You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

qemu-deprecated.texi 15KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352
  1. @node Deprecated features
  2. @appendix Deprecated features
  3. In general features are intended to be supported indefinitely once
  4. introduced into QEMU. In the event that a feature needs to be removed,
  5. it will be listed in this appendix. The feature will remain functional
  6. for 2 releases prior to actual removal. Deprecated features may also
  7. generate warnings on the console when QEMU starts up, or if activated
  8. via a monitor command, however, this is not a mandatory requirement.
  9. Prior to the 2.10.0 release there was no official policy on how
  10. long features would be deprecated prior to their removal, nor
  11. any documented list of which features were deprecated. Thus
  12. any features deprecated prior to 2.10.0 will be treated as if
  13. they were first deprecated in the 2.10.0 release.
  14. What follows is a list of all features currently marked as
  15. deprecated.
  16. @section System emulator command line arguments
  17. @subsection -machine enforce-config-section=on|off (since 3.1)
  18. The @option{enforce-config-section} parameter is replaced by the
  19. @option{-global migration.send-configuration=@var{on|off}} option.
  20. @subsection -no-kvm (since 1.3.0)
  21. The ``-no-kvm'' argument is now a synonym for setting
  22. ``-machine accel=tcg''.
  23. @subsection -usbdevice (since 2.10.0)
  24. The ``-usbdevice DEV'' argument is now a synonym for setting
  25. the ``-device usb-DEV'' argument instead. The deprecated syntax
  26. would automatically enable USB support on the machine type.
  27. If using the new syntax, USB support must be explicitly
  28. enabled via the ``-machine usb=on'' argument.
  29. @subsection -drive file=json:@{...@{'driver':'file'@}@} (since 3.0)
  30. The 'file' driver for drives is no longer appropriate for character or host
  31. devices and will only accept regular files (S_IFREG). The correct driver
  32. for these file types is 'host_cdrom' or 'host_device' as appropriate.
  33. @subsection -net ...,name=@var{name} (since 3.1)
  34. The @option{name} parameter of the @option{-net} option is a synonym
  35. for the @option{id} parameter, which should now be used instead.
  36. @subsection -smp (invalid topologies) (since 3.1)
  37. CPU topology properties should describe whole machine topology including
  38. possible CPUs.
  39. However, historically it was possible to start QEMU with an incorrect topology
  40. where @math{@var{n} <= @var{sockets} * @var{cores} * @var{threads} < @var{maxcpus}},
  41. which could lead to an incorrect topology enumeration by the guest.
  42. Support for invalid topologies will be removed, the user must ensure
  43. topologies described with -smp include all possible cpus, i.e.
  44. @math{@var{sockets} * @var{cores} * @var{threads} = @var{maxcpus}}.
  45. @subsection -vnc acl (since 4.0.0)
  46. The @code{acl} option to the @code{-vnc} argument has been replaced
  47. by the @code{tls-authz} and @code{sasl-authz} options.
  48. @subsection QEMU_AUDIO_ environment variables and -audio-help (since 4.0)
  49. The ``-audiodev'' argument is now the preferred way to specify audio
  50. backend settings instead of environment variables. To ease migration to
  51. the new format, the ``-audiodev-help'' option can be used to convert
  52. the current values of the environment variables to ``-audiodev'' options.
  53. @subsection -mon ...,control=readline,pretty=on|off (since 4.1)
  54. The @code{pretty=on|off} switch has no effect for HMP monitors, but is
  55. silently ignored. Using the switch with HMP monitors will become an
  56. error in the future.
  57. @subsection -realtime (since 4.1)
  58. The @code{-realtime mlock=on|off} argument has been replaced by the
  59. @code{-overcommit mem-lock=on|off} argument.
  60. @subsection -virtfs_synth (since 4.1)
  61. The ``-virtfs_synth'' argument is now deprecated. Please use ``-fsdev synth''
  62. and ``-device virtio-9p-...'' instead.
  63. @subsection -numa node,mem=@var{size} (since 4.1)
  64. The parameter @option{mem} of @option{-numa node} is used to assign a part of
  65. guest RAM to a NUMA node. But when using it, it's impossible to manage specified
  66. RAM chunk on the host side (like bind it to a host node, setting bind policy, ...),
  67. so guest end-ups with the fake NUMA configuration with suboptiomal performance.
  68. However since 2014 there is an alternative way to assign RAM to a NUMA node
  69. using parameter @option{memdev}, which does the same as @option{mem} and adds
  70. means to actualy manage node RAM on the host side. Use parameter @option{memdev}
  71. with @var{memory-backend-ram} backend as an replacement for parameter @option{mem}
  72. to achieve the same fake NUMA effect or a properly configured
  73. @var{memory-backend-file} backend to actually benefit from NUMA configuration.
  74. In future new machine versions will not accept the option but it will still
  75. work with old machine types. User can check QAPI schema to see if the legacy
  76. option is supported by looking at MachineInfo::numa-mem-supported property.
  77. @subsection -numa node (without memory specified) (since 4.1)
  78. Splitting RAM by default between NUMA nodes has the same issues as @option{mem}
  79. parameter described above with the difference that the role of the user plays
  80. QEMU using implicit generic or board specific splitting rule.
  81. Use @option{memdev} with @var{memory-backend-ram} backend or @option{mem} (if
  82. it's supported by used machine type) to define mapping explictly instead.
  83. @subsection -mem-path fallback to RAM (since 4.1)
  84. Currently if guest RAM allocation from file pointed by @option{mem-path}
  85. fails, QEMU falls back to allocating from RAM, which might result
  86. in unpredictable behavior since the backing file specified by the user
  87. is ignored. In the future, users will be responsible for making sure
  88. the backing storage specified with @option{-mem-path} can actually provide
  89. the guest RAM configured with @option{-m} and QEMU will fail to start up if
  90. RAM allocation is unsuccessful.
  91. @subsection RISC-V -bios (since 4.1)
  92. QEMU 4.1 introduced support for the -bios option in QEMU for RISC-V for the
  93. RISC-V virt machine and sifive_u machine.
  94. QEMU 4.1 has no changes to the default behaviour to avoid breakages. This
  95. default will change in a future QEMU release, so please prepare now. All users
  96. of the virt or sifive_u machine must change their command line usage.
  97. QEMU 4.1 has three options, please migrate to one of these three:
  98. 1. ``-bios none`` - This is the current default behavior if no -bios option
  99. is included. QEMU will not automatically load any firmware. It is up
  100. to the user to load all the images they need.
  101. 2. ``-bios default`` - In a future QEMU release this will become the default
  102. behaviour if no -bios option is specified. This option will load the
  103. default OpenSBI firmware automatically. The firmware is included with
  104. the QEMU release and no user interaction is required. All a user needs
  105. to do is specify the kernel they want to boot with the -kernel option
  106. 3. ``-bios <file>`` - Tells QEMU to load the specified file as the firmwrae.
  107. @section QEMU Machine Protocol (QMP) commands
  108. @subsection block-dirty-bitmap-add "autoload" parameter (since 2.12.0)
  109. "autoload" parameter is now ignored. All bitmaps are automatically loaded
  110. from qcow2 images.
  111. @subsection query-block result field dirty-bitmaps[i].status (since 4.0)
  112. The ``status'' field of the ``BlockDirtyInfo'' structure, returned by
  113. the query-block command is deprecated. Two new boolean fields,
  114. ``recording'' and ``busy'' effectively replace it.
  115. @subsection query-block result field dirty-bitmaps (Since 4.2)
  116. The ``dirty-bitmaps`` field of the ``BlockInfo`` structure, returned by
  117. the query-block command is itself now deprecated. The ``dirty-bitmaps``
  118. field of the ``BlockDeviceInfo`` struct should be used instead, which is the
  119. type of the ``inserted`` field in query-block replies, as well as the
  120. type of array items in query-named-block-nodes.
  121. Since the ``dirty-bitmaps`` field is optionally present in both the old and
  122. new locations, clients must use introspection to learn where to anticipate
  123. the field if/when it does appear in command output.
  124. @subsection query-cpus (since 2.12.0)
  125. The ``query-cpus'' command is replaced by the ``query-cpus-fast'' command.
  126. @subsection query-cpus-fast "arch" output member (since 3.0.0)
  127. The ``arch'' output member of the ``query-cpus-fast'' command is
  128. replaced by the ``target'' output member.
  129. @subsection cpu-add (since 4.0)
  130. Use ``device_add'' for hotplugging vCPUs instead of ``cpu-add''. See
  131. documentation of ``query-hotpluggable-cpus'' for additional
  132. details.
  133. @subsection query-events (since 4.0)
  134. The ``query-events'' command has been superseded by the more powerful
  135. and accurate ``query-qmp-schema'' command.
  136. @subsection chardev client socket with 'wait' option (since 4.0)
  137. Character devices creating sockets in client mode should not specify
  138. the 'wait' field, which is only applicable to sockets in server mode
  139. @section Human Monitor Protocol (HMP) commands
  140. @subsection The hub_id parameter of 'hostfwd_add' / 'hostfwd_remove' (since 3.1)
  141. The @option{[hub_id name]} parameter tuple of the 'hostfwd_add' and
  142. 'hostfwd_remove' HMP commands has been replaced by @option{netdev_id}.
  143. @subsection cpu-add (since 4.0)
  144. Use ``device_add'' for hotplugging vCPUs instead of ``cpu-add''. See
  145. documentation of ``query-hotpluggable-cpus'' for additional details.
  146. @subsection acl_show, acl_reset, acl_policy, acl_add, acl_remove (since 4.0.0)
  147. The ``acl_show'', ``acl_reset'', ``acl_policy'', ``acl_add'', and
  148. ``acl_remove'' commands are deprecated with no replacement. Authorization
  149. for VNC should be performed using the pluggable QAuthZ objects.
  150. @section Guest Emulator ISAs
  151. @subsection RISC-V ISA privledge specification version 1.09.1 (since 4.1)
  152. The RISC-V ISA privledge specification version 1.09.1 has been deprecated.
  153. QEMU supports both the newer version 1.10.0 and the ratified version 1.11.0, these
  154. should be used instead of the 1.09.1 version.
  155. @section System emulator CPUS
  156. @subsection RISC-V ISA CPUs (since 4.1)
  157. The RISC-V cpus with the ISA version in the CPU name have been depcreated. The
  158. four CPUs are: ``rv32gcsu-v1.9.1``, ``rv32gcsu-v1.10.0``, ``rv64gcsu-v1.9.1`` and
  159. ``rv64gcsu-v1.10.0``. Instead the version can be specified via the CPU ``priv_spec``
  160. option when using the ``rv32`` or ``rv64`` CPUs.
  161. @subsection RISC-V ISA CPUs (since 4.1)
  162. The RISC-V no MMU cpus have been depcreated. The two CPUs: ``rv32imacu-nommu`` and
  163. ``rv64imacu-nommu`` should no longer be used. Instead the MMU status can be specified
  164. via the CPU ``mmu`` option when using the ``rv32`` or ``rv64`` CPUs.
  165. @section System emulator devices
  166. @subsection bluetooth (since 3.1)
  167. The bluetooth subsystem is unmaintained since many years and likely bitrotten
  168. quite a bit. It will be removed without replacement unless some users speaks
  169. up at the @email{qemu-devel@@nongnu.org} mailing list with information about
  170. their usecases.
  171. @section System emulator machines
  172. @subsection pc-0.12, pc-0.13, pc-0.14 and pc-0.15 (since 4.0)
  173. These machine types are very old and likely can not be used for live migration
  174. from old QEMU versions anymore. A newer machine type should be used instead.
  175. @subsection prep (PowerPC) (since 3.1)
  176. This machine type uses an unmaintained firmware, broken in lots of ways,
  177. and unable to start post-2004 operating systems. 40p machine type should be
  178. used instead.
  179. @subsection spike_v1.9.1 and spike_v1.10 (since 4.1)
  180. The version specific Spike machines have been deprecated in favour of the
  181. generic ``spike`` machine. If you need to specify an older version of the RISC-V
  182. spec you can use the ``-cpu rv64gcsu,priv_spec=v1.9.1`` command line argument.
  183. @section Device options
  184. @subsection Block device options
  185. @subsubsection "backing": "" (since 2.12.0)
  186. In order to prevent QEMU from automatically opening an image's backing
  187. chain, use ``"backing": null'' instead.
  188. @subsubsection rbd keyvalue pair encoded filenames: "" (since 3.1.0)
  189. Options for ``rbd'' should be specified according to its runtime options,
  190. like other block drivers. Legacy parsing of keyvalue pair encoded
  191. filenames is useful to open images with the old format for backing files;
  192. These image files should be updated to use the current format.
  193. Example of legacy encoding:
  194. @code{json:@{"file.driver":"rbd", "file.filename":"rbd:rbd/name"@}}
  195. The above, converted to the current supported format:
  196. @code{json:@{"file.driver":"rbd", "file.pool":"rbd", "file.image":"name"@}}
  197. @section Related binaries
  198. @subsection qemu-nbd --partition (since 4.0.0)
  199. The ``qemu-nbd --partition $digit'' code (also spelled @option{-P})
  200. can only handle MBR partitions, and has never correctly handled
  201. logical partitions beyond partition 5. If you know the offset and
  202. length of the partition (perhaps by using @code{sfdisk} within the
  203. guest), you can achieve the effect of exporting just that subset of
  204. the disk by use of the @option{--image-opts} option with a raw
  205. blockdev using the @code{offset} and @code{size} parameters layered on
  206. top of any other existing blockdev. For example, if partition 1 is
  207. 100MiB long starting at 1MiB, the old command:
  208. @code{qemu-nbd -t -P 1 -f qcow2 file.qcow2}
  209. can be rewritten as:
  210. @code{qemu-nbd -t --image-opts driver=raw,offset=1M,size=100M,file.driver=qcow2,file.backing.driver=file,file.backing.filename=file.qcow2}
  211. Alternatively, the @code{nbdkit} project provides a more powerful
  212. partition filter on top of its nbd plugin, which can be used to select
  213. an arbitrary MBR or GPT partition on top of any other full-image NBD
  214. export. Using this to rewrite the above example results in:
  215. @code{qemu-nbd -t -k /tmp/sock -f qcow2 file.qcow2 &}
  216. @code{nbdkit -f --filter=partition nbd socket=/tmp/sock partition=1}
  217. Note that if you are exposing the export via /dev/nbd0, it is easier
  218. to just export the entire image and then mount only /dev/nbd0p1 than
  219. it is to reinvoke @command{qemu-nbd -c /dev/nbd0} limited to just a
  220. subset of the image.
  221. @subsection qemu-img convert -n -o (since 4.2.0)
  222. All options specified in @option{-o} are image creation options, so
  223. they have no effect when used with @option{-n} to skip image creation.
  224. Silently ignored options can be confusing, so this combination of
  225. options will be made an error in future versions.
  226. @section Build system
  227. @subsection Python 2 support (since 4.1.0)
  228. In the future, QEMU will require Python 3 to be available at
  229. build time. Support for Python 2 in scripts shipped with QEMU
  230. is deprecated.
  231. @section Backwards compatibility
  232. @subsection Runnability guarantee of CPU models (since 4.1.0)
  233. Previous versions of QEMU never changed existing CPU models in
  234. ways that introduced additional host software or hardware
  235. requirements to the VM. This allowed management software to
  236. safely change the machine type of an existing VM without
  237. introducing new requirements ("runnability guarantee"). This
  238. prevented CPU models from being updated to include CPU
  239. vulnerability mitigations, leaving guests vulnerable in the
  240. default configuration.
  241. The CPU model runnability guarantee won't apply anymore to
  242. existing CPU models. Management software that needs runnability
  243. guarantees must resolve the CPU model aliases using te
  244. ``alias-of'' field returned by the ``query-cpu-definitions'' QMP
  245. command.