On Tue, 8 Oct 1996 19:02:57 -0400 (EDT) 
 Andrew Gillham <gillhaa@ghost.whirlpool.com> wrote:

 > I haven't tried the actual 1.2 release kernel on my 4/110 (with P4/CG6),
 > but mine does exhibit an odd line "skip" when the CG6 is probed.  I 
 > vaguely remember a CVS commit message going by for one of the other P4
 > drivers that mentioned reading the eeprom values incorrectly? (Jason?)
 > Perhaps the same problem exists on the CG6.  Doublecheck that the values
 > for length&width are correct for your monitor.  I don't have my docs here
 > at work, but I'll check when I get home.  Possibly the screen width is
 > set funny, so the driver thinks it is "1 column x 1 row" or something
 > similar.. :)

Actually, I _thought_ the cgsix/p4 driver was working fine.  If it's not,
I'd like to know about it so I can help whoever has the hardware track
down the problem...

The problem with the cgfour and rcons is that rcons doesn't seem to
get the pixels right on that framebuffer (it tries to use the overlay
plane, which would be much faster, since it's only 1 bit deep).  I'm
not certain what the real problem is, and since I don't have any 4/110
systems (or p4 cards), I can't really debug it first-hand.

You definitely want to double-check your eeprom settings for rows/columns.
fbrcons_init() uses those values.  See arch/sparc/dev/fb.c, around
line 420.

The reason you see this "skip" is because, on the sun4, there's no
way to get the row/column address used by the terminal emulator
in the PROM.  rcons just does the best it can... I get that on an
obio bwtwo on my sun4/260 as well... it's harmless, and not really
a big deal when you consider how much faster the console becomes :-)

ciao.

Jason R. Thorpe                                       thorpej@nas.nasa.gov
NASA Ames Research Center                               Home: 408.866.1912
NAS: M/S 258-6                                          Work: 415.604.0935
Moffett Field, CA 94035                                Pager: 415.428.6939