Side note: after reverting 196705c9bb I can't get the mouse to skip any
more on that mac mini. But since the bad behaviour wasn't 100% reliable to
begin with, that's not really a guarantee of anything. Two out of three
kids are off on camp this week, so that machine probably won't be getting
a lot of testing ;/
That's with an all-intel chipset, no VIA or Broadcom anywhere.
I do find it interesting that VIA apparently ignores the INACTIVE bit: it
implies that Windows probably doesn't use it, which in turn implies that
nobody has ever tested it in any real life situation. So I would not be
shocked to hear that others have problems too.
But as mentioned, the mouse behaviour was flaky (but very noticeable and
irritating when it did happen), so it's hard to be sure my quick testing
really was convincing. Which is why it's even harder to bisect ;(.
Linus
-
