Bug#270743: radeonfb cannot detect display panel size and breaks
Hi,
first: my email setup was temporarily broken. This is a bundled reply, see the
post scriptum.
Sven Luther wrote:
> I also have a Ibook 2.2, the 12" G3/800Mhz one, from february 2003 or
> thereabout, altough i had the motherboard changed earlier this year due to the
> infamous ibook motherboard defect.
Same here, same defect some time ago.
> Radeonfb on 2.6.8-4 works just fine with the debian kernel, as you can see in
> the attached dmesg output, so really i don't understand what is your problem.
This seems to happen only with some machines. Fortunately. Or
unfortunately? :-)
> Do you perhaps have a second monitor or a TV device attached, or maybe do you
> have some hardware defect that makes the radeon chip think something is bad ?
a) No.
b) No indication with MacOS running. Everything fine with my modified
driver.
Regarding your dmesg, I'm not expert enough to interpret all the
differences. Just one thing:
> PCI: Enabling device 0000:00:10.0 (0086 -> 0087)
> radeonfb: Invalid ROM signature 0 should be 0xaa55
> radeonfb: Retreived PLL infos from Open Firmware
> radeonfb: Reference=27.00 MHz (RefDiv=12) Memory=180.00 Mhz, System=180.00 MHz
> radeonfb: Monitor 1 type LCD found
> radeonfb: EDID probed
The last message is missing on my machine. Broken EDID, it seems.
---- next mail ----
Sven Luther wrote:
> The output is very different from mine, and this is clearly a self compiled
> kernel, so are you sure you did enable the radeon_new driver, and not the
> radeon_old one ?
Hmm, well I built this kernel from kernel-tree-2.6.7 with
PATCH_THE_KERNEL=YES and
$ cp /boot/config-2.6.8-powerpc .config
$ make oldconfig
$ make-kpkg --configure <etc>
I guess this is the best method to be as close to the stock Debian
kernel as possible, right? Though AFAIR there were some new config
questions.
And no, it's the new radeon driver:
CONFIG_DRM_RADEON=m
# CONFIG_FB_RADEON_OLD is not set
CONFIG_FB_RADEON=y
CONFIG_FB_RADEON_I2C=y
CONFIG_FB_RADEON_DEBUG=y
> But then the output level is maybe only because you chose a
> verbose radeonfb ?
dmesg's output of all kernels I tried is available on my web page, here
the link again:
http://www.math.fu-berlin.de/~nschulz/linux/trouble/ibook/
---- next mail ----
Jens Schmalzing wrote:
> This is very strange, since the Mobility M7 in the iBook is supposed
> to work perfectly. Can you rule out hardware issues?
How could I? But nothing's wrong with MacOS X.
> Did you by any chance run that little hack that enables true dual-head
> in Mac OS and Mac OS X?
I did. But the motherboard has been replaced since and I didn't apply it
again.
> > AFAIK radeonfb is required to run X on the iBook, right?
>
> Unfortunately, yes.
Oh, damn it. :-(
> You could file a bug report against the Xserver
> about this and contact upstream as well.
Hm, I'm not sure if I understand you correctly: file a bug against
the xserver because it requires the kernel's radeonfb?
Regards,
Nikolaus
P.S.
Unfortunately my original submitter email address isn't functional.
Please ignore <nikolaus@kunigunde.dynalias.net> and
use <microschulz@web.de> instead, or CC me if necessary.
Mails to 270743@bugs.debian.org don't reach me.
Does this indicate a bug in the BTS, by the way?
Reply to: