Re: Intermittent loss of ftdi-sio link

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

 



On Sat, 20 Nov 2010, Robert Pearce wrote:

> I don't know if this is the right place to ask, but it seems to be
> where the experts are.
> 
> I have a prototype board with an FT4232 on it. Mostly it's working
> fine, but if I leave the PC talking to it (or a device connected to it)
> for a significant time, I sometimes come back to find the link has
> died. Indeed the tty device my software was using no longer exists, and
> the FT4232 has been "bumped" around it.
> 
> (This is kernel 2.6.35.7 with a patch to recognise the device PID)
> 
> I've figured out that the USB side is glitching and the kernel is
> re-registering the device, leaving out the tty that's already open
> because it's not available. The question is why is it glitching?

...

> Now, I should probably say that the "Oxford Vacuum Science"
> "PicoSphere" (the device with the FT4232) is plugged into a front panel
> USB socket on a card reader,

This isn't clear.  Do you mean that you've got a USB port built into a
card reader's front panel and the PicoSphere is plugged into that port?  
In that case, how is the card reader connected to the computer?

>  which I suspect means there's a hub in
> there. And the big blob of usb-storage stuff makes me wonder whether
> it's actually the card reader that's flaky.

No, there's no hub.  The log shows that the high-speed USB host
controller connected to both the PicoSphere and the storage device (the
card reader?) stopped working.

> My next step will be to try it on a different USB socket, once I can
> clear enough space to re-route the 4-metre USB cable I need for that.
> The only thing is that the fault doesn't happen often or predictably,
> so it'll take some time to satisfy myself that way. I guess I'm hoping
> someone here can look at that syslog and say "yeah, those in-Win card
> readers fail like that" or "no, that's not what you think" or something.

This is not a failure of a peripheral USB device; it's a failure of a
controller in the computer.  An "lspci" listing might be useful,
together with an "lsusb -v" listing.

Alan Stern

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


[Index of Archives]     [Linux Media]     [Linux Input]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [Old Linux USB Devel Archive]

  Powered by Linux