1: [[!meta title="Xen HowTo"]]
2:
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
9: fulfilled.
10:
11: Xen supports different styles of guests; see [PV on HVM](https://wiki.xen.org/wiki/PV_on_HVM) and [PVH(v2)](https://wiki.xenproject.org/wiki/PVH_(v2\)_Domu) for upstream documentation.
12:
13: [[!table data="""
14: Style of guest |Supported by NetBSD
15: PV |Yes (dom0, domU)
16: HVM |Yes (domU)
17: PVHVM |current-only (domU)
18: PVH |current-only (domU, dom0 not yet)
19: """]]
20:
21: In Para-Virtualized (PV) mode, the guest OS does not attempt to access
22: hardware directly, but instead makes hypercalls to the hypervisor; PV
23: guests must be specifically coded for Xen.
24:
25: In HVM mode, no guest modification is required; however, hardware
26: support is required, such as VT-x on Intel CPUs and SVM on AMD CPUs.
27: The dom0 runs qemu to emulate hardware.
28:
29: In PVHVM mode, the guest runs as HVM, but additionally can use PV
30: drivers for efficiency.
31:
32: There have been two PVH modes: original PVH and PVHv2. Original PVH
33: was based on PV mode and is no longer relevant at all. PVHv2 is
34: basically lightweight HVM with PV drivers. A critical feature of it
35: is that qemu is not needed; the hypervisor can do the emulation that
36: is required. Thus, a dom0 can be PVHv2.
37:
38: The source code uses PVH and config files use pvh; this refers to PVHv2.
39:
40: At boot, the dom0 kernel is loaded as a module with Xen as the kernel.
41: The dom0 can start one or more domUs. (Booting is explained in detail
42: in the dom0 section.)
43:
44: This HOWTO presumes a basic familiarity with the Xen system
45: architecture, with installing NetBSD on i386/amd64 hardware, and with
46: installing software from pkgsrc. See also the [Xen
47: website](http://www.xenproject.org/).
48:
49: [[!toc]]
50:
51: # Versions and Support
52:
53: In NetBSD, Xen is provided in pkgsrc, via matching pairs of packages
54: xenkernel and xentools. We will refer only to the kernel versions,
55: but note that both packages must be installed together and must have
56: matching versions.
57:
58: Versions available in pkgsrc:
59:
60: [[!table data="""
61: Xen Version |Package Name |Xen CPU Support |xm? |EOL'ed By Upstream
62: 4.11 |xenkernel411 |x86_64 | |No
63: 4.13 |xenkernel413 |x86_64 | |No
64: """]]
65:
66: See also the [Xen Security Advisory page](http://xenbits.xen.org/xsa/).
67:
68: Multiprocessor (SMP) support in NetBSD differs depending on the domain:
69:
70: [[!table data="""
71: Domain |Supports SMP
72: dom0 |No
73: domU |Yes
74: """]]
75:
76: Note: NetBSD support is called XEN3. However, it does support Xen 4,
77: because the hypercall interface has remained identical.
78:
79: Older Xen had a python-based management tool called xm, now replaced
80: by xl.
81:
82: ## Architecture
83:
84: Xen 4.5 and later runs on x86_64 hardware (the NetBSD amd64 port).
85: There is a concept of Xen running on ARM, but there are no reports of this working with NetBSD.
86:
87: The dom0 system should be amd64. (Instructions for i386PAE dom0 have been removed from the HOWTO.)
88:
89: The domU can be i386PAE or amd64.
90: i386PAE at one point was considered as [faster](https://lists.xen.org/archives/html/xen-devel/2012-07/msg00085.html) than amd64.
91:
92: # Creating a dom0
93:
94: In order to install a NetBSD as a dom0, one must first install a normal
95: NetBSD system, and then pivot the install to a dom0 install by changing
96: the kernel and boot configuration.
97:
98: In 2018-05, trouble booting a dom0 was reported with 256M of RAM: with
99: 512M it worked reliably. This does not make sense, but if you see
100: "not ELF" after Xen boots, try increasing dom0 RAM.
101:
102: ## Installation of NetBSD
103:
104: [Install NetBSD/amd64](/guide/inst/)
105: just as you would if you were not using Xen.
106:
107: ## Installation of Xen
108:
109: We will consider that you chose to use Xen 4.13, with NetBSD/amd64 as
110: dom0. In the dom0, install xenkernel48 and xentools48 from pkgsrc.
111:
112: Once this is done, install the Xen kernel itself:
113:
114: [[!template id=programlisting text="""
115: # cp /usr/pkg/xen48-kernel/xen.gz /
116: """]]
117:
118: Then, place a NetBSD XEN3_DOM0 kernel in the `/` directory. Such kernel
119: can either be compiled manually, or downloaded from the NetBSD FTP, for
120: example at:
121:
122: [[!template id=programlisting text="""
123: ftp.netbsd.org/pub/NetBSD/NetBSD-8.0/amd64/binary/kernel/netbsd-XEN3_DOM0.gz
124: """]]
125:
126: Add a line to /boot.cfg to boot Xen:
127:
128: [[!template id=filecontent name="/boot.cfg" text="""
129: menu=Xen:load /netbsd-XEN3_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=512M
130: """]]
131:
132: This specifies that the dom0 should have 512MB of ram, leaving the rest
133: to be allocated for domUs. To use a serial console, use:
134:
135: [[!template id=filecontent name="/boot.cfg" text="""
136: menu=Xen:load /netbsd-XEN3_DOM0.gz;multiboot /xen.gz dom0_mem=512M console=com1 com1=9600,8n1
137: """]]
138:
139: which will use the first serial port for Xen (which counts starting
140: from 1, unlike NetBSD which counts starting from 0), forcing
141: speed/parity. Because the NetBSD command line lacks a
142: "console=pc" argument, it will use the default "xencons" console device,
143: which directs the console I/O through Xen to the same console device Xen
144: itself uses (in this case, the serial port).
145:
146: In an attempt to add performance, one can also add `dom0_max_vcpus=1 dom0_vcpus_pin`,
147: to force only one vcpu to be provided (since NetBSD dom0 can't use
148: more) and to pin that vcpu to a physical CPU. Xen has
149: [many boot options](http://xenbits.xenproject.org/docs/4.13-testing/misc/xen-command-line.html),
150: and other than dom0 memory and max_vcpus, they are generally not
151: necessary.
152:
153: Copy the boot scripts into `/etc/rc.d`:
154:
155: [[!template id=programlisting text="""
156: # cp /usr/pkg/share/examples/rc.d/xen* /etc/rc.d/
157: """]]
158:
159: Enable `xencommons`:
160:
161: [[!template id=filecontent name="/etc/rc.conf" text="""
162: xencommons=YES
163: """]]
164:
165: Now, reboot so that you are running a DOM0 kernel under Xen, rather
166: than GENERIC without Xen.
167:
168: TODO: Recommend for/against xen-watchdog.
169:
170: Once the reboot is done, use `xl` to inspect Xen's boot messages,
171: available resources, and running domains. For example:
172:
173: [[!template id=programlisting text="""
174: # xl dmesg
175: ... xen's boot info ...
176: # xl info
177: ... available memory, etc ...
178: # xl list
179: Name Id Mem(MB) CPU State Time(s) Console
180: Domain-0 0 64 0 r---- 58.1
181: """]]
182:
183: Xen logs will be in /var/log/xen.
184:
185: ### Issues with xencommons
186:
187: `xencommons` starts `xenstored`, which stores data on behalf of dom0 and
188: domUs. It does not currently work to stop and start xenstored.
189: Certainly all domUs should be shutdown first, following the sort order
190: of the rc.d scripts. However, the dom0 sets up state with xenstored,
191: and is not notified when xenstored exits, leading to not recreating
192: the state when the new xenstored starts. Until there's a mechanism to
193: make this work, one should not expect to be able to restart xenstored
194: (and thus xencommons). There is currently no reason to expect that
195: this will get fixed any time soon.
196:
197: ## anita (for testing NetBSD)
198:
199: With the setup so far, one should be able to run
200: anita (see pkgsrc/misc/py-anita) to test NetBSD releases, by doing (as
201: root, because anita must create a domU):
202:
203: [[!template id=programlisting text="""
204: anita --vmm=xl test file:///usr/obj/i386/
205: """]]
206:
207: ## Xen-specific NetBSD issues
208:
209: There are (at least) two additional things different about NetBSD as a
210: dom0 kernel compared to hardware.
211:
212: One is that the module ABI is different because some of the #defines
213: change, so one must build modules for Xen. As of netbsd-7, the build
214: system does this automatically.
215:
216: The other difference is that XEN3_DOM0 does not have exactly the same
217: options as GENERIC. While it is debatable whether or not this is a
218: bug, users should be aware of this and can simply add missing config
219: items if desired.
220:
221: ## Updating NetBSD in a dom0
222:
223: This is just like updating NetBSD on bare hardware, assuming the new
224: version supports the version of Xen you are running. Generally, one
225: replaces the kernel and reboots, and then overlays userland binaries
226: and adjusts `/etc`.
227:
228: Note that one must update both the non-Xen kernel typically used for
229: rescue purposes and the DOM0 kernel used with Xen.
230:
231: ## Converting from grub to /boot
232:
233: These instructions were used to convert a system from
234: grub to /boot. The system was originally installed in February of
235: 2006 with a RAID1 setup and grub to boot Xen 2, and has been updated
236: over time. Before these commands, it was running NetBSD 6 i386, Xen
237: 4.1 and grub, much like the message linked earlier in the grub
238: section.
239:
240: [[!template id=programlisting text="""
241: # Install MBR bootblocks on both disks.
242: fdisk -i /dev/rwd0d
243: fdisk -i /dev/rwd1d
244: # Install NetBSD primary boot loader (/ is FFSv1) into RAID1 components.
245: installboot -v /dev/rwd0d /usr/mdec/bootxx_ffsv1
246: installboot -v /dev/rwd1d /usr/mdec/bootxx_ffsv1
247: # Install secondary boot loader
248: cp -p /usr/mdec/boot /
249: # Create boot.cfg following earlier guidance:
250: menu=Xen:load /netbsd-XEN3PAE_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=512M
251: menu=Xen.ok:load /netbsd-XEN3PAE_DOM0.ok.gz console=pc;multiboot /xen.ok.gz dom0_mem=512M
252: menu=GENERIC:boot
253: menu=GENERIC single-user:boot -s
254: menu=GENERIC.ok:boot netbsd.ok
255: menu=GENERIC.ok single-user:boot netbsd.ok -s
256: menu=Drop to boot prompt:prompt
257: default=1
258: timeout=30
259: """]]
260:
261: ## Upgrading Xen versions
262:
263: Minor version upgrades are trivial. Just rebuild/replace the
264: xenkernel version and copy the new xen.gz to `/` (where `/boot.cfg`
265: references it), and reboot.
266:
267: #Unprivileged domains (domU)
268:
269: This section describes general concepts about domUs. It does not
270: address specific domU operating systems or how to install them. The
271: config files for domUs are typically in `/usr/pkg/etc/xen`, and are
272: typically named so that the file name, domU name and the domU's host
273: name match.
274:
275: The domU is provided with CPU and memory by Xen, configured by the
276: dom0. The domU is provided with disk and network by the dom0,
277: mediated by Xen, and configured in the dom0.
278:
279: Entropy in domUs can be an issue; physical disks and network are on
280: the dom0. NetBSD's /dev/random system works, but is often challenged.
281:
282: ## Config files
283:
284: See /usr/pkg/share/examples/xen/xlexample*
285: for a small number of well-commented examples, mostly for running
286: GNU/Linux.
287:
288: The following is an example minimal domain configuration file. The domU
289: serves as a network file server.
290:
291: [[!template id=filecontent name="/usr/pkg/etc/xen/foo" text="""
292: name = "domU-id"
293: kernel = "/netbsd-XEN3PAE_DOMU-i386-foo.gz"
294: memory = 1024
295: vif = [ 'mac=aa:00:00:d1:00:09,bridge=bridge0' ]
296: disk = [ 'file:/n0/xen/foo-wd0,0x0,w',
297: 'file:/n0/xen/foo-wd1,0x1,w' ]
298: """]]
299:
300: The domain will have name given in the `name` setting. The kernel has the
301: host/domU name in it, so that on the dom0 one can update the various
302: domUs independently. The `vif` line causes an interface to be provided,
303: with a specific mac address (do not reuse MAC addresses!), in bridge
304: mode. Two disks are provided, and they are both writable; the bits
305: are stored in files and Xen attaches them to a vnd(4) device in the
306: dom0 on domain creation. The system treats xbd0 as the boot device
307: without needing explicit configuration.
308:
309: By convention, domain config files are kept in `/usr/pkg/etc/xen`. Note
310: that "xl create" takes the name of a config file, while other commands
311: take the name of a domain.
312:
313: Examples of commands:
314:
315: [[!template id=programlisting text="""
316: xl create /usr/pkg/etc/xen/foo
317: xl console domU-id
318: xl create -c /usr/pkg/etc/xen/foo
319: xl shutdown domU-id
320: xl list
321: """]]
322:
323: Typing `^]` will exit the console session. Shutting down a domain is
324: equivalent to pushing the power button; a NetBSD domU will receive a
325: power-press event and do a clean shutdown. Shutting down the dom0
326: will trigger controlled shutdowns of all configured domUs.
327:
328: ## CPU and memory
329:
330: A domain is provided with some number of vcpus, up to the number
331: of CPUs seen by the hypervisor. For a domU, it is controlled
332: from the config file by the "vcpus = N" directive.
333:
334: A domain is provided with memory; this is controlled in the config
335: file by "memory = N" (in megabytes). In the straightforward case, the
336: sum of the the memory allocated to the dom0 and all domUs must be less
337: than the available memory.
338:
339: Xen also provides a "balloon" driver, which can be used to let domains
340: use more memory temporarily.
341:
342: ## Virtual disks
343:
344: In domU config files, the disks are defined as a sequence of 3-tuples:
345:
346: * The first element is "method:/path/to/disk". Common methods are
347: "file:" for a file-backed vnd, and "phy:" for something that is already
348: a device, such as an LVM logical volume.
349:
350: * The second element is an artifact of how virtual disks are passed to
351: Linux, and a source of confusion with NetBSD Xen usage. Linux domUs
352: are given a device name to associate with the disk, and values like
353: "hda1" or "sda1" are common. In a NetBSD domU, the first disk appears
354: as xbd0, the second as xbd1, and so on. However, xl demands a
355: second argument. The name given is converted to a major/minor by
356: calling stat(2) on the name in /dev and this is passed to the domU.
357: In the general case, the dom0 and domU can be different operating
358: systems, and it is an unwarranted assumption that they have consistent
359: numbering in /dev, or even that the dom0 OS has a /dev. With NetBSD
360: as both dom0 and domU, using values of 0x0 for the first disk and 0x1
361: for the second works fine and avoids this issue. For a GNU/Linux
362: guest, one can create /dev/hda1 in /dev, or to pass 0x301 for
363: /dev/hda1.
364:
365: * The third element is "w" for writable disks, and "r" for read-only
366: disks.
367:
368: Example:
369: [[!template id=filecontent name="/usr/pkg/etc/xen/foo" text="""
370: disk = [ 'file:/n0/xen/foo-wd0,0x0,w' ]
371: """]]
372:
373: Note that NetBSD by default creates only vnd[0123]. If you need more
374: than 4 total virtual disks at a time, run e.g. "./MAKEDEV vnd4" in the
375: dom0.
376:
377: Note that NetBSD by default creates only xbd[0123]. If you need more
378: virtual disks in a domU, run e.g. "./MAKEDEV xbd4" in the domU.
379:
380: Virtual Networking
381: ------------------
382:
383: Xen provides virtual Ethernets, each of which connects the dom0 and a
384: domU. For each virtual network, there is an interface "xvifN.M" in
385: the dom0, and a matching interface xennetM (NetBSD name) in domU index N.
386: The interfaces behave as if there is an Ethernet with two
387: adapters connected. From this primitive, one can construct various
388: configurations. We focus on two common and useful cases for which
389: there are existing scripts: bridging and NAT.
390:
391: With bridging (in the example above), the domU perceives itself to be
392: on the same network as the dom0. For server virtualization, this is
393: usually best. Bridging is accomplished by creating a bridge(4) device
394: and adding the dom0's physical interface and the various xvifN.0
395: interfaces to the bridge. One specifies "bridge=bridge0" in the domU
396: config file. The bridge must be set up already in the dom0; an
397: example /etc/ifconfig.bridge0 is:
398:
399: [[!template id=filecontent name="/etc/ifconfig.bridge0" text="""
400: create
401: up
402: !brconfig bridge0 add wm0
403: """]]
404:
405: With NAT, the domU perceives itself to be behind a NAT running on the
406: dom0. This is often appropriate when running Xen on a workstation.
407: TODO: NAT appears to be configured by "vif = [ '' ]".
408:
409: The MAC address specified is the one used for the interface in the new
410: domain. The interface in dom0 will use this address XOR'd with
411: 00:00:00:01:00:00. Random MAC addresses are assigned if not given.
412:
413: Starting domains automatically
414: ------------------------------
415:
416: To start domains `domU-netbsd` and `domU-linux` at boot and shut them
417: down cleanly on dom0 shutdown, add the following in rc.conf:
418:
419: [[!template id=filecontent name="/etc/rc.conf" text="""
420: xendomains="domU-netbsd domU-linux"
421: """]]
422:
423: # Creating a domU
424:
425: Creating domUs is almost entirely independent of operating system. We
426: have already presented the basics of config files. Note that you must
427: have already completed the dom0 setup so that "xl list" works.
428:
429: Creating a NetBSD PV domU
430: --------------------------
431:
432: See the earlier config file, and adjust memory. Decide on how much
433: storage you will provide, and prepare it (file or LVM).
434:
435: While the kernel will be obtained from the dom0 file system, the same
436: file should be present in the domU as /netbsd so that tools like
437: savecore(8) can work. (This is helpful but not necessary.)
438:
439: The kernel must be specifically for Xen and for use as a domU. The
440: i386 and amd64 provide the following kernels:
441:
442: i386 XEN3PAE_DOMU
443: amd64 XEN3_DOMU
444:
445: This will boot NetBSD, but this is not that useful if the disk is
446: empty. One approach is to unpack sets onto the disk outside of xen
447: (by mounting it, just as you would prepare a physical disk for a
448: system you can't run the installer on).
449:
450: A second approach is to run an INSTALL kernel, which has a miniroot
451: and can load sets from the network. To do this, copy the INSTALL
452: kernel to / and change the kernel line in the config file to:
453:
454: kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
455:
456: Then, start the domain as "xl create -c configfile".
457:
458: Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following
459: line should be used in the config file.
460:
461: disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
462:
463: After booting the domain, the option to install via CDROM may be
464: selected. The CDROM device should be changed to `xbd1d`.
465:
466: Once done installing, "halt -p" the new domain (don't reboot or halt,
467: it would reload the INSTALL_XEN3_DOMU kernel even if you changed the
468: config file), switch the config file back to the XEN3_DOMU kernel,
469: and start the new domain again. Now it should be able to use "root on
470: xbd0a" and you should have a, functional NetBSD domU.
471:
472: TODO: check if this is still accurate.
473: When the new domain is booting you'll see some warnings about *wscons*
474: and the pseudo-terminals. These can be fixed by editing the files
475: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
476: `/etc/ttys`, except *console*, like this:
477:
478: console "/usr/libexec/getty Pc" vt100 on secure
479: ttyE0 "/usr/libexec/getty Pc" vt220 off secure
480: ttyE1 "/usr/libexec/getty Pc" vt220 off secure
481: ttyE2 "/usr/libexec/getty Pc" vt220 off secure
482: ttyE3 "/usr/libexec/getty Pc" vt220 off secure
483:
484: Finally, all screens must be commented out from `/etc/wscons.conf`.
485:
486: It is also desirable to add
487:
488: powerd=YES
489:
490: in rc.conf. This way, the domain will be properly shut down if
491: `xm shutdown -R` or `xm shutdown -H` is used on the dom0.
492:
493: It is not strictly necessary to have a kernel (as /netbsd) in the domU
494: file system. However, various programs (e.g. netstat) will use that
495: kernel to look up symbols to read from kernel virtual memory. If
496: /netbsd is not the running kernel, those lookups will fail. (This is
497: not really a Xen-specific issue, but because the domU kernel is
498: obtained from the dom0, it is far more likely to be out of sync or
499: missing with Xen.)
500:
501: Creating a Linux domU
502: ---------------------
503:
504: Creating unprivileged Linux domains isn't much different from
505: unprivileged NetBSD domains, but there are some details to know.
506:
507: First, the second parameter passed to the disk declaration (the '0x1' in
508: the example below)
509:
510: disk = [ 'phy:/dev/wd0e,0x1,w' ]
511:
512: does matter to Linux. It wants a Linux device number here (e.g. 0x300
513: for hda). Linux builds device numbers as: (major \<\< 8 + minor).
514: So, hda1 which has major 3 and minor 1 on a Linux system will have
515: device number 0x301. Alternatively, devices names can be used (hda,
516: hdb, ...) as xentools has a table to map these names to devices
517: numbers. To export a partition to a Linux guest we can use:
518:
519: disk = [ 'phy:/dev/wd0e,0x300,w' ]
520: root = "/dev/hda1 ro"
521:
522: and it will appear as /dev/hda on the Linux system, and be used as root
523: partition.
524:
525: To install the Linux system on the partition to be exported to the
526: guest domain, the following method can be used: install
527: sysutils/e2fsprogs from pkgsrc. Use mke2fs to format the partition
528: that will be the root partition of your Linux domain, and mount it.
529: Then copy the files from a working Linux system, make adjustments in
530: `/etc` (fstab, network config). It should also be possible to extract
531: binary packages such as .rpm or .deb directly to the mounted partition
532: using the appropriate tool, possibly running under NetBSD's Linux
533: emulation. Once the file system has been populated, umount it. If
534: desirable, the file system can be converted to ext3 using tune2fs -j.
535: It should now be possible to boot the Linux guest domain, using one of
536: the vmlinuz-\*-xenU kernels available in the Xen binary distribution.
537:
538: To get the Linux console right, you need to add:
539:
540: extra = "xencons=tty1"
541:
542: to your configuration since not all Linux distributions auto-attach a
543: tty to the xen console.
544:
545: ## Creating a NetBSD HVM domU
546:
547: Use type='hmv', probably. Use a GENERIC kernel within the disk image.
548:
549: ## Creating a NetBSD PVH domU
550:
551: Use type='pvh'.
552:
553: \todo Explain where the kernel comes from.
554:
555:
556: Creating a Solaris domU
557: -----------------------
558:
559: See possibly outdated
560: [Solaris domU instructions](/ports/xen/howto-solaris/).
561:
562:
563: PCI passthrough: Using PCI devices in guest domains
564: ---------------------------------------------------
565:
566: NB: PCI passthrough only works on some Xen versions and as of 2020 it
567: is not clear that it works on any version in pkgsrc. Reports
568: confirming or denying this notion should be sent to port-xen@.
569:
570: The dom0 can give other domains access to selected PCI
571: devices. This can allow, for example, a non-privileged domain to have
572: access to a physical network interface or disk controller. However,
573: keep in mind that giving a domain access to a PCI device most likely
574: will give the domain read/write access to the whole physical memory,
575: as PCs don't have an IOMMU to restrict memory access to DMA-capable
576: device. Also, it's not possible to export ISA devices to non-dom0
577: domains, which means that the primary VGA adapter can't be exported.
578: A guest domain trying to access the VGA registers will panic.
579:
580: If the dom0 is NetBSD, it has to be running Xen 3.1, as support has
581: not been ported to later versions at this time.
582:
583: For a PCI device to be exported to a domU, is has to be attached to
584: the "pciback" driver in dom0. Devices passed to the dom0 via the
585: pciback.hide boot parameter will attach to "pciback" instead of the
586: usual driver. The list of devices is specified as "(bus:dev.func)",
587: where bus and dev are 2-digit hexadecimal numbers, and func a
588: single-digit number:
589:
590: pciback.hide=(00:0a.0)(00:06.0)
591:
592: pciback devices should show up in the dom0's boot messages, and the
593: devices should be listed in the `/kern/xen/pci` directory.
594:
595: PCI devices to be exported to a domU are listed in the "pci" array of
596: the domU's config file, with the format "0000:bus:dev.func".
597:
598: pci = [ '0000:00:06.0', '0000:00:0a.0' ]
599:
600: In the domU an "xpci" device will show up, to which one or more pci
601: buses will attach. Then the PCI drivers will attach to PCI buses as
602: usual. Note that the default NetBSD DOMU kernels do not have "xpci"
603: or any PCI drivers built in by default; you have to build your own
604: kernel to use PCI devices in a domU. Here's a kernel config example;
605: note that only the "xpci" lines are unusual.
606:
607: include "arch/i386/conf/XEN3_DOMU"
608:
609: # Add support for PCI buses to the XEN3_DOMU kernel
610: xpci* at xenbus ?
611: pci* at xpci ?
612:
613: # PCI USB controllers
614: uhci* at pci? dev ? function ? # Universal Host Controller (Intel)
615:
616: # USB bus support
617: usb* at uhci?
618:
619: # USB Hubs
620: uhub* at usb?
621: uhub* at uhub? port ? configuration ? interface ?
622:
623: # USB Mass Storage
624: umass* at uhub? port ? configuration ? interface ?
625: wd* at umass?
626: # SCSI controllers
627: ahc* at pci? dev ? function ? # Adaptec [23]94x, aic78x0 SCSI
628:
629: # SCSI bus support (for both ahc and umass)
630: scsibus* at scsi?
631:
632: # SCSI devices
633: sd* at scsibus? target ? lun ? # SCSI disk drives
634: cd* at scsibus? target ? lun ? # SCSI CD-ROM drives
635:
636:
637: # Specific Issues
638:
639: ## domU
640:
641: [NetBSD 5 is known to panic.](http://mail-index.netbsd.org/port-xen/2018/04/17/msg009181.html)
642: (However, NetBSD 5 systems should be updated to a supported version.)
643:
644: # NetBSD as a domU in a VPS
645:
646: The bulk of the HOWTO is about using NetBSD as a dom0 on your own
647: hardware. This section explains how to deal with Xen in a domU as a
648: virtual private server where you do not control or have access to the
649: dom0. This is not intended to be an exhaustive list of VPS providers;
650: only a few are mentioned that specifically support NetBSD.
651:
652: VPS operators provide varying degrees of access and mechanisms for
653: configuration. The big issue is usually how one controls which kernel
654: is booted, because the kernel is nominally in the dom0 file system (to
655: which VPS users do not normally have access). A second issue is how
656: to install NetBSD.
657: A VPS user may want to compile a kernel for security updates, to run
658: npf, run IPsec, or any other reason why someone would want to change
659: their kernel.
660:
661: One approach is to have an administrative interface to upload a kernel,
662: or to select from a prepopulated list. Other approaches are pygrub
663: (deprecated) and pvgrub, which are ways to have a bootloader obtain a
664: kernel from the domU file system. This is closer to a regular physical
665: computer, where someone who controls a machine can replace the kernel.
666:
667: A second issue is multiple CPUs. With NetBSD 6, domUs support
668: multiple vcpus, and it is typical for VPS providers to enable multiple
669: CPUs for NetBSD domUs.
670:
671: ## Complexities due to Xen changes
672:
673: Xen has many security advisories and people running Xen systems make
674: different choices.
675:
676: ### stub domains
677:
678: Some (Linux only?) dom0 systems use something called "stub domains" to
679: isolate qemu from the dom0 system, as a security and reliabilty
680: mechanism when running HVM domUs. Somehow, NetBSD's GENERIC kernel
681: ends up using PIO for disks rather than DMA. Of course, all of this
682: is emulated, but emulated PIO is unusably slow. This problem is not
683: currently understood.
684:
685: ### Grant tables
686:
687: There are multiple versions of using grant tables, and some security
688: advisories have suggested disabling some versions. Some versions of
689: NetBSD apparently only use specific versions and this can lead to
690: "NetBSD current doesn't run on hosting provider X" situations.
691:
692: \todo Explain better.
693:
694: ## Boot methods
695:
696: ### pvgrub
697:
698: pvgrub is a version of grub that uses PV operations instead of BIOS
699: calls. It is booted from the dom0 as the domU kernel, and then reads
700: /grub/menu.lst and loads a kernel from the domU file system.
701:
702: [Panix](http://www.panix.com/) lets users use pvgrub. Panix reports
703: that pvgrub works with FFsv2 with 16K/2K and 32K/4K block/frag sizes
704: (and hence with defaults from "newfs -O 2"). See [Panix's pvgrub
705: page](http://www.panix.com/v-colo/grub.html), which describes only
706: Linux but should be updated to cover NetBSD :-).
707:
708: [prgmr.com](http://prgmr.com/) also lets users with pvgrub to boot
709: their own kernel. See then [prgmr.com NetBSD
710: HOWTO](http://wiki.prgmr.com/mediawiki/index.php/NetBSD_as_a_DomU)
711: (which is in need of updating).
712:
713: It appears that [grub's FFS
714: code](http://xenbits.xensource.com/hg/xen-unstable.hg/file/bca284f67702/tools/libfsimage/ufs/fsys_ufs.c)
715: does not support all aspects of modern FFS, but there are also reports
716: that FFSv2 works fine. At prgmr, typically one has an ext2 or FAT
717: partition for the kernel with the intent that grub can understand it,
718: which leads to /netbsd not being the actual kernel. One must remember
719: to update the special boot partition.
720:
721: ### pygrub
722:
723: pygrub runs in the dom0 and looks into the domU file system. This
724: implies that the domU must have a kernel in a file system in a format
725: known to pygrub.
726:
727: pygrub doesn't seem to work to load Linux images under NetBSD dom0,
728: and is inherently less secure than pvgrub due to running inside dom0. For both these
729: reasons, pygrub should not be used, and is only still present so that
730: historical DomU images using it still work.
731:
732: As of 2014, pygrub seems to be of mostly historical
733: interest. New DomUs should use pvgrub.
734:
735: ## Specific Providers
736:
737: ### Amazon
738:
739: See the [Amazon EC2 page](/amazon_ec2/).
CVSweb for NetBSD wikisrc <wikimaster@NetBSD.org> software: FreeBSD-CVSweb
![[BACK]](/icons/back.gif){kind=icon}
Return to howto.mdwn CVS log ![[TXT]](/icons/text.gif){kind=icon}
![[DIR]](/icons/dir.gif){kind=icon}
Up to [NetBSD Developer Wiki] / wikisrc / ports / xen