Comment 123 for bug 1220426

The bug, for me, is fairly easy to reproduce. The trick is to be actually using the device, by that I mean touchpad, for a while. In fact, I've set up a script to search my xorg log in regular intervals for the messages that occur on VT switches, which I only do when it freezes, and made it send to an IRC channel whenever it happened so my friends could see my suffering. They sarcastically told me that it was a great advertisement for nvidia-powered laptops.
I've also noticed that when I plug in a wacom tablet and use it for a while, and then touch the touchpad again after a bit, it very often leads to a freeze, as if the touchpad was just waiting for me to use it again. Though, this could be entirely psychological. I haven't gathered enough data to conclude it is in any way related.

However, I do have gathered enough data to confidently say it does not occur while the touchpad is not being touched. To show this, I've posted a graph here:
https://devtalk.nvidia.com/default/topic/766166/elantech-touchpad-related-nvidia-driver-freeze/#4278245

To reproduce it, I have a suggestion: Let one of your interns or someone else play spacechem for a while, with the touchpad. Spacechem includes enough dragging and dropping for the issue to be triggered. If that isn't enough, take upon the unpleasanr task of playing first person shooters on a touchpad.

Another thing I noticed is that kwin uses 100% CPU when I switch from the frozen X11 to a TTY.