X11 based application does not respond to touch after first touch ( ILI210x Touchscreen)

[Peter Hutterer]

On Thu, Jun 08, 2023 at 12:31:51PM +0530, Vishnu Motghare wrote:
> Hi Peter,
> 
> Thanks for your reply.
> 
> I'm using xf86-input-tslib
> <https://layers.openembedded.org/layerindex/recipe/301409/>  & not
> libinput. can xf86-input-tslib
> <https://layers.openembedded.org/layerindex/recipe/301409/>  handle the
> pressure? Also how to capture events from tslib input driver?

libinput record sits below the drivers so that will work either way
(should be called `kernel record` but oh well). As for tslib - no idea,
sorry, the little that I knew about it I have well and truly forgotten.

Cheers,
  Peter

> 
> Regards,
> Vishnu
> 
> On Thu, 8 Jun 2023 at 08:30, Peter Hutterer <peter.hutterer at who-t.net>
> wrote:
> 
> > On Wed, Jun 07, 2023 at 01:04:02PM +0530, Vishnu Motghare wrote:
> > > Hello All,
> > >
> > > X11-based applications like xterm & xinput_calibrator do not respond to
> > > touch after the first touch.
> > >
> > > We have an ILI210x Touchscreen controller & the kernel version is 5.15,
> > > touch was working fine with the 5.4 kernel version with Xterm &
> > > xinput_calibrator
> > >
> > > This is the  Commit
> > > <
> > https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/drivers/input/touchscreen/ili210x.c?h=v5.10.182&id=3a492e4403ee2e3442397230d68e26e81525af90
> > >
> > >   added
> > > in the kernel to report the pressure value & after this, no X11-based
> > > application works after the first touch
> > >
> > > Do we need to add any setting in xorg.conf to handle the pressure value
> > > reported by  ILI210x Touchscreen?
> >
> > Most likely you're using the xf86-input-libinput driver for touchscreens
> > (unless you explicitly changed it to use the evdev driver). In that case
> > you can check with `libinput debug-events` to see what libinput produces
> > - the xorg libinput driver is a fairly thin wrapper and this should give
> > you a good indication of where things go wrong.
> >
> > libinput record works either way and shows you the events produced by
> > the kernel.
> >
> > also, fwiw libinput does not handle pressure for touch screen events so
> > this patch shouldn't have made a difference in userspace. So there's
> > most likely some other event missing, libinput record should show this.
> > Probably best to file an issue against libinput for that.
> >
> > Cheers,
> >   Peter
> >
> >