On Sun, Dec 09, 2001 at 04:07:27AM +0100, Michel Dänzer wrote: > On Sun, 2001-12-09 at 03:49, Peter Surda wrote: > > - there are minor display distortions (parts of screen remain black, basically > > the kde splash screen is completely fscked, but it sorts out later). > Does the R128BlockHandler() function of the driver you're using contain > a FLUSH_RING(); ? Yes it does. It could be possible that this bug is triggered by tvout int10, because it manifests itself as overwriting certain videoram parts with 0's. OTOH splash screen looks ok if I disable dri ....
> > - when I do a chvt 1;chvt 7, X starts eating all the CPU time and 2d
> > accelerated functions seize working (mouse can be moved but kde panel
> > doesn't pop up). killall X doesn't work, killall -9 X does. When this bug
> > feels like it should become especially mean to me, this killall -9 X causes
> > a complete machine lockup.
> In R128EnterVT(), try moving the R128ModeInit() and R128EngineInit()
> calls before the code guarded by #ifdef XF86DRI.
Yes, you were absolutely right, as always, this wasn't there and moving the
lines fixed the bug for good.
Syslog still complains CONTINUOSLY though, but I think that isn't so important
now.
> Those are fixes I made in DRI CVS some time ago, but at least the first
> one is in XFree86 CVS now.
I see.
Volodya, pls commit the second one. In r128_driver.c, there is a R128EnterVT
function, and you only have (as Michel says) move ModeInit and EngineInit (3
lines) before the #ifdef XF86DRI.
Same goes for xf86 maintainers (Marc, Mike, Michel, whoever feels like doing
it), pls commit in case it isn't there yet.
Now I hope this was the last thing that prevented me from reaching normal
uptimes.
Bye,
Peter Surda (Shurdeek) <[EMAIL PROTECTED]>, ICQ 10236103, +436505122023
--
They called their place of existence the "Universe", not the
"Great Programmer/Universe".
msg02107/pgp00000.pgp
Description: PGP signature
