On Tue, May 09, 2000 at 01:39:11PM +0200, Michel D�nzer wrote: > Sven LUTHER wrote: > > > > On Tue, May 09, 2000 at 11:22:56AM +0200, Michel D�nzer wrote: > > > Brendan J Simon wrote: > > > > (--) PCI: (0:16:0) ATI unknown chipset (0x4c46) rev 2, Mem @ > ^^^^^^ > > > > 0xa4000000/26, 0xa0000000/14, I/O @ 0x0400/8 > > > > > > Are you sure the chip in those Notebooks is from the Rage128 family? > > > > > > If it's in fact from the mach64 family, you'll have to use the fbdev > > > driver for now - you might need a recent kernel which corrects broken OF > > > PCI memory ranges from though. > > > > No it is a Rage128, but a Rage128 mobility chip. I am sure this chips have > > different identifier than the desktop version of the rage128. So this would > > have to be added at the right place (should be not so difficult, please > > report the chip id from /proc/pci, or if a name is given there, look at the > > > kenrel source for the place where this name is defined.) > > The ID is there.
For the Rage 128 mobility ? maybe it is a special apple rage 128 mobility :))) > > Then you have only to add it to the right place in the 4.0 source and > > recompile. > > > > Though i think the best way wouldbe to get the device id, and then contact > > the Xfree rage128 maintainer (on xfree-mga and/or xfree-devel i think). > > mga??? Sorry, i was sleeping ... > I'd say either Keith Packard, Kostas Gewrgiou or Ani Joshi wouldn't be bad. BTW, michel, i enabled sync and clipping accels on the pm3 driver, same functions as pm2 ones, the rest cause problems and will have to be rewriten. but when i enter a bad login in gdm, the gdm requester shivers from right to left and vice versa, about 10-20 pixels, becoming less, before stabilizing. Did you already encounter such a problem ? Friendly, Sven LUTHER
