[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: USB mass storage - JetFlash



leopaul wrote:
> Hi Orjan,
> 
> The problem is solved. Thank you for the help.
> 
> It was my mistake. I was not physically power cycling, rather using the
> on-board switches
> to boot and test through "ktest". Since, the power was constantly available
> to the flashdrive
> and when the unit warm boots, the flashdrive and the units were at different
> levels of operation.
> Therefore was the kernel crash. Please, correct me if I am wrong.

It sounds plausible that the device numbering problem is related to what 
happens electrically on the bus.  For example, it could be that we are 
registering an electrical change on the USB connector like it was just 
connected, but the device is still configured (I don't really know, I'm 
just guessing).  In any case, the kernel shouldn't crash.

> There may situations in the field, which might cause a warm boot. Hence, is
> it possible to
> force reset or initialize the flashdrive (or a USB device) ?

We do reset the host controller at boot time, which disables both ports. 
  Maybe we should wait a while before configuring the HC again, or 
perform some other operation (such as issuing a 'deconfig' command to 
the HC) in etrax_usb_hc_init().  If you're interested, you can read 
chapter 8.4.1 in the Designer's Reference at 
http://developer.axis.com/doc/hardware/etrax100lx/des_ref.html and try 
it out.

I think you'd need to read the USB spec, perhaps connect a USB analyzer, 
and have an oscilloscope handy to understand what happens and come up 
with the correct solution, which would require a little more time than I 
can spare at the moment.  I trust you will be able to continue your work 
in the meantime.

Thanks for bringing this problem to our attention.

-- 
Orjan Friberg
Axis Communications