3.12 kernels get stuck midway and do not get to gdm
Adam Williamson
awilliam at redhat.com
Sat Dec 21 22:22:42 UTC 2013
On Sat, 2013-12-21 at 17:34 +1100, Ankur Sinha wrote:
> On Sat, 2013-12-21 at 17:06 +1100, Ankur Sinha wrote:
> > I'll dig up more logs and stuff and see what's causing it.
>
> It looks like a DRM issue
>
> It booted into runlevel3 just fine, which pointed to an X related error.
> This is what the Xorg log gave me:
>
> > egrep EE Xorg.1.log
> > (WW) warning, (EE) error, (NI) not implemented, (??) unknown.
> > [ 19610.801] Initializing built-in extension MIT-SCREEN-SAVER
> > [ 19611.032] (EE) intel(0): [drm] failed to set drm interface version.
> > [ 19611.032] (EE) intel(0): Failed to become DRM master.
> > [ 19611.032] (EE) Screen(s) found, but none have a usable configuration.
> > [ 19611.032] (EE)
> > [ 19611.032] (EE) no screens found(EE)
> > [ 19611.032] (EE)
> > [ 19611.032] (EE) Please also check the log file at "/var/log/Xorg.1.log" for additional information.
> > [ 19611.032] (EE)
> > [ 19611.038] (EE) Server terminated with error (1). Closing log file.
> > [asinha at ankur-laptop ~]$
>
> The system hung up in a bit, even in run level 3, and I had to power it
> off using the button. Here's a screenshot:
> http://ankursinha.fedorapeople.org/kernel-3.12-failures/drm-hangup.jpg
>
> I'm guessing this confirms this to be a kernel/drm/nouveau issue? X
> wouldn't cause the entire system to hang up in runlevel3? (I've checked
> the kernel bug list and haven't found anything like this yet.) Any hints
> on how I could collect more information on the issue?
>
> Complete Xorg log is here:
> http://ankursinha.fedorapeople.org/kernel-3.12-failures/Xorg.1.log
Well, it's dying in the intel driver, not nouveau...
I'd file a bug against xorg-x11-drv-intel either up or downstream, and
include the contents of 'dmesg' or the journal when booting with
'drm.debug=15' and reproducing the problem.
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net
http://www.happyassassin.net
More information about the test
mailing list