On Wed, Jan 03, 2001 at 01:50:46PM +0100, Geert Uytterhoeven wrote:
> On Tue, 2 Jan 2001, Tom Rini wrote:
> > Hey all. While going through the 2.4 tree and removing dead CONFIG_xxx's for
> > PPC stuff, I noticed clgenfb still had CONFIG_PREP stuff (which may have
> > partily explained why it no longer worked here). I've attached a patch, that
> > with another patch to fix some PCI issues on certain machines, gives me a
> > working (so far, can't test heavily yet tho) framebuffer on my powerstack.
> >
> > Comments?
>
> To me it looks like most of them depend on `big endian', not on `PReP'.
Possibly... I don't have or know anyone with the MacPicasso video card,
which is the only other card which may have this. :)
> BTW, doesn't the Cirrus Logic graphics chip have a big endian aperture? I
> don't like things like green.offset = -3, since it will probably break some
> applications (did you run X?).
I finally got the machine up w/ an nfsroot, and played around a bit. w/o the
patch (ie isPReP always 0) I don't get a console, but a few lines of pixels in
a row, and random pixels here and there. Since the original logic was for
PReP only, I'm inclined to think that keeping it at just _machine == _MACH_prep
is the safest way to go, for the moment.
BTW, with or w/o the patch, XF4 (fbdev) wouldn't quite work right. The logs
didn't indicate anything wrong, but the monitor went black, like X was
starting, but never did. (ShadowFB on / off)
-- Tom Rini (TR1265) http://gate.crashing.org/~trini/ - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Jan 07 2001 - 21:00:19 EST