Nested Virtualization in Xen

---

Introduction

Nested virtualization is the ability to run a hypervisor inside of a virtual machine. The hypervisor that runs on the real hardware is called a level 0 or L0; the hypervisor that runs as a guest on L0 is called level 1 or L1; a guest that runs on the L1 hypervisor is called a level 2 or L2.

Nested virtualization is a technology that has several potential applications:

• End-user virtualization for guests. With nested virtualization, users can run Windows 7's XP compatibility mode (VirtualPC), or install Bromium in guests.

• Development. Many developers have found testing and debugging hypervisor and dom0 code much easier when it's run inside a virtual machine.

• Deployment testing. When experimenting with deployment of large clouds, admins can create "virtual clouds", comprising dozens of virtual machines, to test how cloud orchestration software or other coordination and deployment layers will work at scale, without the need to actually have a large number of dedicated machines.

Running PV guests as an L2 has been supported in Xen since the introduction of HVM guests in Xen 3.0. Support for HVM guests as L2 guests is heavily dependent on architecture-specific support. Nested HVM on AMD CPUs is considered "experimental". Nested HVM on Intel CPUs, as of Xen 4.4, is considered "tech preview". For many common cases, it should work reliably and with low overhead. However, there are some important limitations, and we do not recommend that it be used in a production environment at this time.

Only 64-bit hypervisors are supported at this time.

See below for more details on tested hypervisior / guest combinations, and known issues on Intel CPUs.

---

Quick-start guide

• Make sure you have the right support
  • Xen 4.4 or later
  • Intel CPU with EPT support

• Add the following to your config file:

    hap=1
    nestedhvm=1

Abbreviation

Abbreviation	Description
L0	Xen hypervisor
L1	First level VM which is able to start second level VM
L2	The second level VM which boot up by L1
XEN_on_XEN	L1 XEN based on L0 XEN
VMware_ESX_on_XEN	L1 VMware ESX Server on L0 XEN
VMware_Workstaion_on_XEN	L1 VMware Workstation on L0 XEN
Hyperv_on_XEN	L1 WIN8 with Hyper-v feature enabled on L0 XEN
XP_mode	L1 WIN7 VirtualPC XP_mode on L0 XEN

---

How to use nested

Booting Xen:

Nested feature is added to Xen long time ago. But it is really ready to use only from last year. So it’s better to use the latest Xen,

    git clone git://xenbits.xen.org/xen.git xen.git

NB: When I wrote this Wiki, there are still one patch is missing in upstream Xen if you want to boot up VMware and Hyper-v. You can get the patch from:

http://www.gossamer-threads.com/lists/xen/devel/316993

Booting Dom0:

No special requirement for Dom0. Just use the stable one is enough.

QEMU:

Both Qemu-xen and Qemu traditional are OK.

L1 configuration:

To use nested virtualization, you need to add the following two lines in your guest configuration file:

    hap=1
    nestedhvm=1

NB: We don’t want to support L1 guest with shadow mode, since the performance is very poor. So I recommend you to enable the EPT(hap = 1) in the L1 guest. Besides, there is a known issue that nested is failing to work with L1 shadow mode.

To boot L1 VMware-ESX, VMware-Workstation, Win8-hyperv, you needs to mask CPU bit.

    cpuid = ['0x1:ecx=0xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx']

---

Current status

Testing environment:

	Version
L0 XEN	git commit:58f5bcaf05621810f06bf5b3592e2ae87475053d
L1 XEN	git commit:e423b5cd60ff95ba3680e2e4a8440d4d19b2b13e
L1 KVM	3.12.-rc2
L1 VMware Workstation	10.0
L1 VMware ESX	5.1
L1 Hyperv	Win8 with hyperv feature enabled

L1 configuration:

    hap=1
    nestedhvm=1
    cpuid = ['0x1:ecx=0xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx']

Result:

L1\L2	XP_x86	XP_x64	RHEL6U4_x86	RHEL6U4_x64	WIN7_x86	WIN7_x64	WIN8_x86	WIN8_x64
XEN	V	V	V	V	V	V	V	V
KVM	V	V	V	V	V	V	V	V
VMware-ESX	V	V	V	V	V	V	V	V
VMware-Workstation	V*	V	V	V	V*	V	V	V
Hyperv	V	V	V	V	V	V	V	V

V means guest boots well

V* means it works before, but buggy with latest Xen

---

Know Issues

1. Only L2 EPT/Shadow on L1 EPT is working. L2 EPT/Shadow on L1 Shadow is not supported.

2. Virtual Box fails to boot on top of Xen (L1 panic while booting L2)

3. Hyper-v screen flush issue:

Current the screen flush is not correct after booting hyper-v. Two workarouds:

using remote desktop to connect guest

or set full update mode in Qemu.

4. Using populate-on-demand (memory!=maxmem) or guest paging is not supported, and may allow an L1 admin to DoS L0 hypervisor.

For this reason, Xen nested is still considered "tech preview" and unsuitable for production systems

---

Not Tested

1. L1 stress test, performance test

2. L2 stress test performance test

3. L1 save restore, live migration

4. L2 save restore, live migration

---

Reference Documents

1. Nested Virtualization on Xen http://www-archive.xenproject.org/files/xensummit_intel09/xensummit-nested-virt.pdf

2. Nested Virtualization Update from Intel http://www.slideshare.net/xen_com_mgr/nested-virtualization-update-from-intel