Bug 99561

Summary: xHCI communication problem between VIA 3483 USB 3.0 host controller and Fujitsu ScanSnap ix500 scanner
Product: Drivers Reporter: bjoernv
Component: USBAssignee: Greg Kroah-Hartman (greg)
Status: NEW ---    
Severity: normal    
Priority: P1    
Hardware: All   
OS: Linux   
Kernel Version: 4.0.5 Subsystem:
Regression: No Bisected commit-id:
Attachments: dmesg output with "xhci" debugging enabled after calling "scanimage -L" twice
The output of "lsusb -v"
Kernel 4.0.5 configuration
Output of "lspci -vvv"
Contents of /proc/modules
Content of /proc/cpuinfo

Description bjoernv 2015-06-07 18:49:52 UTC 
Created attachment 179031 [details]
dmesg output with "xhci" debugging enabled after calling "scanimage -L" twice

The communication between an USB 3.0 controller (VIA 3483 chip) and a USB 3.0 scanner Fujitsu ScanSnap ix500 scanner is very unreliable.

The device could be used only once. After one command/action the scanner device seems to be dead. If the scanner device is powered On and Off, it works again for one action.

The scanner runs fine on Linux with other USB 3.0 controllers and on Windows 8.1 will all tested USB 3.0 controllers. The scanner also runs on USB 2.0 controllers.

There is a known unsolved USB 3.0 problem for SANE driven scanners:
https://wiki.archlinux.org/index.php/Sane#USB_3.0

This is my test setup:
1) Connect Scanner on USB 3.0 port
2) Type a scanimage command. "scanimage -L" finds attached scanners:
   $ scanimage -L
   device `fujitsu:ScanSnap iX500:102569' is a FUJITSU ScanSnap iX500 scanner
3) Type the scanimage command again:
   $ scanimage -L

   No scanners were identified. If you were expecting something different,
   check that the scanner is plugged in, turned on and detected by the
   sane-find-scanner tool (if appropriate). Please read the documentation
   which came with this software (README, FAQ, manpages).

I attach some log files. I switched on xHCI debugging with
echo -n 'module xhci_hcd =p' > /sys/kernel/debug/dynamic_debug/control

This bug may be related: https://bugzilla.kernel.org/show_bug.cgi?id=76991
Comment 1 bjoernv 2015-06-07 18:51:19 UTC 
Created attachment 179041 [details]
The output of "lsusb -v"
Comment 2 bjoernv 2015-06-07 18:58:11 UTC 
Created attachment 179051 [details]
Kernel 4.0.5 configuration
Comment 3 bjoernv 2015-06-07 18:59:04 UTC 
Created attachment 179061 [details]
Output of "lspci -vvv"
Comment 4 bjoernv 2015-06-07 18:59:57 UTC 
Created attachment 179071 [details]
Contents of /proc/modules
Comment 5 bjoernv 2015-06-07 19:00:28 UTC 
Created attachment 179081 [details]
Content of /proc/cpuinfo
Comment 6 bjoernv 2015-06-07 19:02:10 UTC 
I tested with sane-backends stable (1.0.24) and with sane-backends (GIT snapshot from 2015-06-07). The bug shows up with both versions.
Comment 7 Greg Kroah-Hartman 2015-06-08 00:52:29 UTC 
On Sun, Jun 07, 2015 at 06:49:52PM +0000, bugzilla-daemon@bugzilla.kernel.org wrote:
> https://bugzilla.kernel.org/show_bug.cgi?id=99561
> 
>             Bug ID: 99561
>            Summary: xHCI communication problem between VIA 3483 USB 3.0
>                     host controller and Fujitsu ScanSnap ix500 scanner

Send this to the linux-usb@vger.kernel.org list please
Comment 8 bjoernv 2015-06-08 08:03:32 UTC 
Thanks. I posted this to linux-usb@vger.kernel.org.