Horizontal mouse wheel (a4tech mouse) with X ?

[Matthias Hopf]

On Oct 26, 07 17:21:49 +0200, Pavel Troller wrote:
>   Thanks for the tip with evtest, I didn't know about this program.
> My "regular" vertical wheel causes the following reports when rotated:
> (two steps both ways)
> Event: time 1193410779.718397, type 2 (Relative), code 8 (Wheel), value -1
> Event: time 1193410779.718399, -------------- Report Sync ------------
> Event: time 1193410779.790386, type 2 (Relative), code 8 (Wheel), value -1
> Event: time 1193410779.790389, -------------- Report Sync ------------
> Event: time 1193410779.950366, type 2 (Relative), code 8 (Wheel), value 1
> Event: time 1193410779.950369, -------------- Report Sync ------------
> Event: time 1193410779.998354, type 2 (Relative), code 8 (Wheel), value 1
> Event: time 1193410779.998356, -------------- Report Sync ------------
> 
> 
> The horizontal wheel reports this:
> Event: time 1193410781.070196, type 2 (Relative), code 8 (Wheel), value 1
> Event: time 1193410781.070198, type 2 (Relative), code 9 (Misc), value 8
> Event: time 1193410781.070200, -------------- Report Sync ------------
> Event: time 1193410781.086188, type 2 (Relative), code 8 (Wheel), value 1
> Event: time 1193410781.086189, type 2 (Relative), code 9 (Misc), value 8
> Event: time 1193410781.086189, -------------- Report Sync ------------
> Event: time 1193410781.310156, type 2 (Relative), code 8 (Wheel), value -1
> Event: time 1193410781.310157, type 2 (Relative), code 9 (Misc), value 8
> Event: time 1193410781.310159, -------------- Report Sync ------------
> Event: time 1193410781.350150, type 2 (Relative), code 8 (Wheel), value -1
> Event: time 1193410781.350151, type 2 (Relative), code 9 (Misc), value 8
> Event: time 1193410781.350152, -------------- Report Sync ------------
> 
> 
> So the reports are different, yes... But it seems that the horizontal wheel
> reports the standard wheel event together with a misc event with value 8.

Eek. That sounds... wrong. Maybe a kernel developer would know better
here, but this doesn't look like events the kernel driver would be able
to push correctly to the X11 driver. I don't have a two-wheel mouse
myself, so I don't know ATM what the events should look like. But I
think there was a (HWeel) type. And the additional Wheel codes are
certainly wrong.

> Is that correct ? Is the X11 mouse driver able to distinguish between them ?
> I've tried both "mouse" (with /dev/input/mice) and "evdev" (with
> /dev/input/event4) drivers and both seem to behave identically.

This typically is a strong indication that the kernel driver is not
understanding the protocol correctly.

HTH

Matthias

-- 
Matthias Hopf <mhopf at suse.de>      __        __   __
Maxfeldstr. 5 / 90409 Nuernberg   (_   | |  (_   |__          mat at mshopf.de
Phone +49-911-74053-715           __)  |_|  __)  |__  R & D   www.mshopf.de