Re: KVM induced panic on 2.6.38[2367] & 2.6.39

[Hugh Dickins <hughd@google.com>]

On Tue, 31 May 2011, Brad Campbell wrote:
> On 31/05/11 18:38, Borislav Petkov wrote:
> > On Tue, May 31, 2011 at 05:26:10PM +0800, Brad Campbell wrote:
> > > On 31/05/11 13:47, Borislav Petkov wrote:
> > > > Looks like a KSM issue. Disabling CONFIG_KSM should at least stop your
> > > > machine from oopsing.
> > > > 
> > > > Adding linux-mm.
> > > > 
> > > 
> > > I initially thought that, so the second panic was produced with KSM
> > > disabled from boot.
> > > 
> > > echo 0>  /sys/kernel/mm/ksm/run
> > > 
> > > If you still think that compiling ksm out of the kernel will prevent
> > > it then I'm willing to give it a go.
> > 
> > Ok, from looking at the code, when KSM inits, it starts the ksm kernel
> > thread and it looks like your oops comes from the function that is run
> > in the kernel thread - ksm_scan_thread.
> > 
> > So even if you disable it from sysfs, it runs at least once.
> > 
> 
> Just to confirm, I recompiled 2.6.38.7 without KSM enabled and I've been
> unable to reproduce the bug, so it looks like you were on the money.
> 
> I've moved back to 2.6.38.7 as 2.6.39 has a painful SCSI bug that panics
> about 75% of boots, and the reboot cycle required to get luck my way into a
> working kernel is just too much hassle.
> 
> It would appear that XP zero's its memory space on bootup, so there would be
> lots of pages to merge with a couple of relatively freshly booted XP
> machines running.

Thanks for the Cc, Borislav.

Brad, my suspicion is that in each case the top 16 bits of RDX have been
mysteriously corrupted from ffff to 0000, causing the general protection
faults.  I don't understand what that has to do with KSM.

But it's only a suspicion, because I can't make sense of the "Code:"
lines in your traces, they have more than the expected 64 bytes, and
only one of them has a ">" (with no "<") to mark faulting instruction.

I did try compiling the 2.6.39 kernel from your config, but of course
we have different compilers, so although I got close, it wasn't exact.

Would you mind mailing me privately (it's about 73MB) the "objdump -trd"
output for your original vmlinux (with KSM on)?  (Those -trd options are
the ones I'm used to typing, I bet not they're not all relevant.)

Of course, it's only a tiny fraction of that output that I need,
might be better to cut it down to remove_rmap_item_from_tree and
dup_fd and ksm_scan_thread, if you have the time to do so.

Thanks,
Hugh

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>