On Sun, 10 Feb 2008, Peter Rasmussen wrote:

> Hi Alan,
> 
> Yes, I realize that a flaky cable definitely is a possibility.
> 
> However, as I always saw the first part in syslog (this particular 
> example is from now, when it started working) like the following:

> <snip>
> 
> It looked like the HW was consistent in its behaviour, but without 
> knowing what it means, I can see it start being different from this and 
> on (the following taken from my first posting when it didn't yet work):
> 
> usb-storage: Status code -32; transferred 0/36

Yes, that particular error code would not be caused by a bad cable.  
Most likely it's some sort of firmware bug in the device.

> My thoughts then becomes:
> 
> Is it possible to distinguish such a failure from a bad implementation 
> of a device's USB stack?

Generally cable (and other low-level hardware) failures show up with 
error codes -71 or -84.  Different error codes indicate implementation 
problems.

> If it is, it would be great to also be told in some way that doesn't 
> require me to know USB so intimately.
> 
> E.g.: Connection Error: You seem to have a bad cable

Unfortunately there's no way to distinguish cable problems from other 
sorts of hardware problems.

>          Registration Error: There are problems with registering the 
> device. It may have a bad XYZ command processing

As Matt mentioned, there are so many ways in which devices mess up
command processing, it's impossible to categorize them in general.  
Some of the most common failings are represented by a list of quirks in 
the usb-storage source file unusual_devs.h.

> I mean, I do have trouble decoding these lines:
> 
> usb-storage: Command INQUIRY (6 bytes)
> usb-storage:  12 00 00 00 24 00
> usb-storage: Bulk Command S 0x43425355 T 0x1 L 36 F 128 Trg 0 LUN 0 CL 6
> 
> But then, what do I know :-)

Well, sure...  That log output was never meant for casual users like 
yourself.  It's debugging information, meant for use by developers.  
That's one reason why it doesn't appear unless you enable 
CONFIG_USB_STORAGE_DEBUG.

Alan Stern

-
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to