Re: [Xen-devel] 2.6.29-rc7 pv_ops dom0 BUG / unable to handle kernel paging request

[Pasi Kärkkäinen <pasik@xxxxxx>]

On Wed, Mar 11, 2009 at 10:52:50PM +0200, Pasi Kärkkäinen wrote:
> On Sun, Mar 08, 2009 at 01:54:01PM +0200, Pasi Kärkkäinen wrote:
> > On Sat, Mar 07, 2009 at 10:09:16PM -0800, Jeremy Fitzhardinge wrote:
> > > Pasi Kärkkäinen wrote:
> > > >Hello!
> > > >
> > > >Latest git tree (updated some hours ago) boots up fine for me, xend can 
> > > >be 
> > > >started etc, but some time after starting kernel compilation I get the 
> > > >following BUG:
> > > >
> > > >http://pasik.reaktio.net/xen/pv_ops-dom0-debug/pv_ops-dom0-bootlog-19-xen331-linux-2.6.29-rc7-bug.txt
> > > >
> > > >(XEN) mm.c:2006:d0 Bad type (saw 28000001 != exp e0000000) for mfn 436a3 
> > > >(pfn 3d0a3)
> > > >(XEN) mm.c:707:d0 Error getting mfn 436a3 (pfn 3d0a3) from L1 entry 
> > > >00000000436a3063 for dom0
> > > >(XEN) mm.c:3640:d0 ptwr_emulate: could not get_page_from_l1e()
> > > >BUG: unable to handle kernel paging request at c01cbd58
> > > >IP: [<c0405d2f>] xen_set_pte+0x8c/0x96
> > > >  
> > > 
> > > Well, that's bad news.  It was trying to map a highmem pte page which 
> > > didn't have the Pinned bit set on its page, but Xen thought it was a 
> > > pinned pte page.  Not sure how it could get into that state, but its 
> > > indicative of general memory corruption.
> > > 
> > > What was going on at the time?  Was dom0 busy?  Were you running some 
> > > domUs?
> > > 
> > 
> > At that time I was compiling a kernel on dom0.. so dom0 was busy.
> > 
> > No other domains running. 
> > 
> 
> Same thing happened again.. with the latest tree (as of today).
> 
> Again I was compiling a kernel in dom0, so dom0 was busy. No other domains
> running. 
> 

The exact same kernel booted on baremetal without Xen works OK, and doesn't
BUG during kernel compilation. 

-- Pasi

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel