Annotation of wikisrc/ports/xen/howto.mdwn, revision 1.149

1.144     maxv        1: [[!meta title="Xen HowTo"]]
1.5       mspo        3: Introduction
1.13      gdt         4: ============
1.1       mspo        5: 
                      6: [![[Xen
1.140     khorben     7: screenshot]](](
1.1       mspo        8: 
1.149   ! maxv        9: Xen is a hypervisor for x86 hardware, which supports running multiple guest
1.58      gdt        10: operating systems on a single physical machine.  Xen is a Type 1 or
                     11: bare-metal hypervisor; one uses the Xen kernel to control the CPU,
                     12: memory and console, a dom0 operating system which mediates access to
                     13: other hardware (e.g., disks, network, USB), and one or more domU
                     14: operating systems which operate in an unprivileged virtualized
                     15: environment.  IO requests from the domU systems are forwarded by the
1.149   ! maxv       16: Xen hypervisor to the dom0 to be fulfilled.
1.12      gdt        17: 
1.149   ! maxv       18: Xen supports different styles of guest:
        !            19: 
        !            20: [[!table data="""
        !            21: Style of guest |Supported by NetBSD
        !            22: PV             |Yes
        !            23: HVM            |Yes
        !            24: PVHVM          |No
        !            25: PVH            |No
        !            26: """]]
        !            27: 
        !            28: In Para-Virtualized (PV) mode, the guest OS does not attempt to access
        !            29: hardware directly, but instead makes hypercalls to the hypervisor; PV
        !            30: guests must be specifically coded for Xen. In HVM mode, no guest
        !            31: modification is required; however, hardware support is required, such
        !            32: as VT-x on Intel CPUs and SVM on AMD CPUs.
        !            33: 
        !            34: There are further features for IOMMU virtualization, Intel's VT-d and
        !            35: AMD's AMD-Vi.  TODO: Explain whether Xen on NetBSD makes use of these
        !            36: features.  TODO: Review by someone who really understands this.
1.12      gdt        37: 
1.123     gdt        38: Generally any machine that runs NetBSD/amd64 will work with Xen and PV
                     39: guests.  In theory i386 computers (without x86_64/amd64 support) can
                     40: be used for Xen <= 4.2, but we have no recent reports of this working
                     41: (this is a hint).  For HVM guests, hardware support is needed, but it
1.149   ! maxv       42: is common on recent machines.
1.118     gdt        43: 
1.27      jnemeth    44: At boot, the dom0 kernel is loaded as a module with Xen as the kernel.
1.12      gdt        45: The dom0 can start one or more domUs.  (Booting is explained in detail
                     46: in the dom0 section.)
                     48: NetBSD supports Xen in that it can serve as dom0, be used as a domU,
                     49: and that Xen kernels and tools are available in pkgsrc.  This HOWTO
                     50: attempts to address both the case of running a NetBSD dom0 on hardware
1.24      gdt        51: and running domUs under it (NetBSD and other), and also running NetBSD
                     52: as a domU in a VPS.
1.12      gdt        53: 
1.142     gdt        54: Xen 3.1 in pkgsrc used to support "PCI passthrough", which means that
1.20      gdt        55: specific PCI devices can be made available to a specific domU instead
                     56: of the dom0.  This can be useful to let a domU run X11, or access some
                     57: network interface or other peripheral.
1.111     wiz        59: NetBSD 6 and earlier supported Xen 2; support was removed from NetBSD
1.104     gdt        60: 7.  Xen 2 has been removed from pkgsrc.
1.54      gdt        61: 
1.12      gdt        62: Prerequisites
1.13      gdt        63: -------------
1.12      gdt        64: 
                     65: Installing NetBSD/Xen is not extremely difficult, but it is more
                     66: complex than a normal installation of NetBSD.
1.15      gdt        67: In general, this HOWTO is occasionally overly restrictive about how
                     68: things must be done, guiding the reader to stay on the established
                     69: path when there are no known good reasons to stray.
1.12      gdt        70: 
                     71: This HOWTO presumes a basic familiarity with the Xen system
1.123     gdt        72: architecture, with installing NetBSD on i386/amd64 hardware, and with
                     73: installing software from pkgsrc.  See also the [Xen
                     74: website](
1.1       mspo       75: 
1.15      gdt        76: Versions of Xen and NetBSD
                     77: ==========================
1.27      jnemeth    79: Most of the installation concepts and instructions are independent
                     80: of Xen version and NetBSD version.  This section gives advice on
                     81: which version to choose.  Versions not in pkgsrc and older unsupported
                     82: versions of NetBSD are intentionally ignored.
1.15      gdt        83: 
1.133     gdt        84: The term "amd64" is used to refer to both the NetBSD port and to the
                     85: hardware architecture on which it runs.  (Such hardware is made by
                     86: both Intel and AMD, and in 2016 a normal PC has this CPU
                     87: architecture.)
1.146     maxv       89: Xen versions
                     90: ------------
1.15      gdt        91: 
1.111     wiz        92: In NetBSD, Xen is provided in pkgsrc, via matching pairs of packages
1.15      gdt        93: xenkernel and xentools.  We will refer only to the kernel versions,
                     94: but note that both packages must be installed together and must have
                     95: matching versions.
1.145     maxv       97: Versions available in pkgsrc:
1.85      gdt        98: 
1.145     maxv       99: [[!table data="""
                    100: Xen Version    |Package Name   |Xen CPU Support        |EOL'ed By Upstream
                    101: 4.2            |xenkernel42    |32bit, 64bit           |Yes
                    102: 4.5            |xenkernel45    |64bit                  |Yes
                    103: 4.6            |xenkernel46    |64bit                  |Partially
                    104: 4.8            |xenkernel48    |64bit                  |No
                    105: 4.11           |xenkernel411   |64bit                  |No
                    106: """]]
1.113     gdt       107: 
1.96      gdt       108: See also the [Xen Security Advisory page](
1.145     maxv      110: Note: Xen 4.2 was the last version to support 32bit CPUs.
1.19      gdt       112: Xen command program
                    113: -------------------
1.79      gdt       115: Early Xen used a program called xm to manipulate the system from the
1.19      gdt       116: dom0.  Starting in 4.1, a replacement program with similar behavior
1.79      gdt       117: called xl is provided, but it does not work well in 4.1.  In 4.2, both
1.127     gdt       118: xm and xl work fine.  4.4 is the last version that has xm.
                    120: You must make a global choice to use xm or xl, because it affects not
                    121: only which command you use, but the command used by rc.d scripts
                    122: (specifically xendomains) and which daemons should be run.  The
1.146     maxv      123: xentools packages provide xl for 4.2 and up.
1.127     gdt       124: 
                    125: In 4.2, you can choose to use xm by simply changing the ctl_command
1.135     gdt       126: variable and setting xend=YES in rc.conf.
1.127     gdt       127: 
                    128: With xl, virtual devices are configured in parallel, which can cause
                    129: problems if they are written assuming serial operation (e.g., updating
1.130     gdt       130: firewall rules without explicit locking).  There is now locking for
                    131: the provided scripts, which works for normal casses (e.g, file-backed
                    132: xbd, where a vnd must be allocated).  But, as of 201612, it has not
                    133: been adequately tested for a complex custom setup with a large number
                    134: of interfaces.
1.19      gdt       135: 
1.147     maxv      136: NetBSD versions
                    137: ---------------
1.15      gdt       138: 
1.142     gdt       139: The netbsd-7, netbsd-8, and -current branches are all reasonable
1.105     gdt       140: choices, with more or less the same considerations for non-Xen use.
1.147     maxv      141: NetBSD 8 is recommended as the stable version of the most recent
                    142: release for production use.
1.146     maxv      143: 
                    144: For developing Xen, netbsd-current may be appropriate.
1.15      gdt       145: 
                    146: As of NetBSD 6, a NetBSD domU will support multiple vcpus.  There is
                    147: no SMP support for NetBSD as dom0.  (The dom0 itself doesn't really
1.105     gdt       148: need SMP for dom0 functions; the lack of support is really a problem
                    149: when using a dom0 as a normal computer.)
1.15      gdt       150: 
1.147     maxv      151: Note: NetBSD support is called XEN3. However, it does support Xen 4,
                    152: because the hypercall interface has remained identical.
1.18      gdt       154: Architecture
                    155: ------------
1.133     gdt       157: Xen itself can run on i386 (Xen < 4.2) or amd64 hardware (all Xen
1.105     gdt       158: versions).  (Practically, almost any computer where one would want to
                    159: run Xen today supports amd64.)
1.99      gdt       160: 
1.133     gdt       161: Xen, the dom0 system, and each domU system can be either i386 or
                    162: amd64.  When building a xenkernel package, one obtains an i386 Xen
1.134     wiz       163: kernel on an i386 host, and an amd64 Xen kernel on an amd64 host.  If
1.133     gdt       164: the Xen kernel is i386, then the dom0 kernel and all domU kernels must
                    165: be i386.  With an amd64 Xen kernel, an amd64 dom0 kernel is known to
                    166: work, and an i386 dom0 kernel should in theory work.  An amd64
                    167: Xen/dom0 is known to support both i386 and amd64 domUs.
                    169: i386 dom0 and domU kernels must be PAE (except for an i386 Xen 3.1
1.135     gdt       170: kernel, where one can use non-PAE for dom0 and all domUs); PAE kernels
                    171: are included in the NetBSD default build.  (Note that emacs (at least)
                    172: fails if run on i386 with PAE when built without, and vice versa,
                    173: presumably due to bugs in the undump code.)
                    175: Because of the above, the standard approach is to use an amd64 Xen
                    176: kernel and NetBSD/amd64 for the dom0.  For domUs, NetBSD/i386 (with
                    177: the PAE kernel) and NetBSD/amd64 are in widespread use, and there is
                    178: little to no Xen-specific reason to prefer one over the other.
1.133     gdt       179: 
                    180: Note that to use an i386 dom0 with Xen 4.5 or higher, one must build
1.135     gdt       181: (or obtain from pre-built packages) an amd64 Xen kernel and install
                    182: that on the system.  (One must also use a PAE i386 kernel, but this is
                    183: also required with an i386 Xen kernel.).  Almost no one in the
                    184: NetBSD/Xen community does this, and the standard, well-tested,
                    185: approach is to use an amd64 dom0.
                    187: A [posting on
                    188: xen-devel](
                    189: explained that PV system call overhead was higher on amd64, and thus
                    190: there is some notion that i386 guests are faster.  It goes on to
                    191: caution that the total situation is complex and not entirely
                    192: understood. On top of that caution, the post is about Linux, not
                    193: NetBSD.  TODO: Include link to benchmarks, if someone posts them.
1.29      gdt       194: 
1.15      gdt       195: Recommendation
                    196: --------------
1.142     gdt       198: Therefore, this HOWTO recommends running xenkernel46, xl, the NetBSD 7
                    199: stable branch, and therefore to use an amd64 kernel as the dom0.
                    200: Either the i386PAE or amd64 version of NetBSD may be used as domUs.
                    202: A tentative replacement recommendation is xenkernel48, xl, and NetBSD
                    203: 8.
1.15      gdt       204: 
1.136     gdt       205: Because bugs are fixed quite often, and because of Xen security
                    206: advisories, it is good to stay up to date with NetBSD (tracking a
                    207: stable branch), with the Xen kernel (tracking a Xen version via
                    208: pkgsrc), and with the Xen tools.  Specifically, NetBSD (-7 and
                    209: -current) got an important fix affecting dom0/domU timesharing in
                    210: November, 2015, and xentools46 got a fix to enable Ubuntu guests to
                    211: boot in December, 2016.
1.15      gdt       213: NetBSD as a dom0
                    214: ================
                    216: NetBSD can be used as a dom0 and works very well.  The following
                    217: sections address installation, updating NetBSD, and updating Xen.
1.19      gdt       218: Note that it doesn't make sense to talk about installing a dom0 OS
                    219: without also installing Xen itself.  We first address installing
                    220: NetBSD, which is not yet a dom0, and then adding Xen, pivoting the
                    221: NetBSD install to a dom0 install by just changing the kernel and boot
                    222: configuration.
1.15      gdt       223: 
1.45      gdt       224: For experimenting with Xen, a machine with as little as 1G of RAM and
                    225: 100G of disk can work.  For running many domUs in productions, far
1.135     gdt       226: more will be needed; e.g. 4-8G and 1T of disk is reasonable for a
                    227: half-dozen domUs of 512M and 32G each.  Basically, the RAM and disk
                    228: have to be bigger than the sum of the RAM/disk needs of the dom0 and
                    229: all the domUs.
1.45      gdt       230: 
1.142     gdt       231: In 2018-05, trouble booting a dom0 was reported with 256M of RAM: with
                    232: 512M it worked reliably.  This does not make sense, but if you see
                    233: "not ELF" after Xen boots, try increasing dom0 RAM.
1.15      gdt       235: Styles of dom0 operation
                    236: ------------------------
                    238: There are two basic ways to use Xen.  The traditional method is for
                    239: the dom0 to do absolutely nothing other than providing support to some
                    240: number of domUs.  Such a system was probably installed for the sole
                    241: purpose of hosting domUs, and sits in a server room on a UPS.
                    243: The other way is to put Xen under a normal-usage computer, so that the
                    244: dom0 is what the computer would have been without Xen, perhaps a
                    245: desktop or laptop.  Then, one can run domUs at will.  Purists will
                    246: deride this as less secure than the previous approach, and for a
                    247: computer whose purpose is to run domUs, they are right.  But Xen and a
1.93      gdt       248: dom0 (without domUs) is not meaningfully less secure than the same
1.15      gdt       249: things running without Xen.  One can boot Xen or boot regular NetBSD
                    250: alternately with little problems, simply refraining from starting the
                    251: Xen daemons when not running Xen.
                    253: Note that NetBSD as dom0 does not support multiple CPUs.  This will
1.51      gdt       254: limit the performance of the Xen/dom0 workstation approach.  In theory
                    255: the only issue is that the "backend drivers" are not yet MPSAFE:
1.140     khorben   256:
1.15      gdt       257: 
1.19      gdt       258: Installation of NetBSD
                    259: ----------------------
1.13      gdt       260: 
1.19      gdt       261: First,
1.27      jnemeth   262: [install NetBSD/amd64](/guide/inst/)
1.19      gdt       263: just as you would if you were not using Xen.
                    264: However, the partitioning approach is very important.
                    266: If you want to use RAIDframe for the dom0, there are no special issues
                    267: for Xen.  Typically one provides RAID storage for the dom0, and the
1.22      gdt       268: domU systems are unaware of RAID.  The 2nd-stage loader bootxx_* skips
1.111     wiz       269: over a RAID1 header to find /boot from a file system within a RAID
1.22      gdt       270: partition; this is no different when booting Xen.
1.19      gdt       271: 
                    272: There are 4 styles of providing backing storage for the virtual disks
1.93      gdt       273: used by domUs: raw partitions, LVM, file-backed vnd(4), and SAN.
1.19      gdt       274: 
                    275: With raw partitions, one has a disklabel (or gpt) partition sized for
                    276: each virtual disk to be used by the domU.  (If you are able to predict
                    277: how domU usage will evolve, please add an explanation to the HOWTO.
                    278: Seriously, needs tend to change over time.)
1.27      jnemeth   280: One can use [lvm(8)](/guide/lvm/) to create logical devices to use
                    281: for domU disks.  This is almost as efficient as raw disk partitions
                    282: and more flexible.  Hence raw disk partitions should typically not
                    283: be used.
1.19      gdt       284: 
1.111     wiz       285: One can use files in the dom0 file system, typically created by dd'ing
1.19      gdt       286: /dev/zero to create a specific size.  This is somewhat less efficient,
                    287: but very convenient, as one can cp the files for backup, or move them
                    288: between dom0 hosts.
                    290: Finally, in theory one can place the files backing the domU disks in a
                    291: SAN.  (This is an invitation for someone who has done this to add a
                    292: HOWTO page.)
1.1       mspo      293: 
1.19      gdt       294: Installation of Xen
                    295: -------------------
1.1       mspo      296: 
1.20      gdt       297: In the dom0, install sysutils/xenkernel42 and sysutils/xentools42 from
1.127     gdt       298: pkgsrc (or another matching pair).  See [the pkgsrc
1.140     khorben   299: documentation]( for help with
1.127     gdt       300: pkgsrc.  Ensure that your packages are recent; the HOWTO does not
                    301: contemplate old builds.
1.20      gdt       303: 
                    304: For Xen 3.1, support for HVM guests is in sysutils/xentool3-hvm.  More
                    305: recent versions have HVM support integrated in the main xentools
                    306: package.  It is entirely reasonable to run only PV guests.
                    308: Next you need to install the selected Xen kernel itself, which is
                    309: installed by pkgsrc as "/usr/pkg/xen*-kernel/xen.gz".  Copy it to /.
                    310: For debugging, one may copy xen-debug.gz; this is conceptually similar
                    311: to DIAGNOSTIC and DEBUG in NetBSD.  xen-debug.gz is basically only
                    312: useful with a serial console.  Then, place a NetBSD XEN3_DOM0 kernel
                    313: in /, copied from releasedir/amd64/binary/kernel/netbsd-XEN3_DOM0.gz
1.75      gdt       314: of a NetBSD build.  If using i386, use
                    315: releasedir/i386/binary/kernel/netbsd-XEN3PAE_DOM0.gz.  (If using Xen
                    316: 3.1 and i386, you may use XEN3_DOM0 with the non-PAE Xen.  But you
                    317: should not use Xen 3.1.)  Both xen and the NetBSD kernel may be (and
                    318: typically are) left compressed.
1.135     gdt       320: In a dom0, kernfs is mandatory for xend to communicate with the
                    321: kernel, so ensure that /kern is in fstab.  (A standard NetBSD install
                    322: should already mount /kern.)
1.20      gdt       323: 
                    324: Because you already installed NetBSD, you have a working boot setup
                    325: with an MBR bootblock, either bootxx_ffsv1 or bootxx_ffsv2 at the
1.135     gdt       326: beginning of your root file system, have /boot, and likely also
1.20      gdt       327: /boot.cfg.  (If not, fix before continuing!)
1.76      gdt       329: Add a line to to /boot.cfg to boot Xen.  See boot.cfg(5) for an
                    330: example.  The basic line is
1.20      gdt       331: 
1.142     gdt       332:         menu=Xen:load /netbsd-XEN3_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=512M
1.20      gdt       333: 
1.142     gdt       334: which specifies that the dom0 should have 512M, leaving the rest to be
1.77      gdt       335: allocated for domUs.  To use a serial console, use
1.142     gdt       337:         menu=Xen:load /netbsd-XEN3_DOM0.gz;multiboot /xen.gz dom0_mem=512M console=com1 com1=9600,8n1
1.77      gdt       338: 
                    339: which will use the first serial port for Xen (which counts starting
1.139     gson      340: from 1, unlike NetBSD which counts starting from 0), forcing
                    341: speed/parity.  Because the NetBSD command line lacks a
                    342: "console=pc" argument, it will use the default "xencons" console device,
                    343: which directs the console I/O through Xen to the same console device Xen
                    344: itself uses (in this case, the serial port).
                    346: In an attempt to add performance, one can also add
1.37      gdt       347: 
                    348:         dom0_max_vcpus=1 dom0_vcpus_pin
                    350: to force only one vcpu to be provided (since NetBSD dom0 can't use
1.111     wiz       351: more) and to pin that vcpu to a physical CPU.  TODO: benchmark this.
1.20      gdt       352: 
1.93      gdt       353: Xen has [many boot
                    354: options](,
1.111     wiz       355: and other than dom0 memory and max_vcpus, they are generally not
1.93      gdt       356: necessary.
1.20      gdt       358: As with non-Xen systems, you should have a line to boot /netbsd (a
1.127     gdt       359: kernel that works without Xen).  Consider a line to boot /netbsd.ok (a
                    360: fallback version of the non-Xen kernel, updated manually when you are
                    361: sure /netbsd is ok).  Consider also a line to boot fallback versions
                    362: of Xen and the dom0 kernel, but note that non-Xen NetBSD can be used
                    363: to resolve Xen booting issues.
                    365: Probably you want a default=N line to choose Xen in the absence of
                    366: intervention.
1.1       mspo      367: 
1.76      gdt       368: Now, reboot so that you are running a DOM0 kernel under Xen, rather
                    369: than GENERIC without Xen.
1.54      gdt       371: Using grub (historic)
                    372: ---------------------
                    374: Before NetBSD's native bootloader could support Xen, the use of
                    375: grub was recommended.  If necessary, see the
1.135     gdt       376: [old grub information](/ports/xen/howto-grub).
1.54      gdt       377: 
1.28      gdt       378: The [HowTo on Installing into
1.140     khorben   379: RAID-1](
1.28      gdt       380: explains how to set up booting a dom0 with Xen using grub with
                    381: NetBSD's RAIDframe.  (This is obsolete with the use of NetBSD's native
1.135     gdt       382: boot.  Now, just create a system with RAID-1, and alter /boot.cfg as
                    383: described above.)
1.28      gdt       384: 
1.21      gdt       385: Configuring Xen
                    386: ---------------
1.53      gdt       388: Xen logs will be in /var/log/xen.
1.76      gdt       390: Now, you have a system that will boot Xen and the dom0 kernel, but not
                    391: do anything else special.  Make sure that you have rebooted into Xen.
                    392: There will be no domUs, and none can be started because you still have
1.102     gdt       393: to configure the dom0 daemons.
1.21      gdt       394: 
1.102     gdt       395: The daemons which should be run vary with Xen version and with whether
1.127     gdt       396: one is using xm or xl.  The Xen 3.1, 3.3 and 4.1 packages use xm.  Xen
                    397: 4.2 and up packages use xl.  To use xm with 4.2, edit xendomains to
                    398: use xm instead.
1.132     gdt       400: For 3.1 and 3.3, you should enable xend and xenbackendd:
1.31      gdt       401: 
1.32      gdt       402:         xend=YES
                    403:         xenbackendd=YES
1.31      gdt       404: 
1.132     gdt       405: For 4.1 and up, you should enable xencommons.  Not enabling xencommons
                    406: will result in a hang; it is necessary to hit ^C on the console to let
                    407: the machine finish booting.  If you are using xm (default in 4.1, or
                    408: if you changed xendomains in 4.2), you should also enable xend:
1.31      gdt       409: 
1.132     gdt       410:         xend=YES # only if using xm, and only installed <= 4.2
1.53      gdt       411:         xencommons=YES
1.31      gdt       412: 
                    413: TODO: Recommend for/against xen-watchdog.
1.27      jnemeth   414: 
1.53      gdt       415: After you have configured the daemons and either started them (in the
1.79      gdt       416: order given) or rebooted, use xm or xl to inspect Xen's boot messages,
1.102     gdt       417: available resources, and running domains.  An example with xl follows:
1.34      gdt       418: 
1.102     gdt       419:         # xl dmesg
1.43      gdt       420:        [xen's boot info]
1.102     gdt       421:         # xl info
1.43      gdt       422:        [available memory, etc.]
1.102     gdt       423:         # xl list
1.43      gdt       424:         Name              Id  Mem(MB)  CPU  State  Time(s)  Console
                    425:         Domain-0           0       64    0  r----     58.1
1.33      gdt       426: 
1.88      gdt       427: ### Issues with xencommons
                    429: xencommons starts xenstored, which stores data on behalf of dom0 and
                    430: domUs.  It does not currently work to stop and start xenstored.
                    431: Certainly all domUs should be shutdown first, following the sort order
                    432: of the rc.d scripts.  However, the dom0 sets up state with xenstored,
                    433: and is not notified when xenstored exits, leading to not recreating
                    434: the state when the new xenstored starts.  Until there's a mechanism to
                    435: make this work, one should not expect to be able to restart xenstored
                    436: (and thus xencommons).  There is currently no reason to expect that
                    437: this will get fixed any time soon.
1.127     gdt       439: ### No-longer needed advice about devices
                    441: The installation of NetBSD should already have created devices for xen
                    442: (xencons, xenevt, xsd_kva), but if they are not present, create them:
                    444:         cd /dev && sh MAKEDEV xen
1.41      gdt       446: anita (for testing NetBSD)
                    447: --------------------------
1.82      gdt       449: With the setup so far (assuming 4.2/xl), one should be able to run
                    450: anita (see pkgsrc/misc/py-anita) to test NetBSD releases, by doing (as
                    451: root, because anita must create a domU):
                    453:         anita --vmm=xl test file:///usr/obj/i386/
                    455: Alternatively, one can use --vmm=xm to use xm-based domU creation
                    456: instead (and must, on Xen <= 4.1).   TODO: confirm that anita xl really works.
1.40      gdt       458: Xen-specific NetBSD issues
                    459: --------------------------
                    461: There are (at least) two additional things different about NetBSD as a
                    462: dom0 kernel compared to hardware.
1.111     wiz       464: One is that the module ABI is different because some of the #defines
1.109     gdt       465: change, so one must build modules for Xen.  As of netbsd-7, the build
                    466: system does this automatically.  TODO: check this.  (Before building
                    467: Xen modules was added, it was awkward to use modules to the point
                    468: where it was considered that it did not work.)
1.40      gdt       469: 
                    470: The other difference is that XEN3_DOM0 does not have exactly the same
                    471: options as GENERIC.  While it is debatable whether or not this is a
                    472: bug, users should be aware of this and can simply add missing config
                    473: items if desired.
1.15      gdt       475: Updating NetBSD in a dom0
                    476: -------------------------
                    478: This is just like updating NetBSD on bare hardware, assuming the new
                    479: version supports the version of Xen you are running.  Generally, one
                    480: replaces the kernel and reboots, and then overlays userland binaries
                    481: and adjusts /etc.
                    483: Note that one must update both the non-Xen kernel typically used for
                    484: rescue purposes and the DOM0 kernel used with Xen.
1.55      gdt       486: Converting from grub to /boot
                    487: -----------------------------
                    489: These instructions were [TODO: will be] used to convert a system from
                    490: grub to /boot.  The system was originally installed in February of
                    491: 2006 with a RAID1 setup and grub to boot Xen 2, and has been updated
                    492: over time.  Before these commands, it was running NetBSD 6 i386, Xen
                    493: 4.1 and grub, much like the message linked earlier in the grub
                    494: section.
1.111     wiz       496:         # Install MBR bootblocks on both disks. 
1.55      gdt       497:         fdisk -i /dev/rwd0d
                    498:         fdisk -i /dev/rwd1d
                    499:         # Install NetBSD primary boot loader (/ is FFSv1) into RAID1 components.
                    500:         installboot -v /dev/rwd0d /usr/mdec/bootxx_ffsv1
                    501:         installboot -v /dev/rwd1d /usr/mdec/bootxx_ffsv1
                    502:         # Install secondary boot loader
                    503:         cp -p /usr/mdec/boot /
1.111     wiz       504:         # Create boot.cfg following earlier guidance:
1.142     gdt       505:         menu=Xen:load /netbsd-XEN3PAE_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=512M
                    506:         menu=Xen.ok:load /netbsd-XEN3PAE_DOM0.ok.gz console=pc;multiboot /xen.ok.gz dom0_mem=512M
1.55      gdt       507:         menu=GENERIC:boot
                    508:         menu=GENERIC single-user:boot -s
                    509:         menu=GENERIC.ok:boot netbsd.ok
                    510:         menu=GENERIC.ok single-user:boot netbsd.ok -s
                    511:         menu=Drop to boot prompt:prompt
                    512:         default=1
                    513:         timeout=30
                    515: TODO: actually do this and fix it if necessary.
1.22      gdt       516: 
1.102     gdt       517: Upgrading Xen versions
1.15      gdt       518: ---------------------
1.110     gdt       520: Minor version upgrades are trivial.  Just rebuild/replace the
                    521: xenkernel version and copy the new xen.gz to / (where /boot.cfg
                    522: references it), and reboot.
                    524: Major version upgrades are conceptually not difficult, but can run
                    525: into all the issues found when installing Xen.  Assuming migration
                    526: from 4.1 to 4.2, remove the xenkernel41 and xentools41 packages and
                    527: install the xenkernel42 and xentools42 packages.  Copy the 4.2 xen.gz
                    528: to /.
1.21      gdt       529: 
1.102     gdt       530: Ensure that the contents of /etc/rc.d/xen* are correct.  Specifically,
                    531: they must match the package you just installed and not be left over
                    532: from some previous installation.
                    534: Enable the correct set of daemons; see the configuring section above.
                    535: (Upgrading from 3.x to 4.x without doing this will result in a hang.)
                    537: Ensure that the domU config files are valid for the new version.
1.110     gdt       538: Specifically, for 4.x remove autorestart=True, and ensure that disks
                    539: are specified with numbers as the second argument, as the examples
                    540: above show, and not NetBSD device names.
1.15      gdt       541: 
1.97      gdt       542: Hardware known to work
                    543: ----------------------
                    545: Arguably, this section is misplaced, and there should be a page of
                    546: hardware that runs NetBSD/amd64 well, with the mostly-well-founded
                    547: assumption that NetBSD/xen runs fine on any modern hardware that
1.129     gdt       548: NetBSD/amd64 runs well on.  Until then, we give motherboard/CPU (and
                    549: sometimes RAM) pairs/triples to aid those choosing a motherboard.
                    550: Note that Xen systems usually do not run X, so a listing here does not
                    551: imply that X works at all.
1.97      gdt       552: 
                    553:         Supermicro X9SRL-F, Xeon E5-1650 v2, 96 GiB ECC
                    554:         Supermicro ??, Atom C2758 (8 core), 32 GiB ECC
                    555:         ASUS M5A78L-M/USB3 AM3+ microATX, AMD Piledriver X8 4000MHz, 16 GiB ECC
                    557: Older hardware:
1.98      gdt       559:         Intel D915GEV, Pentium4 CPU 3.40GHz, 4GB 533MHz Synchronous DDR2
1.129     gdt       560:         INTEL DG33FB, "Intel(R) Core(TM)2 Duo CPU     E6850  @ 3.00GHz"
                    561:         INTEL DG33FB, "Intel(R) Core(TM)2 Duo CPU     E8400  @ 3.00GHz"
1.28      gdt       562: 
1.82      gdt       563: Running Xen under qemu
                    564: ----------------------
                    566: The astute reader will note that this section is somewhat twisted.
                    567: However, it can be useful to run Xen under qemu either because the
                    568: version of NetBSD as a dom0 does not run on the hardware in use, or to
                    569: generate automated test cases involving Xen.
1.84      gdt       571: In 2015-01, the following combination was reported to mostly work:
1.82      gdt       572: 
                    573:         host OS: NetBSD/amd64 6.1.4
                    574:         qemu: 2.2.0 from pkgsrc
                    575:         Xen kernel: xenkernel42-4.2.5nb1 from pkgsrc
                    576:         dom0 kernel: NetBSD/amd64 6.1.5
                    577:         Xen tools: xentools42-4.2.5 from pkgsrc
1.140     khorben   579: See [PR 47720]( for a problem with dom0
1.91      gdt       580: shutdown.
1.84      gdt       581: 
1.28      gdt       582: Unprivileged domains (domU)
                    583: ===========================
                    585: This section describes general concepts about domUs.  It does not
1.33      gdt       586: address specific domU operating systems or how to install them.  The
                    587: config files for domUs are typically in /usr/pkg/etc/xen, and are
1.60      wiki      588: typically named so that the file name, domU name and the domU's host
1.33      gdt       589: name match.
1.111     wiz       591: The domU is provided with CPU and memory by Xen, configured by the
1.33      gdt       592: dom0.  The domU is provided with disk and network by the dom0,
                    593: mediated by Xen, and configured in the dom0.
                    595: Entropy in domUs can be an issue; physical disks and network are on
                    596: the dom0.  NetBSD's /dev/random system works, but is often challenged.
1.48      gdt       598: Config files
                    599: ------------
                    601: There is no good order to present config files and the concepts
                    602: surrounding what is being configured.  We first show an example config
                    603: file, and then in the various sections give details.
                    605: See (at least in xentools41) /usr/pkg/share/examples/xen/xmexample*,
                    606: for a large number of well-commented examples, mostly for running
                    607: GNU/Linux.
                    609: The following is an example minimal domain configuration file
                    610: "/usr/pkg/etc/xen/foo".  It is (with only a name change) an actual
                    611: known working config file on Xen 4.1 (NetBSD 5 amd64 dom0 and NetBSD 5
                    612: i386 domU).  The domU serves as a network file server.
                    614:         # -*- mode: python; -*-
                    616:         kernel = "/netbsd-XEN3PAE_DOMU-i386-foo.gz"
                    617:         memory = 1024
                    618:         vif = [ 'mac=aa:00:00:d1:00:09,bridge=bridge0' ]
                    619:         disk = [ 'file:/n0/xen/foo-wd0,0x0,w',
                    620:                  'file:/n0/xen/foo-wd1,0x1,w' ]
                    622: The domain will have the same name as the file.  The kernel has the
                    623: host/domU name in it, so that on the dom0 one can update the various
                    624: domUs independently.  The vif line causes an interface to be provided,
                    625: with a specific mac address (do not reuse MAC addresses!), in bridge
                    626: mode.  Two disks are provided, and they are both writable; the bits
                    627: are stored in files and Xen attaches them to a vnd(4) device in the
1.111     wiz       628: dom0 on domain creation.  The system treats xbd0 as the boot device
1.48      gdt       629: without needing explicit configuration.
                    631: By default xm looks for domain config files in /usr/pkg/etc/xen.  Note
                    632: that "xm create" takes the name of a config file, while other commands
                    633: take the name of a domain.  To create the domain, connect to the
                    634: console, create the domain while attaching the console, shutdown the
                    635: domain, and see if it has finished stopping, do (or xl with Xen >=
                    636: 4.2):
                    638:         xm create foo
                    639:         xm console foo
                    640:         xm create -c foo
                    641:         xm shutdown foo
1.90      gdt       642:         xm list
1.48      gdt       643: 
                    644: Typing ^] will exit the console session.  Shutting down a domain is
                    645: equivalent to pushing the power button; a NetBSD domU will receive a
                    646: power-press event and do a clean shutdown.  Shutting down the dom0
                    647: will trigger controlled shutdowns of all configured domUs.
                    649: domU kernels
                    650: ------------
                    652: On a physical computer, the BIOS reads sector 0, and a chain of boot
                    653: loaders finds and loads a kernel.  Normally this comes from the root
1.111     wiz       654: file system.  With Xen domUs, the process is totally different.  The
1.48      gdt       655: normal path is for the domU kernel to be a file in the dom0's
1.111     wiz       656: file system.  At the request of the dom0, Xen loads that kernel into a
1.48      gdt       657: new domU instance and starts execution.  While domU kernels can be
                    658: anyplace, reasonable places to store domU kernels on the dom0 are in /
                    659: (so they are near the dom0 kernel), in /usr/pkg/etc/xen (near the
                    660: config files), or in /u0/xen (where the vdisks are).
1.59      gdt       662: Note that loading the domU kernel from the dom0 implies that boot
                    663: blocks, /boot, /boot.cfg, and so on are all ignored in the domU.
1.48      gdt       664: See the VPS section near the end for discussion of alternate ways to
                    665: obtain domU kernels.
1.33      gdt       667: CPU and memory
                    668: --------------
1.48      gdt       670: A domain is provided with some number of vcpus, less than the number
1.111     wiz       671: of CPUs seen by the hypervisor.  (For a dom0, this is controlled by
1.48      gdt       672: the boot argument "dom0_max_vcpus=1".)  For a domU, it is controlled
                    673: from the config file by the "vcpus = N" directive.
                    675: A domain is provided with memory; this is controlled in the config
                    676: file by "memory = N" (in megabytes).  In the straightforward case, the
                    677: sum of the the memory allocated to the dom0 and all domUs must be less
1.33      gdt       678: than the available memory.
                    680: Xen also provides a "balloon" driver, which can be used to let domains
                    681: use more memory temporarily.  TODO: Explain better, and explain how
                    682: well it works with NetBSD.
1.28      gdt       683: 
                    684: Virtual disks
                    685: -------------
1.33      gdt       687: With the file/vnd style, typically one creates a directory,
                    688: e.g. /u0/xen, on a disk large enough to hold virtual disks for all
                    689: domUs.  Then, for each domU disk, one writes zeros to a file that then
                    690: serves to hold the virtual disk's bits; a suggested name is foo-xbd0
                    691: for the first virtual disk for the domU called foo.  Writing zeros to
                    692: the file serves two purposes.  One is that preallocating the contents
                    693: improves performance.  The other is that vnd on sparse files has
                    694: failed to work.  TODO: give working/notworking NetBSD versions for
1.127     gdt       695: sparse vnd and gnats reference.  Note that the use of file/vnd for Xen
                    696: is not really different than creating a file-backed virtual disk for
                    697: some other purpose, except that xentools handles the vnconfig
                    698: commands.  To create an empty 4G virtual disk, simply do
1.39      gdt       699: 
                    700:         dd if=/dev/zero of=foo-xbd0 bs=1m count=4096
1.33      gdt       701: 
1.89      gdt       702: Do not use qemu-img-xen, because this will create sparse file.  There
                    703: have been recent (2015) reports of sparse vnd(4) devices causing
                    704: lockups, but there is apparently no PR.
1.33      gdt       706: With the lvm style, one creates logical devices.  They are then used
1.48      gdt       707: similarly to vnds.  TODO: Add an example with lvm.
                    709: In domU config files, the disks are defined as a sequence of 3-tuples.
                    710: The first element is "method:/path/to/disk".  Common methods are
                    711: "file:" for file-backed vnd. and "phy:" for something that is already
                    712: a (TODO: character or block) device.
                    714: The second element is an artifact of how virtual disks are passed to
                    715: Linux, and a source of confusion with NetBSD Xen usage.  Linux domUs
                    716: are given a device name to associate with the disk, and values like
                    717: "hda1" or "sda1" are common.  In a NetBSD domU, the first disk appears
                    718: as xbd0, the second as xbd1, and so on.  However, xm/xl demand a
                    719: second argument.  The name given is converted to a major/minor by
1.49      gdt       720: calling stat(2) on the name in /dev and this is passed to the domU.
                    721: In the general case, the dom0 and domU can be different operating
1.48      gdt       722: systems, and it is an unwarranted assumption that they have consistent
                    723: numbering in /dev, or even that the dom0 OS has a /dev.  With NetBSD
                    724: as both dom0 and domU, using values of 0x0 for the first disk and 0x1
1.49      gdt       725: for the second works fine and avoids this issue.  For a GNU/Linux
                    726: guest, one can create /dev/hda1 in /dev, or to pass 0x301 for
                    727: /dev/hda1.
1.48      gdt       728: 
                    729: The third element is "w" for writable disks, and "r" for read-only
                    730: disks.
1.28      gdt       731: 
1.127     gdt       732: Note that NetBSD by default creates only vnd[0123].  If you need more
                    733: than 4 total virtual disks at a time, run e.g. "./MAKEDEV vnd4" in the
                    734: dom0.
                    736: Note that NetBSD by default creates only xbd[0123].  If you need more
                    737: virtual disks in a domU, run e.g. "./MAKEDEV xbd4" in the domU.
1.28      gdt       739: Virtual Networking
                    740: ------------------
1.111     wiz       742: Xen provides virtual Ethernets, each of which connects the dom0 and a
1.46      gdt       743: domU.  For each virtual network, there is an interface "xvifN.M" in
                    744: the dom0, and in domU index N, a matching interface xennetM (NetBSD
                    745: name).  The interfaces behave as if there is an Ethernet with two
1.111     wiz       746: adapters connected.  From this primitive, one can construct various
1.46      gdt       747: configurations.  We focus on two common and useful cases for which
                    748: there are existing scripts: bridging and NAT.
1.28      gdt       749: 
1.48      gdt       750: With bridging (in the example above), the domU perceives itself to be
                    751: on the same network as the dom0.  For server virtualization, this is
                    752: usually best.  Bridging is accomplished by creating a bridge(4) device
                    753: and adding the dom0's physical interface and the various xvifN.0
                    754: interfaces to the bridge.  One specifies "bridge=bridge0" in the domU
                    755: config file.  The bridge must be set up already in the dom0; an
                    756: example /etc/ifconfig.bridge0 is:
1.46      gdt       757: 
                    758:         create
                    759:         up
                    760:         !brconfig bridge0 add wm0
1.28      gdt       761: 
                    762: With NAT, the domU perceives itself to be behind a NAT running on the
                    763: dom0.  This is often appropriate when running Xen on a workstation.
1.48      gdt       764: TODO: NAT appears to be configured by "vif = [ '' ]".
1.28      gdt       765: 
1.49      gdt       766: The MAC address specified is the one used for the interface in the new
1.53      gdt       767: domain.  The interface in dom0 will use this address XOR'd with
1.49      gdt       768: 00:00:00:01:00:00.  Random MAC addresses are assigned if not given.
1.33      gdt       770: Sizing domains
                    771: --------------
                    773: Modern x86 hardware has vast amounts of resources.  However, many
                    774: virtual servers can function just fine on far less.  A system with
1.142     gdt       775: 512M of RAM and a 4G disk can be a reasonable choice.  Note that it is
1.33      gdt       776: far easier to adjust virtual resources than physical ones.  For
                    777: memory, it's just a config file edit and a reboot.  For disk, one can
                    778: create a new file and vnconfig it (or lvm), and then dump/restore,
                    779: just like updating physical disks, but without having to be there and
                    780: without those pesky connectors.
1.48      gdt       782: Starting domains automatically
                    783: ------------------------------
1.28      gdt       784: 
1.48      gdt       785: To start domains foo at bar at boot and shut them down cleanly on dom0
                    786: shutdown, in rc.conf add:
1.28      gdt       787: 
1.48      gdt       788:         xendomains="foo bar"
1.28      gdt       789: 
1.111     wiz       790: Note that earlier versions of the xentools41 xendomains rc.d script
                    791: used xl, when one should use xm with 4.1.
1.28      gdt       792: 
                    793: Creating specific unprivileged domains (domU)
                    794: =============================================
1.14      gdt       795: 
                    796: Creating domUs is almost entirely independent of operating system.  We
1.49      gdt       797: have already presented the basics of config files.  Note that you must
                    798: have already completed the dom0 setup so that "xl list" (or "xm list")
                    799: works.
1.14      gdt       800: 
                    801: Creating an unprivileged NetBSD domain (domU)
                    802: ---------------------------------------------
1.1       mspo      803: 
1.49      gdt       804: See the earlier config file, and adjust memory.  Decide on how much
                    805: storage you will provide, and prepare it (file or lvm).
1.111     wiz       807: While the kernel will be obtained from the dom0 file system, the same
1.49      gdt       808: file should be present in the domU as /netbsd so that tools like
                    809: savecore(8) can work.   (This is helpful but not necessary.)
                    811: The kernel must be specifically for Xen and for use as a domU.  The
                    812: i386 and amd64 provide the following kernels:
                    814:         i386 XEN3_DOMU
                    815:         i386 XEN3PAE_DOMU
1.95      gdt       816:         amd64 XEN3_DOMU
1.5       mspo      817: 
1.49      gdt       818: Unless using Xen 3.1 (and you shouldn't) with i386-mode Xen, you must
                    819: use the PAE version of the i386 kernel.
                    821: This will boot NetBSD, but this is not that useful if the disk is
                    822: empty.  One approach is to unpack sets onto the disk outside of xen
                    823: (by mounting it, just as you would prepare a physical disk for a
                    824: system you can't run the installer on).
                    826: A second approach is to run an INSTALL kernel, which has a miniroot
                    827: and can load sets from the network.  To do this, copy the INSTALL
                    828: kernel to / and change the kernel line in the config file to:
1.5       mspo      829: 
1.49      gdt       830:         kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
1.5       mspo      831: 
1.49      gdt       832: Then, start the domain as "xl create -c configname".
1.1       mspo      833: 
1.49      gdt       834: Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following
                    835: line should be used in the config file.
1.1       mspo      836: 
1.3       mspo      837:     disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
1.1       mspo      838: 
                    839: After booting the domain, the option to install via CDROM may be
1.49      gdt       840: selected.  The CDROM device should be changed to `xbd1d`.
1.1       mspo      841: 
1.49      gdt       842: Once done installing, "halt -p" the new domain (don't reboot or halt,
                    843: it would reload the INSTALL_XEN3_DOMU kernel even if you changed the
                    844: config file), switch the config file back to the XEN3_DOMU kernel,
                    845: and start the new domain again. Now it should be able to use "root on
                    846: xbd0a" and you should have a, functional NetBSD domU.
1.1       mspo      847: 
1.49      gdt       848: TODO: check if this is still accurate.
1.1       mspo      849: When the new domain is booting you'll see some warnings about *wscons*
                    850: and the pseudo-terminals. These can be fixed by editing the files
1.5       mspo      851: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
                    852: `/etc/ttys`, except *console*, like this:
1.1       mspo      853: 
1.3       mspo      854:     console "/usr/libexec/getty Pc"         vt100   on secure
                    855:     ttyE0   "/usr/libexec/getty Pc"         vt220   off secure
                    856:     ttyE1   "/usr/libexec/getty Pc"         vt220   off secure
                    857:     ttyE2   "/usr/libexec/getty Pc"         vt220   off secure
                    858:     ttyE3   "/usr/libexec/getty Pc"         vt220   off secure
1.1       mspo      859: 
1.5       mspo      860: Finally, all screens must be commented out from `/etc/wscons.conf`.
1.1       mspo      861: 
                    862: It is also desirable to add
1.49      gdt       864:         powerd=YES
1.1       mspo      865: 
1.5       mspo      866: in rc.conf. This way, the domain will be properly shut down if
1.53      gdt       867: `xm shutdown -R` or `xm shutdown -H` is used on the dom0.
1.1       mspo      868: 
1.92      gdt       869: It is not strictly necessary to have a kernel (as /netbsd) in the domU
1.111     wiz       870: file system.  However, various programs (e.g. netstat) will use that
1.92      gdt       871: kernel to look up symbols to read from kernel virtual memory.  If
                    872: /netbsd is not the running kernel, those lookups will fail.  (This is
                    873: not really a Xen-specific issue, but because the domU kernel is
                    874: obtained from the dom0, it is far more likely to be out of sync or
                    875: missing with Xen.)
1.14      gdt       877: Creating an unprivileged Linux domain (domU)
1.5       mspo      878: --------------------------------------------
1.1       mspo      879: 
                    880: Creating unprivileged Linux domains isn't much different from
                    881: unprivileged NetBSD domains, but there are some details to know.
                    883: First, the second parameter passed to the disk declaration (the '0x1' in
                    884: the example below)
1.3       mspo      886:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
1.1       mspo      887: 
                    888: does matter to Linux. It wants a Linux device number here (e.g. 0x300
1.49      gdt       889: for hda).  Linux builds device numbers as: (major \<\< 8 + minor).
                    890: So, hda1 which has major 3 and minor 1 on a Linux system will have
                    891: device number 0x301.  Alternatively, devices names can be used (hda,
                    892: hdb, ...)  as xentools has a table to map these names to devices
                    893: numbers.  To export a partition to a Linux guest we can use:
1.1       mspo      894: 
1.49      gdt       895:         disk = [ 'phy:/dev/wd0e,0x300,w' ]
                    896:         root = "/dev/hda1 ro"
1.1       mspo      897: 
                    898: and it will appear as /dev/hda on the Linux system, and be used as root
                    899: partition.
1.49      gdt       901: To install the Linux system on the partition to be exported to the
                    902: guest domain, the following method can be used: install
                    903: sysutils/e2fsprogs from pkgsrc.  Use mke2fs to format the partition
                    904: that will be the root partition of your Linux domain, and mount it.
                    905: Then copy the files from a working Linux system, make adjustments in
                    906: `/etc` (fstab, network config).  It should also be possible to extract
                    907: binary packages such as .rpm or .deb directly to the mounted partition
                    908: using the appropriate tool, possibly running under NetBSD's Linux
1.111     wiz       909: emulation.  Once the file system has been populated, umount it.  If
                    910: desirable, the file system can be converted to ext3 using tune2fs -j.
1.49      gdt       911: It should now be possible to boot the Linux guest domain, using one of
                    912: the vmlinuz-\*-xenU kernels available in the Xen binary distribution.
1.1       mspo      913: 
1.111     wiz       914: To get the Linux console right, you need to add:
1.1       mspo      915: 
1.3       mspo      916:     extra = "xencons=tty1"
1.1       mspo      917: 
1.111     wiz       918: to your configuration since not all Linux distributions auto-attach a
1.1       mspo      919: tty to the xen console.
1.14      gdt       921: Creating an unprivileged Solaris domain (domU)
1.5       mspo      922: ----------------------------------------------
1.1       mspo      923: 
1.50      gdt       924: See possibly outdated
                    925: [Solaris domU instructions](/ports/xen/howto-solaris/).
1.5       mspo      926: 
1.1       mspo      927: 
1.52      gdt       928: PCI passthrough: Using PCI devices in guest domains
                    929: ---------------------------------------------------
1.1       mspo      930: 
1.53      gdt       931: The dom0 can give other domains access to selected PCI
1.52      gdt       932: devices. This can allow, for example, a non-privileged domain to have
                    933: access to a physical network interface or disk controller.  However,
                    934: keep in mind that giving a domain access to a PCI device most likely
                    935: will give the domain read/write access to the whole physical memory,
                    936: as PCs don't have an IOMMU to restrict memory access to DMA-capable
1.53      gdt       937: device.  Also, it's not possible to export ISA devices to non-dom0
1.52      gdt       938: domains, which means that the primary VGA adapter can't be exported.
                    939: A guest domain trying to access the VGA registers will panic.
1.53      gdt       941: If the dom0 is NetBSD, it has to be running Xen 3.1, as support has
1.52      gdt       942: not been ported to later versions at this time.
                    944: For a PCI device to be exported to a domU, is has to be attached to
                    945: the "pciback" driver in dom0.  Devices passed to the dom0 via the
                    946: pciback.hide boot parameter will attach to "pciback" instead of the
                    947: usual driver.  The list of devices is specified as "(bus:dev.func)",
1.5       mspo      948: where bus and dev are 2-digit hexadecimal numbers, and func a
                    949: single-digit number:
1.1       mspo      950: 
1.52      gdt       951:         pciback.hide=(00:0a.0)(00:06.0)
1.1       mspo      952: 
1.52      gdt       953: pciback devices should show up in the dom0's boot messages, and the
1.5       mspo      954: devices should be listed in the `/kern/xen/pci` directory.
1.1       mspo      955: 
1.52      gdt       956: PCI devices to be exported to a domU are listed in the "pci" array of
                    957: the domU's config file, with the format "0000:bus:dev.func".
1.1       mspo      958: 
1.52      gdt       959:         pci = [ '0000:00:06.0', '0000:00:0a.0' ]
1.1       mspo      960: 
1.52      gdt       961: In the domU an "xpci" device will show up, to which one or more pci
1.111     wiz       962: buses will attach.  Then the PCI drivers will attach to PCI buses as
1.52      gdt       963: usual.  Note that the default NetBSD DOMU kernels do not have "xpci"
                    964: or any PCI drivers built in by default; you have to build your own
                    965: kernel to use PCI devices in a domU.  Here's a kernel config example;
                    966: note that only the "xpci" lines are unusual.
                    968:         include         "arch/i386/conf/XEN3_DOMU"
1.111     wiz       970:         # Add support for PCI buses to the XEN3_DOMU kernel
1.52      gdt       971:         xpci* at xenbus ?
                    972:         pci* at xpci ?
                    974:         # PCI USB controllers
                    975:         uhci*   at pci? dev ? function ?        # Universal Host Controller (Intel)
                    977:         # USB bus support
                    978:         usb*    at uhci?
                    980:         # USB Hubs
                    981:         uhub*   at usb?
                    982:         uhub*   at uhub? port ? configuration ? interface ?
                    984:         # USB Mass Storage
                    985:         umass*  at uhub? port ? configuration ? interface ?
                    986:         wd*     at umass?
                    987:         # SCSI controllers
                    988:         ahc*    at pci? dev ? function ?        # Adaptec [23]94x, aic78x0 SCSI
                    990:         # SCSI bus support (for both ahc and umass)
                    991:         scsibus* at scsi?
                    993:         # SCSI devices
                    994:         sd*     at scsibus? target ? lun ?      # SCSI disk drives
                    995:         cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
1.1       mspo      996: 
1.28      gdt       998: NetBSD as a domU in a VPS
                    999: =========================
                   1001: The bulk of the HOWTO is about using NetBSD as a dom0 on your own
                   1002: hardware.  This section explains how to deal with Xen in a domU as a
                   1003: virtual private server where you do not control or have access to the
1.70      gdt      1004: dom0.  This is not intended to be an exhaustive list of VPS providers;
                   1005: only a few are mentioned that specifically support NetBSD.
1.28      gdt      1006: 
1.52      gdt      1007: VPS operators provide varying degrees of access and mechanisms for
                   1008: configuration.  The big issue is usually how one controls which kernel
1.111     wiz      1009: is booted, because the kernel is nominally in the dom0 file system (to
                   1010: which VPS users do not normally have access).  A second issue is how
1.70      gdt      1011: to install NetBSD.
1.52      gdt      1012: A VPS user may want to compile a kernel for security updates, to run
                   1013: npf, run IPsec, or any other reason why someone would want to change
                   1014: their kernel.
1.111     wiz      1016: One approach is to have an administrative interface to upload a kernel,
1.68      gdt      1017: or to select from a prepopulated list.  Other approaches are pygrub
1.59      gdt      1018: (deprecated) and pvgrub, which are ways to have a bootloader obtain a
1.111     wiz      1019: kernel from the domU file system.  This is closer to a regular physical
1.59      gdt      1020: computer, where someone who controls a machine can replace the kernel.
1.52      gdt      1021: 
1.74      gdt      1022: A second issue is multiple CPUs.  With NetBSD 6, domUs support
                   1023: multiple vcpus, and it is typical for VPS providers to enable multiple
                   1024: CPUs for NetBSD domUs.
1.68      gdt      1026: pygrub
1.59      gdt      1027: -------
1.52      gdt      1028: 
1.111     wiz      1029: pygrub runs in the dom0 and looks into the domU file system.  This
                   1030: implies that the domU must have a kernel in a file system in a format
1.68      gdt      1031: known to pygrub.  As of 2014, pygrub seems to be of mostly historical
                   1032: interest.
1.52      gdt      1033: 
1.59      gdt      1034: pvgrub
                   1035: ------
                   1037: pvgrub is a version of grub that uses PV operations instead of BIOS
                   1038: calls.  It is booted from the dom0 as the domU kernel, and then reads
1.111     wiz      1039: /grub/menu.lst and loads a kernel from the domU file system.
1.59      gdt      1040: 
1.70      gdt      1041: [Panix]( lets users use pvgrub.  Panix reports
1.71      gdt      1042: that pvgrub works with FFsv2 with 16K/2K and 32K/4K block/frag sizes
                   1043: (and hence with defaults from "newfs -O 2").  See [Panix's pvgrub
1.70      gdt      1044: page](, which describes only
1.74      gdt      1045: Linux but should be updated to cover NetBSD :-).
1.70      gdt      1046: 
                   1047: []( also lets users with pvgrub to boot
                   1048: their own kernel.  See then [ NetBSD
1.74      gdt      1049: HOWTO](
                   1050: (which is in need of updating).
1.59      gdt      1051: 
1.70      gdt      1052: It appears that [grub's FFS
                   1053: code](
                   1054: does not support all aspects of modern FFS, but there are also reports
1.72      gdt      1055: that FFSv2 works fine.  At prgmr, typically one has an ext2 or FAT
1.70      gdt      1056: partition for the kernel with the intent that grub can understand it,
                   1057: which leads to /netbsd not being the actual kernel.  One must remember
1.111     wiz      1058: to update the special boot partition.
1.59      gdt      1059: 
                   1060: Amazon
                   1061: ------
1.143     wiki     1063: See the [Amazon EC2 page](/amazon_ec2/).
1.44      gdt      1064: 
1.65      gdt      1065: TODO items for improving NetBSD/xen
                   1066: ===================================
1.93      gdt      1068: * Make the NetBSD dom0 kernel work with SMP.
                   1069: * Test the Xen 4.5 packages adequately to be able to recommend them as
                   1070:   the standard approach.
                   1071: * Get PCI passthrough working on Xen 4.5
1.65      gdt      1072: * Get pvgrub into pkgsrc, either via xentools or separately.
                   1073: * grub
1.70      gdt      1074:   * Check/add support to pkgsrc grub2 for UFS2 and arbitrary
1.66      gdt      1075:     fragsize/blocksize (UFS2 support may be present; the point is to
1.111     wiz      1076:     make it so that with any UFS1/UFS2 file system setup that works
1.66      gdt      1077:     with NetBSD grub will also work).
1.140     khorben  1078:     See [pkg/40258](
1.65      gdt      1079:   * Push patches upstream.
                   1080:   * Get UFS2 patches into pvgrub.
                   1081: * Add support for PV ops to a version of /boot, and make it usable as
                   1082:   a kernel in Xen, similar to pvgrub.
1.93      gdt      1083: 
                   1084: Random pointers
                   1085: ===============
1.117     gdt      1087: This section contains links from elsewhere not yet integrated into the
                   1088: HOWTO, and other guides.
1.93      gdt      1089: 
                   1090: *
                   1091: *
1.117     gdt      1092: *

CVSweb for NetBSD wikisrc <> software: FreeBSD-CVSweb