Simon 'corecode' Schubert
corecode at fs.ei.tum.de
Tue Jan 31 06:54:10 PST 2006
i'm running into more and more weird things: this time it's sysmouse.
I wondered why X (working with an updated version of the free nv
driver) didn't do the scrollwheel game.
Turns out that /dev/sysmouse is sending bullshit: Unsynced frames,
sometimes one byte too much, sometimes some bytes too few, short reads
and zero reads. No mouse click data at all (dunno if that used to
work). hd /dev/sysmouse *should* display a constant stream when moving
the mouse, but it aborts and the data is bogus. It has been verified
that it occurs at at least one different system.
Output can be debugged with
It *is* sysmouse, because the mouse data correctly arrives at
syscons/console mouse. I suspect there is some bug in the tty code,
but even a kernel from 2005-10-11 didn't show correct behaviour. I
suspect it to have something to do with tty handling, but I got too few
time to debug in-depth at the moment.
I wonder why this worked before at all? Maybe some race condition
which didn't show up with the slower processor before.
Serve - BSD +++ RENT this banner advert +++ ASCII Ribbon /"\
Work - Mac +++ space for low â¬â¬â¬ NOW!1 +++ Campaign \ /
Party Enjoy Relax | http://dragonflybsd.org Against HTML \
Dude 2c 2 the max ! http://golden-apple.biz Mail + News / \Attachment:
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 186 bytes
Desc: "Description: This is a digitally signed message part"
More information about the Kernel