Bug 6285 - Writing to Novatel v620 causes kernel hang
Summary: Writing to Novatel v620 causes kernel hang
Status: REJECTED INSUFFICIENT_DATA
Alias: None
Product: Drivers
Classification: Unclassified
Component: USB (show other bugs)
Hardware: i386 Linux
: P2 normal
Assignee: Greg Kroah-Hartman
URL:
Keywords:
Depends on:
Blocks: USB
  Show dependency tree
 
Reported: 2006-03-26 09:18 UTC by Chris Worley
Modified: 2006-07-10 13:31 UTC (History)
1 user (show)

See Also:
Kernel Version: 2.6.16
Subsystem:
Regression: ---
Bisected commit-id:


Attachments
lspci -vv (6.70 KB, text/plain)
2006-03-26 09:20 UTC, Chris Worley
Details
Bootup serial console capture (8.65 KB, text/plain)
2006-03-26 09:32 UTC, Chris Worley
Details
/proc/config (29.27 KB, text/plain)
2006-03-26 09:36 UTC, Chris Worley
Details

Description Chris Worley 2006-03-26 09:18:29 UTC
Most recent kernel where this bug did not occur: Unknown
Distribution: Linux From Scratch 01/23/06
Hardware Environment: Geode GX1 SC1100, TI 1410, DP83815
Software Environment: 2.6.16 with minimal toolset and pccardctl
Problem Description: Any attempt to write to usb-serial port causes
kernel to hang until card is removed.

Steps to reproduce: boot kernel with card inserted, login as root,
execute /bin/cat AT > /dev/ttyUSB0
Comment 1 Chris Worley 2006-03-26 09:20:14 UTC
Created attachment 7678 [details]
lspci -vv
Comment 2 Chris Worley 2006-03-26 09:32:34 UTC
Created attachment 7679 [details]
Bootup serial console capture
Comment 3 Chris Worley 2006-03-26 09:36:27 UTC
Created attachment 7680 [details]
/proc/config

ACPI is turned on, but appears to not exist on the embedded board.
All PHY are turned on, since I'm not sure which driver works.
Comment 4 Andrew Morton 2006-03-26 14:44:08 UTC
bugme-daemon@bugzilla.kernel.org wrote:
>
> http://bugzilla.kernel.org/show_bug.cgi?id=6285
> 
>            Summary: Writing to Novatel v620 causes kernel hang
>     Kernel Version: 2.6.16
>             Status: NEW
>           Severity: normal
>              Owner: linux-pcmcia@lists.infradead.org
>          Submitter: chris.worley@sbcglobal.net
> 
> 
> Most recent kernel where this bug did not occur: Unknown
> Distribution: Linux From Scratch 01/23/06
> Hardware Environment: Geode GX1 SC1100, TI 1410, DP83815
> Software Environment: 2.6.16 with minimal toolset and pccardctl
> Problem Description: Any attempt to write to usb-serial port causes
> kernel to hang until card is removed.
> 
> Steps to reproduce: boot kernel with card inserted, login as root,
> execute /bin/cat AT > /dev/ttyUSB0
> 

Guys, is this likely to be a USB problem or a PCMCIA one?

Thanks.

Comment 5 Greg Kroah-Hartman 2006-03-26 15:20:01 UTC
Can you do:
  echo 1 > /sys/module/usbserial/parameters/debug
  echo 1 > /sys/module/airprime/paramaters/debug

and then do the "echo AT > /dev/ttyUSB0" thing and attach the kernel log
from that?

And as for it being a USB or PCMCIA issue, this is a very wierd device with an
embedded usb controller and a fake usb-serial device all rigged together to try
to get it to work properly together.  I've had numerous reports of this device
working just fine, and numerous ones of it not working at all.

The joys of working with undocumented hardware...
Comment 6 Chris Worley 2006-04-05 07:50:09 UTC
Traced down to a completely messed up airprime driver file.
Although it looks correct, none of the usbserial functions are being called.
After I did the modprobe vendor product hack, everything worked.  So
something is seriously wrong in the airprime.c file.
Comment 7 Greg Kroah-Hartman 2006-04-05 11:52:09 UTC
Ok, perhaps the airprime driver just is not the proper driver for this device?

Why did you think that this driver would work for this device?

What is the output of /proc/bus/usb/devices with the device plugged in?
Comment 8 Adrian Bunk 2006-07-10 13:31:46 UTC
Please reopen this bug if:
- it is still present in kernel 2.6.17 and
- you can provide the requested information.

Note You need to log in before you can comment on or make changes to this bug.