Re: sky2: early oops in pci_vpd_truncate

[Jesse Barnes <jbarnes@virtuousgeek.org>]

On Wed, 8 Apr 2009 14:00:05 +0200
Ingo Molnar <mingo@elte.hu> wrote:

> 
> * Harvey Harrison <harvey.harrison@gmail.com> wrote:
> 
> > On Mon, 2009-04-06 at 11:29 -0700, Jesse Barnes wrote:
> > > On Fri, 03 Apr 2009 20:32:14 -0700
> > > Harvey Harrison <harvey.harrison@gmail.com> wrote:
> > > 
> > > > On Fri, 2009-04-03 at 18:03 -0700, Jesse Barnes wrote:
> > > > > On Fri, 03 Apr 2009 17:10:19 -0700
> > > > > Harvey Harrison <harvey.harrison@gmail.com> wrote:
> > > > > 
> > > > > > Hi Stephen,
> > > > > > 
> > > > > > I'm seeing an early oops in current Linus -git right after
> > > > > > sky2_probe unfortunately it scrolls away too fast to get it
> > > > > > all, but after a delayed-printk boot it appears to be a NULL
> > > > > > dereference in pci_vpd_truncate.  sysfs also features
> > > > > > prominantly in the backtrace.. is this a known issue?
> > > > > > 
> > > > > > I see there were some sky2 patches this cycle using the vpd
> > > > > > functions, any pointers/patches to try out?
> > > > > 
> > > > > Yeah there's a patch for that in my queue (not pushed to
> > > > > Linus yet though), subject "[PATCH] PCI: Fix oops in
> > > > > pci_vpd_truncate" from Anton Vorontsov.
> > > > > 
> > > > 
> > > > What tree is that?
> > > 
> > > git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci-2.6.git.
> > > It'll be in my for-linus branch as soon as I push it out later
> > > today. I'll probably ask Linus to pick it up later this week.
> > > 
> > 
> > Looks like Ingo hit the same oops as well. CC'd.
> 
> I still have the revert - can i revert the revert from -tip now, 
> it's all fixed in .30-rc1, right?

Yeah Linus pulled the fix, hopefully Harvey can confirm that things are
working now.

-- 
Jesse Barnes, Intel Open Source Technology Center