[Bug 209499] New: NKRO does not work on vortex race 3 keyboard

[Bug 209499] New: NKRO does not work on vortex race 3 keyboard

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=209499

            Bug ID: 209499
           Summary: NKRO does not work on vortex race 3 keyboard
           Product: Drivers
           Version: 2.5
    Kernel Version: 5.8.10-1~bpo10+1
          Hardware: All
                OS: Linux
              Tree: Mainline
            Status: NEW
          Severity: normal
          Priority: P1
         Component: USB
          Assignee: drivers_usb@kernel-bugs.kernel.org
          Reporter: sgf.dma@gmail.com
        Regression: No

Created attachment 292809
  --> https://bugzilla.kernel.org/attachment.cgi?id=292809&action=edit
lsusb and usbmon debug

Hi.

Linux does not see more, than 6 simultaneous keypresses on vortex race 3
keyboard. Though on windows 10 at least 10 simultaneous keypresses work.

lsusb info and usbmon debug is in 'vortex_usb.txt' file (there i press
'qwertyuifg' sequently (next key is pressed approximately after 1-2 seconds)
and then release in reverse order):

Thanks.

-- 
You are receiving this mail because:
You are watching the assignee of the bug.

[Bug 209499] NKRO does not work on vortex race 3 keyboard

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=209499

R0b0t1 (sid@aeam.us) changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |sid@aeam.us

--- Comment #1 from R0b0t1 (sid@aeam.us) ---
I understand your confusion but this is expected. The standard report format
for a USB keyboard is 8 bytes. The first byte contains the modifier key state,
the second is padding, and the remaining 6 bytes contain key codes.

N-key rollover is more a statement of how the switch matrix has been
constructed than how many keys you can use in your OS. There are other ways of
constructing a switch matrix that are marginally cheaper but lead to the
inability to distinguish keys after a certain point; for a typical 84-120 key
keyboard the optimal constructions tend to cap out at 6 simultaneous key
presses, explaining the choices made in the USB standard.

Looking at the lsusb -v output the device shows up as a boot keyboard (slightly
more restricted) and a normal keyboard, but both report the expected number of
bytes. Your keyboard has more interface information than usual but I am not
sure how it is interpreted. It could be sending more bytes back to Windows and
Windows may accept them. You may wish to investigate the keyboard from Windows
using the USBView tool.

-- 
You are receiving this mail because:
You are watching the assignee of the bug.

[Bug 209499] NKRO does not work on vortex race 3 keyboard

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=209499

--- Comment #4 from Dmitriy (sgf.dma@gmail.com) ---
I've added output from usbtreeview, it seems identical to usbview (but if it's
not, then i'll install usbview).

Also, wireshark pcap from the same test as on linux: 'qwertyuifg' sequently and
then release in reverse order.

-- 
You are receiving this mail because:
You are watching the assignee of the bug.