[ANNOUNCE] predictable pointer acceleration

[Tiago Vignatti]

Hi,

Simon Thum escreveu:
> Hello list,
> 
> predictable pointer acceleration in X has arrived. Enjoy!

Nice work, Simon! I did a quickly test here and it seems promiser. Very 
much appreciated.


> The main features:
>       * user-selectable profiles control pointer acceleration
>       * adaptive and constant deceleration
>       * acceleration becomes predictable
>       * no overshoot when X blocks for a short time
           ^^^^^^^^^^^^^^^^^^^^^^^^^^
           I don't think so (see below).


>    * The code could further improve blocking cases using better
> timestamps, such as evdev delivers. However, they can't be passed to DIX
> at the moment. Moreover, Tiago's input thread could have a practically
> equivalent effect. This would be strictly optional anyway since most
> sources don't do timestamps. Opinions?

I can't see a way to improve blocking cases with the predictable pointer 
acceleration.

A good timestamps (i.e. timestamps close to device hw) wouldn't help so 
much the cursor's blink on screen simply because its memory footprint 
would be paged on the disk and this is exactly the point of contention. 
We can't do much here [0]. The other reason of jumping cursor is when 
the server is sleeping uninterruptible -- in the case of silken mouse 
[1]. So the pointer accel wouldn't make any difference here at all with 
our current input flow model.


Thank you,

[0] yeshh we can! Shortcutting DRM and input subsystems in-kernel :)
[1] input-thread rework solve this issue.

-- 
Tiago Vignatti
C3SL - Centro de Computação Científica e Software Livre
www.c3sl.ufpr.br