Port-sparc64 archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
X on XVR-100
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hello,
I got X to work on this card ( it's just a low end Radeon after
all ), there are a few gotchas though:
- - we need to #ifdef all the vgahw stuff out - it won't do any good
anyway since we don't use VGA text mode. We could just allow X to
access PCI IO space instead, just like macppc does, but that would
still not let vgahw do anything useful. I'll commit the necessary
changes after some more testing on macppc.
- - there's some funny interaction between XFree86's radeon driver and
radeonfb. X apparently switches the console back to
WSDISPLAYIO_MODE_EMUL after some driver setup, at this point radeonfb
restores its video mode, resets the drawing engine and redraws the
console. Then X switches back to WSDISPLAYIO_MODE_GRAPHICS and lets
the driver finish its setup. For some reason XFree86's radeon driver
changes resolution in the first setup phase, then radeonfb switches
back and the 2nd phase, relying on the 1st one, is screwed unless
radeonfb and X use the exact same video mode. With genfb things just
work but when X exits the radeon driver fails to restore the video
mode so you can't see the console. I suspect Xorg behaves the same.
I'm not sure how exactly to fix this yet, but tracking down where and
why X temporarily switches back to WSDISPLAYIO_MODE_EMUL sounds like
a good idea to me.
have fun
Michael
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (Darwin)
iQEVAwUBSGmUV8pnzkX8Yg2nAQJjUwf/StPxzU7idO+2e6X8m36sc2671JduX599
EiP4XPWxNUMSELJErI4uYywVrGB0qt8Cc7RxuEu6oeMWdLWxY5GvYqmvl7jHVnes
epvBbs+Ei0VIwwQIc0D1fSTvyVqO/a0loLKUUNiwTcR//S8kdlyODeenoAUkP9+1
utNqkj6WOlrIXMOFqZ6xTvxqsR8IyOyn7L3lYXGarNrUt2fNfaghxnmnWwSVJVi2
x2XsO+sBoC3jej/RUColgryWVMo4LD8rQEb5caoIsWo8YRc4JrmCQ5iF1sA+Vnw7
RhuEZFOd8sunadulgBa9xIWgu0XY5IUj1Yjw7Yvwk9mXgxSWWt1reQ==
=I0NJ
-----END PGP SIGNATURE-----
Home |
Main Index |
Thread Index |
Old Index