1: [[!meta title="Xen HowTo"]]
3: Xen is a Type 1 hypervisor which supports running multiple guest operating
4: systems on a single physical machine. One uses the Xen kernel to control the
5: CPU, memory and console, a dom0 operating system which mediates access to
6: other hardware (e.g., disks, network, USB), and one or more domU operating
7: systems which operate in an unprivileged virtualized environment. IO requests
8: from the domU systems are forwarded by the Xen hypervisor to the dom0 to be
11: This HOWTO presumes a basic familiarity with the Xen system
12: architecture, with installing NetBSD on amd64 hardware, and with
13: installing software from pkgsrc. See also the [Xen
18: # Overview
20: The basic concept of Xen is that the hypervisor (xenkernel) runs on
21: the hardware, and runs a privileged domain ("dom0") that can access
22: disks/networking/etc. One then runs additonal unprivileged domains
23: (each a "domU"), presumably to do something useful.
25: This HOWTO addresses how to run a NetBSD dom0 (and hence also build
26: xen itself). It also addresses how to run domUs in that environment,
27: and how to deal with having a domU in a Xen environment run by someone
28: else and/or not running NetBSD.
30: There are many choices one can make; the HOWTO recommends the standard
31: approach and limits discussion of alternatives in many cases.
33: ## Guest Styles
35: Xen supports different styles of guests.
37: [[!table data="""
38: Style of guest |Supported by NetBSD
39: PV |Yes (dom0, domU)
40: HVM |Yes (domU)
41: PVHVM |current-only (domU)
42: PVH |current-only (domU, dom0 not yet)
45: In Para-Virtualized (PV) mode, the guest OS does not attempt to access
46: hardware directly, but instead makes hypercalls to the hypervisor; PV
47: guests must be specifically coded for Xen.
48: See [PV](https://wiki.xen.org/wiki/Paravirtualization_(PV\)).
50: In HVM mode, no guest modification is required; however, hardware
51: support is required, such as VT-x on Intel CPUs and SVM on AMD CPUs.
52: The dom0 runs qemu to emulate hardware.
54: In PVHVM mode, the guest runs as HVM, but additionally can use PV
55: drivers for efficiency.
56: See [PV on HVM](https://wiki.xen.org/wiki/PV_on_HVM).
58: There have been two PVH modes: original PVH and PVHv2. Original PVH
59: was based on PV mode and is no longer relevant at all. PVHv2 is
60: basically lightweight HVM with PV drivers. A critical feature of it
61: is that qemu is not needed; the hypervisor can do the emulation that
62: is required. Thus, a dom0 can be PVHv2.
63: The source code uses PVH and config files use pvh; this refers to PVHv2.
64: See [PVH(v2)](https://wiki.xenproject.org/wiki/PVH_(v2\)_Domu).
66: At system boot, the dom0 kernel is loaded as a module with Xen as the kernel.
67: The dom0 can start one or more domUs. (Booting is explained in detail
68: in the dom0 section.)
70: ## CPU Architecture
72: Xen runs on x86_64 hardware (the NetBSD amd64 port).
74: There is a concept of Xen running on ARM, but there are no reports of this working with NetBSD.
76: The dom0 system should be amd64. (Instructions for i386PAE dom0 have been removed from the HOWTO.)
78: The domU can be i386PAE or amd64.
79: i386PAE at one point was considered as [faster](https://lists.xen.org/archives/html/xen-devel/2012-07/msg00085.html) than amd64.
81: ## Xen Versions
83: In NetBSD, Xen is provided in pkgsrc, via matching pairs of packages
84: xenkernel and xentools. We will refer only to the kernel versions,
85: but note that both packages must be installed together and must have
86: matching versions.
88: Versions available in pkgsrc:
90: [[!table data="""
91: Xen Version |Package Name |Xen CPU Support |EOL'ed By Upstream
92: 4.11 |xenkernel411 |x86_64 |No
93: 4.13 |xenkernel413 |x86_64 |No
96: See also the [Xen Security Advisory page](http://xenbits.xen.org/xsa/).
98: Older Xen had a python-based management tool called xm, now replaced
99: by xl.
101: ## NetBSD versions
103: Xen has been supported in NetBSD for a long time, at least since 2005.
104: Initially Xen was PV only.
106: NetBSD 8 and up support PV and HVM modes.
108: Support for PVHVM and PVH is available only in NetBSD-current.
110: NetBSD up to and including NetBSD 9 as a dom0 does not run SMP,
111: because some drivers are not yet safe for this. NetBSD-current
112: supports SMP in dom0.
114: NetBSD, when run as a domU, can and does typically run SMP.
116: Note that while Xen 4.13 is current, the kernel support is still
117: called XEN3, because the hypercall interface has not changed
120: # Creating a NetBSD dom0
122: In order to install a NetBSD as a dom0, one first installs a normal
123: NetBSD system, and then pivot the install to a dom0 install by
124: changing the kernel and boot configuration.
126: In 2018-05, trouble booting a dom0 was reported with 256M of RAM: with
127: 512M it worked reliably. This does not make sense, but if you see
128: "not ELF" after Xen boots, try increasing dom0 RAM.
130: ## Installation of NetBSD
132: [Install NetBSD/amd64](/guide/inst/) just as you would if you were not
133: using Xen. Therefore, use the most recent release, or a build from
134: the most recent stable branch. Alternatively, use -current, being
135: mindful of all the usual caveats of lower stability of current, and
136: likely a bit more so.
138: ## Installation of Xen
140: ### Building Xen
142: Use the most recent version of Xen in pkgsrc, unless the DESCR says that it is not suitable.
143: Therefore, choose 4.13.
144: In the dom0, install xenkernel413 and xentools413 from pkgsrc.
146: Once this is done, copy the Xen kernel from where pkgsrc puts it to
147: where the boot process will be able to find it:
149: [[!template id=programlisting text="""
150: # cp -p /usr/pkg/xen413-kernel/xen.gz /
153: Then, place a NetBSD XEN3_DOM0 kernel in the `/` directory. Such
154: kernel can either be taken from a local release build.sh run, compiled
155: manually, or downloaded from the NetBSD FTP, for example at:
157: [[!template id=programlisting text="""
161: ### Configuring booting
163: Read boot.cfg(8) carefully. Add lines to /boot.cfg to boot Xen:
165: [[!template id=filecontent name="/boot.cfg" text="""
166: menu=Xen:load /netbsd-XEN3_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=512M
167: menu=Xen single user:load /netbsd-XEN3_DOM0.gz console=pc -s;multiboot /xen.gz dom0_mem=512M
170: This specifies that the dom0 should have 512MB of ram, leaving the rest
171: to be allocated for domUs.
173: NB: This says add, not replace, so that you will be able to more
174: easily boot a NetBSD kernel without Xen. Once Xen boots ok, you may
175: want to set it as default. It is highly likely that you will have
176: trouble at some point, and keeping an up-to-date GENERIC for use in
177: fixing problems is the standard prudent approach.
179: \todo Explain why rndseed is not set with Xen as part of the dom0
182: ### Console selection
184: See boot_console(8). Understand that you should start from a place of
185: having console setup correct for booting GENERIC before trying to
186: configure Xen.
188: Generally for GENERIC, one sets the console in bootxx_ffsv1 or
189: equivalent, and this is passed on to /boot (where one typically does
190: not set the console). This configuration of bootxx_ffsv1 should also
191: be in place for Xen systems, to allow seeing messages from /boot and
192: use of a keyboard to select a line from the menu. And, one should
193: have a working boot path to GENERIC for rescue situations.
195: With GENERIC, the boot options are passed on to /netbsd, but there is
196: currently no mechanism to pass these via multiboot to the hypervisor.
197: Thus, in addition to configuring the console in the boot blocks, one
198: must also configure it for Xen.
200: By default, the hypervisor (Xen itself) will use some sort of vga
201: device as the console, much like GENERIC uses by default. The vga
202: console is relinquished at the conclusion of hypervisor boot, before
203: the dom0 is started. Xen when using a vga console does not process
204: console input.
206: The hypervisor can be configured to use a serial port console, e.g.
207: [[!template id=filecontent name="/boot.cfg" text="""
208: menu=Xen:losad /netbsd-XEN3_DOM0.gz console=com0;multiboot /xen.gz dom0_mem=512M console=com1 com1=9600,8n1
210: This example uses the first serial port (Xen counts from 1; this is
211: what NetBSD would call com0), and sets speed and parity. (The dom0 is
212: then configured to use the same serial port in this example.)
214: With the hypervisor configured for a serial console, it can get input,
215: and there is a notion of passing this input to the dom0. \todo
216: Explain why, if Xen has a serial console, the dom0 console is
217: typically also configured to open that same serial port, instead of
218: getting the passthrough input via the xen console.
220: One also configures the console for the dom0. While one might expect
221: console=pc to be default, following behavior of GENERIC, a hasty read
222: of the code suggests there is no default and booting without a
223: selected console might lead to a panic. Also, there is merit in
224: explicit configuration. Therefore the standard approach is to place
225: console=pc as part of the load statement for the dom0 kernel, or
226: alternatively console=com0.
228: The NetBSD dom0 kernel will attach xencons(4) (the man page does not
229: exist), but this is not used as a console. It is used to obtain the
230: messages from the hypervisor's console; run `xl dmesg` to see them.
232: ### Tuning
234: In an attempt to add performance, one can also add `dom0_max_vcpus=1 dom0_vcpus_pin`,
235: to force only one vcpu to be provided (since NetBSD dom0 can't use
236: more) and to pin that vcpu to a physical CPU. Xen has
237: [many boot options](http://xenbits.xenproject.org/docs/4.13-testing/misc/xen-command-line.html),
238: and other than dom0 memory and max_vcpus, they are generally not
240: \todo Revisit this advice with current.
241: \todo Explain if anyone has ever actually measured that this helps.
243: ### rc.conf
245: Ensure that the boot scripts installed in
246: `/usr/pkg/share/examples/rc.d` are in `/etc/rc.d`, either because you
247: have `PKG_RCD_SCRIPTS=yes`, or manually. (This is not special to Xen,
248: but a normal part of pkgsrc usage.)
250: Set `xencommons=YES` in rc.conf:
252: [[!template id=filecontent name="/etc/rc.conf" text="""
256: \todo Recommend for/against xen-watchdog.
258: ### Testing
260: Now, reboot so that you are running a DOM0 kernel under Xen, rather
261: than GENERIC without Xen.
263: Once the reboot is done, use `xl` to inspect Xen's boot messages,
264: available resources, and running domains. For example:
266: [[!template id=programlisting text="""
267: # xl dmesg
268: ... xen's boot info ...
269: # xl info
270: ... available memory, etc ...
271: # xl list
272: Name Id Mem(MB) CPU State Time(s) Console
273: Domain-0 0 64 0 r---- 58.1
276: Xen logs will be in /var/log/xen.
278: ### Issues with xencommons
280: `xencommons` starts `xenstored`, which stores data on behalf of dom0 and
281: domUs. It does not currently work to stop and start xenstored.
282: Certainly all domUs should be shutdown first, following the sort order
283: of the rc.d scripts. However, the dom0 sets up state with xenstored,
284: and is not notified when xenstored exits, leading to not recreating
285: the state when the new xenstored starts. Until there's a mechanism to
286: make this work, one should not expect to be able to restart xenstored
287: (and thus xencommons). There is currently no reason to expect that
288: this will get fixed any time soon.
289: \todo Confirm if this is still true in 2020.
291: ## Xen-specific NetBSD issues
293: There are (at least) two additional things different about NetBSD as a
294: dom0 kernel compared to hardware.
296: One is that through NetBSD 9 the module ABI is different because some
297: of the #defines change, so there are separate sets of modules in
298: /stand. In NetBSD-current, there is only one set of modules.
300: The other difference is that XEN3_DOM0 does not have exactly the same
301: options as GENERIC. While it is debatable whether or not this is a
302: bug, users should be aware of this and can simply add missing config
303: items if desired.
305: Finally, there have been occasional reports of trouble with X11
306: servers in NetBSD as a dom0.
308: ## Updating Xen in a dom0
310: Basically, update the xenkernel and xentools packages and copy the new
311: Xen kernel into place, and reboot. This procedure should be usable to
312: update to a new Xen release, but the reader is reminded that having a
313: non-Xen boot methods was recommended earlier.
315: ## Updating NetBSD in a dom0
317: This is just like updating NetBSD on bare hardware, assuming the new
318: version supports the version of Xen you are running. Generally, one
319: replaces the kernel and reboots, and then overlays userland binaries
320: and adjusts `/etc`.
322: Note that one should update both the non-Xen kernel typically used for
323: rescue purposes, as well as the DOM0 kernel used with Xen.
325: ## anita (for testing NetBSD)
327: With a NetBSD dom0, even without any domUs, one should be able to run
328: anita (see pkgsrc/misc/py-anita) to test NetBSD releases, by doing (as
329: root, because anita must create a domU):
331: [[!template id=programlisting text="""
332: anita --vmm=xl test file:///usr/obj/i386/
335: # Unprivileged domains (domU)
337: This section describes general concepts about domUs. It does not
338: address specific domU operating systems or how to install them. The
339: config files for domUs are typically in `/usr/pkg/etc/xen`, and are
340: typically named so that the file name, domU name and the domU's host
341: name match.
343: The domU is provided with CPU and memory by Xen, configured by the
344: dom0. The domU is provided with disk and network by the dom0,
345: mediated by Xen, and configured in the dom0.
347: Entropy in domUs can be an issue; physical disks and network are on
348: the dom0. NetBSD's /dev/random system works, but is often challenged.
350: ## Config files
352: See /usr/pkg/share/examples/xen/xlexample*
353: for a small number of well-commented examples, mostly for running
356: The following is an example minimal domain configuration file. The domU
357: serves as a network file server.
359: [[!template id=filecontent name="/usr/pkg/etc/xen/foo" text="""
360: name = "domU-id"
361: kernel = "/netbsd-XEN3PAE_DOMU-i386-foo.gz"
362: memory = 1024
363: vif = [ 'mac=aa:00:00:d1:00:09,bridge=bridge0' ]
364: disk = [ 'file:/n0/xen/foo-wd0,0x0,w',
365: 'file:/n0/xen/foo-wd1,0x1,w' ]
368: The domain will have name given in the `name` setting. The kernel has the
369: host/domU name in it, so that on the dom0 one can update the various
370: domUs independently. The `vif` line causes an interface to be provided,
371: with a specific mac address (do not reuse MAC addresses!), in bridge
372: mode. Two disks are provided, and they are both writable; the bits
373: are stored in files and Xen attaches them to a vnd(4) device in the
374: dom0 on domain creation. The system treats xbd0 as the boot device
375: without needing explicit configuration.
377: There is not type line; that implicitly defines a pv domU.
379: By convention, domain config files are kept in `/usr/pkg/etc/xen`. Note
380: that "xl create" takes the name of a config file, while other commands
381: take the name of a domain.
383: Examples of commands:
385: [[!template id=programlisting text="""
386: xl create /usr/pkg/etc/xen/foo
387: xl console domU-id
388: xl create -c /usr/pkg/etc/xen/foo
389: xl shutdown domU-id
390: xl list
393: Typing `^]` will exit the console session. Shutting down a domain is
394: equivalent to pushing the power button; a NetBSD domU will receive a
395: power-press event and do a clean shutdown. Shutting down the dom0
396: will trigger controlled shutdowns of all configured domUs.
398: ## CPU and memory
400: A domain is provided with some number of vcpus, up to the number
401: of CPUs seen by the hypervisor. For a domU, it is controlled
402: from the config file by the "vcpus = N" directive.
404: A domain is provided with memory; this is controlled in the config
405: file by "memory = N" (in megabytes). In the straightforward case, the
406: sum of the the memory allocated to the dom0 and all domUs must be less
407: than the available memory.
409: Xen also provides a "balloon" driver, which can be used to let domains
410: use more memory temporarily.
412: ## Virtual disks
414: In domU config files, the disks are defined as a sequence of 3-tuples:
416: * The first element is "method:/path/to/disk". Common methods are
417: "file:" for a file-backed vnd, and "phy:" for something that is already
418: a device, such as an LVM logical volume.
420: * The second element is an artifact of how virtual disks are passed to
421: Linux, and a source of confusion with NetBSD Xen usage. Linux domUs
422: are given a device name to associate with the disk, and values like
423: "hda1" or "sda1" are common. In a NetBSD domU, the first disk appears
424: as xbd0, the second as xbd1, and so on. However, xl demands a
425: second argument. The name given is converted to a major/minor by
426: calling stat(2) on the name in /dev and this is passed to the domU.
427: In the general case, the dom0 and domU can be different operating
428: systems, and it is an unwarranted assumption that they have consistent
429: numbering in /dev, or even that the dom0 OS has a /dev. With NetBSD
430: as both dom0 and domU, using values of 0x0 for the first disk and 0x1
431: for the second works fine and avoids this issue. For a GNU/Linux
432: guest, one can create /dev/hda1 in /dev, or to pass 0x301 for
435: * The third element is "w" for writable disks, and "r" for read-only
439: [[!template id=filecontent name="/usr/pkg/etc/xen/foo" text="""
440: disk = [ 'file:/n0/xen/foo-wd0,0x0,w' ]
443: Note that NetBSD by default creates only vnd. If you need more
444: than 4 total virtual disks at a time, run e.g. "./MAKEDEV vnd4" in the
447: ## Virtual Networking
449: Xen provides virtual Ethernets, each of which connects the dom0 and a
450: domU. For each virtual network, there is an interface "xvifN.M" in
451: the dom0, and a matching interface xennetM (NetBSD name) in domU index N.
452: The interfaces behave as if there is an Ethernet with two
453: adapters connected. From this primitive, one can construct various
454: configurations. We focus on two common and useful cases for which
455: there are existing scripts: bridging and NAT.
457: With bridging (in the example above), the domU perceives itself to be
458: on the same network as the dom0. For server virtualization, this is
459: usually best. Bridging is accomplished by creating a bridge(4) device
460: and adding the dom0's physical interface and the various xvifN.0
461: interfaces to the bridge. One specifies "bridge=bridge0" in the domU
462: config file. The bridge must be set up already in the dom0; an
463: example /etc/ifconfig.bridge0 is:
465: [[!template id=filecontent name="/etc/ifconfig.bridge0" text="""
468: !brconfig bridge0 add wm0
471: With NAT, the domU perceives itself to be behind a NAT running on the
472: dom0. This is often appropriate when running Xen on a workstation.
473: TODO: NAT appears to be configured by "vif = [ '' ]".
475: The MAC address specified is the one used for the interface in the new
476: domain. The interface in dom0 will use this address XOR'd with
477: 00:00:00:01:00:00. Random MAC addresses are assigned if not given.
479: ## Starting domains automatically
481: To start domains `domU-netbsd` and `domU-linux` at boot and shut them
482: down cleanly on dom0 shutdown, add the following in rc.conf:
484: [[!template id=filecontent name="/etc/rc.conf" text="""
485: xendomains="domU-netbsd domU-linux"
488: # domU setup for specific systems
490: Creating domUs is almost entirely independent of operating system. We
491: have already presented the basics of config files in the previous system.
493: Of course, this section presumes that you have a working dom0.
495: ## Creating a NetBSD PV domU
497: See the earlier config file, and adjust memory. Decide on how much
498: storage you will provide, and prepare it (file or LVM).
500: While the kernel will be obtained from the dom0 file system, the same
501: file should be present in the domU as /netbsd so that tools like
502: savecore(8) can work. (This is helpful but not necessary.)
504: The kernel must be specifically built for Xen, to use PV interfacesas
505: a domU. NetBSD release builds provide the following kernels:
507: i386 XEN3PAE_DOMU
508: amd64 XEN3_DOMU
510: This will boot NetBSD, but this is not that useful if the disk is
511: empty. One approach is to unpack sets onto the disk outside of Xen
512: (by mounting it, just as you would prepare a physical disk for a
513: system you can't run the installer on).
515: A second approach is to run an INSTALL kernel, which has a miniroot
516: and can load sets from the network. To do this, copy the INSTALL
517: kernel to / and change the kernel line in the config file to:
519: kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
521: Then, start the domain as "xl create -c configfile".
523: Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following
524: line should be used in the config file.
526: disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
528: After booting the domain, the option to install via CDROM may be
529: selected. The CDROM device should be changed to `xbd1d`.
531: Once done installing, "halt -p" the new domain (don't reboot or halt:
532: it would reload the INSTALL_XEN3_DOMU kernel even if you changed the
533: config file), switch the config file back to the XEN3_DOMU kernel,
534: and start the new domain again. Now it should be able to use "root on
535: xbd0a" and you should have a functional NetBSD domU.
537: TODO: check if this is still accurate.
538: When the new domain is booting you'll see some warnings about *wscons*
539: and the pseudo-terminals. These can be fixed by editing the files
540: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
541: `/etc/ttys`, except *console*, like this:
543: console "/usr/libexec/getty Pc" vt100 on secure
544: ttyE0 "/usr/libexec/getty Pc" vt220 off secure
545: ttyE1 "/usr/libexec/getty Pc" vt220 off secure
546: ttyE2 "/usr/libexec/getty Pc" vt220 off secure
547: ttyE3 "/usr/libexec/getty Pc" vt220 off secure
549: Finally, all screens must be commented out from `/etc/wscons.conf`.
551: One should also run `powerd` in a domU, but this should not need
552: configuring. With powerd, the domain will run a controlled shutdown
553: if `xl shutdown -R` or `xl shutdown -H` is used on the dom0, via
554: receiving a synthetic `power button pressed` signal. In 9 and
555: current, `powerd` is run by default under Xen kernels (or if ACPI is
556: present), and it can be added to rc.conf if not.
558: It is not strictly necessary to have a kernel (as /netbsd) in the domU
559: file system. However, various programs (e.g. netstat) will use that
560: kernel to look up symbols to read from kernel virtual memory. If
561: /netbsd is not the running kernel, those lookups will fail. (This is
562: not really a Xen-specific issue, but because the domU kernel is
563: obtained from the dom0, it is far more likely to be out of sync or
564: missing with Xen.)
566: Note that NetBSD by default creates only xbd. If you need more
567: virtual disks in a domU, run e.g. "./MAKEDEV xbd4" in the domU.
569: ## Creating a Linux domU
571: Creating unprivileged Linux domains isn't much different from
572: unprivileged NetBSD domains, but there are some details to know.
574: First, the second parameter passed to the disk declaration (the '0x1' in
575: the example below)
577: disk = [ 'phy:/dev/wd0e,0x1,w' ]
579: does matter to Linux. It wants a Linux device number here (e.g. 0x300
580: for hda). Linux builds device numbers as: (major \<\< 8 + minor).
581: So, hda1 which has major 3 and minor 1 on a Linux system will have
582: device number 0x301. Alternatively, devices names can be used (hda,
583: hdb, ...) as xentools has a table to map these names to devices
584: numbers. To export a partition to a Linux guest we can use:
586: disk = [ 'phy:/dev/wd0e,0x300,w' ]
587: root = "/dev/hda1 ro"
589: and it will appear as /dev/hda on the Linux system, and be used as root
592: To install the Linux system on the partition to be exported to the
593: guest domain, the following method can be used: install
594: sysutils/e2fsprogs from pkgsrc. Use mke2fs to format the partition
595: that will be the root partition of your Linux domain, and mount it.
596: Then copy the files from a working Linux system, make adjustments in
597: `/etc` (fstab, network config). It should also be possible to extract
598: binary packages such as .rpm or .deb directly to the mounted partition
599: using the appropriate tool, possibly running under NetBSD's Linux
600: emulation. Once the file system has been populated, umount it. If
601: desirable, the file system can be converted to ext3 using tune2fs -j.
602: It should now be possible to boot the Linux guest domain, using one of
603: the vmlinuz-\*-xenU kernels available in the Xen binary distribution.
605: To get the Linux console right, you need to add:
607: extra = "xencons=tty1"
609: to your configuration since not all Linux distributions auto-attach a
610: tty to the xen console.
612: ## Creating a NetBSD HVM domU
614: Use type='hmv', probably. Use a GENERIC kernel within the disk image.
616: ## Creating a NetBSD PVH domU
618: Use type='pvh'.
620: \todo Explain where the kernel comes from.
622: ## Creating a Solaris domU
624: See possibly outdated
625: [Solaris domU instructions](/ports/xen/howto-solaris/).
627: ## PCI passthrough: Using PCI devices in guest domains
629: NB: PCI passthrough only works on some Xen versions and as of 2020 it
630: is not clear that it works on any version in pkgsrc. Reports
631: confirming or denying this notion should be sent to port-xen@.
633: The dom0 can give other domains access to selected PCI
634: devices. This can allow, for example, a non-privileged domain to have
635: access to a physical network interface or disk controller. However,
636: keep in mind that giving a domain access to a PCI device most likely
637: will give the domain read/write access to the whole physical memory,
638: as PCs don't have an IOMMU to restrict memory access to DMA-capable
639: device. Also, it's not possible to export ISA devices to non-dom0
640: domains, which means that the primary VGA adapter can't be exported.
641: A guest domain trying to access the VGA registers will panic.
643: If the dom0 is NetBSD, it has to be running Xen 3.1, as support has
644: not been ported to later versions at this time.
646: For a PCI device to be exported to a domU, is has to be attached to
647: the "pciback" driver in dom0. Devices passed to the dom0 via the
648: pciback.hide boot parameter will attach to "pciback" instead of the
649: usual driver. The list of devices is specified as "(bus:dev.func)",
650: where bus and dev are 2-digit hexadecimal numbers, and func a
651: single-digit number:
655: pciback devices should show up in the dom0's boot messages, and the
656: devices should be listed in the `/kern/xen/pci` directory.
658: PCI devices to be exported to a domU are listed in the "pci" array of
659: the domU's config file, with the format "0000:bus:dev.func".
661: pci = [ '0000:00:06.0', '0000:00:0a.0' ]
663: In the domU an "xpci" device will show up, to which one or more pci
664: buses will attach. Then the PCI drivers will attach to PCI buses as
665: usual. Note that the default NetBSD DOMU kernels do not have "xpci"
666: or any PCI drivers built in by default; you have to build your own
667: kernel to use PCI devices in a domU. Here's a kernel config example;
668: note that only the "xpci" lines are unusual.
670: include "arch/i386/conf/XEN3_DOMU"
672: # Add support for PCI buses to the XEN3_DOMU kernel
673: xpci* at xenbus ?
674: pci* at xpci ?
676: # PCI USB controllers
677: uhci* at pci? dev ? function ? # Universal Host Controller (Intel)
679: # USB bus support
680: usb* at uhci?
682: # USB Hubs
683: uhub* at usb?
684: uhub* at uhub? port ? configuration ? interface ?
686: # USB Mass Storage
687: umass* at uhub? port ? configuration ? interface ?
688: wd* at umass?
689: # SCSI controllers
690: ahc* at pci? dev ? function ? # Adaptec 94x, aic78x0 SCSI
692: # SCSI bus support (for both ahc and umass)
693: scsibus* at scsi?
695: # SCSI devices
696: sd* at scsibus? target ? lun ? # SCSI disk drives
697: cd* at scsibus? target ? lun ? # SCSI CD-ROM drives
700: # Miscellaneous Information
702: ## Nesting under Linux KVM
704: It is possible to run a Xen and a NetBSD dom0 under Linux KVM. One
705: can enable virtio in the dom0 for greater speed.
707: ## Other nesting
709: In theory, any full emulation should be able to run Xen and a NetBSD
710: dom0. The HOWTO does not currently have information about Xen XVM
711: mode, nvmm, qemu, Virtualbox, etc.
713: ## NetBSD 5 as domU
715: [NetBSD 5 is known to panic.](http://mail-index.netbsd.org/port-xen/2018/04/17/msg009181.html)
716: (However, NetBSD 5 systems should be updated to a supported version.)
718: # NetBSD as a domU in a VPS
720: The bulk of the HOWTO is about using NetBSD as a dom0 on your own
721: hardware. This section explains how to deal with Xen in a domU as a
722: virtual private server where you do not control or have access to the
723: dom0. This is not intended to be an exhaustive list of VPS providers;
724: only a few are mentioned that specifically support NetBSD.
726: VPS operators provide varying degrees of access and mechanisms for
727: configuration. The big issue is usually how one controls which kernel
728: is booted, because the kernel is nominally in the dom0 file system (to
729: which VPS users do not normally have access). A second issue is how
730: to install NetBSD.
731: A VPS user may want to compile a kernel for security updates, to run
732: npf, run IPsec, or any other reason why someone would want to change
733: their kernel.
735: One approach is to have an administrative interface to upload a kernel,
736: or to select from a prepopulated list. Other approaches are pygrub
737: (deprecated) and pvgrub, which are ways to have a bootloader obtain a
738: kernel from the domU file system. This is closer to a regular physical
739: computer, where someone who controls a machine can replace the kernel.
741: A second issue is multiple CPUs. With NetBSD 6, domUs support
742: multiple vcpus, and it is typical for VPS providers to enable multiple
743: CPUs for NetBSD domUs.
745: ## Complexities due to Xen changes
747: Xen has many security advisories and people running Xen systems make
748: different choices.
750: ### stub domains
752: Some (Linux only?) dom0 systems use something called "stub domains" to
753: isolate qemu from the dom0 system, as a security and reliabilty
754: mechanism when running HVM domUs. Somehow, NetBSD's GENERIC kernel
755: ends up using PIO for disks rather than DMA. Of course, all of this
756: is emulated, but emulated PIO is unusably slow. This problem is not
757: currently understood.
759: ### Grant tables
761: There are multiple versions of using grant tables, and some security
762: advisories have suggested disabling some versions. Some versions of
763: NetBSD apparently only use specific versions and this can lead to
764: "NetBSD current doesn't run on hosting provider X" situations.
766: \todo Explain better.
768: ## Boot methods
770: ### pvgrub
772: pvgrub is a version of grub that uses PV operations instead of BIOS
773: calls. It is booted from the dom0 as the domU kernel, and then reads
774: /grub/menu.lst and loads a kernel from the domU file system.
776: [Panix](http://www.panix.com/) lets users use pvgrub. Panix reports
777: that pvgrub works with FFsv2 with 16K/2K and 32K/4K block/frag sizes
778: (and hence with defaults from "newfs -O 2"). See [Panix's pvgrub
779: page](http://www.panix.com/v-colo/grub.html), which describes only
780: Linux but should be updated to cover NetBSD :-).
782: [prgmr.com](http://prgmr.com/) also lets users with pvgrub to boot
783: their own kernel. See then [prgmr.com NetBSD
785: (which is in need of updating).
787: It appears that [grub's FFS
789: does not support all aspects of modern FFS, but there are also reports
790: that FFSv2 works fine. At prgmr, typically one has an ext2 or FAT
791: partition for the kernel with the intent that grub can understand it,
792: which leads to /netbsd not being the actual kernel. One must remember
793: to update the special boot partition.
795: ### pygrub
797: pygrub runs in the dom0 and looks into the domU file system. This
798: implies that the domU must have a kernel in a file system in a format
799: known to pygrub.
801: pygrub doesn't seem to work to load Linux images under NetBSD dom0,
802: and is inherently less secure than pvgrub due to running inside dom0. For both these
803: reasons, pygrub should not be used, and is only still present so that
804: historical DomU images using it still work.
806: As of 2014, pygrub seems to be of mostly historical
807: interest. New DomUs should use pvgrub.
809: ## Specific Providers
811: ### Amazon
813: See the [Amazon EC2 page](/amazon_ec2/).
CVSweb for NetBSD wikisrc <wikimaster@NetBSD.org> software: FreeBSD-CVSweb