2:37 p.m.
Dear fellow testers,
New kernel oops again and no X when starting it
kernel-2.6.27-0.215.rc1.git4.fc10.i686
Here's oops
http://www.kerneloops.org/submitresult.php?number=48331
Regards,
Antonio
3:16 p.m.
On Tue, Aug 05, 2008 at 12:37:32PM -0700, Antonio Olivares wrote:
Dear fellow testers,
New kernel oops again and no X when starting it
kernel-2.6.27-0.215.rc1.git4.fc10.i686
Here's oops
http://www.kerneloops.org/submitresult.php?number=48331
nasty... did this just start recently?
Any thoughts, Dave?
r, Kyle
3:18 p.m.
On Tue, Aug 05, 2008 at 04:16:39PM -0400, Kyle McMartin wrote:
On Tue, Aug 05, 2008 at 12:37:32PM -0700, Antonio Olivares wrote:
> Dear fellow testers,
>
> New kernel oops again and no X when starting it
>
> kernel-2.6.27-0.215.rc1.git4.fc10.i686
>
> Here's oops
>
> http://www.kerneloops.org/submitresult.php?number=48331
>
nasty... did this just start recently?
Any thoughts, Dave?
fwiw this is hitting the BUG_ON in drm_open.
mutex_lock(&dev->struct_mutex);
BUG_ON((dev->dev_mapping != NULL) &&
(dev->dev_mapping != inode->i_mapping));
if (dev->dev_mapping == NULL)
dev->dev_mapping = inode->i_mapping;
mutex_unlock(&dev->struct_mutex);
3:47 p.m.
On Tue, Aug 05, 2008 at 04:16:39PM -0400, Kyle McMartin wrote:
On Tue, Aug 05, 2008 at 12:37:32PM -0700, Antonio Olivares wrote:
> Dear fellow testers,
>
> New kernel oops again and no X when starting it
>
> kernel-2.6.27-0.215.rc1.git4.fc10.i686
>
> Here's oops
>
> http://www.kerneloops.org/submitresult.php?number=48331
>
nasty... did this just start recently?
Not exactly and for some value "recently". Tom London filed
https://bugzilla.redhat.com/show_bug.cgi?id=457598
few days ago, and it seems to be that one referenced above, and
I just added
https://bugzilla.redhat.com/show_bug.cgi?id=457971
as this appears to be something different
(NULL pointer dereference at 0000000000000000 in
drm_ati_pcigart_init). All of that is pretty consistent
when "rc1" kernels are in use. Those from "rc0" series for
2.6.27 look like working.
Kernel package changelog has entries like:
* Sat Aug 02 2008 Dave Airlie <airlied(a)redhat.com>
- Fix locking in drm patches
but that does not seem to be so far "it".
Michal
4:58 p.m.
On Tue, Aug 05, 2008 at 02:47:18PM -0600, Michal Jaegermann wrote:
Kernel package changelog has entries like:
* Sat Aug 02 2008 Dave Airlie <airlied(a)redhat.com>
- Fix locking in drm patches
but that does not seem to be so far "it".
Michal
I'll do a scratch build with DRM patches turned off so you can see if
the problems goes away.
I'll send a mail when it's done.
r, Kyle
8:40 p.m.
On Tue, Aug 05, 2008 at 05:58:40PM -0400, Kyle McMartin wrote:
On Tue, Aug 05, 2008 at 02:47:18PM -0600, Michal Jaegermann wrote:
> Kernel package changelog has entries like:
>
> * Sat Aug 02 2008 Dave Airlie <airlied(a)redhat.com>
> - Fix locking in drm patches
>
> but that does not seem to be so far "it".
>
I'll do a scratch build with DRM patches turned off so you can see if
the problems goes away.
Thanks. FYI - kernel-2.6.27-0.226.rc1.git5.fc10.i686 from koji is
even worse that all previous "rc1" I tried. Instead of "just"
killing X and crashing video it either locks up everything (no
traces in logs) or reboots my machine after 'startx' from a console.
More details at https://bugzilla.redhat.com/show_bug.cgi?id=457971#c1
Michal
8:50 a.m.
New subject: kernel-2.6.27-0.215.rc1.git4.fc10.i686 oops and no X :(
On Tue, Aug 5, 2008 at 6:40 PM, Michal Jaegermann <michal(a)harddata.com> wrote:
On Tue, Aug 05, 2008 at 05:58:40PM -0400, Kyle McMartin wrote:
> On Tue, Aug 05, 2008 at 02:47:18PM -0600, Michal Jaegermann wrote:
> > Kernel package changelog has entries like:
> >
> > * Sat Aug 02 2008 Dave Airlie <airlied(a)redhat.com>
> > - Fix locking in drm patches
> >
> > but that does not seem to be so far "it".
> >
>
> I'll do a scratch build with DRM patches turned off so you can see if
> the problems goes away.
Thanks. FYI - kernel-2.6.27-0.226.rc1.git5.fc10.i686 from koji is
even worse that all previous "rc1" I tried. Instead of "just"
killing X and crashing video it either locks up everything (no
traces in logs) or reboots my machine after 'startx' from a console.
More details at https://bugzilla.redhat.com/show_bug.cgi?id=457971#c1
Michal
I'm not sure why, but 0.226 is actually a bit better for me. But I
only get the "kernel BUG at drivers/gpu/drm/drm_fops.c:144!", and the
resulting death of X ("black screen of death").
0.226 boots up to gdm (runlevel 5) with some frequency; I don't think
I got earlier ones to ever boot up to gdm. (I could get it to boot
most of the times by booting to runlevel 3, and then running
'startx').
I continue to track this issue here:
https://bugzilla.redhat.com/show_bug.cgi?id=457598
tom
--
Tom London
11:28 a.m.
New subject: kernel-2.6.27-0.215.rc1.git4.fc10.i686 oops and no X :(
On Wed, Aug 06, 2008 at 06:50:30AM -0700, Tom London wrote:
On Tue, Aug 5, 2008 at 6:40 PM, Michal Jaegermann
<michal(a)harddata.com> wrote:
>
> FYI - kernel-2.6.27-0.226.rc1.git5.fc10.i686 from koji is
> even worse that all previous "rc1" I tried.
>
> More details at https://bugzilla.redhat.com/show_bug.cgi?id=457971#c1
>
I'm not sure why, but 0.226 is actually a bit better for me. But I
only get the "kernel BUG at drivers/gpu/drm/drm_fops.c:144!"
This appears to be a different bug than what I am seeing.
Maybe an underlying cause is the same (locking?) or maybe not.
and the
resulting death of X ("black screen of death").
This is a common final result although 0.226 apparently has extra
attractions too.
Michal
6:58 p.m.
With kernel-2.6.27-0.237.rc2.fc10 (mind you, this is "rc2"
and not "rc1" anymore) X on my machine appears to start again.
X takes a long time, blinking screen like crazy many times,
but on those few times I tried that it eventually started.
Michal
7:56 p.m.
New subject: kernel-2.6.27-0.215.rc1.git4.fc10.i686 oops and no X :(
On Thu, Aug 7, 2008 at 4:58 PM, Michal Jaegermann <michal(a)harddata.com> wrote:
With kernel-2.6.27-0.237.rc2.fc10 (mind you, this is "rc2"
and not "rc1" anymore) X on my machine appears to start again.
X takes a long time, blinking screen like crazy many times,
but on those few times I tried that it eventually started.
Michal
X comes up OK about 3 out of 4 attempts for me with 0.237.
I still get the BUG at drm_fops.c:144 when it black screens.
When it does come up, it seems to work like a champ.
tom
--
Tom London
5751
days inactive
5754
days old
9 comments
4 participants
participants (4)
-
Antonio Olivares -
Kyle McMartin -
Michal Jaegermann -
Tom London