Annotation of wikisrc/ports/xen/howto.mdwn, revision 1.103

1.5       mspo        1: Introduction
1.13      gdt         2: ============
1.1       mspo        3: 
                      4: [![[Xen
1.57      gdt         5: screenshot]](http://www.netbsd.org/gallery/in-Action/hubertf-xens.png)](http://www.netbsd.org/gallery/in-Action/hubertf-xen.png)
1.1       mspo        6: 
1.58      gdt         7: Xen is a hypervisor (or virtual machine monitor) for x86 hardware
1.12      gdt         8: (i686-class or higher), which supports running multiple guest
1.58      gdt         9: operating systems on a single physical machine.  Xen is a Type 1 or
                     10: bare-metal hypervisor; one uses the Xen kernel to control the CPU,
                     11: memory and console, a dom0 operating system which mediates access to
                     12: other hardware (e.g., disks, network, USB), and one or more domU
                     13: operating systems which operate in an unprivileged virtualized
                     14: environment.  IO requests from the domU systems are forwarded by the
                     15: hypervisor (Xen) to the dom0 to be fulfilled.
1.12      gdt        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: 
1.54      gdt        52: NetBSD used to support Xen2; this has been removed.
                     53: 
1.12      gdt        54: Prerequisites
1.13      gdt        55: -------------
1.12      gdt        56: 
                     57: Installing NetBSD/Xen is not extremely difficult, but it is more
                     58: complex than a normal installation of NetBSD.
1.15      gdt        59: In general, this HOWTO is occasionally overly restrictive about how
                     60: things must be done, guiding the reader to stay on the established
                     61: path when there are no known good reasons to stray.
1.12      gdt        62: 
                     63: This HOWTO presumes a basic familiarity with the Xen system
1.16      gdt        64: architecture.  This HOWTO presumes familiarity with installing NetBSD
                     65: on i386/amd64 hardware and installing software from pkgsrc.
1.27      jnemeth    66: See also the [Xen website](http://www.xenproject.org/).
1.1       mspo       67: 
1.15      gdt        68: Versions of Xen and NetBSD
                     69: ==========================
                     70: 
1.27      jnemeth    71: Most of the installation concepts and instructions are independent
                     72: of Xen version and NetBSD version.  This section gives advice on
                     73: which version to choose.  Versions not in pkgsrc and older unsupported
                     74: versions of NetBSD are intentionally ignored.
1.15      gdt        75: 
                     76: Xen
                     77: ---
                     78: 
                     79: In NetBSD, xen is provided in pkgsrc, via matching pairs of packages
                     80: xenkernel and xentools.  We will refer only to the kernel versions,
                     81: but note that both packages must be installed together and must have
                     82: matching versions.
                     83: 
1.102     gdt        84: xenkernel3 provides Xen 3.1.  This no longer receives security patches
                     85: and should not be used.  It supports PCI passthrough, which is why
                     86: people use it anyway. Xen 3.1 supports non-PAE on i386.
                     87: 
                     88: xenkernel33 provides Xen 3.3.  This no longer receives security
                     89: patches and should not be used.  Xen 3.3 supports non-PAE on i386.
1.15      gdt        90: 
                     91: xenkernel41 provides Xen 4.1.  This is no longer maintained by Xen,
1.102     gdt        92: but as of 2014-12 receives backported security patches.  There are no
1.103   ! gdt        93: good reasons to run this version.
1.102     gdt        94: 
                     95: xenkernel42 provides Xen 4.2. This is no longer maintained by Xen, but
                     96: as of 2014-12 receives backported security patches.  The only reason
                     97: to run this is if you need to use xm instead of xl.
1.15      gdt        98: 
1.102     gdt        99: xenkernel45 provides Xen 4.5.  This is new to pkgsrc as of 2015-01 and
                    100: recommended for use as a conservative choice.
1.15      gdt       101: 
1.102     gdt       102: xenkernel46 provides Xen 4.6.  TODO: Probably this is the recommended
                    103: version.
1.85      gdt       104: 
1.96      gdt       105: See also the [Xen Security Advisory page](http://xenbits.xen.org/xsa/).
                    106: 
1.15      gdt       107: Ideally newer versions of Xen will be added to pkgsrc.
                    108: 
1.85      gdt       109: Note that NetBSD support is called XEN3.  It works with Xen 3 and Xen
                    110: 4 because the hypercall interface has been stable.
1.20      gdt       111: 
1.19      gdt       112: Xen command program
                    113: -------------------
                    114: 
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
                    118: xm and xl work fine.  4.4 is the last version that has xm.  You must
                    119: choose one or the other, because it affects which daemons you run.
1.19      gdt       120: 
1.15      gdt       121: NetBSD
                    122: ------
                    123: 
                    124: The netbsd-5, netbsd-6, netbsd-7, and -current branches are all
                    125: reasonable choices, with more or less the same considerations for
                    126: non-Xen use.  Therefore, netbsd-6 is recommended as the stable version
1.29      gdt       127: of the most recent release for production use.  For those wanting to
                    128: learn Xen or without production stability concerns, netbsd-7 is likely
                    129: most appropriate.
1.15      gdt       130: 
                    131: As of NetBSD 6, a NetBSD domU will support multiple vcpus.  There is
                    132: no SMP support for NetBSD as dom0.  (The dom0 itself doesn't really
                    133: need SMP; the lack of support is really a problem when using a dom0 as
                    134: a normal computer.)
                    135: 
1.18      gdt       136: Architecture
                    137: ------------
                    138: 
1.29      gdt       139: Xen itself can run on i386 or amd64 machines.  (Practically, almost
1.99      gdt       140: any computer where one would want to run Xen today supports amd64.)
                    141: 
                    142: Xen, the dom0 kernel, and each domU kernel can be either i386 or
                    143: amd64.  When building a xenkernel package, one obtains i386 on an i386
                    144: host, and amd64 on an amd64 host.  If the xen kernel is i386, then the
                    145: dom0 kernel and all domU kernels must be i386.  With an amd64 xen
                    146: kernel, an amd64 dom0 kernel is known to work, and an i386 dom0 kernel
                    147: should in theory work.  An amd64 xen/dom0 is known to support both
                    148: i386 and amd64 domUs.
                    149: 
1.101     gdt       150: i386 dom0 and domU kernels must be PAE (except for Xen 3.1); these are
                    151: built by default.  (Note that emacs (at least) fails if run on i386
                    152: with PAE when built without, and vice versa, presumably due to bugs in
                    153: the undump code.)
1.99      gdt       154: 
                    155: Because of the above, the standard approach is to use amd64 for the
                    156: dom0.
1.29      gdt       157: 
                    158: Xen 4.2 is the last version to support i386 as a host.  TODO: Clarify
1.100     gdt       159: if this is about the CPU, the xen kernel, or the dom0 kernel having to
                    160: be amd64.
                    161: 
1.29      gdt       162: 
1.89      gdt       163: Stability
                    164: ---------
                    165: 
                    166: Mostly, NetBSD as a dom0 or domU is quite stable.
                    167: However, there are some open PRs indicating problems.
                    168: 
1.91      gdt       169:  - [PR 48125](http://gnats.netbsd.org/48125)
                    170:  - [PR 47720](http://gnats.netbsd.org/47720)
1.89      gdt       171: 
                    172: Note also that there are issues with sparse vnd(4) instances, but
                    173: these are not about Xen.
                    174: 
1.15      gdt       175: Recommendation
                    176: --------------
                    177: 
1.18      gdt       178: Therefore, this HOWTO recommends running xenkernel42 (and xentools42),
1.30      gdt       179: xl, the NetBSD 6 stable branch, and to use an amd64 kernel as the
                    180: dom0.  Either the i386 or amd64 of NetBSD may be used as domUs.
1.15      gdt       181: 
1.36      gdt       182: Build problems
                    183: --------------
                    184: 
                    185: Ideally, all versions of Xen in pkgsrc would build on all versions of
                    186: NetBSD on both i386 and amd64.  However, that isn't the case.  Besides
                    187: aging code and aging compilers, qemu (included in xentools for HVM
1.62      gdt       188: support) is difficult to build.  The following are known to work or FAIL:
1.36      gdt       189: 
1.62      gdt       190:         xenkernel3 netbsd-5 amd64
1.63      gdt       191:         xentools3 netbsd-5 amd64
1.64      gdt       192:         xentools3=hvm netbsd-5 amd64 ????
1.62      gdt       193:         xenkernel33 netbsd-5 amd64
1.63      gdt       194:         xentools33 netbsd-5 amd64
1.36      gdt       195:         xenkernel41 netbsd-5 amd64
                    196:         xentools41 netbsd-5 amd64
1.62      gdt       197:         xenkernel42 netbsd-5 amd64
1.64      gdt       198:         xentools42 netbsd-5 amd64
1.62      gdt       199: 
                    200:         xenkernel3 netbsd-6 i386 FAIL
                    201:         xentools3 netbsd-6 i386
1.63      gdt       202:         xentools3-hvm netbsd-6 i386 FAIL (dependencies fail)
                    203:         xenkernel33 netbsd-6 i386
                    204:         xentools33 netbsd-6 i386
1.36      gdt       205:         xenkernel41 netbsd-6 i386
                    206:         xentools41 netbsd-6 i386
1.63      gdt       207:         xenkernel42 netbsd-6 i386
1.64      gdt       208:         xentools42 netbsd-6 i386 *MIXED
                    209: 
1.69      gdt       210:        (all 3 and 33 seem to FAIL)
                    211:         xenkernel41 netbsd-7 i386
                    212:         xentools41 netbsd-7 i386
                    213:         xenkernel42 netbsd-7 i386
                    214:         xentools42 netbsd-7 i386 ??FAIL
                    215: 
1.64      gdt       216: (*On netbsd-6 i386, there is a xentools42 in the 2014Q3 official builds,
                    217: but it does not build for gdt.)
1.36      gdt       218: 
1.15      gdt       219: NetBSD as a dom0
                    220: ================
                    221: 
                    222: NetBSD can be used as a dom0 and works very well.  The following
                    223: sections address installation, updating NetBSD, and updating Xen.
1.19      gdt       224: Note that it doesn't make sense to talk about installing a dom0 OS
                    225: without also installing Xen itself.  We first address installing
                    226: NetBSD, which is not yet a dom0, and then adding Xen, pivoting the
                    227: NetBSD install to a dom0 install by just changing the kernel and boot
                    228: configuration.
1.15      gdt       229: 
1.45      gdt       230: For experimenting with Xen, a machine with as little as 1G of RAM and
                    231: 100G of disk can work.  For running many domUs in productions, far
                    232: more will be needed.
                    233: 
1.15      gdt       234: Styles of dom0 operation
                    235: ------------------------
                    236: 
                    237: There are two basic ways to use Xen.  The traditional method is for
                    238: the dom0 to do absolutely nothing other than providing support to some
                    239: number of domUs.  Such a system was probably installed for the sole
                    240: purpose of hosting domUs, and sits in a server room on a UPS.
                    241: 
                    242: The other way is to put Xen under a normal-usage computer, so that the
                    243: dom0 is what the computer would have been without Xen, perhaps a
                    244: desktop or laptop.  Then, one can run domUs at will.  Purists will
                    245: deride this as less secure than the previous approach, and for a
                    246: computer whose purpose is to run domUs, they are right.  But Xen and a
1.93      gdt       247: dom0 (without domUs) is not meaningfully less secure than the same
1.15      gdt       248: things running without Xen.  One can boot Xen or boot regular NetBSD
                    249: alternately with little problems, simply refraining from starting the
                    250: Xen daemons when not running Xen.
                    251: 
                    252: Note that NetBSD as dom0 does not support multiple CPUs.  This will
1.51      gdt       253: limit the performance of the Xen/dom0 workstation approach.  In theory
                    254: the only issue is that the "backend drivers" are not yet MPSAFE:
                    255:   http://mail-index.netbsd.org/netbsd-users/2014/08/29/msg015195.html
1.15      gdt       256: 
1.19      gdt       257: Installation of NetBSD
                    258: ----------------------
1.13      gdt       259: 
1.19      gdt       260: First,
1.27      jnemeth   261: [install NetBSD/amd64](/guide/inst/)
1.19      gdt       262: just as you would if you were not using Xen.
                    263: However, the partitioning approach is very important.
                    264: 
                    265: If you want to use RAIDframe for the dom0, there are no special issues
                    266: for Xen.  Typically one provides RAID storage for the dom0, and the
1.22      gdt       267: domU systems are unaware of RAID.  The 2nd-stage loader bootxx_* skips
                    268: over a RAID1 header to find /boot from a filesystem within a RAID
                    269: partition; this is no different when booting Xen.
1.19      gdt       270: 
                    271: There are 4 styles of providing backing storage for the virtual disks
1.93      gdt       272: used by domUs: raw partitions, LVM, file-backed vnd(4), and SAN.
1.19      gdt       273: 
                    274: With raw partitions, one has a disklabel (or gpt) partition sized for
                    275: each virtual disk to be used by the domU.  (If you are able to predict
                    276: how domU usage will evolve, please add an explanation to the HOWTO.
                    277: Seriously, needs tend to change over time.)
                    278: 
1.27      jnemeth   279: One can use [lvm(8)](/guide/lvm/) to create logical devices to use
                    280: for domU disks.  This is almost as efficient as raw disk partitions
                    281: and more flexible.  Hence raw disk partitions should typically not
                    282: be used.
1.19      gdt       283: 
                    284: One can use files in the dom0 filesystem, typically created by dd'ing
                    285: /dev/zero to create a specific size.  This is somewhat less efficient,
                    286: but very convenient, as one can cp the files for backup, or move them
                    287: between dom0 hosts.
                    288: 
                    289: Finally, in theory one can place the files backing the domU disks in a
                    290: SAN.  (This is an invitation for someone who has done this to add a
                    291: HOWTO page.)
1.1       mspo      292: 
1.19      gdt       293: Installation of Xen
                    294: -------------------
1.1       mspo      295: 
1.20      gdt       296: In the dom0, install sysutils/xenkernel42 and sysutils/xentools42 from
                    297: pkgsrc (or another matching pair).
                    298: See [the pkgsrc
                    299: documentation](http://www.NetBSD.org/docs/pkgsrc/) for help with pkgsrc.
                    300: 
                    301: For Xen 3.1, support for HVM guests is in sysutils/xentool3-hvm.  More
                    302: recent versions have HVM support integrated in the main xentools
                    303: package.  It is entirely reasonable to run only PV guests.
                    304: 
                    305: Next you need to install the selected Xen kernel itself, which is
                    306: installed by pkgsrc as "/usr/pkg/xen*-kernel/xen.gz".  Copy it to /.
                    307: For debugging, one may copy xen-debug.gz; this is conceptually similar
                    308: to DIAGNOSTIC and DEBUG in NetBSD.  xen-debug.gz is basically only
                    309: useful with a serial console.  Then, place a NetBSD XEN3_DOM0 kernel
                    310: in /, copied from releasedir/amd64/binary/kernel/netbsd-XEN3_DOM0.gz
1.75      gdt       311: of a NetBSD build.  If using i386, use
                    312: releasedir/i386/binary/kernel/netbsd-XEN3PAE_DOM0.gz.  (If using Xen
                    313: 3.1 and i386, you may use XEN3_DOM0 with the non-PAE Xen.  But you
                    314: should not use Xen 3.1.)  Both xen and the NetBSD kernel may be (and
                    315: typically are) left compressed.
                    316: 
                    317: In a dom0 kernel, kernfs is mandatory for xend to comunicate with the
                    318: kernel, so ensure that /kern is in fstab.  TODO: Say this is default,
                    319: or file a PR and give a reference.
1.20      gdt       320: 
                    321: Because you already installed NetBSD, you have a working boot setup
                    322: with an MBR bootblock, either bootxx_ffsv1 or bootxx_ffsv2 at the
                    323: beginning of your root filesystem, /boot present, and likely
                    324: /boot.cfg.  (If not, fix before continuing!)
                    325: 
1.76      gdt       326: Add a line to to /boot.cfg to boot Xen.  See boot.cfg(5) for an
                    327: example.  The basic line is
1.20      gdt       328: 
1.37      gdt       329:         menu=Xen:load /netbsd-XEN3_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=256M
1.20      gdt       330: 
                    331: which specifies that the dom0 should have 256M, leaving the rest to be
1.77      gdt       332: allocated for domUs.  To use a serial console, use
                    333: 
                    334:         menu=Xen:load /netbsd-XEN3_DOM0.gz console=com0;multiboot /xen.gz dom0_mem=256M console=com1 com1=9600,8n1
                    335: 
                    336: which will use the first serial port for Xen (which counts starting
                    337: from 1), forcing speed/parity, and also for NetBSD (which counts
                    338: starting at 0).  In an attempt to add performance, one can also add
1.37      gdt       339: 
                    340:         dom0_max_vcpus=1 dom0_vcpus_pin
                    341: 
                    342: to force only one vcpu to be provided (since NetBSD dom0 can't use
                    343: more) and to pin that vcpu to a physical cpu.  TODO: benchmark this.
1.20      gdt       344: 
1.93      gdt       345: Xen has [many boot
                    346: options](http://xenbits.xenproject.org/docs/4.5-testing/misc/xen-command-line.html),
                    347: and other tham dom0 memory and max_vcpus, they are generally not
                    348: necessary.
                    349: 
1.20      gdt       350: As with non-Xen systems, you should have a line to boot /netbsd (a
                    351: kernel that works without Xen) and fallback versions of the non-Xen
                    352: kernel, Xen, and the dom0 kernel.
1.1       mspo      353: 
1.76      gdt       354: Now, reboot so that you are running a DOM0 kernel under Xen, rather
                    355: than GENERIC without Xen.
                    356: 
1.54      gdt       357: Using grub (historic)
                    358: ---------------------
                    359: 
                    360: Before NetBSD's native bootloader could support Xen, the use of
                    361: grub was recommended.  If necessary, see the
                    362: [old grub information](/ports/xen/howto-grub/).
                    363: 
1.28      gdt       364: The [HowTo on Installing into
                    365: RAID-1](http://mail-index.NetBSD.org/port-xen/2006/03/01/0010.html)
                    366: explains how to set up booting a dom0 with Xen using grub with
                    367: NetBSD's RAIDframe.  (This is obsolete with the use of NetBSD's native
                    368: boot.)
                    369: 
1.21      gdt       370: Configuring Xen
                    371: ---------------
                    372: 
1.53      gdt       373: Xen logs will be in /var/log/xen.
                    374: 
1.76      gdt       375: Now, you have a system that will boot Xen and the dom0 kernel, but not
                    376: do anything else special.  Make sure that you have rebooted into Xen.
                    377: There will be no domUs, and none can be started because you still have
1.102     gdt       378: to configure the dom0 daemons.
1.21      gdt       379: 
1.102     gdt       380: The daemons which should be run vary with Xen version and with whether
                    381: one is using xm or xl.  The Xen 3.1 and 3.3 packages use xm.  Xen 4.1
                    382: and higher packages use xl.  While is is possible to use xm with some
                    383: 4.x versions (TODO: 4.1 and 4.2?), the pkgsrc-provided rc.d scripts do
                    384: not support this as of 2014-12-26, and thus the HOWTO does not support
                    385: it either.  (Make sure your packages are reasonably recent.)
1.43      gdt       386: 
1.102     gdt       387: For "xm" (3.1 and 3.3), you should enable xend and xenbackendd (but
                    388: note that you should be using 4.x):
1.31      gdt       389: 
1.32      gdt       390:         xend=YES
                    391:         xenbackendd=YES
1.31      gdt       392: 
1.102     gdt       393: For "xl" (4.x), you should enabled xend and xencommons (xenstored).
                    394: Trying to boot 4.x without xencommons=YES will result in a hang; it is
                    395: necessary to hig ^C on the console to let the machine finish booting.
                    396: TODO: explain why xend is installed by the package.
1.31      gdt       397: 
1.53      gdt       398:         xencommons=YES
1.31      gdt       399: 
1.102     gdt       400: The installation of NetBSD should already have created devices for xen
                    401: (xencons, xenevt), but if they are not present, create them:
1.31      gdt       402: 
1.102     gdt       403:         cd /dev && sh MAKEDEV xen
1.86      gdt       404: 
1.31      gdt       405: TODO: Recommend for/against xen-watchdog.
1.27      jnemeth   406: 
1.53      gdt       407: After you have configured the daemons and either started them (in the
1.79      gdt       408: order given) or rebooted, use xm or xl to inspect Xen's boot messages,
1.102     gdt       409: available resources, and running domains.  An example with xl follows:
1.34      gdt       410: 
1.102     gdt       411:         # xl dmesg
1.43      gdt       412:        [xen's boot info]
1.102     gdt       413:         # xl info
1.43      gdt       414:        [available memory, etc.]
1.102     gdt       415:         # xl list
1.43      gdt       416:         Name              Id  Mem(MB)  CPU  State  Time(s)  Console
                    417:         Domain-0           0       64    0  r----     58.1
1.33      gdt       418: 
1.88      gdt       419: ### Issues with xencommons
                    420: 
                    421: xencommons starts xenstored, which stores data on behalf of dom0 and
                    422: domUs.  It does not currently work to stop and start xenstored.
                    423: Certainly all domUs should be shutdown first, following the sort order
                    424: of the rc.d scripts.  However, the dom0 sets up state with xenstored,
                    425: and is not notified when xenstored exits, leading to not recreating
                    426: the state when the new xenstored starts.  Until there's a mechanism to
                    427: make this work, one should not expect to be able to restart xenstored
                    428: (and thus xencommons).  There is currently no reason to expect that
                    429: this will get fixed any time soon.
                    430: 
1.41      gdt       431: anita (for testing NetBSD)
                    432: --------------------------
                    433: 
1.82      gdt       434: With the setup so far (assuming 4.2/xl), one should be able to run
                    435: anita (see pkgsrc/misc/py-anita) to test NetBSD releases, by doing (as
                    436: root, because anita must create a domU):
                    437: 
                    438:         anita --vmm=xl test file:///usr/obj/i386/
                    439: 
                    440: Alternatively, one can use --vmm=xm to use xm-based domU creation
                    441: instead (and must, on Xen <= 4.1).   TODO: confirm that anita xl really works.
                    442:     
1.40      gdt       443: Xen-specific NetBSD issues
                    444: --------------------------
                    445: 
                    446: There are (at least) two additional things different about NetBSD as a
                    447: dom0 kernel compared to hardware.
                    448: 
                    449: One is that modules are not usable in DOM0 kernels, so one must
                    450: compile in what's needed.  It's not really that modules cannot work,
                    451: but that modules must be built for XEN3_DOM0 because some of the
                    452: defines change and the normal module builds don't do this.  Basically,
                    453: enabling Xen changes the kernel ABI, and the module build system
                    454: doesn't cope with this.
                    455: 
                    456: The other difference is that XEN3_DOM0 does not have exactly the same
                    457: options as GENERIC.  While it is debatable whether or not this is a
                    458: bug, users should be aware of this and can simply add missing config
                    459: items if desired.
                    460: 
1.15      gdt       461: Updating NetBSD in a dom0
                    462: -------------------------
                    463: 
                    464: This is just like updating NetBSD on bare hardware, assuming the new
                    465: version supports the version of Xen you are running.  Generally, one
                    466: replaces the kernel and reboots, and then overlays userland binaries
                    467: and adjusts /etc.
                    468: 
                    469: Note that one must update both the non-Xen kernel typically used for
                    470: rescue purposes and the DOM0 kernel used with Xen.
                    471: 
1.55      gdt       472: Converting from grub to /boot
                    473: -----------------------------
                    474: 
                    475: These instructions were [TODO: will be] used to convert a system from
                    476: grub to /boot.  The system was originally installed in February of
                    477: 2006 with a RAID1 setup and grub to boot Xen 2, and has been updated
                    478: over time.  Before these commands, it was running NetBSD 6 i386, Xen
                    479: 4.1 and grub, much like the message linked earlier in the grub
                    480: section.
                    481: 
                    482:         # Install mbr bootblocks on both disks. 
                    483:         fdisk -i /dev/rwd0d
                    484:         fdisk -i /dev/rwd1d
                    485:         # Install NetBSD primary boot loader (/ is FFSv1) into RAID1 components.
                    486:         installboot -v /dev/rwd0d /usr/mdec/bootxx_ffsv1
                    487:         installboot -v /dev/rwd1d /usr/mdec/bootxx_ffsv1
                    488:         # Install secondary boot loader
                    489:         cp -p /usr/mdec/boot /
                    490:         # Create boog.cfg following earlier guidance:
                    491:         menu=Xen:load /netbsd-XEN3PAE_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=256M
                    492:         menu=Xen.ok:load /netbsd-XEN3PAE_DOM0.ok.gz console=pc;multiboot /xen.ok.gz dom0_mem=256M
                    493:         menu=GENERIC:boot
                    494:         menu=GENERIC single-user:boot -s
                    495:         menu=GENERIC.ok:boot netbsd.ok
                    496:         menu=GENERIC.ok single-user:boot netbsd.ok -s
                    497:         menu=Drop to boot prompt:prompt
                    498:         default=1
                    499:         timeout=30
                    500: 
                    501: TODO: actually do this and fix it if necessary.
1.22      gdt       502: 
1.102     gdt       503: Upgrading Xen versions
1.15      gdt       504: ---------------------
                    505: 
1.21      gdt       506: Updating Xen is conceptually not difficult, but can run into all the
                    507: issues found when installing Xen.  Assuming migration from 4.1 to 4.2,
                    508: remove the xenkernel41 and xentools41 packages and install the
                    509: xenkernel42 and xentools42 packages.  Copy the 4.2 xen.gz to /.
                    510: 
1.102     gdt       511: Ensure that the contents of /etc/rc.d/xen* are correct.  Specifically,
                    512: they must match the package you just installed and not be left over
                    513: from some previous installation.
                    514: 
                    515: Enable the correct set of daemons; see the configuring section above.
                    516: (Upgrading from 3.x to 4.x without doing this will result in a hang.)
                    517: 
                    518: Ensure that the domU config files are valid for the new version.
                    519: Specifically: remove autorestart=True, and ensure that disks are
                    520: specified with numbers as the second argument, as the examples above
                    521: show, and not NetBSD device names.
1.15      gdt       522: 
1.97      gdt       523: Hardware known to work
                    524: ----------------------
                    525: 
                    526: Arguably, this section is misplaced, and there should be a page of
                    527: hardware that runs NetBSD/amd64 well, with the mostly-well-founded
                    528: assumption that NetBSD/xen runs fine on any modern hardware that
                    529: NetBSD/amd64 runs well on.  Until then, we give motherboard/CPU/RAM
                    530: triples to aid those choosing a motherboard.  Note that Xen systems
                    531: usually do not run X, so a listing here does not imply that X works at
                    532: all.
                    533: 
                    534:         Supermicro X9SRL-F, Xeon E5-1650 v2, 96 GiB ECC
                    535:         Supermicro ??, Atom C2758 (8 core), 32 GiB ECC
                    536:         ASUS M5A78L-M/USB3 AM3+ microATX, AMD Piledriver X8 4000MHz, 16 GiB ECC
                    537: 
                    538: Older hardware:
                    539: 
1.98      gdt       540:         Intel D915GEV, Pentium4 CPU 3.40GHz, 4GB 533MHz Synchronous DDR2
1.28      gdt       541: 
1.82      gdt       542: Running Xen under qemu
                    543: ----------------------
                    544: 
                    545: The astute reader will note that this section is somewhat twisted.
                    546: However, it can be useful to run Xen under qemu either because the
                    547: version of NetBSD as a dom0 does not run on the hardware in use, or to
                    548: generate automated test cases involving Xen.
                    549: 
1.84      gdt       550: In 2015-01, the following combination was reported to mostly work:
1.82      gdt       551: 
                    552:         host OS: NetBSD/amd64 6.1.4
                    553:         qemu: 2.2.0 from pkgsrc
                    554:         Xen kernel: xenkernel42-4.2.5nb1 from pkgsrc
                    555:         dom0 kernel: NetBSD/amd64 6.1.5
                    556:         Xen tools: xentools42-4.2.5 from pkgsrc
                    557: 
1.91      gdt       558: See [PR 47720](http://gnats.netbsd.org/47720) for a problem with dom0
                    559: shutdown.
1.84      gdt       560: 
1.28      gdt       561: Unprivileged domains (domU)
                    562: ===========================
                    563: 
                    564: This section describes general concepts about domUs.  It does not
1.33      gdt       565: address specific domU operating systems or how to install them.  The
                    566: config files for domUs are typically in /usr/pkg/etc/xen, and are
1.60      wiki      567: typically named so that the file name, domU name and the domU's host
1.33      gdt       568: name match.
                    569: 
                    570: The domU is provided with cpu and memory by Xen, configured by the
                    571: dom0.  The domU is provided with disk and network by the dom0,
                    572: mediated by Xen, and configured in the dom0.
                    573: 
                    574: Entropy in domUs can be an issue; physical disks and network are on
                    575: the dom0.  NetBSD's /dev/random system works, but is often challenged.
                    576: 
1.48      gdt       577: Config files
                    578: ------------
                    579: 
                    580: There is no good order to present config files and the concepts
                    581: surrounding what is being configured.  We first show an example config
                    582: file, and then in the various sections give details.
                    583: 
                    584: See (at least in xentools41) /usr/pkg/share/examples/xen/xmexample*,
                    585: for a large number of well-commented examples, mostly for running
                    586: GNU/Linux.
                    587: 
                    588: The following is an example minimal domain configuration file
                    589: "/usr/pkg/etc/xen/foo".  It is (with only a name change) an actual
                    590: known working config file on Xen 4.1 (NetBSD 5 amd64 dom0 and NetBSD 5
                    591: i386 domU).  The domU serves as a network file server.
                    592: 
                    593:         # -*- mode: python; -*-
                    594: 
                    595:         kernel = "/netbsd-XEN3PAE_DOMU-i386-foo.gz"
                    596:         memory = 1024
                    597:         vif = [ 'mac=aa:00:00:d1:00:09,bridge=bridge0' ]
                    598:         disk = [ 'file:/n0/xen/foo-wd0,0x0,w',
                    599:                  'file:/n0/xen/foo-wd1,0x1,w' ]
                    600: 
                    601: The domain will have the same name as the file.  The kernel has the
                    602: host/domU name in it, so that on the dom0 one can update the various
                    603: domUs independently.  The vif line causes an interface to be provided,
                    604: with a specific mac address (do not reuse MAC addresses!), in bridge
                    605: mode.  Two disks are provided, and they are both writable; the bits
                    606: are stored in files and Xen attaches them to a vnd(4) device in the
                    607: dom0 on domain creation.  The system treates xbd0 as the boot device
                    608: without needing explicit configuration.
                    609: 
                    610: By default xm looks for domain config files in /usr/pkg/etc/xen.  Note
                    611: that "xm create" takes the name of a config file, while other commands
                    612: take the name of a domain.  To create the domain, connect to the
                    613: console, create the domain while attaching the console, shutdown the
                    614: domain, and see if it has finished stopping, do (or xl with Xen >=
                    615: 4.2):
                    616: 
                    617:         xm create foo
                    618:         xm console foo
                    619:         xm create -c foo
                    620:         xm shutdown foo
1.90      gdt       621:         xm list
1.48      gdt       622: 
                    623: Typing ^] will exit the console session.  Shutting down a domain is
                    624: equivalent to pushing the power button; a NetBSD domU will receive a
                    625: power-press event and do a clean shutdown.  Shutting down the dom0
                    626: will trigger controlled shutdowns of all configured domUs.
                    627: 
                    628: domU kernels
                    629: ------------
                    630: 
                    631: On a physical computer, the BIOS reads sector 0, and a chain of boot
                    632: loaders finds and loads a kernel.  Normally this comes from the root
                    633: filesystem.  With Xen domUs, the process is totally different.  The
                    634: normal path is for the domU kernel to be a file in the dom0's
                    635: filesystem.  At the request of the dom0, Xen loads that kernel into a
                    636: new domU instance and starts execution.  While domU kernels can be
                    637: anyplace, reasonable places to store domU kernels on the dom0 are in /
                    638: (so they are near the dom0 kernel), in /usr/pkg/etc/xen (near the
                    639: config files), or in /u0/xen (where the vdisks are).
                    640: 
1.59      gdt       641: Note that loading the domU kernel from the dom0 implies that boot
                    642: blocks, /boot, /boot.cfg, and so on are all ignored in the domU.
1.48      gdt       643: See the VPS section near the end for discussion of alternate ways to
                    644: obtain domU kernels.
                    645: 
1.33      gdt       646: CPU and memory
                    647: --------------
                    648: 
1.48      gdt       649: A domain is provided with some number of vcpus, less than the number
                    650: of cpus seen by the hypervisor.  (For a dom0, this is controlled by
                    651: the boot argument "dom0_max_vcpus=1".)  For a domU, it is controlled
                    652: from the config file by the "vcpus = N" directive.
                    653: 
                    654: A domain is provided with memory; this is controlled in the config
                    655: file by "memory = N" (in megabytes).  In the straightforward case, the
                    656: sum of the the memory allocated to the dom0 and all domUs must be less
1.33      gdt       657: than the available memory.
                    658: 
                    659: Xen also provides a "balloon" driver, which can be used to let domains
                    660: use more memory temporarily.  TODO: Explain better, and explain how
                    661: well it works with NetBSD.
1.28      gdt       662: 
                    663: Virtual disks
                    664: -------------
                    665: 
1.33      gdt       666: With the file/vnd style, typically one creates a directory,
                    667: e.g. /u0/xen, on a disk large enough to hold virtual disks for all
                    668: domUs.  Then, for each domU disk, one writes zeros to a file that then
                    669: serves to hold the virtual disk's bits; a suggested name is foo-xbd0
                    670: for the first virtual disk for the domU called foo.  Writing zeros to
                    671: the file serves two purposes.  One is that preallocating the contents
                    672: improves performance.  The other is that vnd on sparse files has
                    673: failed to work.  TODO: give working/notworking NetBSD versions for
                    674: sparse vnd.  Note that the use of file/vnd for Xen is not really
                    675: different than creating a file-backed virtual disk for some other
1.39      gdt       676: purpose, except that xentools handles the vnconfig commands.  To
                    677: create an empty 4G virtual disk, simply do
                    678: 
                    679:         dd if=/dev/zero of=foo-xbd0 bs=1m count=4096
1.33      gdt       680: 
1.89      gdt       681: Do not use qemu-img-xen, because this will create sparse file.  There
                    682: have been recent (2015) reports of sparse vnd(4) devices causing
                    683: lockups, but there is apparently no PR.
                    684: 
1.33      gdt       685: With the lvm style, one creates logical devices.  They are then used
1.48      gdt       686: similarly to vnds.  TODO: Add an example with lvm.
                    687: 
                    688: In domU config files, the disks are defined as a sequence of 3-tuples.
                    689: The first element is "method:/path/to/disk".  Common methods are
                    690: "file:" for file-backed vnd. and "phy:" for something that is already
                    691: a (TODO: character or block) device.
                    692: 
                    693: The second element is an artifact of how virtual disks are passed to
                    694: Linux, and a source of confusion with NetBSD Xen usage.  Linux domUs
                    695: are given a device name to associate with the disk, and values like
                    696: "hda1" or "sda1" are common.  In a NetBSD domU, the first disk appears
                    697: as xbd0, the second as xbd1, and so on.  However, xm/xl demand a
                    698: second argument.  The name given is converted to a major/minor by
1.49      gdt       699: calling stat(2) on the name in /dev and this is passed to the domU.
                    700: In the general case, the dom0 and domU can be different operating
1.48      gdt       701: systems, and it is an unwarranted assumption that they have consistent
                    702: numbering in /dev, or even that the dom0 OS has a /dev.  With NetBSD
                    703: as both dom0 and domU, using values of 0x0 for the first disk and 0x1
1.49      gdt       704: for the second works fine and avoids this issue.  For a GNU/Linux
                    705: guest, one can create /dev/hda1 in /dev, or to pass 0x301 for
                    706: /dev/hda1.
1.48      gdt       707: 
                    708: The third element is "w" for writable disks, and "r" for read-only
                    709: disks.
1.28      gdt       710: 
                    711: Virtual Networking
                    712: ------------------
                    713: 
1.46      gdt       714: Xen provides virtual ethernets, each of which connects the dom0 and a
                    715: domU.  For each virtual network, there is an interface "xvifN.M" in
                    716: the dom0, and in domU index N, a matching interface xennetM (NetBSD
                    717: name).  The interfaces behave as if there is an Ethernet with two
                    718: adaptors connected.  From this primitive, one can construct various
                    719: configurations.  We focus on two common and useful cases for which
                    720: there are existing scripts: bridging and NAT.
1.28      gdt       721: 
1.48      gdt       722: With bridging (in the example above), the domU perceives itself to be
                    723: on the same network as the dom0.  For server virtualization, this is
                    724: usually best.  Bridging is accomplished by creating a bridge(4) device
                    725: and adding the dom0's physical interface and the various xvifN.0
                    726: interfaces to the bridge.  One specifies "bridge=bridge0" in the domU
                    727: config file.  The bridge must be set up already in the dom0; an
                    728: example /etc/ifconfig.bridge0 is:
1.46      gdt       729: 
                    730:         create
                    731:         up
                    732:         !brconfig bridge0 add wm0
1.28      gdt       733: 
                    734: With NAT, the domU perceives itself to be behind a NAT running on the
                    735: dom0.  This is often appropriate when running Xen on a workstation.
1.48      gdt       736: TODO: NAT appears to be configured by "vif = [ '' ]".
1.28      gdt       737: 
1.49      gdt       738: The MAC address specified is the one used for the interface in the new
1.53      gdt       739: domain.  The interface in dom0 will use this address XOR'd with
1.49      gdt       740: 00:00:00:01:00:00.  Random MAC addresses are assigned if not given.
                    741: 
1.33      gdt       742: Sizing domains
                    743: --------------
                    744: 
                    745: Modern x86 hardware has vast amounts of resources.  However, many
                    746: virtual servers can function just fine on far less.  A system with
                    747: 256M of RAM and a 4G disk can be a reasonable choice.  Note that it is
                    748: far easier to adjust virtual resources than physical ones.  For
                    749: memory, it's just a config file edit and a reboot.  For disk, one can
                    750: create a new file and vnconfig it (or lvm), and then dump/restore,
                    751: just like updating physical disks, but without having to be there and
                    752: without those pesky connectors.
                    753: 
1.48      gdt       754: Starting domains automatically
                    755: ------------------------------
1.28      gdt       756: 
1.48      gdt       757: To start domains foo at bar at boot and shut them down cleanly on dom0
                    758: shutdown, in rc.conf add:
1.28      gdt       759: 
1.48      gdt       760:         xendomains="foo bar"
1.28      gdt       761: 
1.86      gdt       762: Note that earlier versions of the xentools41 xendomains rc.d scripth
                    763: usd xl, when one should use xm with 4.1.
1.28      gdt       764: 
                    765: Creating specific unprivileged domains (domU)
                    766: =============================================
1.14      gdt       767: 
                    768: Creating domUs is almost entirely independent of operating system.  We
1.49      gdt       769: have already presented the basics of config files.  Note that you must
                    770: have already completed the dom0 setup so that "xl list" (or "xm list")
                    771: works.
1.14      gdt       772: 
                    773: Creating an unprivileged NetBSD domain (domU)
                    774: ---------------------------------------------
1.1       mspo      775: 
1.49      gdt       776: See the earlier config file, and adjust memory.  Decide on how much
                    777: storage you will provide, and prepare it (file or lvm).
                    778: 
                    779: While the kernel will be obtained from the dom0 filesystem, the same
                    780: file should be present in the domU as /netbsd so that tools like
                    781: savecore(8) can work.   (This is helpful but not necessary.)
                    782: 
                    783: The kernel must be specifically for Xen and for use as a domU.  The
                    784: i386 and amd64 provide the following kernels:
                    785: 
                    786:         i386 XEN3_DOMU
                    787:         i386 XEN3PAE_DOMU
1.95      gdt       788:         amd64 XEN3_DOMU
1.5       mspo      789: 
1.49      gdt       790: Unless using Xen 3.1 (and you shouldn't) with i386-mode Xen, you must
                    791: use the PAE version of the i386 kernel.
                    792: 
                    793: This will boot NetBSD, but this is not that useful if the disk is
                    794: empty.  One approach is to unpack sets onto the disk outside of xen
                    795: (by mounting it, just as you would prepare a physical disk for a
                    796: system you can't run the installer on).
                    797: 
                    798: A second approach is to run an INSTALL kernel, which has a miniroot
                    799: and can load sets from the network.  To do this, copy the INSTALL
                    800: kernel to / and change the kernel line in the config file to:
1.5       mspo      801: 
1.49      gdt       802:         kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
1.5       mspo      803: 
1.49      gdt       804: Then, start the domain as "xl create -c configname".
1.1       mspo      805: 
1.49      gdt       806: Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following
                    807: line should be used in the config file.
1.1       mspo      808: 
1.3       mspo      809:     disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
1.1       mspo      810: 
                    811: After booting the domain, the option to install via CDROM may be
1.49      gdt       812: selected.  The CDROM device should be changed to `xbd1d`.
1.1       mspo      813: 
1.49      gdt       814: Once done installing, "halt -p" the new domain (don't reboot or halt,
                    815: it would reload the INSTALL_XEN3_DOMU kernel even if you changed the
                    816: config file), switch the config file back to the XEN3_DOMU kernel,
                    817: and start the new domain again. Now it should be able to use "root on
                    818: xbd0a" and you should have a, functional NetBSD domU.
1.1       mspo      819: 
1.49      gdt       820: TODO: check if this is still accurate.
1.1       mspo      821: When the new domain is booting you'll see some warnings about *wscons*
                    822: and the pseudo-terminals. These can be fixed by editing the files
1.5       mspo      823: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
                    824: `/etc/ttys`, except *console*, like this:
1.1       mspo      825: 
1.3       mspo      826:     console "/usr/libexec/getty Pc"         vt100   on secure
                    827:     ttyE0   "/usr/libexec/getty Pc"         vt220   off secure
                    828:     ttyE1   "/usr/libexec/getty Pc"         vt220   off secure
                    829:     ttyE2   "/usr/libexec/getty Pc"         vt220   off secure
                    830:     ttyE3   "/usr/libexec/getty Pc"         vt220   off secure
1.1       mspo      831: 
1.5       mspo      832: Finally, all screens must be commented out from `/etc/wscons.conf`.
1.1       mspo      833: 
                    834: It is also desirable to add
                    835: 
1.49      gdt       836:         powerd=YES
1.1       mspo      837: 
1.5       mspo      838: in rc.conf. This way, the domain will be properly shut down if
1.53      gdt       839: `xm shutdown -R` or `xm shutdown -H` is used on the dom0.
1.1       mspo      840: 
1.92      gdt       841: It is not strictly necessary to have a kernel (as /netbsd) in the domU
                    842: filesystem.  However, various programs (e.g. netstat) will use that
                    843: kernel to look up symbols to read from kernel virtual memory.  If
                    844: /netbsd is not the running kernel, those lookups will fail.  (This is
                    845: not really a Xen-specific issue, but because the domU kernel is
                    846: obtained from the dom0, it is far more likely to be out of sync or
                    847: missing with Xen.)
                    848: 
1.14      gdt       849: Creating an unprivileged Linux domain (domU)
1.5       mspo      850: --------------------------------------------
1.1       mspo      851: 
                    852: Creating unprivileged Linux domains isn't much different from
                    853: unprivileged NetBSD domains, but there are some details to know.
                    854: 
                    855: First, the second parameter passed to the disk declaration (the '0x1' in
                    856: the example below)
                    857: 
1.3       mspo      858:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
1.1       mspo      859: 
                    860: does matter to Linux. It wants a Linux device number here (e.g. 0x300
1.49      gdt       861: for hda).  Linux builds device numbers as: (major \<\< 8 + minor).
                    862: So, hda1 which has major 3 and minor 1 on a Linux system will have
                    863: device number 0x301.  Alternatively, devices names can be used (hda,
                    864: hdb, ...)  as xentools has a table to map these names to devices
                    865: numbers.  To export a partition to a Linux guest we can use:
1.1       mspo      866: 
1.49      gdt       867:         disk = [ 'phy:/dev/wd0e,0x300,w' ]
                    868:         root = "/dev/hda1 ro"
1.1       mspo      869: 
                    870: and it will appear as /dev/hda on the Linux system, and be used as root
                    871: partition.
                    872: 
1.49      gdt       873: To install the Linux system on the partition to be exported to the
                    874: guest domain, the following method can be used: install
                    875: sysutils/e2fsprogs from pkgsrc.  Use mke2fs to format the partition
                    876: that will be the root partition of your Linux domain, and mount it.
                    877: Then copy the files from a working Linux system, make adjustments in
                    878: `/etc` (fstab, network config).  It should also be possible to extract
                    879: binary packages such as .rpm or .deb directly to the mounted partition
                    880: using the appropriate tool, possibly running under NetBSD's Linux
                    881: emulation.  Once the filesystem has been populated, umount it.  If
                    882: desirable, the filesystem can be converted to ext3 using tune2fs -j.
                    883: It should now be possible to boot the Linux guest domain, using one of
                    884: the vmlinuz-\*-xenU kernels available in the Xen binary distribution.
1.1       mspo      885: 
                    886: To get the linux console right, you need to add:
                    887: 
1.3       mspo      888:     extra = "xencons=tty1"
1.1       mspo      889: 
                    890: to your configuration since not all linux distributions auto-attach a
                    891: tty to the xen console.
                    892: 
1.14      gdt       893: Creating an unprivileged Solaris domain (domU)
1.5       mspo      894: ----------------------------------------------
1.1       mspo      895: 
1.50      gdt       896: See possibly outdated
                    897: [Solaris domU instructions](/ports/xen/howto-solaris/).
1.5       mspo      898: 
1.1       mspo      899: 
1.52      gdt       900: PCI passthrough: Using PCI devices in guest domains
                    901: ---------------------------------------------------
1.1       mspo      902: 
1.53      gdt       903: The dom0 can give other domains access to selected PCI
1.52      gdt       904: devices. This can allow, for example, a non-privileged domain to have
                    905: access to a physical network interface or disk controller.  However,
                    906: keep in mind that giving a domain access to a PCI device most likely
                    907: will give the domain read/write access to the whole physical memory,
                    908: as PCs don't have an IOMMU to restrict memory access to DMA-capable
1.53      gdt       909: device.  Also, it's not possible to export ISA devices to non-dom0
1.52      gdt       910: domains, which means that the primary VGA adapter can't be exported.
                    911: A guest domain trying to access the VGA registers will panic.
                    912: 
1.53      gdt       913: If the dom0 is NetBSD, it has to be running Xen 3.1, as support has
1.52      gdt       914: not been ported to later versions at this time.
                    915: 
                    916: For a PCI device to be exported to a domU, is has to be attached to
                    917: the "pciback" driver in dom0.  Devices passed to the dom0 via the
                    918: pciback.hide boot parameter will attach to "pciback" instead of the
                    919: usual driver.  The list of devices is specified as "(bus:dev.func)",
1.5       mspo      920: where bus and dev are 2-digit hexadecimal numbers, and func a
                    921: single-digit number:
1.1       mspo      922: 
1.52      gdt       923:         pciback.hide=(00:0a.0)(00:06.0)
1.1       mspo      924: 
1.52      gdt       925: pciback devices should show up in the dom0's boot messages, and the
1.5       mspo      926: devices should be listed in the `/kern/xen/pci` directory.
1.1       mspo      927: 
1.52      gdt       928: PCI devices to be exported to a domU are listed in the "pci" array of
                    929: the domU's config file, with the format "0000:bus:dev.func".
1.1       mspo      930: 
1.52      gdt       931:         pci = [ '0000:00:06.0', '0000:00:0a.0' ]
1.1       mspo      932: 
1.52      gdt       933: In the domU an "xpci" device will show up, to which one or more pci
                    934: busses will attach.  Then the PCI drivers will attach to PCI busses as
                    935: usual.  Note that the default NetBSD DOMU kernels do not have "xpci"
                    936: or any PCI drivers built in by default; you have to build your own
                    937: kernel to use PCI devices in a domU.  Here's a kernel config example;
                    938: note that only the "xpci" lines are unusual.
                    939: 
                    940:         include         "arch/i386/conf/XEN3_DOMU"
                    941: 
                    942:         # Add support for PCI busses to the XEN3_DOMU kernel
                    943:         xpci* at xenbus ?
                    944:         pci* at xpci ?
                    945: 
                    946:         # PCI USB controllers
                    947:         uhci*   at pci? dev ? function ?        # Universal Host Controller (Intel)
                    948: 
                    949:         # USB bus support
                    950:         usb*    at uhci?
                    951: 
                    952:         # USB Hubs
                    953:         uhub*   at usb?
                    954:         uhub*   at uhub? port ? configuration ? interface ?
                    955: 
                    956:         # USB Mass Storage
                    957:         umass*  at uhub? port ? configuration ? interface ?
                    958:         wd*     at umass?
                    959:         # SCSI controllers
                    960:         ahc*    at pci? dev ? function ?        # Adaptec [23]94x, aic78x0 SCSI
                    961: 
                    962:         # SCSI bus support (for both ahc and umass)
                    963:         scsibus* at scsi?
                    964: 
                    965:         # SCSI devices
                    966:         sd*     at scsibus? target ? lun ?      # SCSI disk drives
                    967:         cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
1.1       mspo      968: 
                    969: 
1.28      gdt       970: NetBSD as a domU in a VPS
                    971: =========================
                    972: 
                    973: The bulk of the HOWTO is about using NetBSD as a dom0 on your own
                    974: hardware.  This section explains how to deal with Xen in a domU as a
                    975: virtual private server where you do not control or have access to the
1.70      gdt       976: dom0.  This is not intended to be an exhaustive list of VPS providers;
                    977: only a few are mentioned that specifically support NetBSD.
1.28      gdt       978: 
1.52      gdt       979: VPS operators provide varying degrees of access and mechanisms for
                    980: configuration.  The big issue is usually how one controls which kernel
                    981: is booted, because the kernel is nominally in the dom0 filesystem (to
1.70      gdt       982: which VPS users do not normally have acesss).  A second issue is how
                    983: to install NetBSD.
1.52      gdt       984: A VPS user may want to compile a kernel for security updates, to run
                    985: npf, run IPsec, or any other reason why someone would want to change
                    986: their kernel.
                    987: 
                    988: One approach is to have an adminstrative interface to upload a kernel,
1.68      gdt       989: or to select from a prepopulated list.  Other approaches are pygrub
1.59      gdt       990: (deprecated) and pvgrub, which are ways to have a bootloader obtain a
                    991: kernel from the domU filesystem.  This is closer to a regular physical
                    992: computer, where someone who controls a machine can replace the kernel.
1.52      gdt       993: 
1.74      gdt       994: A second issue is multiple CPUs.  With NetBSD 6, domUs support
                    995: multiple vcpus, and it is typical for VPS providers to enable multiple
                    996: CPUs for NetBSD domUs.
                    997: 
1.68      gdt       998: pygrub
1.59      gdt       999: -------
1.52      gdt      1000: 
1.68      gdt      1001: pygrub runs in the dom0 and looks into the domU filesystem.  This
1.59      gdt      1002: implies that the domU must have a kernel in a filesystem in a format
1.68      gdt      1003: known to pygrub.  As of 2014, pygrub seems to be of mostly historical
                   1004: interest.
1.52      gdt      1005: 
1.59      gdt      1006: pvgrub
                   1007: ------
                   1008: 
                   1009: pvgrub is a version of grub that uses PV operations instead of BIOS
                   1010: calls.  It is booted from the dom0 as the domU kernel, and then reads
                   1011: /grub/menu.lst and loads a kernel from the domU filesystem.
                   1012: 
1.70      gdt      1013: [Panix](http://www.panix.com/) lets users use pvgrub.  Panix reports
1.71      gdt      1014: that pvgrub works with FFsv2 with 16K/2K and 32K/4K block/frag sizes
                   1015: (and hence with defaults from "newfs -O 2").  See [Panix's pvgrub
1.70      gdt      1016: page](http://www.panix.com/v-colo/grub.html), which describes only
1.74      gdt      1017: Linux but should be updated to cover NetBSD :-).
1.70      gdt      1018: 
                   1019: [prgmr.com](http://prgmr.com/) also lets users with pvgrub to boot
                   1020: their own kernel.  See then [prgmr.com NetBSD
1.74      gdt      1021: HOWTO](http://wiki.prgmr.com/mediawiki/index.php/NetBSD_as_a_DomU)
                   1022: (which is in need of updating).
1.59      gdt      1023: 
1.70      gdt      1024: It appears that [grub's FFS
                   1025: code](http://xenbits.xensource.com/hg/xen-unstable.hg/file/bca284f67702/tools/libfsimage/ufs/fsys_ufs.c)
                   1026: does not support all aspects of modern FFS, but there are also reports
1.72      gdt      1027: that FFSv2 works fine.  At prgmr, typically one has an ext2 or FAT
1.70      gdt      1028: partition for the kernel with the intent that grub can understand it,
                   1029: which leads to /netbsd not being the actual kernel.  One must remember
                   1030: to update the special boot partiion.
1.59      gdt      1031: 
                   1032: Amazon
                   1033: ------
                   1034: 
1.95      gdt      1035: See the [Amazon EC2 page](../amazon_ec2/).
1.44      gdt      1036: 
                   1037: Using npf
                   1038: ---------
                   1039: 
1.81      gdt      1040: In standard kernels, npf is a module, and thus cannot be loaded in a
1.44      gdt      1041: DOMU kernel.
                   1042: 
1.95      gdt      1043: TODO: Explain how to compile npf into a custom kernel, answering (but
                   1044: note that the problem was caused by not booting the right kernel)
                   1045: [this email to
                   1046: netbsd-users](http://mail-index.netbsd.org/netbsd-users/2014/12/26/msg015576.html).
1.65      gdt      1047: 
                   1048: TODO items for improving NetBSD/xen
                   1049: ===================================
                   1050: 
1.93      gdt      1051: * Make the NetBSD dom0 kernel work with SMP.
                   1052: * Test the Xen 4.5 packages adequately to be able to recommend them as
                   1053:   the standard approach.
                   1054: * Get PCI passthrough working on Xen 4.5
1.65      gdt      1055: * Get pvgrub into pkgsrc, either via xentools or separately.
                   1056: * grub
1.70      gdt      1057:   * Check/add support to pkgsrc grub2 for UFS2 and arbitrary
1.66      gdt      1058:     fragsize/blocksize (UFS2 support may be present; the point is to
                   1059:     make it so that with any UFS1/UFS2 filesystem setup that works
                   1060:     with NetBSD grub will also work).
1.70      gdt      1061:     See [pkg/40258](http://gnats.netbsd.org/40258).
1.65      gdt      1062:   * Push patches upstream.
                   1063:   * Get UFS2 patches into pvgrub.
                   1064: * Add support for PV ops to a version of /boot, and make it usable as
                   1065:   a kernel in Xen, similar to pvgrub.
1.93      gdt      1066: * Solve somehow the issue with modules for GENERIC not being loadable
                   1067:   in a Xen dom0 or domU kernel.
                   1068: 
                   1069: Random pointers
                   1070: ===============
                   1071: 
                   1072: TODO: This section contains links from elsewhere not yet integrated
                   1073: into the HOWTO.
                   1074: 
                   1075: * http://www.lumbercartel.ca/library/xen/
                   1076: * http://pbraun.nethence.com/doc/sysutils/xen_netbsd_dom0.html

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