1.\" Copyright (c) 2010 Robert N. M. Watson 2.\" All rights reserved. 3.\" 4.\" This software was developed by SRI International and the University of 5.\" Cambridge Computer Laboratory under DARPA/AFRL contract FA8750-10-C-0237 6.\" ("CTSRD"), as part of the DARPA CRASH research program. 7.\" 8.\" Redistribution and use in source and binary forms, with or without 9.\" modification, are permitted provided that the following conditions 10.\" are met: 11.\" 1. Redistributions of source code must retain the above copyright 12.\" notice, this list of conditions and the following disclaimer. 13.\" 2. Redistributions in binary form must reproduce the above copyright 14.\" notice, this list of conditions and the following disclaimer in the 15.\" documentation and/or other materials provided with the distribution. 16.\" 17.\" THIS SOFTWARE IS PROVIDED BY THE AUTHORS AND CONTRIBUTORS ``AS IS'' AND 18.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 19.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 20.\" ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHORS OR CONTRIBUTORS BE LIABLE 21.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 22.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 23.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 24.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 25.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 26.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 27.\" SUCH DAMAGE. 28.\" 29.Dd April 30, 2015 30.Dt XEN 4 31.Os 32.Sh NAME 33.Nm xen 34.Nd Xen Hypervisor Guest (DomU) Support 35.Sh SYNOPSIS 36To compile hardware-assisted virtualization (HVM) Xen guest support with 37para-virtualized drivers into an amd64 or i386 kernel, 38place the following lines in your kernel configuration file: 39.Bd -ragged -offset indent 40.Cd "options XENHVM" 41.Cd "device xenpci" 42.Ed 43.Sh DESCRIPTION 44The Xen Hypervisor allows multiple virtual machines to be run on a single 45computer system. 46When first released, Xen required that i386 kernels be compiled 47"para-virtualized" as the x86 instruction set was not fully virtualizable. 48Primarily, para-virtualization modifies the virtual memory system to use 49hypervisor calls (hypercalls) rather than direct hardware instructions to 50modify the TLB, although para-virtualized device drivers were also required 51to access resources such as virtual network interfaces and disk devices. 52.Pp 53With later instruction set extensions from AMD and Intel to support fully 54virtualizable instructions, unmodified virtual memory systems can also be 55supported; this is referred to as hardware-assisted virtualization (HVM). 56HVM configurations may either rely on transparently emulated hardware 57peripherals, or para-virtualized drivers, which are aware of virtualization, 58and hence able to optimize certain behaviors to improve performance or 59semantics. 60.Pp 61.Fx 62supports hardware-assisted virtualization (HVM) on both i386 and amd64 63kernels. 64.Pp 65Para-virtualized device drivers are required in order to support certain 66functionality, such as processing management requests, returning idle 67physical memory pages to the hypervisor, etc. 68.Ss Xen DomU device drivers 69These para-virtualized drivers are supported: 70.Bl -hang -offset indent -width blkfront 71.It Nm balloon 72Allow physical memory pages to be returned to the hypervisor as a result of 73manual tuning or automatic policy. 74.It Nm blkback 75Exports local block devices or files to other Xen domains where they can 76then be imported via 77.Nm blkfront . 78.It Nm blkfront 79Import block devices from other Xen domains as local block devices, to be 80used for file systems, swap, etc. 81.It Nm console 82Export the low-level system console via the Xen console service. 83.It Nm control 84Process management operations from Domain 0, including power off, reboot, 85suspend, crash, and halt requests. 86.It Nm evtchn 87Expose Xen events via the 88.Pa /dev/xen/evtchn 89special device. 90.It Nm netback 91Export local network interfaces to other Xen domains where they can be 92imported via 93.Nm netfront . 94.It Nm netfront 95Import network interfaces from other Xen domains as local network interfaces, 96which may be used for IPv4, IPv6, etc. 97.It Nm pcifront 98Allow physical PCI devices to be passed through into a PV domain. 99.It Nm xenpci 100Represents the Xen PCI device, an emulated PCI device that is exposed to 101HVM domains. 102This device allows detection of the Xen hypervisor, and provides interrupt 103and shared memory services required to interact with the hypervisor. 104.El 105.Ss Performance considerations 106In general, PV drivers will perform better than emulated hardware, and are 107the recommended configuration for HVM installations. 108.Pp 109Using a hypervisor introduces a second layer of scheduling that may limit the 110effectiveness of certain 111.Fx 112scheduling optimisations. 113Among these is adaptive locking, which is no longer able to determine whether 114a thread holding a lock is in execution. 115It is recommended that adaptive locking be disabled when using Xen: 116.Bd -unfilled -offset indent 117.Cd "options NO_ADAPTIVE_MUTEXES" 118.Cd "options NO_ADAPTIVE_RWLOCKS" 119.Cd "options NO_ADAPTIVE_SX" 120.Ed 121.Sh HISTORY 122Support for 123.Nm 124first appeared in 125.Fx 8.1 . 126.Sh AUTHORS 127.An -nosplit 128.Fx 129support for Xen was first added by 130.An Kip Macy Aq Mt kmacy@FreeBSD.org 131and 132.An Doug Rabson Aq Mt dfr@FreeBSD.org . 133Further refinements were made by 134.An Justin Gibbs Aq Mt gibbs@FreeBSD.org , 135.An Adrian Chadd Aq Mt adrian@FreeBSD.org , 136and 137.An Colin Percival Aq Mt cperciva@FreeBSD.org . 138This manual page was written by 139.An Robert Watson Aq Mt rwatson@FreeBSD.org . 140.Sh BUGS 141.Fx 142is only able to run as a Xen guest (DomU) and not as a Xen host (Dom0). 143.Pp 144As of this release, Xen PV DomU support is not heavily tested; instability 145has been reported during VM migration of PV kernels. 146.Pp 147Certain PV driver features, such as the balloon driver, are under-exercised. 148