What to blame for ATI lockups after upgrade ?

[Michel Dänzer]

On Mon, 2008-01-07 at 14:34 -0500, Alex Deucher wrote:
> On Jan 7, 2008 1:52 PM, Jimmy Jazz <Jimmy.Jazz at gmx.net> wrote:
> > Frédéric L. W. Meunier a écrit :
> >
> > > I'm using IceWM. But reading
> > > https://sourceforge.net/tracker/?func=detail&atid=100031&aid=1834961&group_id=31
> > > and https://bugs.freedesktop.org/show_bug.cgi?id=13688 , I guess it's
> > > something in XOrg 1.4.0.90. A patch was posted, but someone reported a
> > > new problem with the fix. I'm back to 1.4.0 to see if it happens.
> >
> > That's it! I get the same messages in Xorg log
> >
> > "
> > tossed event which came in late
> > mieqEnequeue: out-of-order valuator event; dropping.
> > "
> >
> > but only when using awn. That can't be an IceWM bug.
> >
> >
> > Also i'm using radeonfb (patched) and radeon/drm drivers with a X800
> > PCIE card and  the last evdev driver that manages both the keyboard and
> > the mouse devices. I didn't check if that is related with the new
> > xf86-input-evdev 1.2.0 input driver. I suggest you to try the old kbd
> > driver or downgrade evdev first instead going back to xorg 1.4.0.
> 
> I'm guessing radeonfb is your problem.  In most cases it does not play
> nice with the radeon driver.

Poor radeonfb gets blamed for everything... A simple test should tell
whether this is indeed a GPU lockup or rather input related: If the
server comes back up after killing it, it can't really be the former.


-- 
Earthling Michel Dänzer           |          http://tungstengraphics.com
Libre software enthusiast         |          Debian, X and DRI developer