I have a Dell E4300 with a trackpoint + touchpad combo device which shows up as: [ 17.881956] input: DualPoint Stick as /devices/platform/i8042/serio1/input/input9 [ 17.898488] input: AlpsPS/2 ALPS DualPoint TouchPad as /devices/platform/i8042/serio1/input/input10 With 2.6.31-rc3 stock it suffers from http://bugzilla.kernel.org/show_bug.cgi?id=12577 which is fixed by http://lkml.org/lkml/2008/12/8/182 Running with that patch to correctly handle the 9-byte packets, the combo device exhibits an interesting failure mode -- the buttons sometimes get stuck. I finally tracked it down to the following sequence: 1. press btn1 on deviceA (either touchpad or trackpoint button) 2. press btn1 on deviceB 3. release btn1 on deviceB 4. release btn1 on deviceA Here's what I get when I do that with i8042.debug=1: bytes received event -------------- ----- 09 00 00 trackpoint click 08 00 00 trackpoint release cf 06 20 39 4a 00 touchpad click cf 06 20 38 4a 00 touchpad release 09 00 00 trackpoint click cf 06 20 39 4a 00 touchpad click cf 06 20 39 4a 00 (*) touchpad release 08 00 00 trackpoint release cf 06 20 39 4a 00 touchpad click cf 06 20 38 4a 00 touchpad release The (*)ed release event has the button state incorrect -- byte 4 should be 0x38 but is 0x39 instead. Running "xinput query-state 'AlpsPS/2 ALPS DualPoint TouchPad'" I see that the button gets released at the first motion event on the device with the stuck button, which works fine if you only have one pointer -- but on this hardware it's easy to have the trackpad button stuck down while moving with the trackpoint. So, what's the appropriate way to quirk this hardware? - synch the buttons on every button-up event? (Prevents dual-pointer mode from working, but that's not really usable with a combo trackpad+touchpoint just from a physical layout point-of-view.) - query the device (how?) when getting a down-on-already-down button? - something else? I suspect that there's a separate failure mode that I haven't tracked down yet, too, because in the above scenario, moving the device with the stuck button always releases it, while I've also had buttons stick down that persisted through vigorous wiggling and clicking of both pointing devices. -andy -- To unsubscribe from this list: send the line "unsubscribe linux-input" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
