File:  [NetBSD Developer Wiki] / wikisrc / ports / xen / howto.mdwn
Revision 1.14: download - view: text, annotated - select for diffs
Tue Dec 23 23:43:27 2014 UTC (4 years, 11 months ago) by gdt
Branches: MAIN
CVS tags: HEAD
fix header levesl.

fix dom0 and domU capitalization.

    1: Introduction
    2: ============
    3: 
    4: [![[Xen
    5: screenshot]](http://www.netbsd.org/gallery/in-Action/hubertf-xens.png)](../../gallery/in-Action/hubertf-xen.png)
    6: 
    7: Xen is a virtual machine monitor or hypervisor for x86 hardware
    8: (i686-class or higher), which supports running multiple guest
    9: operating systems on a single physical machine.  With Xen, one uses
   10: the Xen kernel to control the CPU, memory and console, a dom0
   11: operating system which mediates access to other hardware (e.g., disks,
   12: network, USB), and one or more domU operating systems which operate in
   13: an unprivileged virtualized environment.  IO requests from the domU
   14: systems are forwarded by the hypervisor (Xen) to the dom0 to be
   15: fulfilled.
   16: 
   17: Xen supports two styles of guests.  The original is Para-Virtualized
   18: (PV) which means that the guest OS does not attempt to access hardware
   19: directly, but instead makes hypercalls to the hypervisor.  This is
   20: analogous to a user-space program making system calls.  (The dom0
   21: operating system uses PV calls for some functions, such as updating
   22: memory mapping page tables, but has direct hardware access for disk
   23: and network.)   PV guests must be specifically coded for Xen.
   24: 
   25: The more recent style is HVM, which means that the guest does not have
   26: code for Xen and need not be aware that it is running under Xen.
   27: Attempts to access hardware registers are trapped and emulated.  This
   28: style is less efficient but can run unmodified guests.
   29: 
   30: At boot, the dom0 kernel is loaded as module with Xen as the kernel.
   31: The dom0 can start one or more domUs.  (Booting is explained in detail
   32: in the dom0 section.)
   33: 
   34: NetBSD supports Xen in that it can serve as dom0, be used as a domU,
   35: and that Xen kernels and tools are available in pkgsrc.  This HOWTO
   36: attempts to address both the case of running a NetBSD dom0 on hardware
   37: and running NetBSD as a domU in a VPS.
   38: 
   39: Prerequisites
   40: -------------
   41: 
   42: Installing NetBSD/Xen is not extremely difficult, but it is more
   43: complex than a normal installation of NetBSD.
   44: 
   45: This HOWTO presumes a basic familiarity with the Xen system
   46: architecture.
   47: 
   48: This HOWTO presumes familiarity with installing NetBSD on i386/amd64
   49: hardware and installing software from pkgsrc.
   50: 
   51: For more details on Xen, see [](http://www.xen.org/).
   52: 
   53: Installing Xen with NetBSD as privileged domain (dom0)
   54: ======================================================
   55: 
   56: Note that it doesn't make sense to talk about installing a dom0 OS
   57: without also installing Xen itself.
   58: 
   59: First do a NetBSD/i386 or NetBSD/amd64
   60: [installation](../../docs/guide/en/chap-inst.html) of the 5.1 release
   61: (or newer) as you usually do on x86 hardware. The binary releases are
   62: available from [](ftp://ftp.NetBSD.org/pub/NetBSD/). Binary snapshots
   63: for current and the stable branches are available on daily autobuilds.
   64: If you plan to use the `grub` boot loader, when partitioning the disk
   65: you have to make the root partition smaller than 512Mb, and formatted as
   66: FFSv1 with 8k block/1k fragments. If the partition is larger than this,
   67: uses FFSv2 or has different block/fragment sizes, grub may fail to load
   68: some files. Also keep in mind that you'll probably want to provide
   69: virtual disks to other domains, so reserve some partitions for these
   70: virtual disks. Alternatively, you can create large files in the file
   71: system, map them to vnd(4) devices and export theses vnd devices to
   72: other domains.
   73: 
   74: Next step is to install the Xen packages via pkgsrc or from binary
   75: packages. See [the pkgsrc
   76: documentation](http://www.NetBSD.org/docs/pkgsrc/) if you are unfamiliar
   77: with pkgsrc and/or handling of binary packages. Xen 3.1, 3.3, 4.1 and
   78: 4.2 are available. 3.1 supports PCI pass-through while other versions do
   79: not. You'll need either `sysutils/xentools3` and `sysutils/xenkernel3`
   80: for Xen 3.1, `sysutils/xentools33` and `sysutils/xenkernel33` for Xen
   81: 3.3, `sysutils/xentools41` and `sysutils/xenkernel41` for Xen 4.1. or
   82: `sysutils/xentools42` and `sysutils/xenkernel42` for Xen 4.2. You'll
   83: also need `sysutils/grub` if you plan do use the grub boot loader. If
   84: using Xen 3.1, you may also want to install `sysutils/xentools3-hvm`
   85: which contains the utilities to run unmodified guests OSes using the
   86: *HVM* support (for later versions this is included in
   87: `sysutils/xentools`). Note that your CPU needs to support this. Intel
   88: CPUs must have the 'VT' instruction, AMD CPUs the 'SVM' instruction. You
   89: can easily find out if your CPU support HVM by using NetBSD's cpuctl
   90: command:
   91: 
   92:     # cpuctl identify 0
   93:     cpu0: Intel Core 2 (Merom) (686-class), id 0x6f6
   94:     cpu0: features 0xbfebfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR>
   95:     cpu0: features 0xbfebfbff<PGE,MCA,CMOV,PAT,PSE36,CFLUSH,DS,ACPI,MMX>
   96:     cpu0: features 0xbfebfbff<FXSR,SSE,SSE2,SS,HTT,TM,SBF>
   97:     cpu0: features2 0x4e33d<SSE3,DTES64,MONITOR,DS-CPL,,TM2,SSSE3,CX16,xTPR,PDCM,DCA>
   98:     cpu0: features3 0x20100800<SYSCALL/SYSRET,XD,EM64T>
   99:     cpu0: "Intel(R) Xeon(R) CPU            5130  @ 2.00GHz"
  100:     cpu0: I-cache 32KB 64B/line 8-way, D-cache 32KB 64B/line 8-way
  101:     cpu0: L2 cache 4MB 64B/line 16-way
  102:     cpu0: ITLB 128 4KB entries 4-way
  103:     cpu0: DTLB 256 4KB entries 4-way, 32 4MB entries 4-way
  104:     cpu0: Initial APIC ID 0
  105:     cpu0: Cluster/Package ID 0
  106:     cpu0: Core ID 0
  107:     cpu0: family 06 model 0f extfamily 00 extmodel 00
  108: 
  109: Depending on your CPU, the feature you are looking for is called HVM,
  110: SVM or VMX.
  111: 
  112: Next you need to copy the selected Xen kernel itself. pkgsrc installed
  113: them under `/usr/pkg/xen*-kernel/`. The file you're looking for is
  114: `xen.gz`. Copy it to your root file system. `xen-debug.gz` is a kernel
  115: with more consistency checks and more details printed on the serial
  116: console. It is useful for debugging crashing guests if you use a serial
  117: console. It is not useful with a VGA console.
  118: 
  119: You'll then need a NetBSD/Xen kernel for *domain0* on your root file
  120: system. The XEN3PAE\_DOM0 kernel or XEN3\_DOM0 provided as part of the
  121: i386 or amd64 binaries is suitable for this, but you may want to
  122: customize it. Keep your native kernel around, as it can be useful for
  123: recovery. *Note:* the *domain0* kernel must support KERNFS and `/kern`
  124: must be mounted because *xend* needs access to `/kern/xen/privcmd`.
  125: 
  126: Next you need to get a bootloader to load the `xen.gz` kernel, and the
  127: NetBSD *domain0* kernel as a module. This can be `grub` or NetBSD's boot
  128: loader. Below is a detailled example for grub, see the boot.cfg(5)
  129: manual page for an example using the latter.
  130: 
  131: This is also where you'll specify the memory allocated to *domain0*, the
  132: console to use, etc ...
  133: 
  134: Here is a commented `/grub/menu.lst` file:
  135: 
  136:     #Grub config file for NetBSD/xen. Copy as /grub/menu.lst and run
  137:     # grub-install /dev/rwd0d (assuming your boot device is wd0).
  138:     #
  139:     # The default entry to load will be the first one
  140:     default=0
  141: 
  142:     # boot the default entry after 10s if the user didn't hit keyboard
  143:     timeout=10
  144: 
  145:     # Configure serial port to use as console. Ignore if you'll use VGA only
  146:     serial --unit=0 --speed=115200 --word=8 --parity=no --stop=1
  147: 
  148:     # Let the user select which console to use (serial or VGA), default
  149:     # to serial after 10s
  150:     terminal --timeout=10 serial console
  151: 
  152:     # An entry for NetBSD/xen, using /netbsd as the domain0 kernel, and serial
  153:     # console. Domain0 will have 64MB RAM allocated.
  154:     # Assume NetBSD is installed in the first MBR partition.
  155:     title Xen 3 / NetBSD (hda0, serial)
  156:       root(hd0,0)
  157:       kernel (hd0,a)/xen.gz dom0_mem=65536 com1=115200,8n1
  158:       module (hd0,a)/netbsd bootdev=wd0a ro console=ttyS0
  159: 
  160:     # Same as above, but using VGA console
  161:     # We can use console=tty0 (Linux syntax) or console=pc (NetBSD syntax)
  162:     title Xen 3 / NetBSD (hda0, vga)
  163:       root(hd0,0)
  164:       kernel (hd0,a)/xen.gz dom0_mem=65536
  165:       module (hd0,a)/netbsd bootdev=wd0a ro console=tty0
  166: 
  167:     # NetBSD/xen using a backup domain0 kernel (in case you installed a
  168:     # nonworking kernel as /netbsd
  169:     title Xen 3 / NetBSD (hda0, backup, serial)
  170:       root(hd0,0)
  171:       kernel (hd0,a)/xen.gz dom0_mem=65536 com1=115200,8n1
  172:       module (hd0,a)/netbsd.backup bootdev=wd0a ro console=ttyS0
  173:     title Xen 3 / NetBSD (hda0, backup, VGA)
  174:       root(hd0,0)
  175:       kernel (hd0,a)/xen.gz dom0_mem=65536
  176:       module (hd0,a)/netbsd.backup bootdev=wd0a ro console=tty0
  177: 
  178:     #Load a regular NetBSD/i386 kernel. Can be useful if you end up with a
  179:     #nonworking /xen.gz
  180:     title NetBSD 5.1
  181:       root (hd0,a)
  182:       kernel --type=netbsd /netbsd-GENERIC
  183: 
  184:     #Load the NetBSD bootloader, letting it load the NetBSD/i386 kernel.
  185:     #May be better than the above, as grub can't pass all required infos
  186:     #to the NetBSD/i386 kernel (e.g. console, root device, ...)
  187:     title NetBSD chain
  188:       root        (hd0,0)
  189:       chainloader +1
  190: 
  191:     ## end of grub config file.
  192:           
  193: 
  194: Install grub with the following command:
  195: 
  196:     # grub --no-floppy
  197: 
  198:     grub> root (hd0,a)
  199:      Filesystem type is ffs, partition type 0xa9
  200: 
  201:     grub> setup (hd0)
  202:      Checking if "/boot/grub/stage1" exists... no
  203:      Checking if "/grub/stage1" exists... yes
  204:      Checking if "/grub/stage2" exists... yes
  205:      Checking if "/grub/ffs_stage1_5" exists... yes
  206:      Running "embed /grub/ffs_stage1_5 (hd0)"...  14 sectors are embedded.
  207:     succeeded
  208:      Running "install /grub/stage1 (hd0) (hd0)1+14 p (hd0,0,a)/grub/stage2 /grub/menu.lst"...
  209:      succeeded
  210:     Done.
  211:           
  212: 
  213: Creating unprivileged domains (domU)
  214: ====================================
  215: 
  216: Creating domUs is almost entirely independent of operating system.  We
  217: first explain NetBSD, and then differences for Linux and Solaris.
  218: 
  219: Creating an unprivileged NetBSD domain (domU)
  220: ---------------------------------------------
  221: 
  222: Once you have *domain0* running, you need to start the xen tool daemon
  223: (`/usr/pkg/share/examples/rc.d/xend start`) and the xen backend daemon
  224: (`/usr/pkg/share/examples/rc.d/xenbackendd start` for Xen3\*,
  225: `/usr/pkg/share/examples/rc.d/xencommons start` for Xen4.\*). Make sure
  226: that `/dev/xencons` and `/dev/xenevt` exist before starting `xend`. You
  227: can create them with this command:
  228: 
  229:     # cd /dev && sh MAKEDEV xen
  230: 
  231: xend will write logs to `/var/log/xend.log` and
  232: `/var/log/xend-debug.log`. You can then control xen with the xm tool.
  233: 'xm list' will show something like:
  234: 
  235:     # xm list
  236:     Name              Id  Mem(MB)  CPU  State  Time(s)  Console
  237:     Domain-0           0       64    0  r----     58.1
  238: 
  239: 'xm create' allows you to create a new domain. It uses a config file in
  240: PKG\_SYSCONFDIR for its parameters. By default, this file will be in
  241: `/usr/pkg/etc/xen/`. On creation, a kernel has to be specified, which
  242: will be executed in the new domain (this kernel is in the *domain0* file
  243: system, not on the new domain virtual disk; but please note, you should
  244: install the same kernel into *domainU* as `/netbsd` in order to make
  245: your system tools, like MAN.SAVECORE.8, work). A suitable kernel is
  246: provided as part of the i386 and amd64 binary sets: XEN3\_DOMU.
  247: 
  248: Here is an /usr/pkg/etc/xen/nbsd example config file:
  249: 
  250:     #  -*- mode: python; -*-
  251:     #============================================================================
  252:     # Python defaults setup for 'xm create'.
  253:     # Edit this file to reflect the configuration of your system.
  254:     #============================================================================
  255: 
  256:     #----------------------------------------------------------------------------
  257:     # Kernel image file. This kernel will be loaded in the new domain.
  258:     kernel = "/home/bouyer/netbsd-XEN3_DOMU"
  259:     #kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
  260: 
  261:     # Memory allocation (in megabytes) for the new domain.
  262:     memory = 128
  263: 
  264:     # A handy name for your new domain. This will appear in 'xm list',
  265:     # and you can use this as parameters for xm in place of the domain
  266:     # number. All domains must have different names.
  267:     #
  268:     name = "nbsd"
  269: 
  270:     # The number of virtual CPUs this domain has.
  271:     #
  272:     vcpus = 1
  273: 
  274:     #----------------------------------------------------------------------------
  275:     # Define network interfaces for the new domain.
  276: 
  277:     # Number of network interfaces (must be at least 1). Default is 1.
  278:     nics = 1
  279: 
  280:     # Define MAC and/or bridge for the network interfaces.
  281:     #
  282:     # The MAC address specified in ``mac'' is the one used for the interface
  283:     # in the new domain. The interface in domain0 will use this address XOR'd
  284:     # with 00:00:00:01:00:00 (i.e. aa:00:00:51:02:f0 in our example). Random
  285:     # MACs are assigned if not given.
  286:     #
  287:     # ``bridge'' is a required parameter, which will be passed to the
  288:     # vif-script called by xend(8) when a new domain is created to configure
  289:     # the new xvif interface in domain0.
  290:     #
  291:     # In this example, the xvif is added to bridge0, which should have been
  292:     # set up prior to the new domain being created -- either in the
  293:     # ``network'' script or using a /etc/ifconfig.bridge0 file.
  294:     #
  295:     vif = [ 'mac=aa:00:00:50:02:f0, bridge=bridge0' ]
  296: 
  297:     #----------------------------------------------------------------------------
  298:     # Define the disk devices you want the domain to have access to, and
  299:     # what you want them accessible as.
  300:     #
  301:     # Each disk entry is of the form:
  302:     #
  303:     #   phy:DEV,VDEV,MODE
  304:     #
  305:     # where DEV is the device, VDEV is the device name the domain will see,
  306:     # and MODE is r for read-only, w for read-write.  You can also create
  307:     # file-backed domains using disk entries of the form:
  308:     #
  309:     #   file:PATH,VDEV,MODE
  310:     #
  311:     # where PATH is the path to the file used as the virtual disk, and VDEV
  312:     # and MODE have the same meaning as for ``phy'' devices.
  313:     #
  314:     # VDEV doesn't really matter for a NetBSD guest OS (it's just used as an index),
  315:     # but it does for Linux.
  316:     # Worse, the device has to exist in /dev/ of domain0, because xm will
  317:     # try to stat() it. This means that in order to load a Linux guest OS
  318:     # from a NetBSD domain0, you'll have to create /dev/hda1, /dev/hda2, ...
  319:     # on domain0, with the major/minor from Linux :(
  320:     # Alternatively it's possible to specify the device number in hex,
  321:     # e.g. 0x301 for /dev/hda1, 0x302 for /dev/hda2, etc ...
  322: 
  323:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
  324:     #disk = [ 'file:/var/xen/nbsd-disk,0x01,w' ]
  325:     #disk = [ 'file:/var/xen/nbsd-disk,0x301,w' ]
  326: 
  327:     #----------------------------------------------------------------------------
  328:     # Set the kernel command line for the new domain.
  329: 
  330:     # Set root device. This one does matter for NetBSD
  331:     root = "xbd0"
  332:     # extra parameters passed to the kernel
  333:     # this is where you can set boot flags like -s, -a, etc ...
  334:     #extra = ""
  335: 
  336:     #----------------------------------------------------------------------------
  337:     # Set according to whether you want the domain restarted when it exits.
  338:     # The default is False.
  339:     #autorestart = True
  340: 
  341:     # end of nbsd config file ====================================================
  342: 
  343: When a new domain is created, xen calls the
  344: `/usr/pkg/etc/xen/vif-bridge` script for each virtual network interface
  345: created in *domain0*. This can be used to automatically configure the
  346: xvif?.? interfaces in *domain0*. In our example, these will be bridged
  347: with the bridge0 device in *domain0*, but the bridge has to exist first.
  348: To do this, create the file `/etc/ifconfig.bridge0` and make it look
  349: like this:
  350: 
  351:     create
  352:     !brconfig $int add ex0 up
  353: 
  354: (replace `ex0` with the name of your physical interface). Then bridge0
  355: will be created on boot. See the MAN.BRIDGE.4 man page for details.
  356: 
  357: So, here is a suitable `/usr/pkg/etc/xen/vif-bridge` for xvif?.? (a
  358: working vif-bridge is also provided with xentools20) configuring:
  359: 
  360:     #!/bin/sh
  361:     #============================================================================
  362:     # $NetBSD: howto.mdwn,v 1.14 2014/12/23 23:43:27 gdt Exp $
  363:     #
  364:     # /usr/pkg/etc/xen/vif-bridge
  365:     #
  366:     # Script for configuring a vif in bridged mode with a dom0 interface.
  367:     # The xend(8) daemon calls a vif script when bringing a vif up or down.
  368:     # The script name to use is defined in /usr/pkg/etc/xen/xend-config.sxp
  369:     # in the ``vif-script'' field.
  370:     #
  371:     # Usage: vif-bridge up|down [var=value ...]
  372:     #
  373:     # Actions:
  374:     #    up     Adds the vif interface to the bridge.
  375:     #    down   Removes the vif interface from the bridge.
  376:     #
  377:     # Variables:
  378:     #    domain name of the domain the interface is on (required).
  379:     #    vifq   vif interface name (required).
  380:     #    mac    vif MAC address (required).
  381:     #    bridge bridge to add the vif to (required).
  382:     #
  383:     # Example invocation:
  384:     #
  385:     # vif-bridge up domain=VM1 vif=xvif1.0 mac="ee:14:01:d0:ec:af" bridge=bridge0
  386:     #
  387:     #============================================================================
  388: 
  389:     # Exit if anything goes wrong
  390:     set -e
  391: 
  392:     echo "vif-bridge $*"
  393: 
  394:     # Operation name.
  395:     OP=$1; shift
  396: 
  397:     # Pull variables in args into environment
  398:     for arg ; do export "${arg}" ; done
  399: 
  400:     # Required parameters. Fail if not set.
  401:     domain=${domain:?}
  402:     vif=${vif:?}
  403:     mac=${mac:?}
  404:     bridge=${bridge:?}
  405: 
  406:     # Optional parameters. Set defaults.
  407:     ip=${ip:-''}   # default to null (do nothing)
  408: 
  409:     # Are we going up or down?
  410:     case $OP in
  411:     up) brcmd='add' ;;
  412:     down)   brcmd='delete' ;;
  413:     *)
  414:         echo 'Invalid command: ' $OP
  415:         echo 'Valid commands are: up, down'
  416:         exit 1
  417:         ;;
  418:     esac
  419: 
  420:     # Don't do anything if the bridge is "null".
  421:     if [ "${bridge}" = "null" ] ; then
  422:         exit
  423:     fi
  424: 
  425:     # Don't do anything if the bridge doesn't exist.
  426:     if ! ifconfig -l | grep "${bridge}" >/dev/null; then
  427:         exit
  428:     fi
  429: 
  430:     # Add/remove vif to/from bridge.
  431:     ifconfig x${vif} $OP
  432:     brconfig ${bridge} ${brcmd} x${vif}
  433: 
  434: Now, running
  435: 
  436:     xm create -c /usr/pkg/etc/xen/nbsd
  437: 
  438: should create a domain and load a NetBSD kernel in it. (Note: `-c`
  439: causes xm to connect to the domain's console once created.) The kernel
  440: will try to find its root file system on xbd0 (i.e., wd0e) which hasn't
  441: been created yet. wd0e will be seen as a disk device in the new domain,
  442: so it will be 'sub-partitioned'. We could attach a ccd to wd0e in
  443: *domain0* and partition it, newfs and extract the NetBSD/i386 or amd64
  444: tarballs there, but there's an easier way: load the
  445: `netbsd-INSTALL_XEN3_DOMU` kernel provided in the NetBSD binary sets.
  446: Like other install kernels, it contains a ramdisk with sysinst, so you
  447: can install NetBSD using sysinst on your new domain.
  448: 
  449: If you want to install NetBSD/Xen with a CDROM image, the following line
  450: should be used in the `/usr/pkg/etc/xen/nbsd` file:
  451: 
  452:     disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
  453: 
  454: After booting the domain, the option to install via CDROM may be
  455: selected. The CDROM device should be changed to `xbd1d`.
  456: 
  457: Once done installing, `halt -p` the new domain (don't reboot or halt, it
  458: would reload the INSTALL\_XEN3\_DOMU kernel even if you changed the
  459: config file), switch the config file back to the XEN3\_DOMU kernel, and
  460: start the new domain again. Now it should be able to use `root on xbd0a`
  461: and you should have a second, functional NetBSD system on your xen
  462: installation.
  463: 
  464: When the new domain is booting you'll see some warnings about *wscons*
  465: and the pseudo-terminals. These can be fixed by editing the files
  466: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
  467: `/etc/ttys`, except *console*, like this:
  468: 
  469:     console "/usr/libexec/getty Pc"         vt100   on secure
  470:     ttyE0   "/usr/libexec/getty Pc"         vt220   off secure
  471:     ttyE1   "/usr/libexec/getty Pc"         vt220   off secure
  472:     ttyE2   "/usr/libexec/getty Pc"         vt220   off secure
  473:     ttyE3   "/usr/libexec/getty Pc"         vt220   off secure
  474: 
  475: Finally, all screens must be commented out from `/etc/wscons.conf`.
  476: 
  477: It is also desirable to add
  478: 
  479:     powerd=YES
  480: 
  481: in rc.conf. This way, the domain will be properly shut down if
  482: `xm shutdown -R` or `xm shutdown -H` is used on the domain0.
  483: 
  484: Your domain should be now ready to work, enjoy.
  485: 
  486: Creating an unprivileged Linux domain (domU)
  487: --------------------------------------------
  488: 
  489: Creating unprivileged Linux domains isn't much different from
  490: unprivileged NetBSD domains, but there are some details to know.
  491: 
  492: First, the second parameter passed to the disk declaration (the '0x1' in
  493: the example below)
  494: 
  495:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
  496: 
  497: does matter to Linux. It wants a Linux device number here (e.g. 0x300
  498: for hda). Linux builds device numbers as: (major \<\< 8 + minor). So,
  499: hda1 which has major 3 and minor 1 on a Linux system will have device
  500: number 0x301. Alternatively, devices names can be used (hda, hdb, ...)
  501: as xentools has a table to map these names to devices numbers. To export
  502: a partition to a Linux guest we can use:
  503: 
  504:     disk = [ 'phy:/dev/wd0e,0x300,w' ]
  505:     root = "/dev/hda1 ro"
  506: 
  507: and it will appear as /dev/hda on the Linux system, and be used as root
  508: partition.
  509: 
  510: To install the Linux system on the partition to be exported to the guest
  511: domain, the following method can be used: install sysutils/e2fsprogs
  512: from pkgsrc. Use mke2fs to format the partition that will be the root
  513: partition of your Linux domain, and mount it. Then copy the files from a
  514: working Linux system, make adjustments in `/etc` (fstab, network
  515: config). It should also be possible to extract binary packages such as
  516: .rpm or .deb directly to the mounted partition using the appropriate
  517: tool, possibly running under NetBSD's Linux emulation. Once the
  518: filesystem has been populated, umount it. If desirable, the filesystem
  519: can be converted to ext3 using tune2fs -j. It should now be possible to
  520: boot the Linux guest domain, using one of the vmlinuz-\*-xenU kernels
  521: available in the Xen binary distribution.
  522: 
  523: To get the linux console right, you need to add:
  524: 
  525:     extra = "xencons=tty1"
  526: 
  527: to your configuration since not all linux distributions auto-attach a
  528: tty to the xen console.
  529: 
  530: Creating an unprivileged Solaris domain (domU)
  531: ----------------------------------------------
  532: 
  533: Download an Opensolaris [release](http://opensolaris.org/os/downloads/)
  534: or [development snapshot](http://genunix.org/) DVD image. Attach the DVD
  535: image to a MAN.VND.4 device. Copy the kernel and ramdisk filesystem
  536: image to your dom0 filesystem.
  537: 
  538:     dom0# mkdir /root/solaris
  539:     dom0# vnconfig vnd0 osol-1002-124-x86.iso
  540:     dom0# mount /dev/vnd0a /mnt
  541: 
  542:     ## for a 64-bit guest
  543:     dom0# cp /mnt/boot/amd64/x86.microroot /root/solaris
  544:     dom0# cp /mnt/platform/i86xpv/kernel/amd64/unix /root/solaris
  545: 
  546:     ## for a 32-bit guest
  547:     dom0# cp /mnt/boot/x86.microroot /root/solaris
  548:     dom0# cp /mnt/platform/i86xpv/kernel/unix /root/solaris
  549: 
  550:     dom0# umount /mnt
  551:           
  552: 
  553: Keep the MAN.VND.4 configured. For some reason the boot process stalls
  554: unless the DVD image is attached to the guest as a "phy" device. Create
  555: an initial configuration file with the following contents. Substitute
  556: */dev/wd0k* with an empty partition at least 8 GB large.
  557: 
  558:     memory = 640
  559:     name = 'solaris'
  560:     disk = [ 'phy:/dev/wd0k,0,w' ]
  561:     disk += [ 'phy:/dev/vnd0d,6:cdrom,r' ]
  562:     vif = [ 'bridge=bridge0' ]
  563:     kernel = '/root/solaris/unix'
  564:     ramdisk = '/root/solaris/x86.microroot'
  565:     # for a 64-bit guest
  566:     extra = '/platform/i86xpv/kernel/amd64/unix - nowin -B install_media=cdrom'
  567:     # for a 32-bit guest
  568:     #extra = '/platform/i86xpv/kernel/unix - nowin -B install_media=cdrom'
  569:           
  570: 
  571: Start the guest.
  572: 
  573:     dom0# xm create -c solaris.cfg
  574:     Started domain solaris
  575:                           v3.3.2 chgset 'unavailable'
  576:     SunOS Release 5.11 Version snv_124 64-bit
  577:     Copyright 1983-2009 Sun Microsystems, Inc.  All rights reserved.
  578:     Use is subject to license terms.
  579:     Hostname: opensolaris
  580:     Remounting root read/write
  581:     Probing for device nodes ...
  582:     WARNING: emlxs: ddi_modopen drv/fct failed: err 2
  583:     Preparing live image for use
  584:     Done mounting Live image
  585:           
  586: 
  587: Make sure the network is configured. Note that it can take a minute for
  588: the xnf0 interface to appear.
  589: 
  590:     opensolaris console login: jack
  591:     Password: jack
  592:     Sun Microsystems Inc.   SunOS 5.11      snv_124 November 2008
  593:     jack@opensolaris:~$ pfexec sh
  594:     sh-3.2# ifconfig -a
  595:     sh-3.2# exit
  596:           
  597: 
  598: Set a password for VNC and start the VNC server which provides the X11
  599: display where the installation program runs.
  600: 
  601:     jack@opensolaris:~$ vncpasswd
  602:     Password: solaris
  603:     Verify: solaris
  604:     jack@opensolaris:~$ cp .Xclients .vnc/xstartup
  605:     jack@opensolaris:~$ vncserver :1
  606:           
  607: 
  608: From a remote machine connect to the VNC server. Use `ifconfig xnf0` on
  609: the guest to find the correct IP address to use.
  610: 
  611:     remote$ vncviewer 172.18.2.99:1
  612:           
  613: 
  614: It is also possible to launch the installation on a remote X11 display.
  615: 
  616:     jack@opensolaris:~$ export DISPLAY=172.18.1.1:0
  617:     jack@opensolaris:~$ pfexec gui-install
  618:            
  619: 
  620: After the GUI installation is complete you will be asked to reboot.
  621: Before that you need to determine the ZFS ID for the new boot filesystem
  622: and update the configuration file accordingly. Return to the guest
  623: console.
  624: 
  625:     jack@opensolaris:~$ pfexec zdb -vvv rpool | grep bootfs
  626:                     bootfs = 43
  627:     ^C
  628:     jack@opensolaris:~$
  629:            
  630: 
  631: The final configuration file should look like this. Note in particular
  632: the last line.
  633: 
  634:     memory = 640
  635:     name = 'solaris'
  636:     disk = [ 'phy:/dev/wd0k,0,w' ]
  637:     vif = [ 'bridge=bridge0' ]
  638:     kernel = '/root/solaris/unix'
  639:     ramdisk = '/root/solaris/x86.microroot'
  640:     extra = '/platform/i86xpv/kernel/amd64/unix -B zfs-bootfs=rpool/43,bootpath="/xpvd/xdf@0:a"'
  641:            
  642: 
  643: Restart the guest to verify it works correctly.
  644: 
  645:     dom0# xm destroy solaris
  646:     dom0# xm create -c solaris.cfg
  647:     Using config file "./solaris.cfg".
  648:     v3.3.2 chgset 'unavailable'
  649:     Started domain solaris
  650:     SunOS Release 5.11 Version snv_124 64-bit
  651:     Copyright 1983-2009 Sun Microsystems, Inc.  All rights reserved.
  652:     Use is subject to license terms.
  653:     WARNING: emlxs: ddi_modopen drv/fct failed: err 2
  654:     Hostname: osol
  655:     Configuring devices.
  656:     Loading smf(5) service descriptions: 160/160
  657:     svccfg import warnings. See /var/svc/log/system-manifest-import:default.log .
  658:     Reading ZFS config: done.
  659:     Mounting ZFS filesystems: (6/6)
  660:     Creating new rsa public/private host key pair
  661:     Creating new dsa public/private host key pair
  662: 
  663:     osol console login:
  664:            
  665: 
  666: Using PCI devices in guest domains
  667: ----------------------------------
  668: 
  669: The domain0 can give other domains access to selected PCI devices. This
  670: can allow, for example, a non-privileged domain to have access to a
  671: physical network interface or disk controller. However, keep in mind
  672: that giving a domain access to a PCI device most likely will give the
  673: domain read/write access to the whole physical memory, as PCs don't have
  674: an IOMMU to restrict memory access to DMA-capable device. Also, it's not
  675: possible to export ISA devices to non-domain0 domains (which means that
  676: the primary VGA adapter can't be exported. A guest domain trying to
  677: access the VGA registers will panic).
  678: 
  679: This functionality is only available in NetBSD-5.1 (and later) domain0
  680: and domU. If the domain0 is NetBSD, it has to be running Xen 3.1, as
  681: support has not been ported to later versions at this time.
  682: 
  683: For a PCI device to be exported to a domU, is has to be attached to the
  684: `pciback` driver in domain0. Devices passed to the domain0 via the
  685: pciback.hide boot parameter will attach to `pciback` instead of the
  686: usual driver. The list of devices is specified as `(bus:dev.func)`,
  687: where bus and dev are 2-digit hexadecimal numbers, and func a
  688: single-digit number:
  689: 
  690:     pciback.hide=(00:0a.0)(00:06.0)
  691: 
  692: pciback devices should show up in the domain0's boot messages, and the
  693: devices should be listed in the `/kern/xen/pci` directory.
  694: 
  695: PCI devices to be exported to a domU are listed in the `pci` array of
  696: the domU's config file, with the format `'0000:bus:dev.func'`
  697: 
  698:     pci = [ '0000:00:06.0', '0000:00:0a.0' ]
  699: 
  700: In the domU an `xpci` device will show up, to which one or more pci
  701: busses will attach. Then the PCI drivers will attach to PCI busses as
  702: usual. Note that the default NetBSD DOMU kernels do not have `xpci` or
  703: any PCI drivers built in by default; you have to build your own kernel
  704: to use PCI devices in a domU. Here's a kernel config example:
  705: 
  706:     include         "arch/i386/conf/XEN3_DOMU"
  707:     #include         "arch/i386/conf/XENU"           # in NetBSD 3.0
  708: 
  709:     # Add support for PCI busses to the XEN3_DOMU kernel
  710:     xpci* at xenbus ?
  711:     pci* at xpci ?
  712: 
  713:     # Now add PCI and related devices to be used by this domain
  714:     # USB Controller and Devices
  715: 
  716:     # PCI USB controllers
  717:     uhci*   at pci? dev ? function ?        # Universal Host Controller (Intel)
  718: 
  719:     # USB bus support
  720:     usb*    at uhci?
  721: 
  722:     # USB Hubs
  723:     uhub*   at usb?
  724:     uhub*   at uhub? port ? configuration ? interface ?
  725: 
  726:     # USB Mass Storage
  727:     umass*  at uhub? port ? configuration ? interface ?
  728:     wd*     at umass?
  729:     # SCSI controllers
  730:     ahc*    at pci? dev ? function ?        # Adaptec [23]94x, aic78x0 SCSI
  731: 
  732:     # SCSI bus support (for both ahc and umass)
  733:     scsibus* at scsi?
  734: 
  735:     # SCSI devices
  736:     sd*     at scsibus? target ? lun ?      # SCSI disk drives
  737:     cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
  738: 
  739: Links and further information
  740: =============================
  741: 
  742: -   The [HowTo on Installing into RAID-1](http://mail-index.NetBSD.org/port-xen/2006/03/01/0010.html)
  743:     explains how to set up booting a dom0 with Xen using grub 
  744:     with NetBSD's RAIDframe.  (This is obsolete with the use of
  745:     NetBSD's native boot.)
  746: -   An example of how to use NetBSD's native bootloader to load
  747:     NetBSD/Xen instead of Grub can be found in the i386/amd64 boot(8)
  748:     and boot.cfg(5) manpages.

CVSweb for NetBSD wikisrc <wikimaster@NetBSD.org> software: FreeBSD-CVSweb