Annotation of wikisrc/ports/xen/howto.mdwn, revision 1.42

1.5       mspo        1: Introduction
1.13      gdt         2: ============
1.1       mspo        3: 
                      4: [![[Xen
1.7       mspo        5: screenshot]](http://www.netbsd.org/gallery/in-Action/hubertf-xens.png)](../../gallery/in-Action/hubertf-xen.png)
1.1       mspo        6: 
1.12      gdt         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: 
1.29      gdt        30: Generally any amd64 machine will work with Xen and PV guests.  In
                     31: theory i386 computers without amd64 support can be used for Xen <=
                     32: 4.2, but we have no recent reports of this working (this is a hint).
                     33: For HVM guests, the VT or VMX cpu feature (Intel) or SVM/HVM/VT
                     34: (amd64) is needed; "cpuctl identify 0" will show this.  TODO: Clean up
                     35: and check the above features.
1.19      gdt        36: 
1.27      jnemeth    37: At boot, the dom0 kernel is loaded as a module with Xen as the kernel.
1.12      gdt        38: The dom0 can start one or more domUs.  (Booting is explained in detail
                     39: in the dom0 section.)
                     40: 
                     41: NetBSD supports Xen in that it can serve as dom0, be used as a domU,
                     42: and that Xen kernels and tools are available in pkgsrc.  This HOWTO
                     43: attempts to address both the case of running a NetBSD dom0 on hardware
1.24      gdt        44: and running domUs under it (NetBSD and other), and also running NetBSD
                     45: as a domU in a VPS.
1.12      gdt        46: 
1.20      gdt        47: Some versions of Xen support "PCI passthrough", which means that
                     48: specific PCI devices can be made available to a specific domU instead
                     49: of the dom0.  This can be useful to let a domU run X11, or access some
                     50: network interface or other peripheral.
                     51: 
1.12      gdt        52: Prerequisites
1.13      gdt        53: -------------
1.12      gdt        54: 
                     55: Installing NetBSD/Xen is not extremely difficult, but it is more
                     56: complex than a normal installation of NetBSD.
1.15      gdt        57: In general, this HOWTO is occasionally overly restrictive about how
                     58: things must be done, guiding the reader to stay on the established
                     59: path when there are no known good reasons to stray.
1.12      gdt        60: 
                     61: This HOWTO presumes a basic familiarity with the Xen system
1.16      gdt        62: architecture.  This HOWTO presumes familiarity with installing NetBSD
                     63: on i386/amd64 hardware and installing software from pkgsrc.
1.27      jnemeth    64: See also the [Xen website](http://www.xenproject.org/).
1.1       mspo       65: 
1.19      gdt        66: History
                     67: -------
                     68: 
                     69: NetBSD used to support Xen2; this has been removed.
                     70: 
                     71: Before NetBSD's native bootloader could support Xen, the use of
                     72: grub was recommended.  If necessary, see the
1.27      jnemeth    73: [old grub information](/ports/xen/howto-grub/).
1.19      gdt        74: 
1.15      gdt        75: Versions of Xen and NetBSD
                     76: ==========================
                     77: 
1.27      jnemeth    78: Most of the installation concepts and instructions are independent
                     79: of Xen version and NetBSD version.  This section gives advice on
                     80: which version to choose.  Versions not in pkgsrc and older unsupported
                     81: versions of NetBSD are intentionally ignored.
1.15      gdt        82: 
                     83: Xen
                     84: ---
                     85: 
                     86: In NetBSD, xen is provided in pkgsrc, via matching pairs of packages
                     87: xenkernel and xentools.  We will refer only to the kernel versions,
                     88: but note that both packages must be installed together and must have
                     89: matching versions.
                     90: 
                     91: xenkernel3 and xenkernel33 provide Xen 3.1 and 3.3.  These no longer
1.20      gdt        92: receive security patches and should not be used.  Xen 3.1 supports PCI
1.29      gdt        93: passthrough.  Xen 3.1 supports non-PAE on i386.
1.15      gdt        94: 
                     95: xenkernel41 provides Xen 4.1.  This is no longer maintained by Xen,
                     96: but as of 2014-12 receives backported security patches.  It is a
                     97: reasonable although trailing-edge choice.
                     98: 
                     99: xenkernel42 provides Xen 4.2.  This is maintained by Xen, but old as
                    100: of 2014-12.
                    101: 
                    102: Ideally newer versions of Xen will be added to pkgsrc.
                    103: 
1.26      gdt       104: Note that NetBSD support is called XEN3.  It works with 3.1 through
                    105: 4.2 because the hypercall interface has been stable.
1.20      gdt       106: 
1.19      gdt       107: Xen command program
                    108: -------------------
                    109: 
                    110: Early Xen used a program called "xm" to manipulate the system from the
                    111: dom0.  Starting in 4.1, a replacement program with similar behavior
1.27      jnemeth   112: called "xl" is provided.  In 4.2 and later, "xl" is preferred.  4.4 is
                    113: the last version that has "xm".
1.19      gdt       114: 
1.15      gdt       115: NetBSD
                    116: ------
                    117: 
                    118: The netbsd-5, netbsd-6, netbsd-7, and -current branches are all
                    119: reasonable choices, with more or less the same considerations for
                    120: non-Xen use.  Therefore, netbsd-6 is recommended as the stable version
1.29      gdt       121: of the most recent release for production use.  For those wanting to
                    122: learn Xen or without production stability concerns, netbsd-7 is likely
                    123: most appropriate.
1.15      gdt       124: 
                    125: As of NetBSD 6, a NetBSD domU will support multiple vcpus.  There is
                    126: no SMP support for NetBSD as dom0.  (The dom0 itself doesn't really
                    127: need SMP; the lack of support is really a problem when using a dom0 as
                    128: a normal computer.)
                    129: 
1.18      gdt       130: Architecture
                    131: ------------
                    132: 
1.29      gdt       133: Xen itself can run on i386 or amd64 machines.  (Practically, almost
                    134: any computer where one would want to run Xen supports amd64.)  If
                    135: using an i386 NetBSD kernel for the dom0, PAE is required (PAE
                    136: versions are built by default).  While i386 dom0 works fine, amd64 is
                    137: recommended as more normal.
                    138: 
                    139: Xen 4.2 is the last version to support i386 as a host.  TODO: Clarify
                    140: if this is about the CPU having to be amd64, or about the dom0 kernel
                    141: having to be amd64.
                    142: 
                    143: One can then run i386 domUs and amd64 domUs, in any combination.  If
                    144: running an i386 NetBSD kernel as a domU, the PAE version is required.
                    145: (Note that emacs (at least) fails if run on i386 with PAE when built
                    146: without, and vice versa, presumably due to bugs in the undump code.)
1.18      gdt       147: 
1.15      gdt       148: Recommendation
                    149: --------------
                    150: 
1.18      gdt       151: Therefore, this HOWTO recommends running xenkernel42 (and xentools42),
1.30      gdt       152: xl, the NetBSD 6 stable branch, and to use an amd64 kernel as the
                    153: dom0.  Either the i386 or amd64 of NetBSD may be used as domUs.
1.15      gdt       154: 
1.36      gdt       155: Build problems
                    156: --------------
                    157: 
                    158: Ideally, all versions of Xen in pkgsrc would build on all versions of
                    159: NetBSD on both i386 and amd64.  However, that isn't the case.  Besides
                    160: aging code and aging compilers, qemu (included in xentools for HVM
                    161: support) is difficult to build.  The following are known to fail:
                    162: 
                    163:         xenkernel3 netbsd-6 i386
                    164:         xentools42 netbsd-6 i386 
                    165: 
                    166: The following are known to work:
                    167: 
                    168:         xenkernel41 netbsd-5 amd64
                    169:         xentools41 netbsd-5 amd64
                    170:         xenkernel41 netbsd-6 i386
                    171:         xentools41 netbsd-6 i386
                    172: 
1.15      gdt       173: NetBSD as a dom0
                    174: ================
                    175: 
                    176: NetBSD can be used as a dom0 and works very well.  The following
                    177: sections address installation, updating NetBSD, and updating Xen.
1.19      gdt       178: Note that it doesn't make sense to talk about installing a dom0 OS
                    179: without also installing Xen itself.  We first address installing
                    180: NetBSD, which is not yet a dom0, and then adding Xen, pivoting the
                    181: NetBSD install to a dom0 install by just changing the kernel and boot
                    182: configuration.
1.15      gdt       183: 
                    184: Styles of dom0 operation
                    185: ------------------------
                    186: 
                    187: There are two basic ways to use Xen.  The traditional method is for
                    188: the dom0 to do absolutely nothing other than providing support to some
                    189: number of domUs.  Such a system was probably installed for the sole
                    190: purpose of hosting domUs, and sits in a server room on a UPS.
                    191: 
                    192: The other way is to put Xen under a normal-usage computer, so that the
                    193: dom0 is what the computer would have been without Xen, perhaps a
                    194: desktop or laptop.  Then, one can run domUs at will.  Purists will
                    195: deride this as less secure than the previous approach, and for a
                    196: computer whose purpose is to run domUs, they are right.  But Xen and a
                    197: dom0 (without domUs) is not meaingfully less secure than the same
                    198: things running without Xen.  One can boot Xen or boot regular NetBSD
                    199: alternately with little problems, simply refraining from starting the
                    200: Xen daemons when not running Xen.
                    201: 
                    202: Note that NetBSD as dom0 does not support multiple CPUs.  This will
                    203: limit the performance of the Xen/dom0 workstation approach.
                    204: 
1.19      gdt       205: Installation of NetBSD
                    206: ----------------------
1.13      gdt       207: 
1.19      gdt       208: First,
1.27      jnemeth   209: [install NetBSD/amd64](/guide/inst/)
1.19      gdt       210: just as you would if you were not using Xen.
                    211: However, the partitioning approach is very important.
                    212: 
                    213: If you want to use RAIDframe for the dom0, there are no special issues
                    214: for Xen.  Typically one provides RAID storage for the dom0, and the
1.22      gdt       215: domU systems are unaware of RAID.  The 2nd-stage loader bootxx_* skips
                    216: over a RAID1 header to find /boot from a filesystem within a RAID
                    217: partition; this is no different when booting Xen.
1.19      gdt       218: 
                    219: There are 4 styles of providing backing storage for the virtual disks
                    220: used by domUs: raw partitions, LVM, file-backed vnd(4), and SAN,
                    221: 
                    222: With raw partitions, one has a disklabel (or gpt) partition sized for
                    223: each virtual disk to be used by the domU.  (If you are able to predict
                    224: how domU usage will evolve, please add an explanation to the HOWTO.
                    225: Seriously, needs tend to change over time.)
                    226: 
1.27      jnemeth   227: One can use [lvm(8)](/guide/lvm/) to create logical devices to use
                    228: for domU disks.  This is almost as efficient as raw disk partitions
                    229: and more flexible.  Hence raw disk partitions should typically not
                    230: be used.
1.19      gdt       231: 
                    232: One can use files in the dom0 filesystem, typically created by dd'ing
                    233: /dev/zero to create a specific size.  This is somewhat less efficient,
                    234: but very convenient, as one can cp the files for backup, or move them
                    235: between dom0 hosts.
                    236: 
                    237: Finally, in theory one can place the files backing the domU disks in a
                    238: SAN.  (This is an invitation for someone who has done this to add a
                    239: HOWTO page.)
1.1       mspo      240: 
1.19      gdt       241: Installation of Xen
                    242: -------------------
1.1       mspo      243: 
1.20      gdt       244: In the dom0, install sysutils/xenkernel42 and sysutils/xentools42 from
                    245: pkgsrc (or another matching pair).
                    246: See [the pkgsrc
                    247: documentation](http://www.NetBSD.org/docs/pkgsrc/) for help with pkgsrc.
                    248: 
                    249: For Xen 3.1, support for HVM guests is in sysutils/xentool3-hvm.  More
                    250: recent versions have HVM support integrated in the main xentools
                    251: package.  It is entirely reasonable to run only PV guests.
                    252: 
                    253: Next you need to install the selected Xen kernel itself, which is
                    254: installed by pkgsrc as "/usr/pkg/xen*-kernel/xen.gz".  Copy it to /.
                    255: For debugging, one may copy xen-debug.gz; this is conceptually similar
                    256: to DIAGNOSTIC and DEBUG in NetBSD.  xen-debug.gz is basically only
                    257: useful with a serial console.  Then, place a NetBSD XEN3_DOM0 kernel
                    258: in /, copied from releasedir/amd64/binary/kernel/netbsd-XEN3_DOM0.gz
                    259: of a NetBSD build.  Both xen and NetBSD may be left compressed.  (If
                    260: using i386, use releasedir/i386/binary/kernel/netbsd-XEN3PAE_DOM0.gz.)
                    261: 
                    262: In a dom0 kernel, kernfs is mandatory for xend to comunicate with the
                    263: kernel, so ensure that /kern is in fstab.
                    264: 
                    265: Because you already installed NetBSD, you have a working boot setup
                    266: with an MBR bootblock, either bootxx_ffsv1 or bootxx_ffsv2 at the
                    267: beginning of your root filesystem, /boot present, and likely
                    268: /boot.cfg.  (If not, fix before continuing!)
                    269: 
                    270: See boot.cfg(5) for an example.  The basic line is
                    271: 
1.37      gdt       272:         menu=Xen:load /netbsd-XEN3_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=256M
1.20      gdt       273: 
                    274: which specifies that the dom0 should have 256M, leaving the rest to be
1.37      gdt       275: allocated for domUs.  In an attempt to add performance, one can also
                    276: add
                    277: 
                    278:         dom0_max_vcpus=1 dom0_vcpus_pin
                    279: 
                    280: to force only one vcpu to be provided (since NetBSD dom0 can't use
                    281: more) and to pin that vcpu to a physical cpu.  TODO: benchmark this.
1.20      gdt       282: 
                    283: As with non-Xen systems, you should have a line to boot /netbsd (a
                    284: kernel that works without Xen) and fallback versions of the non-Xen
                    285: kernel, Xen, and the dom0 kernel.
1.1       mspo      286: 
1.28      gdt       287: The [HowTo on Installing into
                    288: RAID-1](http://mail-index.NetBSD.org/port-xen/2006/03/01/0010.html)
                    289: explains how to set up booting a dom0 with Xen using grub with
                    290: NetBSD's RAIDframe.  (This is obsolete with the use of NetBSD's native
                    291: boot.)
                    292: 
1.21      gdt       293: Configuring Xen
                    294: ---------------
                    295: 
                    296: Now, you have a system that will boot Xen and the dom0 kernel, and
                    297: just run the dom0 kernel.  There will be no domUs, and none can be
1.31      gdt       298: started because you still have to configure the dom0 tools.  The
                    299: daemons which should be run vary with Xen version and with whether one
                    300: is using xm or xl.  Note that xend is for supporting "xm", and should
                    301: only be used if you plan on using "xm".  Do NOT enable xend if you
                    302: plan on using "xl" as it will cause problems.
1.21      gdt       303: 
1.31      gdt       304: TODO: Give 3.1 advice (or remove it from pkgsrc).
                    305: 
                    306: For 3.3 (and thus xm), add to rc.conf (but note that you should have
                    307: installed 4.1 or 4.2):
                    308: 
1.32      gdt       309:         xend=YES
                    310:         xenbackendd=YES
1.31      gdt       311: 
1.33      gdt       312: For 4.1 (and thus xm; xl is believed not to work well), add to rc.conf:
1.31      gdt       313: 
                    314:         xend=YES
                    315:         xencommons=YES
                    316: 
                    317: TODO: Explain why if xm is preferred on 4.1, rc.d/xendomains has xl.
1.33      gdt       318: Or fix the package.
1.31      gdt       319: 
1.33      gdt       320: For 4.2 with xm, add to rc.conf
                    321: 
                    322:         xend=YES
                    323:         xencommons=YES
                    324: 
                    325: For 4.2 with xl (preferred), add to rc.conf:
1.31      gdt       326: 
                    327:         TODO: explain if there is a xend replacement
                    328:         xencommons=YES
                    329: 
                    330: TODO: Recommend for/against xen-watchdog.
1.27      jnemeth   331: 
1.42    ! gdt       332: After you have configured the daemons and eitehr started them or
        !           333: rebooted, run the following (or use xl) to inspect Xen's boot
        !           334: messages, available resources, and running domains:
1.34      gdt       335: 
1.33      gdt       336:         xm dmesg
1.35      gdt       337:         xm info
                    338:         xm list
1.33      gdt       339: 
1.41      gdt       340: anita (for testing NetBSD)
                    341: --------------------------
                    342: 
                    343: With the setup so far, one should be able to run anita (see
                    344: pkgsrc/sysutils/py-anita) to test NetBSD releases, by doing (as root,
                    345: because anita must create a domU):
                    346: 
                    347:         anita --vmm=xm test file:///usr/obj/i386/
                    348: 
                    349: Alternatively, one can use --vmm=xl to use xl-based domU creation instead.
                    350: TODO: check this.
                    351: 
1.40      gdt       352: Xen-specific NetBSD issues
                    353: --------------------------
                    354: 
                    355: There are (at least) two additional things different about NetBSD as a
                    356: dom0 kernel compared to hardware.
                    357: 
                    358: One is that modules are not usable in DOM0 kernels, so one must
                    359: compile in what's needed.  It's not really that modules cannot work,
                    360: but that modules must be built for XEN3_DOM0 because some of the
                    361: defines change and the normal module builds don't do this.  Basically,
                    362: enabling Xen changes the kernel ABI, and the module build system
                    363: doesn't cope with this.
                    364: 
                    365: The other difference is that XEN3_DOM0 does not have exactly the same
                    366: options as GENERIC.  While it is debatable whether or not this is a
                    367: bug, users should be aware of this and can simply add missing config
                    368: items if desired.
                    369: 
1.15      gdt       370: Updating NetBSD in a dom0
                    371: -------------------------
                    372: 
                    373: This is just like updating NetBSD on bare hardware, assuming the new
                    374: version supports the version of Xen you are running.  Generally, one
                    375: replaces the kernel and reboots, and then overlays userland binaries
                    376: and adjusts /etc.
                    377: 
                    378: Note that one must update both the non-Xen kernel typically used for
                    379: rescue purposes and the DOM0 kernel used with Xen.
                    380: 
1.22      gdt       381: To convert from grub to /boot, install an mbr bootblock with fdisk,
                    382: bootxx_ with installboot, /boot and /boot.cfg.  This really should be
                    383: no different than completely reinstalling boot blocks on a non-Xen
                    384: system.
                    385: 
1.15      gdt       386: Updating Xen versions
                    387: ---------------------
                    388: 
1.21      gdt       389: Updating Xen is conceptually not difficult, but can run into all the
                    390: issues found when installing Xen.  Assuming migration from 4.1 to 4.2,
                    391: remove the xenkernel41 and xentools41 packages and install the
                    392: xenkernel42 and xentools42 packages.  Copy the 4.2 xen.gz to /.
                    393: 
                    394: Ensure that the contents of /etc/rc.d/xen* are correct.  Enable the
                    395: correct set of daemons.  Ensure that the domU config files are valid
                    396: for the new version.
1.15      gdt       397: 
1.28      gdt       398: 
                    399: Unprivileged domains (domU)
                    400: ===========================
                    401: 
                    402: This section describes general concepts about domUs.  It does not
1.33      gdt       403: address specific domU operating systems or how to install them.  The
                    404: config files for domUs are typically in /usr/pkg/etc/xen, and are
                    405: typically named so that the file anme, domU name and the domU's host
                    406: name match.
                    407: 
                    408: The domU is provided with cpu and memory by Xen, configured by the
                    409: dom0.  The domU is provided with disk and network by the dom0,
                    410: mediated by Xen, and configured in the dom0.
                    411: 
                    412: Entropy in domUs can be an issue; physical disks and network are on
                    413: the dom0.  NetBSD's /dev/random system works, but is often challenged.
                    414: 
                    415: CPU and memory
                    416: --------------
                    417: 
                    418: A domain is provided with some number of vcpus, less than the
                    419: number of cpus seen by the hypervisor.  For a dom0, this is controlled
                    420: by the boot argument "dom0_max_vcpus=1".  For a domU, it is controlled
                    421: from the config file.
                    422: 
                    423: A domain is provided with memory, In the straightforward case, the sum
                    424: of the the memory allocated to the dom0 and all domUs must be less
                    425: than the available memory.
                    426: 
                    427: Xen also provides a "balloon" driver, which can be used to let domains
                    428: use more memory temporarily.  TODO: Explain better, and explain how
                    429: well it works with NetBSD.
1.28      gdt       430: 
                    431: Virtual disks
                    432: -------------
                    433: 
1.33      gdt       434: With the file/vnd style, typically one creates a directory,
                    435: e.g. /u0/xen, on a disk large enough to hold virtual disks for all
                    436: domUs.  Then, for each domU disk, one writes zeros to a file that then
                    437: serves to hold the virtual disk's bits; a suggested name is foo-xbd0
                    438: for the first virtual disk for the domU called foo.  Writing zeros to
                    439: the file serves two purposes.  One is that preallocating the contents
                    440: improves performance.  The other is that vnd on sparse files has
                    441: failed to work.  TODO: give working/notworking NetBSD versions for
                    442: sparse vnd.  Note that the use of file/vnd for Xen is not really
                    443: different than creating a file-backed virtual disk for some other
1.39      gdt       444: purpose, except that xentools handles the vnconfig commands.  To
                    445: create an empty 4G virtual disk, simply do
                    446: 
                    447:         dd if=/dev/zero of=foo-xbd0 bs=1m count=4096
1.33      gdt       448: 
                    449: With the lvm style, one creates logical devices.  They are then used
                    450: similarly to vnds.
1.28      gdt       451: 
                    452: Virtual Networking
                    453: ------------------
                    454: 
                    455: TODO: explain xvif concept, and that it's general.
                    456: 
                    457: There are two normal styles: bridging and NAT.
                    458: 
                    459: With bridging, the domU perceives itself to be on the same network as
                    460: the dom0.  For server virtualization, this is usually best.
                    461: 
                    462: With NAT, the domU perceives itself to be behind a NAT running on the
                    463: dom0.  This is often appropriate when running Xen on a workstation.
                    464: 
                    465: One can construct arbitrary other configurations, but there is no
                    466: script support.
                    467: 
1.33      gdt       468: Sizing domains
                    469: --------------
                    470: 
                    471: Modern x86 hardware has vast amounts of resources.  However, many
                    472: virtual servers can function just fine on far less.  A system with
                    473: 256M of RAM and a 4G disk can be a reasonable choice.  Note that it is
                    474: far easier to adjust virtual resources than physical ones.  For
                    475: memory, it's just a config file edit and a reboot.  For disk, one can
                    476: create a new file and vnconfig it (or lvm), and then dump/restore,
                    477: just like updating physical disks, but without having to be there and
                    478: without those pesky connectors.
                    479: 
1.39      gdt       480: domU kernels
                    481: ------------
                    482: 
                    483: On a physical computer, the BIOS reads sector 0, and a chain of boot
                    484: loaders finds and loads a kernel.  Normally this comes from the root
                    485: filesystem.  With Xen domUs, the process is totally different.  The
                    486: normal path is for the domU kernel to be a file in the dom0's
                    487: filesystem.  At the request of the dom0, Xen loads that kernel into a
                    488: new domU instance and starts execution.  While domU kernels can be
                    489: anyplace, reasonable places to store domU kernels on the dom0 are in /
                    490: (so they are near the dom0 kernel), in /usr/pkg/etc/xen (near the
                    491: config files), or in /u0/xen (where the vdisks are).
                    492: 
                    493: See the VPS section near the end for discussion of alternate ways to
                    494: obtain domU kernels.
                    495: 
1.28      gdt       496: Config files
                    497: ------------
                    498: 
                    499: TODO: give example config files.   Use both lvm and vnd.
                    500: 
                    501: TODO: explain the mess with 3 arguments for disks and how to cope (0x1).
                    502: 
                    503: Starting domains
                    504: ----------------
                    505: 
                    506: TODO: Explain "xm start" and "xl start".  Explain rc.d/xendomains.
                    507: 
                    508: TODO: Explain why 4.1 rc.d/xendomains has xl, when one should use xm
                    509: on 4.1.
                    510: 
                    511: Creating specific unprivileged domains (domU)
                    512: =============================================
1.14      gdt       513: 
                    514: Creating domUs is almost entirely independent of operating system.  We
                    515: first explain NetBSD, and then differences for Linux and Solaris.
                    516: 
                    517: Creating an unprivileged NetBSD domain (domU)
                    518: ---------------------------------------------
1.1       mspo      519: 
                    520: Once you have *domain0* running, you need to start the xen tool daemon
1.5       mspo      521: (`/usr/pkg/share/examples/rc.d/xend start`) and the xen backend daemon
                    522: (`/usr/pkg/share/examples/rc.d/xenbackendd start` for Xen3\*,
                    523: `/usr/pkg/share/examples/rc.d/xencommons start` for Xen4.\*). Make sure
                    524: that `/dev/xencons` and `/dev/xenevt` exist before starting `xend`. You
                    525: can create them with this command:
1.1       mspo      526: 
1.3       mspo      527:     # cd /dev && sh MAKEDEV xen
1.1       mspo      528: 
1.5       mspo      529: xend will write logs to `/var/log/xend.log` and
                    530: `/var/log/xend-debug.log`. You can then control xen with the xm tool.
                    531: 'xm list' will show something like:
1.1       mspo      532: 
1.3       mspo      533:     # xm list
                    534:     Name              Id  Mem(MB)  CPU  State  Time(s)  Console
                    535:     Domain-0           0       64    0  r----     58.1
1.1       mspo      536: 
                    537: 'xm create' allows you to create a new domain. It uses a config file in
                    538: PKG\_SYSCONFDIR for its parameters. By default, this file will be in
1.5       mspo      539: `/usr/pkg/etc/xen/`. On creation, a kernel has to be specified, which
                    540: will be executed in the new domain (this kernel is in the *domain0* file
                    541: system, not on the new domain virtual disk; but please note, you should
                    542: install the same kernel into *domainU* as `/netbsd` in order to make
1.27      jnemeth   543: your system tools, like savecore(8), work). A suitable kernel is
1.5       mspo      544: provided as part of the i386 and amd64 binary sets: XEN3\_DOMU.
1.1       mspo      545: 
                    546: Here is an /usr/pkg/etc/xen/nbsd example config file:
                    547: 
1.3       mspo      548:     #  -*- mode: python; -*-
                    549:     #============================================================================
                    550:     # Python defaults setup for 'xm create'.
                    551:     # Edit this file to reflect the configuration of your system.
                    552:     #============================================================================
1.5       mspo      553: 
1.3       mspo      554:     #----------------------------------------------------------------------------
                    555:     # Kernel image file. This kernel will be loaded in the new domain.
                    556:     kernel = "/home/bouyer/netbsd-XEN3_DOMU"
                    557:     #kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
1.5       mspo      558: 
1.3       mspo      559:     # Memory allocation (in megabytes) for the new domain.
                    560:     memory = 128
1.5       mspo      561: 
1.3       mspo      562:     # A handy name for your new domain. This will appear in 'xm list',
                    563:     # and you can use this as parameters for xm in place of the domain
                    564:     # number. All domains must have different names.
                    565:     #
                    566:     name = "nbsd"
1.5       mspo      567: 
1.3       mspo      568:     # The number of virtual CPUs this domain has.
                    569:     #
                    570:     vcpus = 1
1.5       mspo      571: 
1.3       mspo      572:     #----------------------------------------------------------------------------
                    573:     # Define network interfaces for the new domain.
1.5       mspo      574: 
1.3       mspo      575:     # Number of network interfaces (must be at least 1). Default is 1.
                    576:     nics = 1
1.5       mspo      577: 
1.3       mspo      578:     # Define MAC and/or bridge for the network interfaces.
                    579:     #
                    580:     # The MAC address specified in ``mac'' is the one used for the interface
                    581:     # in the new domain. The interface in domain0 will use this address XOR'd
                    582:     # with 00:00:00:01:00:00 (i.e. aa:00:00:51:02:f0 in our example). Random
                    583:     # MACs are assigned if not given.
                    584:     #
                    585:     # ``bridge'' is a required parameter, which will be passed to the
                    586:     # vif-script called by xend(8) when a new domain is created to configure
                    587:     # the new xvif interface in domain0.
                    588:     #
                    589:     # In this example, the xvif is added to bridge0, which should have been
                    590:     # set up prior to the new domain being created -- either in the
                    591:     # ``network'' script or using a /etc/ifconfig.bridge0 file.
                    592:     #
                    593:     vif = [ 'mac=aa:00:00:50:02:f0, bridge=bridge0' ]
1.5       mspo      594: 
1.3       mspo      595:     #----------------------------------------------------------------------------
                    596:     # Define the disk devices you want the domain to have access to, and
                    597:     # what you want them accessible as.
                    598:     #
                    599:     # Each disk entry is of the form:
                    600:     #
1.5       mspo      601:     #   phy:DEV,VDEV,MODE
1.3       mspo      602:     #
                    603:     # where DEV is the device, VDEV is the device name the domain will see,
                    604:     # and MODE is r for read-only, w for read-write.  You can also create
                    605:     # file-backed domains using disk entries of the form:
                    606:     #
1.5       mspo      607:     #   file:PATH,VDEV,MODE
1.3       mspo      608:     #
                    609:     # where PATH is the path to the file used as the virtual disk, and VDEV
                    610:     # and MODE have the same meaning as for ``phy'' devices.
                    611:     #
                    612:     # VDEV doesn't really matter for a NetBSD guest OS (it's just used as an index),
                    613:     # but it does for Linux.
                    614:     # Worse, the device has to exist in /dev/ of domain0, because xm will
                    615:     # try to stat() it. This means that in order to load a Linux guest OS
                    616:     # from a NetBSD domain0, you'll have to create /dev/hda1, /dev/hda2, ...
                    617:     # on domain0, with the major/minor from Linux :(
                    618:     # Alternatively it's possible to specify the device number in hex,
                    619:     # e.g. 0x301 for /dev/hda1, 0x302 for /dev/hda2, etc ...
1.5       mspo      620: 
1.3       mspo      621:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
                    622:     #disk = [ 'file:/var/xen/nbsd-disk,0x01,w' ]
                    623:     #disk = [ 'file:/var/xen/nbsd-disk,0x301,w' ]
1.5       mspo      624: 
1.3       mspo      625:     #----------------------------------------------------------------------------
                    626:     # Set the kernel command line for the new domain.
1.5       mspo      627: 
1.3       mspo      628:     # Set root device. This one does matter for NetBSD
                    629:     root = "xbd0"
                    630:     # extra parameters passed to the kernel
                    631:     # this is where you can set boot flags like -s, -a, etc ...
                    632:     #extra = ""
1.5       mspo      633: 
1.3       mspo      634:     #----------------------------------------------------------------------------
                    635:     # Set according to whether you want the domain restarted when it exits.
                    636:     # The default is False.
                    637:     #autorestart = True
1.5       mspo      638: 
1.3       mspo      639:     # end of nbsd config file ====================================================
1.1       mspo      640: 
                    641: When a new domain is created, xen calls the
1.5       mspo      642: `/usr/pkg/etc/xen/vif-bridge` script for each virtual network interface
                    643: created in *domain0*. This can be used to automatically configure the
                    644: xvif?.? interfaces in *domain0*. In our example, these will be bridged
                    645: with the bridge0 device in *domain0*, but the bridge has to exist first.
                    646: To do this, create the file `/etc/ifconfig.bridge0` and make it look
                    647: like this:
1.1       mspo      648: 
1.3       mspo      649:     create
                    650:     !brconfig $int add ex0 up
1.1       mspo      651: 
1.5       mspo      652: (replace `ex0` with the name of your physical interface). Then bridge0
1.27      jnemeth   653: will be created on boot. See the bridge(4) man page for details.
1.1       mspo      654: 
1.5       mspo      655: So, here is a suitable `/usr/pkg/etc/xen/vif-bridge` for xvif?.? (a
                    656: working vif-bridge is also provided with xentools20) configuring:
1.1       mspo      657: 
1.5       mspo      658:     #!/bin/sh
1.3       mspo      659:     #============================================================================
1.42    ! gdt       660:     # $NetBSD: howto.mdwn,v 1.41 2014/12/26 13:04:31 gdt Exp $
1.3       mspo      661:     #
                    662:     # /usr/pkg/etc/xen/vif-bridge
                    663:     #
                    664:     # Script for configuring a vif in bridged mode with a dom0 interface.
                    665:     # The xend(8) daemon calls a vif script when bringing a vif up or down.
                    666:     # The script name to use is defined in /usr/pkg/etc/xen/xend-config.sxp
                    667:     # in the ``vif-script'' field.
                    668:     #
                    669:     # Usage: vif-bridge up|down [var=value ...]
                    670:     #
                    671:     # Actions:
1.5       mspo      672:     #    up     Adds the vif interface to the bridge.
                    673:     #    down   Removes the vif interface from the bridge.
1.3       mspo      674:     #
                    675:     # Variables:
1.5       mspo      676:     #    domain name of the domain the interface is on (required).
                    677:     #    vifq   vif interface name (required).
                    678:     #    mac    vif MAC address (required).
                    679:     #    bridge bridge to add the vif to (required).
1.3       mspo      680:     #
                    681:     # Example invocation:
                    682:     #
                    683:     # vif-bridge up domain=VM1 vif=xvif1.0 mac="ee:14:01:d0:ec:af" bridge=bridge0
                    684:     #
                    685:     #============================================================================
1.5       mspo      686: 
1.3       mspo      687:     # Exit if anything goes wrong
                    688:     set -e
1.5       mspo      689: 
1.3       mspo      690:     echo "vif-bridge $*"
1.5       mspo      691: 
1.3       mspo      692:     # Operation name.
                    693:     OP=$1; shift
1.5       mspo      694: 
1.3       mspo      695:     # Pull variables in args into environment
                    696:     for arg ; do export "${arg}" ; done
1.5       mspo      697: 
1.3       mspo      698:     # Required parameters. Fail if not set.
                    699:     domain=${domain:?}
                    700:     vif=${vif:?}
                    701:     mac=${mac:?}
                    702:     bridge=${bridge:?}
1.5       mspo      703: 
1.3       mspo      704:     # Optional parameters. Set defaults.
                    705:     ip=${ip:-''}   # default to null (do nothing)
1.5       mspo      706: 
1.3       mspo      707:     # Are we going up or down?
                    708:     case $OP in
1.5       mspo      709:     up) brcmd='add' ;;
1.3       mspo      710:     down)   brcmd='delete' ;;
                    711:     *)
1.5       mspo      712:         echo 'Invalid command: ' $OP
                    713:         echo 'Valid commands are: up, down'
                    714:         exit 1
                    715:         ;;
1.3       mspo      716:     esac
1.5       mspo      717: 
1.3       mspo      718:     # Don't do anything if the bridge is "null".
                    719:     if [ "${bridge}" = "null" ] ; then
1.5       mspo      720:         exit
1.3       mspo      721:     fi
1.5       mspo      722: 
1.3       mspo      723:     # Don't do anything if the bridge doesn't exist.
                    724:     if ! ifconfig -l | grep "${bridge}" >/dev/null; then
1.5       mspo      725:         exit
1.3       mspo      726:     fi
1.5       mspo      727: 
1.3       mspo      728:     # Add/remove vif to/from bridge.
                    729:     ifconfig x${vif} $OP
                    730:     brconfig ${bridge} ${brcmd} x${vif}
1.1       mspo      731: 
                    732: Now, running
                    733: 
1.3       mspo      734:     xm create -c /usr/pkg/etc/xen/nbsd
1.1       mspo      735: 
1.5       mspo      736: should create a domain and load a NetBSD kernel in it. (Note: `-c`
                    737: causes xm to connect to the domain's console once created.) The kernel
                    738: will try to find its root file system on xbd0 (i.e., wd0e) which hasn't
                    739: been created yet. wd0e will be seen as a disk device in the new domain,
                    740: so it will be 'sub-partitioned'. We could attach a ccd to wd0e in
                    741: *domain0* and partition it, newfs and extract the NetBSD/i386 or amd64
                    742: tarballs there, but there's an easier way: load the
                    743: `netbsd-INSTALL_XEN3_DOMU` kernel provided in the NetBSD binary sets.
                    744: Like other install kernels, it contains a ramdisk with sysinst, so you
                    745: can install NetBSD using sysinst on your new domain.
1.1       mspo      746: 
                    747: If you want to install NetBSD/Xen with a CDROM image, the following line
1.5       mspo      748: should be used in the `/usr/pkg/etc/xen/nbsd` file:
1.1       mspo      749: 
1.3       mspo      750:     disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
1.1       mspo      751: 
                    752: After booting the domain, the option to install via CDROM may be
1.5       mspo      753: selected. The CDROM device should be changed to `xbd1d`.
1.1       mspo      754: 
1.5       mspo      755: Once done installing, `halt -p` the new domain (don't reboot or halt, it
                    756: would reload the INSTALL\_XEN3\_DOMU kernel even if you changed the
1.1       mspo      757: config file), switch the config file back to the XEN3\_DOMU kernel, and
1.5       mspo      758: start the new domain again. Now it should be able to use `root on xbd0a`
                    759: and you should have a second, functional NetBSD system on your xen
                    760: installation.
1.1       mspo      761: 
                    762: When the new domain is booting you'll see some warnings about *wscons*
                    763: and the pseudo-terminals. These can be fixed by editing the files
1.5       mspo      764: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
                    765: `/etc/ttys`, except *console*, like this:
1.1       mspo      766: 
1.3       mspo      767:     console "/usr/libexec/getty Pc"         vt100   on secure
                    768:     ttyE0   "/usr/libexec/getty Pc"         vt220   off secure
                    769:     ttyE1   "/usr/libexec/getty Pc"         vt220   off secure
                    770:     ttyE2   "/usr/libexec/getty Pc"         vt220   off secure
                    771:     ttyE3   "/usr/libexec/getty Pc"         vt220   off secure
1.1       mspo      772: 
1.5       mspo      773: Finally, all screens must be commented out from `/etc/wscons.conf`.
1.1       mspo      774: 
                    775: It is also desirable to add
                    776: 
1.3       mspo      777:     powerd=YES
1.1       mspo      778: 
1.5       mspo      779: in rc.conf. This way, the domain will be properly shut down if
                    780: `xm shutdown -R` or `xm shutdown -H` is used on the domain0.
1.1       mspo      781: 
                    782: Your domain should be now ready to work, enjoy.
                    783: 
1.14      gdt       784: Creating an unprivileged Linux domain (domU)
1.5       mspo      785: --------------------------------------------
1.1       mspo      786: 
                    787: Creating unprivileged Linux domains isn't much different from
                    788: unprivileged NetBSD domains, but there are some details to know.
                    789: 
                    790: First, the second parameter passed to the disk declaration (the '0x1' in
                    791: the example below)
                    792: 
1.3       mspo      793:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
1.1       mspo      794: 
                    795: does matter to Linux. It wants a Linux device number here (e.g. 0x300
                    796: for hda). Linux builds device numbers as: (major \<\< 8 + minor). So,
                    797: hda1 which has major 3 and minor 1 on a Linux system will have device
                    798: number 0x301. Alternatively, devices names can be used (hda, hdb, ...)
                    799: as xentools has a table to map these names to devices numbers. To export
                    800: a partition to a Linux guest we can use:
                    801: 
1.3       mspo      802:     disk = [ 'phy:/dev/wd0e,0x300,w' ]
                    803:     root = "/dev/hda1 ro"
1.1       mspo      804: 
                    805: and it will appear as /dev/hda on the Linux system, and be used as root
                    806: partition.
                    807: 
                    808: To install the Linux system on the partition to be exported to the guest
                    809: domain, the following method can be used: install sysutils/e2fsprogs
                    810: from pkgsrc. Use mke2fs to format the partition that will be the root
                    811: partition of your Linux domain, and mount it. Then copy the files from a
1.5       mspo      812: working Linux system, make adjustments in `/etc` (fstab, network
                    813: config). It should also be possible to extract binary packages such as
                    814: .rpm or .deb directly to the mounted partition using the appropriate
                    815: tool, possibly running under NetBSD's Linux emulation. Once the
                    816: filesystem has been populated, umount it. If desirable, the filesystem
                    817: can be converted to ext3 using tune2fs -j. It should now be possible to
                    818: boot the Linux guest domain, using one of the vmlinuz-\*-xenU kernels
                    819: available in the Xen binary distribution.
1.1       mspo      820: 
                    821: To get the linux console right, you need to add:
                    822: 
1.3       mspo      823:     extra = "xencons=tty1"
1.1       mspo      824: 
                    825: to your configuration since not all linux distributions auto-attach a
                    826: tty to the xen console.
                    827: 
1.14      gdt       828: Creating an unprivileged Solaris domain (domU)
1.5       mspo      829: ----------------------------------------------
1.1       mspo      830: 
                    831: Download an Opensolaris [release](http://opensolaris.org/os/downloads/)
                    832: or [development snapshot](http://genunix.org/) DVD image. Attach the DVD
1.5       mspo      833: image to a MAN.VND.4 device. Copy the kernel and ramdisk filesystem
                    834: image to your dom0 filesystem.
1.1       mspo      835: 
1.3       mspo      836:     dom0# mkdir /root/solaris
                    837:     dom0# vnconfig vnd0 osol-1002-124-x86.iso
                    838:     dom0# mount /dev/vnd0a /mnt
1.5       mspo      839: 
1.3       mspo      840:     ## for a 64-bit guest
                    841:     dom0# cp /mnt/boot/amd64/x86.microroot /root/solaris
                    842:     dom0# cp /mnt/platform/i86xpv/kernel/amd64/unix /root/solaris
1.5       mspo      843: 
1.3       mspo      844:     ## for a 32-bit guest
                    845:     dom0# cp /mnt/boot/x86.microroot /root/solaris
                    846:     dom0# cp /mnt/platform/i86xpv/kernel/unix /root/solaris
1.5       mspo      847: 
1.3       mspo      848:     dom0# umount /mnt
1.5       mspo      849:           
                    850: 
                    851: Keep the MAN.VND.4 configured. For some reason the boot process stalls
                    852: unless the DVD image is attached to the guest as a "phy" device. Create
                    853: an initial configuration file with the following contents. Substitute
                    854: */dev/wd0k* with an empty partition at least 8 GB large.
1.1       mspo      855: 
1.4       mspo      856:     memory = 640
                    857:     name = 'solaris'
                    858:     disk = [ 'phy:/dev/wd0k,0,w' ]
                    859:     disk += [ 'phy:/dev/vnd0d,6:cdrom,r' ]
                    860:     vif = [ 'bridge=bridge0' ]
                    861:     kernel = '/root/solaris/unix'
                    862:     ramdisk = '/root/solaris/x86.microroot'
                    863:     # for a 64-bit guest
                    864:     extra = '/platform/i86xpv/kernel/amd64/unix - nowin -B install_media=cdrom'
                    865:     # for a 32-bit guest
                    866:     #extra = '/platform/i86xpv/kernel/unix - nowin -B install_media=cdrom'
1.5       mspo      867:           
                    868: 
1.1       mspo      869: Start the guest.
                    870: 
1.4       mspo      871:     dom0# xm create -c solaris.cfg
                    872:     Started domain solaris
                    873:                           v3.3.2 chgset 'unavailable'
                    874:     SunOS Release 5.11 Version snv_124 64-bit
                    875:     Copyright 1983-2009 Sun Microsystems, Inc.  All rights reserved.
                    876:     Use is subject to license terms.
                    877:     Hostname: opensolaris
                    878:     Remounting root read/write
                    879:     Probing for device nodes ...
                    880:     WARNING: emlxs: ddi_modopen drv/fct failed: err 2
                    881:     Preparing live image for use
                    882:     Done mounting Live image
1.5       mspo      883:           
1.1       mspo      884: 
                    885: Make sure the network is configured. Note that it can take a minute for
                    886: the xnf0 interface to appear.
                    887: 
1.4       mspo      888:     opensolaris console login: jack
                    889:     Password: jack
                    890:     Sun Microsystems Inc.   SunOS 5.11      snv_124 November 2008
                    891:     jack@opensolaris:~$ pfexec sh
                    892:     sh-3.2# ifconfig -a
                    893:     sh-3.2# exit
1.5       mspo      894:           
1.1       mspo      895: 
                    896: Set a password for VNC and start the VNC server which provides the X11
                    897: display where the installation program runs.
                    898: 
1.4       mspo      899:     jack@opensolaris:~$ vncpasswd
                    900:     Password: solaris
                    901:     Verify: solaris
                    902:     jack@opensolaris:~$ cp .Xclients .vnc/xstartup
                    903:     jack@opensolaris:~$ vncserver :1
1.5       mspo      904:           
1.1       mspo      905: 
1.5       mspo      906: From a remote machine connect to the VNC server. Use `ifconfig xnf0` on
                    907: the guest to find the correct IP address to use.
1.1       mspo      908: 
1.4       mspo      909:     remote$ vncviewer 172.18.2.99:1
1.5       mspo      910:           
1.1       mspo      911: 
                    912: It is also possible to launch the installation on a remote X11 display.
                    913: 
1.4       mspo      914:     jack@opensolaris:~$ export DISPLAY=172.18.1.1:0
                    915:     jack@opensolaris:~$ pfexec gui-install
1.5       mspo      916:            
1.1       mspo      917: 
                    918: After the GUI installation is complete you will be asked to reboot.
                    919: Before that you need to determine the ZFS ID for the new boot filesystem
                    920: and update the configuration file accordingly. Return to the guest
                    921: console.
                    922: 
1.4       mspo      923:     jack@opensolaris:~$ pfexec zdb -vvv rpool | grep bootfs
                    924:                     bootfs = 43
                    925:     ^C
                    926:     jack@opensolaris:~$
1.5       mspo      927:            
1.1       mspo      928: 
                    929: The final configuration file should look like this. Note in particular
                    930: the last line.
                    931: 
1.4       mspo      932:     memory = 640
                    933:     name = 'solaris'
                    934:     disk = [ 'phy:/dev/wd0k,0,w' ]
                    935:     vif = [ 'bridge=bridge0' ]
                    936:     kernel = '/root/solaris/unix'
                    937:     ramdisk = '/root/solaris/x86.microroot'
                    938:     extra = '/platform/i86xpv/kernel/amd64/unix -B zfs-bootfs=rpool/43,bootpath="/xpvd/xdf@0:a"'
1.5       mspo      939:            
1.1       mspo      940: 
                    941: Restart the guest to verify it works correctly.
                    942: 
1.4       mspo      943:     dom0# xm destroy solaris
                    944:     dom0# xm create -c solaris.cfg
                    945:     Using config file "./solaris.cfg".
                    946:     v3.3.2 chgset 'unavailable'
                    947:     Started domain solaris
                    948:     SunOS Release 5.11 Version snv_124 64-bit
                    949:     Copyright 1983-2009 Sun Microsystems, Inc.  All rights reserved.
                    950:     Use is subject to license terms.
                    951:     WARNING: emlxs: ddi_modopen drv/fct failed: err 2
                    952:     Hostname: osol
                    953:     Configuring devices.
                    954:     Loading smf(5) service descriptions: 160/160
                    955:     svccfg import warnings. See /var/svc/log/system-manifest-import:default.log .
                    956:     Reading ZFS config: done.
                    957:     Mounting ZFS filesystems: (6/6)
                    958:     Creating new rsa public/private host key pair
                    959:     Creating new dsa public/private host key pair
1.5       mspo      960: 
1.4       mspo      961:     osol console login:
1.5       mspo      962:            
1.1       mspo      963: 
                    964: Using PCI devices in guest domains
1.14      gdt       965: ----------------------------------
1.1       mspo      966: 
                    967: The domain0 can give other domains access to selected PCI devices. This
                    968: can allow, for example, a non-privileged domain to have access to a
                    969: physical network interface or disk controller. However, keep in mind
                    970: that giving a domain access to a PCI device most likely will give the
                    971: domain read/write access to the whole physical memory, as PCs don't have
                    972: an IOMMU to restrict memory access to DMA-capable device. Also, it's not
                    973: possible to export ISA devices to non-domain0 domains (which means that
                    974: the primary VGA adapter can't be exported. A guest domain trying to
                    975: access the VGA registers will panic).
                    976: 
                    977: This functionality is only available in NetBSD-5.1 (and later) domain0
                    978: and domU. If the domain0 is NetBSD, it has to be running Xen 3.1, as
                    979: support has not been ported to later versions at this time.
                    980: 
                    981: For a PCI device to be exported to a domU, is has to be attached to the
1.5       mspo      982: `pciback` driver in domain0. Devices passed to the domain0 via the
                    983: pciback.hide boot parameter will attach to `pciback` instead of the
                    984: usual driver. The list of devices is specified as `(bus:dev.func)`,
                    985: where bus and dev are 2-digit hexadecimal numbers, and func a
                    986: single-digit number:
1.1       mspo      987: 
1.4       mspo      988:     pciback.hide=(00:0a.0)(00:06.0)
1.1       mspo      989: 
                    990: pciback devices should show up in the domain0's boot messages, and the
1.5       mspo      991: devices should be listed in the `/kern/xen/pci` directory.
1.1       mspo      992: 
1.5       mspo      993: PCI devices to be exported to a domU are listed in the `pci` array of
                    994: the domU's config file, with the format `'0000:bus:dev.func'`
1.1       mspo      995: 
1.4       mspo      996:     pci = [ '0000:00:06.0', '0000:00:0a.0' ]
1.1       mspo      997: 
1.5       mspo      998: In the domU an `xpci` device will show up, to which one or more pci
                    999: busses will attach. Then the PCI drivers will attach to PCI busses as
                   1000: usual. Note that the default NetBSD DOMU kernels do not have `xpci` or
                   1001: any PCI drivers built in by default; you have to build your own kernel
                   1002: to use PCI devices in a domU. Here's a kernel config example:
1.1       mspo     1003: 
1.4       mspo     1004:     include         "arch/i386/conf/XEN3_DOMU"
                   1005:     #include         "arch/i386/conf/XENU"           # in NetBSD 3.0
1.5       mspo     1006: 
1.4       mspo     1007:     # Add support for PCI busses to the XEN3_DOMU kernel
                   1008:     xpci* at xenbus ?
                   1009:     pci* at xpci ?
1.5       mspo     1010: 
1.4       mspo     1011:     # Now add PCI and related devices to be used by this domain
                   1012:     # USB Controller and Devices
1.5       mspo     1013: 
1.4       mspo     1014:     # PCI USB controllers
                   1015:     uhci*   at pci? dev ? function ?        # Universal Host Controller (Intel)
1.5       mspo     1016: 
1.4       mspo     1017:     # USB bus support
                   1018:     usb*    at uhci?
1.5       mspo     1019: 
1.4       mspo     1020:     # USB Hubs
                   1021:     uhub*   at usb?
                   1022:     uhub*   at uhub? port ? configuration ? interface ?
1.5       mspo     1023: 
1.4       mspo     1024:     # USB Mass Storage
                   1025:     umass*  at uhub? port ? configuration ? interface ?
                   1026:     wd*     at umass?
                   1027:     # SCSI controllers
                   1028:     ahc*    at pci? dev ? function ?        # Adaptec [23]94x, aic78x0 SCSI
1.5       mspo     1029: 
1.4       mspo     1030:     # SCSI bus support (for both ahc and umass)
                   1031:     scsibus* at scsi?
1.5       mspo     1032: 
1.4       mspo     1033:     # SCSI devices
                   1034:     sd*     at scsibus? target ? lun ?      # SCSI disk drives
                   1035:     cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
1.1       mspo     1036: 
                   1037: 
1.28      gdt      1038: NetBSD as a domU in a VPS
                   1039: =========================
                   1040: 
                   1041: The bulk of the HOWTO is about using NetBSD as a dom0 on your own
                   1042: hardware.  This section explains how to deal with Xen in a domU as a
                   1043: virtual private server where you do not control or have access to the
                   1044: dom0.
                   1045: 
                   1046: TODO: Perhaps reference panix, prmgr, amazon as interesting examples.
                   1047: 
                   1048: TODO: Somewhere, discuss pvgrub and py-grub to load the domU kernel
                   1049: from the domU filesystem.

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