought that was no longer necessary? Anyway it works now.

Cheers,
-- Bob

Bob McElrath ([email protected]) Univ. of Wisconsin at Madison, Department of Physics