On Aug 1, 2016 1:36 AM, "ac via talk" <talk@gtalug.org> wrote:

On Sun, 31 Jul 2016 14:09:34 -0400 Russell Reiter <rreiter91@gmail.com> wrote:

...

<snip>

...

Soo, am thinking, in your case... motherboard... --> bios? Is xHCI hand-off enabled in bios settings? (we are using intel... - also, btw eHCI hand-off should be disabled) or still usb 3 issue, passthrough? other than that it sounds like a frustrating and entertaining problem (although it is solved for everyone else afaict :) )

...

I'm not sure this is a real solution but this guy plugged in a USB 2 device in the 3.0 port on boot and it initialized the port as 2.0. http://forum.kodi.tv/showthread.php?tid=101903

This could work also, cool easy solution Russel :)

afaict the original problem was solved when the endpoint was recalculated each time (sync xhci), so if not resolved then it should be something else, like bios, usb 3 pass, etc - to reconf as usb 2 should also work?

I think flash drives might be more flexible than the scanner. I usually have to fix xsanes internal ordering with a UDEV rule. Perhaps you only need to send ^U in usb setup packets and subsequent block endpoints will toggle correctly. Can't find my link to that tidbit but there is this; "Re: USB scanner stops working with xhci_hcd URB transfer length is wrong, xHC issue? req. len = 0, act. len = 4294967288" https://groups.google.com/forum/m/#!topic/linux.kernel/W1OQjJG2nEs Russell Sent from mobile.