port-xen/48125: dom0 panic during domU creation attempt

[gson%gson.org@localhost (Andreas Gustafsson)]

>Number:         48125
>Category:       port-xen
>Synopsis:       dom0 panic during domU creation attempt
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    port-xen-maintainer
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Thu Aug 15 12:55:00 +0000 2013
>Originator:     Andreas Gustafsson
>Release:        NetBSD 6.0.1
>Organization:
>Environment:
System: NetBSD gusto.araneus.fi 6.0.1 NetBSD 6.0.1 (XEN3_DOM0) amd64

Architecture: x86_64
Machine: amd64
>Description:

I have an amd64 machine running Xen 4.1.4 with a NetBSD 6.0.1 dom0.
It is successfully hosting one NetBSD 5.2 i386PAE domU residing
on a physical disk (as in disk = [ 'phy:/dev/wd1d,0x1,w' ]).

I have been trying to create a second domU "bloute" which unlike the
existing domU would reside in a disk image file in the dom0 file
system.

So far, all my attempts have failed; when I run "xl create -c
/usr/pkg/etc/xen/bloute" I get the error message "xenconsole: Could
not read tty from store: No such file or directory".  However, the
subject of this bug report is not the failure to create the domU, but
the more serious problem that one of my attempts triggered a panic
in the dom0.

The contents of the configuration file /usr/pkg/etc/xen/bloute were:

  kernel = "/root/netbsd-XEN3PAE_DOMU-5.2"
  memory = 512
  name = "bloute"
  vcpus = 1
  nics = 1
  vif = [ 'mac=00:0c:29:90:62:eb, bridge=bridge0' ]
  disk= [ 'file:/root/bloute.img,xvda,w' ]
  root = "xbd0a"

The following output appeared on the console:

   Aug 15 10:19:29 gusto xenbackendd[438]: Failed to read 
/local/domain/0/backend/qdisk/19/51712/script (No such file or directory)
   Aug 15 10:19:29 gusto xenbackendd[438]: Failed to read 
/local/domain/0/backend/console/19/0/script (No such file or directory)
   uvm_fault(0xffffa000057b8a60, 0x0, 1) -> e
   fatal page fault in supervisor mode
   trap type 6 code 0 rip ffffffff806d0b68 cs e030 rflags 10246 cr2  8 cpl 0 
rsp ffffa0005bb3da20
   panic: trap
   cpu0: Begin traceback...
   printf_nolog() at netbsd:printf_nolog
   startlwp() at netbsd:startlwp
   alltraps() at netbsd:alltraps+0x9f
   kernfs_try_fileop() at netbsd:kernfs_try_fileop+0x64
   VOP_WRITE() at netbsd:VOP_WRITE+0x3c
   vn_write() at netbsd:vn_write+0xf9
   dofilewrite() at netbsd:dofilewrite+0x7d
   sys_write() at netbsd:sys_write+0x62
   syscall() at netbsd:syscall+0xc4
   cpu0: End traceback...

   dump to dev 0,1 not possible
   rebooting...
   (XEN) Domain 0 shutdown: rebooting machine.

>How-To-Repeat:

Don't know.  I made several attempts at creating the domU, and
although they all failed, only the last one caused a dom0 panic.
Since the machine is also hosting a production domU and a successful
reproduction of the bug would cause downtime for that domU, I am
reluctant to try.

>Fix: