I tried to boot some other kernels, and got mixed results:
- 2.6.39-997-generic_amd64 [latest one from drm-intel-next from kernel-ppa]
-> black screen on boot, fan spinning up and lots of disk activity, but doesn't seem to boot
- 2.6.36-020636-generic_amd64 [an older one from 10.10, from the kernel-ppa]
-> initally black screen, flickers, shows log messages for a second, then black forever (but actually boots, can hear login prompt sound)
- 2.6.35-28-generic_amd64 (2.6.35-28.50) [a previous one found on my system]
-> boots fine, and eDP1 is gone from xrandr (yay!), but unstable, no network, various errors in syslog (e.g CPU power or thermal limit exceeded, etc.)
So, I'm back to 2.6.38-8-generic, but it seems more and more unstable wrt intel drm. I often get video freezes when plugging in VGA or changing modes with xrandr, with this in syslog: "BUG: unable to handle kernel NULL pointer dereference".
In case it helps, I'm attaching an extract from syslog with a few of these, including trace, but I suppose this might be a different bug?
I tried to boot some other kernels, and got mixed results:
- 2.6.39- 997-generic_ amd64 [latest one from drm-intel-next from kernel-ppa]
-> black screen on boot, fan spinning up and lots of disk activity, but doesn't seem to boot
- 2.6.36- 020636- generic_ amd64 [an older one from 10.10, from the kernel-ppa]
-> initally black screen, flickers, shows log messages for a second, then black forever (but actually boots, can hear login prompt sound)
- 2.6.35- 28-generic_ amd64 (2.6.35-28.50) [a previous one found on my system]
-> boots fine, and eDP1 is gone from xrandr (yay!), but unstable, no network, various errors in syslog (e.g CPU power or thermal limit exceeded, etc.)
So, I'm back to 2.6.38-8-generic, but it seems more and more unstable wrt intel drm. I often get video freezes when plugging in VGA or changing modes with xrandr, with this in syslog: "BUG: unable to handle kernel NULL pointer dereference".
In case it helps, I'm attaching an extract from syslog with a few of these, including trace, but I suppose this might be a different bug?