Red Hat Bugzilla – Bug 134503
s-c-display doesn't set up synaptics
Last modified: 2014-03-16 22:48:59 EDT
With the import rhpl.mouse added to xconf.py, it does not set up synaptics in xorg.conf. # rpm -q synaptics rhpl system-config-display system-config-mouse synaptics-0.13.5-5 rhpl-0.146-1 system-config-display-1.0.20-1 system-config-mouse-1.2.8-1
Created attachment 104701 [details] /proc/bus/input/devices
You've got two devices provided by synaptics, currently (and historical) we take the first one returned by kudzu.probe in rhpl.mouse. I need to add support to handle this case.
FWIW, at least in my case (Dell Dualpoint (TM)), it's not that we get two devices provided by synaptics, but rather the pointing stick gets one device and the touchpad gets another. My /proc/bus/input/devices looks very similar to the one Bill attached to Comment #1.
I suggest that we mark this as target for FC3, just like bug #116091, since I suspect the most common set up of touchpads is in addition to a regular mouse. Even in notebooks that only have a touchpad, an external mouse is generally supported as well, quite often with a keyboard/mouse PS/2 port.
Thanks for adding to the target bug, slipped my mind. I'm wondering if I can have a udev rule ala the wacom tablets: KERNEL="event*", SYSFS{manufacturer}="WACOM", NAME="input/%k", SYMLINK="input/wacom%e" To create /dev/input/touchpadN (as theoretically we could get ALPS support later). This would also mean not needing to use auto-dev in the synaptics config. That way I don't need to majorly change the rhpl.mouse api for FC3. As everything is in /dev/input/mice anyway as long as the config is sane, multi-device should work (IIRC the config I generate currently works for adding a usb mouse in addition anyway).
Err... /dev/input/mice sure works for moving the pointer around, but tap-to-click, scrolling and all the other niceties of a touchpad will only work if there's a specific section in the xorg.conf that sets this stuff up. I thought you already knew this, but the comment above gave me the impression that maybe you didn't.
Yes I'm aware that you need to use the raw event device for synaptics, I think I was unclear - I was trying to say that for non-synaptics mice configuring for /dev/input/mice is fine. I'm thinking of the configs I'm generationg for wacom where the tablet is an additional input device so multi-device works fine (once I've got that bit merged). Anyway hope to have something in place this week...
Fix commited to HEAD.
*** Bug 135491 has been marked as a duplicate of this bug. ***
rhpl 0.148-1 contains fix
*** Bug 135482 has been marked as a duplicate of this bug. ***
The fix works for me on dual-input synaptics (pointing-stick + touchpad) on two laptops, new IBM ThinkPad T42p and 4-year old HP Omnibook 6000. These were fresh installs of 200410150526. Thanks for getting this support into FC!