Re: F22 System Wide Change: Change xorg input stack to use libinput

[Kevin Kofler]

Jaroslav Reznik posted (and
>  Change owner(s): Hans de Goede <hdegoede(a)redhat.com&gt;
wrote):

>  KDE: limits itself to standard X11 mouse config interfaces, no
> changes
>  needed. 
Not true. We ship kcm_touchpad on the KDE spin, which definitely does depend
on synaptics interfaces (search for "synaptics" in:
https://projects.kde.org/projects/playground/utils/kcm-touchpad/repositor...
).
Anything that does not use the synaptics driver is not considered a
touchpad. And kcm_touchpad is also in the process of becoming a core part of
upstream Plasma. (It is currently in the git.kde.org playground.)

Therefore, porting kcm_touchpad (the rewritten 1.x series we currently ship,
not the old, obsolete, 0.3.x version, please make sure you look at the
correct version) is an essential requirement before we can move away from
the synaptics driver.


An additional objection I have to this change proposal is that libinput
(deliberately) only implements a small subset of the configurability of the
old drivers, and thus, if we are going to remove the old drivers entirely,
we are taking away flexibility from our users.

        Kevin Kofler

F22 System Wide Change: Change xorg input stack to use libinput

[Richard Hughes]

On 13 December 2014 at 01:10, Kevin Kofler <kevin.kofler(a)chello.at&gt; wrote:
>  An additional objection I have to this change proposal is that
> libinput
>  (deliberately) only implements a small subset of the configurability of the
>  old drivers, and thus, if we are going to remove the old drivers entirely,
>  we are taking away flexibility from our users. 
Yes, but the old stuff is unmaintainable.

Richard

F22 System Wide Change: Change xorg input stack to use libinput

[drago01]

On Sat, Dec 13, 2014 at 2:10 AM, Kevin Kofler <kevin.kofler(a)chello.at&gt; wrote:
>  Jaroslav Reznik posted (and
> > Change owner(s): Hans de Goede <hdegoede(a)redhat.com&gt;
>  wrote):
>
> > KDE: limits itself to standard X11 mouse config interfaces, no changes
> > needed.
>
>  Not true. We ship kcm_touchpad on the KDE spin, which definitely does depend
>  on synaptics interfaces (search for "synaptics" in:
>
> https://projects.kde.org/projects/playground/utils/kcm-touchpad/repositor...
> ).
>  Anything that does not use the synaptics driver is not considered a
>  touchpad. And kcm_touchpad is also in the process of becoming a core part of
>  upstream Plasma. (It is currently in the git.kde.org playground.)
>
>  Therefore, porting kcm_touchpad (the rewritten 1.x series we currently ship,
>  not the old, obsolete, 0.3.x version, please make sure you look at the
>  correct version) is an essential requirement before we can move away from
>  the synaptics driver.
>
>
>  An additional objection I have to this change proposal is that libinput
>  (deliberately) only implements a small subset of the configurability of the
>  old drivers, and thus, if we are going to remove the old drivers entirely,
>  we are taking away flexibility from our users. 
Longer term we should just use libinput for input and modesetting +
glamor for video (ddx) .. that way its one code base that supports
everything and not a driver for each piece of hardware.

F22 System Wide Change: Change xorg input stack to use libinput

[Tom Hughes]

On 13/12/14 01:10, Kevin Kofler wrote:

>  An additional objection I have to this change proposal is that
> libinput
>  (deliberately) only implements a small subset of the configurability of the
>  old drivers, and thus, if we are going to remove the old drivers entirely,
>  we are taking away flexibility from our users. 
Indeed. Is it, for example, still the case that the libinput developers 
are refusing to consider things like definable button areas on clickpads 
so you can create a proper middle button?

Tom

-- 
Tom Hughes (tom(a)compton.nu)
http://compton.nu/

F22 System Wide Change: Change xorg input stack to use libinput

[Peter Hutterer]

On Sat, Dec 13, 2014 at 10:15:41AM +0000, Tom Hughes wrote:
>  On 13/12/14 01:10, Kevin Kofler wrote:
>  
>  >An additional objection I have to this change proposal is that libinput
>  >(deliberately) only implements a small subset of the configurability of the
>  >old drivers, and thus, if we are going to remove the old drivers entirely,
>  >we are taking away flexibility from our users.
>  
>  Indeed. Is it, for example, still the case that the libinput developers are
>  refusing to consider things like definable button areas on clickpads so you
>  can create a proper middle button? 
Have you filed a bug for it?

Cheers,
   Peter

F22 System Wide Change: Change xorg input stack to use libinput

[Tom Hughes]

On 15/12/14 21:39, Peter Hutterer wrote:
>  On Sat, Dec 13, 2014 at 10:15:41AM +0000, Tom Hughes wrote:
> > On 13/12/14 01:10, Kevin Kofler wrote:
> >
> >> An additional objection I have to this change proposal is that libinput
> >> (deliberately) only implements a small subset of the configurability of the
> >> old drivers, and thus, if we are going to remove the old drivers entirely,
> >> we are taking away flexibility from our users.
> >
> > Indeed. Is it, for example, still the case that the libinput developers are
> > refusing to consider things like definable button areas on clickpads so you
> > can create a proper middle button?
>
>  Have you filed a bug for it? 
Well based on 
http://who-t.blogspot.co.uk/2014/09/libinput-common-input-stack-for-wayla... 
I didn't think there was any point...

What I did do after writing my email above was to go and half implement 
it, mostly to see if it would be feasible to maintain a locally patched 
libinput.

So if you're interested I have code that (hopefully) allows a middle 
button to exist, but I haven't added any way of configuring that button 
yet so it's size is just hard coded.

Tom

-- 
Tom Hughes (tom(a)compton.nu)
http://compton.nu/

F22 System Wide Change: Change xorg input stack to use libinput

[Peter Hutterer]

On Mon, Dec 15, 2014 at 10:38:50PM +0000, Tom Hughes wrote:
>  On 15/12/14 21:39, Peter Hutterer wrote:
>  >On Sat, Dec 13, 2014 at 10:15:41AM +0000, Tom Hughes wrote:
>  >>On 13/12/14 01:10, Kevin Kofler wrote:
>  >>
>  >>>An additional objection I have to this change proposal is that libinput
>  >>>(deliberately) only implements a small subset of the configurability of the
>  >>>old drivers, and thus, if we are going to remove the old drivers entirely,
>  >>>we are taking away flexibility from our users.
>  >>
>  >>Indeed. Is it, for example, still the case that the libinput developers are
>  >>refusing to consider things like definable button areas on clickpads so you
>  >>can create a proper middle button?
>  >
>  >Have you filed a bug for it?
>  
>  Well based on
> http://who-t.blogspot.co.uk/2014/09/libinput-common-input-stack-for-wayla...
>  I didn't think there was any point... 
tbh, I'm not a fan of the feature at all, but then again a discussion may be
worth having. But I'm not going to have this discussion on this list here.

Please file a bug or bring it up on the wayland-devel list, then we can
weigh up the pros and cons there.

>  What I did do after writing my email above was to go and half
> implement it,
>  mostly to see if it would be feasible to maintain a locally patched
>  libinput.
>  
>  So if you're interested I have code that (hopefully) allows a middle button
>  to exist, but I haven't added any way of configuring that button yet so it's
>  size is just hard coded. 
fwiw, hardcoding the size is fine IMO. The detail is in the various corner
cases though, but if you already have it you're welcome to send it to the
wayland list (or attach it to the bug report).

Cheers,
   Peter

F22 System Wide Change: Change xorg input stack to use libinput

[Tom Hughes]

On 16/12/14 00:15, Peter Hutterer wrote:

>  On Mon, Dec 15, 2014 at 10:38:50PM +0000, Tom Hughes wrote:
> > On 15/12/14 21:39, Peter Hutterer wrote:
> >> On Sat, Dec 13, 2014 at 10:15:41AM +0000, Tom Hughes wrote:
> >>> On 13/12/14 01:10, Kevin Kofler wrote:
> >>>
> >>>> An additional objection I have to this change proposal is that libinput
> >>>> (deliberately) only implements a small subset of the configurability of
> the
> >>>> old drivers, and thus, if we are going to remove the old drivers
> entirely,
> >>>> we are taking away flexibility from our users.
> >>>
> >>> Indeed. Is it, for example, still the case that the libinput developers are
> >>> refusing to consider things like definable button areas on clickpads so you
> >>> can create a proper middle button?
> >>
> >> Have you filed a bug for it?
> >
> > Well based on
> http://who-t.blogspot.co.uk/2014/09/libinput-common-input-stack-for-wayla...
> > I didn't think there was any point...
>
>  tbh, I'm not a fan of the feature at all, but then again a discussion may be
>  worth having. But I'm not going to have this discussion on this list here.
>
>  Please file a bug or bring it up on the wayland-devel list, then we can
>  weigh up the pros and cons there. 
https://bugs.freedesktop.org/show_bug.cgi?id=87402

Tom

-- 
Tom Hughes (tom(a)compton.nu)
http://compton.nu/

F22 System Wide Change: Change xorg input stack to use libinput

[Peter Hutterer]

On Sat, Dec 13, 2014 at 02:10:18AM +0100, Kevin Kofler wrote:
>  Jaroslav Reznik posted (and
>  > Change owner(s): Hans de Goede <hdegoede(a)redhat.com&gt;
>  wrote):
>  
>  > KDE: limits itself to standard X11 mouse config interfaces, no changes
>  > needed.
>  
>  Not true. We ship kcm_touchpad on the KDE spin, which definitely does depend
>  on synaptics interfaces (search for "synaptics" in:
>
> https://projects.kde.org/projects/playground/utils/kcm-touchpad/repositor...
> ).
>  Anything that does not use the synaptics driver is not considered a
>  touchpad. And kcm_touchpad is also in the process of becoming a core part of
>  upstream Plasma. (It is currently in the git.kde.org playground.)
>  
>  Therefore, porting kcm_touchpad (the rewritten 1.x series we currently ship,
>  not the old, obsolete, 0.3.x version, please make sure you look at the
>  correct version) is an essential requirement before we can move away from
>  the synaptics driver.
>  
>  
>  An additional objection I have to this change proposal is that libinput
>  (deliberately) only implements a small subset of the configurability of the
>  old drivers, and thus, if we are going to remove the old drivers entirely,
>  we are taking away flexibility from our users. 
You are welcome to file bugs in the freedesktop.org bugzilla
(Wayland/libinput) for the feature requests you have. Some of them may be
closed as wontfix, but I suspect there are others the case isn't clear-cut.

Oh, and if you do so, don't just go through the synaptics/evdev man pages
and file a bug for every option in there. I want clear use-cases that
justify each features you want us to add.

Cheers,
   Peter