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 Build options
  24 -------------
  25
  26 ``gprof`` builds (since 8.0)
  27 ''''''''''''''''''''''''''''
  28
  29 The ``--enable-gprof`` configure setting relies on compiler
  30 instrumentation to gather its data which can distort the generated
  31 profile. As other non-instrumenting tools are available that give a
  32 more holistic view of the system with non-instrumented binaries we are
  33 deprecating the build option and no longer defend it in CI. The
  34 ``--enable-gcov`` build option remains for analysis test case
  35 coverage.
  36
  37 System emulator command line arguments
  38 --------------------------------------
  39
  40 Creating sound card devices without ``audiodev=`` property (since 4.2)
  41 ''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
  42
  43 When not using the deprecated legacy audio config, each sound card
  44 should specify an ``audiodev=`` property.
  45
  46 Short-form boolean options (since 6.0)
  47 ''''''''''''''''''''''''''''''''''''''
  48
  49 Boolean options such as ``share=on``/``share=off`` could be written
  50 in short form as ``share`` and ``noshare``.  This is now deprecated
  51 and will cause a warning.
  52
  53 ``delay`` option for socket character devices (since 6.0)
  54 '''''''''''''''''''''''''''''''''''''''''''''''''''''''''
  55
  56 The replacement for the ``nodelay`` short-form boolean option is ``nodelay=on``
  57 rather than ``delay=off``.
  58
  59 ``-smp`` ("parameter=0" SMP configurations) (since 6.2)
  60 '''''''''''''''''''''''''''''''''''''''''''''''''''''''
  61
  62 Specified CPU topology parameters must be greater than zero.
  63
  64 In the SMP configuration, users should either provide a CPU topology
  65 parameter with a reasonable value (greater than zero) or just omit it
  66 and QEMU will compute the missing value.
  67
  68 However, historically it was implicitly allowed for users to provide
  69 a parameter with zero value, which is meaningless and could also possibly
  70 cause unexpected results in the -smp parsing. So support for this kind of
  71 configurations (e.g. -smp 8,sockets=0) is deprecated since 6.2 and will
  72 be removed in the near future, users have to ensure that all the topology
  73 members described with -smp are greater than zero.
  74
  75 Plugin argument passing through ``arg=<string>`` (since 6.1)
  76 ''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
  77
  78 Passing TCG plugins arguments through ``arg=`` is redundant is makes the
  79 command-line less readable, especially when the argument itself consist of a
  80 name and a value, e.g. ``-plugin plugin_name,arg="arg_name=arg_value"``.
  81 Therefore, the usage of ``arg`` is redundant. Single-word arguments are treated
  82 as short-form boolean values, and passed to plugins as ``arg_name=on``.
  83 However, short-form booleans are deprecated and full explicit ``arg_name=on``
  84 form is preferred.
  85
  86 ``-no-hpet`` (since 8.0)
  87 ''''''''''''''''''''''''
  88
  89 The HPET setting has been turned into a machine property.
  90 Use ``-machine hpet=off`` instead.
  91
  92 ``-no-acpi`` (since 8.0)
  93 ''''''''''''''''''''''''
  94
  95 The ``-no-acpi`` setting has been turned into a machine property.
  96 Use ``-machine acpi=off`` instead.
  97
  98 ``-async-teardown`` (since 8.1)
  99 '''''''''''''''''''''''''''''''
 100
 101 Use ``-run-with async-teardown=on`` instead.
 102
 103 ``-chroot`` (since 8.1)
 104 '''''''''''''''''''''''
 105
 106 Use ``-run-with chroot=dir`` instead.
 107
 108 ``-singlestep`` (since 8.1)
 109 '''''''''''''''''''''''''''
 110
 111 The ``-singlestep`` option has been turned into an accelerator property,
 112 and given a name that better reflects what it actually does.
 113 Use ``-accel tcg,one-insn-per-tb=on`` instead.
 114
 115 User-mode emulator command line arguments
 116 -----------------------------------------
 117
 118 ``-singlestep`` (since 8.1)
 119 '''''''''''''''''''''''''''
 120
 121 The ``-singlestep`` option has been given a name that better reflects
 122 what it actually does. For both linux-user and bsd-user, use the
 123 new ``-one-insn-per-tb`` option instead.
 124
 125 QEMU Machine Protocol (QMP) commands
 126 ------------------------------------
 127
 128 ``blockdev-open-tray``, ``blockdev-close-tray`` argument ``device`` (since 2.8)
 129 '''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 130
 131 Use argument ``id`` instead.
 132
 133 ``eject`` argument ``device`` (since 2.8)
 134 '''''''''''''''''''''''''''''''''''''''''
 135
 136 Use argument ``id`` instead.
 137
 138 ``blockdev-change-medium`` argument ``device`` (since 2.8)
 139 ''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 140
 141 Use argument ``id`` instead.
 142
 143 ``block_set_io_throttle`` argument ``device`` (since 2.8)
 144 '''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 145
 146 Use argument ``id`` instead.
 147
 148 ``blockdev-add`` empty string argument ``backing`` (since 2.10)
 149 '''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 150
 151 Use argument value ``null`` instead.
 152
 153 ``block-commit`` arguments ``base`` and ``top`` (since 3.1)
 154 '''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 155
 156 Use arguments ``base-node`` and ``top-node`` instead.
 157
 158 ``nbd-server-add`` and ``nbd-server-remove`` (since 5.2)
 159 ''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 160
 161 Use the more generic commands ``block-export-add`` and ``block-export-del``
 162 instead.  As part of this deprecation, where ``nbd-server-add`` used a
 163 single ``bitmap``, the new ``block-export-add`` uses a list of ``bitmaps``.
 164
 165 ``query-qmp-schema`` return value member ``values`` (since 6.2)
 166 '''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 167
 168 Member ``values`` in return value elements with meta-type ``enum`` is
 169 deprecated.  Use ``members`` instead.
 170
 171 ``drive-backup`` (since 6.2)
 172 ''''''''''''''''''''''''''''
 173
 174 Use ``blockdev-backup`` in combination with ``blockdev-add`` instead.
 175 This change primarily separates the creation/opening process of the backup
 176 target with explicit, separate steps. ``blockdev-backup`` uses mostly the
 177 same arguments as ``drive-backup``, except the ``format`` and ``mode``
 178 options are removed in favor of using explicit ``blockdev-create`` and
 179 ``blockdev-add`` calls. See :doc:`/interop/live-block-operations` for
 180 details.
 181
 182 Incorrectly typed ``device_add`` arguments (since 6.2)
 183 ''''''''''''''''''''''''''''''''''''''''''''''''''''''
 184
 185 Due to shortcomings in the internal implementation of ``device_add``, QEMU
 186 incorrectly accepts certain invalid arguments: Any object or list arguments are
 187 silently ignored. Other argument types are not checked, but an implicit
 188 conversion happens, so that e.g. string values can be assigned to integer
 189 device properties or vice versa.
 190
 191 This is a bug in QEMU that will be fixed in the future so that previously
 192 accepted incorrect commands will return an error. Users should make sure that
 193 all arguments passed to ``device_add`` are consistent with the documented
 194 property types.
 195
 196 ``StatusInfo`` member ``singlestep`` (since 8.1)
 197 ''''''''''''''''''''''''''''''''''''''''''''''''
 198
 199 The ``singlestep`` member of the ``StatusInfo`` returned from the
 200 ``query-status`` command is deprecated. This member has a confusing
 201 name and it never did what the documentation claimed or what its name
 202 suggests. We do not believe that anybody is actually using the
 203 information provided in this member.
 204
 205 The information it reports is whether the TCG JIT is in "one
 206 instruction per translated block" mode (which can be set on the
 207 command line or via the HMP, but not via QMP). The information remains
 208 available via the HMP 'info jit' command.
 209
 210 QEMU Machine Protocol (QMP) events
 211 ----------------------------------
 212
 213 ``MEM_UNPLUG_ERROR`` (since 6.2)
 214 ''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 215
 216 Use the more generic event ``DEVICE_UNPLUG_GUEST_ERROR`` instead.
 217
 218 ``vcpu`` trace events (since 8.1)
 219 '''''''''''''''''''''''''''''''''
 220
 221 The ability to instrument QEMU helper functions with vCPU-aware trace
 222 points was removed in 7.0. However QMP still exposed the vcpu
 223 parameter. This argument has now been deprecated and the remaining
 224 remaining trace points that used it are selected just by name.
 225
 226 Human Monitor Protocol (HMP) commands
 227 -------------------------------------
 228
 229 ``singlestep`` (since 8.1)
 230 ''''''''''''''''''''''''''
 231
 232 The ``singlestep`` command has been replaced by the ``one-insn-per-tb``
 233 command, which has the same behaviour but a less misleading name.
 234
 235 Host Architectures
 236 ------------------
 237
 238 BE MIPS (since 7.2)
 239 '''''''''''''''''''
 240
 241 As Debian 10 ("Buster") moved into LTS the big endian 32 bit version of
 242 MIPS moved out of support making it hard to maintain our
 243 cross-compilation CI tests of the architecture. As we no longer have
 244 CI coverage support may bitrot away before the deprecation process
 245 completes. The little endian variants of MIPS (both 32 and 64 bit) are
 246 still a supported host architecture.
 247
 248 System emulation on 32-bit x86 hosts (since 8.0)
 249 ''''''''''''''''''''''''''''''''''''''''''''''''
 250
 251 Support for 32-bit x86 host deployments is increasingly uncommon in mainstream
 252 OS distributions given the widespread availability of 64-bit x86 hardware.
 253 The QEMU project no longer considers 32-bit x86 support for system emulation to
 254 be an effective use of its limited resources, and thus intends to discontinue
 255 it. Since all recent x86 hardware from the past >10 years is capable of the
 256 64-bit x86 extensions, a corresponding 64-bit OS should be used instead.
 257
 258
 259 System emulator machines
 260 ------------------------
 261
 262 Arm ``virt`` machine ``dtb-kaslr-seed`` property (since 7.1)
 263 ''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 264
 265 The ``dtb-kaslr-seed`` property on the ``virt`` board has been
 266 deprecated; use the new name ``dtb-randomness`` instead. The new name
 267 better reflects the way this property affects all random data within
 268 the device tree blob, not just the ``kaslr-seed`` node.
 269
 270 Backend options
 271 ---------------
 272
 273 Using non-persistent backing file with pmem=on (since 6.1)
 274 ''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
 275
 276 This option is used when ``memory-backend-file`` is consumed by emulated NVDIMM
 277 device. However enabling ``memory-backend-file.pmem`` option, when backing file
 278 is (a) not DAX capable or (b) not on a filesystem that support direct mapping
 279 of persistent memory, is not safe and may lead to data loss or corruption in case
 280 of host crash.
 281 Options are:
 282
 283     - modify VM configuration to set ``pmem=off`` to continue using fake NVDIMM
 284       (without persistence guaranties) with backing file on non DAX storage
 285     - move backing file to NVDIMM storage and keep ``pmem=on``
 286       (to have NVDIMM with persistence guaranties).
 287
 288 Device options
 289 --------------
 290
 291 Emulated device options
 292 '''''''''''''''''''''''
 293
 294 ``-device virtio-blk,scsi=on|off`` (since 5.0)
 295 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 296
 297 The virtio-blk SCSI passthrough feature is a legacy VIRTIO feature.  VIRTIO 1.0
 298 and later do not support it because the virtio-scsi device was introduced for
 299 full SCSI support.  Use virtio-scsi instead when SCSI passthrough is required.
 300
 301 Note this also applies to ``-device virtio-blk-pci,scsi=on|off``, which is an
 302 alias.
 303
 304 ``-device nvme-ns,eui64-default=on|off`` (since 7.1)
 305 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 306
 307 In QEMU versions 6.1, 6.2 and 7.0, the ``nvme-ns`` generates an EUI-64
 308 identifier that is not globally unique. If an EUI-64 identifier is required, the
 309 user must set it explicitly using the ``nvme-ns`` device parameter ``eui64``.
 310
 311 ``-device nvme,use-intel-id=on|off`` (since 7.1)
 312 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 313
 314 The ``nvme`` device originally used a PCI Vendor/Device Identifier combination
 315 from Intel that was not properly allocated. Since version 5.2, the controller
 316 has used a properly allocated identifier. Deprecate the ``use-intel-id``
 317 machine compatibility parameter.
 318
 319 ``-device cxl-type3,memdev=xxxx`` (since 8.0)
 320 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 321
 322 The ``cxl-type3`` device initially only used a single memory backend.  With
 323 the addition of volatile memory support, it is now necessary to distinguish
 324 between persistent and volatile memory backends.  As such, memdev is deprecated
 325 in favor of persistent-memdev.
 326
 327 ``-fsdev proxy`` and ``-virtfs proxy`` (since 8.1)
 328 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 329
 330 The 9p ``proxy`` filesystem backend driver has been deprecated and will be
 331 removed (along with its proxy helper daemon) in a future version of QEMU. Please
 332 use ``-fsdev local`` or ``-virtfs local`` for using the 9p ``local`` filesystem
 333 backend, or alternatively consider deploying virtiofsd instead.
 334
 335 The 9p ``proxy`` backend was originally developed as an alternative to the 9p
 336 ``local`` backend. The idea was to enhance security by dispatching actual low
 337 level filesystem operations from 9p server (QEMU process) over to a separate
 338 process (the virtfs-proxy-helper binary). However this alternative never gained
 339 momentum. The proxy backend is much slower than the local backend, hasn't seen
 340 any development in years, and showed to be less secure, especially due to the
 341 fact that its helper daemon must be run as root, whereas with the local backend
 342 QEMU is typically run as unprivileged user and allows to tighten behaviour by
 343 mapping permissions et al by using its 'mapped' security model option.
 344
 345 Nowadays it would make sense to reimplement the ``proxy`` backend by using
 346 QEMU's ``vhost`` feature, which would eliminate the high latency costs under
 347 which the 9p ``proxy`` backend currently suffers. However as of to date nobody
 348 has indicated plans for such kind of reimplementation unfortunately.
 349
 350
 351 Block device options
 352 ''''''''''''''''''''
 353
 354 ``"backing": ""`` (since 2.12)
 355 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 356
 357 In order to prevent QEMU from automatically opening an image's backing
 358 chain, use ``"backing": null`` instead.
 359
 360 ``rbd`` keyvalue pair encoded filenames: ``""`` (since 3.1)
 361 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 362
 363 Options for ``rbd`` should be specified according to its runtime options,
 364 like other block drivers.  Legacy parsing of keyvalue pair encoded
 365 filenames is useful to open images with the old format for backing files;
 366 These image files should be updated to use the current format.
 367
 368 Example of legacy encoding::
 369
 370   json:{"file.driver":"rbd", "file.filename":"rbd:rbd/name"}
 371
 372 The above, converted to the current supported format::
 373
 374   json:{"file.driver":"rbd", "file.pool":"rbd", "file.image":"name"}
 375
 376 ``iscsi,password=xxx`` (since 8.0)
 377 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 378
 379 Specifying the iSCSI password in plain text on the command line using the
 380 ``password`` option is insecure. The ``password-secret`` option should be
 381 used instead, to refer to a ``--object secret...`` instance that provides
 382 a password via a file, or encrypted.
 383
 384 Backwards compatibility
 385 -----------------------
 386
 387 Runnability guarantee of CPU models (since 4.1)
 388 '''''''''''''''''''''''''''''''''''''''''''''''
 389
 390 Previous versions of QEMU never changed existing CPU models in
 391 ways that introduced additional host software or hardware
 392 requirements to the VM.  This allowed management software to
 393 safely change the machine type of an existing VM without
 394 introducing new requirements ("runnability guarantee").  This
 395 prevented CPU models from being updated to include CPU
 396 vulnerability mitigations, leaving guests vulnerable in the
 397 default configuration.
 398
 399 The CPU model runnability guarantee won't apply anymore to
 400 existing CPU models.  Management software that needs runnability
 401 guarantees must resolve the CPU model aliases using the
 402 ``alias-of`` field returned by the ``query-cpu-definitions`` QMP
 403 command.
 404
 405 While those guarantees are kept, the return value of
 406 ``query-cpu-definitions`` will have existing CPU model aliases
 407 point to a version that doesn't break runnability guarantees
 408 (specifically, version 1 of those CPU models).  In future QEMU
 409 versions, aliases will point to newer CPU model versions
 410 depending on the machine type, so management software must
 411 resolve CPU model aliases before starting a virtual machine.
 412
 413 QEMU guest agent
 414 ----------------
 415
 416 ``--blacklist`` command line option (since 7.2)
 417 '''''''''''''''''''''''''''''''''''''''''''''''
 418
 419 ``--blacklist`` has been replaced by ``--block-rpcs`` (which is a better
 420 wording for what this option does). The short form ``-b`` still stays
 421 the same and thus is the preferred way for scripts that should run with
 422 both, older and future versions of QEMU.
 423
 424 ``blacklist`` config file option (since 7.2)
 425 ''''''''''''''''''''''''''''''''''''''''''''
 426
 427 The ``blacklist`` config file option has been renamed to ``block-rpcs``
 428 (to be in sync with the renaming of the corresponding command line
 429 option).
 430
 431 Migration
 432 ---------
 433
 434 ``skipped`` MigrationStats field (since 8.1)
 435 ''''''''''''''''''''''''''''''''''''''''''''
 436
 437 ``skipped`` field in Migration stats has been deprecated.  It hasn't
 438 been used for more than 10 years.
 439