Re: -current on PowerBook G4 667 (TiBook)

[David Brownlee <abs%netbsd.org@localhost>]

On 15 July 2012 15:47, Michael <macallan%netbsd.org@localhost> wrote:
> Hello,
>
> On Sun, 15 Jul 2012 15:17:41 +0100
> David Brownlee <abs%netbsd.org@localhost> wrote:
>
>> On 15 July 2012 14:30, Michael <macallan%netbsd.org@localhost> wrote:
>> > Hello,
>> >
>> >> Mouse operation deteriorates over time.  After a few hours, the mouse
>> >> will lose a coordinate, or be limited to a small region of the upper
>> >> left portion of the screen, and eventually, be stuck in the upper left
>> >> corner of the screen.  I was using the trackpad exclusively at the time,
>> >> but plugging in a USB mouse after the fact didn't change the behavior.
>> >
>> > That's a bug in Xorg's mouse acceleration code, xset m 0 0 fixes it but 
>> > yes, it's incredibly annoying.
>>
>> Interestingly on NetBSD/vax the mouse acceleration seems to end up
>> using uninitialised data and crashing the server - I wonder if its
>> related....
>
> Maybe, I wouldn't be overly surprised. Sparc64 behaves like macppc here and I 
> haven't seen it on shark, so it's probably an endianness bug hiding somewhere.

Thats encouraging (at least for my NetBSD/vax hacking) in that its not
just the VAX or me doing something insane.

If it is the same issue then vax may be helpful in that its an instant
SEGV on startup rather than odd behaviour some non-deterministic time
after start.

Its been a while since I poked at it - I'll try to dig back in again
and determine where its going bang to see if anything is obvious...