xref: /qemu/docs/about/deprecated.rst (revision e4418354)
1.. _Deprecated features:
2
3Deprecated features
4===================
5
6In general features are intended to be supported indefinitely once
7introduced into QEMU. In the event that a feature needs to be removed,
8it will be listed in this section. The feature will remain functional for the
9release in which it was deprecated and one further release. After these two
10releases, the feature is liable to be removed. Deprecated features may also
11generate warnings on the console when QEMU starts up, or if activated via a
12monitor command, however, this is not a mandatory requirement.
13
14Prior to the 2.10.0 release there was no official policy on how
15long features would be deprecated prior to their removal, nor
16any documented list of which features were deprecated. Thus
17any features deprecated prior to 2.10.0 will be treated as if
18they were first deprecated in the 2.10.0 release.
19
20What follows is a list of all features currently marked as
21deprecated.
22
23System emulator command line arguments
24--------------------------------------
25
26``QEMU_AUDIO_`` environment variables and ``-audio-help`` (since 4.0)
27'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
28
29The ``-audiodev`` argument is now the preferred way to specify audio
30backend settings instead of environment variables.  To ease migration to
31the new format, the ``-audiodev-help`` option can be used to convert
32the current values of the environment variables to ``-audiodev`` options.
33
34Creating sound card devices and vnc without ``audiodev=`` property (since 4.2)
35''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
36
37When not using the deprecated legacy audio config, each sound card
38should specify an ``audiodev=`` property.  Additionally, when using
39vnc, you should specify an ``audiodev=`` property if you plan to
40transmit audio through the VNC protocol.
41
42Short-form boolean options (since 6.0)
43''''''''''''''''''''''''''''''''''''''
44
45Boolean options such as ``share=on``/``share=off`` could be written
46in short form as ``share`` and ``noshare``.  This is now deprecated
47and will cause a warning.
48
49``delay`` option for socket character devices (since 6.0)
50'''''''''''''''''''''''''''''''''''''''''''''''''''''''''
51
52The replacement for the ``nodelay`` short-form boolean option is ``nodelay=on``
53rather than ``delay=off``.
54
55``-spice password=string`` (since 6.0)
56''''''''''''''''''''''''''''''''''''''
57
58This option is insecure because the SPICE password remains visible in
59the process listing. This is replaced by the new ``password-secret``
60option which lets the password be securely provided on the command
61line using a ``secret`` object instance.
62
63``-smp`` ("parameter=0" SMP configurations) (since 6.2)
64'''''''''''''''''''''''''''''''''''''''''''''''''''''''
65
66Specified CPU topology parameters must be greater than zero.
67
68In the SMP configuration, users should either provide a CPU topology
69parameter with a reasonable value (greater than zero) or just omit it
70and QEMU will compute the missing value.
71
72However, historically it was implicitly allowed for users to provide
73a parameter with zero value, which is meaningless and could also possibly
74cause unexpected results in the -smp parsing. So support for this kind of
75configurations (e.g. -smp 8,sockets=0) is deprecated since 6.2 and will
76be removed in the near future, users have to ensure that all the topology
77members described with -smp are greater than zero.
78
79Plugin argument passing through ``arg=<string>`` (since 6.1)
80''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
81
82Passing TCG plugins arguments through ``arg=`` is redundant is makes the
83command-line less readable, especially when the argument itself consist of a
84name and a value, e.g. ``-plugin plugin_name,arg="arg_name=arg_value"``.
85Therefore, the usage of ``arg`` is redundant. Single-word arguments are treated
86as short-form boolean values, and passed to plugins as ``arg_name=on``.
87However, short-form booleans are deprecated and full explicit ``arg_name=on``
88form is preferred.
89
90``-drive if=none`` for the sifive_u OTP device (since 6.2)
91''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
92
93Using ``-drive if=none`` to configure the OTP device of the sifive_u
94RISC-V machine is deprecated. Use ``-drive if=pflash`` instead.
95
96``-no-hpet`` (since 8.0)
97''''''''''''''''''''''''
98
99The HPET setting has been turned into a machine property.
100Use ``-machine hpet=off`` instead.
101
102
103QEMU Machine Protocol (QMP) commands
104------------------------------------
105
106``blockdev-open-tray``, ``blockdev-close-tray`` argument ``device`` (since 2.8)
107'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
108
109Use argument ``id`` instead.
110
111``eject`` argument ``device`` (since 2.8)
112'''''''''''''''''''''''''''''''''''''''''
113
114Use argument ``id`` instead.
115
116``blockdev-change-medium`` argument ``device`` (since 2.8)
117''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
118
119Use argument ``id`` instead.
120
121``block_set_io_throttle`` argument ``device`` (since 2.8)
122'''''''''''''''''''''''''''''''''''''''''''''''''''''''''
123
124Use argument ``id`` instead.
125
126``blockdev-add`` empty string argument ``backing`` (since 2.10)
127'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
128
129Use argument value ``null`` instead.
130
131``block-commit`` arguments ``base`` and ``top`` (since 3.1)
132'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
133
134Use arguments ``base-node`` and ``top-node`` instead.
135
136``nbd-server-add`` and ``nbd-server-remove`` (since 5.2)
137''''''''''''''''''''''''''''''''''''''''''''''''''''''''
138
139Use the more generic commands ``block-export-add`` and ``block-export-del``
140instead.  As part of this deprecation, where ``nbd-server-add`` used a
141single ``bitmap``, the new ``block-export-add`` uses a list of ``bitmaps``.
142
143``query-qmp-schema`` return value member ``values`` (since 6.2)
144'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
145
146Member ``values`` in return value elements with meta-type ``enum`` is
147deprecated.  Use ``members`` instead.
148
149``drive-backup`` (since 6.2)
150''''''''''''''''''''''''''''
151
152Use ``blockdev-backup`` in combination with ``blockdev-add`` instead.
153This change primarily separates the creation/opening process of the backup
154target with explicit, separate steps. ``blockdev-backup`` uses mostly the
155same arguments as ``drive-backup``, except the ``format`` and ``mode``
156options are removed in favor of using explicit ``blockdev-create`` and
157``blockdev-add`` calls. See :doc:`/interop/live-block-operations` for
158details.
159
160Incorrectly typed ``device_add`` arguments (since 6.2)
161''''''''''''''''''''''''''''''''''''''''''''''''''''''
162
163Due to shortcomings in the internal implementation of ``device_add``, QEMU
164incorrectly accepts certain invalid arguments: Any object or list arguments are
165silently ignored. Other argument types are not checked, but an implicit
166conversion happens, so that e.g. string values can be assigned to integer
167device properties or vice versa.
168
169This is a bug in QEMU that will be fixed in the future so that previously
170accepted incorrect commands will return an error. Users should make sure that
171all arguments passed to ``device_add`` are consistent with the documented
172property types.
173
174System accelerators
175-------------------
176
177MIPS ``Trap-and-Emul`` KVM support (since 6.0)
178''''''''''''''''''''''''''''''''''''''''''''''
179
180The MIPS ``Trap-and-Emul`` KVM host and guest support has been removed
181from Linux upstream kernel, declare it deprecated.
182
183Host Architectures
184------------------
185
186BE MIPS (since 7.2)
187'''''''''''''''''''
188
189As Debian 10 ("Buster") moved into LTS the big endian 32 bit version of
190MIPS moved out of support making it hard to maintain our
191cross-compilation CI tests of the architecture. As we no longer have
192CI coverage support may bitrot away before the deprecation process
193completes. The little endian variants of MIPS (both 32 and 64 bit) are
194still a supported host architecture.
195
196QEMU API (QAPI) events
197----------------------
198
199``MEM_UNPLUG_ERROR`` (since 6.2)
200''''''''''''''''''''''''''''''''''''''''''''''''''''''''
201
202Use the more generic event ``DEVICE_UNPLUG_GUEST_ERROR`` instead.
203
204
205System emulator machines
206------------------------
207
208Arm ``virt`` machine ``dtb-kaslr-seed`` property
209''''''''''''''''''''''''''''''''''''''''''''''''
210
211The ``dtb-kaslr-seed`` property on the ``virt`` board has been
212deprecated; use the new name ``dtb-randomness`` instead. The new name
213better reflects the way this property affects all random data within
214the device tree blob, not just the ``kaslr-seed`` node.
215
216``pc-i440fx-1.4`` up to ``pc-i440fx-1.7`` (since 7.0)
217'''''''''''''''''''''''''''''''''''''''''''''''''''''
218
219These old machine types are quite neglected nowadays and thus might have
220various pitfalls with regards to live migration. Use a newer machine type
221instead.
222
223
224Backend options
225---------------
226
227Using non-persistent backing file with pmem=on (since 6.1)
228''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
229
230This option is used when ``memory-backend-file`` is consumed by emulated NVDIMM
231device. However enabling ``memory-backend-file.pmem`` option, when backing file
232is (a) not DAX capable or (b) not on a filesystem that support direct mapping
233of persistent memory, is not safe and may lead to data loss or corruption in case
234of host crash.
235Options are:
236
237    - modify VM configuration to set ``pmem=off`` to continue using fake NVDIMM
238      (without persistence guaranties) with backing file on non DAX storage
239    - move backing file to NVDIMM storage and keep ``pmem=on``
240      (to have NVDIMM with persistence guaranties).
241
242Device options
243--------------
244
245Emulated device options
246'''''''''''''''''''''''
247
248``-device virtio-blk,scsi=on|off`` (since 5.0)
249^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
250
251The virtio-blk SCSI passthrough feature is a legacy VIRTIO feature.  VIRTIO 1.0
252and later do not support it because the virtio-scsi device was introduced for
253full SCSI support.  Use virtio-scsi instead when SCSI passthrough is required.
254
255Note this also applies to ``-device virtio-blk-pci,scsi=on|off``, which is an
256alias.
257
258``-device sga`` (since 6.2)
259^^^^^^^^^^^^^^^^^^^^^^^^^^^
260
261The ``sga`` device loads an option ROM for x86 targets which enables
262SeaBIOS to send messages to the serial console. SeaBIOS 1.11.0 onwards
263contains native support for this feature and thus use of the option
264ROM approach is obsolete. The native SeaBIOS support can be activated
265by using ``-machine graphics=off``.
266
267``-device nvme-ns,eui64-default=on|off`` (since 7.1)
268^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
269
270In QEMU versions 6.1, 6.2 and 7.0, the ``nvme-ns`` generates an EUI-64
271identifier that is not globally unique. If an EUI-64 identifier is required, the
272user must set it explicitly using the ``nvme-ns`` device parameter ``eui64``.
273
274``-device nvme,use-intel-id=on|off`` (since 7.1)
275^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
276
277The ``nvme`` device originally used a PCI Vendor/Device Identifier combination
278from Intel that was not properly allocated. Since version 5.2, the controller
279has used a properly allocated identifier. Deprecate the ``use-intel-id``
280machine compatibility parameter.
281
282
283Block device options
284''''''''''''''''''''
285
286``"backing": ""`` (since 2.12)
287^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
288
289In order to prevent QEMU from automatically opening an image's backing
290chain, use ``"backing": null`` instead.
291
292``rbd`` keyvalue pair encoded filenames: ``""`` (since 3.1)
293^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
294
295Options for ``rbd`` should be specified according to its runtime options,
296like other block drivers.  Legacy parsing of keyvalue pair encoded
297filenames is useful to open images with the old format for backing files;
298These image files should be updated to use the current format.
299
300Example of legacy encoding::
301
302  json:{"file.driver":"rbd", "file.filename":"rbd:rbd/name"}
303
304The above, converted to the current supported format::
305
306  json:{"file.driver":"rbd", "file.pool":"rbd", "file.image":"name"}
307
308Backwards compatibility
309-----------------------
310
311Runnability guarantee of CPU models (since 4.1)
312'''''''''''''''''''''''''''''''''''''''''''''''
313
314Previous versions of QEMU never changed existing CPU models in
315ways that introduced additional host software or hardware
316requirements to the VM.  This allowed management software to
317safely change the machine type of an existing VM without
318introducing new requirements ("runnability guarantee").  This
319prevented CPU models from being updated to include CPU
320vulnerability mitigations, leaving guests vulnerable in the
321default configuration.
322
323The CPU model runnability guarantee won't apply anymore to
324existing CPU models.  Management software that needs runnability
325guarantees must resolve the CPU model aliases using the
326``alias-of`` field returned by the ``query-cpu-definitions`` QMP
327command.
328
329While those guarantees are kept, the return value of
330``query-cpu-definitions`` will have existing CPU model aliases
331point to a version that doesn't break runnability guarantees
332(specifically, version 1 of those CPU models).  In future QEMU
333versions, aliases will point to newer CPU model versions
334depending on the machine type, so management software must
335resolve CPU model aliases before starting a virtual machine.
336
337Tools
338-----
339
340virtiofsd
341'''''''''
342
343There is a new Rust implementation of ``virtiofsd`` at
344``https://gitlab.com/virtio-fs/virtiofsd``;
345since this is now marked stable, new development should be done on that
346rather than the existing C version in the QEMU tree.
347The C version will still accept fixes and patches that
348are already in development for the moment, but will eventually
349be deleted from this tree.
350New deployments should use the Rust version, and existing systems
351should consider moving to it.  The command line and feature set
352is very close and moving should be simple.
353
354
355QEMU guest agent
356----------------
357
358``--blacklist`` command line option (since 7.2)
359'''''''''''''''''''''''''''''''''''''''''''''''
360
361``--blacklist`` has been replaced by ``--block-rpcs`` (which is a better
362wording for what this option does). The short form ``-b`` still stays
363the same and thus is the preferred way for scripts that should run with
364both, older and future versions of QEMU.
365
366``blacklist`` config file option (since 7.2)
367''''''''''''''''''''''''''''''''''''''''''''
368
369The ``blacklist`` config file option has been renamed to ``block-rpcs``
370(to be in sync with the renaming of the corresponding command line
371option).
372