On Mon, Dec 01, 2003 at 05:46:29PM +0200, Kasper Rönning wrote: > On Mon, 1 Dec 2003 [EMAIL PROTECTED] wrote:
> > In the old days (under kernel 2.4.x) I had > > framebuffer enabled, and a "vga=" line in my lilo.conf. I can no > > longer use this boot option, as my console is rendered useless if I > > do. After reading around, and also playing around a bit with fbset, I > > have the impression that this may be a bug in the framebuffer > > modules. Here's my situation: > > > > kernel 2.6.0-test9, compiled using gcc3.2 in a Sid environment. > > NeoMagic 2160 controller on a 1024x768 lcd laptop screen (HP Omnibook > > 4100, for those who care). > > vesafb compiled into kernel > > neofb compiled as a module and loaded at boot time. > > > > with "vga=" line in lilo.conf, or when invoking fbset, the console > > screen produces wierd colored barcode-looking lines across part or all > > of the lcd screen, and no text to speak of. > > > > Anyone know if framebuffer is still buggy, and in particular with > > neomagic cards? I see positngs from june and july on various > > kernel-related lists that suggest this. > I got bad results using radeonfb (had it compiled in the kernel), with > only a few white dots in the uppermost line blinking as the machine > booted, then I changed to vesafb and it works great. Since I have a > 1024x768 screen, I use vga=792 in lilo.conf. > So do you think it would make a difference to recompile the kernel with neofb removed and vesafb kept in? matt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
