I think KWin's Wayland input mapping logic needs to be adjusted then?

Possibly.

The KCM does not appear to actually work on wayland though ^^

if (!QX11Info::isPlatformX11()) {
    return;

In D3650#68360, @sitter wrote:

Possibly.

The KCM does not appear to actually work on wayland though ^^

It does. I ported it and it works and adjusts the settings for Wayland if KWin is used.

Ok. Kindly take note of potential adjustment needs then.

Do I need to do something to get this moving forward?

According to my comments in the Wayland port: the value 1.0 means no acceleration. Values below 1.0 mean a de-acceleration or a slow down. But it's still a factor. I'm also afraid that putting a 0.0 there could result in a devision by 0 value.

So in any case your commit message does not match the actual reality. And given that I think you are trying to fix the wrong problem. Might it be that you use the libinput xorg driver? If yes, that's not yet supported and needs porting. But this change has potential of breaking the non-libinput xorg driver then.

From my side something between -1 and -2 for this change.

That's not really answering my question now is it?

Ping?

In D3650#80584, @graesslin wrote:

According to my comments in the Wayland port: the value 1.0 means no acceleration. Values below 1.0 mean a de-acceleration or a slow down. But it's still a factor. I'm also afraid that putting a 0.0 there could result in a devision by 0 value.

So in any case your commit message does not match the actual reality. And given that I think you are trying to fix the wrong problem. Might it be that you use the libinput xorg driver? If yes, that's not yet supported and needs porting. But this change has potential of breaking the non-libinput xorg driver then.

From my side something between -1 and -2 for this change.

Does this mean that 0.0 means not moving at all? That doesn't sound right.

In D3650#112888, @apol wrote:

In D3650#80584, @graesslin wrote:

According to my comments in the Wayland port: the value 1.0 means no acceleration. Values below 1.0 mean a de-acceleration or a slow down. But it's still a factor. I'm also afraid that putting a 0.0 there could result in a devision by 0 value.

So in any case your commit message does not match the actual reality. And given that I think you are trying to fix the wrong problem. Might it be that you use the libinput xorg driver? If yes, that's not yet supported and needs porting. But this change has potential of breaking the non-libinput xorg driver then.

From my side something between -1 and -2 for this change.

Does this mean that 0.0 means not moving at all? That doesn't sound right.

It could mean a crash in X - division by zero. No idea how X handles the situation, but 0 is clearly a wrong value here.