docs/about/deprecated.rst

   1 .. _Deprecated features:
   2
   3 Deprecated features
   4 ===================
   5
   6 In general features are intended to be supported indefinitely once
   7 introduced into QEMU. In the event that a feature needs to be removed,
   8 it will be listed in this section. The feature will remain functional for the
   9 release in which it was deprecated and one further release. After these two
  10 releases, the feature is liable to be removed. Deprecated features may also
  11 generate warnings on the console when QEMU starts up, or if activated via a
  12 monitor command, however, this is not a mandatory requirement.
  13
  14 Prior to the 2.10.0 release there was no official policy on how
  15 long features would be deprecated prior to their removal, nor
  16 any documented list of which features were deprecated. Thus
  17 any features deprecated prior to 2.10.0 will be treated as if
  18 they were first deprecated in the 2.10.0 release.
  19
  20 What follows is a list of all features currently marked as
  21 deprecated.
  22
  23 System emulator command line arguments
  24 --------------------------------------
  25
  26 Creating sound card devices without ``audiodev=`` property (since 4.2)
  27 ''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
  28
  29 When not using the deprecated legacy audio config, each sound card
  30 should specify an ``audiodev=`` property.
  31
  32 Short-form boolean options (since 6.0)
  33 ''''''''''''''''''''''''''''''''''''''
  34
  35 Boolean options such as ``share=on``/``share=off`` could be written
  36 in short form as ``share`` and ``noshare``.  This is now deprecated
  37 and will cause a warning.
  38
  39 ``delay`` option for socket character devices (since 6.0)
  40 '''''''''''''''''''''''''''''''''''''''''''''''''''''''''
  41
  42 The replacement for the ``nodelay`` short-form boolean option is ``nodelay=on``
  43 rather than ``delay=off``.
  44
  45 ``-smp`` ("parameter=0" SMP configurations) (since 6.2)
  46 '''''''''''''''''''''''''''''''''''''''''''''''''''''''
  47
  48 Specified CPU topology parameters must be greater than zero.
  49
  50 In the SMP configuration, users should either provide a CPU topology
  51 parameter with a reasonable value (greater than zero) or just omit it
  52 and QEMU will compute the missing value.
  53
  54 However, historically it was implicitly allowed for users to provide
  55 a parameter with zero value, which is meaningless and could also possibly
  56 cause unexpected results in the -smp parsing. So support for this kind of
  57 configurations (e.g. -smp 8,sockets=0) is deprecated since 6.2 and will
  58 be removed in the near future, users have to ensure that all the topology
  59 members described with -smp are greater than zero.
  60
  61 Plugin argument passing through ``arg=<string>`` (since 6.1)
  62 ''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
  63
  64 Passing TCG plugins arguments through ``arg=`` is redundant is makes the
  65 command-line less readable, especially when the argument itself consist of a
  66 name and a value, e.g. ``-plugin plugin_name,arg="arg_name=arg_value"``.
  67 Therefore, the usage of ``arg`` is redundant. Single-word arguments are treated
  68 as short-form boolean values, and passed to plugins as ``arg_name=on``.
  69 However, short-form booleans are deprecated and full explicit ``arg_name=on``
  70 form is preferred.
  71
  72 ``-no-hpet`` (since 8.0)
  73 ''''''''''''''''''''''''
  74
  75 The HPET setting has been turned into a machine property.
  76 Use ``-machine hpet=off`` instead.
  77
  78 ``-no-acpi`` (since 8.0)
  79 ''''''''''''''''''''''''
  80
  81 The ``-no-acpi`` setting has been turned into a machine property.
  82 Use ``-machine acpi=off`` instead.
  83
  84 ``-async-teardown`` (since 8.1)
  85 '''''''''''''''''''''''''''''''
  86
  87 Use ``-run-with async-teardown=on`` instead.
  88
  89 ``-chroot`` (since 8.1)
  90 '''''''''''''''''''''''
  91
  92 Use ``-run-with chroot=dir`` instead.
  93
  94 ``-singlestep`` (since 8.1)
  95 '''''''''''''''''''''''''''
  96
  97 The ``-singlestep`` option has been turned into an accelerator property,
  98 and given a name that better reflects what it actually does.
  99 Use ``-accel tcg,one-insn-per-tb=on`` instead.
 100
 101 User-mode emulator command line arguments
 102 -----------------------------------------
 103
 104 ``-singlestep`` (since 8.1)
 105 '''''''''''''''''''''''''''
 106
 107 The ``-singlestep`` option has been given a name that better reflects
 108 what it actually does. For both linux-user and bsd-user, use the
 109 new ``-one-insn-per-tb`` option instead.
 110
 111 QEMU Machine Protocol (QMP) commands
 112 ------------------------------------
 113
 114 ``blockdev-open-tray``, ``blockdev-close-tray`` argument ``device`` (since 2.8)
 115 '''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 116
 117 Use argument ``id`` instead.
 118
 119 ``eject`` argument ``device`` (since 2.8)
 120 '''''''''''''''''''''''''''''''''''''''''
 121
 122 Use argument ``id`` instead.
 123
 124 ``blockdev-change-medium`` argument ``device`` (since 2.8)
 125 ''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 126
 127 Use argument ``id`` instead.
 128
 129 ``block_set_io_throttle`` argument ``device`` (since 2.8)
 130 '''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 131
 132 Use argument ``id`` instead.
 133
 134 ``blockdev-add`` empty string argument ``backing`` (since 2.10)
 135 '''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 136
 137 Use argument value ``null`` instead.
 138
 139 ``block-commit`` arguments ``base`` and ``top`` (since 3.1)
 140 '''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 141
 142 Use arguments ``base-node`` and ``top-node`` instead.
 143
 144 ``nbd-server-add`` and ``nbd-server-remove`` (since 5.2)
 145 ''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 146
 147 Use the more generic commands ``block-export-add`` and ``block-export-del``
 148 instead.  As part of this deprecation, where ``nbd-server-add`` used a
 149 single ``bitmap``, the new ``block-export-add`` uses a list of ``bitmaps``.
 150
 151 ``query-qmp-schema`` return value member ``values`` (since 6.2)
 152 '''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 153
 154 Member ``values`` in return value elements with meta-type ``enum`` is
 155 deprecated.  Use ``members`` instead.
 156
 157 ``drive-backup`` (since 6.2)
 158 ''''''''''''''''''''''''''''
 159
 160 Use ``blockdev-backup`` in combination with ``blockdev-add`` instead.
 161 This change primarily separates the creation/opening process of the backup
 162 target with explicit, separate steps. ``blockdev-backup`` uses mostly the
 163 same arguments as ``drive-backup``, except the ``format`` and ``mode``
 164 options are removed in favor of using explicit ``blockdev-create`` and
 165 ``blockdev-add`` calls. See :doc:`/interop/live-block-operations` for
 166 details.
 167
 168 Incorrectly typed ``device_add`` arguments (since 6.2)
 169 ''''''''''''''''''''''''''''''''''''''''''''''''''''''
 170
 171 Due to shortcomings in the internal implementation of ``device_add``, QEMU
 172 incorrectly accepts certain invalid arguments: Any object or list arguments are
 173 silently ignored. Other argument types are not checked, but an implicit
 174 conversion happens, so that e.g. string values can be assigned to integer
 175 device properties or vice versa.
 176
 177 This is a bug in QEMU that will be fixed in the future so that previously
 178 accepted incorrect commands will return an error. Users should make sure that
 179 all arguments passed to ``device_add`` are consistent with the documented
 180 property types.
 181
 182 ``StatusInfo`` member ``singlestep`` (since 8.1)
 183 ''''''''''''''''''''''''''''''''''''''''''''''''
 184
 185 The ``singlestep`` member of the ``StatusInfo`` returned from the
 186 ``query-status`` command is deprecated. This member has a confusing
 187 name and it never did what the documentation claimed or what its name
 188 suggests. We do not believe that anybody is actually using the
 189 information provided in this member.
 190
 191 The information it reports is whether the TCG JIT is in "one
 192 instruction per translated block" mode (which can be set on the
 193 command line or via the HMP, but not via QMP). The information remains
 194 available via the HMP 'info jit' command.
 195
 196 QEMU Machine Protocol (QMP) events
 197 ----------------------------------
 198
 199 ``MEM_UNPLUG_ERROR`` (since 6.2)
 200 ''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 201
 202 Use the more generic event ``DEVICE_UNPLUG_GUEST_ERROR`` instead.
 203
 204 ``vcpu`` trace events (since 8.1)
 205 '''''''''''''''''''''''''''''''''
 206
 207 The ability to instrument QEMU helper functions with vCPU-aware trace
 208 points was removed in 7.0. However QMP still exposed the vcpu
 209 parameter. This argument has now been deprecated and the remaining
 210 remaining trace points that used it are selected just by name.
 211
 212 Human Monitor Protocol (HMP) commands
 213 -------------------------------------
 214
 215 ``singlestep`` (since 8.1)
 216 ''''''''''''''''''''''''''
 217
 218 The ``singlestep`` command has been replaced by the ``one-insn-per-tb``
 219 command, which has the same behaviour but a less misleading name.
 220
 221 Host Architectures
 222 ------------------
 223
 224 BE MIPS (since 7.2)
 225 '''''''''''''''''''
 226
 227 As Debian 10 ("Buster") moved into LTS the big endian 32 bit version of
 228 MIPS moved out of support making it hard to maintain our
 229 cross-compilation CI tests of the architecture. As we no longer have
 230 CI coverage support may bitrot away before the deprecation process
 231 completes. The little endian variants of MIPS (both 32 and 64 bit) are
 232 still a supported host architecture.
 233
 234 System emulation on 32-bit x86 hosts (since 8.0)
 235 ''''''''''''''''''''''''''''''''''''''''''''''''
 236
 237 Support for 32-bit x86 host deployments is increasingly uncommon in mainstream
 238 OS distributions given the widespread availability of 64-bit x86 hardware.
 239 The QEMU project no longer considers 32-bit x86 support for system emulation to
 240 be an effective use of its limited resources, and thus intends to discontinue
 241 it. Since all recent x86 hardware from the past >10 years is capable of the
 242 64-bit x86 extensions, a corresponding 64-bit OS should be used instead.
 243
 244
 245 System emulator machines
 246 ------------------------
 247
 248 Arm ``virt`` machine ``dtb-kaslr-seed`` property (since 7.1)
 249 ''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 250
 251 The ``dtb-kaslr-seed`` property on the ``virt`` board has been
 252 deprecated; use the new name ``dtb-randomness`` instead. The new name
 253 better reflects the way this property affects all random data within
 254 the device tree blob, not just the ``kaslr-seed`` node.
 255
 256 Backend options
 257 ---------------
 258
 259 Using non-persistent backing file with pmem=on (since 6.1)
 260 ''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 261
 262 This option is used when ``memory-backend-file`` is consumed by emulated NVDIMM
 263 device. However enabling ``memory-backend-file.pmem`` option, when backing file
 264 is (a) not DAX capable or (b) not on a filesystem that support direct mapping
 265 of persistent memory, is not safe and may lead to data loss or corruption in case
 266 of host crash.
 267 Options are:
 268
 269     - modify VM configuration to set ``pmem=off`` to continue using fake NVDIMM
 270       (without persistence guaranties) with backing file on non DAX storage
 271     - move backing file to NVDIMM storage and keep ``pmem=on``
 272       (to have NVDIMM with persistence guaranties).
 273
 274 Device options
 275 --------------
 276
 277 Emulated device options
 278 '''''''''''''''''''''''
 279
 280 ``-device virtio-blk,scsi=on|off`` (since 5.0)
 281 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 282
 283 The virtio-blk SCSI passthrough feature is a legacy VIRTIO feature.  VIRTIO 1.0
 284 and later do not support it because the virtio-scsi device was introduced for
 285 full SCSI support.  Use virtio-scsi instead when SCSI passthrough is required.
 286
 287 Note this also applies to ``-device virtio-blk-pci,scsi=on|off``, which is an
 288 alias.
 289
 290 ``-device nvme-ns,eui64-default=on|off`` (since 7.1)
 291 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 292
 293 In QEMU versions 6.1, 6.2 and 7.0, the ``nvme-ns`` generates an EUI-64
 294 identifier that is not globally unique. If an EUI-64 identifier is required, the
 295 user must set it explicitly using the ``nvme-ns`` device parameter ``eui64``.
 296
 297 ``-device nvme,use-intel-id=on|off`` (since 7.1)
 298 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 299
 300 The ``nvme`` device originally used a PCI Vendor/Device Identifier combination
 301 from Intel that was not properly allocated. Since version 5.2, the controller
 302 has used a properly allocated identifier. Deprecate the ``use-intel-id``
 303 machine compatibility parameter.
 304
 305 ``-device cxl-type3,memdev=xxxx`` (since 8.0)
 306 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 307
 308 The ``cxl-type3`` device initially only used a single memory backend.  With
 309 the addition of volatile memory support, it is now necessary to distinguish
 310 between persistent and volatile memory backends.  As such, memdev is deprecated
 311 in favor of persistent-memdev.
 312
 313 ``-fsdev proxy`` and ``-virtfs proxy`` (since 8.1)
 314 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 315
 316 The 9p ``proxy`` filesystem backend driver has been deprecated and will be
 317 removed (along with its proxy helper daemon) in a future version of QEMU. Please
 318 use ``-fsdev local`` or ``-virtfs local`` for using the 9p ``local`` filesystem
 319 backend, or alternatively consider deploying virtiofsd instead.
 320
 321 The 9p ``proxy`` backend was originally developed as an alternative to the 9p
 322 ``local`` backend. The idea was to enhance security by dispatching actual low
 323 level filesystem operations from 9p server (QEMU process) over to a separate
 324 process (the virtfs-proxy-helper binary). However this alternative never gained
 325 momentum. The proxy backend is much slower than the local backend, hasn't seen
 326 any development in years, and showed to be less secure, especially due to the
 327 fact that its helper daemon must be run as root, whereas with the local backend
 328 QEMU is typically run as unprivileged user and allows to tighten behaviour by
 329 mapping permissions et al by using its 'mapped' security model option.
 330
 331 Nowadays it would make sense to reimplement the ``proxy`` backend by using
 332 QEMU's ``vhost`` feature, which would eliminate the high latency costs under
 333 which the 9p ``proxy`` backend currently suffers. However as of to date nobody
 334 has indicated plans for such kind of reimplementation unfortunately.
 335
 336
 337 Block device options
 338 ''''''''''''''''''''
 339
 340 ``"backing": ""`` (since 2.12)
 341 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 342
 343 In order to prevent QEMU from automatically opening an image's backing
 344 chain, use ``"backing": null`` instead.
 345
 346 ``rbd`` keyvalue pair encoded filenames: ``""`` (since 3.1)
 347 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 348
 349 Options for ``rbd`` should be specified according to its runtime options,
 350 like other block drivers.  Legacy parsing of keyvalue pair encoded
 351 filenames is useful to open images with the old format for backing files;
 352 These image files should be updated to use the current format.
 353
 354 Example of legacy encoding::
 355
 356   json:{"file.driver":"rbd", "file.filename":"rbd:rbd/name"}
 357
 358 The above, converted to the current supported format::
 359
 360   json:{"file.driver":"rbd", "file.pool":"rbd", "file.image":"name"}
 361
 362 ``iscsi,password=xxx`` (since 8.0)
 363 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 364
 365 Specifying the iSCSI password in plain text on the command line using the
 366 ``password`` option is insecure. The ``password-secret`` option should be
 367 used instead, to refer to a ``--object secret...`` instance that provides
 368 a password via a file, or encrypted.
 369
 370 Backwards compatibility
 371 -----------------------
 372
 373 Runnability guarantee of CPU models (since 4.1)
 374 '''''''''''''''''''''''''''''''''''''''''''''''
 375
 376 Previous versions of QEMU never changed existing CPU models in
 377 ways that introduced additional host software or hardware
 378 requirements to the VM.  This allowed management software to
 379 safely change the machine type of an existing VM without
 380 introducing new requirements ("runnability guarantee").  This
 381 prevented CPU models from being updated to include CPU
 382 vulnerability mitigations, leaving guests vulnerable in the
 383 default configuration.
 384
 385 The CPU model runnability guarantee won't apply anymore to
 386 existing CPU models.  Management software that needs runnability
 387 guarantees must resolve the CPU model aliases using the
 388 ``alias-of`` field returned by the ``query-cpu-definitions`` QMP
 389 command.
 390
 391 While those guarantees are kept, the return value of
 392 ``query-cpu-definitions`` will have existing CPU model aliases
 393 point to a version that doesn't break runnability guarantees
 394 (specifically, version 1 of those CPU models).  In future QEMU
 395 versions, aliases will point to newer CPU model versions
 396 depending on the machine type, so management software must
 397 resolve CPU model aliases before starting a virtual machine.
 398
 399 QEMU guest agent
 400 ----------------
 401
 402 ``--blacklist`` command line option (since 7.2)
 403 '''''''''''''''''''''''''''''''''''''''''''''''
 404
 405 ``--blacklist`` has been replaced by ``--block-rpcs`` (which is a better
 406 wording for what this option does). The short form ``-b`` still stays
 407 the same and thus is the preferred way for scripts that should run with
 408 both, older and future versions of QEMU.
 409
 410 ``blacklist`` config file option (since 7.2)
 411 ''''''''''''''''''''''''''''''''''''''''''''
 412
 413 The ``blacklist`` config file option has been renamed to ``block-rpcs``
 414 (to be in sync with the renaming of the corresponding command line
 415 option).
 416
 417 Migration
 418 ---------
 419
 420 ``skipped`` MigrationStats field (since 8.1)
 421 ''''''''''''''''''''''''''''''''''''''''''''
 422
 423 ``skipped`` field in Migration stats has been deprecated.  It hasn't
 424 been used for more than 10 years.
 425