http://bugzilla.novell.com/show_bug.cgi?id=470341
http://bugzilla.novell.com/show_bug.cgi?id=470341#c46
--- Comment #46 from andreas bittner 2009-11-23 20:41:34 UTC ---
hmm, too late ;) damn, i was wondering already, i thought you guys wanted me to
get rid of the older Xorg.0.log files....
one more weird thing i am coming across, and maybe we are closing in on this
bug is, that back on my current testsystem (the other 785G based asus board,
radeon hd4200 chipset) i booted the linux system with the avocent kvm switch
being active on my windows box, so the linux box with that asus board booted up
til runlevel 5 with some different information or so being supplied by the kvm
box.
and when i switched the hardware kvm switch over to the by-then-booted linux
box, tadaaa it was running fine in full screen (not even out of boundaries)
1680x1050 resoltuion with the radeonhd driver.
this is really strange. so its maybe even some bios settings (the bios POSTed
also with the kvm switch being active on my other windows box, and then
continued to boot into the test 11.1 system) during POST, how it initializes
some of its tables or maybe its just during runtime of early linux stages
(framebuffer console driver?) or maybe during initial X/kdm/xdm initialisation,
so it can apparently succeed when it doesnt know (or knows different things?)
certain things being normally supplied by/through the kvm switch.
i can attach some logs of the (nonverbose Xorg.0.log during boot without the
kvm port of the linux box being selected, and also verbose Xorg.0.log manually
from runlevel 3) scenario when the signal is being routed through the kvm all
the time, the only difference (the mainboard too, but i can open an additional
bugreport, but somehow this doesnt really make sense, i still strongly believe
that its all some common code in that radeonhd driver and a common bug for all
kinds of radeon hd (at least igp integrated chipset graphics related), as this
whole phenomenon about the display staying dark and all this is always related
to the radeonhd driver module......
so maybe somebody can give some elaborate guess of what the difference might be
of the booting of the system when the kvm switch port for that system isnt
selected on the kvm hardware.
anyways, will provide further logs.
--
Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.