2.6.37-6.15 seems buggered

Andy Whitcroft apw at canonical.com
Mon Nov 22 12:07:57 UTC 2010


On Mon, Nov 22, 2010 at 11:09:10AM +0000, Andy Whitcroft wrote:
> On Mon, Nov 22, 2010 at 09:00:09AM +0000, Andy Whitcroft wrote:
> > On Sun, Nov 21, 2010 at 09:39:03AM -0700, Tim Gardner wrote:
> > > I'm having crashes on boot with my Nvidia based Dell XPSM1710 (using 
> > > Noveau). We should probably get the archive admins to hold off on NEWing 
> > > the binaries.
> > > 
> > > The first indication of something wrong is a stack trace from 
> > > kernel/trace/ftrace.c:1014
> > > 
> > > I also crashes on one of my ATI Radeon based servers, but I can't tell 
> > > if its at the same place.
> > 
> > I am getting indications it may be triggering vile panics in modules
> > as well.  I am suspicious this is the module NXRO patches in my case.
> > Anynow am investigating.
> 
> Yeah in testing here it seems to be the module NXRO patches which are
> blowing things up.  It looks superficially that we are going read-only
> before all linking is complete.  I'll get with you to test the updated
> kernels with your two broken boxed, but reverting that patch cirtainly
> fixes all of the machines I have tested here.

Tim,

Ok, I have published some kernels with just these reverts to confirm
that your issues are the same as mine; and I strongly suspect they are.
Could you test these ones on your problematic systems to confirm this is
the only flaw:

    http://people.canonical.com/~apw/ronx-revert-natty/

Cheers.

-apw




More information about the kernel-team mailing list