On 03/25/2016 01:53 PM, Theo Buehler wrote: > On Fri, Mar 25, 2016 at 01:47:07PM +0100, Matthieu Herrb wrote: >> On Thu, Mar 24, 2016 at 10:34:12PM +0100, Ulf Brosziewski wrote: >>> Are there any reviews, tests, OKs, or NOs pending? And if >>> not, may I ask if someone knows a better way to make progress >>> in this area? I'm at a loss here. As far as I can see, I >>> cannot offer a smaller diff; this one contains the minimum >>> that technically makes sense. And as it just adds the basics, >>> a lot of things remain that could and possibly should be done >>> soon, 6.0 isn't far away here. >> >> I've been running the previous version of your diff without problem. >> However, on my laptop (X2240) I can't figure out if the >> touchpad is supposed to support multitouch or not. >> > > Same here: I've been running the three patches you sent gzipped on my > Thinkpad T450 since you posted them and didn't notice any regression so > far. It just works as I expect it to. > Thanks. Verifying that the new input functions and the modified wsconscomm don't break anything is more or less all that can be done with these versions of my diffs. Synaptics touchpads are running in a mode that doesn't need the MT functions. Only Elantech-v4 models make use of them currently. ubcmtp and hidmt could apply it, but I haven't adapted the drivers, it is a bit less trivial than replacing the wsmouse_input call. If someone wants to do that, or could do the tests for me, please let me know (up to now, the wsmouse_mtframe and wsmouse_id_to_slot functions have only been tested with special versions of the elantech handler).
>> I wanted to do more tests (comparing with Linux for instance) but I've >> lacked time so far. >> >> If you have some small test programs (or recipes based on xinput) >> to suggest I'll try that. > > Yes, it would be very helpful if you could give some good test cases and > instructions what to look for. > I don't have a very systematic way of testing touchpads. Usually I check "gestures" that involve changes of the contact count or - possibly - pointer-control, like two-finger scrolling (with two, one, or alternating fingers moving), click-and-drag actions with two fingers on clickpads (again, with alternating fingers moving), tapping, tap-and-drag, and tap-and-drag with "locked drags". >> Again, your work on this is much appreciated, even if it's true that >> we've not provided feedback... > > Seconded. > > Thanks again.
