Accessing a USB device

[Joseph C. Lininger]

Chuck,
In order to help you out, I need to see the following information. You
can send it to me privately so as not to clutter the list. We can then
post a solution to the list if people are interested. For all these
tests, do them as root after the usb device is plugged in.

1. The output from the command "dmesg"
2. The output of the command "lsusb -v"
3. A complete listing of your /dev directory, including all
subdirectories. This command will give you what I want. "find /dev"

Equal causes can produce very unequal effects.
Joseph C. Lininger
jbahm at pcdesk.net
Verification: 5eab38a77ac40416e075be8f50607ff7

And so it came to pass that on Wed, 22 Jun 2005, Charles Hallenbeck said

> I am trying for the first time to access a USB device and having a
problem
> doing so. The device is my new UPS from American Power Conversion,
which I
> plan to monitor with a program called apcupsd. The manual describes
how to
> configure my system for USB and how to test the USB interface before
running
> the program, and that is where I am stuck.
>
> It appears to me to be an interrupt problem. Here is why I think so.
For the
> past month or two I have noticed a peculiarity that happens within
about a
> minute after booting up the computer. Normally I start logging in
various
> users on difference consoles, and while I am doing that, I get a
spontaneous
> message saying that IRQ 5 is being disabled. The message says
something cute
> like, "Nobody cares!" and then disables IRQ 5. Examining the IRQ
assignments,
> I see that the only device assigned to that IRQ shown in
/proc/interrupts is
> the uhci-hcd driver. That has not been a problem until now, since
until now I
> have not attempted to access a USB device, which requires the uhci-hcd
> driver.
>
> When I plug the cable from my UPS into a USB slot, the device is
correctly
> identified by manufacturer, serial number, etc., but then it says that
it
> fails to register with usbcore. And the test program confirms that the
device
> cannot be accessed.
>
> My kernel is 2.6.11.6, and all the relevant drivers are compiled into
the
> kernel. It looks to me like the uhci-hcd driver has been assigned to
IRQ 5
> but is not responding to that interrupt when it occurs. If someone has
seen
> this kind of thing before, or has any suggestion how to proceed with
> troubleshooting this one, I would sure appreciate hearing about it.
The
> software I plan to run to monitor the UPS is not involved here as yet,
since
> I have to resolve the USB interface to the UPS before there is a hope
in hell
> of that software working. If it would be of value to share any of the
error
> messages or other data, I would be happy to do so.
>
> Any ideas? I have run out.
>
> Chuck
>
>
> --
> The Moon is Full
> But you can still get downloads from http://www.mhcable.com/~chuckh
>
> _______________________________________________
> Speakup mailing list
> Speakup at braille.uwo.ca
> http://speech.braille.uwo.ca/mailman/listinfo/speakup
>